Vol 2 - Winrock International India

Report IVOLUME II: ANNEXUREBiodiesel Manufacturing ProcessesAs part of Preparing Status Reports on Themes Related toTechnical and Scientific Utilization of Biofuel UtilizationDepartment of Science and TechnologyGovernment of IndiaSubmitted byMarch 2012

Principal InvestigatorSomnath BhattacharjeeProject TeamShankar HaldarArvind ReddyNilanjan GhoseSharda GautamAniruddha BhattacharjeePrincipal AdvisorSudhir Singhal

Volume IIANNEXUREAnnexure 1:S.P. Singh, et al., Biodiesel production through the use of different sourcesand characterization of oils and their esters as the substitute of diesel: Areview, Renewable and Sustainable Energy Reviews, 2010Annexure 2: Balat et al., Progress in biodiesel processing, Applied Energy, 2010Annexure 3:Annexure 4:Annexure 5:Annexure 6:Annexure 7:Annexure 8:Annexure 9:Annexure 10:Kraai et al., Novel highly integrated biodiesel production technology in acentrifugal contactor separator device, 2009T. Eevera, et al., Biodiesel production process optimization andcharacterization to assess the suitability of the product for variedenvironmental conditions, Renewable Energy, 2009Masoud Zabeti, et al., Activity of solid catalysts for biodiesel production: Areview, Fuel Processing Technology, 2009Zheyan Qiu, et al, Process intensification technologies in continuousbiodiesel production, Chemical Engineering and Processing: ProcessIntensification, 2010Dennis Y.C. Leung, et al., A review on biodiesel production using catalyzedtrans-esterification, Applied Energy, 2010Ru Yang, et al., One-pot process combining trans-esterification and selectivehydrogenation for biodiesel production from starting material of highdegree of unsaturation, Bioresource Technology, 2010J.M. Marchetti, et al., Biodiesel production from acid oils and ethanol usinga solid basic resin as catalyst, Science direct, 2010A. Deshpande, et al., Supercritical biodiesel production and powercogeneration: Technical and economic feasibilities, Bioresource Technology,2010Annexure 11: Gleason, Rodney J., et al., Biodiesel Fuel and Method of Manufacture, 2009,US Patent Application 20090277077Annexure 12:Annexure 13:West, J., et al., Biodiesel Production Unit, 2010, US Patent Application20100095581Strege, Joshua R., et al., Process For The Conversion Of Renewable Oils ToLiquid Transportation Fuels, 2010, US patent application 20100113848Annexure 14: Parnas, R., et al., Methods and systems for alkyl ester production, 2009,Patent US7544830Annexure 15:Annexure 16:Lichtenberger, P.L., et al., Esterification and transesterification systems,methods and apparatus, 2010, US Patent 7,678,340Mohammed, F., et al., Method of Biodiesel Production, 2009, US PatentApplication 20090038209

Renewable and Sustainable Energy Reviews 14 (2010) 200–216Contents lists available at ScienceDirectRenewable and Sustainable Energy Reviewsjournal homepage: www.elsevier.com/locate/rserBiodiesel production through the use of different sources and characterizationof oils and their esters as the substitute of diesel: A reviewS.P. Singh, Dipti Singh *School of Energy and Environmental Studies, Devi Ahilya University, Khandwa Road, Takshila Campus, Indore 452011, Madhya Pradesh, IndiaARTICLEINFOABSTRACTArticle history:Received 26 May 2009Accepted 13 July 2009Keywords:Alternative fuelsBiodieselPetroleum fuelTransesterificationTriglyceridesThe world is confronted with the twin crises of fossil fuel depletion and environmental degradation.The indiscriminate extraction and consumption of fossil fuels have led to a reduction in petroleumreserves. Petroleum based fuels are obtained from limited reserves. These finite reserves are highlyconcentrated in certain region of the world. Therefore, those countries not having these resourcesare facing a foreign exchange crisis, mainly due to the import of crude petroleum oil. Hence it isnecessary to look for alternative fuels, which can be produced from materials available within thecountry. Although vegetative oils can be fuel for diesel engines, but their high viscosities, lowvolatilities and poor cold flow properties have led to the investigation of its various derivatives.Among the different possible sources, fatty acid methyl esters, known as Biodiesel fuel derived fromtriglycerides (vegetable oil and animal fates) by transesterification with methanol, present thepromising alternative substitute to diesel fuels and have received the most attention now a day. Themain advantages of using Biodiesel are its renewability, better quality exhaust gas emission, itsbiodegradability and the organic carbon present in it is photosynthetic in origin. It does notcontribute to a rise in the level of carbon dioxide in the atmosphere and consequently to the greenhouse effect. This paper reviews the source of production and characterization of vegetable oils andtheir methyl ester as the substitute of the petroleum fuel and future possibilities of Biodieselproduction.ß 2009 Elsevier Ltd. All rights reserved.Contents1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2012. Source of Biodiesel. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2013. Chemical composition of alternative fuels (oil and Biodiesel) and diesel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2014. Processes or method to produce Biodiesel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2034.1. Pyrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2034.2. Dilution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2034.3. Microemulsion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2064.4. Transesterification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2065. Properties of alternative fuel or diesel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2085.1. Properties of oil. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2085.2. Properties of Biodiesel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2095.3. Determination and measurement methods for oil/Biodiesel properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2096. Performance of alternative fuel (oil and Biodiesel) as a fuel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2106.1. Use of oils as fuel and their performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2106.2. Need of blending and its effect on performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2106.3. Use of esters as fuel and their performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211* Corresponding author. Tel.: +91 0731 2460309; fax: +91 0731 2467378/2462366.E-mail address: dipti5683@rediffmail.com (D. Singh).1364-0321/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.rser.2009.07.017

S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216 2017. Problems arises and their solution during the use of alternative fuels. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2127.1. Problem arises during the use of oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2127.2. Problem arises during the use of Biodiesel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2128. Economic viability of Biodiesel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2129. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2131. IntroductionThe depleting reserves of fossil fuel, increasing demands fordiesels and uncertainty in their availability is considered to bethe important trigger for many initiatives to search for thealternative source of energy, which can supplement or replacefossil fuels.One hundred years ago, Rudolf Diesel tested peanut oil as fuelfor his engine for the first time on August 10, 1893 [1]. In the 1930sand 1940s vegetable oils were used as diesel fuels from time totime, usually only in emergency. The first International conferenceon plant and vegetable oils as fuels was held in Fargo,North Dakota in August 1982. The primary concern discussedwere the cost of fuel, the effect of vegetable oil fuels on engineperformance and durability and fuel preparation specification andadditives. Oil production, oil seed processing and extraction alsowere considered in this meeting [2]. Vegetable oils hold promiseas alternative fuels for diesel engines [3,4]. But their highviscosities, low volatilities and poor cold flow properties haveled to the investigation of various derivatives. Fatty acid methylesters, known as Biodiesel, derived from triglycerides bytransesterification with methanol have received the most attention[5,6].The name Biodiesel was introduced in the United States during1992 by the National Soy diesel Development Board (presentlynational bio diesel board) which has pioneered the commercializationof Biodiesel in the United States. Biodiesel can be used inany mixture with petroleum diesel as it has very similarcharacteristics but it has lower exhaust emissions. Biodiesel hasbetter properties than that of petroleum diesel such as renewable,biodegradable, non-toxic, and essentially free of sulfur andaromatics. Biodiesel fuel has the potential to reduce the level ofpollutants and the level of potential or probable carcinogens [7].Ma et al. [8] stated that Biodiesel has become more attractiverecently because of its environmental benefits and fact that it ismade from renewable resources. The raw materials beingexploited commercially by the Biodiesel are the edible fatty oilsderived from rapeseed, soybean, palm, sunflower, coconut, linseed,etc. [9]. In recent years, research has been directed to explore plantbased fuels, have bright future [10]. This chapter focused on thesource of oils, problems associated with the use of oils, productionof Biodiesel from non-edible oil, Physical and chemical propertiesof oils and esters; advantages, disadvantages and challenges.2. Source of BiodieselAlternative diesel fuels made from natural, renewable sourcessuch as vegetable oil and fats [11,12]. The most commonly used oilsfor the production of Biodiesel are soybean [13,14], sunflower[15,16], palm[17], rapeseed [18], canola [19], cotton seed [20] andJatropha [21]. Since the prices of edible vegetable oils are higher thanthat of diesel fuel, therefore waste vegetable oils and non-ediblecrude vegetable oils are preferred as potential low priced Biodieselsources. Use of such edible oil to produce Biodiesel in India is also notfeasible in view of big gap in demand and supply of such oils. UnderIndian condition only such plants can be considered for Biodiesel,which produce non-edible oil in appreciable quantity and can begrown on large scale on non-cropped marginal lands and wastelands. Animal fats, although mentioned frequently, have not beenstudied to the same extent, as vegetable oils because of naturalproperty differences. Animal fats contain higher level of saturatedfatty acids therefore they are solid at room temperature [22]. Thesource of Biodiesel in the form of vegetables oils, non-edible oils,animal fats and some other biomass are listed in Table 1.The source of Biodiesel usually depends on the crops amenableto the regional climate. In the United States, soybean oil is themost commonly Biodiesel feedstock, whereas the rapeseed(canola) oil and palm oil are the most common source forBiodiesel, in Europe, and in tropical countries respectively [68].Asuitable source to produce Biodiesel should not competent withother applications that rise prices, for example pharmaceuticalraw materials. But the demand for pharmaceutical raw material islower than for fuel sources. As much as possible the Biodieselsource should fulfill two requirements: low production costs andlarge production scale. Refined oils have high production costs,but low production scale; on the other side, non-edible seeds,algae and sewerage have low production costs and are moreavailable than refined or recycled oils.The oil percentage and the yield per hectare are importantparameters to consider as Biodiesel source. Productions of nonedible oil seeds percentage of oil content are given in Table 2.3. Chemical composition of alternative fuels (oil and Biodiesel)and dieselFrom a chemical point of view, oils from different sources havedifferent fatty acid compositions. The fatty acids vary in theircarbon chain length and in the number of unsaturated bonds theycontain. Fats and oils are primarily water-insoluble, hydrophobicsubstances in the plant and animal kingdom that are made up ofone mole of glycerol and three moles of fatty acids and arecommonly referred as triglycerides [70] Fig. 1Chemically the oil/fats consist of 90–98% triglycerides and smallamount of mono and diglycerides. Triglycerides are esters of threefatty acids and one glycerol. These contain substantial amount ofoxygen in their structures. When three fatty acids are identical, theproduct is simple triglycerides, when they are dissimilar the productis mixed triglycerides fatty acids which are fully saturated withhydrogen have no double bonds. Those with one missing hydrogenmolecule have one double bond between carbon atoms and arecalled monosaturated. And those with more than one missinghydrogen have more than one double bond and are calledpolyunsaturated. Fully saturated triglycerides lead to excessivecarbon deposits in engines. The fatty acids are different in relation tothe chain length, degree of unsaturation or presence of otherchemical functions. Chemically, Biodiesel is referred to as the monoalkylesters of long-chain-fatty acids derived from renewable lipidsources. Biodiesel is the name for a variety of ester based oxygenatedfuel from renewable biological sources. It can be used in compressionignition engines with little or no modifications [72].Biodiesel is made in a chemical process called transesterification,where organically derived oils (vegetable oils, animal fats andrecycled restaurant greases) are combined with alcohol (usuallymethanol) and chemically altered to form fatty esters such as

202S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216Table 1Source of oil.Vegetable oils Non-edible oils Animal Fats Other SourcesSoybeans [13,14,23] Almond [33] Lard [22] Bacteria [1]Rapeseed [18] Abutilon muticum [34] Tallow [64] Algae [1]Canola [19,24] Andiroba [35] Poultry Fat [64] Fungi [1]Safflower [15,16] Babassu [29,30,35] Fishoil [30,31] Micro algae [65]Barley [25] Brassica carinata [36] Tarpenes [66]Coconut [23,26,27] B. napus [36] Latexes [66]Copra [28] Camelina [37] Cooking Oil (Yellow Grease) [64]Cotton seed [20] Cumaru [35] Microalgae (Chlorellavulgaris) [67]Groundnut [23,28–30] Cynara cardunculus [38]Oat [25] Jatrophacurcas [21,39,40]Rice [25,28,31] Jatropha nana [41]Sorghum [25] Jojoba oil [42]Wheat [25] Pongamiaglabra [43,44]Winter rapeseed oil [32] Laurel [23]Lesquerellafendleri [45]Mahua [46,47]Piqui [23,48,49]Palm [17,50–52]Karang [53]Tobacco seed [54–56]Rubber plant [49,57,58]Rice bran [59–62]Sesame [23]Salmon oil [63]Table 2Production of non-edible oil seeds and bio-residues in India [69].SpeciesOil faction(%)Seed estimate(10 6 tones/y)Oil(tons/ha/y)Castor 45–50 0.25 0.5–1.0Jatropha 50–60 0.20 2.0–3.0Mahua 35–40 0.20 1.0–4.0Sal 10–12 0.20 1.0–2.0Linseed 35–45 0.15 0.5–1.0Neem 20–30 0.10 2.0–3.0Pongamia (Karanja) 30–40 0.06 2.0–4.0Others 10–50 0.50 0.5–2.0Others 10–50 0.50 0.5–2.0methyl ester. Chemically, most Biodiesel consists of alkyl (usuallymethyl) esters instead of the alkanes and aromatic hydrocarbons ofpetroleum derived diesel.Oil, ester and diesel have different number of carbon andhydrogen compound. Diesel has no oxygen compound. It is a goodquality of fuel. On the other hand, in the case of vegetable oilsOxidation resistance is markedly affected by the fatty acidcomposition. The large size of vegetable oil molecules (typicallythree or more times larger than hydrocarbon fuel molecules) andthe presence of oxygen in the molecules suggests that some fuelproperties of vegetable oil would differ markedly from those ofhydrocarbon fuels [3]. Chemical structure of oil, Biodiesel andpetroleum diesel are given Table 3.Fig. 1. Chemical structure of vegetable oil [71].Table 3Chemical structure of oil,ester and diesel.Chemical structure ofMonoglycerideChemical structureof DiglycerideChemical structureof Fat &OilChemical Structureof EsterChemical structureof DieselC 12 H 23

S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216 203Table 4Chemical structure of common fatty acids.Name of fatty acid Chemical name of fatty acids Structure (xx:y) FormulaLauric Dodecanoic 12:0 C 12 H 24 O 2Myristic Tetradecanoic 14:0 C 14 H 28 O 2Palmitic Hexadecanoic 16:0 C 16 H 32 O 2Stearic Octadecanoic 18:0 C 18 H 36 O 2Oleic cis-9- Octadecenoic 18:1 C 18 H 34 O 2Linoleic cis-9,cis-12- Octadecadienoic 18:2 C 18 H 32 O 2Linolenic cis-9,cis-l2,cis-15-Octadecatrienoic 18:3 C 18 H 30 O 2Arachidic Eicosanoic 20:0 C 20 H 40 O 2Behenic Docosanoic 22:0 C 22 H 44 O 2Erucle cis-13-Docosenoic 22:1 C 32 H 42 O 2Lignoceric Tetracosanoic 24:0 C 24 H 48 O 2Petroleum based diesel fuels have different chemical structuresfrom vegetable oils and esters. The former contain only carbon andhydrogen atoms, which are arranged in normal (straight chain) orbranched chain structures, as well as aromatics configuration. Thenormal structures preferred for better ignition quality [73]. Dieselfuel can contain both saturated and unsaturated hydrocarbons, butthe latter are not present in large enough amounts to make fueloxidation problem. Petroleum-derived diesel is composed of about75% saturated hydrocarbons (primarily paraffins including n, iso,and cycloparaffins), and 25% aromatic hydrocarbons (includingnaphthalenes and alkylbenzenes) [74]. The average chemicalformula for common diesel fuel is C 12 H 23 , ranging from approximatelyC 10 H 20 to C 15 H 28 .Vegetable oils contain fatty acid, free fatty acids (generally 1–5%), phospholipids, phosphatides, carotenes, tocopherols, sulfurcompound and traces of water [75]. The fatty acids, which arecommonly found in vegetable oils, are stearic, palmitic, oleic,linoleic and linolenic. Table 4 summarizes the fatty acid compositionof some vegetable oils [4,33,35,64].Triglyceride molecules have molecular weights between 800and 900 and are thus nearly four times larger than typical dieselfuel molecules [76]. The various vegetable oils and esters aredistinguished by their fatty acid compositions. In Tables 5 and 6different fatty acids are shown by their respective number ofcarbon present in the structure of oils and esters.Table 5 summarized the fatty acid composition of fifty-onesamples of vegetable oils and fat. Twenty one fatty acids arescreened in all the samples. The fatty acids which are commonlyfound in vegetable oil and fat are stearic, palmitic, oleic, linoleic.The other fatty acids which are also present in many of the oils andfats are myristic (tetradecanoic), palmitoleic, acachidic, linolenicand octadecatetranoic. There are many other fatty acids which arealso found in oils with the above-mentioned common fatty acids.Lauric fatty acid is present only in bay laurel leaf, coconut, Babassuand Luphea. Myristoleic and ecosenoic fatty acids are found in beeftallow, choice white and poultry fat. Eicosenoic fatty acid is presentin beef tallow, choice white, poultry fat, yellow grease andcamalina. Behenic, lignoceric fatty acids are detected in peanutkernel and crambe. Erucle fatty acid is found only in three oils, i.e.crambe camellia oil and Brassica carinata.In the oil of Cotton, Tobbaco, rapeseed, safflower, sunflower,Sesame, linseed, palm, corn, soyabean, peanut and rice bran, thecontent of the fatty acid is different in the same plant species thatmay be either due to the varital or instrumental difference or in thedifferent parts of plants.Fatty acid profiles of seed oil methyl esters of 75 plant specieshaving 30% or more fixed oil in their seed/kernel were examined(Table 6). Fatty acid compositions were used to predict the qualityof fatty acid methyl esters of oil for use as Biodiesel. Fatty acidmethyl ester of oils of 26 species including Azadirachta indica,Calophyllum inophyllum, Jatropha curcas and Pongamia pinnatawere found most suitable for use as Biodiesel and they meet themajor specification of Biodiesel standards of USA, Germany andEuropean Standard Organization [97].The fatty acid methyl esters of another 11 species meet thespecification of Biodiesel standard of USA only. These selectedplants have great potential for Biodiesel [81].4. Processes or method to produce BiodieselConsiderable efforts have been made to develop vegetable oilderivatives that approximate the properties and performance ofhydrocarbons-based diesel fuels. The problem with substitutingtriglycerides for diesel fuel is mostly associated with high viscosity,low volatility and polyunsaturated characters. These can bechanged in at least four ways: pyrolysis, microemulsion, dilutionand transesterification.4.1. PyrolysisPyrolysis is a method of conversion of one substance intoanother by mean of heat or by heat with the aid of the catalyst inthe absence of air or oxygen [98]. The process is simple, wasteless,pollution free and effective compared with other crackingprocesses. The reaction of thermal decomposition is shown inFig. 2.4.2. DilutionThe vegetable oil is diluted with petroleum diesel to run theengine. Caterpillar Brazil, in 1980, used pre-combustion chamberengines with the mixture of 10% vegetable oil to maintain totalpower without any alteration or adjustment to the engine. At thatpoint it was not practical to substitute 100% vegetable oil for dieselfuel, but a blend of 20% vegetable oil and 80% diesel fuel wassuccessful. Some short-term experiments used up to a 50/50 ratio.Fig. 2. The mechanism of thermal decomposition of triglycerides [99].

Table 5Fatty acid composition of oil.204S. No. Vegetable oil Fatty acid composition (wt.%) Reference12:0 14:0 14:1 16:0 16:1 18:0 20:0 20:1 22:0 24:0 18:1 22:1 18:2 18:3 18:4 6:0 a , 8:0 b , 10:0 cand Others1. Cottonseed – – – 28.7 – 0.9 – – – – 13.0 – 57.4 – – – [33]2. Cottonseed – 0 – 28 – 1 0 – 0 0 13 0 58 0 – – [3]3. Tobacco – 0.09 – 10.96 0.2 3.34 – – – – 14.54 – 69.49 0.69 – 0.69 [78]4. Tobacco – 0.17 – 8.87 0.0 3.49 – – – – 12.4 – 67.75 4.20 – – [79]5. Rapeseed – – – 3.5 – 0.9 – – – – 64.1 – 22.3 8.2 – – [33]6. Rapeseed – 0 – 3 – 1 0 – 0 0 64 0 22 8 – – [3]7. Safflower seed – – – 7.3 – 1.9 – – – – 13.6 – 77.2 – – – [33]8. Safflower – 0 – 9 – 2 0 – 0 0 12 0 78 0 – – [3]9. H.O.Safflower – Tr – 5 – 2 Tr – 0 0 79 0 13 0 – – [3]10. Sunflower – – – 6.4 0.1 2.9 – – – – 17.7 – 72.9 – – – [33]11. Sunflower – 0 – 6 – 3 0 – 0 0 17 0 74 0 – – [3]12. Sesame – – – 13.1 – 3.9 – – – – 52.8 – 30.2 – – – [33]13. Sesame – 0 – 13 – 4 0 – 0 0 53 0 30 0 – – [3]14. Linseed – – – 5.1 0.3 2.5 – – – – 18.9 – 18.1 55.1 – – [33]15. Linseed – 0 – 5 – 2 0 – 0 0 20 0 18 55 – – [3]16. Palm – – – 42.6 0.3 4.4 – – – – 40.5 – 10.1 0.2 1.1 – [33]17. Palm tree – – 35 – 7 – – – – 44 – 14 – – – [35]18. Corn marrow – – – 11.8 2.0 – – – – 24.8 – 61.3 – 0.3 – [33]19. Corn – 0 – 12 – 2 Tr – 0 0 25 0 6 Tr – – [3]20. Tallow – – – 23.3 0.1 19.3 – – – – 42.4 – 2.9 0.9 2.9 – [33]21. Beef tallow 2.73 0.50 22.99 2.86 19.44 0.14 0.33 – – 41.60 – 3.91 0.49 0.36 – [64]22. Soybean – – – 13.9 0.3 2.1 – – – – 23.2 – 56.2 4.3 0 – [33]23. Soya bean – – – 14 – 4 – – – – 24 – 52 – 6 – [29]24. Soya bean – 0 – 12 – 3 0 – 0 0 23 0 55 6 – – [3]25. Peanut kernel – – – 11.4 0 2.4 – – 2.7 1.3 48.3 – 32.0 0.9 4.0 – [33]26. Peanut – 0 – 11 – 2 1 – 2 1 48 0 32 1 – – [3]27. Hazelnutkernel – – – 4.9 0.2 2.6 – – – – 83.6 – 8.5 0.2 0 – [33]28. Walnut kernel – – – 7.2 0.2 1.9 – – – – 18.5 – 56.0 16.2 0 – [33]29. Almond kernel – – – 6.5 0.5 1.4 – – – – 70.7 – 20.0 0 0.9 – [33]30. Olive kernel – – – 5.0 0.3 1.6 – – – – 74.7 – 17.6 0 0.8 – [33]31. Coconut 48.8 19.9 – 7.8 0.1 3.0 – – – 4.4 – 0.8 0 65.7 8.9 b , 6.2 c [33]32. Choice white 1.57 0.36 22.04 5.03 9.95 0.14 0.56 – – 42.45 – 13.17 0.97 0.29 – [64]33. Poultry fat 0.57 0.26 22.76 8.37 5.36 0.00 0.45 – – 42.07 – 17.14 1.07 0.22 – [64]34. Yellow grease 0.70 0.00 14.26 1.43 8.23 0.33 0.48 – – 43.34 – 26.25 2.51 0.47 – [64]35. Babassu 48 16 – 10 – 2 – – – – 14 – 5 – – 5 [35]36. Cumaru – – 23 – 7 – – – – 37 – 29 – – 4 [35]37. Piqui – – 40 – 2 – – – – 47 – 4 – – 7 [35]38. Castor – – – 1.1 0 3.1 – – – – 4.9 – 1.3 0 Ricinoic acid 89.6 [33]39. Poppy seed – – – 12.6 0.1 4.0 – – – – 22.3 – 60.2 0.5 – – [33]40. Bay laurel leaf 26.5 4.5 – 25.9 0.3 3.1 – – – – 10.8 – 11.3 17.6 31.0 – [33]41. Wheat grain – 0.4 – 20.6 1.0 1.1 – – – – 16.6 – 56.0 2.9 1.8 11.4 b [33]42. Crambe – 0 – 2 – 1 2 – 1 1 19 59 9 7 – – [3]43. Rice–bran – 0.4–0.6 – 11.7–16.5 – 1.7–2.5 0.4–0.6 – – 0.4–0.9 39.2–43.7 – 26.4–35.1 – – – [56]44. Sal – – – 4.5–8.6 – 34.2–44.8 6.3–12.2 – – – 34.2–44.8 – 2.7 – – – [56]45. Mahua – - – 16.0–28.2 – 20.0–25.1 0.0–3.3 – – – 41.0–51.0 – 8.9–13.7 – – – [56]46. Neem – 0.2–0.26 – 13.6–16.2 – 14.4–24.1 0.8–3.4 – – – 49.1–61.9 – 2.3–15.8 – – – [56]47. Karanjia – – – 3.7–7.9 – 2.4–8.9 – – – 1.1–3.5 44.5–71.3 – 10.8–18.3 – – – [45]48. Camelina oil – – – 5.4 – 2.6 0.25 16.8 1.4 – 14.3 2.9 2.9 38.4 – – [77]49. Andiroba – – – 27 1 7 – – – – 49 – 16 – – – [29]50 B. carinata crude oil – – – 3.1 – 1.0 0.7 – – – 9.7 42.5 16.8 16.6 – – [37]51 Cuphea 6iscosissima 4.81 6.84 3.33 0.15 1.37 2.05 0.13 4.19 a , 40.24 b , 36.90 c [80]S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216

Table 6Fatty acid composition of different methyl esters.S. No. Vegetable oil methyl ester Fatty acid composition (wt.%) Reference1:0 and 2:0 14:0 10:0 12:0 16:0 16:1 18:0 20:0 20:1 22:0 24:0 18:1 22:1 18:2 18:3 16:1 Osa, U.k. and acidsRhus succedanea Linn – – – – 25.4 – – – – – – 46.8 – 27.8 – – – [81]Annona reticulata Linn – 1.0 – – 17.2 – 7.5 – – – – 48.4 – 21.7 – 4.2 – [82]Ervatamia coronaria Stapf – 0.2 – – 24.4 – 7.2 0.7 0.2 0.2 – 50.5 – 15.8 0.6 0.2 uk 0.2 [83]Thevetia peruviana Merrill – – – – 15.6 – 10.5 0.3 – 0.1 – 60.9 – 5.2 7.4 – – [84]Vallaris solanacea Kuntze – – – – 7.2 – 14.4 1.8 – 0.4 0.5 35.3 – 40.4 – – – [83]Balanites roxburghii Planch – – – – 17.0 – 7.8 – – – – 32.4 – 31.3 7.2 4.3 – [85]Basella rubra Linn – 0.4 – – 19.7 – 6.5 – 0.7 – – 50.3 – 21.6 0.4 0.4 – [86]Canarium commune Linn – – – – 29.0 – 9.7 – – – – 38.3 – 21.8 1.2 – – [87]Cannabis sativa Linn – – – – – – – – – – – 15.0 – 65.0 15.0 Osa5.0 [83]Celastrus paniculatus Linn 2.0&1.7 – – – 25.1 – 6.7 – – – – 46.1 – 15.4 3.0 – – [88]Euonymus hamiltonianuis Wall 8.5 – – – 18.3 – 1.5 – – – – 39.1 – 25.8 5.3 – Uk 1.5 [83]Terminalia bellirica Roxb – – – – 35.0 – – – – – – 24.0 – 31.0 – – osa 10.0 [83]Terminalia chebula Retz – – – – 19.7 – 2.4 0.6 – 0.2 – 37.3 – 39.8 – – – [83]Vernonia cinerea Less – 8.0 – – 23.0 – 8.0 3.0 – 4.0 – 32.0 – 22.0 – – – [83]Corylus avellana – 3.2 – – 3.1 – 2.6 – – – – 88.0 – 2.9 – – Uk 0.2 [83]Momordica dioica Rox – – – – 10.2 – 16.9 – – – – 9.2 – 8.8 – – e.a. 54.9 [89]Aleurites fordii Hemsl – – – – – – – 3.0 – – – 6.5 – 9.0 – – e.a. 81.5 [82]Aleurites moluccana Wild – – – – 5.5 – 6.7 – – – – 10.5 – 48.5 28.5 – Uk 0.3 [82]Aleurites montana Wils – – – – – – – 3.5 – – – 18.2 – 10.7 – – Osa 3.0 e.a. 64.6 [87]Croton tiglium Linn – 11.0 – – 1.2 – 0.5 2.3 – – – 56.0 – 29.0 – – – [87]Euphorbia helioscopia Linn – 5.5 – 2.8 9.9 – 1.1 – – – – 15.8 – 22.1 42.7 – Uk 0.1 [83]Jatropa curcas Linn – 1.4 – – 15.6 – 9.7 0.4 – – – 40.8 – 32.1 – [90]Joannesia princeps Vell – 2.4 – – 5.4 – – – – – – 45.8 – 46.4 – [87]Mallotus phillippinensis Arg – – – – 3.2 – 2.2 – – – – 6.9 – 13.6 – – K.a.72.0 [90]Putranjiva roxburghii – – – – 8.0 – 15.0 3.0 – – – 56.0 – 18.0 – – – [87]Sapium sebiferum Roxb – 4.2 – 0.3 62.2 – 5.9 – – – – 27.4 – – – – – [87]Hydnocarpus kurzii Warb – – – – 4.0 – – – – – – 14.6 – – – – H.a.39.4 G.a.19.5 C.a.22.5 [90]Hydnocarpus wightiana Blume – – – – 1.8 – – – – – – 6.9 – – – – H.a.48.7 G.a12.2 C. 27.0 H. 3.4 [90]Calophyllum apetalum Wild – – – – 8.0 – 14.0 – – – – 48.0 – 30.0 – – – [91]Calophyllum inophyllum Linn – – – – 17.9 2.5 18.5 – – – 2.6 42.7 – 13.7 2.1 – – [90]Garcinia combogia Desr – – – – 2.3 – 38.3 0.3 – – – 57.9 – 0.8 0.4 – – [90]Garcinia indica Choisy – – – 2.5 – 56.4 – – – – 39.4 – 1.7 – – – [87]Garcinia echinocarpa Thw – – – – 3.7 – 43.7 – – – – 52.6 – – – – – [87]Garcinia morella Desr – – – – 0.7 – 46.4 2.5 – – – 49.5 – 0.9 – – – [87]Mesua ferrea Linn – 0.9 – – 10.8 – 12.4 0.9 – – – 60.0 – 15.0 – – – [90]Mappia foetida Milers – – – – 7.1 – 17.7 – – – – 38.4 – – 36.8 – – [87]Illicium verum Hook – 4.43 – – – – 7.93 – – – 63.24 – 24.4 – – – [92]Saturega hortensis Linn – – – – 4.0 – 4.0 – – – 12.0 – 18.0 62.0 – – [83]Perilla frutescens Britton – – – – – – – – – – – 9.8 – 47.5 36.2 Osa 6.5 [83]Actinodaphne angustifolia – 1.9 4.3 87.9 0.5 – – – – – – 5.4 – – – – – [90]Litsea glutinosa Robins – – – 96.3 – – – – – – – 2.3 – – – – U.k. 1.4 [89]Neolitsea cassia Linn – 3.8 3.0 85.9 – – – – – – – 4.0 – 3.3 – – – [93]Neolitsea umbrosa Gamble – 11.5 1.7 59.1 – – – – – – – 21.0 – 6.7 – – – [93]Michelia champaca Linn – – – – 20.7 – 2.5 2.6 – – – 22.3 – 42.5 – 6.9 U.K.2.5 [83]Hiptage benghalensis Kurz – – – – 2.6 – 1.6 2.6 – – – 4.5 – 4.4 – – R.a.84.3 [83]Aphanamixis polystachya Park – – – – 23.1 – 12.8 – – – – 21.5 – 29.0 13.6 – – [82]Azadirachta indica – – – – 14.9 – 14.4 1.3 – – – 61.9 – 7.5 – – – [92]Melia azadirach Linn – 0.1 – – 8.1 – 1.2 – – – – 20.8 – 67.7 – 1.5 – [87]Swietenia mahagoni Jacq – – – – 9.5 – 18.4 – – – – 56.0 – – 16.1 – – [87]Anamirta cocculus Wight & Hrn – – – – 6.1 – 47.5 – – – – 46.4 – – – – – [87]Broussonetia papyrifera Vent – – – – 4.0 – 6.1 3.0 – – – 14.8 – 71.0 1.0 – – [86]Moringa concanensis Nimmo – – – – 9.7 – 2.4 3.3 – – – 83.8 – 0.8 – – – [86]Moringa oleifera Lam – – – – 9.1 – 2.7 5.8 – – – 79.4 – 0.7 0.2 2.1 – [86]Myristica malabarica Lam – 39.2 – – 13.3 – 2.4 – – – – 44.1 – 1.0 – – – [87]S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216 205

206S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216Table 6 (Continued )S. No. Vegetable oil methyl ester Fatty acid composition (wt.%) Reference1:0 and 2:0 14:0 10:0 12:0 16:0 16:1 18:0 20:0 20:1 22:0 24:0 18:1 22:1 18:2 18:3 16:1 Osa, U.k. and acidsArgemone mexicana – 0.8 – – 14.5 – 3.8 1.0 – – – 18.5 – 61.4 – – – [87]Pongamia pinnata Pierre – – – – 10.6 – 6.8 4.1 2.4 5.3 2.4 49.4 – 19.0 – – – [90]Ziziphus mauritiana Lam – – – – 10.4 – 5.5 1.8 2.6 1.2 – 64.4 1.7 12.4 – – – [83]Princepia utilis Royle – 1.8 – – 15.2 4.5 – – – 0.9 32.6 – 43.6 – – Uk1.4 [87]Meyna laxiflora Robyns – – – – 18.8 9.0 – – – – 32.5 – 39.7 – – – [89]Aegle marmelos correa Roxb – 16.6 – – 8.8 – – – – – – 30.5 – 36.0 8.1 – – [82]Salvadora oleoides Decne – 50.7 0.8 35.6 4.5 – – – – – 8.3 – 0.1 – – – [90]Salvadora persica Linn – 54.5 1.0 19.6 19.5 – – – – – 5.4 – – – – [90]Santalum album Linn – – – – 1.9 – 1.0 – – – – 8.6 – 0.8 – – St.a.3.7 Sa.a.84.0 [83]Nephelium lappaceum Linn – – – – 2.0 – 13.8 34.7 4.2 – – 45.3 – – – [93]Sapindus trifoliatus Linn – – – – 5.4 – 8.5 20.7 – 2.1 – 55.1 – 8.2 – – – [81]Schleichera oleosa Oken – – – – 1.6 – 10.1 19.7 – 4.0 – 52.5 0.9 – – G.a.8.4 [82]Madhuca butyracea Mac – – – – 66.0 – 3.5 – – – – 27.5 – 3.0 – – – [94]Madhuca indica JF Gmel – 1.0 – – 17.8 – 14.0 3.0 – – – 46.3 – 17.9 – – – [95]Mimusops hexendra Roxb – – – – 19.0 – 14.0 1.0 – – – 63.0 – 3.0 – – – [87]Quassia indica Nooleboom – – – – 9.0 – – – – – – 36.0 – 48.0 – – osa7.0 [83]Ximenia americana Linn – – – – – – 1.2 – – – – 60.8 – 6.7 – – Osa1.5 X.a.14.626:015.2 [83]Pterygota alata Rbr – – – – 14.5 – 8.5 – – – – 44.0 – 32.4 – – Uk1.0 [85]Holoptelia integrifolia – 3.5 – – 35.1 – 4.5 1.4 – – – 53.3 1.9 Uk1.1 [83]Urtica dioica Linn – – – – 9.0 – – – – – – 14.6 73.7 2.7 – – [96]Tectona grandis Linn – 0.2 – – 11.0 10.2 2.3 – – – 29.5 46.4 0.4 – – [87]Osa: Other saturated acid; u.k.: unknown; e.a.: elaeostearic acid; K.a.: kamlolenic acid; H.a.: hydnocarpic acid; G.a.: gorlic acid C.a.: chaulmoogric acid; C.H.: chaulmoogric homolog; R.a.: ricinolic acid; St.a.: stearolic acid; Sa.a:santalbic acid; G.a.:gadoleic acid; X.a.:ximenic acid.25 part of sunflower oil and 75 parts of diesel were blended asdiesel fuel [100]. The low viscosity is good for better performanceof engine, which decreases with increasing the percentage ofdiesel.4.3. MicroemulsionA micro emulsion define as a colloidal equilibrium dispersion ofoptically isotropic fluid microstructure with dimensions generallyinto 1–150 range formed spontaneously from two normallyimmiscible liquids and one and more ionic or more ionicamphiphiles. [99] They can improve spray characteristics byexplosive vaporization of the low boiling constituents in micelles[101,102]. The engine performances were the same for amicroemultion of 53% sunflower oil and the 25% blend of sunfloweroil in diesel [103]. A microemultion prepared by blending soyabeanoil, methanol, and 2-octanol and cetane improver in ratio of52.7:13.3:33.3:1.0 also passed the 200 h EMA test [104].4.4. TransesterificationTransesterification (also called alcoholysis) is the reaction of afat or oil with an alcohol to form esters and glycerol. A catalyst isusually used to improve the reaction rate and yield. Excess alcoholis used to shift the equilibrium toward the product because ofreversible nature of reaction. For this purpose primary andsecondary monohybrid aliphatic alcohols having 1-8 carbon atomsare used [105].(a) Chemistry of transesterification process: Transesterificationconsist of a number of consecutive, reversible reactions [14,99].The triglycerides are converted step wise to triglycerides,monoglyceride and finally glycerol. A mole of ester librated ateach step (Fig. 3).(i) Alkali catalyzed transesterification: In transesterificationmethod we can use different catalyst. The reaction mechanismfor alkali catalyzed transesterification was formulated in threesteps [106] as explained in Fig. 4. The first step is an attack on thecarbonyl carbon atom of the triglycerides molecule by the anion ofthe alcohol (Methoxide ion) to form a tetrahedral intermediatereacts with an alcohol (methanol) to regenerate the anion ofalcohol (methoxide ion). In the last step, rearrangement oftetrahedral intermediate results in the formation of a fatty acidester and a diglyceride. When NaOH, KOH, K 2 CO 3 or other similarcatalysts were mixed with alcohol, the actual catalysts, alkoxidegroup is formed [107]. Kim et al [108] have developed a process forthe production of Biodiesel from vegetable oils using heterogeneouscatalyst Na/NaOH/Al 2 O 3 . These catalysts showed almostthe same activity under the optimized reaction conditionscompared to conventional homogeneous NaOH catalyst. For analkali catalyzed transesterification, the glycericed and alcoholmust be substantially anhydrous [109] because water makes thereaction partially change to saponification, which produces soap.A number of researchers have worked with feed stokes thathave elevated FFA (free fatty acid) levels [61,110–113]. However,in most cases, alkaline catalysts have been used and the FFAs (Freefatty acids) were removed from the process stream as soap andconsidered waste. Waste greases typically contain from 10 to 25%FFAs. This is far beyond the level that can be converted to Biodieselusing an alkaline catalyst.Fig. 3. Transestrification reaction.

S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216 207Fig. 4. Mechanism of the alkali-catalyzed transesterification of vegetable oils [106,107].Fig. 5. Mechanism of acid catalyzed transesterification.(ii) Acid catalyst transesterification: An alternative process is touse acid catalyst that some researchers have claimed are moretolerant of free fatty acids [114–116]. The mechanism of acidcatalyzed transesterification of vegetable oil (for a monoglyceride)is shown in Fig. 5: However, it can be extended to di- andtriglycerides. The protonation of carbonyl group of the ester leadsto the carbocation, which after a nucleophilic attack of the alcoholproduces a tetrahedral intermediate. This intermediate eliminatesglycerol to form a new ester and to regenerate the catalyst. We canuse acid alkali and biocatalyst in transesterification method. Ifmore water and free fatty acids are in triglycerides, acid catalystcan be used [117]. Transemethylation occur approximately 4000times faster in the presence of an alkali catalyst than thosecatalyzed by the same amount of acidic catalyst [118].(iii) Lipase catalyst transesterification: This transesterificationprocess is like alkali transesterification, only ratio of catalyst andsolvent a stirring time different, and in this transesterification wehave used lipase catalyst. The process is explained in the followingFig. 6.Fig. 6. Flow diagrams comparing biodiesel production using lipase-catalysis [30].Lipases are known to have a propensity to act on long-chainfatty alcohols better than on short-chain ones [119]. Thus, ingeneral, the efficiency of the transesterification of triglycerideswith methanol (methanolysis) is likely to be very low compared tothat with ethanol in systems with or without a solvent.Various studied have been conducted for transesterificationreaction for different catalyst, alcohol and molar ratios at differenttemperature. The results and experimental conditions of somestudies are summarized in Tables 7 and 8.Various types of alcohols – primary, secondary, and straightandbranched-chain – can be employed in transesterification usinglipases as catalysts (Table 8). Linko et al. [128] have demonstratedthe production of a variety of biodegradable esters and polyesterswith lipase as the biocatalyst. In the transesterification of rapeseedoil with 2-ethyl-I-hexanol, 97% conversions of esters was obtainedusing Candida rugosa lipase powder. De et al. [129] investigated theconversion of fatty alcohol esters (C 4 –C 18:1 ) using immobilizedRhizomucor miehei lipase (Lipozyme IM-20) in a solvent-freesystem. The percentage of molar conversions of all correspondingalcohol esters ranged from 86.8 to 99.2%, while the slip meltingpoints of the esters were found to increase steadily with increasingalcohol chain length (from C 4 to C 18 ) and to decline with theincorporation of unsaturation for the same chain length (as fromC 18 to C 18:1 ). Transesterification of the triglycerides sunflower oil,fish oil, and grease with ethanol, i.e. ethanolysis, has also beenstudied. In each case, high yields beyond 80% could be achievedusing the lipases from M. miehei [130], Candida antarctica [131],Pseudomonas cepacia [132], respectively. Nelson et al. [133]

208S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216Table 7Transesterification method.S. No. Sample Catalyst Alcohol Temperature Ration toalcohol to oilResult;YieldsReference1 Microalgae Sulfuric acid Methanol 30 8C 56:1 60% [120]2 Rice bran Two-step acid-catalyzed Sulfuric acid Methanol 60 8C 10:1 < 96%. [121]3 Karang oil KOH,Solid acid catalysts viz.Methanol Tetrahydrofuran 60 8C. 10:1 92% [44]Hb; Zeolite,MontmorilloniteK-10 and ZnO were also used(THF), when used as a cosolvent increased theconversion to 95%.4 Peanut oil NaOH Methanol 50 8C 90% [122]5 Soybean oil Absence of catalyst. Supercritical methanol,CO2 as co-solvent in thereaction system6 Soybean oil Solid super acid catalysts of sulfatedtin and zirconium oxides andtungstated zirconia 2. noctanoic acid7 Sunflower oil Supercritical fluids, Enzyme insupercritical carbon dioxideMethanol 200–300 8C;175–200 8C280 8C 24 andCO 2 /methanol = 0.1,20 h 90% forboth98% [123]Supercritical methanol 200–400 8C 40 (80–100%) [125]and ethanol.(27–30%)8 Canola oil Two-stage process, KOH Methanol 25 8C 6:1 87% [126]9 Sunflowerfrying oilKOH Methanol 25 8C 6:1 90% [127][124]investigated the abilities of lipases in transesterification withshort-chain alcohols to give alkyl esters. The lipase from M. mieheiwas the most efficient for converting triglycerides to their alkylesters with primary alcohols, whereas that from C. antarctica wasthe most efficient for transesterifying triglycerides with secondaryalcohols to give branched alkyl esters. Maximum conversions of94.8–98.5% for the primary alcohols methanol, ethanol, propanol,butanol, and isobutanol, and of 61.2–83.8% for the secondaryalcohols isopropanol and 2-butanol were obtained in the presenceof hexane as a solvent. In solvent-free reactions, however, yieldswith methanol and ethanol were lower than those obtained withhexane; in particular, the yield with methanol decreased to 19.4%.Mittelbach [134] reported transesterilication using the primaryalcohols methanol, ethanol, and 1-butanol, with and withoutpetroleum ether as a solvent. Although the ester yields withethanol and 1-butanol were relatively high, even in reactionswithout a solvent, with methanol only traces of methyl esters wereobtained. Abigor et al. [135] also found that in the conversion ofpalm kernel oil to alkyl esters using p cepacia lipase, ethanol gavethe highest conversion of 72%, while only 15% methyl esters wasobtained with methanol. Lipases are known to have a propensity toact on long-chain fatty alcohols better than on short-chain ones[119]. Thus, in general, the efficiency of the transesterification oftriglycerides with methanol (methanolysis) is likely to be very lowcompared to that with ethanol in systems with or without asolvent.5. Properties of alternative fuel or diesel5.1. Properties of oilThe vegetative oils are mainly characterized by some fuelrelated properties. Some of them are listed in Table 9 [3,139]. Thekinematics viscosity of vegetable oils varies in the range of 30–40cSt at 38 8C. High viscosity of these oils is due to large molecularmass and chemical structure. The fatty acid methyl esters of seedoils and fats have already been found suitable for use as fuel indiesel engine [140] because transesterification provides a fuelviscosity that is close to that diesel.Vegetable oils have high molecular weights in the range of 600–900, which are three or more times higher than diesel fuels. Thevolumetric heating values of these oils are in the range of 39–Table 8Enzymatic transesteritication reactions using various types of alcohols and lipases.S.N. Oil Alcohol Lipase Conversion (%) Solvent Ref.1 Rapeseed 2-Ethyl-1-hexanol C. rugosa 97 None [128]2 Mowrah, Mango, Kernel, Sal C, -C, alcohols M. miehei (Lipozyme IM-20) 86.8–99.2 None [129]3 Sunflower Ethanol M. meihei (Lypozyme) 83 None [130]4 Fish Ethanol C. anturctica 100 None [131]5 Recycled restaurant Grease Ethanol J. cepacia (Lipase PS-30) +C. anturclica (Lipase SP435)85.4 None [132]6 Tallow, Soybean, Rapeseed Primary alcohols a; Secondaryalcohols b A-Methanol,ethanol, propanol, butanol, andisobutanol. B-Isopropanoland 2-butanolM. miehei (Lipozyme IM60) C.antarctica (SP435) M. miehei(Lipozyme IM60) M. miehei(Lipozyme IM60)94.8–98.5;61.2–83.8;19.4–65.5Hexane; Hexane;None; None7 Sunflower Methanol; Methanol; Ethanol P. juorescens 3; 79; 82 None; Petroleum; [134]ether None8 Palm kernel; Oil Methanol; Ethanol I. cepuciu (Lipase PS-30) 15; 72 None; None [135]9 Soybean oil Methanol Rhizomucor miehei92.2% Molar ratio 3.4:1, [136](Lipozyme IM-77) enzymeamount 0.9 BAUN10 Soybean oil Methanol C. antarctica lipase 93.8% >1/2 molar[137]equivalent MeOH11 Sunflower oil Methanol Pseudomonas fluorescens(>90%) oil:methanol (1:4.5) [16](Amano AK)12 Palm oil Methanol Rhizopus oryzae 55% (w/w) Water [138][133]

S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216 209Table 9Oils characteristics.VegetableoilKinematic viscosityat 38 8C (mm 2 /s)Cetane No.(8C)Heating Value(MJ/kg)Cloud Point(8C)Pour Point(8C)Flash Point(8C)Density(kg/l)Carbonresidue (wt.%)Corn 34.9 37.6 39.5 1.1 40 277 0.9095 0.24Cottonseed 33.5 41.8 39.5 1.7 15 234 0.9148 0.24Crambe 53.6 44.6 40.5 10.0 12.2 274 0.9048 0.23Linseed 27.2 34.6 39.3 1.7 15.0 241 0.9236 0.22Peanut 39.6 41.8 39.8 12.8 6.7 271 0.9026 0.24Rapeseed 37.0 37.6 39.7 3.9 31.7 246 0.9115 0.30Safflower 31.3 41.3 39.5 18.3 6.7 260 0.9144 0.25Sesame 35.5 40.2 39.3 3.9 9.4 260 0.9133 0.24Soya bean 32.6 37.9 39.6 3.9 12.2 254 0.9138 0.25Sunflower 33.9 37.1 39.6 7.2 15.0 274 0.9161 0.27Palm 39.6 42.0 – 31.0 – 267 0.9180 0.23Babassu 30.3 38.0 – 20.0 – 150 0.9460 –Diesel 3.06 50 43.8 – 16 76 0.855 –40 MJ/kg, which are low, compared to diesel fuels (about 45 MJ/kg). The presence of chemically bound oxygen in vegetable oilslowers their heating values by about 10%. The cetane numbers arein the range of 34–42. The cloud and pour points of vegetable oilsare higher than that of diesel fuels.5.2. Properties of BiodieselThe characteristics of Biodiesel are close to diesel fuels, andtherefore Biodiesel becomes a strong source to replace the dieselfuels. The conversion of triglycerides into methyl or ethyl estersthrough the transesterification process, reduces the molecularweight to one-third that of the triglyceride and also reduces theviscosity by a factor of about eight and increases the volatilitymarginally. Biodiesel has viscosity close to diesel fuels. Theseesters contain 10–11% oxygen by weight, which may encouragemore combustion than hydrocarbon-based diesel fuels in anengine. The cetane number of Biodiesel is around 50. The use oftertiary fatty amines and amides can be effective in enhancing theignition quality of the finished diesel fuel without having anynegative effect on its cold flow properties. Since the volatilityincreases marginally, the starting problem persists in coldconditions. Biodiesel has lower volumetric heating values (about12%) than diesel fuels but has a high cetane number and ash point(Table 10).Chiu [145] reported the cold flow properties of Biodiesel(B100) and 80% (B80) to 90% Biodiesel in kerosene were evaluatedwith pour point depressants, toward the objective of identifyingapproaches to transport and mix Biodiesel with diesel in coldclimates. Four cold flow improver additives were tested at 0.1–2%in B80, B90, and B100 blends. Two additives significantly—amixture of 0.2% additives, 79.8% Biodiesel, and 20% kerosenereduced the pour point of B100 by 27 8C. The unrefined Biodieselsshowed higher lubricity properties than refined Biodiesel. Thechemical factors influencing the lubricity properties of Biodieselswere investigated by Hu et al. [146]. Methyl ester and monoglyceridesare the main composition that determines the lubricityof Biodiesel that meets international standards. Free fatty acids anddiglycerides can also affect the lubricity of Biodiesel, but not somuch as monoglycerides.Biodiesel is considered clean fuel it has no sulphur no aromaticsand has about 10% built in oxygen, which helps it to burn fully itshigher cetane number improves the ignition quality even whenblended in the petroleum diesel.It is a general property of hydrocarbons that the auto-ignitiontemperature is higher for more volatile hydrocarbons. Therefore,the less volatile middle distillate fractions of crude oil boiling in therange of 250–370 8C are suitable as diesel fuels. The hydrocarbonspresent in the diesel fuels include parafins, naphthenes, olefins andaromatics. Carbon numbers of these hydrocarbons ranges from 12to 18.5.3. Determination and measurement methods for oil/BiodieselpropertiesThe properties of the methyl esters were determined bymethods specified in the Handbook of Analytical Methods for FattyAcid Methyl Esters as Diesel Substitutes [147] or by equivalentstandard methods, which given in Table 11.A considerable amount of experience exists in the U.S. with a20% blend of Biodiesel with 80% diesel fuel (B20). AlthoughBiodiesel (B100) can be used, blends of over 20% Biodiesel withdiesel fuel should be evaluated on a case-by-case basis until furtherexperience is available [148]. Table 12 shows the Comparison ofproperties of diesel and Biodiesel Acid value, viscosity and densitywere determined according to ISO 660, 3104 and 3675, respectively.Ash content, cloud point and pour point were measuredusing BS 2000, cold filter plug point (CFPP) using BS 6188 andiodine value using DIN 53241. Total and free glycerine wereTable 10Physical properties of Biodiesel.Vegetable oilmethyl esterKinematic viscosity(mm 2 /s)Cetaneno.Lower heatingvalue (MJ/kg)Cloudpoint (8C)Pourpoint (8C)Flashpoint (8C)Density (kg/l)ReferencePeanut 4.9 (37.8 8C) 54 33.6 5 – 176 0.883 [141]Soya bean 4.5 (37.8 8C) 45 33.5 1 7 178 0.885 [142]Babassu 3.6 (37.8 8C) 63 31.8 4 – 127 0.875 [126]Palm 5.7 (37.8 8C) 62 33.5 13 – 164 0.880 [143]Sunflower 4.6 (37.8 8C) 49 33.5 1 – 183 0.860 [141]Tallow – – – 12 9 96 – [30]Rapeseed b 4.2 (40 8C) 5 l–59.7 32.8 0.882 (15 8C) [144]Palm b 4.3-4.5(40 8C) 64.3–70 32.4 – – – 0.872–0.877 (15 8C) [144]Soybean b 4.0 (40 8C) 45.7-56 32.7 – – – 0.880 (15 8C) [144]Diesel 3.06 50 43.8 – 16 76 0.85520% BD blend 3.2 51 43.2 – 16 128 0.859

210S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216Table 11ASTM specifications (D6751) for B100.Property ASTM Method Limits UnitsFlash Point D93 130 min 8CWater & Sediment D2709 0.050 max %vol.Kinematic Viscosity (40 8C) D445 1.9–6.0 mm 2 /sSulfated Ash D874 0.020 max. %massSulfur D5453. 0.05 max %massCopper Strip Corrosion D130 No. 3 max.Cetane D613 47 minCloud Point D2500 Report 8CCarbon Residue (100% Sample) D4530 a 0.050 max % massAcid Number D664 0.80 max Mg KOH/gFree Glycerin D6584 0.020 max %massTotal Glycerin D6584 0.240 max %massPhosphorous Content D4951 0.001 max %massDistillation Temperature, Atmospheric Equivalent Temperature (90% Recovered) D1160 360 max 8Ca The carbon residue shall be run on the 100% sample.determined enzymatically [147]. Methanol content was measuredby gas chromatography using a Porapak Q column (Perkin-ElmerLtd., Beaconsfield, Buckinghamshire), 900 mm 3mm at 908C,injector at 110 8C, 25 ml/min nitrogen flow, and flame ionizationdetector. Lubricating oils were screened for the presence of methylesters by thin-layer chromatography [150,151]. If detected theywere determined quantitatively by following the method:lubricating oil (5 g) was saponified [152] and the molar amountof methyl ester in the oil was calculated from the difference in titrevolumes between the original and used oils. The percentage ofmethyl ester in the used oil was determined by dividing the molaramount in the used oil by the molar amount in 5 g methyl ester,also determined by saponification.6. Performance of alternative fuel (oil and Biodiesel) as a fuel6.1. Use of oils as fuel and their performanceMany researchers have concluded that vegetable oils holdpromise as alternative fuels for diesel engines [3,4]. During WorldWar II Seddon [153] experimented with using several differentvegetable oils in Perkins P-6 diesel engine with great success andconcluded that vegetable oils could be used to power a vehicle undernormal operating conditions. However it was noted that much morework was needed before vegetable oils used as a reliable substitutefor diesel fuel. Bruwer et al. [154] studied the use of sunflower seedTable 12Comparison of fuel properties between diesel and Biodiesel.Fuel Property Diesel BiodieselFuel Standard ASTM D975 ASTM PS 121Fuel composition C10-C21 HC C12-C22 FAMELower Heating Value (Btu/gal) 131,295 117,093Kin. Viscosity, @ 40 8C 1.3–4.1 1.9–6.0Specific Gravity kg/l @ 60 8F 0.85 0.88Density, lb/gal @ 15 8C 7.079 7.328Water, ppm by wt. 161 0.05% maxCarbon (wt.%) 87 77Hydrogen (wt.%) 13 12Oxygen, by dif. wt.% 0 11Sulfur (wt.%) .05 max 0.0 - 0.0024Boiling Point (8C) 188-343 182-338Flash Point (8C) 60-80 100-170Cloud Point (8C) -15 to 5 -3 to 12Pour Point (8C) -35 to -15 -15 to 10Cetane Number 40-55 48-65Stoichiometric Air/Fuel Ratio (w/w) 15 13.8BOCLE Scuff (g) 3,600 >7,000HFRR (mm) 685 314Source: [149].oil as a renewable energy source. When operating tractor with 100%sunflower oil instead of diesel fuel, an 8% power loss occurred after1000 h of operation, which was corrected by replacing the fuelinjector, and injector pump. Yarbrough et al. [155] reported that rawsunflower oils were found to be unsuitable fuel, while refinedsunflower oil was found to be satisfactory. Tahir et al. [156] reportedthat oxidation of sunflower oil left heavy gum and wax deposit ontested tractor, which could lead to engine failure. Bettis et al. [157]and Engler et al. [158] reported that the sunflower seed oil isacceptable only for short-term use as a fuel source but long termdurability test indicated sever problems due to carbonization ofcombustion chamber. Goering et al. [159] studied the characteristicproperties of eleven vegetable oils to determine which oil would bebest suited for use as an alternative fuel source. Bacon et al. [160]evaluated the use of several vegetable oils as potent fuel sources andreported that use of these oils caused carbon build up in thecombustion chamber. Schoedder [161] used rapeseed oil as a dieselfuel replacement in Germany with mixed results. Short-term enginetests indicated similar energy outputs in both rapeseed oil and dieselfuel. Initial long-term engine tests showed that difficulties arose inengine operation after 100 h due to deposits on piston rings, valvesand injectors. Auld et al. [162] analyze rapeseed oil and showed arelationship between viscosity and fatty acid chain length. Bettiset al. [157] reported that rapeseed oil contained 94–95% of theenergy content of diesel fuel, and to be approximately 15 timesviscous. Reid et al. [163] evaluated chemical and physical propertiesof 14 vegetable oils. They pointed out that the oils are verydifferently from petroleum-based fuel because of ‘high viscosity’.Engine test showed that carbon deposits in the engine were reducedif the oil was heated prior to combustion. It was also noted thatcarbon deposit levels differed from oils with similar viscositiesbecause of oil composition.6.2. Need of blending and its effect on performanceEngelman et al. [164] reported that 10–50% soybean oil fuelblends with diesel minimize the carbon deposition in combustionchamber. Quick [165] used over 30 different vegetable oils tooperate compression engines and reported that the use of rawvegetable oil fuels can lead to premature engine failure. Blendingvegetable oils with diesel fuel was found to be a method to reducechocking and extend engine life. Sims et al. [166] indicated thatshort-term engine tests with 50% vegetable oil fuel blend had noadverse effects. Carbon deposits on combustion chamber componentswas found to be approximately same as that found in enginesoperated on 100% diesel fuel. The similar results were reported byBartholomew [167], Barsic and Humke [168], Fort et al. [169],Baranescu and Lusco [170], Worgetter [171], Wanger and Peterson

S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216 211[172] and Vander Walt and Hugo [173] working on different oilssuch as pea nut oil, cottonseed oil, sunflower oil, rapeseed oil, andsun flower oil respectively. German et al. [174] reported thatcarbon deposits on the internal engine components were greaterfor the tractor fueled with 50/50 sunflower oil/diesel than thosefueled with a 25/75 sunflower/diesel fuel blend. Sapaun et al. [175]reported that power out puts were nearly the same for palm oil,blend of palm oil and diesel fuel, and 100% diesel fuel. Short-termusing of palm oil fuel showed no adverse effects. Hofman et al.[176] and Peterson et al. [177] indicated that while vegetative oilfuel blends had encouraging results in short term testing, problemsoccurred in long term durability tests. Pestes and Stanislao (1984)[178] used a one to one blend of vegetable oil and diesel fuel tostudy the piston rings deposits. Premature piston ring sticking andcarbon built up due to the use of the one to one fuel blend causedengine failure. These investigators suggested that to reduce pistonring deposits a fuel additive or a fuel blend with less vegetative oilwas needed. The atomization and injection characteristic ofvegetative oils were significantly different from that of diesel fueldue to the higher viscosity of the vegetative oils [179]. Engineperformance tests showed that power output slightly decreasedwhen using vegetable oil fuel blends. Nag et al. [180] showed thatperformance tests using fuel blend as great as 50–50 seed oil fromthe Indian Ambulate plant and diesel fuel exhibited no loss ofpower. Knock free performance with no observable carbondeposits on the functional parts of the combustion chamber wereobserved during these tests. McCutchen [181] compared engineperformance of direct injection engine to indirect injection engineswhen fueled with 30% soybean oil 70% diesel fuel. The resultshowed that indirect injection could be operated on this fuel blendwhile the direct injection engine could not without catastrophicengine failure occurring. The direct injection engines showedinjector coking and piston ring sticking as a result of usingsunflower oil. McDonnell et al. [182] studied the use of a semirefinedrapeseed oil as a diesel fuel extender. Test results indicatedthat the rapeseed oil could serve as a fuel extender at inclusionrates up to 25%. As a result of using rapeseed oil as a fuel, injectorlife was shortened due to carbon build up.6.3. Use of esters as fuel and their performanceUsta [54] showed that tobacco seed oil methyl ester can bepartially substituted for the diesel fuel at most operatingconditions in term of performance parameter and emissionswithout any engine modification and preheating of the blends.Frohlich and Rice [37] reported that in vehicle testes, the reductionin fuel economy with camellia methyl ester did not lead to a morerapid deterioration of the lubricating oil. Kalligeros et al. [183]reported that the engine was fueled with pure marine diesel fueland blends containing two types of Biodiesel, at proportion up to50%, the performance was satisfactory. Monyem and Van Gerpen[184] evaluated the impact of oxidized Biodiesel on engineperformance and emission. A John Deere 4276 turbocharged DIdiesel engine was fueled with oxidized and unoxidized Biodieseland the performance and emission were compared with No. 2diesel fuel. The engine performance of the neat Biodiesels and theirblends was similar to that of No. 2 diesel fuel with the samethermal efficiency, but higher fuel consumption. Machacon et al.[185] studied the operation of the test engine with pure coconut oiland coconut oil diesel fuel blends for a wide range of engine loadcondition was shown to be successful even without enginemodification. It was also shown that increasing the amount ofcoconut oil in the coconut oil—diesel fuel blend resulted in lowersmoke and NO x emission. Tan [26] estimated reduction in net CO 2emission at 77–104 g/MJ of diesel replaced by Biodiesel. Hepredicted reduction in CO 2 emission which are much greater thanvalues reported in recent studied on Biodiesel derived from othervegetable oils, due to both the large amount of potential fuel in theresidual biomass and to the lower energy input the traditionalcoconut farming techniques. Wang [186] shown that the heavytrucks fueled by B35 emitted significantly lower particulate matter(PM) and moderately lower carbon monoxide (CO) and hydrocarbon(HC) than the same trucks fueled by number 2 diesel (D2).Oxides of nitrogen (NO x ) emission from B35 and D2, however, weregenerally in the same level. Clark et al. [187] reported that thederivatives particularly the methyl esters of rapeseed, soybean andother seed oils have been shown to be similar to diesel fuel inperformance. Mittelbach et al. [188] examined that the exhaustfrom combustion of the vegetative oil esters is lower in carbonmonoxide and particulate content but higher in nitrogen oxidethan diesel fuel. Peterson et al. [112] used ethyl ester of rapeseed asa Biodiesel and reported that no problems or unusual events wereencountered with the truck’s operation. The truck was completelyunmodified as to the engine and fuel system. However, theBiodiesels were considerably less volatile than the conventionaldiesel fuel. Peterson et al. [6] stated that the Biodiesel esters can beused directly or as blends with diesel fuel in a diesel engine.Kalam and Masjuki [52] evaluate the effect of anticorrosionadditive in Biodiesel isolated from palm as a palm ester on dieselengines, performance, emission and wear characteristics. Cardoneet al. [36] used the non-food use of Brassica carinata oil for Biodieselproduction. The Biodiesel, produced by transesterification of the oilextracted from the Brassica carinata seeds, displayed physical–chemical properties suitable for the use as diesel car fuel. Rahemanand Phadatare [43] used the blend of the Pongamia pinnata methylester and diesel to evaluate the diesel engine emissions andperformance and found reduction in exhaust emission togetherwith increase in torque, brake power, brake thermal efficiency andreduction in brake-specific, fuel consumption. The blends ofPongamia pinnata esterifide oil is a suitable alternative fuel fordiesel and could help in controlling air pollution. Carraretto et al.[189] checked the operation of a Biodiesel fueled boiler for somemonths and stated that the Biodiesel seems to be a promisingsolution for boilers with only minor adjustments and theperformances are comparable with oil operation. Carbon monoxideemission is reduced but no NO 3 are increased. Shi et al. [190]prepared blend of 20% (v/v) ethanol/methyl soyate and observedthat particulate matters decreased with increasing oxygenatecontent in the fuel but nitrogen oxide (NO x ) emission increased.Spataru and Romig [191] and Schumacher and Bolgert [192] bothstudied several blends of No. 2 diesel and SME or canola methylester (CME) to determine and compare engine emission from theDetroit Diesel corporation (DDC) 6V92TA engine (a type of dieselengine widely used in transit buses and heavy trucks) operated onthose fuels. Grabaski et al. [193] also employed a 1991 DDC series60 engine to determine emission of NO x , CO, HC, and PM that resultfrom blending Biodiesel (methyl soy ester) and conventionaldiesel. The test showed that as the percentage of Biodiesel blend infuel increased, the NO x increased but HC, CO, and PM decreased.Schumacher and Peterson [191,194] both conducted about100,000 mi of road test on Cummins B5 9L engine running on100% Biodiesel in Dodge pickups. No fuel related problems werenoted during the tests on the modification trucks. Both emissiontest results showed that CO, HC, and smoke exhaust emission fromBiodiesel tend to be lower. Chang et al. [195] studied the effects ofusing blends of methyl and isopropyl ester of soybean oil with no 2diesels at several steady state operational conditions in a fourcylinderturbocharged John Deer 4276T engine. Both methyl andisopropyl ester provided significant reductions in PM emission ascompared with no. 2 diesel fuels. Emission of CO, and HC were alsoreduced significantly, but NO x increased by about 12%. The fattyacid methyl esters of seed oils and fats and have already been found

212S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216suitable for use as fuel in diesel engine [140] because transesterificationprovides a fuel viscosity that is close to that diesel fuel. InIndia the prohibitive cost of edible oils prevent their use in Biodiesel,but non edible oils are affordable for Biodiesel production. The maincommodity sources for Biodiesel in India can be non-edible oilsobtained from plant species such as Jatropha curcas (ratanjyot),Pongamia pinnata (Karanj), Callophyllum inophyllum (Nagchampa),Hevca brasiliensis (Rubber), etc. Biodiesel can be blended at any levelwith petroleum diesel to create a Biodiesel blend or can be use in itspure form just like petroleum diesel. The diesel engines operated byBiodiesel are required very little or low engine modification becauseBiodiesel has properties similar to petroleum diesel fuels. It can bestored just like a petroleum diesel fuel and hence does not requireseparate infrastructure.7. Problems arises and their solution during the use ofalternative fuels7.1. Problem arises during the use of oilThere are many problems associated with direct use of oil inengine. A potential solution has been given to overcome from theseproblems caused by use of vegetative oil at the place of petroleumdiesel (Table 13).These effects can be reduced or eliminated using filtered, fryingoil and the blend of 95% used cooking oil and 5% diesel fuel [196].Due to their high viscosity (about 11–17 times higher than dieselfuel) and low volatility, they do not burn completely and formdeposits in the fuel injector of diesel engines [33]. Different wayshave been considered to reduce the high viscosity of vegetable oils:(a) dilution of 25 parts of vegetable oil with 75 parts of diesel fuel,(b) microemulsions with short chain alcohols such as ethanol ormethanol,(c) thermal decomposition, which produces alkanes, alkenes,carboxylic acids,(d) catalytic cracking, which produces alkanes, cycloalkanes andalkylbenzenes, and(e) transesterification with ethanol or methanol.7.2. Problem arises during the use of BiodieselThe technical disadvantages of Biodiesel/fossil diesel blendsinclude problems with fuel freezing in cold weather, reducedenergy density, and degradation of fuel under storage forprolonged periods. One additional problem is encountered whenblends are first introduced into equipment that has a long historyof pure hydrocarbon usage. Hydrocarbon fuels typically form alayer of deposits on the inside of tanks, hoses, etc. Biodiesel blendsloosen these deposits, causing them to block fuel filters. However,this is a minor problem, easily remedied by proper filtermaintenance during the period following introduction of theBiodiesel blend [200].The disadvantages of Biodiesel are:- Constraints on the availability of agricultural feedstockimpose limits on the possible contribution of Biodiesels totransport.- The kinematic viscosity is higher than diesel fuel. This affects fuelatomization during injection and requires modified fuel injectionsystems.- Due to the high oxygen content, it produces relatively high NO xlevels during combustion.- Oxidation stability is lower than that of diesel so that underextended storage conditions it is possible to produce oxidationproducts that may be harmful to the vehicle components.- Biodiesel is hygroscopic. Contact with humid air must beavoided.- Production of Biodiesel is not sufficiently standardised. Biodieselthat is outside European or US standards can cause corrosion, fuelsystem blockage, seal failures, filter clogging and deposits atinjection pumps.- The lower volumetric energy density of Biodiesel means thatmore fuel needs to be transported for the same distance travelled.- It can cause dilution of engine lubricant oil, requiring morefrequent oil change than in standard diesel-fueled engines.- A modified refuelling infrastructure is needed to handleBiodiesels, which adds to their total cost.8. Economic viability of BiodieselBiodiesel has become more attractive recently because of itsenvironmental benefits and the fact that it is made from renewableresources. The remaining challenges are its cost and limitedavailability of fat and oil resources. There are two aspects of thecost of Biodiesel, the costs of raw material (fats and oils) and thecost of processing. The cost of raw materials accounts for 60–75% ofthe total cost of Biodiesel fuel [98].Table 13Known problems, probable cause and potential solutions for using straight vegetable oil in diesels engine [22].Problem Probable cause Potential solutionShort-term Cold weather starting High viscosity, low cetane, and low flash point of vegetable oils Preheat fuel prior to injection. Chemically alter fuel to an esterPlugging and gumming of filters, Natural gums (phosphatides) in vegetable oil. Other ashPartially refine the oil to remove gums. Filter to 4-mmlines and injectorsEngine knocking Very low cetane of some oils. Improper injection timing Adjust injection timing. Use higher compression engines.Preheat fuel prior to injection. Chemically alter fuel to an esterLong-term Coking of injectors onPiston and head of engineHigh viscosity of vegetable oil, incomplete combustion of fuel.Poor combustion at part load with vegetable oilsHeat fuel prior to injection. Switch engine to diesel fuelwhen operation at part load. Chemically alter the vegetableoil to an esterCarbon deposits on Pistonand head of engineExcessive engine wearFailure of engine lubricating oildue to PolymerizationHigh viscosity of vegetable oil, incomplete combustion of fuel.Poor combustion at part load with vegetable oilsHigh viscosity of oil, incomplete combustion of fuel.Poor combustion at part load with vegetable oils.Possibly free fatty acids in vegetable oil. Dilution ofengine lubricating oil due to blow-by of vegetable oilCollection of polyunsaturated Vegetable oil blow-by incrankcase to the point where polymerization Occursvegetable oil to an ester. Increase motor oil changes.Motor oil additives to inhibit oxidationHeat fuel prior to injection. Switch engine to diesel fuelwhen operation at part loads. Chemically alter the vegetableoil to an esterHeat fuel prior to injection Switch engine to diesel fuelwhen operation at part loads. Chemically alter the vegetableoil to an ester. Increase motor oil changes. Motor oil additivesto inhibit oxidationHeat fuel prior to injection. Switch engine to diesel fuelwhen operation at part load. Chemically alter the

S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216 213The use of used cooking oil can lower the cost significantly.However, the quality of used cooking oils can be bad [198].Studies are needed to find a cheaper way to utilize used cookingoils to make Biodiesel fuel. There are several choices, firstremoving free fatty acids from used cooking oil beforetransesterification, using acid catalyzed transesterification, orusing high pressure and temperature [199]. In terms ofproduction cost, there also are two aspects, the transesterificationprocess and by-product (glycerol) recovery. A continuoustransesterification process is one choice to lower the productioncost. The foundations of this process are a shorter reaction timeand greater production capacity. The recovery of high qualityglycerol is another way to lower production cost. Because littlewater is present in the system, the Biodiesel glycerol is moreconcentrated. Unlike the traditional soap glycerol recoveryprocess, the energy required to recover Biodiesel glycerol islow due to the elimination of the evaporation process. Inaddition, the process is also simpler than soap glycerol recoverysince there is a negligible amount of soap in Biodiesel glycerol.This implies that the cost of recovering high quality glycerol fromBiodiesel glycerol is lower than that of soap glycerol and that thecost of Biodiesel fuel can be lowered if a Biodiesel plant has itsown glycerol recovery facility. With the increase in global humanpopulation, more land may be needed to produce food for humanconsumption (indirectly via animal feed). So insufficient landmay also increase the production cost of Biodiesel plants. Theproblem already exists in Asia where vegetable oil prices arerelatively high. The same trend will eventually happen in the restof the world. This is the potential challenge to Biodieselproduction. Biodiesel can be used most effectively as asupplement to other energy forms. Biodiesel is particularlyuseful in mining and marine situations where lower pollutionlevels are important. Biodiesel also can lower US dependence onimported petroleum based fuel.9. ConclusionIn recent years, Biodiesel has become more attractive as analternative fuel for diesel engines because of its environmentalbenefits and the fact that it is made from renewable resources.Several methods available for producing Biodiesel, transesterificationof natural oils and fats are currently the method of choice.The purpose of the process is to lower the viscosity of the oil or fat.Researchers focused mainly the edible oils to produce theBiodiesel because of easily availability and familiarity. Very fewresearchers concentrated on Non edible oil for the same purpose.Non-edible oils can also be utilized for making Biodiesel fuel. Forthe production of Biodiesel fuel, an alkali-catalysis process hasbeen established that gives high conversion levels of oils to methylesters. Enzymatic processes using both extra cellular andintracellular lipases have recently been developed. The cost oflipase production is the main hurdle to commercialization of thelipase-catalyzed process; several attempts have been made todevelop cost-effective systems. In terms of production cost, therealso are two aspects, the transesterification process and byproduct(glycerol) recovery. A continuous transesterificationprocess is one choice to lower the production cost. Thefoundations of this process are a shorter reaction time andgreater production capacity. The recovery of high quality glycerolis another way to lower production cost. Land may be a costincreasing factor for Biodiesel production, because of more andmore land required to live the growing population. To overcomethe land problem, the high yielding Biodiesel plants (non edibleproducing plants) should be grown in marginal and waste landareas.References[1] Shay EG. Diesel fuel from vegetable oils: status and opportunities. BiomassBioenergy 1993;4:227–42.[2] ASAE, Vegetable oil fuels: Proceedings of the international conference onplant and vegetable; 1982.[3] Goering CE, Schwab AW, Daugherty MJ, Pryde EH, Heakin AJ. Fuel propertiesof eleven oils. Trans ASAE 1982;25:1472–83.[4] Bagby MO. Vegetable oils for diesel fuel: opportunities for development. In:International winter meeting of the ASAE; 1987.[5] Zhang Y, Dube MA, McLean DD, Kates M. Biodiesel production from wastecooking oil: 2. Economic assessment and sensitivity analysis. Bioresour.Technol.; submitted for publication.[6] Perkins LA, Peterson CL. Durability testing of transesteriWed winter rape oil(Brassica napus L.) as fuel in small bore, multi-cylinder, DI, CI engines. SAEPaper 911,764; 1991.[7] Martini N, Shell JS. Plant oils as fuels—present state of science and futuredevelopment. Berlin: Spinger; 1998. p. 276.[8] Ma F, Clement LD, Hanna MA. The effect of mixing on transesterification ofbeef tallow. Bioresour Technol 1999;69:289–93.[9] Ksorbitz W. Biodiesel production in Europe and North American, an encouragingprospect. Renew Energy 1999;16:1078–83.[10] Martini N, Shell JS. Plant oils as fuels—present state of science and futuredevelopment. Berlin: Spinger; 1998. p. 276.[11] Ratledge, Boulton CA. Fats and oils. In: Moo Young M., Blanch HN,Dream S, Wang DIC, editors, Comparative Biotechnology in industry,agriculture and medicine, vol. 3. New York: Pergamon Press; 1985, p. 983–1003.[12] Lee I, Johnson LA, Hammond EG. Use of branched-chain esters to reduce thecrystallization temperature of Biodiesel. JAOCS 1995;72:1155–60.[13] Freedman B, Butterfield RO, Pryde EH. Transesterification kinetics of soybeanoil. J Am Oil Chem Soc 1986;63(10):1375–80.[14] Noureddini H, Zhu D. Kinetics of transesterification of soybean oil. J Am OilChem Soc 1997;74:1457–63.[15] Antolín G, Tinaut FV, Briceño Y, CastañoV,Pérez C, Ramírez AI. Optimisa tionof Biodiesel production by sunflower oil transesterification. Bioresour Technol2002;83(2):111–4.[16] Mohamed M, Soumanoua B, Uwe T, Bornscheuer A. Improvement in lipasecatalyzedsynthesis of fatty acid methyl esters from sunflower oil. EnzymeMicrob Technol 2003;33:97–103.[17] Darnoko D, Cheryan M. Kinetics of palm oil transesterification in a batchreactor. J Am Oil Chem Soc 2000;77(12):1263–7.[18] Kusdiana D, Saka S. Kinetics of transestrification in rapseed oil to biodieselfuel as tested in supercritical methanol. Fuel 2001;80:693–5.[19] Zou L, Atkinson S. Characterising vehicle emissions from the burning ofbiodiesel made from vegetable oil. Environ Technol 2003;24:1253.[20] Köse Ö, Tüter M, Aksoy HA. Immobilized Candida antarctica lipase-catalyzedalcoholysis of cotton seed oil in a solvent-free medium. Bioresour Technol2002;83:125–9.[21] Foidl N, Foidl G, Sanchez M, Mittelbach M, Hackel S. Jatropha curcas L. as asource for the production of biofuel in Nicaragua. Bioresour Technol1996;58:77–82.[22] Fangrui Ma, Hanna Milford A. Biodiesel production: a review. BioresourTechnol 1999;7:01–15.[23] Haas MJ. Improving the economics of biodiesel production through the use oflow value lipids as feedstocks: vegetable oil soapstock. Fuel Process Technol2005;86:1087–96.[24] Kulkarni MG, Dalai AK, Bakhshi NN. Utilization of green seed canolaoil for biodiesel production. J Chem Technol Biotechnol 2006;81:1886–93.[25] Ugarte DGT, Ray DE. Biomass and bioenergy applications of the POLYSYSmodeling framework. Biomass Bioenergy 2000;18(4):291–308.[26] Tan RR, Culaba AB, Purvis MRI. Carbon balance implications of coconutbiodiesel utilization in the Philippine automotive transport sector. BiomassBioenergy 2004;26:579–85.[27] Haller, Youssoufin, Hardjosuwito B, Hoesnan A. Rubber Seed Oil Abstr1978;3(7):17.[28] Barnwal BK, Sharma MP. Prospects of biodiesel production from vegetableoils in India. Renew Sust Energy Rev 2005;9:363–78.[29] Srivastava A, Prasad R. Triglycerides-based diesel fuels. Renew Sust EnergyRev 2000;4:111–33.[30] Fukuda H, Kondo A, Noda H. Biodiesel fuel production by transesterificationof oils. J Biosci Bioeng 2001;92:405–16.[31] Kamini NR, Fujii T, Kurosu T, Iefuji H. Production, purification and characterizationof an extracellular lipase from the yeast, Cryptococcus sp. S-2. ProcessBiochem (Oxford UK) 2000;36:317–24.[32] Peterson CL, Auld DL, Korus RA. Winter rape oil fuel for diesel engines:recovery and utilization. JAOCS 1983;60:1579–87.[33] Demirbas A. Biodiesel fuels from vegetable oils via catalytic and non-catalyticsupercritical alcohol transesterifications and other methods: a survey. EnergyConvers Manage 2003;44:2093–109.[34] Mahmud K, et al. Biochemical analysis of oil from seeds of abutilon muticum.Pakistan J Biol Sci 2003;6(22):1880–2.[35] Abreu FR, Lima DG, Hamú EH, Wolf C, Suarez PAZ. Utilization of metalcomplexes as catalysts in the transesterification of Brazilian vegetable oilswith different alcohols. J Mol Catal A Chem 2004;209:29–33.

214S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216[36] Cardone M, et al. Brassica carinata as an alternative crop fror the production ofbiodiesel in Italy: agronomic evalution, fuel production by Transesterificationand characterization. Biomass Bioenergy 2003;25:623–36.[37] Rice Fröhlich. Ind Crops Prod 2005;21:25–31.[38] Encinar JM, Gonzalez JF, Rodriguez JJ, Tejedor A. Biodiesel fuels from vegetableoils: transesterification of Cynara cardunculus L. oils with ethanol.Energy Fuels 2002;16:443–50.[39] Shah S, Sharma S, Gupta MN. Biodiesel preparation by lipase-catalyzedtransesterification of Jatropha oil. Energy Fuels 2004;18:154.[40] Pramanik K. Properties and use of Jatropha curcas oil and diesel fuel blends incompression ignition engine. Renewable Energy 2003;28(2):239–48.[41] Banerji R. Jatropha nana seed oil for energy. Biomass Bioenergy 1991;I(4):247.[42] Canoira L, et al. Biodiesel from Jojoba oil-wax: transesterification withmethanol and properties as a fuel. Biomass Bioenergy 2006;30:76–81.[43] Raheman H, Phadatare AG. Diesel engine emissions and performance fromblends of karanja methyl ester and diesel. Biomass Bioenergy 2004;27:393–7.[44] Sarma AK, et al. A Comprehensive Analysis of Fuel Properties of Biodieselfrom Koroch Seed Oil. Energy Fuels 2005;19:656–7.[45] Goodrum JW, Geller DP. Influence of fatty acid methyl esters from hydroxylatedvegetable oils on diesel fuel lubricity. Bioresour Technol 2005;96:851–5.[46] Puhan S, Vedaraman N, Ram BVB, Sankarnarayanan G, Jeychandran K. Mahuaoil (Madhuca Indica seed oil) methyl ester as biodiesel-preparation andemission characterstics. Biomass Bioenergy 2005;28:87–93.[47] Ghadge SV, Raheman H. Biodiesel production from mahua (Madhuca indica)oil having high free fatty acids. Biomass Bioenergy 2005;28:601–5.[48] Altin R, Cetinkaya S, Yucesu HS. The potential of using vegetable oil fuels asfuel for diesel engines. Energy Convers Manage 2001;42(5):529–38.[49] Ikwuagwu OE, Ononogbu IC, Njoku OU. Production of biodiesel using rubber[Hevea brasiliensis (Kunth. Muell.)] seed oil. Indust Crops Prod 2000;12:57–62.[50] Abdullah A, Bassari MNH, Abdullah A, Basri MNH. Selected readings on palmoil and its uses. Malaysia: PORIM; 2002. p. 25–176.[51] Ooi Yean-Sang, Zakaria Ridzuan, Mohamed Abdul Rahman, Bhatia Subhash.Catalytic conversion of palm oil-based fatty acid mixture to liquid fuel.Biomass Bioenergy 2004;27(5):477–84.[52] Kalam MA, Masjuki HH. Biodiesel from palmoil—an analysis of its propertiesand potential. Biomass Bioenergy 2002;23:471–9.[53] Karmee SK, Chadha A. Preparation of biodiesel from crude oil of Pongamiapinnata. Bioresour Technol 2005;96:1425–9.[54] Usta N. Use of tobacco seed oil methyl ester in a turbocharged indirectinjection diesel engine. Biomass Bioenergy 2005;28:77–86.[55] Karaosmanoglu F. Vegetable oil fuels: a review. Energy Sour 1999;21:221–31.[56] Giannelos PN, Zannikos F, Stournas S, Lois E, Anastopoulos G. Tobacco seed oilas an alternative diesel fuel: physical and chemical properties. Indus CropsProd 2002;16:1–9.[57] Njoku OU, Ononogbu IC, Owusu JY. An investigation of oil of rubber (Heveabrasiliensis). J Rubber Res Inst Sri Lanka 1996;78:52–9.[58] Ramadhas AS, Jayaraj S, Muraleedharan C. Characterization and effect ofusing rubber seed oil as fuel in the compression ignition engines. RenewEnergy 2005;30:795–803.[59] Ramezanzadeh FM, Rao RM, Windhauser M, Prinyawiwatkul W, MarshallWE. Prevention of oxidative rancidity in rice bran during storage. J Agric FoodChem 1999;47:2997–3000.[60] Lakkakula NR, Lima M, Walker T. Rice bran stabilization and rice bran oilextraction using ohmic heating. Bioresour Technol 2004;92:157–61.[61] Cravotto G, Binello A, Merizzi G, Avogadro M. Improving solvent-free extractionof policosanol from rice bran by high intensity ultrasound treatment. EurJ Lipid Sci Technol 2004;106:147–51.[62] Bhattacharyya DK, Chakrabarty MM, Vaidyanathan RS, Bhatachryya AC. Acritical study of the refining of ricebran oil. J Am Oil Chem Soc 1983;60:467–71.[63] Reyes JF, Sepu’lveda MA. PM-10 emissions and power of a Diesel engine fueledwith crude and refined Biodiesel from salmon oil. Fuel 2006;85:1714–9.[64] Goodrum JW, et al. Rheological characterization of animal fats and theirmixtures with #2 fuel oil. Biomass Bioenergy 2003;24:249–56.[65] Nagel N, Lemke P. Production of methyl fuel from miceoalgea. Appl BiochemBiotechnol 1990;24:355–61.[66] Calvin M. Fuel oils from higher plants. Ann Proc Phytochem Soc Eur1985;26:147–60.[67] Scragg AH, Morrison J, Shales SW. The use of a fuel containing Chlorellavulgaris in a diesel engine. Enzyme Microb Technol 2003;33:884–9.[68] Knothe G. Current perspectives on biodiesel. INFORM 2002;13:900–3.[69] Mukunda H.S. A biomass option for enhancing energy security. IISc. Bangalore:National Institute of Advanced Studied; 1999. p. 36.[70] Sonntag NOV. Structure and composition of fats and oils. In: Swern D, editor.4th ed., Bailey’s industrial oil and fat products, vol. 1, 4th ed. New York: JohnWiley and Sons; 1979. p. 1.[71] Ali Y, Hanna MA. Alternative diesel fuels from vegetable oils. BioresourTechnol 1994;50:153–63.[72] Demirbas A. Diesel fuel from vegetable oil via transesterification and soappyrolysis. Energy Sources 2002;24:835–41.[73] Kapur JP, Bhasin SD, Mathur KC. Oilseeds for food and industrial applications.Chemical Age India 1982;33(9):475–82.[74] Agency for Toxic Substances and Disease Registry (ATSDR). Toxicologicalprofile for fuel oils. Atlanta, GA: U.S. Department of Health and HumanServices, Public Health Service; 1995.[75] Markley KS. Fatty acids—their chemistry, properties, production, and uses.New York: Interscience Publishing Inc.; 1960.[76] Goering CE. Soybean oil as diesel fuel. In: Paper presented at the SoybeanUtilization Conference. Center for Alternative Crop and Products. Bloomington,MN: University of Minnesota; 1988.[77] Haas MJ. Improving the economics of biodiesel production through the use oflow value lipids as feedstocks: vegetable oil soapstock. Fuel Process Technol2005;86:1087–96.[78] Baydar H, Turgut I. Variation s of fatty acid composition according to somemorphological and physiological properties and ecological regions in oilseedplants. Turk J Agric Forest 1999;23(Suppl. 1):81–6.[79] Geller DP, et al. Fuel properties of oil from genetically altered Cupheaviscosissima. Ind Crops Prod 1999;9:85–91.[80] Liljedhal JB, Carleton WM, Turnquist PK, Smith DW. Tractors and their PowerUnits, 3rd ed., New York: John Wiley; 1979.[81] Anonymous. The wealth of India: raw materials, vol. IX. New Delhi: Publication& Information Directorate, Council of Scientific & Industrial Research;1972.[82] Anonymous. The wealth of India: raw materials, vol. I (revised). New Delhi:Publication & Information Directorate, Council of Scientific & IndustrialResearch; 1985.[83] Tyagi PD, Kakkar KK. Non-conventional vegetable oils. New Delhi: Batra BookService; 1991.[84] Saxena VK, Jain SK. Thevetia peruviana kernel oil: a potential bacterial agent.Fitoterapia 1990;61(4):348–9.[85] Ghanim A. Hingota: a tree of immense untapped potential. Indian Farming1991;41(7):9–11.[86] Anonymous. The wealth of India: raw materials, vol. II (revised). New Delhi:Publication & Information Directorate, Council of Scientific & IndustrialResearch; 1988.[87] Hilditch TP, Williams PN. The chemical constituents of natural fats, 4th ed.,London: Chapman and Hall; 1964.[88] Sengupta A, Sengupta SC, Mazumdar UK. Chemical investigation on Celastruspeniculatus seed oil. Fett Wissenschaft Technologie 1987;89(3):119–23.[89] Anonymous. The wealth of India: raw materials, vol. VI. New Delhi: Publication& Information Directorate, Council of Scientific & Industrial Research;1962.[90] Bringi NV. Non-traditional oilseeds and oils of India. New Delhi: Oxford & IBHPublishing Co. Pvt. Ltd; 1987.[91] Anonymous. The wealth of India: raw materials, vol.III (revised). New Delhi:Publication & Information Directorate, Council of Scientific & IndustrialResearch; 1992.[92] Anonymous. The wealth of India: raw materials, vol. V. New Delhi: Publication& Information Directorate, Council of Scientific & Industrial Research;1959.[93] Anonymous. The wealth of India: raw materials, vol. VII. New Delhi: Publication& Information Directorate, Council of Scientific & Industrial Research;1966.[94] Bhattacharjee A, Ghosh SK, Ghosh D, Ghosh S, Maiti MK, Sen SK. Identificationof heat stable palmitoyl/oleoyl specific acyl-acyl carrier protein thioesterasein developing seeds of Madhuca butyracea. Plant Sci 2002;163(4):791–800.[95] Singh A, Singh IS. Chemical evaluation of mahua (Madhuca indica [M. longifolia])seeds. Food Chem 1991;40(2):221–8.[96] Anonymous. The wealth of India: raw materials, vol. X. New Delhi: Publication& Information Directorate, Council of Scientific & Industrial Research;1976.[97] Mohibbe Azam M, Waris Amtul, Nahar NM. Prospects and potential of fattyacid methyl esters of some non-traditional seed oils for use as biodiesel inIndia. Biomass Bioenergy 2005;29:293–302.[98] Sonntag NOV. Structure and composition of fats and oils. In: Swern D, editor.4th ed., Bailey’s industrial oil and fat products, vol. 1, 4th ed. New York: JohnWiley and Sons; 1979. p. 1-.[99] Schwab AW, Dykstra GJ, Selke E, Sorenson SC, Pryde EH. Diesel fuel fromthermal decomposition of soybean oil. JAOCS 1988;65:1781–6.[100] Ziejewski M, Goettler H, Pratt GL. International congress and exposition,Detroit, MI; 24–28 February 1986 [Paper No. 860301].[101] Pryde EH. Vegetable oils as fuel alternatives—symposium overview. JAOCS1984;61:1609–10.[102] Ziejewski M, Kaufman KR, Schwab AW, Pryde EH. Diesel engine evaluation ofa nonionic sunflower oil-aqueous ethanol microemulsion. JAOCS 1984;61:1620–6.[103] Ziejewski MZ, Kaufman KR, Pratt GL, Vegetable oil as diesel fuel, USDA, Argic.Rev. Man., ARM-NC-28; 1983, p. 106–11.[104] Goering C.E. Final report for project on effect of non petroleum fuels ondurability of direct-Injection diesel engines under contract 59-2171-1-6-057-0. Peoria, IL: USDA, ARS; 1984.[105] Sprules FJ, Price D. Production of fatty esters. US Patent 2; 1950, p. 366–494.[106] Eckey EW. Esterification and interesterification. JAOCS 1956;33:575–9.[107] Sridharan R, Mathai IM. Transesterification reactions. J Sci Ind Res1974;33:178–87.[108] Kim HJ, Kang BS, Kim MJ, Park YM, Kim DK, Lee JS, et al. Transesterification ofvegetable oil to biodiesel using heterogeneous base catalyst. Catal Today2004;93–95:315–20.

S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216 215[109] Wright HJ, Segur JB, Clark HV, Coburn SK, Langdon EE, DuPuis RN. A report onester interchange. Oil Soap 1944;21:145–8.[110] Mittelbach M, Tritthart P. Diesel fuels derived from vegetable oils. III. Emissiontests using methyl esters of used frying oil. JAOCS 1988;65(7):1185–7.[111] Mittelbach M, Pokits B, Silberholz A. Production and fuel properties of fattyacid methyl esters from used frying oil. In: Liquid fuels from renewableresources—Proceedings of an alternative energy conference. St Joseph, MI:ASAE; 1992. p. 74–8.[112] Peterson CL, Reece DL, Hammond BJ, Thompson JC, Beck S, Commercializationof Idaho biodiesel (HYSEE) from ethanol and waste vegetable oil. ASAEPaper No. 956738; 1995. St. Joseph, MI: ASAE.[113] T. Wimmer, Process for the production of fatty acid esters of lower alcohols,US Patent No. 5,399,731; 1995.[114] Freedman B, Pryde EH. Fatty esters from vegetable oils for use as a diesel fuel.Proceedings of International Conference on Plant & Vegetable oils as Fuels.ASAE, St. Joseph, MI; 1982. p. 17–122.[115] Aksoy HA, Kahraman I, Karaosmanoglu F, Civelekoglu H. Evaluation ofTurkish sulphur olive oil as an alternative diesel fuel. JAOCS 1988;65(6):936–8.[116] Liu K. Preparation of fatty acid methyl esters for gas chromatographicanalysis of lipids in biological material. JAOCS 1994;71(11):1179–87.[117] Keim GI. Process for treatment of fatty glycerides. US Patent 2; 1945, p. 383–601.[118] Formo MW. Ester reactions of fatty materials. J Am Oil Chem Soc 1954;31(11):548–59.[119] Shimada Y, Sugihara A, Minamigawa Y, Higasbiyama K, Akimoto K, FujikawaS, Komemushi S, Tominaga Y. Enzymatic enrichment of arachidonic acid fromMortierella single-cell oil. J Am Oil Chem Soc 1998;75:1213–7.[120] Miao X, Wu Q. Biodiesel production from heterotrophic microalgal oil.Bioresour Technol 2006;97:841–6.[121] Zullaikah S, Lai C-C, Vali SR, Ju Y-H. A two-step acid-catalyzed process for theproduction of biodiesel from rice bran oil. Bioresour Technol 2005;96:1889–96.[122] Feuge RO, Gros AT. Modification of vegetable oils. VII: Alkali catalyzedinteresterification of pea-nut oil with ethanol. J Am Oil Chem Soc1949;26(3):97.[123] Hengwen H, Weiliang C, Jingchang Z. Preparation of biodiesel from soybeanoil using supercritical methanol and CO 2 as co-solvent. Process Biochem2005;40:3148–51.[124] Furuta S, Matsuhashi H, Arata K. Biodiesel fuel production with solid superacidcatalysis in fixed bed reactor under atmospheric pressure. Catal Commun2004;5:721–3.[125] Madras G, Kolluru C, Kumar R. Synthesis of biodiesel in supercritical fluids.Fuel 2004;83:2029–33.[126] Dmytryshyn SL, Dalai AK, Chaudhari ST, Mishra HK, Reaney MJ. Synthesis andcharacterization of vegetable oil derived esters: evaluation for their dieseladditive properties. Bioresour Technol 2004;92:55–64.[127] Tomasevic AV, Siler-Marinkovic SS. Methanolysis of used frying oil. FuelProcess Technol 2003;81:1–6.[128] Linko Y-Y, Liimsii M, Wu X, Uosukainen W, Sappiilii J, Linko P. Biodegradableproducts by lipase biocatalysis. J Biotechnol 1998;66:41–50.[129] De BK, Bhattacharyya DK, Band C. Enzymatic synthesis of fatty alcohol estersby alcoholysis. J Am Oil Chem Soc 1999;76(45):l–453.[130] Selmi B, Thomas D. Immobilized lipase-catalyzed ethanolysis of sunflower oilin solvent-free medium. J Am Oil Chem Soc 1998;75:691–5.[131] Breivik H, Haraldsson GG, Kristinsson B. Preparation of highly purifiedconcentrates of eicosapentaenoic acid and docosahexaenoic acid. J Am OilChem Soc 1997;74:1425–9.[132] Wu WH, Foglia TA, Marmer WN, Phillips JG. Optimizing production of ethylesters of grease using 95% ethanol by response surface methodology. J Am OilChem Soc 1999;76:517–21.[133] Nelson LA, Foglia A, Marmer WN. Lipase-catalyzed production of biodiesel. JAm Oil Chem Soc 1996;73:1191–5.[134] Mittelbach M. Lipase catalyzed alcoholysis of sunflower oil. J Am Oil ChemSoc 1990;67:1170–688.[135] Abigor R, Uadia P, Foglia T, Haas M, Jones K, Okpefa E, Obibuzor J, Bafor M.Lipase-catalysed production of biodiesel fuel from some Nigerian lauric oils.Biochem Soc Trans 2000;28:979–81.[136] Shieh CJ, Liao HF, Lee CC. Optimization of lipase-catalyzed biodiesel byresponse surface methodology. Bioresour Technol 2003;88:103–6.[137] Watanabe Y, Shimada Y, Sugihara A, Tominaga Y. Conversion of degummedsoybean oil to biodiesel fuel with immobilized Candida antarctica lipase. J MolCatal B Enzym 2002;17:151–5.[138] Lara Pizarro Ana V, Park Enoch Y. Lipase-catalyzed production of biodieselfuel from vegetable oils contained in waste activated bleaching earth. ProcessBiochem 2003;38:1077–82.[139] Ali Y, Hanna MA, Cuppett SL. Fuel properties of tallow and soybean oil esters. JAm Oil Chem Soc 1995;72(12):1557–64.[140] Harrington KJ. Chemical and physical properties of vegetable oil esters andtheir effect on diesel fuel performance. Biomass 1986;9:1–17.[141] SEA News Circular. The solvent Extractors Association of India, vol. X; 1996.[142] Rao PS, Gopalakrishnan KV. Vegetable oils and their methylesters as fuels fordiesel engines. Indian J Technol 1991;29(6):292–7.[143] Dunn RO, Bagby MO. Low-temperature properties of triglyceride-baseddiesel fuels: transesterified methyl esters and petroleum middle distillate/ester blends. J Am Oil Chem Soc 1995;72(8):895–904.[144] Varese R, Varese M. Methyl ester biodiesel: opportunity or necessity?INFORM 1996;7:816824.[145] Chiu C. Biodiesel synthesis and impact of cold flow additives. Doctoraldissertation. University of Missouri-Columbia; 2004, 87 pp.[146] Hu Jianbo, Du Zexue, Li Change Xia, Min Enze. Studying on lubricationproperties of biodiesel as fuel lubricity enhancers. Fuel 2005;XX:1–6.[147] Bailer H, de Hueber K.In: Handbook of analytical methods for fatty acidmethyl esters used as diesel fuel substitutes. Vienna, Austria: Institute forChemistry and Technology of Petroleum Products, University of Technology;1994. p. 19.[148] National Biodiesel Board. Specification for Biodiesel (B100); December 2001.[149] Tyson K Shaine. Biodiesel Handling and Use Guidelines. Report No. NREL/TP-580-30004. Golden, CO: National Renewable Energy Laboratory; September2001.[150] Christopherson SW, Glass RL. Preparation of milk fat methyl esters in anessentially nonalcoholic solution. J Dairy Sci 1968;52:1289–90.[151] Freedman B, Pryde EH, Kwolek WF. Thin layer chromatography/flame ionizationanalysis of transesterified vegetable oils. J Am Oil Chem Soc1984;61:1215–20.[152] Official Methods of Analysis. 14th ed., 28.028, Arlington, VA, USA: AOAC;1984.[153] Seddon RH. Vegetable oil in commercial vehicle. Gas Oil Power 1942;(August):136–46.[154] Bruwer JJ, Boshoff BD, Hugo JC, Du Plessis LM, Fuls J, Hawkins C, et al. Theutilization of sunflower seed oil as a renewable fuel diesel engines. AgricultureEnergy 1981;2. pp. 4–81 [Biomass Energy/Crop Production, St.Joseph, MI, ASAE, ASAE Publication].[155] Yarbrough C.M., Lapori W.A., Engler C.R. Compression ignition performanceusing sunflower seed oil. ASAE paper number 81–576. St.Joseph,MI: ASAE;1981.[156] Tahir AR, Lapp HM, Buchanan LC. Sunflower oil as a fuel for compressionignition engines. In: Vegetable oils fuels: Proceeding of the internationalconference on plant and vegetable oils fuels. St. Joseph, MI: ASAE; 1982.[157] Bettis BL, Peterson CL, Auld DL, Driscoll DJ, Peterson ED. Fuel characteristics ofvegetable oil from oil seed crops in the Pacific Northwest. Agron J 1982;74:335–9.[158] Engler CR, Johnson LA, Lepori WA, Yarbought CM. Effect of processing andchemical characteristics of plant oil on performance of an indirect injectiondiesel engines. J Am Oil Chem Soc 1983;60(8):1592–6.[159] Goering CE, Schwab AW, Daugherty MJ, Pryde EH, Heakin AJ. Fuel propertiesof eleven vegetable oils. ASAE paper number 81-3579. St.Joseph MI: ASAE;1981.[160] Bacon DM, Brear ID Moncrieff, Walker KL. The use of vegetable oils in straightand modified formas diesel engine fuel. Beyond the energy crisis—opportunityand challenges, volume III. In: Fazzolre RA, Smith CR, editors, Thirdinternational conference on energy use management, Berlin (West). Oxford:Pergamon Press; 1981. p. 1525–33.[161] Schoedder C. Rapeseed oil as an alternative fuel for arriculture. Beyond theenergy crises—opportunity and challenges, volume III. . In: Fazzolre RA,Smith CR, editors, Third international conference on energy use management,Berlin (West). Oxford: Pergamon Press; 1981. p. 1815–22.[162] Auld DL, Bettis BL, Peterson CL. Production and fuel characteristics ofvegetative oil seeds crop in the Pacific north west vegetable oil fuel. In:Proceeding of International Conference on Plant and Vegetable Oils Fuels. St.Joseph, MI: ASAE; 1982.[163] Reid JF, Hensen AC, Goering CE. Quntifying diesel injection coking withcomputer vision. Trans ASAE 1989;32(5):1503–6.[164] Engelman HW, Guenther DA, Silvis TW. Vegetable oil as a diesel fuel. Dieseland gas engine power division of ASME paper number 78-DGP-19. New York,NY: ASME; 1978.[165] Quick GR. Development in use of vegetable oils as a fuel for diesel engine.ASAE paper number 801525. St. Joseph, MI: ASAE; 1980.[166] Sims REH, Raine RR, Mc Leod RJ. Rapeseed oil as a fuel for diesel engine SAE-Australia. Paper presented at the national conference on fuels from crop ofthe society of Automotive Engineers, Australia; 1981.[167] Bartholomew D. Vegetable oil fuel. J Am Oil Chem Soc 1981;58(4):286–8.[168] Barsic NJ, Humke AL. Vegetable oils: diesel fuel supplements. AutomotiveEng 1981;89(4):37–41.[169] Fort EF, Blumberg PN, Staph HE, Staudt JJ. Evalution of cotton seed oil as dieselfuel. SAE Techanical paper series 820317. Warrendale, PA: SAE; 1982.[170] Baranescu RA, Lusco JJ. Performance, durability and low temperature operationof sunflower oil as a diesel fuel extender. Vegetable oil fuels: Proceeding of theinternational conference on plant and vegetable oils. St. Joseph MI: ASAE; 1982.[171] Worgetter M. Result of long term enfine tests based on rape seed oil fuel.Beyond the energy crisis—opportunity and challenge volume III. In: FazzolreRA, Smith CR, editors, Third international conference on energy use management,Berlin (West). Oxford: Pergamon Press; 1981. p. 1955–66.[172] Wanger GL, Peterson CL. Performance of winter rape (Brassica napus) basedfuel mixtures in diesel engine. Vegetable oil fuels: Proceeding of the internationalconference on plant and vegetable oils fuels. St. Joseph, MI: ASAE;1982.[173] VanDerWalt AN, Hugo FJC, Attempts to prevent injector cooking with sunfloweroil by engine modifications and fuel additives. Vegetable oil fuels—Proceedings of the international conference on plant and vegetable oils asfuels, vol. 4-82. Fargo, ND: ASAE Publication; 1982, p. 230.[174] German TJ, Kaufman KR, Pratt GL, Derry J. Field evaluation of sunflower oil/diesel fuel blend in diesel engines. ASAE paper number 85-3078. St. Joseph,MI: ASAE; 1985.

216S.P. Singh, D. Singh / Renewable and Sustainable Energy Reviews 14 (2010) 200–216[175] Sapaun SM, Masjki HH, Azlam A. The use of palm oil as diesel fuel substitute. JPower Energy (Part A) 1996;210:47–53.[176] Hofman, Kaufman VD, Helgeson D, Dinusson WE. Sunflower for powerNDSU Cooperative Extension service circular AE-735. Fargo, ND;1981.[177] Peterson CL, Auld DL, Thomas VM, Withers RV, Smith SM, Betties BL,Vegetable oil as an agricultural fuel for the pacific north west, Universityof Idaho Experiment station bulletin No. 598. University of Idaho: Moscow,ID; 1981.[178] Pestes MN, Stanislao J. Piston ring deposits when using vegetable oil as a fuel.J Test Eval 1984;12(2):61–8.[179] Ryan III TW, Dodge LG, Callahan TJ. The effect of vegetable oil properties ofinjection and combustion in two different diesel engine. J Am Oil Chem Soc1984;61(10):1610–9.[180] Nag AS, Bhattacharya KBDe. New utilization of vegetable oils. J Am Oil ChemSoc 1995;72(12):1591–3.[181] McCutchen R. Vegetable oil as a diesel fuel—soybean oil. Beyond the energycrisis opportunities and challenge volume III. In: Fazzolre RA, Smith CR,editors, Third international conference on energy use management, Berlin(West). Oxford: Pergamon Press; 1981. p. 1679–86.[182] McDonnell KP, Ward SM, McNully PB, Howard-Hildige R. Result of Engineand vehicle testing of semi refined rapeseed oil. Trans ASAE 2000;43(6):1309–16.[183] Kalligeros S, Zannikos F, Stournas S, Lois E, Anastopoulos G, Teas CH,et al. An investigation of using biodiesel/marine diesel blends on the performanceof a stationary diesel engine. Biomass Bioenergy 2003;24(2):141–9.[184] Monyem A, Van Gerpen JH. The effect of biodiesel oxidation on engineperformance and emissions. Biomass Bioenergy 2001;20:317–25.[185] Machacon HTC, Shiga S, KarasawaT, Nakamura H. Performance and emissioncharacteristics of a diesel engine fueled with coconut oil-diesel fuel blend.Biomass Bioenergy 2001;20:63–9.[186] Wang WG, Lyons DW, Clark NN, Gautam M, Norton PM. Emissions from nineheavy trucks fuelled by diesel and biodiesel blend without engine modification.Environ Sci Technol 2000;34:933–9.[187] Clark SJ, Wagner L, Schrock MD, Piennaar PG. Methyl and ethyl ester asrenewable fuels for diesel engine. J Am Oil Chem Soc 1984;61:1632–8.[188] Mittelbach M, Tritthart P, Junck H. Diesel fuel derived from vegetable oil II:emission tests using rape oil methyl ester. Energy Agric 1985;4:207–15.[189] Carraretto C, Macor A, Mirandola A, Stoppato A, Tonon S. Biodiesel asalternative fuel: experimental analysis and energetic evaluations. Energy2004;29:2195–211.[190] Shi X, Yu Y, He H, Shuai S, Wang J, Li R. Emission characteristics using methylsoyate –ethanol-diesel fuel blends on a diesel engine. Fuel 2005;XX:1–7.[191] Spataru A, Romig C. Emissions and engine performance from blends of soyand canola methyl esters with ARB #2 diesel in a DDC 6V92TA MUI engine.Emission Processes and Control Technologies in Diesel Engines. Society ofAutomotive Engineers Publication No. 1119, Technical Paper No. 952388, pp.179–188. Fuels and Lubricants Meeting and Exposition, Toronto, Ontario,October 16–l 9; 1995.[192] Schumacher L, Borgelt SC. SAE Tech. Pap. Ser.; 1996 [No. 962233].[193] Graboski MS, Ross JD, McCormick RL. SAE Tech. Pap. Ser.; 1996 [No. 961166].[194] Peterson CL, Reece DL, Thompson JC, Beck SM, Chase C. Biomass Bioenergy1996;10(5/6):331–6.[195] Chang DYZ, Van Gerpen JH, Lee I, Johnson LA, Hammond EG, Marley SJ. Fuelproperties and emissions of soybean oil esters as diesel fuel. J Am Oil ChemSoc 1996;73(11):1549.[196] Anon.. Filtered used frying fat powers diesel feet. JAOCS 1982;59:780A–1A.[198] Murayama T. Evaluating vegetable oils as a diesel fuel. INFORM 1994;5(10):1138–45.[199] Kreutzer UR. Manufacture of fatty alcohol based on natural fats and oils. J AmOil Chem Soc 1984;61(2):343–8.[200] Wardle DA. Global sale of green air travel supported using Biodiesel. RenewSust Energy Rev 2003;7:1–64.

Applied Energy 87 (2010) 1815–1835Contents lists available at ScienceDirectApplied Energyjournal homepage: www.elsevier.com/locate/apenergyProgress in biodiesel processingMustafa Balat * , Havva BalatSila Science and Energy Unlimited Company, Trabzon, TurkeyarticleinfoabstractArticle history:Received 26 November 2009Received in revised form 22 January 2010Accepted 22 January 2010Available online 16 February 2010Keywords:TriglycerideBiodieselFuel characteristicsProductionTransesterificationReaction parametersBiodiesel is a notable alternative to the widely used petroleum-derived diesel fuel since it can be generatedby domestic natural sources such as soybeans, rapeseeds, coconuts, and even recycled cooking oil,and thus reduces dependence on diminishing petroleum fuel from foreign sources. The injection andatomization characteristics of the vegetable oils are significantly different than those of petroleumderiveddiesel fuels, mainly as the result of their high viscosities. Modern diesel engines have fuel-injectionsystem that is sensitive to viscosity change. One way to avoid these problems is to reduce fuel viscosityof vegetable oil in order to improve its performance. The conversion of vegetable oils into biodieselis an effective way to overcome all the problems associated with the vegetable oils. Dilution, micro-emulsification,pyrolysis, and transesterification are the four techniques applied to solve the problems encounteredwith the high fuel viscosity. Transesterification is the most common method and leads tomonoalkyl esters of vegetable oils and fats, now called biodiesel when used for fuel purposes. The methylester produced by transesterification of vegetable oil has a high cetane number, low viscosity andimproved heating value compared to those of pure vegetable oil which results in shorter ignition delayand longer combustion duration and hence low particulate emissions.Published by Elsevier Ltd.Contents1. Introduction . . . ..................................................................................................... 18162. Fuel characteristics of biodiesel. . . . . . . .................................................................................. 18162.1. Technical characteristics of biodiesel as a transportation fuel . . . . . . ..................................................... 18162.2. Characteristics of biodiesel as a renewable fuel . . . . . . . . . . . . . . . . . ..................................................... 18172.2.1. The impact of biodiesel fuels on air quality. . . . . . . . .......................................................... 18172.2.2. The net energy balance . . . . . ............................................................................. 18173. Review of biodiesel feedstocks . . . . . . . .................................................................................. 18173.1. Vegetable oils as diesel fuels . . . . . . . . . . . . . ........................................................................ 18173.2. Vegetable oils as biodiesel feedstock . . . . . . . ........................................................................ 18183.2.1. Edible vegetable oils . . . . . . . ............................................................................. 18183.2.2. Non-edible vegetable oils . . . ............................................................................. 18203.3. Animal fats as biodiesel feedstock . . . . . . . . . ........................................................................ 18213.4. Waste cooking oil as biodiesel feedstock. . . . ........................................................................ 18214. Derivatives of triglycerides as diesel fuels . . . . . . . . . . . . . . . . . ............................................................... 18214.1. Dilution with diesel fuel . ........................................................................................ 18224.2. Micro-emulsification of oils . . . . . . . . . . . . . . ........................................................................ 18224.3. Pyrolysis. . . . . . . . . . . . . . ........................................................................................ 18224.4. Transesterification of triglycerides to biodiesel. . . . . . . . . . . . . . . . . . ..................................................... 18234.4.1. Catalytic transesterification methods. . . . . . . . . . . . . .......................................................... 18244.4.2. Non-catalytic transesterification methods . . . . . . . . . .......................................................... 18274.4.3. Effect of different reaction parameters on the biodiesel yield . . . . . . . . . . . . ....................................... 18284.4.4. Downstream processing . . . . ............................................................................. 1830* Corresponding author. Tel.: +90 462 871 3025; fax: +90 462 871 3110.E-mail address: mustafabalat@yahoo.com (M. Balat).0306-2619/$ - see front matter Published by Elsevier Ltd.doi:10.1016/j.apenergy.2010.01.012

1816 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–18355. Conclusion . ........................................................................................................ 1831References . ........................................................................................................ 18311. IntroductionDuring the past 25 years, worldwide petroleum consumptionhas steadily increased, resulting in higher standards of living, increasedtransportation and trucking, and increased use of plasticsand other petrochemicals [1]. In 1985, total worldwide petroleumconsumption was 2807 million tons, but in 2008, the figurereached 3928 million tons [2], with an average annual growth rateof almost 1.5%. However, the petroleum is a finite source for fuelthat is rapidly becoming scarcer and more expensive [3]. At theend of 2008, according to BP’s annual Statistical Review of WorldEnergy [2], the world proven oil reserves were estimated at1.7 10 11 tons with a reserve-to-production (R/P) ratio of42 years. In addition, petroleum-based products are one of themain causes of anthropogenic carbon dioxide (CO 2 ) emissions tothe atmosphere. Today, the transportation sector worldwide is almostentirely dependent on petroleum-derived fuels. One-fifth ofglobal CO 2 emissions are created by the transport sector [4], whichaccounts for some 60% of global oil consumption [5]. Around theworld, there were about 806 million cars and light trucks on theroad in 2007 [6]. These numbers are projected to increase to 1.3billion by 2030 and to over 2 billion vehicles by 2050 [7]. Thisgrowth will affect the stability of ecosystems and global climateas well as global oil reserves [8,9]. There are active research programsto reduce reliance on fossil fuels by the use of alternativeand sustainable fuel sources, and thus to increase the time overwhich fossil fuels will still be available [10]. As an alternative topetroleum-based transportation fuels, bio-fuels can help to reinforceenergy security and reduce the emissions of both greenhousegases (GHGs) and urban air pollutants.The term bio-fuel is referred to as liquid or gaseous fuels forthe transport sector that are predominantly produced from biomass.A variety of fuels can be produced from biomass resourcesincluding liquid fuels, such as bioethanol, methanol, biodiesel,Fischer–Tropsch diesel, and gaseous fuels, such as hydrogen andmethane [11]. Biodiesel, an alternative diesel fuel, is made fromrenewable biological sources such as vegetable oils and animalfats. Biodiesel production is a very modern and technological areafor researchers due to the relevance that it is winning everydaybecause of the increase in the petroleum price and the environmentaladvantages [12]. Biodiesel blends reduce levels of globalwarming gases such as CO 2 . This paper gives detailed knowledgeon the fuel characteristics of biodiesel, describes sustainable feedstockoptions for biodiesel production, and improved conversiontechnologies.2. Fuel characteristics of biodiesel2.1. Technical characteristics of biodiesel as a transportation fuelBiodiesel is a cleaner burning alternative to petroleum-baseddiesel fuel. Just like petroleum-based diesel fuel, biodiesel operatesin the compression ignition (diesel) engines. The successful introductionand commercialization of biodiesel in many countriesaround the world has been accompanied by the development ofstandards to ensure high product quality and user confidence.Some biodiesel standards are ASTM D6751 (ASTM = American Societyfor Testing and Materials) and the European standard EN14214. The biodiesel is characterized by determining its physicaland fuel properties including density, viscosity, iodine value, acidvalue, cloud point, pure point, gross heat of combustion and volatility.In general, biodiesel compares well to petroleum-based diesel(Table 1) [13].The advantages of biodiesel as diesel fuel are its portability,ready availability, renewability, higher combustion efficiency, lowersulfur and aromatic content, higher cetane number and higherbiodegradability [14]. The main disadvantages of biodiesel as dieselfuel are its higher viscosity, lower energy content, higher cloudpoint and pour point, higher nitrogen oxide emission, lower enginespeed and power, injector coking, engine compatibility, high price,and higher engine wear [15]. Biodiesel offers safety benefits overdiesel fuel because it is much less combustible, with a flash pointgreater than 423 K compared to 350 K for petroleum-based dieselfuel [16]. Biodiesel has a higher cetane number (around 50) thandiesel fuel [17], no aromatics, no sulfur, and contains 10–11% oxygenby weight [18]. The cetane number is a commonly used indicatorfor the determination of diesel fuel quality, especially theignition quality. It measures the readiness of the fuel to auto-ignitewhen injected into the engine. Ignition quality is one of the propertiesof biodiesel that is determined by the structure of the fattyacid methyl ester (FAME) component [19]. Viscosity is the mostimportant property of biodiesel since it affects the operation ofthe fuel injection equipment, particularly at low temperatureswhen the increase in viscosity affects the fluidity of the fuel.Biodiesel has a viscosity close to that of diesel fuels. High viscosityleads to poorer atomization of the fuel spray and less accurateoperation of the fuel injectors [20].Elemental composition and relative amounts of compoundspresent in biodiesel and diesel fuel are given in Tables 2 and 3.Due to presence of electronegative element oxygen, biodiesel isslightly more polar than diesel fuel as a result viscosity ofbiodiesel is higher than diesel fuel. Presence of elemental oxygenlowers the heating value of biodiesel when compared the dieselfuel [21].The lower heating value (LHV) is the most common value usedfor engine applications. It is used as an indicator of the energy contentof the fuel. Biodiesel generally has a LHV that is 12% less thanTable 1ASTM standards of maximum allowed quantities in diesel and biodiesel [13].Property Diesel BiodieselStandard ASTM D975 ASTM D6751Composition HC a (C10–C21) FAME b (C12–C22)Kin. viscosity (mm 2 /s) at 313 K 1.9–4.1 1.9–6.0Specific gravity (g/mL) 0.85 0.88Flash point (K) 333–353 373–443Cloud point (K) 258–278 270–285Pour point (K) 238–258 258–289Water (vol.%) 0.05 0.05Carbon (wt.%) 87 77Hydrogen (wt.%) 13 12Oxygen (wt.%) 0 11Sulfur (wt.%) 0.05 0.05Cetane number 40–55 48–60HFRR c (lm) 685 314BOCLE d scuff (g) 3600 >7000abcdHydrocarbons.Fatty acid methyl esters.High frequency reciprocating rig.Ball-on-cylinder lubricity evaluator.

M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835 1817Table 2Elemental analysis of biodiesel and diesel fuels [21].Elements Composition (%)BiodieselDiesel fuelCarbon (C) 79.6 86.4Hydrogen (H) 10.5 13.6Oxygen (O) 8.6 –Nitrogen (N) 1.3 –C/H 7.6 6.5Table 4Average B100 and B20 emissions compared to normal diesel [13].Emission B100 (%) B20 (%)Carbon monoxide 48 12Total unburned hydrocarbons 67 20Particulate matter 47 12Nitrogen oxides 10 2Sulfates 100 20Air toxics 60 to 90 12 to 20Mutagenicity 80 to 90 20Table 3Composition of biodiesel and diesel fuels [21].Type of compounds Biodiesel Diesel fueln-Aliphatic 15.2 67.4Olefinics 84.7 3.4Aromatics – 20.1Naphthenes – 9.1No. 2 diesel fuel on a weight basis (16,000 Btu/lb compared with18,300 Btu/lb). Since the biodiesel has a higher density, the LHVis only 8% less on a volume basis (118,170 Btu/gallon for biodieselcompared with 129,050 Btu/gallon for No. 2 diesel fuel) [22,23].Biodiesel can be used as pure fuel or blended at any level withpetroleum-based diesel for use by diesel engines. The most commonbiodiesel blends are B2 (2% biodiesel and 98% petroleum diesel),B5 (5% biodiesel and 95% petroleum diesel), and B20 (20%biodiesel and 80% petroleum diesel). The technical disadvantagesof biodiesel/petroleum diesel blends include problems with fuelfreezing in cold weather, reduced energy density, and degradationof fuel under storage for prolonged periods [24]. Biodiesel blendsup to B20 can be used in nearly all diesel equipment and are compatiblewith most storage and distribution equipment. These lowlevelblends generally do not require any engine modifications.Higher blends and B100 (pure biodiesel) may be used in some engineswith little or no modification, although the transportationand storage of B100 requires special management [25].2.2. Characteristics of biodiesel as a renewable fuel2.2.1. The impact of biodiesel fuels on air qualityAir pollution is one of the most serious environmental problemsall over the world. The combustion of petroleum-based diesel fuelis a major source of GHG emissions. Apart from these emissions,petroleum-based diesel fuel is also major source of other air contaminantsincluding carbon monoxide (CO), nitrogen oxides(NO x ), sulfur oxides (SO x ), particulate matter (PM), and volatile organiccompounds (VOCs). Biodiesel, a fuel that can be made fromrenewable biological sources, such as vegetable oils, animal fatsand waste cooking oils, may have the potential to reduce the relianceon imported oil and reduce air pollutant emissions from dieselengines [26]. Using biodiesel in a conventional diesel enginesubstantially reduces emissions of HCs, CO, PM, sulfates, polycyclicaromatic hydrocarbons and nitrated polycyclic aromatic hydrocarbons.These reductions increase as the amount of biodiesel blendedinto diesel fuel increases. The best emissions reductions are seenwith B100 (Table 4). NO x emissions increase with the concentrationof biodiesel in the fuel. In October 2002, the EnvironmentalProtection Agency (EPA) [27] assessed the impact of biodiesel fuelon emissions and published a draft report summarizing the results.According to the report soybean-based B20 fuel results in a reductionof 10.1 in total PM, 21.1% in HC, and 11% in CO emissions.These are offset by a 2% increase in NO x emissions.Biodiesel from virgin vegetable oil reduces CO 2 emissions whenused in place of petroleum diesel. This conclusion is based on a lifecycle analysis of biodiesel and petroleum diesel, accounting foremissions for all steps in the production and use of the fuel [28].The use of pure biodiesel in the transport sector lowers the emissionsof CO 2 by 80% [29].2.2.2. The net energy balanceThe energy balance measures the units of energy yielded foreach unit of energy required to produce the fuel. The net energybalance of various ethanol and biodiesel feedstock has been a centerof debate within scientific and policy circles [30]. The energybalance for a bio-fuel production system can be defined as the relationbetween the energy produced (output/kg biodiesel) and theenergy consumed (input/kg biodiesel) for each unit of productand that is an important index for the economic and environmentalfeasibility of a bio-fuel project [31].Biodiesel has the highest energy balance ratio of any liquid fuel.It has been reported that the growth of energy crops for biodieselproduction produces a positive energy balance of 2.5–3.2 as outputof energy from the use of biodiesel exceeds that of the energy inputfor manufacturing the fuel [30]. The Sheehan et al. study left outmany of the fossil energy inputs to achieve an energy balance of3.2 [32]. Most fossil fuels have a negative energy balance meaningit takes more units of energy to recover, transport and process thefuel than it returns in units of fuel energy. Every unit of fossil fuelenergy used to extract and refine crude oil into petroleumdiesel yields only 0.83 units of energy – a negative energy balance[33].3. Review of biodiesel feedstocksIn general, biodiesel feedstock can be categorized into threegroups: vegetable oils (edible or non-edible oils), animal fats, andused waste cooking oil including triglycerides. The study by Johnstonand Holloway [34] evaluated production volumes and pricesacross 226 countries and territories and estimated current worldwideproduction potential at 51 billion liters from 119 countries.Table 5 depicts the list of top ten countries in terms of absolutebiodiesel production. These ten countries collectively account formore than 80% of the total. Among these countries, the averagefeedstock dependence is: 28% for soybean oil, 22% for palm oil,20% for animal fats, 11% for coconut oil, and 5% each for rapeseed,sunflower and olive oils [35].3.1. Vegetable oils as diesel fuelsThe concept of using vegetable oil as a transportation fuel datesback to 1893 when Dr. Rudolf Diesel developed the first dieselengine to run on vegetable oil. Vegetable oil is one of the renewablefuels. Vegetable oils have become more attractive recentlybecause of its environmental benefits and the fact that it is made

1818 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835Table 5Top ten countries in terms of absolute biodiesel potential [34].Country Volume potential (Ml) Production cost (US$ per liter)Malaysia 14,540 0.53Indonesia 7595 0.49Argentina 5255 0.62USA 3212 0.7Brazil 2567 0.62Netherlands 2496 0.75Germany 2024 0.79Philippines 1234 0.53Belgium 1213 0.78Spain 1073 1.71from renewable resources. Vegetable oils have the potential to substitutea fraction of petroleum distillates and petroleum-based petrochemicalsin the near future [20,24]. The basic constituent ofvegetable oils is triglyceride. Vegetable oils comprise 90 to 98% triglyceridesand small amounts of mono- and diglycerides [35].These usually contain free fatty acids (FFAs), water, sterols, phospholipids,odorants and other impurities [15]. Different types ofvegetable oils have different types of fatty acids. The fatty acidsfound in the vegetable oils are summarized in Table 6 [15,18,36–40].The advantages of vegetable oils as diesel fuel are their portability,ready availability, renewability, higher heat content (about 88%of D2 fuel), lower sulfur content, lower aromatic content, and biodegradability.The main disadvantages of vegetable oils as dieselfuel are higher viscosity, lower volatility, and the reactivity ofunsaturated hydrocarbon chains [41].The injection and atomization characteristics of the vegetableoils are significantly different than those of petroleum-derived dieselfuels, mainly as the result of their high viscosities [42]. The vegetableoils, as alternative engine fuels, are all extremely viscouswith viscosities ranging from 9 to 17 times greater than that ofpetroleum-derived diesel fuel [43]. Modern diesel engines havefuel-injection system that is sensitive to viscosity change. Oneway to avoid these problems is to reduce fuel viscosity of vegetableoil in order to improve its performance. The vegetable oils may beblended to reduce the viscosity with diesel in presence of someadditives to improve its properties. Heating and blending of vegetableoils may reduce the viscosity and improve volatility of vegetableoils but its molecular structure remains unchanged hencepolyunsaturated character remains. Blending of vegetable oils withdiesel, however, reduces the viscosity drastically and the fuelhandling system of the engine can handle vegetable oil–dieselblends without any problems [44,45]. The conversion of vegetableoils into FAME is an effective way to overcome all the problemsassociated with the vegetable oils.The most common way of producing biodiesel is the transesterificationof vegetable oils [46]. The methyl ester produced bytransesterification of vegetable oil has a high cetane number, lowviscosity and improved heating value compared to those of purevegetable oil which results in shorter ignition delay and longercombustion duration and hence low particulate emissions. Its useresults in the minimization of carbon deposits on injector nozzles[47].3.2. Vegetable oils as biodiesel feedstock3.2.1. Edible vegetable oilsBiodiesel has been mainly produced from edible vegetable oilsall over the world. More than 95% of global biodiesel productionis made from edible vegetable oils [48]. Global use of edible oils increasedfaster, between marketing years 2004 and 2007, than itsproduction. The estimated increase in edible oil use for biodieselproduction was 6.6 million tons from 2004 to 2007, which wouldattribute 34% of the increase in global consumption to biodiesel[49]. Between 2005 and 2017, biodiesel use of edible oils is projectedto account for more than a third of the expected growth inedible oil use [50]. The largest biodiesel producers were the EuropeanUnion (EU), the United States, Brazil, Indonesia, with a combineduse of edible oil for biodiesel production of about 8.6million tons in 2007 compared to global edible oil production of132 million tons [49,51]. The combined use of edible oils for biodieselin the EU was 6.1 million tons in 2007 compared with about1.0 million tons in 2001 [49].A rough estimate suggests that about 7.8 Mha were used to providebiodiesel feedstocks in the four major producing countries in2007 (Table 7) [52]. Biodiesel production uses around 4.4 Mha ofTable 7Biodiesel production and land use by major producing countries, 2006/2007.CountryBiodieselfeedstocksImpliedfeedstockarea (Mha) aCountrytotal(Mha) aArable landArea(Mha) aArgentina Soybean (100%) 0.73 0.73 28.3 2.6Brazil Soybean (66%) 0.45 0.45 59.1 0.8EU-27 Rapeseed (64%) 2.75 4.33 113.8 3.8Soybean (16%) 1.58USA Soybean (74%) 2.31 2.31 174.5 1.3Total 7.82 7.82 357.7 2.1aMillion hectares.Biodieselshare (%)Table 6Percentage of fatty acid type for different oils.Oil Palmitic (C16:0) Palmitoleic (C16:1) Stearic (C18:0) Oleic (C18:1) Linoleic (C18:2) Linolenic (C18:3) Other acids ReferencesCorn 6.0 – 2.0 44.0 48.0 – – [36]Cottonseed 28.3 – 0.9 13.3 57.5 – – [37]Olive 14.6 – – 75.4 10 – – [36]Palm 42.6 0.3 4.4 40.5 10.1 0.2 1.1 [15,38]Peanut 11.4 – 2.4 48.3 32.0 0.9 9.1 [37]Rapeseed 3.5 0.1 0.9 54.1 22.3 – 0.2 [36]Safflower 7.3 0.1 1.9 13.5 77 – – [36]Soybean 11.9 0.3 4.1 23.2 54.2 6.3 – [15]Sunflower 6.4 0.1 2.9 17.7 72.9 – – [15,38]Tallow 29.0 – 24.5 44.5 – – – [36]J. curcas 14.2 0.7 7.0 44.7 32.8 0.2 – [39]P. pinnata 10.2 – 7.0 51.8 17.7 3.6 – [37]M. indica 24.5 – 22.7 37.0 14.3 – – [40]WCO a 20.4 4.6 4.8 52.9 13.5 0.8 [18]aWCO-waste cooking oil.

M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835 1819arable land in the EU [52]. Replacing 10% of EU diesel with biodieselwould account for around 19% of world edible oil production in2020 which means more land will be planted with crops and moreland somewhere in the world will be converted into farmland,thereby releasing GHG emissions [53]. Industrialized countrieswith bio-fuels targets, such as the United States and the EU countries,are unlikely to have the agricultural land base needed to meettheir growing demand for current production of bio-fuels. Currently,biodiesel is mainly prepared from conventionally grownedible oils such as rapeseed, soybean, sunflower and palm thusleading to alleviate food versus fuel issue [54]. Serious problemsface the world food supply today. The rapidly growing world populationand rising consumption of bio-fuels are increasing demandfor both food and bio-fuels. This exaggerates both food and fuelshortages. The human population faces serious food shortagesand malnutrition [55]. Nearly 60% of humans in the world are currentlymalnourished, so the need for grains and other basic foods iscritical. Growing crops for fuel squanders land, water, and energyresources vital for the production of food for people [56].Rapeseed and sunflower oils are used in the EU, palm oil predominatesin biodiesel production in tropical countries, and soybeanoil is the major feedstock in the United States [57].According to Pahl [58], rapeseed oil has 59% of total global biodieselraw material sources, followed by soybean (25%), palm oil(10%), sunflower oil (5%), and other (1%).Rapeseed used for biodiesel is the EU’s dominant bio-fuel cropwith a share of about 80% of the feedstock [59]. Total EU rapeseedproduction was 15.3 million tons in 2006/2007 [60]. About 65% ofrapeseed oil supply is used for biodiesel production [49]. The EU isnow a net importer of rapeseed oil (0.57 million tons in 2006/2007)[61]. China, a major producer of rapeseed, is also starting to producebiodiesel from this source. Chinese production of rapeseedbiodiesel, which currently totals about 114 Ml per year, is limitedbecause of less advanced technology, but scientists are workingat improving technologies in China to increase their biodiesel productionfrom rapeseed oil [27]. Rapeseed (known in North Americaas canola) produces 1 ton of oil per hectare [62], giving it the highestyield of any conventional oilseed field crop. Comparison ofsome sources of biodiesel is given in Table 8.Soybean is the most common feedstock in the United States andit represents 25% of total global biodiesel raw material sources,according to Pahl [58]. Soybean oil accounts for approximately90% of the biodiesel produced in the United States [63]. The UnitedTable 8Comparison of some sources of biodiesel.Energy cropRapeseed 1Soybean 0.52Palm 5Jatropha a 0.5Microalgae 12aUnder arid conditions.Oil yield (tons per hectare)States is the world leader in soybean production with about 87million tons produced in 2006/2007 crop year [64]. Based on currentUS diesel consumption of 227 billion liters per year, thiswould require more than 500 Mha of land in soybeans or morethan half the total US planted just for soybeans [56]. Soybeans onlyproduce about 0.52 tons of oil per hectare [62]. Typically, oil derivedfrom commercial soybean varieties is composed of approximately12% palmitic acid (C16:0), 4% stearic acid (C18:0), 23%oleic acid (C18:1), 54% linoleic acid (C18:2), and 6% linolenic acid(C18:3) (Table 6). High levels of oleic acid make the oil more resistantto oxidation and hence more suitable for processes requiringhigh oxidative stability at high temperatures as in biodiesel andbiolubricants applications. On the other hand, high linoleic acidcontent (less stable to oxidation) makes the oil to have excellentdrying properties, which is a desirable property for applicationsin paints, inks and varnishes [65]. Of importance are the combustionand flow properties of biodiesel fuel which are determinedprimarily by FFA content of the vegetable oil. Properties of selectedvegetable oils and biodiesel fuels are depicted in Table 9 [14,66–70] and Table 10 [14,66,72–75].Palm oil is produced from the fruit of the oil palm, or ElaeisGuineesis tree, which originated in West Guinea. The tree wasintroduced into other parts of Africa, South East Asia and LatinAmerica during the 15th century [76]. After almost 100 years palmoil is now increasingly being produced on large-scale plantations intropical lowland regions especially in Indonesia and Malaysia aswell as on the Philippines, in Myanmar and Thailand. Today theworld market consists of 38.13 million tons annually, and Indonesiaand Malaysia as largest and second-largest producer of palm oilin the world account together for more than 85% of exports [77].China is the largest consumers of palm oil in the world with totaldomestic consumption in 2006/2007 forecast to reach 5.1 millionTable 9Fuel-related properties of selected vegetable oils.Entry Oil Iodine value CN a CP b (K) PP c (K) FP d (K) Viscosity f (mm 2 /s) References1 Cottonseed 90–119 41.8 274.7 258.0 507 33.5 [66]2 Rapeseed 94–120 37.6 269.1 241.3 519 37.0 [66]3 Safflower 126–152 41.3 291.3 266.3 533 31.3 [66]4 Soybean 117–143 37.9 269.1 260.8 527 32.6 [66]5 Sunflower 110–143 37.1 280.2 258.0 547 37.1 [66]6 Karanja – 29.9 – 278.0 488 43.67 [67]7 Jatropha 101 23 – 267.0 459 35.4 [68]8 Palm 35–61 g 42 g – – 577 h 42.66 h [66,69]9 Palm kernel 14–33 i – – 295.0 j 515 j 115.55 j [70]10 WCO e 141.5 49 – 284.0 485.0 36.4 [14]abcdefCN – cetane number.CP – cloud point.PP – pour point.FP – flash point.WCO – waste cooking oil.Measured at 311 K (entries 1–5) and at 313 K (entries 6–10).g Iodine value and cetane number combined from Ref. [66].h Flash point and viscosity combined from Ref. [69].iIodine value combined from Ref. [70].jPour point, flash point and viscosity combined from Ref. [71].

1820 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835Table 10Properties of selected biodiesel fuels.Entry Biodiesel a CN b CP b (K) PP b (K) FP b (K) Viscosity e (mm 2 /s) References1 Cottonseed 51.2 – 269 383 6.8 [66]2 WCO b 54 – 262 469 5.3 [14]3 Rapeseed 54.4 271 264 357 6.7 [66]4 Safflower 49.8 – 267 453 – [66]5 Soybean 46.2 275 272 444 4.08 [66]6 Sunflower 46.6 273 269 – 4.22 [66]7 Karanja 55 280 277 456 4.66 [72]8 Jatropha 48 – 272 366 5.38 [73]9 PME c 65 289 289 447 4.5 [74]10 PEE d 56.2 281 279 292 4.5 [66]11 Palm kernel – 279 275 440 4.84 [75]a Methyl esters (entries 1–9) and ethyl esters (entries 10 and 11).b Explanations as in Table 9.c PME – palm oil methyl ester.d PEE – palm oil ethyl ester.e Measured at 294 K (entry 1), at 310.8 K (entry 2) and at 313 K (entries 3–11).tons [78]. Other large scale consumers of palm oil include Indonesia,India, Malaysia and the EU. These five major consumers accountfor 55% of global palm oil consumption. Even allowing forincreasing biodiesel usage it is expected that by 2010 no more than2% of global palm oil is likely to be used in biodiesel in the EU [79].The palm oil system generates more biodiesel per hectare than therapeseed system, and has less parasitic demand [80]. Palm, whichcan produce an astounding 4000 l of oil per hectare, has the highestoil yield of plants grown for biodiesel feedstock. There are greatdifferences between palm oil and palm kernel oil with respect totheir physical and chemical characteristics. The major differencebetween palm oil and palm kernel oil (PKO) is in their fatty acidcontent. Palm oil contains mainly palmitic and oleic acids, thetwo common fatty acids, and about 50% saturated fatty acids[14], while PKO contains mainly lauric acid and more than 80% saturatedfatty acids [81].3.2.2. Non-edible vegetable oilsCurrently, more than 95% of the world biodiesel is producedfrom edible oils which are easily available on large scale fromthe agricultural industry [48]. However, continuous and large-scaleproduction of biodiesel from edible oils has recently been of greatconcern because they compete with food materials – the food versusfuel dispute. There are concerns that biodiesel feedstock maycompete with food supply in the long-term [82,83]. Hence, use ofnon-edible vegetable oils when compared with edible oils is verysignificant in developing countries because of the tremendous demandfor edible oils as food, and they are far too expensive to beused as fuel at present [84,85]. The production of biodiesel fromdifferent non-edible oilseed crops has been extensively investigatedover the last few years. Some of these non-edible oilseedcrops include jatropha tree ( Jatropha curcas) [44,86,87], karanja( Pongamia pinnata) [88,89], tobacco seed ( Nicotiana tabacum L.)[90], rice bran [91,92], mahua (Madhuca indica) [93], neem (Azadirachtaindica) [44], rubber seed tree ( Hevea brasiliensis) [94],and microalgae [95–97].J. curcas and P. pinnata are two such trees, which can thrive onany type of soil, need minimum input and management, and havelow moisture demand [98]. Jatropha has been planted in severalarid regions, in these regions it only yields about 0.5 ton per hectare.The seeds contain about 30% oil. The oil fraction of J. curcasconsists of both saturated (14.2% palmitic acid and 7.0% stearicacid) and unsaturated fatty acids (44.7% oleic acid and 32.8%linoleic acid) (Table 6). Jatropha appears to have several advantagesas a renewable diesel feedstock. Because it is both non-edibleand can be grown on marginal lands, it is potentially a sustainablebio-fuel that will not compete with food crops. This is not the casewith bio-fuels derived from soybeans, rapeseed, or palm [62]. Theserious problem with jatropha, in addition to the low yield, is thatit is highly toxic to people and livestock. The residue after removingthe oil cannot be fed to livestock, like what is done with theresidue soybean and maize. The comparatively high price ofJatropha is also still a drawback [99].Karanja (P. pinnata) is native to a number of countries includingIndia, Malaysia, Indonesia, Taiwan, Bangladesh, Sri Lanka andMyanmar. P. pinnata can grow in humid as well as subtropicalenvironments with annual rainfall ranging between 500 and2500 mm [100]. It is a fast-growing leguminous tree with the potentialfor high oil seed production and the added benefit of theability to grow on marginal land. These properties support the suitabilityof this plant for large-scale vegetable oil production requiredby a sustainable biodiesel industry [101]. The seeds of P.pinnata contain around 30–40% of oil [101–103], which has beenidentified as a source of biodiesel. The fatty acid composition ofP. pinnata oil has been reported Akoh et al. [37]. The predominantfatty acid is oleic acid (51.8%) with linoleic acid (17.7%), palmiticacid (10.2%), stearic acid (7.0%), and linolenic acid (3.6%).The two major species of genus mahua found in India are M. indicaand Madhuca longifolia [14]. M. indica is one of the forest basedtree-borne non-edible oils with large production potential of about60 million tons per annum in India [104]. The yield of M. indicaseeds varies (5–200 kg/tree) depending upon size and age of thetree [105]. The mahua tree starts producing seeds after 10 yearsand continues up to 60 years. The kernel constitutes about 70% ofthe seed and contains 50% oil [25]. The mahua oil generally containsabout 20% FFAs and a procedure for converting this mahuaoil to biodiesel is very much required [93]. Rice bran oil is anunderutilized non-edible vegetable oil, which is available in largequantities in rice cultivating countries, and very little researchhas been done to utilize this oil as a replacement for mineral diesel[91].Biodiesel from oil crops is being produced in increasingamounts as a clean-burning alternative fuel, but its production inlarge quantities is not sustainable. Extensive use of vegetable oilsmay cause other significant problems such as starvation in developingcountries. With nearly 60% of humans in the world now currentlymalnourished, the need for grains and other basic food cropscontinues to be critical [106]. Microalgae have long been recognizedas potentially good sources for bio-fuel production becauseof their high oil content and rapid biomass production. In recentyears, use of microalgae as an alternative biodiesel feedstock hasgained renewed interest from researchers, entrepreneurs, and the

M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835 1821general public [107]. Microalgae are classified as diatoms (bacillariophyceae),green algae (chlorophyceae), goldenbrown (chrysophyceae)and blue-green algae (cyanophyceae) [108]. Microalgaeare the untapped resource with more than 25,000 species of whichonly 15 are in use [109]. According to Benemann [110] total worldcommercial microalgal biomass production is about 10,000 tonsper year. Microalgae are photosynthetic microorganisms that convertsunlight, water and carbon dioxide to algal biomass [111].Microalgae contain lipids and fatty acids as membrane components,storage products, metabolites and sources of energy. Algaepresent an exciting possibility as a feedstock for biodiesel, andwhen you realize that oil was originally formed from algae. Ascould be seen from Table 11, algae contain anywhere between 2%and 40% of lipids/oils by weight [112]. The oil contents of somemicroalgae are given in Table 12 [113]. They have much highergrowth rates and productivity when compared to conventional forestry,agricultural crops, and other aquatic plants, requiring muchless land area than other biodiesel feedstocks of agricultural origin,up to 49 or 132 times less when compared to rapeseed or soybeancrops, for a 30% (w/w) of oil content in algae biomass [114]. Theproduction of algae to harvest oil for biodiesel has not been undertakenon a commercial scale, but working feasibility studies havebeen conducted to arrive at the above number. Specially bred mustardvarieties can produce reasonably high oil yields and have theadded benefit that the meal left over after the oil has been pressedout can act as an effective and biodegradable pesticide [14].Table 11Chemical composition of algae on a dry matter basis (%) [112].Species of sample Proteins Carbohydrates Lipids NucleicacidScenedesmus obliquus 50–56 10–17 12–14 3–6Scenedesmus quadricauda 47 – 1.9 –Scenedesmus dimorphus 8–18 21–52 16–40 –Chlamydomonas rheinhardii 48 17 21 –Chlorella vulgaris 51–58 12–17 14–22 4–5Chlorella pyrenoidosa 57 26 2 –Spirogyra sp. 6–20 33–64 11–21 –Dunaliella bioculata 49 4 8 –Dunaliella salina 57 32 6 –Euglena gracilis 39–61 14–18 14–20 –Prymnesium parvum 28–45 25–33 22–38 1–2Tetraselmis maculata 52 15 3 –Porphyridium cruentum 28–39 40–57 9–14 –Spirulina platensis 46–63 8–14 4–9 2–5Spirulina maxima 60–71 13–16 6–7 3–4.5Synechoccus sp. 63 15 11 5Anabaena cylindrica 43–56 25–30 4–7 –Table 12Oil contents of some microalgae [113].MicroalgaBotryococcus braunii 25–75Chlorella sp. 28–32Crypthecodinium cohnii 20Cylindrotheca sp. 16–37Dunaliella primolecta 23Isochrysis sp. 25–33Monallanthus salina >20Nannochloris sp. 20–35Nannochloropsis sp. 31–68Neochloris oleoabundans 35–54Nitzschia sp. 45–47Phaeodactylum tricornutum 20–30Schizochytrium sp. 50–77Tetraselmis sueica 15–23Oil content (wt.% of dry basis)3.3. Animal fats as biodiesel feedstockAnother group of feedstock for biodiesel production is fats derivedfrom animals. Animal fats used to produce biodiesel includetallow [115,116], choice white grease or lard [117], chicken fat[118] and yellow grease [119]. Compared to plant crops, these fatsfrequently offer an economic advantage because they are oftenpriced favorably for conversion into biodiesel [107]. Animal fatmethyl ester has some advantages such as high cetane number,non-corrosive, clean and renewable properties [120]. Animal fatstend to be low in FFAs and water, but there is a limited amountof these oils available, meaning these would never be able to meetthe fuel needs of the world [121].3.4. Waste cooking oil as biodiesel feedstockWaste cooking oil (WCO) is a promising alternative to vegetableoil for biodiesel production. In order to reduce the cost of production,WCO would be a good choice as raw material since it is cheaperthan virgin vegetable oils [122]. Its price is 2–3 times cheaperthan virgin vegetable oils [123]. The conversion of WCO intomethyl esters through the transesterification process approximatelyreduces the molecular weight to one-third, reduces the viscosityby about one-seventh, reduces the flash point slightly andincreases the volatility marginally, and reduces pour point considerably[14,124]. The production of biodiesel from WCO is challengingdue to the presence of undesirable components such as FFAsand water [125].The amount of WCO generated in each country is huge and variesdepending on the use of vegetable oil. Management of WCOs poses asignificant challenge because of its disposal problems and possiblecontamination of the water and land resources. Even though someof this waste cooking oil is used for soap production, a major partof it is discharged into the environment. As large amounts of WCOare illegally dumped into rivers and landfills, causing environmentalpollution, the use of waste cooking oil to produce biodiesel as petroleum-baseddiesel fuel substitute offers significant advantages becauseof the reduction in environmental pollution [18]. Therefore,biodiesel derived from WCO has taken a commercial patent as analternative fuel to petroleum-based diesel fuel for diesel enginesin the markets of Europe and the United States [126]. Every yearmany millions of tons of WCO are collected and used in a varietyof ways throughout the world. This is a virtually inexhaustiblesource of energy which might also prove an additional line of productionfor ‘‘green” companies [127]. An estimate of the potentialamount of WCO from the collection in the EU is approximately0.7–1.0 million tons per year [125]. US restaurants, including allthe fast-food chains, produce an estimated 3 billion gallons ofWCO per year [128]. According to Chhetri et al. [18], approximately135,000 tons per year of WCO is produced in Canada. Turkey producesover 350,000 tons of WCO per year [28]. According to a reportby the USDA Foreign Agricultural Service [129], WCO accountsabout 80,000 tons of China’s biodiesel production.4. Derivatives of triglycerides as diesel fuelsConsiderable efforts have been made to develop vegetable oilderivatives that approximate the properties and performance ofthe hydrocarbon-based diesel fuels. The problems with substitutingtriglycerides for diesel fuels are mostly associated with theirhigh viscosities, low volatilities and polyunsaturated character[35]. Dilution/blending, micro-emulsification, pyrolysis, andtransesterification are the four techniques applied to solve theproblems encountered with the high fuel viscosity. Among allthese techniques, the transesterification seems to be the best

1822 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835choice, as the physical characteristics of fatty acid esters are veryclose to those of diesel fuel and the process is relatively simple.4.1. Dilution with diesel fuelVegetable oils may be used with dilution modification techniqueas an alternative diesel fuel [130]. The problems posed bythe high viscosity of vegetable oil can be minimized by blendingit with diesel. The vegetable oil–diesel fuel blend is a simple wayto reduce the viscosity of neat vegetable oil. This method doesnot require any chemical process [131]. The use of blends of conventionaldiesel fuel with sunflower oil [132], soybean oil [133],cottonseed oil [134], rapeseed oil [135], J. curcas oil [84], P. pinnataoil [89], and WCO [136] has been described.Caterpillar Brazil, in 1980, used pre-combustion chamber engineswith the mixture of 10% vegetable oil to maintain total powerwithout any alteration or adjustment to the engine. At that point itwas not practical to substitute 100% vegetable oil for diesel fuel,but a blend of 20% vegetable oil and 80% diesel fuel was successful.Some short-term experiments used up to a 50:50 ratio [137].Ziejewski et al. [132] investigated the effects of the fuel blend of25% sunflower oil with 75% diesel fuel (25/75 fuel) in a directinjectiondiesel engine. The viscosity of this blend was 4.88 cSt at313 K (maximum specified ASTM value is 4.0 cSt at 313 K). Thisblend was considered not suitable for long-term use in a directinjectionengine.The effect of temperature on viscosities of J. curcas oil and variousblends in the range 298–348 K has been investigated by Pramanik[84]. Results of the study show that the viscosity of J. curcas oil ishigher than diesel oil at any temperature. Viscosity values of 50:50J. curcas oil/diesel fuel and 40:60 J. curcas oil/diesel fuel are close todiesel in the range of 328–333 K and at about 318 K, respectively,whereas the blend containing 30:70 J. curcas oil/diesel fuel has viscosityclose to diesel at the range of 308–313 K.4.2. Micro-emulsification of oilsMicro-emulsification is the formation of microemulsions (co-solvency)which is a potential solution for solving the problem of vegetableoil viscosity. Microemulsions are defined as transparent,thermodynamically stable colloidal dispersions in which the diameterof the dispersed-phase particles is less than one-fourth thewavelength of visible light. Microemulsion-based fuels are sometimesalso termed ‘‘hybrid fuels”, although blends of conventionaldiesel fuel with vegetable oils have also been called hybrid fuels [66].To solve the problem of the high viscosity of vegetable oils,microemulsions with immiscible liquids, such as methanol, ethanoland ionic or non-ionic amphiphiles have been studied. It hasbeen demonstrated that short-term performances of both ionicand non-ionic microemulsions of aqueous ethanol in soybean oilare nearly as well as that of No. 2 diesel fuel [138]. Ziejewski etal. [139] prepared a microemulsion containing 53% (v/v) alkali-refinedand winterized sunflower oil, 13.3% (v/v) 190-proof ethanoland 33.4% (v/v) 1-butanol. This non-ionic microemulsion had a viscosityof 6.31 cSt at 313 K, a cetane number of 25, and an ash contentof less than 0.01%. Lower viscosities and better spray patterns(more even) were observed with an increase of 1-butanol.4.3. PyrolysisThe goal of pyrolysis is the optimization of high-value fuel productsfrom biomass by thermal and catalytic means [140]. The pyrolyzedmaterial can be any type of biomass, such as vegetable oils,animal fats, wood, bio-waste, etc. Conversion of vegetable oilsand animal fats composed predominantly of triglycerides usingpyrolysis type reactions represents a promising option for theproduction of biodiesel [141]. Many investigators have reportedthe pyrolysis of triglycerides to obtain products suitable for dieselengines [142–147].Thermal decomposition of triglycerides produces the compoundsof classes including alkanes, alkenes, alkadienes, aromaticsand carboxylic acids. Different types of vegetable oils produce largedifferences in the composition of the thermally decomposed oil[35]. The mechanism of thermal decomposition of triglycerides inFig. 1 was proposed by Schwab et al. [142]. Mechanisms for thethermal decomposition of triglycerides are likely to be complex becauseof the many structures and multiplicity of possible reactionsof mixed triglycerides. Generally, thermal decomposition of thesestructures proceeds through either a free-radical or carboniumion mechanism. The formation of aromatics is supported by aDiels–Alder addition of ethylene to a conjugated diene formed inthe pyrolysis reaction. Carboxylic acids formed during the pyrolysisof vegetable oils probably result from cleavage of the glyceridemoiety [35,142]. The compositions of pyrolyzed oils are listed inTable 13. The main components were alkanes and alkenes, whichaccounted for approximately 60% of the total weight. Carboxylicacids accounted for another 9.6–16.1% [137].The fuel properties of the liquid product fractions of the thermallydecomposed vegetable oil are likely to approach diesel fuels.Pyrolyzed soybean oil, for instance, contains 79% carbon and11.88% hydrogen [35]. It has low viscosity and a high cetane numbercompared to pure vegetable oils. Soybean oil pyrolyzed distillate,which consisted mainly of alkanes, alkenes and carboxylicacids had a cetane number of 43, exceeding that of soybean oil(37.9) and the ASTM minimum value of 40. The viscosity of the distillatewas 10.2 mm 2 /s at 311 K, which is higher than the ASTMspecification for No. 2 diesel fuel (1.9–4.1 mm 2 /s) but considerablybelow that of soybean oil (32.6 mm 2 /s) [66].The HHV of pyrolysis oil from vegetable oil is pretty high. TheHHV of pyrolysis oil from rapeseed oil (38.4 MJ/kg) is slightly lowerthan that of gasoline (47 MJ/kg), diesel fuel (43 MJ/kg), or petroleum(42 MJ/kg) but higher than coal (32–37 MJ/kg) [15].Fig. 1. The mechanism of thermal decomposition of triglycerides [142].Table 13Compositional data of pyrolysis of oils (wt.%) [137].High oleic safflower Soybean oilN 2 sparge Air N 2 sparge AirAlkanes 37.5 40.9 31.1 29.9Alkenes 22.2 22.0 28.3 24.9Alkadienes 8.1 13.0 9.4 10.9Carboxylic acids 11.5 16.1 12.2 9.6Unresolved unsaturates 9.7 10.1 5.5 5.1Aromatics 2.3 2.2 2.3 1.9Unidentified 8.7 12.7 10.9 12.6

M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835 1823The soaps obtained from the vegetable oils can be pyrolyzedinto hydrocarbon-rich products [146–148]. The saponificationand pyrolysis of sodium soap of vegetable oil proceed as follows[146]:Saponification:Vegetable oils or fats þ NaOH ! RCOONa þ GlycerinPyrolysis of sodium soaps:4RCOONa þð1=2ÞO 2 ! R-R þ Na 2 CO 3 þ CO 2The soaps obtained from the vegetable oils can be pyrolyzed intohydrocarbon-rich products according to Eq. (2) with higher yieldsat lower temperatures. Demirbas [146] investigated the yields ofdecarboxylation products by pyrolysis from sodium soaps of fourvegetable oils. The maximum decarboxylation products were obtainedfrom pyrolysis of sunflower oil (97.5%), corn oil (97.1%), cottonseedoil (97.5%), and soybean oil (97.8%) at 610 K. Fig. 2. showsdistillation curves for sunflower oil and methyl ester and sodiumsoap of sunflower oil.The production of bio-fuels (especially bio-gasoline) from vegetableoils by catalytic cracking is a promising alternative. Severalvegetable oils (e.g. palm, canola, soybean) have been employed inthe process that involves conversion of the oils into bio-fuelsFig. 2. Distillation curves for sunflower oil and methyl ester and Na-soap ofsunflower oil [146].ð1Þð2Þsuitable for gasoline engines using acid catalysts such as transitionmetal catalysts (that yield bio-fuels enriched in diesel fraction –over 50% by weight) and zeolites and mesoporous materials thatgive bio-fuels with higher gasoline fractions – over 40% – withhigher aromatic content [149]. The reaction is normally performedat moderate to high temperatures (573–773 K) using different oilsto catalyst ratios depending on the oil and the catalyst [149]. Catalyticcracking not only increases the yield of gasoline by breakinglarge molecules into smaller ones, but also improves the quality ofthe gasoline: this process involves carbocations and yields alkanesand alkanes with the highly branched structures desirable in gasoline[147].Certain petroleum fractions are converted into other kinds ofchemical compounds. Catalytic isomerization changes straightchainalkanes into branched-chain ones. The cracking process convertshigher alkanes and alkenes, and thus increases the gasolineyield. The process of catalytic reforming converts alkanes andcycloalkanes into aromatic hydrocarbons and thus increases thegasoline yield. The process of catalytic reforming converts alkanesand cycloalkanes into aromatic hydrocarbons and thus providesthe chief raw material for the large-scale synthesis of anotherbroad class of compounds [15].4.4. Transesterification of triglycerides to biodieselAmongst the four techniques, chemical conversion (transesterification)of the oil to its corresponding fatty ester is the most promisingsolution to the high viscosity problem. Sometimes, it is moreconvenient to convert the alkyl group of an ester to another alkylgroup. This process is known as ester exchange or transesterification.Fig. 3 shows the transesterification reaction of triglycerides[150]. Transesterification is the reversible reaction of a fat or oilwith an alcohol (methanol or ethanol) to form fatty acid alkyl estersand glycerol. It can be alkali-, acid-, or enzyme-catalyzed;however, currently the majority of the commercialized technologyresides in transesterification using alkali-catalyzed reaction [137].Mostly, biodiesel is derived from the vegetable oils using sodium orpotassium hydroxide catalytic transesterification (m)ethanol process[151].The transesterification reaction proceeds with catalyst or withoutany catalyst by using primary or secondary monohydric aliphaticalcohols having 1–8 carbon atoms [124]. The alcoholsemployed in the transesterification are generally short chainalcohols such as methanol, ethanol, propanol, and butanol [152].Transesterification reaction is an equilibrium reaction. In this reaction,however, a larger amount of methanol is used to shift reactionequilibrium to right side and produce more methyl esters asFig. 3. Reactions in the transesterification of a triglyceride [150].

1824 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835proposed product [153]. Ethanol is a preferred alcohol in transesterificationreaction compared to methanol because it is derivedfrom agricultural products and is renewable and biologically lessobjectionable in the environment, however, methanol is preferredbecause of its low cost and its physical and chemical advantages(polar and shortest chain alcohol).Generally, the reaction temperature near the boiling point ofthe alcohol is recommended. Nevertheless, the reaction may becarried out at room temperature [126]. The reactions take placeat low temperatures (338 K) and at modest pressures (2 atm,1 atm = 101.325 kPa). Biodiesel is further purified by washing andevaporation to remove any remaining methanol. The oil (87%),alcohol (9%), and catalyst (1%) are the inputs in the production ofbiodiesel (86%), the main output [154]. Pretreatment is not requiredif the reaction is carried out under high pressure(9000 kPa) and high temperature (513 K), where simultaneousesterification and transesterification take place with maximumyield obtained at temperatures ranging from 333 to 353 K at a molarratio of 6:1 [155]. Inputs and mass requirements for the transesterificationprocess are given in Table 14.For a complete reaction, an FFA value of lower than 3% is neededand the other materials should be substantially anhydrous [150].Ideally, it should be kept below 1% [150]. The presence of watergives rise to hydrolysis of some of the produced ester, with consequentsoap formation. Soap formation reduces catalyst efficiency,causes an increase in viscosity, leads to gel formation and makesthe separation of glycerol difficult [156].The physical properties of the primary chemical products oftransesterification are given in Tables 15 and 16. The physicalTable 14Inputs and mass requirements for the transesterification process.InputRequirement/ton biodieselFeedstock1000 kg vegetable oilSteam requirement415 kgElectricity12 kW hMethanol96 kgCatalyst5 kgHydrochloric acid (37%)10 kgCaustic soda (50%)1.5 kgNitrogen 1 Nm 3Process water20 kgproperties of these products indicate that the boiling points andmelting points of the fatty acids, methyl esters, mono, di and triglyceridesincrease as the number of carbon atoms in the carbonchain increase, but decrease with increases in the number of doublebonds. The melting points increase in the order of tri, di andmonoglycerides due to the polarity of the molecules and hydrogenbonding [137].4.4.1. Catalytic transesterification methodsTransesterification reactions can be catalyzed by alkalis[54,157,158], acids [97,159,160] or enzymes [161–163]. The catalytictransesterification of vegetable oils with methanol is animportant industrial method used in biodiesel synthesis. Alsoknown as methanolysis, this reaction is well studied and establishedusing acids or alkalis, such as sulfuric acid or sodiumhydroxide as catalysts. However, these catalytic systems are lessactive or completely inactive for long chain alcohols. Usually,industries use sodium or potassium hydroxide or sodium or potassiummethoxide as catalyst, since they are relatively cheap andquite active for this reaction [164]. Enzymes-catalyzed procedures,using lipase as catalyst, do not produce side reactions, but the lipasesare very expensive for industrial scale production and athree-step process was required to achieve a 95% conversion [165].4.4.1.1. Acid-catalytic transesterification methods. Biodiesel producedby transesterification reaction can be catalyzed by sulfuric,phosphoric, hydrochloric and organic sulfonic acids. Currently,the catalysts more used in biodiesel production are the organicacids, such as the derivates of toluenesulfonic acid and, more often,mineral acids such as sulfuric acid [166]. Although transesterificationusing acid catalysts is much slower than that obtained from alkalicatalysis, typically 4000 times, if high contents of water andFFAs are present in the vegetable oil, acid-catalyzed transesterificationcan be used [167]. The acid-catalyzed reaction commonlyrequires temperatures above 373 K and reaction times of 3–48 hhave been reported, except when reactions were conducted underhigh temperature and pressure [168]. These catalysts give veryhigh yields in the transesterification process. Acid-catalyzedreactions require the use of high alcohol-to-oil molar ratios in orderto obtain good product yields in practical reaction times. However,ester yields do not proportionally increase with molar ratio.For instance, for soybean methanolysis using sulfuric acid, esterTable 15Physical properties of chemicals related to transesterification [137].Name Specific gravity (g/ml) Melting point (K) Boiling point (K) Solubility (>10%)Methyl myristate 0.875 291.0 – –Methyl palmitate 0.825 303.8 469.2 Acids, benzene, EtOH, Et 2 OMethyl stearate 0.85 311.2 488.2 Et 2 O, chloroformMethyl oleate 0.875 253.4 463.2 EtOH, Et 2 OMethanol 0.792 176.2 337.9 H 2 O, ether, EtOHEthanol 0.789 161.2 351.6 H 2 O, etherGlycerol 1.26 255.3 563.2 H 2 O, etherTable 16Melting points of fatty acids, methyl esters and mono-, di-, and triglycerides a [137].Fatty acidMelting point (K)Name Carbons Acid Methyl ester 1-Monoglyceride 1,3-Diglyceride TriglycerideMyristic 14 327.6 336.1 342.8 289.5 266.7Palmitic 16 292.0 303.8 312.3 253.4 238.2Stearic 18 343.7 350.2 254.7 308.4 285.5Oleic 18:1 340.0 349.5 352.6 294.7 270.6Linoleic 18:2 330.2 336.7 346.3 278.7 260.1aMelting point of highest melting, most stable polymorphic form.

M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835 1825formation was sharply improved from 77% using a methanol-to-oilratio of 3.3:1–87.8% with a ratio of 6:1. Higher molar ratios showedonly moderate improvement until reaching a maximum value at a30:1 ratio (98.4%) [13]. Despite its insensitivity to FFAs in the feedstock,acid-catalyzed transesterification has been largely ignoredmainly because of its relatively slower reaction rate [169].The mechanism of the acid-catalyzed transesterification of vegetableoils is shown in Fig. 4. It can be extended to di- and triglycerides.The protonation of carbonyl group of the ester leads to thecarbocation, which after a nucleophilic attack of the alcohol producesa tetrahedral intermediate. This intermediate eliminatesglycerol to form a new ester and to regenerate the catalyst [170].Acid-catalyzed transesterification reactions have been extensivelyreported in the literature. Reaction conditions reported inthe literature using acid catalysts are summarized in Table 17[168,171–173]. Rachmaniah et al. [171] studied the transesterificationof refined soybean oil (>99% triglycerides) and low grade highFFA (60%) rice bran oil with methanol using 10% hydrochloricacid as a catalyst (weight percent of oil) at a temperature of343 K. In the study, all the reactions were carried out with amethanol molar ratio of 20:1. The highest FAME content of 90%was obtained from transesterification of low grade high FFA ricebran oil with a reaction time of 6 h.4.4.1.2. Alkali-catalytic transesterification methods. The transesterificationprocess is catalyzed by alkaline metal alkoxides, andhydroxides, as well as sodium or potassium carbonates. Sodiummethoxide is the most widely used biodiesel catalyst with over60% of industrial plants using this catalyst [174]. Alkaline catalystshave the advantages, e.g. short reaction time and relatively lowtemperature can be used with only a small amount for catalystand with little or no darkening of colour of the oil [68]. They showhigh performance for obtaining vegetable oils with high quality,but a question often arises; that is, the oils contain significantamounts of FFAs which cannot be converted into biodiesels butto a lot of soap [175]. These FFAs react with the alkaline catalystto produce soaps that inhibit the separation of the biodiesel, glycerin,and wash water [176].In the alkali catalytic methanol transesterification method, thecatalyst is dissolved into methanol by vigorous stirring in a smallreactor. The oil is transferred into a biodiesel reactor and thenthe catalyst/alcohol mixture is pumped into the oil. The final mixtureis stirred vigorously for 2 h at 340 K in ambient pressure. Asuccessful transesterification reaction produces two liquid phases:ester and crude glycerol [14].The reaction mechanism for alkali-catalyzed transesterificationwas formulated as three steps [137]. The alkali-catalyzed transesterificationof vegetable oils proceeds faster than the acid-catalyzedreaction. The mechanism of the alkali-catalyzed transesterificationof vegetable oils is shown in Fig. 5. The nucleophilic attack of thealkoxide at the carbonyl group of the triglyceride generates a tetrahedralintermediate (Fig. 5, Step 1), from which the alkyl ester andthe corresponding anion of the diglyceride are formed (Fig. 5, Step2). The latter deprotonates the catalyst, thus regenerating the activespecies (Fig. 5, Step 3), which is now able to react with a secondmolecule of the alcohol, starting another catalytic cycle. Diglyceridesand monoglycerides are converted by the same mechanismto a mixture of alkyl esters and glycerol [14,177,178].In general, alkali-catalyzed transesterification processes arecarried out at low temperatures and pressures (333–338 K and1.4–4.2 bar) with low catalyst concentrations (0.5–2 wt.%) [13].Table 18 shows typical reaction conditions for alkali-catalyzedtransesterification processes in biodiesel synthesis. Alkaline metalalkoxides (as CH 3 ONa for the methanolysis) are the most activecatalysts, since they give very high yields (>98%) in short reactiontimes (30 min) even if they are applied at low molar concentrations(0.5 mol%). However, they require the absence of water,which makes them inappropriate for typical industrial processes[14,177]. Alkaline metal hydroxides (KOH and NaOH) are cheaperthan metal alkoxides, but less active. Nevertheless, they are a goodFig. 4. Mechanism of the acid-catalyzed transesterification [170].Table 17Summary of some studies using acid catalysts.Catalyst Oil Catalyst (wt.%) Temp. (K) Reaction time (h) Yield (%) ReferencesH 2 SO 4 Soybean 1 338 69 >90 [168]HCl Soybean 10 343 45 65 [171]HCl Rice bran 10 343 6 >90 [171]p-TsOH a Corn 4 353 2 97.1 [172]AlCl 3 Canola 5 383 18 98 [173]ap-TsOH – p-toluenesulfonic acid (CH 3 C 6 H 4 SO 3 H).

1826 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835Fig. 5. Mechanism of the alkali-catalyzed transesterification of vegetable oils (B: base) [177,178].Table 18Typical reaction conditions for biodiesel synthesis using homogeneous alkali catalysis[13].FeedstocksAlcohol-to-oil molar ratio(recommended)TemperaturePressureCatalystCatalyst concentration (byweight of lipid feedstock)ConversionsTriglyceride mixtures with low free fattyacid contents (

1828 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835tion is, therefore, an important step for high quality biodiesel fuelproduction [216].Demirbas [213] studied the transesterification reaction undersupercritical methanol employing six potential vegetable oils atvarying molar ratios of alcohol to vegetable oil and reaction temperature.It was found that at molar ratio of methanol to oil of 24,523 K, and 300 s reaction time, the best methyl ester yield was95%. Balat [217] investigated the yields of ethyl esters from vegetableoils via transesterification in supercritical ethanol. The criticaltemperature and the critical pressure of ethanol were 516.2 K and6.4 MPa, respectively. It was observed that increasing reaction temperature,especially supercritical temperatures had a favorableinfluence on ester conversion (Fig. 6). The critical temperaturesand critical pressures of the various alcohols are shown in Table 22.According to the current literature, non-catalytic transesterificationreactions at high temperature and pressure conditions provideimproved phase solubility, decrease mass-transfer limitations,provide higher reaction rates and make easier separation and purificationsteps. Besides, the supercritical transesterification methodis more tolerant to the presence of water and FFAs than the conventionalalkali-catalyzed technique, and hence more tolerant tovarious types of vegetable oils, even for fried and waste oils. Comparisonof the yields in alkaline-catalyzed, acid-catalyzed andsupercritical methanol is given in Table 23 [218].4.4.3. Effect of different reaction parameters on the biodiesel yieldThe main factors affecting transesterification are the molar ratioof glycerides to alcohol, catalyst, reaction temperature and pressure,reaction time and the contents of FFAs and water in oils.Fig. 6. Changes in yield percentage of ethyl esters as treated with subcritical andsupercritical ethanol at different temperatures as a function of reaction time. Molarratio of vegetable oil to ethyl alcohol: 1:40 [217].Table 22Critical temperatures and critical pressures of various alcohols [187].Alcohol Critical temperature (K) Critical pressure (MPa)Methanol 512.2 8.1Ethanol 516.2 6.41-Propanol 537.2 5.11-Butanol 560.2 4.94.4.3.1. Effect of molar ratio and alcohol type. The molar ratio of alcoholto oil is one of the most important variables influencing theconversion into esters. Although the stoichiometric molar ratio ofmethanol to triglyceride for transesterification is 3:1, higher molarratios are used to enhance the solubility and to increase the contactbetween the triglyceride and alcohol molecules [219]. A highermolar ratio is required to drive the reaction to completion at a fasterrate [94]. The molar ratio of 6:1 or higher generally gives themaximum yield (higher than 98 wt.%) [220]. Lower molar ratios requirea longer time to complete the reaction. With higher molar ratiosproduction is increased but recovery is decreased due to poorseparation of glycerol. At optimum molar ratio only the processgives higher yield and easier separation of the glycerol. The optimummolar ratios depend on the type and quality of the vegetableoil used [220].The vegetable oils are transesterified 1:6–1:40 vegetable oil–alcohol molar ratios in catalytic and supercritical alcohol conditions[14]. The sunflower seed oil was transesterified 1:1, 1:3,1:9, 1:20, and 1:40 vegetable oil–ethanol molar ratios in supercriticalethanol conditions [217]. It was observed that increasing molarratio had a favorable influence on ester conversion. Fig. 7 showsthe effect of molar ratio of sunflower seed oil to ethanol on yieldof ethyl ester at 517 K.Acid-catalyzed reactions require the use of high alcohol-to-oilmolar ratios in order to obtain good product yields in practicalreaction times. However, ester yields do not proportionally increasewith molar ratio. For instance, for soybean methanolysisusing sulfuric acid, ester formation sharply improved from 77%using a methanol-to-oil ratio of 3.3:1–87.8% with a ratio of 6:1.Higher molar ratios showed only moderate improvement untilreaching a maximum value at a 30:1 ratio (98.4%) [13].Another important variable affecting the yield of ester is thetype of alcohol to triglyceride. In general, short chain alcohols suchas methanol, ethanol, propanol, and butanol can be used in thetransesterification reaction to obtain high methyl ester yields.Canakci and Van Gerpen [221] investigated the effect of differentalcohol types on acid-catalyzed transesterification of pure soybeanoil. Results of this study are summarized in Table 24.4.4.3.2. Effect of water and FFA contents. The water and FFA contentare key parameters for determining the viability of the vegetableoil transesterification process. In the base-catalyzed transesterificationprocess, the acid value of the vegetable oil should be lessthan 1 and all materials should be substantially anhydrous [14].If acid value is greater than 1, more alkali is injected to neutralizethe FFAs. The presence water has a greater negative effect than thatTable 23Comparison of the yields in alkaline-catalyzed, acid-catalyzed and supercritical methanol [218].Raw material FFA a content (wt.%) Water content (wt.%) Yields of methyl esters (wt.%)Alkaline-catalyzed Acid-catalyzed SCM bRapeseed oil 2.0 0.02 97.0 98.4 98.5Palm oil 5.3 2.1 94.4 97.8 98.9Frying oil 5.6 0.2 94.1 97.8 96.9Waste palm oil >20.0 >61.0 – – 95.8abFFA – free fatty acid.SCM – supercritical methanol.

M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835 1829Fig. 8. Yields of methyl esters as a function of water content in transesterification oftriglycerides [218].Fig. 7. Effect of molar ratio of vegetable oil to ethanol on yield of ethyl ester.Temperature 517 K [217].of the FFAs [126]. Water can cause soap formation and frothing.The resulting soaps can induce an increase in viscosity, formationof gels and foams, and made the separation of glycerol difficult[14,178].In the conventional transesterification of fats and vegetable oilsfor biodiesel production, free fatty acids and water always producenegative effects since the presence of free fatty acids and watercauses soap formation consumes catalyst and reduces catalysteffectiveness. In catalyzed methods, the presence of water has negativeeffects on the yields of methyl esters [14]. However, the presenceof water affected positively the formation of methyl esters insupercritical methanol method. Recently, Kusdiana and Saka [218]studied the effect of water on the yield of methyl esters by thesupercritical methanol treatment (623 K, 43 MPa, 4 min of treatmentwith a methanol to oil molar ratio of 42:1) compared withthose from alkaline- and acid-catalyzed method. In the supercriticalmethanol method the amount of water added into the reactionsystem did not have a significant effect on the conversion. Effect ofwater and FFA contents on the alkaline- and acid-catalyzed andsupercritical transesterification of triglycerides are shown in Figs.8 and 9.4.4.3.3. Effect of reaction temperature and time. The reaction temperatureinfluences the reaction rate and yield of ester. Therefore,generally the reaction is conducted close to the boiling point ofmethanol, 318–338 K at atmospheric pressure [14,100]. Further increasein temperature is reported to have a negative effect on theconversion [222–224]. Iso et al. [222] examined the effect of reactiontemperature on production of propyl oleate at the temperaturerange from 313 K to 343 K with free P. fluorescens lipase. Inthis study, the conversion ratio to propyl oleate was observed highestat 333 K, whereas the activity highly decreased at 343 K. Foonet al. [223] studied the effect of temperature on the transesterificationof palm oil using a molar ratio of oil to methanol of 1:10, catalyzedby NaOMe and NaOH, and at 323 K, 333 K and 343 K,respectively. For reactions using NaOH, the effect of temperatureFig. 9. Yields of methyl esters as a function of free fatty acid content intransesterification of triglycerides [218].was more significant. For the NaOH catalyzed reactions, it is bestto carry them out at 333 K. Darnoko1 et al. [224] studied transesterificationof palm oil with methanol (6:1) and 1% KOH at differentreaction temperatures. According to study results, the best yieldwas obtained at 338 K (82%). Effect of temperature on the ethyl estersproduction from soybean oil in propane medium with lipasefrom Yarrowia lipolytica was investigated by Hildebrand et al.[193]. The experiments were performed in the temperature rangeof 308–338 K at 50 bar, enzyme concentration of 5 wt.%, oil to ethanolmolar ratio of 1:6, and solvent to substrates mass ratio of 2:1.Fig. 10 presents the results of these experiments.The methyl ester conversion rate increases with the reactiontime. Different researchers have reported different reaction timesfor the transesterification process. Kim et al. [225] studied thetransesterification reaction, with methanol (6:1), of vegetable oilsusing NaOH and Na/NaOH/c-Al 2 O 3 as catalysts at 333 K. Theyreached the maximum biodiesel yield within 1 h both for the caseof homogeneous and heterogeneous catalyst system. For thehomogeneous catalyst system, the maximum biodiesel productionyield was higher by 20% than that of the heterogeneous catalystsystem. Results of this study are shown in Fig. 11. Demirbas[213] investigated the changes in yield percentage of methyl estersTable 24The effect of alcohol type on conversion and specific gravity of ester [221].Alcohol type Boiling point (K) Reaction temperature (K) Conversion (%) Specific gravity of biodieselMethanol 338 333 87.8 0.88762-Propanol 355.4 348 92.9 0.87861-Butanol 390 383 92.1 0.8782Ethanol 351.5 348 95.8 0.8814

1830 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835Fig. 10. Effect of temperature on the ethyl esters production from soybean oil inpropane medium with lipase from Yarrowia lipolytica [193].Fig. 11. Effect of reaction time on the biodiesel production yield. Methanol/VOmolar ratio 6:1, reaction temperature 333 K, stirring speed 300 rpm, without cosolvent[225].4.4.3.4. Effect of catalyst content and type. Catalysts used for thetransesterification of triglycerides are classified as alkali, acid, enzyme.Alkali-catalyzed transesterification is much faster thanacid-catalyzed transesterification and is most often used commercially[137]. Stavarache et al. [165] investigated the effect of differentcatalyst concentrations on base-catalyzed transesterificationduring biodiesel production from vegetable oil by means of ultrasonicenergy. The best yields were obtained when the catalystwas used in small concentration, i.e. 0.5% wt/wt of oil.The effect of different catalysts types on methanolysis of RBDpalm oil with a low FFA content of 300 50% yieldHCl (conc.) 1 >300 30% yieldIon exchange resin (H+) 2 >300 Too slowDowex 50 (Na+) 1 >300 Too slowAcid treated florisil 2 >300 Too slowActivated silica gel 1 >300 Not suitablea The following conditions were used: type of oil = RBD palm oil (FFA = 0.05%);ratio of oil-to-solvent (methanol) = 1:15.6; temperature = reflux temperature.Fig. 12. Changes in yield percentage of methyl esters as treated with subcritical andsupercritical methanol at different temperatures as a function of reaction time.Molar ratio of vegetable oil to methyl alcohol: 1:41. Sample: hazelnut kernel oil[213].as treated with subcritical and supercritical methanol at differenttemperatures as a function of reaction time (Fig. 12).Fig. 13. The effect of different catalyst types at different temperatures during theproduction of free and bound ethyl ester (FAEE) from castor oil [227].

M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835 1831separate, water is added at the rate of 5.5% by volume of the feedstockand then stirred for 5 min and the glycerol allowed to settleagain. After settling is complete the glycerol is drained and the esterlayer remains [15].An excess of alcohol is normally used in biodiesel production, tobe sure that the oil or fat used will be fully converted to esters. Butthe amount of alcohol in the system has to be minimized for goodphase separation. Once the glycerin and biodiesel phases havebeen separated, the excess alcohol in each phase is removed witha flash evaporation process or by distillation. In others systems,the alcohol is removed and the mixture neutralized before theglycerin and esters have been separated. In either case, the alcoholis recovered using distillation equipment and is re-used. Care mustbe taken to ensure no water accumulates in the recovered alcoholstream [211,229].5. ConclusionWith recent increases in petroleum prices and uncertaintiesconcerning petroleum availability, there is renewed interest invegetable oil fuels for diesel engines. Vegetable oils have the potentialto substitute a fraction of petroleum distillates and petroleum-basedpetrochemicals in the near future. Fundamentally,high viscosity appears to be a property at the root of many problemsassociated with direct use of vegetable oils as engine fuel.Vegetable oils can be converted into to avoid these viscosity-relatedproblems. Different ways have been considered to reducethe high viscosity of vegetable oils.The transesterfication of vegetable oils by methanol, ethanol,propanol, and butanol has proved to be the most promising process.Biodiesel from transesterification of the vegetable oils is betterthan diesel fuel in terms of sulfur content, flash point, aromaticcontent and biodegradability. Ethanol is a preferred alcohol intransesterification reaction compared to methanol because it is derivedfrom agricultural products and is renewable and biologicallyless objectionable in the environment, however, methanol is preferredbecause of its low cost and its physical and chemical advantages(polar and shortest chain alcohol).The catalytic transesterification of vegetable oils with methanolis an important industrial method used in biodiesel synthesis.Although transesterification using acid catalysts is muchslower than that obtained from alkali catalysis, typically 4000times, if high contents of water and FFAs are present in the vegetableoil, acid-catalyzed transesterification can be used. Alkalinecatalysts have the advantages, e.g. short reaction time and relativelylow temperature can be used with only a small amountfor catalyst and with little or no darkening of colour of the oil.They show high performance for obtaining vegetable oils withhigh quality. However, alkaline catalyzed transesterification hasseveral drawbacks: it is energy intensive, the recovery of glycerolis difficult, the acid or alkaline catalyst has to be removed fromthe product, alkaline wastewater requires treatment, and FFAsand water interfere with the reaction. Enzymatic catalysts like lipasesare able to effectively catalyze the transesterification oftriglycerides in either aqueous or non-aqueous systems. In particular,it should be noted that the byproduct, glycerol, can beeasily recovered with simple separation processes. Nevertheless,enzymatic catalysts are often more expensive than chemical catalysts,so recycling and reusing them is often a must for commercialviability.Recently, there have been some reports on the non-catalytictransesterification reaction employing supercritical methanol conditions.Transesterification of vegetable oils in supercritical methanolare carried out without using any catalyst. Supercriticalmethanol has a high potential for transesterification of triglyceridesto methyl esters for diesel fuel substitute. Besides, the supercriticaltransesterification method is more tolerant to the presenceof water and FFAs than the conventional alkali-catalyzed technique,and hence more tolerant to various types of vegetable oils,even for fried and waste oils.Compared with transesterification, pyrolysis process has moreadvantages. The fuel properties of the liquid product fractions ofthe thermally decomposed vegetable oil are likely to approach dieselfuels. The higher heating value of pyrolysis oil from a vegetableoil is slightly lower than that of gasoline, diesel fuel or petroleum,but higher than coal.References[1] Bunnell DE. Predicting worldwide consumption of petroleum by correlatingGDP growth and energy efficiency. School of business, University of Texas ofthe Permian Basin, Odessa; 2007. [accessed on October 2009].[2] British Petroleum (BP). BP Statistical review of world energy 2009. BP plc,London; June 2009.[3] Demirbas A. Biohydrogen. London: Springer Publishing Co; 2009 [June 30].[4] Goldemberg J. Environmental and ecological dimensions of biofuels. In:Conference on the ecological dimensions of biofuels, Washington (DC); 2008[March 10].[5] International Energy Agency (IEA). Key world energy statistics 2008. OECD/IEA, Paris; 2008.[6] Plunkett JW. Plunkett’s automobile industry almanac 2008: automobile, truckand specialty vehicle industry market research, statistics, trends & leadingcompanies. Houston (Texas): Plunkett Research Ltd; 2007.[7] World Business Council for Sustainable Development (WBCSD). Mobility2030: meeting the challenges to sustainability. The sustainable mobilityproject, Geneva (Switzerland); 2004.[8] Balat M, Balat M. Political, economic and environmental impacts of biomassbasedhydrogen. Int J Hydrogen Energy 2009;34:3589–603.[9] Balat M, Balat H. Recent trends in global production and utilization of bioethanolfuel. Appl Energy 2009;86:2273–82.[10] Namasivayam AM, Korakianitis T, Crookes RJ, Bob-Manuel KDH, Olsen J.Biodiesel, emulsified biodiesel and dimethyl ether as pilot fuels for naturalgas fuelled engines. Appl Energy 2010;87:769–78.[11] Demirbas A. Biofuels sources, biofuel policy, biofuel economy and globalbiofuel projections. Energy Convers Manage 2008;49:2106–16.[12] Marchetti JM, Miguel VU, Errazu AF. Possible methods for biodieselproduction. Renew Sustain Energy Rev 2007;11:1300–11.[13] Lotero E, Goodwin JG, Bruce DA, Suwannakarn K, Liu Y, Lopez DE. Thecatalysis of biodiesel synthesis. Catalysis 2006;19:41–83.[14] Demirbas A. Progress and recent trends in biodiesel fuels. Energy ConversManage 2009;50:14–34.[15] Demirbas A. Biodiesel: a realistic fuel alternative for dieselengine. London: Springer Publishing Co; 2008.[16] Demirbas MF, Balat M. Recent advances on the production and utilizationtrends of bio-fuels: a global perspective. Energy Convers Manage2006;47:2371–81.[17] Balat M, Balat H. A critical review of biodiesel as a vehicular fuel. EnergyConvers Manage 2008;49:2727–41.[18] Chhetri AB, Watts KC, Islam MR. Waste cooking oil as an alternate feedstockfor biodiesel production. Energies 2008;1:3–18.[19] Bamgboye AI, Hansen AC. Prediction of cetane number of biodiesel fuel fromthe fatty acid methyl ester (FAME) composition. Int Agrophys 2008;22:21–9.[20] Balat M. Modeling vegetable oil viscosity. Energy Source Part A2008;30:1856–69.[21] Kulkarni BM, Bujar BG, Shanmukhappa S. Investigation of acid oil as a sourceof biodiesel. Indian J Chem Tech 2008;15:467–71.[22] Van Gerpen J. Business management for biodiesel producers. Nationalrenewable energy laboratory, NREL/SR-510-36242, Golden, Colorado; July2004. [accessed on October2009].[23] Capehart BL. Encyclopedia of energy engineering and technology. Boca Raton(FL): CRC Press, Taylor & Francis LLC; 2007.[24] Balat M. Fuel characteristics and the use of biodiesel as a transportation fuel.Energy Source Part A 2006;28:855–64.[25] Pandey A. Handbook of plant-based biofuels. Boca Raton (FL): CRC Press,Taylor & Francis LLC; 2008.[26] Chien SM, Huang YJ, Chuang SC, Yang HH. Effects of biodiesel blending onparticulate and polycyclic aromatic hydrocarbon emissions in nano/ultrafine/fine/coarse ranges from diesel engine. Aerosol Air Qual Res 2009;9:18–31.[27] US Environmental Protection Agency (EPA). A comprehensive analysisof biodiesel impacts on exhaust emissions. Draft technical report,EPA420-P02-001; October 2002. www.biodiesel.org/resources/../reports/gen/20021001_gen [accessed on October 2009].[28] Nas B, Berktay A. Energy potential of biodiesel generated from waste cookingoil: an environmental approach. Energy Source Part B 2007;2:63–71.

1832 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835[29] Leduc S, Natarajan K, Dotzauer E, McCallum I, Obersteiner M. Optimizingbiodiesel production in India. Appl Energy 2009;86:S125–31.[30] Kurki A, Hill A, Morris M. Biodiesel: the sustainability dimensions. ATTRAPublication; 2006 [#IP281, p. 1–12].[31] Da Costa RE, Lora EES, Angarita EY, Torres E. The energy balance in theproduction of palm oil. Biodiesel – two case studies: Brazil and Colombia.World bioenergy conference, Jonkoping (Sweden), May 30 – June 1; 2006.[32] Sheehan J, Camobreco V, Duffield J, Graboski M, Shapouri H. An overview ofbiodiesel and petroleum diesel life cycles. National renewable energylaboratory, report no. NREL/TP-580-24772, Golden, Colorado (USA); May1998.[33] Tazewell A. Transportation, Energy, and the Environment in North Carolina2008;73(Spring/Summer):27–39.[34] Johnston M, Holloway T. A global comparison of national biodieselproduction potentials. Environ Sci Technol 2007;41:7967–73.[35] Srivastava A, Prasad R. Triglycerides-based diesel fuels. Renew Sustain EnergyRev 2000;4:111–33.[36] Pinto AC, Guarieiro LLN, Rezende MJC, Ribeiro NM, Torres EA, Lopes WA, et al.Biodiesel: an overview. J Braz Chem Soc 2005;16:1313–30.[37] Akoh CC, Chang SW, Lee GC, Shaw JF. Enzymatic approach to biodieselproduction. J Agric Food Chem 2007;55:8995–9005.[38] Demirbas A. Biodiesel fuels from vegetable oils via catalytic and non-catalyticsupercritical alcohol transesterifications and other methods: a survey. EnergyConvers Manage 2003;44:2093–109.[39] Edem DO. Palm oil: biochemical, physiological, nutritional, hematological,and toxicological aspects: a review. Plant Foods Hum Nutr 2002;57:319–41.[40] Bhatt YC, Murthy NS, Datta RK. Use of mahua oil (Madhuca indica) as a dieselfuel extender. J Inst Eng (India) Agri Eng 2004;85:10–4.[41] Demirbas A. Production of biodiesel fuels from linseed oil using methanoland ethanol in non-catalytic SCF conditions. Biomass Bioenergy 2009;33:113–8.[42] Cigizoglu KB, Ozaktas T, Karaosmanoglu F. Used sunflower oil as analternative fuel for diesel engines. Energy Sources 1997;19:559–66.[43] Demirbas A. Relationships derived from physical properties of vegetable oiland biodiesel fuels. Fuel 2008;87:1743–8.[44] Agarwal D, Kumar L, Agarwal AK. Performance evaluation of a vegetable oilfuelled compression ignition engine. Renew Energy 2008;33:1147–56.[45] Rao TV, Rao GP, Reddy KHC. Experimental investigation of pongamia,jatropha and neem methyl esters as biodiesel on CI engine. JJMIE2008;2:117–22.[46] Demirbas MF, Balat M, Balat H. Potential contribution of biomass to thesustainable energy development. Energy Convers Manage 2009;50:1746–60.[47] Hariharan VS, Reddy KV, Rajagopal K. Study of the performance, emission andcombustion characteristics of a diesel engine using sea lemon oil-based fuels.Indian J Sci Technol 2009;2:43–7.[48] Gui MM, Lee KT, Bhatia S. Feasibility of edible oil vs. non-edible oil vs. wasteedible oil as biodiesel feedstock. Energy 2008;33:1646–53.[49] Mitchell D. A note on rising food prices. World bank policy research workingpaper no. 4682, world bank – development economics group (DEC),Washington (DC), August 27; 2008.[50] Food and Agriculture Organization of the United Nations (FAO). The marketand food security implications of the development of biofuel production. FAOcommittee on commodity problems, sixty-seventh session, Rome, April 20–22; 2009.[51] Lustig N. The curse of volatile food prices: policy dilemmas in the developingworld. Tenth annual conference on ‘‘natural resources and development”global development network, Kuwait City (Kuwait), February 1–5; 2009.[52] Trostle R. Global agricultural supply and demand: factors contributing to therecent increase in food commodity prices. USDA economic research service,report WRS-0801, Washington (DC); July 2008.[53] Transport & Environment (T&E). The real impact of growing biofuels:calculating indirect land-use change. Joint-NGO publication fromGreenpeace, birdlife international, friends of the Earth Europe, T&E, EEB,November 5; 2008. [accessed on September 2009].[54] Anwar F, Rashid U, Ashraf M, Nadeem M. Okra (Hibiscus esculentus) seed oilfor biodiesel production. Appl Energy 2010;87:779–85.[55] Pimentel D, Marklein A, Toth MA, Karpoff M, Paul GS, McCormack R, et al.Biofuel impacts on world food supply: use of fossil fuel, land and waterresources. Energies 2008;1:41–78.[56] Pimentel D, Marklein A, Toth MA, Karpoff M, Paul GS, McCormack R, et al.Food versus biofuels: environmental and economic costs. Human Ecol2009;37:1–12.[57] Cao P, Dube MA, Tremblay AY. High-purity fatty acid methyl ester productionfrom canola, soybean, palm, and yellow grease lipids by means of amembrane reactor. Biomass Bioenergy 2009;32:1028–36.[58] Pahl G. Biodiesel: growing a new energy economy. Vermont (USA): ChelseaGreen Publishing Company; 2008.[59] Van der Velde, Aloe A, Bouraoui F. Pan-European regional-scale modelling ofwater and N efficiencies of rapeseed cultivation for biodiesel production.Global Change Biol 2009;15:306–18.[60] European Commission. Agriculture in the European Union: statistical andeconomic information. Directorate-General for Agriculture and RuralDevelopment, Brussels, February 2007.[61] Jacquet F, Bamiere L, Bureau JC, Guinde L, Guyomard H, Treguer D. Recentdevelopments and prospects for the production of biofuels. In: Theconference on ‘‘biofuels, food and feed tradeoffs”, Saint Louis, April 12–13;2007.[62] Rapier R. Renewable diesel. In: Pimentel D, editor. Biofuels, solar and wind asrenewable energy systems: benefits and risks. Springer Verlag; 2008[September 11].[63] Balat M. Prospects for worldwide biodiesel market development. EnergySource Part B 2009;4:48–58.[64] Milbrandt A, Overend RP. The future of liquid biofuels for APEC economies.Asia-Pacific economic cooperation (APEC), APEC energy working group,research report, report no. APEC#208-RE-01.8, Singapore; May 2008.[65] Marvey BB. Sunflower-based feedstocks in nonfood applications:perspectives from olefin metathesis. Int J Mol Sci 2008;9:1393–406.[66] Knothe G, Dunn RO, Bagby MO. Biodiesel: the use of vegetable oils and theirderivatives as alternative diesel fuels. In: ACS symposium series no. 666: fuelsand chemicals from biomass, Washington (DC); 1997. p. 172–208.[67] Nag A. Biofuels refining and performance. Professional (New York): McGraw-Hill; 2007 [December 17].[68] Singh RK, Padhi SK. Characterization of jatropha oil for the preparation ofbiodiesel. Nat Prod Radiance 2009;8:127–32.[69] Cheenkachorn K, Udornthep I. A Development of four-stroke engine oil usingpalm oil as a base stock. In: The 2nd Joint international conference on‘‘sustainable energy and environment (SEE 2006)”, Bangkok (Thailand),November 21–23; 2006.[70] Salunkhe DK, Chavan JK, Adsule RN, Kadam SS. World oilseeds: chemistry,technology, and utilization. New York: Van Nostrand Reinhold; 1992.[71] Musa JJ. Evaluation of the lubricating properties of palm kernel oil. LeonardoEl J Pract Technol 2009;8:107–14.[72] Singh RK, Kumar AK, Sethi S. Preparation of karanja oil methyl ester. In:Offshore world. Mumbai (India): Jasubhai Media Pvt. Ltd; 2006 [April–May].[73] Kywe TT, Oo MM. Production of biodiesel from jatropha oil (Jatropha curcas)in pilot plant. WASET 2009;50:477–83.[74] Nagi J, Ahmed SK, Nagi F. Palm biodiesel an alternative green renewableenergy for the energy demands of the future. In: The international conferenceon construction and building technology (ICCBT 2008), Kuala Lumpur(Malaysia), June 16–20; 2008.[75] Alamu OJ, Akintola TA, Enweremadu CC, Adeleke AE. Characterization ofpalm-kernel oil biodiesel produced through NaOH-catalysedtransesterification process. Sci Res Essays 2008;3:308–11.[76] Abdullah S, Tiong FC. Prediction of palm oil properties using artificial neuralnetwork. IJCSNS 2008;8:101–6.[77] Richter B. Environmental challenges and the controversy about palm oilproduction – case studies from Malaysia, Indonesia and Myanmar. Friedrich–Ebert–Stiftung office for regional cooperation in Asia, Singapore; October 2009. [accessed on January 2010].[78] Malaysian Palm Oil Council (MPOC). Malaysian palm oil: a gift from nature, agift for life. Annual report 2007, Selangor Darul Ehsan, Malaysia; June 2008.[79] Greenergy International Ltd. Palm oil in biodiesel – a Greenergy perspective,London; 2008. [accessed onOctober 2009].[80] Thamsiriroj T, Murphy JD. Is it better to import palm oil from Thailand toproduce biodiesel in Ireland than to produce biodiesel from indigenous Irishrape seed? Appl Energy 2009;86:595–604.[81] Alamu OJ, Waheed MA, Jekayinfa SO. Effect of ethanol–palm kernel oil ratioon alkali-catalyzed biodiesel yield. Fuel 2008;87:1529–33.[82] Refaat AA. Correlation between the chemical structure of biodiesel and itsphysical properties. Int J Environ Sci Technol 2009;6:677–94.[83] Chhetri AB, Tango MS, Budge SM, Watts KC, Islam MR. Non-edible plant oilsas new sources for biodiesel production. Int J Mol Sci 2008;9:169–80.[84] Pramanik K. Properties and use of Jatropha curcas oil and diesel fuel blends incompression ignition engine. Renew Energy 2003;28:239–48.[85] Mahanta P, Mishra SC, Kushwah YS. An experimental study of Pongamiapinnata L. oil as a diesel substitute. Proc Inst Mech Eng J Power Eng2006;220:803–8.[86] Tiwari AK, Kumar A, Raheman H. Biodiesel production from jatropha oil(Jatropha curcas) with high free fatty acids: an optimized process. BiomassBioenergy 2007;31:569–75.[87] Ganapathy T, Murugesan K, Gakkhar RP. Performance optimization ofjatropha biodiesel engine model using Taguchi approach. Appl Energy2009;86:2476–86.[88] Naik M, Meher LC, Naik SN, Das LM. Production of biodiesel from high freefatty acid karanja (Pongamia pinnata) oil. Biomass Bioenergy 2008;32:354–7.[89] Agarwal AK, Rajamanoharan K. Experimental investigations of performanceand emissions of karanja oil and its blends in a single cylinder agriculturaldiesel engine. Appl Energy 2009;86:106–12.[90] Veljkovic VB, Lakicevic SH, Stamenkovic OS, Todorovic ZB, Lazic ML. Biodieselproduction from tobacco (Nicotiana tabacum L.) seed oil with a high contentof free fatty acids. Fuel 2006;85:2671–5.[91] Sinha S, Agarwal AK, Garg S. Biodiesel development from rice bran oil:transesterification process optimization and fuel characterization. EnergyConvers Manage 2008;49:1248–57.[92] Lin L, Ying D, Chaitep S, Vittayapadung S. Biodiesel production from crude ricebran oil and properties as fuel. Appl Energy 2009;86:681–8.

M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835 1833[93] Ghadge SV, Raheman H. Biodiesel production from mahua (Madhuca indica)oil having high free fatty acids. Biomass Bioenergy 2005;28:601–5.[94] Ramadhas AS, Jayaraj S, Muraleedharan C. Biodiesel production from high FFArubber seed oil. Fuel 2005;84:335–40.[95] Demirbas A. Production of biodiesel from algae oils. Energy Source Part A2009;31:163–8.[96] Huang G, Chen F, Wei D, Zhang X, Chen G. Biodiesel production by microalgalbiotechnology. Appl Energy 2010;87:38–46.[97] Gao C, Zhai Y, Ding Y, Wu Q. Application of sweet sorghum for biodieselproduction by heterotrophic microalga Chlorella protothecoides. Appl Energy2010;87:756–61.[98] Mukherjee, P. Remote village electrification through the biofuels route: theWII experience. In: 5th international biofuels conference, New Delhi,February 7–8; 2008.[99] Van Eijck J, Romijn H. Prospects for jatropha biofuels in Tanzania: an analysiswith strategic niche management. Energy Policy 2008;36:311–25.[100] Sharma YC, Singh B, Upadhyay SN. Advancements in development andcharacterization of biodiesel: a review. Fuel 2008;87:2355–73.[101] Scott PT, Pregelj L, Chen N, Hadler JS, Djordjevic MA, Gresshoff PM. Pongamiapinnata: an untapped resource for the biofuels industry of the future?Bioenergy Res 2008;1:2–11.[102] Sugla T, Purkayastha J, Singh SK, Solleti SK, Sahoo L. Micropropagation ofPongamia pinnata through enhanced axillary branching. Vitro Cell Dev BiolPlant 2007;43:409–14.[103] Chopade VV, Tankar AN, Pande VV, Tekade AR, Gowekar NM, Bhandari SR,Khandake SN. Pongamia pinnata: Phytochemical constituents, traditional usesand pharmacological properties: a review. Int J Green Pharm 2008;2:72–5.[104] Ghosal MK, Das DK, Pradhan SC, Sahoo N. Performance study of diesel engineby using mahua methyl ester (biodiesel) and its blends with diesel fuel. AgriEng Int CIGR Ejournal 2008;10:1–9.[105] Puhan S, Vedaraman N, Rambrahaman BV, Nagarajan G. Mahua (Madhucaindica) seed oil: a source of renewable energy in India. J Sci Ind Res2005;64:890–6.[106] Pimentel D, Pimentel M. Corn and cellulosic ethanol cause major problems.Energies 2008;1:35–7.[107] Wen Z, Johnson MB. Microalgae as a feedstock for biofuel production.Communications and marketing, College of Agriculture and Life Sciences,Virginia Polytechnic Institute and State University, Publication 442-886,Blacksburg (VA); 2009.[108] Rengel A. Promising technologies for biodiesel production from algae growthsystems. In: The 8th European symposium of the international farmingsystems association (IFSA), Clermont-Ferrand (France), July 6–10; 2008.[109] Raja R, Hemaiswarya S, Kumar NA, Sridhar S, Rengasamy R. A perspective onthe biotechnological potential of microalgae. Crit Rev Microbiol2008;34:77–88.[110] Benemann JR. Microalgae biofuels: a brief introduction. Benemann associatesand MicroBio engineering, Walnut Creek (CA), January 1; 2009.[accessed on September 2009].[111] Chisti Y. Biodiesel from microalgae beats bioethanol. Trends Biotechnol2008;26:126–31.[112] Becker EW. In: Baddiley J et al., editors. Microalgae: biotechnology andmicrobiology. Cambridge (New York): Cambridge University Press; 1994.[113] Chisti Y. Biodiesel from microalgae. Biotechnol Adv 2007;25:294–306.[114] Mata TM, Martins AA, Caetano NS. Microalgae for biodiesel production andother applications: a review. Renew Sustain Energy Rev 2010;14:217–32.[115] Nelson RG, Schrock MD. Energetic and economic feasibility associated withthe production, processing, and conversion of beef tallow to a substitutediesel fuel. Biomass Bioenergy 2006;30:584–91.[116] Oner C, Altun S. Biodiesel production from inedible animal tallow and anexperimental investigation of its use as alternative fuel in a direct injectiondiesel engine. Appl Energy 2009;86:2114–20.[117] Lu J, Nie K, Xie F, Wang F, Tan T. Enzymatic synthesis of fatty acid methylesters from lard with immobilized Candida sp. 99–125. Process Biochem2007;42:1367–70.[118] Guru M, Koca A, Can O, Cinar C, Sahin F. Biodiesel production from wastechicken fat based sources and evaluation with Mg based additive in a dieselengine. Renew Energy 2010;35:637–43.[119] Diaz-Felix W, Riley MR, Zimmt W, Kazz M. Pretreatment of yellow grease forefficient production of fatty acid methyl esters. Biomass Bioenergy2009;33:558–63.[120] Guru M, Artukoglu BD, Keskin A, Koca A. Biodiesel production from wasteanimal fat and improvement of its characteristics by synthesized nickel andmagnesium additive. Energy Convers Manage 2009;50:498–502.[121] Sheedlo M. A review of the processes of biodiesel production. MMG 445 BasicBiotechnol 2008;4:61–5.[122] Hameed BH, Goh CS, Chin LH. Process optimization for methyl esterproduction from waste cooking oil using activated carbon supportedpotassium fluoride. Fuel Process Technol 2009;90:1532–7.[123] Phan AN, Phan TM. Biodiesel production from waste cooking oils. Fuel2008;87:3490–6.[124] Demirbas A. Biodiesel from waste cooking oil via base-catalytic andsupercritical methanol transesterification. Energy Convers Manage2009;50:923–7.[125] Jacobson K, Gopinath R, Meher LC, Dalai AK. Solid acid catalyzed biodieselproduction from waste cooking oil. Appl Catal B Environ 2008;85:86–91.[126] Canakcı M, Ozsezen AN. Evaluating waste cooking oils as alternative dieselfuel. GU J Sci 2005;18:81–91.[127] Zanzi, R Aguilar PJV, Díaz IL, Moya IH. Biodiesel fuel production. In:Proceedings of the Cubasolar 2006, Cienfuegos (Cuba), April 17–21;2006.[128] Chiras DD. Environmental science. 8th ed. Sudbury (MA): Jones and BartlettPublishers, Inc; 2009.[129] Latner K, O’Kray C, Jiang J. China, peoples republic of bio-fuels an alternativefuture for agriculture 2006. USDA foreign agricultural service, GAIN report-CH6049, August 8; 2006.[130] Ozaktas T. Compression ignition engine fuel properties of a used sunfloweroil–diesel fuel blend. Energy Source 2000;22:377–82.[131] Hariharan VS, Reddy KV, Rajagopal K. Study of the performance, emission andcombustion characteristics of a diesel engine using sea lemon oil based fuels.Indian J Sci Technol 2009;2:43–7.[132] Ziejewski M, Goettler H, Pratt GL. Comparative analysis of the long-termperformance of a diesel engine on vegetable oil based alternative fuels,society of automotive engineers paper no. 860301. SAE, Warrendale (PA);1986.[133] Wibulswas P, Chirachakhrit S, Keochung U, Tiansuwan J. Combustion ofblends between plant oils and diesel oil. Renew Energy 1999;16:1098–101.[134] Krishna BM, Mallikarjuna JM. Properties and performance of cotton seed oil–diesel blends as a fuel for compression ignition engines. J Renew SustainEnergy 2009;1:1–10.[135] Hazar H, Aydin H. Performance and emission evaluation of a CI engine fueledwith preheated raw rapeseed oil (RRO)–diesel blends. Appl Energy2010;87:786–90.[136] Dorado MP, Arnal JM, Gomez J, Gil A, Lopez FJ. The effect of a waste vegetableoil blend with diesel fuel on engine performance. ASAE Trans2002;45:519–23.[137] Ma F, Hanna MA. Biodiesel production: a review. Bioresour Technol1999;70:1–15.[138] Goering CE, Camppion RN, Schwab AW, Pryde EH. Vegetable oil fuels. In:International conference on plant and vegetable oils as fuels, Fargo, NorthDakota. St. Joseph (MI): American Society of Agricultural Engineers; 1982. p.279–86.[139] Ziejewski M, Kaufman KR, Schwab AW, Pryde EH. Diesel engine evaluation ofa nonionic sunflower oil–aqueous ethanol microemulsion. J Am Oil Chem Soc1984;61:1620–6.[140] Balat M, Demirbas MF. Bio-oil from pyrolysis of black alder wood. EnergySource Part A 2009;31:1719–27.[141] Maher KD, Bressler DC. Pyrolysis of triglyceride materials for the productionof renewable fuels and chemicals. Bioresour Technol 2007;98:2351–68.[142] Schwab AW, Dykstra GJ, Selke E, Sorenson SC, Pryde H. Diesel fuel fromthermal decomposition of soybean oil. J Am Oil Chem Soc 1988;65:1781–5.[143] Billaud F, Dominguez V, Broutin P, Busson C. Production of hydrocarbons bypyrolysis of methyl esters from rapeseed oil. J Am Oil Chem Soc1995;72:1149–54.[144] Dandik L, Aksoy A. Pyrolysis of used sunflower oil in the presence of sodiumcarbonate by using fractionating pyrolysis reactor. Fuel Process Technol1998;57:81–92.[145] Lima DG, Soares VCD, Ribeiro BE, Carvalho DA, Cardoso ECV, Rassi FC, et al.Diesel-like fuel obtained by pyrolysis of vegetable oils. J Anal Appl Pyrol2004;71:987–96.[146] Demirbas A. Diesel fuel from vegetable oil via transesterification and soappyrolysis. Energy Source 2002;24:835–41.[147] Demirbas A. Fuel conversional aspects of palm oil and sunflower oil. EnergySource 2003;25:457–66.[148] Demirbas A, Kara H. New options for conversion of vegetable oils toalternative fuels. Energy Source Part A 2006;28:619–26.[149] Luque R, Herrero-Davila L, Campelo JM, Clark JH, Hidalgo JM, Luna D,et al. Biofuels: a technological perspective. Energy Environ Sci 2008;1:513–96.[150] Abdullah AZ, Razali N, Mootabadi H, Salamatinia B. Critical technical areasfor future improvement in biodiesel technologies. Environ Res Lett 2007;2:1–6.[151] Demirbas A. Biodiesel from wood oils in compressed methanol. EnergySource Part A 2009;31:1530–6.[152] Speight JG. Synthetic fuels handbook: properties, process, andperformance. New York: McGraw-Hill; 2008.[153] Demirbas A. Biodiesel production via rapid transesterification. Energy SourcePart A 2008;30:1830–4.[154] Lucia LA, Argyropoulos DS, Adamopoulos L, Gaspar AR. Chemicals and energyfrom biomass. Can J Chem 2006;8:960–70.[155] Barnwal BK, Sharma MP. Prospects of biodiesel production from vegetableoils in India. Renew Sustain Energy Rev 2005;9:363–78.[156] Refaat AA, Attia NK, Sibak HA, El Sheltawy ST, ElDiwani GI. Productionoptimization and quality assessment of biodiesel from waste vegetable oil.Int J Environ Sci Tech 2008;5:75–82.[157] Leung DYC, Wu X, Leung MKH. A review on biodiesel production usingcatalyzed transesterification. Appl Energy 2010;87:1083–95.[158] Kotwal MS, Niphadkar PS, Deshpande SS, Bokade VV, Joshi PN.Transesterification of sunflower oil catalyzed by flyash-based solidcatalysts. Fuel 2009;88:1773–8.

1834 M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835[159] Li Y, Zhang XD, Sun L, Zhang J, Xu HP. Fatty acid methyl ester synthesiscatalyzed by solid superacid catalyst /ZrO 2 –TiO 2 /La 3+ . Appl Energy2010;87:156–9.[160] Miao X, Li R, Yao H. Effective acid-catalyzed transesterification for biodieselproduction. Energy Convers Manage 2009;50:2680–4.[161] Chen Y, Xiao B, Chang J, Fu Y, Lv P, Wang X. Synthesis of biodiesel from wastecooking oil using immobilized lipase in fixed bed reactor. Energy ConversManage 2009;50:668–73.[162] Dizge N, Keskinler B. Enzymatic production of biodiesel from canola oil usingimmobilized lipase. Biomass Bioenergy 2008;32:1274–8.[163] Liu Y, Xin H, Yan Y. Physicochemical properties of stillingia oil: feasibility forbiodiesel production by enzyme transesterification. Ind Crop Prod2009;30:431–6.[164] Macedo CCS, Abreu FR, Tavares AP, Alves MP, Zara LF, Rubim JC, et al. Newheterogeneous metal–oxides based catalyst for vegetable oiltransesterification. J Braz Chem Soc 2006;17:1291–6.[165] Stavarache C, Vinatoru M, Nishimura R, Maed Y. Fatty acids methyl estersfrom vegetable oil by means of ultrasonic energy. Ultrason Sonochem2005;12:367–72.[166] Cardoso AL, Neves SCG, Da Silva MJ. Esterification of oleic acid for biodieselproduction catalyzed by SnCl 2 : a kinetic investigation. Energies 2008;1:79–92.[167] De Oliveira D, Do Nascimento I, Di Luccio M, Faccio C, Dalla Rosa C, Bender JP,et al. Kinetics of enzyme-catalyzed alcoholysis of soybean oil in n-hexane.Appl Biochem Biotechnol 2005;121:231–41.[168] Freedman B, Pryde EH, Mounts TL. Variables affecting the yields of fattyesters from transesterified vegetable oils. J Am Oil Chem Soc1984;61:1638–43.[169] Zhang Y, Dube MA, McLean DD, Kates M. Biodiesel production from wastecooking oil: 1. Process design and technological assessment. BioresourTechnol 2003;89:1–16.[170] Meher LC, Sagar DV, Naik SN. Technical aspects of biodiesel productionby transesterification – a review. Renew Sustain Energy Rev 2006;10:248–68.[171] Rachmaniah O, Ju YH, Vali SR, Tjondronegoro I, Musfil AS. A study on acidcatalyzedtransesterification of crude rice bran oil for biodiesel production.In: youth energy symposium, 19th World energy congress and exhibition,Sydney (Australia), September 5–9; 2004.[172] Guan G, Kusakabe K, Sakurai N, Moriyama K. Transesterification of vegetableoil to biodiesel fuel using acid catalysts in the presence of dimethyl ether.Fuel 2009;88:81–6.[173] Soriano Jr NU, Vendittia R, Argyropoulos DS. Biodiesel synthesis viahomogeneous Lewis acid catalyzed transesterification. Fuel 2009;88:560–565.[174] Huber GW, Iborra S, Corma A. Synthesis of transportation fuels from biomass:chemistry, catalysts, and engineering. Chem Rev 2006;106:4044–98.[175] Furuta S, Matsuhasbi H, Arata K. Biodiesel fuel production with solidsuperacid catalysis in fixed bed reactor under atmospheric pressure. CatalCommun 2004;5:721–3.[176] Canakci M, Gerpen JV. A pilot plant to produce biodiesel from high free fattyacid feedstocks. Trans ASAE 2003;46:945–55.[177] Schuchardt U, Sercheli R, Vargas RM. Transesterification of vegetable oils: areview. J Braz Chem Soc 1998;9:199–210.[178] Demirbas A. Biodiesel production from vegetable oils via catalytic and noncatalyticsupercritical methanol transesterification methods. Progr EnergCombust Sci 2005;31:466–87.[179] Zagonel GF, Peralta-Zamora PG, Ramos LP. Production of ethyl esters fromcrude soybean oil: optimization of reaction yields using a 23 experimentaldesign and development of a new analytical strategy for reaction control.Preprints Symp: Am Chem Soc Div Fuel Chem 2002;47:363–4.[180] Dias JM, Alvim-Ferraz MCM, Almeida MF. Comparison of the performanceof different homogeneous alkali catalysts during transesterification ofwaste and virgin oils and evaluation of biodiesel quality. Fuel 2008;87:3572–8.[181] Rashid U, Anwar F, Knothe G. Evaluation of biodiesel obtained fromcottonseed oil. Fuel Process Technol 2009;90:1157–63.[182] Rashid U, Anwar F, Ansari TM, Arif M, Ahmad M. Optimization of alkalinetransesterification of rice bran oil for biodiesel production using responsesurface methodology. J Chem Tech Biotechnol 2009;84:1364–70.[183] Lubes ZIZ, Zakaria M. Analysis of parameters for fatty acid methyl estersproduction from refined palm oil for use as biodiesel in the single- and twostageprocesses. Malaysian J Biochem Mol Biol 2009;17:5–9.[184] Alamu OJ, Waheed MA, Jekayinfa SO, Akintola TA. Optimal transesterificationduration for biodiesel production from Nigerian palm kernel oil. Agri Eng IntCIGR Ejournal 2007;IX:1–11.[185] El Diwani G, Attia NK, Hawash SI. Development and evaluation of biodieselfuel and by-products from jatropha oil. Int J Environ Sci Technol2009;6:219–24.[186] Canakcı M, Van Gerpen J. Biodiesel production from oils and fats with highfree fatty acids. Trans ASAE 2001;44:1429–36.[187] Demirbas A. Biodiesel production via non-catalytic SCF method and biodieselfuel characteristics. Energy Convers Manage 2006;47:2271–82.[188] Kapilan N, Babu TPA, Reddy RP. Technical aspects of biodiesel and itsoxidation stability. Int J ChemTech Res 2009;1:278–82.[189] Fukuda H, Kondo A, Noda H. Biodiesel fuel production by transesterificationof oils. J Biosci Bioeng 2001;92:405–16.[190] Helwani Z, Othman MR, Aziz N, Kim J, Fernando WJN. Solid heterogeneouscatalysts for transesterification of triglycerides with methanol: a review. ApplCatal A: Gen 2009;363:1–10.[191] Oliveira D, Feihrmann AC, Rubira AF, Kunita MH, Dariva C, Oliveira JV.Assessment of two immobilized lipases activity treated in compressed fluids.J Supercrit Fluids 2006;38:373–82.[192] Rosa CD, Morandim MB, Ninow JL, Oliveira D, Treichel H, Oliveira JV. Lipasecatalyzedproduction of fatty acid ethyl esters from soybean oil incompressed propane. J Supercrit Fluids 2008;47:49–53.[193] Hildebrand C, Rosa CD, Freire DMG, Destain J, Dariva C, De OliveiraD, et al. Fatty acid ethyl esters production using a non-commerciallipase in pressurized propane medium. Ciênc Tecnol Aliment 2009;29:603–8.[194] Tiwari AK, Kumar R, Garg S. Transesterification, modeling and simulation ofbatch kinetics of non-edible vegetable oils for biodiesel production. In:Annual AIChE meeting, San Francisco, November 12–17; 2006.[195] Karam R, Karboune S, St-Louis R, Kermasha S. Lipase-catalyzed acidolysis offish liver oil with dihydroxyphenylacetic acid in organic solvent media.Process Biochem 2009;44:1193–9.[196] Shah S, Sharma S, Gupta MN. Enzymatic transesterification for biodieselproduction. Ind J Biochem Biophys 2003;40:392–9.[197] Balat M. Biodiesel fuel from triglycerides via transesterification – a review.Energy Source Part A 2009;31:1300–14.[198] Van Gerpen J, Shanks B, Pruszko R, Clements D, Knothe G. Biodiesel analyticalmethods: August 2002 – January 2004. National renewable energylaboratory, NREL/SR-510-36240, Colorado; July 2004.[199] Reyes-Duarte D, Lopez-Cortes N, Ferrer M, Plou JF, Ballesteros A. Parametersaffecting productivity in the lipase-catalysed synthesis of sucrose palmitate.Biocatal Biotransfor 2005;23:19–27.[200] Watanabe Y, Shimada Y, Sugihara A, Tominaga T. Conversion of degummedsoybean oil to biodiesel fuel with immobilized Candida antarctica lipase. J MolCatal B: Enzym 2002;17:151–5.[201] Ognjanovic N, Bezbradica D, Knezevic-Jugovic Z. Enzymatic conversion ofsunflower oil to biodiesel in a solvent-free system: process optimizationand the immobilized system stability. Bioresour Technol 2009;100:5146–54.[202] Jeong GT, Park DH. Lipase-catalyzed transesterification of rapeseed oil forbiodiesel production with tert – Butanol. Appl Biochem Biotechnol2007;148:131–9.[203] Linko YY, Lamsa M, Wu X, Uosukainen W, Sappala J, Linko P. Biodegradableproducts by lipase biocatalysis. J Biotechnol 1998;66:41–50.[204] Shao P, Meng X, He J, Sun P. Analysis of immobilized Candida rugosa lipasecatalyzed preparation of biodiesel from rapeseed soap stock. Food BioprodProces 2008;86:283–9.[205] Lv P, Wang X, Yuan Z, Tan T. Conversion of soybean oil to biodiesel fuel withimmobilized candida lipase on textile cloth. Energy Source Part A2008;30:872–9.[206] Shah S, Gupta MN. Lipase catalyzed preparation of biodiesel from jatropha oilin a solvent free system. Process Biochem 2007;42:409–14.[207] Bernardes OL, Bevilaqua JV, Leal MCMR, Freire DMG, Langone MAP.Biodiesel fuel production by the transesterification reaction of soybeanoil using immobilized lipase. Appl Biochem Biotechnol 2007;137–140:105–14.[208] Luo Y, Zheng Y, Jiang Z, Ma Y, Wei D. A novel psychrophilic lipase fromPseudomonas fluorescens with unique property in chiral resolution andbiodiesel production via transesterification. Appl Microbiol Biotechnol2006;73:349–55.[209] Lutz G, Chavarría M, Arias ML, Mata-Segreda JF. Microbial degradation ofpalm (Elaeis guineensis) biodiesel. Rev Biol Trop 2006;54:59–63.[210] Dizge N, Koseoglu DY, Keskinler B. Biodiesel Production by Thermomyceslanuginosus and Rhizomucor miehei immobilised in hydrophilic polyurethanefoam. European congress of chemical engineering (ECCE-6), Copenhagen(Denmark), September 16–20; 2007.[211] Soni SK. Microbes: a source of energy for 21st century. New Delhi(India): New India Publishing Agency; 2007.[212] Saka S, Kusdiana D. Biodiesel fuel from rapeseed oil as prepared insupercritical methanol. Fuel 2001;80:225–31.[213] Demirbas A. Biodiesel from vegetable oils via transesterification insupercritical methanol. Energy Convers Manage 2002;43:2349–56.[214] Kusdiana D, Saka S. Two-step preparation for catalyst-free biodiesel fuelproduction: hydrolysis and methyl esterification. Appl Biochem Biotechnol2004;115:781–91.[215] Saka S, Minami E. A novel non-catalytic biodiesel production process bysupercritical methanol as NEDO ‘‘high efficiency bioenergy conversionproject”. In: The 2nd joint international conference on ‘‘sustainableenergy and environment (SEE 2006)”, Bangkok (Thailand), November 21–23; 2006.[216] Minami E, Saka S. Kinetics in hydrolysis of oils/fats and subsequent methylesterification in two-step supercritical methanol method for biodieselproduction. In: The 2nd joint international conference on ‘‘sustainableenergy and environment (SEE 2006)”, Bangkok (Thailand), November 21–23; 2006.[217] Balat M. Biodiesel fuel production from vegetable oils via supercriticalethanol transesterification. Energy Source Part A 2008;30:429–40.[218] Kusdiana D, Saka S. Effects of water on biodiesel fuel production bysupercritical methanol treatment. Bioresour Technol 2004;91:289–95.

M. Balat, H. Balat / Applied Energy 87 (2010) 1815–1835 1835[219] Noureddini H, Harkey D, Medikonduru VA. Continuous process for theconversion of vegetable oils into methyl esters of fatty acids. J Am Oil ChemSoc 1998;75:1775–83.[220] Ramadhas AS. Biodiesel production technologies and substrates. In: PandeyA, editor. Handbook of plant-based biofuels. Boca Raton (FL, USA): CRC Press,Taylor & Francis Group; 2008.[221] Canakci M, Van Gerpen J. Biodiesel production via acid catalysis. Trans ASAE1999;42:1203–10.[222] Iso M, Chen B, Eguchi M, Kudo T, Shrestha S. Production of biodiesel fuel fromtriglycerides and alcohol using immobilized lipase. J Mol Catal B: Enzym2001;16:53–8.[223] Foon CS, May CY, Mgan MA, Hock CC. Kinetics study on transesterification ofpalm oil. J Oil Palm Res 2004;16:19–29.[224] Darnoko1 D, Cheryan M. Kinetics of palm oil transesterification in a batchreactor. J Am Oil Chem Soc 2000;77:1263–7.[225] Kim HJ, Kang BS, Kim MJ, Park YM, Kim DK, Lee JS, Lee KY. Transesterificationof vegetable oil to biodiesel using heterogeneous base catalyst. Catal Today2004;93–95:315–20.[226] May CY. Transesterification of palm oil: effect of reaction parameters. J OilPalm Res 2004;16:1–11.[227] Meneghetti PSM, Meneghetti MR, Wolf CR, Silva EC, Lima GES, Coimbra DA,et al. Ethanolysis of castor and cottonseed oil: a systematic study usingclassical catalysts. J Am Oil Chem Soc 2006;83:819–22.[228] Dunford NT. Biodiesel production techniques. Food technology factsheet, FAPC-150; 2006. [accessed on November2009].[229] Sundar S, Kumar S. A feasibility study on use of alternative fuels at constructionsite activities. MSc., in energy systems and the environment; September 2006. [accessedon November 2009].

Chemical Engineering Journal 154 (2009) 384–389Contents lists available at ScienceDirectChemical Engineering Journaljournal homepage: www.elsevier.com/locate/cejNovel highly integrated biodiesel production technology in a centrifugalcontactor separator deviceG.N. Kraai, B. Schuur, F. van Zwol, H.H. van de Bovenkamp, H.J. Heeres ∗Department of Chemical Engineering, University of Groningen, Nijenborgh 4, 9747 AG Groningen, The NetherlandsarticleinfoabstractArticle history:Received 19 November 2008Received in revised form 31 March 2009Accepted 17 April 2009Keywords:BiodieselPure plant oilsCentrifugal contactor separatorCINC V02The base catalyzed production of biodiesel (FAME) from sunflower oil and methanol in a continuouscentrifugal contactor separator (CCS) with integrated reaction and phase separation was studied. Theeffect of catalyst loading (sodium methoxide), temperature, rotational frequency and flow rates of thefeed streams was investigated. An optimized and reproducible FAME yield of 96% was achieved at a feedrate of 12.6 mL min −1 sunflower oil and a sixfold molar excess of MeOH (3.15 mL min −1 ) containing thecatalyst (1 wt% with respect to the oil). A jacket temperature of 75 ◦ C and a rotational frequency of 30 Hzwere applied. The productivity under those conditions (61 kg FAME m −3liquid min−1 ) was slightly higher thanfor a conventional batch process. The main advantage is the combined reaction–separation in the CCS,eliminating the necessity of a subsequent liquid–liquid separation step.© 2009 Elsevier B.V. All rights reserved.1. IntroductionConcerns for the environment combined with the current highcrude oil prices have stimulated the interest in biofuels from renewableresources [1–3]. In January 2007 the European Commissionpublished the New Energy Policy for Europe, targeting a 10% shareof biofuels in the transportation sector and raising the share ofrenewable energy to 20% by 2020 [1]. This has stimulated the productionof biofuels in Europe considerably, with biodiesel beingthe most important example. The projected biodiesel consumptionfor 2007 was 3.8 MTOE, a 70% increase compared to 2006[4].Typically, biodiesel, also known as FAME (fatty acid methyl ester)is produced from plant or vegetable oils and fats by transesterificationwith an alcohol (Fig. 1) [5,6]. A wide variety of differentoils and alcohols can be used for the production of biodiesel. Mostfrequently methanol is the alcohol of choice, although higher alcoholslike ethanol, 2-propanol and 1-butanol may be applied as well[7]. Glycerol is a byproduct from the transesterification and has tobe separated from the FAME after the reaction [8]. FAME productionmay either be catalyzed by acidic or basic catalysts. Typicalexamples of homogeneous base catalysts are sodium hydroxide,potassium hydroxide and sodium methoxide [9], well known examplesof homogeneous acidic catalysts are sulfuric acid, phosphoricacid and hydrochloric acid [10]. Both types of catalyzed reactionshave been extensively studied [11–13]. Besides homogeneous acidic∗ Corresponding author. Tel.: +31 50 363 4174; fax: +31 50 363 4479.E-mail address: h.j.heeres@rug.nl (H.J. Heeres).and basic catalysts, enzymes [14] and heterogeneous catalysts havebeen explored as well [11,15].FAME production is most commonly performed in batch operation,although continuous processes are emerging [11,16–21]. Newcatalyst, reactor and process concepts for biodiesel production havebeen reported the last decade. An interesting example is the use ofsupercritical methanol [22]. In this medium, complete conversionis obtained within 5 min of reaction time without the need of acatalyst.We here report the use of highly intensified centrifugal contactorseparator (CCS) equipment (CINC V02 [23]) for biodiesel synthesis.In the CCS device reaction and separation are combined in a singleapparatus, thus making it a good example of process intensification(PI). PI is currently one of the most significant trends in process engineeringand aims at replacing large, energy consuming processes[24] by small, highly integrated processes to reduce the size andenergy consumption of process plants. Some well known examplesof PI are reactive distillation, reactive extraction and the applicationof micro-reactors [24,25].The CINC V02 (Fig. 2) is basically a rotating centrifuge in a staticreactor housing. The immiscible liquids are fed to the CCS wherethey are dispersed in the annular zone between the static housingand the rotating centrifuge. The dispersion is then transferred intothe hollow centrifuge, through a hole in the bottom plate, where thephases are separated by centrifugal forces of up to 900 × g, making itpossible to separate fluids with densities that differ only 10 kg m −3 .Both liquid phases are collected individually, making use of a weirsystem.The CCS was originally designed for waste water cleaning in thenuclear industry [26] and has been used successfully for oil–water1385-8947/$ – see front matter © 2009 Elsevier B.V. All rights reserved.doi:10.1016/j.cej.2009.04.047

G.N. Kraai et al. / Chemical Engineering Journal 154 (2009) 384–389 385Nomenclaturea specific interfacial area (m 2 m −3 )d d drop diameter (m)F MeOH methanol flow rate (mL min −1 )F oil sunflower oil flow rate (mL min −1 )k L mass transfer coefficient (m s −1 ) c viscosity of the continuous phase (N s m −2 )N rotational frequency (Hz)r radius of the rotor (m) c density of the continuous phase (kg m 3 ) d density of the dispersed phase (kg m 3 )T jacket jacket temperature ( ◦ C)v s settling velocity (m s −1 )ω angular momentum (rad s −1 )separation [27] (e.g. for cleaning oils spills [28]), for extraction offermentation broths [29] and several other extraction processes[30–33]. We recently demonstrated the use of the CCS for enantioselectiveextractions of amino acid derivatives [34] and reportedprocess integration in the CCS by combining biphasic (bio)catalyticconversions with the separation of the catalyst and the reactionproducts in the device [35]. These examples clearly illustrate thepotential of the CCS equipment for the combined reaction and separationfor liquid–liquid systems. The use of the CCS for biodieselsynthesis was also mentioned in this communication. We herereport the results of an in depth experimental study to optimizethe biodiesel yield in a CCS. The effects of process variables likethe temperature, the flow rates and the rotational frequency on thebiodiesel yield will be provided and rationalized and the potentialof the CCS for biodiesel manufacture will be discussed.2. Methods2.1. MaterialsMethanol (99.8%) was obtained from Labscan, sodium methoxidesolution 30% in methanol (5.4 M) was obtained from Fluka.D 2 O (99.9%) and CDCl 3 (99.8%) were obtained from Sigma–Aldrich.Hydrochloric acid (37%) was obtained from Merck and the sunfloweroil was purchased from Albert Heijn, the Netherlands.2.2. Equipment and conditionsThe experiments were performed in a CCS of the type CINC V02(CINC stands for Costner Industries Nevada Corporation) equippedwith a heating/cooling jacket, a high-mix bottom plate and a0.925 in. weir. The CCS was operated at ambient pressure and jackettemperatures ranging from 60 to 75 ◦ C using a Julabo MV basis temperaturecontrolled water bath (accuracy ±0.01 ◦ C) connected to thejacket. Two Watson-Marlow 101U/R MK2 peristaltic tube pumpswere used to feed the CCS. The temperature was monitored at eachof the two CCS outlets using CMA B016BT temperature sensors connectedto a PC via a Coachlab II (CMA) interface. The system wasFig. 2. Schematic cross-sectional view of the CINC (courtesy of Auxill, the Netherlands).operated at atmospheric pressure by positioning air inlets in theliquid in and outlets. A schematic illustration of the set-up is shownin Fig. 3. The impeller in the annular zone is for illustrative purposesonly. In a CINC, mixing in the annular zone is the result of rotationof the centrifuge.2.3. Experimental procedureThe CCS was operated in a once-through mode for both liquidswithout recycle of the exit streams. The supply vessels containingthe pure sunflower oil and the methanol with sodium methoxide,respectively were preheated to 60 ◦ C and the water bath was setto the pre-determined temperature (60–75 ◦ C). The centrifuge wasstarted (20–90 Hz) and the CCS was fed with pure sunflower oil(12.6–40 mL min −1 ). The molar ratio of methanol to oil was set at6:1 in all experiments. As soon as the oil started flowing out ofthe heavy phase outlet, the reaction was started by feeding themethanol and sodium methoxide solution (1% (w/w) with regardto sunflower oil) at 3.15–10 mL min −1 . After steady operation wasachieved (typically 10–35 min), a glycerol stream with some unreactedmethanol exited the CCS through the heavy phase exit. FAMEwith unreacted oil left the CCS through the light phase exit. Thedissolved catalyst is present in both outlet phases. Samples weretaken at regular intervals from the light phase outlet and analyzedusing 1 HNMR.2.4. Analytical methodsThe FAME yield was determined using 1 H NMR. Hereto, a 1 mLsample of the light phase was directly quenched by adding 1 mL of0.1 M HCl in water to neutralize the remaining sodium methoxide.The mixture was vigorously shaken and centrifuged for 10 min. Afew drops were taken from the top layer and dissolved in CDCl 3 . Thesamples were analyzed using a 200 MHz Varian NMR. To determinethe FAME yield [36], the intensity of the characteristic signal of theFig. 1. The transesterification of a plant or vegetable oil or fat with methanol.

386 G.N. Kraai et al. / Chemical Engineering Journal 154 (2009) 384–389Fig. 3. Schematic representation of the CINC set-up for biodiesel synthesis. T: temperaturesensors.ester (2.3 ppm) was compared with that of the characteristic signalof the methyl end group (0.9 ppm) present in both sunflower oil andFAME.3. Results and discussion3.1. Exploratory experimentsThe experiments were carried out with sunflower oil and NaOMeas the catalyst. Sunflower oil is readily available in high purity witha low free fatty acid number (FFA). The latter is important as a highFFA number renders the base catalyzed biodiesel synthesis cumbersomedue to the formation of the sodium salts of free fatty acids(soap). The FFA number of the sunflower oil applied in this studywas below 1 wt%. The molar ratio of methanol to oil was set at 6:1in all experiments, in line with batch studies in the literature forsunflower oil and methanol [37].In an exploratory experiment the CCS was fed with sunflower oil(40 mL min −1 ) and a solution of NaOMe in MeOH (10 mL min −1 ,1%(w/w) with regard to sunflower oil). The reaction was performedwith a jacket temperature of 60 ◦ C and a rotational frequency (N)of50 Hz. The profile of the FAME yield in time is shown in Fig. 4.It takes about 15 min to reach the steady-state FAME yield ofabout 65%. The liquid residence time in the CCS is estimated tobe about 3.5 min, based on a total liquid hold-up of 180 mL [38]and a total liquid flow rate of 50 mL min −1 . Thus, steady-state isFig. 5. Effect of T jacket on the yield of FAME. Conditions—N: 30 Hz, F oil : 12.6 mL min −1 ,F MeOH/NaOMe : 3.15mLmin −1 (6:1 molar ratio of methanol to oil), and NaOMe: 1%(w/w) with regard to sunflower oil, line is for illustrative purpose only. Yield is theaverage steady-state yield for at least 2 h runtime.reached in just over 3 residence times. After reaching steady-state,the FAME with residual sunflower oil exited the CCS as the lightphase, whereas the heavy phase consisted of glycerol in MeOH. Togain insights in the reproducibility, a second run was performedunder identical conditions. The results are shown in Fig. 4 as welland imply that reproducibility is good.These exploratory experiments clearly indicate the proof of principleof biodiesel synthesis in a CCS. In the following part of thispaper, the effect of process conditions (temperature, catalyst loading,flow rates, rotational speed of centrifuge) on the biodiesel yieldwill be described, with the objective to obtain 95+% yield in a singlerun.3.2. Effect of reaction temperatureThe effect of the reaction temperature on the FAME yield wasstudied by adjusting the jacket temperature in the range 60–75 ◦ C,while keeping all the other variables constant (N: 30 Hz, F oil :12.6 mL min −1 , F MeOH/NaOMe :3.15mLmin −1 , and NaOMe: 1% (w/w)with regard to sunflower oil).The steady-state FAME yield at the lowest temperature in therange (60 ◦ C) was 87%. At 75 ◦ C, the yield increased to 96%, seeFig. 5 for details. It is well known that the FAME yield is a strongfunction of the temperature, with higher temperatures leading tohigher yields for a given reaction time [35]. Higher jacket temperaturesto further enhance the yield could not be applied due toexcessive methanol evaporation.3.3. Effect of catalyst loadingFig. 4. FAME yield as a function of time. Conditions for both runs—T jacket :60 ◦ C, F oil :40 mL min −1 , F MeOH/NaOMe :10mLmin −1 , (6:1 molar ratio of methanol to oil), N: 50 Hz,and NaOMe: 1% (w/w) with regard to sunflower oil. Yield is the average steady-stateyield for at least 2 h runtime.The amount of catalyst has a profound effect on the rateof biodiesel formation. Typically 1 wt% of catalyst on the oil isapplied. This loading was also used in the exploratory experimentsdescribed above. A number of experiments were performedat higher catalyst loadings (1–1.3% (w/w) with regard to sunfloweroil) with all other conditions at constant value (N: 30 Hz,F oil : 12.6 mL min −1 , F MeOH/NaOMe :3.15mLmin −1 , and T jacket =60 ◦ C).Unfortunately, when using sodium methoxide intakes higher thanthe standard 1% (w/w) on sunflower oil, precipitation of solids wasobserved in the centrifuge. Analyses by NMR show that these solidsare rich in fatty acids and suggest that the solids are sodium saltsof the fatty acids (soap). These may be formed by saponification

G.N. Kraai et al. / Chemical Engineering Journal 154 (2009) 384–389 387Fig. 6. Effect of flow rate variation on the FAME yield. Conditions—F oil :F MeOH/NaOMe :4:1 (6:1 molar ratio of methanol to oil), T jacket :60 ◦ C, NaOMe: 1% (w/w) with respectto sunflower oil, and N: 50 Hz, line is for illustrative purpose only. Yield is the averagesteady-state yield for at least 2 h runtime.of the triglyceride and reaction of the remaining free fatty acidswith NaOMe. Thus, yield enhancement by the application of higherintakes of catalyst is not possible due to solid soap formation in thecentrifuge.3.4. Effect of liquid flow ratesThe effect of the liquid flow rates on the FAME yield was studiedin the range 12.6–40 mL min −1 for sunflower oil. In all cases, themethanol flow rate was 25% of the sunflower flow rate to ensure afixed sixfold molar excess of methanol over sunflower oil [37].The FAME yield drops when increasing the sunflower flow ratefrom 12 to 40 mL min −1 , see Fig. 6 for details. The prime reasonis likely a reduction of the mean residence times of both phasesin the CINC when increasing the flow rates, leading to lower conversions.Recent work by Schuur et al. [39] have shown that theliquid hold-ups at these relatively low feed rates ( 40 Hz is determined solely by the intrinsic kinetics of biodieselsynthesis and the FAME yield is expected to be independent of thevalue of N (kinetic regime). However, this is not the case and theyield is lowered at higher N values. It is well possible that this reductionis related to the changes in the volume of the reactive phase inthe CINC as a function of the value of N. For a biphasic reaction withrelatively fast kinetics, the reaction only takes place in a dispersionconsisting of a continuous and a dispersed phase of small droplets.In the CCS, this dispersion is found in the annular zone as well asin parts of the centrifuge. The volume of the dispersed phase in theannular zone is about constant at these low flow rates and independentof N [39]. However, the volume of the dispersed phase inthe centrifuge is expected to be a function of the rate of rotation N.This may be rationalized by considering the settling velocity of individualdrops in the centrifuge. In a centrifugal settler the settlingvelocity for individual drops is given by Eq. (1) [44]: S,CCS = d2 d ( d − c )ω 2 r18 c(1)The settling velocity of the droplets in the dispersed zone is thusproportional to the difference in density, the angular momentumand the squared drop diameter.Due to the proportionality of the settling velocity with the angularmomentum, we can expect that the dispersed phase volumein the centrifuge will be reduced considerably at high rotationalspeeds, see Fig. 8 for details. Thus, the observed reduction of theFAME yield in the kinetic regime at N > 40 Hz, is likely due to astrong reduction of the volume of the reactive, dispersed phase inthe centrifuge of the CCS.

388 G.N. Kraai et al. / Chemical Engineering Journal 154 (2009) 384–389Fig. 8. Cross-sectional view of the CCS at low (A) and high rotor speeds (B). Hatched: dispersed zone, light grey: lighter phase, and darker grey: heavier phase.3.6. FAME synthesis at optimum settingsWith the effects of the most important process variables on theFAME yield established, experiments at optimum settings were performed(N = 30 Hz, catalyst load of 1% (w/w), jacket temperature75 ◦ C, flow of the oil and of the methanol/methoxide feed of 12.6and 3.15 mL min −1 , respectively). The results for two duplicate runsare shown in Fig. 9.After about 30 min, steady-state is reached and the FAME yieldwas on average 96%. Reproducibility is good, see Fig. 9 for details.3.7. Volumetric production rate in the CCSThe volumetric production rate of FAME production inthe CINC V02 at optimized settings were calculated to be61 kg FAME m −3liquid min−1 . This corresponds with an annual productionrate of about 5.6 ton/year. The largest CCS of the CINC typeavailable is the CINC V20 with a maximum flow through put of757 L min −1 . When assuming that the same volumetric productionrate can be achieved in this CCS and taking a volumetric ratio ofboth devices of about 400, the productivity in the large CINC V20 isestimated to be about 2.2 kton/year.It is of interest to compare the volumetric production rate in theCINC V02 with that for a typical batch process. Literature data implythat 98% yield is typically obtained in 20 min reaction time [45]. Thiscorresponds with a productivity of about 40 kg FAME m −3liquid min−1 .Thus, it appears that the productivity in the CINC V02 is at leastcomparable and likely higher than state of the art batch processes.However, compared to batch the CCS has other advantages,the main being that a liquid–liquid separator after reaction is notrequired as this function is already integrated in the CCS. Fromthe literature [46] it is known that it may take up to 2 h to obtaincomplete phase separation. Furthermore, the obvious advantagesof continuous processes compared to batch are also valid in thiscase (e.g. product consistency and operator effort).4. Conclusions and outlookIn this study, the proof of principle for the continuous biodieselmanufacture in a highly integrated CCS of the type CINC V02 is provided.As such, it demonstrates the potential of CCS equipment to beused for combined reactions and separation for biphasic (catalytic)systems. At optimum conditions a reproducible FAME yield of 96%was achieved. The volumetric production rates are at least comparableto state of the art batch processes. Further improvements arelikely by hardware modifications, and particularly by modificationsof the annular zone to allow for higher flow rates while maintaininga high conversion level. These studies as well as reactor engineeringstudies including detailed kinetics and hydrodynamics features arein progress at the moment.Due to the compact size and flexibility in operation, the CCSequipment is likely very suitable for biodiesel production in mobileunits in developing countries. A cascade of two CINCs in series, onefor biodiesel production and one for a subsequent aqueous washto remove remaining glycerol and catalyst, followed by a methanolstripper, may be a very attractive process option for further explorationand demonstration.ReferencesFig. 9. The yield of FAME in time, duplicate run. (N = 30 Hz, catalyst load of 1% (w/w),jacket temperature 75 ◦ C, flow of the oil and of the methanol/methoxide feed of 12.6and 3.15 mL min −1 , respectively (6:1 molar ratio of methanol to oil), yield is theaverage steady-state yield for at least 2 h runtime.)[1] New Energy Policy for Europe, European Commission, 2007.[2] Ethanol 2020: A Global Market Survey, Next-Generation Trends, and Forecasts,2008.[3] Biodiesel 2020: A Global Market Survey, Feedstock Trends, and Forecasts, 2008.[4] F. Kay-Prolea, Biofuels barometer, Le Journal des Energies Renouvelables 179(2007) 63–75.[5] M. Balat, Production of biodiesel from vegetable oils: a survey, Energy SourcesPart A: Recov. Utilization Environ. Effects 29 (2007) 895–913.[6] J.M. Marchetti, V.U. Miguel, A.F. Errazu, Possible methods for biodiesel production,Renew. Sust. Energy Rev. 11 (2007) 1300–1311.[7] H. Sanli, A. Canakci, Effects of different alcohol and catalyst usage on biodieselproduction from different vegetable oils, Energy Fuels 22 (2008) 2713–2719.[8] G. Knothe, J.H. Van Gerpen, J. Krahl, The Biodiesel Handbook, AOCS Press, Champaign,IL, 2005.[9] A. Singh, B. He, J. Thompson, J. van Gerpen, Process optimization of biodieselproduction using alkaline catalysts, Appl. Eng. Agric. 22 (2006) 597–600.[10] F.R. Ma, M.A. Hanna, Biodiesel production: a review, Bioresour. Technol. 70(1999) 1–15.[11] A.A. Kiss, F. Omota, A.C. Dimian, G. Rothenberg, The heterogeneous advantage:biodiesel by catalytic reactive distillation, Top. Catal. 40 (2006) 141–150.

G.N. Kraai et al. / Chemical Engineering Journal 154 (2009) 384–389 389[12] J. van Gerpen, Biodiesel processing and production, Fuel Process. Technol. 86(2005) 1097–1107.[13] G. Vicente, M. Martinez, J. Aracil, Integrated biodiesel production: a comparisonof different homogeneous catalysts systems, Bioresour. Technol. 92 (2004)297–305.[14] C.C. Akoh, S.W. Chang, G.C. Lee, J.F. Shaw, Enzymatic approach to biodieselproduction, J. Agric. Food Chem. 55 (2007) 8995–9005.[15] M. Di Serio, R. Tesser, L. Pengmei, E. Santacesaria, Heterogeneous catalysts forbiodiesel production, Energy Fuels 22 (2008) 207–217.[16] S. Di Bonaventura, Process and Reactor for Production of Biodiesel, 2008, 2007-BR173.[17] M.A. Dube, A.Y. Tremblay, J. Liu, Biodiesel production using a membrane reactor,Bioresour. Technol. 98 (2006) 639–647.[18] S. Hama, H. Yamaji, T. Fukumizu, T. Numata, S. Tamalampudi, A. Kondo, H. Noda,H. Fukuda, Biodiesel-fuel production in a packed-bed reactor using lipaseproducingRhizopus oryzae cells immobilized within biomass support particles,Biochem. Eng. J. 34 (2007) 273–278.[19] A.P. Harvey, M.R. Mackley, T. Seliger, Process intensification of biodiesel productionby using a continuous oscillatory flow reactor, J. Chem. Technol. Biotechnol.78 (2003) 338–341.[20] J.C. Thompson, B.B. He, Biodiesel production using static mixers, Trans. ASABE50 (2007) 161–165.[21] P. Zhang, B. Wen, Microreactor technology for biodiesel production, in:Abstracts of Papers, 235th ACS National Meeting, New Orleans, LA, UnitedStates, April 6–10, 2008, IEC-097.[22] J. van Gerpen, B. Shanks, R. Pruszko, D. Clements, G. Knothe, Biodiesel ProductionTechnology, 2004, NREL/SR-510-36244.[23] D.H. Meikrantz, L.L. Macaluso, H.W. Sams, C.H. Schardin, A.G. Federici, Centrifugalseparator, 08/779,592 (1998).[24] A. Stanckiewicz, J.A. Moulijn, Re-engineering the Chemical Processing Plant;Process Intensification, Marcel Dekker, Inc., New York, 2004.[25] B.P. Mason, K.E. Price, J.L. Steinbacher, A.R. Bogdan, D.T. McQuade, Greenerapproaches to organic synthesis using microreactor technology, Chem. Rev. 107(2007) 2300–2318.[26] G. Bernstein, D. Grosveno, J.F. Lenc, N.M. Levitz, High-capacity annular centrifugalcontactor, Nucl. Technol. 20 (1973) 200–202.[27] K.T. Klasson, P.A. Taylor, J.F. Walker, S.A. Jones, R.L. Cummins, S.A. Richardson,Modification of a centrifugal separator for in-well oil–water separation, Sep.Sci. Technol. 40 (2005) 453–462.[28] A. Jha, Is it true Kevin Costner is helping to clean up oil spills, Guardian (2003),http://www.guardian.co.uk/film/2003/jul/31/science.features.[29] J.Z. Zhou, W.H. Duan, X.Z. Zhou, C.Q. Zhang, Extraction of hydrocortisone fromthe fermentation liquor with annular centrifugal contactors, Sep. Sci. Technol.41 (2006) 573–581.[30] J.F. Birdwell, J. McFarlane, R.D. Hunt, H. Luo, D.W. DePaoli, D.L. Schuh, S. Dai, Separationof ionic liquid dispersions in centrifugal solvent extraction contactors,Sep. Sci. Technol. 41 (2006) 2205–2223.[31] J.Q. Xu, W.H. Duan, X.Z. Zhou, J.Z. Zhou, Extraction of phenol in wastewaterwith annular centrifugal contactors, J. Hazard. Mater. 131 (2006) 98–102.[32] J. Zhou, W. Duan, X. Zhou, C. Zhang, Application of annular centrifugal contactorsin the extraction flowsheet for producing high purity yttrium, Hydrometallurgy85 (2007) 154–162.[33] J.Q. Zhu, J. Chen, C.Y. Li, W.Y. Fei, Centrifugal extraction for separationof ethylbenzene and octane using 1-butyl-3-methylimidazolium hexafluorophosphateionic liquid as extractant, Sep. Purif. Technol. 56 (2007) 237–240.[34] B. Schuur, J. Floure, A.J. Hallett, J.G.M. Winkelman, J.G. deVries, H.J. Heeres,Continuous chiral separation of amino acid derivatives by enantioselectiveliquid–liquid extraction in centrifugal contactor separators, Org. Process Res.Dev. 12 (2008) 950–955.[35] G.N. Kraai, F. van Zwol, B. Schuur, H.J. Heeres, J.G. de Vries, Two-phase(bio)catalytic reactions in a table-top centrifugal contact separator, Angew.Chem. Int. Ed. 47 (2008) 3905–3908.[36] L.C. Meher, D.V. Sagar, S.N. Naik, Technical aspects of biodiesel productionby transesterification—a review, Renew. Sust. Energy Rev. 10 (2006) 248–268.[37] B. Freedman, E.H. Pryde, T.L. Mounts, Variables affecting the yields of fattyesters from transesterified vegetable-oils, J. Am. Oil Chem. Soc. 61 (1984) 1638–1643.[38] B. Schuur, W.J. Jansma, J.G.M. Winkelman, H.J. Heeres, Determination ofthe interfacial area of a continuous integrated mixer/separator (CINC)using a chemical reaction method, Chem. Eng. Process. 47 (2008) 1484–1491.[39] B. Schuur, G.N. Kraai, J.G.M. Winkelman, H.J. Heeres, Hydrodynamic studies ina centrifugal contactor separator; liquid hold-up, residence time distribution,phase behavior and drop size distribution, Chem. Eng. Process., submitted forpublication.[40] R. Clift, J.R. Grace, M.E. Weber, Bubbles, Drops and Particles, Academic Press,London, 1978.[41] E.E. Michaelides, Particles, Bubbles & Drops: Their Motion, Heat and Mass Transfer,World Scientific, New Jersey, 2006.[42] J.D. Thornton, Science and Practice of Liquid–Liquid Extraction, Clarendon Press,Oxford, 1992.[43] K.R. Westerterp, W.P.M. van Swaaij, A.A.C.M. Beenackers, Chemical ReactorDesigning and Operation, John Wiley & Sons, New York, 1984.[44] R.H. Perry, D.W. Green, J.O. Maloney, Perry’s Chemical Engineers’ Handbook,McGraw-Hill, New York, 1997.[45] J. van Gerpen, B. Shanks, R. Pruszko, D. Clements, G. Knothe, Biodiesel Prod.Technol. (2004), NREL/SR-510-36244.[46] A. Demirbas, Biodiesel fuels from vegetable oils via catalytic and non-catalyticsupercritical alcohol transesterifications and other methods: a survey, EnergyConver. Manage. 44 (2003) 2093–2109.

Renewable Energy 34 (2009) 762–765Contents lists available at ScienceDirectRenewable Energyjournal homepage: www.elsevier.com/locate/reneneBiodiesel production process optimization and characterization to assess thesuitability of the product for varied environmental conditionsT. Eevera * , K. Rajendran, S. SaradhaDepartment of Biotechnology, Periyar Maniammai University, Periyar Nagar, Vallam, Thanjavur, Tamilnadu 613 403, IndiaarticleinfoabstractArticle history:Received 10 September 2007Accepted 17 April 2008Available online 17 June 2008Keywords:TransesterificationBiodieselEdible and non-edible oilsSaponification valueIodine valueCetane numberIn this study, both edible (coconut oil, palm oil, groundnut oil, and rice bran oil) and non-edible oils(pongamia, neem and cotton seed oil) were used to optimize the biodiesel production process variableslike catalyst concentration, amount of methanol required for reaction, reaction time and reaction temperature.The fuel properties like specific gravity, moisture content, refractive index, acid value, iodinenumber, saponification value and peroxide value were estimated. Based on the cetane number and iodinevalue, the methyl esters obtained from palm and coconut oils were not suitable to use as biodiesel in coldweather conditions, but for hot climate condition biodiesel obtained from the remaining oil sources issuitable.Ó 2008 Elsevier Ltd. All rights reserved.1. IntroductionThe increased use of diesel fuel resulted in depletion of its fossilreserves. This triggers for many initiatives to search for alternatefuel, which can supplement or replace such fossil fuel. In recentyears, research has been directed to explore plant-based fuels andplant oils and fats as fuels have bright future [1]. The most commonthat is being developed and used at present is biodiesel, which isfatty acid methyl esters of seed oils and fats and have already beenfound suitable for use as fuel in diesel engine. Biodiesel is found tobe environmentally safe, non-toxic and biodegradable [2].The raw material being exploited commercially by the developedcountries constitutes the edible fatty oils derived fromrapeseed, soybean, palm, sunflower, coconut, linseed, etc. [3]. Useof such edible oil to produce biodiesel in India is not feasible in viewof a big gap in demand and supply of such oils in the country. Increasedpressure to augment production of edible oil has also putlimitation on the use of these oils for production of biodiesel. Undersuch conditions, those crops that produce non-edible oil inappreciable quantities can be grown in large scale in non-croppedmarginal lands and wastelands only considered for biodieselproduction [4].Long list of trees, shrubs and herbs is available plenty in India,which can be exploited for fuel production. In this study we haveutilized both edible and non-edible oils for the biodiesel production* Corresponding author.E-mail address: teevera2000@yahoo.com (T. Eevera).process optimization and fuel property characterization to assessthe suitability of the different methyl esters to the varied environmentalconditions.2. Materials and methodsEdible (coconut oil, palm oil, groundnut oil, rice bran oil, andgingelly oil) and non-edible (pongamia, cotton seed oil and neemoil) oils were used in this experiment.All the oils were first filtered by cloth mainly to remove the dirtand other inert materials from the oil and then placed in a conicalflask equipped with magnetic stirrer, thermometer and condenser.Under agitation the raw oil was heated up to nearer to the boilingpoint to remove the water contaminant present in the oil. After thatoil is allowed to cool down under room temperature, and thetreated oil alone was taken for biodiesel production purpose. Again,under agitation, the above treated oil was heated up to a desiredtemperature on a hot plate. A fixed amount of freshly preparedsodium hydroxide–methanol solution was added into the oils,taking this moment as the starting time of the reaction. When thereaction reached the preset reaction time, heating and stirring werestopped. The products of reaction were allowed to settle overnight.During settling two distinct liquid phases were formed: crude esterphase at the top and glycerol phase at the bottom. The crude esterphase separated from the bottom glycerol phase was then washedby cold or warm de-ionized water several times until the washedwater became clear. The excess methanol and water in ester phasewere then removed by evaporation under atmospheric condition.0960-1481/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.doi:10.1016/j.renene.2008.04.006

T. Eevera et al. / Renewable Energy 34 (2009) 762–765 763After that weight of the ester was taken for product yieldcalculation.The reaction was investigated step by step. The optimal value ofeach parameter involved in the process was determined whilethe rest of the parameters were kept constant. After each optimalvalue was attained, this value was adopted for the optimization ofthe next parameter.Physical and chemical properties of methyl esters were estimated[5,6] under laboratory conditions. The cetane number (CN)and higher heating values (HHVs) were calculated [4,6] from thefollowing equation by using the estimated saponification value (SV)and iodine value (IV).CN ¼ 46:3 þ 5458=SV 0:225IV (1)HHV ¼ 49:43 ½0:041ðSVÞþ0:015ðIVÞŠ (2)Methyl esters yield (Wt.%)10.90.83. Result and discussion3.1. Effect of catalyst concentrationPongamiaNeemGroundnutGingellyPalmCotton seedRice branCoconutThe effect of sodium hydroxide concentration on the transesterificationof the edible and non-edible oils was investigatedwith its concentration varying from 0.5 to 2.5 wt.% (based on theweight of raw oil). The operation conditions during the whole reactionprocess were fixed at the optimal level: reaction temperatureof 55 C, reaction time of 90 min. and 180 and 210 ml of methanolfor edible and non-edible oils, respectively.Experimental results showed changes in ester yield contentwith varied catalyst concentration. As the sodium hydroxideconcentration increased, the conversion of triglyceride as well asthe ester content also increased. Insufficient amount of sodiumhydroxide resulted in incomplete conversion of triglycerides intothe esters as indicated from its lower ester content. The estercontent reached an optimal value when the sodium hydroxideconcentration reached 1.5 wt.%, and further increase in catalystconcentration in all the cases, ester production amount decreasedas shown in Fig. 1. Large amount of soap was observed in excessamount of sodium hydroxide added experiments. This is becauseaddition of excess alkaline catalyst caused more triglycerides’participation in the saponification reaction with sodium hydroxide,resulting in the production of more amount of soap andreduction of the ester yield [7].0.70.5 1 1.5 22.5Catalyst concentration (Wt.%)Fig. 1. Effect of catalyst concentration on methyl esters’ yield.3.3. Effects of methanol amountThe effect of alcohol amount on yield of the transesterificationexperiments was conducted with different amounts of methanolto oil in the range of 120–240 ml. The optimized catalyst concentrationand reaction time as obtained in the above sectionswere adopted. Maximum ester content was obtained at a methanolamount of 180 ml for edible oil and 210 ml for non-edible oils.With further increase in the methanol to oil amount above210 ml, a very little effect on the biodiesel yield was observed13.2. Effect of reaction timeThe reaction time of the transesterification reaction conductedat 55 C was optimized with the highest achievable mixing degree,an excess amount of alcohol (220 ml per liter of oil) and optimalsodium hydroxide concentration of 1.5 wt.% for all the oils.The changes in product composition with reaction time duringthe transesterification of the oils and the distribution of variouscomponents in the reaction system can be clearly seen. When thereaction time reached 90 min, no triglyceride was left in the productmixture, indicating complete conversion. In this experiment, glycerolstarted to separate within 15 min. The ester content increased withreaction time from 15 min onwards and reached a maximum ata reaction time of 90 min at 55 C, and then remained relativelyconstant with increasing further the reaction time (Fig. 2). The resultsindicated that an extension of the reaction time from 90 to 150 minhad no significant effect on the conversion of triglycerides but leadsto a reduction in the product yield. This is because longer reactionenhanced the hydrolysis of esters (reverse reaction of transesterification),resulted in loss of esters as well as causing more fattyacids to form soap.Methyl ester yield (Wt.%)0.90.80.7PongamiaNeemGroundnutGingelly30 60 90 120 150Reaction time (min.)PalmCotton seedRice branCoconutFig. 2. Effect of reaction time on methyl esters’ yield.

764T. Eevera et al. / Renewable Energy 34 (2009) 762–765(Fig. 3). Moreover, it was observed that for high alcohol amountadded the set up required longer time for the subsequent separationstage since separation of the ester layer from the waterlayer becomes more difficult with the addition of a large amountof methanol. This is due to the fact that methanol, with one polarhydroxyl group, can work as an emulsifier that enhances emulsion.Therefore, increasing the alcohol amount to oil is anotherimportant parameter affecting the biodiesel yield and biodieselpurity, apart from catalyst concentration and reaction time. Thisresult is in line with the report of many investigations based onneat vegetable oils [7–9]. However the non-edible like pongamiaand neem require 210 ml of methanol to give maximum esteryield, possibly due to higher viscosity of non-edible oil than edibleoils. In this case more amount of methanol is required toincrease the solubility of the oil in the methanol. This leads tomaximize the ester yield at high methanol concentration level.However, when compared to edible oil ester content yield wasminimal in non-edible oil but glycerol yield was found to be morein non-edible oil when compared to edible oil.Methyl esters yield (Wt.%)10.90.8PongamiaNeemGroundnutGingellyPalmCotton seedRice branCoconut3.4. Effect of reaction temperatureTo study the effect of reaction temperature on methyl esters’formation, the transesterification reaction was carried out underthe optimal conditions obtained in the previous section (i.e. 180 and210 ml of methanol for edible and non-edible oils, respectively, and1.5 wt.% sodium hydroxide). The experiments were conducted attemperature ranging from 40 to 60 Cat5 C interval. The effect ofreaction time was shown in Fig. 4.Experimental results showed that the transesterification reactioncould proceed within the temperature range studied but thereaction time to complete the reaction varied significantly withreaction temperature. It can be seen that a high product yield couldbe achieved at 50 C. With the temperature increase above 55 C,the product yield started to decrease. The reason for this is thathigher temperature accelerates the side saponification reaction oftriglycerides.Methyl ester yield (Wt.%)10.90.80.7PongamiaNeemGroundnutGingelly120 150 180 210 240Alcohol amount (ml)PalmCotton seedRice branCoconutFig. 3. Effect of methanol concentration on methyl esters’ yield.0.740 45 50 55 60Reaction temperature (degree C)Fig. 4. Effect of reaction temperature on methyl esters’ yield.3.5. Methyl ester chemical and physical propertyFuel properties like specific gravity, moisture content, refractiveindex, acid value (AV), iodine value (IV), saponification value (SV)and peroxide value (PV) were estimated. The cetane number (CN)and higher heating values (HHVs) of methyl ester were calculatedbased on the estimated SV and IV (Tables 1 and 2).The estimated SV and IV range from 188 to 259 and 9 to 113,respectively. The CN value among the oils varied from 49.91 to65.34; the HHV ranges in between 38.68 and 40.33 among thedifferent methyl esters.CN is the ability of fuel to ignite quickly after being injected.Better ignition quality of the fuel is always associated with higherCN value. This is one of the important parameter, which is consideredduring the selection of methyl esters for use as biodiesel.For this different countries/organization have specified differentminimum values. Biodiesel standards of USA (ASTM D 6751),Germany (DIN V51606) and European Organization (EN 14214)have set this value as 47, 49, and 51, respectively [10]. In our experimentall the different methyl esters have CN value of higherthan 51 except methyl esters obtained from gingelly (49.91) andgroundnut (50.51) oils.Another important criterion for selection of methyl esters is itsdegree of unsaturation, which is measured as IV. To an extent, theTable 1Physical property of methyl estersS. No. Biodiesel SpecificgravityViscosity(10 6 Ns/m 2 )Moisture content(wt.%)Refractiveindex1 Pongamia 0.942 34.66 0.16 1.44472 Neem 0.942 26.286 0.35 1.44563 Palm 0.870 17.81 0.19 1.44184 Cotton seed 0.874 13.27 0.52 1.44285 Rice bran 0.878 18.95 0.72 1.44876 Gingelly 0.873 13.3 0.24 1.44667 Groundnut 0.878 16.62 0.45 1.44958 Coconut 0.868 12 0.82 1.4271

T. Eevera et al. / Renewable Energy 34 (2009) 762–765 765Table 2Chemical property of methyl estersS. No. Biodiesel AV SV IV PV Moisture content (wt.%) HHV (kJ/g) CN1 Pongamia 0.5 194 84 15.78 0.16 40.216 55.532 Neem 0.6 189 99 16.08 0.35 40.196 52.93 Palm 0.2 201 57 11.31 0.19 40.334 60.624 Cotton seed 0.3 199 76 12.52 0.52 40.131 56.635 Rice bran 0.4 188 100 15.02 0.45 40.222 52.836 Gingelly 0.1 188 113 14.66 0.72 40.027 49.9057 Groundnut 0.2 187 111 19.21 0.24 39.048 50.518 Coconut 0.1 259 9 1.52 0.82 38.676 65.34presence of unsaturated fatty acid component in methyl esters isrequired as it restricts the methyl esters from solidification. However,with higher degree of unsaturation, methyl esters are notsuitable for biodiesel as the unsaturated molecules react withatmospheric oxygen and are converted to peroxide, cross-linking atthe unsaturation site can occur and the material may get polymerizedinto a plastic like body. At high temperature, commonlyfound in an internal combustion engine, the process can get acceleratedand the engine can quickly become gummed up with thepolymerized methyl esters. To avoid this kind of situation, biodieselstandards have set a minimum limit of IV in their specifications. Allthe eight species, which qualify the specification of CN, also meetthe specification of IV. All of them have IV less than 115, the lowestmaximum limit among the three biodiesel standards set byEuropean Organization (EN 14214 [10b]).Generally, methyl esters with higher CN are favored for use asbiodiesel. However, with increase of CN, IV decreases which meansdegree of unsaturation decreases. This situation will lead to thesolidification of methyl esters at higher temperature. To avoid thissituation, the upper limit of CN (65) has been specified in US biodieselstandards (ASTM D 6751-99 & ASTM PS 121-99) [10c,10d].Among the eight methyl esters, which already met the specificationof CN and IV of biodiesel standards, two of them like palm andcoconut esters have low IV (

Fuel Processing Technology 90 (2009) 770–777Contents lists available at ScienceDirectFuel Processing Technologyjournal homepage: www.elsevier.com/locate/fuprocActivity of solid catalysts for biodiesel production: A reviewMasoud Zabeti, Wan Mohd Ashri Wan Daud ⁎, Mohamed Kheireddine ArouaChemical Engineering Department, Faculty of Engineering, University Malaya, 50603, Kuala Lumpur, MalaysiaarticleinfoabstractArticle history:Received 31 October 2008Received in revised form 5 March 2009Accepted 21 March 2009Keywords:Heterogeneous catalystCatalyst activityCatalyst supportBiodieselTransesterificationHeterogeneous catalysts are promising for the transesterification reaction of vegetable oils to producebiodiesel. Unlike homogeneous, heterogeneous catalysts are environmentally benign and could be operatedin continuous processes. Moreover they can be reused and regenerated. However a high molar ratio ofalcohol to oil, large amount of catalyst and high temperature and pressure are required when utilizingheterogeneous catalyst to produce biodiesel. In this paper, the catalytic activity of several solid base and acidcatalysts, particularly metal oxides and supported metal oxides, was reviewed. Solid acid catalysts were ableto do transesterification and esterification reactions simultaneously and convert oils with high amount of FFA(Free Fatty Acids). However, the reaction rate in the presence of solid base catalysts was faster. The catalystefficiency depended on several factors such as specific surface area, pore size, pore volume and active siteconcentration.© 2009 Elsevier B.V. All rights reserved.1. IntroductionFinding an alternative fuel has attracted considerable attraction inrecent years due to limitation of traditional fossil resources and increasingof crude oil prices as well as concern over greenhouse gasemissions. Biodiesel, a biodegradable and renewable form of energy,consisting of mono alkyl esters of fatty acids derived from sourcessuch as vegetable oils and animal fats, has cardinal potential asalternative fuels [1]. In 1911, vegetable oils were first used as fuel for adiesel engine invented by Rudolph Diesel and it was concluded thatthe engines could be fed with vegetable oils [2]. Recently, there hasbeen renewed interest on vegetable oils and animal fats to producebiodiesel since it offers many advantages such as high flash point, highcetane number, low viscosity, high lubricity, biodegradability, environmentallyfriendly due to less carbon monoxide emissions, as well asfewer emission profiles compared to conventional fossil fuels [3]. Itcan be also used in conventional compression-ignition engines withoutthe need for engine modifications [4]. In addition, the fuel eitherused as pure or blend can reduce the particulate emissions fromengine [5].Different varieties of vegetable oils such as, canola [6], palm [7],jatropha [8], palm kernel [9], sunflower [10], and coconut [11] havebeen studied as precursors for biodiesel production. The main concernabout biodiesel production is the high price of vegetable oils comparedto that of fossil based diesel fuel. As a result, in some countries,⁎ Corresponding author. Tel.: +60 3 79675297; fax: +60 3 79675319.E-mail addresses: masoud.zabeti@gmail.com (M. Zabeti), ashri@um.edu.my(W.M.A. Wan Daud), mk_aroua@um.edu.my (M.K. Aroua).non-edible oils such as jatropha or waste cooking oils [12] are preferreddue to their low price. Blending of biodiesel with diesel fuel isanother option to mitigate the high price of biodiesel and in this case,the blended product is designated as Bxx (B30, 30% biodiesel in 70%diesel fuel) [13].There are several methods to produce biofuels from renewablesources and among the earliest is based on pyrolysis processwhereby heavy molecules such as vegetable oils or animal fats areconverted to smaller one by means of heat or heat and catalyst. Thefuel obtained through this process was called “diesel-like fuels” sincethere were variety of components such as olefins and paraffinwhichare similar to the petroleum-based diesel. However, this processrequires high temperatures between 300 and 500 °C and productscharacterization is difficult due to the variety of reaction products[14]. Limaetal.[15] pyrolized soybean oil using zeolite as the catalyst,at reaction temperature of 400 °C in a nitrogen flow andolefins, paraffin, carboxylic acids and aldehydes were observed in theproducts. Ensöz et al. [16] investigated the influence of particle sizeon the pyrolysis of rapeseed by varying the particle size of rapeseedin the range of 0.224–1.8 mm and found that the yields of productsare largely independent of particle size. More than 30 compoundswere detected from pyrolysis of Macauba fruit and was concludedthat the amount of main products decreases by increasing the pyrolysistemperature [17].Transesterification is one of the most commercially useable methodsto produce biodiesel and the process involves a reaction between ester(here triglyceride) and alcohol to form new ester and alcohol. Differenttypes of alcohols such as, methanol, ethanol, propanol and butanol havebeen used. However, methanol and ethanol are the most widely used,particularly methanol owing to its low price and availability [18]. In0378-3820/$ – see front matter © 2009 Elsevier B.V. All rights reserved.doi:10.1016/j.fuproc.2009.03.010

M. Zabeti et al. / Fuel Processing Technology 90 (2009) 770–777771the transesterification of triglyceride to fatty acid alkyl esters threereversible reactions take place consecutively in which diglycerides andmonoglycerides are major intermediate products (Fig. 1) [19].Areviewby Ma and Hanna [20] summarized the parameters that significantlyaffect on the rate of transesterification reaction which include reactiontemperature, alcohol to oil molar ratio, catalyst concentration and typeof catalyst. Transesterification can be performed at different temperatures,and the methyl ester yields increase by rising the reactiontemperature. However, if the temperature reaches to the boiling point ofmethanol, a lot of methanol's bubbles are formed hence inhibit the masstransfer on the phase's interface [21]. In transesterification reaction thestoichiometric ratio of alcohol to oil is three moles of alcohol to one moleof oil. Therefore by using alcohol in excess it favors the formation ofmethyl esters. The alcohol oil molar ratio is dependent on the type ofcatalyst. Using KOH as a homogeneous base catalyst molar ratio ofalcohol to oil required is approximately 6:1. Nevertheless, when acidcatalysts are used more alcohol/oil molar ratio is required [22].Transesterification reaction can be either carried out via non-catalyticor catalytic processes. Base catalysts, acid catalysts and enzymes arethree categories of catalysts which have been studied for biodieselproduction. Enzymatic base production of biodiesel has attracted manyattention in recent years since enzymes tolerate free fatty acid and watercontents in the oil to avoid soap formation and thus purification ofbiodiesel and glycerol is easier [23,24].However,enzymesareexpensiveto be used in a commercial production of biodiesel.Non-catalytic transesterification reaction is slow and normallyneeds high pressures and temperatures to be completed. Kusdianaand Saka [25] subjected rapeseed oil into transesterification process inthe absence of catalyst and in the supercritical methanol and foundthat the amount of water in the reaction does not affect the conversionof oil. Conversely, the presence of certain amount of water increasesthe formation of methyl esters and esterification of free fatty acidstakes place simultaneously in one stage. Although the products can beeasily separated since there is no base or acid catalyst, the supercriticalprocess needs to be carried out at high temperatures (250–400 °C)and pressure (35–60 MPa).Significant amounts of work have been carried out on homogeneousacid and base catalysis transesterification of vegetable oils.Sulphuric acid and hydrochloric acid are normally used as acidcatalysts especially when the oil contains high amount of free fattyacids and water since these catalysts are capable to handle esterificationand transesterification of triglyceride simultaneously. Nonetheless,the process requires a high molar ratio of alcohol to oil andlong reaction time [26]. Besides that, under these conditions all theequipment need to withstand against corrosion due to an acidityenvironment [27]. Sodium hydroxide and potassium hydroxide areexamples of homogeneous base catalysts which are usually used inindustry [28,29]. The transesterification can be performed in a shorterreaction time and relatively modest operation conditions [4]. In spiteof this, the water and free fatty acids content in the oil need to be verylow because of the probability of soap formation which can consumecatalyst thus reducing the methyl ester yields. One of the majordisadvantages of homogeneous catalysts is that they cannot be reusedor regenerated, because the catalyst is consumed in the reaction andseparation of catalyst from products is difficult and requires moreequipment which could result in higher production costs [26]. Besides,Fig. 1. Transesterification reaction of vegetable oil with alcohol to esters and glycerol.the process is not environmentally friendly because a large amount ofwaste water is produced in the separation step [30].Based on the above premises, developing a new solid catalystseems to be an appropriate solution to overcome problems associatedwith homogeneous catalysts. Metal hydroxides [6], metal complexes[31], metal oxides such as calcium oxide [32], magnesium oxide [33],zirconium oxide [34] and supported catalysts [35] have beeninvestigated as solid catalysts. The catalysts are not consumed ordissolved in the reaction and therefore can be easily separated fromthe products. As a result, the products do not contain impurities ofthe catalyst and the cost of final separation could be reduced. Thecatalysts can also be readily regenerated and reused and it is moreenvironmentally benign because there is no need for acid or watertreatment in the separation step [36]. Nevertheless, one of the majorproblems associated with heterogeneous catalysts is the formation ofthree phases with alcohol and oil which leads to diffusion limitationsthus lowering the rate of the reaction [37]. One way of overcomingmass transfer problem in heterogeneous catalysts is using certainamount of co-solvent to promote miscibility of oil and methanol andaccordingly accelerate the reaction rate. Tetrahydrofuran (THF),dimethyl sulfoxide (DMSO), n-hexane and ethanol were used morefrequently as co-solvent in transesterification of vegetable oils withmethanol and solid catalysts. Gryglewicz [38] used CaO as a solid basecatalyst for transesterification of rapeseed oil with methanol andafter 170 min of reaction time methyl ester yields of 93% were obtained.However, by adding certain amount of THF into rapeseed oil/methanol mixture the same yields of 93% were observed after 120 minof reaction time. Another way to promote mass transfer problemsassociated with heterogeneous catalysts is using structure promotersor catalyst supports which can provide more specific surface area andpores for active species where they can anchor and react with largetriglyceride molecules.Studies have been carried out to synthesize and develop new solidcatalysts for transesterification reaction in order to overcomedisadvantages of use of homogeneous catalyst and more likely toreduce the cost of biodiesel production. Bournay et al. [39] invented anew commercial continuous biodiesel production process, where asolid catalyst consists of zinc oxide and alumina was used and resultsshowed that the process does not require any post treatment toremove the catalyst from biodiesel and methyl ester yields, close totheoretical value, were achieved at high pressure and temperature.Moreover, glycerol obtained through this process had a purity ofapproximately 98%.Applications of heterogeneous catalysts for biodiesel productionhave been reviewed by some researchers such as reviews by Di Serioet al. [40] and Lotero et al. [41] and the mechanism of solid acid andbase catalysts and their kinetics for transesterification reaction weremainly focused. However, in this article the application of differenttypes of solid catalysts, particularly metal oxides and supportedmetal oxides, in transesterification of vegetable oils to producebiodiesel was reviewed in order to find the most suitable base andacid catalysts, supports, and the most effective parameters on thecatalyst activity.2. Catalysts2.1. Metal oxidesMany metal oxides include alkali earth metal oxides and transitionmetal oxides have been studied for transesterification process of oils.The structure of metal oxides is made up of positive metal ions (cations)which possess Lewis acid and negative oxygen ions (anions) whichpossess Bronsted base. In methanolysis of oils, it provides sufficientadsorptive sites for methanol, in which the O–H bonds readily break intomethoxide anions and hydrogen cations (Fig. 2). Thus methoxide anionsthen react with triglyceride molecules to form corresponding fatty acid

772 M. Zabeti et al. / Fuel Processing Technology 90 (2009) 770–777that the calcium oxide catalytic performance is quite weak at lowtemperatures since only 5% methyl ester yields were obtained at 60 °Cafter 3 h.methyl esters [42]. There are several metal oxides which have beenstudied and among them are discussed below:2.1.1. Alkali earth metal oxidesFig. 2. Surface structure of metal oxides.2.1.1.1. Magnesium oxide. Magnesium oxide which is produced bydirect heating of magnesium carbonate or magnesium hydroxide hasthe weakest basic strength and solubility in methanol among group ΙΙoxides and has been rarely used for biodiesel production. Nanomagnesium oxide has been used in transesterification of soybean oiland yields of 99% were obtained in 10 min under supercriticaltemperature of 523 °C and high pressure of 24 MPa [33]. The resultsshowed that this catalyst has more activity at higher pressures andtemperatures.2.1.1.2. Calcium oxide. Among alkali earth metal oxides, calciumoxide has attracted many attentions for transesterification reactionsince it possess relatively high basic strength and less environmentalimpacts due to its low solubility in methanol and can be synthesizedfrom cheap sources like limestone and calcium hydroxide. Calciumoxide has also been used as a catalyst for reactions involved in organiccompounds such as isomerization of 5-vinylbicyclo [2.2.1] hept-2-ene(VBH) to 5-ethylidenebicyclo [2.2.1] hept-2-ene (ENB) [43] andsynthesis of 1,3-dialkylurea from ethylene carbonate and amine [44]and showed a high activity for organic reactions. In case of biodieselproduction, Granados et al. [32] used the activated calcium oxide as asolid base catalyst in the methanolysis of sunflower oil to investigatethe role of water and carbon dioxide on the deterioration of thecatalytic performance by contact with air for different periods of time.The study showed that CaO is rapidly hydrated and carbonated in air.No calcium oxide peaks were detected in the samples exposed to airfor more than 20 days. This means that active surface sites of CaO werepoisoned with CO 2 and covered with H 2 O. As a result, in order to avoidreduction of CaO catalytic activity, the catalyst was thermally treatedat 700 °C in order to chemically desorb CO 2 before being used in thereaction. After 100 min of reaction time, 94% conversion was obtainedat 60 °C with alcohol/oil molar ratio of 13:1 and catalyst content of3 wt.% based on the weight of oil. Leaching of active species wasobserved in the reaction media when the catalyst was activated athigh temperature. However, the amount of leaching did not result incatalyst activity reduction and the catalyst was reusable for 8 cycles.It was also found that small amounts of water can improvecatalytic activity of CaO and biodiesel yields because in the presence ofwater, O 2− on the surface of the catalyst extracts H + from watermolecules to form OH − which subsequently extracts H + of methanolto form methoxide anions, which are the real catalyst of transesterificationreaction [45]. By adding 2.03 wt.% water into the reactionmedium of 12:1 alcohol/oil molar ratio and 8 wt.% catalysts, themethyl ester yields exceeded 95% within 3 h of reaction time compareto 80% under anhydrous conditions. No leaching of the active speciesinto the reaction media was observed and the activity of the catalystwas stable after 20 cycles of the reaction.On the other hand, Demirbas [46] described the effect of thesupercritical conditions on the catalytic transesterification of sunfloweroil in the presence of 3 wt.% CaO 60–120 mesh, 40:1 alcohol/oilmolar ratio, at 252 °C and pressure of 24 MPa. And found that after26 min of reaction time 98.9% methyl ester yields were obtained.Unlike Granados et al. [32] and Liu et al. [45], Demirbas [46] believes2.1.1.3. Strontium oxide. Not much works have been carried out usingthis catalyst. Strontium oxide is a highly active metal oxide and seemsto be soluble in the reaction medium. Transesterification of soybeanoil with SrO as a solid base catalyst showed that although the specificsurface area of the catalyst is as low as 1.05 m 2 /g, 90% yields of methylesters were achieved after 30 min of reaction time at 65 °C withalcohol/oil molar ratio of 12 and 3 wt.% catalyst [21]. The catalyst wasstable even after 10 reaction cycles.2.1.2. Transition metal oxidesZirconium oxide, titanium oxide and zinc oxide are among thetransition metal oxides that have attracted attention for biodieselproduction due to their acidic properties. Zirconia has been used invariety of reactions, for example alkylation of isobutane [47] andisomerization of N-butane [48] and showed high activity correlatedwith acidity. Titanium oxide has been also used in reactions, includingdehydration [42]. There have been several reports on the usage ofzirconia as a solid acid catalyst for transesterification of different oilsrather than other transition metal oxides due to its strong acidity andwas found that the acidity is promoted when the surface of thesemetal oxides contains anions like sulfate and tungstate. Jitputti et al.[34] assessed the catalytic performance of zinc oxide and zirconiumoxide as solid acid catalysts in transesterification reaction of palmkernel oil at supercritical methanol and found that after 1 h of reactiontime, using 3 wt.% catalyst and 6:1 molar ratio of alcohol/oil, 86.1%methyl ester yields were obtained for zinc oxide while only 64.5% forzirconium oxide. However, using sulfated zirconia, the yieldsconsiderably increased to 90.3% which might be due to high acidicstrength of sulfate anions on the surface of zirconia.Catalytic activity of ZrO 2 /SO 4 2− and ZrO 2 /WO 3 2− as super acid catalystsin the methanolysis of triacetin was compared by López et al.[49] and the results are shown in Table 1. The results indicate that thespecific surface area and active site concentration play an importantrole in the catalyst activity. Sulfated zirconia showed more activitytoward conversion of triacetin than tungstated zirconia owing to itshigher specific surface area and active phase concentration under thesame conditions of 60 °C and 8 h of reaction time.Comparing activity of TiO 2 /SO 4 2− and ZrO 2 /SO 4 2− as solid strongacid catalysts in the transesterification of cotton seed oil to methylesters, both catalysts showed high activity. However, higher methylester yields were obtained for sulfated titanium oxide due to its higherspecific surface area of 99.5 m 2 /g compare to 91.5 m 2 /g for sulfatedzirconia [50]. It was found that after the introduction of sulfate anionsnew Bronsted acid sites were formed on the catalyst surface. Byintroducing 2 wt.% catalyst and 12:1 alcohol/oil molar ratio, after 8 hof reaction time the methyl ester yields in the presence of TiO 2 /SO 42−and ZrO 2 /SO 4 2− were 90% and 85% respectively.The activity of zirconium oxide functionalized with tungsten oxidewas also studied by López et al. [51] to investigate the effect ofcalcination temperatures on the catalytic properties of ZrO 2 /WO 3 2− asa solid strong acid for transesterification of triacetin. The most activecatalyst was obtained after calcination of 800 °C with the formation ofBronsted acid sites concentration of 161 μmol/g. The catalyst wasactive for both transesterification and esterification reactionsalthough the reaction time was relatively long.Table 1The effect of active site concentration and specific surface area on the catalyst activity.Catalyst BET (m 2 /g) Active site concentration (µmol/g) Conversion (%)2−ZrO 2 /SO 4 134 94 572−ZrO 2 /WO 3 89 54 10

M. Zabeti et al. / Fuel Processing Technology 90 (2009) 770–777773On the other hand, Ramu et al. [52] evaluated the effect of calcinationtemperature on the catalytic performance of ZrO 2 /WO 3 2− aswell as the effect of amount of WO 3 loading on the zirconium oxidefor the esterification of palmitic acid with methanol and discoveredthat the catalyst which was calcined at 500 °C had the most activity.This activity was attributed to the formation of tetragonal phase ofZrO 2 . Beyond 500 °C the tetragonal phase transferred to the monoclinicphase which caused a decrease in the activity from 98%conversion of palmitic acid to 8% for the catalyst which was calcinedat 900 °C. With 5 wt.% loading of WO 3 the acid site concentrationof 1.04 mmol/g was achieved. The acidity which is correlated to theactivity decreased by increasing the amount of WO 3 as a result ofexcess coverage of WO 3 species on ZrO 2 .Furuta et al. [53] studied conversion of soybean oil over tungstatedzirconia and found that the catalyst activity remained stable upto 100 h of use. After reaction time of 8 h at 300 °C, 94% of oil wasconverted with alcohol/oil molar ratio of 40:1. He also comparedactivity of sulfated tin oxide and zirconium oxide under the samereaction conditions and found that WZO showed the highest activityfor conversion of soybean oil whereas 70% and 80% conversions werereached over SZO and STO respectively.Vanadium phosphate solid acid catalyst was also studied fortransesterification reaction [54]. Although the specific surface area ofthe catalyst was low (2–4 m 2 /g), the catalyst was active for thetransesterification of soybean oil with methanol since the methylester yields reached 80% at 150 °C after 1 h of reaction. Furthermore,only 0.2 wt.% of the catalyst and 1:1of alcohol/oil molar ratio wasused in the reaction. A slow deactivation of the catalyst was observedat high reaction temperatures due to the reduction of vanadiumspecies form V 5+ to V 3+ with methanol. However, the catalyst couldbe readily regenerated.Fe–Zn double metal cyanide complex has been tested as a solid acidcatalyst for methanolysis of sunflower oil [55]. The catalyst was preparedby mixing three solutions; aqueous solution of K 4 Fe (CN) 6·3H 2 O, asolution of ZnCl 2 in mixture of distilled water and tert-butanol and asolution of tri-block copolymer, poly ethylene glycol-block-poly propyleneglycol-block-poly ethylene glycol, in mixture of water and tertbutanolwith subsequent calcination at 180 °C. The specific surface areaof the catalyst was 51.6 m 2 /g. When the transesterification reaction wasperformed at 170 °C, with oil/alcohol molar ratio of 1:15 and 3 wt.% ofcatalyst, the oil conversion reached 97% after 8 h of reaction. The catalystactivity was attributed to the Lewis acid active sites of probably Zn 2+ onthe surface of catalyst. Moreover, the catalyst converted the oil with upto 20% of water content which implies the surface hydrophobicity of thecatalyst. The activity of catalyst was successfully tested for esterificationof high amount of FFA in the oil. In addition, the catalyst was stable aftermany cycles since no significant loss of activity was detected.Siano et al. [56] patented the application of Mg–Al hydrotalcite forbiodiesel production from soybean oil and found that the activity ofthe catalyst was affected by the ratio of Mg/Al and the most activecatalyst was obtained with Mg/Al ratio of 3 to 8. The catalyst wassuitable for oils with high amount of water content (10,000 ppm) anda conversion of 92% obtained after 1 h of reaction time using catalystcontent of 5 wt.% and alcohol/oil weight ratio of 0.45 at reactiontemperature of 180 °C.Mg–Al hydrotalcite has been reported as solid base catalyst for thetransesterification of soybean oil with methanol to produce biodiesel[57]. The hydrotalcite was synthesized by co-precipitation method ofaqueous solution of Mg (NO 3 ) 2·6H 2 O, Al (NO 3 ) 3·9H 2 O, NaOH andNa 2 CO 3 under vigorous stirring. The precipitate was then calcined atdifferent temperatures to examine the effect of calcination temperatureson the catalyst performance. The results indicated that byenhancing calcination temperature the catalyst activity and basicityimproved since the optimum catalyst for the reaction obtained aftercalcination at 500 °C. However, this catalyst represented quite lowactivity with only 67% conversion of oil after 9 h of reaction time withalcohol/oil molar ratio of 15:1 and 7.5 wt.% of catalyst content attemperature of 65 °C. On the other hand, Georgogianni et al. [58]compared the catalyst activities of Mg–Al hydrotalcite base catalystsand MgO supported on MCM-41 for transesterification of soybeanfrying oil with methanol and after 24 h of reaction time and underidentical operation conditions conversion of oil achieved 97% and 85%respectively. The greater activity of the hydrotalcite was correlated toits higher basicity although the specific surface area of Mg-MCM 41(1289 m 2 /g) was much higher than the specific surface area of Mg–Alhydrotalcite (82 m 2 /g).Di Serio et al. [59] also investigated the catalytic activity of Mg–Alhydrotalcite for biodiesel production from soybean oil. The catalystwas prepared by co-mixing of 0.8 M solution of Mg (NO 3 ) 2 , 0.2solution of Al (NO 3 ) 3 , NaOH and Na 2 CO 3 followed by calcination at500 °C and based on the results of TPD analysis four different basicsites including weak basic sites related to OH − surface groups,medium basic sites related to oxygen atoms in both Mg 2+ –O 2− andAl 3+ –O 2− , strong basic sites and super-basic sites attributed toisolated O 2− anions were found to be responsible for the basicityand accordingly the activity of the catalyst. Moreover, the activity ofthe catalyst was also correlated to porosity of the catalyst since Mg–Alhydrotalcite with larger pore size showed more activity than MgOwith micropores structure towards transesterification reaction. TheMg–Al hydrotalcite showed high performance at higher reactiontemperatures and when the reaction was performed at 215–225 °Cusing catalyst content of 1 wt.% and methanol/oil weight ratio of 0.45the biodiesel yields of 94% were obtained.3. Catalyst supportsOne of the ways to minimize the mass transfer limitation for heterogeneouscatalysts in liquid phase reactions is using catalyst supports.Supports can provide higher surface area through the existenceof pores where metal particles can be anchored [42]. Supports suchas: alumina, silica, zinc oxide and zirconium oxide have been used inbiodiesel production.3.1. AluminaAluminum oxide exists in forms of porous γ-alumina and ή-aluminaand nonporous crystalline α-alumina has been widely used as a supportin catalysis processes, such as polymerization, reforming, steam reforming,dehydration and hydrogenation owing to its extremely thermaland mechanical stability, high specific surface area, large pore sizeand pore volume and. This also includes in the synthesis of biodiesel.Catalytic activity of titanium oxide–zinc oxide supported by alumina(ATZ) and titanium oxide–bismuth oxide supported by alumina (ATB)for transesterification of colza oil with methanol was studied [60].Bothcatalysts were prepared by co-mixing method and followed bycalcination at 600 °C for 3 h. The specific surface areas of ATZ and ATBwere 62 m 2 /g and 54 m 2 /g respectively. Although the specific surfacearea of both catalysts was high, a long reaction time of 8 h was requiredto reach 94% of methyl ester yields. The reaction took place at temperatureof 200 °C and pressure of 50 bar. Both catalysts were reusableand showed almost the same activity in the second cycle.Al 2 O 3 /ZrO 2 /WO 3 solid acid catalyst was prepared by co-precipitatingmethod and was investigated in the methanolysis of soybean oil [61].The catalyst was compatible for both esterification and transesterificationreaction and under reaction conditions of temperature of 250 °Cand alcohol/oil molar ratio of 40:1 methyl ester yields of 90% wereobtained. This catalyst was also compared to Al 2 O 3 /ZrO 2 ; however,the yields were 80%.Xie et al. [62] evaluated the catalytic efficiency of potassium supportedby alumina as a solid base catalyst in the transesterificationof soybean oil to biodiesel. The catalyst was prepared by an impregnationmethod of aqueous solution of potassium nitrate on

774 M. Zabeti et al. / Fuel Processing Technology 90 (2009) 770–777alumina. The results demonstrated that the catalyst calcined at 500 °Cand loaded by 35 wt.% of KNO 3 was the most active. From Hammettindicator method, the basicity was 6.75 mmol/g and a correlationbetween catalyst performance and basic sites on the surface of catalyst,probably K 2 O and Al–O–K, was detected. The highest conversion of oilreached 87%, after the reaction time of 8 h with molar ratio of methanolto soybean oil of 15:1 and 6.5 wt.% of catalyst.In another work, Xie and Li [63] used potassium iodide supportedby alumina as a solid base catalyst for methanolysis of the same oil tostudy the effect of loading amount of KI on the support and calcinationtemperature on the conversion. The optimum catalyst was obtainedby loading 35 wt.% of KI on the support and calcined at 500 °C for 3 hand from Hammett indicator method the basicity recorded was1.57 mmol/g. These results were directly correlated with theconversion of oil since the maximum conversion reached 96%. Thereaction was performed with oil to methanol molar ratio of 1:15,catalyst content of 2.5 wt.% and 8 h of reaction time. Table 2 representsthe effect of catalyst precursor by comparing the activity of Al 2 O 3 /Ksynthesized by two different potassium precursors. Both reactionshave been carried out with alcohol/oil molar ratio of 15:1, catalystcontent of 2.5 wt.% and 8 h of reaction time. As can be seen from thetable, although the basicity of Al 2 O 3 /KI is less than the basicity ofAl 2 O 3 /KNO 3 , more conversion was obtained using Al 2 O 3 /KI whichimplies the effect of precursor type on the catalyst activity.Kim et al. [64] utilized alumina loaded with sodium and sodiumhydroxide as a solid base catalyst for transesterification of soybean oilwith a mixture of hexane and methanol to biodiesel. The specificsurface area of γ-alumina decreased from 143.1 m 2 /g to 83.2 m 2 /gwhen Na and NaOH were loaded onto the support. However, thebiodiesel yields reached 83% using Al 2 O 3 /Na/NaOH while it was only1% when γ-alumina was used without any modification. The activitywas closely correlated to basic active sites which determined bytemperature programmed desorption of carbon dioxide (TPD-CO 2 ).The catalyst activity was also compared to NaOH homogeneouscatalyst and nearly the same activity was detected.Kaita et al. [65] patented aluminum phosphate solid acid catalystfor transesterification of palm kernel oil with methanol. Two catalystswere prepared by a co-precipitating method of aluminum nitrate andorthophosphoric acid. One calcined at temperature of 400 °C for 3 hand another one used as amorphous. The reaction was carried out at200 °C for 5 h and 69% methyl ester yields were obtained in thepresence of crystalline catalyst compared to 63% yields whenamorphous one was used. Then the crystalline catalyst was moldedinto cylindrical form with 3 mm diameter to study the effect of catalystshapes on catalyst efficiency and as a result, methyl ester yieldsincreased to 90.1%.3.2. Zinc oxideYang and Xie [66] compared the catalyst performance of alkaliearth metals loaded on different catalyst supports for soybean oilconversion to biodiesel and discovered a correlation between loadingamount of catalyst precursor on support and the conversion of oil.They also found that the catalyst performance is depending uponconcentration of basic sites on the surface of the catalyst. DifferentTable 2The effect of precursor type on the catalyst activity.CatalystLoadingamount ofactivematerial(wt.%)Active siteconcentration(mmol/g)Catalystcontentin thereaction(wt.%)Conversion(%)ReferenceAl 2 O 3 /KNO 3 35 6.75 2.5 37 [62]Al 2 O 3 /KI 35 1.57 2.5 96 [63]Both catalysts were activated at 500 °C for 3 h.Table 3Textural properties of functionalized mesoporous silicate [68].Catalyst Surfactant S BET(m 2 /g)catalysts were prepared by impregnation method of water solution ofan alkali earth metal nitrate with specified concentration and calcinedat 600 °C for 5 h and among them ZnO/Sr (NO 3 ) 2 showed the bestcatalytic activity. In this case the most active catalyst was obtained byloading 2.5 mmol of strontium nitrate on zinc oxide which resulted inbasicity of 10.8 mmol/g. After 5 h of reaction time, at 65 °C, with 12:1molar ratio of alcohol/oil and 5 wt.% of catalyst content, the maximumconversion achieved was 93.7%. In order to study the effect of cosolventon the conversion of soybean oil to biodiesel, Yang and Xie [66]applied different co-solvents such as THF, DMSO and n-hexane andTHF was found as the most effective co-solvent since the conversion ofsoybean oil increased to 96.8%.Xie and Huang [35] also applied potassium supported by zinc oxideas a solid base catalyst and the catalyst was prepared by an impregnationmethod of aqueous solution of potassium fluoride followed bycalcination at 600 °C for 5 h. Based on Hammett titration method theconcentration of basic sites reached 1.47 mmol/g when 15 wt.% of KFwas loaded on ZnO. The reaction was performed with alcohol/oilmolar ratio of 10:1 and catalyst content of 3 wt.% and after 9 h of thereaction time 87% of soybean oil was converted. The result is quitesimilar to that of Al 2 O 3 /KNO 3 which was previously described [62].A mixture of Zn/I 2 was used as a heterogeneous catalyst fortransesterification of soybean oil with methanol [9]. The metal wastreated with distilled water and diluted HCL, and iodine was treated bysublimation. A mixture of 5 wt.% Zn and 2.5 wt.% I 2 was applied in thereaction. After 26 h, by using alcohol/oil molar ratio of 42:1, theconversion of oil reached 86%. The influence of co-solvent on thereaction rate was also investigated by introducing co-solvents such asTHF and DMSO into the reaction and a 10% increase in conversion ofsoybean oil was observed after adding a certain amount of DMSO intothe reaction mixture.3.3. SilicateV P(cm 3 /g)D P (°A) H +(meq/g)CDAB-SO 3 H-C12 CDAB 778 0.39 28 0.60SBA-15-SO 3 H-L64 Pluronic L64 820 0.58 27 0.84SBA-15-SO 3 H-P123 Pluronic P123 735 0.67 35 1.44SBA-15-ph-SO 3 H-P123 Pluronic P123 540 0.71 50 0.92Catalyst properties of CaO supported on different types ofmesoporous silica were studied and results from XRD characterizationshowed that after impregnation of CaO, the hexagonal structure ofSBA-15 remained stable while this structure of MCM-41 and fumedsilica was collapsed indicated that only SBA-15 can stabilize calciumoxide species on its surface [67]. The optimum catalyst was obtainedby loading 15 wt.% of calcium oxide on SBA-15 followed by calcinationat 900 °C and using this catalyst the maximum conversion of sunfloweroil achieved 95% within 5 h. Transesterification reaction wascarried out at 60 °C, with 1 wt.% catalyst and methanol to sunflower oilmolar ratio of 12.Lin and Radu [68] patented the application of sulfonic acid supportedby mesoporous silica as a strong solid acid for transesterificationreaction. The method of catalyst synthesis was relatively complicated,whereby Tetraethoxysilane (TEOS) was used as a precursor of silica.Mesoporous silica was synthesized by cetthyldimethylammonium bromidefunctionalized group denoted as CDAB, while those prepared byusing tri-block Pluronic L64 and Pluronic P123 copolymers denoted asSBA-15. The textural properties of the catalysts are summarized inTable 3. As can be seen from the table, SBA-15-SO 3 H-P123 bearsthe most surface acidity. Consequently, the maximum biodiesel yieldsachieved were 85% within 20 h of reaction over SBA-15-SO 3 H-P123while CDAB-SO 3 H with 0.6 meq/g of surface acidity showed the least

M. Zabeti et al. / Fuel Processing Technology 90 (2009) 770–777775Table 4Summarization of the activity of metal oxides and supported catalysts as heterogeneous catalysts for biodiesel production.Catalyst Catalyst characterizations Catalyst preparation Feedstock Operation conditions Results Ref.Nano MgOsolid baseCaO solid baseParticle size=60 nm Not reported SoybeanoilSSA a =32 m 2 /g,MPS b =25–30 nmCaO powder was calcinedat 1000 °C.SunfloweroilCaO solid base SSA=56 m 2 /g Not reported SoybeanoilCa(C 2 H 3 O 2 ) 2 /SBA-15solid baseCa(OCH 3 ) 2solid baseSSA=7.4 m 2 /g, porevolume=0.019 cm 3 /gSSA=19 m 2 /g, pore size=40 nmThe catalyst was prepared by an impregnation Sunflowermethod of aqueous solution of calcium acetate on the oilsupport followed by calcination at 900 °C for 4 h.Calcium methoxide was synthesized by a directreaction of calcium and methanol at 65 °C for 4 h.SrO solid base SSA=1.05 m 2 /g SrO was prepared from calcination of strontiumcarbonate in a muffle furnace at 1200 °C for 5 h.ZnO/Sr(NO 3 ) 2solid baseZnO/Basolid baseZnO/KFsolid baseAl 2 O 3 /KIsolid baseAl 2 O 3 /KNO 3solid baseAl 2 O 3 /Na/NaOHsolid baseVOPO 4·2H 2 Osolid acidBasicity=10.8mmol/gBasicity=14.54 mmol/gBasicity=1.62 mmol/gBasicity=1.57 mmol/gBasicity=6.67 mmol/gThe aqueous solution of Sr (NO 3 ) 2 was loaded onZnO by impregnation method and calcined at600 °C for 5 h.SoybeanoilSoybeanoilSoybeanoilThe catalyst was prepared by an impregnation Soybeanmethod using barium nitrate as precursor on ZnO. oilDried overnight and calcined at 600 °C in air for5h.The supported catalyst was prepared bySoybeanimpregnation method with an aqueous solution of oilKF. Afterward was dried at 393 K and calcined at600 °C in air for 5 h.The catalyst was prepared by an impregnation Soybeanmethod from aqueous solution of potassium iodide, oildried at 120 °C and activated at 500 °C for 3 h.KNO 3 was loaded on alumina by an impregnationmethod from aqueous solution, dried at 393 K for16 h and finally calcined at 500 °C for 5 h.SoybeanoilSSA=83.2 m 2 /g The catalyst was prepared by mixing of γAl 2 O 3 , SoybeanNaOH and metal sodium in a stainless steel reactor oilat 320 °C.SSA=2–4 m 2 /gVanadyl phosphate was obtained from thesuspension of V 2 O 5 in diluted phosphoric acid.Then activated at 500 °C.SoybeanoilZnO solid acid Not reported Pure metal oxide was used. 1-palmkernel oil2-coconutoilZnO/I 2solid acidZrO 2 /WO 32−solid acidZrO 2 /SO 42−solid acidTiO 2 /SO 42−solid acidAl 2 O 3 /PO 43−solid acidAl 2 O 3 /TiO 2 /ZnOsolid acidNot reportedNot reportedNot reportedSSA=99.5 m 2 /gNot reportedSSA=62 m 2 /gThe metal was prepared just by some treatmentwith distilled water and dilute HCl. Iodine wastreated by sublimation.The isopoly tungstated zirconia was prepared bysuspending a known amount of zirconiumoxyhydroxide powder in an aqueous solution ofammonium metatungstate, finally calcined at750 °C.Zirconia powder were immersed in sulfuric acidsolution, filtered, dried and calcined at 500 °C for2h.TiO 2·nH 2 O was prepared from precipitation ofTiCl 4 using aqueous ammonia. Then immersed tosulfuric acid and finally calcined at 550 °C for 3 hto give TiO 2 –SO 2− 4 .SoybeanoilSunfloweroil1-palmkernel oil2-coconutoilCottonseedoilAluminum nitrate hydrated with nine moles of Palmwater dissolved in water and 85% orthophosphoric kernel oilacid was added. The PH was controlled at 7 byaqueous solution of ammonia. The finalprecipitation was filtered out, washed, and dried at383 K for 12 h. finally calcined at 400 °C for 3 h.The catalyst was prepared by co-mixing ofboehmite, titanium gel and zinc oxide in thepresence of nitric acid and water. Calcined at600 °C for 3 h.P=24 MPa,T=250 °C,t=10 min,alcohol/oil =36:1,catalyst content=3%T=60 °C, t=100 min,alcohol/oil =13:1,catalyst content=3%T=65 °C, t=3 h,alcohol/oil =12:1,catalyst content=8%T=60 °C, t=5 h,methanol/oil =12:1,catalyst content=1%T=65 °C, t=2 h,alcohol/oil =1:1,catalyst content=2%T=65 °C, t=30 min,alcohol/oil =12:1,catalyst content=3%T=65 °C, t=5 h,methanol/oil =12:1,catalyst content=5%T=65 °C, t=1 h,methanol/oil =12:1,catalyst content=6%T=65 °C, t=9 h,methanol/oil =10:1,catalyst content=3%Methanol/oil =15:1,catalyst content=2.5%,t=8 hMethanol/oil =15:1,catalyst content=6.5%,t=7 hT=60 °C, t=2 h,methanol/oil =9:1,catalyst content=1 gT=150 °C, t=1 h,alcohol/oil =1:1,catalyst content=2%T=300 °C, t=1 h,alcohol/oil =6:1,catalyst content=3%Conversion=99% [33]Conversion=94% [32]Conversion=95% [45]Conversion=95% [67]Conversion=98% [71]Conversion=95% [21]Conversion=93% [66]Conversion=95% [18]Conversion=87% [35]Methyl esteryields =96%.Methyl esteryields =87%.Methyl esteryields =83%.Methyl esteryields =80%[61][62][64][54]1-methyl ester [34]yields =86.1% 2-methyl esteryields =77.5%T=65 °C, t=26 h, Conversion=96% [9]and 2.5% I 2methanol/oil =42:1,catalyst content=5% ZnT=200 °C, t=5 h, Conversion=97 [52]alcohol/oil =20:1,catalyst content=15%T=200 °C, t=1 h,alcohol/oil =6:1,catalyst content=1%T=230 °C, t=8 h,alcohol/oil =12:1,catalyst content=2%T=200 °C, t=5 h,methanol/oil =5:1,catalyst content=10 gColza oil T=200 °C, t=8 h,alcohol/oil =1:1,atalyst content=6%1-methyl esteryields =90.3% 2-methyl esteryields =86.3%Methyl esteryields =90%.Methyl esteryields =69%.Methyl esteryields =94%[34][50][65][60](continued on next page)

776 M. Zabeti et al. / Fuel Processing Technology 90 (2009) 770–777Table 4 (continued)Catalyst Catalyst characterizations Catalyst preparation Feedstock Operation conditions Results Ref.Al 2 O 3 /ZrO 2 /WO 3solid acidSBA-15-SO 3 H-P123solphonic acidsupported onmesoporous silica.Solid acidNot reporteda : Specific surface area; b : mean pore.SSA=735 m 2 /g, pore volume=0.67cm 3 /gThe catalyst was prepared from mixture ofhydrated zirconia, hydrated alumina andammonium metatungstate and deionized water.Calcined at 900 °C for 1 h.Tetraethoxysilane (TEOS) was used as silicasource. 3-(mercaptopropyl)-trimethoxysilane wasused as mesoporous silica modifier. Pluronic P123(tri-block copolymer) used as surfactant. A cocondensationmethod was applied.SoybeanoilSoybeanoilT=250 °C, t=20 h,methanol/oil =40:1,catalyst content=4 gT=75 °C, t=20 h,methanol/oil =20:1,catalyst content=10%Methyl esteryields =90%Methyl esteryields =85%[61][68]activity with less than 55% methyl ester yields. MCM-41 also wassubjected to the transesterification of soybean oil and yields of 60% wereobtained. All the reactions were carried out at 75 °C, 10 wt.% of catalystand methanol/oil molar ratio of 20:1.Siano et al. [69] studied the effect of loading amount of the catalystprecursor on the support and the effect of the support shaping on theactivity of TiO 2 supported by amorphous silica and found that byincreasing the loading amount of the precursor (titanium tetraisopropoxide)the activity of the catalyst is increased. Moreover, theyused powder form and pellet form of the support to prepare thecatalyst and under the same operating conditions more conversion ofsoybean oil obtained using the catalyst which was prepared bypowder form of silica. They also tested the effect of the reactiontemperature and found that conversion of oil was increased by thereaction temperature. Using TiO 2 supported on the powder form ofsilica the conversion of 88% obtained after 5 1/2 h when the reactionwas performed at temperature of 225 °C using the catalyst content of2.5 wt.% and methanol/oil weight ratio of 0.45.Silica supported phosphoric acid was also investigated for transesterificationof triacetin with methanol [49]. The catalyst activity wasextremely low with less than 10% conversion of triacetin, consistentwith the small specific surface area of 2.6 m 2 /g and weak acid siteconcentration about 986 μmol/g.3.4. Zirconium oxideZirconia-supported heteropoly and isopoly tungsten acids wereapplied in biodiesel synthesis from sunflower oil and the resultindicated that isopoly tungsten was more active than heteropolytungsten since it possesses more Bronsted acid sites [70]. Oil conversionof 97% was reached after 5 h of reaction time under reactionconditions of temperature of 200 °C with catalyst content of 15 wt.%and alcohol/oil molar ratio 15:1.The catalyst activity of some metal oxides and catalyst supports forthe biodiesel production is summarized in Table 4. From this table it canbe concluded that, generally solid catalysts need a longer reaction timethan homogeneous catalysts. Solid acid catalysts either supported orunsupported require higher temperature to give more methyl estersyields, even though solid acid catalysts are active for both transesterificationand esterification reactions are able to convert oils with highamount of FFA. All alkali earth metal oxides except MgO produced highconcentration of methyl esters at lower temperature of 65 °C at amoderate reaction time. Furthermore, from the table it can be observedthat alumina was more preferred as a support than other supports fortransesterification reaction. Therefore modification and optimization ofthese catalysts in order to obtain the most active catalyst for transesterificationreaction or producing a supported catalyst contain basic andacidic species are optimistic ways to find a new heterogeneous catalystto be substituted by homogeneous catalysts for biodiesel production.4. ConclusionEffective factors on catalytic activity of solid catalysts are specificsurface area, pore size, pore volume and active site concentration on thesurface of catalyst. Moreover type of precursor of active materials hassignificant effect on the catalyst activity of supported catalysts. Howeveractive site concentration was found to be the most important factor forsolid catalyst performance. The use of catalyst supports such as alumina,silica and zinc oxide could improve the mass transfer limitation of thethree phase reaction. Furthermore, by anchoring metal oxides insidepores, catalyst supports could prevent active phases from sintering inthe reaction medium. The amount of methyl ester yields and conversionof oils depend on not only catalyst activity but also the type of oils andthe applied operation conditions. By modifying the catalyst activity andsynthesis cost, heterogeneous catalyst could be replaced by homogeneouscatalyst for the biodiesel production.AcknowledgementThis study was carried out with the aid of a research grant fromUniversity Malaya Research Fund.References[1] C.-Y. Lin, H.-A. Lin, L.-B. Hung, Fuel structure and properties of biodiesel producedby peroxidation process, Fuel 85 (2006) 1743–1749.[2] A.C. Pinto, L.L.N. Guarieiro, M.J.C. Rezende, N.M. Ribeiro, E.A. Torres, W.A. Lopes, P.A.de Pereira, J.B. de Andrade, Biodiesel: an overview, Journal of the Brazilian ChemicalSociety 16 (2005) 1313–1330.[3] M.P. Dorado, E. Ballesteros, J.M. Arnal, J. Gómez, F.J. López, Exhaust emissions from adiesel engine fueled with transesterified waste olive oil, Fuel 82 (2003) 1311–1315.[4] M.A. Dubé, A.Y. Tremblay, J. Liu, Biodiesel production using a membrane reactor,Bioresource Technology 98 (2007) 639–647.[5] J.P. Szybist, J. Song, M. Alam, A.L. Boehman, Biodiesel combustion, emissions andemission control, Fuel Processing Technology 88 (2007) 679–691.[6] A.K. Dalai, M.G. Kulkarni, L.C. Meher, Biodiesel productions from vegetable oilsusing heterogeneous catalysts and their applications as lubricity additives, IEEE EICClimate Change Technology Conference EICCCC art 4057358, 2006.[7] S. Baroutian, M.K. Aroua, A.A.A. Raman, N.M.N. Sulaiman, Density of palm oil-basedmethyl ester, Journal of Chemical and Engineering Data 53 (2008) 877–880.[8] A.K. Tiwari, A. Kumar, H. Raheman, Biodiesel production from jatropha oil(Jatropha curcas) with high free fatty acids: an optimized process, Biomass andBioenergy 31 (2007) 569–575.[9] H. Li, W. Xie, Transesterification of soybean oil to biodiesel with Zn/I 2 catalyst,Catalysis Letters 107 (2006) 25–30.[10] G. Arzamendi, E. Arguiñarena, I. Campo, S. Zabala, L.M. Gandía, Alkaline andalkaline-earth metals compounds as catalysts for the methanolysis of sunfloweroil, Catalysis Today (2008) 133–135.[11] R.R. Tan, A.B. Culaba, M.R.I. Purvis, Carbon balance implications of coconutbiodiesel utilization in the Philippine automotive transport sector, Biomass andBioenergy 26 (2004) 579–585.[12] Y. Wang, S.O. Pengzhan Liu, Z. Zhang, Preparation of biodiesel from waste cookingoil via two-step catalyzed process, Energy Conversion and Management 48 (2007)184–188.[13] K. Ropkins, R. Quinn, J. Beebe, H. Li, B. Daham, J. Tate, M. Bell, G. Andrews, Real-worldcomparison of probe vehicle emissions and fuel consumption using diesel and 5%biodiesel (B5) blend, Science of the Total Environment 376 (2007) 267–284.[14] Sonntag, Reactions of fats and fatty acids, Bailey's Industrial Oil and Fat Products,vol. 1, 1979, p. 99.[15] D.G. Lima, V.C.D. Soares, E.B. Ribeiro, D.A. Carvalho, E.C.V. Cardoso, F.C. Rassi, K.C.Mundim, P.A.Z. Suarez, Diesel-like fuel obtained by pyrolysis of vegetable oils,Journal of Analytical and Applied Pyrolysis 71 (2004) 987–996.[16] S. Ensöz, D. Angın, S. Yorgun, Influenceofparticlesizeonthepyrolysisofrapeseed(Brassica napus L.): fuel properties of bio-oil, Biomass and Bioenergy 19 (2000) 271–279.[17] I.C.P. Fortes, P.J. Baugh, Pyrolysis—GC/MS studies of vegetable oils from Macaubafruit, Journal of Analytical and Applied Pyrolysis 72 (2004) 103–111.[18] W. Xie, Z. Yang, Ba-ZnO catalysts for soybean oil transesterification, Catalysis Letters117 (2007) 159–165.

M. Zabeti et al. / Fuel Processing Technology 90 (2009) 770–777777[19] E.W. Eckey, Esterification and interesterification, JAOCS 33 (1956) 575–579.[20] F. Ma, M.A. Hanna, Biodiesel production: a review, Bioresource Technology 70 (1999)1–15.[21] X. Liu, H. He, Y. Wang, S. Zhu, Transesterification of soybean oil to biodiesel usingSrO as a solid base catalyst, Catalysis Communications 8 (2007) 1107–1111.[22] G. Sunita, B.M. Devassy, D.P. Sawant, V.V. Balasubramanian, S.B. Halligudi, Synthesisof biodiesel over zirconia-supported isopoly and heteropoly tungstate Catalysts,Catalysis Communications 9 (2008) 696–702.[23] N. Dizge, C. Aydiner, D.Y. Imer, M. Bayramoglu, A. Tanriseven, B. Keskinler, Biodieselproduction from sunflower, soybean, and waste cooking oils by transesterificationusing lipase immobilized onto a novel microporous polymer,Bioresource Technology 100 (2009) 1983–1991.[24] V. Sivozhelezova, D. Bruzzeseb, L. Pastorinoa, E. Pechkova, C. Nicolini, Increase ofcatalytic activity of lipase towards olive oil by Langmuir-film immobilization oflipase, Enzyme and Microbial Technology 44 (2009) 72–76.[25] D. Kusdiana, S. Saka, Effects of water on biodiesel fuel production by supercriticalmethanol treatment, Bioresource Technology 91 (2004) 289–295.[26] B. Freedman, E.H. Pryde, T.L. Mounts, Variables affecting the yields of fatty estersfrom transesterified vegetable oils, Journal of the American Oil Chemists Society61 (1984) 1638–1643.[27] M. Canakci, J. Van Gerpen, Biodiesel production via acid catalysis, Transactions ofthe American Society of Agricultural Engineers 42 (1999) 1203–1210.[28] S.L. Dmytryshyn, A.K. Dalai, S.T. Chaudhari, Synthesis and characterization of vegetableoil derived esters: evaluation for their diesel additive properties, BioresourceTechnology 92 (2004) 55–64.[29] O.J. Alamu, M.A. Waheed, S.O. Jekayinfa, Effect of ethanol–palm kernel oil ratio onalkali-catalyzed biodiesel yields, Fuel 87 (2008) 1529–1533.[30] G. Vicente, M. Martínez, J. Aracil, Optimization of integrated biodiesel production,part I. a study of the biodiesel purity and yields, Bioresource Technology 98 (2007)1724–1733.[31] F.R. Abreu, D.G. Lima, E.H. Hamú, S. Einloft, J.C. Rubim, P.A.Z. Suarez, New metalcatalysts for soybean oil transesterification, JAOCS Journal of the American OilChemists' Society 80 (2003) 601–604.[32] M.L. Granados, M.D.Z. Poves, D.M. Alonso, R. Mariscal, F.C. Galisteo, R. Moreno-Tost, J. Santamaría, J.L.G. Fierro, Biodiesel from sunflower oil by using activatedcalcium oxide, Applied Catalysis B: Environmental 73 (2007) 317–326.[33] L. Wang, J. Yang, Transesterification of soybean oil with nano-MgO or not insupercritical and subcritical methanol, Fuel 86 (2007) 328–333.[34] J. Jitputti, B. Kitiyanan, P. Rangsunvigit, K. Bunyakiat, L. Attanatho, P. Jenvanitpanjakul,Transesterification of crude palm kernel oil and crude coconut oil by differentsolid catalysts, Chemical Engineering Journal 116 (2006) 61–66.[35] W. Xie, X. Huang, Synthesis of biodiesel from soybean oil using heterogeneousKF/ZnO catalyst, Catalysis Letters 107 (2006) 53–59.[36] T.F. Dossin, M.-F. Reyniers, R.J. Berger, G.B. Marin, Simulation of heterogeneouslyMgO-catalyzed transesterification for fine-chemical and biodiesel industrialproduction, Applied Catalysis B 67 (2006) 136–148.[37] I.K. Mbaraka, B.H. Shanks, Conversion of oils and fats using advanced mesoporousheterogeneous catalysts, JAOCS Journal of the American Oil Chemists' Society 83 (2006)79–91.[38] S. Gryglewicz, Rapeseed oil methyl esters preparation using heterogeneouscatalysts, Bioresource Technology 70 (1999) 249–253.[39] L. Bournay, D. Casanave, B. Delfort, G. Hillion, J.A. Chodorge, New heterogeneousprocess for biodiesel production: a way to improve the quality and the value of thecrude glycerin produced by biodiesel plants, Catalysis Today 106 (2005) 190–192.[40] M. Di Serio, R. Tesser, L. Pengmei, E. Santacesaria, Heterogeneous catalysts forbiodiesel production, Energy Fuels 22 (2008) 207–217.[41] E. Lotero, Y. Liu, D.E. Lopez, K. Suwannakarn, D.A. Bruce, J.G. Goodwin, Synthesis ofbiodiesel via acid catalysis, Industrial & Engineering Chemistry Research 44 (2005)5353–5363.[42] I. Chorkendorff, J.W. Niemantsverdriet, Concepts of Modern Catalysis and Kinetics,Wiley-VCH, Germany, 2003.[43] T. Baba, T. Endou, H. Handa, Y. Ono, Calcium oxide as a catalyst for the isomerizationof 5-vinylbicyclo[2.2.1]hept-2-ene to 5-ethylidenebicyclo[2.2.1]hept-2-ene in the liquid phase, Applied Catalysis A: General 97 (1993) L19–L23.[44] S.I. Fujita, B.M. Bhanage, H. Kanamaru, M. Arai, Synthesis of 1,3-dialkylurea fromethylene carbonate and amine using calcium oxide, Journal of Molecular Catalysis A:Chemical 230 (2005) 43–48.[45] X. Liu, H. He, Y. Wang, S. Zhu, X. Piao, Transesterification of soybean oil to biodieselusing CaO as a solid base catalyst, Fuel 87 (2008) 216–221.[46] A. Demirbas, Biodiesel from sunflower oil in supercritical methanol with calciumoxide, Energy Conversion and Management 48 (2007) 937–941.[47] X. Xiao, J.W. Tierney, I. Wender, Alkylation of isobutane with 2-butene over anionmodifiedzirconium oxide catalysts, Applied Catalysis A: General 183 (1999) 209–219.[48] M. Stöcker, N-butane isomerization catalyzed by antimony pentafluoridesupported on sulfate-treated zirconium oxide, Journal of Molecular Catalysis29 (1985) 371–377.[49] D.E. López, J.G. Goodwin Jr., D.A. Bruce, E. Lotero, Transesterification of triacetin withmethanol on solid acid and base catalysts, Applied Catalysis A: General 295 (2005)97–105.[50] C. He, B. Peng, D. Wang, J. Wang, Biodiesel production by the transesterificationof cottonseed oil by solid acid catalysts, Frontiers of Chemical Engineering in China1 (2007) 11–15.[51] D.E. López, K. Suwannakarn, D.A. Bruce, J.G. Goodwin Jr., Esterification and transesterificationon tungstated zirconia: effect of calcination temperature, Journal ofCatalysis 247 (2007) 43–50.[52] S. Ramu, N. Lingaiah, B.L.A. Prabhavathi Devi, R.B.N. Prasad, I. Suryanarayana, P.S. SaiPrasad, Esterification of palmitic acid with methanol over tungsten oxide supportedon zirconia solid acid catalysts: effect of method of preparation of the catalyst on itsstructural stability and reactivity, Applied Catalysis A: General 276 (2004) 163–168.[53] S. Furuta, H. Matsuhashi, K. Arata, Biodiesel fuel production with solid superacidcatalysis in fixed bed reactor under atmospheric pressure, Catalysis Communications5 (2004) 721–723.[54] M. Di Serio, M. Cozzolino, R. Tesser, P. Patrono, F. Pinzari, B. Bonelli, E. Santacesaria,Vanadyl phosphate catalysts in biodiesel production, Applied Catalysis A: General320 (2007) 1–7.[55] P.S. Sreeprasanth, R. Srivastava, D. Srinivas, P. Ratnasamy, Hydrophobic, solid acidcatalysts for production of biofuels and lubricants, Applied Catalysis A: General314 (2006) 148–159.[56] D. Siano, M. Nastasi, E. Santacesaria, M. Di Serio, R. Tesser, G. Minutillo, M. Ledda,T. Tenore, Process for producing esters from vegetable oils and animal fats byusing heterogeneous catalysts. WO 050925 A1 (2006).[57] W. Xie, H. Peng, L. Chen, Calcined Mg–Al hydrotalcites as solid base catalysts formethanolysis of soybean oil, Journal of Molecular Catalysis A: Chemical 246 (2006)24–32.[58] K.G. Georgogianni, A.P. Katsoulidis, P.J. Pomonis, M.G. Kontominas, Transesterificationof soybean frying oil to biodiesel using heterogeneous catalysts, FuelProcessing Technology 90 (2009) 671–676, doi:10.1016/j.fuproc.2008.12.004(Corrected Proof, Available online 19 January 2009).[59] M. Di Serio, M. Ledda, M. Cozzolino, G. Minutillo, R. Tesser, E. Santacesaria, Transesterificationof soybean oil to biodiesel by using heterogeneous basic catalysts,Industrial Engineering Chemistry Research 45 (9) (2006) 3009–3014.[60] B. Delfort, D. Le Pennec, C. Lendresse, Process for transesterification of vegetableoils or animal oils by means of heterogeneous catalysts based on zinc or bismuth,titanium and aluminum. United State patent 7,151,187 B2 (2006).[61] S. Furuta, H. Matsuhashi, K. Arata, Biodiesel fuel production with solid amorphouszirconiacatalysis in fixed bed reactor, Biomass and Bioenergy 30 (2006) 870–873.[62] W. Xie, H. Peng, L. Chen, Transesterification of soybean oil catalyzed by potassiumloaded on alumina as a solid-base catalyst, Applied Catalysis A: General 300 (2006)67–74.[63] W. Xie, H. Li, Alumina-supported potassium iodide as a heterogeneous catalyst forbiodiesel production from soybean oil, Journal of Molecular Catalysis A: Chemical255 (2006) 1–9.[64] H.J. Kim, B.-S. Kang, M.-J. Kim, Y.M. Park, D.-K. Kim, J.-S. Lee, K.-Y. Lee, Transesterificationof vegetable oil to biodiesel using heterogeneous base catalyst,Catalysis Today 93–95 (2004) 315–320.[65] J. Kaita, T. Mimura, N. Fukuoka, Y. Hattori, Catalyst for transesterification. UnitedState Patent 6,407,269 B2 (2002).[66] Z. Yang, W. Xie, Soybean oil transesterification over zinc oxide modified with alkaliearth metals, Fuel Processing Technology 88 (2007) 631–638.[67] M.C.G. Albuquerque, I. Jiménez-Urbistondo, J. Santamaría-González, J.M. Mérida-Robles, R. Moreno-Tost, E. Rodríguez-Castellón, A. Jiménez-López, P. Maireles-Torres,CaO supported on mesoporous silicas as base catalysts for transesterificationreactions, Applied Catalysis A: General 334 (2008) 35–43.[68] V.S.Y. Lin, D.R. Radu, Use of functionalized mesoporous silicates to esterify fattyacids and transesterify oils. United State Patent 7,122,688 B2 (2006).[69] D. Siano, L. Siano, M. Nastasi, E. Santacesaria, M. Di Serio, Methods for producingesters from vegetable oils and animal fats by using heterogeneous catalysts. WO094986 A1 (2006).[70] G. Sunita, B.M. Devassy, A. Vinu, D.P. Sawant, V.V. Balasubramanian, S.B. Halligudi,Synthesis of biodiesel over zirconia-supported isopoly and heteropoly tungstatecatalysts, Catalysis Communications 9 (2007) 696–702.[71] X. Liu, X. Piao, Y. Wang, S. Zhu, H. He, Calcium methoxide as a solid base catalyst forthe transesterification of soybean oil to biodiesel with methanol, Fuel 87 (2008)1076–1082, doi:10.1016/j.fuel.2007.05.059.

Chemical Engineering and Processing 49 (2010) 323–330Contents lists available at ScienceDirectChemical Engineering and Processing:Process Intensificationjournal homepage: www.elsevier.com/locate/cepReviewProcess intensification technologies in continuous biodiesel productionZheyan Qiu ∗ , Lina Zhao, Laurence WeatherleyChemical and Petroleum Engineering Department, The University of Kansas, Lawrence, KS 66045, USAarticleinfoabstractArticle history:Received 27 October 2009Received in revised form 4 March 2010Accepted 5 March 2010Available online 12 March 2010Keywords:Biodiesel productionTransesterificationMass transferMixingProcess intensificationAs an alternative fuel, biodiesel has been accepted because it is produced from renewable resources. Thereare some technical challenges facing biodiesel production via transesterification, which include longresidence times, high operating cost and energy consumption, and low production efficiency. In recentyears, studies on biodiesel synthesis have focused on development of process intensification technologiesto resolve some of these issues. This contribution will present a brief review of some of technologiesbeing developed and includes description of some of the types of novel reactors and relevant coupledreaction/separation processes. These technologies enhance reaction rate, reduce molar ratio of alcohol tooil and energy input by intensification of mass transfer and heat transfer and in situ product separation,thus achieve continuous product in a scalable unit. Some of these technologies have been commercializedsuccessfully.© 2010 Elsevier B.V. All rights reserved.Contents1. Introduction .......................................................................................................................................... 3232. Novel reactors ........................................................................................................................................ 3242.1. Static mixers .................................................................................................................................. 3242.2. Micro-channel reactors....................................................................................................................... 3242.3. Oscillatory flow reactors ..................................................................................................................... 3252.4. Cavitational reactors ......................................................................................................................... 3252.5. Rotating/spinning tube reactors ............................................................................................................. 3262.6. Microwave reactors .......................................................................................................................... 3273. Reaction/separation coupled technologies .......................................................................................................... 3273.1. Membrane reactors ........................................................................................................................... 3273.2. Reactive distillation .......................................................................................................................... 3283.3. Centrifugal contactors ........................................................................................................................ 3284. Summary ............................................................................................................................................. 3295. Conclusions .......................................................................................................................................... 330References ........................................................................................................................................... 3301. IntroductionThe majority of commercial biodiesel is made by transesterificationof vegetable oils and animal fats with methanol or ethanolin stirred tank reactors in the presence of base or acids catalysts.There are some challenges related to this process as follows:a. Reaction rate can be limited by mass transfer between the oilsand alcohol because they are immiscible;∗ Corresponding author. Tel.: +1 785 864 7174; fax: +1 785 864 4967.E-mail addresses: qiu zhy@yahoo.com, jamesqiu@ku.edu (Z. Qiu).b. Transesterification itself is a reversible reaction and thereforethere is an upper limit to conversion in the absence of productremoval;c. Most commercial processes are run in a batch mode and thus donot gain some of the advantages of continuous operation.In order to overcome these problems, current conventionaltechniques involve long reaction time, high molar ratio of alcoholto oil and catalyst concentration. High operating cost and energyconsumption are required to purify biodiesel and recover excessamount of alcohol and catalysts during downstream processing.Significant amounts of toxic waste water may also be producedduring downstream purification. Long residence times and down-0255-2701/$ – see front matter © 2010 Elsevier B.V. All rights reserved.doi:10.1016/j.cep.2010.03.005

324 Z. Qiu et al. / Chemical Engineering and Processing 49 (2010) 323–330Fig. 1. Experimental setup: (a) static mixer closed-loop system, and (b) internalstructure of static mixers [3] (Courtesy of American Society of Agricultural andBiological Engineers).stream processing time incur low production efficiency. Hence,some process intensification technologies have been developed andapplied to improve mixing and mass/heat transfer between the twoliquid phases in recent years. These technologies either utilize novelreactors or coupled reaction/separation processes. Reaction rate isgreatly enhanced and thus residence time may be reduced. Some ofthe technologies have been applied successfully in commercial production.We do not specifically review intensification technologiesinvolving advanced materials such as new heterogenous catalystsin packed or fluidized bed reactors. The goal of this review is to givean overall summary of process intensification technologies whichenhance physical processes including heat, mass, and momentumtransfer in the context of biodiesel synthesis.2. Novel reactors2.1. Static mixersStatic mixers consist of specially designed motionless geometricelements enclosed within a pipe or a column and create effectiveradial mixing of two immiscible liquids as they flow through themixer. Recently, they have been used in continuous biodiesel synthesisin combination with other equipment [1,2].Thompson and He [3] used a stand-alone closed-loop staticmixer system as a continuous-flow reactor to produce biodieselfrom canola oil with methanol when sodium hydroxide wasused as catalyst. The experimental setup is shown in Fig. 1.The system is composed of two stainless steel static reactors(4.9 mm ID × 300 mm long) including 34 fixed right- and left-handhelical mixing elements. High quality biodiesel which met theASTM D6584 specification was obtained after optimization ofexperimental conditions. The most favorable conditions for completenessof reaction included operation at a temperature of 60 ◦ Cand using a concentration of 1.5% sodium hydroxide catalyst and areaction time of 30 min. The total glyceride content was lower than0.24 wt% when the molar ratio of methanol to oil is 6:1. Boucheret al. [4] also reported a reactor/separator design involving a staticmixer for continuous biodiesel production and product separationas shown in Fig. 2. Pretreated waste canola oil and the solution ofpotassium hydroxide in methanol flow into a static mixer, whichwas exploited as an injector into a reaction chamber with no movingparts. Emulsified reactants were released into the chamberfrom the mixer with decreasing bulk velocity and separated intotwo phases under laminar flow conditions in the main body ofthe reactor. The less dense biodiesel phase separated as an upperFig. 2. Diagram of laminar flow reactor/separator [4] (Courtesy of American ChemicalSociety).layer. The glycerol phase, which had a higher density, settled as thelower layer. This required that the bulk flow velocity be set to avalue which was lower than the settling velocity of glycerol. Thenew reactor system obtained conversions greater than 99% withsimultaneous removal of 70–99% of the glycerol after 6 h continuousrunning. This was achieved at slightly elevated temperatures(40–50 ◦ C), using an overall feed rate of 1.2 L/min, a 6:1 molar ratioof methanol to vegetable oil triglycerides, and a 1.3 wt% catalyst.Static mixers present the advantage of low maintenance and operatingcost and low space requirement because they have no movingparts. However, the mixing process relies mainly on slow, unforcedmolecular inter-diffusion in the laminar regime and therefore reactionsare still slow.2.2. Micro-channel reactorsMicro-channel reactors achieve rapid reaction rates by improvingthe efficiency of heat and mass transfer and utilizing highsurface area/volume ratio and short diffusion distance [5]. Canter[6] reported that biodiesel could be produced in a microreactorat mild conditions. Yields of greater than 90% biodiesel aftera residence time of 4 min were reported. Sun et al. [7] studiedKOH-catalyzed transesterification of unrefined rapeseed oil andcottonseed oil with methanol in capillary microreactors with innerdiameters of 0.25 mm. At a 5.89 min residence time, they obtained a99.4% yield of methyl esters at a catalyst concentration of 1 wt% KOHand using a 6:1 molar ratio of methanol to oil and at a temperatureof 60 ◦ C. Finally, Wen et al. [8] investigated zigzag micro-channelreactors for continuous alkali-based biodiesel synthesis. The configurationof zigzag micro-channel reactors with narrower channelsize and more turns is shown in Fig. 3. This type of reactor wasFig. 3. Representative configuration of a zigzag micro-channel reactor [8] (Courtesyof Elsevier).

Z. Qiu et al. / Chemical Engineering and Processing 49 (2010) 323–330 325cost. The short length-to-diameter of this reactor decreases capitalcost and allows it to be scaleable.2.4. Cavitational reactorsFig. 4. The configuration of oscillatory flow reactor [9] (Courtesy of John Wiley &Sons Inc.).shown to intensify the biodiesel production process by obtainingsmaller droplets compared to those micro-channel reactors with T-or Y-flow structures. At a residence time of 28 s and a temperatureof 56 ◦ C, the yield of methyl ester reached 99.5% in an optimizedzigzag micro-channel reactor using a 9:1 molar ratio of methanolto oil and a catalyst concentration of 1.2 wt% sodium hydroxide.Under these conditions it was reported that less energy consumptionwas needed for the same amount of biodiesel than in the case ofa conventional stirred reactor. This was attributed to the high heattransfer rates achieved in the micro-channel reactor. The microchannelreactor is smaller in size offering reductions in footprintrequirements, construction and operating costs. Another advantageof micro-channel reactors is ease of scale-up which may be readilyachieved by adding more reactors of the same proven dimensions inparallel. This approach can reduce the risks associated with scalingup of conventional reactors.2.3. Oscillatory flow reactorsOscillatory flow reactors are tubular reactors in which orificeplate baffles are equally spaced and produce oscillatory flow usinga piston drive as shown in Fig. 4. When a bulk fluid is introducedinto the reactor, an oscillatory motion interacts with it and intensifiesradial mixing, with enhancements in mass and heat transferwhilst maintaining plug flow. The reactor can achieve long residencetimes because the degree of mixing is not directly dependentupon the Reynolds number of the bulk flow through it, but is mainlyrelated to the oscillatory conditions. Hence the oscillatory flowreactor can be designed with a short length-to-diameter ratio andimproves the economy of biodiesel production due to smaller “footprint”,lower capital and pumping cost, and easier control. Harveyet al. [9] developed a continuous oscillatory flow reactor (OFR)to produce saleable biodiesel from rapeseed oil in a pilot-scaleplant shown in Fig. 5. The reactor is composed of two verticallypositioned jacketed QVF tubes of 1.5 m length and 25 mm internaldiameter. Conversions of biodiesel up to 99% were achieved after30minat50 ◦ C using a molar ratio of methanol to rapeseed oil of1.5 and in the presence of a sodium hydroxide catalyst. Anotherbiodiesel pilot plant using an oscillatory flow reactor was demonstratedby the Polymer Fluids Group in the University of Cambridge[10]. It has been in operation since 2004 and producing at a rate of25 l/h.One of the advantages of this technology is the very low molarratio of methanol to oil they applied. It is lower than the stoichiometricratio (3:1) required and reduces significantly the operatingCavitational reactors use acoustic energy or flow energy togenerate cavitation phenomena, which results in process intensification.During the process of cavitation, the violent collapse of thecavities produced by the pressure changes from sound and flowenergy releases large magnitude of energy over a small location,and brings about very high temperatures and pressures [11]. Apicture of hydrodynamic cavitation is shown in Fig. 6. Cavitationalso intensifies the mass transfer rate by generation of local turbulenceand liquid micro-circulation in the reactor [13,14]. All ofthem contribute to the intensification of processes which underother conditions are limited by mass transfer and heat transfer.Kelkar et al. [15] investigated two different reactors based on acousticand hydrodynamic cavitation and compared their performancein biodiesel synthesis from vegetable oils. At a reaction time of15 min, more than 90% yield of biodiesel was obtained duringtransesterification of vegetable oils with methanol in the presenceof sodium hydroxide in these two reactors. From the pointof view of energy efficiency of mixing, hydrodynamic cavitation(1 × 10 −4 to 2 × 10 −4 g/J) is about 40 times more efficient thanacoustic cavitation (5 × 10 −6 to 2 × 10 −5 g/J) and 160–400 timesmore efficient than conventional mixing methods [15]. Moreover,it is difficult for sonochemical reactors to be scaled up becauseacoustic cavitation relies on a source of vibrations. The scale-up ofthe hydrodynamic cavitation reactors is relatively easier and thusthey provide potential for commercial scale applications becauseextensive information about the fluid dynamics downstream of theconstriction is readily available [16,17].Arisdyne System, Inc. [18] used controlled flow cavitation (CFC)technology to produce higher grade ASTM 6751 B100 biodiesel asshown in Fig. 7. The reactor consisted of one or more controlledflow cavitation zones or flow constrictions achieved by a diaphragmwith one or more orifices. The pressure drop across the constrictionsis controlled by the size of orifices, the flow rate of reactionmixture and, by a localized hydraulic resistance downstream of theconstriction. Precise process control ensures a tightly controlled,repeatable droplet size distribution by cavitation during the process.The conversion of fatty acids to fatty acid alkyl esters reached99% after passing through four cavitation constrictions in serieswith 1:6 molar ratio of fatty acids to methanol at 60 ◦ C only inresidence time of microseconds.Another reactor design which uses the principle of cavitationalmixing is described by Hydro Dynamics, Inc. [19]. Here a Shock-Wave Power Reactor (SPR) based on “controlled cavitation” isdescribed for process intensification of the transesterification reactionin biodiesel production. A commercial SPR is shown in Fig. 8.One difference between the SPR and above the controlled flow cavitationreactor is the mechanism of producing flow cavitation. TheShockWave Power Reactor works by taking a feedstock, methanoland catalyst, into the machine housing, where it is passed throughthe generator’s spinning cylinder. The specific geometry of cavitiesin the cylinder and rotational speed creates pressure differenceswithin the liquids where tiny bubbles form and collapse. The cavitationis controlled so that the bubbles collapse only inside thecavities and away from the metal surfaces and therefore reducethe risk of damaging the material of construction. The shock wavesincrease the surface area of the compounds being mixed so thata higher mass transfer rate occurs. It takes only several seconds tocomplete transesterification of vegetable oils or animal fats. Hence,the reactor allows the use of a variety of feedstocks with a broaderrange of free fatty acid concentrations because short reaction timeleads to less saponification and emulsification.

326 Z. Qiu et al. / Chemical Engineering and Processing 49 (2010) 323–330Fig. 5. Schematic of oscillatory flow reactor for biodiesel production [9] (Courtesy of John Wiley & Sons Inc.).2.5. Rotating/spinning tube reactorsFig. 6. Picture of hydrodynamic cavitation [12] (Courtesy of Roger E.A. Arndt).The rotating, or spinning tube reactor is a shear reactor consistingof two tubes. A scheme of the rotating tube reactor is shownin Fig. 9. One inner tube rotates rapidly within another concentricstationary outer tube. There is a very narrow annular gap betweenthe outer tube and the inner tube. Once reactants are introducedinto the gap, Couette flow is induced and the two liquids are mixedinstantaneously and move through the gap as a coherent thin filmdue to the high shear rate. Couette flow leads to high mass transferrate and very short mixing time. The thin film presents a verylarge interfacial contact area. Hence, the rate of reaction betweenreactants is enhanced. Less reaction time and mixing power inputare required compared to conventional reactors.Four Rivers BioEnergy Company, Inc. [22] utilized the technologyand developed a commercial Spinning Tube in a Tube (STT)system for biodiesel production as shown in Fig. 10. The STT reactoraccelerates the rates of chemical reactions by up to three ordersof magnitude. The transesterification reaction of soybean oil andmethanol for biodiesel production is conducted at a residence timeof 0.5 s.Lodha and Jachuck [23] investigated a rotating tube reactor toproduce biodiesel from canola oil with sodium hydroxide as catalyst.Within 40 s, conversions to biodiesel greater than 98% wereattained under mild operating temperatures in the range 40–60 ◦ C,and at atmospheric pressure.Fig. 7. Schematic of a biodiesel process using a controlled flow cavitation apparatus[18] (Courtesy of Arisdyne System Inc.). 10: reaction mixture; 12: controlled cavitationapparatus; 20: flow-through channel; 21: localized flow constriction; 22:orifice; 23, 24: chamber; 25: outlet; 26: localized hydraulic resistance; 30: reactionproduction.Fig. 8. A commercial SPR reactor for biodiesel production developed by HydroDynamics, Inc. [20] (Courtesy of Hydro Dynamics, Inc.).

Z. Qiu et al. / Chemical Engineering and Processing 49 (2010) 323–330 327Fig. 9. Schematic of a spinning tube reactor [21] (Courtesy of Four Rivers BioEnergyCompany, Inc.).The short residence time allows this reactor to handle feedstockswith high free fatty acid content. The size of the reactor is relativelysmall and it is easy to scale up.2.6. Microwave reactorsMicrowave reactors are units that utilize microwave irradiationto transfer energy directly into reactants and thus accelerate therate of chemical reaction. Thus conversions are achieved in lesstime compared with similar reactors using conventional thermalheating. Recently, microwave reactors have been developed forbiodiesel synthesis. Because the mixture of vegetable oil, methanol,and alcohol contains both polar and ionic components, microwaveirradiation can play an active role in heating reactants to therequired temperature quickly and efficiently [24]. Breccia et al.[25] studied the transesterification of commercial seed oils withmethanol under microwave irradiation. In the presence of a varietyof catalysts, yields greater than 97% were achieved with reactiontimes of less than 2 min. Leadbeater and Stencel [26] usedmicrowave irradiation to enhance production of biodiesel in abatch reactor. They made biodiesel with a quantitative conversionobtained after heating vegetable oil, methanol (1:6 molar ratio) andpotassium hydroxide (1 wt%) to 50 ◦ C in a microwave apparatuscalled MARS and maintaining this temperature for 1 min. Finally,they [24] achieved continuous-flow preparation of biodiesel in theapparatus. The apparatus could produce approximately 6.1 L ofbiodiesel per minute with attainment of 99% conversion. Rudimen-Fig. 10. Spinning Tube in a Tube (STT) system developed by Four Rivers BioEnergyCompany, Inc. [22] (Courtesy of Four Rivers BioEnergy Company, Inc.).tary energy consumption studies showed that the continuous-flowpreparation of biodiesel using microwave heating proved to bemore energy efficient than the conventional synthesis of biodieselin large tank reactors. Leadbeater et al. [27] also used this apparatusto make biodiesel from vegetable oils and 1-butanol at flowrates up to 2.3 L/min with sulfuric acid or potassium hydroxide ascatalyst. Azcan and Danisman [28] reported that reaction times oftransesterification of cottonseed oil with methanol by microwaveirradiation were reduced to 7 min while conventional heating typicallyrequired reaction times of 30 min to obtain similar conversionat 60 ◦ C. In experiments to study microwave assisted transesterificationof rapeseed oil in the presence of 1.0% KOH, 93.7% of biodieselyield and 97.8% of biodiesel purity were obtained at 50 ◦ C using a5 min reaction time [29].All above novel reactors intensify biodiesel production processby improving the mixing of oil and methanol except the microwavereactor. The mixing intensity of static mixers depends on the flowrate of the reactants or superficial velocity besides the geometricconfiguration of mixers. The mean droplet size was 63 m ina static mixer when the superficial velocity was 1.3 m/s [30]. Inthe micro-channel reactor, the droplet size of emulsion decreasesand the specific interfacial area increases and hence reaction rateis enhanced greatly compare to the batch stirred reactors. Meandroplet sizes of down to 1.93 m were reported in a zigzag microchannelreactor at a residence time of 28 s, approximately one thirdthat observed in a batch stirred reactor at the residence time of1h [8]. The degree of mixing is also dependent on the oscillatorymotion or the oscillatory frequency and is not related to the bulkflow. The oscillatory Reynolds number of typically 700 is requiredto achieve good mixing [9]. Cavitational reactors utilizing ultrasonicmixing produced the minimum droplet size of about 140 nm, whichis three times smaller than the droplets obtained with impeller at1000 rpm at the same input energies [31]. In spinning tube reactors,high mixing intensity is achieved through the high shear forces createdin the narrow gap between the stator and the rapidly spinningrotor. The droplet sizes from cavitational reactors and spinning tubereactor are relatively smaller than other reactors.3. Reaction/separation coupled technologies3.1. Membrane reactorsMembrane reactors integrate reaction and membrane-basedseparation into a single process. They can increase the conversionof equilibrium-limited reactions by removing some products fromthe reactants stream via membranes. Dube et al. [32] exploited thepossibility of biodiesel production from canola oil and methanolusing a two-phase tubular membrane reactor as shown in Fig. 11.The pore size of the carbon membrane used in the reactor was0.05 m. The inner and outer diameters were 6 and 8 mm, respectively.Its length was 1200 mm giving 0.022 m 2 surface area. Thetransesterification of canola oil was performed via both acid- orbase-catalysis in the 300 ml membrane reactor in semi-batch modeat 60, 65 and 70 ◦ C and at different catalyst concentrations andfeed flow rates. At high flow rate of methanol/acid catalyst, theyobtained 65% conversion. When base was used as catalyst, conversionsof greater than 95% were attained at different flow rates. Theyfound that the microporous carbon membrane selectively permeatesFAAE, methanol and glycerol while excluding bound glycerides(TG, DG, MG) from passing through. Hence, there are almost nobound glycerides in the permeate stream thus yielding high puritybiodiesel. The reactor was also used to handle feedstock with highFFA content because the membrane retained soap from the sidereaction and also retained other unreactable components. Cao etal. [33] investigated the effects of the pore size of the membrane on

328 Z. Qiu et al. / Chemical Engineering and Processing 49 (2010) 323–330Fig. 12. FAMEs production by esterification with methanol in a reactive distillationcolumn [43] (Courtesy of American Chemical Society).Fig. 11. Schematic of biodiesel production in a membrane reactor [32] (Courtesy ofElsevier).William Douglas [44] patented a process for low-sulfur, lowglycerinbiodiesel fuel by reactive distillation from feedstockscontaining fatty acids, glycerated fatty acids and glycerin.the performance of a semicontinuous reactor. The analysis resultsshow that no oil was found in the permeate stream and all permeatesamples obtained were biodiesel of high purity at 65 ◦ C. It alsoindicated that the oil droplets size in the membrane reactor weregreater than the pore sizes tested. The kinetics of canola oil transesterificationin a membrane reactor was also reported [34]. Thestudy showed that the membrane reactor could enhance reactionrate by the excellent mixing in the membrane reactor loop and thecontinuous removal of product from the reaction medium.3.2. Reactive distillationReactive distillation (RD) is a technique which combines chemicalreactions and product separations in one unit. Currently reactivedistillation has attracted more and more attention and been usedwidely [35–38] due to its many advantages over conventionalsequential processes, such as a fixed-bed-reactor followed by a distillationcolumn. One of the most important advantages of reactivedistillation is that conversion limitation is avoided by continuousin situ product removal for equilibrium-controlled reactions. Integrationof reaction and separation reduces capital investment andoperating costs.For reversible (trans)esterification reaction of vegetable oilswith alcohol, usually an excess amount of alcohol (6:1 or higheralcohol: oil molar ratio) is used to drive the equilibrium to theproduct side in order to obtain high conversion rates and highfinal equilibrium conversion. Since recovery of excess alcohol fromthe esters and glycerol streams involves additional operating cost,application of reactive distillation to biodiesel production may leadto a more efficient process. He et al. [39,40] reported a novelreactor system using RD for biodiesel production from canolaoil and methanol using KOH as catalyst. In this system, reactionat methanol: oil molar ratio of 4:1 yielded 95% conversion ratein about 3 min at 65 ◦ C column temperature. The experimentalconditions of a continuous-flow reactive distillation reactor wereoptimized for biodiesel production [41,42]. The upward-flowingmethanol vapor served as an agitator in the reactant mixture, providinguniform mixing while it was bubbling through the liquidphase on each plate. This process has several advantages over conventionalbiodiesel production processes.(1) Short reaction time and high unit productivity, (2) no excessalcohol requirement, (3) lower capital costs due to the small size ofRD and no need for additional separation units. Fig. 12 shows theFAME production process based on reactive distillation.3.3. Centrifugal contactorsThe centrifugal contactor is another process intensification technologybecause it integrates reaction and centrifugal separationinto a unit. It consists of a mixing zone and a separating zone asshown in Fig. 13. As the rotor in the contactor is rapidly rotatingwithin a stationary cylinder, it can achieve intense mixing and goodmass transfer by high shear stress and quick phase separation byhigh centrifugal force simultaneously. However, the residence timein a conventional centrifugal contactor is as low as about 10 s andcannot allow reaction to reach equilibrium. The researchers [46] atOak Ridge National Laboratory modified the design of a centrifugalcontactor and achieved the control of residence time in mixingzone. They used the updated centrifugal contactor to continuouslyproduce biodiesel via base-catalyzed production. Fig. 14 presentsthe experimental setup for biodiesel production using the modifiedcentrifugal contactor. At 60 ◦ C more than 99% conversion wasachieved after about 1 min using a volumetric phase ratio of 5:1 oilto methoxide and potassium hydroxide as catalyst when rotor wasspinning in 3600 rpm. Good product separation was realized after3 min. The technology has been commercialized with Nu-Energie,LLC [47].Kraai et al. [45] tested the performance of a centrifugal contactseparator for the continuous production of biodiesel from sunfloweroil. At an elevated temperature of 60 ◦ C, a yield of 96% ofFAME was achieved after about 40 min using a 6:1 molar ratio ofmethanol to oil and 1% NaOMe as catalyst at 30 Hz (1800 rpm).Fig. 13. Schematic of a centrifugal contactor [46] (Courtesy of John Wiley & SonsInc.).

Z. Qiu et al. / Chemical Engineering and Processing 49 (2010) 323–330 329Fig. 14. Experimental setup for biodiesel synthesis using centrifugal contactor [47](Courtesy of American Institute of Chemical Engineers).The study showed that the flow rate of two phases affected theconversion due to the change of residence time.4. SummaryBiodiesel production can be enhanced by process intensificationtechnologies. Each technology has the potential to improve productionefficiency and thus reduce operating cost of the process.Table 1 summarizes the advantages of these technologies used incontinuous biodiesel production over traditional stirred tank reactors.A rudimentary assessment of reaction time, energy efficiency,operating/capital cost, the difficulty of temperature control, andcurrent status is presented in the table. Most of these reactors orprocess technologies increase the rate of reaction by intensifyingtransport process and mixing between alcohol and oil. In summary,the static mixers can accomplish effective radial mixing asfluids pass through it. Micro-channel reactors improve heat andmass transfer efficiency due to short diffusion distance and highvolume/surface area. Oscillatory reactors enhance radial mixingand transport process by independent and controlled oscillatorymotion. In cavitational reactors, the collapse of cavity or bubblesproduces high temperature and pressure and turbulence locallyresulting in rapid reaction rate. Couette flow from high shear ratein spinning tube reactor with fine gap leads to high mass transferrate and intensive mixing. Microwave reactors increase reactionrate through direct and efficient heating by microwave irradiationreplacing conventional thermal heating. Membrane reactorsachieve high reaction rate by selective removal of bound glycerinfrom products via a membrane. Reactive distillation improvesbiodiesel production efficiency via in situ separation of productionsby distillation. Rates of transesterification can be intensified due touniform mixing by exploiting high shear forces which may producedin reactor geometries involving for example narrow flowfield between a rotating surface and stationary surface. High reactionrate means less reaction time to reach high equilibrium yieldof biodiesel or complete the conversion of oil, and high selectivitydue to less side reactions allowed. High conversion can be reachedat several minutes or even several seconds of reaction time. Potentially,biodiesel can be made continuously which increases processthroughput.Another common characteristic of process intensification technologiesis the small “footprint” required compared to conventionalequipment. In micro-channel reactors the diameter of channel isonly several hundred micrometers. Oscillatory flow reactors canbe designed in shorter length-diameter ratio. The coupled reaction/separationtechnologies reduce down processing procedures.Reactive distillation systems can realize recovery of alcohol fromproducts simultaneously. Centrifugal contactors may obviate theneed for a separate separation process for recovery of the glycerolfrom biodiesel product. Hence, small size of reactors and less processingprocesses reduce the cost of construction and maintenance.Another advantage from enhanced reaction rate allows less moralratio of alcohol to oil. The oscillatory flow reactor achieved highconversion with a 1.5:1 molar ratio of oil to methanol. These processintensification technologies also improve the flexibility of biodieselproduction. Cavitational reactors and spinning tube in tube reactorsare able to handle different feedstocks with a wider range offree fatty acid concentration due to less reaction time. Efficient heattransfer along with high mass transfer rate brings about less energyconsumption in mixing than conventional stirred reactors and goodtemperature control. Cavitational reactors and microwave reactorshave very high energy efficiency. Finally, these technologies canrealize easy scale-up and commercialization due to small size. As aresult, some technologies have been scaled-up and commercializedsuccessfully.The quality of biodiesel is a crucial issue in commercial production.Biodiesel quality depends on not only reaction processbut also downstream processing like product separation, waterwashing, and methanol recovery. However, the PI technologies forbiodiesel production discussed here focus on reaction or/and productseparation. These novel reactors can achieve low total glycerin(

330 Z. Qiu et al. / Chemical Engineering and Processing 49 (2010) 323–3305. ConclusionsProcess intensification technologies have significant potentialfor enhancement of biodiesel production. Enhancement intransport processes and higher reaction rates provide the scopefor continuous production. Hence higher conversion yields arepossible, under milder conditions and involving reduced molarratios of alcohol to oil, lower reaction temperature and catalystconcentration than conventional stirred reactors. Some processintensification technologies offer the flexibility to process a varietyof feedstocks. Compared to conventional tank systems, thesetechnologies are proved more energy efficient because of enhancedheat transfer. Their small “footprint” allows them to be scaled upeasily and reduces the capital and operating cost and thus increasesprofit.References[1] H. Noureddini, D. Harkey, V. Medikonduru, A continuous process for the conversionof vegetable oils into methyl esters of fatty acids, J. Am. Oil Chem. Soc.75 (1998) 1755–1783.[2] C.L. Peterson, J.L. Cook, J.C. Thompson, J.S. Taberski, Continuous flow biodieselproduction, Appl. Eng. Agric. 18 (2002) 5–11.[3] J.C. Thompson, B.B. He, Biodiesel production using static mixers, Trans. ASABE50 (2007) 161–165.[4] M.B. Boucher, C. Weed, N.E. Leadbeater, B.A. Wilhite, J.D. Stuart, R.S. Parnas,Pilot scale two-phase continuous flow biodiesel production via novel laminarflow reactor-separator, Energy & Fuels 23 (2009) 2750–2756.[5] J. Kobayashi, Y. Mori, S. Kobayashi, Multiphase organic synthesis in microchannelreactors, Chem. Asian J. 1 (2006) 22–35.[6] N. Canter, Making biodiesel in a microreactor, Tribol. Lubr. Technol. 62 (2006)15–17.[7] J. Sun, J. Ju, L. Ji, L. Zhang, N. Xu, Synthesis of biodiesel in capillary microreactors,Ind. Eng. Chem. Res. 47 (2008) 1398–1403.[8] Z. Wen, X. Yu, S.-T. Tu, J. Yan, E. Dahlquist, Intensification of biodiesel synthesisusing zigzag micro-channel reactors, Bioresour. Technol. 100 (2009)3054–3060.[9] A.P. Harvey, M.R. Mackley, T. Seliger, Process intensification of biodieselproduction using a continuous oscillatory flow reactor, J. Chem. Technol.Biotechnol. 78 (2003) 338–341.[10] http://www.ceb.cam.ac.uk/pages/biofuels.html.[11] T.G. Leighton, The Acoustic Bubble, Academic Press, London, UK, 1994.[12] http://cav.safl.umn.edu/images/gallery/safl-02-supercav.jpg.[13] F.R. Young, Cavitation, McGraw Hill, London, UK, 1989.[14] J.L. Luche, Synthetic Organic Chemistry, Plenum Press, New York, USA, 1999.[15] M.A. Kelkar, P.R. Gogate, A.B. Pandit, Cavitation as a novel tool for processintensification of biodiesel synthesis, in: Proceedings of the 6th InternationalSymposium on Catalysis in Multiphase Reactors, Pune, India, 2007.[16] P.R. Gogate, A.B. Pandit, Hydrodynamic cavitation: a state of the art review, Rev.Chem. Eng. 17 (2001) 1–85.[17] P.R. Gogate, A.B. Pandit, Engineering design methods for cavitation reactors. II.Hydrodynamic cavitation, AIChE J. 46 (2000) 1641–1649.[18] O.V. Kozyuk, Apparatus and method for producing biodiesel from fatty acidfeedstock, US 2009/0043118 A1.[19] D.G. Mancosky, D.A. Armstead, T. McGurk, G. Hopkins, K. Hudson, The use of acontrolled cavitation reactor for bio-diesel production, in: Proceedings of theAIChE Spring Meeting, Houston, USA, 2007.[20] http://www.hydrodynamics.com/product pics2.htm.[21] http://www.rccostello.com/STT.html.[22] http://www.kreido.com/downloads/ps stt tech.pdf.[23] H. Lodhar, R.J.J. Jachuck, Intensified biodiesel reaction using continuous rotatingtube reactor technology, in: Proceedings of the AIChE Annual Meeting, Salt LakeCity, USA, 2007.[24] T.M. Barnard, N.E. Leadbeater, M.B. Boucher, L.M. Stencel, B.A. Wilhite,Continuous-flow preparation of biodiesel using microwave heating, Energy &Fuels 21 (2007) 1777–1781.[25] A. Breccia, B. Esposito, G.B. Fratadocchi, A. Fini, Reaction between methanoland commercial seed oils under microwave irradiation, J. Microwave PowerElectromagn. Energy 34 (1999) 3–8.[26] N.E. Leadbeater, L.M. Stencel, Fast, easy preparation of biodiesel usingmicrowave heating, Energy & Fuels 20 (2006) 2281–2283.[27] N.E. Leadbeater, T.M. Barnard, L.M. Stencel, Batch and continuous-flow preparationof biodiesel derived from butanol and facilitated by microwave heating,Energy & Fuels 22 (2008) 2005–2008.[28] N. Azcan, A. Danisman, Alkali catalyzed transesterification of cottonseed oil bymicrowave irradiation, Fuel 86 (2007) 2639–2644.[29] N. Azcan, A. Danisman, Microwave assisted transesterification of rapeseed oil,Fuel 87 (2008) 1781–1788.[30] D. Frascari, M. Zuccaro, A. Paglianti, A pilot-scale study of alkali-catalyzed sunfloweroil transesterification with static mixing and with mechanical agitation,Energy & Fuels 22 (2008) 1493–1501.[31] P. Wu, Y. Yang, J.A. Colucci, E.A. Grulke, Effect of ultrasonication on droplet sizein biodiesel mixtures, J. Am. Oil Chem. Soc. 84 (2007) 877–884.[32] M.A. Dube, A.Y. Tremblay, J. Liu, Biodiesel production using a membrane reactor,Bioresour. Technol. 98 (2007) 639–647.[33] P. Cao, A.Y. Tremblay, M.A. Dube, K. Morse, Effect of membrane pore size on theperformance of a membrane reactor for biodiesel production, Ind. Eng. Chem.Res. 46 (2007) 52–58.[34] P. Cao, A.Y. Tremblay, M.A. Dube, Kinetics of canola oil transesterification in amembrane reactor, Ind. Eng. Chem. Res. 48 (2009) 2533–2541.[35] G.J. Harmsen, Reactive distillation: the front-runner of industrial process intensification:a full review of commercial applications, research, scale-up, designand operation, Chem. Eng. Process. 46 (2007) 774–780.[36] F. Omota, A.C. Dimian, A. Bliek, Fatty acid esterification by reactive distillation.Part 1. Equilibrium-based design, Chem. Eng. Sci. 58 (2003) 3159–3174.[37] F. Omota, A.C. Dimian, A. Bliek, Fatty acid esterification by reactive distillation.Part 2. Kinetics-based design for sulphated zirconia catalysts, Chem. Eng. Sci.58 (2003) 3175–3185.[38] L.G. Matallana, L.F. Gutiérrez, C.A. Cardona, Biodiesel production by reactive distillation,in: Empromer 2005-2nd Mercosur Congress on Chemical Engineeringand 4th Mercosur Congress on Process Systems Engineering, Agosto, 2005.[39] B.B. He, A.P. Singh, J.C. Thompson, A novel continuous-flow reactor using reactivedistillation for biodiesel production, Trans. ASAE 49 (2006) 107–112.[40] A.P. Singh, B.B. He, J.C. Thompson, A continuous-flow reactor using reactivedistillation for biodiesel production from seed oils, in: Proceedings of theASAE/CSAE Annual International Meeting, Ottawa, Canada, 2004.[41] B.B. He, A.P. Singh, J.C. Thompson, Experimental optimization of a continuousflowreactive distillation reactor for biodiesel production, Trans. ASAE 48 (2005)2237–2243.[42] A.P. Singh, B.B. He, J.C. Thompson, Experimental optimization of acontinuous-flow reactive distillation reactor for biodiesel production via transesterification,in: Proceedings of the ASAE Annual International Meeting,Tampa, Florida, 2005.[43] A.A. Kiss, A.C. Dimian, G. Rothenberg, Biodiesel by catalytic reactive distillationpowered by metal oxides, Energy & Fuels 22 (2008) 598–604.[44] M. William Douglas, Production of biodiesel fuels which are low in glycerin andsulfur, WO patent, 2008/112881 A1, 2008; US patent, 2008/0223752 A1, 2008.[45] G.N. Kraai, F. van Zwol, B. Schuur, H.J. Heeres, J.G. de Vries, Two-phase(Bio)catalytic reactions in a table-top centrifugal contact separator, Angew.Chem. Int. Ed. 47 (2008) 3905–3908.[46] J. McFarlane, J.F. Birdwell Jr., C. Tsouris, H.L. Jennings, Process intensification incontinuous base-catalyzed biodiesel production, in: Proceedings of the AIChEAnnual Meeting, 2008.[47] http://www.nu-industries.com/our-technology.php.

Applied Energy 87 (2010) 1083–1095Contents lists available at ScienceDirectApplied Energyjournal homepage: www.elsevier.com/locate/apenergyA review on biodiesel production using catalyzed transesterificationDennis Y.C. Leung * , Xuan Wu, M.K.H. LeungDepartment of Mechanical Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong, ChinaarticleinfoabstractArticle history:Received 17 June 2009Received in revised form 6 October 2009Accepted 7 October 2009Available online 7 November 2009Keywords:BiodieselAlkali-catalyzed transesterificationFeedstockPurificationMass transferBiodiesel is a low-emissions diesel substitute fuel made from renewable resources and waste lipid. Themost common way to produce biodiesel is through transesterification, especially alkali-catalyzed transesterification.When the raw materials (oils or fats) have a high percentage of free fatty acids or water, thealkali catalyst will react with the free fatty acids to form soaps. The water can hydrolyze the triglyceridesinto diglycerides and form more free fatty acids. Both of the above reactions are undesirable and reducethe yield of the biodiesel product. In this situation, the acidic materials should be pre-treated to inhibitthe saponification reaction. This paper reviews the different approaches of reducing free fatty acids in theraw oil and refinement of crude biodiesel that are adopted in the industry. The main factors affecting theyield of biodiesel, i.e. alcohol quantity, reaction time, reaction temperature and catalyst concentration,are discussed. This paper also described other new processes of biodiesel production. For instance, theBiox co-solvent process converts triglycerides to esters through the selection of inert co-solvents thatgenerates a one-phase oil-rich system. The non-catalytic supercritical methanol process is advantageousin terms of shorter reaction time and lesser purification steps but requires high temperature and pressure.For the in situ biodiesel process, the oilseeds are treated directly with methanol in which the catalysthas been preciously dissolved at ambient temperatures and pressure to perform thetransesterification of oils in the oilseeds. This process, however, cannot handle waste cooking oils andanimal fats.Ó 2009 Elsevier Ltd. All rights reserved.Contents1. Introduction . . . ..................................................................................................... 10842. Biodiesel production with catalyzed transesterification . . . . . . ............................................................... 10852.1. Basic chemical reactions . ........................................................................................ 10852.2. Biodiesel production processes. . . . . . . . . . . . ........................................................................ 10862.2.1. Process flow chart . . . . . . . . . ............................................................................. 10862.2.2. Raw materials treatment . . . . ............................................................................. 10862.2.3. Pretreatment of acidic feedstocks . . . . . . . . . . . . . . . . .......................................................... 10872.2.4. Catalyst and alcohol . . . . . . . . ............................................................................. 10872.2.5. Mixing and neutralization . . . ............................................................................. 10882.2.6. Transesterification and separation. . . . . . . . . . . . . . . . .......................................................... 10882.2.7. Refining crude glycerol. . . . . . ............................................................................. 10882.2.8. Purification of crude biodiesel ............................................................................. 10892.2.9. Quality control . . . . . . . . . . . . ............................................................................. 10892.3. Storage of biodiesel product . . . . . . . . . . . . . . ........................................................................ 10902.4. Main factors affecting the yield of biodiesel . ........................................................................ 10912.4.1. Alcohol quantity . . . . . . . . . . . ............................................................................. 10912.4.2. Reaction time . . . . . . . . . . . . . ............................................................................. 10912.4.3. Reaction temperature. . . . . . . ............................................................................. 10912.4.4. Catalyst concentration . . . . . . ............................................................................. 10913. Other processes of biodiesel production. ................................................................................. 10913.1. Biox co-solvent process. . ........................................................................................ 1091* Corresponding author. Tel.: +00 852 2859 7911; fax: +00 852 2858 5415.E-mail address: ycleung@hku.hk (D.Y.C. Leung).0306-2619/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.apenergy.2009.10.006

1084 D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–10953.2. Supercritical alcohol process . . . . . . . . . . ........................................................................... 10913.3. In situ biodiesel process . . . . . . . . . . . . . . ........................................................................... 10914. Conclusions. ........................................................................................................ 1092Acknowledgement . . . . . . . . . . . ........................................................................................ 1092References . ........................................................................................................ 10921. IntroductionDue to the depletion of the world’s petroleum reserves and theincreasing environmental concerns, there is a great demand foralternative sources of petroleum-based fuel, including diesel andgasoline fuels. Biodiesel, a clean renewable fuel, has recently beenconsidered as the best candidate for a diesel fuel substitution becauseit can be used in any compression ignition engine withoutthe need for modification [1]. Chemically, biodiesel is a mixtureof methyl esters with long-chain fatty acids and is typically madefrom nontoxic, biological resources such as vegetable oils [2–24],animal fats [16,17,23,24], or even used cooking oils (UFO) [10]. Table1 shows some feedstocks of biodiesel and their physicochemicalproperties. Vegetable oils are promising feedstocks for biodieselproduction since they are renewable in nature, and can be producedon a large scale and environmentally friendly [25]. Vegetableoils include edible and non-edible oils. More than 95% ofbiodiesel production feedstocks comes from edible oils since theyare mainly produced in many regions and the properties of biodieselproduced from these oils are much suitable to be used as dieselfuel substitute [26]. However, it may cause some problems such asthe competition with the edible oil market, which increases boththe cost of edible oils and biodiesel [11]. Moreover, it will causedeforestation in some countries because more and more forestshave been felled for plantation purposes. In order to overcomethese disadvantages, many researchers are interested in non-edibleoils which are not suitable for human consumption because of thepresence of some toxic components in the oils. Furthermore, nonedibleoil crops can be grown in waste lands that are not suitablefor food crops and the cost of cultivation is much lower becausethese crops can still sustain reasonably high yield without intensivecare [12,26]. However, most non-edible oils contain high freefatty acids. Thus they may require multiple chemical steps or alternateapproaches to produce biodiesel, which will increase the productioncost, and may lower the ester yield of biodiesel below thestandards [14,25,27]. Animal fats contain higher saturated fattyacids and normally exist in solid form at room temperature thatmay cause problems in the production process. Its cost is also higherthan vegetable oils [18]. UFO is not suitable for human consumptionbut is a feedstock for biodiesel production. Its usagesignificantly reduces the cost of biodiesel production. However,the quality of UFO may cause concern because its physical andchemical properties depend on the contents of fresh cooking oiland UFO may contain lots of undesired impurity, such as water,free fatty acids [18,26,28,29]. Since the cost of raw materials accountsabout 60–80% of the total cost of biodiesel production,choosing a right feedstock is very important [18,26]. Also, the yieldand properties of biodiesel products produced from different feedstockswould be quite different from each other. Table 2 showssome physicochemical properties of biodiesel from different feedstocksand their yields under different production conditions[7,9,11,14,16,18,20,22,25–28,30–38]. Although at present biodieselTable 1Feedstocks for biodiesel production and their physicochemical properties.Type ofoilSpeciesMain chemical composition (fattyacid composition wt.%)Density(g/cm 3 )Flashpoint (°C)Kinematic viscosity(cst, at 40 °C)Acid value(mg KOH/g)Heating value(MJ/kg)ReferencesVegetable oilEdible oil Soybean C16:0, C18:1, C18:2 0.91 254 32.9 0.2 39.6 [6,15,18,19]Rapessed C16:0, C18:0, C18:1, C18:2 0.91 246 35.1 2.92 39.7 [3,8,15,18]Sunflower C16:0, C18:0, C18:1, C18:2 0.92 274 32.6 – 39.6 [15,18,20]Palm C16:0, C18:0, C18:1, C18:2 0.92 267 39.6 a 0.1 – [2,11,18]Peanut C16:0, C18:0, C18:1, C18:2, C20:0, 0.90 271 22.72 3 39.8 [9,13,18,20]C22:0Corn C16:0, C18:0, C18:1, C18:2, C18:3 0.91 277 34.9 a – 39.5 [17,18]Camelina C16:0, C18:0, C18:1, C18:2 0.91 – – 0.76 42.2 [5,21]C18:3, C20:0, C20:1, C20:3Canola C16:0, C18:0, C18:1, C18:2, C18:3 38.2 0.4 [8,10]Cotton C16:0, C18:0, C18:1, C18:2 0.91 234 18.2 39.5 [17,18]Pumpkin C16:0, C18:0, C18:1, C18:2 0.92 >230 35.6 0.55 39 [22]abNonedibleoilOthersJatrophacurcasC16:0, C16:1, C18:0, C18:1, C18:2 0.92 225 29.4 28 38.5 [4,7,12]Pongamina C16:0, C18:0, C18:1, C18:2, C18:3 0.91 205 27.8 5.06 34 [14]pinnataSea mango C16:0, C18:0, C18:1, C18:2 0.92 – 29.6 0.24 40.86 [11]Palanga C16:0, C18:0, C18:1, C18:2 0.90 221 72.0 44 39.25 [14]Tallow C14:0, C16:0, C16:1, C17:0, C18:0, 0.92 – – – 40.05 [17,23]C18:1, C18:2Nile tilapia C16:0, C18:1, C20:5, C22:6, other 0.91 – 32.1 b 2.81 – [16]acidsPoultry C16:0, C16:1, C18:0, C18:1, C18:2,C18:30.90 – – – 39.4 [23,24]Used Depends on fresh cooking oil 0.90 – 44.7 2.5 – [10]cooking oilKinematic viscosity at 38 °C, mm 2 /s.Kinematic viscosity at 37 °C, mm 2 /s.

D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095 1085cannot entirely replace petroleum-based diesel fuel, there are severaldistinct advantages of biodiesel over diesel fuel. Biodiesel hashigher combustion efficiency, cetane number than diesel fuel [39].It is biodegradable and more than 90% biodiesel can be biodegradedwithin 21 days [40,41]. Biodiesel has lower sulfur and aromaticcontent than diesel fuel and that means it will not emit lots of toxicgas [42,43]. Moreover, it reduces most exhaust emissions exceptNO x , such as monoxide, unburned hydrocarbons, and particulatematter [44–48].A number of methods are currently available and have beenadopted for the production of biodiesel fuel. There are four primaryways to produce biodiesel (Table 3): direct use and blending of rawoils [49–53], micro-emulsions [54], thermal cracking [55–60], andtransesterification [61]. The most commonly used method for convertingoils to biodiesel is through the transesterification of animalfats or vegetable oils, which forms the basis of the present paper.The objective of this paper is to give an overview on the technologicaladvancement of producing biodiesel through transesterification.The main factors affecting the yield of biodiesel and othertypes of transesterification will also be discussed.2. Biodiesel production with catalyzed transesterificationtransesterification. The simplified form of its chemical reaction ispresented in equationCH 2-O-CO-R 1CH 2-OH R-O-CO-R 1| (Catalyst) |CH-O-CO-R 2+ 3ROH CH-OH R-O-CO-R 2| |CH 2-O-CO-R 3CH 2-OH R-O-CO-R 3(Triglyceride) (Alcohol) (Glycerol) (Mixture of fatty acid esters)ð1Þwhere R 1 ,R 2 ,R 3 are long-chain hydrocarbons, sometimes calledfatty acid chains. Normally, there are five main types of chains invegetable oils and animal oils: palmitic, stearic, oleic, linoleic, andlinolenic. When the triglyceride is converted stepwise to diglyceride,monoglyceride, and finally to glycerol, 1 mol of fatty ester is liberatedat each step [61]. Usually, methanol is the preferred alcoholfor producing biodiesel because of its low cost.Vegetable oils and fats may contain small amounts of water andfree fatty acids (FFA). For an alkali-catalyzed transesterification,the alkali catalyst that is used will react with the FFA to form soap.Eq. (2) shows the saponification reaction of the catalyst (sodiumhydroxide) and the FFA, forming soap and water.2.1. Basic chemical reactionsR 1 —COOHðFFAÞþNaOHsodium hydroxide! R 1 COONa þ H 2 OðsoapÞ ðwaterÞð2ÞCommon vegetable oils or animal fats are esters of saturatedand unsaturated monocarboxylic acids with the trihydric alcoholglyceride. These esters are called triglycerides, which can reactwith alcohol in the presence of a catalyst, a process known asThis reaction is undesirable because the soap lowers the yield ofthe biodiesel and inhibits the separation of the esters from theglycerol. In addition, it binds with the catalyst meaning that morecatalyst will be needed and hence the process will involve a higherTable 2Physicochemical properties of biodiesel from different oil source and the yields under the production conditions.FeedstockKinematicviscosity (cst,at 40 °C)Density(g/cm 3 )SaponificationnumberIodinevalueAcidvalue(mg KOH/g)CetanenumberHeatingvalue(MJ/kg)Production conditions 1T (°C)P(min)MC (wt.%)Yield(%)ReferencesSoybean 4.08 0.885 201 138.7 0.15 52 40 65 90 1:12 CaO|8 (2.03% H 2 Oin >95 [18,20,27,33,35]methanol)Rapeseed 4.3–5.83 0.88– – – 0.25–0.45 49–50 45 65 120 1:6 KOH|1 95– [36]0.88896Sunflower 4.9 0.88 200 142.7 0.24 49 45.3 60 120 1:6 NaOH|1 97.1 [20,38]Palm 4.42 0.86– 207 60.07 0.08 62 34 Room 584 1:11 KF/ZnO|5.52 89.23 [18,20,32]0.9Peanut 4.42 0.883 200 67.45 – 54 40.1 60 120 1:6 NaOH|0.5 89 [9,18,20]Corn 3.39 0.88– 202 120.3 – 58–59 45 80 60 1:9 KOH|2 85– [20,25]0.8996Camelina 6.12–7 a 0.882– – 152– 0.08–0.52 – – Room 60 1:6 KOH|1.5 97.9 [31]0.888157Canola 3.53 0.88– 182 103.8 – 56 45 60 60 1:9 KOH|1 80– [20,25]0.995Cotton 4.07 0.875 204 104.7 0.16 54 45 65 90 1:6 NaOH|0.75 96.9 [20,37]Pumpkin 4.41 0.8837 202 115 0.48 – 38 65 60 1:6 NaOH|1 97.5 [20,22]Jatropha 4.78 0.8636 202 108.4 0.496 61–63 40–42 60 120 1:06 NaOH|1 98 [7,20,25]curcasPongamina 4.8 0.883 – – 0.62 60–61 42 65 180 1:6 KOH|1 97– [14,25–26,34]pinnata98Sea mango – – – – – – – 180 180 1:8 Sulfated zirconiz 83.8 [11]alumina|6Palanga 3.99 0.869 – – – – 41 66 4 1:12 b C 6 H 5 CH 3 +H 3 PO 4 |5 + 5, 85 [14]H 2 SO 4 |6.5; KOH|9(three step)Tallow – 0.856 244.5 126 0.65 59 – 60 1440 1:30 H 2 SO 4 |2.5 98.28 [30]Nile tilapia – – – 88.1 1.4 51 – 30 90 1:9 KOH|2 (ultrasound) 98.2 [16]Poultry – 0.867 251.23 130 0.25 61 – 50 1440 1:30 H 2 SO 4 |1.25 99.72 [30]Usedcookingoil4 – – – 0.15 – – 60 20 1:7 NaOH|1.1 94.6 [28]1 T = reaction temperature (°C); P = transesterification reaction period (min); M = molecular ratio of oil/methanol; C = amount of catalyst (wt.%).a Kinematic viscosity at 20 °C.b Volumetric ratio of oil/methanol.

1086 D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095Table 3Different methods of biodiesel production.Methods Definition Advantage Disadvantage Problems of using inenginesReferencesDirect use andblendingMicro-emulsionsThermal cracking(pyrolysis)TransesterificationDirect use as diesel fuel or blend withdiesel fuelA colloidal equilibrium dispersion ofoptically isotropic fluidmicrostructures with dimensionsgenerally in the 1–150 nm rangeformed spontaneously from twoimmiscible liquids and one or moreionic or non-ionic amphiphilesThe conversion of long-chain andsaturated substance (biomass basis)to biodiesel by means of heatThe reaction of a fat or oil with analcohol in the presence of catalyst toform esters and glycerolLiquid nature-portability Higher viscosity Coking and trumpetformationHeat content (80% of diesel fuel) Lower volatility Carbon depositsReadily available; renewability Reactivity ofOil ring sticking;unsaturatedthickening and gelling ofhydrocarbon chains the lubricating oilBetter spray patterns duringcombustionLower cetane numberIrregular injector needlesticking; incompletecombustionLower fuel viscosities Lower energy content Heavy carbon deposits;increase lubrication oilviscosityChemically similar to petroleumderivedgasoline and diesel fuelRenewability; higher cetanenumber; lower emissions; highercombustion efficiencyEnergy intensive andhence higher costDisposal of byproduct(glycerol andwaste water)[9–13][14,21]– [15–20]– [21–23]cost [62]. Water, originated either from the oils and fats or formedduring the saponification reaction, retards the transesterificationreaction through the hydrolysis reaction. It can hydrolyze the triglyceridesto diglycerides and forms more FFA. The typical hydrolysisreaction is shown in Eq. (3).CH 2-O-CO-R 1CH 2-OH| |CH-O-CO-R 2+ H 2O CH-O-CO-R 2+ R 1-COOH| |CH 2-O-CO-R 3CH 2-O-CO-R 3(Triglyceride) (Water) (Diglyceride) (FFA)ð3ÞHowever, the FFA can react with alcohol to form ester (biodiesel)by an acid-catalyzed esterification reaction. This reaction isvery useful for handling oils or fats with high FFA, as shown inthe equation below:R 1 —COOH þ ROHðFFAÞ ðalcoholÞ! Hþ R—O—CO—R 1 þ H 2 Oðfatty acid esterÞ ðwaterÞNormally, the catalyst for this reaction is concentrated sulphuricacid. Due to the slow reaction rate and the high methanol to oilmolar ratio that is required, acid-catalyzed esterification has notgained as much attention as the alkali-catalyzed transesterification[63].ð4Þ2.2.2. Raw materials treatmentThe raw materials, which can be vegetable oils, animal fats, orrecycled greases, used in the production of biodiesel contain triglycerides,free fatty acids, water, and other contaminants in variousproportions. Some crude vegetable oils contain phospholipidsthat need to be removed in a degumming step. Phospholipids canproduce lecithin, a commercial emulsifier [65]. Liu et al. comparedthe different degumming methods, such as membrane filtration,hydration, acid micelles degumming, supercritical extraction, etc.The characteristics of the raw oils should be investigated beforechoosing the suitable degumming method because differentdegumming methods have their advantages and disadvantages[66]. The oil can be separated through membrane filtration accordingto the average molecular weight or the particle size of phospholipids.Although degumming method can solve the problem,CatalystAlcoholMixingFFA _ 2.5wt%>Oils or FatsNeutralized oilTransesterificationFFA>2.5wt%Pretreatment2.2. Biodiesel production processes2.2.1. Process flow chartToday, most of the biodiesel is produced by the alkali-catalyzedprocess. Fig. 1 shows a simplified flow chart of the alkali-catalystprocess. As described earlier, feedstocks with high free fatty acidwill react undesirably with the alkali catalyst thereby formingsoap. The maximum amount of free fatty acids acceptable in an alkali-catalyzedsystem is below 2.5 wt.% FFA. If the oil or fat feedstockhas a FFA content over 2.5 wt.%, a pretreatment step isnecessary before the transesterification process [64] (see Section2.2.3).Re-neutralizationAlcohol RecoveryCrude GlycerolPhase separationCrude BiodieselPurificationQuality ControlBiodieselFig. 1. Simplified process flow chart of alkali-catalyzed biodiesel production.

D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095 1087its process involves two steps with the use of an organic solvent.Therefore, this method has no superiority on cost because of itscomplicated processing [67,68]. In the hydration process, becauseof phospholipids hydrophilicity, hot water can be added into theoil with stirring. The phospholipids solubility will be significantlyreduced, so it could be separated from the oil by natural settlement.The hydration method features a simple process, easy operation,and high yields refining, but non-hydratable phospholipidscannot be removed by this method, as reported by Verleyen et al.[69]. For removing non-hydratable phospholipids, critic acid orphosphoric acid can be added into the oil which is heated to70 °C, this is called the special micelles degumming method. Panet al. [70] found that the reaction time for this method is about5 min, followed by neutralization through dilute lye. The supercriticalextraction method is employed to separate the phospholipids.Moreover, by refining with supercritical CO 2 extraction, it caneffectively remove the free fatty acids and the peroxidation productsthat are in the crude oil. However, the high pressure requiredin this process will be the biggest challenge in its industrial application[71,72].The free fatty acids are removed in a refining step and excessfree fatty acids can be removed as soaps in a later pretreatmentstep. In addition, deodorization is another important step in theraw material treatment. During this step, steam, at 1–6 mm Hgpressure, is injected into the oil at 490–550 K in order to eliminatefree fatty acids, aldehydes, unsaturated hydrocarbons, and ketones,all of which cause undesirable odors and flavors in the oil [73].Next, in order to determine the percentage of FFA in the oils or fats,titration is performed. As described above, if the percentage of FFAis over 2.5 wt.%, pretreatment is necessary to reduce the content ofFFA. This step also determines the amount of caustic soda requiredin the neutralization step.2.2.3. Pretreatment of acidic feedstocksMany pretreatment methods have been proposed for reducingthe high free fatty acid content of the oils, including steam distillation[74], extraction by alcohol [75], and esterification by acidcatalysis[76]. However, steam distillation for reducing high freefatty acids requires a high temperature and has low efficiency. Becauseof the limited solubility of free fatty acids in alcohol, extractionby alcohol method needs a large amount of solvent and theprocess is complicated. Compared with the two former methods,esterification by acid-catalysis makes the best use of the free fattyacids in the oil and transforms it into biodiesel [75,77]. The commonpretreatment is esterification of the FFA with methanol inthe presence of acidic catalysts (usually sulphuric acid). The catalystscan be homogeneous acid-catalysts or solid acid-catalysts[78]. Compared with the former one, solid acid-catalysts offersome advantages for eliminating separation, corrosion, toxicity,and environmental problems, but the reaction rate is slower [76].As described earlier, free fatty acids will be converted to biodieselby direct acid esterification and the water needs to be removed. Ifthe acid value of the oils or fats is very high, one-step esterificationpretreatment may not reduce the FFA efficiently because of thehigh content of water produced during the reaction. In this case,a mixture of alcohol and sulphuric acid can be added into the oilsor fats three times (three-step pre-esterification). The time requiredfor this process is about 2 h and water must be removedby a separation funnel before adding the mixture into the oils orfats for esterification again [79]. Moreover, some researchers reducethe percent of FFA by using acidic ion exchange resins in apacked bed. Strong commercial acidic ion exchange resins can beused for the esterification of FFA in waste cooking oils but the lossof the catalytic activity maybe a problem [80–84].An alternative approach to reduce the FFA is to use iodine as acatalyst to convert free fatty acids into biodiesel. An obviousadvantage of this approach is that the catalyst (iodine) can be recycledafter the esterification reaction. Li et al. [85] found throughorthogonal tests, under the optimal conditions (i.e. iodine amount:1.3 wt.% of oils; reaction temperature: 80 °C; ratio of methanol tooils: 1.75:1; reaction time: 3 h) that the FFA content can be reducedto

1088 D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095Table 4Advantages and disadvantages at different types of catalysts used in the biodiesel production.Type Example Advantages Disadvantages ReferencesAlkaliHomogeneous NaOH, KOH High catalytic activity, low cost,favorable kinetics, modestoperation conditionsHeterogeneous CaO, CaTiO 3 , CaZrO 3 , CaO–CeO 2 , CaMnO 3 ,Ca 2 Fe 2 O 5 , KOH/Al 2 O 3 , KOH/NaY, Al 2 O 3 /KI,ETS-10 zeolite, alumina/silica supportedK 2 CO 3Noncorrosive, environmentallybenign, recyclable, fewer disposalproblems, easily separation, higherselectivity, longer catalyst lifetimesAcidHomogeneous Concentrated sulphuric acid Catalyze esterification andtransesterification simultaneously,avoid soap formationHeterogeneous ZnO/I 2 , ZrO 2 =SO 2 4 , TiO 2=SO 2 4 , carbon-basedsolid acid catalyst, carbohydrate-derivedcatalyst, Vanadyl phosphate, niobic acid,sulphated zirconia, Amberlyst-15, Nafion-NR50Enzymes Candida antarctica fraction B lipase,Rhizomucor mieher lipaseCatalyze esterification andtransesterification simultaneously,recyclable, eco-friendlyAvoid soap formation, nonpolluting,easier purificationLow FFA requirement, anhydrous conditions,saponification, emulsion formation, morewastewater from purification, disposableLow FFA requirement, anhydrous conditions,more wastewater from purification, highmolar ratio of alcohol to oil requirement,high reaction temperature and pressure,diffusion limitations, high costEquipment corrosion, more waste fromneutralization, difficult to recycle, higherreaction temperature, long reaction times,weak catalytic activityLow acid site concentrations, lowmicroporosity, diffusion limitations, highcost[97,101–104][33,99–101,103,104][97,102,103][82,98,103,104]Expensive, denaturation [27,82,104]2.2.5. Mixing and neutralizationThe purpose of mixing methanol with the catalyst is to producemethoxide which reacts with the base oils. Most of the catalysts(e.g. NaOH, KOH) are in solid form and do not readily dissolve intomethanol, it is best to start agitating the methanol in a mixer andadd the catalyst slowly and carefully [64]. Once the catalyst completelydissolves in the methanol, the methoxide is ready to beadded to the oil. Once the methoxide is added into the oil, a neutralizationreaction will immediately start. Some alkali catalystswill react with rudimental acids during the pretreatment step orwill react with the free fatty acids from the oil. Therefore, more catalystneeds to be added to complete the reaction.2.2.6. Transesterification and separationWhen the catalyst, alcohol, and oil are mixed and agitated in areaction vessel, a transesterification reaction will start. A stirredreactor is usually used as the reaction vessel for continuous alkali-catalyzedbiodiesel production. Recently, there is an increasedinterest in new technologies related to mass transfer enhancement.Maeda et al. [110] and Thanh et al. [111] produced biodiesel fromvegetable oils assisted by ultrasound which is a useful tool forstrengthening the mass transfer of immiscible liquids. Ultrasonicirradiation causes cavitation of bubbles near the phase boundarybetween immiscible liquid phases. The asymmetric collapse ofthe cavitation bubbles disrupts the phase boundary and startsemulsification instantly. Micro jets, formed by impinging one liquidto another, lead to intensive mixing of the system near thephase boundary. With the use of ultrasound biodiesel can be producedwithout heating because the cavitation may lead to a localizedincrease in temperature at the phase boundary and enhancethe reaction [110–112]. Moreover, Wen et al. [113] fabricated anew reaction vessel – Zigzag micro-channel reactor in recent yearsand found that less energy consumption for biodiesel synthesis canbe achieved by using this reactor.Leung and Guo [28] studied the effect of operating conditionson the product yield and pointed out that heating the oil prior tothe mixing can increase the reaction rate and hence shorten thereaction time. During this step, in order to speed up the reaction,mixing brings the oil, the catalyst, and the alcohol into intimatecontact while the temperature is kept just below the boiling pointof the alcohol (i.e. 64.5 °C for methanol). Normally, the reactionpressure is close to the atmospheric pressure to prevent the lossof alcohol, and excess alcohol is used to ensure total conversionof the oil to its esters. As previously mentioned, if the free fatty acidlevel or water level is too high, it may cause problems downstreamwith the saponification and the separation of the glycerol by-product.Therefore, the amount of water and free fatty acids in the feedstockoil should be monitored during the reaction.Once the transesterification reaction is completed, two majorproducts exist: esters (biodiesel) and glycerol. The glycerol phaseis much denser than the biodiesel phase and settles at the bottomof the reaction vessel, allowing it to be separated from the biodieselphase. Phase separation can be observed within 10 min and canbe completed within several hours of settling. The reaction mixtureis allowed to settle in the reaction vessel in order to allow the initialseparation of biodiesel and glycerol, or the mixture is pumpedinto a settling vessel. In some cases, a centrifuge may be used toseparate the two phases [86].Both the biodiesel and glycerol are contaminated with an unreactedcatalyst, alcohol, and oil during the transesterification step.Soap that may be generated during the process also contaminatesthe biodiesel and glycerol phase. Schumacher [114] suggested thatalthough the glycerol phase tends to contain a higher percentage ofcontaminants than the biodiesel, a significant amount of contaminantsis also present in the biodiesel. Therefore, crude biodieselneeds to be purified before use.2.2.7. Refining crude glycerolAlthough biodiesel is the desired product from the reactions,the refining of glycerol is also important due to its numerous applicationsin different industrial products such as moisturizers, soaps,cosmetics, medicines, and other glycerol products [115–117]. Itisone of the few products that has a good reactivity on sump oil,and is extremely effective for washing shearing shed floor, so itcan be used as a heavy duty detergent and degreaser. Whittington[118] reported that glycerol can even be fermented to produce ethanol,which means more biofuel can be produced.According to the statements of Van Gerpen et al. [86], typicallyproduced glycerol is about 50% glycerol or less in composition andmainly contains water, salts, unreacted alcohol, and unused catalyst.The unused alkali catalyst is usually neutralized by an acid.In some cases, hydrochloric or sulphuric acids are added into theglycerol phase during the re-neutralization step and produce saltssuch as sodium chloride or potassium sulphate, the latter can be

D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095 1089recovered for use as a fertilizer [119]. Generally, water and alcoholare removed to produce 80–88% pure glycerol that can be sold ascrude glycerol. In more sophisticated operations, the glycerol isdistilled to 99% or higher purity and sold in different markets [120].After the re-neutralization step, the alcohol in the glycerol phasecan be removed through a vacuum flash process or by other types ofevaporators. Usually, the alcohol vapor is condensed back into liquidand reused in the process. However, the alcohol may containwater that should be removed in a distillation column before thealcohol is returned to the process. The alcohol recovery step is moredifficult when the alcohol that is used, such as ethanol or isopropanol,forms an azeotrope with the water. Gerpen [121] proposed theuse of a molecular sieve to remove the water generated.2.2.8. Purification of crude biodieselAfter separation from the glycerol phase, crude biodiesel ismainly contaminated with residual catalyst, water, unreacted alcohol,free glycerol, and soaps that were generated during the transesterificationreaction [114]. Normally, crude biodiesel enters aneutralization step and then passes through an alcohol stripper beforethe washing step. In some cases, acid is added to crude biodieselto neutralize any remaining catalyst and to split any soap. Soaps reactwith the acid to form water soluble salts and free fatty acids. Gerpen[121] stated that neutralization before the washing step reducesthe materials required for the washing step and minimizes the potentialfor emulsions being formed during the washing step. Unreactedalcohol should be removed with distillation equipmentbefore the washing step to prevent excess alcohol from enteringthe wastewater effluent [86]. The primary purpose of this step is towash out the remnants of the catalyst, soaps, salts, residual alcohol,and free glycerol from the crude biodiesel. Generally, three main approachesare adopted for purifying biodiesel: water washing, drywashing [122], and membrane extraction [123,124]. These approachesare briefly shown in Table 5 and discussed in detail asfollows.2.2.8.1. Water washing. Since both glycerol and alcohol are highlysoluble in water, water washing is very effective for removing bothcontaminants. It also can remove any residual sodium salts andsoaps. The primary material for water washing is distilled warmwater or softened water (slightly acidic) [72,73]. Warm water preventsthe precipitation of saturated fatty acid esters and retards theformation of emulsions with the use of a gentle washing action.Softened water (slightly acidic) eliminates calcium and magnesiumcontamination and neutralizes any remaining alkali catalysts [86].After washing several times, the water phase becomes clear, meaningthat the contaminants have been completely removed. Then,the biodiesel and water phases are separated by a separation funnelor centrifuge [125]. Moreover, because of the immiscibility ofwater and biodiesel, molecular sieves and silica gels, etc., can alsobe used to remove water from the biodiesel [86]. The remainingwater can be removed from the biodiesel by passing the productover heated Na 2 SO 4 (25 wt.% of the amount of the ester product)overnight and then be removed by filtration [126]. However, thereare many disadvantages to this method, including an increasedcost and production time, polluting liquid effluent, product loss,etc. Moreover, emulsions can form when washing the biodieselmade from waste cooking oils or acidic feedstocks because of thesoap formation [127].2.2.8.2. Dry washing. Cooke et al. [122] used dry washing by replacingthe water with an ion exchange resin or a magnesium silicatepowder in order to remove impurities. These two dry washingmethods can bring the free glycerol level down and is reasonablyeffective for removing soaps. Both the ion exchange process andthe magnesol process have the advantage of being waterless andthus eliminate many of the problems outlined above. Althoughthe magnesol process has a better effect on the removal of methanolthan the ion resins, none of the products from this process fulfillthe limits specified in the EN Standard [128,129].2.2.8.3. Membrane extraction. Gabelman and Hwang [123] provedthat the contaminants can be removed by using a hollow fibermembrane extraction, such as polysulfone. In this method, a hollowfiber membrane (1 m long, 1 mm diameter) filled with distilledwater is immersed into the reactor (20 °C). The crude biodiesel ispumped into the hollow fiber membrane (flow rate: 0.5 ml/min;operating pressure: 0.1 MPa). Following this step, biodiesel ispassed over heated Na 2 SO 4 and then filtered to remove anyremaining water [124]. This approach effectively avoids emulsificationduring the washing step and decreases the loss during therefining process. The purity of the biodiesel obtained is about90% and the other properties conform to the ASTM standards[130]. It is a very promising method for purifying biodiesel.2.2.9. Quality controlFor commercial fuel, the finished biodiesel must be analyzedusing sophisticated analytical equipment to ensure it meets inter-Table 5Different approaches for purifying crude biodiesel.ApproachesWater washingDry washingMembraneextractionPrimarymaterialusedDistilledwarmwaterSoftenedwaterIonexchangeresinMagnesiumsilicatepowderPolysulfoneFunction Phases separation Advantage Disadvantage ReferencesPrevents precipitationof saturated fatty acidestersRetards the emulsionformationEliminates calciumand magnesiumcontaminationBrings the freeglycerol level downand removing soapsRemove thecontaminantsSeparation funnel,centrifuge, molecularsieves, silica gels, etc.Very effective inremoving contaminantsIncreased cost and productiontime, liquid effluent, product loss,emulsions formation– Waterless Overruns the limit in the ENStandard– Avoids the emulsionformation and decreasesthe refining lossProbably high cost and lowthroughput due to contaminantsexisted[46,72–74][69,75,76][70,71]

1090 D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095national standards. A few specifications have been set but theASTM D 6751 and EN 14214 standards are the most commonlyused standards. Even in blends with conventional diesel fuel, Mittelbach[131] stated that most people in the industry expect thebiodiesel blending stock to meet the relevant standard beforebeing blended. Some properties in the standard, such as the cetanenumber or density, can reflect the properties of the chemical compoundsthat make up the biodiesel, and other properties provide anindication of the quality of the production process [54]. Generally,biodiesel standards identify the parameters that pure biodieselmust meet before being used as a pure fuel or being blended withdistillate fuels [120,129,130].To ensure safe operation in diesel engines, the most importantaspects of the biodiesel product are the completion of the reaction,the removal of the free glycerol, residual catalyst and alcohol, andthe absence of free fatty acids [39,41,42,132]. As mentioned before,if the transesterification reaction is not complete then triglycerides,diglycerides, or monoglycerides may be left in the final product.Chemically, each of these compounds contains a glycerolmolecule. Fuel with excessive free glycerol may plug the fuel filtersand cause combustion problems in the diesel engine. Therefore, theASTM standard requires the total glycerol to be

D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095 1091The tanks should minimize the possibility of water contaminationand should be cleaned prior to use for biodiesel storage [137].2.4. Main factors affecting the yield of biodiesel2.4.1. Alcohol quantityMany researchers recognized that one of the main factorsaffecting the yield of biodiesel is the molar ratio of alcohol to triglyceride[28,47,61,138]. Theoretically, the ratio for transesterificationreaction requires 3 mol of alcohol for 1 mol of triglyceride toproduce 3 mol of fatty acid ester and 1 mol of glycerol. An excessof alcohol is used in biodiesel production to ensure that the oilsor fats will be completely converted to esters and a higher alcoholtriglyceride ratio can result in a greater ester conversion in a shortertime. The yield of biodiesel is increased when the alcohol triglycerideratio is raised beyond 3 and reaches a maximum. Furtherincreasing the alcohol amount beyond the optimal ratio will not increasethe yield but will increase cost for alcohol recovery [28]. Inaddition, the molar ratio is associated with the type of catalystused and the molar ratio of alcohol to triglycerides in most investigationsis 6:1, with the use of an alkali catalyst [47,138]. Whenthe percentage of free fatty acids in the oils or fats is high, suchas in the case of waste cooking oil, a molar ratio as high as 15:1is needed when using acid-catalyzed transesterification[28,133,139,140].2.4.2. Reaction timeFreedman et al. [141] found that the conversion rate of fattyacid esters increases with reaction time. At the beginning, the reactionis slow due to the mixing and dispersion of alcohol into the oil.After a while, the reaction proceeds very fast. Normally, the yieldreaches a maximum at a reaction time of

1092 D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095of its acyglycerols, the oilseeds are directly treated at ambient temperatureand pressure with a methanol solution in which the catalysthas been previously dissolved. That means that the oil in theoilseeds is not isolated prior to transesterification to fatty acid esters[152–154]. To reduce the alcohol requirement for high efficiencyduring in situ transesterification, the oilseeds need to bedried before the reaction takes place [155]. Milled oilseeds aremixed with alcohol in which the catalyst had been dissolved andthe mixture is heated under reflux for 1–5 h. Two layers are formedaround the time of the completion of the reaction. The lower layer isthe alcohol phase and can be recovered. The upper layer, includingthe crude biodiesel, is washed with water to remove the contaminantsuntil the washing solution is neutral. After the washing step,the upper layer is dried over anhydrous sodium sulfate, then filtered,and the residual product is biodiesel [156]. Haas and Scott[155] found that the final biodiesel product can conform to theASTM standard and the conversion of the oilseed is very high (about98%). Since this method eliminates the need for the isolation of, andpossibly for the refining of, the oilseed lipid, the process could reducebiodiesel production costs, reduce the long size of the productionsystem associated with the pre-extraction, degumming, andmaximize the yield of the biodiesel production. However, this processcannot handle waste cooking oils and animal fats, which canreduce the cost of feedstock [156,157].4. ConclusionsBiodiesel is a clean-burning diesel fuel with a chemical structureof fatty acid alkyl esters. Of the various methods availablefor producing biodiesel, the alkali-catalyzed transesterification ofvegetable oils and animal fats is currently the most commonlyadopted method. Transesterification is basically a sequential reaction.However, when the raw materials (oils or fats) contain a highpercentage of free fatty acids or water, the alkali catalyst will reactwith the free fatty acids to form soaps and the water can hydrolyzethe triglycerides into diglycerides and form more free fatty acids.These are undesirable reactions which reduce the yield of the biodieselproduct. Therefore, after refining the raw materials, theacidic feedstocks should be pre-treated to inhibit the saponificationreaction. There are three primary approaches for reducingthe amount of free fatty acids: esterification of free fatty acids withmethanol, in the presence of acidic catalysts; using iodine as a catalyst;adding glycerol into the acidic feedstock with a catalyst likezinc chloride and heated to a high temperature. The first approachcan eliminate the separation, corrosion, toxicity, and environmentalproblems, but the reaction rate is slower while the advantage ofthe last approach is that no alcohol is needed and the water(formed from the reaction) can be immediately vaporized andvented from the mixture. The transesterification reaction requiresan alcohol as a reactant and a catalyst. The most commonly usedalcohol is methanol while sodium hydroxide and potassiumhydroxide are the most commonly used catalysts. During the reaction,glycerol will be produced as a by-product. Because of itsnumerous industrial applications, the crude glycerol should be refinedwith purity higher than 99% to make it usable. For refiningthe crude biodiesel produced, the product should first be neutralizedand then put through an alcohol stripper before cleaning byeither one of the following approaches: water washing, membraneextraction, and dry washing. A very promising method for washingbiodiesel is the hollow fiber membrane extraction, which effectivelyavoids emulsification during the washing step and decreasesthe refining loss. As a commercial fuel, the finished biodiesel mustbe analyzed using sophisticated analytical equipment to ensurethat it meets international standards even if it has been storedfor a long time.There are four primary factors affecting the yield of biodiesel,i.e. alcohol quantity, reaction time, reaction temperature, and catalystconcentration. To ensure a complete transesterification reaction,the molar ratio of alcohol to triglycerides should be increasedto 6:1 with the use of an alkali catalyst. For used cooking oils or foroils with a high percentage of free fatty acids, a higher molar ratiois needed for acid-catalyzed transesterification. Because the conversionrate of fatty acid esters increases with reaction time butthe yield of the biodiesel product reaches a maximum at an optimalreaction time. Higher reaction temperature can decrease theviscosity of oils, enhancing the reaction rate. The optimal temperatureranged between 50 °C and 60 °C, depending on the oil used.The optimal condition of catalyst concentration is about 1.5 wt.%for NaOH which is the most commonly used catalyst. With increasingconcern over global warming, it is foreseeable that biodieselusage would continue to grow at a fast pace. This will trigger thedevelopment of more sophisticated methods of biodiesel productionand refining to cope with the increasing market demand.AcknowledgementThe authors would like to acknowledge the support of the ICEEand the University Development Fund of the University of HongKong.References[1] Xu G, Wu GY. The investigation of blending properties of biodiesel and No. 0diesel fuel. J Jiangsu Polytech Univ 2003;15:16–8.[2] Abreu FR, Lima DG, Hamú EH, Wolf C, Suarez PAZ. Utilization of metalcomplexes as catalysts in the transesterification of Brazilian vegetable oilswith different alcohols. J Mol Catal A Chem 2004;209:29–33.[3] Azcan N, Danisman A. Microwave assisted transesterification of rapeseed oil.Fuel 2008;87:1781–8.[4] Berchmans HJ, Hirata S. Biodiesel production from crude Jatropha curcas L.seed oil with a high content of free fatty acids. Bioresour Technol2008;99:1716–21.[5] Bernardo A, Howard-Hildige R, O’Connell A, Nichol R, Ryan J, Rice B, et al.Camelina oil as a fuel for diesel transport engines. Ind Crops Prod2003;17:191–7.[6] Byun MW, Kang IJ, Kwon JH, Hayashi Y, Mori T. Physicochemical properties ofsoybean oil extracted from [gamma]-irradiated soybeans. Radiat Phys Chem1995;46:659–62.[7] Chitra P, Venkatachalam P, Sampathrajan A. Optimisation of experimentalconditions for biodiesel production from alkali-catalysed transesterificationof Jatropha curcus oil. Energy Sustain Dev 2005;9:13–8.[8] Demirbas A. Biodiesel fuels from vegetable oils via catalytic and non-catalyticsupercritical alcohol transesterifications and other methods: a survey. EnergyConvers Manage 2003;44:2093–109.[9] Kaya C, Hamamci C, Baysal A, Akba O, Erdogan S, Saydut A. Methyl ester ofpeanut (Arachis hypogea L.) seed oil as a potential feedstock for biodieselproduction. Renew Energy 2009;34:1257–60.[10] Issariyakul T, Kulkarni MG, Meher LC, Dalai AK, Bakhshi NN. Biodieselproduction from mixtures of canola oil and used cooking oil. Chem Eng J2008;140:77–85.[11] Kansedo J, Lee KT, Bhatia S. Cerbera odollam (sea mango) oil as a promisingnon-edible feedstock for biodiesel production. Fuel 2009;88:1148–50.[12] Kumar Tiwari A, Kumar A, Raheman H. Biodiesel production from jatropha oil(Jatropha curcas) with high free fatty acids: an optimized process. BiomassBioenergy 2007;31:569–75.[13] Rao Y, Xiang B, Zhou X, Wang Z, Xie S, Xu J. Quantitative and qualitativedetermination of acid value of peanut oil using near-infrared spectrometry. JFood Eng 2009;93:249–52.[14] Sahoo PK, Das LM. Process optimization for biodiesel production fromJatropha, Karanja and Polanga oils. Fuel 2009;88:1588–94.[15] San Jose Alonso J, Lopez Sastre JA, Romero-Avila C, Lopez E. A note on thecombustion of blends of diesel and soya, sunflower and rapeseed vegetableoils in a light boiler. Biomass Bioenergy 2008;32:880–6.[16] Santos FFP, Malveira JQ, Cruz MGA, Fernandes FAN. Production of biodiesel byultrasound assisted esterification of Oreochromis niloticus oil. Fuel2009;doi:10.1016/j.fuel.2009.05.030.[17] Saraf S, Thomas B. Influence of feedstock and process chemistry on biodieselquality. Process Saf Environ Prot 2007;85:360–4.[18] Singh SP, Singh D. Biodiesel production through the use of different sourcesand characterization of oils and their esters as the substitute of diesel: areview. Renewable and Sustainable Energy Reviews 2009;14:200–16.[19] Srivastava A, Prasad R. Triglycerides-based diesel fuels. Renew Sustain EnergyRev 2000;4:111–33.

D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095 1093[20] Winayanuwattikun P, Kaewpiboon C, Piriyakananon K, Tantong S,Thakernkarnkit W, Chulalaksananukul W, et al. Potential plant oil feedstockfor lipase-catalyzed biodiesel production in Thailand. Biomass Bioenergy2008;32:1279–86.[21] Zubr J. Oil-seed crop: Camelina sativa. Ind Crops Prod 1997;6:113–9.[22] Schinas P, Karavalakis G, Davaris C, Anastopoulos G, Karonis D, Zannikos F,et al. Pumpkin (Cucurbita pepo L.) seed oil as an alternative feedstock for theproduction of biodiesel in Greece. Biomass Bioenergy 2009;33:44–9.[23] Goodrum JW, Geller DP, Adams TT. Rheological characterization of animalfats and their mixtures with #2 fuel oil. Biomass Bioenergy 2003;24:249–56.[24] Liu KKM, Barrows FT, Hardy RW, Dong FM. Body composition, growthperformance, and product quality of rainbow trout (Oncorhynchus mykiss)fed diets containing poultry fat, soybean/corn lecithin, or menhaden oil.Aquaculture 2004;238:309–28.[25] Patil PD, Deng S. Optimization of biodiesel production from edible and nonediblevegetable oils. Fuel 2009;88:1302–6.[26] Gui MM, Lee KT, Bhatia S. Feasibility of edible oil vs. non-edible oil vs. wasteedible oil as biodiesel feedstock. Energy 2008;33:1646–53.[27] Haas MJ. Improving the economics of biodiesel production through the use oflow value lipids as feedstocks: vegetable oil soapstock. Fuel Process Technol2005;86:1087–96.[28] Leung DYC, Guo Y. Transesterification of neat and used frying oil:optimization for biodiesel production. Fuel Process Technol 2006;87:883–90.[29] Upham P, Thornley P, Tomei J, Boucher P. Substitutable biodiesel feedstocksfor the UK: a review of sustainability issues with reference to the UK RTFO. JCleaner Prod 2009;17:S37–45.[30] Bhatti HN, Hanif MA, Qasim M, Ataur R. Biodiesel production from wastetallow. Fuel 2008;87:2961–6.[31] Frohlich A, Rice B. Evaluation of Camelina sativa oil as a feedstock forbiodiesel production. Ind Crops Prod 2005;21:25–31.[32] Hameed BH, Lai LF, Chin LH. Production of biodiesel from palm oil (Elaeisguineensis) using heterogeneous catalyst: an optimized process. Fuel ProcessTechnol 2009;90:606–10.[33] Liu X, He H, Wang Y, Zhu S, Piao X. Transesterification of soybean oil tobiodiesel using CaO as a solid base catalyst. Fuel 2008;87:216–21.[34] Meher LC, Dharmagadda VSS, Naik SN. Optimization of alkali-catalyzedtransesterification of Pongamia pinnata oil for production of biodiesel.Bioresour Technol 2006;97:1392–7.[35] Ramadhas AS, Jayaraj S, Muraleedharan C. Biodiesel production from high FFArubber seed oil. Fuel 2005;84:335–40.[36] Rashid U, Anwar F. Production of biodiesel through optimized alkalinecatalyzedtransesterification of rapeseed oil. Fuel 2008;87:265–73.[37] Rashid U, Anwar F, Knothe G. Evaluation of biodiesel obtained fromcottonseed oil. Fuel Process Technol 2009;90:1157–63.[38] Rashid U, Anwar F, Moser BR, Ashraf S. Production of sunflower oil methylesters by optimized alkali-catalyzed methanolysis. Biomass Bioenergy2008;32:1202–5.[39] Demirbas A. Fuel properties and calculation of higher heating values ofvegetable oils. Fuel 1998;77:1117–20.[40] Mudge SM, Pereira G. Stimulating the biodegradation of crude oil withbiodiesel preliminary results. Spill Sci Technol Bull 1999;5:353–5.[41] Speidel HK, Lightner RL, Ahmed I. Biodegradability of new engineered fuelscompared to conventional petroleum fuels and alternative fuels in currentuse. Appl Biochem Biotechnol 2000;84–86:879–97.[42] Harrington KJ. Chemical and physical properties of vegetable oil esters andtheir effect on diesel fuel performance. Biomass 1986;9:1–17.[43] USEPA. A comprehensive analysis of biodiesel impacts on exhaust emissions.Draft Technical Report. USEPA; 2002.[44] Knothe G, Sharp CA, Ryan TW. Exhaust emissions of biodiesel, petrodiesel,neat methyl esters, and alkanes in a new technology engine. Energy Fuels2006;20:403–8.[45] Lin L, Ying D, Chaitep S, Vittayapadung S. Biodiesel production from crude ricebran oil and properties as fuel. Appl Energy 2009;86:681–8.[46] Lopez JM, Gomez A, Aparicio F, Javier Sanchez F. Comparison of GHGemissions from diesel, biodiesel and natural gas refuse trucks of the City ofMadrid. Appl Energy 2009;86:610–5.[47] Zhang Y, Dube MA, McLean DD, Kates M. Biodiesel production from wastecooking oil: 2. Economic assessment and sensitivity analysis. BioresourTechnol 2003;90:229–40.[48] Canakci M, Erdil A, Arcaklioglu E. Performance and exhaust emissions of abiodiesel engine. Appl Energy 2006;83:594–605.[49] Adams C, Peters J, Rand M, Schroer B, Ziemke M. Investigation of soybean oilas a diesel fuel extender: endurance tests. J Am Oil Chem Soc 1983;60:1574–9.[50] Anon. Filtered used frying fat powers diesel fleet. J Am Oil Chem Soc1982;59:780A–1A.[51] Engler C, Johnson L, Lepori W, Yarbrough C. Effects of processing and chemicalcharacteristics of plant oils on performance of an indirect-injection dieselengine. J Am Oil Chem Soc 1983;60:1592–6.[52] Peterson C, Auld D, Korus R. Winter rape oil fuel for diesel engines: recoveryand utilization. J Am Oil Chem Soc 1983;60:1579–87.[53] Strayer R, Blake J, Craig W. Canola and high erucic rapeseed oil as substitutesfor diesel fuel: preliminary tests. J Am Oil Chem Soc 1983;60:1587–92.[54] Schwab AW, Bagby MO, Freedman B. Preparation and properties of dieselfuels from vegetable oils. Fuel 1987;66:1372–8.[55] Chang C, Wan S. China’s motor fuels from tung oil. Ind Eng Chem1947;39:1543–8.[56] Crossley A, Heyes TD, Hudson B. The effect of heat on pure triglycerides. J AmOil Chem Soc 1962;39:9–14.[57] Niehaus RA, Goering CE, Savage Jr LD, Sorenson SC. Cracked soybean oil asfuel for a diesel engine. Trans Am Soc Agric Eng 1986;29:683–9.[58] Pioch D, Lozano P, Rasoanantoandro MC, Graille J, Geneste P, Guida A. Biofuelsfrom catalytic cracking of tropical vegetable oils. Oleagineux1993;48:289–91.[59] Sonntag A. Reactions of fats and fatty acids. In: Swern D, editor. Bailey’sindustrial oil and fat products. New York: John Wiley & Sons; 1979.[60] Weisz PB, Haag WO, Rodeweld PG. Catalytic production of high-grade fuel(gasoline) from biomass compounds by shape-delective catalysis. Science1979;206:57–8.[61] Ma F, Hanna MA. Biodiesel production: a review. Bioresour Technol1999;70:1–15.[62] Van Gerpen J, Shanks B, Pruszko R, Clements D, Knothe G. Biodieselproduction technology. 1617 Cole Boulevard, Golden, CO: NationalRenewable Energy Laboratory; 2004.[63] Soriano Jr NU, Venditti R, Argyropoulos DS. Biodiesel synthesis viahomogeneous Lewis acid-catalyzed transesterification. Fuel 2009;88:560–5.[64] ISTC. Feasibility report, small scale biodiesel production. In: Center IST, editor.Waste Management and Research Center; 2006.[65] Van Gerpen JH, Dvorak B. The effect of phosphorus level on the total glyceroland reaction yield of biodiesel bioenergy 2002. In: The 10th BiennialBioenergy Conference, Boise; 2002.[66] Liu GR, Huang YQ, Chen GY. Study on the degumming process of acidic oil. In:Zhuang X, editor. International conference on biomass energy technologies.Guangzhou, China; 2008.[67] de Carvalho CC, de Souza MP, da Silva TD, Gonçalves LAG, Viotto LA. Soybeanscrude oil miscella degumming utilizing ceramic membranes: transmembranepressure and velocity effects. Desalination 2006;200:543–5.[68] de Souza MP, Cunha Petrus JC, Guaraldo Gonçalves LA, Viotto LA. Degummingof corn oil/hexane miscella using a ceramic membrane. J Food Eng2008;86:557–64.[69] Verleyen T, Sosinska U, Ioannidou S, Verhe R, Dewettinck K, Huyghebaert A,et al. Influence of the vegetable oil refining process on free and esterifiedsterols. J Am Oil Chem Soc 2002;79:947–53.[70] Pan L, Noli A, Campana A, Barrera M, Tomas M, Anon M. Influence of theoperating conditions on acid degumming process in sunflower seed oil. J AmOil Chem Soc 2001;78:553–4.[71] Eisenmenger M, Dunford N. Bioactive components of commercial andsupercritical carbon dioxide processed wheat germ oil. J Am Oil Chem Soc2008;85:55–61.[72] Chen YC, He SL, Cheng J. Study on extraction and deacidification of oils andfats of natural plants by supercritical carbon dioxide. J Yangtze Univ (Nat SciEd) 2007;4:45–7.[73] Demirbas A, Kara H. New options for conversion of vegetable oils toalternative fuels. Energy Sources Part A: Recovery, Utilization &Environmental Effects. Taylor & Francis Ltd; 2006 [p. 619–26].[74] Mittelbach M, Koncar M. Method for the preparation of fatty acid alkyl esters.United States Patent 5849939. URL:http://www.freepatentsonline.com/5849939.html.[75] Turkay S, Civelekoglu H. Deacidification of sulfur olive oil. I. Single-stageliquid–liquid extraction of miscella with ethyl alcohol. J Am Oil Chem Soc1991;68:83–6.[76] Zhang Y, Lu XH, Yu YL, Ji JB. Study on the coupling process of catalyticesterification and extraction of high acid value waste oil with methanol. In:Zhuang X, editor. International conference on biomass energy technologies.Guangzhou, China; 2008.[77] Pina C, Meirelles A. Deacidification of corn oil by solvent extraction in aperforated rotating disc column. J Am Oil Chem Soc 2000;77:553–9.[78] Di Serio M, Tesser R, Dimiccoli M, Cammarota F, Nastasi M, Santacesaria E.Synthesis of biodiesel via homogeneous Lewis acid catalyst. J Mol Catal A:Chem 2005;239:111–5.[79] Sheng M, Tian DL, Cao GM. Production of biodiesel fuel from wast edible oil.China Academic J 2008:26.[80] Jeromin L, Peukert E, Wollmann G. Process for the pre-esterification of freefatty acids in raw fats and/or oils. Henkel, Kommanditgesellschaft AufAktien;1990. URL:http://www.freepatentsonline.com/EP0192035B1.html.[81] Simone MR, Michele CR, Marcelli GR, Paulo LSJ, Fernanda MBC, Elizabeth RL.Transesterification of vegetable oils promoted by poly(styrenedivinylbenzene)and poly(divinylbenzene). Appl Catal A: Gen2008;349:198–203.[82] Lou W, Zong M, Duan Z. Efficient production of biodiesel from high free fattyacid-containing waste oils using various carbohydrate-derived solid acidcatalysts. Bioresour Technol 2008;99:8752–8.[83] Ozbay N, Oktar N, Tapan NA. Esterification of free fatty acids in waste cookingoils (WCO): role of ion-exchange resins. Fuel 2008;87:1789–98.[84] Shibasaki-Kitakawa N, Honda H, Kuribayashi H, Toda T, Fukumura T,Yonemoto T. Biodiesel production using anionic ion-exchange resin asheterogeneous catalyst. Bioresour Technol 2007;98:416–21.[85] Li C, Liu YH, Luo AX, Ruan RS, Liu CM. New two-step method of producingbiodiesel from waster cooking oil. Cereals Oils Process 2008.[86] Van Gerpen J, Shanks B, Pruszko R, Clements D, Knothe G. Biodieselproduction technology: August 2002–January 2004. Other information:

1094 D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095PBD: 1 July 2004; related information: work performed by Iowa StateUniversity, Renewable Products Development Laboratory, and USDA/NCAUR,2004. Size: 110 p.[87] Canakci M, Van Gerpen J. Biodiesel production via acid catalysis. Trans AmSoc Agric Eng 1999;42:1203–10.[88] Nelson L, Foglia T, Marmer W. Lipase-catalyzed production of biodiesel. J AmOil Chem Soc 1996;73:1191–5.[89] Shimada Y, Watanabe Y, Samukawa T, Sugihara A, Noda H, Fukuda H.Conversion of vegetable oil to biodiesel using immobilized Candida antarcticalipase. J Am Oil Chem Soc 1999;76:789–93.[90] Devanesan MG, Viruthagiri T, Sugumar N. Transesterification of jatropha oilusing immobilized pseudomonas fluorescens. Afr J Biotechnol2007;6:2497–501.[91] Fukuda H, Kondo A, Noda H. Biodiesel fuel production by transesterificationof oils. J Biosci Bioeng 2001;92:405–16.[92] Kaieda M, Samukawa T, Matsumoto T, Ban K, Kondo A, Shimada Y, et al.Biodiesel fuel production from plant oil catalyzed by Rhizopus oryzae lipasein a water-containing system without an organic solvent. J Biosci Bioeng1999;88:627–31.[93] Matsumoto T, Takahashi S, Kaieda M, et al. Yeast whole-cell biocatalystconstructed by intracellular overproduction of Rhizopus oryzae lipase isapplicable to biodiesel fuel production. Appl Microbiol Biotechnol2001;57:515–20.[94] Demirbas A. Current technologies in biodiesel production. In: Biodiesel; 2008.p. 161–73.[95] Azcan N, Danisman A. Alkali catalyzed transesterification of cottonseed oil bymicrowave irradiation. Fuel 2007;86:2639–44.[96] Dias JM, Alvim-Ferraz MCM, Almeida MF. Comparison of the performance ofdifferent homogeneous alkali catalysts during transesterification of wasteand virgin oils and evaluation of biodiesel quality. Fuel 2008;87:3572–8.[97] Dizge N, Aydiner C, Imer DY, Bayramoglu M, Tanriseven A, Keskinler B.Biodiesel production from sunflower, soybean, and waste cooking oils bytransesterification using lipase immobilized onto a novel microporouspolymer. Bioresour Technol 2009;100:1983–91.[98] Dizge N, Keskinler B. Enzymatic production of biodiesel from canola oil usingimmobilized lipase. Biomass Bioenergy 2008;32:1274–8.[99] Shimada Y, Watanabe Y, Sugihara A, Tominaga Y. Enzymatic alcoholysis forbiodiesel fuel production and application of the reaction to oil processing. JMol Catal B: Enzym 2002;17:133–42.[100] Kulkarni MG, Gopinath R, Meher LC, Dalai AK. Solid acid catalyzed biodieselproduction by simultaneous esterification and transesterification. GreenChem 2006;8:1056–62.[101] Qian WW, Yang L, Lu XP. Preparation of biodiesel catalyzed by KF/CaO withultrasonic. In: Zhuang X, editor. International conference on biomass energytechnologies. Guangzhou, China; 2008.[102] da Silva RB, Lima Neto AF, Soares dos Santos LS, de Oliveira Lima JR, ChavesMH, dos Santos Jr JR, et al. Catalysts of Cu(II) and Co(II) ions adsorbed inchitosan used in transesterification of soy bean and babassu oils – A newroute for biodiesel syntheses. Bioresour Technol 2008;99:6793–8.[103] Di Serio M, Cozzolino M, Tesser R, Patrono P, Pinzari F, Bonelli B, et al. Vanadylphosphate catalysts in biodiesel production. Appl Catal A: Gen 2007;320:1–7.[104] Kawashima A, Matsubara K, Honda K. Acceleration of catalytic activity ofcalcium oxide for biodiesel production. Bioresour Technol2009;100:696–700.[105] Lukic I, Krstic J, Jovanovic D, Skala D. Alumina/silica supported K 2 CO 3 as acatalyst for biodiesel synthesis from sunflower oil. Bioresour Technol2009;100:4690–6.[106] Noiroj K, Intarapong P, Luengnaruemitchai A, Jai-In S. A comparative study ofKOH/Al 2 O 3 and KOH/NaY catalysts for biodiesel production viatransesterification from palm oil. Renew Energy 2009;34:1145–50.[107] Sakai T, Kawashima A, Koshikawa T. Economic assessment of batch biodieselproduction processes using homogeneous and heterogeneous alkali catalysts.Bioresour Technol 2009;100:3268–76.[108] Shu Q, Zhang Q, Xu G, Nawaz Z, Wang D, Wang J. Synthesis of biodiesel fromcottonseed oil and methanol using a carbon-based solid acid catalyst. FuelProcess Technol 2009;90:1002–8.[109] Zabeti M, Wan Daud WMA, Aroua MK. Activity of solid catalysts for biodieselproduction: a review. Fuel Process Technol 2009;90:770–7.[110] Stavarache C, Vinatoru M, Nishimura R, Maeda Y. Fatty acids methyl estersfrom vegetable oil by means of ultrasonic energy. Ultrason Sonochem2005;12:367–72.[111] Thanh LT, Okitsu K, Sadanaga Y, Takenaka N, Maeda Y, Bandow H.Ultrasound-assisted production of biodiesel fuel from vegetable oils in asmall scale circulation process. Bioresour Technol 2009;101:639–45.[112] Santos FFP, Rodrigues S, Fernandes FAN. Optimization of the production ofbiodiesel from soybean oil by ultrasound assisted methanolysis. Fuel ProcessTechnol 2009;90:312–6.[113] Wen Z, Yu X, Tu S-T, Yan J, Dahlquist E. Intensification of biodiesel synthesisusing zigzag micro-channel reactors. Bioresour Technol 2009;100:3054–60.[114] Schumacher J. Small Scale Biodiesel Production: An Overview. AgriculturalMarketing Policy; 2007. URL:http://www.ampc.montana.edu/policypaper/policy22.pdf.[115] Wen GD, Xu YP, Ma HJ, Xu ZS, Tian ZJ. Production of hydrogen by aqueousphasereforming of glycerol. Int J Hydrogen Energy 2008;33:6657–66.[116] da Silva GP, Mack M, Contiero J. Glycerol: a promising and abundant carbonsource for industrial microbiology. Biotechnol Adv 2009;27:30–9.[117] Wang Z, Zhuge J, Fang H, Prior BA. Glycerol production by microbialfermentation: a review. Biotechnol Adv 2001;19:201–23.[118] Whittington T. Biodiesel production and use by farmers: is it worthconsidering? Western Australia: Department of Agriculture and Food;2006.[119] Duncan J. Cost of biodiesel production. Energy Efficiency Conserv Auth 2003.[120] NBB. Biodiesel production and quality. Fuel fact sheets. USA: NationalBiodiesel Board; 2007.[121] Gerpen JV. Biodiesel processing and production. Fuel Process Technol2005;86:1097–107.[122] Cooke BS, Abrams C, Bertram B. Purification of biodiesel with adsorbentmaterials 2005.[123] Gabelman A, Hwang S. Hollow fiber membrane contactors. J Membr Sci1999;159:61–106.[124] He H, Guo X, Zhu S. Comparison of membrane extraction with traditionalextraction methods for biodiesel production. J Am Oil Chem Soc2006;83:457–60.[125] Predojevic ZJ. The production of biodiesel from waste frying oils: acomparison of different purification steps. Fuel 2008;87:3522–8.[126] Karaosmanoglu F, Cigizoglu KB, Tuter M, Ertekin S. Investigation of therefining step of biodiesel production. Energy Fuels 1996;10:890–5.[127] Canakci M, Van Gerpen J. Biodiesel production from oils and fats with highfree fatty acids. Trans Am Soc Agric Eng 2001;44:1429–36.[128] Berrios M, Skelton RL. Comparison of purification methods for biodiesel.Chem Eng J 2008;144:459–65.[129] EN. The EN 14214 standard-specifications and test methods; 2003.[130] ASTM. ASTM D6751-biodiesel blend stock specification (B100).[131] Mittelbach M. Diesel fuel derived from vegetable oils, VI: specifications andquality control of biodiesel. Bioresour Technol 1996;56:7–11.[132] Zheng M, Mulenga MC, Reader GT, Wang MP, Ting DSK, Tjong J. Biodieselengine performance and emissions in low temperature combustion. Fuel2008;87:714–22.[133] Ali Y, Hanna M, Cuppett S. Fuel properties of tallow and soybean oil esters. JAm Oil Chem Soc 1995;72:1557–64.[134] Leung DYC, Koo BCP, Guo Y. Degradation of biodiesel under different storageconditions. Bioresour Technol 2006;97:250–6.[135] Lee I, Johnson LA, Hammond EG. Use of branched-chain esters to reduce thecrystallization temperature of biodiesel. J Am Oil Chem Soc1995;72:1155–60.[136] Bondioli P, Gasparoli A, Lanzani A, Fedeli E, Veronese S, Sala M. Storagestability of biodiesel. J Am Oil Chem Soc 1995;72:699–702.[137] Mittelbach M, Gangl S. Long storage stability of biodiesel made from rapeseedand used frying oil. J Am Oil Chem Soc 2001;78:573–7.[138] Freedman B, Butterfield RO, Pryde EH. Transesterification kinetics of soybeanoil. J Am Oil Chem Soc 1986;63:1375–80.[139] Sprules FJ, Price D. Production of fatty esters. United States: NOPCO CHEMCO; 1950. URL: http://www.freepatentsonline.com/2494366.html.[140] Zhang D. Crystallization characteristics and fuel properties of tallow methylesters. Master thesis, Food Science And Technology. USA: University ofNebraska–Lincoln; 1994.[141] Freedman B, Pryde EH, Mounts TL. Variables affecting the yields of fattyesters from transesterified vegetable oils. J Am Oil Chem Soc 1984;61:1638–43.[142] Alamu OJ, Waheed MA, Jekayinfa SO, Akintola TA. Optimal transesterificationduration for biodiesel production from nigerian palm kernel oil. Agric EngInt: CIGR Ejournal 2007; IX.[143] Eevera T, Rajendran K, Saradha S. Biodiesel production process optimizationand characterization to assess the suitability of the product for variedenvironmental conditions. Renew Energy 2009;34:762–5.[144] Ma F, Clements LD, Hanna MA. The effects of catalyst, free fatty acids, andwater on transesterification of beef tallow. Trans Am Soc Agric Eng1998;41:1261–4.[145] Guo Y. Alkaline-catalyzed production of biodiesel fuel from virgin canola oiland recycled waste oils. PhD dissertation, Department of MechanicalEngineering, the University of Hong Kong, Hong Kong; 2005. p. 184.[146] Boocock DGB, Konar SK, Mao V, Sidi H. Fast one-phase oil-rich processes forthe preparation of vegetable oil methyl esters. Biomass Bioenergy 1996;11:43–50.[147] Demirbas A. Biodiesel from vegetable oils via transesterification insupercritical methanol. Energy Convers Manage 2002;43:2349–56.[148] Saka S, Kusdiana D. Biodiesel fuel from rapeseed oil as prepared insupercritical methanol. Fuel 2001;80:225–31.[149] Han HW, Cao WL, Zhang JC. Preparation of biodiesel from soybean oil usingsupercritical methanol and CO 2 as co-solvent. Process Biochem 2005;40:3148–51.[150] Bunyakiat K, Makmee S, Sawangkeaw R, Ngamprasertsith S. Continuousproduction of biodiesel via transesterification from vegetable oils insupercritical methanol. Energy Fuels 2006;20:812–7.[151] Vera CR, D’Ippolito SA, Pieck CL, Parera JM. Production of biodiesel by a twostepsupercritical reaction process with adsorption refining. In: 2nd Mercosurcongress on chemical engineering, 4th Mercosur congress on process systemsengineering; 2005.[152] Harrington KJ, D’Arcy-Evans C. Transesterification in situ of sunflower seedoil. Ind Eng Chem Prod Res Dev 1985;24:314–8.

D.Y.C. Leung et al. / Applied Energy 87 (2010) 1083–1095 1095[153] Haas M, Scott K, Marmer W, Foglia T. In situ alkaline transesterification: aneffective method for the production of fatty acid esters from vegetable oils. JAm Oil Chem Soc 2004;81:83–9.[154] Siler-Marinkovic S, Tomasevic A. Transesterification of sunflower oil in situ.Fuel 1998;77:1389–91.[155] Haas M, Scott K. Moisture removal substantially improves the efficiency ofin situ biodiesel production from soybeans. J Am Oil Chem Soc 2007;84:197–204.[156] Qian JF, Wang F, Liu S, Yun Z. In situ alkaline transesterification of cottonseedoil for production of biodiesel and nontoxic cottonseed meal. BioresourTechnol 2008;99:9009–12.[157] Mondala A, Liang K, Toghiani H, Hernandez R, French T. Biodiesel productionby in situ transesterification of municipal primary and secondary sludges.Bioresour Technol 2009;100:1203–10.

Bioresource Technology 101 (2010) 5903–5909Contents lists available at ScienceDirectBioresource Technologyjournal homepage: www.elsevier.com/locate/biortechOne-pot process combining transesterification and selective hydrogenation forbiodiesel production from starting material of high degree of unsaturationRu Yang a, *, Mengxing Su a , Min Li a , Jianchun Zhang b , Xinmin Hao b , Hua Zhang ba State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, PR Chinab The Quartermaster Institute of General Logistics, Department of CPLA, Beijing 100088, PR ChinaarticleinfoabstractArticle history:Received 6 December 2009Received in revised form 19 February 2010Accepted 23 February 2010Available online 15 March 2010Keywords:BiodieselTransesterificationSelective hydrogenationOne-pot processCetane numberA one-pot process combining transesterification and selective hydrogenation was established to producebiodiesel from hemp (Cannabis sativa L.) seed oil which is eliminated as a potential feedstock by a specificationof iodine value (IV; 120 g I 2 /100 g maximum) contained in EN 14214. A series of alkaline earthmetal oxides and alkaline earth metal supported copper oxide were prepared and tested as catalysts. SrOsupported 10 wt.% CuO showed the superior catalytic activity for transesterification with a biodiesel yieldof 96% and hydrogenation with a reduced iodine value of 113 and also exhibited a promising selectivityfor eliminating methyl linolenate and increasing methyl oleate without rising methyl stearate in theselective hydrogenation. The fuel properties of the selective hydrogenated methyl esters are within biodieselspecifications. Furthermore, cetane numbers and iodine values were well correlated with the compositionsof the hydrogenated methyl esters according to degrees of unsaturation.Ó 2010 Elsevier Ltd. All rights reserved.1. IntroductionBiodiesel has lately emerged as an alternative fuel with wideacceptance because of its reduction of most exhaust emissions, improvedlubricity, higher flash point, improved biodegradability andreduced toxicity over conventional diesel fuel (CDF) (Christos et al.,2010; Knothe, 2006; Moser et al., 2007). It is usually produced bythe transesterification of vegetable oil or animal fat with shortchainalcohols (Su et al., 2010). The vegetable oils being exploitedcommercially constitute the edible fatty oils derived from rapeseed,soybean, palm, sunflower, etc. Owning to the high priceand a big gap in demand and supply of such oils, it is not feasibleto produce biodiesel in developing countries such as China andIndia. This highlights the need for the development of technologiesallowing the use of non-edible vegetable oils. More than 350 oilbearingcrops have been identified (Dorado et al., 2004); oils from75 plant species have been converted to biodiesel and iodine value(IV) and cetane number (CN) varied from 4.8 to 212 and 20.56 to67.47, respectively (Azam et al., 2005). However, regardless ofthe raw materials used, biodiesel marketed for consumption mustcomply with specifications mandating biodiesel quality, mostnotably EN 14214, which contains a specification for iodine value(IV; 120 g I 2 /100 g maximum) and cetane number (CN; 51 minimum),which eliminates the highly poly-unsaturated oil as a potentialfeedstock.* Corresponding author. Tel./fax: +86 10 64436736.E-mail address: yangru@bbn.cn (R. Yang).Biodiesel properties included as specifications in standardswere determined by the nature of the biodiesel components. Asbiodiesel consists of fatty acids esters, not only the structure ofthe fatty acids but also that of the ester moiety derived from thealcohol can influence the fuel properties of biodiesel. These structuralfeatures include degree of unsaturation, chain length, andbranching of the chain (Knothe, 2005), and degree of unsaturationof the fatty acids is the most important factor. The properties of abiodiesel fuel, including iodine value, oxidative stability, cetanenumber, cold-flow properties, viscosity and lubricity, are directlydetermined by the degree of unsaturation of its component fattyesters (Knothe, 2008). Iodine value and cold-flow properties increaseas the number of double bonds in the fatty acid molecularincreases (Moraes et al., 2008; Imahara et al., 2006), while cetanenumber and the oxidative stability decrease with an increasing degreeof unsaturation in the fatty acid chain (Knothe, 2005). Thus,only through modifying and keeping a proper degree of unsaturationcan biodiesel obtained from starting materials of high degreeof unsaturation can meet the specifications. Several approaches arepossible to modify the fatty acid composition of biodiesel. A firstapproach is the use of oils with inherently modified fatty acidprofile by genetic modification giving advantageous fuel properties(Knothe, 2008). A second approach is the fraction distillationobtaining biodiesels with different fatty acid compositions of fattyacids esters (Falk and Meyer-Pittroff, 2004). However, when thegood distillates were selected for biodiesel fuel, the problem ofapplication of the left ones occurs. A third approach is the catalytichydrogenation of the feedstock and/or the fatty acids esters. In0960-8524/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.doi:10.1016/j.biortech.2010.02.095

5904 R. Yang et al. / Bioresource Technology 101 (2010) 5903–5909order to use fatty substances having iodine values greater than120 g I 2 /100 g, more preferably greater than 150 g I 2 /100 g, forthe production of biodiesel, selective hydrogenation is essentialand practical, which reduces the degree of unsaturation withoutincreasing the stearic acid content and limiting cis–trans and positionalisomerisation as much as possible (Federico et al., 2006).Past attempts to reduce the degree of unsaturation of the fattyacids in the highly ploy-unsaturated fatty substances using partialhydrogenation over conventional catalysts, lead to derivatives thatwere unsuited to the manufacture of biodiesel. In deed, the selectivityfor biodiesel production in partial hydrogenation is difficultto achieve with the nickel (Moser et al., 2007) and noble metalbasedcatalysts like platinum (Nohair et al., 2005), rhodium (Nikolaouet al., 2009) and palladium (Fernandez et al., 2005; Pérez-Cadenaset al., 2006), resulted in the formation of significant quantitiesof high melting point saturated fatty acids, which prejudice thecold-weather behavior of the product obtained (Federico et al.,2006). On the contrary, copper catalysts were found to have goodcatalytic activity and selectivity in the selective hydrogenation ofhighly unsaturated vegetable oils. Cu/Al 2 O 3 (Federico et al., 2006)and Cu/SiO 2 (Nicoletta et al., 2002) possess high selectivity forhydrogenating linolenate C18:3 to oleate C18:1 with oleateC18:1 unreduced, therefore, the percentage of saturated is scarcelychanged during the hydrogenation process. However, it should bepointed out that these previous researches applied two separateprocesses, i.e., the hydrogenation of triglycerides or esters andtransesterification for biodiesel production. The employment ofdifferent catalysts and equipments for the separate processes couldincrease the cost for large-scale biodiesel production.Hemp (Cannabis sativa L.), an annual herbaceous plant, has beengrown agriculturally for many centuries for its bast fiber andhempseed oil. (Oomaha et al., 2002). Due to the increasing demandfor hemp fiber for clothing and paper and the development of technologiesfor hemp fiber processing, the cultivated area of hemp increasesremarkably. China currently cultivates hemp over an areaof around 20 thousand hectare, which will increase to about 670thousand hectare in 2020 (Jie, 2009). In the processing of hemp fibers,the hempseed becomes an interesting byproduct. Hempseedcontains 25–35% oil, 20–25% protein, 20–30% carbohydrates, and10–15% insoluble fiber and a rich array of minerals (Oomahaet al., 2002), which could be expressed and refined to obtain thehempseed oil. Using hempseed oil as raw materials for biodieselproduction is one of an important application of hempseed oil.Hempseed oil is a typical highly poly-unsaturated vegetable oilwith an iodine value of 164 g I 2 /100 g, which is eliminated by aspecification for iodine value (IV; 120 g I 2 /100 g maximum) of EN14214 as a potential feedstock for direct transesterification to producebiodiesel. Therefore, the present paper establishes a one-potprocess combining transesterification and selective hydrogenationfor biodiesel production from the highly poly-unsaturated oil ofhempseed. Alkaline earth metal oxides and alkaline earth oxidessupported copper oxide are prepared and examined as catalysts,and cetane numbers and iodine values of selective hydrogenatedmethyl esters of hempseed oil were also investigated.2. Methods2.1. MaterialsRefined hemp (Cannabis sativa L.) seed oil, obtained from XishuangbannaDai autonomous prefecture, Yunnan province, was usedas the starting material since it had an iodine value of 164 g I 2 /100 g which was higher than 120 g I 2 /100 g required by EN14214. The acid value was less than 0.1 mg KOH/g. The compositionsof hempseed oil were analyzed in our laboratory throughtransesterification at 65 °C for 3 h with methanol using 1 wt.%NaOH. According to GC analysis, the fatty acid consisted of palmiticacid 6.64%, stearic acid 2.76%, oleic acid 10.12%, linoleic acid54.31%, linolenic acid (C18:3) 23.76%, and traces of other acids.All chemicals used were of analytical grade, which were purchasedfrom Beijing Chemical Co., Ltd., China.2.2. Catalyst preparationMgO, CaO, SrO, BaO, and CuO catalysts were prepared by theconventional precipitation method from Mg(NO 3 ) 2 6H 2 O, Ca(-NO 3 ) 2 4H 2 O, Sr(NO 3 ) 2 , Ba(NO 3 ) 2 , and Cu(NO 3 ) 2 3H 2 O using ammoniumbicarbonate as precipitate. The catalyst preparation wasattempted into a three neck flask equipped with mechanical stirrerand a dropping funnel. Two aqueous solutions were first prepared.For a typical catalyst preparation, the first was a solution of25.6 g Mg (NO 3 ) 2 6H 2 O, (0.10 mol) in 100 ml distilled water. Thesecond was a solution of 9.6 g ammonium bicarbonate (anhydrous,0.10 mol) in 100 ml distilled water. The first solution was put intothe flask where the second one was added under vigorous stirringat ambient temperature. After that, the mixture was stirred for 4 hto form precipitate. The precipitate was washed with distilledwater, dried under vacuum at 120 °C, and then calcinated at1000 °C in vacuum for 5 h.Alkaline earth oxides loaded with copper oxide were preparedby the chemisorption–hydrolysis (CH) method (Nicoletta et al.,2002). The support was added to a (Cu(NH 3 ) 4 ) 2+ solution preparedby adding NH 4 OH to a Cu(NO 3 ) 2 3H 2 O solution until pH was 9.After 20 min under stirring, the slurry was slowly diluted in orderto allow hydrolysis of the copper complex and deposition of the finelydispersed product to occur. After that, the solids were separatedby filtration, dried overnight at 120 °C. Prior to eachreaction, the as-prepared catalysts were calcined at 1000 °C in vacuumfor 5 h. The loading amounts of copper oxide were calculatedon the basis of the amounts of the initial materials.2.3. Base strength determinationBasic strengths of the catalysts (H ) were determined usingHammett indicators (Kawashima et al., 2009). About 25 mg ofthe catalyst sample was shaken with 5 mL of a solution of Hammettindicator diluted in methanol and left to equilibrate for 2 h.After the equilibration, the color of the catalyst was noted. The followingHammett indictors were used: neutral red (pKBH + = 6.8),phenolphthalein (pKBH + = 9.3), 2,4-dinitroaniline (pKBH + = 15.0),4-nitroaniline (pKBH + = 18.4) and 4-chloroaniline (pKBH + = 26.5).The base strength is quoted as being stronger than the weakestindicator which exhibits a color change, but weaker than the strongestindicator that produces no change.2.4. Reaction procedureThe catalytic performance of alkaline earth metal oxides andalkaline earth metal oxides supported copper oxide were evaluatedin the one-pot process combining transesterification of hempseedoil with methanol and selective hydrogenation of methyl esters.For a typical reaction, 100 ± 0.01 g of hempseed oil and50 ± 0.1 mL of anhydrous methanol (molar ratio of methanol tohempseed oil = 12:1) with 3 ± 0.01 g of catalyst immersed (weightratio of catalyst to hempseed oil = 3%) was charged into a 500 mLstainless steal autoclave, which equipped with a stirrer and a coolingjacket surrounded by a heating mantle controlled by a proportionalIntegra derivative (PID) temperature controller. Theautoclave was purged with nitrogen, and the last traces of N 2 /airwere removed by a number of pressurizing/depressurizing cycleswith H 2 before the reaction and then the autoclave was pressured

R. Yang et al. / Bioresource Technology 101 (2010) 5903–5909 5905and the contents were heated to 180 ± 1 °C with a rate of 5 °C/min,constant in all the experiments with constant stirring at 1000 rpm.The one-pot process was carried out at 3.0 ± 0.01 MPa for 3 h. Atthe end of reaction, the autoclave was cooled to room temperature,vented of H 2 and the product mixture was removed. The catalystwas separated from the reaction solution by filtration. The alcoholphase (glycerin and methanol in excess) was separated from theorganic phase (methyl esters) by decantation, and finally the residualmethanol in the methyl ester phase was eliminated bydistillation.2.5. Product analysisThe products collected were immediately analyzed by a SP-3420A gas chromatograph from Beijing Beifen–Ruili AnalyticalInstrument (Group) Co. Ltd., equipped with a flame ionizationdetector (FID) and a BFSP-0677-02 capillary column (2 m 0.32mm 0.25 lm). Methyl heptadecanoate was used as the internalstandard to determine the amounts of products. The injector temperaturewas 240 °C and the detector temperature was 250 °C. Theanalysis of biodiesel for each sample was carried out by dissolving1 mL of biodiesel sample into 5 mL of acetone and injecting 1 ll ofthis solution in GC. The yield of biodiesel (FAME) with respect tohempseed oil was calculated from the content of methyl estersanalyzed by GC with the following equation:Amount of biodieselðmolÞYield ¼3 Charge amount of oilðmolÞ 100ð%ÞCetane numbers of biodiesel samples were determined in theWaukesha A111462B, which was calibrated with hexadecane and2,2,4,4,6,8,8-heptamethylnonane (HMN) as primary referencefuels, according to GB/T386 method (China-GB/T386, 1991). Iodinevalues of biodiesel samples were determined by AOCS Officialmethods Cd 1-25 (AOCS, 1089). Density, kinematic viscosity at40 °C, flash point in open vase, and pour point were evaluatedaccording to American Society for Testing and Materials standardmethods: ASTM D1298, ASTM D445, ASTM D92, ASTM D97, respectively.The acid value was determined according to the InternationalOrganization for Standardization ISO 660.3. Results and discussion3.1. Catalytic activities of transesterification and hydrogenation in theone-pot processAlkaline earth metal oxides, including MgO, CaO, SrO, and BaO,used as heterogeneous basic catalysts for the transesterification oftriglycerides with methanol have been published. The obtained resultsrevealed that alkaline earth metal oxides became active fortransesterification after a thermal pretreatment and the catalyticactivities were predominant upon their alkalinities (Helwaniet al., 2009; Patil and Deng, 2009; Singh and Fernando, 2007). Asthe basic strength increased in the order of MgO < CaO < SrO < BaO,which was attributed to the decrease of electro-negativity of theconjugated metal cation of alkaline earth metal oxides, the catalyticactivities of transesterification exhibited in the same sequence. SrOhad basic sites stronger than H = 26.5 and gave a biodiesel yield inexcess of 95% at 65 °C within 30 min in the transesterification ofsoybean oil to biodiesel (Liu et al., 2007). MgO, CaO, SrO, and BaOwere also reported to show catalytic activity for hydrogenation ofolefins. The alkaline earth metal oxides became active on evacuatingabove 600 °C and showed maximum activities in the order ofMgO < CaO < BaO < SrO when evacuated around 1000 °C (Hattoriet al., 1975). It is suggested that alkaline earth metal oxides evacuatedat high temperature showed catalytic activity for transesterificationof triglycerides with methanol or hydrogenation of olefins. Inthis paper, therefore, alkaline earth metal oxides were investigatedfor a one-pot process combining transesterification of hempseed oilwith methanol and hydrogenation of hempseed oil methyl esters.Basic strengths and catalytic activities in terms of biodiesel yieldsand the compositions of biodiesels of alkaline earth metal oxideswere listed in Table 1. As shown in Table 1, the basic strength ofalkaline earth metal oxides, which were determined using Hammettindicators, increased in the order Mg < Ca < Sr = Ba. The SrOand BaO samples owned the highest basic strength with H > 26.5and gave high activities for transesterification, whereas the MgOand CaO samples possessed weaker basic sites with H < 26.5 andperformed lower biodiesel yields. It seems that such catalytic activitiesfor transesterification of hempseed oil with methanol at ahydrogen pressure are dependent upon the basic strengths of alkalineearth metal oxides which are consistent with those reported byPatil and Deng (2009). Compared with the biodiesel obtainedthrough transesterification by NaOH at 65 °C (see Table 1), the biodieselsobtained by alkaline earth metal oxides showed no obviousdifferences of iodine value and the amount of un-saturates such aslinolenate (C18:3), linoleate (C18: 2), and oleate (C18:1), indicatedthat alkaline earth metal oxides gave poor activity for hydrogenation.It is likely that alkaline earth metal oxides are effective catalystsfor transesterification of hempseed oil with methanol butcan not efficiently catalyze the hydrogenation of fatty acid esters.Alkaline earth metal oxides are active catalysts for the transesterificationof hempseed oil with methanol at a hydrogen pressure,as discussed above, while copper catalysts were reported tobe active in the hydrogenation of fatty acid methyl esters and triglycerideand selective for the reduction of linolenate (C18:3) tooleate (C18:1) leaving stearate (C18:0) unaffected (Federico et al.,2006; Nicoletta et al., 2002). A series of alkaline earth metal oxidesloaded with 10 wt.% CuO were prepared and tested in the one-potprocess to examine the influence of the support on the catalyticperformances of transesterification and hydrogenation. Basicstrengths and catalytic activities in terms of biodiesel yields andthe compositions as well as iodine values were also listed in Table1. As can be seen, MgO loaded with copper oxide exhibited thesame basic strength as MgO with H in the range of 9.3–15,whereas CaO, SrO, and BaO loaded with copper oxide showed alower basic strength with H in the range of 15–18.4 comparedto those of the pure alkaline earth metal oxides. It suggested thatthe loading of copper oxide led to decreases of basic strengths ofalkaline earth metal oxides. With the reduction of basic strength,thereafter, the biodiesel yields decreased slightly. The as-synthesizedbiodiesels by alkaline earth metal oxides loaded with copperoxide showed lower amounts of methyl linolenate (C18:3) andmethyl linoleate (C18:2), and higher amount of methyl oleate(C18:1), resulting in lower iodine values, which indicated thatalkaline earth metal oxides loaded with copper oxide showed higheractivities toward hydrogenation and higher selectivity forincreasing methyl oleate (C18:1) than alkaline earth metal oxides.Among the catalysts tested, SrO loaded with copper oxide gave thehighest decrease of iodine value from 164 g I 2 /100 g to 113 g I 2 /100 g and presented the lowest methyl linolenate (18:3) andmethyl linoleate (C18:2) contents and the highest amount ofmethyl oleate (C18:1) with the methyl stearate (C18:0) and methylpalmitate (C16:0) unchanged. It seems that alkaline earth metaloxides loaded with copper oxide were both active for the transesterificationof hempseed oil with methanol and selective hydrogenationof methyl esters and Cu/SrO catalysts exhibited the bestcatalytic performance of transesterification of hempseed oil withmethanol and selective hydrogenation of methyl esters under theexperimental conditions used.The influence of the copper oxide loading in Cu/SrO catalysts onthe activities for transesterification and selective hydrogenation

Table 1Basic strengths and catalytic activities in terms of biodiesel yields and the compositions of biodiesel obtained of alkaline earth metal oxides and alkaline earth metal oxides supported copper oxide a .Entry Catalysts b Basic strength (H ) Biodiesel yield (%) C16:0 (%) C18:0 (%) C18:1 (%) C18:2 (%) C18:3 (%) Total satures (%) Total unsatures (%) Total PUFA c (%) IV (gI 2 /100 g)1 d NaOH 15 < H < 8.4 98 (±1.1) 6.64(±0.1) 2.76(±0.1) 10.12(±0.5) 54.31(±0.5) 23.76(±0.4) 9.4 88.19 78.07 164 (±1.0)2 MgO 9.3 < H < 15 79 (±2.2) 6.58(±0.1) 2.78(±0.1) 16.21(±0.9) 50.34(±1.1) 21.67(±1.4) 9.36 88.22 71.01 158 (±1.1)3 CaO 18.4 < H < 26.5 84 (±1.3) 6.60(±0.1) 2.78(±0.1) 12.45(±0.8) 52.07(±0.9) 23.03(±1.2) 9.38 87.55 75.1 160 (±1.2)4 SrO 26.5 < H 96 (±1.5) 6.59(±0.1) 2.77(±0.1) 14.37(±0.9) 51.31(±1.3) 22.52(±0.9) 9.36 88.20 73.83 159 (±1.2)5 BaO 26.5 < H 95 (±2.3) 6.61(±0.1) 2.80(±0.1) 12.54(±1.1) 51.74(±1.7) 23.82(±1.5) 9.41 88.10 75.56 162 (±1.2)6 CuO H < 6.8 71(±2.1) 6.67(±0.1) 2.79(±0.1) 18.72(±1.0) 52.84(±1.8) 16.54(±2.1) 9.46 88.10 69.38 150 (±1.1)7 10%9.3 < H < 15 73 (±3.1) 6.71(±0.1) 2.75(±0.1) 37.65(±1.2) 47.51(±1.4) 3.28(±1.5) 9.46 98.44 50.79 131 (±1.1)Cu/MgO8 10%15 < H < 18.4 77 (±2.5) 6.72(±0.1) 2.76(±0.1) 39.23(±1.1) 44.35(±1.4) 4.47(±1.3) 9.48 88.05 48.82 121 (±1.3)Cu/CaO9 10%15 < H < 18.4 90 (±2.3) 6.65(±0.1) 2.76(±0.1) 17.82(±1.3) 52.34(±0.6) 18.29(±2.1) 9.41 88.45 70.63 153 (±1.2)Cu/BaO10 1% Cu/SrO 15 < H < 18.4 90 (±1.2) 6.63(±0.1) 2.74(±0.1) 16.34(±1.2) 52.19(±1.3) 19.16(±1.3) 9.37 87.69 71.35 154 (±1.2)11 3% Cu/SrO 15 < H < 18.4 92 (±2.6) 6.62(±0.1) 2.75(±0.1) 21.65(±1.1) 50.42(±1.1) 16.06(±1.2) 9.37 88.13 66.48 147 (±1.1)12 5% Cu/SrO 15 < H < 18.4 92 (±2.4) 6.68(±0.1) 2.78(±0.1) 24.96(±1.1) 49.07(±1.2) 14.07(±1.3) 9.46 88.10 63.14 142 (±1.3)13 7% Cu/SrO 15 < H < 18.4 94 (±2.3) 6.62(±0.1) 2.75(±0.1) 28.85(±1.2) 49.03(±0.9) 10.05(±1.1) 9.37 87.93 59.08 135 (±1.3)14 10%15 < H < 18.4 96 (±2.0) 6.64(±0.1) 2.76(±0.1) 46.65(±1.0) 41.73(±1.2) 0.49(±1.0) 9.4 88.87 42.22 113 (±1.3)Cu/SrO15 20%15 < H < 18.4 94 (±1.9) 6.67(±0.1) 2.71(±0.1) 42.17(±1.2) 41.61(±1.1) 3.16(±0.9) 9.38 86.94 44.77 116 (±1.1)Cu/SrO16 30%15 < H < 18.4 91 (±2.3) 6.68(±0.1) 2.74(±0.1) 42.87(±1.3) 41.03(±1.4) 3.88(±1.1) 9.42 87.78 44.91 117 (±1.1)Cu/SrO17 10%15 < H < 18.4 94 (±2.1) 6.64(±0.1) 16.75(±0.1) 18.68(±1.6) 44.98(±2.1) 10.51(±1.3) 23.39 74.17 55.49 121 (±1.3)Ni/SrO18 10%Ni–10%Cu/SrO15 < H < 18.4 95 (±2.5) 6.68(±0.1) 5.86(±0.1) 33.69(±1.8) 46.34(±1.5) 4.99(±1.4) 12.54 85.02 51.33 122 (±1.3)5906 R. Yang et al. / Bioresource Technology 101 (2010) 5903–5909aOne-pot process combining transesterification and selective hydrogenation from hempseed oil with methanol was performed at a hydrogen pressure of 3.0 MPa with molar ratio of methanol to oil = 12:1, and catalyst/oil = 3 wt.% for 3 h for all the reactions except for Entry 1.b Metal oxide loadings (wt.%) are referred to nominal value of CuO or NiO loading on the support.c PUFA means poly-un-saturate fatty acid methyl esters.d The initial composition and iodine value of starting hempseed oil are shown in the first row though a typical transesterification reaction by NaOH at 65 °C for 3 h with molar ratio of methanol to oil = 6:1, and catalyst/oil = 1.0 wt.%. Triplicate runs (standard deviations given in parentheses).

5908 R. Yang et al. / Bioresource Technology 101 (2010) 5903–5909Table 2Fuel properties of methyl esters of hempseed oil and hydrogenated methyl esters of hempseed oil obtained through one-pot process combining transesterification and selectivehydrogenation by 10% CuO/SrO catalysts.Property (units) Biodiesel Hydrogenated biodiesel ASTM D6751EN 14214GB/T20828-2007Density (g/cm 3 ) 0.8720.876 (±0.012) 0.87–0.90 0.860–0.900 0.86–0.90(±0.011)Kinematic viscosity (mm 2 /g) a 4.23 (±0.05) 4.31 (±0.04) 2.5–6.0 3.5–5.0 3.0–8.0Flash point (°C) 131 (±0.9) 137 (±1.2) Min. 130 Min. 120 Min. 130Cetane number 44.0 (±2.3) 55.8 (±2.1) Min. 47 Min. 51 Min. 49Iodine value (g I 2 /100 g) 164 (±1.0) 113 (±1.3) – Max. 120 –Acid number (mg NaOH/g) 0.01 (±0.01) 0.01 (±0.01)

R. Yang et al. / Bioresource Technology 101 (2010) 5903–5909 5909opportunity to extend the high un-saturate materials for biodieselproduction, thus making a wide portfolio of raw materials availablewithout interfering with the food market.ReferencesAzam, M.M., Waris, A., Nahar, N.M., 2005. Prospects and potential of fatty acidmethyl esters of some non-traditional seed oils for use as biodiesel in India.Biomass Bioenergy 29, 293–302.China-GB/T386, 1991. Diesel fuels-Determination of Ignition Quality by the CetaneMethod.Christos, E.P., Anastasia, L., Asimina, K., Nikolaos, K., Achilleas, C., Nikolaos, N., 2010.Optimization of cotton seed biodiesel quality (critical properties) throughmodification of its FAME composition by highly selective homogeneoushydrogenation. Bioresource Technology 101, 1812–1819.Dijkstra, A.J., 2002. On the mechanism of the copper-catalysed hydrogenation; areinterpretation of published data. European Journal of Lipid ScienceTechnology 104, 29–35.Dorado, M.P., Ballesteros, E., López, F.J., Mittelbach, M., 2004. Optimization of alkalicatalyzedtransesterification of Brassica carinata oil for biodiesel production.Energy Fuels 18, 77–83.Falk, O., Meyer-Pittroff, R., 2004. The effect of fatty acid composition on biodieseloxidative stability. European Journal of Lipid Science Technology 106, 837–843.Federica, Z., Rinaldo, P., Nicoletta, R.M., 2009. Selective hydrogenation of alternativeoils: a useful tool for the production of biofuels. Green Chemistry 11, 462–465.Federico, B.P., Nicoletta, R.M., Federica, Z., 2006. A method for the production ofbiodiesel, starting from high iodine number fatty substances. WO 2006111997,A1.Fernandez, M.B., Piqueras, C.M., Tonetto, G.M., Crapiste, G., Damiani, D.E., 2005.Hydrogenation of edible oil over Pd–Me/Al 2 O 3 catalysts (Me = Mo, V and Pb).Journal of Molecular Catalysis A: Chemical 233, 133–139.Frankel, E.N., 2005. Lipid Oxidation, second ed. Oily Press, Bridgewater, UK.Hattori, H., Tanaka, Y., Tanabe, K., 1975. Hydrogenation of olefins catalyzed byalkaline earth metal oxides. Chemistry Letters 4, 659–660.Helwani, Z., Othman, M.R., Aziz, N., Kim, J., Fernando, W.J.N., 2009. Solidheterogeneous catalysts for transesterification of triglycerides with methanol:a review. Applied Catalysis, A: General 363, 1–10.Imahara, H., Minami, E., Saka, S., 2006. Thermodynamic study on cloud point ofbiodiesel with its fatty acid composition. Fuel 85, 1666–1670.Jie, J., 2009. This is the moment of China’ hemp. China Fiber Inspection 6, 46–47.Johansson, L.E., Lundin, S.T., 1979. Copper catalysts in the selective hydrogenation ofsoybean and rapeseed oils: I. The activity of the copper chromite catalyst.Journal of the American Oil Chemists’ Society 56, 974–980.Karabulut, I., Kayahanb, M., Yaprak, S., 2003. Determination of changes in somephysical and chemical properties of soybean oil during hydrogenation. FoodChemistry 81, 453–456.Kawashima, A., Matsubara, K., Honda, K., 2009. Acceleration of catalytic activity ofcalcium oxide for biodiesel production. Bioresource Technology 100, 696–700.Knothe, G., 2005. Dependence of biodiesel fuel properties on the structure of fattyacid alkyl esters. Fuel Processing Technology 86, 1059–1070.Knothe, G., 2006. Analysis of oxidized biodiesel by 1 H NMR and effect of contactarea with air. European Journal of Lipid Science and Technology 108, 493–500.Knothe, G., 2008. ‘‘Designer” biodiesel: optimizing fatty ester composition toimprove fuel properties. Energy Fuels 22, 1358–1364.Liu, X., He, H., Wang, Y., Zhu, S., 2007. Transesterification of soybean oil to biodieselusing SrO as a solid base catalyst. Catalysis Communications 8, 1107–1111.Mccormick, R.L., Graboski, M.S., Alleman, T.L., Herring, A.M., 2001. EnvironmentalScience and Technology 35, 1742–1747.Moraes, M.S.A., Krause, L.E., Cunha, M.E., Faccini, C.S., Menezes, E.W., Veses, R.C.,Rodrigues, M.R.A., Caramão, E.B., 2008. Tallow biodiesel: properties evaluationand consumption tests in a diesel engine. Energy Fuels 22, 1949–1954.Moser, B.R., Haas, M.J., Winkler, J.K., Jackson, M.A., Erhan, S.Z., List, G.R., 2007.Evaluation of partially hydrogenated methyl esters of soybean oil as biodiesel.European Journal of Lipid Science and Technology 109, 17–24.Nicoletta, R., Federica, Z., Michele, G., Sandro, R., Achille, F., Nicola, P., Rinaldo, P.,2002. Environmental friendly lubricants through selective hydrogenation ofrapeseed oil over supported copper catalysts. Applied Catalysis, A: General 233,1–6.Nikolaou, N., Papadopoulos, C.E., Lazaridou, A., Koutsoumba, A., Bouriazos, A.,Papadogianakis, G., 2009. Partial hydrogenation of methyl esters of sunfloweroil catalyzed by highly active rhodium sulfonated triphenylphosphitecomplexes. Catalysis Communications 10, 451–455.Nohair, B., Especel, C., Lafaye, G., Marécot, P., Hoang, L.C., Barbier, C., 2005.Palladium supported catalysts for the selective hydrogenation of sunflower oil.Journal of Molecular Catalysis A: Chemical 229, 117–126.Oomaha, B.D., Busson, M., Godfrey, D.V., Drover, J.C.G., 2002. Characteristics ofhemp (Cannabis sativa L.) seed oil. Food Chemistry 76, 33–43.Patil, P.D., Deng, S., 2009. Transesterification of Camelina sativa oil usingheterogeneous metal oxide catalysts. Energy Fuels 23, 4619–4624.Pérez-Cadenas, A.F., Zieverink, M.M.P., Kapteijn, F., Moulijn, J.A., 2006. Selectivehydrogenation of fatty acid methyl esters on palladium catalysts supported oncarbon-coated monoliths. Carbon 44, 158–193.Singh, A.K., Fernando, S.D., 2007. Reaction kinetics of soybean oil transesterificationusing heterogeneous metal oxide catalysts. Chemical Engineering andTechnology 30, 1716–1720.Su, Z., Yuan-Gang, Z., Yu-Jie, F., Meng, L., Dong-Yang, Z., Thomas, E., 2010. Rapidmicrowave-assisted transesterification of yellow horn oil to biodiesel using aheteropolyacid solid catalyst. Bioresource Technology 101, 931–936.UNE-EN 14214, 2003. Automotive Fuels. Fatty Acid Methyl Esters (FAME) for DieselEngines. Requirements and Test Methods.

iomass and bioenergy 34 (2010) 272–277Available at www.sciencedirect.comhttp://www.elsevier.com/locate/biombioeBiodiesel production from acid oils and ethanol using a solidbasic resin as catalystJ.M. Marchetti*, A.F. ErrazuPlanta Piloto de Ingeniería Química, UNS-CONICET, Camino La Carrindanga km 7, (8000) Bahía Blanca, Argentinaarticle infoArticle history:Received 21 October 2008Received in revised form23 October 2009Accepted 26 October 2009Available online 20 November 2009Keywords:FAEE productionSolid resin catalystAcid oilsabstractIn the search of an alternative fuel to substitute diesel fuel, biodiesel appears as one of themost promising sources of energy for diesel engines because of its environmentaladvantages and also due to the evolution of the petroleum market.Refined oil is the conventional raw material for the production of this biofuel; however, itsmajor disadvantage is the high cost of its production. Therefore, frying oils, waste oils,crude oils and/or acid oils are being tested as alternative raw materials; nevertheless, therewill be some problems if a homogeneous basic catalyst (NaOH) is employed due to the highamount of free fatty acid present in the raw oil.In this work, the transesterification reaction of acid oil using solid resin, Dowex monosphere550 A, was studied as an alternative process. Ethanol was employed to havea natural and sustainable final product. The reaction temperature’s effects, the initialamount of free fatty acid, the molar ratio of alcohol/oil and the type of catalyst (homogeneousor heterogeneous) over the main reaction are analyzed and their effectscompared.The results obtained show that the solid resin is an alternative catalyst to be used toproduce fatty acid ethyl esters (FAEEs) by a transesterification reaction with a finalconversion over 90%. On the other hand, the time required to achieve this conversion isbigger than the one required using conventional technology which employs a homogeneousbasic catalyst. This reaction time needs to be optimized.ª 2009 Elsevier Ltd. All rights reserved.1. IntroductionOver the last few years, as well as projecting for the following15–20 years, the consumption of petroleum based fuels willhave increased from 11.52 millions of tonnes per day in 2005 toan estimate of 15.7 millions of tonnes per day in 2030. Simultaneouslyto this grow it will be associated a decreasing on thefossil fuels’ reserves, producing a net increase on its sellingprice, reaching historical values over US$ 100 per 0.1388 tonnes[1]. Because of this, researches have been done in order to findan alternative substitute for diesel fuels, being biodiesel,defined by ASTM as the mono alkyl ester produced from vegetableoils or animal fats, one of the most promising alternatives.Some of the major advantages of this biofuel are that it isbiodegradable, produced from a renewable source and that itdoes not produce carbon monoxide or any sulfur residues[2–6]. Despite all these advantages, one of the mayor drawbackson the biodiesel production process is its cost; mainlyassociated with the price of the oil used related to its quality[7]. Due to this economic restriction, alternative raw materialsof a lower price/cost, like frying oils, soap stocks, acid oils, andwaste oils, have been studied [8–13].* Corresponding author. Tel.: þ54 291 4861700x267; fax: þ54 291 4861600.E-mail address: jmarchetti@plapiqui.edu.ar (J.M. Marchetti).0961-9534/$ – see front matter ª 2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.biombioe.2009.10.016

iomass and bioenergy 34 (2010) 272–277 273Acid oils present high amounts of free fatty acids (FFAs),varying from 3% to 40% [2,4,9], which make non-recommendableits usage with the conventional technology thatemploys a basic homogeneous catalyst such as sodium orpotassium hydroxides. The basic catalyst and the FFA willinteract to produce soap, according to Equation (1). Thismakes the amount of available catalyst for the transesterificationreaction to be reduced and also complicates thedown streaming separation and the biodiesel purification.a pretreatment). The reactant and the reaction studied inMarchetti et al. [23] are the FFA and the esterification reaction,while for this work the reactant is the triglycerides andthe reaction is the transesterification reaction, both reactionand reactants are different and this work is a complementarystudied for the previous work [23].It has been studied and compared the effects of the mainvariables of the whole process: reaction temperature, molarratio of alcohol/oil and initial amount of free fatty acid. It has(1)Alternative processes for FAEE production have been underdevelopment in order to employ different catalysts: homogeneousones (sulfuric acid for example), as well as heterogeneousones such as enzymes or solid resin. Solid catalystshave been useful to carry out the transesterification reactionof refine oil achieving conversion around 79.6%, 91.07%, 95.1%and 95% in 3, 2, 7 and 3 h respectively for each research work[14–18]. On the other hand, several research works showedthat the esterification reaction can be carried out in thepresence of triglycerides using different catalysts [19–22].Biodiesel is generally produced by transesterification oftriglyceride, but it might also be generated by direct esterificationof free fatty acid; both reactions could be summarizedas followsbeen also compared this alternative with sulfuric acid undersimilar operational conditions.2. Materials and methodsRefined sunflower oil from a commercial brand was employedto carry out this study; also, oleic acid provided by Anedra andethanol anhydrous providedby either Anedra or Cicarelli wereemployed.The acid oil used as raw material was made in the laboratoryby mixing measured amounts of refined oil and oleicacid. Once the mixture was made, a small sample wasanalyzed to reassure the initial amount of free fatty acid; thiswhere R 1 ,R 2 ,R 3 and R 0 denote any hydrocarbon chain.The employment of a homogeneous catalyst, such assulfuric acid, has some drawbacks associated with theprocess itself. The down streaming separation and purificationare more complicated than when a solid catalyst isused.In Marchetti et al. [23] it has been prepared as a mixtureof TG (triglycerides) and FFA (free fatty acid). In that work, ithas been tested this mixture to see how FFAs are convertedinto FAME by the esterification reaction, with the aim ofusing this catalyst as a pretreatment; therefore, no studywas done on the triglycerides. In this work, the samemixture was used and the same reaction conditions wereused, but the conversion of triglycerides was studied to see ifthis catalyst could be used to produce biodiesel (not only aswas done by titration [23]. It was employed a lab scale reactorwith a volume of 500 ml. The system contains a warmer jacketthat allows us to set the reaction temperature with a margin oferror of 0.1 C. Since the reaction was carried out ina temperature range between 30 C and 55 C, water wasemployed as heating stream.The mixture (TG þ FFA) was fed into the reactor to be preheated,when reaching the reaction temperature, the solidcatalyst as well as the ethanol was added to the reactionmedium in order to start the reaction, making a pseudohomogeneousreaction environment. All these experimentshave been done with 200 rpm to have a good mix of thecompounds and eliminate possible mass transfer problems.At specified times, samples were withdrawn from thereaction mixture and washed with water to separate the

274biomass and bioenergy 34 (2010) 272–277alcohol from the mixture and to assure that the reaction wasstopped. Each sample was centrifuged for 15–20 min to havea better separation between the water phase and the oilphase. From the oil phase, a small amount was taken andprepared for a gas chromatography (GC) analysis. It wasemployed a GC Varian 3700 with a high temperature columnDB-5HTJ&W Scientific (15 m long, 0.32 mm I.D. and 0.1 mm forthe film) to quantify triglycerides, diglycerides, monoglyceridesand esters. This technique allows us to analyze thetotal amount of FAEE produced, not only from the TG but alsofrom the FFA.The reproducibility of the analysis has been tested forrandom amount of samples at random time in order toassure that it was a verifiable value the one obtained fromthe GC.3. Results and discussionThe transesterification reaction is a series of consecutivereactions, in which from triglycerides, diglycerides areproduced; the latter one is transformed into monoglycerideswhich, in a following step, are modified into glycerol. In allthese steps FAEE is produced. Fig. 1 shows the maincompounds involved on this reaction when the reactiontemperature was 45 C, an initial amount of free fatty acidsof 19.49% w/w, 2.5% w/w of catalyst, 200 rpm of agitationand a molar ratio alcohol/mixture of 6.1:1. In Fig. 1 it can beseen that while esters are produced, triglycerides areconsumed. Diglycerides and monoglycerides go througha maximum because they are produced and consumedduring the reaction.In order to assure that the catalyst will not interact with themedium in any other way that to carry out the desirablereaction, a simple experiment was done by mixing the solidresin and fatty acid and measuring the acidity of the medium.It was not measured any modification in the acidity of themedium, showing that no soaps have been produced and nocatalyst has been consumed.3.1. Effect of the reaction temperatureVariations over the reaction temperature have been analyzed,three temperatures, 30, 45 and 55 C, have been tested whilethe remains of the operation conditions were set at 200 rpm,molar ratio alcohol/mixture 6.1:1, 2.2 w/w % of catalyst, 9.9%w/w of initial free fatty acids. A higher temperature could notbe tested since this catalyst has a maximum operationaltemperature of 60 C.Fig. 2 shows that when the temperature increases the finalconversion increases as well. Nevertheless, at short reactiontimes, the initial reaction rate follows a non-commonbehavior producing a slower reaction while the reactiontemperature increases. These results might be due to a partialdeactivation of the catalyst as a result of been working close tothe maximum allowed temperature (60 ). A similar effect wasseen when the esterification reaction was studied [23]; asa result, certain deactivation of the catalyst, due to an operationclose to the maximum allowed temperature, wasconfirmed.A similar work was done by Hamad et al. [17] where theyreached similar final conversion using an acid catalyst. Theyhave used ethanol as well but with rapeseed oil as rawmaterial. The reaction reached the desirable conversion in3 h., being a much better performance than from our basecatalyst. While Noiroj et al. [15] reached slightly lowerconversion in 6 h of reaction. The authors used a KOH/NaYcatalyst.3.2. Effect of the molar ratio alcohol/mixtureVariations on the amount of alcohol have been done (N ¼ 4.2:1,5.01:1 and 6.1:1) to study the effect of increasing one of the rawmaterials with respect to the other. The reaction was carriedout at a constant temperature of 45 C, an agitation speed of200 rpm, 2.2% w/w of the catalyst was employed, as well as aninitial amount of FFA of 9.9% w/w.Fig. 3 shows that the final conversion into esters increaseswhen the amount of ethanol is increased. However, the initialFig. 1 – Variations of triglycerides (:), diglycerides (-),monoglycerides (3) and esters (C).Fig. 2 – Influence of the reaction temperature. (-)T[ 30, (C)T [ 45, (:) T[ 55 8C.

iomass and bioenergy 34 (2010) 272–277 277[22] López DE, Goodwing Jr JG, Bruce DA, Furuta S.Esterification and transesterification using modifiedzirconia catalyst. Applied Catalysis A: General 2008;339:76–83.[23] Marchetti JM, Miguel VU, Errazu AF. Heterogeneousesterification of oil with high amount of free fatty acid. Fuel2007;86:906–10.[24] Tomasevic AV, Siler-Marinkovic SS. Methanolysis of usedfrying oils. Fuel Processing Technology 2003;81:1–6.[25] Darnoko D, Cheryan M. Kinetics of palm oiltransesterification in a batch reactor. Journal of theAmerican Oil Chemists Society 2000;77(12):1263–7.[26] Canakci M, Van Gerpen J. Biodiesel production via acidcatalysis. Transactions of the American Society ofAgricultural Engineers 1999;42(5):1203–10.[27] Canakci M, Van Gerpen J. Biodiesel production from oils andfats with high free fatty acids. Transactions of the AmericanSociety of Agricultural Engineers 2001;44(6):1429–36.

Bioresource Technology 101 (2010) 1834–1843Contents lists available at ScienceDirectBioresource Technologyjournal homepage: www.elsevier.com/locate/biortechSupercritical biodiesel production and power cogeneration: Technicaland economic feasibilitiesA. Deshpande, G. Anitescu, P.A. Rice, L.L. Tavlarides *Dept. of Biomedical and Chemical Engineering, Syracuse University, Syracuse, NY 13244, United StatesarticleinfoabstractArticle history:Received 2 January 2008Received in revised form 28 September2009Accepted 10 October 2009Available online 24 November 2009Keywords:BiodieselTechnical analysisEconomic assessmentSupercritical transesterificationAn integrated supercritical fluid technology with power cogeneration to produce biodiesel fuels, with noneed for the costly separations involved with the conventional technology, is proposed, documented fortechnical and economic feasibility, and preliminarily designed. The core of the integrated system consistsof the transesterification of various triglyceride sources (e.g., vegetable oils and animal fats) with supercriticalmethanol/ethanol. Part of the reaction products can be combusted by a diesel power generatorintegrated in the system which, in turn, provides the power needed to pressurize the system and the heatof the exhaust gases necessary in the transesterification step. The latter energy demand can also be satisfiedby a fired heater, especially for higher plant capacities. Different versions of this system can beimplemented based on the main target of the technology: biodiesel production or diesel engine applications,including power generation. The process options considered for biodiesel fuel production estimatebreak-even processing costs of biodiesel as low as $0.26/gal ($0.07/L) with a diesel power generator and$0.35/gal ($0.09/L) with a fired heater for a plant capacity of 15,000 gal/day (56,775 L/day). Both are significantlylower than the current processing costs of approximately $0.51/gal ($0.13/L) of biodiesel producedby conventional catalytic methods. A retail cost of biodiesel produced by the proposed method islikely to be competitive with the prices of diesel fuels.Ó 2009 Elsevier Ltd. All rights reserved.1. IntroductionSignificant investments in biofuels, both in research and development(R&D) and in the construction of new commercial plants,emphasize a continued commitment to alternative energy as availability,volatility, security and environmental impact of the fossilfuels drive the need for new and renewable energy sources. Biodiesel(BD), a mixture of fatty acid alkyl esters, derived from plant andanimal triglycerides (TG) through transesterification (TE) reactionswith an alcohol, is one such fuel that is under a great deal of consideration.While worldwide the TG sources are diverse, over 90%of BD current production in the United States is based on soybeanoil (SO) (Collins, 2006). It has been estimated that BD yields 93%more energy than that invested in its production and, relative tofossil fuels, greenhouse gas emissions are reduced 41% by the BDproduction and combustion while less air pollutants are releasedper net energy gain (Hill et al., 2006). Unfortunately, there are concernsthat BD cannot replace much petrodiesel without impactingfood supplies by the extension of the areas of oilseed cultivation.Yet, there is significant feedstock of waste cooking oil, animal fats(e.g., yellow grease and tallow), non-edible and potentially large* Corresponding author.E-mail address: lltavlar@syr.edu (L.L. Tavlarides).excess of edible oils which do not negatively impact the alimentaryTG supply. Also, by increasing, for example, the production of soybeans,there is a positive feedback on agriculture through higherquantity of soybean meals made available as animal feedstock afterSO extraction (Biodiesel, 2008). This also means more meat forfood and more TG supply for BD production.Although these benefits are very attractive, the final cost of BDproduced by conventional means is still prohibitively high withoutgovernmental subsidies (Radich, 2004; USA Biodiesel Prices, 2009).Compared with conventional acid/base catalytic methods, relativelynew supercritical (SC) methods were found not only beingcapable to reduce TE time from several hours to a few minutesthrough a continuous TE process, but also to require no TG feedpretreatment for the FFA and water (Kusdiana and Saka, 2004).Without any significant pretreatment, nearly all of the TG content,even from low-quality feedstock, could be turned into high-qualityBD fuels.Despite these advantages, given the high number of the competitiveprocess parameters affecting SC TE conversions, it is difficultto successfully control their effects. Indeed, there are not eventwo reported sets of data gathered by SC methods with the claimedoptimum TE conditions being similar (Table 1). Although highqualityBD fuels have been eventually obtained at laboratory-scalethrough often sophisticated processing and purification steps, the0960-8524/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.biortech.2009.10.034

A. Deshpande et al. / Bioresource Technology 101 (2010) 1834–1843 1835Table 1Reported optimum conditions and yield for batch (B) and continuous (C) SC TE processes.Oil/cosolvent T (°C) P (bar) TG/MeOH c s (min) B/C Yield (%) ReferencesaSoybean/CO 2 280 143 24 10 B 98 Han et al. (2005)Canola 450 400 45 4 C Near 100 Iijima et al. (2004)Rapeseed 350 450 42 4 B 95 Saka and Kusdiana (2001); Kusdiana and Saka (2001)Coconut and palm 350 190 42 7 C 96 Bunyakiat et al. (2006)Soybean 310 350 40 25 C 96 He et al. (2007)Soybean 235 62 27 600 B 85 Diasakou et al. (1998)bSoybean/C 3 H 8 280 128 24 10 B 98 Cao et al. (2005)Sunflower 350 200 40 40 B 96 Madras et al. (2004)Castor and linseed 350 200 40 40 B 98 Varma and Madras (2007)bSoybean/C 3 H 8 288 96 64 10 B 99 Hegel et al. (2007)Soybean 290 110 12 B Near 100 D’Ippolito et al. (2007)bWaste oil/C 3 H 8 280 128 24 17 C 95 Kasteren and Nisworo (2007)Soybean 400 100 3 3 C Near 100 This studya CO 2 /MeOH = 0.1.b C 3 H 8 /MeOH = 0.05.c Molar ratio.market price of these fuels from the industrial production facilitiesbeing competitive with that of diesel fuel is the ultimate criterionin the acceptance of any of the proposed methods. Unfortunately,technical and economic feasibility studies of creating a modernalternative to conventional BD industry are scarce (D’Ippolitoet al., 2007; Kasteren and Nisworo, 2007; Lim et al., 2009).We conceptually designed an integrated two-step SC technologyto produce both vegetable oils and BD fuels by extracting theoils with SC CO 2 followed by SC TE with methanol/ethanol in a continuoustubular reactor (Fig. 1). The small amount of CO 2 remainingin the oil in the flash tank exhibits a positive effect ascosolvent to increase the mutual oil–alcohol solubility. Themechanical and electrical power as well as the necessary heat forthese processes is provided by a diesel power generator which willconsume a fraction of the produced fuel. Optionally, the requiredTE heat can be supplied by a fired heater. In this paper only theTE step will be discussed. The overall goal of this study was to developa simplified process for continuous and cost-effective BDproduction based on our laboratory-scale screening experiments(Anitescu et al., 2008). Special emphasis was on technical and economicanalyses performed for the proposed process in order tocompare the processing costs at commercial scale with those ofthe current methods.The proposed SC TE technology to produce BD fuels consists of asimple, non catalytic process with no need for the costly BD–alcohol–catalyst–glycerolseparations required for the acid/base conventionalTE. This simplification is made possible by using aclose to stoichiometric proportion of oil and alcohol and by decomposingthe glycerol byproduct in smaller molecular fuel componentsat high reaction temperatures (Anitescu et al., 2008). Whenthis SC TE method (Fig. 2) is compared with a conventional catalytictechnology (Fig. 3), the simplicity of the former is apparent.The method will use a diesel power generator to provide powerand heat for the upstream processes or a highly efficient fired heaterwhich will consume a fraction of the produced fuel.The proposed integrated system has a potentially wide range ofapplications. First, it is very attractive for farmers since it reducestheir dependency on high petrodiesel prices. Many farmers usethe oil from soybeans (18–20% oil content) as a fuel. However,SO alone is not an acceptable fuel because of its very high viscosity,combustion emissions and damage it renders to the engines, especiallyon long-term usage. Alternatively, further processing the oilto BD with associated power generation, is a suitable solutionespecially for remote farms. Also, as large biofuel plants encounterresidential resistance, smaller and process-contained modules certainlywill have a much better environmental fit. Secondly, thisFig. 1. Schematic of the Integrated Multistage Supercritical Technology System to produce vegetable oils and biofuels.

1836 A. Deshpande et al. / Bioresource Technology 101 (2010) 1834–1843Fig. 2. Flow diagram for supercritical transesterification of vegetable oils to biodiesel.Fig. 3. A typical flow diagram for a conventional base catalytic method to produce biodiesel.system should have great potential to be implemented on transportationvehicles since only minor retrofits of the diesel enginesare required to produce on site the necessary fuel.2. Selection of optimum TE conditionsExperiments with both continuous and batch reactors (Hastelloycoiled pipes and 316 stainless steel, respectively) have beenconducted at laboratory-scale to select ranges of P–T conditions,feedstock composition and residence time under which the highestyield of BD is obtained without requiring costly separation/purificationsteps (Anitescu et al., 2008). The conditions at which glyceroldecomposed in our reaction system without adverselyaffecting the FAME composition of BD fuel have been determinedas 350–400 °C and 100–300 bar. It is noted that the temperatureeffect on TG conversion dominates that of the pressure for SC TEand shortens the residence time to very attractive ranges for commercialapplications.

A. Deshpande et al. / Bioresource Technology 101 (2010) 1834–1843 1837The SC TE global reaction, used for the process design, is givenbelow for a generic composition of TG and lumped glycerol decompositionproducts (GDP):ð1ÞCH 2 -OOCR 1 R 1 COOCH 3| +CH-OOCR 2 + 3 CH 3 OH R 2 COOCH 3 + GDPCH 2-OOCR 3 R 3COOCH 3| +In this global reaction, R 1 , R 2 , and R 3 are the hydrocarbon chainsof the major SO fatty acids (e.g., palmitic, oleic, and linoleic) employedin the BD composition calculations. It is well establishedthat the steps of these TE reactions include TG conversion to diglycerides(DG), then to monoglycerides (MG), and finally to fatty acidmethyl esters (FAME) and glycerol (Diasakou et al., 1998). Our GC–MSD chromatograms (for example, Fig. 4a) revealed negligibleamounts of both DG and MG intermediate TE byproducts. Althoughthe glycerol decomposition products were not fully analyzed, theymainly included ethers and carbonylic compounds. It was estimatedthat these byproducts are smaller molecular organic compoundsthan the main FAME components and can be included inthe BD fuel as described elsewhere (Iijima et al., 2004; Aimarettiet al., 2009; Marulanda et al., 2009).Experiments carried out in our laboratory have shown that reactingstoichiometric quantities of alcohol and oil at 350–400 °C, 100–300 bar for about three minutes gives nearly complete oil conversionto FAME and GDP. In these SC states, in contrast to the BD productionat conventional conditions (1 atm, 60–80 °C, with basecatalysts), the reaction rate is about two orders of magnitude faster,BD components are formed using near stoichiometric quantities ofalcohol, and negligible alcohol and glycerol are left in the final reactionproducts. This situation would simplify the BD production processgreatly since these byproducts no longer have to be separatedfrom the FAME and additional fuel will be produced. Furthermore,the presence of small molecular components in BD such as C8–C14methyl esters, some linear alkanes and glycerol decompositionproducts positively affects the viscosity and the cloud/pour pointof the fuel. For example, at 5 °C, commercial samples of BD (compositiongiven in Fig. 4b), which are richer in saturated FAME, are solidswhile our samples only reach the cloud point. This property changecan also be attributed to the smaller molecular compounds whichexhibit lower melting points. In addition, because the alcohol canbe used only in slight excess of the TE stoichiometric amounts,essentially all of the alcohol reacts, eliminating the need for alcoholseparation and recycling at significant pumping costs.3. Economic analysis3.1. Process designThe aim of this economic assessment was to investigate the feasibilityof BD production coupled or not with power cogeneration(i.e., to determine the BD production and processing costs). Thescope of this study does not include the retail cost of BD and othermeasures of process profitability such as rate of return, pay backperiod and net present value. Economic estimates for two differentsystem configurations were performed to select the most desirablearchitecture based on maximum benefits. We targeted a system tocompetitively produce BD from vegetable oils (e.g., SO) and SCalcohols (e.g., methanol) compared with biofuels from conventionalTE processes and current petrodiesel.3.1.1. Plant capacity selectionAs per National Biodiesel Board reports, annual plant capacitiesof BD production in US are in the range of 50,000 to 90 10 6 gal(189,250–340.65 10 6 L), most of these being below 6 10 6 galor 22.71 10 6 L (National Biodiesel Board; Collins, 2006).Although some economic studies have been performed for smalland medium capacity BD plants employing conventional methods(Bender, 1999; Van Gerpen, 2007; Haas et al., 2006; Zhang et al.,2003), such studies are very scarce for plants employing SC methods(Kasteren and Nisworo, 2007; Lim et al., 2009). When an optimumplant capacity for BD production integrated with powercogeneration has to be selected, very high capacity and one reactormay not be feasible due to the oversized power generator andpumping hardware. A small/medium capacity plant appears to bethe choice for remote locations which are nearby to the source ofraw materials and without available power grid. A plant small enoughto be easily built but large enough to overcome the high laborcosts seems to be the best choice. However, larger capacity plantscan be easily assembled from smaller modules. Accordingly, twoplant capacities for BD production by the proposed SC TE methodwere chosen to be of 0.84 10 6 and 5.19 10 6 gal/year (3.18 10 6 and 19.64 10 6 L/year), respectively. The latter was roundedin the following text to 5 10 6 gal/year (18.925 10 6 L/year).These choices also illustrate the influence of scaling-up factors onBD costs.3.1.2. Process parametersTo execute a process design for a commercial facility for BD productionby the proposed SC method for the desired plant capacity,we needed to scale up the volumetric flow rates for SO and methanolemployed in the laboratory-scale experiments. Once the flowrates were set, a process flow diagram was assembled in CHEMCAD(Chemstations Inc., 2007). Such a process flow diagram of a SC TEsystem, including a source of energy, is shown in Fig. 2. Basically,the simple process design consists of pumps and preheaters totransport and preheat the reactants, respectively, and a reactorwith a heat source to produce FAME. The process is unique in thateither the exhaust gases from a diesel power generator or a Dowthermfluid heated by the combustion of a fraction of BD in a firedheater are used to provide heat for the TE reactions. Also, the reactantswere preheated by the hot TE product stream for the effectiveutilization of the energy.The thermodynamic properties of SO and FAME of the threemain fatty acid chains of SO (palmitic, linoleic, and oleic) were estimatedfrom group contribution methods using CHEMCAD. Vapor–liquid equilibrium compositions were estimated by CHEMCADusing different K models such as UNIFAC and Soave–Redlich–Kwong (SRK) equations of state (EOS). UNIFAC was used for thoseportions of the flow sheet where the reaction mixture is liquid andSRK EOS was used for those portions of the flow sheet where atleast one of the components of reaction mixture is above its criticalpoint. The results of the simulation runs were used for the designand sizing of the process equipment.3.1.3. Design assumptionsFor the process design of equipment the following assumptionswere made: (1) the reverse reactions are considered to be negligible(due to high T–P conditions and glycerol decomposition thereaction equilibrium is shifted to product side); (2) glyceroldecomposition products are soluble in methyl esters and wouldnot adversely affect the combustion of methyl esters; (3) disappearingconversion of SO was taken as 99.5%; (4) pressure dropacross any equipment was neglected; (5) diesel engine efficiency(BD energy to shaft power) was taken as 40%, and the efficiencyof the power generator (shaft power to electricity) was taken as95%; and (6) temperature and pressure of the reaction were selectedas 400 °C and 100 bar, respectively.Mass flow rates and compositions of the process input and outputstreams were quantified using component, mass, and energybalances, stoichiometric relations, and established process param-

1838 A. Deshpande et al. / Bioresource Technology 101 (2010) 1834–1843(a)10057.690807051.4Relative Abundance60504057.257.73058.7201058.8 60.3050 52 54 56 58 60 62 64 66 68 70 72 74Time (min)(b)10057.9908051.6Relative Abundance70605057.458.84030201063.662.9 65.371.4050 52 54 56 58 60 62 64 66 68 70 72 74Time (min)Fig. 4. Main TE products of: (a) soybean oil with methanol and (b) a commercial sample. Main peaks are the methyl esters of: palmitic (51.4/51.6); linoleic (57.2/57.4); oleic(57.6/57.9); linolenic (57.7); and stearic (58.7/58.8) acids. (Analytical GC–MSD method used: cross-linked methyl syloxane column of 30 m 0.25 mm I.D. 0.25 lm filmthickness with helium as carrier gas; the temperature program started at 60 °C hold for 2 min and continued with a subsequent ramp of 3 °C/min to 270 °C; the temperaturesof the injector and GC–MSD interface were 250 °C and 270 °C, respectively).eters. The process conditions and the flow rates of each stream inthe process flow diagram are given in Table 2.3.1.4. Process simulationCHEMCAD was used to design all of the process units except thesoybean oil preheater, SC TE reactor, diesel engine/generator set,and fired heater. While the overall results of process simulationfrom CHEMCAD seemed to be appropriate, the results for steadystate simulation for heat exchangers appeared to be less reliable.To exercise caution, hand calculations were also performed in thiscase and the best of the results were taken into consideration forfinalizing the design. The reactor was sized based on the residence

A. Deshpande et al. / Bioresource Technology 101 (2010) 1834–1843 1839Table 2Process stream conditions for biodiesel production coupled with power generation (the stream numbers are from the process flow diagram shown in Fig. 2).Stream no. T (°C) P (bar) Stream composition (g/min)Oil Alcohol FAME1 a FAME2 b FAME3 c GDP d Total1 25 1 0 558 0 0 0 0 5582 25 100 0 558 0 0 0 0 5583 370 100 0 558 0 0 0 0 5584 25 1 5049 0 0 0 0 0 50495 25 100 5049 0 0 0 0 0 50496 300 100 5049 0 0 0 0 0 50497 308 100 5049 558 0 0 0 0 56078 395 100 50.5 5.58 618 1523 2895 515 56079 395 100 3.1 0.35 38 94 180 32 34710 395 100 47.4 5.23 580 1429 2715 483 526011 370 100 47.4 5.23 580 1429 2715 483 526012 50 100 47.4 5.23 580 1429 2715 483 526013 50 1 47.4 5.23 580 1429 2715 483 526014 330 1 3.1 0.35 38 94 180 32 347abcdMethyl palmitate.Methyl oleate.Methyl linoleate.Glycerol decomposition products.time used in the laboratory experiments and was extended to3.79 min to account for scaling-up factors. Furthermore, an extravolume of 20% was considered to provide sufficient heat transferarea for heating the reactant mixture from 308 to 400 °C withthe exhaust gas. The diesel engine and fired heater were sized toprovide sufficient heat for the TE reactions by consuming part ofthe produced BD (4.53 and 6.05 vol.%, respectively) as fuel. Theelectricity from the engine is used locally or is to be sold to the localgrid for $0.10/kWh.3.2. Economic evaluation3.2.1. Cost of manufacturing (COM)BD production costs were calculated with the chemical engineeringmethods available in the literature (Peters and Timmerhaus,1991; Turton et al., 2003; Ulrich, 1984; Haas et al., 2006;Lim et al., 2009). The base-case design to be scaled up by a factorof approximately 6 for a BD plant with a capacity of 15,000 gal/day (56,775 L/day) was the TE of 2100 gal/day (7,949 L/day) of SOwith 269 gal/day (1,018 L/day) of methanol to produce 2424 gal/day (9,175 L/day) of BD. The plant was assumed to operate 24 h/day for 346 days/year. The capital cost and COM for BD productionof 2424 gal/day (9,175 L/day) with and without power generationestimated using the methods shown above are outlined in the followingparagraphs.First, COM was estimated from the equation:COM ¼ 0:2536 FCI þ 2:2835 C OL þ 1:0309 ðC UT þ C WT þ C RM Þð2ÞFCI is the fixed capital investment and C OL , C UT , C WT , and C RM are thecosts for operating labor, utilities, waste treatment, and raw materials,respectively. Although R&D costs are generally included inestimating COM, for those companies which are not interested inreplacing the existing technology for a certain period of time, theR&D costs can be excluded from COM. Accordingly, these costs werenot considered in the COM equation. Also, the retail cost of BD wasnot considered in this economic assessment, this cost structurebeing similar with the conventional BD cost and highly fluctuatingin time. Van Gerpen (2007) provided the economics for BD productionfor conventional base catalyzed method for 5 10 6 gal/year(18.925 10 6 L/year). In this analysis selling and distribution costswere not considered in estimating COM. To make a comparison betweenthe COM costs of BD for SC method and for a traditionalmethod, we also neglected the selling and distribution costs in estimatingCOM. Various other items in the COM (e.g., maintenance,supervisory and clerical labor, depreciation of the FCI, and administrativecosts) were estimated as fractions of the FCI and of operatinglabor, utilities, or raw material costs.3.2.2. Capital costThe costs of major pieces of equipment, their bare module costsin US dollars (2007) and the total module cost or FCI are given inTable 3. These costs were calculated based on CAPCOST softwareof Turton et al. (2003) and available information from manufacturersof the industrial equipment. Although the estimated prices forthese equipments do not affect the COM costs significantly, whencost information was not available, we always considered a conservativeapproach in estimating these costs. The storage tanks weresized to have 14 days storage capacity. The costs of the power generatorand fired heater have been evaluated from Ulrich (1984).The bare module cost for each piece of equipment includes considerationon high-pressure exposure and direct project expensessuch as the purchased cost of equipment, cost of piping, instrumentation,structural supports, etc., associated with the equipmentas well as the labor cost for installation and indirect costs such asfreight, insurance, engineering fees, and overhead. An additional18% was added for contingency and for the contractor’s fee. Carbonsteel was specified for equipment exposed to temperatures lessTable 3Capital cost of equipment for 2424 gal/day biodiesel production.Equipment Cost in US $ (2007)With power generation With fired heaterMethanol storage tank 46,060 46,060Soybean oil storage tank 63,000 63,000Biofuel storage tank 63,600 63,600Methanol feed pump 33,609 33,609Soybean oil feed pump 64,508 64,508Reactor 78,300 78,300Soybean oil preheater 55,600 55,600Methanol preheater 31,800 31,800Pressure regulating valves 1,000 1,000Power generator/fired heater 10,000 66,000Dowtherm fluid 2000Total equipment cost 447,477 505,477Contingency and fees 80,546 90,986Total F.C.I. 528,023 596,463

1840 A. Deshpande et al. / Bioresource Technology 101 (2010) 1834–1843Table 4Utility costs (C UL ) for biodiesel production.Equipment Utility Power (W) Utility usage Cost ($)Power generator Fired heater Power generator Fired heaterReactor EG/DF a 25,427 5.373 kg/min 8.94 kg/min 0 2000 bMethanol feed pump Electricity 246 2043 kWh/year 2043 kWh/year 0 204Soybean oil feed pump Electricity 1926 15994 kWh/year 15994 kWh/year 0 1599Methanol preheater Biofuel 9965 5.62 kg/min 5.62 kg/min 0 0Soybean oil preheater Biofuel 66,429 5.62 kg/min 5.62 kg/min 0 0Fired heater Biofuel 25,427 0.434 kg/min 0abTotal (C UL ) 0 1813EG stands for exhaust gases and DF for Dowtherm fluid.DF cost is not considered as utility cost but as FCI as it is recycled and can be used in the process for a period of 4–5 years.than 250 °C and stainless steel was specified for equipment exposedto higher temperatures.The capital cost for 5 10 6 gal/year (18.925 10 6 L/year) of BDproduction was estimated by scaling up the capital costs for thebase-case unit using six-tenths rule. Peters and Timmerhaus(1991) suggested scaling up plant capacities up to 10 times thebase-capacity. According with this rule, the BD plant capacity couldhave been scaled up to a capacity of 8.3 10 6 gal/year(31.4 10 6 L/year), but for easier economic comparison betweenconventional and SC methods and other considerations, we choseto analyze the SC BD process economics for 5 10 6 gal/year(18.925 10 6 L/year).3.2.3. Utility costThe components for the cost of utilities for BD production weretaken from Turton et al. (2003) and are shown in Table 4. Due toenergy cogeneration in the process, the utilities cost for the processesof the proposed SC TE technology are very low. More efficientuse of high energy exhaust gas stream with other processstreams can further reduce the energy requirement. The utilitiescosts were scaled linearly when the plant capacity was increasedfrom the base-case to 5 10 6 gal/year (18.925 10 6 L/year).3.2.4. Labor costThe labor cost was calculated considering the number of operatorsper shift (N OL ) for BD production and the adjusting factorfor 24/7 operations (Tables 5 and 6). The plant was assumed torun continuously with 3 shifts/day for 346 days/year (for theTable 5Operators per shift (N OL ).Equipment typeNumber ofequipmentOperators per shift perequipmentOil preheater 1 0.1 0.1Methanol1 0.1 0.1preheaterReactor 1 0.5 0.5Power generator/fired heater1 0.5 0.5Total 1.2Operators per shift were rounded to nearest whole number, N OL =2.Operatorsper shiftremaining days the plant was considered to be in shut down andmaintenance mode). The annual operator salary in the rural areawas considered to be $30,000 ($15/h). Hence, the operating laborcosts (2007) were evaluated to be $270,000/year. Since the processis continuous and fully automated it requires less supervision andthe labor costs were considered to be the same when the plantcapacity was increased to 5 10 6 gal/year (18.925 10 6 L/year).3.2.5. Raw material costsThe raw material costs are for SO and methanol. The oil at thetime of this analysis was available for $2.91/gal or $0.77/L (CMEGroup, 2007) and methanol at $0.93/gal or $0.25/L (Methanex,2007) while the price of petrodiesel fuel was approximately$3.00/gal or $0.79/L (Energy Information Agency, 2007). The totalraw material costs as of August 2007 for 2424 gal/day (9,175 L/day) of BD production for both options are shown in Table 7. Thewaste treatment costs were negligible for this economic analysissince no polluting streams other than the exhaust gases of BD,otherwise from a cleaner combustion, have to be disposed.3.2.6. Total COMThe total COM is given in Table 8 along with revenues from thesale of electricity, and the break-even cost of the net BD availablefor sale. The total COM per year includes the annual freight chargesand the annual interest on the FCI. For the base-case design of2424 gal/day (9,175 L/day) of BD production with power generation,COM was estimated to be approximately $3,333,000/year.The total electricity produced by the power generator is approximately526 MWh/year, but 18 MWh/year are used to drive themethanol and oil feed pumps, leaving approximately 508 MWh/year available for sale. Assuming that the selling price of electricityto commercial or industrial sector is $0.10/kWh, the revenues obtainedfrom electricity would be $51,000. Hence, the net COM wasestimated to be $3,282,000/year. The total amount of BD fuel producedis 839,000 gal/year (3.2 10 6 L/year). From this amount,4.53 vol.% is used for generation of electricity by the diesel-generatorand the remaining 95.47% of BD fuel available for sale is800,000 gal/year (3 10 6 L/year). The break-even cost of BD fuelis ($3,282,000/year)/800,000 gal/year or $4.10/gal ($1.08/L) witha processing cost of $1.02/gal ($0.27/L).Table 6Operating labor cost (C OL ).Operators/shift (N OL )Factor for 24/7 weekoperationAnnual operatorsalary (AOS) ($)Total cost($/year)2 4.236 30,000 270,000C OL = N OL adjusting factor AOS (the N OL adjusting factor = 8.47 9).Table 7Raw materials cost (C RM ).Raw material Requirement (gal/year) Cost (2007 US $/year)Methanol 93,000 86,490Soybean oil 770,000 2,240,700Freight charges 54,830Total C RM 2,382,020

A. Deshpande et al. / Bioresource Technology 101 (2010) 1834–1843 1841Table 8Comparison of biodiesel (BD) cost of manufacturing (COM) for two process options (4.9 and 6.05 vol.% of total BD were consumed in options A and B, respectively).Option BD capacity (gal/day) COM ($/year) Electricity revenues ($/year) Net cost ($/gal) Processing cost ($/gal)A 2424 3,322,708 50,827 4.09 1.0115,000 16,817,463 307,238 3.33 0.26B 2424 3,342,872 4.26 1.1315,000 16,919,737 3.47 0.34Option A: BD production with power generation.Option B: BD production with fired heater.Table 9Comparative production and processing costs (in US $/gal) of biodiesel fuels for conventional and SC methods.Cost components Van Gerpen (2007) This study a This study b This study a This study b5 10 6 gal/year 0.84 10 6 gal/yearOil 3.23 c 2.95 e 3.00 e 2.95 e 3.01 eMethanol 0.13 c 0.12 0.12 0.12 0.12NaOCH 3 catalyst 0.08Neutralizer 0.01Utilities 0.02 0.00 0.002 0.00 0.002Labor 0.10 c 0.066 0.065 0.40 0.41Depreciation 0.15 0.033 0.038 0.068 0.078Interest 0.025 0.029 0.052 0.060Administration + overhead 0.020 0.064 0.066 0.327 0.338Maintenance 0.040 0.020 0.023 0.041 0.047Laboratory charges 0.011 d 0.008 0.008 0.051 0.052Operating supplies 0.011 d 0.003 0.003 0.006 0.007Local taxes + insurance 0.040 d 0.010 0.012 0.022 0.025Patents + royalties 0.12 d 0.10 0.10 0.12 0.12Total production cost 3.96 3.41 3.47 4.16 4.28Electricity revenue 0.00 0.064 0.00 0.06 0.00Net production cost 3.96 3.34 3.47 4.10 4.28Processing cost 0.51 0.26 0.35 1.02 1.14abProcess with power generator.Process with fired heater.c Updated costs to August, 2007.d Additional costs.e For the current study the oil cost is less due to the fact that less oil is used to produce the same amount of fuel with the glycerol decomposition products included.Most of the total COM comes from the raw materials and laborcosts (Table 9). Although little can be done to reduce the raw materialscost, the COM per unit volume of product would be significantlyreduced for a larger size plant since the number ofoperators would not change (2 operators per shift were alreadysignificantly rounded up from a required number of 1.2). Also somesavings in capital cost would be realized with a larger capacityplant. Thus, the COM for a plant capacity of 15,000 gal/day(56,775 L/day) was calculated to a reduced break-even cost of$3.34/gal or $0.88/L (about the same with the current diesel fuelcost) with a processing cost of $0.26/gal ($0.07/L).A second alternative design that was considered was the use ofa direct-fired Dowtherm heater fueled by a fraction of the producedBD to heat the esterification reactor instead of exhaust fromthe diesel-generator. For this option, BD needed for heating thereactants was estimated to be 6.05 % of the total BD. For a processproducing 15,000 gal/day (56,775 L/day) of BD, the break-even costis reduced from $4.28/gal ($1.13/L) for 2424 gal/day (9,175 L/day)to $3.47/gal ($0.92/L). Similarly, the processing costs are reducedfrom $1.14/gal ($0.30/L) to $0.35/gal ($0.09/L).3.2.7. Comparison with conventional TE processesIn order to compare these results with those for a conventionalmethod, we selected the reported costs of Van Gerpen (2007). Similarcosts have been reported to a US Senate Committee by Collins(2006). A few cost items that we included in the production costshave not been considered in the analyses of Turton et al. (2003)(e.g., the interest on the borrowed amount for the capital investment)and Van Gerpen (2007) (e.g., laboratory charges, operatingsupplies, local taxes and insurance, patents and royalties). However,for the reasons cited earlier, we did not include in COM thecosts of R&D, selling, and distribution.A comparison of the production and processing costs of BD fuelsby conventional and SC methods for 5 10 6 gal/year (18.925 10 6 L/year) capacity as well as the costs for the base-capacity of0.84 10 6 gal/year (3.18 10 6 L/year) is provided in Table 9. Forthe latter capacity, these costs are reported for only the SC method.Oil and methanol costs (August 2007) were updated for each columnand freight charges were added to these costs. Various costitems reported in this table were calculated by using Turtonet al. (2003). Interest on capital investment was additionally consideredat the rate of 6% compounded annually for 10 years. Inthe cited reference (Van Gerpen, 2007), the depreciation/interest,labor, maintenance, administration and overhead costs seemed tobe lower for the total capital investment of $5.26 10 6 . Also, thelabor cost was reported there for only one shift and we updatedthis cost for 3 shifts per day. Furthermore, maintenance costs reportedby Kasteren and Nisworo (2007) and in this study were calculatedat 3.8 and 6% of FCI, respectively. However, the complexityand the number of the equipment units for the two methods asshown in Figs. 2 and 3 should render the conventional plant maintenancerather more costly. Laboratory charges, operating supplies,local taxes and insurance, patents and royalties costs (Turton et al.,2003) were additionally considered in the first column of Table 9

1842 A. Deshpande et al. / Bioresource Technology 101 (2010) 1834–1843for comparison purposes on a common base. Net production costwas calculated by subtracting byproduct revenues from the totalproduction cost and the processing cost was calculated by subtractingraw materials costs from the net production cost. The oiland methanol costs are higher in the first column of Table 9 ascompared to those in the columns 2 and 3 since less biodiesel isproduced by conventional method (glycerol is separated out as abyproduct) from the same quantity of raw materials.Furthermore, we can compare our production and processingcost with those reported in the only economic analysis for anotherSC method available in literature (Kasteren and Nisworo, 2007). Atcomparable capacity levels (e.g., 2.5 10 6 gal/year or 9.46 10 6 L/year), our calculated production cost of $3.50/gal ($0.92/L) is higherthan that reported, $2.30/gal ($0.61/L). The main reason for thisdifference has to be considered the lower cost of waste cooking oil.However, our processing cost of $0.43/gal ($0.11/L) is much lowerthan that in the above reference of $1.43/gal ($0.38/L) due to architecturaldifferences of the two technologies and the associated processconditions.Raw materials cost, depreciation, interest, local taxes and insurance,utilities, administration and overhead and maintenance costsfor the process without power generation were found to be higherthan those for the process with power generation (column 3 versuscolumn 2 in Table 9). From the results obtained we can concludethat BD production with power generation seems to be moreadvantageous than that without power generation for the plantcapacities for which this economic analysis was performed. Also,processing cost for BD produced by the proposed SC method isapproximately half of that reported for the conventional methods.The costs for SO (CME Group, 2007) and methanol (Methanex,2007) as well as the diesel fuel cost (Department of Energy,2007) exhibit timely market fluctuations. However, the processingcost is much more stable such that it can be stated that the BDprice is the price of SO and methanol plus the processing cost of$0.26/gal or $0.07/L (for 5 10 6 gal/year or 18.925 10 6 L/yearplant capacity). It is also important to note that since the diesel fuelhas to be used for soybeans cultivation and SO production and thatSO cost produced this way is near the cost of diesel fuel, the BDprice will be always above the price of diesel fuel. Otherwise, thediesel fuel has to be replaced by BD all along these processes.4. SummarySelected ranges of the laboratory-scale TE conditions (e.g.,350–400 °C, 100–300 bar, and alcohol to oil ratios at about stoichiometricvalue of 3:1) providing near complete conversions ofoil to BD products have been employed to design an industrialSC TE technology associated with power cogeneration. The proposedtechnology consists in one-step BD production withoutcomplex separation/purification processes and basically all ofthe glycerol produced will decompose to smaller molecular compoundsblended within BD fuel. SC TE products generated underthe above conditions consist in one homogeneous liquid phasewith viscosity, cloud/pour point and composition superior tothose of current commercial BD fuel. By using close to stoichiometricoil–alcohol mixtures, the power required to recycle large excessof alcohol is eliminated with significant technological andeconomic benefits.An economic analysis of the proposed SC TE process has beenperformed. It was found that the BD processing cost through theproposed technology could be half of that of the actual conventionalmethods. (i.e., $0.26/gal or $0.07/L versus $0.51/gal or$0.13/L (Collins, 2006; Van Gerpen, 2007)). Associated with someother incentives along with higher diesel fuel prices, this newcost-saving technology is likely to make BD production profitable.AcknowledgementsThis study has been funded by NYSERDA through the SponsorAward # 9462/2006 (Judy Jarnefeld, Senior Project Manager) andco-sponsored by Syracuse Center of Excellence in Environmentaland Energy Systems (Edward Bogucz, Director).ReferencesAimaretti, N., Manuale, D.L., Mazzieri, V.M., Vera, C.R., Yori, J.C., 2009. Batch study ofglycerol decomposition in one-stage supercritical production of biodiesel.Energy and Fuels 23, 1076–1080.Anitescu, G., Deshpande, A., Tavlarides, L.L., 2008. Integrated technology forsupercritical biodiesel production and power cogeneration. Energy and Fuels22, 1391–1399.Bender, M., 1999. Economic feasibility review for community scale farmercooperatives for biodiesel. Bioresource Technology 70, 81–87.Biodiesel, 2008. Food versus Fuel: FAQ. (accessed 17.09.09).Bunyakiat, K., Makmee, S., Sawangkeaw, R., Ngamprasertsith, S., 2006. Continuousproduction of biodiesel via transesterification from vegetable oils insupercritical methanol. Energy and Fuels 20, 812–817.Cao, W., Han, H., Zhang, J., 2005. Preparation of biodiesel from soybean oil usingsupercritical methanol and co-solvent. Fuel 40, 347–351.CHEMCAD version 5.6.4, 2007. Chemstations Inc., Houston, TX.CME Group, 2007. http://www.cmegroup.com/trading/commodities/grain-andoilseed/soybean-oil_quotes_openOutcry.html(accessed 16.09.09).Collins, K., 2006. Statement of Keith Collins, Chief Economist, US Department ofAgriculture before the US Senate Committee on Appropriations Subcommitteeon Agriculture, Rural Development, and Related Agencies.Diasakou, M., Louloudi, A., Papayannakos, N., 1998. Kinetics of the non-catalytictransesterification of soybean oil. Fuel 77, 1297–1302.D’Ippolito, S.A., Yori, J.C., Iturria, M.E., Pieck, C.L., Vera, C.R., 2007. Analysis of a twostep,noncatalytic, supercritical biodiesel production process with heatrecovery. Energy and Fuels 21, 339–346.Energy Information Administration, 2007. (accessed 16.09.09).Haas, M.J., McAloon, A.J., Yee, W.C., Foglia, T.A., 2006. A process model to estimatebiodiesel production costs. Bioresource Technology 97, 671–678.Han, H., Cao, W., Zhang, J., 2005. Preparation of biodiesel from soybean oil usingsupercritical methanol and CO 2 as a co-solvent. Process Biochemistry 40, 3148–3151.He, H., Wang, T., Zhu, S., 2007. Continuous production of biodiesel fuel fromvegetable oil using supercritical methanol process. Fuel 86, 442–447.Hegel, P., Mabe, G., Pereda, S., Brignole, E.A., 2007. Phase transitions in a biodieselreactor using supercritical methanol. Industrial and Engineering ChemistryResearch 46, 6360–6365.Hill, J., Nelson, E., Tilman, D., Polasky, S., Tiffany, D., 2006. Environmental, economic,and energetic costs and benefits of biodiesel and ethanol biofuels. Proceedingsof the National Academy of Sciences of the United States of America 103,11206–11210.Iijima, W., Kobayashi,Y., Takekura, K., Taniwaki, K., 2004. The Non-Glycerol Processof Biodiesel Fuel Treated in Supercritical Methanol. ASAE/CSAE AnnualInternational Meeting, Ottawa, Ontario, Canada, Paper No. 046073.Kasteren, J.M.N.V., Nisworo, A.P., 2007. A process model to estimate the costof industrial scale biodiesel production from waste cooking oil bysupercritical transesterification. Resources, Conservation and Recycling 50,442–458.Kusdiana, D., Saka, S., 2001. Kinetics of transesterification in rapeseed oil tobiodiesel fuel as treated in supercritical methanol. Fuel 80, 693–698.Kusdiana, D., Saka, S., 2004. Effects of water on biodiesel fuel production,by supercritical methanol treatment. Bioresource Technology 91, 289–295.Lim, Y., Lee, H.-S., Lee, Y.-W., Han, C., 2009. Design and economic analysis of theprocess for biodiesel fuel production from transesterificated rapeseed oil usingsupercritical methanol. Industrial and Engineering Chemistry Research 48 (11),5370–5378.Madras, G., Kolluru, C., Kumar, R., 2004. Synthesis of biodiesel in supercritical fluids.Fuel 83, 2029–2033.Marulanda, V.F., Anitescu, G., Tavlarides, L.L., 2009. Biodiesel fuels through acontinuous flow process of chicken fat supercritical transesterification. Energyand Fuels, in press.Methanex, 2007. (accessed16.09.09).National Biodiesel Board, XXXX. (accessed 16.09.09).Peters, M.S., Timmerhaus, K.D., 1991. Plant Design and Economics for ChemicalEngineers, fourth ed. McGraw Hill Chemical Engineering Series, New York.Radich, A., 2004. Biodiesel Performance, Costs, and Use. Energy InformationAdministration; (accessed 17.09.09).Saka, S., Kusdiana, D., 2001. Biodiesel fuel from rapeseed oil as prepared insupercritical methanol. Fuel 80, 225–231.

A. Deshpande et al. / Bioresource Technology 101 (2010) 1834–1843 1843Turton, R., Bailie, R.C., Whiting, W.B., Shaeiwitz, J.A., 2003. Analysis, Synthesis, andDesign of Chemical Processes, second ed. Prentice Hall International Series, NewJersey.Ulrich, G.D., 1984. A Guide to Chemical Engineering Process Design and Economics,first ed. John Wiley and Sons, New York.USA Biodiesel Prices.com, 2009. Biodiesel Stations and Prices for the EntireUS. (accessed 17.09.09).Van Gerpen, J.H., 2007. Biodiesel Economics in Oilseeds and Biodiesel Workshop,Great Falls, MT. Available from: (accessed 17.09.09).Varma, M.N., Madras, G., 2007. Synthesis of biodiesel from castor oil and linseedoil in supercritical fluids. Industrial and Engineering Chemistry Research 46,1–6.Zhang, Y., Dube, M.A., Mclean, D.D., Kates, M., 2003. Biodiesel production fromwaste cooking oil: 2. Economic assessment and sensitivity analysis. BioresourceTechnology 90, 229–240.