1. Introduction - Firenze University Press

Proceedings e report 

90

ECOS 2012 

The 25 th International Conference on Efficiency, Cost, 

Optimization and Simulation of Energy Conversion 

Systems and Processes 

(Perugia, June 26 th -June 29 th , 2012) 

edited by 

Umberto Desideri, Giampaolo Manfrida, 

Enrico Sciubba 

firenze university press 

2012

ECOS 2012 : the 25 th International Conference on Efficiency, 

Cost, Optimization and Simulation of Energy Conversion 

Systems and Processes (Perugia, June 26 th -June 29 th , 2012) / 

edited by Umberto Desideri, Giampaolo Manfrida, Enrico 

Sciubba. – Firenze : Firenze University Press, 2012. 

(Proceedings e report ; 90) 

http://digital.casalini.it/9788866553229 

ISBN 978-88-6655-322-9 (online) 

Progetto grafico di copertina Alberto Pizarro, Pagina Maestra snc 

Immagine di copertina: © Kts | Dreamstime.com 

Peer Review Process 

All publications are submitted to an external refereeing process under the responsibility of the FUP 

Editorial Board and the Scientific Committees of the individual series. The works published in the 

FUP catalogue are evaluated and approved by the Editorial Board of the publishing house. For a 

more detailed description of the refereeing process we refer to the official documents published on 

the website and in the online catalogue of the FUP (http://www.fupress.com). 

Firenze University Press Editorial Board 

G. Nigro (Co-ordinator), M.T. Bartoli, M. Boddi, F. Cambi, R. Casalbuoni, C. Ciappei, R. Del Punta, 

A. Dolfi, V. Fargion, S. Ferrone, M. Garzaniti, P. Guarnieri, G. Mari, M. Marini, M. Verga, A. Zorzi. 

© 2012 Firenze University Press 

Università degli Studi di Firenze 

Firenze University Press 

Borgo Albizi, 28, 50122 Firenze, Italy 

http://www.fupress.com/ 

Printed in Italy

ECOS 2012 

The 25 th International Conference on 

Efficiency, Cost, Optimization and Simulation 

of Energy Conversion Systems and Processes 

Perugia, June 26 th -June 29 th , 2012 

Book of Proceedings - Volume VI 

Edited by: 

Umberto Desideri, Università degli Studi di Perugia 

Giampaolo Manfrida, Università degli Studi di Firenze 

Enrico Sciubba, Università degli Studi di Roma “Sapienza” 

ECOS 2012 - THE 25 TH INTERNATIONAL CONFERENCE ON 

EFFICIENCY, COST, OPTIMIZATION, SIMULATION AND ENVIRONMENTAL IMPACT OF ENERGY SYSTEMS 

JUNE 26-29, 2012, PERUGIA, ITALY 

EDITED BY UMBERTO DESIDERI, GIAMPAOLO MANFRIDA, ENRICO SCIUBBA 

FIRENZE UNIVERSITY PRESS, 2012, ISBN ONLINE : 978-88-6655-322-9

Advisory Committee (Track Organizers) 

Building, Urban and Complex Energy Systems 

V. Ismet Ugursal 

Dalhousie University, Nova Scotia, Canada 

Combustion, Chemical Reactors, Carbon Capture and Sequestration 

Giuseppe Girardi 

ENEA-Casaccia, Italy 

Energy Systems: Environmental and Sustainability Issues 

Christos A. Frangopoulos 

National Technical University of Athens, Greece 

Exergy Analysis and Second Law Analysis 

Silvio de Oliveira Junior 

Polytechnical University of Sao Paulo, Sao Paulo, Brazil 

Fluid Dynamics and Power Plant Components 

Sotirios Karellas 

National Technical University of Athens, Athens, Greece 

Fuel Cells 

Umberto Desideri 

University of Perugia, Perugia, Italy 

Heat and Mass Transfer 

Francesco Asdrubali, Cinzia Buratti 

University of Perugia, Perugia, Italy 

Industrial Ecology 

Stefan Goessling-Reisemann 

University of Bremen, Germany 

Poster Session 

Enrico Sciubba 

University Roma 1 “Sapienza”, Italy 

Process Integration and Heat Exchanger Networks 

Francois Marechal 

EPFL, Lausanne, Switzerland 

Renewable Energy Conversion Systems 

David Chiaramonti 

University of Firenze, Firenze, Italy 

Simulation of Energy Conversion Systems 

Marcin Liszka 

Polytechnica Slaska, Gliwice, Poland 

System Operation, Control, Diagnosis and Prognosis 

Vittorio Verda 

Politecnico di Torino, Italy 

Thermodynamics 

A. Özer Arnas 

United States Military Academy at West Point, U.S.A. 

Thermo-Economic Analysis and Optimisation 

Andrea Lazzaretto 

University of Padova, Padova, Italy 

Water Desalination and Use of Water Resources 

Corrado Sommariva 

ILF Consulting M.E., U.K 

iii

Scientific Committee 

Riccardo Basosi, University of Siena, Italy 

Gino Bella, University of Roma Tor Vergata, Italy 

Asfaw Beyene, San Diego State University, United States 

Ryszard Bialecki, Silesian Institute of Tecnology, Poland 

Gianni Bidini, University of Perugia, Italy 

Ana M. Blanco-Marigorta, University of Las Palmas de Gran Canaria, Spain 

Olav Bolland, University of Science and Technology (NTNU), Norway 

Renè Cornelissen, Cornelissen Consulting, The Netherlands 

Franco Cotana, University of Perugia, Italy 

Alexandru Dobrovicescu, Polytechnical University of Bucharest, Romania 

Gheorghe Dumitrascu, Technical University of Iasi, Romania 

Brian Elmegaard, Technical University of Denmark , Denmark 

Daniel Favrat, EPFL, Switzerland 

Michel Feidt, ENSEM - LEMTA University Henri Poincaré, France 

Daniele Fiaschi, University of Florence, Italy 

Marco Frey, Scuola Superiore S. Anna, Italy 

Richard A Gaggioli, Marquette University, USA 

Carlo N. Grimaldi, University of Perugia, Italy 

Simon Harvey, Chalmers University of Technology, Sweden 

Hasan Heperkan, Yildiz Technical University, Turkey 

Abel Abel Hernandez-Guerrero, University of Guanajuato, Mexico 

Jiri Jaromir Klemeš, University of Pannonia, Hungary 

Zornitza V. Kirova-Yordanova, University "Prof. Assen Zlatarov", Bulgaria 

Noam Lior, University of Pennsylvania, United States 

Francesco Martelli, University of Florence, Italy 

Aristide Massardo, University of Genova, Italy 

Jim McGovern, Dublin Institute of Technology, Ireland 

Alberto Mirandola, University of Padova, Italy 

Michael J. Moran, The Ohio State University, United States 

Tatiana Morosuk, Technical University of Berlin, Germany 

Pericles Pilidis, University of Cranfield, United Kingdom 

Constantine D. Rakopoulos, National Technical University of Athens, Greece 

Predrag Raskovic, University of Nis, Serbia and Montenegro 

Mauro Reini, University of Trieste, Italy 

Gianfranco Rizzo, University of Salerno, Italy 

Marc A. Rosen, University of Ontario, Canada 

Luis M. Serra, University of Zaragoza, Spain 

Gordana Stefanovic, University of Nis, Serbia and Montenegro 

Andrea Toffolo, Luleå University of Technology, Sweden 

Wojciech Stanek, Silesian University of Technology, Poland 

George Tsatsaronis, Technical University Berlin, Germany 

Antonio Valero, University of Zaragoza, Spain 

Michael R. von Spakovsky, Virginia Tech, USA 

Stefano Ubertini, Parthenope University of Naples, Italy 

Sergio Ulgiati, Parthenope University of Naples, Italy 

Sergio Usón, Universidad de Zaragoza, Spain 

Roman Weber, Clausthal University of Technology, Germany 

Ryohei Yokoyama, Osaka Prefecture University, Japan 

Na Zhang, Institute of Engineering Thermophysics, Chinese Academy of Sciences, China 

iv

The 25 th ECOS Conference 1987-2012: leaving a mark 

The introduction to the ECOS series of Conferences states that “ECOS is a series of 

international conferences that focus on all aspects of Thermal Sciences, with particular 

emphasis on Thermodynamics and its applications in energy conversion systems and 

processes”. Well, ECOS is much more than that, and its history proves it! 

The idea of starting a series of such conferences was put forth at an informal meeting of the 

Advanced Energy Systems Division of the American Society of Mechanical Engineers 

(ASME) at the November 1985 Winter Annual Meeting (WAM), in Miami Beach, Florida, 

then chaired by Richard Gaggioli. The resolution was to organize an annual Symposium on 

the Analysis and Design of Thermal Systems at each ASME WAM, and to try to involve a 

larger number of scientists and engineers worldwide by organizing conferences outside of the 

United States. Besides Rich other participants were Ozer Arnas, Adrian Bejan, Yehia El- 

Sayed, Robert Evans, Francis Huang, Mike Moran, Gordon Reistad, Enrico Sciubba and 

George Tsatsaronis. 

Ever since 1985, a Symposium of 8-16 sessions has been organized by the Systems Analysis 

Technical Committee every year, at the ASME Winter Annual Meeting (now ASME-IMECE). 

The first overseas conference took place in Rome, twenty-five years ago (in July 1987), with 

the support of the U.S. National Science Foundation and of the Italian National Research 

Council. In that occasion, Christos Frangopoulos, Yalcin Gogus, Elias Gyftopoulos, Dominick 

Sama, Sergio Stecco, Antonio Valero, and many others, already active at the ASME meetings, 

joined the core-group. 

The name ECOS was used for the first time in Zaragoza, in 1992: it is an acronym for 

Efficiency, Cost, Optimization and Simulation (of energy conversion systems and 

processes), keywords that best describe the contents of the presentations and discussions 

taking place in these conferences. Some years ago, Christos Frangopoulos inserted in the 

official website the note that “ècos” (’) means “home” in Greek and it ought to be 

attributed the very same meaning as the prefix “Eco-“ in environmental sciences. 

The last 25 years have witnessed an almost incredible growth of the ECOS community: more 

and more Colleagues are actively participating in our meetings, several international Journals 

routinely publish selected papers from our Proceedings, fruitful interdisciplinary and 

international cooperation projects have blossomed from our meetings. Meetings that have 

spanned three continents (Africa and Australia ought to be our next targets, perhaps!) and 

influenced in a way or another much of modern Engineering Thermodynamics. 

After 25 years, if we do not want to become embalmed in our own success and lose 

momentum, it is mandatory to aim our efforts in two directions: first, encourage the 

participation of younger academicians to our meetings, and second, stimulate creative and 

useful discussions in our sessions. Looking at this years’ registration roster (250 papers of 

which 50 authored or co-authored by junior Authors), the first objective seems to have been 

attained, and thus we have just to continue in that direction; the second one involves allowing 

space to “voices that sing out of the choir”, fostering new methods and approaches, and 

establishing or reinforcing connections to other scientific communities. It is important that our 

technical sessions represent a place of active confrontation, rather than academic “lecturing”. 

In this spirit, we welcome you in Perugia, and wish you a scientifically stimulating, 

touristically interesting, and culinarily rewarding experience. In line with our 25 years old 

scientific excellency and friendship! 

Umberto Desideri, Giampaolo Manfrida, Enrico Sciubba 

vi

CONTENT MANAGEMENT 

The index lists all the papers contained all the eight volumes of the Proceedings of the 

ECOS 2012 International Conference. 

Page numbers are listed only for papers within the Volume you are looking at. 

The ID code allows to trace back the identification number assigned to the paper within 

the Conference submission, review and track organization processes. 

vii

VOLUME VI 

CONTENT 

VI. 1 CARBON CAPTURE AND SEQUESTRATION 

» A novel coal-based polygeneration system cogenerating power, 

natural gas and liquid fuel with CO2 capture (ID 96) 

Sheng Li, Hongguang Jin, Lin Gao 

» Analysis and optimization of CO2 capture in a China’s existing coalfired 

power plant (ID 532) 

Gang Xu, Yongping Yang, Shoucheng Li, Wenyi Liu and Ying Wu 

» Analysys of four-end high temperature membrane air separator in a 

supercritical power plant with oxy-type pulverized fuel boiler (ID 442) 

Janusz Kotowicz, Sebastian Stanisaw Michalski 

» Analysis of potential improvements to the lignite-fired oxy-fuel power 

unit (ID 413) 

Marcin Liszka, Jakub Tuka, Grzegorz Nowak, Grzegorz Szapajko 

» Biogas Upgrading: Global Warming Potential of Conventional and 

Innovative Technologies (ID 240) 

Katherine Starr, Xavier Gabarrell Durany, Gara Villalba Mendez, Laura Talens 

Peiro, Lidia Lombardi 

» Capture of carbon dioxide using gas hydrate technology (ID 103) 

Beatrice Castellani, Mirko Filipponi, Sara Rinaldi, Federico Rossi 

» Carbon dioxide mineralisation and integration with flue gas 

desulphurisation applied to a modern coal-fired power plant (ID 179) 

Ron Zevenhoven, Johan Fagerlund, Thomas Björklöf, Magdalena Mäkelä, 

Olav Eklund 

» Carbon dioxide storage by mineralisation applied to a lime kiln (ID 226) 

Inês Sofia Soares Romão, Matias Eriksson, Experience Nduagu, Johan 

Fagerlund, Licínio Manuel Gando-Ferreira, Ron Zevenhoven 

» Comparison of IGCC and CFB cogeneration plants equipped with CO2 

removal (ID 380) 

Marcin Liszka, Tomasz Malik, Micha Budnik, Andrzej Zibik 

» Concept of a “capture ready” combined heat and power plant (ID 231) 

Piotr Henryk Lukowicz, Lukasz Bartela 

» Cryogenic method for H2 and CH4 recovery from a rich CO2 stream in 

pre-combustion CCS schemes (ID 508) 

Konstantinos Atsonios, Kyriakos D. Panopoulos, Angelos Doukelis, Antonis 

Koumanakos, Emmanuel Kakaras 

» Design and optimization of ITM oxy-combustion power plant (ID 495) 

Surekha Gunasekaran, Nicholas David Mancini, Alexander Mitsos 

» Implementation of a CCS technology: the ZECOMIX experimental 

platform (ID 222) 

Antonio Calabrò, Stefano Cassani, Leandro Pagliari, Stefano Stendardo 

» Influence of regeneration condition on cyclic CO2 capture using pretreated 

dispersed CaO as high temperature sorbent (ID 221) 

Stefano Stendardo, Antonio Calabrò 

…….... Pag. 1 

…….... Pag. 18 

…….... Pag. 35 

…….... Pag. 45 

…….... Pag. 61 

…….... Pag. 73 

…….... Pag. 83 

…….... Pag. 103 

…….... Pag. 116 

…….... Pag. 133 

…….... Pag. 145 

…….... Pag. 158 

…….... Pag. 169 

…….... Pag. 175 

-------------------------------------------------------------------------------------------------- 

ECOS 2012 - THE 25 TH INTERNATIONAL CONFERENCE ON 



EDITED BY UMBERTO DESIDERI, GIAMPAOLO MANFRIDA, ENRICO SCIUBBA 

FIRENZE UNIVERSITY PRESS, 2012, ISBN ONLINE : 978-88-6655-322-9

» Investigation of an innovative process for biogas up-grading – pilot 

plant preliminary results (ID 56) 

Lidia Lombardi, Renato Baciocchi, Ennio Antonio Carnevale, Andrea Corti, 

Giulia Costa, Tommaso Olivieri, Alessandro Paradisi, Daniela Zingaretti 

» Method of increasing the efficiency of a supercritical lignite-fired oxytype 

fluidized bed boiler and high-temperature three - end membrane for 

air separation (ID 438) 

Janusz Kotowicz, Adrian Balicki 

» Monitoring of carbon dioxide uptake in accelerated carbonation 

processes applied to air pollution control residues (ID 539) 

Felice Alfieri, Peter J Gunning, Michela Gallo, Adriana Del Borghi, Colin D 

Hills 

» Process efficiency and optimization of precipitated calcium carbonate 

(PCC) production from steel converter slag (ID 114) 

Hannu-Petteri Mattila, Inga Grigalinait, Arshe Said, Sami Filppula, Carl- 

Johan Fogelholm, Ron Zevenhoven 

» Production of Mg(OH)2 for CO2 Emissions Removal Applications: 

Parametric and Process Evaluation (ID 245) 

Experience Ikechukwu Nduagu, Inês Romão, Ron Zevenhoven 

» Thermodynamic analysis of a supercritical power plant with oxy type 

pulverized fuel boiler, carbon dioxide capture system (CC) and four-end 

high temperature membrane air seprator (ID 411) 

Janusz Kotowicz, Sebastian Stanisaw Michalski 

VI. 2 PROCESS INTEGRATION AND HEAT EXCHANGER NETWORKS 

» A multi-objective optimization technique for co- processing in the 

cement production (ID 42) 

Maria Luiza Grillo Renó, Rogério José da Silva, Mirian de Lourdes Noronha 

Motta Melo, José Joaquim Conceição Soares Santos 

» Comparison of options for debottlenecking the recovery boiler at kraft 

pulp mills – Economic performance and CO2 emissions (ID 449) 

Johanna Jönsson, Karin Pettersson, Simon Harvey, Thore Berntsson 

» Demonstrating an integral approach for industrial energy saving 

(ID 541) 

René Cornelissen, Geert van Rens, Jos Sentjens, Henk Akse, Ton Backx, 

Arjan van der Weiden, Jo Vandenbroucke 

» Maximising the use of renewables with variable availability (ID 494) 

Andreja Nemet, Jiri Jaromír Klemeš, Petar Sabev Varbanov, Zdravko 

Kravanja 

» Methodology for the improvement of large district heating networks 

(ID 46) 

Anna Volkova, Vladislav Mashatin, Aleksander Hlebnikov, Andres Siirde 

» Optimal mine site energy supply (ID 306) 

Monica Carvalho, Dean Lee Millar 

» Simulation of synthesis gas production from steam oxygen 

gasification of Colombian bituminous coal using Aspen Plus® (ID 395) 

John Jairo Ortiz, Juan Camilo González, Jorge Enrique Preciado, Rocío 

Sierra, Gerardo Gordillo 

ix 

…….... Pag. 185 

…….... Pag. 197 

…….... Pag. 205 

…….... Pag. 218 

…….... Pag. 233 

…….... Pag. 251 

…….... Pag. 261 

…….... Pag. 270 

…….... Pag. 287 

…….... Pag. 297 

…….... Pag. 314 

…….... Pag. 327 

…….... Pag. 340

----------------------------------------------------------------------- 

CONTENTS OF ALL THE VOLUMES 

----------------------------------------------------------------------- 

VOLUME I 

I . 1 - SIMULATION OF ENERGY CONVERSION SYSTEMS 

» A novel hybrid-fuel compressed air energy storage system for China’s situation (ID 531) 

Wenyi Liu, Yongping Yang, Weide Zhang, Gang Xu,and Ying Wu 

» A review of Stirling engine technologies applied to micro-cogeneration systems (ID 338) 

Ana C Ferreira, Manuel L Nunes, Luís B Martins, Senhorinha F Teixeira 

» An organic Rankine cycle off-design model for the search of the optimal control strategy 

(ID 295) 

Andrea Toffolo, Andrea Lazzaretto, Giovanni Manente, Marco Paci 

» Automated superstructure generation and optimization of distributed energy supply 

systems (ID 518) 

Philip Voll, Carsten Klaffke, Maike Hennen, André Bardow 

» Characterisation and classification of solid recovered fuels (SRF) and model development 

of a novel thermal utilization concept through air- gasification (ID 506) 

Panagiotis Vounatsos, Konstantinos Atsonios, Mihalis Agraniotis, Kyriakos D. Panopoulos, George 

Koufodimos,Panagiotis Grammelis, Emmanuel Kakaras 

» Design and modelling of a novel compact power cycle for low temperature heat sources 

(ID 177) 

Jorrit Wronski, Morten Juel Skovrup, Brian Elmegaard, Harald Nes Rislå, Fredrik Haglind 

» Dynamic simulation of combined cycles operating in transient conditions: an innovative 

approach to determine the steam drums life consumption (ID 439) 

Stefano Bracco 

» Effect of auxiliary electrical power consumptions on organic Rankine cycle system with 

low-temperature waste heat source (ID 235) 

Samer Maalouf, Elias Boulawz Ksayer, Denis Clodic 

» Energetic and exergetic analysis of waste heat recovery systems in the cement industry 

(ID 228) 

Sotirios Karellas, Aris Dimitrios Leontaritis, Georgios Panousis, Evangelos Bellos, Emmanuel 

Kakaras 

» Energy and exergy analysis of repowering options for Greek lignite-fired power plants (ID 

230) 

Sotirios Karellas, Aggelos Doukelis, Grammatiki Zanni, Emmanuel Kakaras 

» Energy saving by a simple solar collector with reflective panels and boiler (ID 366) 

Anna Stoppato, Renzo Tosato 

» Exergetic analysis of biomass fired double-stage Organic Rankine Cycle (ORC) (ID 37) 

Markus Preißinger, Florian Heberle, Dieter Brüggemann 

» Experimental tests and modelization of a domestic-scale organic Rankine cycle (ID 156) 

Roberto Bracco, Stefano Clemente, Diego Micheli, Mauro Reini 

» Model of a small steam engine for renewable domestic CHP system (ID 31 ) 

Giampaolo Manfrida, Giovanni Ferrara, Alessandro Pescioni 

» Model of vacuum glass heat pipe solar collectors (ID 312) 

Daniele Fiaschi, Giampaolo Manfrida 

x

» Modelling and exergy analysis of a plasma furnace for aluminum melting process (ID 254) 

Luis Enrique Acevedo, Sergio Usón, Javier Uche, Patxi Rodríguez 

» Modelling and experimental validation of a solar cooling installation (ID 296) 

Guillaume Anies, Pascal Stouffs, Jean Castaing-Lasvignottes 

» The influence of operating parameters and occupancy rate of thermoelectric modules on 

the electricity generation (ID 314) 

Camille Favarel, Jean-Pierre Bédécarrats, Tarik Kousksou, Daniel Champier 

» Thermodynamic and heat transfer analysis of rice straw co-firing in a Brazilian pulverised 

coal boiler (ID 236) 

Raphael Miyake, Alvaro Restrepo, Fábio Kleveston Edson Bazzo, Marcelo Bzuneck 

» Thermophotovoltaic generation: A state of the art review (ID 88) 

Matteo Bosi, Claudio Ferrari, Francesco Melino, Michele Pinelli, Pier Ruggero Spina, Mauro 

Venturini 

I . 2 – HEAT AND MASS TRANSFER 

» A DNS method for particle motion to establish boundary conditions in coal gasifiers (ID 

49) 

Efstathios E Michaelides, Zhigang Feng 

» Effective thermal conductivity with convection and radiation in packed bed (ID 60) 

Yusuke Asakuma 

» Experimental and CFD study of a single phase cone-shaped helical coiled heat exchanger: 

an empirical correlation (ID 375) 

Daniel Flórez-Orrego, Walter Arias, Diego López, Héctor Velásquez 

» Thermofluiddynamic model for control analysis of latent heat thermal storage system (ID 

207) 

Adriano Sciacovelli, Vittorio Verda, Flavio Gagliardi 

» Towards the development of an efficient immersed particle heat exchanger: particle 

transfer from low to high pressure (ID 202) 

Luciano A. Catalano, Riccardo Amirante, Stefano Copertino, Paolo Tamburrano, Fabio De Bellis 

I . 3 – INDUSTRIAL ECOLOGY 

» Anthropogenic heat and exergy balance of the atmosphere (ID 122) 

Asfaw Beyene, David MacPhee, Ron Zevenhoven 

» Determination of environmental remediation cost of municipal waste in terms of extended 

exergy (ID 63) 

Candeniz Seckin, Ahmet R. Bayulken 

» Development of product category rules for the application of life cycle assessment to 

carbon capture and storage (537) 

Carlo Strazza, Adriana Del Borghi, Michela Gallo 

» Electricity production from renewable and non-renewable energy sources: a comparison 

of environmental, economic and social sustainability indicators with exergy losses 

throughout the supply chain (ID 247) 

Lydia Stougie, Hedzer van der Kooi, Rob Stikkelman 

» Exergy analysis of the industrial symbiosis model in Kalundborg (ID 218) 

Alicia Valero Delgado, Sergio Usón, Jorge Costa 

» Global gold mining: is technological learning overcoming the declining in ore grades? (ID 

277) 

Adriana Domínguez, Alicia Valero 

xi

» Personal transportation energy consumption (ID305) 

Matteo Muratori, Emmanuele Serra, Vincenzo Marano, Michael Moran 

» Resource use evaluation of Turkish transportation sector via the extended exergy 

accounting method (ID 43) 

Candeniz Seckin, Enrico Sciubba, Ahmet R. Bayulken 

» The impact of higher energy prices on socio-economic inequalities of German social 

groups (ID 80) 

Holger Schlör, Wolfgang Fischer, Jürgen-Friedrich Hake 

VOLUME II 

II . 1 – EXERGY ANALYSIS AND 2 ND LAW ANALYSIS 

» A comparative analysis of cryogenic recuperative heat exchangers based on exergy 

destruction (ID 129) 

Adina Teodora Gheorghian, Alexandru Dobrovicescu, Lavinia Grosu, Bogdan Popescu, Claudia 

Ionita 

» A critical exploration of the usefulness of rational efficiency as a performance parameter 

for heat exchangers (ID 307) 

Jim McGovern, Georgiana Tirca-Dragomirescu, Michel Feidt, Alexandru Dobrovicescu 

» A new procedure for the design of LNG processes by combining exergy and pinch 

analyses (ID 238) 

Danahe Marmolejo-Correa, Truls Gundersen 

» Advances in the distribution of environmental cost of water bodies through the exergy 

concept in the Ebro river (ID 258) 

Javier Uche Marcuello, Amaya Martínez Gracia, Beatriz Carrasquer Álvarez, Antonio Valero Capilla 

» Application of the entropy generation minimization method to a solar heat exchanger: a 

pseudo-optimization design process based on the analysis of the local entropy generation 

maps (ID 357) 

Giorgio Giangaspero, Enrico Sciubba 

» Comparative analysis of ammonia and carbon dioxide two-stage cycles for simultaneous 

cooling and heating (ID 84) 

Alexandru Dobrovicescu, Ciprian Filipoiu, Emilia Cerna Mladin, Valentin Apostol, Liviu Drughean 

» Comparison between traditional methodologies and advanced exergy analyses for 

evaluating efficiency and externalities of energy systems (ID 515) 

Gabriele Cassetti, Emanuela Colombo 

» Comparison of entropy generation figures using entropy maps and entropy transport 

equation for an air cooled gas turbine blade (ID 468) 

Omer Emre Orhan, Oguz Uzol 

» Conventional and advanced exergetic evaluation of a supercritical coal-fired power plant 

(ID 377) 

Ligang Wang, Yongping Yang, Tatiana Morosuk, George Tsatsaronis 

» Energy and exergy analyses of the charging process in encapsulted ice thermal energy 

storage (ID 164) 

David MacPhee, Ibrahim Dincer, Asfaw Beyene 

» Energy integration and cogeneration in nitrogen fertilizers industry: thermodynamic 

estimation of the efficiency, potentials, limitations and environmental impact. Part 1: energy 

integration in ammonia production plants (ID 303) 

Zornitza Vassileva Kirova-Yordanova 

» Evaluation of the oil and gas processing at a real production day on a North Sea oil 

platform using exergy analysis (ID 260) 

Mari Voldsund, Wei He, Audun Røsjorde, Ivar Ståle Ertesvåg, Signe Kjelstrup 

xii

» Exergetic and economic analysis of Kalina cycle for low temperature geothermal sources 

in Brazil (ID 345) 

Carlos Eymel Campos Rodriguez, José Carlos Escobar Palacios, Cesar Adolfo Rodríguez 

Sotomonte, Marcio Leme, Osvaldo José Venturini, Electo Eduardo Silva Lora, Vladimir Melián 

Cobasa, Daniel Marques dos Santos, Fábio R. Lofrano Dotto, Vernei Gialluca 

» Exergy analysis and comparison of CO2 heat pumps (ID 242) 

Argyro Papadaki, Athina Stegou - Sagia 

» Exergy analysis of a CO2 Recovery plant for a brewery (ID 72) 

Daniel Rønne Nielsen, Brian Elmegaard, C. Bang-Møller 

» Exergy analysis of the silicon production process (ID 118) 

Marit Takla, Leiv Kolbeinsen, Halvard Tveit, Signe Kjelstrup 

» Exergy based indicators for cardiopulmonary exercise test evaluation (ID 159) 

Carlos Eduardo Keutenedjian Mady, Cyro Albuquerque Neto, Tiago Lazzaretti Fernandes, Arnaldo 

Jose Hernandez, Paulo Hilário Nascimento Saldiva, Jurandir Itizo Yanagihara, Silvio de Oliveira 

Junior 

» Exergy disaggregation as an alternative for system disaggregation in thermoeconomics 

(ID 483) 

José Joaquim Conceição Soares Santos, Atilio Lourenço, Julio Mendes da Silva, João Donatelli, 

José Escobar Palacio 

» Exergy intensity of petroleum derived fuels (ID 117) 

Julio Augusto Mendes da Silva, Maurício Sugiyama, Claudio Rucker, Silvio de Oliveira Junior 

» Exergy-based sustainability evaluation of a wind power generation system (ID 542) 

Jin Yang, B. Chen, Enrico Sciubba 

» Human body exergy metabolism (ID 160) 

Carlos Eduardo Keutenedjian Mady, Silvio de Oliveira Junior 

» Integrating an ORC into a natural gas expansion plant supplied with a co-generation unit 

(ID 273) 

Sergio Usón, Wojciech Juliusz Kostowski 

» One-dimensional model of an optimal ejector and parametric study of ejector efficiency (ID 

323) 

Ronan Killian McGovern, Kartik Bulusu, Mohammed Antar, John H. Lienhard 

» Optimization and design of pin-fin heat sinks based on minimum entropy generation (ID 6) 

Jose-Luis Zuniga-Cerroblanco, Abel Hernandez-Guerrero, Carlos A. Rubio-Jimenez, Cuauhtemoc 

Rubio-Arana, Sosimo E. Diaz-Mendez 

» Performance analysis of a district heating system (ID 271) 

Andrej Ljubenko, Alojz Poredoš, Tatiana Morosuk, George Tsatsaronis 

» System analysis of exergy losses in an integrated oxy-fuel combustion power plant (ID 64) 

Andrzej Zibik, Pawe Gadysz 

» What is the cost of losing irreversibly the mineral capital on Earth? (ID 220) 

Alicia Valero Delgado, Antonio Valero 

II . 2 – THERMODYNAMICS 

» A new polygeneration system for methanol and power based on coke oven gas and coal 

gas (ID 252) 

Hu Lin, Hongguang Jin, Lin Gao, Rumou Li 

» Argon-Water closed gas cycle (ID 67) 

Federico Fionelli, Giovanni Molinari 

» Binary alkane mixtures as fluids in Rankine cycles (ID 246) 

M. Aslam Siddiqi, Burak Atakan 

xiii

» Excess enthalpies of second generation biofuels (ID 308) 

Alejandro Moreau, José Juan Segovia, M. Carmen Martín, Miguel Ángel Villamañán, César R. 

Chamorro, Rosa M. Villamañán 

» Local stability analysis of a Curzon-Ahlborn engine considering the Van der Waals 

equation state in the maximum ecological regime (ID 281) 

Ricardo Richard Páez-Hernández, Pedro Portillo-Díaz, Delfino Ladino-Luna, 

Marco Antonio Barranco-Jiménez 

» Some remarks on the Carnot's theorem (ID 325) 

Julian Gonzalez Ayala, Fernando Angulo-Brown 

» The Dead State (ID 340) 

Richard A. Gaggioli 

» The magnetocaloric energy conversion (ID 97) 

Andrej Kitanovski, Jaka Tusek, Alojz Poredos 

VOLUME III 

THERMO-ECONOMIC ANALYSIS AND OPTIMIZATION 

» A comparison of optimal operation of residential energy systems using clustered demand 

patterns based on Kullback-Leibler divergence (ID 142) 

Akira Yoshida, Yoshiharu Amano, Noboru Murata, Koichi Ito, Takumi Hashizume 

» A Model for Simulation and Optimal Design of a Solar Heating System with Seasonal 

Storage (ID 51) 

Gianfranco Rizzo 

» A thermodynamic and economic comparative analysis of combined gas-steam and gas 

turbine air bottoming cycle (ID 232) 

Tadeusz Chmielniak, Daniel Czaja, Sebastian Lepszy 

» Application of an alternative thermoeconomic approach to a two-stage vapor compression 

refrigeration cycle with intercooling (ID 135) 

Atilio Barbosa Lourenço, José Joaquim Conceição Soares Santos, João Luiz Marcon Donatelli 

» Comparative performance of advanced power cycles for low temperature heat sources 

(ID 109) 

Guillaume Becquin, Sebastian Freund 

» Comparison of nuclear steam power plant and conventional steam power plant through 

energy level and thermoeconomic analysis (ID 251) 

S. Khamis Abadi, Mohammad Hasan Khoshgoftar Manesh, M. Baghestani, H. Ghalami, Majid 

Amidpour 

» Economic and exergoeconomic analysis of micro GT and ORC cogeneration systems 

(ID 87) 

Audrius Bagdanavicius, Robert Sansom, Nick Jenkins, Goran Strbac 

» Exergoeconomic comparison of wet and dry cooling technologies for the Rankine cycle of 

a solar thermal power plant (ID 300) 

Philipp Habl, Ana M. Blanco-Marigorta, Berit Erlach 

» Influence of renewable generators on the thermo-economic multi-level optimization of a 

poly-generation smart grid (101) 

Massimo Rivarolo, Andrea Greco, Francesca Travi, Aristide F. Massardo 

» Local stability analysis of a thermoeconomic model of an irreversible heat engine working 

at different criteria of performance (ID 289) 

Marco A. Barranco-Jiménez, Norma Sánchez-Salas, Israel Reyes-Ramírez, Lev Guzmán-Vargas 

» Multicriteria optimization of a distributed trigeneration system in an industrial area (ID 154) 

Dario Buoro, Melchiorre Casisi, Alberto de Nardi, Piero Pinamonti, Mauro Reini 

xiv

» On the effect of eco-indicator selection on the conclusions obtained from an 

exergoenvironmental analysis (ID 275) 

Tatiana Morosuk, George Tsatsaronis, Christopher Koroneos 

» Optimisation of supply temperature and mass flow rate for a district heating network 

(ID 104) 

Marouf Pirouti, Audrius Bagdanavicius, Jianzhong Wu, Janaka Ekanayake 

» Optimization of energy supply systems in consideration of hierarchical relationship 

between design and operation (ID 389) 

Ryohei Yokoyama, Shuhei Ose 

» The fuel impact formula revisited (ID 279) 

Cesar Torres, Antonio Valero 

» The introduction of exergy analysis to the thermo-economic modelling and optimisation of 

a marine combined cycle system (ID 61) 

George G. Dimopoulos, Chariklia A. Georgopoulou, Nikolaos M.P. Kakalis 

» The relationship between costs and environmental impacts in power plants: an exergybased 

study (ID 272) 

Fontina Petrakopoulou, Yolanda Lara, Tatiana Morosuk, Alicia Boyano, George Tsatsaronis 

» Thermo-ecological evaluation of biomass integrated gasification gas turbine based 

cogeneration technology (ID 441) 

Wojciech Stanek, Lucyna Czarnowska, Jacek Kalina 

» Thermo-ecological optimization of a heat exchanger through empirical modeling (ID 501) 

Ireneusz Szczygie, Wojciech Stanek, Lucyna Czarnowska, Marek Rojczyk 

» Thermoeconomic analysis and optimization in a combined cycle power plant including a 

heat transformer for energy saving (ID 399) 

Elizabeth Cortés Rodríguez, José Luis Castilla Carrillo, Claudia A. Ruiz Mercado, Wilfrido Rivera 

Gómez-Franco 

» Thermoeconomic analysis and optimization of a hybrid solar-electric heating in a fluidized 

bed dryer (ID 400) 

Elizabeth Cortés Rodríguez, Felipe de Jesús Ojeda Cámara, Isaac Pilatowsky Figueroa 

» Thermoeconomic approach for the analysis of low temperature district heating systems 

(ID 208) 

Vittorio Verda, Albana Kona 

» Thermo-economic assessment of a micro CHP systems fuelled by geothermal and solar 

energy (ID 166) 

Duccio Tempesti, Daniele Fiaschi, Filippo Gabuzzini 

» Thermo-economic evaluation and optimization of the thermo-chemical conversion of 

biomass into methanol (ID 194) 

Emanuela Peduzzi, Laurence Tock, Guillaume Boissonnet, François Marechal 

» Thermoeconomic fuel impact approach for assessing resources savings in industrial 

symbiosis: application to Kalundborg Eco-industrial Park (ID 256) 

Sergio Usón, Antonio Valero, Alicia Valero, Jorge Costa 

» Thermoeconomics of a ground-based CAES plant for peak-load energy production system 

(ID 32) 

Simon Kemble, Giampaolo Manfrida, Adriano Milazzo, Francesco Buffa 

xv

VOLUME IV 

IV . 1 - FLUID DYNAMICS AND POWER PLANT COMPONENTS 

» A control oriented simulation model of a multistage axial compressor (ID 444) 

Lorenzo Damiani, Giampaolo Crosa, Angela Trucco 

» A flexible and simple device for in-cylinder flow measurements: experimental and 

numerical validation (ID 181) 

Andrea Dai Zotti, Massimo Masi, Marco Antonello 

» CFD Simulation of Entropy Generation in Pipeline for Steam Transport in Real Industrial 

Plant (ID 543) 

Goran Vukovi, Gradimir Ili, Mia Vuki, Milan Bani, Gordana Stefanovi 

» Feasibility Study of Turbo expander Installation in City Gate Station (ID 168) 

Navid Zehtabiyan Rezaie, Majid Saffar-Avval 

» GTL and RME combustion analysis in a transparent CI engine by means of IR digital 

imaging (ID 460) 

Ezio Mancaruso, Luigi Sequino, Bianca Maria Vaglieco 

» Some aspects concerning fluid flow and turbulence modeling in 4-valve engines (ID 116) 

Zoran Stevan Jovanovic, Zoran Masonicic, Miroljub Tomic 

IV . 2 - SYSTEM OPERATION CONTROL DIAGNOSIS AND PROGNOSIS 

» Adapting the operation regimes of trigeneration systems to renewable energy systems 

integration (ID 188) 

Liviu Ruieneanu, Mihai Paul Mircea 

» Advanced electromagnetic sensors for sustainable monitoring of industrial processes 

(ID 145) 

Uroš Puc, Andreja Abina, Anton Jegli, Pavel Cevc, Aleksander Zidanšek 

» Assessment of stresses and residual life of plant components in view of life-time 

extension of power plants (ID 453) 

Anna Stoppato, Alberto Benato and Alberto Mirandola 

» Control strategy for minimizing the electric power consumption of hybrid ground source 

heat pump system (ID 244) 

Zoi Sagia, Constantinos Rakopoulos 

» Exergetic evaluation of heat pump booster configurations in a low temperature district 

heating network (ID 148) 

Torben Ommen, Brian Elmegaard 

» Exergoeconomic diagnosis: a thermo-characterization method by using irreversibility 

analysis (ID 523) 

Abraham Olivares-Arriaga, Alejandro Zaleta-Aguilar, Rangel-Hernández V. H, 

Juan Manuel Belman-Flores 

» Optimal structural design of residential cogeneration systems considering their 

operational restrictions (ID 224) 

Tetsuya Wakui, Ryohei Yokoyama 

» Performance estimation and optimal operation of a CO2 heat pump water heating system 

(ID 344) 

Ryohei Yokoyama, Ryosuke Kato, Tetsuya Wakui, Kazuhisa Takemura 

» Performances of a common-rail Diesel engine fuelled with rapeseed and waste cooking 

oils (ID 213) 

Alessandro Corsini, Valerio Giovannoni, Stefano Nardecchia, Franco Rispoli, Fabrizio Sciulli, 

Paolo Venturini 

xvi

» Reduced energy cost through the furnace pressure control in power plants (ID 367) 

Vojislav Filipovi, Novak Nedi, Saša Prodanovi 

» Short-term scheduling model for a wind-hydro-thermal electricity system (ID 464) 

Sérgio Pereira, Paula Ferreira, A. Ismael Freitas Vaz 

VOLUME V 

V . 1 - RENEWABLE ENERGY CONVERSION SYSTEMS 

» A co-powered concentrated solar power Rankine cycle concept for small size combined 

heat and power (ID 276) 

Alessandro Corsini, Domenico Borello, Franco Rispoli, Eileen Tortora 

» A novel non-tracking solar collector for high temperature application (ID 466) 

Wattana Ratismith, Anusorn Inthongkhum 

» Absorption heat transformers (AHT) as a way to enhance low enthalpy geothermal 

resources (ID 311) 

Daniele Fiaschi, Duccio Tempesti, Giampaolo Manfrida, Daniele Di Rosa 

» Alternative feedstock for the biodiesel and energy production: the OVEST project (ID 98) 

Matteo Prussi, David Chiaramonti, Lucia Recchia, Francesco Martelli, Fabio Guidotti 

» Assessing repowering and update scenarios for wind energy converters (ID 158) 

Till Zimmermann 

» Biogas from mechanical pulping industry – potential improvement for increased biomass 

vehicle fuels (ID 54) 

Mimmi Magnusson, Per Alvfors 

» Biogas or electricity as vehicle fuels derived from food waste - the case of Stockholm 

(ID 27) 

Martina Wikström, Per Alvfors 

» Compressibility factor as evaluation parameter of expansion processes in organic Rankine 

cycles (ID 292) 

Giovanni Manente, Andrea Lazzaretto 

» Design of solar heating system for methane generation (ID 445) 

Lucía Mónica Gutiérrez, P. Quinto Diez, L. R. Tovar Gálvez 

» Economic feasibility of PV systems in hotels in Mexico (ID 346) 

Augusto Sanchez, Sergio Quezada 

» Effect of a back surface roughness on annual performance of an air-cooled PV module 

(ID 193) 

Riccardo Secchi, Duccio Tempesti, Jacek Smolka 

» Energy and exergy analysis of the first hybrid solar-gas power plant in Algeria (ID 176) 

Fouad Khaldi 

» Energy recovery from MSW treatment by gasification and melting technology (ID 393) 

Fabrizio Strobino, Alessandro Pini Prato, Diego Ventura, Marco Damonte 

» Ethanol production by enzymatic hydrolysis process from sugarcane biomass - the 

integration with the conventional process (ID 189) 

Reynaldo Palacios-Bereche, Adriano Ensinas, Marcelo Modesto, Silvia Azucena Nebra 

» Evaluation of gas in an industrial anaerobic digester by means of biochemical methane 

potential of organic municipal solid waste components (ID 57) 

Isabella Pecorini, Tommaso Olivieri, Donata Bacchi, Alessandro Paradisi, Lidia Lombardi, Andrea 

Corti, Ennio Carnevale 

xvii

» Exergy analysis and genetic algorithms for the optimization of flat-plate solar collectors 

(ID 423) 

Soteris A. Kalogirou 

» Experimental study of tar and particles content of the produced gas in a double stage 

downdraft gasifier (ID 487) 

Ana Lisbeth Galindo Noguera, Sandra Yamile Giraldo, Rene Lesme-Jaén, Vladimir Melian Cobas, 

Rubenildo Viera Andrade, Electo Silva Lora 

» Feasibility study to realize an anaerobic digester fed with vegetables matrices in central 

Italy (ID 425) 

Umberto Desideri, Francesco Zepparelli, Livia Arcioni, Ornella Calderini, Francesco Panara, Matteo 

Todini 

» Investigations on the use of biogas for small scale decentralized CHP applications with a 

focus on stability and emissions (ID 140) 

Steven MacLean, Eren Tali, Anne Giese, Jörg Leicher 

» Kinetic energy recovery system for sailing yachts (ID 427) 

Giuseppe Leo Guizzi, Michele Manno 

» Mirrors in the sky: status and some supporting materials experiments (ID 184) 

Noam Lior 

» Numerical parametric study for different cold storage designs and strategies of a solar 

driven thermoacoustic cooler system (ID 284) 

Maxime Perier-Muzet, Pascal Stouffs, Jean-Pierre Bedecarrats, Jean Castaing-Lasvignottes 

» Parabolic trough photovoltaic/thermal collectors. Part I: design and simulation model (ID 

102) 

Francesco Calise, Laura Vanoli 

» Parabolic trough photovoltaic/thermal collectors. Part II: dynamic simulation of a solar 

trigeneration system (ID 488) 

Francesco Calise, Laura Vanoli 

» Performance analysis of downdraft gasifier - reciprocating engine biomass fired smallscale 

cogeneration system (ID 368) 

Jacek Kalina 

» Proposing offshore photovoltaic (PV) technology to the energy mix of the Maltese islands 

(ID 262) 

Kim Trapani, Dean Lee Millar 

» Research of integrated biomass gasification system with a piston engine (ID 414) 

Janusz Kotowicz, Aleksander Sobolewski, Tomasz Iluk 

» Start up of a pre-industrial scale solid state anaerobic digestion cell for the co-treatment of 

animal and agricultural residues (ID 34) 

Francesco Di Maria, Giovanni Gigliotti, Alessio Sordi, Caterina Micale, Luisa Massaccesi 

» The role of biomass in the renewable energy system (ID 390) 

Ruben Laleman, Ludovico Balduccio, Johan Albrecht 

» Vegetable oils of soybean, sunflower and tung as alternative fuels for compression 

ignition engines (ID 500) 

Ricardo Morel Hartmann, Nury Nieto Garzón, Eduardo Morel Hartmann, Amir Antonio Martins 

Oliveira Jr, Edson Bazzo, Bruno Okuda, Joselia Piluski 

» Wind energy conversion performance and atmosphere stability (ID 283) 

Francesco Castellani, Emanuele Piccioni, Lorenzo Biondi, Marcello Marconi 

V. 2 - FUEL CELLS 

» Comparison study on different SOFC hybrid systems with zero-CO2 emission (ID 196) 

Liqiang Duan, Kexin Huang, Xiaoyuan Zhang and Yongping Yang 

xv iii

» Exergy analysis and optimisation of a steam methane pre-reforming system (ID 62) 

George G. Dimopoulos, Iason C. Stefanatos, Nikolaos M.P. Kakalis 

» Modelling of a CHP SOFC power system fed with biogas from anaerobic digestion of 

municipal wastes integrated with a solar collector and storage units (ID 491) 

Domenico Borello, Sara Evangelisti, Eileen Tortora 

VOLUME VII 

VII . 1 - BUILDING, URBAN AND COMPLEX ENERGY SYSTEMS 

» A linear programming model for the optimal assessment of sustainable energy action 

plans (ID 398) 

Gianfranco Rizzo, Giancarlo Savino 

» A natural gas fuelled 10 kW electric power unit based on a Diesel automotive internal 

combustion engine and suitable for cogeneration (ID 477) 

Pietro Capaldi 

» Adjustment of envelopes characteristics to climatic conditions for saving heating and 

cooling energy in buildings (ID 430) 

Christos Tzivanidis, Kimon Antonopoulos, Foteini Gioti 

» An exergy based method for the optimal integration of a building and its heating plant. 

Part 1: comparison of domestic heating systems based on renewable sources (ID 81) 

Marta Cianfrini, Enrico Sciubba, Claudia Toro 

» Analysis of different typologies of natural insulation materials with economic and 

performances evaluation of the same buildings (ID 28) 

Umberto Desideri, Daniela Leonardi, Livia Arcioni 

» Complex networks approach to the Italian photovoltaic energy distribution system (ID 470) 

Luca Valori, Giovanni Luca Giannuzzi, Tiziano Squartini, Diego Garlaschelli, Riccardo Basosi 

» Design of a multi-purpose building "to zero energy consumption" according to European 

Directive 2010/31/CE: Architectural and plant solutions (ID 29) 

Umberto Desideri, Livia Arcioni, Daniela Leonardi, Luca Cesaretti ,Perla Perugini, Elena Agabitini, 

Nicola Evangelisti 

» Effect of initial systems on the renewal planning of energy supply systems for a hospital 

(ID 107) 

Shu Yoshida, Koichi Ito, Yoshiharu Amano, Shintaro Ishikawa, Takahiro Sushi, Takumi Hashizume 

» Effects of insulation and phase change materials (PCM) combinations on the energy 

consumption for buildings indoor thermal comfort (ID 387) 

Christos Tzivanidis, Kimon Antonopoulos, Eleutherios Kravvaritis 

» Energetic evaluation of a smart controlled greenhouse for tomato cultivation (ID 150) 

Nickey Van den Bulck, Mathias Coomans, Lieve Wittemans, Kris Goen, Jochen Hanssens, Kathy 

Steppe, Herman Marien, Johan Desmedt 

» Energy networks in sustainable cities: temperature and energy consumption monitoring in 

urban area (ID 190) 

Luca Giaccone, Alessandra Guerrisi, Paolo Lazzeroni and Michele Tartaglia 

» Extended exergy analysis of the economy of Nova Scotia, Canada (ID 215) 

David C Bligh, V.Ismet Ugursal 

» Feasibility study and design of a low-energy residential unit in Sagarmatha Park (Nepal) 

for envirnomental impact reduction of high altitude buildings (ID 223) 

Umberto Desideri, Stefania Proietti, Paolo Sdringola, Elisa Vuillermoz 

» Fire and smoke spread in low-income housing in Mexico (ID 379) 

Raul R. Flores-Rodriguez, Abel Hernandez-Guerrero, Cuauhtemoc Rubio-Arana, Consuelo A. 

Caldera-Briseño 

xix

» Optimal lighting control strategies in supermarkets for energy efficiency applications via 

digital dimmable technology (ID 136) 

Salvador Acha, Nilay Shah, Jon Ashford, David Penfold 

» Optimising the arrangement of finance towards large scale refurbishment of housing stock 

using mathematical programming and optimisationg (ID 127) 

Mark Gerard Jennings, Nilay Shah, David Fisk 

» Optimization of thermal insulation to save energy in buildings (ID 174) 

Milorad Boji, Marko Mileti, Vesna Marjanovi, Danijela Nikoli, Jasmina Skerli 

» Residential solar-based seasonal thermal storage system in cold climate: building 

envelope and thermal storage (ID 342) 

Alexandre Hugo and Radu Zmeureanu 

» Simultaneous production of domestic hot water and space cooling with a heat pump in a 

Swedish Passive House (ID 55) 

Johannes Persson, Mats Westermark 

» SOFC micro-CHP integration in residential buildings (ID 201) 

Umberto Desideri, Giovanni Cinti, Gabriele Discepoli, Elena Sisani, Daniele Penchini 

» The effect of shading of building integrated photovoltaics on roof surface temperature and 

heat transfer in buildings (ID 83) 

Eftychios Vardoulakis, Dimitrios Karamanis 

» The influence of glazing systems on energy performance and thermal comfort in nonresidential 

buildings (ID 206) 

Cinzia Buratti, Elisa Moretti, Elisa Belloni 

» Thermal analysis of a greenhouse heated by solar energy and seasonal thermal energy 

storage in soil (ID 405) 

Yong Li, Jin Xu, Ru-Zhu Wang 

» Thermodynamic analysis of a combined cooling, heating and power system under part 

load condition (ID 476) 

Qiang Chen, Jianjiao Zheng, Wei Han, Jun Sui, Hong-guang Jin 

VII . 2 - COMBUSTION, CHEMICAL REACTORS 

» Baffle as a cost-effective design improvement for volatile combustion rate increase in 

biomass boilers of simple construction (ID 233) 

Borivoj Stepanov, Ivan Pešenjanski, Biljana Miljkovi 

» Characterization of CH4-H2-air mixtures in the high-pressure DHARMA reactor (ID 287) 

Vincenzo Moccia, Jacopo D'Alessio 

» Development of a concept for efficiency improvement and decreased NOx production for 

natural gas-fired glass melting furnaces by switching to a propane exhaust gas fired 

process (ID 146) 

Jörn Benthin, Anne Giese 

» Experimental analysis of inhibition phenomenon management for Solid Anaerobic 

Digestion Batch process (ID 348) 

Francesco Di Maria, Giovanni Gigliotti, Alessio Sordi, Caterina Micale, Claudia Zadra, Luisa 

Massaccesi 

» Experimental investigations of the combustion process of n-butanol/diesel blend in an 

optical high swirl CI engine (ID 85) 

Simona Silvia Merola, G. Valentino, C. Tornatore, L. Marchitto , F. E. Corcione 

» Flameless oxidation as a means to reduce NOx emissions in glass melting furnaces 

(ID 141) 

Jörg Leicher, Anne Giese 

xx

» Mechanism of damage by high temperature of the tubes, exposed to the atmosphere 

characteristic of a furnace of pyrolysis of ethane for ethylene production in the 

petrochemical industry (ID 65) 

Jaqueline Saavedra Rueda, Francisco Javier Perez Trujillo, Lourdes Isabel Meriño Stand, Harbey 

Alexi Escobar, Luis Eduardo Navas, Juan Carlos Amezquita 

» Steam reforming of methane over Pt/Rh based wire mesh catalyst in single channel 

reformer for small scale syngas production (ID 317) 

Haftor Orn Sigurdsson, Søren Knudsen Kær 

VOLUME VIII 

VIII . 1 - ENERGY SYSTEMS : ENVIRONMENTAL AND SUSTAINABILITY ISSUES 

» A multi-criteria decision analysis tool to support electricity planning (ID 467) 

Fernando Ribeiro, Paula Ferreira, Madalena Araújo 

» Comparison of sophisticated life cycle impact assessment methods for assessing 

environmental impacts in a LCA study of electricity production (ID 259) 

Jens Buchgeister 

» Defossilisation assessment of biodiesel life cycle production using the ExROI indicator 

(ID 304) 

Emilio Font de Mora, César Torres, Antonio Valero, David Zambrana 

» Design strategy of geothermal plants for water dominant medium-low temperature 

reservoirs based on sustainability issues (ID 99) 

Alessandro Franco, Maurizio Vaccaro 

» Energetic and environmental benefits from waste management: experimental analysis of 

the sustainable landfill (ID 33) 

Francesco Di Maria, Alessandro Canovai, Federico Valentini, Alessio Sordi, Caterina Micale 

» Environmental assessment of energy recovery technologies for the treatment and 

disposal of municipal solid waste using life cycle assessment (LCA): a case study of Brazil 

(ID 512) 

Marcio Montagnana Vicente Leme, Mateus Henrique Rocha, Electo Eduardo Silva Lora,Osvaldo 

José Venturini, Bruno Marciano Lopes, Claudio Homero Ferreira 

» How will renewable power generation be affected by climate change? – The case of a 

metropolitan region in Northwest Germany (ID 503) 

Jakob Wachsmuth, Andrew Blohm, Stefan Gößling-Reisemann, Tobias Eickemeier, Rebecca 

Gasper, Matthias Ruth, Sönke Stührmann 

» Impact of nuclear power plant on Thailand power development plan (ID 474) 

Raksanai Nidhiritdhikrai, Bundhit Eua-arporn 

» Improving sustainability of maritime transport through utilization of liquefied natural gas 

(LNG) for propulsion (ID 496) 

Fabio Burel, Rodolfo Taccani, Nicola Zuliani 

» Life cycle assessment of thin film non conventional photovoltaics: the case of dye 

sensitized solar cells (ID 471) 

Maria Laura Parisi, Adalgisa Sinicropi, Riccardo Basosi 

» Low CO2 emission hybrid solar CC power system (ID 175) 

Yuanyuan Li, Na Zhang, Ruixian Cai 

» Low exergy solutions as a contribution to climate adapted and resilient power supply 

(ID 489) 

Stefan Goessling-Reisemann, Thomas Bloethe 

» On the use of MPT to derive optimal RES electricity generation mixes (ID 459) 

Paula Ferreira, Jorge Cunha 

xxi

» Stability and limit cycles in an exergy-based model of population dynamics (ID 128) 

Enrico Sciubba, Federico Zullo 

» The influence of primary measures for reducing NOx emissions on energy steam boiler 

efficiency (ID 125) 

Goran Stupar, Dragan Tucakovi, Titoslav Živanovi, Miloš Banjac, Sran Beloševi,Vladimir 

Beljanski, Ivan Tomanovi, Nenad Crnomarkovi, Miroslav Sijeri 

» The Lethe city car of the University of Roma 1: final proposed configuration (ID 45) 

Roberto Capata, Enrico Sciubba 

VIII . 2 - POSTER SESSION 

» A variational optimization of a finite-time thermal cycle with a Stefan-Boltzmann heat 

transfer law (ID 333) 

Juan C.Chimal-Eguia, Norma Sanchez-Salas 

» Modeling and simulation of a boiler unit for steam power plants (ID 545) 

Luca Moliterno, Claudia Toro 

» Numerical Modelling of straw combustion in a moving bed combustor (ID 412) 

Biljana Miljkoviü, Ivan Pešenjanski, Borivoj Stepanov, Vladimir Milosavljeviü, Vladimir Rajs 

» Physicochemical evaluation of the properties of the coke formed at radiation area of light 

hydrocarbons pyrolysis furnace in petrochemical industry (ID 10) 

Jaqueline Saavedra Rueda , Angélica María Carreño Parra, María del Rosario Pérez Trejos, 

Dionisio Laverde Cataño, Diego Bonilla Duarte, Jorge Leonardo Rodríguez Jiménez, Laura María 

Díaz Burgos 

» Rotor TG cooled (ID 121) 

Chiara Durastante, Paolo Petroni, Michela Spagnoli, Vincenzo Rizzica, Jörg Helge Wirfs 

» Study of the phase change in binary alloy (ID 534) 

Aroussia Jaouahdou, Mohamed J. Safi, Herve Muhr 

» Technip initiatives in renewable energies and sustainable technologies (ID 527) 

Pierfrancesco Palazzo, Corrado Pigna 

xxii

ECOS 2012 

VOLUME VI

PROCEEDINGS OF ECOS 2012 - THE 25 TH INTERNATIONAL CONFERENCE ON 



A NOVEL COAL-BASED POLYGENERATION 

SYSTEM COGENERATING POWER, NATURAL 

GAS AND LIQUID FUEL WITH CO2 CAPTURE 

Sheng Li a,b , Hongguang Jin a,* , Lin Gao a 

a Institute of Engineering Thermophysics, Chinese Academy of Sciences, Beijing, China, 100190 

b Graduate school of Chinese Academy of Sciences, Beijing, China, 100049 

*the corresponding author: hgjin@mail.etp.ac.cn 

Abstract: 

In this paper, a novel coal-based polygeneration system has been proposed, in which power, natural gas 

and liquid fuel are cogenerated and 62% of carbon is captured. Through proper adjustment instead of full 

adjustment of the syngas component, the CO/H2 of the unreacted gas in methanol synthesis unit can exactly 

meet the CO/H2 requirement for SNG synthesis, and thus cogeneration of methanol and SNG can be 

realized easily. By adopting partial recycle instead of full recycle of the unreacted gas in chemical island, the 

sharp increase of energy consumption for chemical synthesis can be avoided. On the other hand, part of the 

unreacted gas from methanol synthesis unit, which is hard to be converted, together with the recovery gas in 

chemical islands will be sent to combustion for power generation efficiently. At the same time, 62% of the 

carbon has been captured by low-temperature methanol wash method (LTMW) after concentration. As a 

result, the thermal efficiency of the novel system is around 54.9%, and the exergy efficiency is about 57.3%, 

which is much higher than the IGCC system, or single methanol synthesis system. Compared with the 

conventional single product systems, primary energy saving ratio of this novel polygeneration system can 

reach as high as 10.8 percent. Based on the graphical exergy analysis, the key processes of the new system 

are disclosed and the internal phenomena for high performance are revealed. The promising results 

obtained in this coal-based system may realize both the low-energy-penalty decarburization of coal and highefficient 

coal utilization, and will possibly provide a new option to enforce the safety of energy supply for 

countries with abundant coal resources but lack of natural gas and oils. 

Keywords: 

Polygeneration system, CO2 capture, SNG, Liquid fuel. 

1. Introduction 

Coal plays an important role in the energy supply of the world, especially for countries with 

abundant coal resources. For example, coal supplies over 70% of energy demand in China currently 

[1]. In the foreseeable future, coal will still remain the major resource of energy supply for a long 

time in China [2]. The most challenging environment issue of today is to slow down the 

accumulation of CO2 in the atmosphere, which is the prime criminal of the climate change. At 

present, 22 billion tons of CO2 per year is emitted as a result of the use of fossil fuel, most of which 

come from the combustion of coal [3-6]. How to abate CO2 emission during the coal utilization 

becomes a huge challenging for coal-rich countries. 

Until now, three typical technologic options for CO2 capture in power plants are developed, which 

are famous as pre-combustion, post-combustion and oxy-fuel combustion. There are many studies 

have investigated about these options. For example, the pre-combustion capture option has been 

studied by Consonni and Chiesa, Lozza and Chiesa [7-9]. Oxy-fuel combustion has been studied by 

Chiesa and Lozza, Inui et.al, Yang and Lin [10-12]. And post-combustion for CO2 capture using 

membranes and amine has been studied by Bounaceur et.al and Romeo et.al [13-14] respectively. 

Some other options, such as the integration of polygenration systems with CO2 capture has been 

studied by Jin et.al [6, 15-16], biomass-based energy system with negative CO2 emissions has been 

1

studied by Möllersten et.al, Obersteiner et.al [17-19], and chemical-looping combustion systems 

have been investigated by Jin and Ishida [20]. 

However, great energy penalty must be paid for CO2 capture in most of the above systems except 

for few polygeneration systems or chemical combustion systems et.al, which usually lead to an 

overall thermal efficiency decrease by nearly 7.0~15.0 percent points. Thus, how to reduce the 

energy penalty for CO2 capture in coal-based energy systems is of significant importance. 

Meanwhile, for countries with abundant coal but lack of natural gas and oil, energy security is 

another big issue that needs to be considered. For these counties, production of SNG or alternative 

liquid fuel from coal may be a more wise strategy instead of increasing import increasingly. For 

example, in China, with the rapid economic growth, the liquid fuel demand is increasing sharply as 

a result of the up burst of the number of cars. It is predicted that the demand for liquid fuel will soar 

to 0.45~0.61 billion tons [21-22] whereas the supply of liquid fuel can only keep at about 0.2 billion 

per year [22]. Huge gap exists between liquid fuel demand and supply in China. 

Aiming at the two big issues including CO2 abatement and energy security, the purpose of this 

paper is (1) to integrate a novel coal-based polygeneration system in which power, natural gas and 

liquid fuel are cogenerated and CO2 is captured with low energy penalty; (2) to reveal the internal 

phenomena of key processes in the new system by exergy analysis; (3) and to provide an option for 

production of SNG and alternative liquid fuel from coal to enforce energy security. 

2. PROPOSAL OF THE NEW POLYGENERATION SYSTEM 

2.1. Basic concept of system integration 

Fig. 1. Single methanol production process 

For single methanol production process, it is the main target to convert the raw material into 

products to the maximum extent. To achieve this target, the H2/CO in the syngas must be adjusted 

to be 2:1 and the unreacted gas should be fully recycled. Whereas, the component adjustment of the 

syngas will lead to great portion of exergy destruction and the full recycle of the unreacted gas will 

also requires large amount of work. When the conversion ratio exceeds a certain value, the exergy 

destruction for methanol synthesis will increase sharply if we purely pursue higher conversion of 

the raw material, as shown in Figure 2. This concept that “exhaustion of the active composition of 

the material” causes great energy consumption for methanol production in some way, which is 

about 45GJ/t. 

2

Fig. 2. The exergy destruction of MEOH synthesis with the conversion ratio (circulation ratio) 

Aiming at the lacks of the conventional single methanol production process, by proper adjustment 

instead of full adjustment of the component of the syngas, the CO/H2 of the unreacted gas in 

methanol synthesis unit can exactly satisfy the CO/H2 requirement for SNG synthesis and thus the 

component adjustment for SNG synthesis can be avoided. On the other hand, part of the unreacted 

gas, which is hard to be converted, will be sent to combustion for power. This concept of system 

integration for polygeneration system is part of the principle “Cascade utilization of chemical and 

thermal energy”, which is interrupted in Figure 3. 

Fig. 3. The principle of system integration for polygeneration systems 

In Figure 3, A on the vertical coordinate represents the energy level. The fossil fuel with high 

energy level is first converted to chemicals according to its component in cascade. And then, the 

unreacted, which is hard to be converted, is sent to generate high temperature by combustion for 

high-efficiency combined cycle. By such cascade utilization of fossil fuels, energy systems with 

high efficiency would be expected. 

3

2.2. System integration and innovation 

A conceptual flowsheet of the new polygeneration system is shown in Figure 4. The syngas 

produced in a Texaco gasifier is sent to a Waste Heat Boiler (WHB) to recover its sensible heat 

from 1346 to 231, in which the steam at 120bar will be superheated to 535 and then enters 

the steam turbine. After heat recovery, the syngas together with a 230 steam at 30bar enters the 

shift reactor for component adjustment, in which the mole ratio of H2/CO is adjusted to 2.65. Then 

the syngas leaving the shift reactor is cooled down to 40 after two heat exchangers, in which 

steams of 230 at 9.8bar and 180 at 5bar are generated and will be sent to the sequential cleanup 

unit. In the clean-up unit, most of the sulfides and CO2 in the syngas will be removed through 

the low-temperature methanol wash process. The CO2 from the clean-up unit is high concentrated 

(mole fraction is typically over 99%). Then, the purified syngas, in which the mole content of H2S 

is lower than 10 -6 , with CO and H2 will be sent to the methanol synthesis unit sequentially. 

1 2 3 

4 

1-raw syngas; 2-fresh gas; 3-crude methanol; 4-unrecycled gas for power; 5-unrecycled gas for 

SNG synthesis; 6-cycled gas for SNG synthesis; 7-crude SNG; 8-recovered methanol; 9-distillation 

waste; 10-inlet gas for gas turbine; 11- inlet gas for HRSG; 12- flue; 13-methanol product; 14-SNG 

product; 15-CO2 product 

Fig. 4. New polygeneration system adopting proper composition adjustment and partial-recycle 

scheme with CO2 capture 

Liquid Phase Methanol synthesis (LPMEOH) technology is adopted. The reaction occurs at 76bar, 

and the temperature of the reactor is kept by the evaporation process of cooling water at about 200 

.After the reactor, the crude products will be cooled down to 40 through two heat exchangers, 

in which the steam for shift reaction and a steam saturated at 3.75bar are generated. Then, the crude 

methanol and the unreacted gas will be separated by flash. Crude methanol is sent to a flash drum to 

remove the dissolved gas, and then fed into a three-stage distillation unit. The unreacted gas will be 

divided into three streams: the recycled gas, which will be recycled back to the methanol synthesis 

reactor, the reacting gas for SNG synthesis, and the unrecycled gas, which will be sent to the 

combined cycle subsystem for power. 

High Pressure SNG synthesis technology is adopted in this polygeneration system. The reacting gas 

for SNG synthesis from the methanol synthesis unit is sent to three sequential adiabatic reactors at 

about 79bar. Nickel-based catalyst MCR-2X developed by Topsoe company is used for SNG 

synthesis, which can keep high activity in the range of 300-700. Between each reactor, 

exchangers are placed to recover the high-temperature heat of the products to avoid the 

4 

15 

5 

9 

6 

10 11 

7 

8 

12 

13 

14

ineffectiveness of the catalyst, in which steam at 535 and 120bar is generated and sent to 

combined cycle subsystem for power. After the three reactors, the products is cooled down to 25 

to remove water and then sent to a flash drum to separate a spot of methanol which is mixed in the 

SNG. SNG of over 93% CH4 (mole basis) is produced and the separated mixed methanol is sent for 

combustion in the combined cycle subsystem for power. 

As shown in Figure 4, the new system applies the serial connection between the chemical 

production process and power generation system: all of the syngas produced by gasifier enters the 

chemical production process at first, and after methanol production and SNG production, the 

unreacted gas is then sent to power generation subsystem. 

Compared with the single methanol and SNG synthesis production process, the new system has the 

following key features: (1) Adopting partial adjustment of composition of the fresh gas instead of 

full adjustment to satisfy the H2/CO for both methanol synthesis and SNG synthesis, and thus 

avoiding greater exergy destruction for individual adjustment of composition of fresh gas in single 

production systems; (2) Instead of total recycle of the unreacted gas in methanol synthesis unit, 

partial-recycle scheme is adopted in the polygeneration system, avoiding the sharp increase of 

energy consumption for methanol synthesis, and at the same time realizing the SNG synthesis with 

no adjustment of the composition and efficient power generation; (3) recovery of the sensible heat 

of the syngas and recovery of gases released in chemical synthesis process for power. 

2.3. Description of the reference systems 

For performance comparison, a system of integrated gasification combined cycle (IGCC) for power 

generation is selected as a reference for the application of power generation. A coal-based methanol 

production process which adopts the low-pressure Lurgi methanol synthesis technology is selected 

as the reference for single methanol production system. And a coal-based SNG synthesis process, in 

which High Pressure SNG synthesis technology is applied, is selected as the reference system for 

SNG production. 

Figure 5(a) shows the flowsheet of IGCC system, including the air separation unit (ASU), a Texaco 

slurry-feed and O2-blown gasification unit, clean up unit adopting Selexol process to remove 

sulfids, and a combined cycle unit. 

Figure 5(b) shows the flow sheet of single methanol product process, which can be identified into 

fresh gas preparation subsystem and methanol synthesis subsystem. An ASU unit, a Texaco 

gasifier, and a cleanup unit are included in fresh gas preparation unit. The composition adjustment 

unit for producing fresh gas with proper CO/H2, which can meet the synthesis of methanol, is also 

included in fresh gas preparation subsystem. The methanol synthesis subsystem includes the 

synthesis unit and distillation unit. In methanol synthesis unit, nearly all of the unreacted gas is sent 

back to the reactor to pursue highest conversion of materials. In the distillation unit, the crude 

methanol will be refined as the final product. The steam and work requirements in single methanol 

product process are supplied by a captive power plant with coal-fired boiler, and by recovering the 

surplus heat of the chemical reaction. 

The single SNG product process is shown in Figure 5(c), which can also be divided into fresh gas 

preparation subsystem and SNG synthesis subsystem. Different from the single methanol process, 

in the cleanup unit CO2 and sulfids will be removed by low-temperature methanol wash method. In 

SNG synthesis unit, three sequential reactors adopting TREMP process designed by Haldor Topsoe 

are placed, in which synthesis reaction occurs at about 80bar, above 300. Between reactors, steam 

with high parameters, usually 120bar, 535, will be generated to recover heat from the products in 

case the catalyst was burnt. The superheated steam will supply the work requirement of the single 

SNG product process, the insufficient of which will be provided by a captive power plant with coalfired 

boiler if necessary. 

5

(a) 

(b) 

(c) 

Fig. 5. Flow diagrams of single product systems: (a) IGCC system (b) single methanol product 

process (c) single SNG product process 

3. PERMORMANCE ANALYSIS OF THE NEW SYSTEM 

For evaluating the performance improvement of the polygeneration system, a criterion of Energy 

Saving Ratio (ESR) is defined as follows: 

[ P/ CGmthCmth 

GSNG CSNG ] F 

ESR 

P/ G C G C 

C mth mth SNG SNG 

where P represents the net power output of the polygeneration system, kW; C represents the 

thermal efficiency of the power reference system (IGCC); G represents the mass flow rate (kg/s); 

C represents the energy consumption for unit methanol production in a single methanol 

mth 

production process, kJ/kg; C mth is the energy consumption for unit SNG production in a single SNG 

production process, kJ/kg; F represents the lower heating value (LHV) of total fuel input for the 

polygeneration system, kW. The ESR denotes that how much fossil fuel will be saved if the same 

products are produced in the polygeneration system as that in the reference systems. 

The thermal efficiency and exergy efficiency are defined in formula (2) and (3). 

[ PEmth ESNG 

] 

ESR 

F 

where Emth and ESNG represents the energy output of methanol and SNG respectively, kW. 

6 

(1) 

(2)

[ P EXmth EX SNG] 

ESR 

F 

where EXmth and EXSNG represents the exergy output of methanol and SNG respectively, kW. 

The performance of the new polygeneration system and the reference systems is simulated by 

Aspen Plus software. The thermodynamic properties of syngas and chemicals are calculated by the 

Peng-Robinson and Redlich-Kwong equations respectively. The key parameters in the 

polygeneration system are selected to agree with the reference systems. For example, the inlet 

temperature of gas turbine is 1200, the pressure ratio is 16.5, and the isentropic efficiency for gas 

turbine, high-pressure, middle-pressure and low-pressure steam turbines are selected to be 0.9, 0.88, 

0.89 and 0.86 respectively. Datong coal was assumed as the basis for this study, whose LHV is 

26,710 kJ/kg. The coal analysis data along with some other basic conditions for simulation are 

listed in table 1. 

Table 1. Basic condition for simulation 

C f 

H f 

7 

(3) 

Coal component analysis (weight %) 

68.54 3.97 6.85 0.74 1.08 

Ash f 

W f 

9.98 8.84 

O f 

Condition for gas turbine 

parameter value 

Inlet temperature () 

N f 

S f 

1200 

Pressure ratio 16.5 

Isentropic efficiency 0.9 

Inlet temperature () 

Inlet pressure 

High/Middle/Low (bar) 

Isentropic efficiency 

High/Middle/Low pressure 

Pinch temperature difference of HRSG () 

Condition for steam cycle 

535 

120/39/3.75 

0.88/0.89/0.86 

17

The stream data corresponding to the points indicated in Figure 4 is listed in Table 2. The 

simulation results of the polygeneration system and its reference systems are summarized in Table 

3. 

Table 2. Stream data for the flow sheet in Figure 4 

1 2 3 4 5 6 7 8 9 10 13 14 15 

Pressure(bar) 68.0 27.4 74.5 74.5 74.5 80.0 77.7 77.7 4.9 16.4 1.1 77.7 30.0 

Temperature() 1346 51 40 40 40 423 25 25 92 1200 66 25 40 

Flow rate 

(mol/s) 

Mole fraction,% 

2763 2088 169 523 1045 5301 279 6 2.8 4967 163 273 957 

CO 44.0 27.0 0.3 24.3 24.3 4.3 0.1 1.5 0.1 

CO2 11.7 0.9 0.7 1.6 1.6 2.9 5.4 2.2 34.1 3.0 5.5 99.9 

H2 30.5 71.6 0.3 73.4 73.4 14.1 0.2 0.7 0.2 

CH4 38.7 91.4 4.3 93.2 

COS 0.6 

H2S 0.3 

H2O 12.6 0.4 1.7 38.7 0.1 6.0 0.2 8.2 0.1 

N2 0.3 0.2 74.5 0.2 

O2 14.3 

CH4O 97 0.6 0.6 1.1 2.4 87.3 58.7 99.8 0.7 

C2H6O 0.1 0.1 0.2 4.7 0.1 

Others 0.1 

Table 3. The performance of the polygeneration (PG) system 

Items PG with 

8 

Reference systems 

CO2 capture IGCC Single methanol Single SNG 

process process

Fuel inputkW 728221 203771 237719 374571 

SNG output 

Flow ratekg/s) 

LHVkJ/kg) 

SNG output(kW 

Methanol output 

Flow ratekg/s) 

LHVkJ/kg) 

Methanol outputkW 

4.9 

42228 

206918 

5.2 

19938 

103679 

Electricity output(kW 89456 89456 

Energy consumption for the 

polygeneration systemkW 

Air&O2 compression in ASU 

Air compressor for combustion 

Fresh gas compressor 

Electricity needs for clean-up unit 

Coal cracker 

Pumps 

The compressor of recycled gas in 

methanol synthesis unit 

The compressor of recycled gas in SNG 

synthesis unit 

Work output in gas turbine 

Work output in steam turbine 

36461 

58070 

9395 

518 

3090 

915 

67 

592 

-111652 

-86912 

9 

5.2 

19938 

103679 

4.9 

42164 

206918

Net work output -89456 

Thermal efficiency 54.9% 43.9% 55.2% 

ESR 10.8% 

With 62% of carbon captured in chemical production process, the CO2 emission rate in the new 

polygeneration system is 0.278kg/kWh, which is much lower than 0.763kg/kWh in IGCC system. 

Moreover, compared with the single methanol product process, single SNG product process and 

IGCC system, the thermal efficiency of this polygeneration system is as high as 54.9% and can save 

10.8 percent points of fuel input. This indicates that the new polygeneration system has obvious 

advantages at CO2 capture over the single product systems. 

What most attractive is that the new polygeneration system with CO2 capture even performances 

better than most energy systems without CO2 capture. For example, the thermal efficiency of IGCC 

without CO2 capture is about 43.9%, much lower than 54.9% of the new polygeneration system 

with 62% of carbon captured. 

Fig. 6. The thermal performance of the novel system at different Rc1 (methanol to power ratio) 

Fig. 7. The thermal performance of the novel system at different Rc2 (SNG to power ratio) 

10

Not only at desined Rc (the ratio of chemical products to power) does the new polygeneration 

system performance well, but also at other variable Rc. Results show that when Rc varies from 0.5 to 

3, this new polygeneration system can always save primary energy compared with single product 

systems. And an optimal Rc exists to make the thermal performance of the new polygeneration 

system highest, as shown in Figure 6 and Figure 7. The results illustrated in figures 6 and 7 show 

that the polygeneration system achieve best performance by adopting partial recycle instead of full 

recycle of the unreacted gas. 

To disclose the key process of energy saving and to reveal the high performance of the 

polygeneration system, exergy analysis is applied and listed in Table 4 (at designed Rc). Compared 

with the single product systems, the exergy destruction in the new polygeneration system can 

decrease by 89.2MW assuming the same product output. In the new polygeneration system, the 

internal power consumption, for example the work for compressors is supplied by steam turbines 

and gas turbines instead of the captive power plant of single product systems, and this change can 

decrease exergy destruction by 38.2MW. This is because coal-based steam system is adopted in 

captive power plant, and the huge temperature difference between the combustion temperature of 

coal, as high as 1600, and the 600 steam leads to large amount of exergy destruction of the 

captive power plant. Whereas, in this polygeneration system, the difference between the inlet 

temperature of gas turbine, as high as 1200, and the combustion temperature of fuel is much 

smaller, and thus the exergy destruction is much less. By adopting proper component adjustment 

instead of full adjustment of syngas and abolishing the component adjustment in SNG process, the 

shift process can decrease exergy destruction by 3.1MW. Waste heat boiler instead of quench to 

recover sensible heat of the raw syngas can decrease exergy destruction by 6.0MW. And recovery 

of the chemical emissions to combustion for power can decrease exergy destruction by 15.6MW. 

Other processes, like the ASU or gasification process, can also decrease the exergy destruction for 

the less input of fuel in the new polygeneration system. 

Table 4. The exergy analysis for polygeneration system and reference systems 

Items 

Polygeneration 

system with CO2 

MW 

capture 

% 

11 

Single product systems 

Total of ref IGCC MEOH SNG 

MW 

MW 

MW MW 

Fuel exergy 739.1 100 828.3 206.8 241.3 380.2 

Exergy output 

SNG 215.9 29.2 215.9 215.9 

Methanol 118.4 16.0 118.4 118.4 

Electricity 89.5 12.1 89.5 89.5 

Exergy 

destruction 

ASU 18.8 2.5 21.0 5.6 5.3 10.1

Gasification 100.6 13.6 113.0 31.4 27.0 54.6 

Cooling of syngas 29.3 4.0 35.3 5.0 10.6 19.7 

Shift of syngas 16.7 2.3 19.8 7.7 12.1 

Cleanup 19.0 2.6 23.3 2.1 10.2 11.0 

MEOH synthesis 6.7 0.9 4.1 4.1 

MEOH 

distillation 

4.3 0.6 5.8 5.8 

SNG synthesis 27.8 3.8 20.2 20.2 

Captive power 

plant 

Air compression 

& combustion 

42.0 6.2 36.4 36.4 

Gas turbine 4.6 0.6 9.1 9.1 

Steam turbine& 

pumps 

38.2 29.4 8.8 

9.4 1.3 15.5 10.5 5.0 

HRSG 3.6 0.5 5.4 5.4 

Exergy emission 

Exhaust emission 4.0 0.5 7.0 7.0 

ASU N2 emission 6.5 0.9 11.3 1.2 1.2 8.9 

Emission of 

cleanup unit 

Emission of 

chemical island 

Exergy efficiency 

% 

22.0 3.0 23.5 3.6 6.0 13.9 

57.3 

15.6 15.6 

12

4. DISCUSSION 

The graphic exergy analysis (EUD methodology) for the key processes in the polygeneration 

system is adopted to reveal the internal phenomena of high performance of the key processes. A 

represents for the energy level, and H represents for the enthalpy change. Figure 8(a) illustrates 

the Energy Utilization Diagram for combustion of coal in boiler of captive power plant of single 

product systems. The oxidation of coal (curve Aed1) and the heat recovery in coal economizer (curve 

Aed2) act as the energy donor. The energy acceptors include preheating of air (curve Aea1), the 

preheating of fuel (curve Aea2), preheating (from 90 to 351.6), evaporation, superheating (from 

351.6 to 540) of water in boiler (curve Aea3), and the water preheating in boiler economizer 

(from 50 to 90, curve Aea4). The shaded area between energy donor and acceptor represents for 

the exergy destruction of the process. The area 1 in Figure 8(a) represents the exergy destruction 

in boiler, and 2 for exergy destruction in coal economizer. The average energy level of coal is 

about 1.03, but the average energy level of superheating steam is no more than 0.65. This indicates 

that huge difference exists between the energy level of energy donor and acceptor, which causes 

large amount of exergy destruction in captive power plant of single product systems. 

Fig. 8. (a) EUD for combustion of coal in boiler of captive power plant of single product systems; 

and (b) EUD for the combined cycle in the new polygeneration system 

13 

(b) 

(a)

Figure 8(b) discloses the exergy utilization of combined cycle in the novel polygeneration system. 

The energy donor includes the oxidation of unreacted gas and recovery gas (curve Aed1), the gas 

turbine (curve Aed2) and the heat recovery of gas in HRSG (curve Aed3). The energy acceptors are 

the preheating of fuel gas (curve Aea1), the preheating of air (curve Aea2), and the water heating in 

HRSG (from 30 to 535, curve Aea3). 

Comparing Figure 8(a) with Figure 8(b), the difference of energy level between energy donor and 

acceptors in the polygeneration is much smaller for the adoption of gas turbine and HRSG, and that 

resulted in sharp decrease of exergy destruction in the novel polygeneration system. And this 

example is exactly coincided with the general principle of system integration “Cascade utilization 

of chemical and physical energy”. 

Fig. 9. (a) EUD for quench process in single product systems; and (b) EUD for heat recovery of 

sensible heat of syngas 

14 

(a) 

(b)

Figure 9(a) illustrates the energy utilization of quench process in single product systems, in which 

cooling water of about 30 will be jet to mix with the hot syngas continuously to lower the 

temperature of the syngas very quickly. Figure 9(b) illustrates the energy utilization of recovery of 

sensible heat of syngas in the novel polygeneration system. In both Figure 9(a) and 9(b), the cooling 

down of syngas (curve Aed1) acts as the energy donor and the heat absorption of water acts as the 

acceptor (curve Aeas). It can be found that the average energy level of the acceptor in the novel 

polygeneration system has been increased for just one jet of water in WHB is adopted, whereas the 

average energy level of the acceptor in single product systems has been decreased for the adding of 

the cold water continuously, and thus the exergy destruction in the novel system has been decreased 

by 6.0MW for such process. 

Fig. 10. EUD for recovery of chemical emissions 

Figure 10 discloses exergy utilization in the process of recovery of chemical emissions. Curve Aed 

represents for the oxidation of the recovery chemical emissions. Curve Aea1 represents for the fuel 

heating process, and cure Aea2 represents for the air heating process. In single product systems, the 

chemical emissions will be emitted to atmosphere directly after combustion, whose exergy 

destruction is represented by the area between curve Aed and the horizontal ordinate. Whereas, in 

the novel polygeneration system, the chemical emissions will be recovered for power, and the 

shaded area 1 represents for the exergy destruction of this process, which is much smaller than 

that in single product systems. 

5. CONCLUSIONS 

In this paper, a novel coal-based polygeneration system with CO2 capture, which cogenerates 

power, natural gas and liquid fuel, has been proposed. With 54.9% of thermal efficiency and 62% of 

carbon captured, the primary energy saving ratio of this novel polygeneration system can reach as 

high as 10.8 percent compared with the single product systems. Based on the graphical exergy 

analysis, it is disclosed that abolishing captive power plant, converting the unreacted gas from 

methanol synthesis unit into SNG without composition adjustment, and recovering chemical 

emissions for power play an important role in decreasing exergy destruction in the novel system. 

This novel system has realized “the cascade utilization of chemical and thermal energy of the coal” 

and realized the CO2 separation with low energy penalty. The promissing results obtained in this 

coal-based polygeneration system can realize both the coal decarbonization with low energy penalty 

15

and the clean utilization of coal, and will possibly provide a new path to enforce the safety of 

energy supply. 

ACKNOWLEDGEMENTS 

This work has been supported by Supported by the major international cooperation projects of the 

National Natural Science Foundation of China (Grant No. 50520140517) and supported by NFSC 

Projects (No. 50706052). 

REFERENCES 

[1] BP (British Petroleum), 2009. BP statistical review of world energy 2009. Available from: 

. 

[2] X.M. Ou, X.Y. Yan, X. L. Zhang, 2010. Using coal for transportation in China: Life cycle 

GHG of coal-based fuel and electric vehicle and policy implications, International Journal of 

Greenhouse Gas Control, 4(2010), pp. 878-887. 

[3] IEA (International Energy Agency), 2003. CO2 emissions from fuel combustion 1997-2001, 

IEA/OECD, Paris, France, 2003. 

[4] IPCC (Intergovernmental Panel on Climate Change), 2001. Climate Change 2001, 3rd 

assessment report of the Intergovernmental Panel on Climate Change, Cambridge University 

Press, Cambridge, UK, 2001. 

[5] IPCC (Intergovernmental Panel on Climate Change), 2005. Special Report on Carbon dioxide 

Capture and Storage, Eighth Session of IPCC Working Group III, Montreal, Canada, 2005. 

[6] H. G. Jin, L. Gao, W. Han, 2007. A NOVEL COAL-BASED POLYGENERATION SYSTEM 

OF POWER AND LIQUID FUEL WITH CO2 CAPTURE, Proceedings of GT2007 ASME 

TURBO EXPO 2007: Power for Land, Sea and Air May 14-17, 2007, Montreal, Canada. 

[7] Chiesa, P. and Consonni, S., 1999. Shift reactors and physical absorption for low-CO2 emission 

IGCCs, ASME Trans., Journal of Engineering for gas turbines and power, 121, pp. 295-305. 

[8] Lozza, G. and Chiesa, P., 2002. Natural gas decarbonization to reduce CO2 emission from 

combined cycles-part I: partial oxidation, ASME Trans., Journal of Engineering for gas 

turbines and power, 124, pp. 82-88. 

[9] Lozza, G. and Chiesa, P., 2002. Natural gas decarbonization to reduce CO2 emission from 

combined cycles-part II: steam-methane reforming, ASME Trans., Journal of Engineering for 

gas turbines and power, 124, pp. 89-95. 

[10] Chiesa, P. and Lozza, G., 1998. CO2 Emission Abatement in IGCC Power Plants by Semiclosed 

Cycles, Part A: With Oxygen-blown Combustion. ASME98-GT-384. 

[11] Y. Inui, T. Matsumae, H. Koga, K. Nishiura, 2005. High performance SOFC/GT combined 

power generation system with CO2 recovery by oxygen combustion method. Energy 

Conversion and Management 46 (2005), pp. 1837-1847. 

[12] Q. Yang, Jerry Y.S. Lin, 2006. Fixed-bed performance for production of oxygen-enriched 

carbon dioxide stream by perovskite-type ceramic sorbent., Separation and Purification 

Technology 49 (2006), pp.27–35. 

[13] Roda Bounaceur, Nancy Lape, Denis Roizard, Cecile Vallieres, Eric Favre, 2006. Membrane 

processes for post-combustion carbon dioxide capture: A parametric study, Energy 31 (2006), 

pp. 2556–2570. 

[14] Luis M. Romeo, Irene Bolea, Jesus M. Escosa, 2008. Integration of power plant and amine 

scrubbing to reduce CO2 capture costs, Applied Thermal Engineering 28 (2008), pp. 1039– 

1046. 

[15] H.G. Jin, S.E. Sun, W. Han, L. Gao, 2007. A Novel Multi-functional Energy System for Coproducing 

Coke, Hydrogen and Power, Proceedings of the ASME 2007 International Design 

16

Engineering Technical Conferences & Computers and Information in Engineering Conference 

IDETC/CIE 2007 September 4-7, 2007, Las Vegas, Nevada, USA. 

[16] H.G Jin, W. Han, L. Gao, 2007. A NOVEL MULTI-FUNCTIONAL ENERGY SYSTEM 

(MES) FOR CO2 REMOVAL WITH ZERO ENERGY PENALTY, Proceedings of GT2007 

ASME Turbo Expo 2007: Power for Land, Sea, and Air May 14-17, 2007, Montreal, Canada 

GT-2007-27680. 

[17] Möllersten K., Yan J., 2001. Economic evaluation of biomass-based energy systems with CO2 

capture and sequestration—The influence of the price of CO2 emission quota, World Resour 

Rev, 13(4), pp. 509-525. 

[18] Möllersten K., Yan J., Moreira JR., 2003. Potential market niches for biomass energy with CO2 

capture and storage—opportunities for energy supply with negative CO2 emissions. Biomass 

and Bioenergy, 25(3), pp. 273-285. 

[19] Obersteiner M., Azar Ch., Kauppi P., Möllersten K., Moreira J., Nilsson S., Read P., Riahi K., 

Schlamadinger B., Yamagata Y., Yan J., Van Ypersele J.-P., 2001. Managing Climate Risk, 

Science, 294(2001), pp. 786-787. 

[20] IEA (International Energy Agency), 2001Key World Energy Statistics from the IEA, 2001 

Edition. 

[21] H.G. Jin., Ishida M., 1997. A New Advanced IGCC Power Plant with Chemical-Looping 

Combustion, Proc. of TAIES’97, pp. 548-553, Beijing, 1997. 

[22] W. D. Ni, H.S. Zhen, Z. Li, N. Jiang, 2003. Polygeneration A Very Important Way to 

Overcome Five Challenges in Energy Field of China, POW ER ENGINEERING, (3)2003, pp. 

2245-2251. 

17




Analysis and Optimization of CO2 Capture in a 

China’s Existing Coal-fired Power Plant 

Gang Xu a , Yongping Yang b , Shoucheng Li c , Wenyi Liu d and Ying Wu e 

a North China Electric Power University, Beijing, China, xg2008@ncepu.edu.cn 

b North China Electric Power University, Beijing, China, yyp@ncepu.edu.cn 

c North China Electric Power University, Beijing, China, lishoucheng6363@126.com 

d North China Electric Power University, Beijing, China, lwy@ncepu.edu.cn 

e North China Electric Power University, Beijing, China, 837469236@qq.com 

Abstract: 

In China, pulverized coal-fired power plants provide over 70% of the total electricity, on the other side, make 

up nearly half of the total CO2 emission volume of the whole country. Thus, CO2 capture in these coal fired 

power plants will be extremely important to the effort of CO2 reduction made worldwide. However, to retrofit 

existing power plant for CO2 capture may encounter many constrains from the layout of original process and 

the structure of existing equipments, causing a lot of special problems in process design and bringing deep 

influence on system performance, which in turn requiring special considerations in system integration. In 

view of these factors, this paper carried out the process simulation, characteristics analysis and system 

integration of CO2 capture based on a typical China’s existing coal-fired power plant with supercritical 

parameters. The paper analyzes main constrains encountered in retrofitting existing power plant with CO2 

capture using monoethanolamine (MEA) solution and puts forward several special system integration 

schemes for CO2 capture in an existing 600MW unit of China. The results revealed that, due to the 

constrains of the layout of original process and the structure of existing equipments, efficiency penalty of 

CO2 capture in a existing power plant will be even higher than a re-design new power plant by 3-5%-points. 

However, through the special system integrations, the efficiency of such retrofitting existing power plant can 

increase by 2-4%-points. The research of this paper may provide a feasible technology solution for 

decarburization retrofits of existing power plants, and promote CCS technologies into application. 

Keywords: 

CO2 Capture, Existing Coal-fired Power Plant, Retrofit, Thermal Energy Integration. 


Increasing concentration of CO2 and other greenhouse gases (GHG) is the main reason behind 

alarming environmental phenomena, such as global warming and sea level rising [1-2]. China, one 

of the world’s largest producers of CO2 emissions, is responsible for approximately one fifth of 

global CO2 emissions [3]. 

Different from many industrialized countries, China’s main primary energy is coal, which is a kind 

of cheap but carbon-intensive energy resources. And in China, pulverized coal fired power plants, 

whose total installed capacity is over 700GW, provide nearly 80% of the total electricity, however, 

make up almost half of the total CO2 emission volume of the whole country [4]. Thus, the reduction 

of CO2 emissions in the electricity supply sector of China, especially in these pulverized coal fired 

power plants, will make a significant contribution to the country and even to the whole world. 

Though suffering of high energy and cost penalty, CO2 capture and storage (CCS) is commonly 

considered as a technically feasible method of making deep reductions in carbon dioxide emissions 

from sources such as energy utilization systems, and attracted great attentions worldwide [5-11]. At 

present, there are three basic technologies for capturing CO2 from energy systems: post-combustion 

capture, oxy-fuel combustion capture, and pre-combustion capture. As for CO2 separation process, 

18

generally there are four kinds of methods, that is, absorption (including chemical and physical 

absorption), adsorption, membrane and cryogenic separation [1,12]. For pulverized coal fired power 

plant, post-combustion capture with chemical absorption using an aqueous solution, such as 

monoethanolamine (MEA), is recognized as one of the most feasible technologies for the sake that 

it is suitable for removing CO2 at low concentration, quite mature in technology, and easy to make 

great improvements. [7-8,12-15]. 

During the past few decades, recovering CO2 by chemical absorption has been investigated by 

many researchers [8-9, 16-26]. For example, Alie et al. presented a detailed simulation method for a 

typical CO2 capture process using MEA solvent, and carried out the optimization of key process 

operating variables [8]. Jean-Marc and Pellegrini respectively analyzed the influence of different 

absorbents (MDEA-TETA and ammonia) for regeneration energy [9,16]. Mohammad et al. 

investigated the technical and economic performance of CO2 capture from power plants in detail 

[17,18]. Hetland et al. integrated a full carbon capture scheme onto a 450MW nature gas combined 

cycle power station [19]. Huang et al. conducted the industrial test and techno-economic analysis of 

CO2 capture in Huaneng Beijing coal-fired power station. These researches disclosed the basic 

characteristics of the coal-fired power plants with chemical CO2 absorption process and revealed 

that post-combustion is a good option for the capture of CO2 produced by commercial coal-fired 

power plants [20]. 

Besides, a few researchers are also paying attention to the system integration of CO2 capture 

process with power generation system [21-26]. For example, Sanpasertparnich et al. integrated postcombustion 

capture and storage into a pulverized coal-red power plant [23]. Gibbins et al. put 

forward the CO2 capture ready(CCR) plant, They focus on newly-built plant and propose three 

different turbine options for CCR plant [25]. However, most of such integration researches neglect 

the restrictions of the existing power generation assembly and make great modification in the steam 

system of plant, which may be possible in a newly-built plant with thoroughly redesign but not 

suitable for the existing power plant. In fact, to retrofit existing power plant for CO2 capture may 

encounter many constrains from the layout of original process and the structure of existing 

equipments, causing a lot of special problems in process design and bringing deep influence on 

system performance, which in turn requiring special considerations in system integration. However, 

few studies pay much attention to these special phenomena in the research of large scale CO2 

capture in existing power plants 

In view of the importance of the CO2 reduction of China’s enormous existing power plants, this 

paper carries out the process simulation, characteristics analysis and system integration of CO2 

capture based on an existing supercritical power plant in China. Through this study, the paper 

achieves the following targets: (1) to reveal main constrains encountered in retrofitting existing 

power plant with CO2 capture. (2) to put forward several special system integration schemes for 

CO2 capture in the typical existing power generation unit of China. (3) to provide feasible 

technology solutions for decarburization retrofits of existing power plants, and promote CCS 

technologies into application. 

2. Particularity of CO2 capture in existing power plant 

For the coal-fired power plant which uses chemical absorption method to reduce CO2 emissions, its 

thermal efficiency will decrease by 10-15%-points [23-24,27], to achieve a 90% CO2 recovery ratio. 

Most of such efficiency penalty comes from the energy consumption of CO2 capture process, 

particularly the heat requirement of solvent regeneration. For example, in the amine scrubbing 

process, an energy demand between 3.5 and 4.2 MJ/kgCO2 has been reported for solvent 

regeneration [17,23-24]. For CO2 capture in power plant, such huge amount of heat provided for 

19

solvent regeneration mainly comes from the condensation of the steam extracted from steam 

turbine. 

However, retrofitting of the existing power plant for CO2 capture would encounter many constrains 

and be more complex. Compared with the virtual plant or redesigned newly-built plant, the 

steam/water cycle of an existing power plant can not be made great changes due to the restriction of 

process and devices. 

2.1. Restrictions of steam extraction parameters 

In a chemical absorption process for CO2 capture, the solvent desorption temperature would vary 

with different absorbents. However, most chemical absorption methods need to provide thermal 

energy with temperature in the range of 100 o C and 150 o C for stripping process. Take MEA for 

example, the CO2-rich amine stream, leaving from the absorber bottom, is regenerated by thermal 

treatment at 100 o C up to 140 o C in the stripper, releasing CO2 [17,24-25,28]. The stripper makes use 

of steam extracted from the steam/water cycle of the power plant. As economic consideration, the 

extracted steam at 2.1-3.4bar is suitable to provide the solvent regeneration heat. 

Besides, the amount of extraction steam is enormous, which can be half of the total steam flow of 

LP turbine cylinders[24-25], due to the extensive heat demand of solvent desorption. However, in 

the existing power plant, it is impossible to extract too much steam within low-pressure turbines 

(LPT) due to the constraint of the structure of turbines. The only feasible steam extraction point for 

an existing power plant may be located at the crossover pipe between the intermediate pressure (IP) 

and low pressure (LP) cylinders of the steam turbine, [23,25]. And the crossover pipe is also the 

quite place to extract a large amount of steam for heat supply in many combined heat-and-power 

units[29-32]. 

In most supercritical or ultra supercritical units, the pressure of steam extracted from the IP/LP 

steam turbine can be as high as 9-12 bar [23,33-34], this is far higher than the required parameters 

of stripper for absorbent regeneration, which will bring extra power loss due to steam extraction. 

Figure1 shows the relationship of power loss per kg extracted steam with its pressure. As is shown 

in Fig. 1, the higher the extracted pressure, the higher the specific power loss. When the pressure of 

extracted steam reaches 9-12 bar, its power loss will be almost twice as much as that of the steam 

extracted at 2.1bar. 

Power loss kwh/kg 

0.24 

0.20 

0.16 

0.237 kWh/kg 

0.20 kWh/kg 

0.146 kWh/kg 

2.1 bar 

0.12 

0 2 4 6 8 10 12 

Steam extraction pressure bar 

20 

9 bar 

12 bar 

Fig. 1. Relationship between power loss and steam extraction pressure 

2.2. Off-design conditions of LP turbine due to huge extraction steam 

As is mentioned above, a large amount of heating steam will be extracted from the crossover pipe 

between IP and LP turbines in CO2 capture retrofitting of existing power plants, which leads to LP 

cylinders operating under off-design conditions. In this situation, the steam mass flow rate of LP 

cylinders will drop deeply, which leads to the substantial deviation of steam parameters from the 

rated values.

However, even under the off-design conditions, the performance characteristics of steam parameters 

within turbine will still comply with certain rules. If the steam velocity in a stage of a given stage 

group becomes equal to or greater than the critical velocity, the pressure behind that stage will 

influence the steam parameters in the preceding stages and, with the same clear cross-sectional area, 

the flow rate will depend, only on the steam parameters before the blade cascades of the preceding 

stages and will be determined by the equation G A 

p 

v 

. The ratio of an arbitrary steam flow rate 

through a group of stages to the rated flow rate can be represented in the form: 

G p T x T x 

 

(1) 

G p T x T x 

01 00 00 00 00 

1 

0 00 01 01 01 01 

Where p 00 , T 00 and x 00 are the pressure, temperature and dryness fraction of steam at the rated flow 

rate 0 G , and p 01 , T 01 and x 01 are those under the changed conditions with a new flow rate G. For 

superheated steam, the equation (1) with x 01= 

x 00 =1 can be simplified: 

G p T T 

 

(2) 

G p T T 

01 00 00 

1 

0 00 01 01 

In many cases, it can be taken approximately that the steam temperature in intermediate turbine 

stages remains constant on a change of flow rate. Thus until the steam velocity in a stage remains 

critical, the steam pressure in all preceding stages varies in direct proportion to steam flow rate. 

For cases when none of the stages of a group reach the critical velocity, the relationship between 

pressures and flow rate for an i-th stage under the assumption that T 01 = T 00 =constant can be given 

the following form: 

2 

G 

[( p ) ( p ) ] ( p ) ( 

p ) 

G 

0 

2 2 2 2 

00 i 20 i 01 i 21 i 

(3) 

Writing similar equations for all stages of the group considered and noting that the relative change 

of steam mass flow rate GG 0 is the same in all stages, we can sum the left-and right-hand parts of 

these equations: 

2 

z z 

G 

2 2 2 2 

[( p00 ) i ( p20 ) i ] [( p01) i ( 

p21) 

i ] 

G0 1 1 

(4) 

Since the final pressure of the i-th stage is equal to the initial pressure of the (i+1)-th stage, all 

intermediate values of pressure are cancelled and we have for a group of stages: 

G p p 

G p p 

2 2 2 2 

 

0 

01 

2 

00 

z1 2 

z0 01 z1 

2 

1z0 

21 

(5) 

Where 01 p01 p00 

are the relative pressure before the group of stages, and z pz p00 

is the 

relative pressure behind it. 

Let us introduce a correction factor equal to T00 T 01 in order to account for a probable 

temperature change before the group of stages. Then, the following formula is obtained for a group 

of stages operating with sub-critical velocities of superheated steam: 

G 

T 

G T 

2 2 

 

0 

01 z1 

2 

1z0 

00 

01 

(6)

In conclusion, if a group of turbine stages operates with steam velocities above the critical value, 

the steam flow rate on a change of steam state or one of the steam parameters on a change of flow 

rate can be found by formula (2); if steam velocities of all stages are subsonic, formula (6) is 

applicable. 

Fig. 2 shows the performance curves under different steam extraction proportion. Dotted line 

represents the conditions without consideration of pressure loss, which means that the inlet pressure 

of LP turbine will keep constant with part of steam extracted (see dotted line in Fig. 2a)), and the 

power loss of LP turbine is only caused by the fact that extraction steam doesn’t do any work. (see 

dotted line in Fig. 2b)). Obviously, it is a kind of assumptive operation status and impossible in a 

existing power plant. 

Change ratio of LPT inlet pressure 

1.2 


0.8 

0.6 

0.4 

0.2 

without the consideration of pressure loss 

with the consideration of pressure loss 

8% 17% 25% 33% 42% 50% 58% 

Steam extraction proportion 

22 

Change ratio of LPT power output 

1.0 without the consideration of pressure loss 

with the consideration of pressure loss 

0.9 

0.8 

0.7 

0.6 

0.5 

0.4 

0.3 

8% 17% 25% 33% 42% 50% 58% 

Steam extraction proportion 

(a) (b) 

Fig. 2. Variation trends of LPT performances with different steam extraction proportion: a) LPT 

inlet steam pressure, b) LPT power output. 

On the contrary, the solid line stands for practical operation conditions, that is, the steam turbines 

operate under the off-design condition when part of steam is extracted from the system, complying 

with formula (6). Here, the inlet pressure of LP turbine will decrease with the increase of steam 

extraction proportion (see the solid line in Fig. 2a)), which leads to more power loss of LP 

turbine(see the solid line in Fig. 2b)). This means that the pressure loss of extraction steam will 

bring an additional power loss. (see space between the solid and dotted line in Fig. 2b)). For 

example, when the steam extraction proportion is 50% of total steam flow LP cylinder, its pressure 

loss is approximately 50%(Fig. 2a)), and the power output of LP turbine is only 42%(Fig. 2a)). The 

power loss caused by the fact that the extracted steam doesn’t do work, accounts for nearly 50%. 

Meanwhile, additional power loss caused by pressure loss of extraction steam covers about 8%. In a 

word, large amount of steam extraction not only brings significant reduction of steam flow in LP 

cylinder, but also cause LP cylinder operation to deviate from design condition greatly, leading to 

additional power loss and further decrease of power plant efficiency. 

3. Case study based on existing 600MW Supercritical Unit 

3.1. Base Case: a typical 600MW Supercritical Unit in China 

A typical 600MW coal-fired power generation unit without CO2 capture in China is selected as 

base case. It is a pulverized coal fired power generating unit with a 600MW output adopting a 

supercritical pressure steam/water cycle, and the bituminous coal is selected as fuel. A Schematic 

diagram of the supercritical coal-fired power plant without CO2 capture is shown in Fig. 3.

Fig. 3. 600MW supercritical coal-fired power plant without CO2 capture process 

The selected steam turbine process flow diagram is shown in Fig. 4. The turbine consists of a HP, 

IP and LP sections all connected to the generator with a common shaft. Steam from the exhaust of 

the HP turbine is returned to the boiler for reheating and then sent to the double flow IP turbine. 

Exhaust steam form the IP turbines then flows into the double cylinders/four flows LP turbine 

system. 

Fig. 4. Steam/water cycle of the 600MW supercritical power unit 

And the overall performances of the unit is summarized in Table 1. Besides, the unit is designed to 

generate about 1677.5t/h of steam at nominal conditions of 24.2MPa and 566 o C with reheat steam 

heated to 566 o C, and the exhaust steam pressure of the steam turbine is 5.88kPa. These represent 

the typical parameters of China’s existing power generation unit. 

Fuel Parameters 

Table 1. Overall performance of Base Case 

Coal Heat Input (HHV) MJ/kg, ar 23.92 

Coal Heat Input (LHV) MJ/kg, ar 22.76 

Steam/Water Cycle Parameters 

Existing Steam Turbine Generator Output MW 604.3 

Total Auxiliary Power MW 30.22 

Net Output MW 573.8 

Overall Plant Performance Parameters 

Net Efficiency % 40.28 

23

Net Coal Consumption Rate g/kWh 305 

Net Heat Rate kJ/kWh 8937.1 

Overall Plant CO2 Emissions 

Carbon Dioxide Emissions g/kWh 867.8 

3.2. Amine scrubbing process for post-combustion CO2 capture 

In this study, the process design of CO2 capture was based on a standard monoethanolamine (MEA) 

absorption-desorption method. Fig. 5 shows the process of the CO2 separation unit. As is shown in 

Fig. 5, flue gas from the power plant is first cooled (down to a temperature of 40-50oC)and 

desulphurized in a flue gas desulfurization unit (FGD). Then, after passing through a booster fan, 

flue gas is absorbed by MEA in an absorber. The treated flue gas, from which most of the CO2 gas 

has been separated, is vented to the atmosphere. Rich solvent from the bottom of CO2 absorber is 

delivered to a cross heat exchanger by a pump (P1). Having been heated in the heat exchanger, it is 

delivered to a stripper to desorb CO2 by thermal treatment at 100oC up to 140oC. The CO2-H2O 

stream desorbed from the stripper is condensed and the moisture is removed to get more pure CO2 

in a condenser and a separator (Sp). The lean solvent is delivered to the cross heat exchanger by a 

pump (P2), and then to the absorber after cooling the stream to the designed temperature by a cooler 

(C1). Considering the degradation and volatilization of MEA, makeup of MEA solvent is also 

added to the absorber. Usually the CO2 separated from the stripper is compressed to the required 

pressure and temperature by the multistage compressor (CP) for CO2 transportation. 

Fig. 5. The typical CO2 capture process based on MEA 

The main parameters of the absorption process based on MEA are listed in Table 2. The pressure of 

the stripper is set as 2.1bar, and the stripper temperature is 115oC. The entire flue gas stream enters 

the absorber tower and the CO2 recovery ratio can reach over 90%. And the mass purity of CO2 

can reach 99.8%, which is high enough for CO2 storage or many industrial applications. 

Table 2. Main performance parameters with MEA-based CO2 capture process 

Title Number 

Desorber Pressure (bar) 2.1 

Temperature of reboiler (°C) 115 

CO2 recovery ratio (%) 90 

CO2 lean loading(molCO2/molMEA) 0.3 

24

CO2 rich loading(molCO2/molMEA) 0.45 

Energy consumption of reboiler (MJ/t CO2) 3404 

Energy consumption of condenser (MJ/t CO2) -684 

Mass purity of CO2 (%) 99.8 

Mole purity of CO2 (%) 99.6 

3.3. Capture Case 1: CO2 capture case without considering the 

constraint of existing power plant 

Case 1 is a typical 600MW coal-fired power generation unit combined with CO2 capture system. A 

simplified process flow diagram of Case1 is shown in Fig. 6. As it can be seen from this figure, flue 

gas of the generation unit directly enters the absorber of the capture process, and the thermal energy 

consumed by stripper reboiler is supplied by the steam extracted from the steam turbine subsystem. 

Fig. 6 illustrates the extraction scheme of the water/steam system. Thus, steam with pressure of 2.1 

bar is directly extracted from the IP cylinder of the turbine, supplying thermal energy with 

temperature of 115oC for stripper reboiler. The pressure selected can ensure a reasonable 

temperature differential in the reboiler. However, this scheme neglects the constraint of existing 

power plant. 

Fig. 6. 600MW supercritical coal-fired power plant with CO2 capture 

The power generation in Case 1 is the same as that of base case. Besides, its input fuel, boiler 

capacity, main steam and reheated steam flow rate also equal to that of base case. Neglecting the 

constraint of existing power plant, large amounts of 2.1 bar steam is directly extracted from the 

turbine system to provide heat and energy for MEA regeneration in the stripper without 

consideration of the pressure change in the IP cylinder after the large-scale extraction. 

25

Fig. 7. Steam extraction scheme of Case 1 Fig. 8. Steam extraction scheme of Case 2 

3.4. Capture Case 2: CO2 capture case with consideration of the 

constraint of existing power plant 

Case 2 is generally similar to Case 1 in the process flow. (see Fig. 6 for details). However, Case 2 

makes adequate consideration of the constraint of existing power plant. For example, as shown in 

Fig. 8, the steam extraction point of Case 2 is located at the crossover pipe between the intermediate 

pressure (IP) and low pressure (LP) cylinders of the steam turbine, which may be the only feasible 

point to achieve large amount steam extraction in an existing power plant. Here, the steam pressure 

of the crossover pipe between the IP and LP cylinders can reach 9.32bar, much higher than the 

required steam pressure for absorbent regeneration (about 2.1bar), which will bring extra power loss 

due to steam extraction. Furthermore, the extra pressure loss due to the large amount steam 

extraction and the additional power loss resulted from the extraction are also be well considered in 

Case 2. 

Besides, the steam extraction for amine absorption process can account for approximately 50% of 

the total steam flow exhausted from HP turbine cylinder. It may lead to unstable operation 

conditions and bring about some safety problems. For example, the pressure at the exhaust of the 

existing IP turbine would be dropped to a low level, which results in increased mechanical loading 

of the IP blades, especially the last stages of IP cylinder. Besides, because the flow area of the LP 

turbine cylinder is not variable, such a large decrease in the steam flow may lead to an unstable 

operation condition in the LP turbine. 

3.5. Performance analysis 

The performance analysis of three cases is listed in Table 3, the three cases include: 

Base Case: A typical 600MW supercritical power generation unit, as discussed in Section 3.1; 

Case 1: CO2 capture case without considering the constraint of existing power plant, as discussed 

in Section 3.3; 

Case 2: CO2 capture case with consideration of the constraint of existing power plant, as 

discussed in Section 3.4; 

Table 3. Performance analysis of Base Case and Case 1-2 

26 

Base Case Case1 Case2 

Coal input rate (kg/s) 46.66 46.66 46.66 

CO2 capture amount (kg/hr) - 447423 447423 

CO2 capture rate (%) - 90 90 

Reboiler heat duty (MW) - 420.83 420.83 

Extracted steam flow (kg steam/kg CO2) - 1.489 1.439 

CO2 compression work (kWhe/tonne CO2) - 39.91 39.91 

Power output of steam turbine (MW) 604 517.09 461.57 

Auxiliary work (MW) 30.22 85.66 85.44 

Net power output 573.8 431.43 376.13

Net efficiency (%) 40.28 30.29 26.40 

Efficiency penalty(%-points) - 9.99 13.88 

As is shown in Table 3, due to the fact that Case 1 and Case 2 adopt the same MEA CO2 capture 

process illustrated in the same Base Case, the process configuration and several basic parameters of 

these two cases are similar to each other, such as coal input rate, reboiler heat duty, CO2 capture 

amount and CO2 compression work. 

However, because of the difference of the two capture cases in the extracted locations, the flow rate 

and the parameters of extraction steam, the great differences lie in the power output of steam 

turbine, net power output and net efficiency. In fact, on account of the high extraction pressure and 

large power loss of IP cylinder after steam extraction in Case 2, its steam turbine output is only 

461.57MW, 55.52MW less than that of Case 1, which in turn leads to the obvious drop of net power 

output and net efficiency of Case 2 when compared with Case 1. Eventually, efficiency penalty of 

Case 2 reaches 13.88% points, nearly 4% points higher than that of Case 1. 

Though performance of Case 2 seems worse, more attention are paid to the production process of 

power station in this case, which is closer to the practice. In fact, even for a newly-built power 

station, the same constraints in CO2 capture process will be confronted if it uses the traditional 

station design. In other words, in terms of a practical pulverised coal power plant which adopts the 

chemical absorption method to achieve large-scale decarbonisation, the practical efficiency penalty 

will be much higher than the theoretical analysis if no specific optimization is made. 

In a word, specific optimization in the retrot scheme for CO2 capture in a pulverised coal power 

plants will be very helpful to control the penalty of CO2 capture at a low level, which will be 

discussed in the following section. 

4. Special integration for CO2 capture in existing power plant 

4.1. Add a new letdown steam turbine generator (LSTG) 

Since the steam pressure of the IP-LP crossover pipe (9.32bar) are much higher than the required 

steam pressure for solvent regeneration (about 2.1bar), a new letdown steam turbine generator is 

proposed to utilize the surplus pressure for power generation, as is shown in Fig. 9. 

Fig. 9. The structure of adding a new letdown steam turbine generator 

27

Such improvement is really simple and easy to implement, however, it is very effective to retrieve 

the surplus pressure, which can also make the power plant efficiency increase greatly. Fig.9 gives 

the variation trend of the power output and the extraction steam flow, with LSTG outlet pressure of 

the small turbine changed. As is shown in Fig. 10, with the decline of LSTG outlet pressure, the 

power output of LSTG will increase quickly, while the flow of the extracted steam is also 

increasing. The reason lies in that the temperature and enthalpy of the exhaust steam of LSTG will 

decrease with the drop of LSTG outlet pressure. As a consequence, it needs more extraction steam 

flow so as to provide the same energy. However, the enlargement of extraction steam flow and the 

pressure ratio caused by outlet pressure decrease will contribute to the increase of power output of 

LSTG. 

Extracted steams flow (t/h) 

710 

700 

690 

680 

670 

660 

650 

7 6 5 4 3 2 

New LSTG outlet pressure (bar) 

LSTG output (MW) 

28 

60 

50 

40 

30 

20 

10 

7 6 5 4 3 2 

LSTG outlet pressure (bar) 

(a) (b) 

Fig. 10. Variation trends of steam cycle performances with different LSTG outlet pressures: a) 

Extracted steam flow, b) Power output of new LSTG. 

4.2. Thermal energy integration 

For MEA-based CO2 capture process, on the one hand, the CO2 separation unit needs a lot of 

intermediate temperature steam extracted from steam turbine cycle to regenerate the solvent. On the 

other hand, the CO2 separation unit will also release a large amount of low temperature heat, such 

as the heat released by the CO2-H2O condenser of stripper and the intercooler of CO2 multistage 

compressor (Fig. 6). If these heat can be well employed, the energy consumption of CO2 capture 

will dramatically be reduced. 

Fig. 11 reports the basic information of the heat integration of steam turbine cycle with CO2 capture 

process. As is shown in Fig. 11, the main integration measures are: 

(1) the thermal energy released by CO2 cooler of MEA stripper are used for condensed water 

heating (HE3); 

(2) the thermal energy released by CO2 compressor intercoolers are used for condensed water 

heating (HE4); 

(3) After recovered surplus pressure within LSTG, the extracted steam is first sent to heat the 

condensed water (HE2), then sent to the reboiler of stripper (HE1).

Fig. 11. Schematic diagram of heat integration of steam turbine cycle with CO2 capture process 

As mentioned above, steam extracted from the IP-LP crossover pipe is 9.32bar, 695t/hr. And at the 

outlet of LSTG, the extracted steam is 3 bar, 244°C. However, the reboiler temperature should not 

exceed 135°C, otherwise the degradation of MEA and corrosion will be sharply aggravated. 

Therefore, the surplus heat of the extracted steam can be used to heat the condensed water (HE2), 

before it is sent to the reboiler of stripper (HE1). 

To recover the thermal energy of the CO2 capture system, the condensed water out of condenser is 

divided into two parts. One part, which accounts for about 45% of total flow, is sent to absorb the 

heat of CO2 cooler (HE3). The other part is used for recovering the inter-cooling heat of CO2 

compression (HE4). Such an integration scheme can totally replace the original low-pressure 

regenerative heater system to raise the temperature of the condensed water up to approximately 

155°C, before it enters the deaerator (DEA). 

4.3. Throttling one of LP cylinder of steam turbine (Case 5) 

As mentioned above, the flow rate of steam extraction for MEA solvent regeneration is almost half 

of the total inlet steam flow of original LP cylinder. In other words, the flow rate of low pressure 

steam after extraction is approximately equal to one of the two IP cylinder in design conditions. 

In view of this, if we let the low pressure steam after extraction flow into one LP cylinder, the LP 

cylinder can be considered to operate under design conditions and the flow rate similarly equals the 

design flow, as a result of which, the large-scale extra pressure drop caused by steam extraction can 

be avoided. 

However, for a conventional existing power plant, all of the turbine cylinder rotors are connected in 

the same shaft and it is impossible to completely clutch them from machine. Thus, we propose that 

through throttling, most of steam flow into one of the LP cylinders while only a small amount of 

steam enters the other cylinder for heat dissipation. Fig. 12 shows the process scheme. As it can be 

seen in the figure, the huge pressure drop of the LP cylinder can be avoided, which can ensure the 

increment of the power output of steam turbine as well as the net efficiency. 

29

RH1 

HP T IP T LP T LP T 

DEA 

RH2 RH3 RH5 RH6 RH7 RH8 

5 

New LSTG 

HE4 (intercooler 

of CO2 compressor) 

30 

HE2 

HE1 ( Reboiler of 

MEA Stripper) 

Throttling one LP cylinder 

4 

COND 

HE3 (CO2 cooler 

of MEA Stripper) 

3 2 

Fig. 12. The structure of clutched LP turbine option 

4.4. Results analysis 

The performance analysis of three cases of this section is listed in Table 4, the three cases include: 

Case 3: based on Case 2, Add a new letdown steam turbine generator (LSTG), as discussed in 

Section 4.1; 

Case 4: based on Case 3, adopting thermal energy integration, as discussed in Section 4.2; 

Case 5: based on Case 4, Throttling one of LP cylinder, as discussed in Section 4.3; 

Table 4. Performance analysis of Case 3-5 

Base Case Case 3 Case 4 Case 5 

Coal input rate (kg/s) 46.66 46.66 46.66 46.66 

CO2 capture amount (kg/hr) - 447423 447423 447423 

CO2 capture rate (%) - 90 90 90 

Reboiler heat duty (MW) - 420.83 420.83 420.83 

CO2 compression work (MW) - 39.91 39.91 39.91 

Outlet pressure of LSTG (bar) - 3.0 3.0 3.0 

Extracted steam flow (kg steam/kg CO2) - 1.553 1.553 1.553 

Heat recovery from CO2 capture(MW) - 0 73.05 73.05 

Power output of steam turbine (MW) 604 450.18 459.9 474.23 

Power output of LSTG (MW) - 43.69 44.59 44.59 

gross power output (MW) 493.87 504.49 518.82 

Auxiliary work (MW) 30.22 85.51 85.51 85.51 

Net power output 573.8 408.36 418.97 433.31 

Net efficiency of plant (%) 40.28 28.67 29.41 30.42 

Efficiency penalty (%-points) - 11.61 10.87 9.86 

As is shown in Table 4, compared with Case 2, when a new letdown steam turbine generator was 

added, the net efficiency of Case 3 can be increased by 2.27%-points, rising from 26.40% to

28.67%. In addition, through heat integration of steam turbine cycle with CO2 capture process, the 

net efficiency of Case 4 can be further increased by 0.74%-points compared with Case 2, rising 

from 28.67% to 29.41%. Furthermore, if one of LP cylinder of steam turbine can be throttled to let 

most of steam entering the other LP cylinder, the huge pressure drop of the LP cylinder can be 

avoided, which can make system net efficiency further increase by 1.01%-points, rising from 

29.41% to 30.42%. And through adopting all of three system integration measures, Case 5 presents 

the best performance. Compared with Base Case, the efficiency penalty of Case 5 is only 9.86%points, 

even less than that of Case 2. 

5. Exergy analysis 

To reveal the internal phenomena of the new integration system, an exergy analysis is performed for 

both Case 5 and Case 2 with CO2 capture. The results are listed in Table 5. The exergy analysis is 

also based on the assumption that the same quantity of coal was consumed in all of the Cases. 

As shown in Table 5, the exergy efficiency of Case 5 is 30.42%, which is 4.02% points higher than 

that of Case 2, 9.86% lower than that of Base Case. Comparing the exergy distributions of Case 5 

and Case 2, we find that the exergy of the net electricity has increased by 57.18 MW and the total 

exergy loss of the CO2 capture unit is obviously decreased. Otherwise, the exergy loss of power 

generation system is also reduced. And the detailed distribution of exergy loss of these two units is 

given in Table 5. 

Compared with the Case 2, the exergy loss of the CO2 capture unit of Case 5 is decreased 

remarkably 30.25MW. The reason lies in thermal utilization of CO2 capture, it makes heat exergy 

utilization rate improve greatly. In comparison, the exergy loss of power generation system of Case 

2 is 24.88 MW higher than Case 5. 

From the above analysis, we conclude that through thermodynamic system integration of 

steam/water system and CO2 recovery and cascade utilization of energy, the key problem of high 

energy penalty for CO2 capture may be improved in the Case 5 and favorable thermal and 

environment performances can be achieved. However, some challenges still exist, such as the 

complexity of the system, as well as the possible high investment of the system, which will be 

further studied in our following work. 

Table 5. Exergy analysis of Case 2-5 and Base Case 

BaseCase Case2 Case3 Case4 Case5 

MW MW MW MW MW 

Exergy input of coal 1424.53 1424.53 1424.53 1424.53 1424.53 

Exergy output 

Net electricity 573.80 40.28% 376.13 26.40% 408.36 28.67% 418.97 29.41% 433.31 30.42% 

Separated CO2 84.89 5.96% 84.89 5.96% 84.89 5.96% 84.89 5.96% 

Exergy loss 

CO2 recovery unit: 

CO2 separation 71.60 5.03% 71.60 5.03% 71.60 5.03% 71.60 5.03% 

CO2 compression 5.16 0.36% 5.16 0.36% 5.16 0.36% 5.16 0.36% 

Heat exergy 36.85 2.59% 36.85 2.59% 6.61 0.46% 6.61 0.46% 

Subtotal 113.61 7.98% 113.61 7.98% 83.36 5.85% 83.36 5.85% 

Power generation system: 

Boiler(Fuel combustion) 749.97 52.65% 749.97 52.65% 749.97 52.65% 749.97 52.65% 749.97 52.65% 

HTP 14.31 1.00% 14.31 1.00% 14.31 1.00% 14.31 1.00% 14.31 1.00% 

IPT 8.46 0.59% 8.27 0.58% 8.27 0.58% 8.27 0.58% 8.27 0.58% 

LPT 25.12 1.76% 12.16 0.85% 8.87 0.62% 12.17 0.85% 10.99 0.77% 

High temperature heater 5.23 0.37% 5.27 0.37% 5.27 0.37% 5.27 0.37% 5.27 0.37% 

Low temperature heater 6.00 0.42% 13.23 0.93% 4.82 0.34% 3.93 0.28% 4.36 0.31% 

Condenser 30.08 2.11% 20.93 1.47% 17.05 1.20% 21.94 1.54% 19.80 1.39% 

Throttling 15.82 1.11% 11.29 0.79% 16.29 1.14% 4.51 0.32% 

Other equipments 7.57 0.53% 5.60 0.39% 3.11 0.22% 3.20 0.22% 3.20 0.22% 

Subtotal 846.74 59.44% 845.56 59.36% 822.98 57.77% 835.37 58.64% 820.68 57.61% 

Exergy efficiency 40.28% 26.40% 28.67% 29.41% 30.42% 

31

6. Conclusions 

The process simulation, characteristics analysis and system integration of CO2 capture based on a 

typical China’s existing coal-fired power plant with supercritical parameters are carried out in this 

paper. From the work completed in this study, some important conclusions can be drawn out and a 

few of interesting integration measures are put forward. 

(1) When an existing power plant is transformed into a CO2 capture plant using chemical absorption 

methods, some special problems will be met with, which is very different from the virtual plant. 

On the one hand, it will be difficult to find a suitable extraction point for the large amount of 

steam which has to be supplied to the CO2 capture process. On the other hand, some component 

of the existing power plant, especially the steam turbine, will significantly deviate from their 

original design conditions because of a large amount of steam extracted from steam/water cycle, 

resulting in a large efficiency penalty. 

(2) When retrofitting existing power plant, due to the constraint of existing equipments, the energy 

penalty of CO2 capture will tend to be even higher. For example, because the parameters of 

extraction steam doesn't match with the steam parameters for CO2 capture process, it will be 

certain to bring additional power loss. Eventually, efficiency penalty of CO2 capture in an 

existing power plant (Case 2) can be 4% points higher than that of a redesigned new power 

plant(Case 1). 

(3) In this study, through the special (unique) system integrations(Case 3, Case 4, Case 5), the 

efficiency of existing 600MW supercritical power plant increased by 4.02%, rising from 

26.40%(Case 2) to 30.42%(Case 5). The overall studies in this report show that if MEA 

absorption is adopted to recover CO2 from flue gas of a power plant, with a CO2 recovery ratio 

of 90% the efficiency penalty for the power plant will be 9.86% points, and the extraction steams 

flow is 1.553 kg steams/kg CO2. 

Acknowledgments: 

The paper is supported by National Nature Science Fund of China (No. 51006034, No. 51025624), 

and the National Major Fundamental Research Program of China (No. 2009CB219801, No. 

2011CB710706). 

Reference 

[1] Working Group III of the Intergovernmental Panel on Climate Change (IPCC), IPCC Special 

Report on Carbon Dioxide Capture and Storage. Cambridge, United Kingdom: Cambridge 

University Press; 2005. 

[2] Ralph EH, Simsa, Rogner HH, Gregory K. Carbon emission and mitigation cost comparisons 

between fossil fuel, nuclear and renewable energy resources for electricity generation. Energy 

Policy 2003;31(13):1315-1326. 

[3] United Nations Statistics Division, Greenhouse Gas Emissions, August 2009. Available at: 

http://unstats.un.org/unsd/environment/air_CO2_emissions.htm. 

[4] China Electricity Council. The Current status of air pollution control for coal-fired power plants 

in China 2009. China Electric Power Press, Beijing; 2009. 

[5] Park SK, Kim TS, Sohn JL, Lee YD. An integrated power generation system combining solid 

oxide fuel cell and oxy-fuel combustion for high performance and CO2 capture. Applied Energy 

2011;88:1187–1196. 

[6] Sun R, Yingjie L, Hongling L, Shuimu W, Chunmei L. CO2 capture performance of calciumbased 

sorbent doped with manganese salts during calcium looping cycle. Applied Energy 2011. 

[7] Kanniche M, Gros-Bonnivard R, Jaud P, Valle-Marcos J, Amann JM, Bouallou C. Precombustion, 

post-combustion and oxy-combustion in thermal power plant for CO2 capture. 

32

Applied Thermal Engineering 2010;30:53–62. 

[8] Alie C, Backham L, Croiset E, Douglas PL. Simulation of CO2 capture using MEA scrubbing: 

a flowsheet decomposition method. Energy Convers Manage 2005;46(3):475–87. 

[9] Jean-Marc G. Amann, Bouallou C. CO2 Capture from Power Stations Running with Natural 

Gas (NGCC) and Pulverized Coal (PC): Assessment of a New Chemical Solvent Based on 

Aqueous Solutions of N-MethylDiEthanolAmine + TriEthylene TetrAmine. Energy Procedia. 

2009. p. 909-916. 

[10] Yukun H, Jinyue Y. Characterization of ue gas in oxy-coal combustion processes for CO2 

capture. Applied Energy 2011. 

[11] Hao L, Yingjuan S. Predictions of the impurities in the CO2 stream of an oxy-coal combustion 

plant. Applied Energy 2010;87:3162–3170. 

[12] Parsons Infrastructure & Technology Group, 2002, Updated cost and performance estimates for 

fossil fuel power plants with CO2 removal. Report under Contract No. DE-AM26-99FT40465, 

USA. 

[13] Rao AB, Rubin ES. A technical, economic, and environmental assessment of amine based CO2 

capture technology for power plant greenhouse gas control. Environmental Science and 

Technology 2002;36(20):4467-4475. 

[14] Singh D, Croiset E, Douglas PL, Douglas MA. Techno-economic study of CO2 capture from an 

existing coal-fired power plant: MEA scrubbing vs. O2/CO2 recycle combustion. Energy 

Conversion and Management. 2003. p. 3073-3091. 

[15] Alstom Power Inc., ABB Lummus Global Inc., Alstom Power Environmental Systems and 

American Electric Power, 2001, Engineering feasibility and economics of CO2 capture on an 

existing coal-fired power plant. Report under Contract No. PPL-01-CT-09, USA. 

[16] Pellegrini G, Strube R, Manfrida G. Comparative study of chemical absorbents in postcombustion 

CO2 capture. Energy 2011;35:851–857. 

[17] Mohammad RM, Abu-Zahra, Schneiders LHJ. CO2 capture from power plants Part I. A 

parametric study of the technical performance based on monoethanolamine. International 

journal of greenhouse gas control. 2007. p. 37– 46. 

[18] Mohammad RM, Abu-Zahra, Niederer JPM, Feron PHM, Versteeg GF. CO2 capture from 

power plants Part II. A parametric study of the economical performance based on 

monoethanolamine. International journal of greenhouse gas control. 2007. p. 135– 142. 

[19] Hetland J, Kvamsdal HM, Haugen G, Major F, Karstad V, Tjellander G. Integrating a full 

carbon capture scheme onto a 450 MWe NGCC electric power generation hub for offshore 

operations: Presenting the Sevan GTW concept. Applied Energy 2009;86:2298–2307. 

[20] Bin H, Shisen X, Shiwang G, Lianbo L, Jiye T, Hongwei N, Ming C, Cheng Jian. Industrial test 

and techno-economic analysis of CO2 capture in Huaneng Beijing coal-fired power station. 

Applied Energy 2010;87:3347–3354. 

[21] Doukelis A, Vorrias I, Grammelis P, Kakaras E. Partial O2-fired coal power plant with postcombustion 

CO2 capture: A retrotting option for CO2 capture ready plants. Fuel 2009;88: 

2428–2436. 

[22] Yu YS, Li Y, Li Q, Jiang J, Zhang ZX. An innovative process for simultaneous removal of CO2 

and SO2 from flue gas of a power plant by energy integration. Energy Conversion and 

Management 2009;50:2885–2892. 

[23] Sanpasertparnich T, Idem R, Bolea I, deMontigny D, Tontiwachwuthikul P. Integration of postcombustion 

capture and storage into a pulverized coal-fired power plant. International Journal 

of Greenhouse Gas Control 2010;14:499–510. 

[24] Cifre PG, Brechtel K, Hoch S, Garcia H, Asprion N, Hasse H, Scheffknecht G. Integration of a 

chemical process model in a power plant modelling tool for the simulation of an amine based 

33

CO2 scrubber. Fuel 2009;88:2481–2488. 

[25] Lucquiaud M, Gibbins J. Retrotting CO2 capture ready fossil plants with post-combustion 

capture. Part 1: requirements for supercritical pulverized coal plants using solvent-based flue 

gas scrubbing. Power and Energy. 2009. p. 213-226. 

[26] National Energy Technology Laboratory of USA. Carbon Dioxide Capture from Existing Coal- 

Fired Power Plants. DOE/NETL-401/110907, Final Report; 2007. 

[27] Göttlicher G., Pruschek R. COMPARISON OF CO2 REMOVAL SYSTEMS FOR FOSSIL- 

FUELLED POWER PLANT PROCESSES. Energy Conversion and Management 1997;38: 

S173-S178. 

[28] Hanna K, Hallvard FS, Mikko A. CO2 capture from coal-fired power plants based on sodium 

carbonate slurry; a systems feasibility and sensitivity study. International Journal of 

Greenhouse Gas Control 2009;3:143-151. 

[29] Boleslaw Z, Radoslaw S. Energy analysis of technological systems of natural gas fired 

combined heat-and-power plants. Applied Energy 2003;75:43-50. 

[30] Bjorn R. Combined heat-and-power plants and district heating in a deregulated electricity 

market. Applied Energy 2004;78:37-52. 

[31] Jianye Z, Shu X. Type Selection and Parameter Determination Method for Unit of 

Cogeneration Project. Electrical Equipment 2008;9(1):72-75. 

[32] Yongxiu H, Peiji Z, Yixing Z. Cost study of heat supply and electricity supply of cogeneration 

project. Electric Power 2004;37(6):63-66. 

[33] Jianqun X, Gang L, Ling L, Keyi Z, Yongfeng S. Thermodynamic model of HP-IP leakage and 

IP tuibine efficiency. Applied Thermal Engineering 2011;31:311-318. 

[34] Shintaro I, Yasushi M, Yasuyoshi K, Satoshi N, Takumi H, Yasunobu N. Study of steam, 

helium and supercritical CO2 turbine power generations in prototype fusion power reactor. 

Nuclear Energy 2008;50:325-332. 

34

Abstract: 




Analysis of four end high temperature membrane 

air separator in a supercritical power plant with 

oxy type pulverized fuel boiler 

Janusz Kotowicz a , Sebastian Michalski b 

a Silesian University of Technology, Poland, janusz.kotowicz@polsl.pl 

b Silesian University of Technology, Poland, sebastian.michalski@polsl.pl, 

In this article computational algorithm and exemplary results for a model of an air separation unit (ASU) with 

"four end" high temperature membrane (HTM) were presented. First, the software environment for building of 

a "four end" membrane separator model was chosen. Then, a model of an air separation unit was created 

and preliminary calculations were made on that model. The air separation unit structure consists of a "four 

end" membrane, heat exchanger, electrical generator, air compressor and expander. Parameter that 

determines all flows in the ASU model is the oxygen mass flow rate. This mass flow rate is approximately the 

same as oxygen mass flow rate feeding oxy boiler working in a 460 MW power plant. The most important 

step of this paper was the integration of a model of pulverized fuel boiler in the oxy-combustion technology 

and the air separation unit model by sending flue gas from boiler to ASU. The characteristics of ASU such as 

power and efficiency as a function of the oxygen recovery rate were made. Maximal value of oxygen 

recovery rate was calculated. The difference between optimal compressor pressure ratio of the autonomic 

gas turbine and of the air separation unit are presented in this paper. 

Keywords: 

Pulverized fuel boiler, oxy-combustion technology, "four end" membrane separator 


Currently appearing world trend to reduce emissions of harmful substances such as greenhouse 

gases into the environment is changing a direction of the energy technologies [1]. Particularly 

important is the development of low emission coal technologies that play a significant role in the 

balance sheets of many countries including Poland, in which a significant part of electricity is 

generated in coal-fueled power plants. Additionally, during the production of electricity in coalfueled 

power plants carbon dioxide emission per produced electricity unit is higher than in other 

power generation technologies, for example, about 2.5 times more than in gas-steam blocks fueled 

with natural gas. The two most important directions of research aiming to reduce the emissions 

from coal-fueled power plants may be mentioned. The first one is the optimization of a power plant 

within its structure and work parameters. The second direction of development of low emission coal 

technologies is finding new and optimization of the already known low-energy carbon capture 

technologies [2]. The currently developed carbon capture technologies are as follows: 

pre-combustion technology 

post-combustion technology 

oxy-combustion technology 

Oxy-combustion technology is based on fuel combustion in an atmosphere with increased oxygen 

concentration in order to eliminate nitrogen from the flue gas. In this technology, flue gas that 

35

leaves boiler is composed mainly of carbon dioxide and steam, so the separation of carbon dioxide 

from flue gas is based on the low energy-consuming condensation process [3÷4]. The oxycombustion 

technology is now the most promising solution for carbon energetic technologies [5÷7]. 

Currently, most advanced is a cryogenic air separation technology. Membrane air separation 

technologies are also considered, in particular air separation units (ASU) with high temperature 

membranes (HTM) [8]. 

Among the currently investigated high-temperature separation membranes, a "three-end" and "four 

end" membrane-types should be distinguished [9]. The oxygen flow through the membrane is 

caused by the oxygen partial pressure difference on both sides of a membrane. "four-end" type 

membrane used for air separation is immersed on one side by compressed air, while the other side 

of a membrane is immersed by flue gas from boiler. Oxygen mass flow rate permeating through the 

membrane depends on a membrane constant ( C ), oxygen partial pressure on the feed side of the 

membrane ( pO 2 _ F ) and oxygen partial pressure on the permeate side ( pO 2 _ Per ). The relationship 

between these quantities is expressed by the following formula: 

p O 2 _ F 

j 

O 2 C ln , mol/sm 

 

pO 

2 _ Per 

2 (1) 

The membrane constant (C) in (1) depends, among other, on the thickness of the membrane and the 

membrane working temperature. 

Figure 1 shows a power plant diagram with average integration with air separation unit (ASU) that 

contains "four end" high temperature membrane for air separation. It should be noted that for the 

increase of the oxygen partial pressure on the feed side of membrane the compressor with pressure 

ratio k is used. Steam cycle of this power plant is composed of a steam turbine, four low pressure 

and three high pressure feed-water heaters, condenser, deaerator, condensate pump, feed-water 

pump, one low-pressure and one high-pressure flue gas-water heat exchangers and one low pressure 

retentate-water heat exchanger. The steam turbine consist of three parts: high-pressure, 

intermediate-pressure and low-pressure. Between the intermediate and high-pressure part of the 

steam turbine steam is reheated. Water heated in steam cycle is directed first to the two parallel 

economizers and then the water is directed to the boiler. One of the economizers is fed with the flue 

gas and second with the permeate. Flue gas with the temperature at 850 o C leaving the boiler are 

subjected to high temperature filtration. Then, part of the flue gas is supplied to the air separation 

unit. The remaining flue gas is cooled by water and then compressed. The air separation unit is 

composed of the "four-end" high temperature separation membrane, counter-flow permeate-air heat 

exchanger, economizer, permeate fan, air compressor, expander and electric generator. The flue gas 

supplied to ASU is flowing to the separation membrane, where is enriched in oxygen. The gas 

leaving the membrane (called permeate) heats compressed air in a counter-current permeate-air heat 

exchanger. Then, the permeate is cooled to a temperature of 320 o C in the economizer and is 

supplied by the fan as an oxidant to the boiler`s combustion chamber. The air drawn to the ASU is 

compressed and then heated to a temperature of 750 o C in permeate-air heat exchanger. Then, the 

air flows to the separation membrane (feed) where the oxygen is separated. The gas leaving the 

membrane, that consists mostly of nitrogen, is called retentate. The retentate enters the expander 

and then is cooled by a feed water in the steam cycle. Expander drives the air compressor, and the 

excess of the mechanical power is used to generate electricity. 

36

Fig. 1. Solution diagram of „four end”, average integration [9] 

The paper includes the analysis and comparison with a classic gas turbine of an air separation unit 

(ASU). The computations results obtained with ASU model were compared with the results 

obtained using an autonomous gas turbine model. The air separation model will be used to build the 

models of oxy-combustion power plants. 

2. Model of the air separation unit (ASU) and assumptions for 

calculations 

Air separation unit structure consists of: counter-flow air heater (APH), air compressor (C), 

expander (EX), electric generator (G) and "four-end" type membrane (M). The expander drives the 

air compressor. Depending on the assumed quantities the expander and compressor can give or take 

electricity from the grid. The structure of the air separation unit is shown in Figure 2. The 

characteristic basic quantities of the air separation unit and autonomous gas turbine are gathered in 

Table 1. 

Fig. 2. Scheme of the air separation unit (ASU) 

37

Table 1. Characteristic quantities for investigated air separation unit (ASU) 

Name Symbol Value Unit 

Ambient pressure 

Ambient temperature 

Membrane working temperature 

pot 

tot 

tmem 

101.3 kPa 

o 

20 C 

o 

750;850 C 

Stream of separated oxygen m O2 

107.56 kg/s 

Oxygen recovery rate R 40÷100 % 

Compressor pressure ratio k 2÷30 - 

Compressor isentropic efficiency iS 0.88 - 

Expander isentropic efficiency iT 0.9 - 

Generator efficiency g 0.99 - 

It was assumed for the calculations that the air taken from environment is a dry gas consisting of 

21% oxygen and 79% nitrogen (volumetric composition). 

The characteristic quantities gathered in Table 1 were used for computations performed on a "fourend" 

membrane air separation unit model, made in GateCycle TM software. The built-in components 

were used to build the air separation unit model. The quantity that determinates the value of the 

mass flow rate in the entire ASU model is a mass flow rate of oxygen. This mass flow rate is 

approximately the same as the oxygen mass flow rate feeding an oxy boiler working in a 460 MW 

power plant. It was assumed that through the membrane flows pure oxygen. 

The structure of an autonomous gas turbine as opposed to a structure of ASU shown in Figure 2 

does not contain a "four-end" type membrane. In the autonomous gas turbine model the assumption 

concerning a compressor, expander and air heater are the same as in the ASU model. The air mass 

flow rate in both models are the same, the difference is only in the mass flow rate and composition 

of gas flowing into the expander. In the ASU model the composition and mass flow rate of the gas 

is different from air because some oxygen is separated from air in the membrane. In the 

autonomous gas turbine model the mass flow rate and composition of gas flowing into the expander 

is the same as mass flow rate and composition of air leaving the air heater. 

3. The results of calculations of air separation unit and 

autonomous gas turbine 

The air mass flow rate depends on the separated in membrane oxygen mass flow rate ( m O2 ), oxygen 

recovery rate (R) and mass content of oxygen in the air ( g O2air 

). The relationship between these 

quantities is as follows: 

m 

O2 1a , (2) 

R 

gO2air 

m 

Next the air is compressed by the compressor. Effective power required to drive the compressor 

depends on the air mass flow rate ( 1a m ), the air temperature ( 1a T ), the average specific heat ( ~c p ), 

K 

1 

the compressor pressure ratio ( K ), the heat capacity ratio contained in the factor ( K 

), 

K 

the compressor isentropic efficiency ( iK ) and the compressor mechanical efficiency ( mK ). The 

equation showing the relationship between these quantities is as follows: 

38

K 

 

~ 

 

K 1 

eK 1a p K 1a 

iK 

mK 

 

 

N m 

c T 

, (3) 

The mass flow rate flowing through the expander is lower than the oxygen mass flow rate separated 

in the membrane by the mass flow rate flowing through the compressor. This mass flow rate 

depends on the oxygen mass flow rate separated from the air in the membrane ( m O2 ) and the air 

flow rate ( m 1a ). The relationship between these quantities is as follows: 

m m 

m 

, (4) 

4a 

1a 

O2 

The expander effective power depends on the retentate mass flow rate ( a m 4 ), the retentate 

temperature ( T 4a ), the average specific heat ( ~c p ), the compressor pressure ratio ( 

K 

K ),the 

reduction factor of compressor pressure ratio ( ), the heat capacity ratio contained in the factor 

1 

( T ), the expander isentropic efficiency ( iT ) and the expander mechanical efficiency 

T 

( mT ). The equation showing the relationship between these quantities is as follows: 

~ T 

N m 

c T 1 ( ) 

 

,(5) 

eT 

4a 

p 4a K iT mT 

T 

Figure 3 shows computed gross electric power as a function of oxygen recovery rate. The curve in 

this figure is determined for K =20 and t 3a =750 o C. The gross electrical power depends on the 

expander gross power ( N eT ), the compressor gross power ( N eK ) and the generator efficiency ( g ). 

The relationship between these quantities is as follows: 

elTG 

N eT eK g 

N N , (6) 

Electrical power of gas turbine, MW 

80 

60 

40 

20 

0 

-20 

40 45 50 55 60 65 70 75 80 85 90 95 100 

Oxygen recovery rate, % 

Fig. 3. Electrical power achieved or required for power the turbine set in ASU model as a function 

of oxygen recovery rate for K =20 and t 3a =750 o C 

39

Figure 4 shows a graph of the electric power generation efficiency for K =20 and t 3a =750 o C. This 

efficiency depends on the gross electrical power ( N elTG ) and the heat supplied to the unit ( Q d ). The 

relationship between these quantities is as follows: 

N 

elTG el , (7) 

Qd 

The heat supplied to the unit depends on the air mass flow rate ( 2a m ), the air enthalpy ( h 2a ), the 

retentate mass flow rate ( 4a m ) and retentate enthalpy ( h 4a ). The relationship between these 


Q m 

h m 

h , (8) 

d 

4a 

4a 

Gross efficiency of gas turbine, % 

24 

20 

16 

12 

8 

4 

0 

-4 

-8 

-12 

2a 

2a 

-16 

40 45 50 55 60 65 70 75 80 85 90 95 100 


Fig. 4. Gross efficiency of ASU model as a function of oxygen recovery rate for K =20 and 

t 3a =750 o C 

It should be noted that the gross electrical power (Fig. 3) at a large oxygen recovery rate is below 

zero. Equation for the maximal oxygen recovery rate using (2) and (5) is as follows: 

 

 

 

~ 

K 

1 T 1 c 

 

1a 

p K K 

1 

R gr 1 

 

 

T 

 

~ 

, (9) 

 

gO2 

T3a 

iK iT 

mK 

mT 

cp 

1( 

K) 

T 

 

Figure 5 shows graph of maximal oxygen recovery rate as a function of a compressor pressure ratio. 

Computations made for this graph were carried out with the use of GateCycle TM software for the 

two different membrane working temperature ( t 3a =750 o C and t 3a =850 o C). 

40


105 

95 

85 

75 

65 

55 

45 

35 

10 12 14 16 18 20 22 24 26 28 30 

Air compressor pressure ratio, - 

Fig. 5. Maximal oxygen recovery rate in the ASU model as a function of the compressor pressure 

rate for two different membrane work temperature 

Figure 6 shows a graph of gross electric power of the air separation unit (marked as "ASU" on a 

graph) as a function of the compressor pressure ratio for the two different oxygen recovery rate 

(50% and 90%). The same figure also shows the gross electric power of the autonomous gas turbine 

(marked as "TG" on graph) as a function of the compressor pressure ratio for the same oxygen 

recovery rate. 

Electrical power of gas turbine, MW 

200 

180 

160 

140 

120 

100 

80 

60 

40 

20 

0 

-20 

TG R=90% 

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 

Air compressor preassure ratio, - 

Fig 6. Gross electrical power for both units as a function of compressor pressure ratio for two 

different oxygen recovery rate 

It should be noted that the optimal compressor pressure ratio in the air separation unit and the 

autonomous gas turbine are not equal. Using (3) and (5) the optimal compressor pressure ratio in air 

separation unit can easily be determined: 

41

1 

opt(N 

eTG 

) k O2 K T 

opt(N 

eTG 

) 

k 1 R 

kl 

g , (10) 

42 

opt(N 

eTG 

In the (8) optimal compressor pressure ratio ( k ) is determined in the same way as for a 

kl 

classic gas turbine. Calculated optimal compressor pressure ratio values, due to generated electricity 

from the turbine set are gathered in Table 2. 

Table 2. The optimal compressor pressure ratio values, due to generated electricity from turbine set 

Name Symbol Value 

The oxygen recovery rate R, % 50 70 90 

Optimal compressor pressure ratio k_ASUopt, - 

for ASU 

Optimal compressor pressure ratio 

for TG 

k_T Gopt, - 

) 

6.4 5.8 5.3 

7.8 7.8 7.8 

The optimal compressor pressure ratio values, due to the gross efficiency as well as for the 

autonomous gas turbine depends on the optimal compressor pressure ratio values, due to electricity 

opt(N 

) 

generated from the turbine set ( k 

eTG ) and the gross efficiency of the electricity generation of 

the unit ( elTG ). The relationship between these quantities is as follows: 

 

opt( 

eTG 

) 

k 

opt(N ) 1 

K 

 

T 

eTG k 

 

 

1 elTG 

 

, (11) 

mK 

1 

delTG 

Equation (11) was determined using (3), (5) and (7), with the condition 0 

d 

. 

Figure 7 shows a graph of gross efficiency of electricity generation as a function of the compressor 

pressure ratio for the two different oxygen recovery rates in the air separation unit (50% and 90%) 

and for autonomous gas turbine. 

Gross efficiency of gas turbine, % 

40 

36 

32 

28 

24 

20 

16 

12 

8 

4 

0 

-4 

-8 

-12 

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 

Air compressor preassure ratio, - 

Fig. 7. Gross efficiency of both units as a function of compressor pressure ratio for two different 

oxygen recovery rate 

k

Calculated optimal compressor pressure ratio values, due to the gross efficiency are gathered in 

Table 3. 

Table 3. The optimal compressor pressure ratio values, due to gross efficiency of both units 

Name Symbol Value 

Oxygen recovery rate R, % 50 70 90 

Optimal compressor pressure ratio k_ASUopt, - 

for ASU 

11.9 10.2 8.8 

Optimal compressor pressure ratio k_T Gopt, - 

for TG 

17.6 17.6 17.6 

4. Summary 

In this paper the air separation unit with "four-end" high-temperature membrane (HTM) was 

analyzed. 

For the analysis of the air separation unit and of the autonomous gas turbine power and efficiency 

characteristic as a function of oxygen recovery rate and compressor pressure ratio were determined. 

The characteristics for both units are summarized and compared in the figures 6 and 7. 

The maximal oxygen recovery rate as a function of the compressor pressure ratio for the two 

different membrane work temperature were determined. This quantity separates the area of work in 

which we get the extra power from expander in the air separation unit from the area of work in 

which we must deliver additional power to drive the compressor. 

The optimal compressors pressure ratio due to a power of turbine set and due to the efficiency of 

electricity generation for different oxygen recovery rates in the air separation unit and in the 

autonomous gas turbine were determined. The optimal values of compressor pressure ratio in the air 

separation unit are decreasing with the increase of the oxygen recovery rate. In the case of the 

autonomous gas turbine, these values are constant. 

Acknowledgements 

The results presented in this paper were obtained from research work co-financed by the National 

Centre for Research and Development within a framework of Contract SP/E/2/66420/10 – Strategic 

Research Programme – Advanced Technologies for Energy Generation: Development of a 

technology for oxy-combustion pulverized-fuel and fluid boilers integrated with CO2 capture. 

LITERATURE 

[1] Chmielniak T., The role of various technologies in achieving emissions objectives in the 

perspective of the years up to 2050. Rynek Energii, 2011;92:3-9 

[2] Chmielniak T., ukowicz H., Kochaniewicz A., Trends of modern power units efficiency 

growth. Rynek Energii, Nr 6(79), 2008, 14-20. 

[3] Kotowicz J., Janusz-Szymaska K., Influence of CO2 separation on the efficiency of the 

supercritical coal fired power plant. Rynek Energii, 2011, 2 (93), 8-12. 

[4] Liszka M., Zibik A.: Coal – fired oxy – fuel power unit – Process and system analysis. Energy, 

35 (2010), 943 – 951. 

[5] Toftegaard M.B., Brix J., Jensen P.A., Glarborg P., Jensen A.D., Oxy-fuel combustion of solid 

fuels. Progress in Energy and Combustion Science, 2010;36:581-625 

[6] Dillon D.J., White.V., Allam R.J., Wall R.A., Gibbins J., Oxy-combustion Process for CO2 

Capture from Power Plant. Mitsui Babcock Energy Limited,2005 

43

[7] Buhre B.J.P., Elliott L.K., Sheng C.D., Gupta R.P. and Wall T.F., Oxy-fuel combustion 

technology for coal-fired power generation. Progress in Energy and Combustion Science, 

2005;31:283-307 

[8] Pfaff I., Kather A., Comparative thermodynamic analysis and integration issues of CCS steam 

power plants based on oxy-combustion with cryogenic or membrane based air separation. 

Energy Procedia 1 (2009) 495-502. 

[9] Stadler H. et al.: Oxyfuel coal combustion by efficient integration of oxygen transport 

membranes. International Journal of Greenhouse Gas Control 5 (2011) 7-15. 

44




Analysis of potential improvements 

to the lignite-fired oxy-fuel power unit 

Marcin Liszka a , Jakub Tuka b , Grzegorz T. Nowak c , Grzegorz Szapajko d 

a Institute of Thermal Technology, Silesian University of Technology, 44-100 Gliwice, 

Konarskiego 22, Poland, marcin.liszka@polsl.pl 

b Institute of Thermal Technology, Silesian University of Technology, 44-100 Gliwice, 

Konarskiego 22, Poland, jakub.tuka@polsl.pl, CA 

c Institute of Thermal Technology, Silesian University of Technology, 44-100 Gliwice, 

Konarskiego 22, Poland, grzegorz.t.nowak@polsl.pl 

d Institute of Thermal Technology, Silesian University of Technology, 44-100 Gliwice, 

Konarskiego 22, Poland, grzegorz.szapajko@polsl.pl 

Abstract: 

One of the most promising technologies for coal-to-electricity conversion considering CO2 removal is 

the oxy-fuel process. As the energy efficiency of oxy-fuel system is ca. 10 percentage points lower 

than traditional pulverized-coal power unit, it is desirable to look for ways to partial recovery of 

efficiency loss due to CO2 capture. The main goal of the presented paper is the energy analysis of 

several structural improvements of the oxy-fuel power unit based on waste energy recovery and waste 

product (nitrogen) utilization. Five case studies have been proposed and analysed. First, the reference 

lignite-fired, oxy-fuel power unit have been proposed and simulated. It is composed of boiler, flue gas 

recirculation loop, steam cycle, air separation unit and CO2 purification and compression system. The 

subsystems are however only slightly integrated representing the state of the art for oxy-fuel systems 

proposed in recently published pre-feasibility studies. Remaining cases are focused on ideas for 

efficiency improvements like heat recovery from the flue gas to combusted oxygen, heat recovery from 

the flue gas compression train to steam cycle, as well as, lignite drying by waste nitrogen leaving the 

air separation unit. First two of these ideas have been already investigated in the literature, while the 

use of waste nitrogen for coal drying seems to be innovative and promising. As the nitrogen leaving air 

separation unit is completely dry, its potential for lignite drying is higher than for ambient air. Moreover, 

the risk of ignition and explosion in the dryer is minimised. The last, fifth case is the summarize of all 

structural improvements. The Thermoflex software has been used as simulation tool for all analysed 

cases. The proposed improvements based on waste heat recovery within the oxy-fuel power unit may 

bring substantial rise of the net electric efficiency. The reference, not thermally integrated plant, 

achieves the efficiency of 29.55%, while in case of highly integrated plant (almost all waste heat is 

utilized) the efficiency increases to 32,98%. 

Keywords: 

Oxy-fuel, CCS, integration, lignite, drying 


One of the most promising technologies for coal-to-electricity conversion considering CO2 

removal is the oxy-fuel process. As the energy efficiency of oxy-fuel system is ca. 10 

percentage points lower than traditional pulverized-coal power unit [1, 2], it is desirable to 

look for ways to partial recovery of efficiency loss due to CO2 capture. The CCS-related drop 

of efficiency is mainly caused by the necessity of ASU and CO2 purification and compression 

system installation. On the other hand, the utilisation of waste energy generated within these 

subsystems is possible. 

The main goal of the presented paper is thus the energy analysis of several structural 

improvements of the oxy-fuel power unit based on waste energy recovery and waste product 

45

(nitrogen) utilization. Five case studies have been proposed and analyzed. The first one is a 

reference, lignite-fired oxy-fuel power unit. It is composed of boiler, flue gas recirculation 

loop, steam cycle, air separation unit and CO2 purification and compression system. The 

subsystems within the reference case have been only slightly integrated. It represents 

therefore, the state of the art for oxy-fuel systems proposed in recently published 

prefeasibility studies [3, 4]. Remaining cases are focused on ideas for efficiency 

improvements like heat recovery from the flue gas to combusted oxygen, heat recovery from 

the flue gas compression train to steam cycle, as well as, lignite drying by waste nitrogen 

leaving the air separation unit. First two of these ideas have been already investigated in the 

literature, while the use of waste nitrogen for coal drying seems to be innovative and 

promising. The results dealing with heat recovery from the flue gas to combusted oxygen 

have been presented in [2]. Authors concluded that it is better to use the waste heat to preheat 

boiler feed water than oxidizer. 

Considering the ASU-waste nitrogen as drying agent for lignite, it is important to mention, 

that this gas is completely dry and have therefore higher potential for moisture absorption 

than ambient air. Moreover, the risk of ignition and explosion in the dryer is minimized. 

The last, fifth case summarize all structural improvements applied in previous cases. 

2. Case studies 

General assumptions for all analysed cases have been summarized in Table 1. As it has been 

already mentioned case no. 1 is the reference. 

Table 1. Analyzed case studies 

Heat recovery from the 

flue gas to combustion 

oxygen 

Heat recovery from the flue 

gas compression train to 

steam cycle 

46 

Lignite drying by waste 

nitrogen 

Case 1 NO NO NO 

Case 2 YES NO NO 

Case 3 NO YES NO 

Case 4 NO NO YES 

Case 5 YES YES YES 

2.1 Reference case (Case 1) 

The flow sheet of the reference case has been presented in Fig. A1a and A1b in the appendix 

A. The reference case structure composes of boiler, flue gas treatment and conditioning line, 

CO2 compression system, ASU and supercritical steam cycle. The recirculation loop is of hot 

and wet type which means that part of the flue gas which goes back to the furnace to control 

the flame temperature has not been cooled down and the moisture has not been condensed. In 

accordance to [2], oxy-fuel power units equipped in hot or cold recirculation loop do not 

differ in energy efficiency. 

Main parameters of the reference case plant have been presented in Table 2. Volumetric 

content of oxygen in oxidizer entering the combustion chamber has been assumed to 23,5% 

which enables for keeping similar adiabatic flame temperature and heat exchange rates in the 

boiler furnace as for conventional air-combustion system [5].

The net amount of the flue gas leaving the boiler island is of high CO2 concentration (ca 82,5 

%vol), however further inert gas separation (mainly O2 and N2) is necessary. Inert gas 

separation line is of cryogenic type. The inert gas separation line has not been physically 

modelled within the current paper as it is not related to the analysed waste heat recovery 

systems. After [3], it has been just assumed that there is some required flue gas pressure 

before the cryogenic separation unit. The flue gas pressure drop within the separation 

installation, as well as, separation effectiveness have also been assumed as constant values. 

The auxiliary power consumption related to inert gas separation is determined as power of 

flue gas compressors. Finally, the CO2 content in flue gas leaving the oxy-fuel system is 

nearly 96% and is the same as reported in [3]. 

The steam cycle is of single reheat supercritical design. Its structure and parameters represent 

the best available technology for today plants – the parameters of live / reheated steam are: 

600OC, 28,5MPa / 620OC, 5MPa. 

2.2 Heat recovery from the flue gas to combustion oxygen (Case 2) 

The idea of case 2 has been shown in Fig. 1, where the oxygen heater has been added between 

ASU and boiler. The oxygen taken from ASU exhaust is preheated up to 260 0 C by the flue 

gas flowing within the main recirculation loop. The detailed flow sheet of Case 2 has been 

presented in Fig. A2 in appendix A. The steam cycle is the same as for the reference case, so 

the Fig. A1b in appendix A refers also to case 2. 

Figure 1. Idea for heat recovery from the flue gas to combustion oxygen (case 2) 

2.3 Heat recovery from the flue gas compression train to steam 

cycle (Case 3) 

The idea for heat recovery from the flue gas compression train to steam cycle which is the 

essence of case 3 is presented in Fig. 2. Recovered heat replaces the LP heat regeneration 

within the steam cycle in ca 70%. The detailed flow sheet of Case 3 has been presented in Fig. 

A3a and A3b in appendix A. 

47

Figure 2. Idea for heat recovery from the flue gas compression train to steam cycle (case 3) 

Comparing to the reference case, the additional flue gas coolers located behind appropriate 

compression stages have been added. The cooling media within these new coolers is steam 

cycle condensate. The flue gas is finally cooled to the required, possible lowest temperature 

before following compression stage by cooling water as for the reference case. 

2.4 Lignite drying by ASU-waste nitrogen (Case 4) 

The idea for lignite drying by ASU-waste nitrogen (case 4) is presented in Fig. 3. The plant 

structure includes fluidized-bed lignite dryer fed by dry nitrogen taken from ASU exhaust. 

Additionally, the heat exchanger is located inside the bed to enhance the heat transfer to the 

fuel being dried. Such a technology has been tuned for utilization of low-temperature waste 

heat and follows the commercially proved idea presented in [6]. The coal dryer applied within 

power system presented in [6] uses however air as drying agent and assumes integration with 

conventional air-firing boiler. 

It has been assumed, that the dry nitrogen and water for the in-bed heat exchanger are 

preheated by the compressed air from the ASU air compression line. The detailed flow sheet 

of Case 4 has been presented in Fig. A4 in appendix A. The steam cycle is the same as for the 

reference case, so the Fig. A1b in appendix A refers also to Case 4. Detailed information 

about streams entering fuel drier has been shown in Table 2. 

Table 2. Detailed information about streams entering fuel dryer 

Stream (No at Fig.3) Temperature, 0 C Mass flow, kg/s 

Nitrogen entering preheater (158) 30 566,9 

Nitrogen entering fuel dryer (161) 70 566,9 

Nitrogen leaving fuel dryer (155) 35 575,3 

Lignite entering fuel dryer (159) 25 219,2 

Lignite leaving fuel dryer (2) 30 204,5 

48

Figure 3. Idea for lignite drying by ASU-waste nitrogen (case 4) 

3. Simulation model 

All the analysed cases have been modelled using the Thermoflex software [7]. The schemes 

used for modelling purposes follow structural configurations presented in Figs A1-A4 in 

appendix A. 

Selected assumed parameters of analysed power units have been reported in Table 3. Fuel 

specification has been shown in Table 4. Parameters presented in Tables 2 and 3 are kept 

constant for all analyzed cases. 

Table 3. Assumed design parameters for simulation modelling 

Cooling water temperature rise at each cooler (ASU air, 

CO2-rich flue gas) and in ST condenser 

K 10 

Minimal temperature difference (pinch) at each cooler K 20 

Live (HP) steam pressure MPa 28,5 

Live (HP) steam flow rate kg/s 620 

Live (HP) steam temperature 

0 

C 600 

Reheated (MP) steam temperature 

0 

C 620 

ST condenser pressure 

kPa 

5 

ST polytrophic efficiency 

% 

89 

ASU compressors polytrophic efficiency 

% 

86 

CO2-rich flue gas compressors polytrophic efficiency % 

92 

Content of oxygen in oxidizer entering the boiler 

combustion chamber 

% (vol) 23,5 

ASU oxygen purity 

% (vol) 95 

Final CO2 pressure for transport pipeline 

Ambient parameters: 

MPa 

13 

Temperature 

0 

C 15 

Pressure MPa 0,1013 

relative humidity % 60 

49

Table 4. Assumed lignite parameters (as received, mass shares) 

Total moisture % 45,00 

Ash % 20,00 

Carbon % 23,80 

Hydrogen % 2,17 

Nitrogen % 0,25 

Sulphur % 0,70 

Oxygen % 8,08 

LHV kJ/kg 8340 

4. Results 

The energy efficiency calculated for a whole oxy-fuel power unit has been selected as a main 

assessment factor of the proposed improvements. Calculated efficiencies have been presented 

in Fig. 4. Moreover, the more detailed results of simulation have been collected in Table 6. 

Temperature profiles for waste heat utilizing units have been shown in Figs 5 and 6. 

Table 5. Lignite parameters after drying (mass shares) 

Total moisture % 41,05 

Ash % 21,44 

Carbon % 25,51 

Hydrogen % 2,321 

Nitrogen % 0,268 

Sulphur % 0,750 

Oxygen % 8,661 

LHV kJ/kg 9116 

Table 6. Results of simulation 

Case 1 

Case 5 

Parameter Unit (reference) Case 2 Case 3 Case 4 (cumulative) 

Live steam flow rate kg/s 620 620 620 620 620 

Chemical energy flow rate (LHV 1875,9 1856,2 1875,9 1816,5 1798,6 

based) MW 

Generator terminal power MW 842,5 842,5 879,7 842,5 872,7 

Net power output MW 554,3 558,8 590,2 560,7 593,2 

Auxiliary power consumption 

including: 

MW 288,2 283,8 289,5 281,8 279,5 

- ASU MW 130,7 129,3 130,7 127,3 125,7 

- CO2 compression and 

86,1 87,6 88,6 86,3 85,2 

purification line MW 

- steam cycle MW 6,0 6,0 6,9 6,0 6,8 

- others MW 65,6 60,8 63,3 62,2 61,8 

Total waste heat recovered MW 0,0 40,2 187,5 30,4 240,3 

Net plant efficiency % 29,55 30,10 31,46 30,87 32,98 

50

Energy efficiency, % 

34,00 

33,00 

32,00 

31,00 

30,00 

29,00 

28,00 

27,00 

29,55 

30,10 

51 

31,46 

30,87 

32,98 

Case 1 Case 2 Case 3 Case 4 Case 5 

Fig. 4. Energy efficiency of oxy-fuel power unit 

Case number 

Fig. 5. Temperature profiles in oxygen preheater (case 2)

Fig. 6. Temperature profiles in compressors intercoolers preheating steam cycle condensate 

(case 3) 

Rise of energy efficiency, 

% points 

4 

3,5 

3 

2,5 

2 

1,5 

1 

0,5 

0 

0,55 

1,91 

Case 1 Case 2 Case 3 Case 4 Case 5 Theoretical 

Case number 

sum of Cases 

2-4 

Fig. 7. Increase of energy efficiency related to the reference case 

The influence of proposed improvements on the energy efficiency of analysed oxy-fuel power 

unit is clearly visible. The highest increase, nearly 2 percentage points, brings heat recovery 

from the flue gas compression train to steam cycle (case 3). The substantial increase of 

efficiency (1.3 % point) is caused also by lignite drying (case 4). The parameters of lignite 

after dryer have been presented in Table 5. Heat recovery from the flue gas to combustion 

oxygen (case 2) provides relatively low (0.5 % point) increase of plant efficiency. Case no 5 

which includes all proposed modifications is, as expected the best one, achieving the 3.4% 

52 

1,32 

3,43 

3,78

points efficiency increase related to the reference plant. This increase is however lower than 

sum of efficiency changes obtained for cases 2, 3 and 4. 


The proposed improvements based on waste heat recovery within the oxy-fuel power unit 

may bring substantial rise of the net electric efficiency. The reference, not thermally 

integrated plant, achieves the efficiency of 29.55%, while in case of highly integrated plant 

(almost all waste heat is utilized) the efficiency increases to 32,98%. 

As results from the temperature profiles in heat exchangers where waste heat is recovered, 

there is still some potential for increase of waste energy use by e.g. replacing also part of the 

high-pressure steam regeneration. 

Comparing to literature studies, the conclusion drawn in [2] on better results obtained by 

preheating the steam cycle condensate comparing to preheating of oxidizer has been 

confirmed within the current research. 

The proposed lignite drying by ASU-waste nitrogen is promising technology from the energy 

efficiency and exploitation safety (explosion risk) points of view. Considering practical 

aspects presented in [6] for similar air-fed dryers, the verification of dryer dimensions and 

capital cost will be crucial for final assessment of this technology. 

It is important to mention, that the efficiency increase in case no. 5 (relative to the reference 

plant – case no 1) which simultaneously cumulates all improvements introduced in cases 2, 3, 

and 4 is less than the sum of efficiency increases obtained in cases 2, 3 and 4. The reason is 

that each single improvement causes decrease of waste heat production. Moreover, single 

cases are partially using the same waste heat of the flue gas which is not possible within the 

cumulated case no 5. 

Acknowledgments 

This work has been prepared in framework of the task of research: "Development of oxycombustion 

technologies for pulverized-coal and fluidized bed boilers integrated with carbon 

dioxide capture” funded by the Polish National Centre for Research and Development within 

the strategic program of research and development: "Advanced energy generation 

technologies". 

Nomenclature 

ASU air separation unit 

CCS carbon capture and storage 

E 

chfuel chemical energy of lignite, calculated on LHV basis, MW 

HP high pressure 

LHV lower heating value, kJ/kg 

MP medium pressure 

LP low pressure 

N elN net electric power of the system, MW 

ST steam turbine 

53

Appendix A 

Fi 

gure A1a. Thermoflex flow sheet of the reference plant (case no. 1) including boiler, flue gas treatment and conditioning line, CO2 compression 

system and ASU 

54

Figure A1b. Thermoflex flow sheet of reference steam cycle (the same for cases no. 1, 2 and 4) 

55

Figure A2. Thermoflex flow sheet of case no. 2 including boiler, flue gas treatment and conditioning line, CO2 compression system, ASU and 

oxygen preheater (red box) 

56

Figure A3a. Thermoflex flow sheet of case no. 3 including boiler, flue gas treatment and conditioning line, CO2 compression system, ASU and 

heat recovery exchangers (red boxes) 

57

Figure A3b. Thermoflex flow sheet of steam cycle in case no. 3 

58

Figure A4. Thermoflex flow sheet of case no. 4 including boiler, flue gas treatment and conditioning line, CO2 compression system, ASU and 

fuel dryer with heat exchangers (red boxex) 

59

References 

[1] CO2 Capture Ready, IEA Report Number 2007/4, May 2007 

[2] Eriksson T., Sippu O., Hotta A., Oxyfuel CFB Boiler as a Route to Near Zero CO2 Emission 

Coal. Foster Wheeler 

[3] Oxy Combustion Processes for CO2 Capture from Power Plant, IEA Report Number 2005/9, 

July 2005 

[4] Pulverized Coal Oxycombustion Power Plants, DOE/NETL-2007/1291, Final Report August 

2008 

[5] Hack H., Fan Z., Seltzer A., Eriksson T., Sippu O., Hotta A., Development of Integrated Flexi- 

Burn Dual Oxidant CFB Power Plant. The 33rd International Technical Conference on Coal 

Utilization & Fuel Systems, June 1 - 5, 2008 – Clearwater, Florida, USA 

[6] Bullinger Ch.W., Sarunac N., Lignite Fuel Enhancement. Final Technical Report Great River 

Energy, 2010 

[7] Thermoflow, Inc., 29 Hudson Road , Sudbury, MA 01776 USA, http://www.thermoflow.com 

60

Abstract: 




Biogas Upgrading: Global Warming Potential of 

Conventional and Innovative Technologies 

Katherine Starr a , Xavier Gabarrell Durany b , Gara Villalba Mendez b , Laura Talens 

Peiro c and Lidia Lombardi d 

a Sostenipra, Department of Chemical Engineering, Xarxa de Referència en Biotechnologia (XRB) de 

Catalunya, Universitat Autònoma de Barcelona, Bellaterra, Spain, Katherine.starr@uab.cat, CA 

b Sostenipra, Institute de Ciència i Technologua Ambientals (ICTA), Department of Chemical 

Engineering, Xarxa de Referència en Biotechnologia (XRB) de Catalunya, Universitat Autònoma de 

Barcelona, Bellaterra, Spain, Xavier.gabarrell@uab.cat, gara.villalba@uab.cat, 

c INSEAD - Campus Europe, Fontainebleau, France, laura.talenspeiro@insead.edu 

d Dipartimento di Energetica, Universita delgi Studi di Firenze, Firenze, Italy lidia.lombardi@unifi.it 

Biogas upgrading technologies provides an alternative source of methane and their implementation in waste 

management systems can help reduce the greenhouse effect. This paper uses a life cycle assessment 

(LCA) to study eight technologies, six of which are already on the market and the two others are novel 

technologies that use carbon mineralization in their process in order to not only remove CO2 but also store it. 

The two technologies are under development in the frame of the UPGAS-LOWCO2 LIFE08/ENV/IT/000429 

project (upgas.eu) and include alkaline with regeneration (AwR) and bottom ash upgrading (BABIU). These 

technologies utilize waste from municipal solid waste incinerators rich in calcium to store CO2 from biogas. 

Among all conventional technologies, high pressure water scrubbing and chemical scrubbing with amine had 

the lowest CO2 impacts. The results of the two novel technologies show that BABIU saves 10% more CO2 

than AwR. An uncertainty analysis and a material flow analysis showed that the placement of these two 

novel technologies is an important factor (for CO2 emissions and availability of waste) and therefore they 

should be located close to a MSWI that produces sufficient waste. 

Keywords: 

Biogas, Carbon Capture, Carbon Mineralization, Life Cycle Assessment, Sustainability. 


Among the renewables, the biogas industry in the EU is growing, reaching about 8.3 Mtoe in 2009 

with more than 6000 biogas plants. The main source is agriculture (52%), then landfills (36%) and 

sewage plants (12%) [1]. 

Biogas can be fed with a variety of bio-materials which can be waste or energy crops. Biogas 

produced in anaerobic digestion plants (AD-plants) or landfill sites is primarily composed of 

methane (CH4) and carbon dioxide (CO2) with smaller amounts of hydrogen sulphide (H2S) and 

ammonia (NH3). Trace amounts of hydrogen (H2), saturated or halogenated carbohydrates and 

oxygen (O2) are occasionally present in the biogas. Usually the gas is saturated with water vapour 

and may contain dust particles and organic silicon compounds (e.g. siloxanes). 

Biogas from anaerobic digestion plants (AD-plants) or landfill sites can be directly used for the 

production of heat and steam, electricity, vehicle fuels and chemicals. Alternatively, it can be 

further upgraded to increase the methane concentration, by removing CO2 and other impurities, in 

order to be suitable as a substitute for natural gas in the already established distribution grid. This 

gas can now be regarded as biomethane and is of a quality where it can fed into the natural gas 

distribution grid or be used as a vehicle fuel. This option is gaining more interest throughout Europe 

61

and there are currently several different commercial technologies for reducing the concentration of 

CO2 in biogas. 

There are four different types of upgrading technologies which removes CO2 and they include 

absorption, adsorption, membrane separation and cryogenic separation. For the absorption 

processes a reagent is used to absorb CO2. Within absorption one can find high pressure water 

scrubbing (HPWS) which uses water, chemical scrubbing (AS) which uses an amine based solvent 

such as diethanolamine (DEA), and organic physical scrubbing (OPS) which uses a commercial 

blend of polyethylene glycol. Under adsorption CO2 is normally adsorbed onto a medium such as 

activated carbon and then removed through changes in pressure, as in the case of pressure swing 

adsorption (PSA). For membrane separation (MS) a selective membrane is used to separate CO2 

from the biogas. Cryogenic separation (Cry) separates CH4 and CO2 through a decrease in 

temperature which causes a change in the physical state of the gases [2]. The marketed technologies 

use varying techniques to process the gas but what they do have in common is that they do not 

permanently store the CO2, instead it is sent back into the atmosphere or used for industrial 

purposes if it meets quality requirements [3]. 

Currently, under the framework of the UPGAS-LOWCO2 LIFE08/ENV/IT/000429 project, there 

are two novel upgrading technologies under development additionally storing the separated CO2 

through carbon mineralization. These technologies use wastes from municipal solid waste 

incinerators (MSWI) rich in calcium compounds to fix CO2 and thus form calcium carbonate 

(CaCO3). The two technologies that are being developed, and are currently in the pilot plant stage, 

are alkaline with regeneration (AwR) – developed jointly 

"Tor Vergata" in Italy [4,5] - and the bottom ash for biogas upgrading 

(BABIU) – developed by the University of Natural Resources and Life Sciences in Austria [6,7]. 

The AwR process, which is a continuous process, absorbs the CO2 using an alkaline solution of 

potassium hydroxide (KOH). This solution is regenerated at a rate of 70% when put into contact 

with air pollution control residues (APC) which is rich in calcium. Once the CO2 is adsorbed into 

the APC the biogas (from here referred to as biomethane) is free of impurities. BABIU, which is a 

batch process, uses a direct solid-gas phase interaction. Biogas is pumped through a column 

containing bottom ash (BA) rich in calcium, CO2 is absorbed in the BA and thus the resulting 

biomethane has a high concentration of CH4. 

In this study the amount of greenhouse gases created and saved by implementing these technologies 

is analyzed through a life cycle assessment (LCA). Eight technologies that were described above 

are examined and they include AwR, BABIU, PSA, HPWS, OPS, Cry, MS, and AS. LCA is a 

useful tool to determine the environmental impact of technologies. While it is often applied to 

technologies that are on the market, it is often used during the development phase in order to help 

create a more environmentally sound process [8]. While LCAs have various indicators that can be 

selected, the Global Warming Potential was chosen as the focus of the study as one of the roles of 

biogas upgrading technologies could be considered to be reducing CO2 emissions from anaerobic 

digesters or landfills. 

These results are then compared with a Material Flow Analysis (MFA), which quantifies the flows 

and stocks of a system, in order to determine the applicability of the novel technologies. 

2. Methodology 

A life cycle assessment (LCA) was run according to the ISO 14040 [9]. A material flow analysis 

(MFA) was conducted for the waste flow of Spain as a complement to the LCA. 

62

2.1. Life Cycle Assessment 

2.1.1. Goal and Scope 

The goal of this study is to determine the global warming potential (GWP) of biogas upgrading 

technologies. By accounting the GWP, we can identify the process that diverts the highest amount 

of greenhouse gases from being emitted into the atmosphere. 

2.1.2. Functional Unit 

The functional unit used for this study is 1 kWh of biomethane upgraded from biogas which is 

composed of 50% CH4 and 50% CO2. This hypothetical composition is applied as it allows one to 

disregard any prior gas treatment. 

2.1.3. System Boundaries 

The system boundaries include the electricity used to treat the gas, the production of any reagents 

used, the amount of biogas that is upgraded, the amount of methane lost during the process either 

through the treatment (know as methane slip) or lost within the waste gas. Fig. 1 demonstrates the 

boundaries for the LCA and the uncertainty analysis. 

Landfill 

LCA 

Uncertainty Analysis, transport 

Pre-treated 

biogas 

Energy 

Production 

Reagent 

production 

Reagent 

Transport 

Infrastructure 

Methane loss 

CO 2 removal 

process 

Fig. 1. System boundaries 

The processes excluded for the LCA and the uncertainty analyses are the generation of the biogas in 

landfills and its pre-treatment, and the infrastructure for the CO2 removal process and to manage the 

waste generated. The transport of the reagents was excluded from the LCA study, but it was 

included in an uncertainty analysis discussed in section 3.3.2. 

2.1.4. Literature Review 

The technologies that were chosen for the study are: AwR, BABIU, HPWS, PSA, AS, Cry, MS and 

OPS [10]. 

63 

Biomethane 

Waste gas/ 

Carbonated 

waste 

Infrastructure 

waste 

Transport 

Injection into 

Natural gas 

grid

2.2. Life Cycle Inventory 

A life cycle inventory was conducted on the eight chosen technologies. Information on the AwR 

and BABIU process was obtained through direct email communication and information request 

forms sent to the Universities developing these technologies, in the framework of the ongoing Life 

project. Actually, the information for the AwR and BABIU have to be considered preliminary as it 

is the results of the laboratory analysis phase of the project and has been upscaled to industry size. 

Information for the HPWS was obtained through email communications and questionnaires 

received from representatives of two manufacturers, Greenlane Biogas (part of the Flotech Group) 

and DMT Environmental Technologies. Information for the other technologies was obtained 

through literature review. The median point was chosen for information that had more than one 

value. 

Information for reagents used in certain processes was not obtainable and therefore was not 

included in the study, as in these cases their impact could be considered negligible [10]. 

Data for the LCA was complemented by the Ecoinvent 2.2 [11] and GaBi PE databases [12] and 

inventory data for Spain was used. The inventory data used can be found in Table 1. 

Table 1. Life cycle inventory data for biogas upgrading technologies per 1 kWh of biomethane 

(functional unit) 

Inputs Electricity 

(kWh) [11] 

Properties Biomethane 

purity (%) 

BABIU AwR HPWS PSA OPS AS MS Cry reference 

0.017 0.009 0.042 0.051 0.060 0.024 0.068 0.070 [2,3,13- 

22] 

KOH (kg) [11] 0.087 [19] 

H2O (kg) [11] 1.468 0.025 [19,21,22] 

N2 (kg) [12] 0.015 [20] 

DEA (kg) [11] 0.0002 [23] 

BA (kg) 8.890 [20] 

APC (kg) 1.018 [19] 

Diesel (kg) [11] 0.002 [20] 

Biogas (m3) 0.203 0.206 0.203 0.209 0.210 0.202 0.233 0.203 

Heat (kWh) [11] 0.031 0.109 [14,17,24] 

Methane loss 

(%) 

90.3 96.7 98 97.5 97 99 85 98 [2,3,14,16 

-22,25] 

0.78 2.3 1 3.5 4 0.1 13.5 0.65 [2,3,13- 

16,18- 

22,25] 

2.3. Life Cycle Impact Assessment 

The LCA was run using the program GaBi 4.4. The impact indicator selected for this study is the 

Global Warming Potential, 100 years [g CO2 equiv.] from the CML 2001 method [26]. For this 

impact indicator positive values mean that CO2 is being emitted and therefore is considered as a 

negative impact on the environment. Meanwhile negative values mean that CO2 is removed from 

the environment and therefore is seen as a positive impact to the environment, or as a CO2 savings. 

The following assumptions were taken into consideration. The methane that is upgraded (also 

referred to as biomethane) and used as a substitute for natural gas down the line is considered as a 

CO2 savings. The CO2 originally contained in the biogas can either be considered CO2 neutral if it is 

released back into the environment or as a savings if it is stored. The methane slip (methane loss) of 

each process is considered as a CO2 emission. 

As the methane slip and the final biomethane concentration is a property that is inherent to each 

technology, a sensitivity analysis was performed to ensure that the end results were independent of 

64

these factors. A sensitivity analysis was also preformed to evaluate possible changes once the novel 

technologies reach industrial scale. As well, two uncertainty analyses were also performed to 

explore the effects on CO2 emissions in: the regeneration rate in AwR, the distance between a 

municipal solid waste incinerator and AwR and BABIU facilities, and the effect of the country 

where the upgrading plant is located. 

2.4 Material Flow Analysis 

BABIU and AwR are currently being developed with the goal of applying it to waste treatment 

processes (Anaerobic Digesters (AD) and landfills) while using waste from another waste treatment 

process (MSWI). Therefore it is important to study the flows of waste to see whether there would be 

enough Bottom ash (BA) and air pollution control (APC) residues from MSWI for BABIU and 

AwR, respectively. 

Therefore a MFA was conducted on the municipal waste flows of Spain in 2008. This data was 

obtained through literature reviews and personal communications with people in the field [27-31]. 

Once the waste flow was determined three scenarios were planted and explored. 

Fig. 2. Urban waste flow of Spain for2008 

The amount of organic matter (OM) within the flow of unsorted waste was calculated at 41% [27]. 

For the potential amount of biogas generated the following assumptions were made: AD generates 

115m 3 of biogas per t of OM [32], with a capture rate of 100%; and landfills generate 170 m 3 of 

CH4 per t of OM [33], with a capture rate of 30%. The potential amount of BA produced was 

calculated as 20% of the total waste in MSWI. The potential electricity that can be generated in 

MSWI was estimated to be around 0.52 MWh/t of waste and was determined based on information 

provided for a MSWI in Barcelona in 2008 [34]. 

3. Results and Discussion 

3.1. Life Cycle Assessment 

Table 2 shows the g of CO2 saving by each of the technologies under study. The amount of CO2 

saved varies from 1400 g to almost 2000 g. The BABIU process has the lowest global warming 

potential (GWP) and actually the largest potential CO2 savings, 1980 g of CO2 eq. In general all the 

other processes generate about 10% more CO2 emissions than BABIU, except for OPS and MS 

which generate 15% and 25% more emissions, respectively. 

65

Table 2. Global warming potential of biogas upgrading technologies 

Upgrading process Global Warming Potential (g of CO2 Eqv.) 

BABIU -1977 

AwR -1794 

HPWS -1766 

AS -1761 

Cry -1758 

PSA -1714 

OPS -1691 

MS -1489 

Percentage of total impact 

20% 

0% 

-20% 

-40% 

-60% 

-80% 

-100% 

Upgrading Technologies 

Fig. 3. Breakdown of the global warming impact of biogas upgrading technologies 

Fig. 3 demonstrates the role that each component plays in the carbon balance of each technology. 

The biomethane processed and the CO2 stored account for the CO2 savings while the production of 

reagents, electricity and any methane slip contribute to CO2 emissions. 

The amount of CH4 processed and turned into biomethane saves the largest amount and accounts 

for the fact that these technologies overall save CO2 rather than contribute to climate change, as was 

demonstrated in Table 2. All the processes do emit CO2 but the amount saved compensates for this 

impact. Both the BABIU and the AwR process store CO2 and therefore this contributes to an extra 

savings of 198 g and 204 g of CO2 respectively. The BABIU process had the greatest savings as it 

not only processes a large amount of biogas but it also produces a relatively small amount of CO2. 

While AwR stores more CO2 than BABIU it doesn’t have as high of an overall CO2 savings due to 

the production of KOH which counts for 8% of AwR’s GWP. 

For only two of the upgrading technologies, HPWS and Cry, the electricity used produced the 

largest amount of CO2 emissions. For AS the production of required heat was the largest source of 

emissions. Meanwhile, for all the other technologies BABIU, PSA, OPS and MS, the methane slip 

that occurs during the upgrading process had the highest negative impact. In the case of MS, the 

66 

Heat 

Biomethane obtained 

Methane slip 

Reagent 

Electricity 

CO2 stored

methane slip contributes to 13% of the overall impact. For these technologies if the methane loss is 

reduced then their GWP would improve. 

3.2. Sensitivity Analysis 

Each technology has a final biomethane concentration and methane slip that is inherent to each 

process. It is therefore of interest to determine whether these characteristics affect their CO2 

balance. A sensitivity analysis done for all the 8 technologies showed that there is no correlation 

between the GWP of the technologies and the percentage of methane loss nor the final biomethane 

concentration. 

The data obtained for the two novel technologies, BABIU and AwR consist of laboratory scale data 

that was scaled up to industrial scale. Therefore one can rightfully assume that once these 

technologies are developed to the industrial level that the data may not be the same. Though in 

Table 1 it is possible to see that values such as biogas input, electricity use, biomethane purity and 

methane loss for BABIU and AwR fall within the range established by the other six technologies 

that are currently on the market. From Fig. 3 one can see that the electricity use and methane loss in 

play a small role in the overall CO2 impact of the technologies. Therefore one can assume that while 

there may be changes once the technologies are commercialized, the effect on the GWP would not 

be significant. This assumption is supported by a sensitivity analysis conducted where the amount 

of electricity used by both AwR and BABIU was increased to 0.07 kWh (which is the higher end of 

the electricity use by commercialized technologies). Applying this new value only reduced the CO2 

savings by less than 1.5 %. 

3.3. Uncertainty Analysis 

3.3.1. Reagent use in AwR 

As was seen in Fig. 3, one of the largest sources of CO2 for the AwR is the production of the 

alkaline reagent KOH. Currently, the regeneration rate is around 70%, therefore it was decided to 

study if improving the regeneration rate would improve the technology enough so that it could be 

comparable to BABIU and others on the market. As well NaOH is another base that is of interest 

for this process therefore it was also used in this comparison. The AwR using each base at different 

regeneration rates were compared to BABIU, AS and HPWS. 

Global Warming Potential (g of CO 2 equivalent) 

-1400 

-1500 

-1600 

-1700 

-1800 

-1900 

-2000 

-2100 

Regeneration rate of base (%) 

0 20 40 60 80 100 

Fig. 4. Comparison of the global warming potential of using KOH and NaOH at varying 

regeneration rates in AwR 

67 

HPWS 

AS 

BABIU 

AwR using KOH 

AwR using NaOH

As can be seen in Fig. 4 even if for AwR the regeneration rate of both KOH and NaOH is improved 

to 99%, BABIU is still the technology with the greatest CO2 savings. This is due to the fact that the 

AwR process has a slightly higher methane slip than BABIU. Though, since both of these 

technologies are in the development stage the methane slip may improve for both before 

commercialization. 

Using NaOH instead of KOH will result in a greater CO2 savings for AwR. While using KOH, 

AwR passed HPWS at a 65% regeneration rate but NaOH passed HPWS at a 40% regeneration rate. 

If the regeneration rates of either bases is improved a greater CO2 savings is achieved, though if the 

regeneration rate is not improved and NaOH is substituted for KOH then an additional savings of 71 

g can be achieved. 

3.3.2. Transport distance and location of technology 

A variable in the implementation of the novel technologies that could affect the final CO2 emissions 

generated is the location of where the technology is installed. This pertains to both the distance 

between the upgrading plant and a municipal solid waste incinerator (MSWI), and the country 

where the upgrading plant is located. 

As the novel technologies depend on waste coming from MSWI it is important to determine how 

the distance between the MSWI and the location of the upgrading technology affects the GWP. As 

well, large amounts of the waste are needed to run the system, for BABIU it requires 9 kg of bottom 

ash (BA) and 1 kg of air pollution control residues (APC) for AwR, per functional unit of 1 kWh of 

biomethane. It was decided to explore the impact related to transport by truck on a small scale with 

a distance up to 300km. 

The electricity production mix of the country where the technology is installed could have an effect 

on the GWP. For the LCA study the inventory data used was for Spain. We decided to use also the 

electricity production mix for Italy as the pilot plant of BABIU and AwR are presently located 

there. 

BABIU and AwR were compared to HPWS and AS which are the marketed technologies that 

showed the greatest CO2 savings. Though to ensure proper comparability, the energy mixes of both 

Spain and Italy were used for all four technologies. As well a travel of 50km by truck was applied 

to any additional reagents used for AwR, BABIU and the amine used in AS. 

Global Warming Potential (g of CO 2 equivalent) 

-1500 

-1550 

-1600 

-1650 

-1700 

-1750 

-1800 

-1850 

-1900 

-1950 

-2000 

Kilometers travelled 

0 25 50 75 100 125 150 175 200 225 250 275 300 

BABIU transport of BA HPWS AwR transport of APC AS transport: 50 km amine 

Fig. 5. Comparison of global warming potential of distance of transport of bottom ash for BABIU 

and APC of AwR. 

68

As can be seen in Fig. 5 the impact of the distance travelled becomes increasingly significant when 

the amount of waste (APC for AwR and BA for BABIU) transported is increased. From 0 to 125 

km the BABIU process still shows the greatest CO2 savings. At around 145 km the AwR process 

and the BABIU process have the same CO2 savings. At distances greater than 145 km the AwR 

achieves a greater CO2 savings than BABIU, but at the same time they both have a lower CO2 

savings than HPWS and AS. When the distance between the MSWI and a BABIU plant reaches 

around 1315 km the impact from transport becomes higher than any CO2 savings and the process 

begins to have a negative impact on the environment. For AwR, this point is reach at a much further 

distance of around 10475 km. 

As the other part of the study, it was determined that comparatively the country where the system is 

implemented does not have a large effect on the GWP. Overall Spain has a greater CO2 savings than 

Italy but one could state that the effect is negligible. This difference exists due to the fact that Spain 

uses more nuclear and solar energy than Italy [11]. Only in HPWS is it possible to note a difference 

and that is because out of all the 4 technologies the HPWS uses the most energy, therefore 

highlighting better the difference between the two. 

3.4 Material Flow Analysis 

Both BABIU and AwR use waste coming from MSWI in order to remove CO2 from biogas which 

comes from landfills or anaerobic digesters (AD). Therefore it is of interest to determine how much 

BA and APC would be needed and whether enough could be generated. To obtain a general idea, 

the waste flow of Spain in all of 2008 was studied and the hypothetical situation was applied where 

all of the biogas generated was upgraded through either BABIU or AwR. This was considered as 

scenario 1. 

Fig. 2, which demonstrates the waste flow in Spain, highlights the fact that most of the unsorted 

waste goes to either the landfill or for composting. On the other hand, Spain currently does not treat 

a lot of its waste through AD or MSWI. 

Table 2. Scenarios for implementation of BABIU and AwR based on municipal waste flow of Spain 

in 2008 

Scenario 1 

Waste 

received 

(t) 

Estimated 

biogas 

production 

(m3) 

Estimated electricity production 

potential (MWh) 

MSWI BABIU AwR 

69 

BA from 

MSWI 

needed for 

BABIU (t) 

APC 

from 

MSWI 

needed 

for AwR 

(t) 

Anaerobic 

digester 

624,036 37,651,670 185,476 182,570 1,648,882 185,857 

Landfill 9,419,352 393,917,300 1,940,447 1,910,077 17,250,844 1,944,459 

Possible 

BA 

production 

(t) 

MSWI 1,890,000 984,007 378,000 

Scenario 2 

Anaerobic 

digester 

9,283,654 1,067,620,203 5,259,207 5,176,815 46,754,354 5,269,998 

MSWI 6,672,517 3,473,970 1,334,503 

Scenario 3 

Anaerobic 

digester 

624,036 37,651,670 185,476 182,570 1,648,882 185,857 

MSWI 11,309,352 5,888,085 2,261,870 

From Table 2 it can be seen that under scenario 1 not enough waste is treated through MSWI to 

supply sufficient BA or APC to treat all of the biogas emitted from AD and landfills. It might be 

possible to have enough APC to treat biogas from AD using AwR, but there would not be enough to

treat the biogas from landfills and in both cases there would not be enough BA to treat the biogas 

using the BABIU process. 

In an ideal situation countries would have citizen that are engaged enough to ensure that all organic 

material (OM) is selectively collected. In scenario 2 all of this OM is treated in the AD and all 

unsorted non OM waste would be sent to the MSWI. While in this scenario the production of biogas 

is around 2.5x higher, this would in turn require almost 47,000,000 t of BA for the BABIU process 

and 5,000,000 t of APC for the AwR, which could not be satisfied as only 6,000,000 t of waste 

would be treated through MSWI. 

Scenario 3 therefore focuses on increasing the amount of BA and APC generated by sending the 

unsorted waste that would have gone to the landfill to the MSWI instead. In this case there would 

only be biogas coming from AD. Applying this scenario could generate enough APC for AwR and 

even enough BA for BABIU. As well, the potential electricity generated through MSWI is greater 

than the potential electricity from biomethane obtained through upgrading landfill biogas. While 

this situation seems like the best possible choice, given the current infrastructure of waste 

management in Spain, it would not be feasible to implement. Currently there are not enough MSWI 

plants to handle the additional waste. 

4. Conclusion 

Out of the technologies that are currently on the market the HPWS and AS showed the greatest 

potential CO2 savings followed by Cry. In the former and later processes the impact of electricity 

used plays the largest role in the CO2 emissions generated, while for AS the production of heat 

played this role. In the lower end of the spectrum are located PSA, OPS and at last place MS. For 

all of these three technologies the impact due to the methane slip plays the largest role. If the 

technologies are improved in these areas then its potential CO2 savings could possibly be improved. 

The BABIU process showed the overall greatest potential CO2 savings. Though if one starts to 

factor in the distance between the MSWI and the location where the technology is installed, then it 

rapidly decreases in CO2 savings due to the high amount of BA that must be transported. Therefore 

in order for the BABIU technology to keep its position as best technology, it must be installed 

within 125 km of a MSWI. As well since BABIU requires a large amount of BA it was found that 

applying it as a biogas upgrading solution for all of Spain is not realistic. Therefore based on these 

two studies the installation of BABIU should be applied at a local scale where an AD plant or 

landfill can be found close to a MSWI. Therefore it is dependent on whether or not there is a MSWI 

close enough that produces sufficient BA. Meanwhile AwR, which uses less APC per functional 

unit, has more of a leeway in both the distance from a MSWI and the production capacity of the 

MSWI. 

The production of the KOH used in AwR plays a large role in its CO2 impact. If the KOH is 

changed to NaOH then its impact is reduced. AwR can currently obtain a base regeneration rate of 

70%, if this is improved then the GWP is improved as well, though it cannot yet achieve the same 

CO2 savings as for BABIU. 

These novel technologies show a great potential savings mainly due to the fact that they also store 

the CO2 from the biogas. If the CO2 removed from the current technologies is stored then they may 

also show similar savings, though it would be necessary to factor in the impact of the storage 

technology as well. 


The authors of this study would like to thank the Life + 2008 programme for its financial support. 

70

References 

[1] Eurobserv’er, The State of Renewable Energies in Europe. Liébard A., Editor; 2010. Technical 

report: 10 th EurOberv’ER Report. 

[2] Petersson A., Wellinger A., Biogas upgrading technologies – developments and innovations. 

IEA Bioenergy; 2009 Oct. 

[3] Lems R., Dirkse E., Making pressurized water scrubbing the ultimate biogas upgrading 

technology with the DMT TS-PWS system. DMT Dirkse Milieutechniek Environmental 

Technology. 2009. 

[4] Baciocchi R., Corti A., Costa G., Lombardi L., Zingaretti D., Storage of carbon dioxide 

captured in pilot-scale biogas upgrading plant by accelerated carbonation of industrial residues. 

Energy Procedia 2011; 4: 4985-4992. 

[5] Baciocchi R., Costa G., Faraoni G., Lombardi L., Olivieri T., and Zingaretti D., Pilot-Scale 

Carbonation of APC Residues for CO2 Storage and Alkali Regeneration Downstream and 

Absorption Column. Thirteenth International Waste Management and Landfill Symposium; 

2011 Oct 3-7; Cagliari, Italy. 

[6] Mostbauer P., Lenz S., Upgrading of lean landfill gas using MSWI bottom ash. Proceedings of 

Eleventh International Waste Management and Landfill Symposium; 2007 Oct 1-5; Sardinia, 

Italy. CISA publisher. 

[7] Olivieri T., Lombardi L., Mostbauer P., Demonstration Scale Upgrading of Landfill Gas with 

the Use of Bottom Ash –BABIU Process Design and Preliminary Results. Thirteenth 

International Waste Management and Landfill Symposium; 2011 Oct 3-7; Cagliari, Italy. 

[8] Gabarrell X., Font M., Vicent T., Caminal G., Sarra M., Blanquez P., A comparative life cycle 

assessment of two treatment technologies for Grey Lanaset G textile dye : biodegradation by 

Trametes versicolor and granular activated carbon adsorption. International Journal of Life 

Cycle Assessment 2012. In press. 

[9] ISO 14040 Environmental management - Life cycle assessment - Principles and framework. 

2006. 

[10] Starr K., Gabarrell X., Villalba G., Talens L., Lombardi L., Life cycle assessment of biogas 

upgrading technologies. Waste Management 2012; 32(5): 991-9. 

[11] Swiss Center for Life Cycle Inventories. Ecoinvent Data, The Life Cycle Inventory Data. 

Version 2.2. 2010 

[12] PE international, PE international database. Extensions: Ib Inorganic Intermediates, XIV 

Construction Materials. Last updated: 2010 

[13] MT- BIOMETHAN. Gas Treatments, Efficiency - Available at: 

[accessed 02.12.2010] 

[14] Badger G., Zach C., Biogasaufbereitungssysteme zur Einspeisung in das Erdgasnetz - ein 

Praxisvergleich. Munich, Germany: BASE Technologies GmbH, Fraunhofer-Institut für 

Umwelt-, Sicherheits- und Energietechnik, Bavarian Association for the Promotion of Solar 

Energy; 2008 Technical Report No: 02 

[15] Pertl A., Mostbauer P., Obersteiner G. Climate balance of biogas upgrading systems. Waste 

Manage. 2010; 30:92-99. 

[16] Urban, W., Biogas upgrading to pipeline quality - technology and costs. Presentation at 

Connecting Clean Mobility conference; 2007 Nov 14-15; Arnhem, Netherlands. 

[17] Bekkering J., Broekhuis A.A., van Gemert W.J.T., Optimisation of a green gas supply chain – 

A review. Bioresour. Technol. 2010; 101; 450-456. 

[18] de Arespacochaga N., Bouchy L., Cortina J., Larrotcha E., Sanchez M., Peregrina C., Sewage 

biogas to biomethane for grid injection: challenges and opportunities in Europe. ORBIT 2010 

(Organic Resources for the Carbon Economy) Conference; 2010 Jun 29 –Jul 3; Crete, Greece. 

71

[19] Olivieri, T., Lombardi, L., Costa, G., Upgas environmental assessment forms. 2010, personal 

communications. 

[20] Mostbauer, P., BABIU LCI data. 2010, personal communications. 

[21] Rowntree, S., High pressure water scrubber - request for information, GreenLane Biogas. 2010, 

personal communications. 

[22] Kruit, J., PWS - Request for information, DMT Dirkse Milieutechniek Environmental 

Technology. 2010, personal communications. 

[23] Bailey D.W., Feron P.H.M., Post-combustion Decarbonisation Process. Oil & Gas Science and 

Technology - Rev. IFP 2005; 60:461-474. 

[24] Beil M., Beyrich W., Holzhammer U., Krause T. BIOMETHAN. Fraunhofer-Institut für 

Windenergie und Energiesystemtechnik IWES/ Bereich Bioenergie-Systemtechnik. 

Fachagentur Nachwachsende Rohstoffe e. V. (FNR). Gülzow-Prüzen, Germany. 2012 

[25] de Hull J., Maassen J.I.W., van Meel P.A., Shazad S., Vaessen J.M.P., Comparing different 

biogas upgrading techniques. Eindhoven, Netherlands: Eindhoven University of Technology 

and DMT Dirkse Milieutechniek; 2008 Jul 3. 

[26] Guinée J.B., Gorrée M., Heijungs R., Huppes G., Kleijn R., Koning A.d., Oers L.v., Wegener 

Sleeswijk A., Suh S., Udo de Haes H.A., Bruijn H.d., Duin R.v., Huijbregts M.A.J., Handbook 

on Life Cycle Assessment. Operational Guide to the ISO Standards. Dordrecht: Kluwer 

Academic Publishers; 2002. 

[27] Farreny R., Huella de carbono del sector residuos panel de control / escenario actual. Inèdit 

Innovació per la sostenibilitat. 2011 Nov, Personal Communications. 

[28] Ministerio de Medio Ambiente, y Medio Rural y Marino, Perfil Ambiental de España 2010. Ch 

2.7 Residuos. Gobierno de España. 2010; Spain. Technical report: 770-11-221-7 

[29] Instituto Nacional de Estadística, Encuesta sobre recogida y tratamiento de residuos. Residuos 

urbanos. 2008; Madrid, Spain. 

[accessed 22.11.2011] 

[30] Observatorio de la Sostenibilidad en España, Sostenibilidad en España, 2010. Capítulo 13. 

Residuos urbanos. 2010; Spain. Technical Report: 770-10-275-3 

[31] Ministerio de Medio Ambiente, y Medio Rural y Marino, Plan Nacional Integrado de Residuos 

(PNIR). Gobierno de España. 2008 Dec; Spain. 

[32] Vicent T. 2011 Nov 18, Personal Communications. 

[33] USEPA, LandGEM- Landfill gas Emissions Model, Version 3.02. U.S. Environmental 

Protection Agency, Office of Research and Development, National Risk Management and 

Research Laboratory, and Clean Air technology Center. 2005 May. 

[34] Tersa (Tractament i selecció de residus), Memoria 2008 Anexos. Datos de producción. 2009; 

Barcelona 

72




CAPTURE OF CARBON DIOXIDE USING GAS 

HYDRATE TECHNOLOGY 

Beatrice Castellani a , Mirko Filipponi b , Sara Rinaldi a and Federico Rossi b 

a CRB Biomass Research Center, University of Perugia, Perugia, Italy, castellani@ipassnet.it 

b CIRIAF, University of Perugia, Perugia, Italy, filipponi.unipg@ciriaf.it CA 

Abstract: 

According to IPCC Fourth Report, carbon dioxide emissions from the combustion of fossil fuels have been 

identified as the major contributor to global warming and climate change. To reduce these environmental 

concerns, there is a considerable R&D effort in all technical fields to capture carbon dioxide and 

subsequently lower the emissions. 

One of the new approaches for capturing carbon dioxide is based on gas hydrate crystallization. Gas 

hydrates have a large capacity for the storage of gases which also resemble an attractive method for gas 

filtration. 

Gas hydrates are crystalline solids, in which low molecular weight guest molecules are trapped inside cages 

of hydrogen-bonded water molecules. These crystals are stable under high pressures and low temperatures. 

The basis of the separation is the selective partition of the target component between the hydrate phase and 

the gaseous phase. It is expected that carbon dioxide is preferentially encaged into the hydrate crystal phase 

compared to the other components. 

In the present paper, after a comparison of gas hydrates with existing capture technologies, a novel 

apparatus for gas hydrate production is illustrated and results of a first set of experimental applications of the 

reactor for CO2 hydrate formation and separation are presented. Results are a basis for setting up a 

procedure for CO2 separation and capture. 

Keywords: 

Carbon dioxide capture, Gas hydrate, Gas separation, Promoters, Water spraying. 


Carbon capture and sequestration (or storage) - known as CCS - has attracted interest as a measure 

for mitigating global climate change because large amounts of carbon dioxide emitted from fossil 

fuel use are potentially available to be captured and stored or prevented from reaching the 

atmosphere. 

A variety of capture processes have been developed for removing or isolating carbon dioxide from a 

gaseous stream. These processes include absorption, adsorption, membrane separation, cryogenic 

fractionation [1-3]. 

After separation, the captured carbon dioxide must be definitely stored: methods under study 

include storage in depleted oil reserves, salt formations, terrestrial ecosystems and geological 

formations or direct injection into the deep ocean [4]. To be considered viable, a storage method 

must provide stable and long-term storage, be environmentally safe and cost-effective. 

One of the biggest issues is the high energy consumption for CO2 separation. It has been estimated 

that, the cost of separation and disposal of CO2 from existing coal-fired, air-blown boilers would 

increase the cost of electricity by about 75% [5]. The cost of CO2 separation may reduce the power 

generation efficiency from 38 to 26% [6]. 

Therefore, the major research interest is in the development of new less energy intensive processes. 

Several works on estimation of energy consumption for CCS processes are available in literature. In 

every CCS process, separation of CO2 from a flue gas mixture and its compression are the largest 

contributors to the cost [2,6,7]. 

73

Aaron and Tsouris [8] reviewed in detail all processes available for recovery of CO2 from a flue gas 

mixture including some which are still at the laboratory stage and concluded that the absorption 

with monoethaolamine (MEA) is the best method. However the regeneration of the solvent makes 

the absorption process energy-intensive. 

The need to reduce costs motivates further research into the subject. One of the novel separation 

techniques for subsequent storage or utilization of CO2 is through gas hydrate formation [8]. 

Hydrate technology for gas separation seems to be cheaper in case of a CO2 rich source gas and gas 

separation by hydrate technology will be competitive in application fields where the inlet gas has a 

high pressure such as the oil and gas industry [9]. 

Gas hydrates are crystalline solids, in which low molecular weight guest molecules are trapped 

inside cages of hydrogen-bonded water molecules. 

The three most commonly occurring hydrate structures are Structure I (sI), Structure II (sII) and 

Structure H (sH), all with individual crystal structures (see Fig.1). The three structures are formed 

by five different water cavities, the 5 12 , 5 12 6 2 , 5 12 6 4 , 5 12 6 8 and the 4 3 5 6 6 3 [10]. In its pure form, the 

unit cell of the sI hydrate contains two 5 12 and six 5 12 6 2 cavities while a unit cell of the sII hydrate 

contains sixteen 5 12 and eight 5 12 6 4 cavities. Both of these unit cell lattice structures belong to the 

cubic type. The sH hydrate structure is more complex and contains three 5 12 , two 4 3 5 6 6 3 and one 

5 12 6 8 cavities [10]. This hydrate structure forms a hexagonal unit cell. 

Fig. 1. Hydrate structures. 

A given hydrate structure is typically determined by the size and shape of the guest molecule. 

Carbon dioxide is known to form structure I. Each cavity may encapsulate one - or in rare cases 

more - guest molecules of proper sizes. It is the presence of the guest molecule that stabilizes the 

crystalline water structure at temperatures well above the normal freezing point. 

Gas hydrate technology may be used as a CO2 separation method but also in marine sequestration 

applications, where the replacement of CO2 in marine methane hydrate fields is carried out. Results 

of this process will be sequestration of CO2 and release of methane for energetic purposes [4]. 

As a separation method, gas hydrates may be used with treated flue gas from power plants in which 

CO2 is separated from N2 and O2; with synthesis gas from integrated coal gasification power plants, 

74

in which CO2 is separated from H2, but also in biogas upgrading processes, in which CO2 must be 

separated from biomethane [1;9;11]. 

The basis of the separation is the selective partition of the target component between the hydrate 

phase and the gaseous phase. It is expected that CO2 is preferentially encaged into the hydrate 

crystal phase compared to the other components. For instance, the equilibrium pressure of N2 

hydrate is three times greater than that of CO2. This difference allows to separate CO2 from treated 

flue gas, that is a CO2-N2 mixture [12]. 

Flue gas from power plants usually contains from 15% to 20% mol. of CO2 and are released at 

atmospheric pressure. The gas/hydrate equilibrium pressure for this kind of gas mixture is relatively 

high. For example, the equilibrium pressures for a gas mixture containing CO2 at 17.61% mol. are 

7.6 MPa and 11.0 MPa at 274 K and 277 K, respectively [13]. These pressures are not compatible 

with the industrial reality, since the operative cost will be expensive if it is necessary to compress 

the gas to the hydrate formation pressure. 

In addition, evaluations of energy consumption for gas separation processes by the clathrate hydrate 

formation indicate that hydrate separation process is competitive – compared to other conventional 

separation processes - under lower pressure conditions, as well as in case of lower hydrate 

formation heat [14]. 

Consequently, the main challenge is to obtain a decrease in the operating pressure. This task can be 

achieved using specific compounds called promoters that allows to reach milder conditions for 

hydrate formation. A suitable promoter is essential to help in reducing the hydrate formation 

pressure and the energy consumption. 

Conventionally, water-soluble additives are classified either as kinetic or as thermodynamic 

additives. Thermodynamic additives consist of organic compounds and have the tendency to 

displace the equilibrium conditions towards higher temperatures or lower pressures. Kinetic 

additives consist typically of surfactant molecules and have the effect to accelerate hydrate 

formation [10,12]. 

Several studies report a significant reduction of hydrate equilibrium pressures at a given 

temperature by adding small amounts of tetrahydrofuran (THF) in the aqueous phase. Kang et al. 

[13] and Linga et al. [15] found that the equilibrium pressure of hydrates in the presence of this 

additive is considerably lower than the case without the additive. 

Another promoter is Sodium dodecyl sulfate (SDS), which seems the best commercially available 

surfactant to be used for enhancement of hydrate formation [16] and has already been investigated 

in various works [17-20]. It was found that a small concentration of SDS added to the aqueous 

phase drastically increases the kinetics of hydrate formation. 

Recently, Liu et al. [21] and Torré et al. [22] showed that THF and SDS used in combination are 

efficient additives for promoting CO2 hydrate formation. 

According to previous works [23-26] a continuous production of hydrates is feasible, provided that 

the technology assures an optimal contact between gas and liquid phases. 

The choice of the correct gas-liquid mixing method, together with the proper promoter, is crucial 

for producing hydrates in a continuous manner suitable for scale-up to industrial settings. 

The apparatus described in the present work allows the use of aqueous solutions with additives for 

rapid hydrate production. 

In particular, the reactor was designed to maximize interfacial area between reactants. A first set of 

hydrate formation experiments indicated that mass transfer barriers and thermal effects that 

negatively affect conversion of reactants into hydrate are minimized, resulting in fast hydrate 

production and good storage capacity [20]. 

In the present paper, an improved configuration of the apparatus and its application to CO2 hydrate 

formation are presented and discussed. Experiments on formation of hydrates from pure CO2 are 

75

preparatory to further applications, such as biogas upgrading and CO 2 replacement in methane 

hydrates. 

2. Experimental apparatus 

The experimental apparatus consists of a high-pressure cylindrical AISI 304 stainless steel vessel 

with internal diameter of 200 mm, an internal length of 800 mm and a total internal volume of 25 l. 

It has been designed for pressure values up to 120 bar and provided with a safety valve (Fig. 2). 

Fig. 2. Schematic diagram of the experimental apparatus. 

To remove reaction heat and to ensure a rather constant internal temperature, the reactor was 

provided with copper cooling coils wrapped around the outside of its vessel wall and with an 

internal heat exchanger constituted by a finned tube. 

The external cooling coils were coated with a thermally insulating paste and a metallic sheet to 

minimize thermal dispersion from coils to external environment. The cooling medium is ethylene 

glycol – water solution supplied by an air-cooled chiller (GC-LT Eurochiller). 

Two side flanges and a bottom- smaller - flange are used to seal the reactor. One side flange has 

appropriate ports for access to the interior. The five ports are used for inserting 2 temperature 

sensors, for gas inlet and outlet and for pressurized aqueous solution recirculation. The temperature 

sensors are mineral insulated type T thermocouples (accuracy class 1) and measure the temperature 

inside the vessel in the lower part and in the upper part, near spray nozzles (see Fig. 1). The gas 

inlet line is equipped with a pressure sensor, that is a digital piezo-resistive manometer (Kobold - 

accuracy class 0.5) and a mass flow meter (Bronkhorst Hi-Tech) to measure flow rate of gas 

injected in the reactor. Gas is supplied directly by gas bottles through a pressure-reducing valve, 

that provides adjustment of the pressure to the gas injection line. Gas is injected through a manifold 

on which five check valves are mounted. 

The bottom flange has two ports, the former is used for the initial aqueous solution uploading and 

the latter is connected to a pump for recirculation. The pump flows the aqueous solution, previously 

uploaded inside the reactor, to the atomizing manifold. Water is atomized by six hydraulic nozzles 

mounted on the internal part of the manifold itself. Such devices for water spraying allow control of 

76

dimensions of water droplets and water flow. The manifold is equipped with pressure gauge for the 

measure of differential pressure and with a thermocouple for the measure of recirculated solution 

temperature. 

Voltage signals from pressure transducers and temperature sensors are collected by a software for 

data acquisition on a personal computer. The installation of the apparatus is shown in Fig. 3. 

Fig. 3. Installation of the experimental apparatus. 

2.2. Materials 

Carbon dioxide (99% purity) was supplied by Air Liquide Italia Service. Tap water was used to 

prepare solutions. SDS (purity >99%) and THF (purity >99.8%) were from Sigma-Aldrich. 

2.3. Procedure for hydrate formation 

The reactor was designed to produce hydrates in a rapid manner, with hydrate formation times of 

few minutes. Moreover the new - improved - configuration is suitable for hydrate formation both 

through bubbling gas into the liquid phase and through spraying aqueous solution into the gas 

phase. In this set of experiments, the reactor was used to produce carbon dioxide hydrates through 

spraying aqueous solution into the gas phase according to the procedure described below. 

The established amount of aqueous solution is firstly uploaded and the reactor is filled with carbon 

dioxide from gas bottles until the internal pressure equals the experimental pressure and then 

cooled. Gas is bubbled into the liquid phase through 5 check valves. The temperature is controlled 

in order to achieve relatively uniform values inside the reactor. When the experimental conditions 

are reached, aqueous solution is flowed by the recirculation pump through the nozzles. 

Since the flow rate of aqueous solution through the nozzles depends on the differential pressure on 

nozzles themselves, the water spraying is continued for several minutes until the established total 

amount of aqueous solution is injected. During the experiment pressure and temperature data are 

collected every 5 seconds. 

Each experiment is carried out with a constant internal pressure. When the internal pressure 

decreases because of hydrate formation, gas is injected into the reactor to re-establish the correct 

pressure value. 

77

Investigations were carried out in batch conditions. Therefore, at the end of each experiment, gas is 

vented out from gas outlet port. Flange is opened both for visual observations and for taking hydrate 

samples out. 

In fact, several samples are taken directly out from the reactor. Hydrate storage capacity is 

determined putting hydrate samples inside a custom built dissociation vessel. It is a cylindrical AISI 

304 stainless steel vessel with a volume of 1.4 lt. It was designed and built to carry out the 

dissociation of samples of gas hydrate formed. 

After sealing the vessel, the dissociation starts and gas pressure and temperature after dissociation 

are measured. To calculate number of gas moles Eq.(1) was used: 

P V = Z n R T (1) 

where P is the gas partial pressure in the vessel at the end of dissociation, V is the volume of gas in 

the vessel, n is number of the gas moles, T is the temperature in K at the end of dissociation, R is 

the universal gas constant, and Z is the compressibility factor, which can be calculated using 

Benedict-Webb-Rubin equation of state. 

As the number of gas moles is calculated, hydrate storage capacity, measured both in %wt of CO2 

and in Nm 3 /m 3 , can be determined, since hydrate density is known. 

In the calculation of hydrate storage capacity, the contribute of CO2 solubility in water was also 

taken into account. 

3. Results and discussion 

A first set of experimental runs were carried out for CO2 hydrate production. Effects of additives, 

such as THF and SDS, were tested. The amount of additives was chosen according to the optimal 

ranges of concentration found in literature [15, 18, 19]. 

Typical profiles of internal pressure and temperature for an experimental run of 15 minutes are 

shown in Fig. 3. Those profiles are for experimental run 2 in Table 1. 

In particular, internal temperature is calculated as the average of the two temperature values 

measured by two thermocouples in two different positions. 

All experimental runs were carried out with an internal pressure of 3 MPa and with initial 

temperature values of ca. 3 °C. With an experimental pressure of 3 MPa, the equilibrium 

temperature of carbon dioxide hydrates is ca. 280 K [27], therefore experiments were carried out 

with a not negligible subcooling as a driving force for the process. 

Before starting aqueous solution recirculation and spraying, a slight decrease in pressure values was 

observed and ascribed to carbon dioxide solubility. Therefore, only when constant values of 

temperature and pressure were reached, recirculation started and continued for 15 minutes (runs 

1,2,3 in Table 1) or 30 minutes (Runs 4,5 in Table 1). 

As a result of the hydrate formation, which is an exothermic process, internal temperature increases 

after ca. four minutes. Heat removal and temperature control is an issue, especially for applications 

in scaled-up systems, in which constancy and uniformity of internal temperature are difficult to 

achieve. With the improvements brought to the temperature control system, variations were kept 

within 1 °C, as shown in the temperature profile. 

Moreover, internal and external heat exchangers of the apparatus allow to achieve also relatively 

uniform values of temperature inside the entire internal volume. 

In Fig.4 it can be noted that internal pressure is constant for the first minute and then decreases 

smoothly for 6-7 minutes. This can be ascribed to gas consumption due to hydrate formation. The 

following four peaks result from gas injection for re-establishing the fixed experimental pressure. 

After each peak, a rapid decrease in internal pressure, due to formation of gas hydrates, is 

observable. 

78

Temperature (°C) 

8 

7 

6 

5 

4 

3 

2 

1 

0 

16 

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 

Time (min) 

79 

Internal Temperature 

Internal Pressure 

Fig. 4. Pressure and temperature profile with elapsing time for CO2 hydrate formation in SDS 300 

ppm- experimental run 2 in Table 1. Water spraying starts at t = 2 min. 

In Fig. 5 pressure profiles for runs 1, 2, 3 in Table 1 are reported. These three runs help to determine 

the effects of two different promoters (SDS and THF). It is evident that in absence of promoter, 

pressure decreases, due to gas capture in the hydrate structure, are not observable. On the other 

hand, in presence of promotor, pressure starts decreasing after 4-5 minutes, with an induction time 

shorter than those observed in other experiments described in literature, as already proved in our 

previous applications [20]. 

Pressure (bar) 

32 

31 

30 

29 

28 

27 

26 

no promotor 

SDS 300 ppm 

THF 1%wt 

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 

Time (min) 

Fig. 5. Pressure profiles with elapsing time for CO2 hydrate formation with and without 

promotors- experimental runs 1-2-3 in Table 1. Water spraying starts at t = 1 min. 

Several samples were taken directly from the reactor. Hydrate storage capacity was determined 

putting hydrate samples inside a custom-built dissociation vessel, as described in section 2.3. 

32 

30 

28 

26 

24 

22 

20 

18 

Pressure (bar)

Hydrate storage capacity was measured in CO2wt% (weight percentage) and in Nm 3 /m 3 (Nm 3 of gas 

entrapped in 1 m 3 of hydrates), since these two parameters represent, in a macroscopic and 

engineering approach, the capability of hydrates to entrap gases in their structure. 

Results of experimental applications are reported in Table 1. Initial temperature is calculated as the 

mean value of those measured in the lower and upper part of the reactor, before starting 

recirculation and spraying. 

Table 1. Results of experimental applications for CO2 hydrate formation. 

Run T initial, Promotor Water spraying CO2 Wt% Storage 

°C 

time, min 

capacity 

Nm 3 /m 3 

1 1.7 - 15 1.3 6.4 

2 3.0 SDS 300 

ppm 

15 8.0 47.4 

3 2.2 THF 1%wt 15 5.3 30.9 

4 4.0 SDS 300 

ppm 

30 9.2 62.3 

5 3.8 THF 1%wt 30 7.5 44.6 

For comparison, a first experiment without additives was carried out (Run1). Results show that in 

this case gas content is very small and the effect of gas solubility is not negligible. 

Both SDS and THF promote formation of gas hydrates with a short reaction time, suitable for 

industrial in-continuo applications (Run 2-3). Fig. 6 shows the picture of the hydrates formed in 

Run 2. 

Fig. 6. Carbon dioxide hydrates on the internal heat exchanger. 

Gas storage increased of 15%, in case of SDS, and 41%, in case of THF, if the spraying time was 

doubled (Run 4-5). 

Gas storage capacity values are consistent with or rather greater - in case of SDS - than that 

observed by other authors [9]. 

80


In the present paper, a novel apparatus for gas hydrate production is illustrated and results of a first 

set of experimental applications of the reactor for CO2 hydrate formation are presented. 

Improvements on reactor design allowed to overcome issues relating to thermal effects and mass 

transfer barriers, resulting in a rapid CO2 hydrate formation with reaction times of even 15 minutes 

with additive promotion. 

Carbon dioxide hydrate formation was carried out in mild operating conditions, such as pressure 

values of 30 bar and temperature of 2-3 °C. The maximum value of storage capacity was 62.3 

Nm 3 /m 3 in presence of SDS with a reaction time of 30 minutes. Gas storage capacity values are 

consistent with or rather greater - in case of SDS - than that observed by other authors. 

Results on CO2 hydrate formation are preparatory to investigation on other applications of industrial 

interest. In particular, our research activities will focus on CO2 separation from gas mixture, 

especially in case of biogas upgrading, and CO2 replacement in methane hydrates. 


The authors would like to thank Consorzio IPASS Scarl, Italy for providing laboratory personnel 

and materials. 

References 

[1] A. Scondo, A. Sinquin, Effect of additives on CO2 capture from simulated flue gas by hydrates 

formation in emulsion –Proceedings of the 7th International Conference on Gas Hydrates 

(ICGH 2011), Edinburgh, Scotland, United Kingdom, July 17-21, 2011. 

[2] Davison J, Thambimuthu K, Technolgies for Capture of Carbondioxide. In: 7th International 

Conference on Greenhouse Gas Control Technologies (GHGT-7); Vancouver, 2004. 

[3] E.S. Kikkinides, R.T. Yang, S.H. Cho, Concentration and recovery of CO2 from flue-gas by 

pressure swing adsorption, Ind. Eng. Chem. Res. 1993;32:2714–2720. 

[4] R. Sivaraman, The Potential Role of Hydrate Technology in Sequestering Carbon Dioxide. Gas 

Tips, 2003 

[5] C.A. Hendriks, K. Blok,W.C. Turkenburg, The Recovery of Carbon Dioxide from Power Plants 

in Climate and Energy, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1989. 

[6] A. Chakma, A.K. Mehrotra, B. Nielsen, Comparison of chemical solvents for mitigating CO2 

emissions from coal-fired power-plants, Heat Recovery Syst 1995; 15:231–240. 

[7] Herzog HJ, Drake E, Adams E, CO2 Capture, Reuse, and Storage Technologies for Mitigating 

Global Climate Change, White Paper Final Report 1997 DOE: 66. 

[8] Aaron D, Tsouris C, Separation of CO2 from flue gas: A review. Separation Science and 

Technology 2005; 40:321-348. 

[9] Nguyen Hong Duc, F. Chauvy, J.M. Herry - CO2 capture by hydrate crystallization – A 

potential solution for gas emission of steelmaking industry – Energy conversion and 

management 2007, 48:1313-1322. 

[10] Sum, A. K., Koh, C. A., Sloan, E. D. Clathrate Hydrates: From Laboratory Science to 

Engineering Practice. Ind. Eng. Chem. Res. 2009; 48: 7457-7465. 

[11] S. Arca, L. Poletti, R. Poletti, E. D’Alessandro. Upgrading of biogas technology through the 

application of gas hydrates. Proceedings of the 7th International Conference on Gas Hydrates 


[12] Sloan, E. D., Koh, C.A., 2008. Clathrate hydrates of natural gases, third ed. CRC Press, Taylor 

&Francis Group, Boca Raton. 

81

[13] Kang S.P., Lee H., Lee C.S., Sung W.M. Hydrate phase equilibria of the guest mixtures 

containing CO2, N2 and tetrahydrofuran. Fluid Phase Equilibria, 2001;85(1-2):101-109. 

[14] H. Tajima, A. Yamasaki, F. Kiyono, Energy consumption estimation for greenhouse gas 

separation processes by clathrate hydrate formation, Energy 2004; 29:1713–1729 

[15] Linga P., Adeyemo A. and Englezos P. Medium-pressure clathrate hydrate/membrane hybrid 

process for postcombustion capture of carbon dioxide. Environmental Science & Technology, 

2008;42(1):315-320. 

[16] Di Profio, P., Arca, S., Germani, R., Savelli, G., 2005. Surfactant promoting effect on clathrate 

hydrate formation: are micelles really involved? Chemical Engineering Science 60, 4141-4145. 

[17] Kalogerakis N, Jamaluddin AKM, Dholabhai PD, Bishnoi PR. Effect of surfactants on hydrate 

formation kinetics. (SPE 25188). Proceedings of SPE International Symposium on Oilfield 

Chemistry, New Orleands, 1993. 

[18] Zhong, Y., Rogers, R.E., 2000. Surfactant effects on gas hydrate formation. Chemical 

Engineering Science 55, 4175-4187. 

[19] Ganji, H., Manteghian, M., Sadaghiani Zadeh, K., Omidkhah, M.R., Rahimi Mofrad, H., 2007. 

Effect of different surfactants on methane hydrate formation rate, stability and storage capacity. 

Fuel 86, 434-441. 

[20] L. Brinchi, B. Castellani, M. Filipponi, F. Rossi, G. Savelli – Investigation on a novel reactor 

for gas hydrate production - Proceedings of the 7th International Conference on Gas Hydrates 


[21] Liu N, Gong G, Liu D, Xie Y. Effect of additives on carbon dioxide hydrate formation. 

Proceedings of the 6th International Conference on Gas Hydrates (ICGH 2008); Vancouver, 

2008. 

[22] Torré JP, Dicharry C, Ricaurte M, Daniel-David D, Broseta D. CO2 capture by hydrate 

formation in quiescent conditions: in search of efficient kinetic additives. Energy Procedia 

2011;4:621-628. 

[23] M. Mork, Gudmundsson. Hydrate crystallization rate in a continuous stirred reactor: 

experimental results and a bubble-to-crystal model. In: Proceedings of the 4th international 

conference on gas hydrates, vol. 2; 2002. p. 813–8. 

[24] Iwasaki et al. Continuous natural gas hydrate pellet production by process development unit. - 

Proceedings of the 5th international conference on gas hydrates, vol. 4; 2005. p. 4003. 

[25] Hideo Tajima, et al. Continuous gas hydrate crystallization process by static mixing of fluids. - 

Proceedings of the 5th international conference on gas hydrates, vol. 1; 2005. p. 1010. 

[26] Dwain F. Spencer. US Patent: methods of selectively separating CO2 from a multicomponent 

gaseous stream; 2000. 

[27] G. K. Anderson, Enthalpy of dissociation and hydratation number of carbon dioxide hydrate 

from the Clapeyron equation – J. Chem. Thermodynamics 2003;35:1171-1183 

82

Abstract: 




Carbon dioxide mineralisation and integration 

with flue gas desulphurisation applied to a 

modern coal-fired power plant 

Ron Zevenhoven a , Johan Fagerlund a , Thomas Björklöf a,b , 

Magdalena Mäkelä c , Olav Eklund c 

a Åbo Akademi University, Dept. of Chemical Engineering, Åbo / Turku, Finland, 

ron.zevenhoven@abo.fi (CA), johan.fagerlund@abo.fi (presenter) 

b currently with Neste Jacobs, Porvoo, Finland 

c Åbo Akademi University, Dept. of Geology & Mineralogy, Åbo / Turku, Finland, 

magma@abo.fi, oleklund@abo.fi 

For Finland, carbon dioxide mineralisation was identified as the only option for CCS (carbon dioxide capture 

and storage) application. Unfortunately it has not been embraced by the power sector, partly because the 

most suitable mineral resources are found in central and northern Finland while most fossil-fuel fired 

electricity production is located in southern Finland. One interesting source-sink combination, however, is 

formed by the magnesium silicate resources at Vammala, located ~ 85 km east of the 565 MWe coal-fired 

Meri-Pori power plant on the country’s southwest coast, producing 2.5 Mt/y CO2. Between 2008 and 2010 

the companies Fortum and TVO considered retrofitting the Meri-Pori power plant with CCS. Due to absence 

of geological storage options within Finland, the CO2 would be shipped to the North Sea for injection into 

saline aquifers. The project was, however, discontinued. This paper assesses sequestration of Meri-Pori 

power plant CO2 by mineralisation, using the Vammala mineral resources and the mineralisation process 

under development at Åbo Akademi University. That process implies Mg(OH)2 production from magnesium 

silicate-based rock, followed by gas/solid carbonation of the Mg(OH)2 in a pressurised fluidised bed. 

Included here are results on experimental work, i.e Mg(OH)2 production, with rock material from locations 

close to Meri-Pori. Results suggest a total CO2 fixation capacity ~ 50 Mt CO2 for the Vammala site, although 

production of Mg(OH)2 from rock from the site is challenging as a result of varying magnesium silicate 

mineral types (serpentine, amphibole, pyroxene). Finally, as carbon dioxide mineralisation without CO2 precapture 

could be directly applied to flue gases that contain sulphur oxides, we report from experimental work 

where carbonation of Mg(OH)2 with CO2 is compared with CO2-SO2-O2 gas mixtures. Results show that 

SO2 readily reacts with Mg(OH)2, providing an opportunity to simultaneously capture SO2 and CO2. Ideally, 

this could make separate flue gas desulphurisation redundant. 

Keywords: 

CO2 mineral sequestration, Large-scale application, Coal-fired power plant, Desulphurisation 


Between 2008 and 2010 Fortum and TVO explored the possibility of retrofitting the Meri-Pori coal 

combustion power plant with CO2 capture technology. Due to a lack of geological storage options 

within Finland, the CO2 was to be shipped to the Danish North Sea, by ship, for injection into saline 

aquifers. The project was, however, discontinued in October 2010 [1-5]. The Meri-Pori power plant 

(1994) is a 565 MWe coal fired power plant with a thermal efficiency of 43%, producing 2.5 Mt/y 

CO2, or 820 kgCO2/MWh. The plan was to capture and store 1.2 Mt/y of this. For the capture, both 

oxy-fuel combustion and amino acid salt technology (Siemens) were considered, the latter being 

deemed more convenient (easier to retrofit). The CO2 would have been transported, as a cooled 

liquid (-50 ºC, ~7 bar), with two or three tanker ships over a distance of 1000 km to the closest 

storage site. 

The biggest obstacle on Meri-Pori’s path to CCGS (carbon capture and geological storage) was 

considered to be the energy demand of the capture process. Post-combustion capture required 2-4 

83

GJ/tCO2 during pilot scale tests, with 90% capture efficiency. Apart from the capture, other cost 

factors considered are: 

Compression and cooling of the CO2 

Shipping 

Intermediate storage facilities 

Injection into underground storage (including pressurisation to 120 bar) 

This would lower the efficiency of the power plant by 10-13 percentage units. In terms of cost (and 

by deduction, energy), the capture was estimated to account for the largest part, with 50 – 80% of 

the total costs related to CCGS. Transporting the CO2 was estimated to account for 5 – 35% of the 

total costs, and storage for 5 – 25%. 

This paper explores another possibility for sequestering the CO2 emitted by the Meri-Pori power 

plant, namely CO2 mineralisation, using mineral resources located not too far from the power plant. 

There are several motivations for this: 

It is known for quite some time (and repeatedly confirmed) that underground storage capacity is 

not available in Finland [e.g., 6,7], while the same appears to hold for the Baltic region in 

general (apart from CCGS capacity in Poland, a country with a lot of coal-derived CO2 

emissions) [8] 

Finland has vast resources of magnesium silicate-based mineral resources; assessments by the 

Geological Survey of Finland typically mention 2-3 Gt CO2 storage capacity in minerals of the 

Outokumpu-Kainuu region of central Finland alone [9-11] 

Underground storage capacity in west-Russia may seem attractive but export of CO2 to outside 

the European Economic Area is prohibited under the EU directive on CCS (which in fact 

addresses only CCGS) [12] 

Implementation of the above-mentioned EU directive on CCS in Finnish legislation is ongoing 

and may result in CO2 underground storage being forbidden within Finland’s borders [13,14] 

A five-year (2011-2015) research program on CCS is commencing in Finland coordinated by 

Cleen Oy [15]. (The work reported here is outside that program, however.) 

Finland (and at the moment primarily Åbo Akademi University, ÅA) has an extensive track 

record on CO2 mineral sequestration R&D, with process routes that use either both magnesium 

silicate-containing rock [e.g. 16,17, based on presentations at ECOS2010] or steelmaking slags 

moving from lab-scale to demonstration scale [18]. 

Below, the feasibility of CO2 mineralisation applied to CO2 produced at the Meri-Pori power plant 

using four types of minerals and the staged process route that is under development at ÅA is 

assessed. Moreover, the combined removal and trapping of SO2 and CO2 from the flue gas is 

investigated in an experimental study at the end of this paper. CO2 pre-capture would be omitted 

from the CCS chain. 

2. Considering the mineralisation option 

Given that Finland does not possess underground storage sites, CCGS will always entail large 

transport distances and export of the CO2 to Norway, Denmark or Poland. The location of the Meri- 

Pori plant is indicated in Fig. 1. 

Onshore pipeline transport of CO2 is significantly cheaper than shipping, for distances up to 1500 

km (see Fig. 4.6 in [19]). With proven Mg-rich serpentine (Mg3Si2O5(OH)4) deposits as close as 

~85 km from Meri-Pori, in Vammala, and olivine-type material located at Meri-Pori itself, not only 

could transport costs be minimised, but the whole capture process could potentially be omitted if the 

carbonation can be applied to the flue gas directly. Current CO2 mineral sequestration R&D focuses 

more and more on direct carbonation with the CO2 containing gas, removing the expensive and 

complicated (especially for gases that contain oxygen) capture step from the CCS chain. Compared 

to pumping CO2 into saline aquifers, current mineralisation technology comes with an energy 

84

penalty, but without separate capture and long transport distances the energy use should at least be 

comparable to the capture step of CCGS, of the order of 2-5 GJ/ton CO2 (mainly heat). 

In recent years, research into CO2 mineralisation has taken a giant leap forward as demonstrated by 

the rate at which new process routes are suggested, patented and in several promising cases 

developed to large-scale application [20]. Many of these processes do not require pure CO2, but can 

be run with flue gases directly, such as the process routes suggested by Nottingham University [21], 

Hunwick [22] and also the ÅA route, to be described below in more detail. 

Fig 1. The location of the Meri-Pori power plant, indicated as , and ultramafic rock in Finland. 

Also shown a photo of the nickel mine at Hitura and its location, and ultramafic rock findings in 

southern Finland at Vammala and Suomusjärvi (circled). 

3. Mineralisation of CO2 

3.1 - General 

Carbon dioxide mineralisation is the general term describing the sequestration of CO2 by reacting it 

with Mg- or Ca-containing compounds, to produce stable carbonates. Magnesium in particular is 

abundant in the earth’s crust, as silicates such as serpentinite and olivine. Calcium also has a 

potential to store significant amounts of CO2, although calcium silicates are not as abundant as 

magnesium silicates. In general, the exothermic reaction between magnesium or calcium silicates 

and CO2 can be described by reaction 1. 

(Mg,Ca)xSiyOx+2y+zH2z(s) + xCO2(g) x(Mg,Ca)CO3(s) + ySiO2(s) + zH2O(l/g) (1) 

85

These reactions occur in nature over geological timescales (hundreds of thousands of years). 

Research has focused on improving the reaction rates by treating the mineral rock by thermal, 

mechanical or chemical means (Chapter 7 in [19], [20], [23]). Due to the exceptionally large scale 

of CCS processes, all additives must be recovered, and the energy input minimised. Strong acids, 

such as HCl and H2SO4 are able to dissolve the rock rapidly, but are difficult to recover. Hence, 

most promising processes incorporate a combination of weaker acids or ammonium salts and 

thermal and/or mechanical treatment to produce more reactive magnesium containing species. 

Examples of such processes can be found in the literature. 

Both the process developed by Hunwick [22] and a similar one by Maroto-Valer and co-workers 

[21] utilise ammonia to capture CO2. Hunwick claims to react ammonium bicarbonate directly with 

serpentinite, whereas Maroto-Valer extracts magnesium from mineral with ammonium bisulfate, 

before carbonating the MgSO4 with ammonium bicarbonate. The latter reports over 90% conversion 

of Mg to carbonates. 

A process under development at ÅA uses recoverable ammonium sulphate (AS) salt to extract Mg 

from grinded serpentinite rock, under elevated temperatures. The extraction has been shown 

conversions of up to 70% of Mg into either reactive Mg(OH)2 or MgSO4, depending on the desired 

intermediate. Mg(OH)2 reacts directly with CO2 under elevated temperature and pressure, whereas 

MgSO4 can be reacted with aqueous ammonium (bi-) carbonate, in both cases producing 

magnesium carbonates. Here, the route that involves Mg(OH)2 carbonation in a pressurised 

fluidised bed, aiming at obtaining the reaction heat from the carbonation step at a useful 

temperature level is considered – see Fig. 2 for an overview of the process route. 

Magnesium silicate 

mineral 

AS + Mg-silicate 

solid-solid 

reaction 

NH 3 

MgSO 4 etc. 

AS 

Ammonium 

sulphate recycling 

HEAT 

Magnesium 

(and iron) 

extraction 

Residue 

Mg(OH) 2 

Iron oxide 

( steel/iron industry) 

Steam 

Pressurized 

fluidized bed 

> 20 bar, > 500°C 

In the first process step, (preheated) serpentinite rock is thermally treated with ammonium sulphate 

(AS) at 400 – 500 °C and atmospheric pressure for 10 – 60 minutes. A significant amount of the 

magnesium, Mg, in the rock is thus converted to sulphate, MgSO4, which is highly soluble in water. 

Unfortunately, MgSO4 cannot be directly converted with CO2 to MgCO3, but in an aqueous solution 

it can be converted to Mg(OH)2. After cooling, the solid from the reaction with AS is slurried in 

water, leaving behind unreacted mineral and insoluble reaction products, e.g., silica. The pH of the 

filtrate solution is raised to 8 – 9, precipitating iron and calcium (from the mineral, see Table 1 

below) as FeOOH and Ca(OH)2, respectively, while increasing the pH further to 10 – 11 

precipitates Mg(OH)2. For the Finnish Hitura mineral, the preferable conditions for extraction of 

Mg (and Fe) to MgSO4 (and FeSO4) are temperatures 400 – 440 °C, for 30 – 60 minutes at S/AS = 

86 

CO 2 

Steam 

activation 

Fig. 2. A schematic illustration of the mineral carbonation process under development at ÅA. 

3.2 – The ÅA process route 

3.2.1. Mg(OH) 2 production 

Magnesium 

carbonate 

MgCO 3

0.5 – 0.7 kg/kg, resulting typically in 60 – 66% extraction of Mg. Lower temperatures and longer 

reaction times give a higher (relative) extraction of iron. Ammonia vapour, NH3, released during the 

thermal step is collected and used to give the necessary pH increases for precipitation. It is 

thereafter recovered for regeneration of the AS salt downstream, using heat from another process 

step. Nonetheless, the recovery of solid ammonium sulphate from the aqueous form incurs a not 

insubstantial energy penalty. More detail on the procedure is given by Nduagu et al. [24,25]. 

3.2.2. Mg(OH) 2 carbonation 

The Mg(OH)2 produced as described above is converted into MgCO3 in a pressurised fluidised bed 

(PFB) reactor at pressures > 20 bar and temperatures 450 – 600 °C. (Some more detail on the set-up 

is given in section 7.2). Results on conversion levels obtained under varying conditions 

(temperature, pressure, water content of the gas, time, fluidisation velocity) are reported elsewhere 

[16,26,27] for both a synthetic, commercial Mg(OH)2 material and Mg(OH)2 produced from 

Finnish or Lithuanian serpentinites. (A few tests were made under supercritical CO2 conditions, 

pressure > 74 bar, which showed significantly lower conversion levels and rates, suggesting that 

little benefit should be expected from operating at such pressure levels.) It was found that the 

Mg(OH)2 materials produced from the serpentinites are much more reactive (as a result of a ~10× 

larger specific surface of ~45 m 2 /g vs. ~5 m 2 /g), giving conversion levels of 50% within 15 minutes 

for ~300 µm particles. 

The product gas from the carbonator is a hot, pressurised mixture of CO2 and H2O, the solids 

obtained will be partly recycled for further carbonation conversion. Unfortunately, although the 

carbonation reaction is rapid it levels off at carbonation levels up to 65% (the best result obtained so 

far) [27], which appears to be the result of calcination of Mg(OH)2 to MgO. However, it is noted 

that in order for Mg(OH)2 to carbonate, dehydroxylation (i.e. calcination) has to occur. Apparently, 

carbonation at some point becomes slower than dehydroxylation, resulting in a partially calcined 

and carbonated product. Thus, below it is assumed that with ~2/3 of the Mg(OH)2 produced also 

being carbonated the necessary amount of it is 150% of the stoichiometric amount. 

3.2.3. Process energy input requirements 

Since CCS is one of the solutions to what is in fact an energy problem, routes that lead to the 

production of large amounts of CO2 while producing the power and heat for the CCS process are 

obviously not viable. The Meri-Pori plant produces 820 g CO2/kWh electricity, thus CCS with an 

electricity consumption of 1/0.82 = 1.22 kWh = 4.39 MJ/kg CO2 would have a zero net output of 

both electricity and CO2. The use of electricity in CCS processes should be avoided although some 

power consumption will follow from gas compression and crushing/grinding of solid material. 

Fortunately, part of the energy input of a CCS processes would be in the form of heat and at ~ 43% 

thermal efficiency the Meri-Pori plant produces similar amounts of electricity and (waste) heat. 

CCS routes based on CO2 mineralisation appear to be more dependent on heat as energy input than 

the “conventional” route that involves underground storage of CO2, while – as done in the ÅA route 

– the heat output from the carbonation reaction can be benefitted from. (Therefore the higher 

temperature of the carbonation step in the ÅA route, ~500 °C, compared to the earlier suggested 

process route from the Albany Research Center (ARC), currently NETL Albany, in the US, results 

in a better LCA (life cycle assessment) performance of the ÅA route compared to the ARC route 

[28]. The ARC route is based on one-step carbonation in pressurised aqueous solutions at ~150 bar, 

~185 °C [29,30].) At the same time, CO2 mineralisation routes that involve electrochemical steps 

(electrolysis, fuel cells) are very unlikely to have a net CO2 fixation effect [31]. 

As presented at ECOS2010 [17] a quick assessment of energy input requirements for the ÅA route 

can be made based on the reaction heat QE or HE needed for Mg extraction from rock and the heat 

QC or HC released by Mg(OH)2 carbonation. Besides this, crushing/grinding of rock contributes to 

only a few % of the energy input requirements while process integration and optimisation will result 

in improvements to the energy efficiency [17]. 

87

With Mg extraction conversion XE = XMg(OH)2prod and Mg(OH)2 carbonation conversion XC = 

XMg(OH)2carb the net heat input requirements is equal to 

H 

(2) 

E 

Q (MJ/kg CO2 

) XC 

HC 

X E 

with HE = 234.6 kJ/mol Mg extracted (value for 480 °C) and HC = -59.5 kJ/mol Mg carbonated 

(value for 550 °C) as in [17]. For serpentinite (rock mainly composed of serpentine) found at Hitura 

composed of ~84%-wt serpentine, ~13%-wt iron oxides as FeO and ~3%-wt calcium silicates the 

heat input requirements are given in Table 1 for XE and XC ranging from 25 to 100% [17]. 

Table 1. Process energy input requirements (MJ/ kg CO2) according to (2) 

Mg(OH)2 carbonation efficiency 

Mg extraction 

efficiency 

25% 50% 75% 90% 95% 100% 

25% 21.33 20.65 20.32 20.11 20.05 19.98 

50% 10.66 9.99 9.65 9.45 9.38 9.31 

75% 7.11 6.43 6.10 5.89 5.83 5.76 

90% 5.92 5.25 4.91 4.71 4.64 4.57 

95% 5.61 4.94 4.60 4.40 4.33 4.26 

100% 5.33 4.66 4.32 4.12 4.05 3.98 

Of course, incomplete Mg extraction would not have a heat penalty (an endothermic reaction that 

doesn’t occur won’t give an energy penalty) and thus only the last row of Table 1 would apply. At 

the same time, if Mg extraction conversion XE 400 °C. 

Also, temperatures > 400 °C give increased thermal decomposition of the AS salt, with an energy 

penalty. Therefore for a case with XE = 0.75 and XC = 0.75 the heat input requirements for a Hitura 

serpentinite-type material (see below) are 4.32 < Q < 6.10 MJ/kg CO2, and presumably closer to the 

higher value Note that this is heat of ~ T = 450 °C = 723 K: for surroundings temperature T° = 

15 °C = 288 K this corresponds to exergy equal to Ex(Q) = (1-T/T°)·Q = 2.6 – 3.7 MJ/kg. Using the 

exergy of heat allows for comparing it in calculations with power input requirement P, for which 

the exergy Ex(P) = P. 

3.2.4. CO2 mineralisation applied directly on power plant flue gas 

The capture of CO2 from flue gases that contain oxygen and other problematic species is more 

complicated than CO2 (and H2S) stripping from natural gas, and is hard to accomplish against an 

energy penalty lower than 3 – 4 MJ/kg CO2 captured [32]. This is one main reason why CO2 

mineralisation R&D increasingly focuses on avoiding CO2 separation and would operate on the 

CO2-containing gas directly. Energy input requirements for CSM (carbon storage by mineralisation) 

would be of the same order as those for only the capture step of “conventional” CCS. (An LCA 

study on this approach for CO2 mineralisation applied to natural gas – fired electricity production in 

Singapore, using the ÅA route with serpentinite rock purchased from Australia, and considering 

both CO2 capture and operating with the flue gas directly was recently reported [33].) In that case a 

gas with Y%-vol CO2 must be compressed to a total pressure of ~ 20 / (Y/100) bar, which is 

integrated with expansion of a carbonation product gas mixture (in which CO2 is replace by H2O) at 

the same 20 / (Y/100) bar, at ~500 °C, to atmospheric pressure. 

Moreover, the mineralisation of CO2 from a flue gas may be combined with sulphur capture: 

Mg(OH)2 may also react with SO2 (and SO3) present in the flue gas which would then allow for 

removing the FGD unit from a power plant that uses a sulphur-containing fuel. This is addressed in 

sections 7 and 8 below. 

88

3.3 – Utilising waste heat 

All existing mineralisation processes require significant amounts of energy (usually as heat) in 

order to achieve sufficient reaction kinetics and/or favourable thermodynamic circumstances. 

Especially those processes, where large amounts of low quality heat are required, could benefit 

from access to sources of waste heat. Such sources may be low quality steam or flue gas from the 

very power plant the CO2 would be extracted from. 

In the relative vicinity of Pori (at ~ 60 km to the south), the Olkiluoto nuclear power plant is 

located, which produces large amounts of waste heat. Presently this waste heat is rejected into the 

sea as 29500 kg/s of cooling water used to condense low quality steam at a temperature of around 

200 °C, from each of the two reactors already in operation. The heat carried off with the cooling 

water amounts to approximately 1600 MW per reactor. A fraction of that waste heat would be 

enough to carbonate all the CO2 emissions from the Meri-Pori power plant [34]. Unit 3 is under 

construction, for operation in 2014 to generate 2700 MW waste heat besides 1600 MW electricity. 

The ÅA mineralisation process requires heat at above 400 °C. But even with a heat source of a 

lower than required temperature, significant savings could be achieved using a heat pump. This, of 

course, would require that the mineralisation could be performed close to the heat source. 

4. Mineral resources and characterisation 

4.1 – The rock resources considered suitable 

Finland has large amounts of serpentine available. The suitability of any given mining site as a 

source of mineral for carbonation depends on the magnesium (and calcium) contents of the rock. In 

addition, the distance from the power plant and total amount of rock, as well as possible nearby 

sources of waste heat are of importance. Many mines in relative proximity to Meri-Pori are in 

operation or have previously been in operation producing large amounts of serpentinite as mine 

tailings and overburden. Table 2 presents a shortlist of sites with their respective rock compositions. 

Table 2. Composition (as oxides) of the Finnish rock considered for mineralisation 

Location MgO %-wt CaO %-wt *Fe2O3 %-wt SiO2 %-wt Al2O3 %-wt Other %-wt 

Suomusjärvi 16.2 8.6 11.4 47.6 10 6.2 

Hyvinkää / 

Mäntsälä** 

9.8 11.4 13.2 45.9 13.6 6.1 

Lammi / 

Asikkala** 

4.9 6.8 8.0 58.0 15.6 6.7 

Kaipola / 

7.2 

Kuhmoinen** 

7.5 7.7 55.2 14.9 7.5 

Vammala 14.5 5.6 12.5 49.5 8.8 9.1 

Hitura 38.1 0.5 14.8 32.6 0.4 13.6 

* Calculated, presumably a mixture of FeO and Fe2O3, partly (?) Fe3O4. ** Not a mining site 

The Suomusjärvi and Vammala rock and, for comparison with a “better” material, the Hitura nickel 

mine mining tailings are central for this work – see the locations of these in Fig. 1. The other 

southern-Finland rock types (Hyvinkää, Lammi and Kaipola) won’t be further addressed here [35]. 

For the work reported here the materials and chemical compositions as given in Table 3 are used. 

Table 3. Composition (as oxides) of the Finnish rocks for mineralisation as tested 

89

Working names MgO %-wt CaO %-wt *Fe2O3 %-wt SiO2 %-wt Al2O3 %-wt Other %-wt 

Hitura # 36.2 0.5 14.4 24.8

4.5 – Suomusjärvi olivine deposits rock 

Two samples for testing were taken from the region around Suomusjärvi, located roughly half-way 

between Turku and Helsinki – see Fig.1. This is at the south end of the Vammala nickel belt [41]. 

One sample (Suomusjärvi-1) is a side-material from the Salittu quarry where macadam is mined for 

roadmaking. The other sample (Suomusjärvi-2) is an olivine-hornblendite actually from Nummi- 

Pusula (~20 km east of Suomusjärvi), containing ~50% olivine (Mg,Fe)2SiO4 [35]. Data on 

amounts of material is not yet available. 

5. Production of Mg(OH)2 from the rocks considered suitable 

Table 4 summarises the results of Mg(OH)2 production from the materials listed in Table 3, using 

the procedure described in section 3.2.1. For most tests, 2 g rock + 3 g ammonium sulphate powder 

were mixed, with rock particle size 75-125 µm and using an Al sample cup. 

Table 4. Production efficiency of Mg(OH)2 and FeOOH from the Finnish rocks tested 

Hitura #,§ 

“ 

“ 

“ 

Vammala-1 ¤ 

“ 

“ 

Vammala-2 ¤ 

“ 

“ 

Satakunta § 

olivine 

T 

°C 

400 - 440 

480 

440 

550 

420 

460 

520 

420 

460 

520 

480 

550 

Time 

min 

30 – 60 

30 

40 

30 

20 

40 

30 

20 

< 30 

< 30 

pH levels 

- / - 

8-9 / 11-12 

9.5 / 11.5 






91 

FeOOH 

g/g rock 

0.006 

0.037 

0.008 

0.02 

0.02 

0.09 

0.04 

0.04 

0.055 

% Fe 

extracted 

4 

23 

5 

13 

13 

58 

22 

20 

30 

2 

3 

Mg(OH)2 

g/g rock 

~ 0.34 

0.23 

0.39 

0.13 

0.005 

0.035 

0.035 

0.075 

0.070 

0.070 

Suomusjärvi-1 ¤ 520 20 9.9 / 11.9 0.025 23 - 0 

Suomusjärvi-2 ¤ 520 20 9.5 / 11.9 0.055 46 0.03 14 

%Mg 

extracted 

~ 65 

44 

74 

25 

3 

18 

18 

27 

25 

25 

14 

1 

# Ref. [23], § Ref [24], ¤ Ref. [35], * Calculated, presumably a mixture of FeO and Fe2O3, i.e. Fe3O4. 

The amount of rock, mrock, needed to sequester a unit mass of CO2 can be calculated from the MgO 

content of the rock (%MgO) and the extraction of the magnesium from it, producing Mg(OH)2, 

XMg(OH)2 prod(%): 

mrock 

40.3 1 1 

( kg / kg) 

 

m 44 % MgO 100 X (%) /100 

CO2 Mg( OH )2 prod 

1 1 

0.916 

% MgO 100 X (%) /100 

Mg ( OH )2 prod 

(The values 40.3 and 44 give the molar masses (kg/kmol) of MgO and CO 2, respectively.) This 

assumes complete carbonation of Mg(OH)2 to MgCO3; as the best experimental result obtained for 

that so far is 65% 2/3, also the requirement of 1½× the amount calculated with (3) is given in the 

results presented below. 

(3)

The number of years it would take to consume the suitable rock material msite at a certain location 

can be calculated for the sites mentioned above, for a CO2 sequestration rate of 1.2 Mt/year (as was 

the plan with the Fortum / TVO project): 

msite 

t ( year) 

 

 

m year m 

rock / 

m 

m_rock / m_CO 2 

(kg/kg) 

30 

25 

20 

15 

10 

5 

rock 

CO2 

(4) 

where mrock/mCO2 (kg(kg) is given by (3). 

6. Feasibility of the considered deposits for Meri-Pori CO2 

6.1 – Hitura mine rock 

As noted in section 4.2, the rock available at Hitura has a theoretical capacity to sequester more 

than 400 Mt CO2. Figure 3 shows the amount of rock needed to carbonate 1 t CO2 as a function of 

Mg extraction, for full and partial (65%) carbonation extent. With Mg(OH)2 production levels of the 

order of 65- 70% obtained for the rock processing according to Nduagu et al., the process would 

require ~ 5 ton rock / ton CO2 and could sequester 1.2 Mt CO2/y during a period of ~150 years with 

65% Mg/OH) carbonation. According to Romão et al. [17] - see also Table 1 – the heat 

requirements would be ~ 6 GJ/t CO2 which can be reduced somewhat by heat integration. It would 

be reduced to ~ 4 GJ/t CO2 if extraction and carbonation levels > 90% can be realised. 

100% Mg(OH) 2 carbonatio n 

0 

0 

0.00 0.20 0.40 0.60 0.80 1.00 

Mg(OH) 2extraction from rock Mg (-) 

m 

m 

site 

CO2 

 

m 

/ year 

m 

450 

400 

350 

300 

250 

200 

150 

100 

50 

rock 

CO2 

operaration for 1.2 Mt CO 2 / y ear 

(years) 

92 


(kg /kg) 

50 

40 

30 

20 

10 

65% Mg(OH) 2 carbonation 

0 

0 

0.00 0.20 0.40 0.60 0.80 1.00 

Mg(O H) 2 extraction from rock Mg (-) 

Fig. 3 Rock consumption rate and availability at Hitura depending on extraction of Mg from the 

rock, for full (left) or partial (right) carbonation of the Mg(OH)2 produced. 

The distance of around 300 km from Meri-Pori to the Hitura site would add a few €/t (on-shore 

pipeline) transport costs to this CCS option, making a site like Stormi-Vammala more attractive. 

6.2 – Vammala (Stormi-Vammala) nickel mine rock 

As presented in section 4.3, the rock available at Stormi-Vammala has a theoretical capacity to 

sequester ~50 Mt CO2. Figure 4 shows the amount of rock needed to carbonate 1 t CO2 as a 

function of Mg extraction, for full and partial (65%) carbonation extent. With Mg(OH)2 production 

levels of ~ 25% obtained for the rock processing [35] the process would require ~ 12 ton rock / ton 

CO2 and could fix 1.2 Mt/y during a period of < 10 years only, with 65% Mg(OH)2 carbonation. 

The heat requirements would be in the range of 4.5 – 20 GJ/t CO2 depending on whether nonreactive 

material behaves as inert or not. These values can be reduced somewhat by heat integration 

but the most urgent need for improvement is the extraction of Mg and producing more Mg(OH)2 

from the rock material. With the current result the contribution of crushing and grinding the rock 

m 

site 

1. 

2 Mt CO / year 

2 

300 

250 

200 

150 

100 

50 

operaration for 1.2 Mt CO 2 / year 

(years)

material will change from a few % to a significant energy penalty. Again, note that the rocks 

collected and analysed are from a rock mass nearby, not at, the Stormi mine [35]. 


(kg/kg) 

40 

30 

20 

10 

100% Mg(OH) 2 carbonation 

0 

0.00 0.20 0.40 0.60 0.80 1.00 

Mg(OH) 2extraction from rock Mg (-) 

50 

45 

40 

35 

30 

25 

20 

15 

10 

5 

0 

operaration for 1.2 Mt CO 2 / y ear 

(years) 

93 


(kg /kg) 

60 

50 

40 

30 

20 

10 

65% Mg(OH) 2 carbonatio n 

0 

0 

0.00 0.20 0.40 0.60 0.80 1.00 

Mg(OH) 2 extraction from rock Mg (-) 

Fig. 4 Rock consumption rate and availability at Stormi-Vammala depending on extraction of Mg 

from the rock, for full (left) or partial (right) carbonation of the Mg(OH)2 produced. 

6.3 – Pori olivine deposits rock 

A small number of experiments were carried out with olivine from Åheim in Norway, which 

confirmed (with < 10% of present Mg extracted) the hypothesis that the method developed by 

Nduagu et al. is not well applicable to olivines. Tests on the Satakunta olivine diabase gave no good 

result: Mg extraction was a disappointing 15% of the material’s Mg content which is already quite 

low at only 5.5%. Work at ÅA is ongoing to further analyse the application of the Mg(OH)2 

production method on minerals like olivine and enstatite, besides serpentine. 

6.4 – Suomusjärvi olivine deposits rock 

The experimental results showed that Mg(OH)2 could be produced only from the rock material 

Suomusjärvi-2, (Nummi-Pusula), at an extraction of ~14% of the ~21%-wt of MgO in the material. 

This implies that ~ 3 t Mg(OH)2 can be produced from ~100 t of rock, which is ~10-15× the amount 

of rock needed compared to a Hitura-type serpentinite and crushing / grinding energy needs become 

a significant cost factor. As noted above, more work is needed on extending the capabilities of the 

Nduagu et al. route to Mg(OH)2 production from “low quality” (< 20%-wt MgO) type of rock. 

7. Combined SO2 capture and CO2 mineralisation: scope 

7.1 – Background 

Very few studies have been conducted using Mg(OH)2 for sulphur and/or CO2 capture, mainly due 

to its limited operational temperature range [42,43], but there are some studies that consider the use 

of MgO based sorbents [44,45]. Still, a similar material that has been much more studied is calcium 

oxide (e.g. [42,43,46-48]). Although calcium-based species are considered not abundant enough for 

large scale CO2 sequestration at levels that mitigate global warming and climate change [49], it is 

being widely studied for the use of separating CO2 from flue gases by means thermal cycling (see 

e.g. a review by Stanmore and Gilot [50]). The idea is to carbonate CaO in one fluidised bed (FB) 

reactor and then decompose the formed CaCO3 in another FB releasing a pure stream of CO2, while 

simultaneously recovering the CaO for re-use in the first reactor. In contrary to these studies, 

carbonation of Mg(OH)2 for CO2 sequestration purposes does not require the recycling of the 

reactant, and the product from the carbonation unit is ready for re-use or final disposal. This is 

beneficial as the continuous carbonation-calcination cycling has been shown to reduce (often 

quickly) the performance of the used material (CaO, in most cases) [50]. 

35 

30 

25 

20 

15 

10 

5 

operaration for 1.2 M t CO2 / year 

(years)

In the case of simultaneous sulphation and carbonation experiments, sulphate formation has been 

found to dominate when Ca(OH)2 or CaO is exposed to a SO2 (1 ppm) and CO2 (6 ppm) containing 

gas [47]. Although, the conditions were different from those studied here, it was also noted that, 

SO2 readily reacts with Ca species at dry conditions, while some humidity was needed to form 

carbonate. On the other hand, Stanmore and Gilot [50] commented in a review article that 

carbonation is initially much faster than sulphation and only in the longer run does sulphation 

become the principal reaction. In contrast to this study, however, it should be noted that the sorbents 

considered were Ca-based and that the conditions did not incorporate elevated pressures. 

Despite the fact that SO2 readily reacts with various Ca-based sorbents, the conversion levels 

obtained are typically in the range of 30–50% [42,50] leaving room for considerable improvement. 

The reason for the low conversion levels has been attributed to the closing of pores at the surface of 

the reacting particles due to the larger molar volume of calcium sulphate than that of either calcium 

oxide or –carbonate [50]. 

To date, most of the experiments using the ÅA mineral carbonation process have only considered 

the use of pure pressurised CO2. However, in order for the process to become a realistic alternative, 

it is apparent that it needs to work with diluted CO2 streams as well, such as industrial flue gases. In 

this paper we present the results from a number of experiments using CO2 containing a small 

amount of sulphur dioxide and oxygen, both of which are common components in typical industrial 

flue gases. If successful, simultaneous carbonation and sulphation may motivate the removal of flue 

gas desulphurisation (FGD) equipment from sulphur-containing fossil fuel-fired power plants. 

7.2 – Materials and methods 

The materials used for the gas-solid carbonation experiments consists of two different types of 

Mg(OH)2, one commercially obtained (Dead Sea Periclase Ltd.) and one derived from Finnish 

serpentinite according the ÅA process described briefly above. From here on, these will be referred 

to as DSP-Mg(OH)2 and serp-Mg(OH)2 respectively. 

DSP-Mg(OH)2 has already been studied extensively [27] and is also used for reference purposes in 

this paper. However, the difference between serpentinite-derived Mg(OH)2 and DSP-Mg(OH)2 is 

apparent from surface analysis and typically serp-Mg(OH)2 has a much higher specific surface area 

(~50 vs. ~5 m 2 /g) and porosity (0.24 cm 3 /g vs. 0.024 cm 3 /g) than DSP-Mg(OH)2. For this reason, 

serp-Mg(OH)2 offers a much greater potential in form of reactivity and reaction extent than DSP- 

Mg(OH)2. 

The gas used in the carbonation experiments has been a high purity (99.999%-vol) CO2 bottle and a 

pre-mixed CO2-O2-SO2 bottle with 90%, 8% and 2%-vol of each component respectively. For some 

experiments steam was added to the gas stream. The amount of SO2 in the gas-mixture was varied 

between 0 and 20 000 ppmv (parts per million, volumetric). 

A more thorough description of the methods used to carbonate Mg(OH)2 can be found elsewhere 

[26,27], but for purposes of continuity, a short summary is also given here. 

The experimental setup for gas/solid carbonation at ÅA, see Fig 5, consists of a small (height 

~0.5 m, inner diameter ~1.5 cm) pressurised fluidised bed (PFB) that is operated by preheating the 

incoming fluidisation gas and by maintaining the reactor at the target conditions during each 

experiment. The PFB is operated as a bubbling fluidised bed and run in batch mode (max. temp. 

~600 °C, max pressure ~100 bar). After each experiment the particles are blown out and collected 

by a cyclone for easy removal. 

94

Fig. 5 Schematic diagram of the PFB setup at ÅA [26,27]. 

The appropriate fluid velocity is regulated/maintained by two flow controllers that allow for the 

mixing of two gas streams. In addition a HPLC-pump is used for adding water to the system prior to 

the preheater when steam is required in the reaction gas. 

In contrary to previous studies, where a simple method of reacting the carbonated product with 

hydrochloric acid could be used for determining the carbonate content [51], the now both 

carbonated and sulphated Mg(OH)2 samples were analysed for elemental carbon and sulphur using 

an ELTRA elemental analyser. 

7.3 – Thermodynamics 

We have recently concluded that the reaction between Mg(OH)2 and CO2 is likely taking place in 

accordance to the following overall equations [27]: 

2 2 2 

Mg OH MgO H O* MgO + H O(g) 

(5) 

MgO H2O* + CO 2(g) MgCO 3 + H2O(g) (6) 

The equations have been simplified to emphasize the importance of H2O, but it is clear that the 

elementary reactions taking place on the surface of Mg(OH)2 particles are more complex than what 

the simple direct carbonation of Mg(OH)2 would suggest. To further highlight the role of H2O, the 

assumed intermediate product of Mg(OH)2 dehydroxylation is MgO H 2O* 

, which corresponds to 

magnesium capable of forming carbonate. In the absence of water the reaction between MgO and 

CO2 is slow and likewise if the temperature for Mg(OH)2 dehydroxylation is not exceeded very 

little carbonation will take place [52]. Adding sulphur dioxide to the reaction gas will compete with 

the formation of magnesium carbonate, but to what extent is highly dependent on the SO2 

concentration. In any case, a SO2 concentration at < 0.5%-vol is much lower than that of CO2 in a 

typical flue gas. Similar conclusions have also been established for calcium-based species [53]. 

In a study by Hartman and Svoboda [44] a number of alternative reaction mechanisms between 

magnesium species and SO2 were suggested. However, in the reaction conditions investigated here, 

95

the formation of MgSO3 is found thermodynamically infeasible [44] (see also Fig. ), but in the 

presence of oxygen the following reaction has been suggested [44,47]: 

1 

MgO + SO (g) + O (g) MgSO 

2 2 4 

(7) 

2 

Increasing the concentration of oxygen further, increases the amount of sulphur trioxide in the gas 

and thus the reaction between MgO and SO3 also needs to be considered [44]: 

MgO + SO 3(g) MgSO4 

(8) 

In addition to the reactions above, the possibility of MgCO3 reacting with SO2 (or SO3) to form 

MgSO4 cannot be ignored. The conversion of MgCO3 to MgSO4 is given by the equation below: 

1 

MgCO + SO (g) + O (g) MgSO + CO (g) 

3 2 2 4 2 

(9) 

2 

From thermodynamic equilibrium calculations (HSC Chemistry 5.11 software) it can be concluded 

that both Reactions (8) and (9) are thermodynamically favoured under the experimental conditions 

investigated here. It appears that MgSO4 is stable up to 640 °C for SO2 and SO3 concentrations 

above 0.1 ppmv. Furthermore, as long as the CO2/SO2 ratio is below 10 10 , sulphation of MgCO3 is 

also feasible as can be seen from Fig. 6. In other words, even if the concentration of SO2 is only 

0.1 ppbv (parts per billion, volumetric) in CO2, sulphate is stable. In order to perform the 

equilibrium calculations, the amount of oxygen in the gas was arbitrarily chosen to be 3.5%-vol. 

Fig. 6. Equilibrium concentrations of SO2 (Reaction 3) and SO3 (Reaction 4), together with the 

equilibrium ratio, CO2/SO2 (Reaction 5) as a function of temperature based on Gibbs energy 

minimisation calculations (HSC Chemistry 5.11). O2 (arbitrarily chosen) in the gas phase: 

3.5%-vol. 

96

8. Combined SO2 capture and CO2 mineralisation: results 

8.1 – Test results with DSP-Mg(OH) 2 

In accordance with results from experiments with similar calcium-based species [46,47] and the 

thermodynamic calculations shown in Fig. 8, the reactivity of Mg(OH)2 towards SO2 (and SO3) is 

significant as seen in Fig. 7. Typically, the concentration of SO2 in a flue gas is much lower than 

20 000 ppmv, hence the most interesting result in Fig. 7 are the ones showing the influence of SO2 

concentrations below 5 000 ppm (= 0.5%-vol). 

Fig. 7. The influence of SO2 concentration (0–20 000 ppmv) on the reactivity of DSP- 

Mg(OH)2 (125–212 m) under a total pressure of 25 bar. The experiment time was 15 minutes. PO 2 

= 4·PSO 2 in dry CO2. 

Interestingly the influence of SO2 is much stronger than that of CO2, although it was present in 

much lower concentrations (90%-vol CO2 vs. 2%-vol SO2). However, it should be noted that SO2 is 

a much stronger acid than CO2 [53] and likely has a stronger affinity for the basic surface of the 

Mg(OH)2 particles. The reason why carbonation is inhibited considerably in the presence of even 

small amounts (500 ppmv) of SO2 is unclear, but evident from the data shown in Fig. 7. 

In addition, the small increase in carbonate formation between experiments performed under 1 000 

and 5 000 ppmv SO2 is interesting, but could be the result of slightly differing temperature 

conditions inside the fluidised bed. 

Because all of the experiments shown in Fig. 7 were performed for 15 minutes, the results represent 

a kind of steady state and product layer diffusion (much slower than the initial kinetics) is necessary 

for further reactivity. The initial reactivity of DSP-Mg(OH)2 is comparatively fast and can be seen 

in Fig. 8 from a set of experiments performed under similar conditions (510 °C, 25 bar), but for 

different durations. However, it should be noted that 2%-vol SO2 in the gas-phase is very high 

compared to industrial standards. 

Although the influence of temperature can be seen from Fig 7, it is easier to compare its effect on 

both carbonation and sulphation from Fig. 9. Increasing the temperature beyond 520 °C, results in 

the decomposition of MgCO3, which is why this is the maximum temperature in the graphs below. 

97

Fig. 8. Reactivity of DSP-Mg(OH)2 at 510 °C, 25 bar. SO2 concentration 20 000 ppmv (=2%-vol). 

It is likely that a further increase in temperature would result in a slightly higher sulphation degree 

after which pore closure is expected. The porosity of the used DSP-Mg(OH)2 material is relatively 

low (0.024 cm 3 /g) and it has been noted to limit the overall conversion attainable during 

carbonation experiments [27]. 

Fig. 9. Carbonate- (left) and sulphate (right) conversion using DSP-Mg(OH)2 as a function of 

temperature at 25 bar and PO 2 = 4·PSO 2 in dry CO2. 

It can be noted that sulphur contents between 500 and 5 000 ppmv resulted in similar carbonation 

extents, but not sulphation extents. Clearly the formation of sulphates is primarily driven by 

reactions (3) and (4), while the transformation of carbonate to sulphate, reaction (5), is less 

pronounced. A similar conclusion was reached by Wang et al. [54], who studied the simultaneous 

sulphation and carbonation of CaO in a CO2 rich gas (oxy-fule combustion) in the presence and 

absence of steam at temperatures between 600–800 °C. 

8.2 – Test results with serpentinite-derived Mg(OH) 2 

In order to investigate the effect of pore closure another Mg(OH)2 material was used for the 

experiments. This material (serpentinite-derived) is more porous and has a higher surface area than 

that of DSP-Mg(OH)2. The results from only three such experiments, shown in Fig. 10, indicated 

that both carbonation and sulphation increased, likely due to the increased particle surface area. 

However, more experiments are required before any further conclusions from the data can be made. 

In addition, two experiments with steam in the gas-phase were performed, but both experiments 

(1%-vol, 2%-vol steam) resulted in lower overall conversion of the material. Thus it seems that, 

dehydroxylation is a necessary precursor to both carbonation and sulphation (as also suggested by 

Reactions 5 and 6). This is a topic for future work, while scale-up tests are ongoing on producing 

larger batches of serp-Mg(OH)2. 

98

Fig. 10 Carbonate- (left) and sulphate (right) conversion using serpentinite-derived Mg(OH)2 as a 

function of temperature at 25 bar and PO 2 = 4·PSO 2 in dry CO2. For reference purposes, the results 

using DSP-Mg(OH)2 at similar conditions (500 and 1 000 ppmv SO2) have also been included. 


A study was made on the application of CO2 mineral carbonation for the sequestration of CO2 from 

the Meri-Pori power plant in Finland, after a plan for “convential” CCS, (or rather CCE, carbon 

capture and export, to Norway/Denmark) was cancelled. Serpentinite rock at Hitura, a nickel mine 

at ~ 300 km from Meri-Pori would be useful, with good and well-defined process chemistry and 

energy efficiency, and is available in much more than sufficient amounts. Serpentinite rock 

available at a shorter distance of ~70-80 km at Vammala appears to be available in a reasonably 

sufficient amount but the production of Mg(OH)2 for subsequent carbonation must be improved 

before it can be considered. Still more improvement and R&D work is needed on olivine-containing 

dunites (>90% olivine), peridotites (60-100% olivine) and olivine diabase – type rock: the latter is 

available in the immediate surroundings of the Meri-Pori power plant but the Mg(OH)2 production 

method isn’t (so far) able to extract more than a few % of the materials Mg content, which is also 

much lower than for the serpentinites considered. Also the possibility to co-capture SO2 and CO2 

was evaluated. It was found that the reactivity of Mg(OH)2 towards SO2 in the presence of CO2 at 

pressurised conditions is significant even under low SO2 partial pressures. As a consequence, the 

possibility to replace a conventional flue gas desulphurisation unit with a combined CO2 and SO2 

scrubber could be considered. However, in order to be more conclusive, additional experiments are 

required. 


This work was supported by KH Renlund’s Foundation funding 2010 for project “Large scale CO2 

mineralisation in Finland: broadening the mineral resources horizon". Vilija Vaijegaite and Agne 

Babarskaite, visiting from Kaunas University of Technology, Lithuania, are acknowledged for the 

experimental work on the Vammala and Suomusjärvi rocks. Experience Nduagu and Inês Romão of 

ÅA are acknowledged for comments and feedback. The authors would also like to acknowledge 

Nordkalk and in particular Thomas Nyberg for the analysis of carbonated/sulphated samples. 

Nomenclature 

AS ammonium sulphate 

Ex exergy, J or W 

H enthalpy, J/mol 

99

m mass, kg 

P power, J or W 

Q heat, J or W 

t time, year 

T remperature, °C or K 

X degree of conversion, - 

y year 

Y %-vol CO2 in flue gas 

difference 

Subscripts and superscripts 

C Carbonation 

E Extraction (of Mg) 

Mg(OH)2carb Mg(OH)2 carbonation 

Mg(OH)2prod Mg(OH)2 production 

site rock deposit site 

References 

[1] Fortum and Polish PGE cooperate on CCS technology. Carbon Capture and Storage Journal, on-line 

version, April 17, 2009, Available at: [Accessed: 25.1.2012] 

[2] Siemens, Fortum and TVO cooperate on Finnish project. Carbon Capture and Storage Journal, on-line 

version, October 23, 2009, Available at: [Accessed: 

25.1.2012] 

[3] Fortum, TVO and Maersk join for Finnish project. Carbon Capture and Storage Journal, on-line version, 

October 23, 2009, Available at: [Accessed: 25.1.2012] 

[4] Fortum scraps CCS project in Finland. Carbon Capture and Storage Journal, on-line version, November, 

2010, Available at: [Accessed: 25.1.2012] 

[5] Iso-Trykkäri, M., Meri-Pori CCS project. Presented at the ClimBus seminar on CCS, Espoo, Finland, 

Nov. 11, 2010. Avialable at: 

[Accessed: 25.1.2012] 

[6] Koljonen, T., Siikavirta, H., Zevenhoven, R., Savolainen. I. CO2 capture, storage and reuse potential in 

Finland, Energy 2004; 29:1521-1527 

[7] Teir, S. et al. Potential for carbon capture and storage (CCS) in the Nordic region, VTT Research notes 

2556, 2010 (223 p). Available at: [Accessed: 


[8] Shogenova, A. Insight into storage of CO2. Presentation at the Cleen CCSP kick-off seminar, Espoo, 

Finland, Sept. 6, 2011. 

[9] Aatos S., Sorjonen-Ward, P., Kontinen, A., Kuivasaari, T., Outlooks for utilisation of serpentine and 

serpentinite (in Finnish: Serpentiinin ja serpentiniitin hyötykäyttönäkymiä) Geological Survey of 

Finland (GSF), Report No. M10.1/2006/3, Kuopio, Finland (2006). 

[10] Teir, S., Aatos, S., Kontinen, A., Zevenhoven, R., Isomäki, O.-P. 2006. Silicate mineral carbonation as a 

possible sequestration method of carbon dioxide in Finland. (in Finnish, English summary): 

Silikaattimineraalien karbonoiminen hiilidioksidin loppusijoitus-menetelmänä Suomen oloissa) Materia 

2006;63 (1):40-46. 

[11] Aatos, S., Kujanpää, L., Teir, S. CO2 capture and geological storage applications in Finland. Geological 

Survey of Finland, 2011; Special Paper 49, 187-193 

[12] Directive 2009/31/EC of the European Parliament and of the Council of 23 April 2009 on the 

geological storage of carbon dioxide. Available at: [Accessed: 25.1.2012] 

100

[13] Ekdahl, E., Idman, H. Statement on the applicability of Directive 2009/31/EC in Finland for Finnish 

Ministry of the Environment, K/468/42/2010, March 14, 2011. 

[14] Proposal for implementation of legislation on CCS, Finnish Ministry of the Environment, (in Finnish 

and in Swedish), July 12, 2011, Available at < http://www.ymparisto.fi/ 

default.asp?node=3662&lan=fi#a1> [Accessed: 25.1.2012] 

[15] Finlands Cleen Oy CCSP (Carbon Capture and Storage Program). Available at: 

[Accessed: 


[16] Fagerlund, J., Nduagu, E., Romão. I., Zevenhoven, R. CO2 fixation using magnesium silicate minerals. 

Part 1: Process description and performance. Energy 2012;41:184-191 

[17] Romão, I., Nduagu, E., Fagerlund, J., M. Gando-Ferreira, L., Zevenhoven, R. CO2 Fixation Using 

Magnesium Silicate Minerals. Part 2: Energy Efficiency and Integration with Iron-and Steelmaking. 

Energy, 2012;41:203-2011 

[18] Mattila, H.-P., Grigalinait, I, Said, A., Filppula, S., Fogelholm, C.-J., Zevenhoven, R. Process 

efficiency and optimization of precipitated calcium carbonate (PCC) production from steel converter 

slag. Submitted to ECOS2012, Perugia, Italy, June 2012; resubmitted after minor revision 

[19] IPCC Special Report on Carbon Dioxide Capture and Storage B. Metz, O. Davidson, H. de Coninck, M. 

Loos, L. Meyer, Working Group III of the IPCC, Cambridge Univ. Press (2005) Available at: 

 

[20] Zevenhoven, R., Fagerlund, J., Songok, J.K. CO2 mineral sequestration - developments towards largescale 

application. Greenhouse Gases: Science and Technology. 2011; 1:48-57 

[21] Wang, X, Maroto-Valer, M. Dissolution of Serpentine using recyclable ammonium salts for CO2 mineral 

carbonation. FUEL 2011; 90(3): 1229-1237 

[22] Hunwick, R.J. A new, integrated approach to mineralisation-based CCS, Modern Power Systems, 2009; 

November:25-28 

[23] Sipilä, J., Teir, S., Zevenhoven, R. Carbon dioxide sequestration by mineral carbonation – Literature 

Review Update 2005-2007. Åbo Akademi Univ., Heat Engineering Lab. report VT 2008-1, Turku, 

Finland (2008). Available at: 

[24] Nduagu, E., Björklöf, T., Fagerlund, J., Wärnå, J., Geerlings, H., Zevenhoven, R. Production of reactive 

magnesium from magnesium silicate for the purpose of CO2 mineralization. Part 1. Application to 

Finnish serpentinite. Minerals Engineering, 2012;30:75 - 86 

[25] Nduagu, E., Björklöf, T., Fagerlund, J., Mäkelä, E., Salonen, J., Geerlings, H., Zevenhoven, R. 

Production of reactive magnesium from magnesium silicate for the purpose of CO2 mineralization. Part 

2. Mg extraction modeling and application to different Mg silicate rocks. Minerals Engineering, 2012, 

2012;30:87-94 

[26] Fagerlund, J., Zevenhoven, R. An experimental study of Mg(OH)2 carbonation. Int. J. of Greenhouse 

Gas Control 2011; 5:1406-1412 

[27] Fagerlund, J., Carbonation of Mg(OH)2 in a pressurised fluidised bed for CO2 sequestration [PhD 

thesis], Turku, Finland: Åbo Akademi University / Chemical Engineering (2012) Available at: 

[Accessed: 1.5.2012] 

[28] Nduagu E, Bergerson, J., Zevenhoven, R. Life cycle assessment of CO2 sequestration in magnesium 

silicate rock - a comparative study. Energy Conv. & Manage. 2012; 55:116-126 

[29] O'Connor, W.K., Dahlin, D.C., Rush, G.E., Gerdemann, S.J., Penner, L.R., Nilsen, R.P., Aqueous 

mineral carbonation: Mineral availability, pretreatment, reaction parametrics, and process studies, 

DOE/ARC-TR-04-002, Albany Research Center, Albany, (OR) USA (2005) 

[30] Gerdemann, S.J., O’Connor, W.K., Dahlin, D.C., Penner, L.R., Rush, H. 2007. Ex Situ Aqueous Mineral 

Carbonation Environ. Sci. Technol. 2007;41:2587-2593 

[31] Björklöf, T., Zevenhoven, R. Energy efficiency analysis of CO2 mineral sequestration in magnesium 

silicate rock using electrochemical steps. Chem. Eng. Res. & Des. 2012; accepted / in press, available 

on-line: doi: 10.1016/j.cherd.2012.02.001 

101

[32] Zahra, A. Carbon dioxide capture from flue gas: Development and evaluation of existing and novel 

process concepts [PhD Thesis] Delft, the Netherlands: Delft University of Technology2009. 

[33] Khoo, H.H., Sharatt, P.N., Bu, J., Borgna, A., Yeo, T.Y., Highfield, J., Björklöf, T.G., Zevenhoven, R. 

Carbon capture and mineralization in Singapore: preliminary environmental impacts and costs via LCA. 

Ind. & Eng. Chem Res. 2011; 50:11350 - 11357 

[34] Nuclear plant units Olkiluoto 1 and Olkiluoto 2 (in Finnish: Ydinvoimalaitos yksiköt Olkiluoto 1 ja 

Olkiluoto2) Available at: [Accessed: 

2.2.2012] 

[35] Mäkelä, M. Storing of carbon dioxide by mineral carbonation in Southern Finland (in Finnish; 

Hiilidioksidin sitominen mineraalikarbonaatiolla Etelä-Suomessa) [MSc thesis]. Turku, Finland: 

University of Turku / Geology and Mineralogy; 2011. 

[36] Hitura - Nickel Database Available at: [Accessed: 25.1.2012] 

[37] Stormi - Nickel Database Available at: [Accessed: 25.1.2012] 

[38] Belvedere Resources Ltd, A European Nickel Producer Available at: [Accessed: 2.2.2012] 

[39] Hamalainen, Arja,. The Postjotnian diabases of Satakunta. (in Finnish, English summary): Satakunnan 

Postjotuniset Diabaasit) Geological Survey of Finland, Report of Investigation 1987; 76:173-178. 

Available at: < http://arkisto.gtk.fi/tr/tr76/tr76_pages_173_178.pdf > 

[40] Norilsk Nickel Harjavalta (in Finnish only) Available at: [Accessed: 2.2.2012] 

[41] Nickel in Finland. Available at: [Accessed: 25.1.2012] 

[42] Anthony, E. J., and Granatstein, D. L., Sulfation phenomena in fluidized bed combustion systems, 

Progress in Energy and Combustion Science, 2001;27(2): 215-236. 

[43] Gupta, H., and Fan, L., Carbonation-Calcination Cycle Using High Reactivity Calcium Oxide for 

Carbon Dioxide Separation from Flue Gas, Ind Eng Chem Res, 2002;41(16): 4035-4042. 

[44] Hartman, M., and Svoboda, K., Physical properties of magnesite calcines and their reactivity with sulfur 

dioxide, Ind. Eng. Chem. Proc. Des. Dev., 1985; 24(3):613-621. 

[45] Han, K. K., et al., Efficient MgO-based mesoporous CO2 trapper and its performance at high 

temperature, J. Hazard. Mater., 2012; 203–204: 341-347. 

[46] Snow, M. J. H., Longwell, J. P., and Sarofim, A. F., Direct sulfation of calcium carbonate, Ind Eng 

Chem Res, 1988:27(2): 268-273. 

[47] Elfving, P., Panas, I., and Lindqvist, O., In situ IR study on the initial sulphition and carbonation of 

Ca(OH)2 and CaO by SO2 polluted air, Atmos. Environ., 1996:30(23): 4085-4089. 

[48] Chen, H., and Zhao, C., Development of a CaO-based sorbent with improved cyclic stability for CO2 

capture in pressurized carbonation, Chem. Eng. J., 2011;171(1): 197-205. 

[49] Lackner, K. S., Butt, D. P., and Wendt, C. H., Progress on binding CO2 in mineral substrates, Energy 

Convers. Mgmt., 1997;38(Supplement 1): 259-264. 

[50] Stanmore, B. R., and Gilot, P., Review — calcination and carbonation of limestone during thermal 

cycling for CO2 sequestration, Fuel Process Technol, 2005; 86(16): 1707-1743. 

[51] Fagerlund, J., et al., Gasometric Determination of CO2 Released from Carbonate Materials, J. Chem. 

Educ., 2010:87(12): 1372-1376. 

[52] Highfield, J., et al., The promoter effect of steam in gas-solid CO2 mineralisation, Proc. ICCDU-XI, 

Dijon, France, June 2011. 

[53] Lu, H., and Smirniotis, P. G., Calcium Oxide Doped Sorbents for CO2 Uptake in the Presence of SO2 at 

High Temperatures, Ind Eng Chem Res, 2009;48(11):5454-5459. 

[54] Wang, C., Jia, L., and Tan, Y., 2011, Simultaneous Carbonation and Sulfation of CaO in Oxy-Fuel CFB 

Combustion, Chem. Eng. Technol., 2011;34(10): 1685-1690. 

102

Abstract: 




Carbon dioxide storage by mineralisation applied 

to a lime kiln 

Inês Romão a,b , Matias Eriksson c,d , Experience Nduagu a , Johan Fagerlund a , 

Licínio M. Gando-Ferreira b and Ron Zevenhoven a 

a Åbo Akademi University, Dept. of Chemical Engineering, Åbo / Turku, Finland, 

iromao@abo.fi (CA) 

b University of Coimbra, Dept. of Chemical Engineering, Coimbra, Portugal 

c Nordkalk Corporation, Pargas / Parainen, Finland 

d Umeå University, Sweden 

This paper describes a design, for a pilot-scale application, of a two-staged process that is under study at 

Åbo Akademi University (ÅA), for Carbon dioxide Storage by Mineralisation (CSM). The ÅA route implies the 

production of brucite (besides Ca- and Fe- based by-products) from a magnesium/calcium silicate rock, 

using recoverable ammonium sulphate (AS), followed by carbonation of the Mg(OH)2 in a pressurised 

fluidised bed at ~ 500°C, 20-30 bar CO2 partial pressure. An assessment is reported for operating the CSM 

process on waste heat from a limekiln (lime production: 210 t/day) in Pargas, Southwest Finland, i.e. without 

external energy input apart from what is needed for crushing the rock to the required particle size (a few % of 

the overall CSM process energy requirement) and compressing the flue gas to be treated. Part of the off-gas 

from the limekiln (CO2 content ~21%-vol) will be processed without a CO2 separation step. The feature of 

operating without CO2 separation makes CSM an attractive and cost-competitive option when compared to 

conventional CCS involving underground storage of CO2. An exergy analysis is used to optimise process 

layout and energy efficiency, and at the same time maximise the amount of CO2 that can be bound to 

MgCO3 given the amount of waste heat available from the kiln. Also, experimental results are reported for 

producing Mg(OH)2 (and Fe,Ca(OH)2) from local rock material. 

Keywords: 

CO2 mineral sequestration, Scale-up, Lime kiln. 


1.1 CO2 mineralisation 

For Finland, carbon dioxide storage by mineralisation (CSM) was identified as the only option for 

CCS (carbon dioxide capture and storage) application. It is a permanent storage option and has an 

estimated storage potential that is much larger than underground storage as pressurized CO2 [1-2], 

the currently most intensively studied option for CO2 disposal. The purpose of CSM is to promote 

CO2 fixation by metal oxides into thermodynamically stable carbonates while benefiting of the 

exothermicity of the carbonation reaction: 

(Ca,Mg)*ySiO2*zH2O (s) + CO2 (g) → (Ca,Mg)CO3 (s) + ySiO2 (s) + z H2O (g) + HEAT (R1) 

Magnesium in particular is abundant in the earth’s crust, as silicates such as serpentinite and 

olivine. These reactions occur in nature over geological timescales (hundreds of thousands of 

years). Research has focused on improving the reaction rates by treating the mineral rock by 

thermal, mechanical or chemical means [1-4]. Due to the exceptionally large scale of CCS 

processes, all additives must be recovered, and the energy input minimised. 

103

A process under development at ÅA uses recoverable ammonium sulphate (AS) salt to extract Mg 

from grinded serpentinite rock, at elevated temperatures. The extraction has shown conversions of 

up to 70 % of Mg into either reactive Mg(OH)2 or MgSO4, depending on the desired intermediate. 

Mg(OH)2 reacts directly with CO2 under elevated temperature in a pressurised fluidised bed. The 

ÅA process may also be used as a scheme to capture the CO2 directly from a flue gas stream. The 

direct mineralisation of flue gas instead of separated and compressed CO2, eliminates the need of 

expensive and energy intensive processes to isolate and compress CO2, thus significantly lowering 

the materials and energy requirements for the overall CCS process chain. Besides, the simultaneous 

CO2 separation and capture avoids the main risks associated with geological storage (potential 

leakage of CO2 into the atmosphere) and the costs associated with monitoring [5]. Hence, CSM is a 

promising, safe and permanent CO2 fixation route and for that reason has been studied at ÅA since 

2006. 

In addition, metal and mineral processing and papermaking sectors (but not the power sector) have 

shown much interest in using CSM for CO2 emissions mitigation. The prospect of simultaneously 

making use of and (in some cases) upgrading/stabilising process’ by-products and waste materials 

(ashes e.g) is another interesting benefit [6,7]. 

This paper explores the possibility of running the ÅA CSM process on waste heat provided by a 

limekiln (lime production: 210 t/day) in Pargas, Southwest Finland (Nordkalk Corporation) and 

assess the performance of a pilot plant, with direct mineralisation of flue gases. Nordkalk is one of 

the European leading producers of lime emitting a total of 0.79 MtCO2/a [8]. Along with the 

limestone mining comes significant amounts of diopside. The use/upgrading of this by-product by 

CSM is discussed at the end of this paper. 

The serpentinite rock material, used in this particular simulation, comes from a nickel mine located 

in Hitura, Central Finland, 500 km from Pargas. This mine has significant resources of magnesium 

silicate rocks (

HEAT 

Fig. 1. Scheme and main reactions of the magnesium silicates’ carbonation process. Adapted from 

[15] 

2. Aspen Plus ® simulations – Mineralisation of serpentine with 

flue gas 

2.1. Aspen Plus ® model – Hitura’s (Finland) serpentine 

The ÅA CSM process was simulated using Aspen Plus® Software. The model follows the same 

scheme as the ones presented in earlier publications [12,16-17] but differs on the CO2 inlet stream 

of the carbonation step and on the source of heat supplied to the endothermic stages. 

The S-DECOMP block represented in the Aspen Plus model (Figure 3) simulates the Solid/Solid 

extraction reactor. It operates at 440ºC and atmospheric pressure. Experimental results, on 

Portuguese samples, show that the presence of water greatly enhances the rate of magnesium 

extraction and lowers the temperature by ≈50°C compared to reacting dry matter. [16-17]. 

In this block, AS is mixed with Hitura’s serpentinite (86% Mg3Si2O5(OH)4, 13% FeO, 1% CaSiO3) 

and water at a ratio of 3:2:1. In the simulations 80% of Magnesium 1 , 60% of iron and 100% of 

calcium are assumed to be extracted. 

It is expected that, under the referred operating conditions, the following reactions occur: 

Mg3Si2O5(OH)4 + 3 (NH4)2SO4 = 3 MgSO4 + 2 SiO2 + 6 NH3 (g) + 5 H2O (g) (R1) 

FeO + (NH4)2SO4 = FeSO4 + 2 NH3 (g) + H2O (g) (R2) 

1 Although experimental results with Hitura’s samples so far had a maximum of 74% of Mg extraction. 

105

CaSiO3 + (NH4)2SO4 = CaSO4 + SiO2 + 2 NH3 (g) + H2O (g) (R3) 

(NH4)2SO4 (s) → 2NH3(g) + SO3 (g) + H2O(g) (R4) 

As the large amounts of NH3 and water released are cooled, the NH3 dissolves in the water to 

produce a 30-40% solution of NH4OH (Block TANK 1). A small make-up stream of NH3 or 

NH4OH will be necessary due to possible losses in the overall process. The appropriate temperature 

for this vessel depends on the amount of water present in the gases and, in this case, it was 

estimated to be ~40°C (roughly the boiling temperature of an aqueous solution with ≈30% NH3). 

The NH4OH produced in this vessel is later used to cautiously raise the pH in both precipitation 

steps (blocks PPI and PPII). After cooling, the hot solid from the S-DECOMP block is dissolved in 

water (block DISSOLUT). Unreacted serpentine, iron oxide, wollastonite and insoluble reaction 

products (silica, e.g) settle and are separated, most likely by a filtration process, forming the 

RESIDUE. In the aqueous solution at ~80⁰C, the XSO4 salts formed in the first step are converted to 

X 2+ species. 

For the sequential precipitations, pH is the most important operation parameter. The solution must 

be rigorously kept at pH=9-9,5 (block PPI) so that only iron and calcium are precipitated and, 

consequentially, magnesium recovery is maximised in the second precipitation stage (block PPII) at 

a pH of ~9,5-11,5. This pH may slightly vary depending on the mineral and reaction conditions. 

The iron is precipitated in the form of FeOOH (goethite), (although in the simulation Fe(OH)2 is 

used due to some database problems in Aspen Plus®) calcium as Ca(OH)2 and magnesium as 

Mg(OH)2. The NH3 present in the gases from the Solid/Solid Reactor may not be enough to 

increase the pH of the second precipitation stage to 11,5. For that reason it may be necessary to add 

NH3 to the block PPII. 

Although AS is a cheap chemical (cheaper than ammonia or sulphuric acid) its recovery is 

compulsory, not only for environmental reasons but also to make the process economically viable. 

After being separated from the precipitated Mg(OH)2, the AS is recovered from the aqueous 

solution through a concentration/crystallisation process (block EVAPORAT). The recovered wet 

AS is then fed back to S-DECOMP reactor, thereby closing the AS cycle and process loop. 

The Mg(OH)2 is directly carbonated with flue gas from the limekiln, in a pressurised fluidised bed 

(block CARBONAT). In order to achieve a CO2 partial pressure of ~ 20 bar, the flue gas, with 

~21%-vol CO2, must be compressed to at least 80 bar. The block MSCOMPRESS (multistage 

compressor) represents a series of 6 polytropic compressors with 80% of efficiency. Mg(OH)2 

reacts with gaseous CO2 to form MgCO3 and water vapour at 500°C and a total pressure of 80 bar. 

The stream leaving the CARBONAT reactor contains MgCO3(s), H2O(g), O2, N2 and unreacted 

Mg(OH)2(s) and CO2(g). The gases are separated from the solids (most likely) by a cyclone. The 

MgCO3/Mg(OH)2 solid product mix may be re-circulated until the Mg(OH)2 content becomes 

sufficiently low (few %). Decompressing and cooling the gaseous stream leaving the separation unit 

allows for the recovery of some of the energy input for the initial flue gas compression. This stream 

contains water vapour, produced in the carbonation reaction. 

Mg(OH)2 (s) + CO2 (g) → MgCO3 (s) + H2O(g) (R4) 

In order to avoid the presence of a liquid phase inside the isentropic expansion turbine, the 

decompression cannot go under ~5.6 bar. The rest of stream’s exergy/heat content is recovered in a 

heat exchanger. The main assumptions taken to design the Aspen model are summarized in Table 1. 

In this simulation, the heat required for this process is provided by ~10.3 m 

106 

3 n [7] of flue gases 

coming from the limekiln located in Pargas. Currently, the heat content of those gases is utilised to 

supply district heating for the city of Pargas. In the Aspen Plus® model the flue gas is defined as 

two different utilities. Utility FG500 is the flue gas leaving the limekiln at 500⁰C and is used to 

provide heat for the Mg, Ca and Fe elements extraction and to pre-heat the CO2 rich gas entering 

the carbonator. After cooling to 440⁰C, the flue gas FG440 is then used to provide heat for the

(raw) materials pre-heating and AS recovery. The specifications for these two utilities are listed in 

Table 2. 

Table 1. Main assumptions taken in the Aspen Plus ® model 

Block Assumption 

S-SDECOM 

Operating conditions: T= 440ºC, P=1bar 

AS/S/W=3:2:1 

Extraction %: Mg = 80%, Fe = 60%, Ca = 100% 

DISSOLUT T=80ºC; Total conversion of XSO4 species to X 2+ 

TANK 1 

PPI 

PPII 

CRYST 

CARBONAT 

T=40ºC; Complete dissolution of NH3 into water producing a 

30~40% solution of NH4OH 

Total precipitation of Fe and Ca as Fe(OH)2 and Ca(OH)2 

pH =9-9.5; T=30ºC 

Precipitation of Fe and Ca as Fe(OH)2 and Ca(OH)2 

pH =9.5-11.5; T=30ºC 

Water is evaporated so that the recovered product has 75% -w/w of 

AS and 25 %-w/w of water. 

Carbonation with Flue Gas from the limekiln 

pCO2~=20 bar, P=80 bar 

90% of Mg(OH)2 carbonation according to reaction R4 

MSCOMPRE 6 polytropic compressors with 80% efficiency and intercoolers 

TURBINE Isentropic; Decompression to ~5.6 bar 

Table 2. Data for the limekiln’s flue gas defined as a utility in the Aspen Plus ® model 

Utility ID 

Composition 

(mol %) 

107 

TIN 

(°C) 

TOUT 

(°C) 

FG500 

H2O 5.9 

CO2 21.7 

500 460 

FG440 

O2 6.9 

440 350 

N2 65.5 

2.2. Results – Mass and exergy 

The aim of this simulation is to evaluate the performance of a pilot scale mineral carbonation plant 

running on waste heat from the limekiln. In order to process 600 kg/hr of flue gas (7 m 3 at 500 ⁰C 

and 80 bar) i.e, ~190 kg of CO2, 500 kg/hr of serpentinite and 750 kg/hr (later recovered) of AS are 

required. The amounts of residue produced are quite significant, 298 kg/hr. However, this material 

is very rich in SiO2 (>70%) making its future processing for Si recovery an auspicious possibility. 

Ideally, the ~36 kg/hr of iron and calcium hydroxide products are redirected to the steelmaking 

industry. Although this appears to be a modest quantity of by-product, note that, at a larger scale, 

the mineralisation of, e.g., all the CO2 emissions of a single steel company operating in Finland 

(Ruukki), is enough to replace up to 18% of iron ore raw material with FeOOH [12]. 

Out of the 500 kg/hr of processed serpentinite it is possible to produce 275.3 kg/hr of Mg(OH)2. 

Assuming a 90% carbonation efficiency, 275 kg/hr of MgCO3/Mg(OH)2, with a content of 90% 

MgCO3, are produced. Figure 2 presents a schematic diagram of the process including mass balance 

results.

750 kg (NH 4) 2SO 4 

250 kg H 2O 

550 kg SERPENTINE 

83% Mg 3Si 2O 5(OH) 4 

13% FeO 

1% CaS iO 3 

SOLID/SOLID 

DECOMPOSITION 

440ºC 

8.32 m 3 n/s at 500ºC 

3.1 m 3 n/s at 440ºC 

900 kg H 2O 

FLUE GAS USED FOR CO 2 CAPTURE: 

600kg/hr, 0.13 m 3 n/s 

CO 2 CAPTURED 185 kg/hr 

2.97 kg Serp/kg CO 2 

TOTAL FLUE GAS USED AS AN UTILITY: ~9.4 m 3 n/hr 

GAS 

191 kg NH 3 

407 kg H 2O 

43 kg SO 3 

SOLIDS 

561 kg MgSO 4 

60 kg FeSO 4 

1 kg CaSO 4 

211 kg SiO 2 

8 kg (NH 4) 2SO 4 

90 kg Unr. Serp 

Mg 2+ Fe 2+ Ca 2+ 

NH 4 + SO4 - 

RESIDUE 

211 kg SiO 2 (70%) 

300 kg insoluble material 

Fig. 2. Mass results for 80% magnesium extraction and 90% Mg(OH)2 carbonation using Hitura 

nickel mine serpentinite rock (kg values are for one hour operation). 

3. Mineralisation of diopside with flue gas 

This section assesses the applicability of the ÅA process using the in-site available diopside as the 

metal oxide source, instead of nickel mine tailing from Hitura (500 km from Pargas). The Aspen 

Plus® model presented in section 2 simulates a case where serpentinite rock is used as the mineral 

source for Mg. An alternative would be to use a diopside material that is produced by Nordkalk as 

by-product from its limestone quarry. Therefore the reactivity of this material was assessed by the 

performing of some experimental tests using the same experimental procedure developed by 

Nduagu [9], before embarking on furthers Aspens Plus® simulations. 

3.1. Diopside characterisation and experimental work 

A sample of diopside provided by the Nordkalk’s facility located in Pargas, was ground to obtain a 

size distribution of 125-250 μm and analysed using X-ray Diffraction (XRD) and X-ray 

Fluorescence (XRF). The results are reported in Table 4. 

NH 4 + 

H 2O 

SO 4 - 

109 

Mg 2+ NH 4 + 

SO 4 - 

PPI 

0.53 kg 

Ca(OH) 2 

36 kg Fe(OH) 2 

0.1 kg 

NH 3 

NH 4 + SO4 - 

PPII 

272 kg 

Mg(OH) 2 

~900 kg 

H 2O 

80ºC pH=7~8 

pH=10~11 103C 

FLUE GAS (DRY) 

600 kg (0.13m 3 n/s) 

23.5% CO 2, 

74.5% N 2 , 25% O 2 

0.2 m 3 n/s 

at 440ºC 

1.1 m 3 n/s at 500ºC 

PFB 

500ºC 

80 bar 

EVAP 

3.4 m 3 n/s at 440ºC 

AS RECOVERED 

375 kg (NH 4) 2SO 4 (s) 

375 kg (NH 4) 2SO 4 (aq) 

250 kg H 2O 

490 kg GAS 

15.4% H 2O, 74.3% N 2 

1.0% CO 2, 9.1% O 2 

381 kg MgCO 3

Fig 3- Aspen Plus® model 

110

Table 3. Summary of the process’ exergy. 

EQUIPMENT 

Energy 

Q (J/s) 

T 

(°C) 

Exergy as heat 

(MJ/hr) MJ/kgCO2 

SS DECOMPOSITION 463146 450 1003 5.44 

DISSOLUTION -52259 80 -35 -0.19 

PPI -52455 30 -9 -0.05 

PP2 -14218 30 -3 -0.01 

TANK 1 -82000 40 -137 -0.74 

EVAPORAT 630066 103 532 2.88 

CARBONATOR -5500 500 -12 -0.07 

HX-1 (S+AS+WT pre-heat) 434378 430 923 5.01 

HX-2 (FG pre-heat) 65050 450 141 0.76 

HX-3 (Mg(OH)2 pre-heat ) 27517 260 46 0.25 

HX-4 (Gaseous Products ) -472663 440 -1014 -5.50 

HX-5 (solid Products) -93161 440 -200 -1.08 

HX-6 (wt from AS recovery) -612765 103 -517 -2.80 

HX-7 (gas from turbine) -73507 187 -99 -0.54 

Utilities Multistage Compressor 

(Water [15–90] °C) 

stage 1 -30105 90 -22 -0.12 

Total Exergy Process 2 

Exergy provided by the FG 3 

Exergy Process + FG 4 

stage 2 -23328 90 -17 -0.09 

stage 3 -23512 90 -18 -0.09 

stage 4 -23878 90 -18 -0.10 

stage 5 -24599 90 -18 -0.10 

111 

Flue Gas (kg/hr) Flue Gas NTP (m 3 n/s) 

FG500 FG440 FG500 FG440 

37442 

4992 

11235 

13723 

525 2.85 Total 42434 24958 9.42 6.65 

2645 14.34 

-2120 -11.49 

2 Total exergy as heat (120~450⁰C) needed to run the process, without integration with flue gas. 3 Exergy that the flue gas provides to the endothermic blocks. 4 Total exergy of the process when 

flue gas provides heat to all the endothermic stages. This final value may lead to the erroneous conclusion that the heat content of the flue gas is enough to process more serpentine. This “surplus” of 

exergy comes from streams at T ≤ 400⁰C making its heat content unsuitable for the processing of rock material at ~440⁰C. On the other hand this heat is appropriate for district heating. 

891 

8.32 


3.40 

3.08 

0.20

Table 4. Elemental and XRD analysis of the diopside sample. 

Elemental Analysis (%) Structural Analysis 

(XRD) 

CaO SiO2 TiO2 Al2O3 Fe2O3 MgO K2O Na2O Others 

15.6 50.9 0.4 12.2 4.5 4.9 3.1 2.1 6.3 

112 

Hedenbergite, 

Orthoclase, Albite 

(Calcium), Muscovite, 

Clinochlore 

A total of twelve experiments were done in order to determine the reactivity of the rock. The 

diopside (D) was mixed with solid ammonium sulphate (D:AS=2g:3g) and placed in an oven at 

different temperatures, 250-500°C, for 30 minutes. The products of this extraction step were 

dissolved in water and the concentrations of magnesium, iron, aluminium, calcium, sodium, 

potassium and sulphur in the solution were measured by an ICP-OES analysis. In six of the 

experiments, the influence of water was studied by adding 3 ml of water to the diopside/AS mixture 

(D:AS:W=2g:3g:1ml). 

3.2 Experimental results 

À priori, the composition of the diopside does not seem to be very suitable for direct application of 

the ÅA procedure mainly due to its dramatically low content of MgO, but also due to its high 

percentage of aluminium and other alkaline elements. In earlier experiments [16] the increase of 

aluminium content did not appear to favour the extraction reactions. 

Disappointingly, also the extraction results are very discouraging. Besides its low Mg content, the 

diopside’s reactivity is extremely low making the upgrading of this material doubtful. Water does 

not appear to play a key role on the materials reactivity. In fact, contrary to what occurs with the 

serpentine minerals [17], the evolution of the extraction rates with temperature is quite similar both 

in the presence and absence of water. 

Even if it was possible to extract all the X elements of the rock material, the applicability of the ÅA 

route presents several challenges. The extraction of all the magnesium is energy consuming due to 

the presence (in greater percentages) of other elements (mainly Al, Fe, Na, K, and Ca). In fact, a 

quick calculation allows for concluding that only 1/4 of the overall reactions’ heat is used for Mg 

extraction. A successful extraction of Ca would imply a second carbonation stage (probably in an 

aqueous solution). And finally, the presence of alkali metals (which are highly water soluble and do 

not precipitate) would make the AS recovery very difficult implying the application of extra 

separation stages to preclude the recirculation/accumulation of those unwanted elements. In Table 5 

a brief comparison of the pros and cons of using serpentinite vs diopside for CO2 mineralisation is 

presented. As these results show that the diopside seems not to be a suitable material, no Aspen Plus 

simulations were made for that. 

Extraction % 

12 

10 

8 

6 

4 

2 

DRY Extraction 

0 

200 250 300 350 400 450 500 550 

Temperature ( C) 

Al 

Fe 

Mg 

Ca 

K 

Na 

0 

200 250 300 350 400 450 500 550 

Temperature ( C) 

Fig.4 Experimental results concerning the extraction Fig.5 Experimental results concerning the 

Extraction % 

12 

10 

8 

6 

4 

2 

WET Extraction 

Al 

Fe 

Mg 

Ca 

K 

Na

of six X elements from the diopside material in 

absence of water. 

113 

extraction of six X elements from the diopside 

material in presence of water. 

Table 5 – Pros and cons of utilising serpentinite or diopside as raw materials in the ÅA CSM 

process. 

Availability Must be purchased 

(500 km away) 

Extraction 

temperature 

Reactivity 

Iron/Calcium/ 

Aluminium byproduct 

Mg(OH)2 

product 

Serpentinite Diopside 

Available at the plant site 

~400 – 440 °C ≥ 500°C making the use of waste heat 

questionable. 

Good reactivity 

(So far ~60 – 70 % of Mg 

extraction but with good 

potential for 80%) 

Low content of Al 

hydroxides 

Suitable for the steelmaking 

industry. 

~0.5kgMg(OH)2/kg of 

serpentinite processed 

(89% Mg extraction) 

Crushing energy is 

~12% of the exergy input 

(as electricity): 

0.19MJ/kgCO2 captured 

AS recovery Final solution contains 

NH4 + , SO4 2- and water 

Very low reactivity (

the ÅA CSM route. The content of Mg and Ca in the material make it unsuitable for CO2 

mineralisation in general because the large amounts of material needed will give rise to excessive 

crushing and grinding costs. 

The energy penalty on the district heating supply, which arises from CO2 capture, may be reduced 

by running this during low demand hours, for example during the summer. The same integration 

concept may be applied to other industries, making use of steam, at the right temperature, from a 

steam cycle of a power plant or some other nearby process which could supply most of the heat 

needed for CSM. 


The authors want to acknowledge Cleen Ltd. and Tekes (the Finnish Funding Agency for 

Technology and Innovation) for their financial support for the research via the Cleen 

CCSP project (2011-2015) 

Nomenclature 

AS – Ammonium Sulphate 

CCS – Carbon Capture and Storage 

CSM – Carbon storage by Mineralisation 

D – Diopside 

E0 – Energy required for pulverisation [kWh/t] 

F – Size of feed particles [μm] 

MC – Mineral Carbonation 

PFB – Pressurised Fluidised Bed (reactor) 

P – Size of the pulverized particles [μm] 

W – Water 

Wi– work index [kWh/t] 

X – Mg, Ca, Fe, Al, Na, K 

References 

[1] IPCC Special Report on Carbon Dioxide Capture and Storage B. Metz, O. Davidson, H. de 

Coninck, M. Loos, L. Meyer, Working Group III of the IPCC, Cambridge Univ. Press (2005) 

Available at: 

[2] Lackner, K.S., A guide to CO2 sequestration, Science Vol 300, 1677-1678, 13 June 2003 

[3] Zevenhoven, R., Fagerlund, J., Songok, J.K. CO2 mineral sequestration - developments towards 

large-scale application. Greenhouse Gases: Science and Technology. 2011; 1:48-57 

[4] Sipilä, J., Teir, S., Zevenhoven, R. Carbon dioxide sequestration by mineral carbonation – 

Literature Review Update 2005-2007. Åbo Akademi Univ., Heat Engineering Lab. report VT 

2008-1, Turku, Finland (2008). Available at: 

[5] Zevenhoven, R., Fagerlund, J., Songok, J.K. CO2 mineral sequestration - developments towards 

large-scale application. Greenhouse Gases: Science and Technology. 2011; 1:48-57 

[6] Teir, S. Fixation of carbon dioxide by producing carbonations from minerals and steelmaking 

slags. PhD thesis, Helsinki Univerisity of Technology, Espoo Finland, 2008 

[7] Eloneva, S., Teir, S., Revitzer, H., Salminen, J., Said, A., Fogelholm, C.-J., Zevenhoven, R. 

Reduction of CO2 emissions from steel plants by using steelmaking slags for production of 

marketable calcium carbonate. Steel Research International 2009; 80:415-421 

114

[8] Confidential Report on Lime Production, Nordkalk, May 2011 

[9] Nduagu, E. Mineral carbonation: preparation of magnesium hydroxide [Mg(OH)2] from 

serpentinite rock. M.Sc. (Eng.) Thesis, Åbo Akademi University, Finland, 2008 

[10] Nduagu, E., Björklöf, T., Fagerlund, J., Wärnå, J., Geerlings, H., Zevenhoven, R. Production of 

reactive magnesium from magnesium silicate for the purpose of CO2 mineralization. Part 1. 

Application to Finnish serpentinite. Minerals Engineering, 2012; 30:75-86 

[11] Nduagu, E., Björklöf, T., Fagerlund, J., Mäkelä, E., Salonen, J., Geerlings, H., Zevenhoven, R. 

Production of reactive magnesium from magnesium silicate for the purpose of CO2 

mineralization. Part 2. Mg extraction modeling and application to different Mg silicate rocks. 

Minerals Engineering, 2012; 30:87-94 

[12] Fagerlund, J., Zevenhoven, R. An experimental study of Mg(OH)2 carbonation. Int. J. of 

Greenhouse Gas Control 2011; 5:1406-1412 

[13] Fagerlund, J., Carbonation of Mg(OH)2 in a pressurised fluidised bed for CO2 sequestration 

PhD thesis, Åbo Akademi University, Turku Finland (2012) Available at: 

 

[14] Fagerlund, J., Zevenhoven, R. The effect of SO2 on CO2 mineral sequestration applied directly 

to a flue gas. Submitted to ECOS2012, Perugia, Italy, June 2012; paper under review 

[15] Romão, I., Nduagu, E., Fagerlund, J., M. Gando-Ferreira, L., Zevenhoven, R. CO2 Fixation 

Using Magnesium Silicate Minerals. Part 2: Energy Efficiency and Integration with Iron-and 

Steelmaking. Energy, 2012; 41:203-211 

[16] Romão, I., Fagerlund, J., Gando-Ferreira, L. M., Zevenhoven, R. CO2 Sequestration with 

Portuguese serpentine. Proceedings of 3rd International Conference on Accelerated 

Carbonation for Environmental and Materials Engineering – ACEME10, 29 Nov - 1 Dec, 2010, 

Turku, Finland; 77-87; Ed. Ron Zevenhoven 

[17] Romão, I., Gando-Ferreira, L. M, Morais, I., Silva, M.V.G., Fagerlund, J., Zevenhoven, R., 

CO2 sequestration with Portuguese serpentinite and metaperidotite. 11th International 

Conference on Energy for a clean environment – CLEAN AIR 2011, 5-8th July 

[18] Bond, F:C:, The third Theory Communition, Trans AIME, 1952; 193: 484-494 

115

Abstract: 




Comparison of IGCC and CFB cogeneration 

plants equipped with CO2 removal 

Marcin Liszka a , Tomasz Malik b , Micha Budnik c , Andrzej Zibik d 

a Institute of Thermal Technology, Silesian University of Technology, 44-100 Gliwice, Konarskiego 22, 

Poland, marcin.liszka@polsl.pl, 

b Institute of Thermal Technology, Silesian University of Technology, 44-100 Gliwice, Konarskiego 22, 

Poland, tomasz.malik@polsl.pl, CA 

c Institute of Thermal Technology, Silesian University of Technology, 44-100 Gliwice, Konarskiego 22, 

Poland, michal.budnik@polsl.pl, 

d Institute of Thermal Technology, Silesian University of Technology, 44-100 Gliwice, Konarskiego 22, 

Poland, andrzej.ziebik@polsl.pl 

The introduction of CO2 removal processes into coal-fired power units causes usually generation of waste 

heat which is not possible to utilize within steam cycle. Normally, the waste heat is rejected to cooling water 

and then to the environment. As the temperature of waste heat carriers is usually moderately high (ca. 80 - 

100C), there is a potential possibility for using them in district heating systems. The main goal of the present 

paper is thus the energy and CO2 emission analysis of large-scale CHP plants equipped with CO2 removal 

and utilizing waste heat generated within the plant. Two case studies have been formulated. First of them is 

dealing with the CFB plant equipped with a tap-backpressure steam turbine and post-combustion chemical 

CO2 absorption. The steam necessary for CO2 solvent (MEA) regeneration is taken from the steam turbine 

exhaust, while district heat is produced mainly in CO2 dehumidifier and CO2 compression train. The second 

case is dealing with an IGCC equipped with the pre-combustion CO2 removal by physical absorption. The 

district heat is then produced using classical final flue gas cooler located in HRSG, syngas cooler, as well as, 

compression trains of ASU air, nitrogen and CO2 product. For both analyzed cases, the peak-load district 

heat production using steam turbine extraction is also possible. Both CFB and IGCC plants have been 

modelled on the Thermoflex software. The reference, CFB-based CHP plant without CO2 removal has also 

been modelled. The district heat production and district water parameters have been fixed for all analyzed 

cases to the same values. The energy utilization factor, exergy efficiency and electricity-to-heat ratio have 

been calculated for both plants as main assessment factors. The methodology of alternative electricity 

production (equivalent power unit) has been involved for calculation of CO2 emissions. The obtained results 

indicates, that IGCC plant has better thermodynamic indicators than CFB-based unit. Moreover, the CO2 

emission considering system interconnections within the electricity production network is negative for both 

the CFB and IGCC plants equipped with CCS. When comparing exergy efficiency, the highest value is 

achieved for the reference CFB plant (without CO2 capture). The decrease of exergy efficiency caused by 

CO2 capture and compression is ca. 8 percentage points, but in case of IGCC CHP plant the exergy 

efficiency plant is only 3 points lower than for the reference system. 

Keywords: 

IGCC, CFB, CHP, CCS, waste heat 


The CO2 removal processes integrated with coal-fired power units cause significant drop of energy 

efficiency and economic profitability of the overall power generation process. The decrease of 

power generation efficiency is externalized usually by increased amount of waste heat rejected to 

the environment. The waste heat coming from the CO2 removal and compression installations is 

often of moderate temperature (ca. 80-100C), and therefore its utilization within the power cycle or 

for external purposes could be possible. On the other hand, the decrease of CO2 emission without 

its removal is also possible. The combined heat and power production (CHP) is a good example 

116

where the thermodynamic integration of processes leads inherently to higher effectiveness, fuel 

saving and therefore decrease of CO2 emission. 

Combining the availability of a moderate-temperature waste heat at the CO2 removal facility with 

inherently high effectiveness of the coal-fired CHP plant, the idea for the CHP system equipped 

with CCS unit and utilizing waste energy for district heat production has been proposed and 

analysed within the current paper. It was expected, that the high-level integration of power cycle 

with CO2 removal unit and district water heat exchangers will the partial recovery of CCS energy 

expenses make possible. 

Within the current paper two different CHP plants incorporating presented idea have been proposed 

and investigated. First is the classical Rankine-based steam unit equipped with back-pressure steam 

turbine, circulated fluidized-bed boiler and CO2 removal unit based on chemical absorption in 

monethanolamine. Second plant configuration is based on IGCC structure equipped with precombustion 

CO2 removal based on physical absorption in Selexol. Both plants have been scaled to 

the same rated district heat production (110 MWt). 

The idea for rather small CHP units equipped with CO2 removal and waste heat recovery is 

relatively new. There is a lot of literature references for CHP systems (non CCS) based on CFB 

boilers, as well as, for large coal-fired power units equipped with post-combustion CCS. The 

problem of waste heat recovery from the chemical CO2 absorption units is discussed e.g. in [1] 

In case of IGCC CHP system, described in literature installations use biomass mostly. Experience 

gained in IGCC CHP plants based on biomass gasification may be however transferred into similar 

plants based on coal. Several IGCC CHP plants do exist, however usually biomass or wastes are 

gasified, there is no plant which uses hard coal as a main feedstock. The SVZ plant in Schwarze 

Pumpe is one of such installations where coal and municipal wastes mix is gasified in BGL (British 

Gas and Lurgi) gasifier [2]. Installation produces electricity and methanol. This installation was 

however unprofitable, which caused that in year 2007 has been closed [2]. Another example of 

IGCC CHP plant is installation in Varnamo in Sweden. This biomass-based unit reaches 6MWe 

electric and 9MWth thermal power, while the energy efficiency in cogeneration mode is 83% [3]. 

Plant in Varnamo was operating since 1996 till 2000, then it was closed for the same reason as SVZ 

plant [3]. Plant which uses lignite as a main feedstock for IGCC has been built in Versova (Czech 

Republic) [4]. There are 26 Lurgi and 1 Siemens gasifier in operation. Electricity and liquid fuels 

are the main products of this installation, however hot flue gas is used for district heat production 

[4]. In all recognized IGGC CHP installations district heat is produced in conventionally as for 

combined cycles – using flue gas from the heat recovery steam generator (HRSG) exhaust or steam 

turbine extraction. 

2. Case studies 

Two case studies of CCS-integrated CHP units have been analyzed: 

IGCC CHP plant with waste heat recovery equipped with Selexol-based CO2 absorption, 

CFB CHP plant with waste heat recovery equipped with tap-backpressure steam turbine and 

MEA-based CO2 absorption. 

For comparison purposes the same fuel parameters have been used for both analysed systems. Fuel 

composition has been presented in Table 1. 

Table 1. Fuel composition 

Coal parameters (as received) 

c, carbon - 0,5263 

h, hydrogen - 0,0343 

o, oxygen - 0,1102 

n, nitrogen - 0,0075 

s, sulphur - 0,0104 

117

moisture - 0,2237 

ash - 0,0876 

LHV MJ/kg 20,16 

2.1 IGCC plant 

IGCC plants are usually designed as electricity or chemicals production plants, however they have a 

high potential for district heat production. Present IGCC units (especially with CO2 capture) give an 

opportunity for recovering waste heat at levels suitable for district heat production as well as for 

using classical heat sources e.g. tap-steam turbine or outlet of HRSG. 

Proposed IGCC CHP plant configuration is schematically presented in Fig.1. The system is based 

on that presented in [5] which has been optimized and redesigned towards IGCC CHP plant. It is 

composed of cryogenic air separation unit (ASU), dry-feed gasifier with syngas cooler, water gas 

shift reactors (WGSR), acid gas removal (AGR) unit and finally the combined cycle. 

Fig. 1. Concept of IGCC CHP plant 

The entrained flow dry-feed gasifier has been chosen for syngas production. Its operating pressure 

has been set to 4,238 MPa. The temperature of the syngas leaving the reactor reaches 1639 o C. The 

syngas is cooled down primarily by its partial recirculation and then by the production of high 

pressure steam in a radiant/convection heat exchanger. The temperature of the syngas leaving the 

cooler and entering the scrubber is equal to 335C. The gasification reactor consumes oxygen of 

95% purity. The coal feeding system is based on ASU nitrogen. The syngas treating and 

conditioning line is composed of an inter-cooled WGSR reactor. The syngas temperature after the 

first WGSR reactor has been assumed to 460C while behind the second one to 360C. 

The heat rejected from the syngas during the conditioning processes is recovered to the steam cycle 

and for district heat production. The AGR unit is based on a two-stage Selexol process. The syngas 

stream leaving the second WGSR is cooled down, dehydrated and supplied to the first stage of 

AGR unit, where 99% of H2S is stripped. The separated H2S is then sent to the Claus plant for 

118

conversion to elemental sulphur. The H2S-free gas is supplied to the second stage of AGR where it 

is brought in contact with a low-temperature lean Selexol solvent for CO2 absorption. The reverse 

CO2 desorption process occurs in flash drums operating on 3 different pressure steps. Desorbed 

CO2 is compressed and pumped into the transportation pipeline. 

The hydrogen-rich (ca 90%) syngas leaving the AGR is diluted by ASU nitrogen and used as a gas 

turbine (GT) fuel. An F-class GT has been selected. As already mentioned, the steam turbine 

receives steam from both the HRSG and syngas cooling equipment. All compressors within the 

ASU and AGR islands have been assumed as electric motor – driven machines. Table 2 summarizes 

the assumptions/specifications concerning the input/output for the system. 

Waste heat recovery concept involves final flue gas cooler located in the HRSG, syngas cooler in 

syngas treating and conditioning line prior to AGR unit, as well as, intercoolers in compression 

trains of ASU air, nitrogen and CO2 product. All waste heat recovery exchangers are marked in red 

boxes in Fig.1. 

For further transportation, the CO2 stream is compressed to 13 MPa. 

Table 2. Assumed parameters for IGCC CHP Plant 

Parameter IGCC CHP plant 

Pressure in the reactor MPa 4,238 

Temperature of syngas at scrubber outlet 

119 

O C 180 

Nitrogen to fuel ratio (for fuel transportation), mass basis - 0,175 

Steam to fuel ratio, mass basis - 0,005 

Oxygen to fuel ratio, mass basis - 0,66 

Specific electricity consumption kWh/Mgfuel 67 

Syngas treating and conditioning line 

CO conversion ratio at 1 st WGSR % 63,6 

CO conversion ratio at 2 nd WGSR % 86,2 

H2O to CO ratio at 1 st WGSR - 2,34 

H2O to CO ratio at 2 nd WGSR - 4 

Syngas temperature prior to AGR 

Acid gas removal and CO2 compression 

O C 35 

Effectiveness of H2S removal % 99 

Effectiveness of CO2 removal % 90 

Process steam consumption kg/s 1,81 

CO2 capture miscellaneous auxiliary load kWh/MgCO2 60 

Gas turbine 

Combustor exit temperature 

O C 1300 

Compressor pressure ratio - 15,5 

Steam cycle 

Live (HP) steam temperature 

O C 565 

Live (HP) steam pressure MPa 12,8 

MP steam pressure MPa 3,95

2.2 CFB plant 

For the current moment, one of the closest to commercial application method for CO2 capture in 

classic coal-fed CHP units is post-combustion absorption using MEA or similar solvent. The main 

energy requirement for this type of process is heat demand for desorption of CO2 from the amine 

solution. 

Proposed CHP plant with CO2 capture is equipped with CFB boiler, tap-backpressure steam turbine 

and CO2 capture unit – Fig. 2. CFB boiler produces live steam which has typical parameters for 

modern CFB boilers installed recently in Poland (560C, 16,1MPa). Boiler is equipped with 

economizer, convective and radiant superheater. Steam expands in tap-backpressure steam turbine. 

Steam cycle is equipped with four heat recovery exchangers The heat necessary for the CO2 

removal unit is taken from the steam turbine exhaust as a back-pressure steam. The rest of steam 

flow available at the turbine outlet is supplied to the district water heater which operates in parallel 

with other heaters utilizing waste heat from the CO2 absorption and compression units - district 

water is preheated basically in CO2 dehumidifier and CO2 compression train (inter-coolers). 

Fig. 2. Concept of CFB CHP plant 

Hot flue gas leaving the economiser is primarily cooled down in rotating heat exchanger, where 

combustion air is preheated. Then in electrostatic precipitator fly ash is removed and finally prior to 

CO2 absorption process, flue gas is cooled down to 40C and desulphurized for final required SO2 

concentration (10 ppmv). Absorption unit is composed of absorber and stripper columns. In 

absorber column flue gases are brought in contact with MEA solvent, rich solvent is then injected 

into a stripper column where occurs its regeneration and CO2 is separated. For further 

transportation, the CO2 stream is compressed to 13 MPa. 

For comparison purposes, the typical CFB-based CHP plant without CCS has also been studied. 

The district heat production within the non-CCS CFB CHP plant takes place in two heat exchangers 

connected to steam turbine outlet (base load) and extraction (peak-load). The structure and thermal 

120

parameters within the boiler island and steam cycle are the same expecting the stem turbine outlet, 

where steam pressure is lower than for the CCS case, as the required temperature for district heat 

exchanger is lower than for CO2 desorption process. 

Crucial parameters of both considered CFB CHP plants have been presented in Table 3. 

Table 3. Assumed parameters for CHP Plants with CFB boiler 

Parameter Unit CHP plant without 

CO2 capture 

121 

CHP plant with CO2 

capture 

Boiler type Natural circulation Natural circulation 

Live steam pressure MPa 16,1 16,1 

Live steam temperature C 560 560 

Dry step efficiency of steam turbine (HP, % 

MP) 

91 91 

Dry step efficiency of steam turbine (LP) % 85 85 

Temperature of flue gases behind rotary air 

heater 

C 130 130 

Temperature of preheated air C 260 260 

Furnace exit temperature C 850 850 

Excess air ratio % 20 20 

Steam pressure at steam turbine exhaust MPa 0,084 0,24 

Heat demand for the CO2 absorption unit 

(per 1 kg of CO2 removed) 

kJ - 4000 

CO2 removal effectiveness % - 90 

3. Assessment factors 

For the comparison purposes, each of three analysed units (IGCC with CCS, CFB with CCS and 

CFB reference – without CCS) has been evaluated from the energy, exergy and CO2 emission 

points of view. 

Energy utilization factor has been calculated as defined in (1), representing the proportions of 

energy balance rather than a true thermodynamic assessment. 

EUF 

Q 

N 

E 

dh elN (1) 

chfuel 

Energy analysis does not take into account different quality of products. Exergy is a measure of 

energy quality or thermodynamic irreversibility related to the isolated devices or whole analyzed 

system [6-8]. Exergy analysis is suitable for the thermodynamic comparison of co-product or multiproduct 

systems. The exergy efficiency of each analyzed CHP system has been calculated in 

accordance to (2). 

X 

dh NelN 

X 

(2) 

X 

fuel 

The exergy of produced heat is expressed by increase of exergy of district water as presented in (3) 

where subscript “II” characterize hot water, while subscript “I” characterize cold water. 

h h T 

s s ) 

X dh m 

dh II I a ( II I 

(3)

The CO2 emission factor which indicates the net CO2 emission per 1 GJ of produced heat has also 

been calculated in accordance with (4). Considering dual product situation (heat and electricity) it 

has been assumed that the main product is heat and the emission assigned to heat is calculated as 

difference between total emission from the CHP unit and emission avoided in other power stations 

due to production of specified amount of electricity in considered CHP unit. The scheme for 

avoided emission calculus has been illustrated in Fig. 3. The equivalent power plant is assumed to 

be coal-fired, non-CCS, supercritical unit of net electric efficiency equal to 44%. 

m 

CO 2 m 

CO 2 _avoid 

Q 

 

 

(4) 

dh 

Fig. 3. Scheme for calculus of avoided CO2 emission 

4. Simulation model 

All analyzed CHP systems have been modelled using the Thermoflex software [9]. Thermoflex 

contains models of typical energy conversion devices like compressors, turbines or heat exchangers, 

as well as, agglomerated multi-device models such as steam boiler or gasifier islands. The modeling 

approach combines the physical and empirical modeling. 

The flow sheets of analyzed IGCC, CFB and CFB reference systems used for simulation purposes 

are presented in Figs A.1, A.2 and A.3 in appendix A. 

Main assumptions taken for the simulation process for all considered systems are as follows: 

the same district heat power (ca. 110MWth), 

the same parameters of district water 62,6C/37C (annual average), 

the same feedstock (hard coal), 

CO2 stream compressed to 13 MPa for both CO2 capture installations, 

the same temperature of cooling water at battery limits (20C). 

122

The calculated parameters in characteristic points of the analyzed systems are presented in Tables 

A.1, A.2, A.3 in appendix A. 

5. Results 

The crucial calculated parameters at system boundaries of each analysed CHP unit have been 

presented in Table 3. 

For the same district heat production, obtained net and gross electricity production is the highest for 

IGCC unit. The CO2 emission for both plants equipped with CO2 capture is relatively low. Lack of 

energy consumption for the CO2 capture installation in case of reference CFB plant results in the 

highest value of energy utilization factor achieved. EUF indicator for the IGCC CHP plant is lower 

than for other systems as it does not take into account different quality of products. 

When comparing exergy efficiency, the highest value is achieved for the reference CFB plant 

(without CO2 capture). The decrease of exergy efficiency caused by CO2 capture and compression 

is ca. 8 percentage points, but in case of IGCC CHP plant the exergy efficiency plant is only 3 

points lower than for the reference system. 

Table 3. Results of simulation analysis 

Parameter Unit 

CFB CHP plant 

without CO2 capture 

(reference) 

123 

CFB CHP 

plant with 

CO2 capture 

IGCC CHP 

plant with CO2 

capture 

Total district heat 

production MWth 110,22 110,26 110,14 

including: 

- waste heat MWth 0 53,24 29,22 

- steam-fed heat 

exchangers MWth 110,22 57,02 80,92 

Fuel flow rate kg/s 9,34 10,40 13,02 

Fuel chemical energy 

(LHV-based) MW 188,29 209,66 262,48 

Gross electricity 

production MWe 64,70 64,10 113,30 

Net electricity production MWe 60,10 50,80 78,40 

Electric power 

consumption for CO2 

compressors MW - 6,10 7,00 

CO2 emission from CHP 

plant (total) kg/s 18,46 2,06 2,52 

CO2 mass flow rate for 

transportation (captured) kg/s - 18,49 22,20 

EUF (Eq. 1) % 0,90 0,77 0,72 

Fuel exergy input MW 205,24 228,53 286,11 

District water exergy 

increase (Eq. 3) MW 10,36 10,36 10,36 

Exergy efficiency (Eq. 2) - 0,35 0,27 0,32 

CO2 emission factor per 

GJ of produced district 

heat (Eq. 4) kgCO2/GJt 48,86 -81,59 -131,95

The CO2 emission factor calculated in accordance with (4) is below zero for both cases with CO2 

capture due to effect of coupling CO2 capture and cogeneration effectiveness. Better values has 

been obtained for IGCC CHP plant. Values of emission factor have been achieved for assumed net 

electric efficiency of equivalent power station equal to 44%. For better recognition of the impact of 

equivalent plant efficiency on CO2 emission, sensitivity analysis has been prepared as presented in 

Fig. 4. 

Emission factor (kgCO2/GJ) 

100,0 

50,0 

0,0 

-50,0 

-100,0 

-150,0 

-200,0 

0,34 0,35 0,36 0,37 0,38 0,39 0,40 0,41 0,42 0,43 0,44 0,45 0,46 

el_equiv 

IGCC CHP CHP with CO2 capture CHP without CO2 capture 

Fig. 4. Sensitivity analysis of equivalent power station efficiency on CHP emission factor. 

6. Concluding remarks 

Two configurations of coal-fed CHP plants with massive CO2 removal and waste heat recovery 

systems have been proposed and analysed assuming constant district heat demand. 

Obtained results of exergy analysis indicate, that the drop of efficiency due to CO2 removal and 

compression (ca. 8 % points for CFB CHP) is lower than for large-scale coal-fired power units (ca. 

10-12 % points for similar CCS heat demand as reported in [10,11]), as the waste heat recovery lets 

for partial cancellation of negative impact of CCS on overall plant efficiency. 

An IGCC CHP unit with CO2 capture has significantly better emission factors and thermodynamic 

excellence when comparing to CHP plant with CFB boiler and post-combustion CO2 capture 

process. Advantages of IGCC CHP unit are mainly due to commonly known effect of Brayton and 

Rankine cycles integration, as well as, due to advantage of pre-combustion CO2 removal over postcombustion 

system (referring to assumed in this paper energy demands for both installations). 

Negative value of CO2 emission factor which arises from applied CO2 removal processes and 

substituting of electricity produced in other power stations (system advantage of cogeneration), 

enables potentially for substituting emission from power facilities, where decreasing emission is 

economically ineffective or impossible (e.g. peak-load boilers). 

Future work should be focused on off-design analysis of both proposed CHP systems to evaluate 

change of supply of waste heat as function of variable ambient temperature and district heat 

demand. The comparative economic analysis reflecting the costs of district heat production, as well 

as, costs of avoiding the CO2 emission should also be done. 

124


This work has been prepared in framework of the task of research, "Development of coal 

gasification technology for high efficient production of fuels and electricity" funded by the Polish 

National Centre for Research and Development within the strategic program of research and 

development: "Advanced energy generation technologies". 


Centre of Research and Development in the framework of Contract SP/E/1/67484/10 – Strategic 

Research Program – Advanced technologies for energy generation: Development of a technology 

for highly efficient zero-emission coal-fired power units integrated with CO2 capture. 

Nomenclature 

AGR Acid gas removal 

ASU Air separation unit 

Echfuel chemical energy of coal, calculated on LHV basis, W 

EUF Energy utilization factor 

DHX District heat exchanger 

GT Gas turbine 

h specific enthalpy, kJ/kg 

h a specific enthalpy at ambient parameters, kJ/kg 

HP High pressure 

HRSG Heat recovery steam generation 

IGCC Integrated gasification combined cycle 

LHV Lower heating value 

m mass flow rate, kg/s 

MP Medium pressure 

N net electric power of the system, W 

elN 

N elG gross electric power of the system, W 

Qdh district heat flow rate, W 

r enthalpy of vaporization, kJ/kg 

s specific entropy, kJ/kgK 

sa 

specific entropy at ambient parameters, kJ/kgK 

ST Steam turbine 

ambient temperature, K 

Ta 

WGSR Water gas shift reactor 

X chemical exergy of coal, W 

fuel 

X dh 

exergy of produced district heat, W 

zi 

molar share of i-th compound 

X exergy efficiency 

125

Appendix A 

Fig A.1. Simulation model of IGCC CHP unit with waste heat recovery (Thermoflex) 

126

(a) 

127

(b) 

Fig. A.2. Simulation model of CFB CHP plant with CO2 capture (Thermoflex): a) boiler and power cycle, b) CO2 capture and compression 

installation 

128

Fig. A.3. Simulation model of CFB CHP plant, without CO2 capture (Thermoflex) 

129

Table A.1. Calculated parameters at selected points of the plant structure for IGCC CHP case 

(point numbers correspond to those in Fig. A.1) 

Water/Steam 

T, C p, bar m, kg/s Quality 

Syngas 

Stream 

number 

Stream number 

154 565 128 40,26 Superheated 235,4C 

152 565 39,5 35,02 Superheated 315,4C 

101 321,92 8 39,93 Superheated 151,5C 

123 211,96 3 35,67 Superheated 78,4C 

67 72,53 0,3478 35,67 0,9703 

126 122,8 43,16 8,52 Sub cooled 132,1C 

127 125,38 154,06 40,34 Sub cooled 218,9C 

160 122,06 11,22 4,92 Sub cooled 62,9C 

T, C 

p, 

bar 

m, 

kg/s 

H2 CO2 CO N2 H2O H2S COS Ar Other 

130 

mole fractions 

41 600 41,17 22,74 21,73 6,64 47,80 8,18 15,2 0,38 0,031 0,03 0,009 

47 179,88 40,44 24,44 19,29 5,9 42,44 7,26 24,73 0,33 0,0275 0,02 0,0025 

204 220 38,89 24,44 19,29 5,9 42,44 7,26 24,73 0,33 0,0275 0,02 0,0025 

20 358,39 35,06 35,82 38,31 29,72 1,37 4,67 25,68 0,23 0,0004 0,01 0,0096 

219 149,61 30,24 35,82 38,31 29,72 1,37 4,67 25,68 0,23 0,0004 0,01 0,0096 

29 58,9 28,79 27,79 51,21 39,73 1,83 6,23 0,66 0,31 0,0005 0,02 0,0095 

1 27,22 27,03 5,12 84,5 2,11 3,02 10,29 0,042 0,0005 0,0008 0,03 0,0067 

201 143,73 26,5 17,33 53,3 1,33 1,91 42,59 0,03 0,0003 0,0275 0,03 0,7822 

Table A.2. Calculated parameters at selected points of the plant structure for CHP plant with CO2 

capture case 

(point numbers correspond to those in Fig. A.2a and Fig. A.2b) 

Water/Steam 

Stream number 


16 560 161 81,14 Superheated 212,1C 

20 344,8 39,8 76,54 Superheated 94,8C 

31 287,1 25,3 71,59 Superheated 62,6C 

32 194,6 10,9 63,32 Superheated 10,9C 

1 158,8 6 61,19 0,977 

29 126,1 2,4 35,99 0,936 

33 126,1 2,4 25,09 0,936 

17 245 172,6 81,34 Sub cooled 108,5C 

15 470 162,6 81,14 Superheated 121,3C

Flue Gas 

Stream number 

255 37 20 1030 Sub cooled 175,4C 

256 37 20 42,3 Sub cooled 175,4C 

257 37 20 987,7 Sub cooled 175,4C 

247 49,35 20 1030 Sub cooled 163C 

248 62,6 20 1030 Sub cooled 149,8C 

T, C p, bar m, kg/s mash, 

kg/s 

131 

CO2 N2 H2O SO2 Ar O2 


37 850 1,0427 95,41 1,241 14,44 71,11 10,40 0,007 

38 323,13 1,0353 95,41 1,241 14,44 71,11 10,40 0,007 

0,86 3,18 

0,86 3,18 

40 130,3 1,0328 98,83 1,241 13,93 71,34 10,06 0,007 0,86 3,8 

44 130,3 1,0132 98,83 0,002 13,93 71,34 10,06 0,007 0,86 3,8 

167 35 1,0132 76,65 0 1,71 87,68 4,88 0 1,06 4,67 

218 30 130 18,49 0 99,97 0 0,03 0 0 0 

Table A.3. Calculated parameters at selected points of the plant structure for reference CHP plant 

without CO2 capture case 

(point numbers correspond to those in Fig. A.3) 

Water/Steam 

Stream number 

Flue Gas 

Stream number 


16 560 161 72,7 Superheated 212,1C 

20 344,8 39,8 68,6 Superheated 94,8C 

32 287,1 25,3 64,2 Superheated 62,6C 

33 194,6 10,9 60,5 Superheated 10,9C 

24 158,8 6 53,9 0,977 

31 105,6 1,2 53,7 0,913 

29 95 0,85 53,7 0,901 

41 105,6 1,2 0,2 0,913 

17 245 172,6 72,9 Sub cooled 108,5C 

15 470 162,6 72,7 Superheated 121,3C 

T, C p, bar m, kg/s mash, 

kg/s 

CO2 N2 H2O SO2 Ar O2 


5 850 1,0378 85,73 1,134 14,44 71,1 10,41 0,007 

8 329 1,0303 85,73 1,134 14,44 71,1 10,41 0,007 

0,86 3,18 

0,86 3,18 

66 130 1,0228 94,97 1,134 13,01 71,71 9,49 0,007 0,86 4,92 

18 130 1,0132 94,97 0,002 13,01 71,71 9,49 0,007 0,86 4,92

References 

[1] Pfaff I., Oexmann J., Kather A., Optimised integration of post-combustion CO2 capture process 

in greenfield power plants. Energy 2010;35:4030-4041 

[2] Rakowski J., Przegld zagadnie technologicznych zwizanych ze zgazowaniem paliw staych 

dla potrzeb energetycznych. Energetyka 2003 September (in Polish) 

[3] Stahl K., Neergaard M., IGCC Power Plant for Biomass Utilisation, Värnamo, Sweden, 

Biomass and Bioenergy 1998:15(7):205-211. 

[4] Report from Vresova: 12 years of operating experience with the world's largest coal-fuelled 

IGCC. (PLANT OPERATING EXPERIENCE). Available at: < 

http://goliath.ecnext.com/coms2/gi_0199-9669125/Report-from-Vresova-12-years.html> 

[accessed 1.10.2008] 

[5] Liszka M., Malik T., Manfrida G., Energy and Exergy Analysis of Hydrogen-Oriented Coal 

Gasification with CO2 Capture. Proceedings of the 24th International Conference on Efficiency, 

Cost, Optimization, Simulation, and Environmental Impact of Energy Systems; 2011 Jul 4-7; 

Novi Sad, Serbia. 

[6] Szargut J., Exergy Method. Technical and Ecological Applications, WIT Press, Southampton- 

Boston 2005. 

[7] Uson S., Valero A., Thermoeconomic Diagnosis of Energy Systems, Prensas Universitarias de 

Zaragoza 2010. 

[8] Bejan A., Tsatsaronis G., Moran M., Thermal Design & Optimisation, John Wiley & Sons, 

New York 1996. 

[9] Thermoflow, Inc., 29 Hudson Road , Sudbury, MA 01776 USA, Available 

at: 

[10] Davison J., Performance and costs of power plants with capture and storage of CO2. Energy 

2007;32:1163-1176 

[11] Liszka M., Zibik A., Thermoeconomic comparison of coal-based oxy-fuel and postcombustion 

CO2 capture - case study for Polish conditions. Proceedings of 25th Annual 

International Pittsburgh Coal Conference; 2008 Sep 29 - Oct 2; Pittsburgh, PA, USA. 

132




Concept of a “capture ready” combined heat and 

power plant 

Piotr Lukowicz a , Lukasz Bartela b 

a Silesian University of Technology, Gliwice, Poland, piotr.lukowicz@polsl.pl 

b Silesian University of Technology, Gliwice, Poland, lukasz.bartela@polsl.pl 

Abstract: 

In case of the CHP plants with backpressure turbines and auxiliary condensing units, there is no possibility of 

an additional steam extraction for CO2 chemical sequestration. This is due to the fact that those turbines are 

designed to operate in conditions where the mass flow rate directed to district heating heat exchangers 

varies from 0 to 90 percent. Such plants require an additional heat source to be installed to cover the heat 

consumption of the CO2 sorbent regeneration system. It is possible to build an additional power unit with a 

backpressure turbine to heat the sorbent up. However one must consider high costs of such an operation 

and must provide additional space for the installation. The paper presents an analysis of a CHP (Combined 

Heat and Power) plant with the increased steam flow rate by the amount needed for the sequestration 

process. 

Advantages of this solution would be an increased efficiency through increased blade length and higher heat 

power until integration with CO2 sequestration plant. Two variants are analyzed, one with a single lowpressure 

unit, second with two parallel connected LP units. In the second case, a possibility of removing one 

of the LP parts after integration with CO2 installation. 

A parametric study of the units efficiency as a function of sorbents energy consumption ratio is carried out. 

Keywords: 

Combined Heat and Power, Carbon Capture and Storage. 


Sorbent regeneration process requires amount of heat that depends on the type of used chemical 

agent. The heat is carried by steam extracted from a turbine bleed, Fig. 1. In CHP plans with 

extraction-condensing turbines there exists no possibility of an extra steam extraction for the 

chemical absorption process due to the construction reasons of those turbines, for which it is usually 

assumed that the steam flow rate varies, depending on the needed heating power, from 0 to 95% of 

the steam supplied to an inlet of a LP (Low Pressure) turbine unit. As a result, there is an 

insufficient amount of steam that could feed an absorber column. Therefore, such a CHP plant does 

not meet the requirements of a “capture ready” plant, which state that a “capture ready” plant 

should be designed in the way that will allow it to be equipped with an CO2 capture installation as 

soon as commercial technology is available. In practice it all comes to the providence of the 

required space needed to build the installation. The lack of an extra steam extraction, in case of the 

typical plants, implies an introduction of an additional heat source that will provide the required 

heat to the sorption process. One of the mostly considered options is an introduction of an 

additional small CHP plant with an backpressure turbine to heat the sorbent in a CCS (Carbon 

Capture and Storage) unit, Fig.1. However, such a solution is costly, and needs some extra space for 

the installation itself. Furthermore it will also emit CO2. Most probably coal would be used here as 

fuel, since only-biomass boiler would not provide enough steam. Such concepts should be 

considered, however the most reasonable solution seems to be such a design of a plant that would 

include the heating needs of a CC (Carbon Capture) installation from the very beginning. Which 

means, the plant would be capable of producing more steam that is required only for district heating 

and electrical power production. Thus, any bigger modifications of an existing plant to couple it 

133

with a CC installation could be avoided. Most practical would be an available additional steam 

extraction with parameters desired for the adsorption process. Such a solution could also be 

economically beneficial. Flow rate of steam feeding the turbine results from a CC installation heat 

consumption and desired electric and district heating power production. 

Fig. 1. Scheme of CHP plant integration with a CO2 capture unit. A - absorption column, D - 

desorption column, C - compressor, HP – high pressure turbine, LP – low pressure turbine, B – 

boiler, FTU – flue gas treatment unit 

2. Design calculations 

2.1. Basic thermal parameters of steam in the cycle 

The higher the steam parameters that feeds a turbine are, the higher the cycle efficiency is and also 

the CO2 emitted to the atmosphere shrinks. A steam boiler of given below parameters is considered 

in the project: 

Table 1. Steam and water thermal parameters in the cycle 

Unit Value 

Fresh steam pressure at boilers outlet MPa 30.3 

Fresh steam temperature at boilers outlet °C 653 

Fresh steam pressure at turbines inlet MPa 30 

Fresh steam temperature at turbines inlet °C 650 

Reheated steam pressure at boilers outlet MPa 6 

Reheated steam temperature at boilers outlet °C 672 

Reheated steam temperature at turbines inlet °C 670 

Boilers feeding water temperature °C 310 

Coal of composition c=0.599, h=0.038, s=0,01, n=0,012, o=0,05, p=0.2, w=0.09 and LHV of 23 

MJ/kg was considered as fuel. 

134

2.2. Design Cases 

Aim of this research is to design a CHP plant that will meet the requirement of the UE parliament 

directive 2001/80/WE. According to which any fossil fired power unit of electric power of 300MW 

and above has to undergo an analysis of CO2 storage possibility, economical and technical aspects 

of CO2 transport as well as modernization towards CC installation coupling to a power unit. 

Therefore it was assumed that the CHP plant will have 305MWe power. Electric power of currently 

working polish CHP units does not exceed 145MWe. For an assumed nominal electric power, 

heating power depends on the amount of steam that can be directed to a heating station. Two design 

cases for a CHP plant were considered: 

1. A “capture-ready” plant. Its design allows to build on an CC unit in future. 

2. A plant that will be equipped with a CC unit already from the beginning. 

Fig. 2 shows a scheme of CHP plant with an extraction point intended to feed a CC unit. Steam 

used for district heating purposes is taken from the outlet of the IP part of turbine which also 

supplies the CC unit. 

Fig. 2. Scheme of CHP plant with a steam extraction point for CO2 capture purposes 

An important role plays here the appropriate choice of the structure of a plant. There are various 

possibilities like i.e.: 

1. With a single-flow IP and one single-flow LP part of turbine. 

2. With a symmetric double-flow IP and one or two connected parallel LP parts of turbine. 

The second option allows shutting down one of the LP parts, after integration with a CC unit. The 

chosen structure was selected on the basis of an analysis of different LP regeneration 

configurations. 

135

2.3. Basic CHP plant indices. 

The basic indices are based upon following equation: 

Cycle efficiency: 

Where: - heat delivered to the cycle, – heat extracted from cycle 

Plant gross efficiency: 

Where: – gross electric power, – fuel flow, - lower heating value 

Plant net efficiency: 

Where: – in-house load (auxiliaries) 

Heat rate: 

CHP plant efficiency: 

Where: - heating capacity 

(2) 

(3) 

(4) 

(5) 

2.4. Selection of steam extraction point for CO2 separation process. 

The appropriate placing of an extraction point is dictated by the heat consumption of a given 

sorbent and the required temperature for the sorbent regeneration. 

136 

(1)

There are currently studies carried out on new types of sorbents in order to decrease their 

regeneration heat consumption (i.e. for MEA from 4,5 MJ/kg CO2 to 3 MJ/kg [2]). Because of large 

amount of steam required it is only possible to extract it from IP to LP line. Thermal parameters at 

this point should fulfill sorbents regeneration process requirements and on the other hand be low 

enough to avoid thermal degeneration of the sorbent. The desired pressure should be then equal to 

p=ps(T+Ts)/(1- ) 

Where: T – Sorbent upper temperature, Ts – upper temperature difference, – pressure loss 

coefficient. 

If one decides to use amine sorbents, then it is necessary to heat them to ca. T = 127 °C. If Ts = 5 

K and = 0.02, then steam of pressure p = 0.3 MPa is needed to feed the sorption unit. 

2.5. Results of the design calculations. 

It was assumed in computations that the plants nominal working conditions is pure condensation, 

which means no steam is provided for district heating. Table 2 shows calculated results for 

considered variants of nominal electric power of 305 MW each. Computations were carried out 

under assumption that the house load is equal to 7.5 % of plants gross power. They were also 

performed for two values of sorbents heat consumption. 

Table 2. Results of design calculations 

Only condensation With CC unit steam 

consumption 

Fresh steam flow rate, kg/s 209 245,2 231,5 

Sorbents energy consumption, 

MJ/kgCO2 

- 3,89 2,83 

CO2 separation steam flow 

rate, kg/s 

0 101,3 70,298 

Steam feeding LP turbine flow 140,12 69,48 90,916 

rate, kg/s 

Cycle efficiency, % 51,23 43,48 46,14 

Gross cycle efficiency, % 49,35 42,09 44,59 

Net cycle efficiency, % 45,65 38,94 41,24 

Boiler efficiency, % 94,41 94,41 94,41 

Heat rate, kJ/kWh 6886,6 8074,1 7711,8 

Fuel flow rate, kg/s 26,87 31,524 30,06 

CO2 flow rate, kg/s 59,398 69,686 65,793 

Percentage of removed CO2, 

% 

0 90 90 

Steam mass flow rate produced in boiler depends on the chosen variant. In case of a “capure-ready” 

CHP plant it is smaller than that in case of a plant already equipped with a CC installation. It is 

related to the bigger mass flow rate through LP turbine compared to that in the previous case, which 

is responsible for generating more mechanical power. Sorbents heat consumption has high impact 

on work indices of a CHP plant that is already build with a CCS installation. Its reduction by 1 

MJ/kgCO2 causes an efficiency growth of 2% points. Whereas in case of “capture-ready” plant the 

efficiency is higher by 4.5 % points compared to a plat with CC unit of sorbents energy 

consumption equal to 2.83 MJ/kgCO2 and about 6.5 % points of sorbents energy consumption 

equal to 3.89 MJ/kgCO2. Calculations included only sorbents energy needs and left out sorbent 

pump losses and CO2 compressors work. 

137

3. Off-design calculations 

The main task of off-design calculations is to analyze their working conditions for different thermal 

and electrical load. Examinations were carried out under assumption that plants operate under 

nominal boiler load (table 2). Results of the calculations for the different variants are shown in Fig. 

3. Fig. 3 shows electric power of a CHP plant with an operating CC unit of CO2 capture efficiency 

90 % and with a turned off district heating unit. Gross electric power of the first variant plant will 

decrease after integration with a CC unit from 305 to 256 MW for sorbents energy consumption of 

3,89 MJ/kgCO2 and to 269 MW for sorbents of 2,83 MJ/kgCO2 energy consumption. Fig. 4 

illustrates values of the gross efficiency of the plants, corresponding to their design points. Gross 

efficiency value of variant W1-2,8 and W1-3,89 is the expected efficiency of plant after its 

integration with the CC unit. Fig. 5 shows peak heating capacities of the discussed variants and 

decrease in electric power resulting from lower steam flow rate in the LP turbine, which is to be 

directed to the district heating heat exchanger. Fig.6 illustrates the influence of the sorbents heat 

consumption rate on the efficiency of CHP plants by their peak heating capacities. A simple relation 

can be seen here: the smaller the heat consumption of a sorbent is the more heat (steam) can be 

directed to the district heating heat exchanger and is then considered as useful. 

One of the most important characteristic values of CHP plant is electric power produced in cogeneration 

and in condensation, while they play an important role in economic effectiveness of a 

CHP plant. In case of a CHP plant integrated with a CC unit it is also advisable to include power 

generated by the steam used in the CC unit. Fig. 7-12 show shares of each steam flow (for heating, 

sorbent regeneration and condensation) in power generation. 

Fig. 3. Electric power 

138

Fig. 4. Gross efficiency of plants 

Fig. 5. Electric power at peak heating capacities 

139

Fig. 6. CHP plant efficiency 

Fig. 7. Electric power produced by steam used for: condensation, heating and CO2 absorption 

140

Fig. 8. Percentage share of electric power produced by steam directed to condenser, district 

heating and CO2 capture purposes for the first variant – W1 by different energy 

consumption rates of sorbent (2,8 MJ/kgCO2 and 3,89 MJ/kgCO2). 

Fig. 9. Percentage share of electric power produced by steam directed to condenser, district 

heating and CO2 capture purposes for the second variant – W2 by different energy 

consumption rates of sorbent (2,8 MJ/kgCO2 and 3,89 MJ/kgCO2). 

Fig. 10. Electric power as function of heating capacity 

3.1. Heating unit operation 

Heating unit is fed from the intermediate pressure turbine outlet. Before integration with a CC unit 

outlet water temperature can be regulated only by means of steam throttling at the low pressure 

turbines inlet. After integration, pressure at the intermediate turbines outlet has to be kept on a level 

that will provide proper temperature for sorbents regeneration. Regulation is realized then by 

141

amount of steam delivered to the heating unit, adjusted by a throttling valve and a by-pass on the 

heated water side. 

Enthalpy [kJ/kg] 

4200 

4000 

3800 

3600 

3400 

3200 

3000 

2800 

2600 

2400 

2200 

Fig. 11. Expansion line in the turbine 

50 MPa 

30 

20 

10 

05 

5.0 

2000 

5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 

Entropy [kJ/kgK] 

Table 3 contains results of a “capture-ready” CHP plant off-design calculations after integration 

with a CC unit with sorbent of 2.83 MJ/kgCO2. Computations were carried out for the nominal 

steam flow of fresh steam and two types of heated water outlet temperature regulation by thermal 

power equal to 108400 kW, which is 57% of its maximal value. In case of steam throttling 

regulation the regulation valve remains closed. Throttling starts at point 05 and continues towards 

point 15, Fig. 11. During by-pass regulation no throttling is used so water is heated by steam of 

parameters at point 05. Division of water stream between the heat exchanger and by-pass results 

from the desired water temperature. 

142 

0.85 

2.0 

1,0 

15 06 

Table 3. Steam and water parameters at the heating unit 

Quantity Unit 

0.90 

0,5 

0.95 

0.2 

x=1 

0.1 

Regulation method 

throttling bypass 

Heating capacity kW 108400. 108400. 

Fresh steam flow rate kg/s 209.0 209.0 

Steam flow rate m15 kg/s 43.245 45.511 

Steam pressure p15 MPa 0.300 0.300 

Steam pressure p16 MPa 0.116 0.297 

Water temperature T200 

o 

C 55.8 55.800 


o 

C 98.7 128.216 


o 

C 98.7 98.7 

Water flow rate m200 kg/s 602.877 602.877 

Watre flow rate m201 kg/s 0 247.178 

0.05 

100 

0.03 

0.01 

0.005 

0.03 

0.001 

800 O C 

700 

600 

500 

400 

300 

200

Gross electric power kW 244619. 244962. 

CHP plant efficiency % 57.13 57.19 

Results of performed analysis show that the efficiency of the CHP plant is only slightly influenced 

by the way of regulation. By-pass regulation provides higher electric power by 1.4 per mill and 

efficiency by 0.06 percentage point higher than that obtained when using throttling. 

This kind of regulation is also used in the second variant. 

4. Modernization of the power plant after its integration with a CO2 

capture facility 

A CHP plant will need to be modernized after being coupled with a CO2 capture unit. This problem 

is covered in [3 - 6]. It concerns especially turbine, regeneration system and condensers cooling [3 - 

5]. Therefore, machinery and utilities of such a plant should be chosen appropriately during the 

design stage taking those changes into consideration. Considered structure (first variant) of a CHP 

plant was chosen in the way that will reduce required after the integration changes as much as 

possible. In fact, they are limited to the low pressure turbine, which will be mostly affected by 

changed operation conditions. 

5. Conclusions and remarks 

Aim of this work was a design of a CHP plant structure, that will meet the requirements of the UE 

Parliament and Council directive UE 2001/80/WE. 

Presented structure of a “capture ready” CHP plant will work with possibly high efficiency before 

and after the integration with a CO2 capture unit, providing costs of modernization to be low as 

possible. 

By decreasing the sorbents energy consumption by 1 MJ/CO2 one can raise the efficiency by ca. 2 

percentage points. 

Peak power of the heating unit is limited by the sorbents energy consumption rate. 

Calculation results show that the heating unit regulation method does not affect the plants efficiency 

significantly. 

Performed analysis includes only the sorbents energy consumption rate and omits other auxiliary 

power needs. 



Centre of Research and Development in the framework of Contract SP/E/1/67484/10 – „Strategic 

Research Programme – Advanced Technologies for obtaining energy: Development of a technology 

for highly efficient zero-emission coal-fired Power units integrated with CO 2 capture”. 

References 

[1] Duan L., Zhao M., Xu G., Yang Y.: Integration and optimization on the coal fired power plant 

with CO2 capture using MEA. 24 th International Conference ECOS 2011, Novy Sad, 4- 

7.07.2011, Conference papers, p. 582-593. 

[2] Abu-Zahra M.R.M., Schneiders L.H.J., Niederer J.P.M., Feron P.H.M., Versteeg G.F., CO2 

capture from power plants. A parametric study of the technical performance based on 

monoethanolamine. Part I. International Journal of Greenhouse Gas Control 2007; p. 47-46 

[3] IEA Greenhouse Gas R&D Programme (IEA GHG), “CO2 capture ready plants”, 2007/4, May 

2007. 

[4] Lucquiaud M., Chalmers H., Gibbins J.: Capture-ready supercritical coal-fired power plants and 

flexible post-combustion CO2 capture. Energy Procedia 1(2009), p. 1411-1418. 

143

[5] Lukowicz H., Mroncz M.: Analysis of the possibilities of steam extraction from a condensing 

turbine 900 MW for the carbon dioxide separation system. Archiwum Energetyki, vol. 

XLII(2012), nr. 1, p.1–13. 

144




Cryogenic Method for H2 and CH4 recovery from a 

rich CO2 stream in pre-combustion CCS schemes 

K. Atsonios 1,2 , K. D. Panopoulos 2 , A. Doukelis 1 , A. Koumanakos 1 , E. Kakaras 1,2 

1 Laboratory of Steam Boilers and Thermal Plants, National Technical University of Athens, Heroon 

Polytechniou 9, 15780, Athens, Greece, atsonios@central.ntua.gr, adoukel@central.ntua.gr, 

akouman@central.ntua.gr, ekak@central.ntua.gr 

2 Institute for Solid Fuels Technology and Applications, Centre for Research and Technology Hellas, 

4th km. N.R. Ptolemais – Kozani, 50200 Ptolemais, Greece, panopoulos@certh.gr 

Abstract: 

Pre-combustion carbon capture technology based on coal gasification or methane reforming followed by 

water gas shift reactors assisted with Pd-alloys membranes (WGS-MR) is considered very promising for the 

production of a rich hydrogen stream that can be combusted in combined cycles. 

However, the recovery of the total H2 content is not feasible and a part of it remains in the retentate side. 

The requirement for upstream high pressure operation of the necessary reforming step has a drawback: 

complete reforming of the CH4 is not allowed; thus small but significant amounts of it remain in the rich CO2 

stream. These CH4 amounts not only affect the efficiency of the process but also are against regulations for 

the allowed composition of carbon dioxide for storage. Therefore an efficient purification step before its 

compression is of high importance. This work models a cryogenic method for combustibles separation from a 

rich-CO2 stream and evaluates its effects on pre-combustion carbon capture systems’ efficiency. The 

modeling study is performed in AspenPlusTM. An investigation of the operating parameters is presented as 

well as how other parameters of the Purification & Compression Unit (PCU) affect performance. 

Keywords: 

Hydrogen production, ATR, WGS Membrane Reactor, CO2 purification, cryogenic separation 


Pre-combustion capture is one of the proposed options for carbon dioxide removal for which great 

interest has been shown lately [1, 2]. Having sequestrated carbon in the form of CO2 before the fuel 

combustion and at high pressures, enables using the resulting fuel gas in various applications like 

the production of power, pure hydrogen or other chemicals. 

In CO2 pre-combustion capture schemes based on Combined Cycles using Natural Gas (NGCC) or 

Integrated with Coal Gasification (IGCC) the main strategy is to reform or gasify the original fuel 

towards a reformate fuel. The standard proposed method for this carbon reduction is then to perform 

water gas shift, transform CO into CO2 and use amine absorption for capture. An alternative is to 

employ H2 permeable reactors assisting the shift reaction: this option has advantages, such as high 

hydrogen recovery by pushing the shift reaction and high H2 purity. A schematic of such an 

operation is shown in Figure 1: the reformate gas stream is fed to the reactor, CO is shifted with 

simultaneous H2 recovery on the permeate side through Pd-alloys membranes. The permeation 

mechanism of Pd-alloys is through the disassociation of H2 on the membrane surface and 

transportation through the metal structure as atomic hydrogen [9, 10]. Some of the main 

disadvantages of this technique are the high costs, potential poisoning at low temperatures and the 

fact that the hydrogen recovery is not perfect and considerable amounts of hydrogen remain at the 

retentate side (from 2 to 10% of the total hydrogen [11]). Besides hydrogen, other combustibles are 

present such as residual CH4 that was not reformed, as well as CO traces that have not been shifted. 

Since the retentate stream contains the whole amount of CO2 that is led to storage, it is essential to 

increase CO2 purity by removing the other species, like H2O, H2, CH4, CO, etc. On one hand, 

145

water removal is feasible by cooling the stream under the dew point so that the water is condensed. 

On the other hand, special attention is paid to the other species. The conventional suggestion in 

order to eliminate these compounds is to combust them [12], but this has to be accomplished with 

pure oxygen as the oxidant agent. Technically this would require the use of catalysts to achieve high 

combustion rates with such a lean fuel and stoichiometric ratio very close to 1. 

In all carbon capture schemes, it is of paramount importance that the internal energetic consumption 

of the CO2 compression train is minimised [3-8] by purifying the rich-CO2 stream. Huang et al. [6] 

and Posch et al. [8] adopted a cryogenic method for CO2 purification for oxy fuel capture schemes, 

which was proven to be energetically and costly inefficient because of high energy consumptions 

due to the cooling loads, while the recovered gas could not be further utilized and was directed to 

the stack. It was also shown that applying a distillation column rather than flash separators provided 

better CO2 stream purification, accompanied by a higher energy penalty. 

The present study suggests an alternative choice to handling the combustibles by recovering them 

with cryogenic separation techniques. Based on differences in thermodynamic properties as far as 

the dew point of each component is concerned, the retentate stream is cooled down and the CO2 is 

separated in flash separators or a distillation column. A parametric investigation for the best 

operation of the plant is performed and a comparison is made with the conventional option for 

purification with oxy combustion. 

2. Plant Description 

Figure 1. Process flowsheet diagram of the total power system 

The outline of the system is presented in Figure 1. A H2 rich fuel is produced from natural gas 

reforming in an Autothermal Reactor (ATR), and CO is further shifted in a High Temperature 

Water Gas Shift Reactor (HT-WGS). The autothermal conditions are met using a rich oxygen 

stream (95% purity) which is produced in an Air Separation Unit (ASU). The maximization of 

hydrogen production and purification is performed in Water Gas Shift -Membrane Reactor where it 

is assumed that the Hydrogen Recovery Factor (HRF) is equal to 98% and the operating 

temperature 400°C is (base case). The nitrogen stream that is produced from the ASU is utilized as 

sweep gas, increasing the hydrogen recovery driving force in the membranes. The H2-fuel mixture 

is fed to the power plant island, which consists of a Gas Turbine combined with a Heat Recovery 

Steam Generator (HRSG). A more detailed description of the process can be found in a previous 

study [13]. 

146

The base case for the CO2 purification and compression option is also shown in Figure 1: The 

retentate stream after the membranes is expanded and catalytically combusted in an oxy-fired 

combustor. Chemical equilibrium predicts that complete combustion of the remaining combustible 

species is feasible with almost no oxygen surplus [13] but this is not practically easily achievable. 

The required oxygen depends on how the upsteam units (ATR and WGS performance) are operated: 

for example assuming TATR=1050°C and S/CATR=1.5 for the base case, the required amount of 

oxygen for the post combustor increases the total oxygen production in the ASU by 17.5%. The hot 

flue gases deliver heat in a secondary Heat Recovery Steam Generator where the feedwater is 

transformed to high-pressure superheated steam. Next, the water content of the flue gases is 

removed in a flash separator, and the almost pure CO2 is then compressed and pumped. The 

operating parameters of the ATR and WGSMR play an important role in the amount of heat present 

in the retentate stream. The present study is focused in the CO2 stream purification and 

compression block based on two separation methods, (a) flash separator and (b) distillation column 

(see Fig. 2). Apart from CO2, the retentate stream mainly consists of H2O, H2, CH4, N2 and Ar (see 

Table 1). 

Table 1. Retentate stream T, P and composition 

This gas stream is firstly expanded and then is cooled down to around 220°C. Some of the heat is 

recovered for generating superheated intermediate pressure steam at 315°C. Expanding the gas has 

a triple positive effect: firstly, the mixture is separated more easily at lower pressures. Secondly, the 

manufacturing cost of equipment such as the evaporator is significantly lower if they operate at 

lower pressures. Thirdly, as the content of water in the retentate stream is about 25% w/w, the CO2 

rich mass flow rate that is compressed is less than the corresponding stream that is expanded, 

contributing positively to the total power balance. 

After that, cooling water is used to bring the stream to water condensation conditions at 28°C. The 

next part of the Purification and Compression Units differs for the two proposed schemes: 

2.1 Scheme 1: Double flash separation – internal cooling 

This system is auto-refrigerated with no additional cooling system required (Figure 2). Flash 

separations are performed at two different temperatures, and at the same pressure level. Before each 

flash, there is a Heat Exchanger that cools the inlet stream. The required cooling loads are taken 

from the final steams as can be seen in Figure 2a. The rich-CO2 liquid steams are throttled 

adiabatically and their temperature is reduced (Joule–Thomson effect). The level of throttling has 

been set so as to permit heat transfer at the two Heat Exchangers, without temperature crossovers, 

assuming a minimum temperature approach =3°C. The final streams come out of the 

Purification Unit at the temperature of 18°C and the rich-CO2 stream enters the Compression Unit, 

where it is compressed in a three stage inter-cooled compressor up to 80 bar to supercritical 

conditions. Then, it is cooled, liquefied and pumped up to 28°C/110bar and is transported for 

storage. 

Since the temperature at the 2 nd Flash is -54.5°C, (near the triple point of CO2) the parameters that 

determine the system’s efficiency are the expander outlet pressure and the temperature of the 1st 

flash separation. 

147

Figure 2. Sketch of the suggested Purification and Compression Unit with cryogenic fuel recovery 

with flash separators. 

The expander outlet pressure determines the CO2 capture rate: Lower outlet pressures result in 

better CO2 recovery rate. On the other hand, if the pressure outlet of the rich CO2 stream is low, the 

energy duty for compression is considerable. 

2.2 Scheme 2: Separation by distillation column – external & internal 

cooling 

Figure 3. Schematic of the suggested Purification and Compression Unit with cryogenic fuel 

recovery with distillation column. 

The concept of combustibles recovery with distillation mainly consists of a distillation column, a 

heat exchanger, a flash separator and an external refrigeration system (Figure 3). 

After the Heat Exchanger, the inlet stream is partly condensed and is separated at the flash 

separator. The liquid stream is further cooled by means of an expansion Joule–Thomson effect, 

while the gas stream is cooled through expansion. Both streams enter the distillation column to 

make combustibles separation more effective. The cooling loads for the column’s condenser are 

obtained from an external refrigerant and are dependent to the desirable rate of Carbon Capture 

Efficiency. The corresponding heat at the reboiler can be obtained by the refrigerant inter-cooling 

148

while it is compressed. The distillation column outlet streams can be used to assist the cooling of 

the retentate stream after water removal (reducing energy consumption). 

The refrigeration system comprises a two stage compressor in order to provide cooling loads at two 

different temperatures: at -20°C for the retentate stream cooling and at around -65°C (depending on 

recovery rate that determines the dew point of the outlet gas stream) to fulfil the condenser duties at 

the distillation column. The most suitable cooling medium that is employed in the modelling is 

R1150 (ethylene – C2H4). Assuming that the temperature approach of all heat exchangers is 3°C, 

the pressure levels of the evaporation are 2.26 and 23.47 bar and that for condensing is 27 bar. 

Finally, the compressor’s polytropic efficiency is assumed to be equal to 0.82. Since the present 

cooling medium cannot reject its heat directly to the ambient while it is condensed, a secondary 

auxiliary cooling cycle is required. The cooling medium in this refrigerant cycle is the commercial 

R134a and the COP of this cycle is assumed 3.59. 

The thermochemical properties of the PCU block are calculated according to the Peng- Robinson 

equation of state [8]. The technical data of the system are presented at the Table 1. 

Table 1. Process model specifications 

3. Results and Discussion 

3.1 Process parameters investigation 

In order to come up with concluding remarks about the efficiency of the proposed systems, the 

comparison of exergetic efficiencies of the schemes is performed. The exergy balance of the PCU 

block is shown in Figure 5a. The exergy input comes from the retentate stream while any power or 

heat duties are considered as part of the exergy outputs. 

The exergy dissipation for the cases under investigation is depicted by Grassmann diagrams, which 

are displayed in Figures 5b-d. Arrows that are at the upper side of the main exergy arrow 

correspond to exergy that is not lost (power production or heat recovery at the steam cycle) whereas 

the remaining arrows refer to exergy losses (irreversibilities). As far as the process of the oxycombustion 

option is concerned, heat recovery and power generation are exergy that is not lost. 

The retentate stream at the exit of the burner is at high temperature and a large fraction of the 

exergy input is utilised for high enthalpy superheated steam generation. 

The ASU consumption for extra oxygen production is not taken into account in the exergy balance 

and its effect on total efficiency is investigated below. On the other hand, both separation options 

can recover almost half of the total exergy by recovering the combustible content. 

149

a b 

Figure 5. a) Exergy balance of the PCU and Grassmann diagram of PCU with b) expander and 

post combustor, c) flash separation and d) distillation column. 

The corresponding gains from heat recovery are lower in the combustibles separation cases, as the 

temperature inlet at the heat exchanger is 400°C instead of 708°C in the oxycombustion case. 

Consequently, the steam quality differs from case to case, as well as the corresponding exergy 

utilisation. The final CO2 stream in the oxy-combustion case has larger exergy fraction than the 

separation options, due to increased CO2 content from CH4 combustion. Other losses correspond to 

irreversibilities from various processes like heat exchanger, valves, expanders, separators etc. 

According to Figure 5a, the exergetic efficiency of the Unit can be defined as: 

where, PPCU>0 when the Unit operation yields power and PPCU

Table 2. Comparison of the three CO2 purification and compression options 

The most important streams of the process are shown in Table 3. The second method of separation 

is more efficient as far as rec is concerned, resulting in higher purity CO2-rich stream. What is 

more, in this case, the CO2 content in the recovered gas stream is lower. 

Table 3. Outlet streams from PCU for the three purification methods 

It should be mentioned that high recovery rates do not mean high efficiency of the total plant. The 

way that the recovered combustibles are exploited plays significant role on the choice of the most 

suitable technique. To this end, the thermodynamic comparison of the purification options under 

investigation is completed by the process integration and the investigation of their effects on total 

plant operation. 

3.2 Effect of purification methods on total plant operation 

Unlike other cryogenic CCS applications, due to the fact that the recovered gas stream has a 

considerable amount of chemical energy, it can be fed back to the system for increased energy 

efficiency. The proposed alternatives for its utilisation are either to reform or to burn it. 

Although returning them to the ATR would decrease fuel consumption, the inert compounds that 

are contained in the stream (mainly N2 and Ar) would accumulate in the reactor since there is no 

way to escape from the system. Given that there is not an available method to remove them, this 

option is abandoned. The alternative choice suggests the injection to the GT combustor. In this case, 

the carbon capture rate of the system is less than the capture rate at the PCU due to the produced 

CO2 from the combusted CH4 in the GT. Given that the system under investigation is aimed to the 

maximization of the final H2 rich fuel and not the maximization of electrical power production, a 

different approach concerning the recovered fuel should be adopted. 

151

The basis of comparison of each method from performance perspective can be the parameter named 

SPECCA (Specific Energy Consumption for CO2 Avoided). It expresses the additional fuel energy 

in MJ that is required to avoid 1 kg of CO2: 

where E is the CO2 emission rate, in kgCO2/kWhel, and the net electrical efficiency of the plants. 

The reference plant (REF) is referred to the corresponding NGCC plant without Carbon Capture. 

The net efficiency of the reference plant is REF=58.3% with specific CO2 emission rate 

EREF=354.3grCO2/kWh. 

Among the most important parameters for total plant specification are the operating parameters of 

the ATR and the WGS-MR. Table 4 summarizes the characteristics of the base case model for the 

three purification methods under investigation: 

Table 4. Base case results for the three purification methods (CCR=90%) 

It is clear from Table 4 that oxy combustion of the retained combustibles (namely CH4 that is not 

reformed, CO that is not shifted and H2 not recovered at the membranes) is the most efficient 

method for the CO2-rich gas treatment in terms of energy efficiency. However, the specific quantity 

of the produced H2 that enters the GT combustor is increased by 11% in the cases of cryogenic 

separation. In other words, for systems dedicated to H2 production, cryogenic separation methods 

are considered to be more efficient in terms of H2 production yield. Additionally, the oxygen 

demand is reduced by 10% in these cases, implying a smaller ASU. However, the more complex 

purification system compensates this feature. The high heat recovery rate in the oxy-combustion 

case results in the increased power production in the ST (c. 2% increase). 

3.2.1 Effect of Carbon Capture Rate (CCR) 

It should be mentioned that Carbon Capture Efficiency (CCE) does not coincide with the Carbon 

Capture Rate (CCR) of the total plant because it is independent of the final usage of the recovered 

gases. In this study, this stream is fed to the GT to be combusted. As a result, the CCR is also 

dependent on the recovered CH4 and CO. 

Figure 6 provides useful information about the effect of Carbon Capture Rate on the plant 

performance. 

152

Figure 6a. Impact of CCE on PCU Figure 6b. Effect of CCR on the power plant 

performance for the two purification methods. for the two purification methods. 

For low CCE the flash separation method has positive effect on energy balance as the PCU 

produces electrical power instead of consuming (Figure 6a). This is owed to the expansion of the 

CO2-rich stream. However, the combustibles are not recovered as effectively as in the case of the 

distillation column method and therefore the latter is exergetically more efficient. As the CCE 

approaches 100%, energy duties for both methods tend to be equal and the divergence of the 

corresponding exergy efficiency increases. The effect of CCR on the overall plant efficiency and the 

corresponding SPECCA (Figure 6b) is the same for both combustibles separation options 

(especially for CCR

Figure 7a. Impact of S/CATR on PCU Figure 7b. Effect of S/CATR on the power 

performance for the two purification plant for the two purification methods. 

methods (CCE=90%). 

3.2.3 Effect of Hydrogen Recovery Factor (HRF) of the WGS-MR 

This parameter refers to the ability of the reactor to make hydrogen available at the opposite side 

(permeate side), where the H2-rich fuel is produced. Recent developments at Pd-alloys membranes 

combined with WGS catalysts showed that hydrogen recovery is achievable at rates greater than 

90% [14] and in some cases even close to 100% [11]. Hydrogen recovery factor strongly affects 

membranes cost as the higher the hydrogen recovery the larger membrane area required [15]. 

Figure 8a. Impact of HRF on PCU Figure 8b. Effect of HRF on the power plant 

performance for the two purification methods for the two purification methods. 

(CCE=90%). 

Figure 8b shows that cryogenic methods may be beneficial for membranes with low HRF (>90%) 

as the variation of this parameter does not affect the total efficiency, unlike the oxy combustion 

option. Membranes with high HRF favor the application of oxy combustor instead of combustibles 

recovery. 

154

3.2.4 Effect of the ATR operational temperature (TATR) 

The ATR operating temperature is of high importance as it plays significant role on the syngas 

quality: according to the chemical equilibrium, the higher the temperature, the higher the methane 

conversion rate. On the other hand, in order to secure the autothermal conditions in the reactor, 

more oxidant agent is required in the case that the temperature is high. 

Furthermore, materials stability limits the maximum operating temperature to around 1100°C. 

Figure 9a. Impact of TATR on PCU Figure 9b. Effect of TATR on the power plant 

performance for the two purification methods for the two purification methods (CCR=90%). 

(CCE=90%). 

The ATR operating temperature is of high importance as it plays significant role on the syngas 

quality: according to the chemical equilibrium, the higher the temperature, the higher the methane 

conversion rate. On the other hand, in order to secure the autothermal conditions in the reactor, 

more oxidant agent is required in the case that the temperature is high. 

Furthermore, materials stability limits the maximum operating temperature to around 1100°C. 

As is shown in Figure 9a, at low ATR temperatures, cryogenic systems consume more energy for 

CO2 purification due to the increased presence of methane in the retentate stream. 

Recovery and exergy efficiencies for separation with a distillation column are greater than those of 

the corresponding flash separation method. This feature has a positive effect on the total system 

when the combustion of the recovered fuel satisfies the required CO2 capture rate. In this case, for 

TATR = 950°C, part of the methane is selected not to be recovered in order to meet the goal of 

CCR=90%. Consequently, the net efficiency drops considerably, (Figure 9b). At high temperatures, 

the oxy combustion option is more efficient than the cryogenic options by ca. 1%. 


This study investigates the cryogenic method as an alternative choice for the rich-CO2 stream 

purification after membrane separation, instead of simply combusting the retained combustibles. 

Two proposed cryogenic systems are investigated: flash separation with internal cooling and 

separation with distillation column. In the first case, electrical power is produced while the 

separation efficiency is quite high, as 62.6% of the combustibles heat input is recovered. On the 

other hand, separation by a distillation column may result in the complete separation of 

combustibles, providing high purity in the final CO2 stream (>99%). 

155

However, more energy duty is required due to the external cooling system. 

The sensitivity analysis showed that the cryogenic methods can overbalance any ‘weak’ operating 

mode of the hydrogen block, such us low methane conversion rates at the ATR and low hydrogen 

recovery rates at the membranes. Nevertheless, as far as the total system efficiency is concerned, the 

oxy combustion option is preferable as it can combine both high capture rates and performance. 

Future work that correlates membrane area and investment cost of the whole plant would finally 

determine under which conditions a cryogenic recovery system is required. 


The authors would like to gratefully acknowledge the support of the European Commission 

(CACHET II, FP7 Project No. 241342). 

Abbreviations 

ASU Air Separation Unit 

ATR Autothermal Reformer reactor 

COP Coefficient of performance 

GT Gas Turbine 

HRSG Heat Recovery Steam Generator 

HRF Hydrogen Recovery Factor 

HT-WGS High Temperature Water Gas Shift reactor 

LHV Lower Heating Value 

NG Natural Gas 

PCU Purification & Compression Unit 

ST Steam Turbine(s) 

S/CATR Steam-to-Carbon Ratio in ATR 

WGS-MR Water Gas Shift Membrane Reactor 

References 

[1] Olajire A. CO2 capture and separation technologies for end-of-pipe applications - A review. 

Energy 2010;35: 2610-2628. 

[2] Kanniche M., Gros-Bonnivard R., Jaud P., Valle-Marcos J., Amann J.M., Bouallou H. 

Precombustion, post-combustion and oxy-combustion in thermal power plant for CO2 capture. 

Applied Thermal Engineering 2010;30:53–62. 

[3] Romeo L.M., Lara Y., González A, Reducing energy penalties in carbon capture with Organic 

Rankine Cycles, Applied Thermal Engineering 31 (2011) 2928-2935 

[4] Zanganeh K.E., Shafeen A., Salvador C., Beigzadeh A. Abbassi M., CO2 processing and 

multipollutant control for oxy-fuel combustion systems using an advanced CO2 capture and 

compression unit (CO2CCU), Energy Procedia 4 (2011) 1018–1025 

[5] Kakaras E., Koumanakos A., Doukelis A., Giannakopoulos D., Vorrias I., Oxyfuel boiler 

design in a lignite-fired power plant, Fuel 86 (2007) 2144–2150 

[6] Huang Y., Wang M., Stephenson P., Rezvania S., McIlveen-Wright D., Minchener A., Hewitt 

N., Dave A., Fleche A., Hybrid coal-fired power plants with CO2 capture: A technical and 

economic evaluation based on computational simulations, Fuel paper in press 

156

[7] Romeo L. M., Bolea I., Lara Y., Escosa J. M., Optimization of intercooling compression in 

CO2 capture systems, Applied Thermal Engineering 29 (2009) 1744–1751 

[8] Posch S., Haider M., Optimization of CO2 compression and purification units (CO2CPU) for 

CCS power plants, Fuel paper in press 

[9] Colin A. Scholes, Kathryn H. Smith, Sandra E. Kentish, Geoff W. Stevens, CO2 capture from 

pre-combustion processes - Strategies for membrane gas separation, International Journal of 

Greenhouse Gas Control 2010;4:739–755. 

[10] John J. Marano, Jared P. Ciferino, Integration of Gas Separation Membranes with IGCC - 

Identifying the right membrane for the right job, Energy Procedia 1 (2009) 361–368 

[11] Mendes D., Chibante V., Zheng J.M., Tosti S., Borgognoni F., Mendes A., Madeira L. M., 

Enhancing the production of hydrogen via water gas shift reaction using Pd-based membrane 

reactors, J Membrane Sci 35 (2010) 12596- 12608 

[12] Dijkstra J.W., Jansen D., Novel concepts for CO2 capture, Energy 29 (2004) 1249–1257 (In: 

6th International Conference on Greenhouse Gas Control Technologies). 

[13] Atsonios K., Panopoulos K., Doukelis A., Kakaras E., Exergy Analysis of a Hydrogen fired 

combined cycle with natural gas reforming and membrane assisted shift reactors for CO2 

capture, presented in ECOS 2011 in Novi Sad, Serbia, July 4-7 2011. 

[14] Jan Wilco Dijkstra, Johannis A.Z. Pieterse, Hui Li, Jurriaan Boon, Yvonne C. van Delft, 

Gunabalan Raju, Gerard Peppink, Ruud W. van den Brink, Daniel Jansen. Development of 

membrane reactor technology for power production with pre-combustion CO2 capture. Energy 

Procedia 2001;4:715–722 

[15] Criscuoli A., Basile A., Drioli E., Loiacono O., An economic feasibility study for water gas 

shift membrane reactor. Journal of Membrane Science 181 (2001) 21–27 

157

Abstract: 




Design and Optimization of ITM Oxy-Combustion 

Power Plants 

Surekha Gunasekaran a , Nicholas D. Mancini a and Alexander Mitsos a 

a Massachusetts Institute of Technology, Cambridge, United States 

surekhag@mit.edu, mancini@mit.edu, amitsos@alum.mit.edu (CA) 

Oxy-fuel combustion using an oxygen ion-transport membrane (ITM) is a promising alternative to the existing 

cryogenic air separation method, which incurs heavy thermodynamic and economic penalties. The 

performance of ITM-based power plant systems depends on the operating conditions, geometric structure of 

the reactor and the integration approach of ITM to the existing power plant system. A detailed study of these 

factors is required to perform an optimization analysis. In this paper, an intermediate-fidelity ITM model is 

used to study the performance of ITM reactors under different operating conditions and flow configurations. 

Using this model, ITM-based oxy-combustion power cycles are investigated. This article focuses on the 

results of an optimization analysis of the AZEP 100 cycle, which consists of a Brayton-like topping cycle and 

a triple pressure heat recovery steam generation bottoming cycle. The effects of power plant operating 

parameters are analyzed, namely the outlet pressure of pumps, turbines, valves and de-aerator, the split 

fractions of splitters, flow rates, and the outlet temperatures of heat exchangers. The optimization study has 

resulted in an increase of 2.92 percentage points which is important with respect to the feasibility of ITMbased 

oxy-combustion power plants compared to alternatives. 

Keywords: 

Oxy-fuel combustion, Ion-transport membrane, Zero-emission power cycle, Power cycle efficiency. 

1. CCS and ITM Technology 

Global warming and anthropogenic emissions of CO2 have motivated the search for more efficient 

and economically feasible environment-friendly technologies for power generation, which 

contributes to about 65% of total anthropogenic CO2 emissions [1]. Carbon-dioxide capture and 

sequestration (CCS) allows for the use of fossil fuels for power generation without the detrimental 

effects of associated CO2 emissions. The most conventional CCS technique is post-combustion 

capture, which is energy-intensive and expensive [2]. 

In the oxy-combustion method, O2 is separated from air prior to the combustion of the fuel-air 

mixture and fuel oxidation occurs in a nitrogen free environment, typically with large recirculation 

of exhaust gases to control the temperature. The flue gas consists only of CO2 and H2O, from which 

CO2 can be separated simply by condensation. Thus, the penalty associated with separation of CO2 

from the flue gas is greatly reduced [3]. At present, large scale separation of O2 from air is done 

using cryogenic air separation methods. The major disadvantages of this method are that it is energy 

intensive, and has low second law efficiency [4]. A promising alternative is the use of ion-transport 

membranes (ITM), which operate based on chemical potential differences, and use a high 

temperature mixed-conducting (ionic and electronic) ceramic membrane [5]. This technology is 

motivated by the fact that the penalties incurred are much lower than the additional power 

requirement for cryogenic air separation [6]. 

158

2. Assessment of flow configurations 

2.1. Intermediate-Fidelity ITM Modelling 

A spatially-distributed quasi two dimensional model developed based on the fundamental 

conservation laws (mass and energy balance), semi empirical oxygen transport equations and fuel 

oxidation kinetics is used in this work. Reference [7] gives a detailed description of this model. A 

brief overview of the results obtained using this model is presented here. 

2.1. Separation-Only Flow Configuration 

Co-current and counter-current mode separation-only ITMs are compared for various flow rates, 

while keeping the sweep ratio the same for a fixed ITM size [8]. 

The partial pressure difference between feed and permeate steam along the length of the reactor 

remains constant for counter-current flow, but decreases for co-current flow. One major advantage 

of the counter-current arrangement is that it has a higher recovery ratio due mostly to a higher 

average wall temperature. Heat transfer coefficient and wall temperature distribution are important 

factors for the optimization of an ITM power plant. 

2.2. Reactive Separation 

The reactive mode arrangement combines the separation of O2 from air with combustion. The main 

promise is to increase the driving force by maintaining a very low partial pressure of O2 on the 

permeate side. In principle, this can eliminate the thermodynamic penalty for separation: the 

reaction drives the separation. 

Reference [8] discusses important co-current simulation results. Initially, the temperature rises 

slowly due to the slow oxygen transport, which is the rate-determining step for the oxidation of the 

fuel. The values of the permeate, feed stream, and membrane temperatures are close to each other 

because of the high heat transfer coefficient between the permeate and feed streams. As the 

temperature increases gradually, the chemical reaction speeds up, accelerating the increase of the 

local temperature. This ultimately results in hot spots, i.e., locations with high temperatures. Since 

ITM membranes have maximum allowed temperature limits, the combustion process needs to be 

controlled by manipulating the diluent to fuel flow ratio in order to ensure membrane stability. 

Although partial pressure difference of O2 between permeate and feed sides is higher for a reactivemode 

arrangement when compared to the separation-only mode, the diffusive flux of O2 does not 

increase significantly. This is due to the greater dependence of the diffusive flux on temperature 

than on the partial pressure difference. Results of counter-current reactive mode simulations show 

that the counter-current arrangement increases localized heating effects. The temperature overshoot 

occurs because the local heat release rate due to combustion exceeds the local heat transfer capacity. 

It is interesting to note that the flow direction does not affect the main driving force (partial pressure 

difference across the membrane) of reactive type ITM. Analogous to a single stream heat 

exchanger, in which one of the temperatures is independent of flow direction, the partial pressure of 

the permeate side is zero irrespective of the flow arrangement. 

3. ITM Oxy-Combustion Power Cycles 

An optimization analysis of ITM and a study of its thermodynamic performance require an 

understanding of the effect of power plant operating conditions, geometric structure of the reactor 

and the specific way in which ITM is integrated with the power plant system, on its performance. In 

particular, the ITM model must predict the effect of ITM size, flow configuration and operation on 

performance (e.g., recovery ratio of oxygen) and constraints (e.g., maximal temperature); moreover, 

it is essential to accurately calculate the pressure drop in the ITM. In this section, the 

implementation of the ITM model used is described along with the optimization formulation. 

159

3.1. JACOBIAN-ASPEN Interface 

ASPEN Plus ® and JACOBIAN are used to simulate an ITM integrated power cycle. A power plant 

flow sheet is constructed using ASPEN Plus ® , which has pre-defined unit operation models for heat 

exchangers, turbo-machines, splitters, chemical reactors, etc. JACOBIAN [9] is an equationoriented 

modelling and simulation program that is used to model the ITM. JACOBIAN and ASPEN 

Plus® are linked using the USER2 model block in ASPEN Plus® as shown in Fig. 1. 

Fig. 1. JACOBIAN -Based ITM Model in ASPEN Plus ® using USER 2 Block [10] 

3.2. ITM-Power Cycle Flow Sheet 

Due to its high performance and compatibility, the advanced zero emission power plant (AZEP) is 

the most commonly studied ITM-based power plant in the literature [6, 12, 13, 15]. The AZEP 

concept can be used for both (essentially) zero emission cycles and partial emissions cycles [6, 13, 

15]. Partial emissions cycles have an additional afterburner which increases the efficiency, but also 

increases the CO2 emissions. 

The focus of this article is optimization of an AZEP 100 power cycle [6, 10, 13, 15]. The cycle is 

sized to produce a net electric power of 500 MW. For partial emission cycles, the base flow sheet is 

the same as the AZEP 100 with the exception of the inclusion of an afterburner after the hightemperature 

heat exchanger in order to increase the gas turbine inlet temperature to the maximum 

possible limit (assumed) of 1573 K. Reference [10] gives a detailed description of various ITMpower 

cycles and the assumptions. 

160

3.3. AZEP 100 

Fig. 2. AZEP 100 Top Cycle Process Flow Diagram in ASPEN Plus ® with a JACOBIAN based ITM 

model [10] 

The topping cycle of the AZEP 100 is a Brayton-like cycle with an ITM air separation unit and a 

combustor. The air is compressed and split into two streams - "AIRMCM" and "AIRREST". The 

feed stream to the ITM is preheated by the recycled combustion products with a heat exchanger 

network (see "LHEX-ITM-HHEX" shown in Fig. 2). "AIRMCM" is preheated to 973 K in the heat 

exchanger "LHEX". This preheated feed stream provides oxygen to the permeate stream in the 

ITM. The "AIRRES" exiting the ITM (O2 depleted stream) is further heated by the combustion 

products "RECYCLED" (which serves as the permeate stream in the ITM) and is expanded in the 

gas turbine. A part of "AIRREST" is used to cool the gas turbine and a part is used to regenerate 

thermal energy from the combustion products in the heat exchanger "BHEX". The permeate stream 

contains O2 (from the feed stream) necessary to burn the required amount of fuel. 

The temperature of combustion products is limited by the temperature limit of the high temperature 

heat exchanger "HHEX". A design specification control loop is implemented to maintain the 

temperature of the combustion products at 1473 K by varying the split fraction of the compressed 

air (splitter "B3"). Another design specification control loop varies the recycle ratio (split fraction 

of splitter "B6") to maintain a minimum approach temperature in "LHEX" without any temperature 

cross over in the heat exchanger network "LHEX-ITM-HHEX". The degrees of freedom for the 

topping cycle include the inlet flow rates and ITM size. 

The "PRODBOTM" stream after extraction of thermal energy in "BHEX" and the "GTEXH" (outlet 

from the turbine) are fed to the bottoming cycle for extraction of work from the thermal energy of 

these streams. A standard triple pressure stream generator cycle with pressure levels at 100, 25 and 

5 bars is used as bottoming cycle for the AZEP 100 (Fig. 3). The outlet stream "TOCPU" is fed to 

the compression and purification unit to separate H2O and CO2. 

161

Fig. 3. Triple Pressure Bottoming Steam cycle: Pressure Levels are 100, 25, 5 bar [10] 

3.4. Optimization of AZEP 100 

A local optimum is determined with sequential quadratic programming (SQP) using the inbuilt 

ASPEN Plus ® Optimizer. Overall, the objective is maximal efficiency keeping the ITM size fixed. 

The power cycle has 14 optimization variables – 5 variables for the topping cycle, and 9 for the 

bottoming cycle. To overcome numerical difficulties and limitations of ASPEN Plus ® , the power 

cycle is optimized in two steps. First, only the topping cycle is optimized. Then, only the bottoming 

cycle is optimized using the inlet streams to the bottoming cycle – "GTEXH" and "PRODBOTM" – 

as input specifications, fixed to the optimum operating condition of the topping cycle. In principle, 

this two-stage method does not guarantee local optimization of the entire cycle [25]. However, as 

the efficiency of the gas turbine "GT" in the top cycle is higher than that of the bottoming cycle, it is 

more efficient to extract work through the gas turbine in the topping cycle than the bottoming cycle. 

In other words, to attain maximum efficiency of the total power plant, maximum possible power 

extraction should take place in the top cycle, transporting minimum thermal energy to the 

bottoming cycle. Thus, sequential optimization of the topping cycle followed by the bottoming 

cycle is believed to give the optimum of the entire power cycle. 

3.4.1. Optimization of Top Cycle 

The one-dimensional intermediate fidelity ITM model makes it impossible to optimize the ITM 

geometry. Attempting to minimize the pressure drop would result in an infinite number of permeate 

and feed channels, which are extremely small in length. Moreover, the ITM is an expensive 

component, so optimization of the ITM size would require accurate estimates for its cost which are 

not available since ITM is a very new technology. Hence, the topping cycle is optimized by varying 

operational parameters such as mass flow rate, temperature and split fractions, while the ITM size is 

kept fixed. More specifically, the degrees of freedom are the mass flow rates of "AIRMCM" and 

"AIRREST", the split fraction of "B10", and the cold side outlet temperature of "BHEX". Varying 

"AIRMCM" varies fuel flow rate since flue flow rate is stoichiometrically related to the amount of 

O2 separated in ITM to ensure complete combustion. 

As the mass flow rate of "AIRMCM" increases, the amount of oxygen separated by a fixed size 

ITM also increases. This corresponds to an increase in fuel flow rate and greater compressor power 

(decrease in efficiency). At the same time, the power output from the turbine increases (increase in 

efficiency). The combination of these opposing effects provides scope for optimization. 

162

As the mass flow rate of "AIRREST" increases, the compressor power and flow rate through the 

turbine increase, thus increasing the power output from gas turbine "GT". At the same time, 

increase in "AIRREST" flow rate, keeping the "B10" split fraction constant, decreases the input 

temperature to the turbine, which decreases the output power. A similar effect is seen for the 

variation of the "B10" split fraction. For fixed "AIRREST" flow rate, increased direct flow through 

the gas turbine results in lower gas turbine "GT" inlet temperatures. However, this results in smaller 

flow rates through "BHEX" and thus, the maximum energy in the heat exchanger is not extracted. 

When the flow to the heat exchanger is large, the air inlet flow temperature to the gas turbine 

decreases. The interplay of these effects emphasizes the importance of optimization. 

As aforementioned, it is advantageous to extract the maximum possible power from the gas turbine 

and transport less thermal energy to the bottoming cycle. Therefore, the air outlet temperature from 

the heat exchanger "BHEX" must be the maximum possible, while satisfying the minimum pinch. 

Fixing the pinch which occurs at the hot end to 10 K, for inlet temperature of combustion products 

1473.15 K, the outlet temperature of air must be 1463.15 K. However, it is seen that as air flow rate 

through the heat exchanger "BHEX" is increased, air outlet temperature begins to decrease after a 

point. Though larger air-flow through "BHEX" is expected to produce more power in gas turbine 

"GT", other deteriorative factors such as a higher air compression power required, and lower air 

outlet temperature in "BHEX" also come into play. It is seen from the optimization results of Table 

1, the local optimum occurs at lower flow rates, with maximum possible air outlet temperature from 

"BHEX". 

The amount of "GT" turbine blade cooling required is chosen according to performance maps [11 

Chart 5.16] and is specified as an optimization constraint. Performance maps specify the amount of 

cooling air required based on the turbine inlet temperature. As the optimizer searches for a local 

optimum, the turbine inlet temperature varies. Therefore, a constraint is added to meet the turbine 

blade cooling requirements. The average of the lower and upper limit of this range is chosen for our 

optimization studies. 

Based on the performance map: 

For faster convergence of the ASPEN Plus ® optimizer, the design specifications of the topping cycle 

have been implemented as optimization constraints. The split fractions of "B6" and "B3", which are 

treated as design specification variables earlier, are now treated as optimization variables. This 

means, in addition to the actual optimization constraints, design specifications also become 

optimization constraints, and in the process, design specification variables now become 

optimization variables. The minimum approach temperature for "BHEX" is also treated as an 

optimization variable. 

Table 1. Results of Optimization of Topping cycle 

Variables Units Before Optimization After Optimization 

Molar flow rate of AIRREST kmol/s 5.665 9.28 

Molar flow rate of AIRMCM kmol/s 37.335 49.18 

Split fraction of B6 (BLDPROP) 0.1267 0.1189 

Split fraction of B10 (Stream 2) 0.73 0.7169 

Air outlet temperature from BHEX K 1463.15 1463.15 

ITM ΔP feed/permeate (%) 

ITM Recovery ratio 

Efficiency 

163 

1.1/0.6 

29.1 

25.33 % 

1.5/0.9 

30.51 

26.07%

As seen from Table 1, the mass flow rates of both "AIRMCM" and "AIRREST" have decreased. 

This implies lesser oxygen separation across the ITM and lower fuel intake. Split fractions of "B6" 

vary to attain minimum approach temperature in heat exchanger "LHEX" without any temperature 

crossover in the heat exchanger network "LHEX-ITM-HHEX" (implemented as one of the 

constraints). The air outlet temperature of "BHEX" does not change as 1463.15 K is the maximum 

temperature of air that can be reached for the specified minimum approach temperature of "BHEX". 

The first law efficiency, which is defined as the ratio of power output to the product of heating value 

of the fuel and the fuel flow rate, increases by 0.74 percentage points (from 25.33 % to 26.07 %) as 

a result of the optimization. 

Fig. 4. Variation of efficiency with air mass flow rate through "LHEX" 

Fig. 4 shows the effect of varying the air flow rate through the heat exchanger "LHEX" on the 

efficiency of the topping cycle, while the topping cycle is optimized by changing the other variables 

with the solution of the previous run as the initial guess value for the present run. Clearly, in the 

range 30 - 55 kmol/sec, only one local optimum exits and hence the point that has been reported has 

the maximum efficiency, although this does not prove that it is the global optimum. Thus global 

optimization is capable of improving the cycle efficiency even higher. 

3.4.2. Optimization of Bottoming Cycle 

The bottoming cycle is a triple pressure HRSG. The input specifications of the streams "GTEHX" 

and "PRODBOTM" are specified using the results obtained from the optimized top cycle, and the 

bottoming cycle is then independently optimized. 

The variables considered for optimization of the bottoming cycle include the three pressure levels, 

the discharge pressure of the turbine and the condenser pump, and the split fraction of the three 

splitters. Lowering the temperature of stream "EXHEXIT" which leaves the heat exchanger 

"ECON" and increasing the temperature of streams "HPSTM" and "IPSTM" which exits the heat 

exchanger "HPSP" increases the efficiency as the heat input to the bottoming cycle increases. Thus, 

the outlet temperature of the heat exchangers "ECON" (stream "EXHEXIT") and "HPSP" (streams 

"HPSTM" and "IPSTM") are increased and decreased respectively, to the maximum possible extent 

such that there are no temperature crossovers in any of the heat exchangers. At the optimal point, 

the pinch value for heat exchangers "ECON" and "HPSP" are observed to be 4.8 K and 4.4 K 

respectively. This can be done only for the AZEP 100 as "GTEXH" is pure air without any CO2 

emissions. This is not true in the case of partial emission cycles. For partial emission cycles, there is 

a limit on "ECON" outlet temperature since low temperatures can cause acid condensations. 

164

The discharge pressure of "LASTST" is fixed to be equal to the saturation pressure for an 

atmospheric temperature which is assumed to be 25 °C. 

Unlike the topping cycle, implementing design specifications as constraints does not work well for 

the bottoming cycle. It is found that implementing design specifications and optimization 

constraints separately for the bottoming cycle makes the optimizer convergence easy. The 

optimization constraints include vapor quality of the outlet stream from the de-aerator be equal to 

zero and the inlet pressure of the valve "B27" be greater than its outlet pressure. 

Table 2. Results of Optimization of Bottoming cycle 

Variables Units Before Optimization After Optimization 

Outlet pressure of HPPMP bar 100 104.9 

Outlet pressure of IPPMP bar 25 24.5 

Outlet pressure of LPPMP bar 5 8.6 

Discharge pressure of LPST bar 0.3 0.33 

Outlet pressure of CONDPUMP bar 0.2 0.29 

Split fraction of B29 (LPFW) 0.1663 0.2025 

Split fraction of B29 (LPIPFW) 0.083 0.060 



Outlet Temperature of air from ECON K 400.4 381.15 

Outlet Temperature of HPSTM from HPSP K 460 501 

Outlet Temperature of IPSTM from HPSP K 460 485 

Efficiency 23.47 % 25.65 % 

Optimization of the bottoming cycle increases its efficiency (bottoming cycle efficiency is defined 

as the ratio of power output from the bottoming cycle to the product of fuel flow rate and heating 

value) from 23.47 % to 25.65 % - an increase of 2.18 percentage points. The total efficiency of the 

power plant thus increases by 2.92 percentage points. This is a significant improvement in the 

efficiency, which plays an important role in determining the feasibility of AZEP cycles, see also the 

discussion in the next section. A summary of results from the optimization of the topping and 

bottoming cycles is shown in Table 3. 

Table 3. Summary of Optimization of Top and Bottoming cycle 

Efficiency Before Optimization After Optimization Increment in Percentage points 

Top cycle 25.33 % 26.07 % 0.74 

Bottoming cycle 23.47 % 25.65 % 2.18 

Entire power plant 48.8 % 51.72 % 2.92 

165

4. Conclusion 

Fig. 5. Partial Emission ITM cycle analysis 

Reference [10] gives a detailed analysis of partial emission cycles along with a new metric (see Fig. 

5) which compares plants at the fleet level. Different variations of partial emission ITM cycles have 

been proposed to increase the efficiency. Improving the efficiency of a partial emission cycle by a 

small fraction is not very useful if the CO2 emissions simultaneously increase by a large amount. 

Fig. 5 shows the first law efficiency and specific CO2 emissions for different cycles which have 

been studied in the literature [6, 12-24]. Theoretically it is possible to get 100% CO2 separation for 

AZEP 100 cycle with large number of compression and cooling stages in the CO2 compression and 

purification unit, but this involves unreasonable capital costs. Therefore, practically, CO2 

compression and purification units are not 100% efficient as seen in Fig. 5, which shows a small 

CO2 emission for the AZEP 100 cycle simulation result. A linear combination of the AZEP 100 and 

a combined cycle is also shown for comparison. In Fig. 5, AZEP 100 cycles from literature assume 

100% CO2 separation and hence have zero CO2 emissions .The black dotted line represents the 

stipulated locus of partial emission cycles from AZEP 100 to AZEP 72 [10]. The green solid line 

represents the linear combination of a zero emission cycle and a combined cycle with an efficiency 

of 65 %. The green line shows the efficiency and the specific CO2 emission for different 

combinations of the AZEP 100 and the best efficiency reported for a combined cycle without carbon 

capture. For the partial emission cycle to be feasible, it should have better efficiency than the 

combination of an AZEP 100 and a combined cycle for the same specific CO2 emission [10]. This 

implies that for the partial emission cycles to be considered feasible, they must lie above the green 

line, which is not the case for virtually all partial emission cycles proposed. As can be seen from 

Fig. 5, after optimization the AZEP 100 cycle lies above the green line. This means after 

optimization the AZEP 100 cycle which nominally has complete capture (zero CO2 emission) is 

viable even after the imperfections of the CO2-H2O separation is accounted for this cycle, but not 

for the literature results. Note that fuel-cell based processes can achieve substantially higher 

performance, but require completely different technology. The improvements obtained for the 

nominally zero emissions cycle herein, suggest that optimization of partial emission cycles may 

increase their efficiency to a significant extent and make them viable. 

166


The authors would like to thank the King Fahd University of Petroleum and Minerals (KFUPM) in 

Dhahran, Saudi Arabia, for funding the research reported in this article through the Center for Clean 

Water and Energy at Massachusetts Institute of Technology (MIT) and KFUPM under project R2- 

CE-08. 

References 

[1] I.E.A., CO2 emissions from fuel combustion, IEA Statistics, 2011. 

[2] Steenveldt R., Berger B, Torp T., CO2 capture and storage: closing the knowing-doing gap. 

Chemical Engineering Research and Design 2006;84(9):739-63. 

[3] Wall TF., Combustion processes for carbon capture. Proceedings of the Combustion Institute 

2007;31(1):31-47. 



Energy Procedia 2009;1(1):495e502. 

[5] Gellings PJ., Bouwmeester HM., The CRC handbook of solid state electrochemistry. CRC 

Press; 1997. Ch. 14, pp. 16-30. 

[6] Sundhvist S., Julsrud S., Viegeland B., Naas T., Budd M., Winkler D., Development and testing 

of AZEP reactor components. In proceedings of International Journal of Greenhouse Gas 

Control, 2007,1,180 -187. 

[7] Mancini ND., Mitsos A., Ion transport membrane reactors for oxy-combustion-Part I: 

intermediate fidelity modeling. Energy 2011;36(8):4701-20 

[8] Mancini ND., Mitsos A., Ion transport membrane reactors for oxy-combustion-Part II: analysis 

and comparison of alternatives. Energy 2011;36(8):4721-39 

[9] Numerica Technology, JACOBIAN Modeling and Optimization Software – Available at: 

[accessed 8.01.2009]. 

[10] Mancini ND, Mitsos A., Conceptual design and analysis of ITM oxy-combustion power 

cycles. Phys. Chem. Chem. Phys., 2011, 13, 21351-21361 

[11] P. P. Walsh, P. Fletcher, Gas turbine performance, Blackwell Science, Malden Mass., 2004 

[12] Colombo, K.E., Bolland, O., Kharton, V.V., Stiller, C., Simulation of an oxygen membrane 

based combined cycle power plant: Part-load operation with operational and material 

constraints, Energy and Environmental Science 2009, 2 (12), pp. 1310-1324. 

[13] Kvamsdal H., Jordal K., Bolland O., A quantitative comparison of gas turbine cycles with 

CO2 capture Energy, 2007, 32, 10–24. 

[14] Liu S., Gavalas GR., Membr J., Oxygen selective ceramic hollow fiber membrane. Journal of 

Menbrane Science., 2005, 246, 103–108. 

[15] Anantharaman R., Bolland O., Asen K., Novel cycles for power generation with CO2 capture 

using OMCM technology. Energy Procedia, 2009, 1, 335-342. 

[16] Yantovski E., Gorski J., Smyth B., J. Elshof., Zero-emission fuel-fired power plants with ion 

transport membrane. Energy, 2004, 29, 2077-2088. 

[17] Fiaschi D., Gamberi F., Bartlett M., Griffin T., The air membrane -ATR integrated gas 

turbine power cycle: A method for producing electricity with low CO2 emissions. Energy 

Conversation and Management, 2005, 46, 2514-2529. 

[18] Pfaff I and Kathe A., Comparative thermodynamic analysis and integration issues of CCS 

steam power plants based on oxy-combustion with cryogenic or membrane based air separation. 

Energy Procedia, 2009, 1, 495–502. 

167

[19] Yantovski E., Shokotov M., Shokotov V., McGovern J., Kirsten Foy., Elaboration of Zero 

Emissions Membrane Piston Engine System (ZEMPES) for Propane Fuelling. Proceedings of 

Fourth Annual Conference on Carbon Capture and Sequestration DOE/NETL; 2005, May 2-5. 

[20] Walker ME., Abbasian J., Chmielewski DJ., Castaldi MJ., Dry Gasification Oxy-combustion 

Power cycle. Energy Fuels 2011, 25, 2258-2266. 

[21] Dillon DJ., Panesar RS., Wall RA., Allam RJ., WhiteV., Gibbins J., Haines MR., Oxycombustion 

processes for CO2 capture from advanced supercritical PF and NGCC plant. 

Proceedings of seventh conference on greenhouse gas control technologies (CHGT-7), 

Vancouver, BC, Canada, 2004 

[22] Lozza G., Romano M., Giuffrida A., Thermodynamic Performance of IGCC with Oxy- 

Combustion CO2 capture. In proceedings of First International Conference on Sustainable 

Fossil Fuels for Future Energy - S4FE 2009. 

[23] Gou C., Cai R., Hong Hui., A novel hybrid oxy-fuel power cycle utilizing solar thermal 

energy. Energy 32 (2007) 1707-1714 

[24] Tan. X., Wang Z., Meng B., Meng X., Li K., Pilot-scale production of oxygen from air using 

perovskite hollow fiber membranes. J. Membr. Sci., 2010, 352, 189-196 

[25] Bertsekas DP., Nonlinear Programming. Athena Scientific; 1999. 

[26] Petrakopoulou F., Tsatsaronis G., Morosuk T., Exergoeconomic Analysis of an Advanced 

Zero Emission Plant. Journal of Engineering for Gas Turbines and Power , 2011, 133, 113001- 

12 

168

Abstract: 




Implementation of a CCS technology: the 

ZECOMIX experimental platform 

Calabrò Antonio a , Cassani Stefano a , Pagliari Leandro a 

Stendardo Stefano a 

a ENEA, Italian National Agency for New Technologies, Energy, and the Sustainable Economic 

Development Via Anguillarese, 301, S. Maria di Galeria, 00123, Rome, Italy 

One of the key challenges for the implementation of CO2 Capture and Storage (CCS) is the reduction of the 

CO2 capture costs derived by their applications in actual power plants. The research of new technologies 

based on more efficient materials and more efficient design for integration of CO2 capture technologies in 

power cycles, is a promising way to ensure, in the medium term, costs and energy performances comparable 

to the actual power plant without CCS. ZECOMIX experimental platform represents the ENEA proposal for 

this challenge and it can be count among the Research Infrastructures more innovative in Europe. The aim 

of this work is to present the first results of commissioning plant tests. Therefore this work is focused on the 

planning of future experimental activities in order to demonstrate the feasibility of the high temperature solid 

sorbent CO2 sorption process, applied to a syngas derived from coal gasification. 

Keywords: 

Keywords: Ca-based sorbent, carbon capture, solid chemical looping, gasification. 


In July 2005 the activities of the project ZECOMIX (Zero Emission Coal Mixed Technology) 

started officially, within the framework of a program funded by the Italian University and Research 

Minister. The main aim of the project is to demonstrate, through a series of modelling and 

experimental activities, the feasibility of a new innovative process for production of electricity and 

hydrogen "zero emission" by coal. In the ZECOMIX project different processes (e.g. coal 

gasification, clean-up of syngas, capture and sequestration of CO2, combustion of hydrogen in gas 

turbine) have been coupled. The integration of these processes is the key factor to achieve higher 

cycle performance and optimization (fig.1) of the platform at hand. Preliminary studies on high 

temperature decarbonization syngas [1-3] and thermodynamic H2/O2 cycle [4,5] as led us to 

propose ZECOMIX cycle with a net electrical efficiency close to 50% [6]. The analysis was 

subsequently refined with the cooperation of major Italian Universities and Research Institutes, and 

the results have fully confirmed the encouraging prospects that the cycle had announced: at least 

ten-point performance higher than a current post-combustion capture technology coal plant [4,5]. 

The design and realization of a complex experimental platform to test new process for CO2 capture 

and hydrogen production was certainly the more challenging tasks from a technical and financial 

point of view. A large numbers of advanced research issues is related to the proposed platform (coal 

hydro-gasification, simultaneous high temperature CO2 and H2S capture with solid sorbents, 

hydrogen/steam fuelled micro-turbine). Preliminary studies have permitted us the commencement 

of several research lines in the fields of advanced system modelling [2,6] for testing coal 

gasification, and the study of materials for CO2 capturing [1-2]. On the other hand the great knowhow 

gained with years allows us to enter with a leading role in the most important European 

scientific organizations, in order to coordinate the research on CCS technologies considered the 

most promising option for sustainable use of fossil fuels in the near future. The experimental 

ZECOMIX pilot plant has been designed and constructed to test each single unit integrated in the 

platform. It is an advanced and flexible facility particularly oriented to the experimental 

169

investigation and the developing of process and component modeling. The plant has been presented 

in an international framework (e.g. CSLF, ZEP, ECCSEL, EERA) and consistent with advanced 

technological initiatives of the European and worldwide scientific community. Moreover 

ZECOMIX plant has been evaluated and inserted in the first Italian Roadmap towards the Large 

research Facilities conceived by the Ministry of Research. 

2. Conceptual description of the ZECOMIX platform 

Details of the ZECOMIX concept are presented in [6] whereas alternative configurations are 

proposed in [4,5] . In Fig 1(a) we have reported the main areas of the platform and how they are 

connected each other; in Fig 1(b) the layout of the platform constructed is showed. This work 

highlights our efforts in the realisation of the plant layout showed in Fig 1(b). The plant (see Fig 

2(a)) can be broken into : 

Unit based on a 50 kg/h coal fluidised bed; 

Syngas decarbonising reactor. It has been conceived as a fluidised bed and designed in order to 

decarbonise 100 Nm3/h of coal syngas leaving the gasifier or a gaseous mixture produced by 

means adequate cylinders; 

100 kWe micro-turbine modified to accommodate a mixture of H2 and H2O. 

Fig. 1. (a) Simplified schematic of the ZECOMIX concept; (b) Layout of the ZECOMIX platform. 

A number of devices are installed to further treat the syngas (e.g. scrubbing, drying and 

compression) to make it suitable as fuel for the micro-turbine. Moreover a steam generator has been 

170 

a 

b

envisaged for operating the gasifier, the carbonator and the micro-turbine as well. The plant has 

been designed to test several experimental configurations: 

Gasification test: the yielded syngas is diverted directly to the scrubbing device and then to the 

flare stack; 

Carbonator test: the syngas is produced by mixing pure gases leaving bottles. This test allows us 

to investigate the decarbonising process over a number of syngases or flue gases; the inlet 

temperature of the syngas could be set up to 600 °C by means an adequate electric heater; 

Micro-turbine test: in this test the turbine is fuelled with the syngas produced at the mixing unit; 

Base-case test: such a test is carried out with the syngas produced by means of the gasifier and 

driven to the carbonator. Having completed the decarbonising process the syngas undergoes a 

scrubbing process and fed the micro-turbine or burnt in the flare stack. When the micro-turbine 

runs the syngas is compressed up to 6 bar by means of a compressing device and an amount of 

water steam is injected into the micro-turbine’s combustion chamber to moderate operating 

temperature. Such a device has been already tested by the Italian company “Ansaldo Ricerche” 

retrofitting an existing micro-turbine named Turbec T100; 

Regeneration test: in this configuration the carbonator operates in regeneration mode firing by a 

number of adequate burners placed at the bottom of the reaction chamber to heat the sorbent up 

to the regeneration temperature. 

The layout of experimental configuration and the control of the whole plant is carried on by a 

particular Distributed Control System (DSC) interfaced software through several synoptic schemes 

with regard to the plant configuration (Fig 2(b)). 

a b 

Fig. 2. (a) Picture of the ZECOMIX platform; (b) Control station of the platform. 

3. General description of the platform main components. 

The detailed design of the plant started in early 2007 and it was realized with the help of Ansaldo 

Energy, and the first supply contracts were launched in April 2008. Civil infrastructures are 

composed of plant foundations, storage for solid material (dolomite, catalyst and coal), security 

bunker for H2 and CO gas cylinders, platforms of CO2, N2 and O2 gas tanks, and the basement of 

gas mixing station. Moreover, a control room building has been realized which includes the 

electrical cabinets of main equipments and the DCS. Simultaneously to the progress of civil works, 

the order of main equipments of the plant were carried out: carbonation reactor, fluidized bed 

gasifier, steam generator, syngas compressor, gas pre-heaters, gas clean up scrubber, flare stack, 

and, a micro-turbine adapted to high content hydrogen syngas. 

Gasifier: Hydrogasification reactor as presented in [6] operates at pressure up to 60 bar, 

incompatible with the project design atmospheric pressure; thus, it has been replaced by an 

oxygen/steam atmospheric gasifier, more robust and flexible than hydro-gasification reactor (see 

Fig 3(a)). The coal gasifier is a fluidized bubbling bed reactor designed in collaboration with 

171

University of L’Aquila and Germanà&Co. Engineering. The coal feed system has been design in 

order to feed a nominal load of 50 kg/h of coal. The system is formed by a 2 m 3 coal silo, which 

permits a stationary operation up to 36 hours, and two worm drives: a dosage worm drive in order 

to control the coal mass flow, and a second one to introduce mechanically the coal to the interior of 

gasifier. Steam and oxygen are feed on different points of the reactor, in order to control the solids 

hydrodynamics, and the reaction rate all over the reactor. Dolomite is added to the coal enhancing 

the fluidization, controlling temperature and capturing the H2S formed in the coal gasification. A 

syngas composed of H2, CO, CO2 and steam at temperature of 800°C is obtained. This syngas is 

sent to a regenerative heat exchanger reducing its temperature around 600°C. After this point can be 

introduced into the carbonation reactor, or can be clean-up in a scrubber, after a second cooling step 

to 350°C. Methane is mixed with the yielded syngas in order to emulate syngas composition of a 

syngas leaving hydro-gasification reactor. Moreover water is injected into this gaseous mixture in 

order to carry on steam methane reforming and CO-shift reactions 

Carbonator: The carbonation reactor is a cylinder with 1 m diameter and 4.5 m height cylindrical 

chamber (Fig. 3(b)). Reactor wall have 30 cm thickness and two layers refractory enclosure. At the 

bottom of the reactor there are two burners in order to calcine the sorbent in the regeneration phase 

at 900 °C. The distributor gas plate is situated above the burners and realized as a series of tubes 

placed perpendicularly to the orifices in order to broke the gas jets producing a quasi-homogenous 

velocity field. Moreover, this design allows us to reduce the attrition of nozzles and particles, and 

prevent the clogging of the particles in the orifice. The fluidized bed reactor is loaded with Ni-based 

catalyst, necessary for the steam methane reforming, and Ca-based sorbent in order to capture CO2. 

The addition of the sorbents serves to decarbonize the flue gas and to enhanced the steam methane 

reforming. This process is called Sorption Enhanced-Steam Methane Reforming and allows us to 

obtain a very high hydrogen content syngas, improving the methane conversion at relatively low 

temperatures (550-600°C). When the solid sorbent reaches at its ultimate conversion it is sent back 

to the regeneration step. Regeneration is done by means of oxy-combustion of methane in order to 

calcine the spent sorbent. High-concentrated CO2 stream is released and sent to final disposal. 

Subsequently, a cooling process is accomplished in order to return to initial carbonation condition 

starting another CO2 capture cycle. 

a b 

c 

Fig. 3. (a) Fluidised bed gasifier; (b) Carbonator; (c) Turbec T100 micro-turbine with the modified 

burner. 

Micro-turbine and syngas scrubbing unit: The power generation is produced by means a microturbine 

adapted to high content hydrogen syngas (Fig. 3(c)). Original turbine is a Turbec T100, with 

100 kWe of power output, retrofitting with a hydrogen burner ARI100T2 developed by Ansaldo 

Energy. Flue gas in the turbine outlet are mainly nitrogen and steam, achieving the “ carbon zero 

172

emission” scope of the cycle. Micro-turbine is very sensible to the impurities. Then syngas 

scrubbing unit is fitted upstream the micro-turbine. Therefore, the scrubber is able to treat 400 

Nm3/h in three stages: a Venturi scrubber, a spray tower wet scrubber, and a packed tower wet 

scrubber. The outlet particle diameter is lower than 1 m. The flare stack has been designed to the 

correct combustion of syngas from carbonator reactor or gasifier. 

Steam generator: The steam generator have 200kWt power output, and it is provided with an 

accumulator which permits a high flexibility mass flow production as a function of the 

configuration tests. It is possible to obtain three different mass flows at different pressure for every 

equipment: 30 kg/h at 1 atm for the coal gasification, 90 kg/h at 1 atm to the carbonatator, and 60 

kg/h at 6 bar to the micro-turbine. 

Syngas compressor: The syngas compressor has three intercool centrifugal stages in order to 

maintain hydrogen under 130°C and water is drained. Then hydrogen stream is pre-heated and 

mixed with steam to the turbine conditions. 

Heaters: Three heaters are installed in the platform. An oxygen heater has been fitted in order to 

prevent condensation of water in the oxygen/steam mix. A nitrogen heater is needed to control the 

temperature in the heating/cooling processes in carbonation reactor between carbonation/calcination 

reactions. A heater of the syngas entering the carbonator is installed. Such a component is needed 

when synthetic gas is produced in the mixing station and send to the carbonator. In this case the 

syngas has to be heated from environmental temperature up to the carbonation temperature (600 

°C). 

The platform is being commissioning. Several tests has been carried out in order to prove DCS 

system, steam generator, hydrogen compressor, pre-heater and flare. The main objectives which 

have been accomplished are mainly related to coal gasification, high temperature CO2 capture by 

means of solid sorbents, fluidynamics of bubbling bed, combustion of hydrogen. 

4. Future activities in the framework of the ZECOMIX project 

In order to supply the experimental platform with a number of solid fuels (e.g. biomass and a blend 

of coal) a feeding system as flexible as possible is envisaged. Such a technological option will 

permit us to study the performances of the hydrogen production from solid fuels by varying the 

ratio of the different blends. Particularly the co-gasification of coal and biomass could be 

experienced and the main parameters affecting such a process will be studied. Such an experimental 

investigation will give information on how to mix coal and biomass and how the quality of the 

feedstock affects the production of electricity from hydrogen. Implementation of new sensors and 

probes (thermocouples and flux-meters) are envisaged. This improvement will permit us to increase 

the potential of the acquisition system and then the remote control of the process at hand with online 

monitoring and dynamic measurement. Particularly during the warm up of decarbonisation 

reactor the control of the refractory temperature is needed in order to know the evolution of the 

temperature profile into the refractory wall of the reactor and the characteristic time for heating up 

the reactor. In order to manage the energetic system at hand in an economically viable way, 

optimization of hydrogen and electricity production should be taken into account. These outputs of 

the experimental platform strongly depend on the correct blend of both feedstock and energy inputs 

(e.g quality and quantity of the solid fuel to be gasified and heat demand of calciner for sorbent 

regeneration as well). There is a lack of dependable economic and operative parameters, due to the 

intrinsic novelty of the proposed process. A valuable tool for addressing such a problem could be 

modeling each single main component (gasifier, carbonator, micro-turbine) to simulate and 

optimize ZECOMIX system. When thermo-economic constraints are included the optimization 

problem will increase significantly. Moreover if the number of variables governing the hydrogen 

and power production is relatively low an adequate and accurate DoE (design of experiment) is 

suggested as a valuable tool to determine how the main parameters affect the electricity production. 

Finally, the activities scheduled for the power unit at hand have been proposed in order to test the 

173

hydrogen combustion technology at different scale levels and grades of integration with the whole 

experimental platform. In particular a dynamic model of the micro-turbine has been developed and 

integrated in a commercial power plant simulator. Such a model is a valuable tool for the simulation 

of a the micro-turbine in a number of operative conditions. Numerical simulation are needed to 

estimate the dynamic behavior of the power unit when the fuel supply switches from natural gas to 

hydrogen during full and half load as well. The experimental programmed activities have been 

enthusiastically welcomed by a number of European projects and research alliances to promote the 

co-operation between researchers. ZECOMIX project is involved in the ECCSEL initiative in the 

framework of CCS research facilities to share knowledge through network of researchers with the 

common interest to improve the assessment methodology for decarbonised energy production. 

Moreover the platform is involved in the EERA sub-programme for developing a methodology and 

comparing process performances in the field of CO2 capture.A complete schedule of activities has 

been planned are: 

1. hot tests on carbonator in order to study heat transfer in the gas-particle-wall system and start-up 

procedure. Modeling activities are planned to predict the thermal behavior of the carbonator 

reactor. Start-up of H2 micro-turbine will be performed as well. 

2. tests of SE-SMR reactor fed with synthetic syngas from cylinders. Performance of CO2 capture 

sorbents and catalyst. Study of instabilities in the process between calcination and carbonation 

phase. Integration with micro-turbine. Optimizing of operational conditions. 

3. carbonation/calcination cycling with gasification integration. Studies of influence of gasification 

on flue gas composition. Effect of gasification on hydrogen production. Optimization of the 

process. Study of loss efficiency due to clean-up of gases. 

Acknowledgment 

The authors are grateful to Prof. Pier Ugo Foscolo and Prof. Antonino Germanà, of University of 

L’Aquila, for carbonator and gasifier design, and SO.IM.I. company for constructing the 

experimental platform. 

References 

[1] Stendardo S., Calabrò A., Experiments on CaO–CaCO3 loop for high temperature CO2 capture. 

ENEA Technical Report 2008 No.: EHE08039 TER-ENEIMP. 

[2] Stendardo S., Foscolo P.U., Carbon dioxide capture with dolomite: A model for gas–solid 

reaction within the grains of a particulate sorbent. Chem Eng Science 2009;64:2343-2352. 

[3] Stendardo S., Di Felice L., Gallucci K., Foscolo P.U., CO2 capture with calcined dolomite: the 

effect of sorbent particle size. Biomass Conv and Biorefinery 2011;1(3):149-161. 

[4] Galeno G., Spazzafumo G., ZECOMIX: Performance of alternative layouts. International 

Journal of Hydrogen Energy 2010;35:9845-9850. 

[5] Romano MC, Lozza G. 2010. Long-term coal gasification-based power plants with near-zero 

emissions. Part A: Zecomix cycle. Int Journal of Greenhouse Gas Control 2010;4:459-468. 

[6] Calabrò A., Deiana P., Fiorini P., Girardi G., Stendardo S., Possible optimal configurations for 

the ZECOMIX high efficiency zero emission hydrogen and power plant. Energy 2008;33:952- 

962. 

174

Abstract: 




Influence of regeneration condition on cyclic CO2 

capture using pre-treated dispersed CaO as high 

temperature sorbent 

Stendardo Stefano a , Calabrò Antonio a 

a ENEA, Italian National Agency for New Technologies, Energy, and the Sustainable Economic 

Development Via Anguillarese, 301, S. Maria di Galeria, 00123, Rome, Italy 

In this experimental investigation the effect of calcination temperature and atmosphere composition on CO2 

uptake of a solid sorbent have been analysed. The sorbent were synthesized by means of a CaO hydrolysis 

technique to generate sorbents with 75% and 85% of active phase CaO. The material at hand also contains 

a calcium aluminate phase acting as a binder of the active phase. Pre-treatment was accomplished in a 

thermo-gravimetric analyser (TGA) exposed in an atmosphere of 86% N2 and 14% CO2 under 600 °C. The 

as-synthesised sorbent and the pre-treated sorbent have been characterised by scanning electron 

microscope, nitrogen physic-sorption tests, and multi-cycling carbonation-calcination test in TGA (160 

cycles). Here, the CO2 uptake took place at programmed temperature (600 °C) with three different 

regeneration condition tested: 

a) mild condition: regeneration under 900 °C with 14% CO2 and 86 % N2; 

b)moderately severe condition: regeneration under 1000 °C with 14% CO2 and 86 % N2; 

c) severe condition: regeneration under 1000 °C with 86% CO2 and 14 % N2. 

The experimental results show significant improvement in the stability of the CO2 uptake capacity over 

multiple cycles when comparing the synthetic sorbents to natural dolomite. For an instance, the 75% CaO 

synthetic sorbent shows a good reversibility for the CO2 uptake (0.13 g-CO2/g-sor) up to 150th cycle under 

severe condition. 

Keywords: 

Keywords: solid sorbent, carbon capture, carbonate chemical looping, self-reactivation. 


Carbon capture technologies are expected to be a promising route to meet the objective of a low 

carbon electricity production and the increasing of coal foreseen by the scientific and industrial 

community. Hydrogen production from renewable energy sources, coal and biomass is a priority for 

Italian medium and long term energy policy. Thus, in Italy technologies for CO2 capture are 

considered a main topic to be studied and demonstrated, and represent a significant opportunity for 

industries. In order to promote carbon capture technologies, ENEA has constructed an experimental 

platform, named ZECOMIX, to investigate both the gasification of coal and the separation of CO2 

from synthetic gas fuel or flue gas. Particularly the decarbonisation of the gaseous stream happens 

by means of a carbonate looping (CaL) technology with a CaO based solid sorbent. When the CaO 

is converted to the calcium carbonate, the spent solid sorbent is sent back to the regeneration 

process where an active sorbent is regenerated for a new carbonate looping. As reported in the 

scientific literature: [1-4], when naturally occurring material as calcite or dolomite are used as CO2 

acceptor in a CaL, there is a decay of reversibility. The ideal CO2 sorbent in a CaL should show a 

number of properties: high and stable CO2 uptake capacity throughout continuous decarbonisingregeneration 

cycling, fast reaction kinetics, uptake capacity and kinetics close to theoretical 

maximum values, and also mechanical stability and sintering resistance. In an attempt to achieve 

this goal, researchers have developed novel synthetic sorbents based on e.g. CaO dispersed on 

calcium aluminate ceramic supports. [5]. In this work a number of experimental results on the 

175

eversibility of a pre-treated synthetic solid sorbent are reported. In particular the uptake of CO2 

through a multi-cycling carbonation-regeneration has been evaluated. The main goal of this work is 

analysing the influence of regeneration of the tailored material on its performance. Recent papers 

[6-7] have demonstrated that pre-treating dolomite or calcite increases stability and reversibility of 

the material. However those experimental tests were focused on pre-treating naturally occurring 

sorbent while material regeneration occurs in 100% nitrogen atmosphere. As the aim of CO2 

capture is obtaining a high-concentrated CO2 stream, there is a need to investigate the behaviour of 

solid sorbent when the regeneration step is conducted in presence of carbon dioxide. Then the 

material at hand was regenerated in high concentration of CO2 near to realistic condition required 

for CCS technologies and compared with that regenerated in low concentration of CO2. Besides the 

influence of regeneration condition on the subsequently CO2 uptake, the aim of this work is to see 

whether the positive effect of heating pre-treatment persists also for the synthetic sorbent presented 

here. 

2. Material and method 

2.1. Synthesis of solid sorbent 

The synthesis was performed according to [5] where powdered CaO (about 99.8%, after 

calcination) and aluminum nitrate Al(NO3)3·9H2O (> 98%) were used as precursors in the synthesis 

of the CaO-Ca12Al14O33 sorbent. A wet method was used to ensure an intimate mixing of starting 

materials: Distilled water was used containing 2-propanol as surfactant. CaO was calcined at 900°C 

for 2 h in the presence of air to remove humidity and decompose any traces of CaCO3 into CaO. 

The amounts of CaO and aluminium nitrate were chosen such that the mass ratio of CaO to binder 

phase was 75:25 and 85:15. The compounds were added to the water and the mixture was stirred at 

75 °C and 700 rpm. After 60 minutes stirring, the solution was dried at 120 °C for 18 h to obtain a 

dried cake. In order to form the binder Ca12Al14O33 the material was ground and heated up to 850 °C. 

Before reaching the temperature for Ca12Al14O33 formation the sorbent precursor was maintained at 

500 °C for 180 minutes to evaporate nitric oxides and produce Al2O3 in a controlled mode. After 

cooling to room temperature the material was ground again in a mortar (at this stage some water 

could be added) and heated up to 850 °C for 90 minutes in order to react Al2O3 with CaO to form 

Ca12Al14O33. The solid-solid reaction to form Ca12Al14O33 will require several hours, but the 

reaction rate can be improved by second grinding of the manufactured material as specific surface 

area is maintained higher and fresh surface is brought in contact. After having completed the final 

calcination and the subsequent cooling to the room temperature the sorbent material was grounded 

and sieved. The powder used in this investigation had particle sizes in the range 180 to 500 µm. 

2.2. Characterisation of solid sorbent 

A number of experimental investigation at a lab scale have been performing to characterise a 

synthetic solid sorbent to accomplish the uptake of CO2 from a gaseous mixture. The reactivity and 

CO2 uptake capacity during cycling of the sorbent were analysed by using a GC-10 Mettler-Toledo 

thermo-gravimetric analyzer (TGA). This apparatus can measure minute mass changes of solid 

samples placed in a furnace with a variable and well-controlled temperature and gas atmosphere. 

Blank runs were conducted with an empty crucible to record the disturbances in the mass change 

readings when moving the experiment from calcinations to carbonation process. To avoid the effect 

of the sample size on carbonation and regeneration processes, such as the external mass transfer 

resistance of CO2 through the sample, a ~2.90 mg samples were used. The experimental protocol 

used for performing the solid chemical looping consisted of three steps: 

1) Regeneration phase. This phase is conducted by heating the sample up to a programmed 

temperature at 100 °C/min rate (maximum heat rate allowed by this experimental apparatus). A 

mixture of nitrogen and carbon dioxide flows over the sample. During this phase the CaCO3 is 

176

decomposed into CaO. The sample dwells for 15 min at 1000 °C. Three representative 

regeneration condition were used: 

o Mild condition: temperature 900 °C, atmosphere composition 14/86 %v (CO2/ N2); 

o Moderately severe: temperature 1000 °C, atmosphere composition 14/86 %v (CO2/ N2); 

o Severe: temperature 1000 °C; atmosphere composition 86/14 %v (CO2/ N2); 

2) Cooling and carbonation phase. The temperature is then lowered with the rate of 100 °C/min 

to reach 600 °C selected for CO2 sorption. The atmosphere remains 84% v/v nitrogen and 16% 

carbon dioxide. When the temperature goes below 700-750 °C the CO2 begins reacting with 

CaO to form CaCO3. 

3) Isotherm carbonation phase. Having completed the cooling phase, the temperature is 

maintained constant at 600 °C for 20 min exposing the solid specimen to a reacting atmosphere 

composed of 25 % CO2 and 75 % N2. 

As reported earlier here, the synthesised material was thermal pretreated to see whether the 

advantage found for the dolomite and calcite [6-7] persists on the investigated material. Thus solid 

specimens have been dwelling at 600 °C for 80 minutes, exposing it to a controlled atmosphere 

composed of 14 % O2 and 86 %N2. Then the sample was pre-treated in the TGA and subsequently 

undergone the multi-cycling regeneration-carbonation process to avoid influences on the 

investigated specimen (e.g reaction between sample and CO2 in the room atmosphere). Comparison 

between as synthesised sorbent and thermal pretreated sorbent are presented in this work. 

Particularly the capacity to retain carbon was investigated by cycling the material up to 150 cycles. 


Fig. 1 (a-b) show conversion-time curves for different numbers of regeneration-carbonation cycles 

regenerated in moderately severe condition. At the beginning of each single cycle, besides a short 

nucleation period, a linear kinetically-controlled mass growth was found followed by a transition to 

a much slower reaction rate. This transition was found to be smoother when compared with that 

occurred during the carbonation of naturally occurring sorbent investigated in [3]. In other words, 

no plateau is reached during the last slower phase and the surface reaction continues to play a key 

role in the whole process. Such a smooth transition was found in [6] where experimental test on 

CO2 capture for pre-treated limestone were presented. Moreover, Fig 1(b) shows that with increase 

in cycle number the initial slope of the conversion-time curves increases from the first to the tenth 

cycle. The augment of the slope could likely be explained with the migration of the CaO grains 

from the core of the sorbent particle to the outer. As a consequence more specific calcium oxide 

surface would be exposed enhancing the reaction between active phase (CaO) and CO2. The 

increase of the sample weight with first cycles could likely explained with the phenomenon, named 

self-reactivation, previously found in [6]. Analogous effect was observed also in the field of 

chemical looping combustion where a particular O2-carrier showed a gain in reactivity during 

reduction/oxidation cycle performed in a thermo gravimetric analyzer. After the 10 th cycle the linear 

growth shows the same slope up to the 80 th cycle carbonation-regeneration loop of the experimental 

test. Moreover After that cycle, as reported in Fig 1(a), TG curves were found to be similar denoting 

a good reversibility through the multi-cycling CO2 capture. In addition, Fig 1(a) shows that with the 

increase of cycling up to 10 th carbonation-regeneration step the sample mass achieved after each 

single loop was increased when compared with the previous loop. In Fig 2 the self-reactivation 

phenomenon is presented in term of the uptake of CO2 referred to the initial sample weight. In order 

to compare the self reactivation of synthetic sorbent to the performance of naturally occurring 

sorbent, dolomite sample (Bianchi dolomite: 55.6% CaCO3 and 44.2% MgCO3) was selected and 

exposed to the same condition reported in Fig. 1. 

177

Fig. 1. TG curves collected for specimens subjected to moderately severe regeneration up to 80 

cycles: (a) mass growth shows no plateau; (b) self reactivation effect: the slope of initial linear 

mass growth increases with the number cycle. 

In particular, the uptake of the two sorbents for each single cycle was evaluated according to the 

following ratio: 

uptake = Δm m 

(1) 

CO2 Nth 

0 

where ΔmNth represents the mass augment of the solid specimen at each the N-th cycle and m0 

represents the weight of the sample inserted in TGA. 

Uptake [g-CO2/g-sor] 

0,25 

0,20 

0,15 

0,10 

0,05 

0,00 

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 

Cycle number [-] 

Fig. 2. Comparison between 75 % CaO synthetic sorbent and “Bianchi” dolomite in term of 

carbon retain capacity. 

178 

75/25 synthetic 

sorbent

Even if the dolomite shows a larger uptake when compared to that of synthetic sorbent, a decay in 

the capacity to retain carbon during the first few cycles was found for Bianchi dolomite whereas the 

synthetic sorbent shows higher uptake while the material is cycled during CO2 capture tests. As 

shown in Fig 3, the self reactivation was observed also for the 85% CaO sorbent. Interestingly, 

during the first cycles the uptake of CO2 for the 85% CaO sorbent is greater to that of 75% CaO 

sorbent because there is more active phase (CaO). But as the cycling runs the pore plugging occurs 

likely due to the reduced amount (15%) of binder leading to more CaO being less dispersed and 

thus more inaccessible to CO2. 

Uptake [g-CO2/g-sor] 

0,20 

0,15 

0,10 

0,05 

0,00 

0 20 40 60 80 

Cycle number [-] 

Fig. 3. Effect of CaO load: increasing the load of active phase a decreasing of capacity was 

observed 

In fact during repeated cycling the grains grow and sinter together via formed 'necks' as reported in 

Fig 4. You can see the genesis of a neck structure for a 75% CaO sorbent. The circle (a) focuses the 

neck between two different bodies, at a later stage the neck increases its cross section (b) and when 

the neck reaches at its maximum size the two bodies merge together in a single body (c). This 

structural alteration could lead the blockage of pores and the formation of isolated volumes 

throughout the particle and it is likely an important factor leading to reduced uptake kinetics and 

capacity. 

Fig. 4. Micrograph of sorbent after 60 carbonation/regeneration cycle 

75% CaO 

sorbent 

Improvement of sorbent activity was also observed for the sorbent obtained by chemical 

pretreatment of the sorbent precursor. By adding water after the second grinding (see Sec 2.1) the 

obtained sorbent shows greater uptake when compared with the sorbent whose precursor was dry 

179

(see Fig 5). Carbon dioxide uptake of the former sorbent was ~94% higher that the latter at the 

second cycle. In subsequent cycles the uptake was at least ~60% higher than that obtained with dry 

precursor. 

Fig. 5. Comparison between material obtained with hydrated precursor and dry precursor when 

subjected to moderately severe regeneration. 

Such an improvement could be likely be explained with the formation of calcium hydroxide in the 

sorbent precursor due to the presence of water and the subsequently calcination (for the binder 

formation see Sec 2.1) leaving more pore volume. In fact the molar volume of Ca(OH)2 is greater 

than that of CaO. Thus when the hydrated precursor would undergo calcination process water vapor 

would be formed and it would leave the particle producing extra pore volume. Moreover the 

migration of water vapor towards the outer part could likely create cracks throughout the sorbent 

particle exposing more specific surface area to the carbon dioxide. Consequently the specific 

surface area is higher and more CaO surface is brought in contact with the CO2. Moreover as Fig 6 

suggests the chemical treatment of the sorbent precursor leads to a major change in the pore size 

distribution. The formation of larger and smaller pores was indeed found as reported in Fig. 6 where 

the BJH curves for the two kinds of sorbent are presented 

dV/dlog(D) Pore Volume (cm³/ 

g·nm) 

0,0018 

0,0016 

0,0014 

0,0012 

0,001 

0,0008 

0,0006 

0,0004 

0,0002 

0 

Dry 

precursor 

Hydrated 

precursor 

1 10 100 

Pore size [nm] 

Fig. 6. Pore size distribution of material obtained with hydrated precursor and dry precursor. 

As you can see the sorbent synthesized with dry precursor shows an uni-modal pore size 

distribution (average size 30 nm) whereas for the material obtained from hydrated precursor a wider 

180

pore size distribution was observed. Particularly, the formation of larger pores (100 nm) were 

detected which permits CO2 to get the inner core of the particle with a major CaO utilization 

whereas smaller pores play a key role to the rapid carbonation of the sorbent material. Then the 

experimental results presented in the remainder of this work are collected from the sorbent obtained 

from the hydrated precursor. 

Fig. 7. Carbon capture capacity of 85% CaO sorbent when subjected to different regeneration 

condition: (a) mild regeneration: 900 °C, 14/86 %v (CO2/ N2); (b) moderately severe regeneration: 

1000 °C, 14/86 %v (CO2/ N2); (c) severe regeneration: 1000 °C, 86/14 %v (CO2/ N2). 

Fig 7 shows the influence of regeneration condition on the 85% CaO sorbent activity. In particular, 

you can see that with the increase of both temperature and the amount of CO2 in the atmosphere 

during the sorbent regeneration the self reactivation period decreases. When the sorbent is 

regenerated with mild condition, the self reactivation period is extended up to ~60 th (Fig 7 (a)). As 

reported in Fig. 7 (b), if the regeneration temperature is increased up to 1000 °C the self reactivation 

181

period is reduced to the first 15 cycles. When the presence of CO2 is increased up to 86% as in the 

severe condition, the material did not present considerable self reactivation period (see Fig 7 (c)). In 

addition while the regeneration temperature and the CO2 content in the atmosphere are increased the 

ability to retain carbon decreases in later cycles. 

Fig. 8. Carbon capture capacity of 75% CaO sorbent when subjected to different regeneration 

condition: (a) mild regeneration: 900 °C, 14/86 %v (CO2/ N2); (b) moderately severe regeneration: 

1000 °C, 14/86 %v (CO2/ N2); (c) severe regeneration: 1000 °C, 86/14 %v (CO2/ N2). 

Moreover Fig 7 shows improvement of sorbent activity due to the thermal pretreatment confirming 

results published elsewhere [6-7] where naturally occurring sorbents (limestone, dolomite) showed 

better uptake capacity when exposed to thermal pretreatment. Here, thermal pretreatment of the 

manufactured sorbent with a short carbonation period (80 minutes at 600 °C under 14/86 % v/v 

182

CO2/N2), indeed, leads to some improvement in the CO2 uptake in subsequent cycles. Moreover 

CO2 uptake for sorbent treated with a further carbonation period and a previous calcination step are 

presented as well. As reported in Fig 7 (a) the single pretreated sorbent shows a self reactivation 

period as for the untreated material but the ability of reacting with the carbon dioxide is superior to 

the latter one. Particularly for the single pretreated sorbent a ~68 % higher uptake when compared 

to the as synthesized sorbent was observed at the second cycle. In consequent 

carbonation/regeneration cycles the uptake was at least ~22% higher than that without thermal 

pretreatment. Instead a loss of activity was found for the untreated and single treated material at the 

later cycles (beyond the 60 

183 

th cycle). The double treated sorbent shows no significant self 

reactivation period but a very good reversibility for 150 cycles was detected. The increase of 

activity was found to be maximum at the third cycle, ~110%, whereas in later cycles the augment 

was ~26% higher than the untreated sorbent. With the increase in the regeneration temperature 

sorbent material shows a decreasing activity with cycling and an increase in the loss of reversibility. 

On the contrary, the positive effect of thermal treatment persists: ~49 % higher uptake when 

compared to the as synthesized sorbent was found at the second cycle whereas the performance was 

at least ~8% higher with respect to the untreated sorbent (Fig 7 (a)). The double treated material 

shows neither positive effect of thermal treatment at later cycles nor self activation at the beginning: 

even if a small self reactivation effect was detected at later cycles (in particular beyond 70 th cycle). 

Finally, Fig 7 (c) shows the performance of the material subjected to severe regeneration condition. 

When the molar fraction of CO2 is increased up to 86% the investigated material did not show self 

regeneration at all. Besides an initial short period where the material shows stability in CO2 

capturing, thermal treatment did not influence positively the reversibility through the material 

cycling particularly beyond the 20 th cycle. 

Finally, Fig 8 shows the influence of regeneration condition on the 75% CaO sorbent activity. As 

for the 85 % CaO sorbent, self reactivation effect is reduced when increasing both temperature and 

carbon dioxide during the regeneration step. When the material is regenerated at lower temperature 

(900 °C) and low CO2 content (14 % CO2) the self reactivation stage is expanded up to ~60 th cycle 

for the as synthesized sorbent and up to ~30 th cycle for the single and double treated sorbent (see 

Fig 8 (a)). As for the 85 % CaO sorbent, when the regeneration temperature is increased up to 1000 

°C the self reactivation period was observed up to ~15 th cycle (Fig 8 (b)). It is worth to note that, on 

the contrary of 85 % CaO sorbent, when the untreated material is regenerated in severe condition 

self reactivation effect was observed (Fig 8 (c)). The positive effect of thermal pretreatment was 

also confirmed for the 75 % CaO sorbent. As reported in Fig 8 (a) the capability of reacting with 

CO2 is greater when compared to the as synthesized material. For the single pretreated specimen a 

~43 % higher uptake when compared to the as synthesized sorbent was observed at the second 

cycle. In later cycles the performance was at least ~1% higher than that without thermal 

pretreatment. A loss of activity was observed for the single and double pretreated specimen beyond 

~30 th cycle, even if a minor self reactivation effect was found for the double treated sorbent beyond 

~70 th cycle (as observed earlier here for the 75 % CaO sorbent in Fig 7 (b)). For the double treated 

sorbent the augment of activity was found to be maximum at the third cycle, ~63%, whereas in later 

cycles the augment was at least ~4% higher than the untreated sorbent. When the temperature is 

increase from 900 °C to 1000 °C at the same CO2 molar fraction (16 %) the self reactivation period 

was found to shrink to 15 th cycle. Thermal treatment confirms its positive effect on increasing the 

CO2 capture capability: ~39 % higher capacity at the second stage whereas the uptake was at least 

~6 % in the remainder of the cycling when compared to as synthesized sorbent. The double treated 

sorbent has instead a negative influence on the material: the CO2 uptake decreases drastically below 

the performance of the untreated material. Perhaps the most remarkable result is reported in Fig 8 

(c) where the sorbent activity subjected to severe regeneration condition is showed. On the contrary 

of 85 % CaO sorbent, the as synthesized 75 % CaO sorbent shows self reactivation period at the 

beginning of the cycling test when severe regeneration condition are used. Besides an initial 

decrease of the performance, the single pretreated sorbent shows a continuously increment of its

CO2 capture capacity up to 150 th cycle. In order to confirm such an unexpected behavior another 

experimental run has been accomplished confirming the previous observation (see black dots in Fig 

8 (c)) with a good reproducibility. Finally for the double treated specimen no self reactivation 

period was found at the beginning where a loss of activity was observed followed by a period of 

good reversibility; beyond such a period a self reactivation was observed reaching a maximum at 

~150 th cycle. 

4. Conclusion 

The multi-cycling CO2 sorption-desorption tests on synthetic CaO-Ca12Al14O33 sorbent show that 

the reversibility of the CO2 uptake in repeated cycles is significantly improved compared that of 

dolomite. Preliminary experiments aimed at increasing the capacity to retain carbon show that 

exposing the material to carbon dioxide at 600 °C for 80 minutes prior the multi-cycling 

experiments the sorbent activity is increased. Self reactivation period is found at the beginning of 

the multi-cycling experiments or at later cycles reaching a maximum or a plateau in CO2 uptake. 

Unexpected results are detected for the 75 % CaO sorbent which shows a continuously increase in 

CO2 capture up to 150 th cycle under severe regeneration condition. The good reversibility showed 

by this material in severe regeneration condition make it as a good candidate for CO2 acceptor in a 

carbonate looping. In fact, in such a technology option a high concentrated CO2 stream is required 

for final disposal [8]. In fact, carbon dioxide is collected at the outlet of the calciner where 

regeneration of the sorbent happens. As a consequence the ideal CO2 acceptor should withstand 

high CO2 concentration in the calciner to achieve a good regeneration extent to begin properly 

another carbon capture cycle. 

References 

[1] Abanades J.C., Anthony E.J., Lu D.Y., Salvador C., Alvarez D., Capture of CO2 from 

combustion gases in a Fluidized Bed of CaO. American Institute of Chemical Engineers Journal 

2004, 50(7): 1614-1622. 

[2] Grasa GS, Abanades JC.,CO2 capture capacity of CaO in long series of carbonation/calcination 

cycles. Ind Eng Chem Res 2006; 45:8846-8851. 

[3] Gallucci K., Stendardo S., Foscolo P.U., CO2 capture by means of dolomite in hydrogen 

production from syn gas. Int J Hydrogen Energy 2008;33:3049-‐3055. 

[4] Delgado J., Aznar M.P., Corella J., Calcined dolomite, magnesite, and calcite for cleaning hot 

gas from a fluidized bed biomass gasifier with steam: life and usefulness. Ind Eng Chem Res 

1996;35:3637-‐3643. 

[5] Li Z., Cai N., Huang Y., Han H., Synthesis, Experimental Studies, and Analysis of a New 

Calcium-Based Carbon Dioxide Absorbent. Energy Fuels 2005;19:1447-1452. 

[6] Manovic V., Anthony E.J., Thermal Activation of CaO-Based Sorbent and Self-Reactivation 

during CO2 Capture Looping Cycles. Environ. Sci. Technol. 2008;42:4170–4174. 

[7] Chen Z., Song H.S., Portillo M., Lim C.J., Grace J.R., Anthony E.J., Long-Term 

Calcination/Carbonation Cycling and Thermal Pretreatment for CO2 Capture by Limestone and 

Dolomite. Energy & Fuels 2009;23:1437-1444. 

[8] Calabro`A., Deiana P., Fiorini P., Girardi G., Stendardo S., Possible optimal configurations for 

the ZECOMIX high efficiency zero emission hydrogen and power plant. Energy 2008;(33) 

952–962. 

184

Abstract: 




Investigation of an innovative process for biogas 

up-grading – pilot plant preliminary results 

Lidia Lombardi a , Renato Baciocchi b , Ennio Carnevale a , Andrea Corti c , 

Giulia Costa b , Tommaso Olivieri a , Alessandro Paradisi a and Daniela Zingaretti b 

a Dipartimento di Energetica, Università degli Studi di Firenze - Firenze, Italy, lidia.lombardi@unifi.it 

b Dipartimento di Ingegneria Civile – University of Roma Tor Vergata - Roma, Italy, 

baciocchi@ing.uniroma2.it 

c Dipartimento di Ingegneria dell’Informazione,Università degli Studi di Siena - Siena, Italy, 

corti@dii.unisi.it 

Biogas up-grading treatments aimed at producing biomethane to fuel vehicles or to inject into the gas grid, 

are applications that are gaining increasing interest throughout Europe. Several different commercial 

methods are available for separating carbon dioxide from biogas. In this work an innovative carbon dioxide 

removal method that, differently fromthe currently employed commercial techniques, allows also to capture 

and store the separated CO2is investigated.This process, named Alkali absorption with Regeneration 

(AwR),consists in a first step in which CO2is separated from the biogas by chemical absorption with an alkali 

aqueous solution followed by a second step in which the spent absorption solution is regeneratedfor reuse in 

the first step of the upgrading process and the captured CO2is stored in a solid and thermodynamically stable 

form. The latter process is carried outcontacting the spent absorption solution, rich in carbonate and 

bicarbonate ions, with a waste material characterized by a high content of calcium hydroxide andleads to the 

precipitation of calcium carbonate and to the regeneration of the alkali hydroxide content of the solution.The 

proposed processes werefirst investigated by preliminary laboratory and simulation analysis. On the basis of 

the results of these tests, air pollution control (APC) residues from Waste-to-Energy plants were selected as 

the waste material to use for the regeneration step and a pilot-scale regeneration plant to place downstream 

of an existing absorption column installed at a landfill site was designed and built.In this paper the layout of 

the plants, their operating conditions and the results obtained bypreliminary pilot-plant tests are reported.This 

study was conducted in the framework of theUPGAS-LOWCO2 (LIFE08/ENV/IT/000429) Life+ project. 

Keywords: 

Biogas upgrading, carbon dioxide capture, accelerated carbonation, air pollution control residues,alkali 

absorption. 


Among renewable energy sources, the biogas industry is growingin the EU, reaching about 8,3 

Mtoe in 2009 with more than 6.000 biogas plants. The main source is agriculture (52%), then 

landfills (36%) and sewage plants (12%)[1]. 

Biogas produced in AD-plants – fed with a variety of bio-materials such as waste or energy crops - 

or landfill sites is primarily composed of methane (CH4) and carbon dioxide (CO2) with smaller 

amounts of hydrogen sulphide (H2S) and ammonia (NH3). Trace amounts of hydrogen (H2), 

nitrogen (N2), saturated or halogenated carbohydrates and oxygen (O2) are also occasionally present 

in biogas. Usually, the gas is saturated with water vapour and may contain dust particles and 

organic silicon compounds (e.g.:siloxanes).The heating value of biogas is determined mainly by the 

methane content of the gas. 

There are four basic ways of biogas utilization:heat and steam production, electricity production 

and/or co-generation, use asvehicle fuel andproduction of chemicals. Biogasutilisation strategies 

may vary depending on National factors such astaxation, subsidies, availability of gas and heat 

grids. Worldwide, biogas is mainly used for electricity production whereas in Sweden and in 

Switzerland a growing amountof biogas is used in the transport sector. The major driver defining 

185

the way for biogas utilisation is the compensation of the energy, i.e. electricity or (upgraded) 

biogas. Most of the European countries have increased feed-in tariffs for electricity. However, using 

biogas as vehicle fuel or injecting the gas into the gas grid are applications that are gaining more 

and more interest. After proper upgrading - i.e. removal of carbon dioxide and tracecontaminants - 

biogas canbefed into the natural gas distribution grid. The deregulation of the natural gas market in 

Europe has opened the possibility to find new customers for upgraded biogas via the gas grid. There 

is no international technical standard for biogas injection but some countries have developed 

national standards and procedures for biogas injectioninto the natural gas grid, such Sweden, 

Switzerland, Germany and France [2]. The standards have been set to avoid contamination of the 

gas grid or atend use. In the standards requirements onWobbe index values and limits on the 

concentration of certain components such as sulphur, oxygen, dustand the water dew point, as well 

asaminimum methane concentration of 96% are reported. These demands are in most cases 

achievable applying existing upgrading processes. In some cases landfill gas can be difficult to 

upgrade to sufficient quality due to highnitrogen contents. 

There are several different commercial methods for reducing the carbon dioxide content of 

biogas[3]. The most common are High Pressure Water Scrubbing (HPWS), amine scrubbing, and 

Pressure Swing Adsorption (PSA) on activated carbon. New technologies are, for example, 

cryogenic upgrading, molecular sieves and separation membranes. When removing carbon dioxide 

from the biogas stream small amounts of methane are also removed. It is important to keep these 

methane losses low for economical and environmental reasons and to maximize thegasenergy 

content. 

The commercial technologies available for biogas upgrading have the common feature of removing 

carbon dioxide from biogas without focusing on the fate of the separated carbon dioxide, which is 

usually re-emitted into the atmosphere during the system regeneration phase. For example, when 

CO2 removal is achieved by means of absorption with a liquid solution, the load solution containing 

the absorbed CO2 is regenerated emitting CO2 to the atmosphere. The same happens during 

theregeneration of activated carbon when PSA is applied. In thecases in which absorption takes 

place using water without regeneration, CO2 is discharged with the spent solution and released to 

the atmosphere. It should be anyhow pointed out that these CO2emissions are of biogenic originand 

should hence not be accounted for as an effective contribution to greenhouse gas emissions. 

As a matter of fact, the innovative aspect proposed in this work, developed in the framework of the 

European Life+ project UPgrading of landfillGAS for LOWeringCO2 emissions (UGAS- 

LOWCO2), is to develop a biogasupgrading process that can not only capture but also definitely 

store the separated CO2in a solid form [4]. The subtraction of carbon dioxide of biogenic origin 

from the atmosphere can contribute as a negative emission (sink) to the overall greenhouse gases 

balance. In addition, the proposed method may allow to achieve also other specific environmental 

benefits that will be highlighted in the following paragraphs. 

2. Alkali absorption with regeneration process concept 

This method – named Alkali absorption with Regeneration (AwR) - is based on CO2 chemical 

absorption by means of an alkali aqueous solution followed by regenerationof the spent solution 

using Air Pollution Control (APC) residues. 

CO2 is first physically absorbed in the liquid solution and here it reacts with the alkaline compound 

producing carbonate (CO3 2- ) and bicarbonate ions (HCO3 - ) (chemical absorption) [5][6][7]. The 

alkaline reactants that can be used in the absorption process are potassium hydroxide (KOH) or 

sodium hydroxide (NaOH). The reactions that take place during the absorption step are the 

following: 

2 KOH + CO2 K2CO3 + H2O (1) 

or 

186

2 NaOH + CO2 Na2CO3+ H2O (2) 

The load solution – i.e. the solution containing the carbonate/bicarbonate ions – can be chemically 

regenerated by contacting it withcalcium hydroxide (Ca(OH)2) in solid form. In this step, poorly 

soluble calciumcarbonate(CaCO3) precipitation takes place (carbonation reaction), thus the 

CO2separated from the biogas can be permanently storedin a chemically inert and 

thermodynamically stable form, whereas KOH or NaOH is recovered for the first step of the 

upgrading process. The basic reactions that take place during the regeneration phase are the 

following: 

Ca(OH)2 + K2CO3 CaCO3()+ 2KOH (3) 

or 

Ca(OH)2 + Na2CO3 CaCO3()+ 2NaOH (4) 

Since the use of calcium hydroxide for such a process would not make sense from a carbon dioxide 

mitigation perspective, as it is manufactured by calcination of limestone releasing carbon dioxide 

into the atmosphere, in order to obtain a net reduction of carbon dioxide emissions, in this project 

industrial waste residues were chosen as alternative alkalinity sources. Several studies have in fact 

shown the feasibility of using different types of industrial residues, characterized by a high content 

of calcium hydroxide phases, such as bottom ash and air pollution control (APC) residues from 

waste incineration or steel slag, to sequester CO2. Such a capture process is known as accelerated 

carbonation of natural minerals or industrial residues [8][9][10]. In case of using industrial residues, 

the accelerated carbonation allows also improving the leaching behaviour of the residues 

[11][12][][]. 

From preliminary investigations, APC residues, which are the product of incineration flue gas 

treatment with calcium-based products, were selected for the regeneration process owing to their 

chemical, physical and mineralogical composition [][][]. Figure 1 shows the schematic layout ofthe 

AwR process. 

Fig. 1. Conceptual layout of the AwR process. 

3. Results of the preliminary investigation phase 

At a preliminary level the two steps of the AwR process were studied separately. In particular, the 

absorption step was first investigated by means of computer simulations and carrying out some 

absorption tests on the pilot plant – which was already available from a previous research 

187

projectand is described in the following[]. The regeneration step was instead investigated by 

laboratory testing [][]. 

One of the main outcomesof the preliminary analysis concerned the definition of the maximum 

concentrations of the alkali compoundsto use in the absorption process. In fact,for increasing initial 

concentrations of KOH or NaOHin the absorption solution and hence of K2CO3 or Na2CO3 in the 

solution resulting from the absorption treatment, the yield of the regeneration process showed to 

decrease owing to the combination of two negative effects: a decrease in KOH or NaOH 

regeneration efficiency and an increase in solution losses during the solid separation process after 

regeneration []. So it was concluded that a maximum of 4 eq/l of carbonate ionscould be acceptable 

in the solution to be regenerated in order to allow for high regeneration yields during the second 

stage of the process[]. Hence, assuming a complete conversion of KOH or NaOH during the 

absorption step, the above mentioned condition would correspond to an absorption solution with a 

18-20 % wt. concentration of KOH or NaOH. As a matter of fact, such concentration values are 

definitely lower than those used in previous pilot-scale absorption experiments aimed at treating 20 

Nm 3 /h of landfill gas in which mass concentrations ofover 50%wt. of KOH were adopted[][]. Under 

those conditions with a liquid flow rate of 60 l/h characterized by a KOH concentration of 48-53% 

wt, CO2 removal efficiencies of 83-97% were achieved[][].At least 97% CO2 removal efficiency is 

required to obtain an acceptable upgraded biogas quality (CH4> 98%) starting from 50% in vol. 

CH4 and 50% in vol. CO2.Consequently - keeping the same landfill gas and solution flow rates as 

those defined in the original design of the pilot plant - the achievable CO2 removal efficiency 

obtained using alkali concentration values of 12-25% would be far lower than the target value, as is 

evident in Table 1, where the results obtained by both the simulations and the pilot plant tests are 

compared. In this case the simplified simulations of the absorption process were carried out using 

Aspen Plus[25]. The layout of the simulation was based on one absorption column, modeled using a 

radfrac unit, with two entering streams (biogas and absorbing solution) and two exiting streams 

(CH4 enriched biogas and load solution). 

Then the simulation layout was modified, considering more than one absorption column, with the 

aim of evaluating the number of consecutive absorption stages, required to reach the biomethane 

quality in output, according to the German standards [2]. From these simulations, it was estimated 

that to reach the required removal efficiency, three absorption stages, with an entering solution flow 

rate of 60 l/h and KOH or NaOHconcentrations of 12-14% wt. would be required to process 20 

Nm 3 /h of landfill gas. 

The existing pilot absorption column represents the first absorption stage of the three stage 

process.For this reason, the results ofthe pilot tests that will be reported in the following paragraphs 

are to be considered representative, for the moment, only of the first of the three absorption stages. 

Table 1. CO2 removal efficiency obtained from simulations and pilot plant tests, considering 20 

Nm 3 /h entering landfill gas with 50% CH4 and 50% CO2 in vol.; absorption solution flow rate 60 

l/h. 

Mass 

concentration 

CO2 removal efficiency % - 

Simulation 

188 

CO2 removal efficiency % - Pilot 

plant test 

KOH 

11,9 35,8 29,7±1,45 

17,0 51,9 41,8±3,5 

25,5 78,5 60,5±4,9 

NaOH 

11,0 35,7 32,9±1,1 

20,0 59,4 56,0±3,7

Concerning the regeneration step, specificlaboratory experiments were preliminarily carried out to 

characterize the APC residues and to investigate their capability of regeneratingthe spent 

absorptionsolution rich in carbonate/bicarbonate ions[].The amount of calcium phases available for 

the regeneration reaction was estimated as the difference between the total Ca and the Ca as CaCO3 

content of the ash, and consisted mainly of Ca(OH)2 and CaClOH[]. The quite high chloride content 

(around22% by weight) of the ash, mainly as calcium hydroxychlorideproved to hinder the 

regeneration reaction. As a matter of fact, for every mole of CaOHCl, as shown in Eq. (5), only 1 

mol of KOH can be produced, differently from calcium hydroxide thatallows to regenerate 2 mol of 

potassium hydroxide (see Eq. (3)): 

CaOHCl + K2CO3KCl + KOH + CaCO3 (5) 

Since CaOHCl is more soluble than Ca(OH)2, it is more readilyavailable for reacting with 

potassium carbonate than Ca(OH)2;hence an increase in the amount of ashes added to the solution 

produceda reduction of the efficiency of the regeneration reaction[].To improve the alkali 

regeneration yield, a washing pretreatmentof the APC residues was hence tested in order to remove 

mostof the phases responsible of decreasing the total buffering capacityof the solution. 

Furthermore, in order to improve the overall leaching behavior of the solidmaterial so to comply 

with the disposal criteria for non hazardouswaste landfilling, the effects of a second washing 

treatment appliedto the residues after the regeneration step were investigated. In this case, due tothe 

removal of part of the KOH contained in the ash, the pH of theresidues decreased, but remained still 

well above values indicatingsolubility control by calcite. The mobility of most of the 

testedcompounds (Zn, Pb, Sb, Cr and SO4 2- ) appeared anyhow to decreaseafter this latter treatment, 

resulting lower than the limit values fornon hazardous waste disposal[]. 

The appropriate operating conditions able to maximize KOH or NaOHregenerationwere defined[][], 

including pre- and post-washing of the residues, as reported in Table 2. The APC residues were 

added to the solution to be regenerated on the basis of their calcium content as Ca(OH)2, applyinga 

1,2 ratio with respect to the carbonate ions content of thesolution.In this way it was possible to 

reach a 90% efficiency of the reaction in terms of KOH or NaOH regeneration. In addition, the 

solid product showed to be mainly made up by calcite and a CO2 storagecapacity of above 300 g/kg 

solid product was obtained.It should be noted however, that after the regeneration reaction the 

liquid solution must be separated from the solid product and that disregarding the type of method 

useda complete recovery of the solution is not possible and therefore a lower overallfinal recovery 

of KOH or NaOH should be anticipated. 

Table 2. Operating conditions selected for the regeneration step. 

Pre-treatment 

Type Washing 

L/S 5 l/kg 

Time 15 min 

Regeneration 

Ca/CO3 2- ratio 1,2 molCa/mol CO3 2- 

Temperature 55 °C 

Time 60 min 

Post-treatment 

Type Washing 

L/S 5 l/kg 

Time 15 min 

189

4. Pilot plant tests 

In order to demonstrate the technical feasibility of the proposed process, an integrated pilot plant for 

CO2absorption and regeneration of the spent solution with CO2 storage was designed and built. The 

pilot plant is located at the research laboratory of the University of Florence hosted at the landfill 

site managed by one of the partnersof the UPGAS-LOWCO2 project. 

4.1. Absorption pilot plant 

The absorption pilot plant (Figure 2 (i)) consists of a packed column where an aqueous solution of 

KOH or NaOH reacts with the carbon dioxide contained in the landfill gas, which is directly 

extracted from the landfill [][][][][]. 

The alkali compound aqueous solution is fed to the column top, while the landfill gas is fed to the 

bottom of the column. The landfill gas is extracted from a collection station in the landfill and flows 

into the column by means of a side channel blower. The column 

fillingconsistsofSulzerlaboratoryDX packing, with a diameter of80mmandoverall heightof990cm, in 

stainlesssteel. The column was originally designed to process about 20-25 Nm 3 /h of landfill gas and 

40-60 l/h of absorbingsolution. 

(a) 

(b) 

(i) (ii) 

Fig. 2. Pictures of the pilot plant: i) absorption column; ii) regeneration reactor and its main 

components: a) regeneration reactor with mixer and heating jacket, b) bottom section of the reactor 

with filter tensioning system, c) vacuum filtration pump, d) liquid collection tank, e) control panel 

and f) steel frame. 

4.1.1. Monitoring equipment 

Input and output gas flow ratesaremeasured by means of a volumetric flow meter (Fluidwell – 

F110) able to work in the range from 2,5 to 35 m 3 /h.Input and output volumetric gas composition is 

measured too by means of a portable gas analyzer (input with Geotechnical Instruments - GA 94- 

and output with Geotechnical Instruments-GA2000) which measures CH4and CO2by infra-red 

absorption and O2 by internal electro-chemical cells.Input and output differential pressure is 

measured by a diaphragm pressure transducer (Delta Ohm-HD 408T 100MBG) able to work in the 

range from -100 to +100 mbar relative to the atmospheric pressure. Atmospheric pressure is 

measured by means of a barometric pressure transducer (Delta Ohm HD 9908 BARO) able to work 

190 

(a) 

(b) 

(e) 

(f) 

(c) 

(d)

in the range from 700 to 1100 mbar. Input and output gas temperature is measured by means of Ktypethermocouples. 

Gas flow rate, pressure and temperature are measured and registered in a quasicontinuous 

manner (every 10 seconds). The measurement instruments are controlled by a 

programmable automation controller (Compact Field Point – National Instrument) composed by 

rugged I/O modules and intelligent communication interfaces. The composition is measured every 

60 seconds and is directly registered by the gas analyzers. 

4.2. Regenerationpilot plant 

The regeneration pilot plant was designed and built with the aim of using the same reactor to 

perform in batch mode the pre-washingtreatment of the residues, the regeneration/carbonation 

reaction and the post-washing treatment, as well as the liquid/solid separation step after each of the 

three operations. For the separation step it was decided to apply vacuum filtration, the same method 

adopted for the lab-scale experiments. As shown in Figure 2 (ii), the plant is made up by: the 

regeneration reactor, which includes a paddle type mixer and an external heating jacket; the filter 

medium, which is fitted on the bottom of the reactor in a custom made tensioning system; the 

vacuum filtration system, made up by the pump and filtered liquid collection tank; the control panel 

with switches for activating all equipment (mixer, heating system and vacuum pump) and a display 

for setting the heating temperature; a stainless steel support system on wheels on which all units and 

equipment are placed and manoeuvred. 

The APC residues and the liquid medium (distilled water for the pre- and post-washing treatments 

and the spent solution exiting from the absorption step for the carbonation process) are mixed in the 

regeneration reactor tank. In the first step of pre-washing, distilled water and APC residues are 

mixed and kept in the tank for the required time; then filtration starts and at the end of this process a 

solid cake remains at the bottom of the reactor, while the filtered water is collected in the liquid 

collection tank and, from there, discharged. In the second step, the solution coming from the 

absorber is added to the cake previously formed at the bottom of the reactor. The slurry is mixed 

and after the required reaction time the filtration starts again. The liquid phase which accumulates in 

the collection tank is the regenerated solution, which is reused – after proper make up addition – in 

the absorption column. The carbonated cake remaining at the bottom of the reactor after the second 

filtration step is washed with distilled water and after a third filtration step, the final solid product is 

extracted from the bottom of the reactor while the filtration liquid is removed from the collection 

tank and disposed of. 

4.3. AwR operation test: procedure and preliminary results 

In this paragraph the procedure adopted and the results obtained from the preliminary operational 

tests carried out on the pilot-scale AwR plant are described. The operating conditions selected for 

these tests are reported in Table 3. 

Table 3. Operating conditions selected for the preliminary AwR pilot plant tests. 

Alkali compound KOH 

Absorbing solution flow rate [l/h] 60 

Absorption operation time [min] 10 

Volume of load solution to be regenerated 1 [l] 8 

Alkali mass concentration [%] 11,9% 

Alkali normality [eq/l] 2,35 

Reactive Ca to CO3 -- ratio [Camol /CO3 2- mol] 1,2 

1 The volume of liquid produced by the column during a 10 minutes absorption operation, discarding the solution 

generated during the first 2 minutes, in order to consider steady functioning. 

191

Concerning the absorption reaction, after the preparation of the absorbing solution – obtained 

mixing the appropriate amount of KOH and water - the experiment was started. Temperature, 

pressure, flow rate and composition were continuously measured and recorded for the inflow and 

outflow gas. The average flow rate of entering landfill gas was 18,03 Nm 3 /h, while the average 

exiting flow was 15,07 Nm 3 /h. Entering average concentration in vol. of CH4 was 50,66% while 

CO2 was 37,48% in vol. In the exiting gas the average CH4concentration was 60,61%, while the 

average CO2content was 28,76%. The calculated CO2 removal efficiency was about 35,86%. 

The spent solution was collected at the outlet of the column in a bucket after the first two minutes of 

the reaction. After ten minutes the absorption experiment was stopped. 

The spent solution sample was titrated, showing a complete conversion of KOH to K2CO3 with a 

total concentration of 2,35 eq./l (Figure 3 (ii)). 

Before the testing phase, the properties of the APC residues to use for the test were assessed by 

laboratory analysis. Basically this characterisation was necessary to determine the amount of APC 

residues to be used for the regeneration experiments. Specifically the total Ca(OH)2 content of the 

washed ash was estimated as 55,5% wt, while the weight loss of the material measured upon the 

washing pre-treatment (mainly due to NaCl and CaClOH dissolution) was of 43,2 % by weight. 

Based on the above mentioned characteristics and the conditions reported in Table 3, the required 

amount of washed APC residues for the regeneration test was calculated to be equal to 1,5 kg; 

hence considering the weight loss of the material consequent to the washing pre-treatment, it was 

estimated that 2,64 kg of untreated APC ash would be necessary for the complete regeneration test. 

Based on this, the amount of distilled water required for the washing pre-treatment with a L/S ratio 

of 5 l/kg was calculated (13,23 l). 

Prior to the beginning of the experiment the reactor was washed and dried and the filtering material 

was substituted. The valve at the bottom of the reactor was closed and the required amounts of APC 

residues and of distilled water were weighed and fed into the reactor. 

After the introduction of the residues and distilled water into the reactor, the mixer was activated. 

The mixing was maintained for 15 minutes at ambient temperature (internal temperature 24,3 °C), 

then the vacuum pump was turned on and connected to the liquid collection tank and the valve at 

the bottom of the reactor was opened. During the liquid separation phase the mixing of the solution 

was continued in order to help the filtration process. After about 100 minutes the mixer was stopped 

since the liquid separation appeared to be complete. The filtered washing solution was collected 

from the tank, which was then cleaned and dried and samples of the washing solution were taken. 

The absorption spent solution (8 l)wasthen poured into the reactor and mixed with the washed APC 

residues cake. After 1 hour of reaction time at 55°C, the vacuum pump was turned on and 

connected to the liquid collection tank and the valve at the bottom of the reactor was opened. Also 

during this liquid separation phase the mixing of the solution was continued in order to help the 

filtration process. This separation step proved faster than the previous one and after 45 minutes the 

filtration appeared to be complete (Figure 3 (i)). Samples of the solid product of the regeneration 

process were taken and right after that the final washing treatment was performed. At the same time 

the regenerated solution was emptied from the collection tank and a sample of it was directly 

titrated. Around 1 eq./l of KOH were regenerated out of a total buffering capacity of 1,6 2 eq/l, hence 

the regeneration yield(ability of the reaction to obtain again the initial compound) was about 62,5%. 

This value was quite lower than the one measured in the lab scale tests (78-92%)[][]. Through a 

preliminary mass balance it is possible to estimate the overall regeneration efficiency (mass of 

recovered KOH with respect to the initial amount of KOH entering in the absorption column) which 

resulted of about 41%. This overall regeneration efficiency is quite low and needs to be increased in 

2 Due to dilution with the humidity of the washed cake, the total buffering capacity of the regenerated solution was 

lower than the spent solution one(1,6 vs. 2,35 eq./l) as shown in Figure 3 (ii). 

192

order to increase the competitiveness of this method compared to other upgrading processes from an 

environmental and economical point of view[][][]. 

The amount of distilled water necessary for the final washing treatment was estimated assuming 

that the product of the regeneration phase had a humidity of 50% wt. and a weight increase of 40% 

compared to the washed residues characterized by a density of 0,99 kg/l (data determined in the labscale 

tests). On the basis of these assumptions, the amount of water required to set the L/S ratio to 5 

l/kg was calculated to be 8,42 kg. The distilled water was poured into the reactor and the mixer was 

switched on. After 15 minutes the vacuum pump was turned on and connected to the liquid 

collection tank and the valve at the bottom of the reactor was opened. During the liquid separation 

phase the mixing of the solution was continued in order to help the filtration process. The 

liquid/solid separation appeared to be complete after 45 minutes. The bottom of the reactor was then 

opened and the solid cake was collected: roughly 2,64 kg of solid humid product were recovered. 

In conclusion, the outcome of this preliminary pilot-scale test was considered satisfactory, since no 

great operating difficulties were encountered and all three steps of the regeneration process were 

carried out. The liquid/solid separation step after the first washing treatment proved to be as 

expected the most critical one in terms of time; in addition some ash particles were found in the 

liquid collection tank, hence it was decided to check if other types of filtering materials could be 

used to optimize this process. As for the outcome of the regeneration process, the lower 

regeneration yield was ascribed to the less constant temperature and especially to poor mixing at the 

bottom of the tank that did not allow the complete reaction of Ca(OH)2. To improve this aspect it 

was decided to try to verify if the shape of the mixer could be modified. 

Finally, the amount of CO2 stored per unit of mass of solid product after the regeneration process 

was of about 220 g/kg solid material, and showed to increase after the final washing treatment 

owing to the dissolution of residual soluble phases. 

(i) (ii) 

Fig. 3. Picture of the solid cake obtained at the end of the filtration process after the 

regeneration/carbonation reaction (i); titration curves of the spent and regenerated solution (ii). 


An innovative method for removing carbon dioxide from landfill gas – with the final aim of 

upgrading its quality to that of natural gas– was proposed and investigated. With respect to 

commercial methods for biogas upgrading, the proposed process presents two additional 

environmental benefits: i) the carbon dioxide separated from the methane is permanently stored by 

accelerated carbonation of an alkaline waste material(air pollution control residues from waste 

incineration flue gas treatment), ii) the series of treatments applied during the regeneration process 

has shown to improve the leaching behaviour of the residues. This process, named Alkali absorption 

193

with Regeneration (AwR), consists in a first step in which CO2 is separated from the biogas by 

chemical absorption with an alkali aqueous solution followed by a second step in which the spent 

absorption solution is regenerated for reuse in the first step of the upgrading process and the 

captured CO2 is stored in a solid and thermodynamically stable form. The processes were 

investigated first by simulations and laboratory testing and then in order to verify the feasibility of 

the method at a larger scale, a pilot-scale AwR plant for treating 20 Nm 3 /h of biogas was designed, 

built and installed at a landfill site. From the results of preliminary tests, the regeneration yield 

achieved at pilot scale appeared to be lower than the yields obtained in the laboratory, so the main 

efforts for the next testing phase will focus on the optimization of plant operation in order to 

increase the overall efficiency of KOH or NaOH recovery and possibly allow to improve the 

competitiveness of this process compared to traditional commercial biogas upgrading methods. 


The authors wish to acknowledge the Life+ Programme and European Commission for co-funding 

the activities of the UPGAS-LOWCO2 project (LIFE08/ENV/IT/000429) www.upgas.eu. 

References 

[1] Eurobserv’er. Biogas Barometer. November 2010. Available at 

[accessed 10.02.2012]. 

[2] Persson M., Jönsson O. and Wellinger A. (2006). Biogas Upgrading to Vehicle Fuel Standards 

and Grid Injection. IEA Bioenergy. Available at 

[accessed 10.02.2012]. 

[3] Petersson A. and Wellinger A. (2009). Biogas upgrading technologies – developments and 

innovations. IEA Bioenergy. Available at 

[accessed 10.02.2012]. 

[4] Lombardi, L., Baciocchi R., Carnevale E., Corti A., Costa G., Gabarrell X., Mostbauer P., 

Olivieri F., Olivieri T., Paradisi A., Starr K., Villalba G., Zingaretti D.. Innovative processes for 

biogas upgrading. In Proceedings of Sardinia 2011 Thirteenth International Waste Management 

and Landfill Symposium 3 - 7 October 2011 S. Margherita di Pula (Cagliari), Sardinia, Italy. ISBN 

978-88-6265-000-7 

[5] Corti A. (2004). Thermoeconomic evaluation of CO2 alkali absorption system applied to semiclosed 

gas turbine combined cycle, Energy 29 (3) march 2004, 415-426, Elsevier ldt. 

[6] Corti A., Beconi B.M., Lombardi L. (2001). Alkali absorbing for CO2 removal: thermoeconomic 

comparison between carbonate and sodium hydroxide based processes. InProceedings of 

IcheaP-5 The Fifth Italian Conference on Chemical and Process Engineering, Firenze, Italy - 20-23 

May, 2001, AIDIC, Milano 

[7] Baciocchi R., Storti G., Mazzotti M. (2006). Process design and energy requirements for the 

capture of carbon dioxide from air. Chemical Engineering and Processing; 45(12):1047-1058. 

[8] Huijgen W.J.J. (2007). Carbon dioxide sequestration by mineral carbonation. Thesis, Energy 

research Centre of the Netherlands, The Netherlands. 

[9] Huijgen, W. J. J., Witkamp, G. J., &Comans, R. N. J. (2005). Mineral CO2 sequestration by 

steel slag carbonation. Environmental Science and Technology, 39, 9676–9682. 

[10] Li X., FernándezBertos M., Hills C.D., Carey P.J., Simon S. (2007). Accelerated carbonation 

of municipal solid waste incineration fly ashes. Waste Management, 27, 1200–1206 

[11] Huijgen, W. J. J., &Comans, R. N. J. (2006). Carbonation of steel slag for CO2 sequestration: 

leaching of products and reaction mechanisms. Environmental Science and Technology, 40, 2790– 

2796. 

194

[12] Van GervenT.,VanKeer, ArickxS.,Jaspers M., Wauters G., Vandecasteele C. 

(2005).Carbonation of MSWI-bottom ash to decrease heavy metal leaching, in view of recycling. 

Waste Management, 25, 291–300 

[13] Costa G., Baciocchi R., Polettini A., Pomi R., Hills C.D., Carey P.J. (2007). Current status and 

perspectives of accelerated carbonation processes on municipal waste combustion residues, 

Environmental Monitoring and Assessment, 135:55-75. 

[14] Baciocchi R., Costa G., Polettini A., Pomi R. (2008). An insight into the effect of accelerated 

carbonation on metal release from incinerator ash. In Proceedings of the 2nd International 

Conference on Accelerated Carbonation for Environmental and Materials Engineering, Rome 1-3 

October 2008. 

[15] Baciocchi R. Corti A., Costa G., Lombardi L., Zingaretti D. (2010). Storage of carbon dioxide 

captured in a pilot-scale biogas upgrading plant by accelerated carbonation of industrial residues. 

International Conference on Greenhouse Gas Technologies (GHGT-10). 19th-23rd September 

2010, RAI, Amsterdam, The Netherlands. Energy Procedia (2011) 4: 4985–4992. 

[16] Baciocchi R., Costa G., Lombardi L., Verginelli I., Zingaretti D. (2010). Storage of carbon 

dioxide captured in a pilot-scale biogas upgrading plant by accelerated carbonation of industrial 

residues. In Proceedings of the Third International Conference on Accelerated Carbonation for 

Environmental and Material Engineering. ACEME10. Turku, Finland, Nov. 29 – Dec. 1 2010. 

ISBN 978-952-12-2505-5 - ISBN 978-952-12-2506-2 (pdf version). 

[17] Baciocchi R. Costa G., Lombardi L., Zingaretti D. (2010). Storage of carbon dioxide captured 

in a pilot-scale biogas upgrading plant by accelerated carbonation of industrial residues. In 

Proceedings of Venice 2010, Third International Symposium on Energy from Biomass and Waste. 

Venice, Italy; 8-11 November 2010. CISA, Environmental Sanitary Engineering Centre, Italy. 

ISBN 978-88-6265-008-3. 

[18] Lombardi L., Carnevale E., Corti A., Olivieri T.. Biogas up-grading through carbon dioxide 

absorption with alkali solution. In Proceedings of Sardinia 2011 Thirteenth International Waste 

Management and Landfill Symposium 3 - 7 October 2011 S. Margherita di Pula (Cagliari), 

Sardinia, Italy. ISBN 978-88-6265-000-7. 

[19] Baciocchi R., Costa G., Gavasci R., Lombardi L., Zingaretti D.. Regeneration of a spent 

alkaline solution from a biogas upgrading unit by carbonation of APC residues. 

ChemicalEngineering Journal; 179:63– 71. 

[20] Baciocchi R., Costa G., Faraoni G., Lombardi L., Olivieri T., Zingaretti D.. Pilot-scale 

carbonation of APC residues for CO2 storage and alkali regeneration downstream an absorption 

column for biogas upgrading. In Proceedings of Sardinia 2011 Thirteenth International Waste 


Sardinia, Italy. ISBN 978-88-6265-000-7. 

[21] Lombardi L. and Corti A. (2009). Carbon Dioxide Capture: Technical Review and Proposal of 

an Alternative Absorption Process. In: Theodore B.N. (ed.), Flue Gases – Research, Technology 

and Economics, Chapter 8, Nova Science Publishers (ISBN 978-1-60692-449-5). 

[22] Lombardi L., Corti A. Carnevale E., (2008). Carbon Dioxide Capture from Landfill Gas. In 

Proceedings of Second International Conference on Accelerated Carbonation for Environmental and 

Materials Engineering. Rome (Italy) 1-3 October 2008 (pp. 17-26). 

[23] Lombardi L., Carnevale E., Corti A. (2008). Landfill gas quality up-grading through carbon 

dioxide capture: environmental and economic evaluations. In Proceedings of 16th European 

Biomass Conference and Exhibition. Valencia (SP) 2-6 June 2008. 

[24] Lombardi L. Corti A. Carnevale E. Baciocchi R. Zingaretti D. (2010). Carbon dioxide removal 

and capture for lanfill gas up-grading. International Conference on Greenhouse Gas Technologies 

(GHGT-10). 19th-23rd September 2010, RAI, Amsterdam, The Netherlands. Energy Procedia 

(2011) 4:465–472. 

195

[25] Aspen Plus 2004.1. Cambridge, MA, USA: Aspen Technology Inc.; 2005 

[26] Lombardi L, Carnevale E, Carpentieri M, Corti A (2007). Carbon dioxide capture from 

landfill gas. In Proceedings of ISWA/NRVD World Congress 2007, Amsterdam, The Netherlands, 

24-27 September, 2007. 

[27] Starr K., GabarrellDurany X., Villalba Mendez G., TalensPeiró L., Lombardi L.. Life cycle 

assessment of biogas upgrading technologies. Waste Management (2012), doi:10.1016/ 

j.wasman.2011.12.016 

[28] Starr K., GabarrellDurany X., Villalba Mendez G., TalensPeiró L., Lombardi L.. CO2 balance 

of biogas upgrading technologies. In Proceedings of Sardinia 2011 Thirteenth International Waste 


Sardinia, Italy. ISBN 978-88-6265-000-7 

[29] Starr K., GabarrellDurany X., Villalba Mendez G., TalensPeiro L., Lombardi L., Biogas 

Upgrading: Environmental Comparison of Conventional and Innovative Technologies. Submitted to 

ECOS 2012. 

196

Abstract: 




Method of increasing the efficiency of a 

supercritical lignite-fired oxy-type fluidized bed 

boiler and high-temperature three - end 

membrane for air separation 

Janusz Kotowicz a , Adrian Balicki a 

a Institute of Power Engineering and Turbomachinery, Silesian University of Technology, ul. 

Konarskiego 18 44-100 Gliwice, Poland, adrian.balicki@polsl.pl 

In this paper a thermodynamic analysis of a supercritical power plant supplied with lignite was made. The 

power plant consists of: a steam cycle at constant power of 600 MW, live steam parameters at 

600 °C/29 MPa and reheated steam parameters at 620 °C/5 MPa; supercritical OXY type circulating fluidized 

bed boiler and air separation unit. An air separation unit is based on a three – end type high temperature 

membrane. Models of the steam cycle, circulating fluidized bed boiler and air separation unit were built using 

a commercial computer program GateCycle and inhouse codes. After the integration of the listed above 

models, CFB boiler thermal efficiency as a function of the oxygen recovery ratio in the high temperature 

membrane for the variant without and with fuel drying were determined. The calculated thermal efficiency for 

the variant with fuel drying increases from 76% to 87% with increasing oxygen recovery ratio from 0.45 to 

0.9. 

Keywords: 

Thermodynamic analysis, Supercritical oxy-fuel boiler, High temperature membranes, Oxy-combustion, 

Lignite drying 


The necessity of meeting the standards of flue gas emissions, introduced by both, EU and national 

legislators, requires the continuous development of the technologies, which allow for maximum 

limitation of the greenhouse gases emission. The group of these technologies includes oxy– 

combustion technology, which, thanks to an almost complete elimination of nitrogen from the 

process, substantially limits the flue gas stream generated by the power unit [1]. For uninterrupted 

operation of the OXY type boiler it is necessary to provide a constant stream of oxidizer with the 

highest possible oxygen content, which mixed with a stream of recirculated exhaust gas as an 

oxidizing is mixture fed to the combustion chamber. The production of the technical oxygen with 

the use of both currently available on an industrial scale cryogenic technology and considered 

alternative technologies, is associated with the significant power consumption of the process, and 

thus also with a decrease of the efficiency of the block [2]. For boilers fed with lignite, just like in 

classical solutions, a possible way to improve the efficiency of electricity generation is to pre-dry 

the fuel prior to the injection into the combustion chamber. The presence of large amounts of water 

in the fuel reduces the lower heating value, enforces necessity to provide greater fuel mass flow to 

the boiler and, which is particularly important in systems with wet type recirculation, increases the 

probability of condensation of moisture from the flue gases. The drying process requires the 

delivery of large quantities of the drying medium, of which a function is to raise the temperature 

and the evaporation of the greatest possible amount of moisture from the fuel. In the classical 

solutions as a drying medium hot air or exhaust gases leaving the boiler is used. In OXY type 

197

systems for fuel drying waste stream of a mixture of nitrogen and oxygen from the air separation 

unit can be used. 

2. Assumptions 

A model of a supercritical boiler with circulating fluidized bed working in oxy - combustion 

technology was built using commercially available computer program GateCycle and in-house 

codes. Scheme of the model built (with ideological scheme of steam cycle) is shown in Figure 1. 

Fig. 1. Scheme of the CFB boiler integrated with lignite dryer, steam cycle, ASU and CCS 

installations 

At the stage of adopting assumption for its construction it was decided to use provided by the 

GateCycle program the fluidized bed boiler block, which consists of: furnace chamber (AC), 

evaporator (EVAP) and last sections of the live steam superheater (PP II) and reheated steam 

superheater (PW II). In the direction of flue gas flow this block ends with particle separator 

(cyclone). After the cyclone gas/gas type heat exchanger (PPO), in which air from ASU installation 

is heated to the temperature of 850°C, was placed. Subsequently following exchangers were placed: 

live steam superheater (PP I), reheated steam superheater (PW I), economizer (ECO), recirculated 

198

exhaust gas heater (PRS) and nitrogen heater (PA). The last devices in the direction of flue gas flow 

are: electrostatic precipitator (EF), flue gas fan (W1) and the flue gas dryer (OS) [3,4]. 

Table. 1. Main assumptions for OXY type CFB boiler 

Lower Heating Value kJ/kg 9960 

Feedwater flow kg/s 431.02 

Feedwater temperature ºC 297 

Feedwater temperature at the outlet of ECO II °C 340 

Steam temperature at the outlet of the evaporator °C 480 

Live steam temperature at the outlet of the boiler ºC 604.9 

Live steam pressure at the outlet of the boiler MPa 30.1 

Reheated steam flow at the outlet of the boiler kg/s 364.82 

Reheated steam temperature at the outlet of the boiler ºC 622.4 

Reheated steam pressure at the outlet of the boiler MPa 5.12 

Temperature difference at the cold side of ECO I K 55 

Oxidant excess ratio - 1.2 

Oxygen content in oxidizer fed to the boiler % 30 

Temperature difference at the hot end of recirculated flue gas heater 

K 30 

PRS 

Ambient pressure kPa 101.32 

Ambient temperature ºC 15 

The boiler is supplied with lignite composed of: C - 28.60%, S - 0.95%, N - 0.25%, H - 2.20%, O - 

8.00%, ash – 17.50%, moisture – 42,5% [3]. Main assumptions for the CFB boiler adopted for the 

calculations are shown in Table 1. 

In the present model a three – end type high temperature membrane was treated as a black box, 

where at a given composition of the permeate (in this case 100% composed of oxygen) the recovery 

ratio of oxygen was a decision variable [2,5,6]. Recovery ratio of oxygen during the calculation was 

changed from the value of 45% to 90%. Heat losses were not included, so each of the streams 

within the membrane has a temperature of 850°C. The pressure losses within the same membrane 

were not assumed as well. Chosen assumptions for the calculations for an air separation unit are 

shown in Table 2. 

Oxygen recovery ratio is defined as the ratio of oxygen in the stream which permeated through the 

membrane (permeate) to a stream of oxygen contained in the air feeding the membrane (feed). 

Table. 2. Main assumptions for ASU 

Air pressure at the outlet of compressor kPa 1400 

Feed temperature °C 850 

Permeate pressure kPa 42.5 

Oxygen content in the permeate % 100 

Permeate temperature °C 850 

Retentate temperature °C 850 

Oxygen temperature at the inlet to the vacuum pump °C 20 

Air compressor isentropic efficiency - 0.88 

Expanders isentropic efficiency - 0.90 

Vacuum pump isentropic efficiency - 0.88 

Temperature difference at the hot end of oxygen heater PU K 40 

Lignite dryer model was built as a simple heat exchanger in which the fuel stream is dried by the 

drying medium (a mixture of nitrogen and the oxygen derived from ASU). Drying medium is pre- 

199

heated in a heat exchanger placed in the path of exhaust gas within the convective pass of the CFB 

boiler. Chosen assumptions for a lignite dryer are shown in Table 3. 

Table. 3. Main assumptions for lignite dryer 

Lignite temperature at the inlet to the dryer °C 15 

Drying medium temperature at the outlet of the dryer °C 130 

Minimum temperature difference between lignite and drying medium K 20 

Temperature difference at the hot end of drying medium heater PA K 30 

Drying medium pressure at the outlet of the drying medium fan kPa 108 

3. Methodology and computational algorithm 

Lower heating value of lignite with a given in section 2 of this paper composition was calculated 

using the following Dulong formula variant: 

W 

d 

o 

340. 

80 c 1427. 

70 h 

92. 

90 s 

25. 

50 

8 

200 

kJ 

w 9 h 

 

kg 

where: c , s, 

o, 

h, 

p, 

w - mass fractions of: carbon, sulfur, oxygen, hydrogen, nitrogen, ash and 

moisture in the fuel. 

In the calculation process for lignite dryer it was necessary to determine the specific heat of coal 

before and after drying. For this purpose the following correlation was used: 

c c c 

s 

c o c h c n c p c w 

c 

w c s o h n p H 2O 

where: c c - specific heat of carbon (0.71 kJ·kg-1 ·K -1 ), 

c s - specific heat of sulfur (0.71 kJ·kg -1 ·K -1 ), 

c o - specific heat of oxygen (0.92 kJ·kg-1·K -1 ), 

c h - specific heat of hydrogen (14.304 kJ·kg-1·K -1 ), 

c n - specific heat of nitrogen (1.04 kJ·kg-1·K -1 ), 

c p - specific heat of ash (0.8 kJ·kg-1·K -1 ), 

c H 2O 

- specific heat of moisture (4.19 kJ·kg-1·K -1 ). 

The amount of heat that can be used to evaporate the moisture was determined from the formula: 

Q 

Q 

Q 

I 

I 

odp 

7a 

8a 

where: 7a , Q8a 

Q 

1c 

2c 

- heat flux of drying medium feeding and leaving lignite dryer, 

, I I 

1c 

2c 

- physical enthalpy flux of raw lignite and lignite leaving dryer. 

The mass flow of moisture evaporated in the drying process from the fuel is determined by the 

formula: 

m 

Q 

odp 

H O_odp 

2 ( r cp 

T 

) 

(3) 

(4) 

(2) 

(1)

where: r - enthalpy of vaporization, kJ·kg -1 , 

p c - specific heat of water vapor, (1.88 kJ·kg-1 ·K -1 ), 

T - temperature increase of water vapor derived from the lignite. 

After evaporation of the moisture contained in the fuel the new lower heating value of the fuel must 

be determined: 

1 

w w w 

2c 1c 2c 

W W 

 

r 

w 

 

 

 

w 

 

d_2c d_1c 

1 

1c 1 

1c 

where: d W - lignite lower heating value, kJ·kg-1 , 

w - moisture content in lignite, kg H2O·kg lignite -1 . 

The whole computational process aims at setting a new stream of lignite fed to the boiler, that it 

was possible to determine the thermal efficiency of the boiler [7]: 

 

k 

 

 

m5s h h m 

h 

h 

5s 

where: 5s 8s ,m 

 

1s 

m 

1c 

W 

8s 

d_1c 

8sl 

m – streams of live and reheated steam, kg·s-1 , 

6s 

h 5s – live steam enthalpy at the outlet of the boiler, kJ kg -1 , 

h 1s – feedwater enthalpy at the inlet to the boiler, kJ kg -1 , 

h h – reheated steam enthalpy at the inlet and outlet of the boiler, kJ kg -1 , 

8s 

6s 

m 1c – raw lignite stream fed to the boiler, kg s -1 , 

W d_1c – lower heating value of lignite fed to the boiler, kJ kg -1 . 

4. Results of calculations 

Using a model with the assumptions that were made for calculations the impact of drying of lignite 

supplied to the boiler on selected characteristics of the system were determined. 

The first stage of the calculations was to determine the characteristics of the calculated lower 

heating value of fuel as a function of the changing oxygen recovery ratio at high temperature 

membrane. The results of this analysis are shown in figure 2. For the variant without drying the fuel 

heating value is maintained at a constant level of 9960 kJ/kg. In a variant with fuel drying the whole 

stream of the available nitrogen – oxygen mixture was used for drying. The amount of the drying 

medium flow varies inversely to changes in oxygen recovery ratio. That means that the lower 

oxygen recovery ratio the greater the flow of drying medium, and thus a greater degree of fuel 

drying. As it can be seen, the highest lower heating value of the fuel equal to 18210 kJ/kg was 

obtained for oxygen recovery ratio 0.45, then the lower heating value decreases to reach the value 

12911 kJ/kg for the oxygen recovery ratio equal to 0.90. 

201 

(5) 

(6)

Fig. 2. Lower heating value of fuel as a function of the oxygen recovery ratio in the membrane 

The next step was to examine how the flow of the lignite supplied to the boiler changes as a 

function of the oxygen recovery ratio at high temperature membrane. The lignite stream fed to the 

boiler, at constant parameters of the circulating agent, is significantly influenced by the load of the 

heat exchanger, in which air stream delivered to the high-temperature membrane is heated. 

Fig. 3. Lignite stream supplied to the boiler as a function of the oxygen recovery ratio in the 

membrane 

The lower the oxygen recovery ratio in the membrane the greater the intake air stream needed to 

keep the assumed excess air ratio in the combustion chamber and hence, a greater load on the heat 

exchanger. As it can be seen in figure 3, for the variant without drying the amount of the supplied 

coal for the oxygen recovery ratio from 0.45 to 0.9 it decreases from 186 kg/s to 147 kg/s, while in 

202

the case of the drying of the fuel in the same range of oxygen recovery ratio stream of coal 

decreases from 156 kg/s to 136 kg/s. 

As a result of the process of producing technical oxygen in the air separation unit based on three - 

end type high-temperature membrane in comparison with the classic coal boiler, a significant 

reduction in thermal efficiency of the boiler can be observed. The results of the analysis of the 

impact of the oxygen recovery ratio in the membrane on the boiler thermal efficiency are shown in 

figure 4. In the variant without drying of the fuel for the oxygen recovery ratio equal to 0.45 boiler 

thermal efficiency is only 63.6%. When drying of the lignite was implemented, for the same oxygen 

recovery ratio thermal efficiency of the boiler increased by over twelve percentage points, to 75.8%. 

Fig. 4. Boiler thermal efficiency as a function of the oxygen recovery ratio in the membrane. 

For the variant with fuel drying, despite a higher LHV of fuel for low volumes of recovery of 

oxygen recovery ratios in the membrane, the boiler thermal efficiency increases with the oxygen 

recovery ratio up to the value of 86.8%. For the oxygen recovery ratio equal to 0.9 in the variant 

with fuel drying an increase in thermal efficiency of the boiler compared to the variant without 

drying by six percentage points was observed. 

Summary 

This paper presents a model of the supercritical circulating fluidized bed OXY type boiler 

integrated with the air separation unit based on three - end type high temperature membrane and the 

installation of fuel drying and results of studies of the impact of lignite drying on a number of 

characteristics, such as mass flow of fuel supplied to the boiler and boiler thermal efficiency. A 

sensitivity analysis of a built model shows that a change of oxygen recovery ratio at a high 

temperature membrane causes an increase in thermal efficiency of the boiler. For oxygen recovery 

ratio from 0.45 to 0.9 for the installation with a working lignite dryer, the boiler thermal efficiency 

increases from 75.8% to 86.8%. The increase in thermal efficiency of the boiler in comparison to 

the installation without lignite dryer ranges from 12 percentage points for the oxygen recovery ratio 

equal to 0.45 to 6 percentage points for the oxygen recovery ratio equal to 0.9. Basing on the 

outcome results it can be concluded that the fuel drying in OXY type systems fed by lignite leads to 

a significant increase in the boiler thermal efficiency. In addition to an increase in efficiency, the 

use of fuel drying leads to a reduction of water stream supplied to the system and thus helps to 

reduce the likelihood of condensation of moisture from the exhaust gases. 

203






References 

[1] Chmielniak T., Kosman G., ukowicz H., Integracja instalacji wychwytu CO2 z 

kondensacyjnymi blokami energetycznymi. Rynek Energii, 2008, 6 (79), 75-81. 

[2] Pfaff I., Kather A.: Comparative thermodynamic analysis and integration issues of CCS steam 


Energy Procedia, 1 (2009), 495-502. 

[3] The technical report of step 6.1 in research topic: "Numerical simulations and systemic analysis 

of oxy - burning," the research task 2 " Development of a technology for oxy-combustion 

pulverized-fuel and fluid boilers integrated with CO2 capture." in the strategic program of 

research and development, "Advanced Technologies for Energy Generation ". 

[4] Skorek – Osikowska A., Bartela .: Model of a supercritical oxy-boiler - analysis of the 

selected parameters. Rynek Energii, 2010, 5 (90), 69 – 75. 

[5] Castillo R.: Thermodynamic analysis of a hard coal oxyfuel power plant with high temperature 

three-end membrane for air separation. Applied Energy, 88 (2011), 1480-1493. 


membranes. International Journal of Greenhouse Gas Control, 5 (2011), 7-15. 


35 (2010), 943 – 951. 

204

Abstract: 

PROCEEDINGS OF ECOS 2012 - THE 25TH INTERNATIONAL CONFERENCE ON 



Monitoring of carbon dioxide uptake in 

accelerated carbonation processes applied to air 

pollution control residues 

Felice Alfieri a , Peter J. Gunning b , Michela Gallo c , Adriana Del Borghi d , Colin D. 

Hills e 

a Department of Chemical and Process Engineering “G.B. Bonino”, University of Genoa, Genoa, Italy, 

felice.alfieri@unige.it (CA) 

b Centre for Contaminated Land Remediation, University of Greenwich, Medway, Chatham Maritime, 

United Kingdom, peter@c8s.co.uk 

c Department of Chemical and Process Engineering “G, B, Bonino”, University of Genoa, Genoa, Italy, 

michela.gallo@unige.it 

d Department of Chemical and Process Engineering “G, B, Bonino”, University of Genoa, Genoa, Italy, 

adry@unige.it 

e Centre for Contaminated Land Remediation, University of Greenwich, Medway, Chatham Maritime, 

United Kingdom, c.d.hills@greenwich.ac.uk 

The application of Accelerated Carbonation Technology (ACT) has potential for the sequestration of carbon 

in waste and geological materials. ACT also has potential to be supported by carbon credit mechanisms 

based upon the amount of carbon sequestered from industrial emissions. For this to happen, the routine 

monitoring of CO2 sequestered into the solid phase is required for the planning and operation of any 

accelerated carbonation plant. The present paper reports the preliminary results from an assessment of 

existing methods for measuring CO2 imbibed into a solid by an accelerated carbonation processes. 

Laboratory-scale experiments were carried out to evaluate the accuracy of methodologies for measuring 

mineralised carbon including: loss on ignition, acid digestion and total carbon analysis. The CO2 reactivity of 

several wastes from municipal incineration known as Air Pollution Control residues (APCr) were also 

included in the study. A detailed characterisation of the materials being carbonated, using X-ray diffraction 

(XRD), X-ray fluorescence (XRF), thermogravimetric analysis (TGA) and ion chromatography was carried 

out. The results of this study showed that monitoring CO2 during accelerated carbonation is made difficult by 

the complex mineralogy of materials such as APCrs. As such, the presence of calcium bearing species and 

polymorphs of calcium carbonate formed varied between the materials investigated. The use of an acid 

digestion technique was not subject to interference from the chemistry or mineralogy of an ash. Among the 

investigated methods, acid digestion gives the most promising results as it provided robust data on the 

amount of carbon imbibed during processing. 

Keywords: 

Accelerated carbonation technology (ACT), Air pollution control residues (APCr), CO2 uptake. 


Carbonation is a natural phenomenon occurring when gaseous carbon dioxide (CO2) reacts with 

substrate materials, resulting in the production of carbonate salts. Carbonation can be accelerated 

using management techniques such as accelerated carbonation technology (ACT) working under a 

gaseous, carbon dioxide (CO2)-rich environment [1]. Chemical stability and leaching behaviour of 

materials such as alkaline combustion residues is improved and carbonated materials can be 

diverted from landfill into beneficial use as engineering media [1-4]. 

The accelerated carbonation of alkaline combustion residues is an attractive Carbon Capture and 

Storage (CCS) option. These residues such as Air Pollution Control residues (APCr), are capable of 

combining with significant amounts of CO2, and are often generated by processes also producing 

205

large amount of CO2 [1,5-7]. APCr are produced from dry and semi-dry scrubber systems fitted to 

municipal incinerator flue stacks, which involve the injection of an alkaline powder or slurry to 

remove acid gases, particulates and condensation/reaction products. Fabric filters in baghouses are 

used after the scrubber systems to remove fine particulates (baghouse filter dust). APCr also include 

the solid phase generated by wet scrubber systems (scrubber sludge) [4]. These particulates residues 

can contain large amounts of reactive calcium species coming from the alkaline sorbents commonly 

used [8,9]. 

The amount of CO2 sequestered during industrial-scale carbonation has potential to be traded as a 

commodity. Companies, governments, or other entities buy carbon offsets in order to comply with 

caps on the total amount of carbon dioxide they are allowed to emit. This market exists in order to 

achieve compliance with obligations of Annex 1 Parties under the Kyoto Protocol, and of liable 

entities under the European Emissions Trading Scheme (EU-ETS). Carbon Capture and Storage is 

being introduced in the EU-ETS in 2013, initially for geological storage [10, 11]. It is anticipated 

that a carbon market will eventually provide a financial incentive for the minimization of CO2 

emissions from a wider range of industrial processes. In addition to geological storage of carbon, 

processes that encourage the beneficial re-use of captured carbon (e.g. in solid materials) by 

technologies such as ACT, will be supported. 

The monitoring of the amount of carbon sequestered by carbonation processes, also known as CO2 

uptake, is a key aspect of process planning and operation. Different methods to measure the CO2 

uptake by accelerated carbonation are reported in literature. In several works the CO2 uptake was 

assessed by calcimetry [12, 13], by thermo-gravimetric analysis [1, 14, 15, 16], or by gravimetric 

methods [14, 16, 18]. 

In order to ensure that the emission reductions claimed during the life time of an accelerated 

carbonation plant are verifiable and permanent, reliable methods for the monitoring of CO2 uptake 

are currently needed that are both accurate and economical. This investigation evaluates the 

suitability of three methods: loss on ignition, acid digestion and total carbon analysis. A validation 

of these analytical methods has been carried out and presented in to ensure that future 

measurements in routine analysis will be close enough to the unknown true value for the CO2 

uptake. 

2. Material and methods 

2.1. Accelerated Carbonation of APCr 

An accelerated carbonation treatment was applied to seven APCr samples (APCr 1-7) supplied by 

different incinerators in the UK. About 100 g of each APCrs were mixed with water (30% to 40% 

w/w) and treated with 100% CO2 in static reaction vessels (20 x 10 cm) held at atmospheric 

pressure for 72 hours. 

In order to investigate the effect of the accelerated carbonation on mineralogy, the APCrs were 

analysed by X-ray diffraction (XRD). A Siemens D500 diffractometer with a CuK radiation 

source at 40 kV and 30 mA was used for analysis. The APCr samples were prepared as powder 

tablets and scanned between 5° and 65° 2, with a step size of 0.02° each lasting 1.2 seconds. Peak 

identification and interpretation of the X-ray diffractograms was achieved using 

DIFFRACplus EVA software (Bruker AXS). 

Thermo-gravimetric analysis (TGA) and differential thermo analyses (DTA) was performed on both 

untreated and carbonated APCrs using a Stanton-Redcroft STA-780 Series analyser. Approximately 

10 mg of material were placed in an alumina crucible (4mm of diameter). The temperature was 

raised between 20 °C and 1000 °C at a constant heating rate of 10 °C/min. 

206

2.2. Synthetic standards 

To test the accuracy of the three methods, ten synthetic standards (STD1-10) representative of the 

mineralogical composition of APCr were formulated. Analytical grade reagents; calcium carbonate 

(CaCO3), portlandite (Ca(OH)2), lime (CaO), anhydrite (CaSO4), gypsum (CaSO4.2H2O), bassanite 

(CaSO4.0.5H2O), halite (NaCl), sylvite (KCl) and quartz (SiO2) were combined according to Table 

1. The produced synthetic standards were stored in a desiccated environment to avoid possible 

alteration due to atmospheric humidity. 

STD1 to STD4 were formulated with a high percentage of reactive calcium phases (portlandite and 

lime) and a low content of calcium carbonate, simulating the composition of an untreated APCr. 

Other standards (STD7, STD9 and STD10) were formulated without reactive calcium species and 

with higher percentage of calcium carbonate, simulating the composition of a carbonated APCr. 

The influence of the other phases including gypsum and anhydrite were also investigated. 

Table 1. Percentage mineralogical composition of synthetic standards 

Standard 

ID 

CaCO3 Ca(OH)2 CaO CaSO4 

CaSO4 

.2H2O 

CaSO4 

.0.5H2O NaCl KCl SiO2 

(Calcite) (Portlandite) (Lime) (Anhydrite) Gypsum Bassanite Halite Sylvite Quartz 

STD1 0.2 - 20.8 6.5 14.1 14.9 22.3 6.1 15.1 

STD2 14.1 21.9 10.4 18.8 - - 20.0 9.9 5.0 

STD3 20.7 37.2 - - 10.0 - 15.9 5.6 10.6 

STD4 20.6 24.4 - 6.5 - 6.2 31.8 10.5 - 

STD5 24.7 5.0 - - - 9.5 14.6 17.1 29.3 

STD6 40.0 6.9 9.9 5.4 3.9 17.1 4.9 9.2 2.7 

STD7 36.1 - - 4.5 - 14.8 26.9 17.8 - 

STD8 50.9 4.3 5.1 4.0 15.9 19.8 - - - 

STD9 51.3 - - 8.2 5.3 10.1 14.9 10.2 - 

STD10 55.6 - - 19.9 - - 14.6 - 9.9 

The synthetic standards were tested using the three methods to assess their carbon dioxide content 

[CO2 (%)]. All tests were conducted in triplicate for each material. The carbon dioxide uptake 

[CO2,uptake (%)] can be calculated as difference between carbon dioxide content of treated sample 

[CO2,treated (%)] minus the carbon dioxide content of untreated sample [CO2,untreated (%)] according to 

eq. (1). 

CO CO (%) CO (%) 

(1) 

2, uptake(%) 

2, 

treated 

2, 

untreateted 

A validation process was carried out. Carbon dioxide content [CO2(%)] in synthetic standards was 

measured and compared with the expected values. The accuracy and precision of the methods was 

evaluated. Equation (2) was used to assess the relative error of mean ( REm) [19]: 

RE m 

Z T 

(2) 

T 

where Z is the analytical result and T is the calculated true value. Precision was assessed by 

sample standard error of the mean ( SEm) (3), where is the standard deviation according to (4), n 

is the number of measurements and xi are the observed value for the sample and xm is the mean 

value of these measurements: 

SE m 

 

(3) 

n 

207

N 

1 

i ( x x 

i 

n 1 

m 

) 

2.3. Carbonation measuring methods 

Three different experimental methods were used to assess the CO2 content of the synthetic 

standards. These are summarised in figure 1. 

Acid Digestion 

5 grams of material 

20 ml of HCl 37% 

Acid – sample reaction 

(30 min) 

Weight Loss 

detemination 

Sample grinding 

Sample drying 

(105 °C) 

Loss on Ignition 760 / 980 °C Total CarbonAnalysis 

10 grams of material 

Thermal treatment 

(2 hours at 550 °C) 

Weight Loss 

detemination 

Thermal treatment 

(2 hours at 760 / 980 °C) 

Weight Loss 

detemination 

Fig. 1. Experimental methods 

208 

150 milligrams of 

material 

(4) 

CO 2 determination 

by IR analizer (5 min) 

2.3.1. Loss on ignition (LOI) 

About 10 grams of

a dilute hydrochloric acid solution (37% w/w). The pots were tightly closed and a tiny hole was 

drilled in the lid to allow the liberated gaseous CO2 to escape. The entire apparatus was reweighed, 

before the contents of the syringe were flushed into the pot. Reweighing the apparatus continued 

until a constant value was achieved. 

WBefore 

W 

After 

AD(%) 

x100 

(7) 

W 

Sample 

Where WBefore is the mass of the apparatus before digestion, WAfter is the mass of the sample after 

digestion, and Wsample is the mass of the oven dried material. 

2.3.3. Total carbon analysis (TCA) 

TCA was determined using a Hach Lange TOC IL-550 analyser. In the analyser, the sample is 

heated in a tube furnace under a stream of oxygen. Carbon present is converted to carbon dioxide, 

and the concentration carried in the exhaust gas leaving the furnace is quantified. Analytical grade 

calcium carbonate was used for the calibration of the instrument. 

3. Results and discussions 

3.1. APCr characterization 

X-Ray Diffraction was used to analyse the seven APCrs, which were composed of eight main 

phases (see table 2). XRD analysis confirms the presence of calcite (CaCO3) for all the residues. 

Portlandite (Ca(OH)2) was detected in four APCrs, which was found to correlate with the TGA 

observation. Similarly, a unique thermal interval was found in the six APCrs for the presence of 

calcium hydroxide-chloride (CaClOH). Halite (NaCl), sylvite (KCl), lime (CaO), anhydrite (CaSO4) 

and quartz (SiO2) were also identified. Accelerated carbonation of the APCrs resulted in 

mineralogical change (see table 3). The disappearance of the carbon dioxide-reactive phases 

(portlandite, lime, CaClOH) was observed, with the formation of new calcium carbonate in the form 

of vaterite. 

Table 2. Mineralogical composition of untreated APCrs 

APCr1 

APCr2 

APCr3 

APCr4 

APCr5 

APCr6 

APCr7 

Ca(OH)2 CaO CaClOH CaCO3 CaCO3 NaCl KCl CaSO4 SiO2 

(Portlandite) (Lime) (Calcite) (Vaterite) (Halite) (Sylvite) (Anhydrite) (Quartz) 

Table 3. Mineralogical composition of accelerated carbonated APCrs 

APCr1 

APCr2 

APCr3 

APCr4 

APCr5 

APCr6 

APCr7 

Ca(OH)2 CaO CaClOH CaCO3 CaCO3 NaCl KCl CaSO4 SiO2 

(Portlandite) (Lime) (Calcite) (Vaterite) (Halite) (Sylvite) (Anhydrite) (Quartz) 

209

DTA analysis of the untreated APCrs show three clear events occurring between 400 – 440 °C, 465 

– 550 °C, and 550 – 760 °C. This is illustrated in the DTA curves for APCr7 (see figure 2). 

According to Bodenan and Deniard [8] the peak in the 400 – 440 °C interval corresponds to 

Ca(OH)2 decomposition and the peak in the 465– 550 °C interval is related to that of calcium 

hydroxide-chloride (CaClOH).The 550–760 °C interval corresponds to CaCO3 decomposition. 

In the accelerated carbonated APCrs, the Ca(OH)2 and CaClOH signatures are absent, and there is 

an increase in the CaCO3 peak. DTA was used to assess the calcium phases present in the remaining 

six untreated and accelerated carbonated APCrs (see tables 4 and 5). 

Fig. 2. DTA of untreated and accelerated carbonated APCr7 

Table 4. Calcium phases of the untreated APCrs by DTA analysis 

APCr1 

APCr2 

APCr3 

APCr4 

APCr5 

APCr6 

APCr7 

Ca(OH)2 CaClOH CaCO3 

(Portlandite) (calcium hydroxide-chloride) 

TGA analyses of APCrs show slightly differing behaviours between untreated and treated materials 

up to 500 °C. These differences are due to the portlandite and calcium hydroxide chloride that were 

found to be present only in untreated APCrs. They are responsible for two small decomposition 

steps at 400 °C and 500 °C respectively, due to bound water loss. In accelerated carbonated APCrs 

these steps are not present, although a gradual decomposition process starts at temperatures lower 

than 300 °C. Most significant is a sudden change in mass in the region 550–760 °C. This mass 

210

change is significantly greater in the accelerated carbonated APCrs. A second major event begins at 

800 °C (see figure 3). 

Table 5. Calcium phases of the accelerated carbonated APCrs by DTA analysis 

APCr1 

APCr2 

APCr3 

APCr4 

APCr5 

APCr6 

APCr7 

Ca(OH)2 CaClOH CaCO3 

(Portlandite) (calcium hydroxide-chloride) 

Fig. 3. TGA of untreated and accelerated carbonated APCr7 

3.2. Carbonation Measuring Methods 

The CO2 content of the synthetic standards was measured by AD, TCA, LOI 550-760 °C and LOI 

550-980 °C and compared with known compositions. SEm and REm for all the developed methods 

and for all the tested standards are summarised in Table 4 and Table 5. 

211

Table 4. Relative Error (%) of the mean (REm) of carbon content determination for synthetic 

standards. 

Z –T / T 

STD1 STD2 STD3 STD4 STD5 STD6 STD7 STD8 STD9 STD10 

AD 2991 32,1 7,6 6,8 5,6 1,2 3,4 5,3 4,7 13,5 

TCA 515 37,75 39,70 35,2 15,0 21,8 6,8 19,7 16,5 7,1 

LOI 550-760 °C 1420 125,3 107,4 94,6 70,8 68,5 43,6 51,7 25,9 62,4 

LOI 550-980 °C 10919 132,3 82,9 83,8 89,2 31,8 360,7 3,1 35,8 6,1 

Table 5. Standard Error (%) of the mean ( SEm) of carbon content determination for synthetic 

standards. 

SEm 

STD1 STD2 STD3 STD4 STD5 STD6 STD7 STD8 STD9 STD10 N 

AD 0,4 0,2 0,9 1,2 0,8 0,9 1,1 0,1 0,9 0,2 3 

TCA 0,2 0,2 0,5 0,7 0,5 0,5 0,6 0,1 0,5 0,1 3 

LOI 550-760 °C 0,2 0,4 0,8 0,7 1,3 2,6 2,7 2,7 0,6 4,2 4 

LOI 550-980 °C 0,3 0,8 0,1 1,3 2,4 0,9 0,4 0,2 0,8 0,9 3 

3.2.1 Loss on ignition (LOI) results 

The results of the LOI analyses are shown in figure 4. LOI methods were found to have a poor 

degree of accuracy (see table 4) and a weak linear relationship between observed and calculated 

CO2 content (see figure 4). Overlaps with thermal events associated with other phases, or the 

presence of CO2 bound as different polymorphs with the effect of broadening the decomposition 

temperature range, can be considered as possible mechanisms of affecting the measurement. 

Fig. 4. CO2(%) in synthetic standards measured by LOI (550-760) and LOI (550-980). 

212

The significant endothermic event occurring in the range 550 – 760 °C is due to calcium carbonate 

decomposition. 

Treated APCrs have different thermal behaviours compared to untreated APCrs. The first 

decomposition step starts slower and at a lower temperature (Fig. 6), affecting the accuracy of the 

tested LOI methods. This decomposition step could be due to decomposition of amorphous or finely 

divided vaterite. This polymorph of calcium carbonate is the main mineralogical product of the 

accelerated carbonation process. One study [20] conducted by TGA-TDA coupled with an on-line 

gas chromatography confirms CO2 leaving the carbonated APCr samples at temperatures lower than 

500 °C. 

TGA analysis of STD7 identified further events in the 760 – 980 °C range (see figure 3). Analysis 

of Analar grade NaCl and KCl shows that decomposition of these minerals takes place within this 

range (see figure 5). Consequently, the presence of NaCl and KCl together with CaCO3 can result in 

the overestimation of CO2(%) by LOI 550-980 °C. CaSO4 and his hydrated forms (CaSO4.0.5H2O 

and CaSO4.2H2O), does not affect the LOI measurements. CaSO4.H2O profile shows weight loss 

below 500 °C, related to the bound-water loss (see figure 5). 

Fig. 5. Thermo-gravimetric profiles of STD7 compared to 100% NaCl and 100% KCl. 

3.2.2 Acid digestion (AD) results 

The results of the AD analyses are shown in figure 6. AD results show a very good linear 

relationship between observed and calculated CO2 content (R 2 value 0.9824), low values of SEm 

(Table 5), and low values of REm (Table 4) except for STD1 and STD2 characterized by a low CO2 

content (less than 5%). 

During the digestion process, a ‘rebound’ in the weight of the sample is observed (see figure 7), 

which may be ascribed to evaporation of the dilute acid, followed by condensation. Hydrochloric 

acid can produce three different exothermic reactions with the APCr or synthetic standards (8, 9, 

10). Particularly when using concentrated acid, a heating of the samples occurs due to the 

exothermic neutralization reactions. 

213

CaO HCl CaCl H O 

193kJ 

/ mol (8) 

( s) 

2 ( aq) 

2( 

aq) 

2 ( l ) 

2( 

s) 

2HCl( aq) 

CaCl 2 2H 

( aq) 

2O( 

l) 

214 

H reaction 

CaOH 

127. 

9kJ 

/ mol (9) 

3( 

s) 

2HCl( aq) 

CaCl2 

2CO 

( ) 

2 2H 

aq 

( s) 

2O( 

l) 

H reaction 

CaCO 

15. 

3kJ 

/ mol (10) 

H reaction 

In order to minimize any loss by evaporation, the acid must be diluted to 30% w/w to minimise 

heating of the system. 

Fig. 6. CO2(%) in synthetic standards measured by AD. 

Fig. 7. Rebound effect in AD measurements. 

3.2.3 Total Carbon Analysis (TCA) results

The results of the TCA analyses are shown in figure 10. The correlation coefficient shows a good 

relationship between the observed and calculated values. However, the accuracy of the TCA results 

were found to be lower than those for AD (see table 4). 

Fig. 8. CO2 (%) in synthetic standards measured by TCA. 


The accurate measurement of bound carbon dioxide in reactive materials, including wastes, is 

important for the purposes of assessing Carbon Capture and Storage. Municipal incineration APC 

residues are mineralogically complex materials. XRD and TGA-DTA have identified the presence 

of calcium-bearing minerals e.g. lime, portlandite, calcium hydroxide-chloride and calcium 

carbonate. Since different reaction mechanisms and carbonation pathways can be involved, the 

accurate measurement of imbibed carbon dioxide becomes challenging. 

APCrs subjected to accelerated carbonation show a decrease in the minimum temperature and a 

broadening of the range at which calcium carbonate decomposes. TGA-DTA analyses have shown 

that calcium carbonate decomposition starts at temperatures lower than 500°C for treated APCr, at 

which point overlap with thermal events associated with calcium hydroxide and calcium hydroxide 

chloride occurs. Consequently, this has a significant effect upon the measured values obtained with 

thermo-gravimetric methods using fixed temperature ranges. 

Among the tested methods, acid digestion has been shown to have good precision and a linear 

correlation between measured and calculated CO2 contents. Total carbon analysis measurements 

show a good linear correlation with calculated results, but overall accuracy is lower while loss on 

ignition was found to be unreliable due to different thermal behaviour between APC residues before 

and after carbonation. 

Obtained CO2 uptake values, corrected for the mass balance (input/output) of the carbonation 

process, can be integrated during the operational life of the plant through a continuous monitoring 

process. This gives the opportunity of monitoring the total amount of CO2 sequestered and the 

related amount of potential carbon credits. 

References 

215

[1] M. Fernández Bertos, S.J.R. Simons, C.D. Hills, P.J. Carey, A review of accelerated 

carbonation technology in the treatment of cement-based materials and sequestration of CO2, 

Journal of Hazardous Materials 2004;B112:193–205 

[2] Peter J. Gunning, Colin D. Hills, Paula J. Carey, Accelerated carbonation treatment of industrial 

wastes, Waste Management 2010;30:1081–1090. 

[3] M. Fernández-Bertos, X. Li, S. J. R. Simons, C. D. Hills and P. J. Carey, Investigation of 

accelerated carbonation for the stabilisation of MSW incinerator ashes and the sequestration of 

CO2, Green Chem. 2004;6:428-436. 

[4] D. Amutha Rani, A.R. Boccaccini, D. Deegan, C.R. Cheeseman, Air pollution control residues 

from waste incineration: Current UK situation and assessment of alternative technologies, 

Waste Management 2008;28:2279–2292 

[5] E.-E. Chang, Shu-Yuan Pan, Yi-Hung Chen, Hsiao-Wen Chu, Chu-Fang Wang, Pen-Chi 

Chiang (2011), CO2 sequestration by carbonation of steelmaking slags in an autoclave reactor, 

Journal of Hazardous Materials 2011;186(1):558-564. 

[6] Renato Baciocchi, Andrea Corti, Giulia Costa, Lidia Lombardi, Daniela Zingaretti, Storage of 

carbon dioxide captured in a pilot-scale biogas upgrading plant by accelerated carbonation of 

industrial residues, Energy Procedia 2011;4:4985–4992. 

[7] Valentina Prigiobbe, Alessandra Polettini, Renato Baciocchi, Gas–solid carbonation kinetics of 

Air Pollution Control residues for CO2 storage, Chemical Engineering Journal 2009;148:270– 

278. 

[8] F. Bodenan, Ph. Deniard, Characterization of flue gas cleaning residues from European solid 

waste incinerators: assessment of various Ca-based sorbent processes, Chemosphere 

2003;51:335–347. 

[9] T. Van Gerven, D. Geysen, L. Stoffels, M. Jaspers, G. Wauters, C. Vandecasteele, Management 

of incinerator residues in Flanders (Belgium) and in neighbouring countries. Waste 

Management 2005;25:75–87 

[10] EUROPA- the official web site of the European Union. Official Journal of the European 

Communities. Directive 2009/29/EC of the European Parliament available at: 

http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2009:140:0063:0087:en:PDF. 

[accessed 24.01.2012]. 

[11] UNFCCC, CMP 6 - Cancun: Carbon dioxide capture and storage in geological formations as 

clean development mechanism project activities 

http://unfccc.int/files/meetings/cop_16/application/pdf/cop16_cmp_ccs.pdf . [accessed 

24.01.2012]. 

[12] G. Cappai, S.Cara, A.Muntoni, M.Piredda, Application of accelerated carbonation on MSW 

combustion APC residues for metal immobilization and CO2 sequestration. Journal of 

Hazardous Materials Article in Press (2011). 

[13] R. Baciocchi, G. Costa, E. Di Bartolomeo, A. Polettini, R. Pomi, Comparison of different 

process routes for stainless steel slag carbonation. Energy Procedia 2009;1:4851–4858 

[14] Deborah N. Huntzinger, John S. Gierke, Lawrence L. Sutter, S. Komar Kawatra, Timothy C. 

Eisele, Mineral carbonation for carbon sequestration in cement kiln dust from waste piles. 

Journal of Hazardous Materials 2009;168:31–37. 

[15] V. Morales-Flórez, A. Santos, A. Lemus, L. Esquivias, Artificial weathering pools of 

calcium-rich industrial waste for CO2 sequestration, Chemical Engineering Journal 

2011;166:132–137. 

[16] Xiaomin Li, Marta Fernandez Bertos, Colin D. Hills, Paula J. Carey, Stef Simon, 

Accelerated carbonation of municipal solid waste incineration fly ashes Waste Management 

2007;27:1200–1206 

216

[17] Lei Wanga, Yiying Jin, Yongfeng Nie, An investigation of accelerated and natural 

carbonation of MSWI fly ash with a high content of Ca, Journal of Hazardous Materials 

2010;174:334–343 

[18] Frederic Clarens, Fidel Grandia, Sandra Meca, Lara Duro, Joan de Pablo. Determination of 

CO2 sequestration capacity and stabilization of MSWI fly ash through accelerated carbonation. 

Proceedings of the third international conference on Accelerated Carbonation for 

Environmental and Materials Engineering, Turku (Finland), 29 November - 1 December 2010 

[19] A. Gustavo Gonzalez, M. Angeles Herrador, A practical guide to analytical method 

validation, including measurement uncertainty and accuracy profiles, Trends in Analytical 

Chemistry 2007;26(3):227-238. 

[20] Renato Baciocchi, Alessandra Polettini, Raffaella Pomi, Valentina Prigiobbe, Viktoria 

Nikulshina Von Zedwitz and Aldo Steinfeld (2006). CO2 Sequestration by Direct Gas-Solid 

Carbonation of Air Pollution Control (APC) Residues. Energy & Fuels; 2006(20):1933-1940. 

217

Abstract: 




Process efficiency and optimisation of 

precipitated calcium carbonate (PCC) production 

from steel converter slag 

Hannu-Petteri Mattila a , Inga Grigalinait b , Arshe Said c , Sami Filppula d , 

Carl-Johan Fogelholm e and Ron Zevenhoven f 

a Thermal and Flow Engineering Laboratory, Åbo Akademi University, Turku, Finland, 

hmattila@abo.fi, CA 

b Thermal and Flow Engineering Laboratory, Åbo Akademi University, Turku, Finland, 

currently at Feyecon, Weesp, The Netherlands, ingagrigaliunaite@feyecon.com 

c Department of Energy Technology, School of Engineering, Aalto University, Espoo, Finland, 

arshe.said@aalto.fi 

d Thermal and Flow Engineering Laboratory, Åbo Akademi University, Turku, Finland, sfilppul@abo.fi 

e Department of Energy Technology, School of Engineering, Aalto University, Espoo, Finland, 

carl-johan.fogelholm@aalto.fi 

f Thermal and Flow Engineering Laboratory, Åbo Akademi University, Turku, Finland, rzevenho@abo.fi 

CO2 emissions could be reduced with capture and storage (CCS) methods. CCS causes costs for the 

industry, creating an extra barrier for implementation of these techniques. A recently developed mineral 

carbonation process producing a valuable product covering process implementation costs is advanced 

towards commercial scale. In this two-step process, calcium-rich steel converter slag is treated with an 

aqueous ammonium salt solution to selectively extract the calcium. Then the dissolved calcium is removed 

from the process liquid as pure precipitated calcium carbonate (PCC) by introducing gaseous CO2 to the 

system. This concept would utilise waste material from steel industry, and spare the natural resources used 

for conventional PCC production. Experimental results have shown that papermaking grade PCC can be 

produced with ammonium nitrate, chloride or acetate solvents. The used solvent has also been regenerated 

and recycled between the process stages, which reduces the need for fresh solvent and lowers the process 

costs. To maximise the PCC production in a continuous process, a thermodynamic process model in Aspen 

Plus is now constructed. The modelling results are validated with a series of semi-continuous experiments. It 

is found that by using staged carbonate precipitation the yield of pure product is increased. Ammonium 

nitrate is observed to enable the highest conversion of calcium from steel slag to calcium carbonate also in 

continuous operation. At ambient conditions the process steps generate heat but this low temperature heat 

cannot be utilised. The mixing and pumping energies for process operation are small in comparison to the 

heat duties in chemical reactors. Preliminary results on solid-liquid separation studies, on washing of the 

solid outlet streams, as well as on recovery and make-up needs of process chemicals are discussed. The 

work is a continuation of work presented by Said et al. at 22 nd ECOS in Brazil 2009. 

Keywords: 

Ammonium salt solution, Chemical thermodynamics, Mineral carbonation, pH swing process, Steel 

converter slag. 


It is a widely accepted fact that the climate on Earth is changing because of human activities. One 

major factor causing global warming is the high emission rate of so-called greenhouse gases. Of 

these, carbon dioxide is an especially problematic one, since modern economies are based on 

combustion of fossil, carbon-based fuels. Thus the amount of CO2 emissions is also high compared 

to other greenhouse gas emissions like N2O that are possibly more harmful to the climate calculated 

per mass only [1]. 

Various methods have been proposed to diminish CO2 emissions. Apart from improving the 

efficiencies of existing combustion processes, choosing CO2 neutral fuels and decreasing the need 

218

of power and electricity on global scale, carbon capture and storage (CCS) technologies are seen as 

one means of reducing carbon dioxide emissions. In CCS, carbon is removed from the gases at 

different stages in combustion or gasification processes depending on the chosen technology. The 

capture techniques are already on a relatively mature level, especially for oxygen-free gases, since 

they have during several decades been applied for industrial processes to produce pure CO2 [2]. 

Thus, one major threshold in CCS applications is currently the lack of a permanent, leakage-free 

storage [1]. 

Mineral carbonation, where the carbon dioxide gas is reacted directly with some magnesium or 

calcium containing compounds, is one option for safe storage. Simultaneously it offers the 

possibility of removing the separate capture stage. However, the mineralization reactions with 

natural materials are very slow by their nature, and thus development is still needed before 

industrial applications [3]. 

Apart from this, there are industrial waste materials, in which calcium or magnesium is present in 

more reactive forms [3-7]. For example, steel slags from steelmaking plants have been found to 

react with suitable solvents in such a way that calcium is selectively extracted from the slag matrix. 

This selective reaction enables the utilization of the produced carbonates in other industrial 

processes. Thus, two waste streams, CO2 in flue gases and calcium in steel slags can be used for 

generating a marketable calcium carbonate product [3, 6, 7]. 

In this paper, a two-step pH swing process (Fig. 1) operating at ambient temperature and pressure is 

discussed. In this concept, calcium from steel converter slag is first leached out with ammonium 

chloride, nitrate or acetate solution under slightly acidic conditions, and then, after separating the 

slag residue, the dissolved calcium is carbonated in a separate reaction vessel with a feed of gaseous 

CO2. This CO2 gas could be absorbed to the process liquid directly from the flue gas stream from 

e.g. a lime kiln or a steel plant, and thus there would be no need for a separate capture unit. 

Naturally, the concentration of CO2 in the flue gas would be a crucial factor for the process kinetics. 

In the carbonation step the ammonium salt solvent is regenerated and can be recycled back to the 

extraction reactor after the precipitated carbonate product has been separated. 

Fig. 1. A principal process scheme of the two-step pH swing process. 

In this text especially the efficiencies and losses of different process steps are evaluated based on 

both thermodynamic modelling and experimental work. The development and scale-up of the 

process from laboratory scale batch tests towards industrial applications is discussed too [8-9]. 

The general target of the current study is to produce calcium carbonate that would meet the 

demands of papermaking applications; if the product could be utilised as filler or a coating pigment, 

Precipitated Calcium Carbonate (PCC), the profits of selling this material would according to 

preliminary calculations cover the costs of the carbonation process. This in turn would remove or 

lower the financial threshold of the utilization of CCS technologies, which usually bring large costs 

to CO2 generating industries or power plants. Also, the financial incentive would enable a larger 

scale demonstration of this mineral carbonation concept [3, 6, 9]. To ensure a high quality product, 

the CO2 feed should be free of sulphur and particulate matter. 

219

2. Process development 

2.1. Background 

During earlier studies [6, 10, 11], several batch experiments have been performed to verify the 

effects of different process changes such as temperature and CO2 pressure on the process chemistry. 

It was found that the reaction steps, which both generate some heat (see Section 4), could be 

performed at room temperature (20–30 °C). An increase in temperature to 70 °C had hardly any 

effect on the extraction stage kinetics. In carbonation the shape of the precipitated particles changes 

with the temperature. This restricts the applicable carbonation temperatures below 30–40 °C. The 

effect of carbon dioxide pressure was limited to changes in process kinetics, the precipitation rate 

being slower at lower partial pressures of CO2. Thus, 20–30 °C was chosen as a suitable process 

operation temperature interval. At these temperatures the kinetic reaction rates are acceptable [10] 

and solubility and volatility of gaseous components such as NH3 and CO2 are beneficial for the 

process. 

Steel converter slag was reported to contain calcium as free lime (CaO), larnite (Ca2SiO4) and 

various calcium-iron compounds that seem not to react with ammonium salt solvents. The 

dissolution reactions of lime and larnite are presented as (R1) and (R2). Carbonate precipitation 

chemistry can be summarised as (R3) and (R4). X in the reaction equations represents Cl - , NO3 - or 

CH3COO - (acetate), depending on the chosen salt. 

( s) 

2NH 

Xaq 

H Ol 

CaX 

aq 2NH 

OHaq 

(R1) 

CaO 4 

2 

2 

4 

aq H Ol 

CaX 

aq CaO 

SiO ( s) 

2NH 

OHaq 

2CaO SiO2 

( s) 

2NH4 

X 

2 

2 

2 

(R2) 

4 

aq CO ( g) 

NH CO aq H Ol 

2NH4 OH 2 

4 3 

(R3) 

2 

CO aq CaX aq CaCO ( s) 

2NH 

X aq NH 4 2 3 

2 

3 

4 

2 

(R4) 

In the batch experiments it was also observed that if solvents with molarities higher than 1.0 mol/L 

or solid-to-liquid ratios higher than 100 g/L were used, also some iron and manganese was extracted 

from the steel slag, decreasing the purity and whiteness of the process solution and the produced 

carbonates [10-12]. Thus, these specifications were used in the current work. Mainly ammonium 

chloride solutions were used, both in modelling work and in experiments, since the available data 

were most complete for this solvent, and also because it is cheaper than the other two ammonium 

salts. Some observations of experimental work with ammonium nitrate will be presented in later 

sections. 

In larger scale, the process should preferably be operated on continuous basis to achieve 

presumably lower operational costs and better adaptability to changes in feedstock quality. 

Problems arising especially from the continuous operation as identified already by [9, 13] are the 

losses of solvent components (NH3 and water vapour, ammonium and calcium salt precipitates) and 

dissolution of excess carbon dioxide as bicarbonate and carbonate ions in the carbonation step. The 

solvent losses cause an unnecessary increase in process costs, both as a need of a solvent make-up 

but also as a need for purification of precipitates and purged gases. On the other hand, if excess 

dissolved carbon species are recycled from carbonation to the extraction unit, solid calcium 

carbonate is precipitated on the slag particles, lowering the overall production rate of pure carbonate 

product (PCC). These problems exist to some extent also in a batch type process. 

2.2. Process modelling 

Thermodynamic modelling and simulation software Aspen Plus 7.2 were utilised to study the 

possibilities to decrease losses of both carbonate product and solvent components. The software 

220

also provided information on sizes and compositions of different process streams. The model design 

is shown in Fig. 2. It consists of one extraction step “EXTRACTO”, and two carbonation steps, 

“CARBONAT” and “SETTLER”. All these reactors are so called RGIBBS units, which calculate 

the output by minimizing Gibbs’ free energy of the system. Carbonation is divided in two stages to 

enhance the precipitation rate of calcium (see Section 3.1.). 

25 ton/h dry steel converter slag, containing 5%-wt CaO and 59%-wt Ca2SiO4, the remaining 36%wt 

consisting of inert compounds, is fed to the extraction reactor together with an ammonium salt 

solution (~1 mol/L NH4Cl). The solid-to-liquid ratio used in modelling is approximately 100 g slag 

in one litre of solvent [10-11]. 

In the first carbonate precipitation unit, “CARBONAT”, 85-100% of the calcium-rich solution from 

extraction unit is put in contact with 25 ton/h flue gas containing 20% CO2 and 80% N2, being 

approximately the composition of lime kiln flue gases. The 300 °C flue gas is fed to the process via 

a cooling unit to estimate the released heat, when the gas is brought down to room temperature 

(20°C). The flue gas feed amount is adjusted so that approximately 5% of the CO2 gas leaves the 

system unreacted. After the first step the solution is flashed to 0.5 bar in “VAPORSEP”, thus 

removing the non-reacted gases. Re-pressurisation to 1.0 bar is done by a separate “PUMP” unit. 

The “SETTLER” unit is used to increase the pH of the once carbonated solution, so that the 

chemical equilibrium can be shifted to favour additional precipitation of calcium carbonate. This is 

done by introducing 0-15% of the calcium-rich solution coming from the extraction unit directly to 

“SETTLER” as a bypass stream of “CARBONAT”. At the same also some calcium is added to the 

“SETTLER” reactor, shifting the equilibrium even further towards carbonate. After this second 

precipitation unit the solution is again flashed to 0.5 bar in “VAPORSE2” and pressurised to 1.0 bar 

with “PUMP2” to remove dissolved gases. 

Fig. 2. A detailed process scheme of the pH swing process as simulated with Aspen Plus. 

Solids are separated from the process solution with “RESIDSEP” for slag residue and “PRODSEP” 

and “PRODSEP2” for carbonate product. These units are based plainly on percentages of separation 

efficiency specified for each component. They can be thickeners, filters, hydrocyclones and 

combinations of these. Later in this paper some studies on gravitational separation of steel slag are 

presented. Both the slag residue (“RESIWASH”) and produced PCC (“PRODWASH”) are washed 

221

to dissolve the chloride salts precipitated on these particles. To demonstrate the recovery of 

vaporised ammonia from the carbonation steps, an additional RGIBBS unit, “NH3SCRUB” is 

introduced to the model. In this unit an HCl solution is reacted with NH3 vapour to produce aqueous 

NH4Cl. These aspects are discussed in more detail in Section 5. 

The solvent liquid is recycled in the process. Also make-up streams for ammonia and water are 

included in the model to maintain the balance with vaporization losses. The “CO2SEPAR” unit is 

not active in this model, although it could be used to artificially decrease the concentrations of 

carbon species in the recycled stream. 

2.3. Experimental work 

A set of five experiments was performed to provide information that could not be obtained from the 

modelling work, but also to confirm that the modelling results can be applied for predicting the 

behaviour of the process in practice. Thus, the product yields and conversion rates from the Gibbs 

energy minimization performed in Aspen Plus software and experimental work could be compared 

with each other, but also the purity and crystal shape of PCC product, properties that are difficult to 

model, were studied in experiments. In future, the exact experimental result values may be used in 

modelling work to get better estimates for the large scale stream sizes and energy needs of the 

process. 

The tests were done at ambient conditions (20°C, 1 bar), with a mechanical stirrer (170 rpm) in the 

extraction reactor, using 1 mol/L ammonium chloride or ammonium nitrate solution, and 

maintaining the slag-to-liquid ratio at approximately 100 g/L during extraction. After each 

experiment the process equipment was modified according to the obtained results before the next 

experiment (Tables 1 and 2). 

Table 1. Experimental parameters 

Experiment 1 2 3 4 5 

Reaction time, min 75 135 180 255 255 

Ammonium salt NH4Cl NH4Cl NH4Cl NH4Cl NH4NO3 

Volume of ammonium salt, ml 750 1500 2500 3500 2500 

System filled in the beginning No No Yes Yes Yes 

Initial slag amount, g 50 50 50 50 50 

Total slag amount used, g 50 70 90 110 110 

Table 2. Amount of equipment used in experimental set-ups 


Filters 2 4 4 4 4 

Settlers 0 1 0 1 0 

Pumps 2 4 5 4 4 

H2O lock 0 1 1 1 1 

pH adjustment unit 0 0 1 1 1 

CO2 removal unit 0 0 1 1 1 

Table 3. Volumes, volume flows and residence times in experiments 4 and 5 

Unit Volume, ml Volume flow, ml/min Residence time, min 

Extraction 500 20 25 

Carbonation 300 15 20 

pH adjustment 500 20 25 

CO2 removal 500 20 25 

Filter equipment 600 N/A N/A 

222

Fig. 3. Process scheme for experiments 4 and 5. 1 – Extraction step; 2 – Carbonation step; 3 – 

water lock; 4 – pH adjustment unit; 5 – CO2 removal unit; 6 – Settler (not used in experiment 5); 7 

– Filtration step; 8 – Pump. 

Fig. 4. Process equipment used in experiments 4 and 5. 

Figs. 3 and 4 represent the process equipment of experiments 4 and 5, for which the most successful 

and advanced setup was constructed. Volume flows and residence times for these tests are shown in 

Table 3. In experiment 4 the whole system was filled with ammonium chloride solution, of which 

0.5 L with 50 g steel converter slag (size fraction 125-250 µm) was fed to the extraction reactor. 5 g 

slag was added to the extractor with 15 minutes interval during the whole experiment. To allow for 

the calcium concentration in the extractor to increase towards a constant level, the solution was 

allowed to react 5 minutes before the pumps were switched on. Fine particles were continuously 

removed from extraction reactor via an overflow of 20 ml/min. The overflow was filtered to prevent 

solid particles from entering the carbonation stages. 75% of the filtrate (15 ml/min) was pumped to 

the 1 st carbonation reactor, where a small overpressure (0.02 bar) of carbon dioxide (99.9996%, Oy 

Aga Ab) was maintained with a water lock. CO2 was introduced through a Waters 10 micron Hplc 

solvent inlet filter (WAT025531) to provide an even gas distribution. Pure CO2 gas was used 

because of the impracticality of actual flue gas, but as mentioned above, this has been found to 

mainly affect the process kinetics, if the flue gas is sulphur-free and contains no solid particles. 

223

The remaining 25% (5 ml/min) of the filtrate from extraction was fed to the 2 nd precipitation stage 

to adjust the pH of this step to approximately 7. The neutralised solution was transferred by 

gravimetric flow into a vessel, where 99.9999% nitrogen (Oy Aga Ab, 0.5 L/min) was bubbled 

through the liquid to remove the unreacted CO2. Finally, the slurry was allowed to settle in another 

vessel, after which the overflow was filtered with 0.45µm/47mm membrane (Pall Life Sciences, 

Supor-450) and the particle-free solution (20 ml/min) was recycled back to the extraction reactor. 

Since the product was not continuously removed from the system, this process can be described as 

semi-continuous. For comparison, the experiment was repeated with ammonium nitrate solvent, 

other parameters unchanged. The settler, however, was not used in the nitrate experiment, since it 

was observed to have no effect in experiment 4. 

Samples were taken of both the slag residue and the produced PCC. They were dried overnight at 

105°C and analysed with SEM/EDX to get an approximate elemental composition. The 

concentrations of dissolved calcium ions were measured after each step with an ion selective 

electrode (NICO 2000). 

3. Conversion efficiencies of calcium and CO2 to PCC 

3.1. Modelling results 

Theoretically, based on presented assumptions on extraction chemistry and slag composition, 44.7% 

of the calcium present in the feed can be extracted. However, if only one carbonation unit is used, 

almost 7% of this extracted calcium will precipitate as a mixture with slag residue in the extraction 

reactor. Table 4 represents the effect of different bypass fractions on the efficiencies of both 

calcium and carbon dioxide conversions. As can be seen, according to the model already a 5% 

bypass stream is enough to reduce the losses by 40%. With a 15% bypass practically all extracted 

calcium could be utilised as marketable product. On the other hand, the amount of sequestered CO2 

remains constant, since the total amount of CaCO3 precipitate is not changing. 

Table 4. Modelled conversion efficiency calculations of calcium and CO2 to PCC with various 

process configurations 

Bypass fraction % 0 5 10 15 

Reactive calcium feed kmol/h (ton/h) 108 (4.3) 

PCC product kmol/h (ton/h) 101(10.1) 104(10.4) 107(10.7) 108(10.8) 

PCC in residue kmol/h (ton/h) 7(0.7) 4(0.4) 1(0.1) 0(0) 

Loss of marketable calcium % of reactive input 6.5 3.7 0.9 0.0 

Conversion of total Ca to PCC % 42 43 44 45 

CO2 feed kmol/h (ton/h) 114 (5.0) 

CO2 not sequestered kmol/h (ton/h) 6 (0.3) 

% 5.3 

Table 5. Modelled changes in solution pH values at different stages of the process 

Bypass fraction % 0 5 10 15 

pH after extraction 10.61 10.70 10.83 11.13 

pH after 1 st carbonation 7.68 7.64 7.61 7.61 

pH after 2 nd carbonation 7.71 7.87 8.18 8.68 

224

Fig. 5. Theoretical dependency of dissolved Ca 2+ equilibrium concentration at 20°C on pH and 

CO2 pressure modified from [14]. 

The chemical explanation for this behaviour can be summarised in Table 5 and Fig. 5. When the pH 

of solution is increased, the concentration of dissolved calcium at thermodynamic equilibrium is 

decreased. Consequently, the precipitation of CaCO3 is favoured at higher pH values. In the Aspen 

model, the increase of pH after both carbonation steps at different bypass rates is almost one pH 

unit from 0 to 15% bypass, resulting in the above described shift in marketable carbonate 

production efficiency. Thus, pH measurement and adjustment can be used for process control 

purposes. 

3.2. Experimental results 

Experimentally obtained conversion rates (Table 6) are remarkably lower than modelling results 

(Table 4). This can be partially explained by experimental inaccuracies, the main reason being the 

lower extraction efficiency from the used steel converter slag. As discussed in [10], this might be 

due to a long storage time in laboratory conditions. However, in general it can be observed that by 

introducing the pH adjustment unit with a 25% bypass stream passing the first carbonation unit, the 

total conversion of calcium from slag to precipitated product can be enhanced. Compared to a case 

with only two reactors, one extractor and one carbonator (Experiment 1), the setup described in 

Figs. 3 and 4 (Experiment 4) was able to increase the conversion efficiency by 13 %-units. By 

changing the ammonium chloride solvent to ammonium nitrate an additional increase of 7 %-units 

was measured (Experiment 5). When carbon dioxide concentration was decreased before the pH 

adjustment unit (Experiment 3), the calcium conversion was 5 %-units lower than in experiment 4. 

Nevertheless, results from experiment 2 show that only by introducing an additional settler unit, 

compared to experiment 1, a similar conversion efficiency as with the pH adjustment system can be 

obtained. From the measured calcium concentrations (Table 7) it can be seen that this is actually 

due to a more efficient extraction conversion, depending on the variation of the input slag 

composition and on the different operation of pumps during the experiments, which resulted in 

longer residence times during extraction. 

Fig. 6 shows the recorded pH values from experiments 3-5, where the pH adjustment unit was used. 

It can be seen that the levels are stabilising quite well to constant values both in extraction and in 

carbonation, especially in experiments 4 and 5. Still, the extraction pH is slightly decreasing with 

time, indicating carbonate precipitation also in this reactor. The stable carbonation pH at 

approximately 6.3 with a calcium concentration of 0.004-0.007 mol/L is quite well corresponding to 

the theoretical values of Fig. 5 (pH 6.3 gives 0.002 mol/L), when the experimental inaccuracy is 

taken into account as well. Also, the pH values obtained from Aspen Plus are systematically 1.3-2 

units too high compared to the experimental results. 

225

Table 6. Experimental conversions of steel slag into PCC 


Reactive calcium feed g/h 5.8 4.5 4.3 3.7 3.7 

PCC product (carbonator) g/h(% of total) 2.5(100) 4.2(75) 1.9(51) 1.4(31) 2.0(35) 

PCC product (pH adjustment) g/h(% of total) - - 0.8(20) 2.3(51) 3.5(60) 

CaCO3 (other system) g/h(% of total) - 1.4(25) 1.1(29) 0.8(18) 0.3(5) 

CaCO3 (total) g/h 2.5 5.6 3.8 4.5 5.8 

Conversion of reactive Ca % 17 49 35 48 62 

Conversion of total Ca % 8 22 16 21 28 

Table 7. Experimental Ca 2+ concentrations after and before extraction reactor 


Ca 2+ after extraction reactor mol/L 0.026 0.041 0.012 0.015 0.018 

Ca 2+ before extraction reactor mol/L 0.007 0.019 0.004 0.007 0.004 

226 

Fig. 6. pH during experiments 3-5 in extraction 

(up left), carbonation (right) and pH adjustment 

(down left) units. 

A remark that can be made based on Fig. 6 is that the pH in the pH adjustment unit has been quite 

unstable during the experiments. Thus, applying a process control device to steer the bypass rate as 

proposed in the previous section would improve the performance of the experimental setup. 

4. Energetic properties of the process 

Based on the presented Aspen model, heat duties of different process steps can be estimated. In 

Table 8 values are shown for the case with a 15% bypass, which was in above discussion found to 

result in highest production rate of pure PCC. It can be observed that all process steps, except for 

the flashing units, are exothermic, i.e. release heat. However, as the process is operated at low

temperature, this heat can in practice not be utilised. Only exception is the cooling unit for flue gas, 

where the gas is cooled from 300°C to 20°C. The maximum amount of reversible work from these 

process steps, when the ambient temperature is assumed to be 20°C, is shown as exergy in Table 8. 

Calculations were made using (1) for heat Q [15]. 

0 

T 

(1) 

Exergy ( Q) 

( 1 

) * Q 

T 

Table 8. Heat duties, inlet temperatures and pressures as well as exergies of heat in different 

process steps according to Aspen Plus model (Fig. 2) with 25 ton/h slag feed and a 15% bypass 

Process unit Q(kW) in(°C) p(bar) Exergy (Q)@T 0 =293.15K(kW) 

EXTRACTO -1638 20 1 0 

CARBONAT -2614 20 1 0 

COOLER -2010 300 1 982 

VAPORSEP 258 20 0.5 - 

SETTLER -62 20 1 0 

VAPORSE2 215 20 0.5 - 

SUM -5851 982 

It must also be noted that the heat duties for process units where gas absorption into process 

solution, or vaporization of solvent liquid is occurring, include also these values, besides of the 

actual heats of chemical reactions. If the results are presented as specific values, the result becomes 

66 W/kg slag for the extraction step. During carbonation, 390 W/kg PCC is generated. This value is 

a sum of all units listed in Table 8, excluding the extractor. 

Two energy-related topics not yet discussed are the energy inputs needed for pumping the process 

liquids and mixing the contents of the reactor units, especially during extraction, where a large 

amount of solids is blended with a liquid stream. From the presented Aspen model the power 

consumption of the two pumps to pressurise the liquid streams after flashing is 2.8 kW for “PUMP” 

and 3.4 kW for “PUMP2”, calculated according to P=p*V , where p is the pressure change (0.5 

bar) and V is the volume flow. Compared to heat duties in reactors, these are quite small numbers. 

In general, also other pumping equipment will be needed for the process, but since the pressure 

losses can be assumed to remain low with small distances and height differences, even for flows of 

200-250 m 3 /h. 

For mixing, the energy dissipation from the stirrer can be calculated with (2) [16], where Po is 

Power number, n is the frequency of rotation for the mixer (1/s), d is the diameter of the mixer (m), 

and V is the total volume (m 3 ). 

3 5 

Po n d 

 

(2) 

V 

Power number can be estimated from Reynolds number of mixing according to (3) [16], where is 

the density of the solution and is the viscosity of slag-ammonium salt solution mixture, calculated 

from (4) [17] with p as the volume fraction of slag (m 3 /m 3 ) and p the viscosity of the dispersed 

material. Also is assumed that the viscosity of ammonium salt solution will not differ noticeably 

from viscosity of pure water. 

2 

nd 

Rev (3) 

 

mixture 

227

p 

p 

1 1. 

5 

p 

H 1 1. 

5 

2O 

p 

mixture 

H 

, using 

2O 

H 2O 

p 

1 

1 

p 

p 

With input values H2O = 1002.7*10 -6 Pa*s, slag=2604 kg/m 3 and solvent =1008 kg/m 3 , (4) yields 

mixture= 1102*10 -6 Pa*s. Together with =1074 kg/m 3 , and with experimental values n=5 1/s and 

d=0.09 m, Rev =39471. In other words, the mixing situation is clearly turbulent. Power number in a 

vessel without baffles for a turbine stirrer is then estimated to 1.25, resulting finally to energy 

dissipation of 0.076 W/kgsolution. In a fully baffled reactor, resulting to better mixing, the power 

number would be approximately 6, yielding to 0.369 W/kgsolution [16]. 

If it is further assumed that the ratio of mixer diameter and vessel volume is maintained constant in 

scale-up, this means that plainly mixing of slag and ammonium salt solvent in extraction requires 

20.8 kW/25 ton slag, or 0.8 kW/ton slag. Related to PCC production, this equals to 1.9 kW/ton 

PCC. If the experimental residence times (Table 3) would be used for the larger scale, this would 

mean 0.8 kW/ton slag * (25/60) h = 0.33 kWh/ton slag. However, these numbers do not include the 

electrical losses of the system (90% efficiency would yield to 23.1 kW/ 25 ton slag), nor they are 

optimised considering the mixer dimensions and reactor size relation. When compared to heat 

duties of the different reactors (Table 8), it can be seen that also the mixing energy has a small, yet 

significant role for the overall energy balance of the process. 

For the carbonation reactor, the energy dissipation in mixing would be lower since the high gas 

flow through the reactor lowers the solution density and viscosity. Also, depending on the gas inlet 

arrangement, the mixing caused by the gas flow itself could be sufficient to obtain the required 

dispersion of solution components. 

5. Studies on additional process units 

5.1. Separation equipment 

As mentioned in Section 2.2, various options exist for continuous separation of solids from the 

aqueous streams. For steel slag separation, a gravitational settling tube (Fig. 7) was tested 

experimentally [18]. The utilised plastic tube with a circular cross-section had a volume of 8.64 L 

(length 110 cm, diameter 10 cm), and it was used with a feed flow of 40 L/h. The underflow was 

maintained at ~7 L/h, a value high enough to guarantee a steady stream for the concentrated particle 

suspension, resulting thus to an overflow of 33 L/h. 

The feasibility of using the settling tube was tested with spherical Ballotini glass beads (p = 2634 

kg/m 3 , d p = 163 µm) in water. Tests were performed with the settling tube positioned horizontally 

(0°) and at an angle of 10°. The beads settled within 0-22 cm from the feed orifice on the horizontal 

tube and approximately within the same range in the tube with 10° angle. The distances where the 

settling occurred were observed visually during the experiments. Because of the large size and high 

density of the glass beads the separation efficiency was ~100%. 

To obtain an estimate with separation efficiencies with the actual steel slag, tests were performed 

with the steel converter slag (p = 2604 kg/m 3 , d p = 96 µm) that was used also as an input material 

for the modelling work described in Section 2. Angles of 30° and 45° were used with water as the 

liquid phase, but 45° was tested also with 1 mol/L NH4Cl. Sedimentation was observed between 0- 

83 cm (0-80 cm for 45°), although the determination was quite challenging due to a very turbid 

suspension. 

After the process had reached a steady-state, a sample was taken from the outflow to determine the 

concentration of particles. The samples were filtrated, dried in an oven at 105 °C and weighed. 

Calculated from (5), the separation efficiencies were 99.9% for both 30° and 45° experiments. 

228 

H 2O 

(4)

c 

 

in 

c 

c 

in 

out 

100% 

However, a SEM picture taken of an overflow sample revealed that the particle size in this fraction 

was approximately 1-2 µm, indicating thus that for the actual slag particles the separation efficiency 

had been 100%, in accordance with the glass particle experiments. Only a fraction of small particles 

released from the slag was leaving the separation unit through the overflow. Thus, according to 

these tests, a gravitational settler could be utilised for steel slag separation in a continuous process, 

but it should be accompanied with a filter to remove also the fraction of dissolved, micrometre scale 

particles. At the same, this filter could also be used to protect the pump needed to transport the 

process liquid from extraction to carbonation. In a larger scale process, the height of the settler tube 

would directly affect the needed pumping effect. Optimisation of the required tube dimensions and 

angle for particles of a known size and density with a known flow is on-going. 

Fig. 7. A 30° inclined settling tube 0 minutes (left) and 2 minutes (right) from the start of the test. 

5.2. Washing steps 

As discussed above, losses of ammonia and water vapour as well as ammonium chloride precipitate, 

mixed with slag residue or the PCC product, should be minimised to increase the profitability of the 

process. During the described experiments it has been measured that residual slag exiting the 

process had moisture content of approximately 10%. Moreover, the chlorine level in dry residue 

was 5.5%-wt according to SEM/EDX measurements. For the modelled process, this would mean 

that 1670 kg/h ammonium chloride (~0.15 kg per kg PCC product) would be lost with the residue, 

assuming that all chlorine is bound with ammonia. 

However, from the experimental results of applying one washing stage for the residue, with 

washing water amount of 0.50 m 3 /t residue, the reduction in calcium content was found to be 0.65 

mol, while approximately 1.5 mol chlorine was dissolved. Thus, it would mean that only 14% of the 

chlorine would have been bound as NH4Cl, the rest being CaCl2. In other words, 230 kg/h 

ammonium chloride and 1485 kg/h calcium chloride would be lost with the slag residue. 

According to experimental results the one-stage wash would lower the chlorine level in the residue 

to 0.7%-wt, corresponding to 28 kg/h NH4Cl and 177 kg/h CaCl2, if calculated with the same 

distribution as above. The used washing liquid of 10 m 3 /h would then contain 0.003 mol/L Cl - , 

0.001 mol/L Ca 2+ and 0.0004 mol/L NH4 + , that could be re-utilised in the process by evaporating 

the water in such extent that the need for make-up water for the process would be satisfied at the 

same. This, however, would bring an extra energy penalty to the process. 

Regarding the produced PCC, washing tests will be done as a part of future work. It can be 

mentioned, that for unwashed PCC produced with 1 mol/L ammonium chloride, the experimental 

chlorine levels have been below 1 %-wt, being still too high for a marketable product. Using 

ammonium nitrate solvent instead of ammonium chloride would remove the problem of chlorine 

precipitates. 

229 

(5)

Calculating from the Aspen Plus model with 15% bypass of the 1 st carbonation unit, the 

vaporization losses of ammonia would be 18 kg/h (1.1 kmol/h) with PCC production of 10795 kg/h. 

This amount, 0.4% of the total ammonium species present, is of the same order of magnitude what 

has been obtained experimentally by FTIR measurements [13]. However, also this ammonia could 

be captured, e.g. by using an HCl-based scrubber for the flue gas stream leaving the system (Fig. 2). 

The need for make-up water to replace the moisture lost with streams of solid matter and water 

vapour flashed from the separation units depends on many variables, and has not yet been 

quantitavely studied. However, assuming a 20% moisture content for the PCC product and 10% for 

slag residue, the need for make-up water for the above discussed process would be 5830 kg/h. This 

amount could easily be taken from washing units, but it requires further optimisation. In general, all 

the values presented in this section are based on preliminary studies, and thus need to be verified by 

additional experimental studies and analysis as a central part of future work. Only then can the 

usage of fresh make-up chemicals be optimised too. 


The studied pH swing process possesses potential for commercial application. Based on both 

modelling and experimental work, precipitated calcium carbonate can be produced with a 

continuously operating system instead of batch reactors. By choosing the solvent with highest 

conversion efficiency (ammonium nitrate) and by dividing the precipitation step in two stages, the 

conversion of calcium in the input material to a valuable product can be enhanced. Energetically the 

process will require some mixing and pumping power, but the chemical reaction steps are 

exothermic. The recovery and separation units for process chemicals and for solids need to be 

studied in more detail in future work, but it seems to be possible to use a gravitational settler to 

separate slag residues from the process liquid. Also, washing the output streams will aid the 

recovery of chemicals, and thus decrease the need for make-up material. However, additional 

studies on particle quality, shape, purity and whiteness of the produced PCC must be performed to 

guarantee that a marketable product is manufactured. 


The authors want to acknowledge Cleen Ltd. and Tekes (the Finnish Funding Agency for 

Technology and Innovation) for their financial support for the research via the Cleen CCSP project 

(2011-2015). H.-P. Mattila also acknowledges the Graduate School in Chemical Engineering for 

support for his work. 

Nomenclature 

c concentration, g/L 

d diameter of the mixer, m 

d mean diameter, µm 

n frequency of rotation, 1/s 

P power, kW 

p pressure, bar 

Po power number, – 

Q heat duty, kW 

Rev Reynolds number of mixing 

T temperature, K 

T 0 ambient temperature, K 

V volume, m 3 

V volume flow, m 3 /h 

230

Abbreviations 

CCS Carbon Capture and Storage 

FTIR Fourier Transform Infrared Spectroscopy 

PCC Precipitated Calcium Carbonate 

SEM/EDX Scanning Electron Microscopy – Energy Dispersive X-ray 

Greek symbols 

difference 

energy dissipation, W/kgsolution 

viscosity, Pas; efficiency, % 

temperature, °C 

density, kg/m 3 

Subscripts and superscripts 

in at inlet 

p particle 

References 

[1] IPCC, IPCC Special Report on Carbon Dioxide Capture and Storage, Cambridge University 

Press, New York, 2005. 

[2] Herzog, H., An Introduction to CO2 Separation and Capture Technologies, MIT Energy 

Laboratory, 1999. Available at: 

[accessed 27.1.2012]. 

[3] Teir, S., Fixation of carbon dioxide by producing carbonates from minerals and steelmaking 

slags. [Dissertation]. Espoo, Finland: Helsinki University of Technology; 2008. 

[4] Baciocchi, R., Costa, G., Polettini, A., Pomi, R., Prigiobbe, V., Comparison of different 

reaction routes for carbonation of APC residues, Energy Procedia 2009;1: 4851-4858. 

[5] Velts, O., Uibu, M., Kallas, J., Kuusik, R., CO2 mineral trapping: Modeling of calcium 

carbonate precipitation in a semi-batch reactor. Energy Procedia 2011;4: 771-778. 

[6] Eloneva, S., Reduction of CO2 Emissions by Mineral Carbonation: Steelmaking Slags as 

Raw Material with a Pure Calcium Carbonate End Product. [Dissertation]. Espoo, Finland: 

Aalto University, School of Science and Technology; 2010. 

[7] Kodama, S., Nishimoto, T., Yamamoto, N., Yogo, K., Yamada, K., Development of a 

new pH-swing CO2 mineralization process with a recyclable reaction solution. Energy 

2008;33: 776-784. 

[8] Eloneva, S., Teir, S., Savolahti, J., Fogelholm, C.-J., Zevenhoven, R., Co-utilization of CO2 

and calcium silicate-rich slags for precipitated calcium carbonate production (Part II). In: 

Mirandola, A., Arnas, O., Lazzaretto, A., editors. ECOS 2007: Proceedings of the 20 th 

International Conference on Efficiency, Cost, Optimization, Simulation and Environmental 

Impact of Energy Systems; 2007 Jun 25-28; Padova, Italy. Volume II:1389-1396. 

[9] Said, A., Eloneva, S., Fogelholm, C.-J., Mattila, H.-P., Zevenhoven, R., Process simulation 

of utilization CO2 and steelmaking slags to form Precipitated Calcium Carbonate (PCC). In: 

ECOS 2009: Proceedings of the 22 nd International Conference on Efficiency, Cost, 

Optimization, Simulation and Environmental Impact of Energy Systems; 2009 Aug 31- Sep 3; 

Foz du Iguacu, Brazil. pp. 1261-1270. 

[10] Mattila, H.-P., Grigalinait, I., Zevenhoven, R., Chemical kinetics modeling and process 

parameter sensitivity for precipitated calcium carbonate production from steelmaking slags. 

Chem. Eng. J. 2012;192: 77-89. 

231

[11] Mattila, H.-P., Experimental studies and process modeling of aqueous two-stage steel slag 

carbonation. M.Sc. Thesis. Turku, Finland: Åbo Akademi University; 2009. 

[12] Mattila, H.-P., Grigalinait, I., Said, A., Fogelholm, C.-J., Zevenhoven, R., Production of 

papermaking grade calcium carbonate from steelmaking slag – product quality and 

development of a larger scale process. Accepted to be presented at SCANMET IV conference, 

Luleå, Sweden, Jun 2012. 

[13] Eloneva, S., Mannisto, P., Said, A., Fogelholm, C.-J., Zevenhoven, R., Ammonium saltbased 

steelmaking slag carbonation: Precipitation of CaCO3 and ammonia losses assessment. 

Greenhouse Gases: Science and Technology 2011;1(4): 305-311. 

[14] Wiklund, A., Sipilä, J., Eloneva, S., Zevenhoven, R., Assessment of Shell work on 

carbonation of calcium based materials. Åbo Akademi University, Heat Engineering 

Laboratory; 2008. Report. 

[15] Szargut, J., Exergy Method: Technical and Ecological Applications. Southampton, UK: 

WIT Press; 2005. 

[16] Beek, W. J., Muttzall, K. M. K., van Heuven, J. W., Transport Phenomena, 2 nd edition. 

Guildford, UK: Biddles Ltd; 2000. 

[17] Kunitz, M., An empirical formula for the relation between viscosity of solution and 

volume of solute. J.Gen.Physiol. 1926, 9:715-25. 

[18] Davis, R. H., Gecol, H., Classification of concentrated suspensions using inclined 

settlers. International Journal of Multiphase Flow 1996; 22(3): 563-574. 

232

Abstract: 



June 26-29, 2012, Perugia, Italy 

Production of Mg(OH)2 for CO2 emissions 

removal applications: parametric and process 

evaluation 

Experience Nduagu a* , Inês Romão a,b , Ron Zevenhoven a 

a Åbo Akademi University, Åbo/Turku, Finland, enduagu@abo.fi. CA 

b University of Coimbra, Coimbra, Portugal 

Technological processes that accelerate natural and geochemical weathering of abundantly available Mgsilicate 

minerals have the potential for large-scale, safe and permanent storage of CO2. One of these CO2 

sequestration routes involves as a first step the production of reactive Mg(OH)2 from Mg-silicates using 

recoverable ammonium sulfate (AS) salt. This route avoids the very slow kinetics of carbonating magnesium 

silicates. A recently identified Mg(OH)2 production process involves a closed loop, staged process of Mg 

extraction followed by Mg(OH)2 precipitation and reagent (AS) recovery. This process has been applied to 

different Mg-silicate (serpentinite and olivine rocks in particular) minerals from worldwide locations, having 

varying physical and chemical properties. Experimental results showed some dependence of Mg extraction 

and mass of the Mg(OH)2 product on the reaction parameters: mass ratio of Mg-silicate mineral (S) to AS 

salt reacted, reaction temperature (T) and time (t). This paper statistically evaluates the contribution of these 

effects and their interactions using a 2n-1 factorial experimental design. Both Mg(OH)2 production and 

carbonation were simulated using Aspen Plus® software while process heat integration was done by pinch 

analysis. Process energy evaluation, on an exergy basis, gives 3.88 GJ of energy requirement for 1t-CO2 

sequestered (for Finnish serpentinite). This value is ~ 0.5 GJ/t-CO2 (10 % points) less than the energy 

requirement of the process in a previous model. The results of this analysis would be beneficial for 

optimization and pilot scale studies of this process. 

Keywords: 

Mg-silicates, Magnesium hydroxide, CO2 mineralization, Process evaluation. 


Weathering of alkaline silicate rocks plays a significant role in absorbing atmospheric CO2 [1]. 

Alkaline and alkaline-earth silicate mineral deposits are abundant and larger than fossil 

resources[2]. A resource of this magnitude, over 300,000 Gt of Mg-based silicate minerals[3] 

provides significant amounts of base ions for the natural process of neutralizing atmospheric CO2 

emissions. However, natural weathering has very slow kinetics and occurs on geological 

(multimillion-year) timescales [4]. So, it becomes foolhardy to rely on natural weathering in 

reducing or stabilizing atmospheric CO2 emissions. The goal of meeting both current and future 

energy demands in a “carbon neutral” manner has therefore spurred research that aims at 

accelerating the kinetics of the reaction of mineral silicates and CO2. This geochemical option of 

carbon (dioxide) capture and storage is known as CO2 mineralization or mineral carbonation. 

The direct carbonation chemistry of Mg silicates is exothermic, and potentially allows for a process 

with a zero or negative overall energy input [5]. Mg silicates, for example, serpentine and olivine 

which are abundantly available (with a combined capacity of ~ 200,000 Gt[3]) reacts with CO2 

according to (1) and (2)[6]. 

233

Mg2SiO4(s) + 2CO2(g) →2MgCO3(s) + SiO2(s), 

∆H (298 K)= - 69...-109 kJ/mol CO2 (1) 

Mg3Si2O5(OH)4 + 3CO2 (g) →3MgCO3(s) + 2SiO2(s) + 2H2O(l), 

∆H (298 K) = -46… -64 kJ/mol CO2 (2) 

Direct gas/solid carbonation of Mg-silicates appears simple but suffers from slow chemical kinetics 

and poor energy economy even at elevated temperatures and pressures. Surprisingly, most of the 

routes presented in the literature do not take benefit of the exothermic nature of the overall mineral 

carbonation chemistry. A staged process of CO2 mineralization via production of Mg(OH)2 

followed by gas/solid carbonation is the major focus of the mineralization research at Åbo Akademi 

University (hereafter ÅA), Finland. This route allows for a good process heat integration utilizing 

the exothermic heat produced from Mg(OH)2 carbonation to drive the upstream Mg(OH)2 process. 

Mg(OH)2 produced in the first step can be used to capture and store CO2 via the following ways: 

i. carbonation using a high temperature pressurized fluidized bed (FB) reactor (480-600 o C,

an XRD analysis showing that a combination of FeO and Fe2O3 compounds (which of course could 

be summed up to be Fe3O4) is present in the same rock sample. 

2.2. Method for producing Mg(OH)2 from Mg-silicate minerals 

This section describes the method for producing Mg(OH)2 from Mg-silicate minerals, a procedure 

that has been previously reported in literature[6, 15, 16]. The process route of producing Mg(OH)2 

involves a staged, closed loop process of Mg extraction using recoverable AS salt. The process 

schematic is presented in Fig.1. 

Figure 1. Schematic of process route for the production of Mg(OH)2 from Mg-silicate minerals. 

After Nduagu et al.[17]. 

Mg is extracted from the reaction of Mg-silicate rocks and AS salt at 270-550 o C (depending on the 

rock type) in an oven/reactor. Information on the ranges of reaction parameters tested is presented 

in Table 1. The reaction in the oven produces MgSO4, SiO2, water vapor and recoverable gaseous 

NH3. For details of the reactions and thermodynamics of the (Mg/Fe/Ca) extraction, refer to the 

Appendix section (Table A2 and Fig. A1). Mg/Fe/Ca sulfates obtained from the extraction reaction 

are leached in water at room temperature and pressure conditions. The elemental amounts of Mg, Fe 

and Ca and other metals extracted were determined by ICP-OES analysis. 

Increasing the pH of (Mg/Fe/Ca) sulfates-rich solution (using the recovered ammonia) results in the 

precipitation of hydroxides or oxy-hydroxides. Of major interest are Fe, which is precipitated as 

goethite (FeOOH) and Mg, precipitated as Mg(OH)2[16]. At pH ranges of 8–9 and 11–12 Fe and 

Mg respectively precipitate out of the solution. FeOOH by-product could be a useful raw material in 

the iron- and steelmaking industry. Due to the high concentration of iron compounds in these 

minerals (in different oxide forms), iron oxide by-product stream may be a useful raw material for 

the iron- and steel-making industry[17-19]. In Finland, for instance, the iron and steel sector is the 

largest point-source CO2 producer. Thus, integrating steel industry’s CO2 emissions with 

mineralization is crucial and would result in emissions reduction, and in the replacement of raw 

materials (iron ore) using the iron oxide by-products. However, we have shown earlier[17] that the 

processing of Fe together with Mg in a CO2 sequestration process comes with a significant energy 

penalty. The results showed that the contribution of iron to the energy requirement of CO2 

sequestration increases by 70%, 30% and 16% points for rocks containing Fe as Fe3O4, Fe2O3 and 

FeO compounds respectively as compared to an iron-free rock. 

After filtering precipitated Fe/Mg/Ca (oxy)hydroxides from the solution, AS salt is then recovered 

via crystallization. The following crystallization techniques may suffice: evaporative, mechanical 

vapor recompression (MVR) or anti-solvents (especially alcohols)[17]. This study focuses on MVR. 

The Mg(OH)2 thus produced from Mg-silicate mineral rocks is then used to sequester CO2 in the 

form of thermodynamically stable magnesium carbonates. 

3. Parametric evaluation by 2 n-1 experimental design 

This section studied the extent to which the reaction parameters affect Mg extraction, and in 

extension their effects on Mg(OH)2 production. These parameters include elemental Mg to Fe ratio 

235

(Mg/Fe) of the mineral rock, Mg-silicate to AS mass ratio (S/AS), reaction temperature (T), time (t), 

and the interaction of these effects. 

It is important to point out the nature of the test data (statistical details are presented in Table 1). 

The initial batch of tests were done using mostly Finnish serpentinite between 2008 and 2009, and 

were reported in [15] and [16]. The aim at that time was to prove that Mg(OH)2 can be produced 

from Finnish serpentinite, and efficiently too. After this, efforts were channeled towards applying 

the method to different Mg-silicate rocks from worldwide locations[6, 20-22]. 

Table 1. Statistical overview of the parameter values tested in 82 experiments. 

Parameters Minimum Maximum Median Average Standard Deviation 

Mg/Fe (kg/kg) 0.31 5.90 2.16 2.81 1.32 

S/AS (kg/kg) 0.40 4.0 0.67 0.85 0.6 

T ( o C) 270 550 475 457 63 

t (min) 10 120 22 32 27 

Clearly, earlier tests did not focus on identifying reaction trends as experiments were performed at 

varying reaction conditions chosen almost at random - targeting to cover a broad range of each 

parameter. However, after testing a range of values of each of the factors, it now becomes necessary 

to identify which parameters have the most significant effects on Mg extraction and Mg(OH)2 

production. More so, parameter cross-correlation effects would be determined as well. A better 

understanding of these effects and their interactions is essential for optimization of Mg(OH)2 

production from Mg-silicate minerals for the purposes of fixing CO2 as carbonate(s). 

Due to the range of values parameters considered (Table 1) the choice of a reasonable reference 

point was important in order to design a two-level fractional factorial design. We used a reference 

level “0” condition to classify each factor according to levels: high (+1) or low (-1) (in Table 2). 

The first “0” level was chosen to reflect the median value of the parameters while a second “0” 

level was chosen at values of the parameters at near optimal conditions. The response parameter 

(dependent variable) in this analysis is % Mg extraction (% Mg ext). The % Mg extraction is the 

amount of Mg (grams) extracted from the Mg-silicate rock divided by the total amount of Mg 

(grams) present in the Mg-silicate, expressed as percentage. The motivation for focusing on the 

parametric analysis of Mg extraction is the fact that the amount of Mg(OH)2 produced from the 

process strongly correlates with values for Mg extraction[16]. 

Table 2. Reference level “0” conditions for evaluation of factors and their interactions. 

Parameters 

Levels 

Mg/Fe (kg/kg) 

A 

high 

(+1) 

low 

(-1) 

S/AS (kg/kg) 

B 

high 

(+1) 

236 

low 

(-1) 

T ( o C) 

C 

high 

(+1) 

low 

(-1) 

t (min) 

D 

high 

(+1) 

Condition Iᵝ > 2.16 ≤ 2.16 ≤ 1 > 1 ≥ 480 < 480 > 25 ≤ 25 

Condition IIᵝ > 2.16 ≤ 2.16 ≤ 0.67 > 0.67 ≥ 440 < 440 > 60 ≤ 60 

ᵝ Condition I reflects the median of the data. ᵝ Condition II is chosen at near optimal 

experimental conditions. “+1” and “-1” are the high and low levels respectively. 

3.1 Fractional factorial design 

Fractional factorial design (2 n-1 , where n represents the number of parameters) enables the analysis 

of only a subset of treatment combinations, while still obtaining a meaningful result that is 

statistically representative of the entire data set. In this analysis n=4 (A, B, C and D in Table 3) and 

the objective function is Y which represents % Mg extraction. The fractional factorial design is 

constructed by partitioning the runs into two blocks; one block, which is a contrast of the other, is 

completely sacrificed [23]. Instead of using a full 2 n design with 16 design points, the 2 n-1 design 

low 

(-1)

with only 8 design points was chosen at points ABCD=I (1, ab, ac, ad, bc, bd, ad and abcd). Design 

points having ABCD=-I (a, b, c, d, abc, abd and bcd) which are considered as complementary to the 

points with ABCD=I were excluded in the 2 n-1 factorial design (as illustrated in Table 3). At this 

stage, the third and fourth order interaction effects of the parameters (ABC, ABD, ACD, BCD and 

ABCD) were also neglected in order to avoid ambiguity. 

Table 3. 2 4-1 factorial design. 

Effects and interactions 

Treatment Mg/Fe S/AS T t AB AC AD ABCD Observation 

(A) (B) (C) (D=BCD) (=CD) (=BD) (=BC) (=I) Y 

1 − − − − + + + + --ab 

+ + − − + − − + --ac 

+ − + − − + − + --ad 

+ − − + − − + + --bc 

− + + − − − + + --bd 

− + − + − + − + --cd 

− − + + + − − + --abcd 

+ + + + + + + + --- 

The estimated effect (see (3)) of each design factor is represented mathematically as the average at 

the high level (+) of the factor minus the average at the low level (-) of the factor. 

Effect=Contrast/(n ’ 2 n-1 ) (3) 

Where n and n ’ are the number of factors and replicates respectively, and Contrast is the sum of the 

values of each factor at its high level minus the sum of the values of the same factor at its low level. 

The significance of any parameter or the interaction of parameters was determined at 95 % (α=0.05) 

confidence level. This value is determined by using a student t-test to obtain t-values and assessing 

that with the probability (p value) associated with the test statistic. MINITAB ® statistical software 

[24] was used to analyze the data from experimental tests using the 2 n-1 (and a 2-level) factorial 

design. 

3.2. Mg extraction: parametric effects and interactions 

3.2.1 Effect of Mg/Fe ratio of rock types 

Thirteen different Mg-silicate minerals (nine serpentinite and four olivine rocks) were studied in a 

total of eighty-four tests performed at varying reaction conditions. The results showed a huge 

difference in reactivity of serpentinites and olivines using the method applied in this study. Based 

on maximum extraction values obtained so far for each rock type, serpentinite is about 5x as 

reactive as olivine (see Fig.2). This confirms previous results for two olivine-containing rocks (from 

Åheim, Norway and Satakunta, Finland) samples tested and found not to be suitable for Mg 

extraction[6]. It was observed that the olivine rocks tested had a harder texture, smaller internal 

Brunauer, Emmett and Teller (BET) surface area as well as pore volume. The range of % Mg 

extraction between the maximum (Max.) and minimum (Min.) in Fig.2 is due to results obtained at 

wide range of reaction conditions. 

At varying reaction conditions the factors and interactions that have a significant effect on the 

extent of Mg extraction were obtained (see Table 4). While keeping constant some parameters and 

varying others, the effects of changing levels of each parameter was determined. This sensitivity 

analysis was performed in order to determine if the parameters are important or not. If any 

parameter was found to contribute significantly to Mg extraction, it was important to determine the 

levels to which that factor is significant. 

237

Figure 2. Effect of Mg/Fe ratio of the Mg-silicate rocks on % Mg extraction. The figure on the left 

shows results from both serpentinite and olivine rocks while the one on the right is for only 

serpentinite rocks. 

Table 4. Sensitivity analysis for the factors affecting Mg extraction by varying the conditions. 

# Selected conditions 

1 

2 

3 

4 

5 

6 

7 

8 

A>2.16 g/g, B≤1 g/g, 

C>480 o C, D>60 min 

A>2.16 g/g, B≤1 g/g, 

C>440 o C, D>60 min 

A>2.16 g/g, B≤0.67 g/g, 

C>480 o C, D>60 min 

A>2.16 g/g, B≤0.67 g/g, 

C>440 o C, D>60 min 

A>2.16 g/g, B≤1 g/g, 

C>480 o C, D>25 min 

A>2.16 g/g, B≤1 g/g, 

C>440 o C, D>25 min 

A>2.16 g/g, B≤0.67 g/g, 

C>480 o C, D>25 min 

A>2.16 g/g, B≤0.67 g/g, 

C>440 o C, D>25 min 

Factors Interactions 

Mg/Fe (g/g) 

A 

S/AS (g/g) 

B 

238 

T ( o C) 

C 

T (min) 

D 

AB 

AC 

AD 

R 2 

√ 27% 

√ 31% 

22% 

25% 

√ √ β √ 47% 

√ √ √ 52% 

√ √ √ 45% 

√ √ √ 50% 

√ and √ β represent positive and negative effect of the factors/interactions respectively. R 2 is the 

regression coefficient obtained for each condition. 

It is obvious from Fig. 2 that serpentinite rocks with a Mg/Fe ratio ≥ 2.16 show an exceptionally 

(>2x) higher % Mg extraction than others. This was the reason why Mg/Fe ratio level was 

benchmarked at > 2.16 (Table 4) in the sensitivity analysis. Given the information deductible from 

Fig. 2, it was more interesting to understand the effects and interaction effects of the more reactive 

minerals with Mg/Fe ratio > 2.16. 

Our goal is to obtain a process condition that allows us to design a Mg extraction reactor that 

operates at optimal reaction conditions. The reactor should be able to process different types of 

serpentinite minerals with varying Mg/Fe ratios, minimal amounts of AS salt reagent (slightly less 

than 1 g/g), temperatures < 440 o C and reaction time ≤ 60 min. The combination of parameters in 

Table 4 which mostly suits this goal is condition 6 which also has the highest regression coefficient 

(R 2 =52%). It is arguable that the R 2 value obtained is not sufficient enough to describe a process; 

however, it is not surprising that a system as complicated as the one simulated here would give a

comparatively low R 2 . The results reported here contain tests performed on the reaction of solids 

(solid state reaction) with multivariate parameters. Solid state reactions are less predictable than 

those involving other states/phases. We assume that not all the factors contributing to Mg extraction 

have been identified and studied. Some other factors like particle size difference between the 

reacting Mg-silicate mineral and AS salt, heat and mass transfer, geometry and size of the reactants 

and their containers may affect solid/solid reactions. These are the main subjects of ongoing 

investigation as we embark on the next stage - pilot scale development. 

3.2.2 Effect of amount of reagents (S/AS ratio) 

By varying S/AS ratio of the tests between ≤0.67 g/g and ≤1 g/g, its effect on the extent of Mg 

extraction was evaluated. For conditions 5-8 (Table 4), at 95 % (α = 0.05) significance level, S/AS 

ratio has a significant positive effect on Mg extraction. The results obtained show that an increase in 

S/AS salt ratio above both the level of 0.67 g/g or 1 g/g does not significantly affect Mg extraction 

beyond a reaction time of 60 min (see conditions 1 to 4 in Table). In other words, a change in the 

amount of AS salt reagent levels is more important when the reaction time is less than 60 min. 

Increasing S/AS salt from low (-1) to high (+1) levels results in a 10% point increase in Mg 

extraction. The effects of S/AS ratio, those of the parameters and their interactions can be visualized 

from Fig.3 which is plotted for condition 6. 

Figure 3.Main effects and interaction for Mg extraction under Condition I 

3.2.3 Effect of reaction temperature and time 

The effect of reaction temperature is not important under most of the conditions evaluated, but 

shows negative dependence on Mg extraction above 480 o C (condition 4 in Table 4). 

Figure 4. Effect of temperature (left) and time (right) on Mg extraction 

239

Figure 4 shows that an increase in temperature results in a reduction in % Mg extraction is already 

at 440 o C (i.e. within the 401-450 o C temperature range). For the selected reaction condition 6, the 

effect due to temperature is almost flat (see Fig.3). However, reaction time has a significant effect 

on magnesium extraction; an increase in reaction time from low (-1) to high (+1) levels leads to a 

15 percent points’ increase in magnesium extraction. More so, reaction time significantly affects 

Mg extraction at all the conditions modeled except when t > 60 min and S/AS ≤0.67 g/g (conditions 

3 and 4). Besides, this effect of reaction time on Mg ext seems not straightforward from Fig.4; more 

investigation is needed. 

3.2.4 Interaction effects 

Under the conditions modeled, the interaction effects of Mg/Fe-S/AS ratios and T-t are significant at 

95 % significance level. The interaction effects presented in Fig.3 show that increasing the reaction 

time from high (+1) to low (-1) (above 25 min) levels significantly increases (by 30 % point) the 

value for Mg extraction if the reaction temperature are kept below 480 o C. Above this temperature, 

no increase in Mg extraction is possible, presumably due to thermal decomposition of AS above at 

high temperatures leading to the formation of sulfur trioxide gas, which could alter the 

thermodynamics [16]. On the other hand, increasing S/AS ratio levels from high (+1) to low (-1) (≤1 

g/g) at both high (+1) and low (-1) levels Mg/Fe leads to a significant increase in % Mg ext. But, the 

% Mg ext values obtained with high (+1) level of Mg/Fe (>2.16 g/g) are higher. This confirms 

previous results which showed that rocks with high Mg/Fe ratios respond better to Mg extraction 

than those with low Mg/Fe ratios[6]. 

4. Process evaluation using exergy and pinch analysis 

4.1 Process simulation 

The Mg(OH)2 production, AS recovery and Mg(OH)2 carbonation were modeled using Aspen 

Plus® software. The process flow diagram is presented in Fig.5. Pinch analysis was done using 

Aspen Energy Analyzer®. 

4.1.1 Mg, Fe and Ca extraction 

The base property method used for this simulation is the ELECTRTL method. The solid state 

reaction of serpentinite and AS salt was simulated using a stoichiometric reactor (REACTOR) with 

the extraction equations and thermodynamics specified as presented in the Appendix section ((R1), 

(R3) and (R5) in Table A2). The serpentinite feed has its composition simulated after the Finnish 

serpentinite which contains ~83 %-wt Mg3Si2O5(OH)4, ~14 %-wt Fe2O3 and ~1 %-wt CaSiO3. The 

AS feed (AS-1) is a product from the MVR section, where AS salt is crystallized. The specified 

conversion of this reactor is 100% – meaning that serpentine and AS feed react completely to form 

products. This assumption is based on the best case scenario of the extraction reaction which is the 

aim of an ongoing optimization study. However, all the scenarios have previously been explored 

using life cycle analysis (LCA)[25]. 

4.1.2 Dissolution of extraction products 

The product stream from the reactor (PRDTS) was separated in a solid/gas separator (SEP-1) into a 

solid stream (SOLIDS-1) and a gas stream (GASES) before cooling. The dissolution of the solid 

products was modeled using a stoichiometric reactor (CONVTR) and an RGibbs reactor 

(DISSOLUT) respectively. At the CONVTR the solid compounds were converted to aqueous 

compounds before dissociating into anions and cations at the DISSOLUT. The DISSOLUT 

simulated the dissolution reactions of MgSO4, FeSO4, Fe2(SO4)3 and CaSO4 in water streams at 40 

°C by calculating both the phase and chemical equilibrium based on Gibbs free energy 

minimization. The water stream (DISS-H2O) used for dissolution is made up of the following: a 

recycled water stream (MVR-H2O) from the MVR section, a water stream (WATER) recovered from 

the separation of the GASES stream into H2O and NH3 gas. After dissolution, the mixture is 

240

separated by filtration into a solid stream (SIO2), containing mainly silica and a liquid stream (DIS- 

PRDT) of mainly Fe- and Mg-sulfate compounds. 

4.1.3 Precipitation of FeOOH and Mg(OH)2 

The stoichiometric reactors, PREP-1 and PREP-2 were used for precipitation of FeOOH and 

Mg(OH)2 respectively, and the following reactions and thermodynamics (4) - (5) specified: 

Table 5. Chemical reactions and thermodynamics of the precipitation stage 

# Precipitation reactions ∆Hr (T=313K) 

4 Fe2(SO4)3(s)+6NH3(g)+4H2O(l)→2FeOOH(s)+3(NH4)2SO4(aq) -720 kJ/mol Fe 

5 MgSO4(s)+2NH3(g)+2H2O(l) →Mg(OH)2(s)+(NH4)2SO4(aq) -85 kJ/mol Mg 

The pH of Fe- and Mg-rich solution stream (DIS-PRDT) was increased (using the recovered NH3 

gas from the flash separator SEP-4) in stages of 8–9 and 11–12 to precipitate hydroxides of iron and 

magnesium respectively. Ca(OH)2 precipitates together with Fe in the first precipitation stage. 

Aqueous AS is formed at both precipitation stages (see (4) and (5)). Products of the precipitation 

stages, FeOOH and Mg(OH)2 were separated by filtration while aqueous AS passes through a 

converter (CONVTR-2) to the MVR section for crystallization before it is recycled. The role of the 

CONVTR-2 was to combine anions and cations in stoichiometric amounts into aqueous compounds. 

The application of the MVR crystallization method to this process has been reported earlier [17, 26]; 

however, this paper revisits the MVR crystallization application in the pinch analysis section 

(section 4.2). 

4.1.4 Mg(OH)2 carbonation 

The reaction of CO2 and Mg(OH)2 is exothermic, and at suitable conditions forms 

thermodynamically stable MgCO3 and superheated steam. Sequestration of CO2 using the gas/solid 

route as being developed at ÅA [7, 8, 27] provides utilizable energy at high temperatures (480- 550 

o C, ∆H ~ -59.5 kJ/mol Mg) and pressure conditions. Pressures can vary from 20 bars to 80 bars 

depending on the concentration of CO2 – pure and concentrated or in flue gas stream [9, 10] . For 

simulation purposes, it was assumed that the sequestration plant stores 1 ton/h CO2 (~ 8000 t/y). 

As shown in Figs. 5 and 6, at high carbonation conversion (> 90%) the exothermic heat of 

carbonation makes the temperature of outlet stream of the reactor hotter than those of the inlet 

streams (according to (6)). This energy is at the same time sufficient enough to heat up the reactants 

(Mg(OH)2 and CO2) and as well provide energy to the process (heat or power depending on what it 

is designed to achieve). The carbonation section in Fig. 5 produces both heat and power while that 

of Fig.6 produces only heat. 

241

Figure 5. Process flow diagram of Mg(OH)2 production and carbonation simulated using ASPEN PLUS software. 

242

Figure 6. Mg(OH)2 carbonation flow sheet producing utilizable heat. 

, (T2> T1x, T1y) (6) 

where z - MgCO3, s - H2O, x - Mg(OH)2, y - CO2, - heat of formation, ṅ - molar amount of 

compound, Cp - specific heat capacity, T1- inlet temperature and T2 - outlet temperature. In this 

case, the molar amounts associated with each term on the left side of (6) cancel out (since they are 

equal). If the inlet temperatures of the reactants are same (i.e. T1x=T1y=T1), (6) reduces to (7). 

, (T2> T1) (7) 

It was assumed that CO2 was delivered at 20 o C, 20 bars from stream CO2-IN. The CO2 pressure of 

this stream looks optimistic; however, this value was based upon the fact that the CO2 capture and 

purification unit would be located nearby the CO2 sequestration site. This in essence provides 

compressor power savings that are required for CO2 compression to pipeline transport pressures of 

~ 150 bars. The Mg(OH)2 product separated by filtration in SEP-3 was dried by heating to 150 o C 

before entering the carbonation reactor (CARBONAT). The CO2 stream (CO2-IN) was heated to 520 

o C before entering the CARBONAT. The exothermic nature of carbonation reaction led to a higher 

temperature of the products than that of the reactants. This was beneficial since power and heat 

were intended to be extracted from the steam and MgCO3 products using a turbine (TURBINE) and 

heat exchangers respectively. More importantly, given the resulting temperature difference, the 

outlet streams of the reactor can then be used to heat up the inlet streams. 

4.2 Pinch analysis 

Pinch analysis has become a useful energy targeting and design tool for thermal and chemical 

processes and utilities[28]. This method enables the plotting of composite and grand composite 

curves using temperature versus enthalpy axes[29]. These curves provide an insight on the process 

heat availability and requirements. 

These basic rules were followed while applying the pinch analysis[28]: 

1. Separate the system into two independent sections – above and below the pinch, and do not 

transfer heat across the pinch. 

243

2. Only cold utility is needed below the pinch. Heating of streams at the section below the 

pinch incurs a heat penalty. 

3. Only hot utility is needed above the pinch. Cooling of streams at the section above the pinch 

incurs an energy penalty. 

Table 6. Heating and cooling requirements of the process implemented using Pinch Analysis 

Inlet T Outlet T Flow rate Enthalpy mCP 

Streams 

Cold Streams 

o 

C 

o 

C kg/h MJ/h 

o 

MJ/ C-h 

SERP-1 25 400 2529 909 2.4 

AS-AQ1 

Stream 1 

Stream 2 

40 

107 

107 

115 

10926 

10926 

8413 

9381 

125.6 

1173 

CO2-IN 20 520 1000 530 1.06 

MG 

Hot streams 

40 150 1325 220 2 

SOLIDS-1 400 40 4703 1603 4.45 

GASES 400 25 1863 3559 9.49 

MGCO3 532 40 1916 1160 2.36 

STEAM - UT 246 40 520 1146 5.56 

Figure 7. Working cycle (B) and Aspen process flow sheets (A&B) of mechanical vapor 

recompression (MVR) crystallization of AS salt. Modified after Nduagu et. al[17]. 

More rules were applied during designing of an optimal heat exchanger network as implemented 

using the Aspen Plus model in Fig.5. These include: 

1. For pinch matches, above the pinch the CPcold ≥ CPHot while below the pinch CPHot ≥ CPcold. CP 

is a value calculated by dividing the enthalpy of a stream by the difference in temperatures of the 

outlet and the inlet streams (see Table 6). 

244

2. ∆T of 10 o C was the set minimum temperature difference. 

3. Solid streams were not matched with solid stream as solid/solid heat exchange may be 

problematic. 

A B 

Fig 8. Hot and cold composite curves of the process shown in red and blue colors respectively 

Two scenarios, with and without the MVR crystallization were compared. In the case of without 

MRV, the AS-water stream (AS-AQ1 in Fig. 5, Fig.7A) was heated step-wisely, 40 o C 76 o C 

115 o C. At 115 o C, all the water in the stream was evaporated with virtually no heat recovery. In 

Fig. 7A the MVR was simulated with two crystallization vessels, allowing for an operation in two 

different temperature regimes (107 and 115 o C) while Fig.7C used only one crystallizer. However, 

compressing the water vapor stream from 1 to 2 bars (points 12 in Fig. 7) increases the enthalpy 

as well as the temperature of the stream to a level it can transfer heat to saturated water at 100 o C. 

The temperature-enthalpy plot (composite curve) of the process was first plotted (in Fig. 8A), 

assuming that a complete evaporative crystallization of the AS salt was carried out. The upper and 

lower pinch points are 40 o C and 50 o C. In this case, the latent heat added to evaporate water from 

AS-water mixture would be lost as low grade heat. Most part of that heat is represented in Fig.8A as 

QLv (~ 9.4 GJ/h). QH is the value of other low temperature heating (sensible heat) required. The 

thick black arrow in the Fig. 8A pointing towards the cold composite curve (in blue color) gives an 

insight to the temperature and enthalpy values of the hot utility required. The gap between the hot 

and cold steams needs to be closed in order to optimize heat recovery. In order to achieve this, the 

temperature and the enthalpy of the hot stream must be such that allows for a heat transfer to the 

cold stream (saturated water at about 100 o C) while maintaining a minimum ∆T of 10 o C. Applying 

MVR closes the gap by compressing low grade steam, consequently increasing its enthalpy and 

temperature and forming superheated steam. The heat from the superheated steam is then used to 

produce more saturated steam. This modification changes the pinch point from 40 - 50 o C to 400 - 

410 o C (Fig. 8) and reduces the hot utility requirements from 12290 MJ to 93 MJ. In achieving this, 

however, a power penalty of 330 kWh/t-CO2 is incurred. 

4.3 Exergy analysis 

Exergy analysis was used to evaluate the process modeled based on the results from heat exchanger 

network implemented through pinch analysis. At any specified surroundings temperature (here T0 = 

15°C = 288 K), using exergy provides a standard basis for calculating the amount of valuable 

energy[30] that can be extracted from a heat stream and comparing heat with power input 

requirement P, for which the exergy Ex(P) = P. For example, ~ 9.1 Gt/t-CO2 heat requirement of the 

245

extraction process at 400 °C (~ 623 K) corresponds to an exergy equal to Ex(Q) = (1-T/T0)·Q = 9.1 

– 3.9 GJ/t-CO2. (T/T0)·Q is the exergy destruction, ED. 

Table 7. Energy (Q), exergy destruction (ED) and requirement (EQ) of the process in GJ/t-CO2 

246 

Q ED EQ 

Mg(OH)2 production 

Kiln 9.09 3.89 5.20 

DISS 0.48 0.46 0.02 

PREP1 1.10 1.06 0.04 

PREP 2 -10.5 -9.70 -0.84 

MVR Compressor 1.18 

Sep-4 -0.89 -0.65 -0.24 

Heat exchangers -2.40 -1.73 -0.67 

Total -1.99 -6.67 4.68 # 

Mg(OH)2 Carbonation 

Turbine -0.24 

-0.24 

Heat exchangers -1.78 -1.22 -0.55 

Total -2.02 -1.22 -0.79 

Net 

3.88 # 

# 

These values are lowered by ~0.45 GJ/t-CO2 if the Fe form in mineral is assumed to be FeO 

instead of the Fe2O3 used here. 

The exergy destruction of a system, which is the measure of the amount by which the value of the 

resource is consumed or degraded, is shown as (8) while the exergy flow is presented in (9); 

(8) 

where (S-S0) is the entropy change, T0 is the ambient temperature and (H-H0) the enthalpy change. 

The results obtained here are compared with the results of a previous model [17] where no pinch 

analysis was done. 

The application of pinch analysis and the heat exchanger network as implemented in the Aspen Plus 

model (Fig.5) resulted in a ~ 0.5 Gt/t-CO2 (~ 10% points) reduction in the exergy requirement of 

producing Mg(OH)2. Mg(OH)2 carbonation unit provides ~ 17% points energy offset to the process. 

When the Mg(OH)2 production and carbonation units are integrated, the process requirements of the 

process becomes 3.88 Gt/t-CO2. This value becomes 3.4 GJ/t-CO2 (reducing by another ~0.5 GJ/t- 

CO2) if the compound form of Fe in mineral is assumed to be FeO instead of the Fe2O3 used here. 


This paper investigated the influence that reaction parameters has on the production of Mg(OH)2 by 

analyzing the effects of Mg/Fe ratio, S/AS ratio, T and t on Mg extraction. Once produced Mg(OH)2 

could be used to sequester carbon by direct reaction with flue gases or CO2 derived from power or 

chemical plants. Notable among the results presented in this paper is the fact that olivine rocks are 

5x less as reactive as their serpentinite counterparts. It was also obvious that serpentinite rocks with 

Mg/Fe < 2.16 were less (>2x) reactive than others. This validates previous results which showed 

that an increase in Mg/Fe ratio increases Mg extraction. Reaction time has a significant effect on 

magnesium extraction as an increase in t above 25 minutes results in a 15 percent points’ increase in 

Mg extraction, but this effect tends to diminish after 60 min. On the other hand, Mg extraction 

shows a negative dependence on reaction temperature; T > 440 o C do not favor Mg extraction. This 


(9)

appears to be due to thermal decomposition of ammonium sulfate leading to the formation of sulfur 

oxide(s), which could alter the thermodynamics of the extraction reactions. 

The application of pinch analysis and the heat exchanger network as implemented in an Aspen Plus 

model resulted in a ~ 0.5 Gt/t-CO2 (~ 10% points) reduction in the exergy requirement of producing 

Mg(OH)2. When the Mg(OH)2 production and carbonation units are integrated the process 

requirements of the process becomes 3.88 Gt/t-CO2. Carbonating Mg(OH)2 in the carbonation unit 

provides a ~17% points energy offset to the entire process. The overall energy requirement of the 

process reduces by another ~0.5 GJ/t-CO2 if the compound form of Fe in mineral is assumed to be 

FeO instead of the Fe2O3 used here. 


This work was supported by the Academy of Finland program “Sustainable Energy” (2008-2011). 

Further support came from KH Renlund Foundation (2007-2009). Financial support from Åbo 

Akademi University’s Graduate School for Chemical Engineering (GSCE) is also acknowledged. 

Appendix 

Apendix A - Tables 

Table A1 . Composition of magnesium silicate minerals tested. 

Elemental composition (%-wt) Mg/Fe 

Rock type and locations Mg Fe Si Ca Al (kg/kg) 

Serpentinite rocks 

N.S. Wales serp. (Aus) 23.0 4.80 19.5 0.00 0.50 4.8 

Donia serp. (Portugal) 22.0 5.01 19.4 0.18 0.88 4.4 

7 Fontes serp. (Portugal) 23.3 5.77 19.5 0.09 1.02 4.0 

Bragança serp. (Portugal) 21.6 5.70 19.6 0.00 0.60 3.8 

Finnish serp. (Fin) 21.8 10.1 11.6 0.30 0.00 2.2 

Lithuania serp. (Lit) 18.9 12.3 15.9 0.90 0.10 1.5 

Olivine rocks 

Åheim olivine (Nor) 29.6 5.00 19.5 0.10 2.80 5.9 

Suomusjärvi-2 (Fin) 12.60 8.32 20.71 5.93 3.71 1.5 

Vammala-2 (Fin) 16.88 12.87 18.37 1.00 0.69 1.3 

Vammala-1 (Fin) 11.58 10.77 21.03 6.43 1.85 1.1 

Suomusjärvi-1(Fin) 8.14 7.62 23.46 5.57 5.72 1.1 

Satakunta olivine (Fin) 3.30 10.7 21.9 6.30 8.50 0.3 

Reactions (A1)-(A5) represent the thermodynamics of reactions involving Mg3Si2O5(OH)4, Fe- and 

Ca-based compounds; iron could be found as FeO/Fe2O3/Fe3O4 and calcium is present as 

CaSiO3[16, 17]. It can be seen from the thermodynamic compositions of possible products of the 

reactions (see also Fig.B1) that MgSO4 is the dominant solid product of the reaction. 

247

Table A2. Extraction equations and thermodynamics 

# Extraction reactions 

(A1) 

(A2) 

(A3) 

(A4) 

(A5) 

Appendix B – Figures 

248 

T (K) 

∆G

Mg-silicate Magnesium silicate mineral. 

MVR Mechanical vapor recompression 

P Pressure, atm 

Q Heat, J 

R&D Research and development 

T Temperature, K 

t time 

S Magnesium silicate mineral 

S/AS Mg-silicate to ammonium sulfate ratio 

∆S Change in entropy, J/mol-K 

W Work, J/s 

Greek symbols 

Δ difference 

∑ sum 

References 

[1] Kojima T, Nagamine A, Ueno N, Uemiya S. Absorption and fixation of carbon dioxide by rock 

weathering. Energy Conversion and Management. 1997;38:S461-S6. 

[2] Lackner K. Climate change: a guide to CO2 sequestration. Science. 2003;300(5626):1677. 

[3] Lackner K. Carbonate chemistry for sequestering fossil carbon. Annu Rev Energy Environ. 

2002;27:193–232. 

[4] Hilley GE, Porder S. A framework for predicting global silicate weathering and CO2 drawdown 

rates over geologic time-scales. Proceedings of the National Academy of Sciences. 

2008;105(44):16855-9. 

[5] Zevenhoven R, Teir S, Eloneva S. Heat optimisation of a staged gas-solid mineral carbonation 

process for long-term CO2 storage. Energy. 2008;33(2):362-70. 

[6] Nduagu E, Björklöf T, Fagerlund J, Mäkilä E, Salonen J, Geerlings H, et al. Production of 

magnesium hydroxide from magnesium silicate for the purpose of CO2 mineralization – Part 2: Mg 

extraction modeling and application to different Mg silicate rocks. Minerals Engineering. 

2012;30:87-94. 

[7] Fagerlund J, Nduagu E, Romão I, Zevenhoven R. A stepwise process for carbon dioxide 

sequestration using magnesium silicates. Frontiers of Chemical Engineering in China. 

2010;4(2):133-41. 

[8] Fagerlund J, Teir S, Nduagu E, Zevenhoven R. Carbonation of magnesium silicate mineral using 

a pressurised gas/solid process. Energy Procedia. 2009;1(1):4907-14. 

[9] Romão I, Eriksson M, Nduagu E, Fagerlund J, Gando-Ferreira L, Zevenhoven R. Carbon 

dioxide storage by mineralisation applied to an industrial-scale lime kiln. 25th International 

Conference on Efficieny, Costs, Optimisation, Simulation and Environmental Impact of Energy 

Systems (ECOS 2012). Perugia, Italy.2012. 

[10] Fagerlund J, Zevenhoven R. The effect of SO2 on CO2 mineral sequestration applied directly to 

a flue gas. 25th International Conference on Efficieny, Costs, Optimisation, Simulation and 

Environmental Impact of Energy Systems (ECOS 2012). Perugia, Italy.2012. 

[11] Zhao L, Sang L, Chen J, Ji J, Teng HH. Aqueous Carbonation of Natural Brucite: Relevance to 

CO2 Sequestration. Environmental Science & Technology. 2009;44(1):406-11. 

[12] Kheshgi HS. Sequestering atmospheric carbon dioxide by increasing ocean alkalinity. Energy. 

1995;20(9):915-22. 

249

[13] Teir S, Kuusik R, Fogelholm CJ, Zevenhoven R. Production of magnesium carbonates from 

serpentinite for long-term storage of CO2. International Journal of Mineral Processing. 2007;85(1- 

3):1-15. 

[14] Rinne J. CO2 sequestration potential of industrial by-products and waste materials in Finland 

[M.Sc.(Eng) Thesis]: Helsinki University of Technology, Finland, 2008. 

[15] Nduagu E. Mineral carbonation: preparation of magnesium hydroxide [Mg(OH)2] from 

serpentinite rock [MSc. (Eng) Thesis]. Turku: Åbo Akademi University, Finland, 2008. 

[16] Nduagu E, Björklöf T, Fagerlund J, Wärnå J, Geerlings H, Zevenhoven R. Production of 

magnesium hydroxide from magnesium silicate for the purpose of CO2 mineralisation – Part 1: 

Application to Finnish serpentinite. Minerals Engineering. 2012;30:75-87. 

[17] Nduagu E, Fagerlund J, Zevenhoven R. Contribution of iron to the energetics of CO2 

sequestration in Mg–silicates-based rock. Energy Conversion and Management. 2012;55(0):178-86. 

[18] Romão I, Nduagu E, Fagerlund J, Gando-Ferreira LM, Zevenhoven R. CO2 Fixation Using 

Magnesium Silicate Minerals. Part 2: Process Energy Efficiency and Integration with iron-and 

steelmaking. Energy – the Int J (special edition ECOS2010). 2012;41 203-11 

[19] Lackner K, Doby P, Yegulalp T, Krevor S, Graves C. Integrating Steel Production with 

Mineral Carbon Sequestration. New York, NY: Columbia University, Department of Earth and 

Environmental Engineering, New York, NY; 2008. p. Medium: ED; Size: 75 pages, 1914 KB. 

[20] Romão I, Ferreira LMG, Fagerlund J, Zevenhoven R. CO2 sequestration with Portuguese 

serpentinite. Conference CO2 sequestration with Portuguese serpentinite, Turku, Finland. p. 77-87. 

[21] Stasiulaitiene I, Fagerlund J, Nduagu E, Denafas G, Zevenhoven R. Carbonation of serpentinite 

rock from Lithuania and Finland. Energy Procedia. 2011;4(0):2963-70. 

[22] Nduagu E, Zevenhoven R. Production of magnesium hydroxide from magnesium silicate for 

the purpose of CO2 mineralisation and increasing ocean alkalinity: effect of reaction parameters. 

Accelerated Carbonation for Environmental and Materials Engineering (ACEME10). Turku, 

Finland 2010. p. 31-40. 

[23] Montgomery D. Design and analysis of experiments. Wiley New York; 1997. 

[24] MINITAB®. MINITAB statistical software, Release 16 for windows, State College, 

Pennsylvania. MINITAB®, is a registered trademark of Minitab Inc2010. 

[25] Nduagu E, Bergerson J, Zevenhoven R. Life cycle assessment of CO2 sequestration in 

magnesium silicate rock – A comparative study. Energy Conversion and Management. 

2012;55(0):116-26. 

[26] Björklöf T. An Energy Efficiency Study of Carbon Dioxide Mineralization [MSc. (Eng) 

Thesis]. Turku: Åbo Akademi University, Finland, 2010. 

[27] Fagerlund J, Nduagu E, Romão I, Zevenhoven R. CO2 fixation using magnesium silicate 

minerals. Part 1: Process description and performance. Energy – the Int J (special edition 

ECOS2010). 2012;41:184-91. 

[28] Kemp IC. Pinch analysis and process integration: a user guide on process integration for the 

efficient use of energy. Oxford, UK.: Butterworth-Heinemann, 2007. 

[29] Linnhoff B. Pinch analysis: a state-of-the-art overview: Techno-economic analysis. Chemical 

engineering research & design. 1993;71(5):503-22. 

[30] Kotas TJ. The exergy method of thermal plant analysis. Stoneham, MA, US.: Butterworth 

Publishers, 1985. 

250

Abstract: 




Thermodynamic analysis of a supercritical power 

plant with oxy type pulverized fuel boiler, carbon 

dioxide capture system (cc) and four-end high 

temperature membrane air separator 

Janusz Kotowicz a , Sebastian Michalski b 

a Silesian University of Technology, Poland, janusz.kotowicz@polsl.pl, 

b Silesian University of Technology, Poland, sebastian.michalski@polsl.pl, 

In this paper the analysis of a supercritical power plant was made. Power of the power plant is 460 MW. The 

parameters of life steam are at 29 MPa/600 oC and of the reheated steam 4.8 MPa/600 oC. Power plant is 

equipped with the following units: oxy type pulverized fuel boiler, "four-end" high temperature membrane 

(HTM) air separator and carbon dioxide capture system (CC). With the assumption of a constant gross 

power of the analyzed power plant the thermal efficiency of the boiler and power consumption of all 

mentioned above units were calculated. These parameters were designated as a function of the recovery 

rate of oxygen in the HTM. This allowed to make the characteristic of efficiency as a function of recovery 

rate. The net efficiency increased from 34.8% to 36.7% with a change of oxygen recovery rate from 0.45 to 

0.9. The effect of membrane working temperature on the efficiency characteristics was also analyzed. 

Integration of CC, HTM air separator and steam cycle was proposed for the increase of the efficiency of a 

power plant. The theoretical analysis was carried out and appropriate calculations were made for this 

integration. 

Keywords: 

OXY type pulverized fuel boiler, air separation, four-end HTM (High Temperature Membrane) 


Currently observed in the world trend in efforts to reduce emissions, especially of greenhouse gases, 

contributes to significant changes in the direction of the development of energy technologies [1]. 

This is very important for the development of coal technologies, due to importance of these fuels in 

the energy balances of many countries, including Poland, as well as due to significant emission of 

CO2 per electricity production unit. In the area of coal technology there are two main research 

directions aiming to bring down the reduction of CO2 unit emission, thus, in consequences, to the 

reduction of global emission: 

search of low-energy consuming carbon capture technologies (including searching for new 

technologies, optimization of known technologies, also in the area of integration of the CCS unit 

with a power plant), 

increasing the efficiency of electricity generation, including optimization of a power plant, both 

in the area of its structure, as well as in area of operation parameters [2]. 

Among carbon capture technologies three directions are developed: 

pre-combustion technology, 

post-combustion technology, 

oxy-combustion technology. 

251

In the area of coal technologies all of these solutions can be used. However, the first solution is 

predisposed for IGCC systems [3], in which there is the possibility of generating carbon dioxide 

before combustion of synthesis gas. The fuel before entering the combustion chamber is subjected 

to a carbon sequestration. Due to the lower gas stream from which the carbon dioxide is removed 

such a separation process is connected with less energy consumption. The next technology is based 

on removing carbon dioxide from flue gases leaving the power system. The post-combustion 

technology is predisposed for the conventional coal-fired power plants. In the area of postcombustion 

technology, as in the case of pre-combustion, the research on absorption and adsorption 

techniques, as well as membrane and cryogenic separation are realized [4÷5]. Among of clean coal 

technology large hopes are associated with oxy-combustion technology, of which the principal 

purpose is combustion of coal in the oxygen-rich atmosphere in order to eliminate from the exhaust 

gases the inert gas (nitrogen). In this case the exhaust gases leaving the steam boiler consists mainly 

of carbon dioxide and steam, so the carbon capture process is much less energy intensive. Currently 

in the research area of oxy-combustion technology, the solutions aiming for decreasing the energy 

consumption connected with oxygen production in the air separation unit are searched for [6÷10]. 

The results presented in the paper were realized within the framework of the Strategic Project 

”Advanced Technologies for Energy Generation: Oxy-combustion technology for PC and FBC 

boilers with CO2 capture”. In the paper the results of the analysis of the steam cycles of energy 

generation units are shown. These steam cycles will be the basis for creation of the models of the 

whole oxy-combustion power plants. In the paper the results of analysis including the influence of 

different solutions of steam cycles, and thus their assumed parameters, on the energy effectiveness 

evaluation indicators are shown. 

2. Model of the air separation unit integrated with the oxy type 

pulverized boiler and assumptions for calculations 

Air separation unit structure consists of: a counter-current air heater (APH), an air compressor (C), 

an expander (EX), a generator (G) and a "four end" type membrane (M). The expander drives the 

air compressor. Depending on the assumed quantities the expander and compressor can give or take 

electricity from the grid. The structure of the air separation unit is shown in Figure 1. 

Fig. 1. Scheme of the air separation unit (ASU) 

252

Oxy type pulverized boiler structure consist of: a combustion chamber (CC), an evaporator divided 

into two parts (EVAP), a counter-current air heater (APH), three recirculated flue gas heaters 

(RHX1, RHX2 and RHX3), a live steam superheater (LSSH), a reheated steam superheater (RSSH), 

an economizer (ECO), an electrostatic precipitator (EP), a flue gas extractor fan (F1), a flue gas 

dryer (FGD) and a recirculated flue gas fan (F2). The structure of air separation unit integrated with 

the oxy type pulverized boiler is shown in Figure 1. The basic characteristic quantities of integrated 

models are gathered in Table 1. 

Fig.2. Scheme of air the separation unit (ASU) integrated with OXY type pulverized boiler 

253

Tabel 1. Characteristics quantities for considered air separation unit integrated with the oxy type 

pulverized boiler 

Name Symbol Value Unit 

Excess air coefficient 1.2 - 

Live steam flow rate, Reheated steam flow rate 6s 8s ,m m 336.05, 

284.42 

kg/s 

°C, kPa 

Temperature, Pressure of live steam leaving the boiler t6s, p6s 604.90, 

30100.00 

Temperature, Pressure of reheated steam leaving the 

boiler 

254 

t8s, p8s 

602.40, 

4918.03 

°C, kPa 

Temperature of feed water t1s 297.00 °C 

Temperature of recycled feed gas leaving PRS t17g 320 °C 

Efficiency of heat exchangers in the boiler heat 

wck 99.8 % 

exchangers Moisture content in the flue gas leaving flue gas dryer (H2O)14g 10 % 

Share of oxygen in the mixture of the recycled flue gas 

and oxygen from ASU supplied to the combustion 

chamber 

xO2,per 0.3 kmol, O2 

kmol, gw 

Ambient pressure, Pressure of flue gas supplied to the 

carbon dioxide capture unit 

pot , p16g 101.325 kPa 

Ambient temperature tot 20 

o 

C 

Fan isentropic efficiency, Compressor isentropic 

efficiency 

iF,iS 0.88 - 

Expander isentropic efficiency iT 0.9 - 

Generator efficiency g 0.99 - 

Share of slag in the ash u 40 

% 

Share of fly ash in the ash upl 60 

% 

Share of carbon element from coal in the ash c 0.5 

% 

Membrane work temperature, Temperature of flue gas 

at the inlet to the separation membrane 

tmem , t19g 850 

o 

C 

Specific temperature difference of the permeaterecycled 

flue gas heat exchanger 

TsP-S=t1g- 50 K 

Oxygen recovery rate R 40÷100 - 

Compressor pressure ratio k 15 - 

Energy consumption of electrostatic precipitator - 1 kJ/kg,fg 

Energy consumption of coal mill - 64.8 kJ/kg,fuel 

Energy consumption of carbon dioxide capture unit - 316.03 kJ/kg,fg 

It was assumed that the air taken from environment is a dry gas consisting of 21% oxygen and 79% 

nitrogen (volumetric composition). 

The characteristic quantities gathered in Table 1 were used for computations performed on the 

model of the oxy-type pulverized boiler integrated with the air separation unit, made in the 

t18g

GateCycle TM software. The built-in components were used to build the integrated model. It was 

assumed that through the membrane flows pure oxygen. The auxiliary power rate of steam cycle, 

the live steam thermodynamic parameters, reheated steam thermodynamic parameters, mass flow 

rates of live steam and reheated steam were assumed as for a 460 MW power plant. The gross 

power of steam turbine is constant, despite electricity generated in the air separation unit. 

Energy consumption of the carbon dioxide capture unit was calculated with use of the model of 

carbon dioxide capture unit. The structure of this model is shown in Figure 3. The energy 

consumption value is correct for the specific composition of the flue gas. 

Rys.3. Scheme of the carbon dioxide capture unit (CC) 

3. The results of calculations of air separation unit integrated 

with the oxy type pulverized boiler 

The air mass flow rate depends on the separated in membrane oxygen mass flow rate ( m O2 ), oxygen 

recovery rate (R) and mass content of oxygen in the air ( g O2air 

). The relationship between these 


m 

O2 

m 

1a 

(1) 

R 

g 

O2air 

Next the air is flowing to the compressor. Effective power required to drive the compressor depends 

on the air mass flow rate ( 1a m ), the air temperature ( ~c 1a 

), the 

255 

T ), the average specific heat ( p K 

1 

compressor pressure ratio ( K ), the heat capacity ratio contained in the factor ( K 

), 

K 

the compressor isentropic efficiency ( iK ) and the compressor mechanical efficiency ( mK ). The 

equation showing the relationship between these quantities is as follows: 

N m 

eK 

1a 

 

 

 

 

 

K 

~ 

 

K 1 

 

c T 

p 

K 

1a 

iK 

 

mK 

 

 

(2) 

The mass flow rate of gas flowing through the expander is lower than the mass flow rate of gas 

flowing through the compressor. This mass flow rate depends on the oxygen mass flow rate 

separated from the air in the membrane ( m O2 ) and the air flow rate ( m 1a ).The relationship between 

these quantities is as follows: 

m m 

m 

4a 

1a 

O2 

(3) 

The expander effective power depends on the retentate mass flow rate ( a m 4 ), the retentate 

temperature ( T 4a ), the average specific heat (~c p ), the compressor pressure ratio ( 

K 

K ), the 

reduction factor of compressor pressure ratio ( ), the heat capacity ratio contained in the factor 

1 

( T ), the expander isentropic efficiency ( iT ) and the expander mechanical efficiency 

T 

). The equation showing the relationship between these quantities is as follows: 

( mT

N m 

eT 

4a 

 

~ T 

cp 

T4a 

1 ( K) 

iT 

mT 

T 

The gross electrical power of the air separation unit depends on the expander gross power ( N eT ), 

the compressor gross power ( N eK ) and the generator efficiency ( g ). The relationship between 

these quantities is as follows: 

N 

elTG 

N eT eK g 

N 

(5) 

Figure 4 shows a graph of boiler thermal efficiency as a function of oxygen recovery rate of the air 

separation unit. It should be noted that the boiler thermal efficiency increases from about 55% for 

low oxygen recovery rates to about 83% for high recovery rates. The boiler thermal efficiency 

depends on the live steam flow rate ( 6s m ), the reheated steam flow rate ( m 8s ), the enthalpy of live 

seam leaving the boiler ( 6s h ), the enthalpy of feed water at inlet to the boiler ( h 2s ), the enthalpy of 

reheated steam leaving the boiler ( h 8s ),the enthalpy of reheated steam at the inlet to the boiler ( h 7s ), 

the fuel mass flow rate ( 1c m ) and fuel lower heating value ( W dp ). The equation showing the 

relationship between these quantities can be written as: 

m 6s h6s 

h2 

s m 

8s h8s 

h7s 

 

k 

(6) 

m 

W 

Boiler thermal efficiency, % 

85 

80 

75 

70 

65 

60 

1c 

dp 

55 

40 45 50 55 60 65 70 75 80 85 90 95 100 


Fig.4. Thermal efficiency of the oxy type boiler as a function of the air separation unit oxygen 

recovery rate 

The thermal efficiency of steam cycle depends on the heat supplied to the steam cycle ( d Q ) and the 

heat discharged from the steam cycle ( w Q ).The relationship between these quantities is as follows: 

Q 

Q 

d w 

obp 

Q 

(7) 

d 

The gross efficiency of electricity generation is calculated in terms of the generator electrical power 

driven by the steam turbine. The electricity generated in other units is included in their auxiliary 

recovery rates. The gross efficiency of electricity generation depends on the steam turbine electrical 

power ( el 

256 

N ), the fuel flow rate ( 1c m ) and lower heating value ( W dp ).The equation showing the 


(4)

N 

el 

el, brutto 

(8) 

m1c 

Wdp 

The net efficiency of electricity generation depends on the gross efficiency of electricity generation 

( el, brutto), 

the auxiliary power rate of air separation unit ( pw, ASU ), the auxiliary power rate of carbon 

dioxide capture unit ( pw, CCS ), the auxiliary power rate of steam cycle ( pw, bloku) 

and the auxiliary 

power rate of the boiler ( pw, koto ). The equation showing the relationship between these quantities is 

as follows: 

el, nettoel, 

brutto ( 

1 

pw, ASU 

pw, CCS 

pw, bloku 

pw, koto ) 

Auxiliary power rates except the auxiliary power rate of the air separation unit, depends on 

auxiliary power of unit ( N pw, i ) and the steam turbine electric power ( N el ).The equation showing the 


 

pw, i 

N 

pw, i 

N 

el 

(10) 

The auxiliary power rate of the air separation unit depends on the mechanical power used to drive 

the compressor ( N SN ), the mechanical power generated in expander ( N TR ) and the steam turbine 

electrical power ( N el ).The equation showing the relationship between these quantities is as follows: 

N N 

SN TG 

pw, ASU 

(11) 

Nel 

Figure 5 shows a graph of auxiliary power rates of the boiler, steam cycle and carbon dioxide 

capture unit as a function of oxygen recovery rate. Figure 6 shows a graph of the auxiliary power 

rate of the air separation unit. It should be noted that the auxiliary power rate of the air separation 

unit, unlike the other auxiliary power rates, has a negative value in the studied range of oxygen 

recovery rate. 

Auxiliary power rate, % 

12 

11 

10 

9 

8 

7 

6 

5 

4 

Steam cycle 

3 

40 45 50 55 60 65 70 75 80 85 90 95 100 

257 

(9) 


Fig.5. The auxiliary power rates of the carbon dioxide capture unit (CCS), boiler and steam cycle 

as a function of the oxygen recovery rate

Auxiliary power rate of ASU, % 

40 45 50 55 60 65 70 75 80 85 90 95 100 

0 

-5 

-10 

-15 

-20 

-25 

-30 

-35 

-40 


Fig.6. The auxiliary power rate of the air separation unit as a function of the oxygen recovery rate 

Figure 7 shows a graph of the gross and net efficiency of electricity generation as a function of 

oxygen recovery rate. This characteristics were determined with the use of equations (8) and (9). 

Efficiency of electricity generation, % 

43 

41 

39 

37 

35 

33 

31 

29 

40 45 50 55 60 65 70 75 80 85 90 95 100 


Fig.7. Gross and net efficiency of electricity generation as a function of the air separation unit 

oxygen recovery rate 

258

Summary 

In this paper the air separation unit integrated with the oxy type pulverized boiler was analyzed. The 

oxy type boiler supplies the live steam and the reheated stem to the steam cycle. The gross power of 

the steam turbine generator of the steam cycle is equal to 460 MW. 

For the analysis the characteristics of the boiler thermal efficiency, auxiliary power rate of the steam 

cycle, carbon dioxide capture unit and air separation unit as a function of oxygen recovery rate in 

the "four-end" type separation membrane were determined. 

It should be noticed that the value of the boiler thermal efficiency shown in Figure 4 is increasing 

from about 55% (for 40% of the oxygen recovery rate) to about 83% (for 100% of the oxygen 

recovery rate). This characteristic is nonlinear and the acceleration of growth of the boiler thermal 

efficiency decreases with increasing oxygen recovery rate. 

The auxiliary power rate of the carbon dioxide capture unit and the auxiliary power rate of the oxytype 

pulverized boiler shown in Figure 5 decreases with increasing the oxygen recovery rate. This 

characteristics are nonlinear. The auxiliary power rate of steam cycle is the same for all oxygen 

recovery rates, because the power of steam turbine is held at a constant level. The auxiliary power 

rate of the air separation unit shown in Figure 6 unlike the other auxiliary power rates has a negative 

value in the studied range of oxygen recovery rate. This means that the expander generates more 

power than the power needed to drive the compressor. The auxiliary power rate of the air separation 

unit increases with the increase of the oxygen recovery rate. 

It should be noticed that the value of the net efficiency of electricity generation shown in Figure 7 is 

increasing from about 35% (for 40% of the oxygen recovery rate) to about 38% (for 100% of the 

oxygen recovery rate). This characteristic is nonlinear and the acceleration of growth of net 

efficiency of electricity generation decreases with increasing the oxygen recovery rate. This graph 

indicates that the studied integrated models have the highest net overall efficiency for the oxygen 

recovery value equal to 100%. 






LITERATURE 

[1] Chmielniak T., The role of various technologies in achieving emissions objectives in the 

perspective of the years up to 2050. Rynek Energii, 2011;92:3-9 

[2] Chmielniak T., ukowicz H., Kochaniewicz A., Trends of modern power units efficiency 

growth. Rynek Energii, Nr 6(79), 2008, 14-20. 

[3] Badyda K., Kupecki J., Milewski J., Modelling of integrated gasification hybrid power systems. 

Rynek Energii, 2010;88:74-79. 

[4] Kotowicz J., Janusz-Szymaska K., Influence of CO2 separation on the efficiency of the 

supercritical coal fired power plant. Rynek Energii, 2011, 2 (93), 8-12. 


35 (2010), 943 – 951. 

[6] Toftegaard M.B., Brix J., Jensen P.A., Glarborg P., Jensen A.D., Oxy-fuel combustion of solid 

fuels. Progress in Energy and Combustion Science, 2010;36:581-625 

[7] Dillon D.J., White.V., Allam R.J., Wall R.A., Gibbins J., Oxy-combustion Process for CO2 

Capture from Power Plant. Mitsui Babcock Energy Limited,2005 

259

[8] Buhre B.J.P., Elliott L.K., Sheng C.D., Gupta R.P. and Wall T.F., Oxy-fuel combustion 

technology for coal-fired power generation. Progress in Energy and Combustion Science, 

2005;31:283-307 



Energy Procedia 1 (2009) 495-502. 


membranes. International Journal of Greenhouse Gas Control 5 (2011) 7-15. 

260

Abstract: 




A multi-objective optimization technique for coprocessing 

in the cement production 

Maria Luiza Grillo Renó a , Rogério José da Silva b , Mirian de Lourdes Noronha 

Motta Melo c , José Joaquim Conceição Soares Santos d 

a Federal University of Itajubá, Itajubá, Brazil, malureno@yahoo.com.br 

b Federal University of Itajubá, Itajubá, Brazil, rogeriojs@unifei.edu.br 

c Federal University of Itajubá, Itajubá, Brazil, mirianmottamelo@unifei.edu.br 

d Federal University of Espírito Santo, Vitória, Brazil, jose.j.santos@ufes.br 

Cement is produced by blending different raw materials and by burning them at high temperatures. The 

Portland cement production is a process that demands high energy consumption; therefore the process 

needs a great amount of fuel. Moreover, industrial waste has been used by Portland cement industries as a 

secondary fuel through a technique called co-processing. For energy conservation in this work, mineralizers 

are added into the clinker formation. The mineralizers promote the decrease the temperature in the kiln and 

improve the quality of the clinker. The purpose of the present work is to provide an analysis of an optimal 

production point through an optimization technique with multiple objective functions, in which genetic 

algorithms, sequential quadratic programming are applied. 

Keywords: 

Optimization, Portland cement, Co-processing, Mineralizers. 


The cement industry is connected to the environment. The production process requires energy and 

this causes to emissions. Information on energy consumption including secondary fuels in the 

cement industry is relatively well known. Fossil fuels (e.g. coal, oil or natural gas) are the 

predominant fuels used in the cement industries. However, low-grade fuels such as petrol coke and 

waste derived fuels have been increasingly utilised in the recent years. 

The use of waste derived fuels, as alternative secondary fuels, has been showed itself as a possible 

path so that the cement industries reduce its production costs and reduce the consumption of fossil 

fuels. Moreover, for the residues industries generators is the expected solution for the disposition 

demanded by the environmental legislation. Now the spectrum of residues co-processed in the 

cement industry is varied being added the list of alternative fuels used since the decade of eighty: 

used oils, waste of the re-refining process of lubricating oils, used tyres, shavings of tyres, residues 

of solvents, dregs of paintings, urban residues and treatment muds and others. 

Besides, it is well known that the most energy-demanding phases of cement manufacturing is the 

clinkering process, which consumes 80% of the total energy for pyro processing. Thus a potential 

for energy conservation, which could be rapidly and economically accomplished by promoting the 

clinker formation at lower temperature in the kiln by using mineralizers that promote certain 

reactions and fluxes that lower the melting point of clinker liquid phase. Then this work proposes 

the combination of mineralizers and alternative fuel in cement production with main to reduce the 

production costs and environmental impacts. 

261

1.1 - Cement Manufacturing 

The cement manufacturing process consists broadly of quarrying, crushing and grinding, burning, 

and grinding with gypsum. Two basic processes, the wet process and the dry process, are used for 

cement manufacturing. In the wet process, proper proportions of the raw materials are mixed with 

water. The mixture is called slurry. While in the dry process there is not water with raw materials. 

Both processes of the raw materials are proportioned, mixed, grounded and pulverized, and then 

pumped into a rotary kiln. Inside the kiln, the raw materials are subjected to a thermal treatment 

process consisting of consecutive steps of drying/preheating, calcination, and sintering (or 

clinkerisation, at temperatures up to 1450°C). 

The burnt product “clinker” is cooled down with air to 100-200°C and is transported to intermediate 

storage. After, a few percentage of natural or industrial gypsum is added to clinker to regulate the 

setting time of cement. Finally, the finished product, known as the Portland cement, is stored in 

large storage bins called silos, from which is fed to an automatic packing machine. 

Fig. 1 – Cement manufacturing process [1] 

1.2 – Emissions of cement production 

The atmospheric emission of the cement plant includes particles, NOx, SO2, CO2 and lower amounts 

of CO, organic compounds such as metals (mainly adhered to particles), as well as other minor 

pollutants [2]. 

The SO2 emissions are primarily determined by the content of volatile sulphur in raw materials and 

fuel used. Sulphur dioxide emissions result from the combustion of fuel and raw materials. In 

addition, SO2 emissions can result from decomposition of alkali or calcium sulphates that are 

trapped within the volatile cycles of the kiln [3]. 

Sulphur input is usually absorbed within a range of 50-90 % of input. Modern kiln system of 

preheat and pre-calciner types tend to have higher rates of SO2 absorption. The unabsorbed portion 

of the sulphur input is emitted from the kiln system as SO2 and SO3. 

262

NOx emissions from cement kilns depend on both the kiln type and the fuel type. In general, preheater 

and precalciner kilns have lower NOx emission rates than long dry and wet kilns, due to the 

higher fuel efficiency and lower firing rates in the kiln firing zone [4]. 

NOx is formed during fuel combustion by oxidation of molecular nitrogen of combustion air as well 

as nitrogen compounds of the fuel. Significant oxidation of molecular nitrogen of combustion air 

takes place in oxidizing flames with a temperature above 1200°C. The NOx formed in this way is 

named thermal NOx, otherside the NOx formed by oxidation of nitrogen compounds in the fuel 

which is named fuel NOx. 

1.3 – Co-processing 

Cement companies are introducing the industrial waste co-processing from a perspective that 

combines environmental policies with the interests of companies. Industrial wastes with residual 

energy and low content of chlorides and heavy metals can be appropriate to provide part of the 

energy required to make cement. 

The process of clinker burning in rotary kiln creates favourable conditions for the use of industrial 

waste like alternative fuel. These conditions are high temperature, alkaline environment, oxidizing 

atmosphere, large heat-exchange surface, good mixture of gases and products, and sufficient time 

(over 2 seconds) for the disposal of hazardous wastes [5]. The Table 1 presents examples of 

alternative fuels used in cement industry. 

Table 1 – Examples of Alternative Fuels 

Alternative Fuel Examples 

gaseous Landfill gas, pyrolysis gas 

Liquid Pasty wastes, solvents, waste oils, greases 

Solid Paper, used tires, rubber wastes, plastics 

The used tires are material residual special, based on higher heating value (28–32 MJ/kg of tyres), 

they are excellent sources of energy, mainly when used as secondary fuels. The high temperature, 

the high time of residence, the high effect of absorption of the raw material in the pre-heating and 

the incorporation of the ashes generated to the clinker, are favourable conditions for burning of 

tyres in rotary kilns of clinker production, so that, is an adequate form to final disposition for these 

wastes. Besides, due to the high calorific value of the tires, the co-processing contributes to a 

decrease in consumption of others fossil fuels utilized (as coal, petroleum coke and fuel oil), saving 

the natural resources [6]. 

1.4 – Mineralizers 

The use of mineralizers in the cement industry is widely known. The incorporation of compounds 

other than those usual in low proportions improves the clinkering conditions as well as decreases 

the maximum clinkering temperature or improves the phase formation in the clinker without 

altering the final properties of the product [7]. The mineralizing properties of the compounds CaF2 

and CaSO4 have already been described in the literature. Their properties are different when they 

are added separately or jointly in the raw materials. 

It has been verified that the combined addition of CaF2 and CaSO4 to raw clinker materials results 

in a decrease of the maximum clinkering temperature to approximately 1350°C, in a clinker with a 

good proportion of alit and a cement with satisfactory mechanical properties [7]. The reduction of 

temperature by 100°C is expected to result in a saving of fuel by 80-100 kcal/kg of clinker [8]. 

263

2. Numerical model 

The purpose of this section is to provide an analysis of an optimal production point through 

optimization problem with multi-objective functions and constraints. The multi-objective functions 

are the clinker production cost, NOx and SO2 emissions. The constraints refer to operational 

parameters and clinker quality. 

2.1. Objective functions 

Clinker production cost function: 

The clinker production cost taking into consideration raw materials cost as well the energy 

consumption requested for grinding. This last can be written with respect to the specific surface of 

the cement, silica modulus and electricity cost (Eq. (1)). 

E ce A 

( ) 

exp BS 

(1) 

Where: 

A 5,76( 

SM ) -5,82 

B -0,2( SM) 

0,98 

SM = Silica Modulus; ce = electricity cost; S = specific surface of cement 

Silica modulus (SM) is the ratio of the silicates oxide with the sum of the ferric oxide and alumina 

oxide in the clinker. The Silica Modulus has influence on the burning of raw materials, clinker 

granulometry and liquid phase. The raw materials and fuel costs are written as linear combination, 

Eq. (2): 

n 

 

C p ( 

x ) 

Where: 

i1 

i i 

pi = raw materials and fuels medium cost. 

xi = raw materials and fuel quantity. 

The raw materials of clinker production are limestone (x1), clay (x2), sand (x3), and iron ore (x4). 

The fuels used are mineral coal (x5), petroleum coke (x6), and the alternative fuel used is the scrap 

tires (x7). The mineralizers applied in this work are CaF2 (x8) and CaSO4 (x9), these mineralizers 

proceed from industrial wastes and it is considered as revenue for the cement industry, with a 

symbolic income of US$ 1/ton. The final cost equation (clinker production cost) is presented in Eq. 

(3): 

f ( x) 1.25x 1.45x 0.87 x 10x 63.11x 40x 1x 1x 1x 

1 1 2 3 4 5 6 7 8 9 

ce SM e 

Where 

( 0.2( SM) 0.98)* 

S 

*{(5.76( ) 5.82)* } 

264 

(2) 

(3)

0.61x 58.86x 83.67x 4.4x 4.03x 1.97x 18.80x 1.41x 

SM 

0.32x 21.31x 5.6x 92.58x 6.975x 0.9207x 0.09x 

SO2 and NOx emissions functions: 

265 

2 

S 4 cm / kg 

In this optimization model the SO2 and NOx emissions functions refer to S and N contained in fuels 

(mineral coal, petroleum coke and used tires). Table 2 shows the composition of S and N in fuels 

used. Eq. (4) and (5) represent the SO2 and NOx emissions functions respectively. The composition 

is obtained in [9], [10], [11]. 

Table 2 – N and S composition in the fuels 

Mineral coal 

(%) 

f ( x) 0.055x 0.05x 0.0123x 

Petroleum coke 

(%) 

Tires used 

(%) 

N 1.2 1.5 0.36 

S 5.5 5 1.23 

2 5 6 7 

f ( x) 0.012x 0.015x 0.0036x 

3 5 6 7 

2.2. Constraints 

1 2 3 4 5 7 8 9 

1 2 3 4 5 7 9 

Equation (6) up to Eq. (8) represents the operational order restrictions for SiO2, Al2O3 and Fe2O3 

content, respectively. The data were obtained in Brazil cement industry. 

0.0061x 0.59x 0.837x 0.044x 0.0403x 0.0192x 0.188x 0.0141x 0.216 

1 2 3 4 5 7 8 9 

0.0015x 0.171x 0.047x 0.027 x 0.017 x 0.0079x 0.0009x 0.062 

1 2 3 4 5 7 9 

0.0017x 0.042x 0.0095x 0.90x 0.053x 0.0013x 0.025 

1 2 3 4 5 7 

Equation (9) and Eq. (10) represent the silica modulus that present an adequate value for interval 

between 2.3 and 2.7. Eq. (11) and Eq. (12) relate to alumina modulus. This modulus influences 

mainly on the burning process by acting on speed of the reaction of limestone and silica. The values 

for this modulus are within the range of 1.3 and 2.7. Eq. (13) and Eq. (14) refer to the lime 

saturation factor. A high factor of lime saturation causes burning difficulties. Acceptable values for 

this factor are between 0.9 and 1. 

-0.216x19.85x2 70.8x3 208.53x4 12.01x5 0.20x7 18.80x8 1.20x9 0 

(9) 

-0.254x11.32x2 68.55x3 245.57x4 14.8x50.57 x7 18.80x8 1.17 x9 

0 

(10) 

-0.071x1 11.65x2 3.42x3 114.12x4 5.146x50.63x7 0.09x9 0 

(11) 

(4) 

(5) 

(6) 

(7) 

(8)

-0.309x15.77x2 2.085x3 240x4 12.53x5 0.452x7 0.09x9 0 

(12) 

51.57x1157.1x2 210.64x3 70.33x4 13.69x5 4.77x7 34x8 29.3x9 0 

(13) 

51.37x1-175.71x2-234.65x3-78.15x4-15.37x5-5.41x7-39.21x8 28.9x9 0 

(14) 

The total feeding of fuel must satisfy the specific heat consumption. This value would be 3600 

kJ/kg clinker, but the additions of mineralizers reduce to 3181 kJ/kg clinker. Eq. (15) presents the 

specific heat consumption; the coefficients of equation are Lower Heating Value of fuels. Eq. (16) 

denotes the consumption of used tires. Eq. (17) and (18) are the mineralizers’ limits for CaF2 and 

CaSO4. These limits are obtained in work [8] that established 1 % (mass) of each mineralizer. 

27670x 36425x 32100x 3181 

32100x7 795 

x 0.01634 

8 

x 0.01634 

9 

5 6 7 

3. Optimization techniques 

In this work, the nonlinear problem defined with multi-objective functions and constraints was 

solved using Genetic Algorithms (GAs) and Sequential Quadratic Programming (SQP). 

3.1. Genetic Algorithms (GAs) 

Genetic algorithms (GAs) work on the principle of “survival of the fittest”. They have been 

extensively applied by many optimization problems. In GAs, the decision variables are encoded in a 

string form. The encoded solutions are called chromosomes and the elements of the chromosomes 

are called genes. 

Depending on the nature of the problem the encoded solution may include binary digits or real 

numbers. An initial population is created and the fitness (the objective function value) of the 

population members is evaluated. Genetic operators (mutation and crossover) are applied to keep 

the gene pool diverse that aids the inclusion of better fitted members for quick convergence [12]. 

3.2. Sequential Quadratic Programming (SQP) 

The main idea in SQP is to obtain a search direction by solving a quadratic program, that is, a 

problem with a quadratic objective function and linear constraints. This approach is a generalization 

of Newton’s method for unconstrained minimization [13], and it is used to solve a nonlinear 

program in the following form: 

Minimize f ( x) 

Subject to g( x) 0 

Where g is a vector of m constraint functions gi. Applying Newton’s method to the corresponding 

optimality conditions, the Lagrange for this problem is obtained as: 

266 

(15) 

(16) 

(17) 

(18) 

(19)

T 

L( x, ) f( x) g( x) 

(20) 

Thus, the formula for Newton’s method is: 

xk1 xk pk 

v 

k1 k k 

where pk and vk are obtained as the solution to the following linear system: 

2 pk Lx ( k, k) Lx 

( k, k) 

vk This linear system has the form: 

2 

xxL( xk, k) g( xk) pk 

xLx ( k, k) 

 

 

 

T 

g( x ) 0 

 

vk gx ( k) 

k 

 

 

These formulas are used in the method for constrained optimization. This system of equations 

represents the first-order optimality conditions for the optimization problem: 

1 T 2 

T 

MinimizepxxLx ( k, k) p p xLx 

( k, k) 

2 

 

 

subjecttogx ( ) pgx ( ) 0 

T 

k k 

 

With v k is the vector of Lagrange multipliers. This optimization problem is a quadratic program, 

where the quadratic function is a Taylor series approximation to the Lagrange at (xk, k), and the 

constraints are a linear approximation to g(xk +p) = 0. In the SQP method, at each iteration a 

quadratic program is solved to obtain (pk,vk), which are used to update (xkk), and the process 

repeats at the new point. 

4. Results 

Numerical results for SQP and GA are obtained using the SQP function GA function from 

optimization toolboxes in a computer package [14]. The SQP algorithm shows fast convergence 

rates. Nevertheless the GA has a computer time of 10 seconds. The GA obtain a global optimal, and 

the SQP a local optimal. The results of objective functions and optimal points are presented in 

Table 4. 

267 

(21) 

(22) 

(23) 

(24)

Table 4 – Results of the optimization model 

Functions (US$/ton clinker) 

GA f1 = 8,54 f2 = 0,0047 f3 = 0,00125 

SQP f1 = 8,39 f2 = 0,0038 f3 = 0,0011 

GA 

SQP 

x1 = 0,823 

x2 = 0,1149 

x3 = 0,1115 

x1 = 1,0783 

x2 = 0,3505 

x3 = 0 

x4 = 0,0801 

x5 = 0,03369 

x6 = 0,05448 

x4 = 0,0094 

x5 = 0 

x6 = 0,0704 

Variables (kg/kg clinker) 

x7 = 0,0082 

x8 = 0,5284 

x9 = 0,8534 

x7 = 0,0192 

x8 = 0,0088 

x9 = 0,0163 

The numerical results for SQP and GA conclude that the SQP method presents better solutions 

since the objective functions present minor values. Moreover, the GA did not attend all restrictions 

as well as the restrictions regarding the quantity of mineralizers 

Therefore, the optimization technique mostly applied for this optimization problem is the SQP 

because of minor results and it attends all restrictions. Both techniques showed emissions values 

(SO2 and NOx) smaller than the ones in European laws. The European laws values are: 

SO2 : 0,02 – 7,00 kg/ton clinker [15] 

NOx : 0,4 – 6 kg/ton clinker [15]. 


In this work were presented a multi-objective optimization technique for co-processing in the 

cement production. SQP and GA optimizations were applied in order to obtain optimal point and 

objective functions. 

The results of SO2 and NOx emissions, as well as the cost function were better when it was applied 

the Sequential Quadratic Programming method. Besides the emissions were smaller than the values 

that regulates in Europe. 

Future works can be developed focusing the reduction of others emissions, mainly the carbon 

dioxide and heavy metal provide of co-processing. The application of others mineralizers e its 

influence in energy consumption is also important for cement production that looking for reducing 

the fossil fuel consumption and production costs. 


The authors would like to acknowledge FAPEMIG – Fundação de Amparo à Pesquisa do Estado de 

Minas Gerais, CNPq – Conselho Nacional de Pesquisa and CAPES – Coordenação de 

Aperfeiçoamento de Pessoal de Nível Superior for the finanacial support that allowed to conduct 

this study. 

References 

[1] SAMAC, Sand & Cement Cogeneration Plant – Available at: 

[accessed 12.16.2011]. 

[2] Schuhmacher M., Doming J.L., Garreta, J., Pollutants emitted by a cement plant: health risks 

for the population living in the neighborhood. Environmental Research 2004; 95:198-206. 

268

[3] Greer, W.L., SO2/NOx Control Compliance with Environmental Regulations. IEE Transactions 

on Industry Applications 1989; 25:475-485. 

[4] McQueen, A.T., Bortz, S.J., Hatch, M.S., Leonard, R.L., Cement Kiln NOx Control. IEE 

Transactions on Industry Applications 1995; 31:36-44. 

[5] Mokrzycki, E., Uliasz-Bochénczyk, A., Alternative Fuels for the Cement Industry. Applied 

Energy 2003; 74:95-100. 

[6] Souza F.A., Utilização de Pneus Pós-Consumo como Combustível em Fornos de Cimento. In: 

CETESB, Seminário Nacional Sobre Reuso/Reciclagem de Resíduos Sólidos Industriais; 2000; 

São Paulo. 

[7] Giménez-Molina, S., Blanco-Varela, M.T., Solid state phases relationship in the CaO-SiO, 

Al2O, CaF2-CaSO4 system. Cement and Concrete Research 1995; 25:870-882. 

[8] Raina, K., Janakiraman, L.K., Use of Mineralizer in Black Meal Process for Improved 

Clinkerization and Conservation of Energy. Cement and Concrete Research 1998; 28:1093- 

1099. 

[9] Lawrence, L.B., Joseph, R.H., Chemical Analyses of Middle and Upper Pennsylvanian Coals 

form south eastern Kansas – Available at: [accessed 3.16.2006]. 

[10] Salomon V.G., Evaluation of the Effects of the Presence of Heavy Metals in Waste Coprocessed 

when Used as Alternative Fuels and Raw Materials in Cement Industry [dissertation]. 

Itajubá, Brazil: Universidade Federal de Itajubá; 2002. 

[11] Kaplan E., Nedder N., Petroleum Coke Utilization for Cement Kiln Firing. In: Cement 

Industry Technical Conference, IEEE-IAS/PCA; 2001 29 Abr – 03 Mai; Vancouver, Canada. 

Academic Press: 251-263. 

[12] Jha, M.K., Abdullah, J., A Markovian approach for optimization highway life-cycle with 

genetic algorithms by considering maintenance of roadside appurtenances. Journal of the 

Franklin Institute 2006; 343:404-419. 

[13] Nash S.G., Sofer, A., A Linear and Nonlinear programming. McGraw-Hill, NY: 1996. 

[14] MatLab 7.0, MathWorks, USA – Available at: [accessed 

6.22.2004]. 

[15] Carpio R.C., Optimization in Co-processing Waste in Cement Industry involving Costs, 

Quality and Environmental Impact [Thesis]. Itajubá, Brazil: Universidade Federal de Itajubá; 

2005. 

269




Abstract: 

Comparison of options for debottlenecking 

the recovery boiler at kraft pulp mills – 

Economic performance and CO2 emissions 

Johanna Jönsson a , Karin Pettersson b , Simon Harvey c and Thore Berntsson d 

a Div. of Heat and Power Technology, Dept. of Energy and Environment, Chalmers University of 

Technology, Göteborg, Sweden, johanna.jonsson@chalmers.se 

b Div. of Heat and Power Technology, Dept. of Energy and Environment, Chalmers University of 

Technology, Göteborg, Sweden, karin.pettersson@chalmers.se, 

c Div. of Heat and Power Technology, Dept. of Energy and Environment, Chalmers University of 

Technology, Göteborg, Sweden, simon.harvey@chalmers.se 

d Div. of Heat and Power Technology, Dept. of Energy and Environment, Chalmers University of 

Technology, Göteborg, Sweden, thore.berntsson@chalmers.se 

The trend in the European pulp and paper industry is toward fewer mills with larger capacity. As a 

result, a number of existing mills will increase their production capacity. For increased production 

capacity in kraft pulp mills, the recovery boiler is often a bottleneck.This paper compares three 

different options for debottlenecking the recovery boiler and utilizing a potential mill steam surplus at a 

typical Scandinavian kraft pulp mill, when increasing the production capacity by 25%:1) Upgrading the 

recovery boiler, 2) Lignin extraction and 3) Black liquor gasification (as a booster). For black liquor 

gasification (BLG) two options for using the product gas are considered: production of electricity or 

DME motor fuel. Furthermore, both BLG and upgrading of the recovery boiler are assumed to be 

possible to combine with carbon capture and storage (CCS). The extracted lignin is assumed to either 

be valued as wood fuel or as oil. The different options are evaluated and compared with respect to 

annual net profit and global CO2 emissions for four different future energy market scenarios.The 

results show that, generally, BLG with motor fuels production and lignin extraction with lignin valued as 

oil achieve the best economic performance. Upgrading the recovery boiler and capture and store CO2 

from the boiler flue gases gives the highest CO2 emissions reduction but is only an economically 

attractive option in scenarios with a high CO2 emissions charge. 

Keywords: 

Kraft pulp mill, Biorefinery, Black liquor gasification, Lignin extraction, Carbon capture and 

storage, Energy market scenarios. 


With increased concern for climate change and increasing energy prices, the need for energy 

efficiency measures is high on the agenda for both industrial and political decision makers. 

The industry sector, which stands for almost 30% of the energy use in Europe [1], is a key 

player in the transition towards a more sustainable European energy system. The pulp and 

paper industry is the sixth largest industrial energy user in Europe, using approximately 

121 TWh of electricity and 365 TWh of thermal energy during 2007 [2]. Due to its high use 

of biomass and possibility to achieve an energy surplus, the kraft pulp industry has the 

potential to become a major contributor in reducing global CO2 emissions through increased 

delivery of energy products[3-10]. 

The potential for reduced process steam demand in the kraft pulp industry through improved 

process integration and by installing new efficient equipment has in previous research shown 

to be up to approximately 30-35%[11-14]. For most market kraft pulp mills implementation 

of such measures would lead to a significant steam surplus that can be utilized in different 

270

ways. It could for example be used to cover the heat demand of production processes for 

additional products (such as electricity production or capture of CO2)or it permits the 

introduction of processes which reduces the steam production whilst producing new products 

(such as extraction of lignin or black liquor gasification where the product gas is used to 

produce motor fuels or electricity). Throughout this paper these two options are referred to as 

“utilisation of a potential steam surplus”.In a previous study by the authors [18], it was shown 

that making investments in steam saving measures and thereby enabling production of 

additional products generally is very profitable and contributes to reduction of global CO2 

emissions. 

The trend in the pulp and paper industry is toward fewer mills with larger capacity. This 

means that some mills will be closed down, while the remaining mills will increase their 

production capacity [15]. For increased production capacity in pulp mills, the recovery boiler 

is often a bottleneck. In a study by Axelsson et al.[16], lignin extraction was found to be an 

economically attractive alternative for debottlenecking the recovery boiler, in comparison to 

upgrading the recovery boiler (and steam turbines) at a kraft pulp mill. The economic 

performance of lignin extraction compared to increased electricity generationfor utilization of 

a potential steam surplus is better for mills investigating investment options in connection 

with increased production capacity [8, 16]. Another approach to achieve debottlenecking of 

the recovery boiler is to introduce a black liquor gasifier as a booster. This approach has been 

investigated by for example Berglin and Andersson[17], who concluded that a black liquor 

gasifier using the product gas for steam generation yields a better economic return than 

investing in a new recovery boiler (it was assumed that the existing recovery boiler could not 

be rebuilt). 

2. Objective 

This paper compares three different options for debottlenecking the recovery boiler and 

utilizing a potential steam surplus at a typical Scandinavian kraft pulp mill when increasing 

the production capacity by 25%: 

1. Upgrading the recovery boiler 

2. Lignin extraction 

3. Black liquor gasification (as a booster) 

For black liquor gasification two options for the product gas are considered; production of 

electricity (black liquor gasification combined cycle, BLGCC) or DME motor fuel (black 

liquor gasification with motor fuel production, BLGMF). Furthermore, both black liquor 

gasification and upgrading of the recovery boiler are assumed to be possible to combine with 

carbon capture and storage (CCS),where excess steam is used to cover the heat demand of the 

capture process. The extracted lignin is assumed to either be valued as wood fuel or as oil. 

The different options are evaluated and compared with respect to annual net profit and global 

CO2 emissions for four different future energy market scenarios. A further analysis of how 

different parameters such as policy instruments and investment costs affect the different 

technologies is also included. The results are compared with the results from the previous 

study by the authors where no production increase was considered; see [18]. 


This work follows a methodology previously developed and described by one of the authors 

[6, 9]. The methodology enables changes in a studied energy system to be analyzed in a 

systematic way. The main steps in this methodology are briefly described below: 

271

1. Define the studied system (in this case the kraft pulp mill). 

2. Define possible system changes to be evaluated (in this case the investments in 

energy efficiency measures together with the different options for debottlenecking 

the recovery boiler). 

3. Define the surrounding system used for evaluation of the studied system (in this case 

the different energy market scenarios describing energy market prices, policy 

instruments and associated CO2 emissions for marginal use of energy carriers). 

4. Construct a model for simulation of energy flows in the overall system defined in 

steps 1-3 (in this case the simulation model is constructed using the energy systems 

modelling tool reMIND). 

5. Define the performance indicator(s) to be used for evaluating the system changes 

considered (in this case the net annual profit and global CO2 emissions). 

6. Use the simulation model to optimize the system based on the selected performance 

indicator for given settings in the surrounding system (defined by the aim of the 

study in question). 

7. Vary key settings in the surrounding system to see how/whether the optimal solution 

is affected by these changes (in this study we calculate the optimal solution for a 

number of possible energy market conditions defined by energy market scenarios). 

8. Fix certain parameters in order to investigate how close other solutions of interest are 

to the optimal solution (in this case solutions for all of the studied energy related 

technologies are obtained together with a sensitivity analysis showing the effect of 

changes in different system parameters) 

9. Analyze the results in relation to the aim of the study (as defined by steps 1-3). 

For constructing the model of the studied system and the surrounding system (described in 

step 3 above) the energy systems modelling tool reMIND, based on mixed-integer linear 

programming, is used. The reMIND tool has previously been used and described by e.g. 

Karlsson [19]. With this tool a simulation model of an energy system can be specified using a 

graphical interfaceand pre-defined equations. The constructed model can then be used for 

optimization purposes. In this study the objective is to minimize the total annual system cost 

of the studied energy system (the mill), assuming given conditions in the surrounding system 

(the energy market, including policy instruments) and the objective function can be defined as 

follows: 

min Z = rI tot - B tot + C tot (1) 

wherer = Capital recovery factor (0.2) 

I tot = Total investment cost (energy efficiency measures and technologiesfor utilizing the steam surplus and 

debottlenecking the recovery boiler) 

B tot = Revenue of sold energy products including policy instruments (electricity, district heating, bark, 

captured CO 2, biofuels etc.) 

C tot = Running costs (electricity, chemicals etc.) 

For the system studied in this paper the different parameters, that is the investment costs, 

running costs and revenues for the different technologies, are described in Section 4. 

Asdescribed above, reMIND is constructed for minimization of the annual system cost. The 

different investment options studied in this paper are profitable when the annual system cost, 

Z, is negative. Hereafter the annual system cost is therefore referred to as the system’s annual 

net profit. 

272

4. The studied system and input data 

The studied system consists of a kraft pulp that is planning to increase its production capacity 

and has the possibility to invest in energy efficiency measures that reduce the mill steam 

demand. The mill is further described in Section 4.1. The studied system is connected to a 

surrounding system in which the imported and exported energy and material streams are 

priced and the CO2 emissions associated with the imported and exported energy products are 

calculated. Section 4.2 presents the data for the surrounding system (energy market 

scenarios). Fig. 1 shows a schematic representation of the studied energy system and the 

surrounding system. 

Fig. 1.A schematic representation of the studied system and surrounding system. Solid lines 

represent flows that are relevant for all studied cases, whereas dotted lines represent possible 

flows. 

4.1. The studied system 

The studied mill is a model mill, developed within the national Swedish research programme 

“Future Resource Adapted pulp Mill” (FRAM), representing an average Scandinavian market 

kraft pulp mill[20]. Table 1 presents an overview of key mill data. Axelsson et al. [11]have 

shown that the mill can achieve a steam surplus through investments in process integration 

and new efficient equipment, hereafter denoted energy efficiency measures. The energy 

efficiency measures in which the studied mill can invest in order to achieve the steam surplus 

are described in e.g. [6, 18]. The steam savings of MP and LP steam are approximately 

0-20 % and 35-50 % respectively (depending on which energy efficiency measures are 

chosen) at a total cost of approximately 10-17 MEUR. 

In this study, the mill is assumed to increase its production by 25% (typical number used in 

other studies, e.g. Axelsson et al. [16]) to 1250 ADt/d. The use of steam, electricity and oil is 

assumed to increase in direct proportion to the production. It is assumed that the recovery 

boiler is the bottleneck and to handle the increased amount of black liquor, the mill has to 

invest in either a black liquor gasifier (connected to a DME plant or a gas turbine combined 

273

cycle) or a lignin extraction plant, or upgrade the recovery boiler.Fig. 2 shows the main 

energy and material streams for the mill with a 25% increased pulp production. 

Table 1.Key mill data. 

25% production 

increase 

Kraft pulp production, design [ADt/d] 1000 1250 

Process thermal energy use b [GJ/ADt] 14.3 17.9 

Steam use MP/LP [t/h] 69/190 86/238 

Electricity use/production [MW] 33/24 41/ c 

Oil use in lime kiln [MW] 22 28 d 

Biomass surplus (bark sold) [MW] 31 39 

a Earlier work by the authors compare different energy-related technologies for utilisation of excess steam and 

heat for the mill in its original design, not considering any production increase, see Jönsson et al. [18]. 

b Excluding steam conversion to electricity in the back-pressure turbine. 

c Depends on which option that is used for debottlenecking the recovery boiler. 

d In case of black liquor gasification, the oil usage increases. 

274 

Original design a 

Fig. 2. A schematic representation of the main energy and material streams for the mill with a 

25% increased pulp production. Solid lines represent flows that are relevant for all studied 

cases, whereas dotted lines represent possible flows. 

Table 2 presents the different possible outcomes studied (cases). The option to invest in a new 

mini-recovery boiler has not been included, because it is always more expensive than the 

option to rebuild the recovery boiler (which gives the same result, increased capacity for 

burning the black liquor). In theory, all of the considered investment alternatives could be 

adopted at the same time. In practice, however, this will not be the case since the investment 

costs for the technologies are scale-dependent (see Table 3). Thus, the optimal solution will 

most likely consist of investments in energy efficiency measures together with only one of the 

investment alternatives for debottlenecking the recovery boiler.

Table 2. Presentation of the different possible outcomes (cases) including case code, case 

description and references. 

Case code Case description Key data 

from 

Black liquor gasification 

Lignin extraction 

Recovery boiler 

upgrade 

BLGMF 70% of the BL to the BLGMF/DME plant. [21] 

BLGMF:CCS 70% of the BL to the BLGMF/DME plant. The 

removed CO2 (part of the BLGMF process) is 

compressed and sent for storage. If profitable, a part 

of the CO2 from the recovery boiler (and bark boiler) 

flue gases can also be captured by absorption. 

BLGCC 70% of the BL to the BLGCC plant. Possible 

investment in a new back-pressure steam turbine 

and/or a condensing steam turbine. 

BLGCC:CCS 70% of the BL to the BLGCC plant. Possible 

investment in a new back-pressure steam turbine 

and/or a condensing steam turbine. The BLGCC 

plant is modified to include CO2 capture (different 

gas cleaning including a water gas shift reactor). If 

profitable, a part of the CO2 from the recovery boiler 

(and bark boiler) flue gases can also be captured by 

absorption. 

Lignin:Wood 

fuel 

Lignin is extracted from the black liquor, which leads 

to a decrease of the energy content in the BL to the 

RB (max 53%). The lignin is sold as fuel priced as 

wood chips. 

Lignin:Oil Lignin is extracted from the black liquor, which leads 

to a decrease of the energy content in the BL to the 

RB (max 53%). The lignin is sold as fuel or 

feedstock priced as fuel oil. 

RBU:Electricity The RB is upgraded so it can handle the total amount 

of BL, 125%. Investment in a new back-pressure 

steam turbine and a condensing steam turbine. 

RBU:CCS The RB is upgraded so it can handle the total amount 

of BL, 125%. Investment in a CO2 capture plant 

connected to the recovery boiler flue gases and a new 

back-pressure steam turbine. 

275 

[21], own 

calculations 

[22, 23] 

[21-23], 

own 

calculations 

For all cases, there is also a possibility to invest in heat exchangers and/or heat pumps for 

district heating production (DH). It is assumed to be possible to deliver between 10 and 

50 MW of district heating depending on season 1 . The falling bark can be fired in an existing 

bark boiler and/or be sold. The possibility to import external wood fuel is not considered; the 

steam use has to be met by internal resources. Thus, this study shows what can be done at a 

kraft pulp mill without importing external biomass, something that could be of interest in the 

future when the supply of biomass may be low due to its limited availability. 

For technical reasons, the minimum load of the recovery boiler is set to 55% of the maximum 

load. For both of the cases with black liquor gasification, the mill’s steam balance is not the 

1 The district heating demand that can be supplied with mill excess heat varies according to a duration heat 

load curve with a top load of 50 MW. 

[8] 

[8] 

[16] 

[24]

limiting factor for the maximum size of the plants – this factor is the minimum load of the 

recovery boiler (in the BLGMF case, however, the maximum size set by the minimum load of 

the recovery boiler is close to the maximum size set by the steam balance). However, for 

lignin extraction the steam balance limits the size of the plant. Black liquor gasification leads 

to an increased consumption of fuel oil in the lime kiln 2 . 

Extracted lignin can be used in the lime kiln to replace fossil fuel oil, or be sold as a wood 

fuel or oil replacement (both for replacement of oil as a fuel and for replacement of oil as a 

feedstock in production of materials and chemicals). Therefore, two different lignin cases are 

considered: one where lignin is valued as wood chips and one where it is valued as fuel oil. 

The captured (green) CO2 is compressed and delivered to a storage location (for costs, see 

Table 3). It is assumed to generate an income corresponding to the charge for emitting fossil 

CO2. Table 3presents the cost related to the different investment options.The investment cost 

for the not yet commercial technologies, that is black liquor gasification, lignin extraction and 

CO2 capture and storage, are assumed to be for the “N th plant”. Energy-related operating costs 

have been omitted in the table, since they vary depending on energy market scenario. The 

capitalrecovery factor is set to 0.2 3 . 

Table 3. Investment and operating costs for different units needed for implementation of the 

different cases studied. 

Investment and operating costs a Based 

on 

BLGMF/BLGMF with CO2separation b 

7055BL 0,6 M€, 9.7 k€/yr/BL [21] 

BLGCC b,c 5952BL 0,6 M€, 1.0 k€/yr/BL [22] 

BLGCC b,c with CO2 separation 6365BL 0,6 M€, 1.0 k€/yr/BL [21, 22] 

Lignin extraction plant d 

7.2LR 0,6 M€, 5.8 €/MWh [8] 

Recovery boiler upgrade including cost for 

lost production 

35.4 M€ [16] 

Back-pressure steam turbine e 

1.3P 0,6 M€ [8] 

Condensing steam turbine e 

2.4P 0,6 M€ 

f, g 

CO2 separation plant for RB flue gases 2.3CO2 0.7 f, h 

CO2compressor 

, 4% of investment cost 

1.1P 

[24] 

0.7 , 4% of investment cost 

Transportation and storage of CO2 8 €/tonne 

Heat pump for DH i 

0.11Q M€ [25] 

Heat exchanger steam-DH i 0.68+0.033Q M€ [6] 

Heat exchanger heat-DH i 0.059+0.042Q M€ 

a 

All values in 2008 money value. All investment costs have been recalculated to 2008 money value using 

Chemical Engineering’s Plant Cost Index (CEPCI). 

b 

BL refers to the flow of black liquor in MW. 

c 

Excluding the steam turbine/s. 

d 

LR refers to the lignin extraction rate in kg/s. 

e 

P refers to the power output in MW. 

f 

Operating cost CO2 absorber and CO2 compressor: 4% of investment cost. 

g 

CO2 refers to the CO2 capture rate in kg/s. 

h 

P is the compressor electricity demand. 

i 

Q refers to the heat supplied by the heat pump or heat exchanged in the heat exchanger in MW. 

2 Due to a different composition of the green liquor, the load of the lime kiln increases. 

3 A capital recovery factor of 0.2 is equivalent to e.g. an economic lifetime of 10 years and an interest rate of 

15% or an economic lifetime of 6 years and an interest rate of 5%. 

276

4.2. Energy market scenarios 

The future economic performance, as well as the global emissions of CO2, associated with the 

different cases studied is dependent on the development of the energy market. Consequently, 

to identify robust investment options, their performance should be evaluated for varying 

future energy market conditions. Here, energy market scenarios are used to reflect a variety of 

possible future energy market conditions. To achieve reliable results from an evaluation using 

energy market scenarios, the energy market parameters within a given scenario must be 

consistent, i.e. the energy prices must be related to each other (i.e. accounting for energy 

conversion technology characteristics and applying suitable substitution principles). 

Consequently, a systematic approach for constructing such consistent scenarios is facilitated 

by the use of a suitable calculation tool. In this work the Energy Price and Carbon Balance 

Scenarios tool (the ENPAC tool) developed by Axelsson and Harvey [26] was used. The 

scenarios reflect different future energy market conditions for the “average” years 2020 and 

2030 and are based on two fossil fuel price levels (low and high) and two CO2 emission 

charge levels (low and high).The big difference between the two time periods is that for the 

period with 2020 as its average year, it is assumed that infrastructure for CCS is not 

established, and therefore the options for CCS are excluded. Tables 4 and 5 present scenario 

data used. A further description of the ENPAC tool and the scenarios used in this work can be 

found in Jönsson et al. [18]. 

Table 4. Key data for the four energy market scenarios used for 2020. 

Scenario input data 1 2 3 4 

Fossil fuel price level a Low Low High High 

CO2 charge level Low High Low High 

CO2 charge [€/tonne CO2] 26 67 26 67 

Green electricity policy instrument [€/MWh] 26 26 26 26 

Resulting prices and values of policy instruments [€/MWh] 

Electricity 63 85 66 95 

DME 47 58 70 81 

Bark/by-products/wood chips b 22 37 23 38 

Heavy fuel oil (incl. CO2) 37 49 53 65 

District heating 15 28 18 28 

Biofuel policy instrument 49 61 28 41 

Resulting marginal/alternative technologies and their CO2 emissions [kg/MWh] 

Electricity 

722 345 722 722 

marginal production of electricity 

CP NGCC CP CP 

Biomass 

225 237 225 225 

marginal user of biomass 

CP/DME CP/DME CP/DME CP/DME 

District heating production 

224 380 224 224 

alternative heat supply technology CCHP/GB CCHP/GB 

Transportation 

273 273 

alternative transportation technology Diesel Diesel 

a 

Oil prices: Low: 62 USD/barrel, High: 100 USD/barrel. 

b 

In the past years the prices of wood by-products and chips have been very similar. 

277 

CCHP/GB 

273 

Diesel 

CCHP/GB 

273 

Diesel

Table 5. Key data for the four energy market scenarios used for 2030. 

Scenario input data 1 2 3 4 

Fossil fuel price level a Low Low High High 

CO2 charge level Low High Low High 

CO2 charge [€/tonne CO2] 35 109 35 109 

Green electricity policy instrument [€/MWh] 26 26 26 26 

Resulting prices and values of policy instruments [€/MWh] 

Electricity 68 90 74 98 

DME 57 77 88 109 

Bark/by-products/wood chips b 27 52 30 56 

Heavy fuel oil (incl. CO2) 45 67 67 89 

District heating 19 49 27 56 

Biofuel policy instrument 46 67 20 41 

Resulting marginal/alternative technologies and their CO2 emissions [kg/MWh] 

Electricity 

679 129 679 129 

marginal production of electricity 

CP CP CCS CP CP CCS 

Biomass 

227 244 227 244 

marginal user of biomass 

CP/DME CP/DME CP/DME CP/DME 

District heating production 

242 468 242 468 

alternative heat supply technology CCHP/GB CCHP/GB 

Transportation 

273 273 

alternative transportation technology Diesel Diesel 

a 

Oil prices: Low: 74 USD/barrel, High: 126 USD/barrel. 

b 

In the past years the prices of wood by-products and chips have been very similar. 

4.3. Sensitivity analysis 

Table 6 presents the parameters included in the sensitivity analysis. 

Table 6. Parameters included in the sensitivity analysis. 

Parameter Denotation Cases 

District heating 

production 

278 

CCHP/GB 

273 

Diesel 

CCHP/GB 

273 

Diesel 

a 1. The amount of district heating possible to deliver is 

increased by 100%, to between 20-100 MW 

depending on season. 

2. No possibility to deliver district heat. 

Capital recovery factor b The capital recovery factor is changed from 0.2 to 0.1. 

Investment costs for 

non-commercial 

technologies 

Green electricity 

policy instrument 

support level 

Biofuel policy 

instrument support 

level 

Recovery boiler 

upgrade cost 

c The investment costs for the non-commercial 

technologies, i.e. BLG, lignin extraction and CO2 

capture and storage, are increased by 25%. 

d 1. The support is increased by 50%. 

2. The support is decreased by 50%. 

3. No support is considered. 

e 1. The support is decreased by 50%. 

2. No support is considered. 

f The investment cost for upgrading the recovery boiler 

is lowered from 35.4 to 27.4 M€.

By using the described energy market scenarios, a sensitivity analysis regarding energy and 

CO2 prices is automatically carried out. It is, however, difficult to include all parameters 

affecting the results, for example the level of different policy instruments in the scenarios and 

the effect of the estimate of the investment cost. Therefore an extended sensitivity analysis is 

performed in addition to that inherent in using the scenarios for some key parameters to 

identify their impact on investment decision variables. The variation of the policy instruments 

are performed without considering the relationship between these parameters, and other 

parameters, within the scenarios. 


Figs 3 and 4 present a comparison of the different studied options for debottlenecking the 

recovery boiler and utilise pulp mill excess steam and heat for the energy market of 2020 and 

2030. Each technology option is represented by a shaded area where the larger focal point 

show the original solution gained for the technology and the smaller points show the solutions 

gained from the sensitivity analysis. Since non-energy related costs and revenues (e.g. raw 

material costs and sales of pulp) are omitted in this study no information can be found in the 

absolute values; it is only the comparisons between the different cases that are 

interesting.RBU:Electricity is used as a baseline for the comparison, represented by the 

intersection by the x- and y-axises. Consequently, solutions positioned in the lower right 

quadrant both have a better economic performance and are associated with lower global CO2 

emissions thanRBU:Electricity. 

279

Global CO 2 emissions [ktonnes/yr] 

25% increased pulp production in 2020 (no CCS) 

Scenario 1 (Low/Low) Scenario 2 (High/Low) 

Annual net profit [kEuro/yr] 

e2 

100 

-30000 

d3 d2 

0 

c 

-10000 

a2 

c 

f 

b 

d1 b 

a1 10000 30000 50000 


d3 

c 

d2 

a2 

a1 

e1 

c b 

-100 

d1 

c 

b 

a1 

a2 

b 

280 

Global CO 2 emissions [ktonnes/y r] 


a2 

100 

a2 

-30000 

e2 

c 

d3 

0 f 

d1 

d3 d2 d2 b b 

-10000 c 10000 

c 

d1 

e1 b c 

b 

30000 50000 

b 

a1 

-100 

Scenario 3 (Low/High) Scenario 4 (High/High) 


e2 

100 

c 

-100 

a2 

b 

e1 c 

a2 

a1 

b 

Global CO 2 emissions [ktonnes/y r] 

-30000 

d2 c 

0 

d3 

-10000 

a2 

d3 d2 

d1 

f d1 

a1 10000 

b 

c 

b 

30000 50000 -30000 

d3 d2 

0 

c 

-10000 

a2 

d3 

d2 

f 

d1 b 

10000 

a1 

c 

b 

30000 50000 

c 

d1 

a1 

b 

c 

a1 

b 

100 

-100 

a1 


Fig. 3. Results for the different studied cases and the sensitivity analysis for the average year 

of 2020.For each studied case, the larger centre point represents the optimal solution for that 

case given the energy market scenario prices. The smaller points show how the optimal 

solution shifts when changing certain parameters in the sensitivity analysis. The shaded areas 

show the span between the solutions given in the sensitivity analysis.RBU:Electricity is used 

as a baseline for the comparison, represented by the intersection by the x- and y-axises. 

e2 

c 

a2 

b 

e1 

c 

a2 

a2 

a1 

a1 

b



a2 

f d1 

d3 -600 

d2 

a1 

c 

-800 

25% increased pulp production in 2030 

Scenario 1 (Low/Low) Scenario 2 (High/Low) 


200 

no CCS 

c a2 b 

c 

d3 0 

b 

no CCS d2a1d1 

b a2 

-20000 10000 

e2 

e1 c 

a1 

-200 

40000 

b 

70000 

a2 

d2 

d1 

d3 c 

a1 

-400 

b 

a2 

d1f 

d3 -600 

d2 a1 

-800 

b 

281 



200 

a2 

d3 d2 d1 b 

0 a2 

no CCS a2 b 

-20000 10000 

a1 

c b 

c 

-200 

a1 

b 

a1 

40000 

noCCS 

70000 

a2 

-400 

-600 

-800 

e2 

a2 

e1 

a2 

c 

d3 

d1 

d2 c f 

d3,2 d1 

a1 

a1 

Scenario 3 (Low/High) Scenario 4 (High/High) 


200 

-20000 

no CCS 

c a2 

b 

b 

d1 b c 

d3 0 

d2 

a1 a2 

10000 

no CCS e2 

e1 c 

a1 

-200 

b 

40000 

b 

70000 

a2 

d3 d2 d1 

c a1 

-400 

b 

b 



200 

a2 

-400 

-600 

-800 

c 

b 

b 

d3 

d2 

0 

noCCS 

-20000 

-200 

d1 

b 

a2 

10000 

a1 

c b 

a1 

a2 

c 

40000 

noCCS 

b 

a1 

a2 

70000 

e2 

a2 

e1 

a2 

d3 

d1 

c 

d1 

a1 

f 

d2 

c 

d3,2 

a1 

BLGMF:CCS Lignin:Wood fuel RBU:Electricity 

BLGCC:CCS Lignin:Oil RBU:CCS 

Fig.4. Results for the different studied cases and the sensitivity analysis for the average year 

of 2030. For each studied case, the larger centre point represents the optimal solution for that 

case given the energy market scenario prices. The smaller points show how the optimal 

solution shifts when changing certain parameters in the sensitivity analysis. The shaded areas 

show the span between the solutions given in the sensitivity analysis.RBU:Electricity is used 

as a baseline for the comparison, represented by the intersection by the x- and y-axises. 

5.1. General characteristics for, and comparison of, the studied 

alternatives (cases) 

Using a BLG booster as a means of debottlenecking the recovery boiler yields very different 

results for both economic performance and global CO2 emissions depending on whether the 

product gas is used for electricity or motor fuel production. The BLGMF case shows the best 

economic performance of the two cases but is very sensitive to changes of the level of support 

c 

b 

b 

a1 

a1 

b 

b

for biofuels and the investment cost. Fairly large amounts of CO2 can, however,be separated 

at a low cost. This has a positive effect on both the global CO2 emissions and the economic 

performance for BLGMF 2030 (where it is assumed that an infrastructure for CCS is 

established). The BLGCC case is considerably less profitable compared to BLGMF, but also 

less profitable than RBU:Electricity due to a large investment cost in relation to the additional 

amount of electricity produced. BLGCC shows the greatest global CO2 emission reduction 

potential in all scenarios for the year 2020. However, for the year 2030 RBU:CCS shows the 

greatest CO2 emission reduction potential in all scenarios. 

For the cases of lignin extraction as a means to debottleneck the recovery boiler, lignin valued 

as wood fuel shows a poor economic performance with the exception of Scenarios 2 and 4 

(with high wood fuel prices). Lignin valued as oil, however, has a very good economic 

performance, even in the scenarios with a low oil price (Scenarios 1 and 2). Furthermore, it is 

not highly influenced by any of the parameters studied outside the scenarios, and can 

therefore be said to be a fairly robust investment. As stated, lignin can replace oil both as a 

fuel and as a feedstock for production of material or chemicals. If, for example, lignin is 

upgraded to a material at the mill, it is of course possible to get an evenhigher price than the 

oil price depending on the type of material. However, the cost and energy demand for 

upgrading will then also have to be taken into consideration. 

Investing in a recovery boiler upgrade and new turbines gives quite similar results for both 

global CO2 emissions and net annual profit in the different scenarios. The possibility to 

capture CO2 from the boiler flue gases yields a large CO2 reduction potential. However, the 

profitability of capturing the CO2 is strongly dependent on the CO2 charge – e.g. it is only for 

the scenarios having the highest CO2 charge, Scenarios 2 and 4, that RBU:CCS is more 

profitable than RBU:Electricity. 

5.2. Levels of support for green electricity and biofuels 

From the results it can be concluded that the level of the green electricity support does not 

significantly affect the economic performance of RBU:Electricity. For BLGCC the effect is 

more pronounced, but is small compared to the effects of other parameter changes such as the 

level of annuity factor. The relatively small effects are partly due to the assumed design of the 

green electricity support system where only new production capacity is entitled to support. 

The support for biofuels varies between the scenarios and is set at a level so that a stand-alone 

biofuel production plant has the same willingness to pay for wood fuel as a coal power plant, 

which is assumed to be a large volume price-setting user of wood fuel. This results in very 

substantial levels of support for biofuels in Scenarios 1 and 2, which of course is questionable. 

The resulting CO2 emission reduction in relation to the cost for society is relatively low, and 

the money might be better used elsewhere. The sensitivity analysis (point e1) shows for 

example the consequences of a 50% reduction of the level of support, which in Scenarios 1 

and 2 results in a more reasonable level of the support. At this level, BLGMF is only the most 

profitable option in Scenario 4 2030. 

5.4. CCS 

For 2030 CCS was assumed to be commercially available. The possibility to capture CO2 both 

from the recovery boiler flue gases and, if in operation, from the bark boiler flue gases gives a 

large CO2 reduction potential. However, the profitability of capturing the CO2 is strongly 

dependent on the CO2 charge; e.g. it is only for the scenarios having the highest CO2 charge, 

Scenarios 2 and 4, that RBU:CCS is more profitable than RBU:Electricity. If CCS is 

available, this improves the global CO2 emissions effect and the economic performance for 

BLGMF both in absolute terms and in relation to the other technologies. For Scenarios 2 and 

282

4 the CO2 emissions effect for BLGMF is further improved due to the occurrence of CCS on 

the margin in the power sector. If CCS is available, BLGCC gives almost as large CO2 

emission reductions as RBU:CCS, however, due to a higher capital intensity it always shows 

a poorer economic performance. 

5.5. Comparison with results for unchanged production capacity 

Here, the results of this paper are compared with the results from a previous study of the same 

mill where the production capacity remained unchanged (presented in [18]). Comparing the 

results from this paper with those previous results it becomes clear that the BLG cases, 

BLGMF and BLGCC (both with and without CCS), benefit from economy of scale. Thus, 

they have a better economic performance when increasing the production capacity than when 

the production capacity remains unchanged. For the BLGMF case, this means that if the 

production capacity is increased, BLGMF becomes more profitable than Lignin:Oil for some 

of the scenarios where lignin prices as oil were more profitable in case of unchanged 

production. For the RBU:Electricity and RBU:CCS cases, the production capacity increase 

affects the economic performance in a negative way. This is due to the fact that an additional 

investment in an upgrading of the recovery boiler has to be made. Thus, a production capacity 

increase benefits lignin extraction and BLG since these technologies unload the existing 

recovery boiler and thereby an investment in an upgrade of the recovery boiler can be 

avoided. 


This paper compares different technologies for utilisation of a potential steam surplus and 

debottlenecking the recovery boiler of a typical Scandinavian kraft pulp mill assuming a pulp 

production increase of 25%. The technologies are compared with respect to annual net profit 

and global CO2 emissions for four different energy market scenarios using two time frames, 

2020 (where CCS is not available, neither for the mill nor in the surrounding power system) 

and 2030. Based on the results and discussion the following main conclusions can be drawn: 

For all energy market scenarios, both year 2020 and 2030, BLGMF and lignin 

extraction where the lignin is priced as oil have a better economic performance than 

upgrading the recovery boiler (and existing steam turbines). 

The BLGMF case generally has the best economic performance, but is contrary to 

lignin extraction very sensitive to changes of several parameters, especially the level 

of support for biofuels and the investment cost. 

Extraction of lignin that can be priced as oil has a very good economic performance 

and it is not highly influenced by any of the parameters studied outside the scenarios 

and can therefore be said to be a fairly robust investment. The CO2 emissions 

reduction from lignin extraction is also fairly stable between the scenarios. 

For the year 2020, where there are assumed to be no possibilities for CCS, BLGCC 

generally gives the highest CO2reduction potential. For the year 2030, where there is 

assumed to be an established infrastructure for CCS, upgrading the recovery boiler and 

investing in CCS coupled to the boiler flue gases render the highest CO2 reduction 

potential, followed by BLGCC and BLGMF, where CCS also can be included. 

283

The possibility to capture CO2 from the recovery boiler flue gases gives a large CO2 

reduction potential. However, the profitability of capturing the CO2 is strongly 

dependent on the CO2 charge – e.g. it is only for the scenarios with a high CO2 charge 

that CCS coupled to the boiler flue gases is more profitable than investments in new 

turbines (in connection to upgrading the recovery boiler). CCS decreases the global 

CO2 emissions and increases the economic performance for BLGMF and BLGCC 

both in absolute terms and in relation to the other technologies. 

BLGMF and BLGCC benefit from economy of scale and thus have a better economic 

performance when increasing the production capacity than when the production 

capacity remains unchanged. For the BLGMF case this means that for some of the 

scenarios with increased production BLGMF becomes more profitable than extracting 

lignin that can be priced as oil, which was the most profitable choice if the production 

was unchanged. For increased electricity production and CCS in connection with the 

recovery boiler, the production capacity increase affects the economic performance in 

a negative way due to the fact that an additional investment in an upgrading of the 

recovery boiler has to be made. 

Nomenclature 

ADt Air Dried tonne 

BL Black Liquor 

BLG Black Liquor Gasification 

BLGCC Black Liquor Gasification Combined Cycle 

BLGMF Black Liquor Gasification with Motor Fuel production 

CCS Carbon Capture and Storage 

CHP Combined Heat and Power 

CCHP Coal Combined Heat and Power 

CP Coal Power 

DH District Heat 

DME Dimethyl Ether 

GB Gas Boiler 

HP High Pressure (steam) 

LP Low Pressure (steam) 

MP Medium Pressure (steam) 

NGCC Natural Gas Combined Cycle 

RB Recovery Boiler 

RBU Recovery Boiler Upgrade 


The work has been carried out under the auspices of the Energy Systems Programme, which 

is financed by the Swedish Energy Agency (SEA). The work was co-funded by the Black 

Liquor Gasification Programme, which is financed by SEA, MISTRA, Chemrec, 

Smurfit Kappa, Södra Cell, SCA Packaging, Sveaskog and the County Administrative Board 

284

of Norrbotten. The work was also co-funded by the Södra Foundation for Research, 

Development and Education. 

References 

[1] Eurostat. Energy, transport and environment indicators. Eurostat pocketbooks. 

Luxembourg, 2009. 

[2] CEPI. Key Statistics 2008 -European Pulp and Paper Industry. Brussels: Confederation of 

European Paper Industries, 2009. 

[3] Costa A, Bakhtiari B, Schuster S, Paris J. Integration of absorption heat pumps in a Kraft 

pulp process for enhanced energy efficiency. Energy. 2009;34(3):254-60. 

[4] Costa A, Paris J, Towers M, Browne T. Economics of trigeneration in a kraft pulp mill for 

enhanced energy efficiency and reduced GHG emissions. Energy. 2007;32(4):474-81. 

[5] Consonni S, Katofsky RE, Larson ED. A gasification-based biorefinery for the pulp and 

paper industry. Chemical Engineering Research and Design. 2009;87(9):1293-317. 

[6] Jönsson J, Algehed J. Pathways to a sustainable European kraft pulp industry: Trade-offs 

between economy and CO2emissions for different technologies and system solutions. Applied 

Thermal Engineering. 2010;30(16):2315-25. 

[7] Marinova M, Mateos-Espejel E, Jemaa N, Paris J. Addressing the increased energy 

demand of a Kraft mill biorefinery: The hemicellulose extraction case. Chemical Engineering 

Research and Design. 2009;87(9):1269-75. 

[8] Olsson MR, Axelsson E, Berntsson T. Exporting lignin or power from heat-integrated 

kraft pulp mills: A technoeconomic comparison using model mills. Nordic Pulp and Paper 

Researsh Journal. 2006;21(4):476-84. 

[9] Jönsson J, Svensson I-L, Berntsson T, Moshfegh B. Excess heat from kraft pulp mills: 

Trade-offs between internal and external use in the case of Sweden--Part 2: Results for future 

energy market scenarios. Energy Policy. 2008;36(11):4186-97. 

[10] Andersson E, Harvey S. System analysis of hydrogen production from gasified black 

liquor. Energy. 2006;31(15):3426-34. 

[11] Axelsson E, Olsson MR, Berntsson T. Heat Integration Opportunities in Average 

Scandinavian Kraft Pulp Mills: Pinch Analysis of Model Mills. Nordic Pulp and Paper 

Research Journal. 2006;4(21):466-74. 

[12] Laaksometsä C, Axelsson E, Berntsson T, Lundström A. Energy savings combined with 

lignin extraction for production increase: Case study at a eucalyptus mill in Portugal. Clean 

Technologies and Environmental Policy. 2009;11(1):77-82. 

[13] Wising U, Berntsson T, Stuart P. The potential for energy savings when reducing the 

water consumption in a Kraft Pulp Mill. Applied Thermal Engineering. 2005;25(7):1057-66. 

[14] Axelsson E, Olsson MR, Berntsson T. Opportunities for process-integrated evaporation 

in a hardwood pulp mill and comparison with a softwood model mill study. Applied Thermal 

Engineering. 2008;28(16):2100-7. 

[15] CEPI. Annual Statistics 2007 -European Pulp and Paper Industry. Brussels: 

Confederation of European Paper Industries, 2008. 

[16] Axelsson E, Olsson MR, Berntsson T. Increased capacity in kraft pulp mills: Lignin 

separation and reduced steam demand compared with recovery boiler upgrade. Nordic Pulp 

and Paper Research Journal. 2006;21(4):485-92. 

285

[17] Berglin N, Andersson L. Process Integration of Black Liquor Gasification for 

Incremental Recovery Capacity. Conference Process Integration of Black Liquor Gasification 

for Incremental Recovery Capacity, Whistler, B.C., Canada, June 11-14, 2001. 

[18] Jönsson J, Pettersson K, Harvey S, Berntsson T. Comparison of options for utilization of 

a potential steam surplus at kraft pulp mills – Economic performance and CO2 emissions. 

International Journal of Energy Reseach. Accepted for publication. 

[19] Karlsson M. The MIND method: A decision support for optimization of industrial energy 

systems - Principles and case studies. Applied Energy. 2011;88(3):577-89. 

[20] FRAM. FRAM Final report – Application area: Model mills and system analysis. 

FRAM report No 70, 2005. 

[21] Ekbom T, Berglin N, Lögdberg S. Black Liquor Gasification with Motor Fuel Production 

- BLGMF II Nykomb Synergetics, 2005. 

[22] KAM. Ecocyclic Pulp Mill -"KAM", Final Report 1996-2002. In: STFI, editor. KAMreport 

A100. Stockholm, 2003. 

[23] Berglin N, Lindblom M, Ekbom T. Preliminary economics of black liquor gasification 

with motor fuels production. Colloquium on Black Liquor Combustion and Gasification, Park 

City, Utah, May13-16, 2003. 

[24] Hektor E, Berntsson T. Future CO2 removal from pulp mills - Process integration 

consequences. Energy Conversion and Management. 2007;48(11):3025-33. 

[25] Bengtsson C. Novel Process Integration Opportunities in Existing Kraft Pulp Mills with 

Low Water Consumption. Göteborg: Chalmers University of Technology, 2004. 

[26] Axelsson E, Harvey S. Scenarios for assessing profitability and carbon balances of 

energy investments in industry. AGS Pathways report 2010:EU1. Göteborg, Sweden: AGS, 

The alliance for global sustainability. Pathways to sustainable european energy systems, 2010. 

286

Abstract: 




Demonstrating an integral approach for industrial 

energy saving 

René Cornelissen a , Geert van Rens b , Jos Sentjens c , Henk Akse d , Ton Backx e , Arjan 

van der Weiden f , Jo Vandenbroucke g 

a CCS B.V., Deventer, The Netherlands, cornelissen@cocos.nl, 

b CCS B.V., Deventer, The Netherlands, vanrens@cocos.nl, CA 

c Jacobs consultancy, Leiden,The Netherlands, Jos.Sentjens@jacobs.com 

d Traxxys, Woerden, The Netherlands, henk.akse@traxxys.com 

e Eindhoven University of Technology, Eindhoven, The Netherlands, a.c.p.m.backx@tue.nl 

f NL Agency,Utrecht, The Netherlands, arjan.vanderweiden@agentschapnl.nl 

g Nyrstar, Budel, The Netherlands, Jo.Vandenbroucke@nyrstar.com 

The reduction of energy consumption in industry is getting increasingly more difficult. In this article an 

integral approach is used to perform an industrial energy saving study at a Zinc manufacturing plant. The 

approach is a combination of exergy analysis, pinch analysis, process intensification and control 

engineering. It was found that exergy analysis at the level of process functions can act as a focal point for 

more detailed studies, like process intensification, control engineering and exergy itself. Optimising on 

control engineering as a part of an energy saving study has the advantage of tackling process control issues, 

while saving energy. It was found that the structured method of the integral approach ensures a broad range 

of solutions for both the short term and the long term, of which 12 were elaborated into simple business 

cases. 

Keywords: 

Integral approach, Energy, Exergy, Pinch, Industry. 


Natural resources are becoming increasingly scarce. Oil prices are fluctuating, but have an upward 

tendency. Furthermore, many oil reserves are located in countries which are not always politically 

stable. Income from oil can increase the political instability. Therefore, a shift away from oil is 

required. Coal could be an alternative, as coal reserves are distributed more widely over the world. 

However, consumption of coal leads to more polluting emissions, like CO2, which is believed to 

cause the greenhouse effect. 

Instead of focussing on alternative fuels to replace fossil fuels, this paper focuses on the reduction 

of energy use; to be more precise on the reduction of energy consumption in industrial processes. 

Companies with a long history of energy saving, find it increasingly difficult to come up with 

additional measures for energy saving with an attractive pay-back period. The conventional 

analyses focus on optimisation based on energy balances, however, other tools are available for 

energy optimisation as well. When these tools are used, they are often applied haphazardly, in parts 

of the process that are expected to cause the biggest losses, or expected to achieve the biggest gains. 

A systematic approach to energy saving is generally not used. 

287

Therefore NL Agency (an agency of the Dutch ministry of Economic affairs, agriculture and 

innovation) supported an approach that uses a multitude of disciplines. This may lead to new ideas, 

and more clever ways of saving energy. This approach consists of an exergy analysis, a pinch 

analysis, a process intensification scan and an analysis of the control system of the process. This 

approach was used in a number of pilot projects. The approach is illustrated using one of those pilot 

projects, being a Zinc-manufacturing plant. 


2.1. Overview of the approach 

The integral approach used aspects in the field of process engineering, systems and control 

engineering and heat integration. The project partners were selected for complimentary knowledge, 

and each partner had their own way of evaluating a process. CCS was selected for its knowledge in 

the field of exergy, Jacobs Consultancy for its experience in the field of pinch-analysis, Traxxys for 

the process intensification, and Eindhoven University of Technology for the systems and control 

engineering. To enhance the knowledge transfer, several plenary meetings were held to discuss the 

process and generate solutions for the brainstorm session. 

The project was split in five phases: 

1. The analysis of the process 

2. Generation of solutions 

3. Selection of the solutions 

4. Technical elaboration of selected solutions 

5. Making business cases of selected solutions 

The integral approach is characterised by a relatively thorough analysis phase. About half of the 

man-hours were spent in the analysis phase. Without a proper and profound evaluation of the 

process, no proper solutions are expected. In the analysis phase Nyrstar provided the project 

partners with a mass and energy balance of the plant and indicated bottlenecks within the process. 

This formed the base for the analyses on exergy and pinch. 

In this case exergy was used at a functional level of a group of devices and not at the level of every 

single device or process unit. This enabled the use of exergy as a focal point. The exergy analysis 

identified opportunities for process improvement and improvement for thermal integration. This is 

why an exergy analysis may serve as an indicator for the need for a pinch-analysis. In this study it 

was used as a focal point for the process intensification options. Both during the pinch-analysis and 

the process intensification scan (PI-scan) the process was studied in detail. Pinch-analysis is a way 

of improving heat-integration of the plant. The PI-scan looks at improvement of the process, by the 

application of new technology or by using different (more energy-friendly) production methods. 

The used method is indicated in Fig. 1. For more simple processes, the pinch-analysis, exergyanalysis 

and PI-scan could be performed in parallel. 

288

Fig. 1. Schematic overview of the Integral Approach 

2.2. Analysis phase 

2.2.1. Exergy analysis 

Exergy is known for more than 50 years. Rant was the first to introduce the term exergy [1]. The 

approach was further developed by amongst others Szargut [1,2] and Kotas [3]. Amongst others 

Tsatsaronis and Valero [4-6] took the concept one step further by focussing on thermo-economic 

costs. 

Although the concept of exergy is widely known in the academic world and the chemical industry, 

its use is rather limited outside the academic world and new chemical plants. This is in our opinion 

partly caused by the limited availability of data, which makes it difficult to make an energy balance, 

let alone one that closes, and as a result makes it difficult to perform an exergy analysis. However in 

industry, and specifically for existing plants the use of exergy may lead to new insights too. 

In this specific example it was opted to use an exergy analysis on a function level, and use exergy as 

a tool to pinpoint the process steps with the biggest losses and destruction of exergy. These process 

steps are then subsequently studied in more detail, by for example the process intensification scan. 

The analysis was performed by using the software tool OptiJoule, which is in-house developed by 

CCS. 

289

2.2.2. Pinch analysis 

Pinch analysis is a method to find the ideal way of integrating excess heat and heat demand. For 

more information on the pinch analysis method, please refer to literature, for example [7]. It has 

been used in many industrial plants, see e.g. [8,9] as examples of recent publications. The analysis 

consists of the following stages: 

Data extraction, including evaluation of the heat and mass balances 

Creating the ‘Composite Curves’ and ‘Grand Composite Curves’, resulting in design targets. 

This gives insight in the present heat situation and the possibilities for improvement. 

Analysis of the heat exchangers network and identification of new heat integration 

measures. 

2.2.3. Process intensification scan 

The process intensification scan addresses perceived bottlenecks within the process, and attempts to 

solve them using process intensification options. The performed PI-scan is a 3-phase entity, which 

is described in more detail in [10]. The first phase is a quick scan which identifies the current 

bottlenecks in the process and delivers a list with possible spots where PI may solve the bottleneck. 

The second phase is a more detailed study, which develops into business cases. The third phase is 

the actual contracting phase. In the Netherlands 42 of these quick scans (i.e. phase 1) have been 

performed up to 2011 [10]. In the present approach loss of exergy is added as a bottleneck to phase 

1. The result of the process intensification scan should be a substantial saving of the fuel and 

feedstock use. For new installations or refurbishment of an existing plant, the investment cost may 

be reduced significantly as well by using PI. By using process intensification, options can be 

discovered that require a complete redesign of the installation or even R&D, which are more longterm 

solutions, and solutions that can be implemented immediately. 

2.2.4. Analysis of control system 

The analysis of the control system is performed to examine where in the production process fuel 

and feedstock can be saved by using more advanced control systems. Additionally, difficulties in 

the present operation of the process were taken into account. More advance control systems that 

improve both energy use and process operation were preferred. In the analysis the focus was on fast 

control systems that require a limited amount of calculation power and on the development of 

estimation routines for control systems if a limited amount of measured data was available. 

2.3. Solution and Selection phase 

In the solution phase, process improvements were generated during a brainstorm session. During 

this brainstorm session employees of Nyrstar participated as well. Ideally this brainstorm session 

takes place after the pinch-analysis and the PI-scan and after the detailed analysis of the control 

system. In this case the approach slightly deviated from the ideal approach in order to reduce the 

number of plenary meetings. The additional options for process improvement were added later to 

the list of process improvements. Other ideas of participants that came up after the brainstorm 

session were added to the list as well. 

The brainstorm session led to a rather long list of ideas for process improvement. The list contained 

132 ideas in total. Rating all ideas on selection criteria would have been a tedious job, as all 132 

ideas need to be discussed. Therefore the long-list was reduced. To reduce the long list every 

technical expert including the technical experts of Nyrstar was asked to list their 7 favourite ideas 

independently, based on their experience and the strategy of Nyrstar. The number of 7 was chosen 

to ensure that the list was on one hand not too long, and on the other hand sufficiently long, to still 

include less likely candidates, that might prove to be interesting. The mix of backgrounds ensured 

290

sufficient diversity of the ideas in terms of disciplines and also meant that the less likely ideas could 

be filtered out before rating all the ideas. 

Given the 7 technical experts, the longest possible list was 49 options. Because some solutions were 

selected twice or more, this resulted in a shorter list of 39 ideas. These ideas were rated on a number 

of ranking criteria. 

The chosen ranking criteria were: 

magnitude of the (expected) energy saving 

exergy level of the energy saving 

a qualitative estimate of the required investment 

the influence on the operation of the production process 

the extent of modifications needed within the process 

the timeframe in which the solution is expected to be technically feasible. 

Rating based on the ranking criteria was done by discussing the merits of the solutions within the 

entire project team. The scoring possibilities and criteria are given in Table 1. Note that the 

quantitative units need to be different for each study, as these savings are clearly not possible for a 

smaller plant. 

Table. 1. Schematic overview of the ranking criteria 

Magnitude (expected) energy saving 

2MW 

Exergy level (expected) energy saving 

0 3 6 9 

Heat (100°C) Electricity and gas 

Required investment (e stimate) 

3 6 9 

2.000 k€ 

10 7 4 1 

Influence on operation of production process 

Positive (more stable) No influence Slightly more complex Loss of robustness 

13 10 7 4 

Extent of modifications needed within the process 

Easy modifications Slight modifications Significant modifications Major modifications 

10 7 4 1 

Time horizon within which the solution can be implemented 

Direct (10 years 

10 7 4 1 

Each ranking criterion had a weight factor accompanying it. In this way the ranking criteria can be 

matched to the company’s strategy. The magnitude of the expected energy saving for example had a 

higher weight factor than the timeframe in which the solution is expected to be technically feasible, 

i.e. long-term solutions with big energy savings would be rated higher than short-term solutions 

with mediocre energy savings. The weight factors varied from 1 to 5. The score for each criterion 

was multiplied by the weight factor. All scores were added up to find the best scoring solutions. 

Based on available capacity and similarity of solutions, the twelve highest ranking ideas were 

elaborated into business case. 

291

3. Case study 

3.1. Plant and process 

The plant under consideration is the Budel Plant of Nyrstar. The main products of the plant are 

high-purity zinc (SHG-zinc), zinc alloys and sulphuric acid. A side product is Budel Leach Product; 

a product containing silver and lead, that was present in the zinc ore, which is used for the 

production of silver. The plant uses the so-called Roasting, Leaching, Electrolysis process. The 

scheme is given below. 

Fig 2. Roasting-Leaching Electrolysis Zinc production process at Budel plant. Reproduced with 

permission from [11] 

The zinc ore is roasted in an air-blown furnace, thereby converting ZnS into ZnO and SOx. The 

zinc-oxide is cooled and subsequently leached in sulphuric acid, thereby forming ZnSO4 (solute) 

and water. Other metals are removed from the liquor of H2SO4 and Zn. The main metals are 

contained in the Budel Leach Product, but also pure cadmium, and copper and cobalt containing 

cakes are removed from the process. Very pure zinc is formed on electrodes in the electrolysis 

process, where ZnSO4 (solute) and water are converted into Zn, O2 and H2SO4. 

The heat in the SOx-containing roast gas is first used to produce steam. Solids and other impurities 

are removed, and subsequently sulphuric acid is produced. Additional process units are a water 

purification plant and the melting and casting process. 

The majority of the consumed energy is electricity. The majority of the electricity is consumed in 

the electrolysis section. The primary energy use of the entire plant is about 8000 TJ per year. 

292

3.2. Results 

3.2.1. Exergy analysis 

The process can be characterised as a process with excess heat that becomes available at (relatively) 

low temperatures. Because of the location of the Budel plant the heat cannot be used elsewhere. 

From the exergy analysis it can be concluded that all the additional energy is not required from 

exergy point of view. The exergy present in the zinc ore is in theory sufficient to fuel the entire 

process. The destruction of exergy and the losses are illustrated in Fig 3. 

Fig 3. Exergy destruction and exergy losses for each section of the plant 

Clearly the biggest exergy loss is the electrolysis section. The loss is mainly caused by resistance 

and overpotential in the electrolysis section, which leads to the generation of low-temperature heat. 

However the losses in the leaching section and the sulphuric acid plant are significant as well. The 

losses in the leach section are caused by the exothermic reactions. The heat is released to the 

cooling water. Additionally steam is used to fulfil a low-temperature heat demand. This indicates 

that there should be opportunities to optimise the heat consumption, which was confirmed during 

the pinch analysis. The losses in the sulphuric acid plant are caused by the low temperature heat 

(80°C), that is released during the exothermic reaction. The exergy loss in the steam boiler and 

roasting section are in part caused by the suboptimal operation of the roasting process (from exergy 

point of view), and in part by the fact that any combustion process results in exergy loss. 

3.2.2. Pinch analysis 

293

Te 1000 

mp 

era 

900 

tur 800 

e 

[°C 700 

] 

600 

Temperatuur, °C 

500 

400 

300 

200 

100 

Stoom Steam>electricity > elektriciteit 

met With voorverwarming; 

preheating, potential 

Potentieel 191 GJ/h 191 GJ/h 

0 Stoom Steam>electricity > elektriciteit 

0 zonder 50 voorverwarming; 

100 150 

Without preheating, 

Potentieel 180 GJ/h 

200 250 

Enthalpy, GJ/h 

300 350 400 450 

potential 191 GJ/h 

Fig 4. Grand composite curve of the pinch analysis 

294 

Base_case 

Case 1 - mod. reactor pre-heat 

Case 2 - no reactor preheat 

voorverwarming Preheating above boven pinch pinch point 

verbetert does not increase electricity 

elektriciteitspotentieel potential 

niet 

The pinch analysis learns that there is an excess of heat. No heat source is required except for the 

zinc ore. The aim of the pinch analysis is, therefore, to maximise the amount of steam that can be 

used for power production. Nyrstar has a high power consumption, which is required during the 

entire year. The potential for energy saving is 180 GJ per hour without reactor preheating or 191 GJ 

per hour if the reactor is preheated. 

Two scenarios for heat integration have been made. In scenario 1 the thermal energy saving is 73 

GJ per hour leading to an electrical potential of 4.6 MWe and an investment of € 1.8 million. In 

scenario 2 the thermal energy saving is 92 GJ per hour leading to an electrical potential of 5.3 MWe 

with an associated investment of € 2.1 million 

3.2.3. Process intensification 

The results of the process intensification scan are solutions or solution directions, and thereby differ 

from regular analyses. 

The process intensification scan suggests to increase the understanding of the processes in the 

boiler, by completing the mass and energy balances, making a chemical analysis of the scaling in 

the bed and determining the effect of the process parameters on the bed. On the basis of this 

information an improved redesign of the roaster and boiler can be made. 

For the gas cleaning section a process simulation study is suggested. For the long term new 

technologies can be considered. The leaching and purification section have different steps, which 

occur at different pH. It is proposed to investigate the possibilities of using membranes to extract 

water, when the pH needs to be decreased, i.e. acidity increased. It is expected that the process will 

become more stable and easier to control. Demands on the material of the membrane will be an 

important research parameter.

In the electrolysis the ZnSO4-solution is transformed into Zn metal. This is one of the large energy 

consuming steps. The recommendations from the process intensification study, have led to renewed 

thinking on the electrolysis process. 

3.2.4. Process control 

The present process control system is mainly by hand and quasi static, i.e. the process is 

automatically controlled on the level of a single process unit for which a desired value is manually 

set, but the process is not automatically controlled for a combination of process units that interact 

with each other. There is no feedback from the output to the input and the systems to be controlled 

have a very complex interaction. It is suggested to apply a process controller that incorporates 

feedback by application of an Advanced Process Controller. Additionally it is suggested to use a 

planning system and a real-time optimizer to determine the optimal operation points of the plant. 

In the electrolysis section the production capacity is proportional to the power requirement, and 

therefore it is desirable to adjust the production capacity with the electricity price. Because of this, 

the production capacity in the leaching and purification section may have to vary. The present 

control systems have controllers which have relative large steps, which means that a small deviation 

will lead to bigger changes in the input than required. This can be changed easily by using filters, 

that allow the process to operate more smoothly. 

3.3. Selected measures 

From the proposed measures, twelve have been selected to be worked out in business cases. About 

half of the measures were in the area of improved heat integration. The pay-back period for each 

solution ranges between 4.5 and 30 years (including steam turbine), basically because heat 

integration itself does not lead to energy efficiency, because heat is available in excess and no steam 

turbine is present to convert the saved steam. It is demonstrated that the present system of direct 

coupling between the steam turbine and SO2-blower is not ideal. When revision of the blowers is 

required an investment in a steam turbine is strongly recommended. If the steam turbine would be 

dimensioned for a larger steam flow than present, or alternatively is supported by a gas boiler, the 

presence of the steam turbine will act as a catalyst for further reduction in heat consumption, as in 

this case heat or steam will have a monetary value. 

Other plans focused on the improvement of the roaster oven and electrolysis section. Interesting 

opportunities could be created by improvement of the process control systems. These solutions 

potentially could be used at more sites. Further study should show the feasibility of these ideas. 

It is noteworthy, that the exergy losses in the gas cleaning system are significant, even though the 

removal of SOx from the exhaust gas is state of the art. Further process development is required. 


The integral approach, is a well-structured method for performing energy saving studies. The 

relatively long analysis phase gives additional insight in the process regarding the loss of the quality 

of energy, which is often a precursor to the actual energy loss. The exergy analysis and pinch 

analysis were used on the level of process functions. The exergy analysis acted as a focus for more 

detailed analyses. On the detailed level, a process intensification scan and a process control analysis 

were performed. Exergy can be applied on the detailed level as well, and thereby indicate where the 

losses are taking place, but in this study it was opted not to. 

295

Application of the approach led to a list of 132 possible process improvement options. It included 

options which could be implemented immediately, but it also provided options for long term 

improvement. Of these improvement options 12 were elaborated into business cases, of which 

several are now under consideration. 


We acknowledge AgentschapNL, division energy and climate for the financial support of and 

participation in this project. 

References 

[1] Szargut J., Morris D.R., Steward F.R., Exergy analysis of thermal, chemical and metallurgical 

processes. New York: Hemisphere Publ. Corp; 1988. 

[2] Szargut J., Exergy Method, Technical and ecological applications, Southampton, UK : 

Wittpress; 2005 

[3] Kotas T.J., The Exergy Method of Thermal Plant Analysis. London, UK: Butterworths; 1985 

[4] Tsatsaronis G., 1993, Thermoeconomic analysis and optimization of energy systems. Prog. 

Energy Combust, Vol 19, pp. 227-257. 

[5] Valero A., Lozano M.A. and Munoz M., A general theory of exergy saving I. On the exergetic 

Cost. In: Gaggioli R.A. Computer-aided engineering and energy systems Vol. 3: Second Law 

Analysis and Modelling, AES Vol 2-3, The American Society of mechanical engineers, New 

York, USA. 

[6] Morusuk T. and Tsatsaronis G. How to calculate the parts of exergy destruction in an advanced 

exergetic analysis. In: Zibik A., Kolenda Z. and Stanek W., editors. ECOS 2008: Proceedings 

of the 21st International Conference on Efficiency, Cost, Optimization, Simulation and 

Environmental Impact of Energy Systems; 2008 Jun 24-27; Cracow-Gliwice, Poland. 

[7] Linhoff B., Townsend D.W, Boland D., Hewitt G.F., Thomas B.E.A., Guy A.R., Marsland 

R.H., A user guide on Process integration for the efficient use of energy. Revised first edition, 

Rugby, UK: Institution of Chemical engineers; 1994. 

[8] Martinez-Patiño J., Verda V., Serra L.M., Picón-Núñez N. and Hernández-Figueroa M.A., 

Composite curves in direct and indirect heat exchange network for simultaneous heat and mass 

transfer system: Analysis and applications. In: ECOS 2010: Proceedings of the 23 rd 

international conference on efficiency, cost, optimization, simulation and environmental impact 

of energy systems; 2010 Jun 14-17; Lausanne, Switzerland 

[9] Dubliauskaite M., Becker H., Maréchal, F. Utility optimization in a brewery process based on 

energy integration methodology. In: ECOS 2010: Proceedings of the 23 rd internationa l 

conference on efficiency, cost, optimization, simulation and environmental impact of energy 

systems; 2010 Jun 14-17; Lausanne, Switzerland 

[10] Akse H.N., Technology Outlook. Woerden, The Netherlands: Traxxys Innovation & 

Sustainability, Technical Report No.: 2011 008 2 20-07-11. Available at:< 

http://traxxys.com/downloads/Files/Technology%20Outlook%20- 

%20Traxxys%20Report%202011%20008%202%2020-7-11.pdf> [accessed 04.05.2012]. 

[11] Nyrstar, Budel Fact Sheet Available at 

http://www.nyrstar.com/operations/Documents/NYR1288%20BUDEL%20230911.pdf 

[accessed 04.05.2012]. 

296




Maximising the Use of Renewables with Variable 

Availability 

Andreja Nemet a , Jií Jaromír Klemeš a *, Petar Sabev Varbanov a , Zdravko Kravanja b 

a Centre for Process Integration and Intensification - CPI 2 , Research Institute of Chemical and Process 

Engineering, Faculty of Information Technology, University of Pannonia, Veszprém, Hungary, 

klemes@cpi.uni-pannon.hu (CA) 

b Faculty of Chemistry and Chemical Engineering, University of Maribor, Maribor, Slovenia, 

Abstract: 

A problem connected with the exploitation of renewable energy sources, such as wind and sun is, their 

fluctuating availability. The accelerating development has been very substantial for techniques, 

methodologies and equipment for exploiting solar energy [1]. The integration of renewables into an 

energy system needs an approach that accounts for the variations in energy supply availability, as well 

as for those of the demands. Dynamic models could be used for modelling precisely is intermittency. 

They are usually employed to solve servo- and regulatory tasks in process control. Dynamic models 

have been used to model solar thermal plants [2-4], but only a few models have been dedicated to 

estimating available energy from solar sources [5] and they usually evaluate only a part of the whole 

capture system – e.g. the thermal storage [6]. However, dynamic models are unsuitable for design or 

long-horizon operational optimisation. 

In the present work, the Heat Integration [7] for batch processes based on Time Slices [8, 9] is 

extended to the integration of solar thermal energy with certain variations. The main steps involve 

partitioning the measured/forecasted heat availability profile using a large number of candidate time 

boundaries and then approximating it by a piecewise-constant profile using high-precision. The 

approximation profile is obtained by subjecting the candidate superset of time-boundaries to MILP 

optimisation thus minimising the integral inaccuracy. The integration of solar thermal energy can be 

performed for each Time Slice, after the optimal number of Time Slices has been selected with 

approximated constant load. Using heat storage, the heat can then be transferred between Time 

Slices. 

Keywords: 

Variations of Renewables, Renewable Availability Curve, Solar Thermal Energy Integration, Time 

Slices, Heat Integration 


An accelerated development of techniques, methodologies and equipment for exploiting solar 

energy has been taking place recently. This helps to improve the existing technology. An example is 

solar-based water desalination [1]. A lot of attention has been focused on photovoltaic panels for 

producing electricity. There is also a significant potential for utilising solar irradiation as heat. 

Generally, thermal solar capture offers a higher efficiency compared to photovoltaic panels. 

The integration of renewables into a process system needs a specific approach due to the variations 

in energy supply availability from renewable sources, and fluctuations in the users’ energy 

demands. Two approaches can be used for integrating renewables and accounting for this 

variability: 

(i) A dynamic model formulation, followed by dynamic optimisation 

(ii) A multi-period model involving steady-states, associated with time intervals. 

297

The advantage of dynamic models is that they accurately describe the system behaviour. They are 

usually employed to solve servo- and regulatory tasks during process control. There are dynamic 

models that describe plants using solar thermal-energy as a utility [2]. Several other models 

estimate solar irradiation [3, 4], and just a few models estimating the available solar thermal-energy 

and available electricity [5]. Typically, such models only evaluate a part of the whole capture 

system, for example, just thermal-energy storage [6]. The main drawback of dynamic models is that 

they are not favourable for design or long-horizon operational optimisation since these models are 

complex and presently computationally intensive. 

Models assuming steady-states are simpler and yet still capable of describing the systems with 

acceptable accuracy. As some of the variables are discretised, the computational time becomes 

much shorter. 

During batch processes, energy demands vary over time. In order to account for these variations, 

Batch Process Integration was formulated by Kemp and Deakin [7] who developed two models: (i) 

The Time Average Model, where the heat-loads are averaged throughout the time horizon, and (ii) 

The Time Slice Model, where the Time Slices are obtained by combining the starting and ending 

time points of the involved process streams. During each Time Slices Heat Integration is performed 

in the same manner as the continuous processes. A detailed description can be found in [8]. The 

batch process scheduling method using the MILP formulation with heat integration was another 

step in exploiting batch process heat integration [9]. A similar formulation of the problem was also 

used to design a HEN [10]. In addition, a different, more combinatorial approach has been 

developed for batch process scheduling based on the S graph, [11] where also scheduling influence 

is presented on the HEN synthesis. To enable the integration of heat-storage into the system design, 

combined with pinch analysis, another combinatorial approach was subsequently introduced using 

time decompositions of the processes [12]. Majozi [13] developed a mathematical model for 

optimising energy usage for a multi-purpose batch plant. The evolution of a batch heat exchanger 

network was described by Foo et al. [14]. The methodology of time decomposition was recently 

extended by Varbanov and Klemeš [15] for analysing Total Sites using the integration of 

renewables. A Comprehensive review of Process Integration, including batch, has been presented 

by Friedler [16, 17]. 

Muster-Slawitsch et al. [18] presented the annual load curve for a renewable energy source. While 

adequate for that work, the approach is not appropriate for the integration of solar thermal energy as 

developed in this work. The reason is that clustering of the loads leaves the temporal sequence out 

of consideration, accounting only for the load horizon. It lumps loads from different time intervals 

within the overall horizon (one year) at certain temperature and load levels. For the current problem 

accounting for the temporal sequence is essential, as the heat supply and demand streams may be 

active during different time intervals, since a batch process is considered. 

Ludig et al. [19] modelled a power system, investigating 14 different technologies for producing 

electricity. They evaluated optimal technology-mix from the viewpoint of cost. An interesting part 

of this work was how they dealt with the variations of renewable energy sources e.g. wind, hydro, 

solar. They created equal-length time slices and averaged the load of supply within each time slice. 

In contrast, in the present work the time durations of the TSs and the supply load are the result of a 

two-stage optimisation. It is a systematic approach compared to the heuristic used previously. 

The focus of previous work in the field of varying heat supply and demand was either on a variation 

of the process demand or the energy availability from renewable sources. This current work 

accounts for both aspects. An analogy from batch process integration is used. TSs, with loads 

assumed to be constant, was developed for varying the availability of solar thermal-energy. The 

procedure for integrating solar thermal-energy covers several steps: 

Heat recovery within batch processes 

Identifying the number of TSs and the values of the TS boundaries for solar irradiation 

Estimation of the supply loads 

298

Combination of TSs for process demand and solar thermal-energy 

Integration of solar thermal-energy within each combined TS 

Estimation of storage size 

The trade-off between the inaccuracy and number of TSs should be evaluated, in order to determine 

the number of TSs. 

2. Determining the number of time slices 

2.1. Problem formulation 

The approximation inaccuracy decreases with an increasing number of TSs. On the other hand, the 

aim is to minimise the number of TSs, in order to simplify the computations during the following 

steps of the integration procedure. Therefore, the task can be defined as obtaining the minimum 

number of TSs with acceptable accuracy. The solar irradiation (G) measurements or the temporal 

variation of the captured heat flow could be used to identify the TS for solar energy availability. 

2.2. Approximation of the irradiation profile 

This procedure is based on optimising the load-levels and selecting items from a discrete superset of 

candidate time boundaries. These represent the measured Solar Irradiation – G data in dependence 

of time, t[h], by constructing a high-precision piecewise-constant profile [20] (Fig 1). 

G [W/ 700 m2 ] 

600 

500 

400 

300 

200 

100 

0 

4 6 8 10 12 14 16 18 t 20 [h] 

299 

measured data 

discretization 

Fig 1: Discretisation of the measured profile/ input data for optimising the number of TSs 

When using a large number of time-intervals, the inaccuracy of this transformation is minimised 

and can be ignored. However, such high-accuracy would require very intensive computation 

Therefore, the piecewise–constant load profile to be obtained has to contain a significantly smaller 

number of TSs. The supply is approximated separately at each time-interval by the minimisation of 

any inaccuracy represented by those areas occurring between the approximated and real inputsupply 

profiles (Fig 2). 

The boundaries of the time-intervals are the candidate boundaries for the final TSs. If there is a 

difference between two consecutively approximated supply levels, the time-boundary is also a TS 

boundary. When two time-intervals are joined into one TS, the approximated supply-levels should

e equal at both time intervals and the time-interval period boundary candidate is deselected as a TS 

boundary (Fig 3). 

Fig 2: Determining the inaccuracy between the input 

and approximated supply 

2.3. MILP model formulation 

300 

Fig 3: Acceptance/ rejection of the candidate time 

period boundary as a TS boundary 

A two-stage MILP model has been developed for minimising the number of TSs at acceptable 

inaccuracy. During the first stage, the number of TSs is minimised, depending on the tolerance of 

inaccuracy specified by the models’ users. During the second stage, the inaccuracy is minimised at 

a fixed minimum number of TSs, determined during the first optimisation stage. 

Initially there is NI number of time-intervals and, hence, NI +1 boundaries of time-intervals indexed 

by the following index and set: i for the time-boundaries of the time-intervals, iI. 

The difference between the real input-supply and approximated-supply is calculated during each 

time period separately: 

SD RS – AS , iI, (1) 

i i i 

Because the difference, SDi can have a positive or negative value, it can be represented as the 

difference between the positive variables PDi and NDi: 

SD PD – ND , iI, (2) 

i i i 

Note that, when the SDi has a positive value, the NDi is zero, as a result of minimising the 

inaccuracy. When SDi has a negative value, the PDi is zero. For minimal inaccuracy the difference 

between the real and approximated supply should be the lowest possible. 

EDi PDi NDi, 

iI, (3) 

In (3) the positive value is obtained for the difference between real and approximated supply load. 

Further equations relate to the accepting / rejecting of the time-interval boundary as a TS boundary. 

The decision is made by the binary variable yi. When there is a positive (4) or negative difference 

(5) between the two consecutively-approximated supply loads, there is a TS boundary and the value 

of yi is 1. If there is no difference between these supplies, there is no TS boundary and the value of 

yi is 0. 

ASi1ASi LV yi 

, iI, i NI + 1, (4) 

ASi1ASi LV yi 

, iI, i NI + 1, (5) 

In order to present the selected TS boundaries, the binary variable is multiplied by the observed 

time-period boundary:

TSi yi ti1, iI, i NI + 1, (6) 

The number of TSs is obtained from (7). One is added to the sum of the selected TS boundaries, as 

the TS boundaries at the beginning and end of the observed time-horizon were excluded within the 

model: 

NTS yi 

1, 

(7) 

iI, iNI 1 

The inaccuracy during each time-interval is determined by multiplying the positive difference 

between the real and approximated supplies with the time-horizon of the time-interval. 

INi EDi( ti ti1) , iI, i NI + 1, (8) 

The overall inaccuracy is a result of summating the inaccuracies over the time-intervals: 

INA IN , (9) 

iI, iNI 1 

i 

and this overall inaccuracy is constrained and should be less than or equal to the fraction of the 

initial amount of solar irradiation presented as an area (A0) below the measured profile of Fig 1: 

INA A , (10) 

 

0 

A (( t t) RS 

0 i1 i i 

iI, iNI 1 

2.4. Optimisation procedure 

) 

, 

(11) 

Optimisation is performed over two stages. During the first stage of optimisation, Equations (1-11) 

are used with the objective of minimising the number of TSs as follows: 

min zI NTS , (12) 

This step requires specifying the acceptable error-level (tolerance) . The procedure applies multi- 

objective optimisation by the –constraint method, where one objective is considered in the 

objective function and the other is inserted in the model as an –constraint. The result from 

optimisation is the minimal number of TSs, min NTSI required to meet any constraint about the 

inaccuracy limit (10). 

However, after the first stage, the inaccuracy is not optimal. In order to obtain a further reduction in 

inaccuracy, in the second stage of optimisation the same model using equations (1–11) is used 

together with an additional equation (13), which fixes the number of TSs, and the objective as 

expressed in (14) , 

NTS min NTSI 

, (13) 

min zII INA 

, (14) 

Multi-objective optimisation could also be performed over one stage, with the so called weighted 

sum method as sometimes this is faster. In this case, the objective function would be a weighted 

sum of NTS and INA with a high enough weight w (e.g. 10,000) for NTS, in order for the minimised 

NTS to have priority over the minimum of INA. 

z wNTS INA , (15) 

301

2.5. Selecting the number of TSs 

Selecting the number of TSs depends on the accuracy required. Fig 4 presents the obtained Pareto 

results from the multi-objective optimisation, where minimal numbers of TSs are shown vs. 

different tolerances selected. 

INA [%] 20 

18 

16 

14 

12 

10 

8 

6 

4 

2 

0 

0 10 20 30 40 NTS 50 

Fig. 4: Selecting an acceptable inaccuracy 

As can be seen from Fig 4, by increasing the tolerance the number of TSs decreases but, however, 

the inaccuracy becomes too high. On the other hand, if the tolerance is too small, the number of TSs 

might become too high and, hence, the further steps of integration would be too complex; however, 

no significant improvement may be achieved. The exact trade-off between the number of NTS and 

INA depends on the users INA tolerance. As can be seen from Fig 4, generally, a tolerance of 

between 5 and 10 % should be acceptable as with increase NTS the inaccuracy is not increased 

significantly. 

3. Estimating the supply-loads 

The supply of the loads is determined separately in each TS. Estimation of the supply-loads depends 

on the capture system. Different kinds of systems are possible, or even a system coupled with a heat 

pump [21]. A simplified scheme for capturing was assumed during this work (Fig 5). The heattransfer 

from the collectors in this model can be (i) direct or (ii) indirect. Direct heat-transfer is 

feasible when solar thermal-energy is available and there is a demand within the evaluated TS. If 

the amount of heat is higher than the demand or the heat-transfer is unfeasible in one TS, then the 

heat is transferred to storage. This heat will be available for covering any heat-demand in the 

following TSs. The indirect heat-transfer is the described transfer through storage. 

302

Fig 5: Simplified scheme for the integration of solar thermal-energy 

In order to determine the heat-load for the direct transfer of solar thermal-energy to the process, the 

irradiation load was multiplied by: 

The area of collectors and 

The efficiency of the solar collector system. 

The area usually depends on the investment and the available area of the collectors. 

Solar collector-efficiency varies significantly with changes in the quantities of solar radiation (G), 

ambient air temperature ( T A), 

and the average internal fluid temperature ( T C ) [22]: 

a ( T – T ) a 

( T 

G 

– T ) 

2 

C O – 1 C A 2 C A , (16) 

 

when O is the optical efficiency of the collector and a1 and a2 are the solar collector thermal loss 

coefficients, which are usually determined experimentally. During the first stage of the evaluation 

the average fluid temperature can be assumed to be the arithmetic average of the collectors’ inlet 

and outlet temperatures [22]: 

T 

C 

TinTout , (17) 

2 

The efficiency also depends on the ambient air temperature. It is the average temperature of the air 

in each TS separately. 

4. Integration of solar thermal-energy 

4.1. Combining the supply and demand 

The first step when combining the supply and demand is determining the TSs for any fluctuating 

load of solar thermal-energy. However, many processes have fluctuating demands. Therefore the 

solution is to also create TSs, as developed for the batch processes [7]. A combination of these two 

types of TSs can be seen in Fig 6. 

303

A) Time 

Slices for 

solar 

thermal 

energy 

B) Time 

Slices for 

the 

varying 

heat 

demand 

C) 

Combined 

Time 

Slices 

HD1 

TS1 

HD2 

cTS1 

cTS2 

TS2 

cTS3 

Fig 6: A Gantt chart for those TSs for supplying A) Solar thermal-energy, B) Heat demand and C) 

A combination of for both. 

The TS boundaries for solar thermal-energy and those processes with varying demand are joined 

together into combined TS boundaries. 

304 

cTS4 

4.2. Integration with the Grand Composite Curve 

Integration of the solar thermal-energy should be performed after the combined Time Slices (cTSs) 

are obtained. The Grand Composite Curve [23, 24] can be used for the integration of solar-thermal 

energy (Fig 7). 

TS3 

time 

cTS5 

Fig 7: Integration of solar thermal energy in one combined TS [23, 24] 

This is not, however, the only option. The use of Total Site analysis [25] and especially a Total Site 

with renewable sources of energy, including solar thermal-energy [26, 27], would be an efficient 

approach when analysing heat recovery and the integration of solar thermal energy. 

time 

time

4.3. Hierarchy for covering heat demand 

Within each Time Slice, there are three different sources regarding utilities. The following 

hierarchy [26, 27] should be followed in order to cover the heat demand: 

i) Heat recovery should be maximised. 

ii) The use of solar thermal energy via direct heat-transfer from collectors – immediately, when 

available. 

iii) Usage of the energy from the storage-indirect heat-transfer of solar thermal energy. 

iv) A backup utility with constant availability is required. 

Fig 8: Transferring solar heat from one combined TS to another [7, 27] 

Integration using the direct transfer of solar thermal-energy within TS is then performed after heat 

recovery. If there is unused heat from the solar-source, it is transferred to storage. The solar 

thermal-energy can be unused for different reasons. One is a surplus and the other is a higher 

demanded temperature than the temperature of the heat available from the solar-source. The stored 

heat will be available in other TSs (Fig 8). This is an indirect way of using the solar thermal-energy. 

When all the available solar thermal heat from the direct and indirect transfers is integrated, the rest 

of the demand should be covered by those utilities with constant availability. 

5. Case Study 

5.1. Heat recovery 

In this case study, the varying demand was presented by the batch process [7]. The streams are 

presented in Table 1. The Time Slices from the heat demand were the starting and ending times of 

the streams or changes in the loads for heat demand. 

Stream No 

and type 

TS 

[°C] 

Table 1: Streams for Case Study [7] 

TT 

[°C] 

305 

CP 

[kW/°C] 

tstart 

[h] 

1 Cold 25 110 10 12 16 

2 Cold 55 115 8 6 24 

3 Hot 140 35 4 0 12 

4 Hot 130 15 3 6 19 

tend 

[h]

The first step of procedure is to perform heat recovery within the batch process using a Problem 

Table Algorithm. The Grand Composite Curves for each TS obtained separately are presented in 

Fig 9. 

In the first TS there was an excess of heat, which could be used in the following TS. In the second 

TS there was a heat recovery pocket (Fig 9). There was also an excess of heat; however the 

temperature of the available heat was quite low, below 60 °C. There was a significant heat demand 

in the TS of between 12 h and 16 h. There was also some heat surplus; however, its temperature 

was too low, 10 °C, to be usable in the following TS. In the TS of between 16 h and 19 h, the 

demand was also significant and there was also an opportunity to store the heat, but the temperature 

was low. In the last TS, there was only heat demand and no heat surplus. Only after maximising the 

heat recovery a solar thermal energy should be integrated to the process. 

5.2. Creating a TS for solar thermal-energy 

The input real-supply profile is presented in Fig 1 (in section 2.2). This presents the daily 

irradiation. The data was taken as for a typical summer day in Central Europe. The time-period of 

the irradiation was from 5-22 to 19-22 as there was no irradiation before or after this period. It was 

a 14 h time-horizon and the measurements were taken every 15 mins. This resulted in 56 

measurements [25]. The discretisation of the irradiation can be seen in Fig 1. The results were 

obtained in 49 s on Intel(R) Core (TIM) i3 CPU processor. 

T*[°C] 140 

120 

100 

80 

60 

40 

20 

0 

T*[°C] 140 

120 

100 

80 

60 

40 

20 

0 

0 - 6 h 

0 200 H 400 [kW] 

16 - 19 h 

T*[ 140 °C] 

120 

100 

80 

60 

40 

20 

0 100 200 300H 400 [kW] 

0 

6 - 12 h 

heat recovery pocket 

0 100 200 H [kW] 300 

T*[°C] 140 

120 

100 

80 

60 

40 

20 

0 

0 200 400 H [kW] 600 

306 

19- 24 h 

Fig 9: GCCs for each TS separately 

T*[°C] 140 

120 

100 

80 

60 

40 

20 

0 

12 - 16 h 

0 250 500 750 H 1000 [kW]

The optical efficiency of the tube collectors was 0=76 %, the coefficients a1=1.53 W °C -1 m -2 and 

a2=0.0003 W °C -2 m -2 [22], the inlet and the outlet temperatures of the solar collector media were 70 

and 90 °C, and the area of the solar collectors was 150 m 2 . 

The selected acceptable tolerance in this Case Study was 10 %. The optimal number of TSs, 

obtained by the proposed MILP model, was 8. It was an important achievement, as the initial 

number of time-intervals from the measurements was 56. The minimal inaccuracy at this number of 

TSs was 9.4 %. The TS determined for the irradiation can be seen in Fig 10. 

G [W/ 700 m2 ] 

600 

500 

400 

300 

200 

100 

0 

4 6 8 10 12 14 16 18 20 t [h] 

Fig 10: TS boundaries for irradiation. 

307 

measured data 

discretization 

result of 

optimisation 

The results, obtained for TS for irradiation, suggested 8 TSs. However, capture of the heat was 

impossible when determining the efficiency of the capture system in the first and last TSs, as the 

irradiation was too low. For this reason the number of TSs with a constant load of supply was, in 

this case, 6. Fig 11 presents the final approximated load-profile for the supply of solar thermalenergy, 

and the TS boundaries. 

G [W/ 700 m2 ] 

600 

500 

400 

300 

200 

100 

5.3. Combining the TSs 

0 

5 7 9 11 13 15 17 19 t [h] 

Fig 11: TSs and approximated loads for solar thermal energy 

After obtaining TSs the (i) heat demand variations and (ii) solar thermal energy supply were joined. 

In order to combine them, the time-boundaries from both TSs were listed and any duplicates (if 

existing) were eliminated. As can be seen in Fig 12, in this case study there were 5 TSs (with 6 time 

boundaries) from the heat demand and 6 (with 7 time boundaries) from the solar thermal energy 

supply. Combining them resulted in 12 cTSs (with 13 time boundaries). This case study clearly 

showed how important it is to reduce the number of TSs for solar thermal-energy supply.

cTSs 

demand TS 

solar TSs 

0 2 4 6 8 10 12 14 16 18 20 22 t 24 [h] 

Fig 12: Combining the TS together from the Solar TSs and heat demand TSs 

5.4. Integration of solar thermal energy 

In all the cTSs the heat recovery was done first, as described in the hierarchy for covering heat 

demand (section 4.3). The hot utility requirement after heat recovery, HUR, and the excess of heat, 

HE, are shown in Table 2. The next step was to integrate the available solar thermal-energy, 

HSTE, within the observed TS in order to determine the load of the direct heat-transfer of the solar 

thermal energy, HDTE. The load of heat demand at the feasible temperature of the heat- transfer 

was also obtained. From these two calculated loads, the amount of exchanged heat and the heat load 

transferred to or from the storage of solar thermal-energy could have been also be specified. 

Another source of heat could have been also be excess-heat, which could have also been stored if 

the temperature allowed for it. For simplicity, an isothermal storage was assumed. As the timehorizon 

when using the storage was short, it was not far from a real situation. As backup, at least 

one hot, HHU, and one cold, HCU, utility were required, with constant availability. 

Table 2: Determining the load of solar thermal energy supply and the utility with constant load 

cTS After recovery Solar thermal 

energy 

Duration 

h 

HUR 

kW 

HE 

kW 

HSTE 

kW 

308 

HDHT 

kW 

HE 

kW 

Storage 

from 

HSTE 

kW 

Constant available 

utility 

HHU 

kW 

HCU 

kW 

0:00-6:00 - 420 - - 220 - 200 

6:00-6:22 - 285 - - - - - 285 

6:22-7:37 - 285 57.1 - - 57.1 - 285 

7:37-8:52 - 285 273 - - 273 - 285 

8:52-10:07 - 285 461 - - 461 - 285 

10:07-12:00 - 285 607 - - 607 - 285 

12:00-14:22 1045 60 607 510 - 97 535 60 

14:22-16:00 1045 60 402 402 - -108 535 60 

16:00-16:07 285 150 402.1 100 - 302.1 185 150 

16:07-17:22 285 150 157.4 100 - 57.4 185 150 

17:22-19:00 285 150 - - - -100 185 150 

19:00-24:00 480 - - - - -160 320 - 

As can be seen from Table 2, not all of the heat demand could have been covered from solar 

thermal-energy, because the temperature of the capture was often lower than some of the heat 

demands. This was also a reason for using a utility with constant availability. The amount of hot 

utility needed after the recovery was 7,435 kWh. This amount was calculated by multiplying the

load by the time horizon of the TS. 2,000 kWh could have been covered by direct heat-transfer from 

solar thermal-energy and 1,140 kWh could have been covered from indirect heat-transfer using 

storage. This means that the demand 3,140 kWh could have been covered by solar thermal-energy, 

which is 42.2% of the overall heat demand. The rest of the demand, 4,295 kWh, should still been 

covered from the utility with constant availability. However, the dependency on fossil fuels should 

be decreased as much as possible, since this energy source has an impact on the environment. 

5.5. Determination of storage size 

In order to estimate the storage size, the amount of the heat stored or used should be determined in 

each cTS separately e.g. the heat stored at the cTS1 is H = 220 kW6 h=1320 kWh (Table 2, 

storage column and Fig 13). These calculated amounts of heat are presented in the boxes of Fig 13. 

The cumulative amount of stored heat is represented by the numbers outside the boxes (Fig 13). Fig 

13A presents the initial cumulative heat stored. As can be seen, in the last cTS12 the amount of 

stored heat is more than zero. This indicates that smaller storage would also be sufficient. The 

smallest storage, at which the heat recovery remains the same, would be when the cumulative 

amount stored at the last cTS is equal to zero (Fig 13B). 

The storage from this case study should be large enough to store 1,032.4 kWh of heat. The result 

was determined by the maximal amount of heat within the cascade. However, in order to obtain a 

proper trade-off besides the rate of heat recovery, also the investment of the storage should be 

analysed. 

Fig 13: Cascading the amount of heat in storage through different Time Slices at A) maximal 

storage and B) reduced storage 

6. Conclusions and future work 

309

In the presented paper a framework for the integration of solar thermal-energy with processes 

featuring varying demand was developed. By applying this framework, the amount of solar thermalenergy 

can be determined, which can be potentially used within the process. 

As part of the algorithm, the current work offers a systematic procedure capable of identifying Time 

Slices with an assumed constant solar thermal-energy supply. It is an important step because, to 

date, Time Slices have mostly been detected heuristically, usually with equal lengths. However, the 

solar irradiation varies unevenly and the inaccuracy in such a model is high. This current model 

enables the user to set the accuracy wanted at the stage of analysis. Higher accuracy will result in a 

larger number of TSs. 

The presented case study used utility demand as a base case. It illustrated that the demand for a 

utility with constant availability (usually a fossil fuel) can be reduced by up to 27 % by utilising 

solar thermal-energy directly, without any storage. A further decrease of up to 15 % (on the same 

basis) can be achieved by introducing thermal-energy storage. The combined reduction in hot utility 

resulting from these two steps is about 42 %. This is a significant decrease, which should be 

encouraging enough taking solar thermal-energy in consideration, during the designing of a utility 

system. 

The formulated algorithm offers a simple and fast approach with the accuracy openly available as a 

degree of freedom for the user. Another important step for achieving better solutions is the 

simultaneous evaluation of heat supply and demand. 

As future work, the computer-aided synthesis of the developed framework will be pursued. As a 

further methodological development, shifting process operations in time (rescheduling) should be 

considered, in order to achieve as high a usage of direct transfer from solar thermal-energy as 

possible. 

7. Nomenclature 

HCU cold utility requirement, with constant availability, kW 

HDTE direct heat-transfer the solar thermal energy to process, kW 

HE 

excess of heat after heat recovery, kW 

HHU hot utility requirement, with constant availability, kW 

HSTE available solar thermal energy, kW 

HUR utility requirement after heat recovery, kW 

A0 initial overall amount of irradiation, kWh m -2 

a1 solar collector thermal loss coefficients, W °C -1 m -2 

a2 solar collector thermal loss coefficients, W °C -2 m -2 

ASi approximated supply over time-interval i, W m -2 

CP heat capacity flowrate, kW °C -1 

cTS combined time slice 

EDi the positive and negative difference between the real and approximated supplies 

together over time-interval i, W m -2 

G solar irradiation, W m -2 

I set of time boundaries 

i time boundaries of time intervals 

INA overall inaccuracy, kWh 

INi inaccuracy within time-interval i, kWh 

310

LV large value, maximum difference between real and approximated supplies 

NDi positive differences between the real and approximated supplies over time-interval i, 

W m -2 

NI number of time-intervals 

NTS number of Time Slices 

PDi positive differences between the real and approximated supplies over time-interval i, 

W m -2 

RSi real supply irradiation over time-interval i, W m -2 

SDi supply difference over time-interval i, W m -2 

t time, h 

TA ambient air temperature, °C 

TC average internal fluid temperature, °C 

tend ending time of the heat demand, h 

ti time period boundary of time-interval i, h 

Tin inlet temperature for collectors, °C 

Tout outlet temperature for collectors, °C 

TS Time Slice 

TS supply temperature of streams, °C 

TSi Time Slice boundary, h 

tstart starting time of the heat demand, h 

TT target temperature of streams, °C 

yi binary variable, selection as to whether the time period boundary is a TS boundary 

z objective function 

zI first-stage objective function 

zII second-stage objective function 

tolerance, % 

0 optical efficiency of the collector, % 

efficiency of the collector, % 

C 


Financial support is gratefully acknowledged from the EC FP7 project "Intensified Heat Transfer 

Technologies for Enhanced Heat Recovery - INTHEAT", Grant Agreement No. 262205 and 

Társadalmi Megújulás Operatív Program (TÁMOP-4.2.2/B-10/1-2010-0025). 

References 

[1] Gude VG, Nirmalakhandan N, Deng S. Desalination using solar energy: Towards 

sustainability. Energy 2011; 36: 78 – 85. 

[2] Chaabene M, Annabi M. Dynamic thermal model for predicting solar plant adequate energy 

management. Energy Conversion and Management 1998; 39: 349 – 355. 

[3] GeoModel Solar s.r.o. ; 2011 [Accessed 28/9/2011]. 

[4] KGA Associates, ; 2011 [Accessed 28/9/2011]. 

311

[5] Erdil E, Ilkan M, Egelioglu F. An experimental study on energy generation with a 

photovoltaic (PV)– solar thermal hybrid system. Energy 2008; 33: 1241– 1245. 

[6] Mawire A, McPherson M, van den Heetkamp RRJ. Simulated energy and exergy analyses of 

the charging of an oil–pebble bed thermal energy storage system for a solar cooker. Solar 

Energy Materials & Solar Cells 2008; 92: 1668–1676. 

[7] Kemp IC, Deakin AW. The Cascade Analysis for Energy and Process Integration of Batch 

Processes, Part 1: Calculation of Energy Targets. Chem Eng & D 1989; 67: 495-509. 

[8] Klemeš J, Linnhoff B, Kotjabasakis E, Zhelev TK, Gremouti I, Kaliventzeff B, et al. Design 

and Operation of Energy Efficient Batch Processes, EC Project EURO BATCH Final Report, 

EC Brussels; 1994. 

[9] Zhao XG, O’Neill BK, Roach JR, Wood RM, Heat integration for batch processes Part 1: 

Process Scheduling Based on Cascade Analysis. Trans IChemE 1998; 76: 685 – 699. 

[10] Zhao XG, O’Neill BK, Roach JR, Wood RM. Heat integration for batch processes Part 2: 

Heat Exchanger Network Design. Trans IChemE 1998; 76: 700 – 710. 

[11] Adonyi R, Romero J, Puigjaner L, Friedler F. Incorporating heat integration in batch process 

scheduling. Applied Thermal Engineering 2003; 23: 1743–1762. 

[12] Pourali O, Amidpour M, Rashtchian D., Time decomposition in batch process integration, 

Chemical Engineering and Processing 2006; 45: 14 – 21. 

[13] Majozi T. Minimization of energy use in multipurpose batch plants using heat storage: an 

aspect of cleaner production. Journal of Cleaner Production, 2009; 17: 945 – 950. 

[14] Foo DCY, Chew YH, Lee CT. Minimum units targeting and network evolution for batch heat 

exchanger network, Applied Thermal Engineering, 2008; 28: 2089-2099. 

[15] Varbanov P, Klemeš J. 2010, Total Sites Integrating Renewables with Extended Heat Transfer 

and Recovery. Heat Transfer Engineering, 31(9), 733-741. 

[16] Friedler F. Process integration, modelling and optimisation for energy saving and pollution 

reduction, Chemical Engineering Transactions 2009; 18: 1-26. 

[17] Friedler F. Process integration, modelling and optimisation for energy saving and pollution 

reduction. Applied Thermal Engineering 2010; 30: 2270-2280. 

[18] Muster-Slawitsch B, Weiss W, Schnitzer H, Brunner C. The green brewery concept e Energy 

efficiency and the use of renewable energy sources in breweries. Applied Thermal 

Engineering 2011, 31, 2123-2134. 

[19] Ludig S, Haller M, Schmid E, Bauer N. Fluctuating renewables in a long-term climate change 

mitigation strategy. Energy 2011, 36 (11) 6674-6685 

[20] European Commission, [Accessed: 23/08/2011]. 

[21] Xi C, Hongxing Y, Lin L, Jinggang W, Liu W. Experimental studies on a ground coupled 

heat pump with solar thermal collectors for space heating. Energy 2011; 36: 5292 – 5300. 

[22] Atkins MJ, Walmsley MRW, Morrison AS. Integration of solar thermal for improved energy 

efficiency in low-temperature-pinch industrial processes. Energy2010; 35: 1867 – 1873. 

[23] Townsend DW, Linnhoff B. Heat and power networks in process design. Part II: Design 

procedure for equipment selection and process matching. AIChE Journal 1983; 29(5): 748- 

771. 

[24] Linnhoff B, Townsend DW, Boland D., Hewitt GF, Thomas BEA, Guy AR, Marsland RH. A 

user guide on process integration for the efficient use of energy. IChemE, Rugby UK;1982, 

last edition 1994. 

[25] Klemeš J, Dhole VR, Raissi K, Perry SJ, Puigjaner L. Targeting and Design Methodology for 

Reduction of Fuel, Power and CO2 on Total Sites. Appl Therm Eng 1997; 17: 993 – 1003. 

312

[26] Perry S, Klemeš J, Bulatov I. Integrating waste and renewable energy to reduce the carbon 

footprint of locally integrated energy sectors. Energy 2008; 33: 1489-1497. 

[27] Varbanov PS, Klemeš JJ. Integration and management of renewables into Total Sites with 

variable supply and demand. Computers and Chemical Engineering 2011; 35: 1815 – 1826. 

313

PROCEEDINGS OF ECOS 2012 - THE 25 TH INTERNATIONAL CONFERENCEON 



Methodology for the Improvement of Large 

District Heating Networks 

Anna Volkova a , Vladislav Mashatin b , Aleksander Hlebnikov c and Andres Siirde d 

Abstract: 

a Tallinn University of Technology, Estonia, anna.volkova@ttu.ee, CA 

b Tallinn University of Technology, Estonia, vladislav.mashatin@dalkia.ee 

c Tallinn University of Technology, Estonia, aleksandr.hlebnikov@ttu.ee 

d Tallinn University of Technology, Estonia, andres.siirde@ttu.ee 

The purpose of this paper is to offer a methodology for the evaluation of large district heating 

networks. The methodology includes an analysis of heat generation and distribution based on 

the models created in the TERMIS and EnergyPro environment. For the approbation of 

proposed methodology the data on large-scale Tallinn district heating system was used as a 

basis of case study. The effective operation of district heating system, both at the stage of heat 

generation and heat distribution, can reduce the cost and price of heat supplied to the 

consumers. It can become an important factor for increasing the number of district heating 

consumers and demand for the heat load, which in turn will allow installing new cogeneration 

plants, using renewable energy sources and heat pump technologies. 

Keywords: 

District heating, DH, energy efficiency, energy systems, simulation, pipes, cogeneration 


The properly operating district heating systems can provide possible improvement of energy 

efficiency, reduce emissions, improve energy security and competitiveness and creating of new 

jobs. A district heating network includes the infrastructure for centralised heat production and 

distribution to the consumers for providing space heating and hot tap water in a wider area. The 

district heating system can be considered energy efficient and cost-effective only at optimal 

operation conditions and minimum heat loss. 

One of the actions mentioned in the EU strategy Energy 2020 is to increase the uptake of high 

efficiency district heating systems. A high efficiency district heating system can only be provided 

when efforts are concentrated on the whole energy chain, from energy production, via distribution, 

to final consumption [1]. 

There are more than 5000 district heating systems in Europe, currently supplying more than 9% of 

total European heat demand. District heating systems are mainly used in the northern European 

countries, such as Sweden and Finland [2]. As regards to Latvia, Lithuania and Estonia, the 

percentage of district-heated households is around 60-75%. 

The main advantages of district heating are efficiency, reliability and cleanness compared to the 

individual heating systems. Efficiency can be reached when heat is produced simultaneously with 

electricity in the cogeneration process. For larger heat generation units there are more options of 

flue gas cleaning available than for small scale boilers. District heating is a good solution for the 

areas with high population density and multiplied welling houses, because the investments per 

household can be reduced. Due to the fact that the connection to a single-family house is rather 

expensive, district heating is a less attractive solution for the countryside. 

314

Large district heating networks supply heat usually in big cities, since the level of heat consumption 

in large areas is high. Large district heating systems are typical for Estonia. District heating is used 

in all bigger cities in Estonia, including the capital Tallinn [3]. 

The purpose of this paper is to offer a methodology for the improvement of large district heating 

networks. For the approbation of the offered methodology, the data on Tallinn district heating 

system was used as a basis of the case study. 

District heating systems offer a potential for renewable heat generation technologies. The most 

popular renewable energy source for heat generation is biomass, which includes agricultural, forest, 

and manure residues and in extent, urban and industrial wastes, which under controlled burning 

conditions, can generate energy, with limited environmental impacts [4-6]. Geothermal as 

renewable energy can be used for district heating system and geothermal district heating has been 

given increasing attention in many countries during the last decade [7]. The expansion of district 

heating will help utilize heat production from above mentioned renewable energy sources [8]. 

The developed district heating systems promote cogeneration development. When cogeneration 

plant supplies heat to district heating system, its capacity is defined by maximum heat load of this 

system [9, 10]. In some cases thermal storage unit is attached to cogeneration plant for efficient 

operation of district heating system [11]. Cogeneration plant with district heating provides an 

alternative energy production and delivery mechanism that is less resource intensive, more efficient 

and provides greater energy security than many popular alternatives [12]. 


As it was mentioned before only an optimally operated district heating system can be considered 

energy efficient. The efficiency of operation should be evaluated both relative to heat generation 

(boiler houses and cogeneration plants) and heat distribution networks (pre-insulated pipes). 

2.1. Evaluation and improvement of heat distribution 

Improvement of district heating network is a complex task where many parameters should be taken 

into account. There are three ways to improve a district heating system by reducing the heat loss: 

The low investment scenario assumes reduction of supply temperature and increased water flow. 

This can be possible only in case the network pipe dimensions are larger than required. In this case 

the pressure will grow, which means that the number of damaged pipes may increase. The increased 

pressure can also be a problem for the customer systems. Additional pumping capacity is required 

in power plants. 

The medium investment scenario assumes replacement of pipe insulation. The insulation can be 

replaced when the steel casing of pipe is in good condition, otherwise the pipe should be fully 

replaced. Selection of insulation thickness is a complex task where many parameters should be 

taken into account: material and work cost, thermal conductivity of new and old insulation, pipe 

diameter, environmental temperature and water temperature and so on. 

The high investment scenario assumes reconstruction of pipelines with the installation of preinsulated 

pipes and increasing or decreasing their diameter, if needed. The new diameter should be 

selected very carefully whereas considering the future network development possibility. As a matter 

of fact, it is possible to replace all the pipes only in small networks; otherwise the project cost will 

be too high. 

It is not possible to carry out such improvement without creating a virtual model and trying all 

possible scenarios, especially in large networks where many heat suppliers can work together in 

different combinations. For the evaluation of heat distribution, a special model was created using 

the commercial TERMIS software [13]. Simulation can be done using other software like Bently sis 

HYD or Zulu Thermo, but TERMIS is considered to be the most advanced, powerful and extensive 

315

district energy network simulation platform for improving system design and operation. Different 

types of improvement of European district heating systems were made using commercial TERMIS 

[13-15]. TERMIS is a hydraulic modelling software tool, which gives an overview and control of 

district energy network by simulating the flow, pressure and thermal behaviour. With TERMIS, it is 

possible to reduce energy loss and reduce the CO2 emission [13]. 

Before creating the model, it is necessary to create a database, which should include the data on all 

the pipes with their dimension, insulation, coordinates, roughness and single pressure loss 

description; the consumer data like seasonal consumption of heat and tap water; environmental data 

like the air temperature for overhead pipelines and soil temperature for subsurface pipelining to 

calculate the heat loss. 

2.2. Evaluation and improvement of heat production 

Usually in large-scale district heating systems various energy sources are used: large and small 

boiler houses and cogeneration plants. Both fossil fuel and wood fuel can be used for heat 

production. The operation efficiency of boiler houses depends on the age of installed equipment. 

The renovated or new boiler houses have higher efficiency and are easily operated. 

As regards to cogeneration plants, especially those based on wood fuel, the efficiency begins to fall 

when the load is less than 70 %. Besides, the investments in cogeneration plants operation are much 

higher than in boiler houses. That is why, it is more important to operate the cogeneration plants at 

the maximum load. The boiler houses are often used as peak demand covering units. 

The following indicators should be used for the evaluation of heat production: type of production 

unit (boiler house or cogeneration plant), heat capacity (for cogeneration plant the electrical 

capacity, additionally), age of a heat production unit, fuel type (fossil fuel of renewable fuel), 

energy efficiency, and shut-downs. 

It is important to find a right solution in the operation strategy for all heat production units. Priority 

should be given to CHP production. Boiler houses are used only in case the heat supplied from a 

cogeneration plant is insufficient. 

Different types of modelling tools for the economic analysis and optimal operation of cogeneration 

plants have been developed in recent years. As the examples SEA/RENUE, CHP sizer, Ready 

Reckoner, EnergyPro can be mentioned [16]. EnergyPRO was chosen for evaluation of heat 

production in the district heating system, because it is modelling software which allows carrying out 

detailed technical and financial analyses of energy projects. For the optimisation of cogeneration 

plants, the priority in EnergyPRO software tool is that the cogeneration plant meets the heat demand 

for the period being analysed [17]. 

A simple model, which was created using the EnergyPro software, can be applied to determine the 

optimal operating strategy. The current situation and development scenarios can be compared, using 

the following parameters: heat production, fuel consumption, electricity production, operation time. 

3. Case Study 

3.1. Tallinn Municipality District Heating System 

Tallinn is the capital of Estonia located on the northern coast of the country. Tallinn is the largest 

city in the country with about 415,000 inhabitants. 

District heating networks in Estonia are mostly old and in poor condition. The state of the district 

heating networks of Tallinn is typical for the rest of Estonian district heating systems. In Tallinn the 

heat is supplied to the consumers through a 429-kilometre long heating network including 119 km 

of pre-insulated pipes (27.7%), 22.2km is a pipeline with the renovated PUR insulation; 46% of the 

316

whole pipeline network is canal pipes and 8.2 % overhead pipeline. Other pipelines are in tunnels 

and undergrounds. The diameter of main pipeline is up to 1200mm. The peak heat load of Tallinn 

district heating system was 640 MW (-22.6 C) in the 2010/2011 heating season while in the 

2009/2010 heating season it had been higher reaching 695 MW (-23.4 C). The minimum heat load 

during the summer period is 55-65 MW [3]. The district heating systems of Tallinn were mostly 

constructed in 1960-1980 and their average age is 23 years as of 2012. The district heating systems 

of Tallinn consist of three connected districts of central heat supply where one of them is divided 

into two smaller districts, and 26 local boiler houses. Currently two cogeneration plants and three 

large-scale boiler houses supply heat to the districts of Tallinn. Almost the whole district heating 

network belongs to the Tallinna Küte company[18]. The Tallinn district heating network is shown 

in Fig. 1. 

Fig.1. Tallinn district heating network. 

Most of the pipelines were built during the rapid industrial growth of the city and thus the pipelines 

were oversized with a view of future development. After the collapse of Soviet Union many 

industries were closed. At the moment there are two main problems in the network: bad insulation 

and oversized pipelines; as a result, heath losses are high. According to the Tallinna Küte AS 

development plans, the relative heat loss should be reduced by 20%. 

3.1. Heat distribution 

3.1.1. Model description 

A model was created for the Tallinn district heating network. The model was designed for 9868 

pipes, over 3658 consumers and 9800 nodes with the geographic information included. 

Different scenarios were simulated for the hydraulic and heat loss analyses: 

-current consumption and temperature schedule; 

-current consumption and maximum temperature decrease by 15 °C; 

- consumption reduced by 20% and current temperature schedule; 

- consumption reduced by 20% and maximum temperature decrease by 20 °C. 

In the fourth scenario the temperature is decreased by 20°C and due to the reduced consumption, 

the water flow can be increased further. For the heat loss analysis the average seasonal temperatures 

317

were decreased by 10°C.All scenarios were calculated twice: for the maximum consumption at - 

22°C to analyse the hydraulics and for the average seasonal parameters to analyse the heat loss. 

3.1.2. Input data and assumptions 

All the data has been taken from the Tallinna Küte GIS and converted to fit the TERMIS model. 

Tallinna Küte has also a large statistical database on different parameters in the critical points and 

consumption of each household during the last ten years. The parameters in critical points are 

required for model tuning; the number of points depends on the network. The GIS data and other 

databases can easily be interconnected by using Model Manager. Depending on the model, the 

estimated or average seasonal consumption can be used while the average seasonal consumption is 

more justified in most cases. First simulation can be made when the plant parameters like water 

flow, pressure and temperature are given. After first simulation with the adjustment factors applied, 

the simulation results should be identical to the known parameters in critical points. Only in this 

case the input parameters can be changed and it can be assumed that the simulation result is correct. 

3.1.3. Results 

The results of model simulation are shown in Tables 1 to 2. Table 1 shows the results at the 

maximum consumption and should be used for hydraulic parameters analysis, the Table 2 shows the 

seasonal average results and should be used for heat loss analysis. 

Table 1. Simulation results for the maximum consumption at -22°C 

Outdoor temp. -22°C today -15°C -20% -20°C /-20% 

Production, MW 678 670 558 546 

Consumption, MW 600 600 480 480 

Heat loss, MW 78 70 78 66 

Heat loss, % 11.5% 10.4% 14.0% 12.1% 

Water flow, t/h 11180 14070 8920 11890 

As it can be seen in Table 2,with changing the yearly average temperature by 10°C, it is possible to 

reduce the average relative heat loss for a heating season by 1.1% points that makes about 23.2GWh 

(for the 5800h heating season) or over 4200t/CO2 in case the consumption stays at the same level as 

today. In case the consumption will be reduced for 20% in the future, the relative heat loss can be 

reduced by 1.3% compared to the current temperature schedule. However, the relative loss would 

be higher compared with the present consumption. A possible solution in this case the temperature 

lowering could be higher, especially, as it can be seen, water flow is only 6,5%. 

Table 2. Simulation results for average seasonal parameters 

Season average today -10°C -20% -10°C /-20% 

Production, MW 324 320 267 263 

Consumption, MW 279 279 223 223 

Heat loss, MW 45 41 44 40 

Heat loss, % 13.9% 12.8% 16.5% 15.2% 

Water flow, t/h 7280 9260 5800 7340 

318

It should be mentioned that the total water flow and pressure difference in the network will grow. It 

means that more powerful pumps should be used. Electricity consumption will grow, but the heat 

savings will be bigger than the increase of pumping cost. Besides, in case of Tallinn, most of the 

pipes are oversized and the growth of pressure difference is not so rapid. 

3.2. Heat production 

3.2.1. Model description 

The model of current situation was built using the EnergyPro software. The components included in 

this model are shown in the Table 3. The model consists of three sites. Site 1includes 2 boiler 

houses operated during the heating season and a heat consumer. Site 2 includes a heat consumer, 

which is supplied by the heat produced in Site 1 and Site 3. Site 3 includes 2 energy units: the 

Tallinn CHPP where heat and electricity are cogenerated and Iru Plant where only two boilers are 

operated with no electricity generation. Besides, a heat consumer is included in Site 3. During the 

summer period only the Tallinn CHPP is operated supplying heat for hot water production to the 

whole district heating system. During the winter period all energy units are operated while the heat 

produced in Sites 1 and 3 is used to cover the heat demand of these sites and supplied to Site 2 also. 

The description of the model components is presented in Table 3. 

Table 3. Components of the Tallinn DH model 

Site 1 

Heat sources 

Heat load 

Site 2 

Heat sources 

Mustamäe boiler 

house 

Natural gas, heat capacity 100 MW, fuel input 106, 

working time, 15/09-15/05 

Kadaka boiler house Natural gas, heat capacity 129 MW, fuel input 138, 

working time, 15/09-15/05 

Site 2 When the boiler houses in Site 1 are shut down, heat is 

supplied via Site 2 

Demand in Mustamäe 

District 

Annual heat demand is 693 GWh, 11% of the demand is 

hot water heating load, 89% of the demand depends 

linearly on ambient temperature during the heating 

period. The data on the ambient temperature in Tallinn 

for 2010 were used for simulation 

Site 2 During the heating period the heat produced in 

Mustamäe and Kadaka boiler houses is supplied to Site 

2 

Site 1 Heat produced in Site 1 (by Mustamäe and Kadaka 

boiler houses) during the heating season is supplied to 

Site 2 

Site 3 Heat produced in Site 3 (by the Tallinn CHPP and Iru 

Plant) during the heating season and in summer supplied 

to Site 2 

Heat load Demand in Kesklinna 

District 

Annual heat demand is 380 GWh, 11% of the demand 

is hot water heating load and 89% of the demand has a 

linear dependence on the ambient temperature during 

the heating period. The data on the ambient temperature 

in Tallinn for 2010 were used for the simulation. 

319

Site 3 

Heat sources 

Heat load 

Site 1 During the summer period, the heat supplied from Site 3 

is distributed in Site 1 

Tallinn CHPP Wood, heat capacity 65 MW, electrical capacity 25 

MW, fuel input 125, working time year-round supply 

Iru Plant Natural gas, 353 MW 

Demand in Lasnamäe 

District 

Annual heat demand is 561 GWh, 11% of the demand is 

hot water heating load and 89% of the demand has a 

linear dependence on the ambient temperature during 

the heating period. The data on the ambient temperature 

in Tallinn for 2010 were used for simulation. 

Site 2 During the heating period and in summer the heat 

produced in Tallinn CHPP and Iru Plant is supplied to 

Site 2, during the summer time when other heat 

generation units are shut down, the heat is supplied via 

Site 2 to Site 3 

Transmissions 

Transmission 1 Heat from Site 1 can be supplied to Site 2 and from Site 2 to Site 1, the maximum 

capacity 200 MW, loss 10% 

Transmission 2 Heat from Site 3 can be supplied to Site 2 and from Site 2 to Site 3, the maximum 

capacity 173 MW, loss 10% 

The sites of the model are shown in Figs 2 to 4. 

Fig. 2. Model of Tallinn district heating system, Site 1. 

320

Fig. 3. Model of Tallinn district heating system, Site 2. 

Fig. 4. Model of Tallinn district heating system, Site 3. 

On the territory of Iru Plant a waste incineration plant is planned to be built where electricity and 

heat will be generated from the municipal waste (Site 3). To forecast the possible operating process, 

two scenarios were simulated, with and without a new incineration unit. 

The future changes in Site 3 are shown in Fig. 5. 

Fig. 5. Model of Tallinn district heating system, Site 3 (with an incineration plant). 

Data about the new unit are presented in Section 3.2.2. 

321

3.2.2. Input data and assumptions 

Actual data on operating heating plants were used (Table3). For the heat demand simulation, it was 

assumed that 89% of heat demand is linearly dependent on ambient temperatures and 11% of the 

demand goes to cover the hot water consumption. The average annual heat demand for last three 

years (2008-2010),which is 1691 GWh, was taken as a basis for the calculation 

As it was mentioned above, the simulation was made for two scenarios: with and without a new 

waste incineration plant. The data on the waste incineration plant used in the simulation are shown 

in Table 4. 

Table 4.Parameters of waste incineration plant [19, 20] 

Waste incineration plant 

Fuel Waste 

Heat value of fuel 10.5 MJ/kg 

Fuel input 80.5 MW 

Heat capacity 50 MW 

Electricity capacity 17 MW 

For simulating the operation, one more indicator should be included. This indicator is the priority of 

unit operation. The assumed priorities according to the operation strategy are shown in Table 5. 

In both cases the highest priority is the Tallinn CHPP. Tallinn CHPP is a plant, which was launched 

in cogeneration mode in 2009. Wood chips are used as a fuel for electricity and heat production. 

The Mustamäe and Kadaka boilers have almost the same parameters and that is why they have the 

same priority. The Iru Plant is owned by another company and will be bought from the owner. 

When the incineration plant will start to operate, its priority will be very high, because it is an 

environmentally friendly and energy efficient technology. 

Table 5.Operation strategy priorities 

Unit Priority 

(current 

situation) 

Priority 

(with the 

incineration plant) 

322 

Partial load 

allowed 

Tallinn CHPP 1 1 yes 

Iru Plant (boiler) 4 4 yes 

Mustamäe boiler 2 3 yes 

Kadaka boiler 2 3 yes 

Incineration plant (CHP) - 2 yes 

3.2.3. Results 

The results of simulation are shown in Table 6. 

The simulation showed that in case the incineration plant is used additionally, electricity generation 

will increase by 43 %. The consumption of fossil fuel - natural gas will decrease by 20%. 

Table 6.Simulation results for heat production 

Indicators Without the incineration plant With incineration plant

Heat production (GWh) 1,854.60 1,857.30 

Tallinn CHPP 438.00 438.00 

Incineration plant CHP 278.40 

Mustamäe 

boiler house 711.00 667.10 

Kadaka boiler house 394.40 354.70 

Iru Plant 311.20 119.10 

Electricity production (GWh) 219.00 313.60 

Tallinn CHPP 219.00 219.00 

Incineration plant CHP 94.66 

Fuel consumption (GWh) 2,314.00 2,467.40 

Natural gas 1,438.00 1,144.80 

Wood 876.00 876.00 

Waste 446.60 

Working hours 

Tallinn CHPP 8760 8760 

Incineration plant CHP 5568 

Mustamäe 

boiler house 5616 5616 

Kadaka 

boiler house 5064 4008 

Iru Plant 5136 2544 

The year-round operation of the system is shown graphically in Fig. 6. Fig. 6 shows that Tallinn 

CHPP operates all year round. The Mustamäe boiler house operates during all the heating period, 

but Kadaka boiler house and Iru Plant are used for peak loads. 

Fig. 6. Heat load of Tallinn District Heating Network. Simulation of the current situation. 

323

Fig.7 shows the operation forecast for the second scenario when the incineration plant is added. The 

Tallinn CHPP operates all year round. The incineration plant works during all the heating period at 

full load. The Mustamäe boiler house operates during all the heating period, but at partial load. The 

Kadaka boiler house and Iru Plant work less than in the first scenario. 

Fig.7. Heat load of Tallinn District Heating Network. Simulation of the scenario with an additional 

waste incineration plant. 

Conclusion 

The reliability and cost-efficiency of district heating depends on the efficiency of its operation. In 

this paper a methodology for the assessment and efficiency increasing of heat production and 

distribution was offered. 

As a case study, the Tallinn district heating system was analysed. The Tallinn district heating 

system includes 4 heat plants, which cover the heat demand in 3 districts. For the evaluation of 

district heating network, a model was created using the TERMIS software. For the heat loss 

analysis and hydraulic analysis, four different scenarios were simulated: the current situation, 

decreased maximum flow temperature by 15°C, decreased consumption by 20% and decreased 

temperature by 20°C with the decreased consumption by 20%. As a result, the decreasing of relative 

heat loss by 23.2GWh during the heating season compared to the current situation was gained. 

For the evaluation of energy production, a model was created using the EnergyPro software. The 

system was split into three sites. This model was used for the simulation of two scenarios: the 

current situation and the case where anew incineration cogeneration plant will be installed. The 

actual data for the last years were used for the simulation. The results of simulation showed that 

according to the current situation, the cogeneration plant should work all year round, the boiler 

houses should operate during the heating period and the Iru Plant boiler should be used only for the 

peak heat load. In case when the incineration plant is added, it can operate at full load during more 

than 5500 hours per year. Heat generation in this plant will decrease the consumption of natural gas 

324

y 20%. The amount of electricity production in cogeneration mode will increase by 43%. The 

evaluation of district heating demand was made mainly from the technical point of view. 

The effective operation of district heating system, both at the stage of heat generation and heat 

distribution, can reduce the cost and price of heat supplied to the consumers. It can become an 

important factor for increasing the number of district heating consumers and demand for the heat 

load, which in turn will allow installing new cogeneration plants, using renewable energy sources 

and heat pump technologies. 


This work has been partly supported by the European Social Fund within the researcher mobility 

program MOBILITAS (2008-2015), MJD10 and DoRa program. 

References 

[1] EUROPA- the official web site of the European Union [internet]. Brussels, Belgium: 

Communication from the Commission to the European Parliament, the Council, the European 

Economic and Social Committee and the Committee of the Regions. Energy 2020. A strategy 

for competitive, sustainable and secure energy. Available at 

>. [accessed 01.12.2011]. 

[2] Rezaie B, Rosen MA. District heating and cooling: Review of technology and potential 

enhancements. Applied Energy 2012; 93: 2-10. 

[3] Hlebnikov A., Volkova A., Džuba O., Poobus A., Kask U. Damages of the Tallinn District 

Heating Networks and indicative parameters for an estimation of the networks general 

condition In:Proceedings of 12th International Symposium on District Heating and Cooling , 

Estonia, Tallinn, 5.-7. September, 2010. - 277.-282. pp 

[4] Vallios I., Tsoutsos T., Papadakis G. Design of biomass district heating systems. Biomass and 

Bioenergy 2009; 33(4): 659-678 

[5] Difs K., Wetterlund E., Trygg L., Söderström M. Biomass gasification opportunities in a 

district heating system. Biomass and Bioenergy 2010, 34(5): 637-651 

[6] Holmgren K., Gebremedhin A. Modelling a district heating system: Introduction of waste 

incineration, policy instruments and co-operation with an industry. Energy Policy 2004; 

32(16): 1807-1817 

[7] Hepbasli A. A review on energetic, exergetic and exergoeconomic aspects of geothermal 

district heating systems (GDHSs) Energy Conversion and Management 2010; 51(10): 2041- 

2061 

[8] Lund H., Möller B., Mathiesen B.V., Dyrelund A. The role of district heating in future 

renewable energy systems. Energy 2010; 35(3):1381-1390 

[9] Tveit T.-M., Savola T., Gebremedhin A., Fogelholm C.-J. Multi-period MINLP model for 

optimising operation and structural changes to CHP plants in district heating networks with 

long-term thermal storage. Energy Conversion and Management 2009;50(3):639-647 

[10] Volkova A., Hlebnikov A., Siirde A. Methodology for defining of eligible capacity for wood 

fuel based cogeneration plants in small towns in Estonia. Journal of Energy and Power 

Engineering: 2011; 5(6):481 – 489 

[11] Barelli L., Bidini G., Pinchi E.M. Implementation of a cogenerative district heating system: 

Dimensioning of the production plant Energy and Buildings:2007;39(6):658-664 

[12] Kelly S., Pollitt M. An assessment of the present and future opportunities for combined heat 

and power with district heating (CHP-DH) in the United Kingdom Energy Policy 2010, 

38(11):6936-6945 

325

[13] TERMIS Operation User Guide, 7-Technologies, 2011 

[14] Tol H.I., Svendsen S. Improving the dimensioning of piping networks and network layouts in 

low-energy district heating systems connected to low-energy buildings: A case study in 

Roskilde, Denmark, Energy 2012: 38(1):276–290 

[15] Li H., Dalla Rosa A., Svendsen S. Design of low temperature district heating network with 

supply water recirculation, In: Proceedings of 12th International Symposium on District 

Heating and Cooling , Estonia, Tallinn, 5.-7. September, 2010:73-80 

[16] Hinojosa L.R., Day A.R., Maidment G.G., Dunham C. A Comparison of combined heat and 

power feasibility models, Applied Thermal Engineering 2007;27 (13): 2166–2172 

[17] User's Guide energyPRO, EMD, 2008 

[18] Tallina Küte www.soojus.ee 

[19] Report of environmental assessment. Waste use for combined heat and power production on 

the territory of Iru Plant. (Kütusena jäätmeid kasutava soojus- ja elektrienergia koostootmisploki 

rajamine Iru elektrijaama territooriumile, KMH hindamise aruanne), EestiEnergia 2007 

[20] IRU WtE - technical overview Jäätme energia plokk WTE, 2010, CNIM 

326




Optimal mine site energy supply 

Monica Carvalho a , Dean Millar b 

a (CA) Energy Renewables & Carbon Management Group, Mining Innovation Rehabilitation and Applied 

Research Corporation (MIRARCO), Sudbury, Ontario, Canada. {mcarvalho@mirarco.org} 

a,b Bharti School of Engineering, Laurentian University, Sudbury, Ontario, Canada. 

{dmillar@mirarco.org} 

Abstract: 

This paper reports on early work and concept development for Optimal Mine Site Energy Supply, where the 

specific energy supply requirements and constraints for mineral production operations are considered 

against methodologies that have been applied for other sectors and in other energy policy regimes. The 

primary motivation for this research is to help ensure that Canadian mineral producers will achieve reduced 

production costs through improvements in the efficiency with which they consume energy resources. Heat 

has not yet been considered for the mining sector in an integrated manner, which makes polygeneration of 

great interest. Through extension of proven methodologies, the ‘most adequate’ configurations of energy 

supply equipment that satisfice the energy requirements of mine sites both on and off transmission and 

distribution systems are identified by the optimization process. The methodology that optimizes configuration 

of polygeneration systems for mine sites has not been reported before. The variety of mining circumstances, 

temporal variations in energy prices, institutional inertia, and conservatism in design for mines are some of 

the reasons. This paper reviews some aspects of precedent energy management practice in mineral 

operations, which highlights energy challenges characteristic of the sector and sets out the initial formulation 

of optimal mine site energy supply. The review indicates the additional benefits of energy supply systems for 

mine sites that concurrently meet all utilities. 

Keywords: 

Polygeneration, renewable energy, heat integration, energy management, mineral sector. 


Mining is first and foremost a source of mineral commodities that all countries find essential for 

maintaining their economies and improving their standards of living. Mined materials are needed to 

construct roads and hospitals, to build automobiles and houses, to make computers and satellites, to 

generate electricity, and to provide many other goods and services [1]. Both energy consumption 

within the mining industry and energy prices, are rising and thus there increased need to reduce 

consumption, and improve primary energy utilization to maintain competitiveness within the 

mining sector by reducing input costs. Generally, energy supply security and reduced emissions can 

be achieved through [2]: i) improvement in energy efficiency; ii) energy savings; iii) higher 

proportion of renewable energy in supply systems; and iv) process-wide integration. 

Given the significance of energy costs in operating expenses, efficiency of energy production and 

use must be improved in the energy-intensive mining sector [3]. Governments and mining 

associations recognize the importance of improving energy efficiency, and are working together to 

implement more energy-efficient technologies. Energy efficiency makes sense for mining 

operations because it can reduce production cost while simultaneously realizing additional benefits 

including reduction in the greenhouse gas emissions. While not yet well utilized, process integration 

and polygeneration are promising tools which reach the double objective of increasing the 

efficiency of utilization of natural resources, and also of reducing the environmental impact [4]. 

327

Polygeneration is a term used to describe a generalization of the cogeneration concept where two 

(co-generation) or more (poly-generation) energy services are simultaneously provided through use 

of highly-integrated energy systems. An immediate advantage of polygeneration is its 

thermodynamically efficient use of fuel. Polygeneration systems utilize otherwise wasted thermal 

energy, and can use it for space heating, industrial process needs, or as an energy source for another 

system component. This “cascading” use of energy is what distinguishes polygeneration systems 

from conventional separate electric and thermal energy systems (e.g., a powerplant and a 

lowpressure boiler), and from simple heat recovery strategies [5]. 

The deployment of polygeneration systems in mine sites aims at increasing the efficient use of 

natural resources by combining different technologies, process integration, and energy resources, an 

objective which may render mineral production operations compliant with the new energy 

management standard, ISO 50001 [6]. 

Advantages of polygeneration systems have been demonstrated in literature [7] [8], as energy 

efficiency is associated with economic savings and sparing of the environment, as less fuel is 

consumed and consequently less pollution is generated. Such integrated energy systems could play 

an important role in the gap between fossil fuel-based energy systems and renewable energy-based 

systems. Polygeneration is a fully developed technology that has a long history of use in many types 

of industry, particularly in pulp and paper, petroleum and chemical industries, where there is a large 

demand for both heat and electricity at each site [9]. In recent years, the greater availability and 

choice of suitable technology options means that polygeneration can become an attractive and 

practical business proposition. 

In recent years, the analysis and design tools for energy systems and energy management have 

undergone development. In particular, the synthesis and design of energy systems for the industrial 

sector has become increasingly elaborate, with numerous possibilities for energy sources and 

technological options. This increase in complexity allows for more flexible systems but at the same 

time increases difficulties when designing the polygeneration system itself. 

This paper reports on early work and concept development for Optimal Mine Site Energy Supply, 

where the specific requirements and constraints of mineral production operations are considered 

against methodologies that have been applied for other sectors and in other energy policy regimes. 

Through these extensions, the ‘most adequate’ configurations of energy supply equipment that 

satisfice (as coined by Herbert Simon [10]) the energy requirements of mine sites in different 

scenarios and conditions of constraint can be identified. 

This paper also presents a critical review of precedent energy management practice in mineral 

operations, where the thrust has been independent deployment of beneficial technologies. The 

review indicates the additional benefits of energy supply systems for mine sites that concurrently 

meet all utilities. 

2. Energy in mineral operations 

South Africa’s Department of Minerals and Energy estimates that the mining industry uses 6% of 

all the energy consumed in South Africa. In Brazil, the largest single energy consumer is mining 

giant Vale, which accounts for around 4% of all energy used in the country. In the US State of 

Colorado, mining has been estimated to account for 18% of total industrial sector energy use, while 

overall in the US it is calculated that the mining industry uses 3% of industry energy [11]. A secure 

and reliable supply of energy is thus critical for all mining operators to meet their production 

requirements. For most, energy constitutes a major operating expense and its generation and 

distribution requires substantial capital investment. To minimize costs, it is important to recognize 

that energy is a controllable operating cost [12]. 

328

Canada is one of the world’s leading mining countries and ranks among the largest producers of 

minerals and metals [13]. Mines, quarries, and primary metal and mineral manufacturing facilities 

(the mining sector) are distributed across every province and territory [14]. Industrial energy prices 

increased 58% for electricity and 310% for heavy fuel oil in Canada, from 1990 to 2008 [15]; these 

increases partly illustrate the financial incentive for energy management which aims to allow 

companies to reduce economic risks resulting from rising energy costs balanced against a need for 

security of energy supply to ensure continuous production [16]. Between 1990 and 2008, total 

energy use in Canada has risen by 25.7%, with the mining industry increasing its energy 

consumption by 137.7% in the same period [15]. 

It has been reported [17] that the metal mining industry in the United States has the potential to 

reduce energy consumption by about 61% from current practice to the best-estimated practical 

minimum energy consumption. This reduction was made up of a 21% reduction by implementing 

best practices and a 40% reduction from research and development that improves energy efficiency 

of mining and mineral processing technologies. Governments, especially in countries with large 

mining sectors, are imposing standards for energy efficiency. Australia’s miners are obliged to 

comply with the Equipment Energy Efficiency program for energy efficiency. In South Africa, the 

DME set a target in 2007 for the mining industry to reduce energy demand by 15% by 2015 [11]. In 

China, a vigorous program was launched in 2004 aimed at reducing energy intensity by 20% over 

the period between 2006 and 2010 [18]. 

Energy management is the judicious and effective use of energy to maximize profits (and minimize 

costs) and enhance competitive positions [19], while meeting energy demand when and where it is 

needed (the energy utility). This can be achieved by adjusting and optimizing energy systems and 

procedures so as to reduce energy requirements per unit of output while maintaining or reducing 

total costs of producing the output from these systems [20]. 

Energy management activities are often categorized into supply side management or demand side 

management activities. As mineral production businesses are frequently vertically integrated 

businesses that hold their own generation [21] and/or transmission [22] and/or distribution assets, as 

well as maintaining control of their own demand centers, both sides of the energy system (supply 

and demand) are of concern in energy management practice. 

Demand Side Management (DSM) can be defined [23] as the planning, implementation, and 

monitoring of distribution network utility activities designed to influence customer use of electricity 

in ways that will produce desired changes in the load shape. The goal of DSM is to smooth out 

peaks and valleys in energy demand to make better use of energy resources and defer the need to 

build new power plants. 

Supply-side management (SSM) refers to actions taken to ensure the generation, transmission and 

distribution of energy are conducted efficiently [24]. Effective SSM actions will usually increase 

the efficiency with which demand centers are supplied, allowing installed generating capacity to 

provide electricity at lower cost (permitting lower prices to be offered to consumers) and reducing 

environmental emissions per unit of end-use electricity provided. 

The potential economic benefits of a high energy consumption intensity for mineral production (see 

for example [25]) lead to consideration of local (mine site) elements of supply side energy 

management. One example of supply side energy management is the adoption of energy supply 

technologies including renewable energy and polygeneration. As heat has not yet been considered 

for the mining sector in an integrated manner, polygeneration is of great interest for the sector. 

Renewables and polygeneration options are the new pathways that are subject to the current 

investigations. 

3. Polygeneration technology 

329

The increase in energy utilization efficiency is, without doubt, the main advantage of producing 

different energy services (heating, coolth, and electricity) in one installation from the same energy 

source. Furthermore, polygeneration schemes can generate many configurations and thus allow for 

ample design flexibility that accommodates specific regional conditions [26]. However, choosing 

the correct size and design of a polygeneration system is a key factor for the success of the project: 

undersized systems do not realize the profit of exploiting the whole polygeneration potential of the 

site, and if the system is oversized low or negative primary energy savings may be obtained [27]. 

In the case that the energy supply system has already been built, the optimization procedure will 

encompass only the operational strategy. However, if external conditions change (energy demands, 

utility prices, etc.), a retrofit adopting additional equipment is added to the existing system (the 

configuration of which then comprises a constraint on optimization). For new systems, in addition 

to the optimal dispatch of energy supply plant the optimal system configuration must also be 

identified (essentially the specification of equipment in rating and number) [28] [29]. 

A general framework has been established [30] [31] [32] [33] [34] to identify optimal combinations 

of energy conversion and delivery technologies, as well as operating rules for systems installed in 

tertiary sector buildings. A reference system for production of electricity, heating and coolth to 

attend the demands of a hospital is shown in Figure 1, where all electricity is purchased from a 

utility company owned electricity distribution grid to either meet the electricity demand directly or 

produce cooling in mechanical chillers driven by electrical motors, and heating demand is produced 

by a natural gas boiler. The aforementioned framework described as an energy superstructure [30] 

is shown in Figure 2, containing all technology alternatives that may be adopted (but not their 

ratings or numbers). D, S, P and W refer to, respectively, demand, sale, purchase and waste (loss) of 

a utility. Within Figure 1 and 2, the horizontal or vertical lines essentially represent physical 

distribution systems into which the technologies indicated can feed in energy of a specific form. 

Site loads for energy in that specific form (a specific energy utility) are supplied from that 

distribution system, including further energy conversion technologies that convert the supplied 

utility into another form (which in turn supplies another distribution required by the site loads). 

Figure 1 Reference system. 

330

Electricity 

distribution 

(EE) 

D 

Ep 

Es 

P 

S 

P 

Natural gas 

GT + Heat 

recovery 

Gas engine 

+ Heat 

recovery 

Steam, 180°C 

331 

Steam 

boiler 

Steam/Hot 

Water HX 

Hot 

Water 

Boiler 

Hot Water, 90°C 

Hot Water/ 

Cooling 

Water HX 

Cooling water, ambient + 5°C 

2x effect 

1x effect 

EE EE EE 

Abs 

Abs 

Mechanical EE 

Chiller 

Chiller 

Chiller 

Chilled water, 5°C 

Cooling 

tower 

Figure 2 Superstructure of energy supply system for a hospital. 

D 

Heating 

Ambient 

Air 

D 

W 

Coolth 

4. Optimal Mine Site Energy Supply – OMSES initial formulation 

For energy services superstructures such as those presented in Figures 1 and 2, following precedent 

practices [32] [34]], the decision variables for an optimization problem formulated to determine the 

optimal mine site energy supply may be characterized as follows: 

i) binary variables (denoting whether a technology is/is not installed); 

ii) integer variables (denoting the number of installed units of a technology); 

iii) continuous variables (denoting the energy flows between utilities). The latter include 

connections to distribution systems across the site boundary (to the utility company’s 

electrical and natural gas distribution systems – where appropriate) and energy flows 

between distributions and site loads. 

The constraints that may be taken to apply to the optimization process are manifold: 

1. Energy conversion technology constraints. These are constraints that reflect an energy 

conversion mass and energy balance across each technology of a specific type. A graphical 

example representation of such a constraint is presented in Figure 3 which shows the energy 

(and inferred mass) balance across a gas turbine equipped with a heat recovery unit. 

2. Technology installation limit constraints. These constraints apply a threshold on the 

maximum number of units of a given technology type that may be installed. Such a 

constraint is useful to reflect practical considerations such as the amount of land footprint 

available to accommodate technology of a specific type. This also articulates a capacity 

constraint for the specific technologies. 

3. Utility balancing off constraints. These are constraints that ensure that the net sums of 

energy flows from each of the indicated distributions are zero, which also ensure that energy 

supply (in all its forms) meets on-site demand. In the event that hourly on-site demand data

are available for an entire year, 8760 such constraints feature in the problem formulation, for 

a specific utility. In practice, various heuristics may be deployed to reduce the number of 

these constraints (e.g., considering only 12, 24 hour periods, each representative of a typical 

day in a typical month – see Figure 4). 

4. Carbon dioxide equivalent emissions constraints. These are constraints that express the idea 

that the total emissions associated with the production of energy from the energy services 

system falls below a given threshold. In prior formulations [e.g., [32] [33]], such emissions 

include those attributable to the materials (manufacture and installation of the technologies 

at the site) and operation of the system. 

5. Energy market constraints. These include inter alia thresholds on the quantities of energy 

that may be procured externally, exported off-site, or wasted. 

Figure 3: Showing (LHS) a cut-away diagram of a Capstone C600 gas turbine [35] and (RHS) the 

energy flows for this technology normalized to the electrical energy output. 

The objective function for the optimization process is defined in economic terms, expressed as a 

minimum total cost of meeting the on-site energy demand. Various components are considered in 

its formulation. Principally, the total cost is decomposed into an annuitized fixed cost element and a 

variable cost element. 

The fixed cost element thus accounts for the discounting process and the time value of money, as 

well as the replacement / major overhaul interval of the technology. The variable cost element is 

primarily expressed in terms of the unit cost of procurement of energy in a particular form, supplied 

to the site, multiplied by the volume of that energy used in a specified time period. Through 

constraints taken to apply in the optimization process, this cost element may reflect particular 

operational strategies imposed, for example, a user-defined decision to operate bespoke cogeneration 

plant components at full load. 

As it is within the scope of the energy services infrastructure to export energy in various forms off 

the site, revenues arising from any such sales are treated as negative costs. 

Given the nature of the decision variables indicated, in prior deployments Mixed Integer 

Programming (MIP) has been found effective in establishing optimal configurations and equipment 

operating strategies, this technique having been broadly applied in production planning, sequencing 

processes, distribution and logistics problems, refinery planning, power plant scheduling, and 

process design. MIP captures the complexity of polygeneration systems in synthesis problems such 

as that described for optimal on-site energy supply. Formulation of the problem in MIP compatible 

terms presents opportunities to benefit from significant recent advances in the mathematical 

programming field such as those found in [36]. 

332

5. Characteristic challenges of OMSES 

Mining operations in Northern Canada can also face a particular energy challenge given the lack of 

grid (electric and gas) capacity and limited infrastructure. In addition, it is typical for underground 

mines that they require progressively more energy to access and extract the minerals as they mine 

deeper. Deep mines (nominally >2000 m) acquire substantial cooling loads as they age, and 

meeting such important energy demands is becoming economically and technically important. 

Opportunities for savings offered by adoption of polygeneration have not been investigated in depth 

previously for the specific cases of mine sites, possibly for the following reasons: i) wide variety of 

technology options for the provision of energy services; ii) temporal variations (diurnal, seasonal 

and inter-annual) in energy prices; iii) temporal variations (diurnal, seasonal) in climate; iv) 

temporal variation (diurnal, seasonal and inter-annual) in energy demand; and v) institutional 

inertia, and conservatism in mine design for mines. 

Diurnal, seasonal and inter-annual variability of thermal loads in the mining sector in Canada 

increase the complexity of a generic energy supply systems and solutions of high sophistication are 

required for operation to be economically attractive. Mine air heating and cooling loads are 

seasonally complementary (see Figure 4). The timing of the highest cooling loads in summer 

seasons, if serviced by a mechanical chiller system, is at odds with electricity tariff peak times. 

Figure 4 Heating and cooling demand profile projections for a 2500 m deep mine located in Canada 

at latitude 46 North. 

In Figure 4, thermal demand profile projections are illustrated for a study period of one year, 

distributed in 12, each day taken as typical for each month and divided into 24 hourly periods (at 

2500 m depth, base temperatures were: for heating = 18°C; for cooling = 30°C). 

Applying optimization techniques to the problem of mine-site energy supply presents some unique 

challenges: variability of energy loads which will always remain variable, but may be highly 

predictable (e.g., those due to winding and groundwater pumping activity in the case of electricity), 

the need to produce from deeper, hotter levels, as in Canadian climates with extreme climatic 

variation, and in remote areas with no connection to the electric or gas grids. These characteristics 

are not seen as insurmountable and offer potential for innovation as indicated below. 

333

6. Extensions to core methodology 

6.1 Integration of renewable energy technologies 

Primary energy inputs from new and renewable energy resources and technologies are considered, 

with characteristics of intermittency and variability, alongside conventional energy supply 

technologies, as shown in Figure 5. Potential advantages are set out by Trapani & Millar in [37], for 

remote mining operations. Renewable energies are introduced as available utilities in the synthesis, 

design and operation of energy systems, which is generically applicable to renewable energy 

integration studies for other industries too. 

P 

Electricity 

distribution 

(EE) 

Electricity 

Storage 

Wind 

Turbine 

PV 

Hydro 

electricity 

D 

P 

P 

S 

P 

P 

P 

Biomass (gaseous) 

Biomass (liquid) 

Biomass (solid) 

Fuel oil / Diesel 

Natural gas 

GT + Heat 

recovery 

Fischer- 

Tropsch 

Gas engine 

+ Heat 

recovery 

334 

Gasifier 

Steam 

boiler 

Steam/Hot 

Water HX 

Hot Water, 90°C 

Steam, 180°C 

Cooling water, ambient + 5°C 

2x effect 

1x effect 

EE EE EE 

Abs 

Abs 

Mechanical EE 

Chiller 

Chiller 

Chiller 

Gas 

burner 

Heat 

exchanger 

Wgt 

Wge 

Wgt 

EE 

Wgt 

EE 

Diesel 

engine 

stationary 

Fan 

Compres 

sor 

Wde 

Liq Biom 

Storage 

Fuel 

Storage 

Hot Water 

Boiler 

Hot Water/ 

Cooling 

Water HX 

Cooling 

tower 

Chilled water, 5°C 

Wde 

Wge 

Ventilation air 

Wde 

Wge 

Heat 

exchanger 

Figure 5 Superstructure of an energy supply system for a mine. 

Motive 

Power 

D 

Steam 

D 

Heating 

D 

Ambient 

Air 

D 

L 

Coolth 

D 

Intake 

ventilation air 

D 

Compressed 

air @ 15 bar

6.2 Integration of energy storage technologies 

Energy storage systems will also be considered as utilities on the supply side and as dispatchable 

loads on the demand side. On-site energy storage technologies are included to compensate for the 

variable and intermitent characteristics of renewable energy sources. Integration of storage 

technologies into the energy supply optimization process may introduce less constraint into the 

resulting energy supply system, and consequently could lower energy supply cost, equivalent CO2 

emissions, or both. Many of the technical challenges in reformulating the mathematical 

optimization procedures to accommodate intermittent and variable renewable energy supply 

utilities, may be reapplied in consideration of energy stores acting as energy supply components. 

6.3 Part load operation of all technologies 

A common practice to facilitate the operation of a system is to consider that cogeneration modules 

operate at full load when in service. This study will go a step further by considering that energy 

balance data varies with load conditions for all technologies. As a consequence of considering part 

load operation, a more heterogeneous range of technologies may emerge in optimal solutions. 

6.4 Consideration of work from prime movers as a new utility 

On mineral production sites, some processes that form part of the electricity base load 

fundamentally constitute a demand for energy in the form of work (Wde, Wgt, and Wge in Figure 5) , 

which could be met by the prime mover of a polygeneration system. Examples of the latter could 

include surface pumping, production of compressed air, and the running of main ventilation fans on 

surface. 

6.5 Work in progress 

A database and models of the potentially installable equipment is currently being adapted through 

the addition of equipment with larger rating (commensurate with that needed for large industrial 

applications, such as mining). It contains investment, operation, performance and environmental 

information on each technology that may be considered. The latter informs Life Cycle Analysis that 

sets and coordinates criteria in the multiobjective optimization (economic/environmental trade-off). 

Energy demands are characterized for mine sites, and the collaborators will also supply data on the 

purchase/sale tariffs of the different utilities (when available). The exemplar consumer centers to be 

used in this extension are located in Northern Ontario (Canada), with energy demands varying 

seasonally and diurnally. The consumer centers are mineral productions operations that are 

connected to energy supply infrastructure and those that are not. 

7. Closure 

This contribution outlines the priorities of investigation, development and demonstration of new 

concepts and technologies to improve energy efficiency and reduce final consumption of primary 

energy in the mining sector, considering the life cycle holistically. Due to the energy-intensive 

nature of mining operations, energy is a significant component of total operating costs and a 

reduction in energy consumption may directly result in cost savings. It is expected that energy 

procurement costs will increase in the future adding to the incentive to effectively manage energy 

[12] in the sector. 

The analysis of thermal process integration has contributed to the improvement of the efficiency of 

cogeneration systems used in the industry sector, and its application to polygeneration systems with 

335

thermal storage and possible support of renewable energies will reveal the most cost effective or 

low carbon configuration and operational strategy. Scenarios with substantial economic potential in 

which renewables are advantageous, alone or in combination with cogeneration systems have been 

identified, while a priori both are energy production systems that compete. 

The design techniques developed will facilitate the collaboration of equipment manufacturers in the 

development and commercialization of modular systems. The development of new products 

(modular systems) and the techniques involved will contribute to a greater competitiveness of the 

participating companies and, what is equally important, to decrease the energy costs of the 

consumer center. 

Although the research is applied and has a Canada focus, in terms of the case studies adopted, the 

work is of global scientific importance. The local dimension, even the specific industry focus, is just 

a way in which the relevance of the science will be demonstrated. Theoretical extensions (storage, 

renewables) are of generic applicability for all integration studies. The mining sector is just used as 

an example to drive and exemplify the methodological development process. 


This work was developed within the framework of research projects Smart Underground 

Monitoring and Integrated Technologies for Deep Mining (SUMIT, funded by Ontario Research 

Fund for Research Excellence Funding - Round 5), Optimal Mine Site Energy Supply (OMSES, 

funded by Natural Sciences and Engineering Research Council of Canada through an Industrial 

Research & Development Fellowship), and Low Carbon Mine Site Energy Initiatives 

(LOWCARB, funded by the Research Fund for Coal & Steel - European Commission). Thanks are 

extended to DeBeers Canada, Hearst Community, and the Ontario Geological Survey (unit of 

Ministry of Northern Development and Mines). 

References 

1. NRC. Evolutionary and revolutionary technologies for mining. National Research Council, 

Committee on Technologies for the Mining Industries. Washington, U.S. 2001. 

2. LIU, P.; GEORGIADIS, M. C.; PISTIKOPOULOS, E. N. Advances in Energy Systems 

Engineering. Ind. Eng. Chem. Res., v. 50, n. 9, p. 4915–4926, 2011. 

3. MILLAR, D. L. et al. Enabling advanced energy management practice for minerals operations. 

Montreal: Proceedings of the Annual Canadian Institute of Mining, Metallurgy and 

Petroleum (CIM) Conference and Exhibition. 2011. 

4. SERRA, L. M. et al. Polygeneration and efficient use of natural resources. Energy, v. 34, p. 

575-586, 2009. 

5. STUICKEL, J. J. Industrial and Commercial Cogeneration. Office of Technology Assessment 

Arquive, 1983. See also: . Accessed on: 11 Jan 

2012. 

6. ISO50001. Energy Management Systems - requirements with guidance. International 

Organization for Standardization (ISO). Geneva. 2011. 

7. MAGLORIE, P.; HETEU, T.; BOLLE, L. Economie d´énergie en trigénération. International 

Journal of Thermal Sciences, v. 41, p. 1151-1159, 2002. 

8. CHICCO, G.; MANCARELLA, P. Distributed multi-generation: A comprehensive review. 

Renewable and Sustainable Energy Reviews, v. 13, p. 535-551, 2007. 

336

9. LOZANO, M. A.; CARVALHO, M.; SERRA, L. M. Allocation of economic costs in 

trigeneration systems at variable load conditions. Energy and Buildings, v. 43, n. 10, p. 

2869-2881. 

10. CARMEL, Y.; BEN-HAIM, Y. Info-gap robust-satisficing model of foraging behavior: do 

foragers optimize or satisfice? American Naturalist, v. 166, p. 633-641, 2005. 

11. MCIVOR, A. Energy in Mining. Cleantech (special edition fuel cells), v. 5, p. 16-19, 2010. 

12. EA. Energy efficiency & greenhouse gas reduction. Environment Australia, 2002. See also:: 

. Accessed on: 06 Jan 2012. 

13. MAC. 

http://www.miningworks.mining.ca/miningworks/media_lib/documents/Facts_Figures_200 

7_E.pdf. Mining Association of Canada - Facts & Figures. A report on the state of the 

Canadian Mining Industry., 2007. Acesso em: 5 jan. 2012. 

14. NRC. Mining sector performance. Natural Resources Canada - Final report of the Federal, 

Provincial and Territorial social licence task group., 2010. See also:: 

. 

Accessed on: 05 jan. 2012. 

15. NRC. Industrial energy prices and background indicators. Natural Resources Canada, 2010. See 

also: 

. 

Accessed on: 10 Jan 2012. 

16. LEVESQUE, M. Understanding mine heat. M.A.Sc. dissertation (Natural Resources 

Engineering). School of Graduate Studies, Laurentian University. Sudbury, Canada, p. 140. 


17. NORGATE, T.; HAQUEA, N. Energy and greenhouse gas impacts of mining and mineral 

processing operations. Journal of Cleaner Production., v. 18, p. 266-274, 2010. 

18. FANG, H.; WU, J.; ZENG, C. Comparative study on efficiency performance of listed coal 

mining companies in China and the US. Energy Policy, v. 37, n. 12, p. 5140-5148, 2009. 

19. CAPEHART, B.; TURNER, W. C.; KENNEDY, W. J. Guide to energy management. 5th 

edition. ed. [S.l.]: The Fairmont Press, 2005. 

20. ABDELAZIZ, E. A.; SAIDUR, R.; MEKHILEF, S. A review on energy saving strategies in 

industrial sector. Renewable and Sustainable Energy Reviews., v. 15, n. 1, p. 150-168, 


21. INCO, V. Vale Inco Limited. 100 years of Clean Renewable Power. See also:: 


22. LARMOUR, A. Empowering remote communities. Northern Ontario Business, v. 27, n. 4, p. 9, 

2007. 

23. QURESHI, W. A.; NAIR, N. C.; FARID, M. M. Impact of energy storage in buildings on 

electricity demand side management. Energy Conversion and Management., v. 52, n. 5, p. 

2110-2120, 2011. 

24. UNIDO. Suply-side management. United Nations Industrial Development Organization - 

Sustainable energy regulation and policymaking for Africa. See also: . 

Accessed on: 10 Jan 2012. 

25. MUDD, G. M. Global trends and environmental issues in nickel mining: Sulfides versus 

337

laterites. Ore Geology Reviews, v. 38, n. 1-2, p. 9-26, 2010. 

26. LIU, P.; GEROGIORGIS, D. I.; PISTIKOPOULOS, E. N. Modeling and optimization of 

polygeneration energy systems. Catalysis Today, v. 127, p. 347-359, 2007. 

27. EUROPE, C. Deploying large-scale polygeneration in industry (D-PLOY), 2008. See also:: 

. Accessed on: 11 

Jan 2012. 

28. YOKOYAMA, R. et al. Development of a General-Purpose Optimal Operational Planning 

System for Energy Supply Plants. Journal of Energy Resources Technology, v. 16, p. 290- 

296, 1994. 

29. LOZANO, M. A.; RAMOS, J. C.; SERRA, L. M. Cost optimization of the design of CHCP. 

Energy , v. 35, n. 2, p. 794 - 805, 2010. 

30. CARVALHO, M. Thermoeconomic and environmental analyses for the synthesis of 

polygeneration systems in the residential-commercial sector. Zaragoza: PhD thesis; 

Department of Mechanical Engineering, University of Zaragoza, Spain, 2011. 

31. CARVALHO, M.; LOZANO, M. A.; SERRA, L. M. Multicriteria synthesis of trigeneration 

systems considering economic and environmental aspects. Applied Energy, v. 91, n. 1, p. 

245-254, 2012. 

32. CARVALHO, M.; SERRA, L. M.; LOZANO, M. A. Optimal synthesis of trigeneration systems 

subject to environmental constraints. Energy, v. 36, n. 6, p. 3779-3790, 2011. 

33. CARVALHO, M.; SERRA, L. M.; LOZANO, M. A. Geographic evaluation of trigeneration 

systems in the tertiary sector. Effect of climatic and electricity supply conditions. Energy, v. 

36, n. 4, p. 1931-1939, 2011. 

34. LOZANO, M. A. et al. Structure optimization of energy supply systems in tertiary sector 

buildings. Energy and Buildings, v. 41, n. 10, p. 1063-1075, 2009. 

35. CORPORATION, C. T. Capstone product catalog, P0112 CAP11. Chatsworth, CA, U.S. 2010. 

36. GROSSMANN, I. E.; CABALLERO, J. A.; YEOMANS, H. Advances in mathematical 

programming for the synthesis of process systems. Latin American Applied Research, v. 

30, n. 4, p. 263-284, 2000. 

37. TRAPANI, K.; MILLAR, D. L. Powering mines in the McFauls lake area using large scale 

arrays of floating PV panels. Proceedings of the Mining and Environment V Conference. 

Sudbury, Canada: 2011. 

38. WUA, L.; KANEKOB, S.; MATSUOKAB, S. Dynamics of energy-related CO2 emissions in 

China during 1980 to 2002: The relative importance of energy supply-side and demand-side 

effects. Energy Policy, v. 34, n. 18, p. 3549–3572, 2006. 

39. STRICKLAND, C.; NYBOER, J. Cogeneration Potential in Canada Phase 2. Natural Resources 

Canada. Ottawa. 2002. 

40. MCKINSEY&CO. Reducing US greenhouse gas emissions: How much at what cost? 

McKinsey & Company, 2007. See also: 


41. IEA. Cogeneration and District Energy. International Energy Agency. Paris, p. 60. 2009. 

42. HOUWING, M. et al. Uncertainties in the design and operation of distributed energy resources: 

The case of micro-CHP systems. Energy, v. 33, p. 1518-1536, 2008. 

43. FISKEL, J. Designing Resilient, Sustainable Systems. Environmental Science and Technology, 

338

v. 37, p. 5330-5339, 2003. 

44. DOE. International Energy Outlook 2011. U.S. Department of Energy - U.S. Energy 

Information Administration, 2011. See also: 


339

Abstract: 




Simulation of Synthesis Gas Production from 

Steam Oxygen Gasification of Colombian 

Bituminous Coal using Aspen Plus ® 

John Ortiz a , Juan González b , Jorge Preciado c , Rocío Sierra d and Gerardo Gordillo e 

a Universidad de los Andes, Bogotá, Colombia: jj.ortiz24@uniandes.edu.co, 

b Universidad de los Andes, Bogotá, Colombia, jc.gonzalez141@uniandes.edu.co 

c Universidad de los Andes, Bogotá, Colombia, je.preciado40@uniandes.edu.co 

d Universidad de los Andes, Bogotá, Colombia, rsierra@uniandes.edu.co 

e Universidad de los Andes, Bogotá, Colombia, g.gordillo43@uniandes.edu.co 

The growing energy demand, sustained increase in fossil fuel prices, and need for environmental protection 

have forced to consider alternatives to the traditionally energy resources employed. Synthesis gas (syngas) 

produced by gasification of coal, biomass, petroleum coke, or solid waste has proven to be a useful clean 

fuel due to its lower emissions of sulfurs and nitrous oxides compared to other fuels. In this work, a 

simulation of syngas production from a Steam-Oxygen Gasification (SOG) process that uses the GE/Texaco 

technology was performed using Aspen Plus ® . For the simulation, the average proximate and ultimate 

compositions of bituminous coal obtained from the Colombian Andean region were employed. The obtained 

simulation was applied to conduct sensitivity analyses in key parameters. The information obtained allows 

the selection of critical operating conditions leading to improve system efficiency and environmental 

performance. The results of the parameter analysis indicate that the oxygen to carbon ratio is a key variable 

as it affects significantly both the LHV and thermal efficiency of the process. On the other hand, the process 

becomes almost insensitive to SDG values higher than 2. Finally, a thermal efficiency of 62.2% can be 

reached. This result corresponds to a slurry solid concentration of 0.65, a WGS process SDG of 0.59, and a 

LTS reactor operating temperature of 473 K. At these fixed variables, a syngas with H2 molar composition of 

92.2% and LHV of 12 MJ N m -3 was attained. 

Keywords: 

Aspen Plus ® Simulation, Coal, Steam-Oxygen Gasification, Synthesis Gas. 


Currently, in a world of high and growing energy demands and increasing oil prices, alternative and 

sustainable raw material resources are being sought. Ideally, these sources would be used for either 

energy generation or as valuable chemical feedstocks for variety of chemical processes and 

synthesis. Among different alternatives, syngas has shown to be a favorable option. Syngas is 

considered a clean fuel with environmental advantages compared to other fossil fuels because the 

sulfur oxides (SOX), nitrous oxides (NOX) and CO2 emissions are considerably lower [1, 2]. 

Syngas is an important building block in chemical, oil and energy industries due to its applications: 

1) as a feedstock for the production of several chemicals such as hydrogen, ammonia, methanol, and 

Fischer-Tropsch products [3], 2) as a fuel in a gas turbine to produce electricity [4], 3) as a cell fuel 

for mobile sources [5], 4) as an electricity supplier through solid oxide fuel cells [3], and 5) as a 

primary fuel. Syngas, which is mainly a mixture of hydrogen (H2) and carbon monoxide (CO), is 

mainly used as a chemical substance rather than a fuel, representing 50 billion US$ market for 40 

Tg (40 Mt) annual production nowadays [6]. Hydrogen could help to satisfy the world energy 

demand. Recent reports show that global energy consumption grew 5.6% in 2010, the largest yearly 

increase since 1973 [7]. 

340

There are many alternatives for hydrogen production from liquid and gaseous hydrocarbons such as 

thermo-catalytic cracking, steam reforming and plasma arc decomposition [6]. Moreover, from 

solid feedstocks, H2 can be produced through the gasification of coal, biomass, petroleum coke, or 

solid waste. Nearly 50% of the global hydrogen is generated through natural gas reforming, 30% 

from oil/naphtha reforming, 18% from gasification, 3.9% from water electrolysis and 0.1% from 

other sources [6]. Coal gasification is a promising way to obtain H2 because the production 

techniques have achieved maturity and are commercially available. Moreover, the relatively high 

global resources of coal and its widespread availability worldwide make his resource a promising 

option [8]. In addition, this process has environmental advantages: 1) SOX can be processed into a 

marketable by-product, 2) ash can be liquefied into a slag that passes toxicity issues, 3) CO2 can be 

held and recovered in the loops of gasifiers for remediation/reuse, and 4) gasifiers can be modified 

such that wide product flexibility is easily obtained [9]. 

The steam-oxygen gasification (SOG) process is the only commercialized method of gasification 

used to manufacture several chemicals from coal. The Wabash River Coal Gasification Repowering 

Plant, near to West Terre Haute, Indiana (USA), has proven since November of 1995 the successful 

application of H2 production by coal gasification. This plant uses H2, from SOG process, in a gas 

combustion turbine generator to produce electricity. It generates around 292 MW of electric power. 

With this production, this plant is one of the largest single-train gasification combined cycle plants 

operating commercially in the world [10]. 

Table 1. Ultimate and proximal analysis of Guaduas Formation’s coal (HHV = 30,634 kJ kg -1 ) data 

taken from [11] 

w/w (%) 

Proximate analysis 

Moisture 4.12 

Ash 5.61 

Fixed carbon 67.8 

Volatiles 

Ultimate analysis* 

22.4 

Carbon 70.7 

Hydrogen 5.29 

Nitrogen 1.58 

Chloride 2.35 

Sulfur 1.57 

Ash 5.61 

Oxygen 7.91 

*dry basis 

Two thirds of the total fuel fossil reserves in the world are coal and will last for more than 150 years 

[12]. Coal is in fact one of the main resources in Colombia. It is estimated that 0.7% of the world 

proved coal reserves, which corresponds to 6.7 Pg (6700 Mt), are in Colombian territory [11]. 

Colombia has several coal formations over its territory. The main ones are: Cerrejón, Los Cuervos, 

Guaduas, Umir, Cerrito, and Amagá. The Guaduas formation’s coal, located in the center of 

Colombia, is characterized by a bituminous coal with high volatiles and low sulfur and ash content 

(Table 1). which is advantageous for a gasification use [13]. Therefore, coal from Guaduas 

formation was selected for this study. 

Aspen Plus® has been widely employed to simulate chemical processes in a wide number of fields 

including but not limiting to the petroleum industry, chemical processes and biomass gasification. It 

also can be used to model steady state processes handling solid carbons materials in multiple unit 

operations. Therefore, many coal and biomass conversion processes have been simulated using 

Aspen Plus as integrated coal gasification combined cycle (IGCC) power plant [14], biomass 

341

gasification [15], hybrid biomass gasification [16], hydrogen production from biomass gasification 

[17], and coal combustion [18]. Additionally, proximate and ultimate analysis properties of solid 

coal are specified to provide a fairly rigorous simulation of the gasifier performance [19]. 

The purpose of this study is to simulate and analyze through Aspen Plus the coal gasification 

process and subsequent processing for the hydrogen-rich syngas production, using the most 

commercialized and referenced available technologies. A sensitive analysis of the variables with 

high impact over the key process parameters is performed to identify important process efficiency 

improvements (yield and energy) and environmental performance. 

1.1. Gasification Technologies 

There are three main types of coal gasification technologies: fixed-bed, fluidized-bed, and 

entrained-flow gasification. Table 2 summarizes key parameters for these gasification technologies. 

Among these processes, entrained-flow gasification is the commercially preferred technology due to 

its versatility and lower environmental impact [4, 13, 14, 20] . 

Table 2. Main features of industrial gasifiers 

Gasifier type Main features 

Entrained-flow Particle size below 0.1 mm 

High operating temperature (> 1473 K) 

High operating pressure (3 to 12 MPa) 

High oxidant demand 

Short residence time (0.5 to 10 s) 

Ash is removed as molten slag 

Fluidized-bed Particle size between 6 and 10 mm 

Uniform temperature distribution 

High operating temperature (1073 to 1323 K) 

Lower carbon conversion 

Ash is removed as slag or dry 

Fixed-bed Coarse particles (6 to 50 mm) 

Low operating temperature (698 to 1088 K) 

Low oxidant demand 

Resident time above 600 s 

Ash is removed as slag or dry 

Many commercial technologies in entrained-flow gasification reactors are available nowadays such 

as: GE/Texaco, Shell, and ConocoPhillips. GE/Texaco and Shell entrained-flow gasification 

reactors are used in about 75% of the gasification plants throughout the world [13]. In this study, 

the GE/Texaco gasifier has been selected because: 1) it is profusely discussed in the literature, 2) 

high coal conversion is reported, and 3) the resulting syngas is free of tars, phenols and paraffins. 

Additionally, the GE/Texaco gasifier is leader worldwide with 145 reactors in commercial 

operation and 85 in planning, engineering, or under contract agreements in 15 different countries 

[21]. 

2. Process Description 

In the SOG process, coal-water slurry is gasified with O2 from the air separation unit (ASU) to 

produce a gas mainly composed by CO and H2. It is necessary to increase the H2 concentration by a 

342

sour water-gas shift (WGS) process followed by an acid gas removal. H2-rich syngas is obtained 

after water condensation in the resulting gas. The SOG simplified process flow diagram is shown in 

Fig. 1: 

Fig. 1. SOG simplified process flow diagram. 

There are many technologies for separating air into its main components. The application of either 

one depends on the process requirements. For lower volumes of O2 and/or N2 (< 1.6 kg s -1 ), 

pressure swing adsorption or membrane processes are preferred [22]; whereas, for producing large 

quantities of gaseous products, cryogenic air separation technology is currently the most efficient, 

especially when high purity products are required [23]. 

The cryogenic process consists on several unit operations that compress, purify and separate air into 

its principal components. First, impurities (H2O, CO2, among others) are removed in a prepurification 

unit, located downstream of the air compression. Secondly, the air is cooled down to 

cryogenic temperatures (from 123 K to 463 K, depending on the operating pressure [24]) and goes 

into the air separation unit. Then, a multi-column cryogenic distillation is usually used for 

separating O2 and N2 [22]. Several configurations of rectifying columns and heat exchangers are 

made according to the requirements of the process [25]. 

A double column system is widely used in air separation processes. Air enters into the high pressure 

column (HPC) and provides two reflux streams that feed the low pressure column (LPC) [26]. At 

the top of the LPC, a pure gaseous nitrogen stream is obtained while liquid oxygen is evaporated at 

the bottom of this column to deliver a pure oxygen stream [25]. The two columns are built in a 

single tower for the commercial application, considering the use of a condenser-reboiler as a heat 

exchange unit [24]. 

The gasification process is developed using a GE gasifier with a gas-water quench system. Guaduas 

formation coal is wet-milled to a particle size about 100 µm and mixed with water to produce 

slurry. Coal slurry and O2 stream from the ASU unit are fed in the top of the pressurized reactor 

through burners. The coal reacts exothermally with O 2 at high temperature (> 1473 K) and high 

pressure (> 7 MPa) to produce syngas and slag [27]. The hot gas is contacted directly with water 

where the slag is solidified. The quenching process cools the syngas and generates a water-saturated 

gas product, leaving the quench chamber at a temperature between 473 K and 573 K. The resulting 

syngas is mainly free of particulate matter and water-soluble contaminants such as NH3, HCN and 

chlorides [20]. 

To increase the H2 concentration, the WGS process is employed to convert mostly CO into H2. This 

process consists of two reactors in series with intercooling. A high temperature (HTS) reactor (573 

K - 873 K) as an initial stage followed by a low temperature (LTS) reactor (453 K – 523 K). The 

HTS reactor feed is heated by the effluent of the LTS to control the operating temperature. 

Additionally, the effluent of the HTS is cooled producing high pressure steam and then it is fed in 

343

the second reactor. In this unit, syngas and steam are mixed with a steam to dry gas ratio (SDG) 

depending on the feed syngas water content and the required H2 to CO ratio. 

In the WGS reaction, chemical equilibrium favors products at low temperature; therefore, a catalyst 

is required to enhance the reaction rate. A catalyst typically made of sulfided Co/Mo on aluminum 

support reacts with the sulfurs, producing metal sulfides which activates the catalyst [28, 29]. 

Carbonyl sulfide (COS) is converted to H2S making the sulfur removal easier due to the WGS 

process location before the acid-gas removal process. 

For conditioning of the gas leaving the LTS, the Rectisol process is used. It employs methanol 

(CH3OH) as solvent to clean up the syngas. The high selectivity of methanol for H2S over CO2 at 

low temperatures (211 K to 233 K) and the ability to remove COS are the main advantages of the 

process. Besides, it allows a deep sulfur removal (< 0.1 ppmv H2S + COS) [30]. 

There are many possible process configurations for Rectisol, depending on the process 

requirements. A selective H2S removal configuration was used in the simulation. In this 

configuration, the raw syngas feeds up the main absorber in which CH3OH absorbs most of the 

impurities produced in gasification process such as CO2, H2S, COS, HCN and NH3 [31]. Thereafter, 

the solvent passes through a regeneration process, where these components are desorbed by 

reducing the pressure, stripping and/or boiling up the solvent. The regenerated and recirculated 

solvent is free of sulfur compounds but still contains some CO2. The acid gas leaving the solvent 

regeneration units is suitable for the Claus process [32]. 

3. ASPEN PLUS ® Simulation Model 

In order to model the process, the following assumptions were considered: 1) the process is in 

steady state, 2) the coal feed flow rate is 12500 (kg h -1 ), 3) the reactors are perfectly insulated, 4) 

heat losses are neglected, and 5) coal tar is not modeled; char only contains carbon and ash. Main 

unit operations modeled in Aspen Plus ® are shown in Table 3: 

Table 3 Main blocks used in the process 

Aspen 

Unit operation Plus Comments/specifications 

model 

ASU RadFrac LPC: Rigurous distillation model, first stage to separate N2 and O2. 

SN 40, RR 12.3, BR 41.3, partial-vapor condenser, TSP 0.14 MPa, 

CPD 0.005 MPa. 

HPC: Rigurous distillation model, second stage to separate N2 and 

O2. SN 26, RR 0.5, BR 1.0, partial-vapor condenser, TSP 0.6 MPa, 

CPD 0.05 MPa. 

Coal Gasification RGibbs Specification of the possible products: CO, CO2, C, H2, H2O, CH4, 

SO2, H2S, S, CS2, COS, N2, NH3, HCN, O2, NO2, NO3. 

HTS reactor REquil Specification of the stoichometric reactions. OP 3.8 MPa, OT 623 K. 

LTS reactor REquil Specification of the stoichometric reactions. OP 0.5 MPa, OT 473 K. 

CH3OH absorber Radfrac Rigorous absorption of H2S, SO2, COS, NH3, HCN. SN 10, TSP 3.2 

MPa. 

SN: Stage number; RR: Reflux ratio; BR: Boil up ratio; TSP: Top stage pressure; CPD: Column pressure drop; OT: 

Operating temperature; OP: Operating pressure. 

3.1. Physical Property Method 

344

The Soave-Redlich-Kwong equation of state with Kabali-Danner mixing rules (SRKKB) was 

selected to calculate all thermodynamic properties for the conventional components in the overall 

process. Additionally, the HCOALGEN and DCOALIGT models were used to calculate enthalpy 

and density for coal and ash (non-conventional components) [14]. 

3.2. Chemical Reactions 

Gasification reactions occur above 873 K; at this temperature or higher, the kinetic barrier is 

minimized and reactor products are found around equilibrium. Therefore, in the simulation, a free 

kinetics model was implemented [33]. In this model, the equilibrium approach was employed by 

neglecting the hydrodynamic complexity of the gasifier. Gasification products are estimated 

employing the RGibbs model which uses Gibbs free energy minimization to calculate the chemical 

equilibrium of a list of conventional components. The gasification products are taken from the most 

important coal gasification reactions (Table 4) [18]. As RGibbs only estimate chemical equilibrium 

of conventional compounds, it is necessary to decompound solid coal (a nonconventional 

compound) on its constituting components. This is done by using the RYield model and specifying 

the yield distribution according to the Guaduas coal ultimate analysis. By this approach, satisfactory 

results have been obtained for many researchers from gasification simulation using Aspen Plus ® 

[14, 16, 34, 35]. 

The HTS and LTS reactors are simulated using the REquil model. The WGS reaction (R4) and COS 

hydrolysis (R14) are obtained specifying the stoichiometric reactions [4]. 

Table 4. Main process reactions 

Reaction Reaction name 

4. Sensitivity Analysis 

345 

Heat of reaction 

(kJ mol -1 ) 

Carbon combustion -393 R1 

Carbon combustion -221 R2 

Boudouard +173 R3 

Steam gasification +131 R4 

Water gas Shift -412 R5 

Steam reforming - 206 R6 

, Methanation -165 R7 

Sulfur combustion - 297 R8 

H2S formation -207 R9 

Reaction 

number 

CS2 formation +115 R10 

COS formation +63 R11 

NH3 formation -46 R12 

NO2 formation +66 R13 

COS hydrolysis -34 R14 

Sensitivity analysis was performed with the aim to analyze and optimize overall operating 

conditions in the process. 

The chosen variables were: 1) O2 to coal mass ratio, 2) mass solid concentration in coal slurry, 3) 

LTS reactor operating temperature, and 4) steam to dry gas molar ratio (SDG) in the WGS process.

The variables effect was evaluated over the next key process parameters: 1) syngas molar 

composition upstream and downstream the WGS process, 2) overall CO conversion in the WGS 

reactors, 3) lower heating value (LHV) of H2 rich-syngas, and 4) thermal efficiency ( ). 

4.1. Thermal efficiency 

As the best performance, which is also the most economic option, is sought; this discussion starts 

showing the results obtained during the sensitivity analysis for thermal efficiency ( ). This is an 

indicator of the overall process performance [14]. Thermal efficiency was calculated considering 

the hydrogen-rich syngas output energy divided by the thermal energy of the coal used as raw 

material and the energy requirements for auxiliary equipment (ASU, Rectisol, etc) as follows: 

Table 5. Variables effect on and H2-rich syngas LHV 

Variable TE, % LHV, MJ kg -1 LHV, MJ Nm -3 H2 molar fraction in H2rich 

syngas 

O2 to carbon ratio † 

0,160 34,1 55,5 20,9 0,561 

0,320 42,2 69,0 15,1 0,806 

0,480 52,0 79,8 13,0 0,895 

0,640 62,6 83,4 12,0 0,922 

0,800 60,1 92,7 10,8 0,977 

0,960 54,5 97,3 10,7 0,983 

Coal slurry concentration (% w/w) ‡ 

86,21 61,3 61,8 10,8 0,926 

75,47 60,4 87,0 10,8 0,971 

65,01 59,9 95,1 10,8 0,979 

56,34 59,2 97,0 11,0 0,974 

50,00 58,4 93,4 11,5 0,958 

LTS reactor temperature (K)** 

453,15 59,5 99,3 10,7 0,983 

473,15 59,9 95,1 10,8 0,979 

473,25 59,9 95,0 10,8 0,979 

498,15 60,4 88,6 10,8 0,971 

523,15 61,5 81,5 10,8 0,962 

SDG gas molar ratio in WGS †† 

0,694 59.9 95,1 10,8 0,979 

0,972 58.6 102,5 10,8 0,986 

1,768 55.3 106,3 10,8 0,989 

2,564 52.4 107,3 10,8 0,990 

3,360 49.8 107,7 10,8 0,990 

3,917 48.2 107,9 10,8 0,991 

† Solid concentration in coal slurry: 0.65, WGS process SDG : 0.59, LTS reactor operating temperature: 473 K and HTS 

reactor operating temperature: 623 K. 

‡ 

O2 to coal ratio: 0.8, WGS process SDG : 0.59, LTS reactor operating temperature: 473 K and HTS reactor operating 

temperature: 623 K as fixed variables. 

** 

O2 to coal ratio: 0.8, solid concentration in coal slurry: 0.65, WGS process SDG : 0.59, and HTS reactor operating 

temperature: 623 K as fixed variables. 

346 

(1)

†† O2 to coal ratio: 0.8, solid concentration in coal slurry: 0.65, LTS reactor operating temperature: 473 K and HTS 

reactor operating temperature: 623 K as fixed variables. 

According to Chen and co-workers [34], the LHV of syngas (kJ N m -3 ) can be defined as: 

Table 5 summarizes the results obtained from the sensitivity analysis of the variables which 

presents higher effect on the and LHV of the H2-rich syngas. Additionally, the H2 molar 

fraction in the final process stream is reported. The effect of those variables will be analyzed 

individually in the next subsections: 

4.2. Oxygen to Carbon Mass Ratio Effect 

(a) 

(b) 

Fig. 2. Effect of the O2 to carbon ratio on (a) the syngas molar composition upstream WGS reactors 

and (b) molar flow rate downstream WGS reactors: CO (), H2 (), CO2 (), and 

adiabatic temperature () with solid concentration in coal slurry: 0.65, WGS process SDG : 

0.59, LTS reactor operating temperature: 473 K and HTS reactor operating temperature: 

623 K as fixed variables. 

Fig. 2 (a) and (b) summarizes the results obtained in the syngas molar composition and the 

gasification temperature in the gasifier as functions of a wide variation of the O2 to carbon mass 

347 

(2)

atio, as well as the shift-syngas flow rate after WGS. As expected, the increase in O2 to coal ratio 

favors exothermic reactions, therefore, an increase in gasifier operating temperature is achieved. 

However, as shown Fig. 2(a), there is a turning point in the operating gasifier temperature at an O2 

to carbon ratio close to 0.8. This is due to the differences in the heat released from partial 

combustion and complete combustion [33]. This turning point appears when the maximum CO and 

H2 concentration is reached. Beyond this point the CO2 increases because of the complete 

combustion while the CO and H2 compositions decrease. 

As shown in Fig. 2(b), the maximum H2 flow rate downstream the WGS reactors was obtained with 

an O2 to carbon ratio of 0.8. At this rate, CO concentration in the syngas leaving the gasifier is 

maximized while CO2 concentration is minimized. As a consequence, H2 production is favored in 

the WGS reactors. 

Surprisingly, the maximum thermal efficiency was 62.6% and was obtained for an O2 to carbon 

ratio of 0.64 (see Table 5). When O2 to carbon ratio was fixed at 0.8, the thermal efficiency was 

60.0%, decreasing 17% with respect to the maximum. That efficiency fall is caused by the 

increment in O2 flow rate. As a result, energy requirements for ASU process penalize the 

despite LHV increment. 

As suggested by the results presented in Table 5, there is a linear correlation between the syngas H2 

composition and its LHV. As the H2 composition is raised LHV also increases. This is because H2 

is the main contributor, over CH4 and CO, to the syngas heating value. 

4.3. Coal Slurry Concentration Effect 

(a) 

348

(b) 

Fig. 3. Effect of the coal slurry concentration on (a) the syngas composition upstream WGS reactor 

and (b) molar flow rate downstream the WGS reactors: CO (), H2 (), CO2 () and the 

adiabatic temperature () with O2 to coal ratio: 0.8, WGS process SDG : 0.59, LTS reactor 

operating temperature: 473 K and HTS reactor operating temperature: 623 K as fixed 

variables 

Fig. 3 (a) and (b) summarizes the results obtained in the syngas molar composition and the 

gasification temperature in the gasifier as functions of a wide variation of the coal slurry 

concentration, as well as the shift-syngas flow rate after WGS. Fig. 3(a) shows that lower steam 

flow leads to a slight raise in H2 concentration and a significant increment of CO at the gasifier 

downstream. An increase in solid concentration results in a higher gasifier temperature. Hence, at 

higher temperatures Boudouard reaction (R3) and steam gasification (R4) are favored and CO 

production is increased. Downstream the WGS reactors, the H2 maximum flow is obtained with a 

solid concentration of 65% approximately, as shown in Fig. 3(b). Beyond this value, the H2 flow 

decreases because the CO conversion in the WGS reactors is limited by the steam flow rate. 

As shown in Table 5, the coal slurry concentration has slight effect on the (< 5% change within 

the range). Nevertheless, higher solid concentration does affect the LHV, as the WGS conversion is 

decreased and final H2 composition decrease moderately. 

4.4. WGS reactor operating temperature effect 

349

Fig. 4. Effect of operating temperature on overall CO conversion in the shift reactors with O2 to 

coal ratio: 0.8, solid concentration in coal slurry: 0.65, WGS process SDG : 0.59, and HTS 

reactor operating temperature: 623 K as fixed variables. 

As shown in Fig. 4, when the LTS reactor is operated at low temperatures, the CO conversion is 

promoted due to the exothermic nature of the WGS reaction (R5). Low operating temperatures are 

preferred to obtain higher CO to H2 conversion but it implies a decrease in the reaction rate and 

catalytic activity. 

When the LTS reactor operating temperature was raised from 473 K to 523 K, the cleaned-syngas 

LHV felt from 99.33 MJ kg -1 to 81.54 MJ N kg -1 . This is a decrease of 18% (Table 5). Nevertheless, 

the increases 3.3%. The CO2 flow in the shift-syngas drops as the LTS reactor temperature is 

increased (from 7.43 Mg h -1 to 7.19 Mg h -1 ). Therefore, less acid gas is removed in the Rectisol 

absorber from the shift-syngas. The energy requirement necessary to achieve the CH3OH 

recuperation is lower compared with higher CO2 concentration downstream the LTS reactor (from 

4.20 GJ h -1 to 4.16 GJ h -1 ). 

4.5. Effect of steam to dry gas molar ratio. 

As shown in Fig. 5, the total CO conversion grows inversely exponential with the steam to dry gas 

ratio (SDG), approaching asymptotically to an extent of CO conversion of 1. Besides, the SDG is 

dependent of the operating temperature. At lower temperatures, higher conversion with lower steam 

flow fed to the HTS reactor is reached. The maximum conversion of 0.999 was achieved when a 3.9 

SDG was set at 473 K. However, the CO conversion keeps mainly constant after a SDG of 2, 

whilst a 0.996 conversion was attained. Past this value, to increase the CO conversion 0.12%, it is 

required to raise the SDG by 40%. Therefore, to get a conversion over 0.996 is preferred to 

manipulate other variables as it becomes almost insensitive to the SDG in ratios greater than 2. 

Fig. 5. Effect of the steam to dry gas molar ratio over overall CO conversion, when an operating 

temperature of 473 K (), 523 K () and 573 K () is established in the LT-WGS reactor 

with O2 to coal ratio: 0.8, solid concentration in coal slurry: 0.65, and HTS reactor 

operating temperature: 623 K as fixed variables. 

Cleaned-syngas LHV and , increases inversely exponential and decreases lineally respectively 

with an increasing of the SDG at a LTS operating temperature of 473 K (Table 5). Cleaned-syngas 

LHV increment is due to the CO conversion raising. Therefore, as in the CO conversion, after a 

SDG of 2 the LHV is almost constant. On the other hand, drop is caused by the steam flow 

350

feed increment in WGS process. As a result, in the Rectisol process higher acid gas flow required to 

be treated increasing the energy consumption. 

4.6. Optimal Syngas composition 

According to Table 5, the highest efficiency is reached at a slurry solid concentration of 0.65, a O2 

to carbon ratio of 0.64, a WGS process SDG of 0.59, a LTS reactor operating temperature of 473 K. 

Table 6 shows the H2-rich syngas composition at these conditions. 

Table 6. H2-rich syngas molar fraction 

As shown in Table 6, the syngas product is suitable for gas turbines since H2 molar fraction is 

92.2%. The operation of gas turbines using syngas with hydrogen fuel concentrations (>90%) has 

been demonstrated at several facilities in the United States [36]. Nevertheless, the co-sequestration 

of CO2 and H2S with the Rectisol process has proven to be a success as a high CO2 + H2S capture is 

obtained. The CO2 composition in the H2-rich syngas is 6 ppm as well as H2S and COS are found as 

traces. Furthermore, tail gas CO2 molar fraction is over 98% with a H2S concentration is 0.26% 

mol. As a result, this tail gas can be advantageous for enhanced oil production in sour fields as the 

environmental impact and processing costs will not be significant [37]. 


Component H2-rich syngas molar fraction 

H2O 6,47x10 -11 

H2 

N2 

0,922 

7,13x10 -3 

3,77x10 -24 

CL2 

CO 9,58x10 -3 

6,00x10 -6 

CO2 

CH4 

0,062 

H2S 3,53x10 -20 

COS 2,89x10 -20 

1,74x10 -7 

NH3 

HCN 2,20x10 -42 

CH4OH 4,15x10 -6 

In this paper, a SOG simulation was proposed using Aspen Plus® to estimate syngas production by 

coal gasification. Sensitivity of the process for different operating variables was then analyzed. As a 

result, a maximum thermal efficiency of 62.6% was reached. This maximum corresponds to a slurry 

solid concentration of 0.65, a O2 to carbon ratio of 0.64, a WGS process SDG of 0.59, a LTS 

reactor operating temperature of 473 K. At these fixed conditions, a H2-rich syngas of 92.2% molar 

composition and LHV of 12 MJ N m 

351 

-3 was attained. 

The thermal efficiency is found to be (1) insensitive to coal slurry concentration and LTS reactor 

operating temperature, (2) moderately sensitive to SDG in the WGS process and (3) most sensitive 

to oxygen to carbon ratio. An excessive increase in the O2 flow rate causes a fall in thermal 

efficiency. This behavior is caused as the energy requirements for ASU process and Rectisol 

process penalized the despite LHV increment. 

The lower heating value of the H2-rich syngas results to be (1) moderately sensitive to the LTS 

reactor temperature and coal slurry concentration, (2) most sensitive to O2 to carbon ratio. 

Nonetheless, a SDG higher than 2 is necessary for a complete CO conversion. Beyond this ratio, the

SDG has slight effect on the syngas composition (

[5.] Giunta P, Amadeo N, Laborde M. Simulation of a low temperature water gas shift reactor using 

the heterogeneous model/application to a pem fuel cell. Journal of power sources. 

2006;156(2):489-96. 

[6.] Dincer I. Green methods for hydrogen production. International Journal of Hydrogen Energy. 

2012;37(2):1954-71. 

[7.] BP. BP Statistical Review of World Energy June 2011. 2011 Available at: 

http://www.bp.com/statisticalreview. [accessed 10.02.2012]. 

[8.] Staczyk K, Kapusta K, Wiatowski M, widrowski J, Smoliski A, Rogut J, et al. 

Experimental simulation of hard coal underground gasification for hydrogen production. Fuel. 

2012;91(1):40-50. 

[9.] Jones RH, Thomas GJ. Materials for the Hydrogen Economy: CRC; 2007. 

[10.] Energy USDo. Pioneering Gasification Plants. 2011 Available at: 

http://www.fossil.energy.gov/programs/powersystems/gasification/gasificationpioneer.html. 

[accessed 10.02.2012]. 

[11.] Colombia MdMyEad. El carbón Colombiano; Recursos, Reservas y Calidad 2004. 

[12.] Kopyscinski J, Schildhauer TJ, Biollaz SMA. Methanation in a fluidized bed reactor with high 

initial CO partial pressure: Part II— Modeling and sensitivity study. Chemical Engineering 

Science. 2011;66(8):1612-21. 

[13.] Minchener AJ. Coal gasification for advanced power generation. Fuel. 2005;84(17):2222-35. 

[14.] Emun F, Gadalla M, Majozi T, Boer D. Integrated gasification combined cycle (IGCC) 

process simulation and optimization. Computers & Chemical Engineering. 2010;34(3):331-8. 

[15.] Nikoo MB, Mahinpey N. Simulation of biomass gasification in fluidized bed reactor using 

ASPEN PLUS. Biomass and Bioenergy. 2008;32(12):1245-54. 

[16.] Ramzan N, Ashraf A, Naveed S, Malik A. Simulation of hybrid biomass gasification using 

Aspen plus: A comparative performance analysis for food, municipal solid and poultry waste. 

Biomass and Bioenergy. 2011;35(9):3962-9. 

[17.] Shen L, Gao Y, Xiao J. Simulation of hydrogen production from biomass gasification in 

interconnected fluidized beds. Biomass and Bioenergy. 2008;32(2):120-7. 

[18.] Liu B, Yang X, Song W, Lin W. Process Simulation Development of Coal Combustion in a 

Circulating Fluidized Bed Combustor Based on Aspen Plus. Energy & Fuels. 2011;25(4):1721- 

30. 

[19.] Robinson PJ, Luyben WL. Simple Dynamic Gasifier Model That Runs in Aspen Dynamics. 

Industrial & Engineering Chemistry Research. 2008;47(20):7784-92. 

[20.] Chiesa P, Consonni S, Kreutz T, Robert W. Co-production of hydrogen, electricity and CO2 

from coal with commercially ready technology. Part A: Performance and emissions. 

International Journal of Hydrogen Energy. 2005;30(7):747-67. 

[21.] Davison TY. GE Gasification Project Update. Gasification 11 Conference: GE Energy; 2012. 

[22.] Smith AR, Klosek J. A review of air separation technologies and their integration with energy 

conversion processes. Fuel Processing Technology. 2001;70(2):115-34. 

[23.] Vinson DR. Air separation control technology. Computers & Chemical Engineering. 

2006;30(10–12):1436-46. 

[24.] AG L. Cryogenic Air Separation, History and Technological progress. Pullach: 2009 

Available at: http://www.lindele.de/process_plants/air_separation_plants/documents/L_2_1_e_09_150dpi.pdf. 

[accessed 

10.02.2012]. 

[25.] Chen P-C, Chiu H-M, Chyou Y-P, Yu C-S. Processes Simulation Study of Coal to Methanol 

Based on Gasification Technology. Engineering and Technology. 2010;65:988-96. 

353

[26.] Castle WF. Air separation and liquefaction: recent developments and prospects for the 

beginning of the new millennium. International Journal of Refrigeration. 2002;25(1):158-72. 

[27.] Damartzis T, Zabaniotou A. Thermochemical conversion of biomass to second generation 

biofuels through integrated process design—A review. Renewable and Sustainable Energy 

Reviews. 2011;15(1):366-78. 

[28.] Topsøe H. Sulphur resistant/sour water-gas shift catalyst. Lyngby 2009. 

[29.] Bell D, Towler B. Coal Gasification and Its Applications: ELSEVIER; 2010. 

[30.] Korens N, Simbeck D, Wilhelm D. Process screening analysis of alternative gas treating and 

sulfur removal for gasification. Prepared for DOE/NETL. 2002. 

[31.] Heil S, Brunhuber C, Link K, Kittel J, Meyer B. Dynamic Modelling of CO2-removal units 

for an IGCC power plant. Proceedings 7th Modelica Conference; Sep. 20-22; Como, Italy: The 

Modelica Association; 2009. p. 77-85. 

[32.] Larson ED, Consonni S, Katofsky RE, Consulting N, Burlington I, Iisa K, et al. A Cost- 

Benefit Assessment of Gasification-Based Biorefining in the Kraft Pulp and Paper Industry 

Volume 1 Main Report. Princeton University, Princeton, NJ. 2006. 

[33.] Yuehong Z, Hao W, Zhihong X. Conceptual design and simulation study of a co-gasification 

technology. Energy Conversion and Management. 2006;47(11–12):1416-28. 

[34.] Chen C, Jin Y-Q, Yan J-H, Chi Y. Simulation of municipal solid waste gasification in two 

different types of fixed bed reactors. Fuel. [Available online 24 August 2011]. 2011. 

[35.] Wang Y, Dong W, Dong L, Yue J, Gao S, Suda T, et al. Production of Middle Caloric Fuel 

Gas from Coal by Dual-Bed Gasification Technology. Energy & Fuels. 2010 

2010/05/20;24(5):2985-90. 

[36.] Walton SM, He X, Zigler BT, Wooldridge MS. An experimental investigation of the ignition 

properties of hydrogen and carbon monoxide mixtures for syngas turbine applications. 

Proceedings of the Combustion Institute. 2007;31:3147-54. 

[37.] Davison J, Bressan L, Domenichini R. CO2 capture in coal-based IGCC power plants. 7th 

International Conference in Greenhouse Gas Control Technologies; Vancouver2004. 

354

1. Introduction - Firenze University Press

Create successful ePaper yourself

Delete template?

Save as template?