analysis of a pilot-scale anaerobic baffled reactor treating domestic ...

More documents

Recommendations

Info

The original objective to develop a dynamic mathematical model of biochemical processes for the purposes of designing ABR systems was over-optimistic, given the quality of the experimental data available. These data were mined for design information about process limitations (upflow velocity in this case) and for CH4 and sludge production estimates. More sophisticated design tools can only be developed with the assistance of additional experimental data describing the relationships between observed sludge accumulation or retention and applied upflow velocity, alkalinity and organic loading rate. Nevertheless, a model was developed using the WEST 1 platform to identify what further information would be required to build a biochemical model for design purposes. 6.5.1 Model construction The model was constructed as a series of 8 continuous stirred tank reactors with a solids retention factor (Figure 6.8) i.e. a small fixed fraction of the concentration of particulate species was allowed to leave each compartment. Figure 6.8: WEST ® representation of the ABR flow configuration. Each element represents a continuous stirred tank reactor with sludge retention. The biochemical model of Siegrist et al. (1993) was used to describe anaerobic conversions in the modelled ABR. Figure 6.9 shows the flow diagram for carbon catabolism in the Siegrist model. The Siegrist model has a number of limitations: • Only one category of influent particulate COD is described, represented as Biopolymers in Figure 6.9 Hydrolysis of this component yields amino acids, sugars and fatty acids in a fixed ratio. The stoichiometry of this process is fixed in the Siegrist model, although it is possible to manually alter the stoichiometry. There is no mechanism for allowing the ratio between the different hydrolysis products to vary during a simulation. • The model was constructed and calibrated for mesophilic sewage sludge digestion, and therefore default parameter values for kinetic constants and stoichiometry may not be appropriate for domestic wastewater • Protolysis and deprotolysis of volatile fatty acids are not included in the model; therefore it is not possible to simulate extreme acidification (pH < 6) of a digester (Siegrist et al., 1993). 1 WEST: Worldwide Engine for simulation, Training and Automation. This software is built and supported by Hemmis. 154
Key: COD-bearing component Micro-organism population Figure 6.9: Flow diagram showing route taken during catabolic degradation of COD in the anaerobic digestion model proposed by Siegrist et al. (1993). Details of implementation and results of the model were presented in Foxon et al. (2004) and Foxon and Buckley (2007). Copies of these conference papers are included in Annexure 1 on the CD enclosed in the back cover of this thesis. 6.5.2 Results of biochemical modelling Acetate The modelling exercise was extremely useful in that the process of building and manipulating the model identified concepts that proved important in analysing and understanding the experimental data. These included the following: • Except during start-up and when the reactor is already full of solids, a pseudo-steady state exists where the ABR effluent does not change substantially with time, despite ongoing accumulation of solid material • During stable digestion, the bulk of accumulating material is inert particulate material • It is possible to continue accumulating solid material, while washing out active microorganisms 6.5.2.1 Simulation results In both the 2004 and 2007 studies, the biochemical models were able to describe ammonia and alkalinity generation, when appropriate feed characteristics were specified. However, the simulations were not able to predict pH and solids dynamics in the compartments of the ABR with any reliability: • They did not accurately depict the dynamics of soluble COD components, and components that have an effect on the pH value in each compartment (Foxon and Buckley, 2007) 155 Biopolymers Amino acids and sugars Fatty acids Acidogens Propionate Acetoclastic methanogens H 2 + CO 2 Propionate oxidising acetogens CO 2 + CH 4 Fatty acids oxidising acetogens Hydrogenotrophic methanogens CH 4 + H 2 O
Page 1:
ANALYSIS OF A PILOT-SCALE ANAEROBIC
Page 5:
i DECLARATIONS I, Katherine Maria F
Page 8 and 9:
My deepest thanks must be extended
Page 11 and 12:
vii TABLE OF CONTENTS ANALYSIS OF A
Page 13 and 14:
5.3 Feed characteristics and loadin
Page 15 and 16:
xi LIST OF TABLES Table 1.1: List o
Page 17 and 18:
xiii LIST OF FIGURES Figure 1.1: Gr
Page 19 and 20:
Figure 5.1 Installation of the ABR
Page 21 and 22:
Figure 6.8: WEST® representation o
Page 23 and 24:
xix LIST OF ABBREVIATIONS A-HRT App
Page 25 and 26:
1 1 INTRODUCTION The work presented
Page 27 and 28:
obtained from anaerobic systems rel
Page 29 and 30:
• Lalbahadur (MTech, 2005) perfor
Page 31 and 32:
1.7 ORGANISATION OF THE THESIS This
Page 33 and 34:
9 2 LITERATURE REVIEW A detailed re
Page 35 and 36:
• Disintegration of composite par
Page 37 and 38:
Homoacetogens are one of the most v
Page 39 and 40:
Most of the control in anaerobic di
Page 41 and 42:
therefore the sensitivity of the ov
Page 43 and 44:
∆Gº (W) [kcal] -10 -8 -6 -4 -2 L
Page 45 and 46:
For systems with low potential for
Page 47 and 48:
design of the digester in which the
Page 49 and 50:
2.2.2 Expanded granular sludge bed
Page 51 and 52:
fact that for most non-ideal flow s
Page 53 and 54:
hydrolysis steps to convert them to
Page 55 and 56:
from the clarified zone between the
Page 57 and 58:
Table 2.4: Typical pathogen surviva
Page 59:
Table 2.5: Studies using anaerobic
Page 62 and 63:
• There is no requirement for bio
Page 64 and 65:
Kennedy and Barriault (2005) perfor
Page 66 and 67:
methanogenic (Akunna and Clark, 200
Page 68 and 69:
and 6 h with no significant differe
Page 70 and 71:
2.5.1.12 Granule formation in ABRs
Page 72 and 73:
• A baffled reactor is described
Page 74 and 75:
2.5.3.4 Scanning electron microscop
Page 76 and 77:
spacing on flow patterns in a singl
Page 78 and 79:
Figure 3.5: Orthographic projection
Page 80 and 81:
PLC calculated the fraction of that
Page 82 and 83:
epresent buulk conditionns. Compart
Page 85 and 86:
The rreactor was initially commmiss
Page 87 and 88:
Phase I Phase II Phase III Phase IV
Page 89 and 90:
Table 4.1: Characteristics of degri
Page 91 and 92:
Table 4.2: Pilot-scale ABR approxim
Page 93 and 94:
COD [mgCOD/ℓ] 2000 1800 1600 1400
Page 95 and 96:
4.3.5 Phosphorus Figure 4.8 present
Page 97 and 98:
Observation of sludge in compartmen
Page 99 and 100:
Another interesting observation is
Page 101 and 102:
4.4.3 pH pH measurements were perfo
Page 103 and 104:
esult in low pH values. The reasons
Page 105 and 106:
4.5 SUMMARY: PHASE I - OPERATION AT
Page 107 and 108:
than the optimal range for methanog
Page 109 and 110:
5 EXPERIMENTAL PHASES II-IV: KINGSB
Page 111 and 112:
sludge on a number of occasions. Wh
Page 113 and 114:
On days 99 and 100 of Phase III, a
Page 115 and 116:
5.3.2 Analysis of variations in fee
Page 117 and 118:
Table 5.2: Pilot-scale ABR average
Page 119 and 120:
5.4.1.2 Phase III In Phase III, (Fi
Page 121 and 122:
Phase II: Alkalinity [mgCaCO3/ℓ]
Page 123 and 124:
Eq. 5-1 However, at a COD/SO4 2- ra
Page 125 and 126:
wastewater was probably higher than
Page 127 and 128: 5.5.2.2 Damping Figure 5.12 and Fig
Page 129 and 130: • The outflow COD values were not
Page 131 and 132: • Information about the volume of
Page 133 and 134: As with the total volume of sludge
Page 135 and 136: It is not clear why the rate of acc
Page 137 and 138: most compartments was exactly 6.5,
Page 139 and 140: Soluble COD [mg/ℓ] 600 500 400 30
Page 141 and 142: pathogens and parasites, which may
Page 143 and 144: % Probe of DAPI % Probe of DAPI % P
Page 145 and 146: later compartments at all. Furtherm
Page 147 and 148: and 4, with decreasing observations
Page 149 and 150: • The mechanism of sludge build-u
Page 151: Table 5.6: Inflow and outflow chara
Page 154 and 155: Behling et al. (1997) studied the p
Page 156 and 157: organisms in previous compartments.
Page 158 and 159: velocities. Unfortunately, these tw
Page 160 and 161: Thus for a fixed sludge load, at hi
Page 162 and 163: sampling was not possible since in
Page 164 and 165: Wentzel (2006) recommends a COD to
Page 166 and 167: 6.2.1.4 Calculation of CH4 producti
Page 168 and 169: organically bound N) or that N accu
Page 170 and 171: All VFA is consumed in the process,
Page 172 and 173: However, these assumptions are clea
Page 174 and 175: The calculated pH value is compared
Page 176 and 177: influent alkalinity will increase d
Page 180 and 181: • It was not possible to simulate
Page 182 and 183: employed for a medium strength wast
Page 184 and 185: 6.6.4 Alternative baffle design Thi
Page 186 and 187: 6.7.3 Monitoring requirements The m
Page 188 and 189: compartments (Section 6.1.1) pH val
Page 190 and 191: generally about anaerobic digestion
Page 192 and 193: 7.1.1.6 Effect of sludge age on bio
Page 194 and 195: 7.1.2.4 Critical design parameters
Page 196 and 197: 172
Page 198 and 199: Batstone D. J., Keller J., Angelida
Page 200 and 201: Gopala K. G. V. T. (2007). Treatmen
Page 202 and 203: MacLeod F. A., Guiot S. R. and Cost
Page 204 and 205: Sasse L. (1998). Decentralised wast
Page 206 and 207: Wanasen (2003). Upgrading conventio
Page 209 and 210: METTHODS OFF SAMPLINNG AND ANNALYSI
Page 211 and 212: 3.6 Enumeration of total coliforms
Page 213 and 214: that the means are the same, or con
Page 215 and 216: And the significance of the regress
Page 217 and 218: ADDITIONAL RESULTS 193 APPENDIX A3
Page 219 and 220: Table A3. 1 cont. Phase I Phase II
Page 221 and 222: each of the eight compartments were
Page 223 and 224: 3.2 Pathogen indicator organisms A
Page 225 and 226: APPLIED VS. APPARENT HYDRAULIC RETE
Page 227 and 228: 19 h apparent HRT 14 h apparent HRT
Page 229 and 230:
205 APPENDIX A5 CALCULATION OF SOLI
Page 231 and 232:
i.e. the average SRT of an accumula
Page 233:
X X = P Inert ⋅ where Xinert is t
Page 236 and 237:
iodegradable organic material may b
Page 238 and 239:
Substituting Eq. A6- 1 in Eq. A6- 7
Page 240 and 241:
1.5 Implementation of steady-state
Page 242 and 243:
X = 7⋅ Y = 7 Z = 7⋅ A = 7⋅ (
Page 244 and 245:
Dama, P., Bell J., Naidoo, V., Foxo
Page 246:
Lalbahadur T. (2004) Characterisati
show all

analysis of a pilot-scale anaerobic baffled reactor treating domestic ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?