Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

External membrane bioreactor for the anaerobic treatment of low strength municipal wastewater Di, Donghong 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-0078a.pdf [ 10.38MB ]
Metadata
JSON: 831-1.0063245.json
JSON-LD: 831-1.0063245-ld.json
RDF/XML (Pretty): 831-1.0063245-rdf.xml
RDF/JSON: 831-1.0063245-rdf.json
Turtle: 831-1.0063245-turtle.txt
N-Triples: 831-1.0063245-rdf-ntriples.txt
Original Record: 831-1.0063245-source.json
Full Text
831-1.0063245-fulltext.txt
Citation
831-1.0063245.ris

Full Text

EXTERNAL MEMBRANE BIOREACTOR FOR THE ANAEROBIC TREATMENT OF LOW STRENGTH MUNICIPAL WASTEWATER by DONGHONG DI B.Sc. Tianjin University, China, 1990 M.Sc. Tianjin University, China, 1993 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Civil Engineering) THE UNIVERSITY OF BITISH COLUMBIA April 2007 ©Donghong Di, 2007 A B S T R A C T In this study, the workability of a newly designed anaerobic external membrane bioreactor (AnMBR) was examined. During the acclimatization period, the system was operated at a temperature of 25 to 42 °C and a pH of 6.8 to 7.5. The influent COD was 510 (±13) mg/L. After the acclimatization, the system was maintained at 25 °C and a pH of 6.3 to 7.0. The influent COD was 290 (±18) mg/L. The hydraulic retention time (HRT) was decreased in a stepwise manner from an initial value of 61 hours to 2 hours. As the HRT was decreased, the corresponding volumetric organic loading rate (OLR) was increased from 0.22 ± (0.04) kg COD/m3day to 2.84 ± (0.11) kg COD/m3-day. Results showed that the anaerobic external MBR was capable of treating low strength municipal wastewater under controlled conditions. The COD removal ef ficiencies were 84 (±4) % and 60 (±14) % for the acclimatization period and the steady-state operation period, respectively. The decrease in COD removal efficiency may have been caused by short retention time of organic material in the reactor due to the short HRTs, the inhibition of biomass activity by the high OLR, a decrease of the soluble COD portion of the influent, and the accumulation of VFAs in the reactor. An HRT of 8 hours and OLR of 1.6 kg COD/mJ«day was identified as the optimum operational cond itions for the system. The permeability of the membrane limited the system treatment capacity. Inorganic membrane permeability was superior to the organic membrane permeability. When operated at trans-membrane pressure (TMP) of 207 kPa and cross flow velo city (CFV) of 3.3 m/s, the organic membrane exhibited high and stable permeate flux. Membrane permeability decreased with an increase of sludge concentration at TMP of 276 kPa, but was independent of sludge concentration at TMP of 83 kPa. Organic membrane cleaning with sodium hypochlorite resulted in an effective flux recovery. TABLE OF CONTENTS A B S T R A C T i i T A B L E OF C O N T E N T S -. i i i LIST O F T A B L E S v LIST O F F I G U R E S vi A C K N O W L E D G E M E N T S v i i i C H A P T E R 1 I N T R O D U C T I O N 1 C H A P T E R 2 L I T E R A T U R E R E V I E W 4 2.1 Anaerobic Process 4 2.1.1 Critical parameters affecting process monitoring and control 6 2.2 Anaerobic Process for the Treatment of Municipal, Low -strength Wastewater at L o w Temperatures 12 2.3 Anaerobic Membrane Bioreactor ( A n M B R ) Treatment System 15 2.3.1 Anaerobic M B R experience 15 2.3.2 Anaerobic M B R treatment of municipal wastewater 19 2.4 Mechanisms Governing the Efficiency of the Membrane Component of the External Anaerobic M B R 21 2.4.1 Mechanisms governed by the membrane system 22 2.4.2 Mechanisms governed by the operational parameters 26 2.4.3 Mechanisms governed by the characteristic of the mixed liquid 31 2.5 Summary and Objectives of the Research 33 2.5.1 Parameters critical to the system operation control 34 2.5.2 Parameters critical for the system performance evaluation 35 2.5.3 Objectives 35 C H A P T E R 3 M E T H O D S A N D M A T E R I A L S 37 3.1 System Design 37 3.2 Experimental Program 44 3.2.1 Monitoring variables 45 iii 3.2.2 Experimental variables 46 3.3 Bioreactor Inoculation and Acclimatization 50 3.4 Membrane Cleaning 52 CHAPTER 4 RESULTS AND DISCUSSIONS 54 4.1 General System Operation Conditions and Control 54 4.1.1 Hydraulic retention time and volumetric organic loading rate 54 4.1.2 Temperature 55 4.1.3 pH 57 4.2 Biological Performance of Anaerobic MBR 59 4.2.1 Chemical oxygen demand (COD) 59 4.2.2 Volatile fatty acids (VFAs) 66 4.2.3 Gas production and composition 71 4.2.4 Biomass concentrations and partic le size distributions 78 4.3 Membrane Component Performance of Anaerobic MBR 84 4.3.1 Initial testing 84 4.3.2 Permeate flux of membranes 86 4.3.3 Factors affecting the performance of membrane component 90 4.3.4 Short-term off-line filtration tests 101 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS 110 REFERENCES 113 APPENDIX A CALCULATION OF COD REMOVAL EFFICIENCY 127 APPENDIX B CALCULATIONS FOR DISSOLVED METHANE 128 APPENDIX C EXPERIMENTAL DATA 129 iv L I S T O F T A B L E S Table 2.1 Summary of information related to anaerobic processes treating low-strength wastewater at low or ambient temperatures 13 Table 3.1 Characteristics of influent 38 Table 3.2 Nominal design information for membrane modules 40 Table 3.3 Monitoring program to track ongoing proc ess acclimation and performance 45 Table 3.4 Specifications of the gas chromatograph for a nalysis of volatile fatty acids 48 Table 4.1 System operation conditions - HRT and OLR 55 Table 4.2 Average COD concentrations in the influent and the effluent and removal efficiencies in different operational periods 60 Table 4.3 The initial testing results of membrane permeate flux at different TMP 85 Table 4.4 Measurements of cross-flow velocity under different operational conditions 86 Table 4.5 The permeate flux recovery by membrane cleaning 100 Table C - l System operational conditions 129 Table C-2 COD measurement results 137 Table C-3 Biogas production and compositions 145 Table C-4 The concentrations of VFAs in influent and effluent 159 Table C-5 External membranes' permeate flux and permeability 173 LIST OF FIGURES Figure 2-1 Pathway of anaerobic biode gradation 5 Figure 2-2 Total COD and soluble COD removal efficiencies as function of OLR 7 Figure 2-3 Effect of temperature on COD removal at different HRTs 10 Figure 2-4 Total COD and soluble COD removal efficiencies as function of HRT. 11 Figure 3-1 Schematic diagram of anaerobic external membrane bioreactor system 41 Figure 3-2 Anaerobic external membrane bioreactor system - Reactor. 42 Figure 3-3 Anaerobic external membrane bioreactor system - Membrane components. ...43 Figure 4-1 Anaerobic external membrane bioreactor operating temperature 57 Figure 4-2 pH values in influent and effluent samples 58 Figure 4-3 COD concentrations in influent and effluent samples 59 Figure 4-4 COD removal efficiency as a function of OLR 61 Figure 4-5 COD removal efficiency as a function of HRTs 62 Figure 4-6 The effect of temperature on the COD removal efficiency at different HRTs. ...65 Figure 4-7 The concentration of acetic and propionic acids in influent a nd effluent 66 Figure 4-8 The influence of VFA concentration on COD removal efficiency. 68 Figure 4-9 The effluent concentration of VFAs as a function of OLR. „ 69 Figure 4-10 The effluent concentration of VFAs as a function of HRT. 70 Figure 4-11 Gas production in anaerobic external membrane MBR 71 Figure 4-12 Anaerobic external membrane MBR biogas composition 73 Figure 4-13 Revised gas productions in anaerobic external membrane MBR 73 Figure 4-14 The influence of OLR on methane production 75 Figure 4-15 The influence of HRT on the methane production 77 Figure 4-16 Biomass concentrations in the anaerobic external membrane MBR 78 Figure 4-17 Biomass concentrations as a function of organic loading rate 80 Figure 4-18 External AnMBR sludge particle size distributions during acclimatization period. 82 vi Figure 4-19 External AnMBR sludge particle size distributions during steady state operation period 83 Figure 4-20 Submerged AnMBR sludge size distributions 83 Figure 4-21 The external membrane system permeate flux (Jv) 87 Figure 4-22 Permeate flux of organic membrane PCI and Koch with operation time 92 Figure 4 -23 The effect of cross-flow velocity on membrane performance 93 Figure 4-24 The effect of high CFV flushing on the permeate flux of Koch2 94 Figure 4-25 Membrane performance at different combinations of TMP and CFV. 95 Figure 4-26 The effect of gas injection on the permeate flux of Koch2 membrane and Membralox membrane 96 Figure 4-27 The effect of sludge concentration on the membrane permeabili ty. 98 Figure 4-28 Filtration test apparatus setup 101 Figure 4-29 Filtration results of the aerobic sludge at a concentration of 6 g/L 104 Figure 4-30 Filtration resistance of the aerobic sludge at a concentration of 6 g/L 104 Figure 4-31 Filtration results of the anaerobic sludge at a concentration of 14 g/L 105 Figure 4-32 Filtration results of the anaerobic sludge at a concentration of 5 g/L 106 Figure 4-33 Filtration resistance of the anaerobic sludge at a concentration of 14 g/L 107 Figure 4-34 Filtration resistance of the anaerobic sludge at a concentration of 5 g/L 107 Figure 4-35 Permeability of membrane under different sludge concentrations 108 vii A C K N O W L E D G E M E N T S I would like to acknowledge the assistance, support and encouragement that I have received throughout my research. Without their cooperation, this study would not be possible. • Firstly, I would like to convey my gratitude to my supervisors, Dr. Eric Hall and Dr. Pierre Berube, for their encouragement, understanding and firm support throughout the course of the research. • Water Environment Research Foundation (WERF), for sponsoring this project. • Jeffery Chen, my project partner. I enjoyed the time working w ith him. • Bill Liang, for his wonderful ideas and diligent work for the equipment buildup. • Frederic Koch, for his immense help and support in pilot plant. • Paula Parkinson, for her help with analytical work, always with a smile. • Susan Harper, for her technical and logistic help. • Colleen Chan and Kazi Parvez Fattah, for their invaluable opinion on my thesis writing. • My daughter Anna and my husband Bin, for their care and spiritual support. V I M CHAPTER 1 INTRODUCTION Anaerobic biological processes have received high attention in wastewater treatment owing to their attractive advantages of lower energy consumption, lower sludge production and biogas recovery. However, since the growth rates of anaerobic biomass are much lowe r than those of the aerobic biomass, the need to retain a sufficient quantity of active biomass in the reactor limits the wide use of conventional anaerobic biological systems for the treatment of municipal wastewaters (Fakhru'l-Razi, 1993; Wen et al, 1999). The treatment performance of conventional anaerobic systems is highly dependant on the performance of a gravity sedimentation tank. In addition, the effluent quality is of high concern, which limits the use of conventional anaerobic biological systems. Although an anaerobic biological treatment system can effectively remove the bulk ofthe organic contaminants present in the wastewater, it is typically not effective in removing residual levels of organic contaminants. Anaerobic effluents, due to residual organic contaminants, normally need to be poli shed by a downstream aerobic biological reactor (Wen et al, 1999; Baek and Pagilla, 2003). As an alternative technology for the enhancement and improvement of conventional anaerobic systems for wastewater treatment, the anaerobic membrane bioreactor (An MBR) has emerged and been studied. In an AnMBR system, microfiltration or ultrafiltration membranes are used instead of a gravity sedimentation tank to separate the solids from the treated water. By completely retaining suspended solids within the reactor, the AnMBR system shows advantages over the conventional anaerobic biological system (Stephenson et al, 2000; Visvanathan et al, 2000; Hernandez et al, 2002; Stuckey and Hu, 2003). These advantages include the following. • Biomass concentration is independent of the settling characteristics of these solids; • system start-up time is short due to retention of all microbial seed material; • sludge retention time (SRT) can be easily controlled; l • volumetric loading rate capacity is highly improved, thereby reducing the volume of the bioreactor; • residual organic contaminants can be easily hydrolyzed and biodegraded due to prolonged retention in the bioreactor. This improves effluent quality; and • the complete retention of bacteria and pathogens reduces the necessity of disinfection. With the increasing interest in AnMBR technology for wastewater treatment, a few research studies have been conducted on AnMBRs (Kataoka et al, 1992; Chung et ai, 1998). With the geometric world population growth, and with most of population distributed in areas where the temperature in the winter time is relatively low, it is very meaningful to develop a low energy-consuming technology for municipal wastewater treatment at low temperature. Therefore, an AnMBR that can be effectively op erated in temperate regions is, no doubt, very promising. However, little information has been reported regarding the performance and efficiency of AnMBRs when operated under lower temperature conditions. The technical and economical feasibility of applying AnMBR systems for treatment of relatively low strength wastewaters is likely to depend on the necessity of achieving a high level of performance at lower temperatures. The project entitled "Membrane Bioreactors for Anaerobic Treatment of Conventional a nd Medium Strength Wastewater" was initiated in early August 2003 by the Water Environment Research Foundation (WERF). The objective of the research was to develop new design and operational information from laboratory experimental studies for system scale-up to pilot scale. The pilot scale study will provide further information for assessing the technical and economical feasibility of treating relatively low and medium strength wastewater using a membrane bioreactor system operating under anaerobic conditi ons at low temperature. In order to achieve the objective, both a submerged internal membrane AnMBR and external membrane AnMBR were built and operated at the target conditions. The present study 2 presents the external membrane AnMBR operational results and discusses the critical bioreactor parameters and membrane system parameters affecting the performance and efficiency ofthe entire external membrane AnMBR system. 3 C H A P T E R 2 L I T E R A T U R E R E V I E W 2.1 Anaerobic Process Generally, anaerobic treatment processes are favored for treating wastewater with biodegradable Chemical Oxygen Demand (COD) concentrations greater than 2000 mg/L (Hall, 1992; Tchobanoglous et al., 2003). However, numerous studies and practices have demonstrated the feasibility of anaerobic processes for the treatment of low-strength wastewater (Draaijer et al., 1991; Schellinkhout and Collazos, 1991; Collins et al., 1998; Angenent et al., 2000; Elmitwalli et al., 2002; Foresti 2002). The overall anaerobic conversion of biodegradable organic substrates to the end products carbon dioxide (CO2) and methane (CH4) is a complex biogenic process involving a number of microbial populations, often linked by their individual substrate and product specificities. This conversion takes place in three stages: (i) hydrolysis of organic polymers to intermediate organic monomers such as sugars, fatty acids, and amino acids; (ii) the production of lower fatty acids, hydrogen, bicarbonate, pyruvate, and alcohols from organic monomers and (iii) the generation of C H 4 and CO2 from organic acids (Winkler, 1981; Pohland, 1992). Gujer and Zehnder (1983) proposed a six -step system as shown in Figure 2.1, to describe the process in more detail. Two groups of microorganisms, acid-producing and methane-producing bacteria are, in principle, responsible for the overall anaerobic conversion of biodegradable organics. Acid-producing bacteria are responsible for converting heterogeneous substrate into fatty acids. The primary acids produced during acid fermentation are acetic, propionic, and butyric. These acids are subsequently decomposed by methane-producing bacteria, resulting in the production of methane. 4 ORGANIC POLYMERS CARBOHYDRATES PROTEINS, LIPIDS 1 AMINO ACIDS SUGARS HIGHER FATTY ACIDS ALCOHOLS Methanosarcina \ 5 Methanothrix Hydrogen-oxidizing methanogens 1. Hydrolysis 2. Fermentation 3. Anaerobic (B) Oxidation 4. Anaerobic Oxidation 5. Decarboxylation of Acetate 6. Hydrogen Oxidation Figure 2-1 Pathway of anaerobic biodegradation (Adapted from Gujer and Zehnder, 1983) 5 2.1.1 Critical parameters affecting process monitoring and control Volumetric organic loading rate Volumetric organic loading rate (OLR) of an anaerobic process is used as a yardstick to evaluate the efficiency of utilization of the available reactor volume. A higher organic loading rate indicates that more wastewater can be processed per unit available volume. Organic loading rate is not in itself a fundamental design parameter that directly controls process performance. However, all of the complex interactive effects of the process can be lumped together through this parameter. Much research has been conducted to investigate the effect of organic loading rate on the efficiency of a process (Convert! et al., 1993). Hickey et al. (1987) found that severe inhibition of methane production (>70% reduction of methane produced compared to controls) occurred at high organic loading rates, resulti ng in an accumulation of hydrogen and VFAs; however, at lower loading rates, inhibition was less severe and hydrogen accumulated to a level only slightly above controls. By running an anaerobic fluidized bed reactor (FBR) to treat municipal wastewater enriched with glucose, Converti et al. (1993) found that the organic degradation efficiency and the methane production rate were adversely affected by the increase of organic loading rate from 4 to 24 kg COD/m3«day, suggesting a maximum operational value for this parameter. Their results proved the existence of an organic loading rate threshold, beyond which the excessive substrate will cause an increase in the concentration of organic acids, which may behave as competitive inhibitors of digestion. Vartax et al. (1994) arrived at the same conclusion through experiments on attached -film anaerobic digesters. They found that methane production increased as OLR was reduced under low temperature conditions (10 °C). Yu and Anderson (1996) demonstrated that total COD and soluble COD removal efficiencies were functions of OLR as shown in Figure 2.2. 6 90 60 1 ' ' ' 1 0.5 1 1.5 2 2.5 OLR (kgCOD/m 3«day) — COD Soluble CODj Figure 2-2 Total COD and soluble COD removal efficiencies as function of OLR. (Adapted from Yu and Anderson, 1996) pH As indicated in Figure 2.1, the methanogenic bacteria are crucial to the anaerobic stabilization process, constituting a major final step in the transfer of electrons from the various donor species. The aceticlastic methanogens, Methanosarcina and Methanothrix, are of primary importance since approximately 66% to 72% of the methane produced is derived from the acetate in a typical anaerobic microbial conversion, when an abundant source of organic substrate is present (Jeris and McCarty, 1965; Pohland, 1992). However, these anaerobic microorganisms exhibit optimal activities under limited ranges of pH. Aceticlastic methanogens can be adversely affected by the accumulation of hydrogen (Jones, 1991). Most anaerobic conversion processes operate best at near neutral pH. Deviations from this optimum, if not introduced by the influent, are usually caused by the excess production and accumulation of acidic or basic conversion products such as organic fatty acids. Moreover, the intensity of hydrogen ion will affect the toxic or inhibitory characteristics of other potential inhibitors, including both organic and inorganic species (Pohland, 1992; Hall, 1992). Low pH and excessive volatile fatty acid production and accumulation are considered to be conditions that are more inhibitory to methane-producing 7 bacteria than to acid-producing bacteria. The optimum pH range for methanogens is 6.5 -7.5, whereas it is 5.0 - 6.5 for acid-producing bacteria. When the pH is below 6, the growth of methane-producing bacteria falls rapidly, but the acid producers can also continue to produce fatty acids, despite pH depression, thereby aggravating the environmental conditions further. pH has been found to influence system treatment capacity. Perez -Garcia et al. (2005) investigated the effect of the influent pH conditions in an upflow fixed -film reactor for anaerobic thermophilic treatment of wine-distillery wastewater and demonstrated that the neutral or alkaline operation conditions favor the anaerobic biomass ac tivities, therefore, the treatment process. Their results showed that with similar total COD removal efficiency, the reactor could operate at organic loading rate (OLR) around 10.5 kg COD/m3«day with an alkaline influent, but only operate at OLR around 5.6 kg COD/m3*day with an acidic influent. Temperature The wastewater temperature significantly affects the economics and feasibility of anaerobic treatment. The production of methane, the important parameter dominating the economic feasibility, has been shown to be strongly temperature-dependent since the activities of methane producers are significant affected by the operational temperature. Reaction rates generally increase with temperature up to 60 °C (Pohland, 1992). The biomass activities obtained at thermophilic temperatures (55 - 60 °C) are generally 25-50% higher than those at mesophilic temperatures (near 35 °C) (Henze and Harremoes, 1983; Zinder, 1988). However, there are many potential problems associated with thermophilic processes, including low bacterial growth yields and high endogenous death rates (Buhr and Andrews, 1977; Henze and Harremoes, 1983). There have been relatively few species of thermophilic anaerobic bacteria identified (Zinder, 1988), and a small change of temperature can result in a significant change in the level of each species in anaerobic biomass (Ahring et al., 2001). Therefore, it is very possible that the ecological stability of a thermophilic process may be inadequate for steady state operation under frequently shifting conditions. 8 In order to achieve more optimal biological reaction rates and a more stable treatment process as well, reactor temperatures of 25 to 35 °C are generally preferred. (Tchobanoglous et al, 2003). When temperature decreases, the system treatment efficiency decreases significantly. Methanogenic bacteria appear to be active at temperatures of 8 -10 °C, but biomass activities and anaerobic treatment capacities may be reduced to 10 to 20 % of the values obtained at 35 °C (Switzenbaum and Jewell, 1980; Kennedy and Van den Berg, 1982; Grin et al, 1985). Langenhoff and Stuckey (2000) found that every 10 °C drop of temperature could result in about a 10 % decrease in COD removal efficiency when a full scale anaerobic baffled reactor (ABR) operated at a co nstant hydraulic retention time (HRT). Despite discussions above, anaerobic treatment can be applied at lower temperatures and has been sustained at 10 to 20 °C in suspended and attached growth reactors (Vartax et al.,, 1994; Angenent et al, 2001; Bodik et al, 2002 a,b). At the lower temperatures, slower reaction rates occur and thus longer sludge retention time (SRT), larger reactor volumes, and lower organic COD loadings are needed to maintain a high treatment efficiency (Collins et al, 1998). Chu et al (2005) investigated the performance of an expanded granular sludge bed (EGSB) reactor for treating domestic wastewater during a seven-month period under moderate to low temperatures (11 - 25 °C). They demonstrated that system efficiency in terms of COD removal rate was greatly affected by temperature. As shown in Figure 2.3, in the HRT range of 3.5 - 5.7 hours, COD removal efficiency was relatively independent of HRT when the reactor was operated at a temperature higher than 15 °C. However, when the temperature was reduced to 11 °C, an increase of COD removal from 76 to 81% required an increase in HRT from 3.5 to 5.7 hours. This increase in HRT led to a large reactor volume and a low treatment capacity per unit volume of available reactor. Similarly, Langenhoff and Stuckey (2000) observed that the COD removal efficiency of anaerobic baffled reactor was independent of HRT at a higher operating temperature of 35 °C, although the HRT was changed over a large range from 80 hours to 10 hours. 9 100 95 90 2 85 a o u 80 75 3.5 4.5 HRT (hours) 25 0C o 20 0C A 15 0C A 11 OC 5.5 Figure 2-3 Effect of temperature on COD removal at different HRTs. (Adapted from Chu et al, 2005) Viraraghavan and Kikkeri (1990) compared anaerobic filters operating at different temperatures and found that a reactor running at 30 °C consistently achieved higher levels of performance than the other reactors operating at 12.5 °C and 21 °C. Temperature effects on organic removal efficiency were not found to be pronounced at high HRTs. The reactor operating at 12.5 °C was the most affected by changes in HRT. Hydraulic retention time Hydraulic retention time (HRT) is an important operational variable that can be easily manipulated. A short HRT encourages the growth of acid formers and concurrently suppresses the growth of methane producers (Elefsiniotis and Oldham, 1994). The results of effluent VFA concentration measurements obtained by Lew et al. (2003) support this point as well. Lew et al. (2003) found that at longer HRTs the effluent VFA concentration was close to zero, however, at short HRTs (shorter than 3 hours) the effluent VFA concentration increased. Also in their study, the COD removal efficiency was significantly affected by the HRT. For HRTs shorter than 3 hours (corresponding to an influent organic 10 load higher than 5.0 kg COD/m3*day) the COD removal efficiency decreased to 54.3%, while at longer HRTs (from 3 to 24 hours) the COD removal was always constant (around 82%). They attributed the deteriorated effluent quality to the incomplete organic degradation due to the very short HRT. By operating a modified anaerobic baffled reactor treating pre -settled municipal wastewater at ambient temperature, Yu and Anderson (1996) demonstrated that total COD and soluble COD removal efficiencies were functions of HRT as shown in Figure 2.4. 90 , , 60 I ' 1 1 ' 1 1 1 1 3 4 5 6 7 8 9 10 11 HRT (hours) • COD Soluble COD Figure 2-4 Total COD and soluble COD removal efficiencies as function of HRT. (Adapted from Yu and Anderson, 1996) 2.2 Anaerobic Process for the Treatment of Municipal, Low -strength Wastewater at Low Temperatures Extensive research has been conducted to investigate and evaluate the performance and efficiency of anaerobic process es treating lower strength wastewaters at lower temperatures . Stensel and Strand (2004) recently examined more than 50 reports over the period from 1981 to 2002 to investigate the effects of temperature, wastewater strength and OLR on treatment performance of anaerobic reactors. Thirteen lab and pilot scale anaerobic reactors were operated at temperatures equal to or less than 20 °C to treat municipal wastewater. An effluent COD of equal to or less than 65 mg/L was achieved. Most of the reactors were fixed film or upflow anaerobic sludge bed (UASB) reactors. Particular UASB reactors have received extensive attention for the treatment of municipal wastewater at low or ambient temperatures. Singh and Viraraghavan (2003), and Lew et al. (2003) ran the lab scale UASBs at temperatures as low as 6 °C and 10 °C, respectively. At these conditions, they observed COD removal efficiencies of 40% and 44%, respectively. From the results, Lew et al. (2003) concluded that solids accumulation in the reactor would be pronounced due to the low bacterial hydrolytic activity at low temperature. Singh and Viraraghavan (2003) concluded that the poor performance was due to the presence of anaerobic soluble microbial products in the effluent which could not be completely degraded under low temperature conditions at low HRTs. Their conclusions imply that complete retention of feed TSS will be a key factor leading to a successful anaerobic treatment process operation at low temperatures. Seghezzo et al. (2002) operated a pilot scale UASB reactor treating settled municipal wastewater at a mean ambient temperature of 16.5 °C. During two years of operation, UASB reactor efficiently treated low -strength domestic sewage at ambient temperature with total and suspended COD removal efficiencies of approximately 70 and 80%, respectively. In addition to fixed-film and UASB reactors, extensive research has been carried out to examine the performance of other anaerobic reactor configurations. Table 2.1 12 Table 2.1 Summary of information related to anaerobic processes treating low-strength wastewater at low or ambient temperatures Influent Reactor Operating temperature (°C) HRT (hours) COD removal efficiency (%) Reference Low-strength soluble wastewater with constant COD of 600 mg/L Anaerobic Migrating Blanket Reactor (AMBR) 15-20 59 (Total COD) 73 (Soluble) Angenent etai, 2001 Synthetic municipal wastewater with COD of 300 mg/L Up-tlow Anaerobic Filter (UAF) 9-23 10, 20 46-92 Bodik etal., 2001(b) Anaerobic Sequencing Batch Reactor (AnSBR) 10-46 56-88 Synthetic wastewater with COD of380-850 mg/L Membrane-coupled Expended Granular Sludge Bed (EGSB) 15 -25 3.5-5.7 85-96 Chu et al., 2005 11 76-81 Synthetic wastewater with COD of 1000, 800, 600, 400 mg/L Anaerobic Sequencing Batch Reactor (AnSBR) 35,25, 20 48, 24,16,12 80-90 Ndon and Dague, 1997 15 <80 Diluted wastewater with COD of 500 mg/L Anaerobic Baffled Reactor (ABR) 20 80 70 Langenhoff and Stuckey, 2000 10 60 Raw municipal sewage Fluidized Bed Reactor (AFBR) 20-5 1.5 70 Sanz and Fdz-Polanco, 2003 Pre-settled municipal wastewater with COD of 370 -520 mg/L Modified Anaerobic Baffled Reactor (ABR) 18-28 10-4 83-68 Yu and Anderson, 1996 Domestic wastewater with total COD of 300 - 600 mg/L Two-step system consisting of a Filter and a Hybrid Reactor 13 4 + 8 71 Elmitwalli et ai, 2002 summarizes some research work investigating the performance of anaerobic reactor s treating low-strength wastewater at low or ambient temperature. From the studies presented above, the following statements can be made. • Anaerobic bioreactors show good potential for treating low-strength wastewater at low temperatures. However, the results are mostly based on bench-scale experiments and pilot plants. No information about large commercial scale systems operating under these conditions is provided or referenced by the authors . • The success obtained with phase-separated systems in the treatment of domestic sewage at low temperatures indicate s the importance of the hydrolysis step. This implies an advantage for the anaerobic membrane bioreactor due to the complete retention of solids, giving particulate or colloidal COD enough time for degradation. In a two-step anaerobic filter plus hybrid reactor (AF+AH) system operated by Elmitwalli et al. (2002), the COD removal efficiencies obtained at a temperature of 13 °C were similar to that in a one step UASB reactor in tropical countries ( Draaijer etal, 1991; Schellinkhout etal, 1991). • By allowing the microorganisms to acclimate to the new temperature, a gradual temperature decrease does not have a great effect on COD removal efficiencies, but as temperature decreases a longer HRT is needed. • The start-up period is considered to be a crucial step for the stable operation of anaerobic reactors. It is possible to start up an anaerobic reactor at low temperature. Viraraghavan and Kikkeri (1990) found that the start-up of an anaerobic filter was possible at 21 °C without any adverse effect on its future performance. However, high quality inoculum is required. Without inoculation the start-up period can last up to 6 months. With inoculation, even with very poor inoculum, this period can be greatly shortened. • Reactors coupled with a membrane for liquid-solids separation exxhibit a higher level of performance than other reactors at similar operation conditions. Except for the Chu et al (2005) study listed in Table 2.1, Wen et al (1999) study referenced by Stensel and Strand (2004) also showed higher COD removal efficiency (>92%) in a 14 temperature range of 12 - 27 UC. • Although considered to be low-strength wastewater, municipal sewage is quite complex due to the high fraction of particulate COD, and the presence of fatty compounds, proteins and detergents (Foresti, 2002). These characteristics impose limitations on the anaerobic process with respect to COD removal efficiency, and also in terms of the maximum organic (OLR) and hydraulic loading rates (HLR) to be applied. These limitations impose the need for post -treatment in many situations. • The need for development of low-cost systems to face financial constraints is the main factor leading to the implementation of anaerobic technology. For low-strength wastewaters, especially municipal sewage, the energy balance is favorable only if heating can be avoided. It is only in tropical regions where the reactors can be operated at ambient temperatures that are usually higher than 20°C (Foresti, 2002). 2.3 Anaerobic Membrane Bioreactor (AnMBR) Treatment System 2.3.1 Anaerobic MBR experience There is widespread interest in the anaerobic membrane bioreactor (MBR) technology for wastewater treatment. Work in 1959 at the University of California, Los Angeles, by Sourirajan and Loeb on asymmetric, thin-skin, cellulose acetate membranes is generally credited as launching the modern era of membrane use. Since then, the development of membrane technology used in wastewater treatment has continued, leading to the emergence and development of the anaerobic MBR. • In 1978, Grethlein (1978) reported results from a research project involving the coupling of an anaerobic reactor to a membrane module for sanitary wastewater treatment in the United States. This is considered to be the first known anaerobic membrane project. 15 In 1982, Dorr-Oliver introduced an AnMBR process, the Membrane Anaerobic Reactor System (MARS), for high strength industrial wastewaters with CODs greater than 15,000 mg/L (Sutton et al, 1983; Sutton and Evans, 1983; Sutton, 1986 a,b). In 1983, Choate et al. reported the first full scale application of the technology in England, in which an ultrafiltration (UF) membrane unit was coupled to an existing suspended growth anaerobic reactor installed at Tenstar Products in Ashford, U.K. to treat wheat flour processing wastewater. The system operated successfully for 2 years with no membrane failures. In 1985, the Japan Ministry of International Trade and Industry began the Aqua-Renaissance 90 R&D project, "New Wastewater Treatment System" , which included development of AnMBR step for treatment of separated municipal wastewater solids (Kirmura, 1991; Kiriyama et ai, 1992). In 1985, researchers at the University of Newcastle developed an AnMBR system using a crossflow MEMSEP MF unit (Saw et al, 1986). Subsequent research indicated significant flux declines occurred due to entrapment of biomass in the microporous membrane structure (Anderson et ai, 1986). In 1987 and 1988, South African researchers piloted an AnMBR system referred to as the anaerobic digestion ultrafiltration (ADUF) process. The system used tubular ultrafiltration membranes manufactured in Paarl by Bintech Ltd. In the early 1990s, the ADUF process was applied to a variety of food processing and beverage wastewaters (Ross etal, 1990; Ross etal, 1994; Botha etal, 1992). In the early 1990s, AnMBR systems were applied to sludge treatment. In Durban, South Africa it was shown that by applying an MBR to separate sludge from the liquid stream, the performance of the digester was enhanced (Pillay et al, 1994). In the early 1990s, the application of an anaerobic MBR for waste treatment and resource recovery in the management of livestock waste slurries was first reported in Norway (Bilstad et al., 1992). In the late 1990s, the application of an anaerobic MBR to municipal, low strength wastewater treatment caused extensive attention. Research showed the promising 16 future of this technology (Wen et al., 1999; Baek and Pagilla, 2003; Stuckey and Hu, 2003). Sutton, Berube and Hall (2004) recently examined 56 reports documenting anaerobic MBR system design and performance information. The information from these references related to this study is presented below. • Although these reported studies from the period from 1978 to 2003 cover a wide variety of wastewater treatment applications, most of these research reports focused on high strength industrial wastewater, as evident from the number of reports (50 out of 56). The study of anaerobic MBRs for municipal, low strength wastewater treatment became increasingly popular in the late 1990s. Anaerobic MBR application to municipal wastewater treatment has been investigated in the Far East, Europe and South Africa. Although there are no known commercial, large scale, anaerobic MBR systems in operation treating municipal wastewater, active research with respect to this application has recently been reported in China, England and the U.S. (Wen et al., 1999; Baek and Pagilla, 2003; Stukey and Hu, 2003). • Anaerobic MBR systems for the treatment of a wide variety of industrial wastewater s at the laboratory, pilot and full scale level have been reported. These systems generally involved external MBR configurations, in which an ultrafiltration (UF) or microfiltration (MF) membrane is located outside to the bioreactor. Tubular and flat sheet membranes represent the most popular external membrane configurati on. In most of the laboratory and pilot scale studies, the research focused on the investigation of the efficiency of the membrane component and /or the mechanisms governing membrane fouling. The membrane component is used as a separation measure for the complete retention of solids but operated under conditions to maximize the permeate flux, which is a key factor affecting the economic feasibility of the system. Therefore, little information about optimum membrane operating conditions can be obtained from these studies. Also, there is little information about the performance of 17 membranes in large pilot scale and full scale applications in recent years. Most of the experiments were completed at temperatures above 35 °C. A few were carried out in the thermophilic temperature range. Consequently, little information was provided regarding the performance and efficiency of anaerobic MBRs when operated at low or ambient temperatures. Gas production in most experiments conducted at a temperature of 35 °C was in the range of 0.2 to 0.3 m3 CH4/kg COD removed, which is lower than the theoretically calculated 0.35 m3 CHVkg COD removed at the same temperature. Gas production in the experiments operating at 55 °C was higher than that of experiments operating at a temperature of 35 °C. This might be caused by higher anaerobic biomass activities and lower methane solubility at higher temperature. Hydraulic retention time (HRT) did not have much influence on gas production in terms of methane volume per unit COD removed, when the system was operated at high temperature. Good effluent quality can be achieved with effluent COD and BOD5 values less than 100 and 50 mg/L, respectively. Most of studies achieved COD and BOD removal exceeding 90 percent. Compared to the other anaerobic reactor configurations, the performance of the MBR has been largely improved. This is mainly because of the absolute retention of the biomass produced in the reactor together with the retention of organic solids originating in the feed. The reported results imply that MBR technology represents an ideal reactor configuration for achieving the highest quality effluent feasible in the anaerobic treatment of wastewater. Laboratory, pilot and full scale anaerobic MBR systems have been operated at OLRs ranging from less than 5 to over 30 kg COD/m3*day. In general, the higher OLRs reported were based on the treatment of readily biodegradable wastewater utilizing external MBRs in which a high concentration of MLSS (i.e., 20 to 60 g/L) was maintained in the reactor. The effect of variation of OLR on COD removal efficiency is not obvious. In an extreme situation, the COD removal only decreased 3% from 99% to 96% when the OLR increased from 1 to 19.7 kg COD/m3«day. Anaerobic MBR systems have been operated at HRTs ranging from less than 0.5 to 15 days. In general, the COD removal was above 90% except at extremely low HRT. When one 18 system was operated at an HRT of around 0.5 days, the COD removal efficiency was only greater than 70% (Guan et al., 2000). One exception is the experiment conducted by Okamura (1994), which achieved COD removal efficiency of 92% at an HRT of 0.5 days and a temperature of 52 °C. The performance of phase separated systems was better than single phase systems when they were operated under similar conditions (Kataoka et al, 1992; Yushina and Hasegawa, 1994). 2.3.2 Anaerobic MBR treatment of municipal wastewater Although anaerobic treatment has been widely used for municipal wastewater, as discussed in Section 2.2, the use of the anaerobic MBR configuration is still limited. However, due to the advantages of the anaerobic MBR approach, the ap plication of this technology for the treatment of municipal wastewater has generated extensive attention and active research as discussed in Section 2.3.1. In a review by Sutton et al. (2004), five laboratory or pilot scale studies were referenced, providi ng general information about the design, operation and performance of anaerobic MBR treating municipal wastewater. The information from these five references follows. • In 1978, Grethlein (1978) reported his laboratory research work, the first known research project involving a membrane-coupled anaerobic reactor to treat domestic wastewater. In this study, external flat sheet and tubular reverse osmosis (RO) membrane modules were coupled to a septic tank. The author examined the membrane flux maintenance and the effect of the membrane on the septic tank. It was shown that the BOD reduction in the septic tank was 90% and the specific rate of reaction for the organic component measured as BOD was 2 4 mg/L«day. The entire remaining BOD in effluent was in a solubl e form. The effluent was clear with zero turbidity, and free of E. coli. The system was fed 11 to 19 L wastewater per day, equal to an HRT of 9.6 to 5.6 days. No operational temperature was mentioned. • In 1992, Naoaki et al. (1992) reported on research work on three pilot scale 19 anaerobic membrane bioreactors treating low-strength wastewater. Although their results focused on the examination of bacterial charact eristics, they did give some information about the system configuration and performance. Plant A was used for domestic sewage and consisted of a hydrolization reactor and a UASB reactor coupled to an external ultrafiltration (UF) module for methane fermentation. The hydrolization reactor was run at an HRT of 3 days and at a temperature of 35 °C. The UASB was run at an ordinary temperature that was not specified. The COD and BOD removal efficiencies for the complete system were 86.6% and 83%, respectively. Plant B was used for municipal sewage and consisted of a hydrolization reactor coupled to a UF membrane module and a fluidized bed reactor (FBR) coupled to a MF membrane module for methane fermentation. The hydrolization reactor was run at a HRT of 5 days and at a temperature of 32 °C. The operating temperature was not specified. The COD and BOD removal efficiencies for the complete system were 91.4% and 89.7%, respectively. Plant C was used for soybean-processing wastewater, beyond the topic of this section. The discussion implied that the long retention time and the high reactor biomass concentration achievable by membrane separation provided an advantage for the interaction between bacteria and their dependant associates regarding substrate incorporation. A significant laboratory research on the application of a submerged membrane MBR for the anaerobic treatment of synthetic municipal wastewater has been done by Stuckey and Hu at Imperial College in London (Stuckey and Hu, 2003). This anaerobic system was operated at a bioreactor HRT as low as 3 hours and at a temperature of 35 °C. The final effluent carbonaceous BOD5 (CBOD5) and total suspended solid (TSS) values of less than 30 mg/L confirmed the feasibility of an anaerobic MBR system for treating degritted and screened municipal wastewater. A laboratory experimental study on the application of an external membrane MBR for anaerobic treatment of municipal wastewater was completed by Baek and Pagilla at the Illinois Institute of Technology in Chicago (Baek and Pagilla, 2003). The system was operated at a low bioreactor HRT of 12 hours and at a temperature of approximately 32 °C. The achieved effluent CBOD5 and TSS values of less than 30 20 mg/L demonstrated the feasibility of an anaerobic MBR system for treatin g primary treated municipal wastewater. • A six month laboratory study of a reactor combining a UASB and a Mitsubishi hollow fiber, UF membrane module for the treatment of municipal wastewater was performed by Wen and others at Tsinghua University in Beijing (Wen et al, 1999). The system was operated at a low bioreactor HRT of 4 hours, and at a temperature as low as 12 °C. The effluent C B O D 5 and TSS values of less than 30 mg/L implied the feasibility of an anaerobic MBR system for treating municipal wastewat er at low temperatures. In addition to the research referenced by Sutton et al (2004), a seven month laboratory study of an anaerobic MBR has been reported recently (Chu et al, 2005). An expanded granular sludge bed (EGSB) reactor coupled with hollow fi bre membrane filtration was operated by Chu and others at Dalian University of Technology in Dalian, China for the anaerobic treatment of synthetic domestic wastewater. The system was run at an HRT of 3.5 to 5.7 h, and at a temperature in the range of 11-25 °C. With temperatures above 15 °C, the system was capable of removing 85 -96% of total COD and 83 -94% of total organic carbon (TOC). At 11 °C, by increasing the HRT from 3.5 to 5.7 hours, the total COD removal efficiency was increased from 76 to 81%. The results further confirm the potential of anaerobic MBR technology for the treatment of municipal wastewater at low temperatures. 2.4 Mechanisms Governing the Efficiency of the Membrane Component of the External Anaerobic MBR In the review by Sutton et al. (2004), the authors classified mechanisms that impact the permeate flux in an AnMBR into three general categories: those that are governed by the membrane itself, those that are governed by the operational parameters of the membrane and those that are governed by the characteristics of the mixed liquor being filtered. The 21 characteristics of the membrane material (e.g. polymeric versus ceramic, charge, pore size), the membrane packing density (i.e. membrane area per unit volume) and the membrane configuration (i.e. external or submerged) are specific parameters for a given membrane product. Although some operating conditions (e.g. surface shear, operating trans-membrane pressure TMP, operating temperature) are relatively fixed for a specific membrane, they can be typically varied within a specific range. Although the other operating parameters such as the organic loading rate (OLR), the sludge retention time (SRT) and the hydraulic retention time (HRT) are more specific to the biological component of an AnMBR, they can significantly affect the permeate flux of the membrane component. The characteristics of the mixed liquor being filtered such as sludge concentration, can also significantly affect the permeate flux. The information related to an external AnMBR, as it relates to this project, is discussed below. 2.4.1 Mechanisms governed by the membrane system Membrane material The fouling mechanisms governing the performance of organic membranes are different than those for inorganic membranes. According to studies by Kang et al. (2002), Lee et al. (2001) and Choo and Lee (1996a), for organic membranes, the external fouling due to cake layer formation is more significant than internal fouling caused by the adsorption of soluble and/or particulate material within the pore structure of a membrane . Choo et al. (1996a) reported that the cake layer that forms on organic membranes in an AnMBR consisted of both biological/organic solids and inorganic precipitates (generally referred to as struvite), and that the inorganic precipitation was especially responsible for hardening the cake layer at the membrane surface where the strong binding and solidification led to pronounced external fouling. Choo et al. (2000) reported that periodic backfeeding of acidic wastewater nearly doubled the permeate flux for the organic membrane, implying that inorganic precipitation is an essential part of the foulant on the organic membrane surface. However, they also reported that there was little difference in the organic 22 membrane permeate flux before and after the struvite formation was suppressed, implying struvite is not a big contributor to organic membrane fouling. Their results are somewhat mutually contradictory. For an inorganic membrane, internal fouling predominates and a cake layer typically does not form on an inorganic membrane (Yoon et al, 1999; Kang et al, 2002). Yoon et al. (1999) attributed the extensive internal fouling that occurs in inorganic membranes to the precipitation of struvite. Using Scanning Electron Micros cope (SEM) image analysis, they observed almost no microbial cake layer deposited on the inorganic membrane, but the images revealed struvite precipitates within the pore structure of the inorganic membrane. Studies by Kaiig et al. (2002) and Choo et al. (2000) further demonstrated that the internal foulant material in inorganic membranes was struvite. They reported that the flux improvement of the ceramic membrane was pronounced after the struvite formation was suppressed. However, Elmaleh and Abdelmoumni (1997) reported a contradictory result. They observed that the formation of a cake layer was the principal mechanism governing the reduction in the permeate flux through an inorganic membrane. The structural difference between organic and inorganic membra nes may be a factor that causes different permeate flux. For the two types of membranes investigated by Kang et al. (2002), the inorganic membranes had a smooth surface and a pore diameter of 0.14 pm, while the organic membranes had a rougher, fibrous surface and a pore diameter of 0.2 pm. Hydrophobic nature and charge of the membrane Some contradictory results have been published with regard to the effect of the hydrophobicity of a membrane material in an AnMBR on the permeate flux. On one hand, some research has shown that a hydrophilic membrane surface is beneficial for increasing permeate flux. Sainbayar et al. (2001) reported that the permeate flux through a hydrophobic membrane could be increased through graft polymerization, which introduces hydrophilic functional groups on the membrane surface. Choo et al. (2000) also reported that graft polymerization of an organic membrane led to a significant flux improvement. 23 However, there is an optimal degree of grafting. Choo et al. (2000) reported that the effectiveness of chemical modification for fouling control was most significant at 70% in the degree of grafting. On the other hand, Choo and Lee (1996b) observed that the extent of fouling was lower for membrane materials that were more hydrophobic in natur e. These results suggest that surface hydrophobicity on its own does not govern membrane fouling (Choo and Lee 1996b). The membrane surface charge plays a significant role in membrane fouling. Shimizu et al. (1989) reported that negatively charged inorganic membranes fouled less rapidly than non-charged or positively charged membranes during the filtration of an anaerobic broth. They attributed the difference to a stronger electrical repulsion between negatively charged colloids in the broth and the membrane surface. Also, Kang et al. (2002) reported that the charge that a membrane adopted during the cleaning process significantly affected the extent to which the permeate flux could be recovered. However, when filtering protein solutions, Fane et al. (1983) reported that the impact of the membrane surface charge was negligible when the ionic concentration of the solution being filtered was high. Nominal pore size As an important parameter of a membrane system, the nominal pore size of a membrane can also significantly affect the permeate flux. The research by Elmaleh and Abdelmoumni (1997) indicated that the optimal membrane pore size was a function of the particle size of mixed liquor being filtered. They reported that the permeate flux obtained from a membrane with a nominal pore diameter of approximately 0.45 pm was highest when filtering an anaerobic mixed liquor, while the highest permeate flux was obtained from a membrane with a nominal pore diameter of approximately 0.15 pm when filtering a mixed microbial population of methanogens. Choo and Lee (1996b) reported that the optimal pore size for an AnMBR was 0.1 pm when filtering anaerobic digestion broth. Chung et al. (1998) reported that the permeate flux achieved by the membrane with a nominal pore size of 0.22 pm was three times higher than that achieved by the membrane with a pore size of 0.6 pm. However, the initial permeate flux achieved by a membrane with a larger pore 24 size was greater than that achieved by a membrane with a smaller pore size (Saw et al, 1986; Imasaka et al, 1989). However, the decrease of permeate flux for a membrane with a larger pore size is much faster than that for a membrane with a smaller pore size ( Saw et al, 1986; Imasaka et al, 1989; Wen et al, 1999). Chang et al (1994) found that although the initial flux of a 0.4 urn membrane was much higher than that of a 0.05 pm membrane, the flux produced by the membrane with a 0.05 pm pore size was always higher than that of the 0.4 pm membrane after the operation reached steady state. He et al. (1999) reported that membranes with a larger molecular weight cut -off fouled more rapidly and to a greater extent. Imasaka et al. (1989) attributed the increase in the rate of fouling with an increase in the membrane pore size to an increase in internal pore fouling. Membranes with a larger nominal pore size are more readily clogged by macro-colloids, which can completely block the entrance ofthe pores. In contrast, if the membrane pore size is much smaller than that of the particles to be filtered, the particles will not be able to plug the pores, but to simply roll off under the shear forces generated by the flow. The pore size had no impact on the extent of cake fouling, but the resistance due to clogging increased with an increase in pore size. A different result was observed by Hernandez et al. (2002) when they investigated the retention of granular sludge at high hydraulic loading rates in an AnMBR with immersed fdtration. Suction pressure for membranes with a nominal pore size of 10 pm increased more rapidly than that for membrane with a pore size of 100 pm regardless of operation at low or high OLR. The discrepancy can be explained by the differences in the mechanisms that govern the fouling of microfdtration membranes (pore size greater than 10 pm) and those that govern the fouling of ultrafiltration membranes. The foulant layer formed on microfiltration membranes is more susceptible to collapsing under elevated trans-membrane pressures due to the coarse membrane surface (Saw et al, 1986). In addition, when the pore size of a membrane is much greater than the size of particles being filtered, the particles can pass through the pore easily but block the entrance of the pore. 25 2.4.2 Mechanisms governed by the operational parameters Cross-flow velocity (CFV") An increase of cross-flow velocity can significantly decrease the resistance due to concentration polarization and the resistance due to cake layer formation (Grethlein, 1978; Choo and Lee, 1998). In some extreme case, the total fouling resistance can be reduced to virtually zero (Elmaleh and Abdelmoumni, 1997). In general, permeate flux increases with an increase of cross-flow velocity (Elmaleh and Abdelmoumni, 1997, 1998; Imasaka et al., 1989). However, a permeate flux plateau was observed once the shear stress caused by the cross-flow velocity reached a certain level. Choo et al. (2000) reported that a plateau was reached at a Reynold's Number of approximately 2000, after which no further reduction in the resistance could be achieved by increasing the cross -flow velocity. Elmaleh and Abdelmoumni (1997, 1998) found that the introduction of baffles to the tubular membrane promoted turbulence, which had a similar effect on the permeate flux as an increase in the cross-flow velocity. Saw et al. (1986) also reported that the permeate flux in an AnMBR increased to a greate r extent with an increase in the cross -flow velocity, when the flow through the membrane was turbulent. However, although the permeate flux can be increased by increasing the cross-flow velocity, this increase comes at a cost. Bourgeous et al. (2001) reported that an increase in the cross-flow velocity from 1 to 2 m/s increased the permeate flux by 20%, as well as the power cost for the system by 58%. In addition, Lee et al. (1995) observed that the high cross-flow velocity required to generate high shear conditions can generate large axial pressure gradients, resulting in a non-uniform trans-membrane pressure (TMP) in tubular membrane systems. As a consequence, some sections of the membrane can be under non -optimal TMP conditions. The cross-flow velocity cannot always be optimized solely for the benefit of improved permeate flux in an AnMBR. The high shear stress caused by high cross -flow velocity significantly affects the biomass activity (Brockmann and Seyfried, 1996; Ghyoot and Verstraete, 1997), reducing the floe size in the mixed liquor being filtered (Choo and Lee, 1998), increasing the cell lysis that results in an decrease in the overall activity of the 26 biomass (Choo and Lee, 1996a), and releasing high concentrations of extracellular polymeric substances (EPS) into the bioreactor (Kim et al, 2001). Chu et al. (2005) found that EPS tends to accumulate on the membrane surface in an AnMBR, filling the void spaces between the particles in the cake layer. This results in a drastic reduction of permeate flux compared to that expected for filtration of a porous cake layer. For inorganic membranes, internal fouling can dominate, especially at high cross -flow velocities (Kang et ai, 2002). Although the extent of internal fouling is typically considered to be independent of the cross-flow velocity, internal fouling can increase slightly as the cross-flow velocity increases (Choo and Lee, 1998; Choo et al., 2000). This increase in the extent of internal fouling can be attributed to the thinning of the cake layer, which is considered to behave as a so-called self-rejective dynamic membrane (Imasaka et ai, 1989), subsequently increasing the passage of foulants into the membrane pores. Trans-membrane pressure (TMP) Beaubien et al. (1996) described the relationship between permeate flux and trans-membrane pressure by Equation 2.1: J= , AP' x Eq2.1 In this equation, J is the permeate flux (um/s), APt the applied trans-membrane pressure (Pa) and u the permeate viscosity (Pa s). The total resistance of the membrane is grouped into two; one encompasses all membrane-solute interactions presumed unaffected by operating parameters (R m) and one is a function of the operating conditions of the system (Rg)-Rg = BAP, Eq. 2.2 Where, p depends on the variables affecting the mass transfer properties of the system. From Equation 2.1, two pressure regions can be distinguished, a low pressure zone where the hydraulic resistance of the membrane is the governing factor (R m »|3AP t) and a high pressure zone where the gel layer resistance was dominant (j3APt » Rm)- In the low 27 pressure zone, the permeate flux increases linearly with an increase of pressure. The solids concentration only affects the permeate flux at low concentrations (i.e. less than 2.5 g/L). Higher solids concentrations do not have much effect on permeate flux. Membrane permeability does not depend on fluid shear stress. In the high pressure zone, surface shear caused by cross -flow velocity significantly affects the permeate flux. The permeate flux increases linearly with an increase in cross-flow velocity. However, the magnitude of the increase in the permeate flux is lower at higher MLSS concentrations. Beaubien et al. (1996) attributed the lower magnitude of the increase in the permeate flux to the higher rate of mass transfer towards the membrane and/or to the increase in the viscosity of the mixed liqu id that occurred at higher MLSS concentrations. The trans-membrane pressure has little effect on the permeate flux. Saw et al. (1986) also observed that the permeate flux in an ultrafiltration membrane remained constant with an increase in the trans -membrane pressure in the high pressure region. However, they found that the permeate flux in a microfiltration membrane decreased with an increase in the trans-membrane pressure in this pressure region. They attributed this discrepancy to the different foulant layer structures formed on ultrafiltration membranes and microfiltration membranes. In the latter case, the foulant layer was more susceptible to collapsing under elevated trans -membrane pressures. However, Elmaleh and Abdelmoumni (1997) reported a decrease in the permeate flux with an increase in the trans-membrane pressure for both filtration ranges, and they attributed the decrease to the compaction of the foulant layer. Beaubien et al. (1996) suggested an optimal trans-membrane pressure which increases with an increase in the cross-flow velocity, but was independent of the suspended solids concentration. No significant improvement in filtration performance could be expected from operating at higher trans-membrane pressures once the optimal trans-membrane pressure had been reached, since increased membrane fouling would result. 28 Permeate flux and its recovery Membrane permeate flux is also a key factor affecting the membrane fouling rate in terms of trans-membrane pressure change. Wen et ai (1999) found that trans-membrane pressure increased slowly with the lower flux than with the higher flux. When the trans-membrane pressure up to 70 kPa was taken as a limited value, the operation period could be kept over 2 weeks with a flux of 5 Lmh, 8 days longer than that observed during the operation at a flux of 10 Lmh. In addition, the membrane relaxation cycle also play ed an important role in keeping a stable operation, ln order to investigate the influence of an intermittent suction mode on the change of trans-membrane pressure, Wen et ai (1999) ran their experiment with different intermittent operational modes and found that the system was most effectively operated in a mode of 4 minutes on and 1 minute off. Continuous operation over 2 weeks could be ensured under this condition before the trans-membrane pressure reached the limit of 70 kPa. It is widely accepted that caustic solutions are effective in removing organic/biological foulants that mainly form on an organic membrane surface, while acidic solutions are effective in removing inorganic foulants that mainly contribute to the fouling of inorganic membranes (Sutton et ai, 2004). However, some different results have been reported. Lee et al. (2001) reported that cleaning with alkali solution (1 N NaOH) alone did not enhance organic membrane flux sufficiently, even though it was expected to enhance permeate flux by removing biological fouling effectively. When both the 1 N NaOH solution and 1 N HCI solution were used in series to clean a fouled membrane, the permeate flux increased greatly. This implies that the inorganic foulant is the main component of a foulant layer on an organic membrane surface. This was further confirmed by Choo et ai (2000) and Kang et ai (2002). They reported a consistent permeate flux recovery of an organic membrane by back-flushing the membrane exclusively with an acidic solution. These results suggest that the suppression of struvite, the component of a cake layer formed on organic membranes (Kang et ai, 2002), will be beneficial for the control of organic membrane fouling. 29 On the other hand, a number of studies have reported an inconsistent permeate flux recovery by back-flushing with an acidic solution for an inorganic membrane, regardless of the type of acidic solution used (Yoon et al, 1999; Choo et al., 2000; Kang et al., 2002). These results are somewhat counter intuitive since the internal pore fouling in inorganic membranes has been attributed mainly to struvite (Yoon et ai, 1999; Kang et al, 2002), which is soluble under acidic conditions. Chang et al. (1994) even reported a negative flux recovery when ceramic membranes were cleaned with hydrogen peroxide followed by nitric acid. These research groups attribute the poor recoveries observed when back-flushing an inorganic membrane with acid ic solutions to a change in the membrane surface charge. At low pH, anionic ligands replace the membrane surface hydroxyl groups, resulting in a positive charge on the membrane surface and, therefore, promoting the adsorption of the foulants remaining in the cleaning solution. Operating temperature Higher operating temperatures can have beneficial impacts on the permeate flux since the viscosity of the permeate decreases with an increase in temperature. Hogetsu et al. (1992) reported an increase in the permeate flux of over 30% when the operating temperature was increased from 40°C to 47°C. Beolchini et al. (2005) also reported a positive effect of temperature on the permeate flux. Operating at the same trans-membrane pressures, the permeate flux of a tubular ceramic membrane at 40 °C was 30- 60 % higher than that at 30 °C. Wang et al. (2005) reported that high temperature treatment was effective in reducing fouling layer specific resistance. Faster foulant consolidation of the fouling layer, but lower specific resistance and thus higher steady state flux, occurred when a microfiltration membrane was operated at 40 °C. Schiener et al. (1998) and Fawehinmi et al. (2004) observed a decrease in the concentration of soluble microbial products in a conventional anaerobic bioreactor as the operating temperature increased. The beneficial impacts of elevated temperature can be explained by the reduction of soluble microbial products in mixed liquor being filtered. However, the study conducted by Zoh and Stenstrom (2002) showed a marginal increase of permeate flux as the temperature increased from 15 °C to 40 °C. The authors realized that their results were contradictory 30 with other research, but they did not give any explanation. Most of the results presented in this review are from studies that were performed using an AnMBR operating at temperatures in excess of 30°C. Vazquez and Benavente (2003), through atomic concentration percentages and X-ray photoelectron spectroscopy (XPS) analysis, found a chemical degradation of the membrane surface around 40 °C, causing changes in the membrane structure. This phenomenon can explain the changes of permeate flux with temperature changes in a high range. However, a number of studies have been performed at low or ambient temperatures. Wen et al. (1999) were able to maintain a relatively high permeate flux in an AnMBR operating at temperatures ranging from 14 to 25°C over an extended period of time, when the membrane was operated with a relaxation period. Kiriyama et al. (1992) also successfully operated an AnMBR at temperatures ranging from 20 to 25 °C. 2.4.3 Mechanisms governed by the characteristic of the mixed liquid Suspended solids concentration The adverse effect of mixed liquid suspended solids (MLSS) concentration in an AnMBR has been reported. To maintain a constant permeate flux in an AnMBR treating synthetic wastewater, the TMP required at an MLSS concentration of 35 g/L was over two times greater than that required at an MLSS concentration of 7 g/L ( Stuckey and Hu, 2003). In another anaerobic treatment process with membrane fdtration for digested sludge, the permeate flux at steady state operation log-linearly decreased with an increase in the MLSS concentration (Saw et al., 1986). In an AnMBR study conducted by Kitamura et al. (1996) treating distillery wastewater, the permeate flux also decreased with an increase in the MLSS concentration. However, as discussed in the section on trans-membrane pressure, the impact of solids concentration was reported to be different under different operational conditions (Beaubien et al, 1996). At a relatively low TMP, the impact of MLSS concentration on the permeate flux was significant when MLSS concentration was 31 less than 2.5 g/L, while the impact became moderate when the MLSS concentration was higher. At a relatively high TMP, the increase of MLSS concentration decreased the magnitude of the permeate flux increase caused by an increase in the cross-flow velocity. The decrease of permeate flux can be attributed to the higher rate o f mass transfer towards the membrane and/or to the increase in the viscosity of the mixed liquor that occurs at higher MLSS (Liibbecke et al., 1995; Beaubien et al., 1996; Mallevialle et al, 1996) Colloidal particles and soluble products Compared to large suspended solids, fine colloids have a lower back -diffusion rate so that fine colloids tend to migrate and accumulate on the membrane surface to a greater extent than larger suspended solids. In addition, the smaller the particle size, the more compact will be the foulant layer formed on the membrane surface (Choo and Lee, 1996b, 1998). Therefore, fine colloids play a significant role in increasing the hydraulic resistance of a membrane filtration process. Choo and Lee (1996b) found that the polarization index at the membrane surface for the colloid s was much higher than that for the soluble material when they examined the effect of anaerobic digestion broth component on membrane permeability. They speculated that the membrane performance could be improved by degrading the colloids into soluble material or by agglomerating the colloids into coarser particles. The result of a study conducted by Choo et al. in 2000 confirmed their speculation. To control the deposition of organics and fine colloids onto the po lymeric membrane, they added powdered activated carbon (PAC) into the bioreactor, giving rise to a reduction of specific cake resistance of biosolids through the sorption and/or coagulation of dissolved and colloidal matter: However, some research studies found that the production of soluble microbial products (SMP) could increase the membrane fouling (Langenhoff et al, 2000; Defrance et al., 1999). The result of a study by Langenhoff et al. (2000) showed that the production of SMP in a conventional anaerobic bioreactor treating synthetic wastewater was higher when the colloidal material of the wastewater was higher. Defrance and Jaffrin (1999) suggested 32 that the relative contributions of suspended solids (SS), colloids, and soluble materials to the fouling resistance were 65%, 30% and 5%, respectively, in an aerobic MBR. Harada et al. (1994) also reported that the soluble component of the mixed liquor played a significant role in the formation of a foulant layer on the membrane surface in an anaerobic UF MBR. Lesjean et al. (2005) investigated the correlation between membrane fouling and soluble/colloidal organic substances in aerobic membrane bioreactors for municipal wastewater treatment, and found that a linear relationship existed between the fouling rate of the membrane and the concentration of polysaccharides (one of the extracellular polymeric substances) in the sludge. Van Houten et al. (2001) reviewed recent developments in AnMBR technology and suggested that membrane fouling in anaerobic systems and aerobic systems are governed by different mechanisms since anaerobic mixed liquor contains more fine colloids than aerobic mixed liquor. However, no specific data or results were presented. The impact of SMP concentration on the membrane fouling in aerob ic MBR systems has been investigated extensively (Fawehinmi et ai, 2004; Lee et al, 2001a; Chang and Lee, 1998; Wisniewski and Grasmick, 1998). However, there is little information about the impact of SMP on the membrane performance in an AnMBR. In a sub merged AnMBR study conducted by Stuckey (2003), the type of SMPs produced in the reactor was different from that of SMPs in the permeate. This suggests that the membrane can retain some ofthe soluble microbial products. Since some of the soluble microbial products are retained, they are likely to contribute to the formation of a foulant layer on the membrane surface. 2.5 Summary and Objectives of the Research Although anaerobic MBR technology has gained extensive attention and research has been conducted to examine the system treatment efficiency and investigate the m embrane fouling mechanisms, the research and application of AnMBR for low-to-medium strength 33 municipal wastewater treatment are very limited. The potential of AnMBR treating low-to-medium strength municipal wastewater as an alternative to conventional approaches has been demonstrated, however, a significant amount of research is still indispensable in order to assess the technical and economic feasibility of this technology for the treatment of lower strength wastewaters at lower temperatures. From the literature reviewed and compiled, critical parameters to the acquisition of high biological and membrane performance are identified to be examined in the experimental process. 2.5.1 Parameters critical to the system operation control Organic loading rate for the external membrane AnMBR The bioreactor COD OLR will dictate the treatment performance and capacity, which are commonly used to evaluate the technical and economic feasibility of a system. The OLR of 5 to 8 kg COD/m3«day is anticipated as an operational target. However, the incremental OLR should be in a step-wise manner so that the seed sludge can acclimatize itself to the new operational conditions. System operating temperature The information obtained from the literature clearly indicates that the system operating temperature and pH will significantly affect the system performan ce. The technical and economic feasibility of applying AnMBR for treatment of relatively low strength wastewater is likely to depend on the necessity of achieving a high leve 1 of performance at lower temperatures. The target operational temperature should be an ambient temperature around 20 °C. Suspended solids concentration and particle size distribution The mixed liquor suspended solids concentration possibly has an adverse effect on the membrane permeate flux, however, maintaining a high MLSS concentration is critical to an anaerobic treatment process, particularly under low temperature operation conditions. 34 The colloidal and soluble components of mixed liquor are critical to the permeability of membrane modules. Therefore, the effect of MLSS concentration on the system performance and the particle size distribution need to be investigated. 2.5.2 Parameters critical for the system performance evaluation Effluent COD and COD removal efficiency The effluent COD and COD removal efficiency are the most important parameters to evaluate the feasibility of AnMBR treating municipal wastewater. A successful AnMBR system should demonstrate a low effluent COD and a high COD removal efficiency. Gas production and composition The economic feasibility of AnMBR system is likely to depend on the energy recover during the treatment process. Membrane permeate flux Membrane permeate flux limits the AnMBR system treatment capacity which is one of the factors determining the economic feasibility of the system. Therefore, the factors affecting membrane permeability should be closely examined in the experimental process. These factors include membrane material and pore size, cross -flow velocity, trans-membrane pressure, gas injection, suspended solids concentration, and membrane chemical cleaning method. 2.5.3 Objectives The information obtained in this laboratory scale study will help develop the design and operation guidelines for a pilot scale treatment system, which can furthe r provide design and operation information for system scale-up to the commercial scale. Therefore, the following aspects will be assessed in this study. 35 1. The effect of selected AnMBR operating and design parameters on the AnMBR system biological performance. 2. The effect of selected AnMBR operating and design parameters on the AnMBR system membrane component performance . 3. The membrane fouling mechanism(s) in an AnMBR treating municipal wastewater and the options for reducing the membrane fouling and maximizing the membrane permeate flux. 36 C H A P T E R 3 M E T H O D S A N D M A T E R I A L S This chapter deals with the methods and materials that were incorporated in this research work. All the experiments were carried out at the University of British Columbia's pilot plant, located at the south end of the campus, and the Environmental Engineering Laboratory of the Civil Engineering Department. The pilot plant gets its sewage from the north-south trunk sewer that conveys wastewater generated at UBC and its associated housing development. Currently, a timer-operated trash pump in the sump lifts wastewater into two large storage tanks at ground level four times per day (3 AM, 9 AM, 3 PM, and 9 PM). These two tanks serve as a reservoir which feeds the pilot plant. The raw sewage was buffered daily by the addition of approximately 500 mg/L of sodium bicarbonate to compensate for the low alkalinity (approximately 100 mg/L) in the Vancouver area (to bring the pH up to around 7). All the effluent and discharges from the experimental treatment systems are returned to the main sew er. 3.1 System Design The design of the anaerobic external membrane bioreactor system was based on knowledge gained from the UBC pilot plant and the membrane manufacturers. In this project, effluent from the UBC pilot plant primary clarifier was used as AnMBR influent. During the acclimatization period, the primary effluent was supplemented with sodium acetate. The characteristics of the primary clarifier effluent are shown in Table 3.1. Three commercial membranes were used for this project. The membrane information provided by the manufacturers is presented in Table 3.2. Conceptual and preliminary designs of the configuration had to be made first by following the existing design criteria of the up-flow anaerobic sludge blanket (UASB) process and theoretical conceptual ideas. Modification s to the system were made during the operation. The schematic diagram of the system and actual experimental setup are shown in Figure 3.1, Figure 3.2 and Figure 3.3, respectively. 37 Table 3.1 Characteristics of influent Range Average Total COD (mg/L) 200 - 480 310 ± 10 Soluble COD (mg/L) 50 - 293 130 ± 10 TSS (mg/L) 44 - 148 86 ± 6 Total P (mg/L) 2.82-5.86 4.2 ±0.1 Ortho-P (mg/L) 1.40-3.82 2.8 ±0.2 TKN (mg/L) 25.9-44:2 32 ±1 NH4-N (mg/L) 20.8 - 36.1 29 ± 1 NO3-N (mg/L) 0.035 -0.309 0.09 ± 0.02 pH 6.6-7.6 7.2 ±0.1 Alkalinity (mg/L as CaC03) 175 -325 250 ± 34 Temperature (°C) 16.8-23.1 20.3 ±0.3 The main reactor, constructed of transparent polymethyl methacrylate to allow visual inspection of the mixed liquor inside, was a 120 cm tall column with 20 cm diameter. Two external piping loops were installed to accommodate the inorganic membrane (Membralox) and the organic membrane (PCI or Koch), respectively. Two Moyno 500 pumps were used to pump the mixed liquor through the membrane modules with the desired cross flow velocity (CFV). A 1 to 100 RPM Masterflex pump, with one standard Masterflex pump head was used as feed pump that was controlled by a timer. Liquid level control Three liquid level sensors were used as part of the system automatic control measures. The low liquid level sensor was used for the Moyno pumps' emergency shut-down. In the event that the reactor liquid level decreased to the low limit, the low level sensor shut down the Moyno pumps to avoid pump burn-out. The high liquid level sensor was used for feed pump control in the event of effluent system failure, in which case liquid level could 38 increase to fill the head space of the reactor, causing the mixed liquor to spill out . Once the reactor liquid level increased to the high limit, the high level sensor would shut down the feed pump. The middle level sensor controlled solenoid valves on the permeate lines and thereby controlled permeate recirculation, by which the desired reactor liquid volume was maintained. 39 Table 3.1 Nominal design information for membrane modules PCI Membralox Koch Configuration External, tubular Material PVDF co-polymer Ceramic PVDF co-polymer Nature Hydrophilic and negatively charged Overall dimension Long 1200 mm Diameter 50 mm Long 300 mm Diameter 25 mm Long 1100 mm Diameter 25 mm Surface area (m2) 0.06 0.0062 0.028 Module number 1 2 4 Pore size (um) <0.1 0.05 0.005-0.01 Permeate flux (Lmh) 100 161 Cross-flow velocity (m/s) 2.3-3.5 2.7-3.4 TMP (kPa) Max. inlet pressure 1000 200 Max. inlet pressure 620 Operation Continuous operation under constant pressure S c r e e n T e r n p e ra tu re s e n s o r © F e e d p u m p L e v e l s e n s o r s a n d c o n t r o l l e r G a s c o u n t e r 0 f f - g a s G a s s a m p l i n g 5 p H c o n t r o l l e r a n d b u f f e r t a n k M e m b ra n e '—&-W a s t e / d r a i n p o r t T S o l e n o i d v a l v e s P e r m e a t e t a n k s Figure 3-1 Schematic diagram of anaerobic external membrane bioreactor system. Figure 3-2 Anaerobic external membrane bioreactor system - Reactor. 42 Figure 3-2 Anaerobic external membrane bioreactor system - Membrane components. Temperature control To maintain a high temperature during the initial acclimatization phase, a Visi-therm Deluxe submersible aquarium heater 'was installed inside the reactor to heat the mixed liqu or. The reactor was wrapped with insulation to reduce heat loss. The heater was controlled by a thermostat and a temperature sensor. The thermostat was initially set to a temperature of 32 °C, almost the highest temperature the heater could operate at. The insulation jacket was removed on January 13 t h, 2005 due to high operational temperature. Instead, a copper coil with cooling water was put around the reactor to decrease the operational temperature. On March 25 t h, 2005, a stainless steel cooling coil was installed within the reactor instead of the aquarium heater to further decre ase the temperature so that reactor temperature was close to the project target of 25 °C. The cooling fluid used was tap water with an average temperature of 15 °C, and coolant circulated only when activated by the temperature controller. The process of temperature reduction was gradually implemented to avoid rapid changes in temperature as the anaerobic bacteria are sensitive to temperature fluctuations. pH control A Cole-Parmer series 7142 pH control/pump system was initially installed for pH control. A pH probe located on the mixed liquor recirculation line was used for communication between the reactor and the Cole-Parmer system. However, the pH control system was not used in this project because the raw sewage was buffered adequately with sodium bicarbonate. 3.2 Experimental Program The sampling and analysis schedule shown in Table 3.3 was developed to obtain a consistent evaluation of the performance of the external membrane MBR system. This schedule was maintained from the beginning of September, 2004 or near the start-up of the external membrane reactor system, until June 15, 2005 the end of the project. All onsite readings (i.e. 44 temperature, pH, gas count, and permeate flux) were usually taken in the morning and only once per day. Samples taken to the lab for analysis were transported in a timely fashion (usually within 20 minutes) or preserved in accordance with Standard Methods (Greenberg et al, 1998). The results from the monitoring program are presented in the following chapter. Table 3.3 Monitoring program to track ongoing process acclimation and performance Mon. Tues. Wed. Thur. Fri. Sat. Sun. Temperature X X X X X X X pH X X X X X X X Gas Count X X X X X X X Jv X X X X X X X COD X X X VFA X X X Gas X X X Composition X X X TSS/VSS X 3.2.1 Monitoring variables The following variables were monitored throughout the experimental period. Temperature The temperature in the reactor was measured with a digital thermometer with the probe inside the reactor. The thermometer was calibrated against a mercury thermometer, accurate to 1 °C. The readings were taken every day according to the monitoring program. pH The pH values of the influent and effluent were measured daily with a Fisher Scientific Accument pH meter (model 25). A three point calibration was done with buffers of pH 4, 7, and 10. The sample was constantly stirred with the pH probe when the reading was being taken. 45 Gas production The produced biogas was counted by a Gas Flow Totalizer (patent number: 04064750). The Gas Flow Totalizer was calibrated by injecting air with a syringe. The volume of injected air that trigged one count was recorded. The biogas volume was calculated by multiplying the count number by the volume per count. Trans-membrane pressure (TMP) Trans-membrane pressure was indicated from pressure gauges installed in the mixed liqu or recirculation lines and was controlled manually by a globe valve located on the mixed liquor recirculation lines downstream of the membrane modules. Permeate flux Permeate through the membranes was collected at the sample ports on the permeate lines over a defined period of time. The volumes of the collected permeate were then measured by a graduated cylinder. The unit mL/min of permeate flux was converted to Lmh by taking the membrane surface area into account. 3.2.2 Experimental variables Sample collection and preparation Grab samples for analyses were gathered from three locations. 1. Influent samples were collected from the sample port of the primary clarifier. 2. Mixed liquor samples were collected from the sample port on the mixed liquor recycle line. 3. Effluent samples were collected from the sample ports on the permeate lines. Collected samples were prepared for analysis within 20 minutes (Greenberg et ai, 1998). The influent was filtered through a 45 mm diameter 0.45 pm HA filter paper using a syringe. 46 When the influent was filtered, care was taken to ensure that there was no tear in the filter paper during filtration. In case a tear occurred (which happened a few times), the used filter paper was examined carefully after it was taken out from the filtering unit. The filtrate was used to measure the soluble chemical oxygen demand (soluble-COD), volatile fatty acids (VFAs), and nutrients. The permeate of effluent samples were analysed directly without filtration since the pore size of membranes was smaller than 0.45 pm. Samples were preserved according to Standard Methods (Greenberg et al, 1998). Total and soluble chemical oxygen demand (COD and soluble -COD) The closed reflux colorimetric method was used as outlined in Standard Method 5220D (Greenberg et al, 1998). COD standards of 50 mg/L, 100 mg/L, 200 mg/L, and 400 mg/L were used as calibration standards. The samples, blanks, and calibration standards were then digested at approximately 150 °C for about two hours on a HACH block digester. After cooling, a calibration curve was determined on a HACH DR2000 spectrophotometer by measuring the absorbance ofthe blanks and standards at 600 nm. Then the absorbance of the samples were measured and compared to the regression equation for the calibration curve. Volatile fatty acids (VFAs) About 1 mL aliquots of the prepared samples were dispensed into 2 mL clear glass GC vials (HP model 5181-3375) and preserved using 0.1 mL, 2% phosphoric acid (H3PO4). After this, the vials were stacked in properly labeled racks and refrig erated at 4 °C. The VFA analyses were done in the .Environmental Engineering Laboratory of the Department of Civil Engineering. The method used was described in the gas chromatograph (GC) manual as Supelco Bulletin 751. The samples were analyzed by a Hewlett Packard 5880A series gas chromatograph along with a Hewlett Packard 7672A programmable auto sampler. The specifications and settings of the instrument are shown in tabular form in Table 3.4 47 Table 3.4 Specifications of the gas chromatograph for analysis of volatile fatty acids Instrument Gas Chromatograph Model number Hewlett-Packard 5880-A Automated sampler Hewlett-Packard 7672-A Oven temperature 120 °C Injection port temperature 180 °C Detector temperature 200 °C Detector type Flame Ionization Detector (FID) Carrier gas Helium Gas flow rate 20 mL/min Column material Glass Column dimension 1.2 m length, 2 mm diameter. Column packing 60/80 CarbopackC/0.3% Carbowax 20M/0.1% H 3PO4 (Source from Gas Chromatograph manual) All samples were analyzed for acetic acid, propionic acid, iso -butyric acid, and butyric acid. From previous work, it was also found that volatilization of VFA can occur even at very low temperature (-2 °C). Therefore, the time period between sampling and analysis was kept to a maximum of 7 days. Total suspended solids (TSS) and volatile suspended solids (VSS) The samples were filtered through pre-dried Whatman 934-AH 1.5 pm glass microfibre filters by using a stainless steel filtration apparatu s and a vacuum pump. The filter papers were then dried at 103 to 105 °C for at least two hours and then allowed to cool inside a desiccator. The TSS concentration was calculated by evaluating the weight contribution by the residue and by taking into account the volume used during filtration, as explained by Method Number 2540D (Greenberg et al, 1998). The filters holding the suspended solids were fired in Lindberg 51800 series Moldafherm Box Furnace at 550 °C for about two hours. The loss of mass during ignition was used to 48 determine VSS. Modification for TSS/VSS tests Due to the small particle size caused by shear imposed on the mixed liquid by the Moyno pumps, the standard method for the measurement of TSS and VSS was no t suitable for this project. A modification was developed for the standard method at the end of January 2005. A thermo IEC Multi (RT) Series multipurpose centrifuge was used to separate the supernatant and suspended solids by centrifuging the mixed liqu or at 20,000 RCF for 30 minutes. Any solids remaining in the supernatant were defined as soluble. The supernatant was extracted and dried for dissolved solids concentration. The suspended solids concentration was obtained by subtracting the soluble solids concentration from the total solids concentration which was determined by weighing the residue remaining after the mixed liquor was dried at 103 to 105 °C. Sludge particle size distribution Sludge particle size distribution was measured by a Malvern Mastersizer 2000 particle analyzer and Hydro 2000S sample dispersion accessory. A thoroughly mixed sample was fed into the mixer/injector unit located on Hydro 2000S sample dispersion accessory, where the mixed liquor sample was diluted with de -ionized water and injected into the analyzer. The analyzer analyzed the sample for three times and then showed the results with an average in a graphical form. Gas composition The biogas produced by the system was collected with a glass gas collector installed on the off-gas line, and the gas collector was then brought to the laboratory for gas composition measurement. A gas sample was taken from the gas collector using a 1 -mL Hamilton syringe with Chaney adapter and a 2.5-cm needle. The injection syringe was flushed twice before the sample was injected into the Fisher Hamilton Model 29 gas partitioner. Helium was used as the carrier gas through two chromatographic columns, pack ed with a liquid phase coated on a 49 solid support known as DEHS and 42 -60 mesh molecular sieve for column 1 and column 2, respectively. A thermal conductivity detector then sensed the differences in conductivity of the separated components, which were amplified and integrated for quantification. Calculation of the gas concentration was made by comparing the peak areas of measurable samples against those of standards, expressed in volume percentage. 3.3 Bioreactor Inoculation and Acclimatization It has long been recognized that the acclimatization process has a great impact on the success or failure of anaerobic bioreactor system operation. In order to achieve the desired level system performance, it is extremely important to understand the conceptual development of the process and to acclimatize the system properly. The process of acclimatization of an anaerobic culture started a t the beginning of August, 2004. One of the main points stressed was the need for the reactor to be inoculated with a high quality methanogenic sludge. Initially, anaerobic sludge from a high rate UASB wastewater treatment system was used for inoculation. The seed sludge was obtained from the Quesnel River Pulp Mill in Quesnel, British Columbia. Aerobic mixed liquor was also obtained from the UBC pilot plant's enhance d biological phosphorus removal system. The starting seed used was a combination of 19 L aerobic sludge and 1 L anaerobic sludge. The mixed sludge concentration was 3.65 g/L. The aerobic sludge was left idle initially for dissolved oxygen removal prior to the anaerobic sludge addition. The acclimatization process utilized a fed batch reactor operating mode during the first 12 days of operation. During that time, 6 L of sewage was added every day while the same volume of permeate from the membrane was removed to keep a constant water level and a reactor liquid volume of 20 L. One additional litre of anaerobic seed sludge from the pulp mill was added to the reactor every four days. 50 The reactor was eventually switched to continuous operation after the first 12 days of fed batch operation. At that point, the rate of feed was determined by the perme ate flux rate of the membranes, or about 120 L/day. With an average influent COD of 3 00 mg/L, the organic loading rate of the system was about 1.8 kg COD/m3d. During the entire acclimatization period, the system was operated at a temperature between 32 -36 °C with a pH value of about 6.5. After 20 days of acclimatization, no gas production or COD removal was detected. The reactor was then reseeded with screened anaerobic digester sludge from a local wastewater treatment plant on Day 33. The total solids concentration in the digested sludge was 1.82%, and the volatile solids fraction was 73.6%. The average temperature of the sludge was 37.5 °C, and an average pH of 7.2. To maintain robust biogas production, solid sodium acetate mixed with 1 L of wastewater was added into the reactor every day, to give a concentration of sodium acetate in the reactor of 300 mg/L. The fed batch mode was used again for the first 14 days. The reactor initially accommodated seed sludge 28.26 L, and the water level of the reactor was at 90 cm, leaving some head space for biogas and shower head used for preventing foam. The initial seed sludge volume was used as large as possible i n order to avoid frequent exposure of the anaerobic biomass to a high shear environment . Also the maximum sludge volume can allow the anaerobic seed material (i.e. solids from a conventional municipal sludge digester) added at start -up to gradually adapt to new operating conditions. Semi-continuous influent feeding was started on September 10th, 2004, which was designated as "Day 1" of operation. The value of pH and temperature were controlled strictly thereafter. Volumetric organic loading rate (OLR) was used as the main controlling factor for system operation. 51 3.4 M e m b r a n e Cleaning Membrane chemical cleaning is an effective technique for membrane flux recovery. Membrane cleanings were carried out when the permeate flux could no longe accommodate the desired feed rate, or when mechanical failure occurred leaving mixed liquor idle inside the membrane. Different membrane cleaning procedures were explored to determine the optimal cleaning method for the external membrane operated under the anaerobic condition. Membrane cleaning required the complete shutdown of the reactor, often for a few hours, depending on the type of cleaning carried out. Table 3.5 provides an overview of the different types of membrane cleaning procedures used during this study along with their type designation. 52 Table 3.5 Membrane cleaning procedures Cleaning type Procedure details A 1. Water rinse. 2. Sodium hydroxide at pH 10 - 10.5 with 15 min permeation off and 10 min permeation on. 3. 200 mg/L sodium hypochlorite at pH 10 to 10.5 with 20 min permeation. 4. Water rinse until permeate pH neutral. 5. Nitric acid at pH 1.5 - 2 with 15 min permeation off and 10 min permeation on. 6. Repeat step 4. Al Follow the first 4 steps of Procedure A A2 Follow the first 4 steps of Procedure A and change sodium hypochlorite concentration to 1000 mg/L. B 1. Water Rinse. 2. 200 - 300 mg/L sodium hypochlorite at pH 10-10.5 with 10 min permeation. 3. 2% sodium hydroxide with 30 min permeation off and 30 min permeation on. 4. Water rinse until permeate pH neutral. 5. 2% nitric acid with 20 min permeation off and 20 min permeation on. 6. Water rinse until permeate pH neutral. C 1. Water rinse for 10 min. 2. 300 mg/L sodium hypochlorite with 30 min permeation. 3. Water rinse. Cleanings were usually performed at a liquid temperature of 40 - 50 C. 53 CHAPTER 4 RESULTS AND DISCUSSIONS The results of the experiments perf ormed in this study are discussed in this chapter. Three topics are addressed: system operation conditions and control, biological treatment performance, and membrane component performance. The raw data obtained from laboratory measurements were screened prior to statistical analysis. Q-testing was performed on all data sets to determine whether a discordant point could be discarded with 90% confidence. The data presented herein are mainly shown graphically. In cases where the average values of the measurements are used, confidence intervals at a 95% confidence level are shown with error bars. 4.1 General System Operation Conditions and Control 4.1.1 Hydraulic retention time and volumetric organic loading rate As mentioned in Chapter 3, OLR was used as the main controlling factor for the system operation. The volumetric organic loading rate (OLR) was increased in a stepwise manner. The increase of OLR was implemented by an increase of influent flow rate and/or a decrease of reactor liquid volume. However, due to the fluctuation of influent COD concentration, the OLR could not be held constant, but fluctuated around an average value in the different operational periods. The OLR values were calculated from grab samples taken for the evaluation of total influent COD. During the period of acclimatization (from Day 1 to Day 214), the influent domestic sewage was augmented with 300 mg/L of sodium acetate (which has an approximate COD of 240 mg/L). The reactor was then fed intermittently with a timer-controlled pump, to dispense the desired influent volume each day. The subsequent removal of Sodium acetate supplementation of the feed was terminated on day 224. By controlling the reactor liquid level and the influent flow rate, hydraulic retention time (HRT) could be maintained at a specific value. Table 4.1 presents the HRT and corresponding OLR during the whole operational period. 54 Table 4.1 System operation conditions - HRT and OLR Phase Duration (day) HRT (hours) OLR (kg COD/m3«day) Sodium acetate addition 1 1 - 17 61 0.22 ± 0.04 Y 2 18-41 24 0.55 ± 0.03 Y 3 42-91 15 0.79 ±0.05 Y 4 92-106 13 0.96 ±0.13 Y 5 107-144 8 1.51 ±0.08 Y 6 145-183 10 1.16 ±0.07 Y 7 184-214 8 1.53 ±0.11 Y 8 215-246 4 1.68 ±0.12 N 9 247-281 2 2.84 ±0.11 N For the convenience of discussion, the whole experimental program was divided into 9 phases based on different HRTs. In Phase 6 (from Day 145 to Day 183), the OLR was limited by the fdtration capacity ofthe membranes. The initial HRT of 8 hours in the system was subsequently increased to 9, 10 and then 11 hours, soon after the flux of the PCI membrane decreased to zero due to irreversible fouling, causing the OLR to decrease. However, soon after the installation of the Koch membranes on Day 184, the system regained sufficient filtration capacity. Thereafter, the HRT in Phase 7 was decreased to 8 hours and the OLR was increased to approximately the same value as in Phase 5. After removal of sodium acetate from t he influent, the system was run at very short HRTs (Phases 8 and 9) and the OLR was close to the target OLR of 3 kg COD/m3«day. 4.1.2 Temperature Since methanogenesis is strongly temperature-dependent, the temperature during the start-up phase is critical for anaerobic sludge acclimatization. Although the objective of this study was to operate the anaerobic MBR at a relatively low temperature successfully, the 55 process. The temperature was decreased only after the anaerobic sludge from a municipal wastewater treatment plant digester was completely adapted to the new feed and exhibited a stable biological performance. The profile of the operational temperature is presented in Figure 4.1. The system was maintained at a higher temperature at the beginning. During the first 20 days, the temperature was maintained at an average of 30 °C. In the next three month period, the temperature of the system was stabilized between 30 °C to 35 °C. This slight increase of temperature was due to the slight increase of mixed liquor recirculation pump speed, which introduced more heat to the system. The temperature in the reactor was above 35 UC between Day 102 to Day 139, even above 40 °C on some days. This was due to the substantial increase in pump speed required to meet the cross -flow velocity requirements of the membranes. Another reason was the increase in the ambient temperature to around 30 °C due to the operation of a heater inside the pilot plant trailer. Several measures were taken to decrease the reactor temperature as discussed in Chapter 3 (Section 3.1). In addition, the heater for the trailer was turned off in order to decrease the ambient temperature. As a result, the AnMBR bioreactor temperature was gradually decreased to around 30 °C. On Day 197, a stainless steel coil was installed inside the reactor to circulate cooling water. The reactor temperature was gradually decreased to 25 °C and the operation was stabilized at this temperature (except on some days on which cooling system problems occured). 56 45 20 -I , , , , , 1 0 50 100 150 200 250 300 Days Figure 4-1 Anaerobic external membrane bioreactor operating temperature. 4.1.3 pH As discussed in Chapter 2, anaerobic bacteria are very pH-sensitive. The optimum pH for methanogens falls in the range of 6.5 - 7.5 (Tchobanoglous et al., 2003). The pH during the experimental program is shown in Figure 4.2. Greater attention was given to the effluent pH since it was more representative of the mixed liquor pH inside the reactor. Therefore, the influent pH was not measured until Day 78. The municipal wastewater was buffered by the addition of sodium bicarbonate because the feed used in this study was shared with another ongoing project at the UBC pilot plant which required bicarbonate buffering. The average influent pH of 7.6 (±0.1) during the acclimatization period (Day 1 to Day 214) was higher than the average value 7.2 ( ±0.1) after the removal of sodium acetate. This was caused by the addition of sodium acetate to the influent used in the study which was obtained from the UBC pilot plant primary clarifier. Since sodium acetate is composed of strong-base cation and weak-acid radical, it can contribute to the higher pH of in fluent. 57 8.5 8.0 • •Acetate addition terminated X 6.5 • o <SJ>3030O s o « e o o o o o o o o o o o • < » o o >-o o o o-o-o o 6.0 0 50 100 150 200 250 300 Days • InfpH o Eff pH Figure 4-2 pH values in influent and effluent samples. During the first 20 days of the start-up period, the effluent pH gradually decreased from 7.5 to 6.9. Anaerobic sludge from a well-established digester is highly buffered in order to keep the alkalinity concentrations in the range from 2000 to 4000 mg/L as CaCO 3 required to maintain the pH at or near neutral (Malina, 1992; Tchobanoglous et al, 2003). The influent alkalinity was 250 ± 34 mg/L as CaC03 in this experiment. During the first period of operation, the digester sludge alkalinity was gradually washed out of AnMBR by the wastewater feed, leading to a decrease in system pH. However, later the pH leveled off around 7.0 and did not seem to drop any further. After about two months of stable operation, the effluent pH was stabilized at 7.0 to 7.5 until the removal of the acetate supplementation. The average pH of the effluent during the acclimatization period was 7.2. Figure 4.2 illustrates the decrease in the pH of the reactor following the termination of sodium acetate supplementation of the influent. After this point, the effluent pH was in the range of 6.4 to 7.0 with an average of 6.7. The decrease in effluent pH may have been caused by several factors. The first was the decrease in influent pH. Another factor may 58 have been the increase in OLR, which resulted in the accumulation of VFAs, lowering the effluent pH. 4.2 Biological Performance of Anaerobic M B R 4.2.1 Chemical oxygen demand (COD) The measured concentrations of COD in the AnMBR influent and effluent are shown in Figure 4.3. The effluent COD shown here is the average value of the permeates from the three membranes used in the system. 800 700 600 S 500 t 400 300 O 1 u 200 100 1 • • • % • Cfj Acetate adclition te rmina ted - c P -50 100 150 Days 200 250 300 A E f f . C O D • Inf. T o t a l C O D • Inf. So lub le C O D Figure 4-3 COD concentrations in influent and effluent samples. From Day 1 to Day 214, (acclimatization period), when sodium acetate was added to the influent, the total COD ranged from 320 to 730 mg/L in the influent and from 45 to 170 mg/L in the effluent. The total COD removal eff iciency was above 80%. From Day 215 to Day 281, acetate supplementation was omitted from the feed and the influent used was composed only of pre-settled municipal wastewater. In this period, the total COD ranged 59 from 230 to 380 mg/L in the influent and from 50 to 180 mg/L in the effluent. The COD removal efficiency decreased to slightly higher than 60%. Table 4.2 lists the average COD concentrations in the influent and the effluent and removal efficiencies in these two periods. Table 4.2 Average COD concentrations in the influ ent and the effluent and removal efficiencies in different operational periods Period Influent COD (mg/L) Effluent COD (mg/L) COD removal (%) Acclimatization 510 ± 13 84 ±5 84 ± 4 Steady-state operation 290 ± 18 110 ± 14 60 ± 14 The reactor start-up period was very short, and within 12 days the average effluent COD was less than 100 mg/L. Quick start-up was also noted by Manariotis and Grigoropoulos (2002), who reported that an anaerobic baffled reactor treating low -strength wastewater at a HRT of 24 hours and temperature of 26 °C achieved an effluent COD of less than 100 mg/L within 15 days from start-up. The influence of OLR on COD removal efficiency COD removal efficiency (the detailed calculation is in Appendix A) as a function of OLR is shown in Figure 4.4. The OLR was increased from 0.22 (±0.04) to 2.84 (±0.11) kg COD/m3-day during the whole study period. Within the OLR range from 0.22 (±0.04) to 1.53 (±0.11) kg COD/m3»day, the COD removal efficiency was in between 78 (±18) % and 85 (±20) %, relatively independent of the operational OLR. In this operational period, the influent was supplemented sodium acetate, which contributes soluble COD in the influent. 60 110 ~ 100 40 I 1 ' ' 1 1 : 1 0.00 0.50 1.00 1.50 2.00 2.50 3.00 OLR ( kg COD/m3«day) Figure 4-4 COD removal efficiency as a function of OLR. This indicates that the reactor had not reached its loading capacity in the early part of the study and shows the potential capacity of anaerobic MBR treating municipal wastewater. This is quite different with the observations from other anaerobic bioreactor configurations. In the experiment conducted by Yu and Anderson (1996) with a combined reactor of UASB, hybrid UASB-AF and ABR, although within the OLR range from 0.92 to 1.42 kg COD/m3«day, the COD removal efficiency was in a similar range from 83% to 73%, it decreased linearly with an increase of OLR. An adverse proportional relationship between COD removal efficiency and organic loading rate was also observed by Converti et al. (1992) in a fluidized bed reactor treating semi-synthetic municipal wastewater, although the OLR in the range from 5 to 25 kg COD/m3*day was much higher than the OLR used in the study by Yu and Anderson (1996). After the supplementation of sodium acetate was terminated in the present study, when OLR was increased from 1.68 (±0.12) kg COD/m3«day to 2.8 (±0.11) kg COD/m3«day, COD removal efficiency decreased from 69 (±15) % to 54 (±10) %. Similar to the result obtained by Converti et al. (1992), the COD removal efficiency was adverse proportional to the organic loading rate. 61 Since an OLR of up to 1.53 (±0.11) kg COD/m -day could be implemented without compromising the system biological treatment performance, the suggested highest OLR for an anaerobic external MBR operating at a temperature of 25 °C is 1.6 kg COD/m3,day. The influence of HRT on COD removal The effect of HRT on COD removal efficiency is shown in Figure 4.5. 70 60 50 40 30 HRT (hours) 20 10 Figure 4-5 COD removal efficiency as a function of HRTs. It is interesting to note that, based on average values, a t applied HRTs between 8 hours and 24 hours, there was essentially no variation of COD removal efficiency. The COD removal efficiency was relatively independent of HRT. This is not unusual in an anaerobic MBR. A similar relationship between the COD removal efficiency and HRT has been reported by Chu et al. (2005). They showed that when treating municipal wastewater at temperatures higher than 15 °C and in the HRT range of 3.5 - 5.7 hours, COD removal efficiency was relatively independent of HRT. However, this is uncommon with other anaerobic reactor studies, in which HRT always had an impact on the COD removal. Converti et al. (1992), using a fluidized bed anaerobic reactor treating enriched municipal wastewater, showed that 62 COD removal efficiency increased more than 10% when the HRT was increased from 8 hours to 24 hours. This difference between an anaerobic MBR and the other anaerobic bioreactor configurations is probably due to the capacity of the AnMBR to retain colloidal COD. At an initial start-up HRT of 61 hours, the COD removal efficiency was slightly lower than that observed at an HRT between 24 hours and 8 hours. This is mainly attributed to COD washed out of the seed sludge, which dominated the effluent COD during the start -up period. Therefore, the COD removal efficiency at an HRT of 61 hours is not really relevant. After the termination of sodium acetate addition in influent, the d ecrease of HRT impaired the COD removal efficiency. When the HRT was decreased from 4 hours to 2 hours, the COD removal efficiency was decreased from 69 (±15) % to 54 (±10) % . This impact on COD removal efficiency imposed by the change of HRT is slightly higher than those reported from other types of anaerobic reactor. Elmitwalli et al. (2002) reported that the COD removal efficiency was improved by 12% from 58.6 (±7.7) % to 70.6 (±7.4) % when the HRT was increased from 2+4 hours to 4+8 hours in a two-step anaerobic filter/anaerobic hybrid system treating municipal wastewater. Yu and Anderson (1996) also reported a moderate effect of HRT change on COD removal efficiency. Their results showed that the COD removal efficiency increased by 10% from 68% to 78% with an increase of HRT from 4 hours to 8 hours in a combined anaerobic reactor (UASB, fixed film and hybrid UASB-AF) treating municipal wastewater. It should be noted that in the present study even at an HRT of 4 hours, the reactor was still able to remove 69 (±15) % of the influent COD. The high COD removal efficiency at relatively low HRT may be due to the retention of particulate and colloidal COD in the system due to the membrane separation. Concurrent with the decrease of HRT from 8 hours to 2 hours was a decrease of soluble COD to total COD ratio in the influent. The soluble influent COD accounted for 67 ( ±2) % of the total influent COD in the acclimatization period, and this value decreased to 46 (±6) 63 % without acetate supplementation. Since hydrolysis of soluble acetate is not required, much faster conversion rates to methane can be obtained (Harper and Pohland, 1986). Although with membrane filtration, the influent suspended solids will be completely retained in the reactor, adequate solids retention time is required to mineralize the majority of the particulates completely. Unfortunately, due to the termination of acetate addition, the influent flow was significantly increased in order to keep the OLR at a previous level or even higher, leading to a rapid accumulation of particulate COD inside the reactor and a decrease of HRT. At a short HRT, once the particulates were hydrolyzed to soluble high molecular weight organics, these intermediate products would be washed out with the effluent. Therefore, the decrease in soluble COD to total COD ratio of the influent may be one of contributors to the significant decrease in COD removal efficiency from 84 (±4) % in the acclimatization period to 60 (±14) % in the steady-state operation period. The results from the present study indicate that the optimal HRT for the anaerobic external membrane MBR was 8 hours, giving the system the greatest treatment capacity without impairment of the treatment efficiency. The influence of temperature on COD removal efficiency It has long been recognized that temperature strongly affects the rates of the anaerobic conversion processes, so the quality of bioreactor effluent fluctuates with changes in temperatures (Yu and Anderson, 1996; Ndon and Dague, 1997; Bodik et ai, 2002b). Although a gradual decrease in temperature does not affect the COD removal efficiency significantly, a greater HRT will be needed. However, from the result shown in Figure 4.6, we can see that the COD removal efficiency was independent of the operational temperature at HRTs of 8, 15 and 24 hours. 64 1 0 0 9 5 § 7 0 U 6 5 6 0 I 1 1 1 1 2 0 . 0 2 5 . 0 3 0 . 0 3 5 . 0 4 0 . 0 4 5 . 0 Temperature (°C) I A HRT=8 hours B H R T = 1 5 hours • H R T = 2 4 hours Figure 4-6 The effect of temperature on the COD removal efficiency at different HRTs. In all the references found so far (Chu et al, 2005; Sutton et al, 2004; Wen et al, 1999) for municipal wastewater treatment using an anaerobic MBR, only Wen et al (1999) and Chu et al, (2005) reported an effect of temperature on the COD removal efficiency. The results ofthe present study are consistent with those of Wen et al. (1999) but contrary to those of Chu et al. (2005), although both studies ran experiments at a similar temperature range (12 to 27 °C and 11 to 25 °C, respectively). Three issuses are important to the understanding of the results of the present study. 1) The strong suspended solids retaining capacity of the reactor. This is one of important factors for high-rate anaerobic wastewater treatment at relatively low temperature. 2) The addition of sodium acetate in the influent, in which the COD contributed by acetate addition accounted for 35 - 55% of the total COD. Without the requirement of hydrolysis, the removal rate of COD existing in the form of acetate is faster than that of particulate and/or colloidal COD even if the temperature decreases 17 °C (from 42 65 to 25 UC). 3) Although the operational temperature was reduced from 42 to 25 °C, the operational temperatures in the study still fall into the temperature range that is preferred to support more optimal biological reaction rates and to provide more stable treatment. This necessitates the investigation of further decreases in temperature under identical other operational conditions. 4.2.2 Volatile fatty acids (VFAs) Volatile fatty acids are intermediate products in an anaerobic treatment process. T he concentrations of acetic and propionic acids in influent and effluent are shown in Figure 4.7. 1000.0 —. s, < to, > 100.0 b IO.O o U 1.0 ^ ^ ^ ^ ^ . . — — % n • A • A • ^ • . • j* • • • a „ a . " -acetateaddition terminated - a — a - r a -?stP_-r 50 100 150 200 Time (days) 250 300 * Inf. Acetate a Eff. Acetate • Inf. propionic • Eff. propionic Figure 4-7 The concentration of acetic and propionic acids in influent and effluent. In the acclimatization period, with the addition of sodium acetate to the raw wastewater, the acetic acid concentrations in the influent and the effluent were 183 (±7) and 5 (±1) mg/L, 66 respectively. In the steady-state operational period, since the sodium acetate was removed from the feed, the acetic acid concentration in the influent decreased significantly to 15 (±1) mg/L, but the acetic acid concentration in the effluent increased to 25 (±6) mg/L, even higher than that in the influent. The influent propionic concentrations during the acclimatization period and the steady -state operational period exhibited little difference, averaging 3.6 (±0.3) mg/L and 2.1 (±0.6) mg/L, respectively. The propionic concentration in the first period was a little higher because the semi-synthetic influent was retained in a storage tank for up to 24 hours, causing some organics to decompose. However, the effluent propionic concentrations in these two periods were significantly different. In the acclimatization period, there was almost no propionic acid in the effluent. In the steay-state operational period the propionic acid concentration gradually increased with an average concentration of 6.7 (±1.3) mg/L. In effect, the methanogens and the acidogens form a mutually beneficial relationship in an anaerobic wastewater bio-treatment process (Tchobanoglous et al, 2003). If process upsets occur and the methanogenic organisms do not utilize the hydrogen produced by acidogens fast enough, the propionate and butyrate fermentation will be slowed with the accumulation of volatile fatty acids in the anaerobic reactor, resulting in a high concentration of VFAs in the effluent, which can be considered representative of the VFA concentration inside the reactor. Therefore, the treatment efficiency will be decreased in terms of COD removal. The influence of VFA concentration on the COD removal efficiency in the present study is illustrated in Figure 4.8. With an increase of VFA concentrations the COD removal efficiency decreases. The influence of VFA concentration on the COD removal efficiency is more obvious at VFA concentrations greater than 10 mg/L and this can be explained by higher percentage of propionic acid in total VFAs. 67 o U 20 10 0 0 10 20 30 40 50 60 The concentration of VFAs in effluent (mg/L) Figure 4-8 The influence of VFA concentration on COD removal efficiency. According to the individual VFA analysis results, when the total VFA concentrat ion was less than 10 mg/L, it was mainly composed of acetic acid. When the total VFA concentration was above 10 mg/L, propionic acid accounted for 21 (±3) % of the total VFA concentration. Since propionic acid has inhibitory effects on methanogenic process es at low concentrations, its concentration in an anaerobic bioreactor has been taken as one of the most important indicators of the stability of an anaerobic treatment process (Renard et al., 1991). Some researchers have monitored individual volatile fatty acids, particularly propionic acid, to direct loading adjustment s so that the startup of the anaerobic process will proceed without instability problems, and to maintain steady -state operating periods with acceptable performance (Renard et al, 1991). Monitoring total volatile acids can be implemented by instruments on-line or by semi-continuous sampling (Powell and Arches, 1989; Renard et al., 1991). Alkaline consumption for pH control can be used as an indirect measure of total VFAs (Denac et al, 1990). The influence of OLR on the concentration of VFAs One of the difficulties encountered in an anaerobic treatment process is the volatile fatty 68 acids accumulation coupled to a pH decrease as the load increases (Butcher, 1989). The concentration of VFAs in the effluent as a function of the OLR is shown in Figure 4.9. The results presented indicate that an increase in organic loading rate leads to a proportional increase in the volatile fatty acids level in the reactor as a result ofthe intrinsic slow growth rate of methanogens compared with the acidogens. E, < u. > o B O C O u 50 45 40 35 30 25 20 15 10 5 T y = 18.01 x-13.58 ¥ / R 2 = 0.86 • I 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 OLR (kgCOD/m j«day) Figure 4-9 The effluent concentration of VFAs as a function of OLR. However, compared to the anaerobic treatment process for high strength wastewater, the influence of OLR on the VFA concentration in the reactor in this study is relatively small. Converti et al (1992) reported an exponential increase in VFA concentration with an increase of OLR from 4 kg COD/m3-day to 24 kg COD/m3-day. The different observations can be explained by the different OLR range applied. In fact, from the trend of their results, we can infer that at OLR of less than 4 kg COD/m 3-day, the relationship between OLR and VFAs was similar to that found in the present study. 69 The influence of HRT on the concentration of VFAs The HRT has great influence on the accumulation of VFAs in an anaerobic reactor. A short HRT encourages the growth of acid formers and concurrently suppresses the growth of methane producers (Elefsiniotis and Oldham, 1994). This can cause poor performance of the treatment system, one of the indicators of which is an increase in the effluent total VFA concentration (Fongastitkul et al., 1994). The effluent VFA as a function of HRT in the present study is shown in Figure 4.10. 60 50 - J "3D s o 40 b 30 c < > 20 10 30 25 20 15 10 HRT (hours) Figure 4-10 The effluent concentration of VFAs as a function of HRT. From this result, we can see that with the acetate addition, although the HRT was decreased from 24 hours to 8 hours, the concentration of VFAs in effluent was not significantly affected. The effluent concentration of VFAs at an HRT greater than 10 hours was slightly lower than that at an HRT of 10 and 8 hours. When the acetate addition was terminated, however, the decrease of HRT significantly affected the concentration fof VFAs. The concentration of VFAs in effluent increased rapidly with a decrease of HRT. Although the change in feed is one important factor resulting in the current results, we can not rule out 70 that HRT of 10 hours is a critical hydraulic retention time for the anaerobic external membrane MBR. Beyond this, the intermediate VFA in the anaerobic process can be significantly converted to gaseous products. Kim et al. (2002) also reported a large effect of HRT on the production of VFAs in a membrane-coupled anaerobic VFAs fermentor system. Over the range of HRT 8 to 96 hours, the VFA concentration decreased with an increase of HRT above 12 hours, and a maximum in VFA concentration was obtained at an HRT of 12 hours, indicating favorable conditions for the growth of acid-producing bacteria. The low concentration of VFAs at longer HRTs were attributed to the conversion of soluble VFAs to gaseous products. 4.2.3 Gas production and composition Total biogas and methane production in th e entire experimental program are presented in Figure 4.11. a o u on .2 o O 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 • • Acetate addition terminated • • • m • m p-• 1 1 i i * • • 50 100 150 Days 200 250 300 Total gas • Methane Figure 4-11 Gas production in anaerobic external membrane MBR. 7! During the study, the average total biogas production and methane production decreased sharply from 0.27 m3/kg COD removed and 0.22 m3/kg COD removed in the acclimatization period to 0.07 m3/kg COD removed and 0.04 m3/kg COD removed in the steady-state operational period, respectively. These values of methane production are significantly lower than the theoretical conversion rate of 0.4 m 3 CrL/kg COD converted (Tchobanoglous et ai, 2003). Lower rates have also been reported with municipal wastewater (0.09 to 0.12 m3 CH4/kg COD removed, Yu and Anderson 1996; 0.16 to 0.21 m 3 biogas/kg COD removed, Kobayashi et al., 1983 and Lettinga et al, 1983). The big gap between the experimental results in the present study and the theoretical value can be explained as follows. First of all, the temperature used for the theoretical calculation is 35 °C, which is higher than the average temperature used in the present study. The same situation exists in the other studies, which were operated at ambient temperatures below 30 °C. In addition, in the theoretical calculation, the methane production is expressed in terms of the COD converted to methane. However, in the present study, the methane production was converted to the methane volume in terms of COD removed. The removed total COD may be greater than the COD converted to methane due to the retention of particulate COD by the membrane filtration. However, the los of soluble methane in the liquid is the main contributor for the low measured methane production when low strength wastewater is involved (Manariotis and Grigoropoulos, 2002). Singh and Viraraghavan (2003) even estimated that approximately 50% of methane produced could be lost as dissolved methane. According to the predictive model developed by Masoudi et al. (2003) for gas solubility calculations, at equilibrium conditions and atmospheric pressure, the dissolved methane concentration could be 21.7 mL/L and 24.3 mL/L at 32 °C and 25 °C, respectively. Based on methane solubility and the measured percentage content in the biogas (Figure 4.12), the biogas and methane productions were revised, as shown in Figure 4.13. The detailed calculation is in Appendix B. 72 CO 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Acetate addition terminated • a J n • • • • : • d b rf • • • • • • • • B • m " 1 n ° • • • on , • • • ° a n _ • • • • • ^A.^ j^ ftA fit A-A*— 50 100 150 Days 200 250 300 A C02 A 02 • N2 • CH4 Figure 4-12 Anaerobic external membrane MBR biogas composition. 0.60 o o U 0.50 £ 0.40 c ^ .2 "g S o 0.30 "2 E O cu c a en •a a> C/l cu 0.20 0.10 0.00 Acetate addition —• terminated • • • • • • • • • • • • • " a • • • • • • _ • . • T J I • ' • 0 • • - • • • % a n a . • • • ' n D 50 100 150 Days 200 250 Biogas • Methane 300 Figure 4-13 Revised gas productions in anaerobic external membrane MBR. the acclimatization period, the average revised total gas and methane yield values were 73 0.32 m3/kg COD removed and 0.27 m3/kg COD removed, respectively. In the steady-state operational period, they were 0.38 m3/kg COD removed and 0.17 m3/kg COD removed, respectively. The revised methane yield was still lower than the theoretical methane conversion rate. This implies that there were some unknown factors affecting the methane production notably. The measured biogas composition is presented in Figure 4.12. During the acclimatization period, methane accounted for more than 85% of the total biogas, and this composition was quite stable. The decreases in methane percentage shown were due mainly to system mechanical maintenance and/or membrane cleaning, which required nitrogen injection to purge the air that was introduced. From the speed of recovery of the methane content, we can infer that the production rate of methane was high. The high and relatively stable methane composition in the biogas in this period indicates that the system was coping well with the increasing loading rates, and that the methanogenic bacteria had adapted well to the process conditions of the study. This assumption can be verified by looking at the effluent COD concentration and effluent VFAs concentration in Figure 4.3 and Figure 4.7, respectively. However, in the steady-state operational period, the methane content continually decreased while the nitrogen content increased. The cessation of sodium acetate addition to the feed was one of the reasons. The 300 mg/L sodium acetate addition to the influent contributed approximately 240 mg/L soluble COD to the feed. Although this amount of COD was compensated by the COD in the raw sewage, the latter was mainly high moleculer weight organic material that needed to be hydrolyzed and fermented into acetate before the methane-producers could utilize it. This process needs a relatively long hydraulic retention time to permit conversion of organics in the reactor. However, in this study the removal of sodium acetate was coincident with the short HRT. The content of nitrogen did not seem to decrease to below 10% by volume during the acclimatization period and it increased sharply in the steady-state operational period. 74 Dissolution of nitrogen gas from the influent because of its reduced partial pressure in the reactor headspace could account for the substantial presence of nitrogen in the biogas. Particularly, the rapid increase of influent flow rate from 50 L/d in the acclimatization period to 108 L/d, and then to 143 L/d in the steady-state operational period, introduced significant amounts of dissolved nitrogen. Reduced methane production in the steady -state operational period is one of factors as well. The influence of OLR on methane production In Figure 4.14, a revised methane production graph for the study is plotted against organic loading rates. Q O U WD a. •a o a 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 0.00 0.50 1.00 1.50 2.00 2.50 3.00 OLR (kg COD/m3»day) Figure 4-14 The influence of OLR on methane production. During the acclimatization period, the methane production was steady in between 0.23 (±0.04) m3/kg COD removed and 0.30 (±0.07) m3/kg COD removed when OLR was no more than 1.53 kg COD/m3«day, indicating the potential capacity of the system. During the steady-state operation period, with the increases in OLR from 1.68 kg COD/m3*day to 2.83 kg COD/m3»day, the methane production slightly decreased from 0.18 (±0.04) m3/kg COD removed to 0.15 (±0.03) m3/kg COD removed. The significant decrease of methane 75 production from acclimatization period to steady -state opertion period can be attributed to the change of influent characteristics. And this can also be attributed to the higher loading rates favoring an increase in growth rate for the acidogenic bacteria over the methanogenic bacteria in the anaerobic bioreactor. Even with this condition, the anaerobic reactor performed rather well, indicating that the methanogenic bacteria had adapted well to the new process conditions and a new equilibrium must have been established between the bacterial populations in the reactor, resulting in the process continuing with no sign of impending failure. As shown in Figure 4.12, the methane content continuously decreased after day 213, on which the organic loading rate was increase to 1.68 kg COD/m 3«day. Decreasing methane content with increasing OLR has been reported when high strength wastewater was involved (Hickey et ai, 1987; Fakhru'l-Razi, 1995). Fakhru'l-Razi (1995) observed that the methane content decreased from 78.9% to 57.1% when OLR was increased from 2 kg COD/m3*day to 20 kg COD/m3'day. The authors attributed the decrease of methane content to the inhibition at higher loading rates to methanogenic bacteria. The influence of HRT on methane production In Figure 4.15, the revised methane production variation graph i s plotted against hydraulic retention time. 76 e o u M 0.40 0.35 0.30 0.25 o "O 3 0> o o fe E 0.20 0.15 0> 0.10 0.05 0.00 30 25 20 15 HRT (hours) 10 Figure 4-15 The influence of HRT on the methane production. The methane production was relatively steady during the acclimatization period, although the HRT decreased from 24 to 8 hours. When the system was operated at steady-state, the HRT continuously decreased to 4 and then to 2 hours, the methane production decreased to 1.8 (±0.04) m3/kg COD removed and then to 0.15 (±0.03) m3/kg COD removed. The methane production decreased sharply when the system operation switched from acclimatization to steady-state, which was characterized by the termination of acetate addition. The charateristics of influent change due to the termination of acetate addition is one of contributors to the decrease of methane production. However, t he higher methane production achieved at the longer HRTs was mainly due to the complete COD removal, which can be seen from the COD removal efficiency at higher HRTs shown in Figure 4.5. The extended HRT leaves the organic substrates in the reactor for complete anaerobic digestion, resulting in higher conversion efficiency to methane. 77 4.2.4 Biomass concentrations and particle size distributions The initial biomass concentrations (i.e. mixed liquor suspended solids, MLSS and mixed liquor volatile suspended solids, MLVSS) at the start -up ofthe experiment were about 4 g/L and 3.7 g/L, and they decreased gradually as presented in Figure 4.16. 16 14 J M 12 B •2 10 U B o _ n • n • "H-fl-• 8 H . 50 100 150 Days 200 250 300 MLSS a MLVSS Figure 4-16 Biomass concentrations in the anaerobic external membrane MBR. The reason that the concentrations were decreasing at the beginning of operation may be due to the accidental wasting that occurred during operation and maintenance despite an elevated SRT that resulted from minimal wasting. However, the subsequent significant decrease was not due to an actual reduction in the biomass concentration, but to a deficiency in the analytical protocol. Due to the high shear imparted by the high speed mixed liquor recirculation pumps, the distribution of biomass floe size was mainly located between 0.1 and 1 pm as shown in Figure 4.18. With a 1.5 pm glass fibre filter used for suspended solids measurements, noticeable amounts of sludge passed through the filter during analysis, causing the erroneous results. The three peaks in Figure 4.16 seen on day 70, 97 and 99 may reflect inconsistent filtering techniques, bu t the overall MLSS trend is toward 1 to 2 g/L. A new measurement method was developed a s introduced in Chapter 3, 78 section 3.2.2 Experimental variables. In the modified method, a high speed centrifuge was used to separate the sludge and supernatant. The solids left in the supernatant after centrifugation were defined as soluble solids, which were then subtracted from the total mixed liquor solids (TS) to estimate the total suspended solids, i.e. MLSS. After the new measurement method was introduced, the MLSS concentration steadily increased from 6 to 10 g/L. A significant increase in MLSS concentration can be detected after the termination of acetate supplementation. The big jump from the measurement of 9.4 g/L on day 222 to the measurement of 14.5 g/L on day 234 can be accounted for by the extensive accumulation of suspended solids introduced with the feed and not by the biomass growth. After day 215, when the sodium acetate supplementation was terminated, in order to sustain the OLR as same as or even higher than the level prior to the termination of supplemental acetate, the influent flow was increased significantly from 50 L/d to 108 L/d. The average suspended solids concentration in the influent was approximately 0.1 g/L. The suspended solids introduced by the influent every day was thus 10.8 g/d. Due to the significant decrease of HRT from 8 hours to 4 hours, and the decrease of pH values from 7.2 to 6.7 (shown in section 4.1.3), the treatment capacity of the biomass decreased, which resulted in a decrease of COD removal efficiency. If the suspended solids hydrolysis rate cannot keep up with the rate of suspended solids loading, massive suspended solids may not be hydrolyzed by the biomass. Assuming that the treatment capacity of biomass was the same as that prevailing prior to the termination of supplemental acetate addition, i.e. the biomass in the reactor could only decompose the amount of suspended solids introduced by 50 L influent each day although the influent was increased to 108 L/d, suspended solids of 5.8 g introduced by the influent each day was accumulated inside the reactor. This accumulated 5.8 g/d suspended solids made MLSS concentration increase 0.4 g/L each day. Therefore, after 11 days from day 222 to day 234, the mixed liquor suspended solids was increased by 4.4 g/L. This theoretical calculated increase of 4.4 g/L is app roximately the same as the practical discrepancy between the MLSS concentrations on day 222 and day 234. 79 The decrease of MLSS concentrations after day 234 was due mainly to periodic wasting to keep sludge retention time (SRT) at 50 days. With the membrane filtration, it is easy to maintain a desired SRT. 10.0 8.0 6.0 Dil 05 > — s 35 4.0 -2.0 0.0 • • • • • . • * n • n ° „ n • r£ • a • 1 • 3 " D -D S « m • • 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 OLR (kgCOD/m3«day) I MLSS nMLVSS Figure 4-17 Biomass concentrations as a function of organic loading rate. Figure 4.17 shows the biomass concentrations with the increasing organic loading rates. The mixed liquor suspended solids (MLSS) increased with increasing OLR except some low points resulting from the defective measurement method. The mixed liquor volatile suspended solids (MLVSS) increased with the increase of OLR when OLR was less than 1.0 kg COD/m3'day, and it was fairly constant at about 6 g/L when OLR was greater than 1.0 kg COD/m3,day, independent of the increase of OLR. This fact indicates that the biomass yield reached equilibrium. The increased inert part may account for the decrease of COD removal efficiency and gas production at higher OLR. Sizes of the mixed liquor solids and the amount of colloidal particles and soluble material 80 product have great impact on membrane performance. Due to the recirculation pumping that is required in an external MBR, the biomass is exposed to a high shear force. As a result of the high shear force, the floe will be disintegrated. Figure 4.18 and Figure 4.19 present the sludge particle size distributions during the acclimatization period and the steady-state operation period. The main peaks were in the range of 0.1 to 0.7 pm for both periods along with subtler peaks. The sludge particle size distributions show slight change after the supplementary acetate was terminated. However, the slight change might not have been caused by the effects of acetate withdrawal, but rather, by the way the particle analyzer was set up. A Malvern Mastersizer 2000 particle analyzer was used for the measurements of sludge particle size distributions as introduced in Chapter 3, section 3.2.2 Measurement variables. Before measurements, specific information regarding the type of particle analyzed must be entered into the analyzer. However, due to the limited information available for the particles analyzed, estimated values were first used and the results were obtained as shown in Figure 4.18. Coincidentally, another setting for the analysis was selected after the removal of acetate and a better theoretical fit to the observed value was obtained. Therefore, the new setting was used and the measurement results were obtained as shown in Figure 4.19. By definition, the solids contained in the filtrate that passes through a filter with a nominal pore size of 2.0 pm or less are classified as dissolved (APHA/AWWA/WEF, 1998). Therefore, from Figure 4.18 and Figure 4.19 it can be seen that dissolved solids in the external AnMBR system mixed liquor represent a significant portion of the MLSS. This fact illustrated the defect in the standard method for VSS/SS concentration measurements in the present study. Thus, after the defect was dis covered, dissolved solids were redefined and the standard method for MLVSS/MLSS concentration measurements was modified as described in Chapter 3. Typically, the size of colloidal particles is in the range from 0.01 to 1.0 pm (Tchobanoglous et al., 2003). Therefore, most of the dissolved solids were colloidal. The external AnMBR sludge floe size distributions we re quite different from the submerged AnMBR sludge floe size distributions, as presented in Figure 4.20. The submerged AnMBR 81 sludge particle size distribution peaks were mainly located at around 10 pm. The dissolved solids, particularly the colloidal solids, count only a small portion of MLSS. The difference between the sludge particle size distributions for the external and submerged anaerobic MBRs is likely caused by the different magnitude of the shear conditions within these two systems. Particle Size Distr ibution 5 4.5 u 4 j / 1 • 3.5 ..... t / ^ Jul \ \ \ fe J ( f / me (' NJ . i ; Volu 1 N) i i / I \\ \^\ 1.5 j •7 1 ) $ 0.5 7 lz 0.01 0.1 1 10 Particle Size (pm) 100 1000 3000 — External sludge feb21 - Average, Sunday, February 20, 2005 11:49:52 PM — Bio Mass External Feb 28 - Average, Sunday, February 27, 2005 11:32:57 PM — External Sludge March 7 2005 - Average, Monday, March 07, 2005 11:31:45 PM — Bio Mass External March 21 - Average, Tuesday, March 22, 2005 1:50:09 AM — Exteranl sludge March 28 - Average, Tuesday, March 29, 2005 1:56:56 A M — Bio Mass EX Apr 11 - Average, Monday, April 11, 2005 4:27:42 PM Figure 4-18 External AnMBR sludge particle size distributions during acclim atization period. g2 Particle Size Distribution 5 4.5 4 3.5 3 CD E D 2.5 O > 2 1.5 1 0.5 01 10 Particle Size (pm) 1000 3000 -Ext sludge May 13 - Average, Friday, May 13, 2005 1:32:33 PM -external sludge June 18th - A\«rage , Sunday, July 24, 2005 9:44:38 PM -External sludge June 27th - Average, Sunday, July 24, 2005 10:06:40 PM -External sludge July 4th - Average, Sunday, July 24, 2005 10:44:06 PM -External sludge Aug. 0 2 - 2 - Average, Tuesday, August 02, 2005 1:10:58 PM Figure 4-19 External AnMBR sludge particle size distributions during steady state operation period. Particle Size Distribution Particle Size (\m) — S u b sludge Feb21 - Average, Sunday, February 20, 2005 11:39:29 P M — Bio Mass Submerged Feb 28-Average, Sunday, February 27, 2005 11:23:50 P M —submerged sludge March 7 2005 - Average, Monday, March 07, 2005 11:39:58 P M —sub sludge march 21 - Average, Tuesday, March 22, 2005 2:00:33 AM —sub sludge March 28 - Average, Tuesday, March 29, 2005 1:44:58 AM — Bio Mass SUB Apr 11 - Average, Monday, April 11, 2005 4:34:40 PM — B i o Mass Apr 28 sub sludge - Average, Thursday, April 28, 2005 11:46:13 A M —sub sludge May 4 - Average, Thursday, May 05, 2005 2:29:41 P M Bio Mass submerged May 9 - Average, Monday, May 09, 2005 1:24:37 P M Sub sludge May 16 - Average, Monday, May 16, 2005 1:42:28 PM — Sub Sludge June 6 2005 - Average, Monday, June 06, 2005 12:52:33 P M — submerged sludge June 20th - Average, Monday, June 20, 2005 4:10:04 P M f.l ihrnnrnnn 1 nil irlnn ii inn 18th—A>re-vin Si inrlnw ,lnlw94 Q-fifV4fl PKA Figure 4-20 Submerged AnMBR sludge size distributions. 83 4.3 Membrane Component Performance of Anaerobic M B R 4.3.1 Initial testing To help uncover any unforeseen problems in the design and to determine the initial operational conditions of the system, initial testing with tap water and aerobic sludge was completed by using the new constructed reactor. The aerobic sludge used was obtained from the UBC pilot plant membrane bioreactor for the treatment of municipal wastewater with enhanced biological phosphorus removal. Permeate flux of membrane modules and flow rate of mixed liquor recirculation loops were measured. This gave good estimated flow values, on which OLR calculations were based, and through these values, subsequent rea ctor modifications were made. Worth noting is the modifications made to the baffles inside the reactor to accommodate higher OLR values. The determination of operating trans -membrane pressure (TMP) The permeate flux values of the membrane modules measured under different operating trans-membrane pressures with different filtration media are presented in Table 4.3. The permeate flux increased with TMP when the membranes were tested with tap water. However, when membranes were tested with aerobic mixed liquor, the permeate flux reached a maximum value at a TMP of 276 kPa (i.e. limiting for conditions). Above this TMP, the permeate flux decreased with further increases of TMP. Therefore, the initial operating TMP was selected to be 207 kPa. A TMP below the limiting flux condition was selected to leave some leeway to the increase the TMP in order to investigate the effect of TMP on the membrane performance. 84 Table 4.3 The initial testing results of membrane permeate flux at different TMP Membrane modules Trans-membrane pressure (kPa) Permeate flux (Lmh) With ta j water With aerobic mixed liquor Original membrane Rewetted membrane PCI 138 170 310 96 207 300 520 150 276 450 660 176 345 530 780 162 414 570 1168 154 Membralox 138 580 155 207 330 1026 213 276 435 1297 232 345 580 1606 213 414 770 1936 155 The determination of cross-flow velocity (CFV) According to the membrane information provided by the manufacturers of PCI and Koch, the cross-flow velocity should be maintained between 2.3 to 3.4 m/s during operation. The initial operational crow-flow velocity in the present study was chosen to be 2.5 m/s. The cross-flow velocity of the external membrane system was determined by measuring the flow rate of the recirculation pumps at different trans-membrane pressures and pump capacities. The results of the findings are presented in Table 4.4. Based on the measurement resu Its, the initial pump capacity settings were determined to be 30% for the Membralox membrane, 45% for PCI membrane and 60% for Koch membrane. The practical operating pump capacities used were slightly higher with 35% for the membralox membrane and 65% for the Koch2 membrane since the measurement results were obtained from the tap water test. 85 Table 4.4 Measurements of cross-flow velocity under different operational conditions Membrane Pressure (kPa) Pump setting Flow rate (L/min) C F V (m/sec) Membralox 207 30% 6.6 2.9 40% 8.8 3.8 276 30% 5.7 2.5 40% 8.4 3.6 PCI 207 35% 11.6 2.3 40% 12.8 2.5 45% 13.8 2.7 276 45% 12.4 2.4 50% 13.4 2.6 Koch2 207 50% 11.6 2.5 60% 14.2 3.0 276 50% 9.2 2.0 60% 12.6 2.7 4.3.2 Permeate flux of membranes The major design considerations for the application of membrane separation in biological reactors are the membrane flux and the ability to prevent fouling of the membrane to sustain acceptable flux rates (Tchobanoglous et ai, 2003). Therefore, the desired and sustainable permeate flux of membranes is an important indicator of anaerobic MBR performance. Figure 4.21 shows the permeate flux over time for three different external membranes in terms of L/m2«hr (Lmh). 86 150 100 50 0 TMP = 2G?kPrt , o o Q „ T\fP = :7(5 trPa *. Acetate removed QD D OO > O o O G GQSQfJO v o o o ^ pr> <sf•1 « O O CHDO0D o o o cm CIXJB <g> o o x K o c h ! at 207 kPa ° x S5 1 1— 1 *—1 1 1 1 <&>**o<t-v —v o 1 e 0 50 100 150 200 250 300 Days X PCI o Mem & Koch <> Kochl Figure 4-3 The external membrane system permeate flux (Jv). The permeate flux of inorganic membrane, Membralox It can be seen that the inorganic membrane (Membralox) perform ed substantially better than the two organic membranes. During the first three months of operation, the permeate flux steadily decreased from 106 Lmh to 48 Lmh. Due to the low permeate flux, chemical cleaning was carried out on Day 98. However, the permeate flux only recovered to 68 Lmh after chemical cleaning. The flux increase observed after Day 101 resulted from an increase in CFV achieved by increasing recirculation pump setting from 35% to 45% on that day and again to 60% on Day 103. After Day 105, the permeate flux greatly increased with the increase of TMP from 207 to 276 kPa. However, this increase in permeate flux only lasted for a short period. Beyond Day 108, the permeate flux decreased exponentially at a TMP of 276 kPa and after about 70 days of operation the flux dropped below 100 Lmh. This might be due to more compact cake layer forming on the membrane surface at a higher TMP. Although membrane cleaning was carried out on Day 148, the permeate flux decreased again quickly. On Day 192, in order to keep a relative stable permeate flux, the recirculation pump setting was increased slightly from 60% to 65%. However, it seems that the increased pump speed did not play a significant role, and the permeate flux continuously decreased to 77 Lmh. Membrane cleaning was carried out again oh Day 212 in order to improve the permeate flux to meet the requirement of permeate flux increase due to the increase of influent flow after acetate removal. However, this cleaning did not achieve the expected results. It is interesting to note that the permeate flux increased to an average of 110 Lmh after the removal of supplemental acetate. The permeate flux of organic membrane. PCI The permeate flux of the PCI membrane decreased steadily from the initial value of 37 Lmh to zero over the duration of the project. Since the low permeate flux was not sufficient to sustain the target OLR, chemical cleaning was carried out on Day 89 and Day 99. Although the permeate flux recovered to close to the initial value, it decreased to even lower values within a few days. The recirculation pump setting was subsequently increased from 60% to 85% on Day 100 in order to increase the CFV. This increase in CFV improved the permeate flux significantly as can be seen in the Figure 4.21 when the peak flux was above 50 Lmh. 88 Unfortunately, the flux then decreased rapidly and back to normal within a few short days. Although the TMP was increased from 207 to 276 kPa on Day 105, unlike the inorganic membrane, the permeate flux of PCI was not affected by the TMP increase. The permeate flux decreased continuously and s ubsequent cleaning did not help and the flux never made a significant recovery. This is likely the result of inadequate CFV provided to the membrane over time. Due to the change of recirculation pump, although the pump setting was used at 85%, the new pump was unable to meet the flow requirements for a proper CFV. Another reason for the poor flux recovery could be the fact that the nominal pore size ofthe membrane was about 0.1 pm, which was very close to the size of the biomass (as shown in Figure 4.18 and 4.19), further discussion was given in section 4.3.3 Factors affecting the performance of membrane component. In addition, the surface charge of the membrane may have been altered during the process of cleaning. On Day 143, permeate flux for the PCI membrane finally decreased to zero after a cleaning two days prior. This membrane was removed and replaced with the new Koch membranes, referred to as Koch2. The permeate flux of organic membrane. Koch and Koch2 The Koch membrane was installed on Day 106 in series with the inorganic Membralox membrane. The flux was relatively steady at an average flux of 40 Lmh until Day 150. Thereafter, the permeate flux of the Koch membrane decreased steadily until the membrane provided essentially no contribution to the permeate flux requirement of the AnMBR system. Thorough chemical cleaning was not helpful to the permeate flux recovery. The Koch2 membranes were installed on Day 184 to replace the PCI membrane . The initial flux of Koch2 decreased quite rapidly within the first 10 days of operation until it was almost the same as that of Ko ch. However, the permeate flux of Koch2 progressed in an upward trend after Day 198 and eventually recovered. After Day 199, the permeate flux of Koch2 membrane remained substantially higher than that of the Koch membrane. The difference may have resulted from a change of operational conditions. Initially, the Koch2 89 membrane was operated at the same operational conditions as the Koch membrane, at a TMP of 276 kPa and recirculation pump setting of 60%. After 10 days of operation, the TMP and pump setting for Koch2 membrane were changed to 207 kPa and 65%, respectively. Therefore, a combination of proper TMP and CFV may affect the performance of Koch2 membrane, as examined in section 4.3.3 Factors affecting the performance of membrane component. As for the inorganic membrane, Membralox, the Koch2 membrane flux also increased slightly after the termination of supplemental acetate. The permeate flux of Koch2 membrane suffered a setback on Day 224 as shown in Figure 4.21, due to a failed pump motor. A subsequent membrane cleaning with bleach solution of 300 mg/L recovered the flux to near original values. Membrane cleaning events are listed and discussed in the following subsection entitled Membrane cleaning. 4.3.3 Factors affecting the performance of membrane co mponent Membrane material and pore size The results in Figure 4.21 show that the permeate flux of the inorganic membrane was significantly higher than that of the organic membranes. Ghyoot and Verstraete (1997) also observed that the permeate flux in an An MBR with a ceramic membrane was significantly higher than that which could be achieved with an organic membrane. Kang et al. (2002) observed a slower decline in permeate flux of an inorganic membrane over time than that observed for organic membrane. As stated in the literature, a cake layer typically does not form on an inorganic membrane (Kang et al., 2002; Yoon et al., 1999). As examined in the section 4.3.4 Short-term off-line filtration tests, compared to aerobic sludge, the filtration resistance of anaerobic sludge was mainly due to cake layer formation. Therefore, the superior performance of inorganic membrane may have been due to the absence of a cake layer on the membrane surface. The difference between the cake layer presence on an organic membrane surface and on an inorganic membrane surface may have been mainly 90 been mainly due to the structural differences between these two types of membranes. As examined by Kang et al. (2002), the organic membrane had a rougher, fibrous surface which favors cake layer formation on the membrane surface. Kang et al. (2002) also attributed the absence of cake layer formation on the inorganic surface to the weak adsorption of inorganic material to the foulant s. The ionic ligand on the membrane surface was not easily replaced by the ionic ligand ofthe materials in mixed liqu or. Therefore, the cake layer buildup on an inorganic membrane surface is relatively rare. However, the cross-flow velocity at the membrane surface can also affect the presence of a cake laye r on a membrane surface (Choo and Lee, 1998). In the present project, an organic Koch membrane was installed in series with an inorganic Membralox membrane. The flow rates and the TMP of these two membranes were identical. However, due to different cross section areas, the CFV along the organic Koch membrane surface was only about 50% of the CFV along the inorganic Membralox membrane surface. The cross -flow velocity effect is detailed in the following subsection. In this experimental program, the inorganic Membralox membrane was used for the whole experimental period without decline of the permeate flux, while the organic membranes, PCI and Koch, were unable to contribute significantly to the permeate flux of the system after about 100 days operation. Therefore, the inorganic membrane had a longer service life cycle than the organic membrane. As one of intrinsic characteristics of a membrane, pore size greatly affects membrane performance. A large pore size membrane has an initial high permeate flux, but it fouls rapidly (Elmaleh and Abdelmoumni, 1997; Chung et al, 1998; He et al, 1999). Figure 4.22 shows the effect of membrane pore size on permeate flux. Organic membranes PCI and Koch are made of the same material p olyvinylidene fluoride (PVDF), but they have different pore sizes. The PCI membrane has a pore size of 0.1 pm, and the Koch membrane has a smaller size of 0.005 to 0.01 pm. Operating at the same TMP and CFV, initially the PCI membrane exhibited a higher permeate flux than the Koch membrane. However, the fouling rate for the PCI membrane was much faster than for the Koch membrane. During 91 40 days of operation, the permeate flux of the PCI decreased exponentially from 55 Lmh to 6 Lmh, but the permeate flux of the Koch membrane was stable at around 40 Lmh. 60 50 E 40 •= 30 I 20 10 • A A A A A A A . A A-A-A -A A A A-P C I 0.1 u m A K o c h 0.005-0.01 u m 10 15 20 25 30 35 40 45 Days Figure 4-22 Permeate flux of organic membrane PCI and Koch with operation time. Imasaka et al (1989) reported that under the same operating condition of cross-flow filtration, neither the pore size of the membran e nor the membrane structure influenced the properties of the cake layer. Therefore, the permeate resistance of the cake layer does not differ with pore size, but the resistance due to plugging increases with an increase in pore size. As shown in Figure 4.18 and 4.19, the biomass particle size distribution in the present study was dominated by particles with diameters of about 0.1 pm, which was quite close to the pore size of the PCI membrane. This similarity between membrane pore size and the biomass particle size may easily cause the irreversible pore plugging, resulting in rapid fouling for the PCI membrane. Cross-flow velocity The special feature of the external membrane filtration process is the dynamic cross-flow. The filtration process causes an increase in concentration of solids on the membrane surface. 92 The material is transferred convectively in turbulent flow to the membrane surface. The transport from the membrane surface, and back to the bulk liquid, occurs partly through diffusion. If the back transport rate is smaller than the convective rate, a cake layer builds up on the membrane surface (Brockmann and Seyfried, 1996). To control fouling, it is necessary to maintain a high liquid velocity across the membrane to maintain a high back transport rate. Figure 4.23 shows the effect of cross-flow velocity on the membrane performance. With an increase in CFV, the permeate flux of the PCI and Membralox membranes increased linearly. 160 140 120 100 80 60 40 20 y = 15.194x +51.367 y - 8.7327x + 20.349 = 0.9812 3 4 Cross-flow velocity (m/s) * Membralox • PCI Figure 4 -23 The effect of cross-flow velocity on membrane performance. To further investigate the effect of CFV on permeate flux, a flushing test was conducted with the Koch2 membrane from March 25 to April 5, 2005. In this test, the cross flow velocity was increased from 3.3 m/s to 4.3 m/s, by increasing recirculation pump setting from 65% to 85% for about 5 minutes. The effect of flushing on the permeate flux is presented in Figure 4.24. 93 60 a . 44 40 I 1 1 1 1 1 1 . 1 0 1 2 3 4 5 6 7 8 9 Days • Jv before flush * Jv after flush Figure 4-24 The effect of high CFV flushing on the permeate flux of Koch2. From Figure 4.24 it can be seen that during the 8 days of testing, high CFV flushing maintained membrane permeate flux at a steady state. The increase of permeate flux after flushing can mainly be attributed to the reduction of cake layer thickness by high velocity scouring of cake from the membrane surface. Trans-membrane pressure The trans-membrane pressure is the driving force in the filtration process, but high pressure alone can not achieve efficient flux rates. Figure 4.25 shows the membrane performance at different combinations of TMP and CFV. 94 80 • co -rxP- • cb cs _0 C % 3 10 20 30 40 50 Days 60 70 80 90 Koch at 276 kPa 2.7 m/s CFV • Koch2 at 207 kPa 3.3 m/s CFV Figure 4-25 Membrane performance at different combinations of TMP and CFV. The combination of a minimum pressure with the necessary cross-flow velocity makes effective filtration possible. In a cross-flow filtration process, the cake layer build-up commences with the initial operation of the membrane pump and is completed within several hours, and is thus an integral part ofthe filtration process (Brockmann and Se yfried, 1996). In-order to obtain a high flux, the cake layer should be loosely packed or have large pores. The trans-membrane pressure is one ofthe decisive factors that affects the cake layer density. An increase in pressure only temporarily achieves hi gher flux. Initially, the higher pressure forces more water through the filter cake, but the cake layer is compressed in the process. As a result, the resistance of the filter increases and the flux decreases. The high velocity flow inside tubular membranes can create enough shear to reduce the cake layer thickness. Therefore, although the Koch2 membrane was driven by lower TMP of 207 kPa than the Koch membrane at a TMP of 276 kPa, the permeate flux of the Koch2 membrane was higher and more stable than that of the Koch membrane, due to the higher CFV for the Koch2 operation. 95 Gas injection To investigate the effect of gas injection on the permeate flux, an experiment was conducted from day 250 to day 276. In this experiment, biogas in the headspace of the reactor was injected by a Masterflex pump into the mixed liquor recirculation line, mixing with the fluid. Figure 4.26 presents the experimental results. When operation was carried out with gas injection over a 7 day period, both the Koch2 and Membralox membranes exhibited a steady permeate flux. When'the gas injection was turned off, the permeate flux of both membranes decreased. The permeate flux of the organic Koch2 membrane was affected most significantly. However, when the gas injection was turned on again, it seems that the inorganic membrane was not affected. Although the permeate flux of the organic Koch2 membrane increased gradually, the permeate flux of the inorganic Membralox remained at the same value. E j x s IS cu cu E 140 120 100 80 60 40 With gas injection Without gas injection With gas injection 10 15 Days 20 25 30 Koch2 -»- Membralox Figure 4-26 The effect of gas injection on the permeate flux of Koch2 membrane and Membralox membrane. The effect of gas injection on the permeate flux may be caused by two impacts: the promotion of turbulence in the gas-liquid mixture in a two-phase flow system, and the 96 resulting influence on the formation of a cake layer on the membrane surface. Particles . deposited on the membrane surface can, in principle, be removed with a turbulent flow of water parallel to the membrane surface (Verberk et al, 2002). By adding gas into the feed stream during the mixed liquor recirculation, the flow becomes highly turbulent, thus promoting cleaning. In cross-flow ultrafiltration, a stable cake layer of a certain thickness and porosity can be formed on the membrane surface by a dding gas continuously to the feed stream (Cabasssud et al., 1997; Laborie et al, 1997, 1998). Although in the present study the gas injection did not increase the permeate flux of the inorganic Membralox membrane, a stable permeate flux was sustained. The different effect of gas injection on the organic Koch2 membrane and the inorganic Membralox membrane may be due to the structural differences between the two types of membranes and the operational conditions. The inorganic membrane has a smooth surface w hile the organic membrane has a rougher, fibrous surface (Kang et al, 2002). Therefore, the interaction between the turbulent fluid and the organic membrane surface may have been stronger than the interaction between the turbulent fluid and the inorganic membrane surface. As a result, the development of a cake layer would hardly take place on the surface of the inorganic membrane. Also, if there is less of a cake on the inorganic membrane, the more shear would not have a large impact. In addition, in this experiment, the inorganic membrane was operated at a TMP of 276 kPa, but the organic membrane at a TMP of 207 kPa. At higher trans -membrane pressure, the cake layer formed on the membrane surface is more compact, so that it is not affected easily by turbulent shear. Suspended solids concentration As one of mixed liquor characteristics, the suspended solids concentration in an anaerobic MBR has a significant impact on membrane permeability, which is defined as the permeate flux provided by unit pressure. Figure 4.27 presents the effect of mixed liquor MLSS concentration on the membrane permeability. The permeability of the Membralox and Koch membranes decreased with an increase of MLSS concentration when the membranes were operated at a TMP of 276 kPa. This would 97 be a result of increased liquid viscosity associated usually with an increased solids concentration. An increased liquid viscosity can create lower turbulence at the membrane surface, causing the decrease in mass transfer away from the membrane . This observation is similar to the findings by Saw et al. (1986) and Kitamura et al. (1996). 25.00 | — ,. MLSS concentration (g/L) I • Membralox • Koch I Figure 4-27 The effect of sludge concentration on the membrane permeability . From the slope of the trendline, the effect of the suspended solids concentration on the inorganic membrane was judged to be greater than on the organic membrane. This may be explained by the membrane pore size. The Koch membrane has 0.005 - 0.01 pm membrane pore size, which is five times smaller than that of Membralox's 0.05 pm. It has been extensively verified that membranes with a larger pore size foul more rapidly and to a greater extent (Choo and Lee, 1996b; Chung et al, 1998; He et al, 1999). However, on the other hand, the performance ofthe inorganic Membralox membrane was superior to that of the organic Koch membrane. This might be related to the morphology o f and the cross-flow velocity along the membrane surface. As discussed in the previous section Gas injection, the rougher surface of an organic membrane is more beneficial for the formation of a cake layer on the membrane surface. Although the flow rate passing through these two 98 membranes was the same, the CFV along membrane surface was different due to the different cross sectional area of the tubular membranes. The CFV through the Koch was 2.7 m/s, which is much smaller than the 5.4 m/s CFV for the Membralox unit. Membrane cleaning A number of studies have reported that it is possible to consistently recover the permeate flux by membrane cleaning with caustic solution for organic fouling and acidic solution for inorganic fouling (Lee et al, 2001b). However, in the present project, the permeate flux recovery by chemical cleaning, according to the procedures provided by manufacturers with some modifications, did not result in complete flux recovery. Table 4.5 presents the permeate flux before and after membrane cleaning and the cleaning procedures used that were listed in Table 3.5. The cleaning sequence of sodium hydroxide followed by nitric acid only resulted in a very short term flux recovery, followed by a faster fouling rate that is represented by the decrease of permeate flux. A surface charge interaction between the membrane and solutes in the chemical solution could be a possible explanation for inorganic membrane (Kang et a l , 2002). The removal of cake layer on the membrane surface by chemical cleaning temporarily increased the permeate flux, but the change of the membrane surface characteristics might dominate. Therefore, the permeate flux decreased again within a very short term after the membrane cleaning. The results of cleaning for Koch2 verified this assumption. In the process of Koch2 membrane cleaning, only sodium hypochlorite was used. The permeate recovery for Koch2 membrane was quite satisfactory and the flux was stable during the remaining experimental period. 99 Table 4.5 The permeate flux recovery by membrane cleaning Membrane Cleaning event Jv-1 (Lmh) Jv-2 (Lmh) Jv-3 (Lmh) D (days) Cleaning procedure 1 37 20 29 4 PCI 2 29 15 38 1 A 3 15 55 2 4 5 10 1 1 106 48 68 * Membralox 2 242 116 169 4 B 3 82 102 ** 1 46 29 49 1 Al Koch 2 11 11 3 11 9 A2 . Koch2 1 65 24 49 *** C 2 49 49 67 *** * Due to the increase of CFV, the permeate flux did not decrease after the cleaning. ** The permeate flux slightly increased and held to the end of the project. *** The need for cleaning was due to the system mechanical failure, leaving the sludge idle inside the membrane. The permeate flux was quite stable until the end of the project. Iv-1: the flux of clean membrane Iv-2: the flux of used membrane that needed to be cleaned Iv-3: the flux of membrane right after cleaning D: duration during which Jv-3 declined to Jv-2 100 4.3.4 Short-term off-line filtration tests To obtain a better understanding of the difference in membrane fouling mechanisms between the anaerobic and aerobic MLSS, off-line filtration testing was conducted on new fiber membranes. The filtration setup in Figure 4.2 8 was adopted from Geng (Z. Geng, Civil Engineering, UBC, Vancouver, B.C., pers. comm., 2005), a Ph.D student in the Civil Engineering Department of University British Columbia working on membrane fouling characteristics. This work utilized a 2 L reactor that supported four strands of hollow fiber membranes (total surface area: 0.004 m2) with nitrogen gas sparging for the filtration test. Membranes used for this part of the research were new PVDF Zenon hollow fiber membranes with 0.04 pm pore size, possessing surface chara cteristics that are hydrophilic and negatively charged in nature. A digital pressure gauge was connected on the permeate line to detect pressure changes over time. The system was driven by a Masterflex positive displacement pump with variable flow velocity control. Figure 4-28 Filtration test apparatus setup. 101 In order to be consistent with Geng's filtration testing procedure, the operating conditions of this experiment were identical to the ones she used. The Masterflex flow rate was set to 2.2 mL/min, equal to a membrane permeate flux of 33 Lmh. The air/nitrogen gas sparging intensity was set at 1.8 L/min. This test was conducted three times with different sludge types and concentrations, as shown in Table 4.6. Table 4.6 Sludge type and concentration in different test Test Sludge type and concentration 1 Aerobic sludge with 6 g/L sludge concentration 2 Anaerobic external membrane MBR sludge with concentration of 14 g/L. 3 Anaerobic external membrane MBR sludge diluted with de-ionized distilled water to the concentration of 5 g/L. The dependence of permeate flux on applied trans -membrane pressure can be expressed by the following equation (Choo and Lee, 1996 a,b; Kang et al., 2002). j _ TMP \x.{Rm + Rc + Rp + Ra) Where J is the permeate flux, TMP is the applied trans-membrane pressure, and u is the permeate viscosity. Rm is the intrinsic membrane resistance, Rc is the resistance due to a cake layer formed on the membrane surface, Rp is the resistance due to plugging in the membrane pores, and Ra is the resistance due to adsorption of organic species. The total resistance encountered during the filtration of sludge is due to Rm plus the resistance invoked by the activated sludge (Rs); which is composed of the three parameters as described above: Rc, Rp and Ra. Rm can be obtained by subjecting new membrane modules to filtration with de-ionized distilled water (DDW). Rs can be obtained from the difference in resistance observed through filtration tests with DDW and the mixed liquor. In a filtration process, Rc can be eliminated with the removal of bulk sludge along with relaxation and gas sparging. Rp is usually reduced by the process of back-flushing. Ra is hydraulically 102 irreversible and can only be removed through chemical cleaning. The relationship among these resistances is illustrated by aerobic sludge filtration test result s in Figure 4.29 and Figure 30. Aerobic sludge filtration tests The repeatability of the filtration equipment and procedure was tested by comparing the filtration characteristics of the aerobic sludge with previous observations made by Geng. The aerobic sludge was obtained from the UBC pilot plant membrane bioreactor for the treatment of municipal wastewater with enhanced biologica 1 phosphorus removal. The results of the aerobic system then set a benchmark (Figure 4.29 and Figure 30) to which anaerobic mixed liquor filtration results could be compared. New membrane modules were submerged in activated sludge and operated under vacuum with aeration. The trans-membrane pressure built up within 30 minutes and then leveled off (Phase I). After 120 minutes with a constant TMP, the sludge was replaced with DDW, followed immediately by a five minute membrane relaxation period prior to Phase II. During this phase, a significantly lower TMP was observed. After a quick 30 second back flushing, the TMP decreased again, but only slightly (Phase III). The permeate flux was sustained at 33 Lmh during the testing period. The results of this short-term test indicate that the total resistance was mainly due to the intrinsic membrane resistance, Rm, which accounted for 87% of the total resistance, R. The resistance invoked by aerobic sludge, Rs, was largely contributed by cake layer formed on the membrane surface, which accounted for about 56%ofRs and 7% of R. The plugging resistance, Rp and the adsorption resistance, Ra only represented a small portion of Rs, at 31% and 17%, respectively. 103 Figure 4-29 Filtration results of the aerobic sludge at a concentration of 6 g/L. Figure 4-30 Filtration resistance of the aerobic sludge at a concentration of 6 g/L. Anaerobic sludge filtration tests Anaerobic sludge tests were conducted to investigate the filtration characteristics of the sludge in the AnMBR. In the first of two filtration tests, the mixed liquor used was taken 104 from the external membrane AnMBR at a concentration of 14 g/L, and in the second, was the dilution of the same sludge in DDW at a concentration of 5 g/L in order to simulate the aerobic sludge concentration for the purpose of comparison. New membrane modules were submerged in activated sludge and operated under vacuum with nitrogen gas sparging. The test results are shown in Figure 4.31 and Figure 4.32. a. 2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 T i m e (mins) 1 6 0 T M P - D D W - o - T M P - S l u d g e ( 1 4 g / L ) - A - J V 1 8 0 2 0 0 Figure 4-31 Filtration results of the anaerobic sludge at a concentration of 14 g/L. 105 Figure 4-32 Filtration results of the anaerobic sludge at a concentration of 5 g/L. During filtration of the high concentration anaerobic sludge (Figure 4.31), the TMP increased rapidly for the first 10 minutes and then stabilized at around 40 kPa with a slightly variation to the end of 120 minutes. During that same period, the membrane flux rapidly decreased to half of its initial value. The results of the filtration of the low concentration sludge (Figure 4.32) was slightly different. The TMP increased rapidly for the first 20 minutes to a much higher level than for the high concentration sludge, but subsequently gradually decreased between 20 to 80 minutes and leveled off to the end of Phase 1. Although there were still slight flux decreases during the filtration, the extent was smaller than the results of the sludge with high concentration. The Masterflex pump speed was set at a consistent level to try to keep a constant permeate flux. However, the variation of the permeate flux indicated that the pump may not have been suitable for use with anaerobic sludge. Due to the variation of TMP and permeate flux at the same time, filtration resistance results shown in Figure 4.33 and Figure 4.34 were used to compare filtration characteristics of the 106 anaerobic sludge with that of the aerobic sludge, and also to examine the effect of anaerobic sludge concentration on the filtration process. 180 200 Time (mins) -•- DDW -m- Sludge (14 g/L) Figure 4-33 Filtration resistance of the anaerobic sludge at a concentration of 14 g/L. 3.5 0 20 40 60 80 100 120 140 160 180 200 Time (mins) -»-DDW -m- Sludge ( 5 g/L) Figure 4-34 Filtration resistance of the anaerobic sludge at a concentration of 5 g/L. 1 0 7 Comparing the anaerobic sludge filtration results with the aerobic filtration results (Figure 4.30), it was established that the anaerobic sludge developed a total filtration resistance 10 times greater than that of the aerobic sludge under similar MLSS concentrations despite identical operating conditions for both the aerobic and anaerobic filtration tests. This is due perhaps to the greater resistance imposed by the anaerobic sludge. Unlike the filtration with aerobic sludge, the total resistance was mainly contributed by the resistance due to a cake layer, Rc, which was significantly greater than the intrinsic membrane resistance itself (about 14 times). The resistance due to adsorption and pore plugging only accounted for a very small portion and was thus considered negligible for the sludge from the external membrane AnMBR. From the results shown in Figure 4.33 and Figure 4.34, a conclusion can be drawn that the filtration resistance of the sludge from the external membrane AnMBR was independent of the sludge concentration. This can be further illustrated with membrane permeability under different anaerobic sludge concentrations as shown in Figure 4.3 5. 16 T 14 II 'e? 12 Q 0 20 40 60 80 100 120 Time (mins) -•- at sludge cone, of 14 g/L -o- at sludge cone, of 5 g/L Figure 4-35 Permeability of membrane under different sludge concentrations. 108 The permeability of the membranes decreased rapidly over the first 20 minutes of the filtration tests, and then leveled off. Comparing the two curves, we can see that the membrane permeabilities exhibited no obvious difference with the very different sludge concentrations. Therefore, the permeability of membrane appears not to have been affected by suspended solids concentration. Obviously, this result is contradictory to the results obtained from external AnMBR operation in Figure 4.27. The results from the present study are consistent with those of Beaubien et al. (1996). Their results showed that at low trans-membrane pressures (less than 83 kPa), sludge concentrations of more than 2.5 g/L had almost no effect on the permeability of a membrane, but at trans -membrane pressures higher than 83 kPa, the permeability of membrane decreased with an increase of suspended solids concentration. 109 C H A P T E R 5 C O N C L U S I O N S A N D R E C O M M E N D A T I O N S This-thesis focuses on the evaluation of an external membrane anaerobic bioreactor. The biological performance, membrane component performance and the factors which affected these performances were investigated. According to the experimental results, the following conclusions can be made. • The external membrane anaerobic MBR used in the present project demonstrated satisfactory biological treatment performance when it treated low strength municipal wastewater at a temperature of 36 - 25 °C. The measured COD removal efficiency was as high as 86%, and declined to 55% when the HRT was decreased to as low as 2 hours. • Hydraulic retention time (HRT) and organic loading rate (OLR) significantly affected the reactor performance. Based on our experiments, an HRT of 8 hours and an OLR of 1.6 kg COD/m3-day can be identified as the optimum operational conditions for the system. • COD removal efficiency was independent of temperature at 36 - 25 °C. • The ratio of influent soluble COD to total COD exerted a significant effect on COD removal efficiency. • Taking the loss of dissolved methane in permeate into account, in the acclimatization period, the average total gas and methane production were 0.32 m 3/kg COD removed and 0.27 m3/kg COD removed, respectively. In the steady -state operation period, these were 0.38 m3/kg COD removed and 0.17 m3/kg COD removed, respectively. The production of methane was lower than the theoretically calculated amount of 0.4 m3/kg COD converted. • It was easy to maintain suspended solids in the reactor due to the mem brane separation. The suspended solids concentration in the reactor increased steadily. no • The sludge particle size distribution of the external membrane AnMBR was quite different from that of a submerged membrane AnMBR due to the high shear force imposed on sludge by the recirculation pump. The sludge particle size distribution of external AnMBR was dominated by particles ranging from 0.1 to 1 pm in diameter,. • The performance of the membrane component is critical to the system efficiency. With high permeate flux, the system could have more capacity for the wastewater treatment. • Although the organic PCI membrane with a relatively large pore size exhibited high initial permeate flux, it fouled faster than the organic Koch membrane with a smaller pore size. • The permeate flux of both the organic and inorganic membranes was significantly impacted by the cross flow velocity. The permeate flux of membranes increased linearly with the increase of cross flow velocity. By rapidly increasing the CFV for a short time to flush the membrane surface, the membrane fouling could be effectively reduced. • The combination of a minimum pressure with the necessary cross-flow velocity makes effective filtration possible. The Koch membrane exibited higher and more stable permeate flux when operated at a TMP of 207 kPa and a CFV of 3.3 m/s than when operated at a TMP of 276 kPa and a CFV of 2.7 m/s. • At a low operating pressure (less than 83 kPa), suspended solids concentration did not affected membrane permeability. However, under an operating pressure of 276 kPa, the membrane permeability linearly decreased with an increase in suspended solids concentration. • The chemical cleaning of membrane with caustic solution and acidic solution did not give an effective permeate flux recovery. However, chemical cleaning with hypochlorite solution only effectively recovered the organic Koch membrane permeate . flux, and the recovered flux could be maintained for reasonably long periods of operation. Based on the experience gained from this study, the following recommendations are made in order to facilitate future research. in Due to time limitations, a further decrease of operational temperature was not conducted in this study. In the future study, the target operational temperature should be set at below 20 ° C . The cross-flow velocity through membranes should be controlled and measured precisely during future operation. The shear force imposed on the membrane surface should be measured as well. In order to determine the exact reasons that chemical cleaning with caustic and acidic solution did not effectively recover the permeate flux, the change of membrane surface characteristics should be further investigated. Recirculation pumps, feed pumps, level sensors, and gas line should be examined and maintained periodically. The contribution of different sludge components to membrane fouling should be examined by fdtration test in order to further understand the membrane fouling mechanisms in anaerobic MBR. 112 R E F E R E N C E S Ahring, B.K., Ibrahim, A.A. and Mladenovska, Z. (2001). "Effect of temperature increase from 55 to 65°C on performance and microbial population dynamics of an anaerobic reactor treating cattle manure," Water Research, Vol. 35, No. 10, p. 2446-2452. Anderson, G.K., Saw, CB. and Fernandes, M. I.A.P. (1986). "Application of porous membranes for biomass retention in biological wastewater treatment processes," Process Biochemistry, December, p. 174. Angenent, L.T., Banik, G.C. and Sung, S. (2000). "Psychrophilic anaerobic pretreatment of low-strength wastewater using the anaerobic migrating blanket reactor," Proceedings of the WEF 73rd Annual Conference & Exposition, Anaheim, California, CD-ROM, October. Angenent, L.T., Banik, G.C. and Sung, S. (2001). "Anaerobic migrating blanket reactor treatment of low-strength wastewater at low temperatures," Water Environ Res. Vol. 73, p. 567-574. APHA/AWWA/WEF (1998). "Standard methods for the examination of wastewater, 20 t h edition," Washington, USA. Baek, S.H. and Pagilla, K. (2003). "Anaerobic membrane bioreactor for dilute municipal wastewater treatment," Proceedings of the WEF 49 Annual Conference & Exposition, Los Angeles, California, October. Beaubien, A., Baty, M., Jeannot, R, Francoeur, E. and Manem, J. (1996) "Design and Operation of Anaerobic Membrane Bioreactors: Development of a Filtration Testing Strategy," Journal of Membrane Science, Vol. 109, p. 173. Beolchini, F., Cimini, S., Mosca. L., Veglio, F. and Barba, D. (2005). "Microfiltration of bovine and ovine milk for the reduction of microbial content: Effect of some operating conditions on permeate flux and microbial reduction ," Sep Sci Technol., Vol. 40, p. 757-772. 113 Bilstad, T., Madland, M., Espedal, E. and Hanssen, P.H. (1992). "Membrane separation of raw and anaerobically digestd pig manure," Water Science & Technology, Vol.25, No. 10, p. 19. Bodik, I. Herdova, B. and Drtil, M. (2002a). "The use of upflow anaerobic filter and AnSBR for wastewater treatment at ambient temperature," Water Res., Vol. 36, p. 1084-1088. Bodik, I., Kratochvil, K., Herdova, B., Tapia, G. and Gasparikova, E. (2002b) "Municipal wastewater treatment in the anaerobic-aerobic baffled reactor at ambient temperature," Water Science & Technology, Vol. 46, No. 8, p. 127. Botha, G.R., Sanderson, R.D. and Buckley, CA. (1992). "Brief historical review of membrane development and membrane applications in wastewater treatment in Southern Africa," Water Science & Technology, Vol. 25, p. 10. Bourgeous, K.N., Darby, J.L. and Tchobanoglous, G. (2001 ). "Ultrafiltration of wastewater effects of particles, mode of operation, and backwash effectiveness," Water Research, Vol. 35, No. 1, p. 77. Brockman, M. and Seyfried, C.F. (1996). "Sludge activity and cross-flow microfiltration -a non-beneficial relationship," Water Science & Technology, Vol. 34, No. 9, p. 205. Buhr, H.O. and Andrews, J.F. (1977). "The thermophilic anaerobic digestion process," Water Research, Vol.11, p. 129-143. Butcher, G.J., (1989) "Experiences with anaerobic digestion of wheat starch processing waste," International Biodeterioration Vol.25 pp 71-77 Cabassud, C , Laborie, S. and Laine, J.M. (1997). "How slug flow can improve ultratiltration flux in organic hollow, fibre", Journal of Membrane Science, 128, 93-101. Chang, In-Soung, Choo, K.H., Lee, C. H., Pek, U.H., Koh, U.C., Kim, S.W. and Koh, J.H. (1994). "Application of ceramic membrane as a pretreatment in anaerobic digestion of alcohol-distillery wastes," Jour. Mem. Sci., Vol. 90, p. 131-139. 114 Chang LS. and Lee CH. (1998) "Membrane filtration characteristics in membrane coupled activated sludge system - The effect of physiological states of activated sludge on membrane fouling," Desalination, Vol. 120, p. 221. Choate, W.T., Houldsworth, D. and Butler G.A. (1983). "Membrane enhanced anaerobic digestors," Proceedings 37th Industrial Waste Conference, Purdue University, West Lafayette, Indiana, Ann Arbor Science, Ann Arbor, Michigan, p. 661. Choo, K.H., Kang, I.J., Yoon, S.H., Park, H., Kim, J.H., Adlya, S. and Lee, CH. (2000). "Approaches to membrane fouling contro 1 in anaerobic membrane bioreactors," Water Science & Technology, Vol. 41, No. 10-11, p. 363. Choo, K. and Lee, C. (1998) "Hydrodynamic Behaviour of Anaerobic Biosolids During Crossflow Filtration in the Membrane Anaerobic Bioreactor," Water Research, Vol. 32, No. 11, p. 3387. Choo, K.H. and Lee, CH. (1996 a). "Membrane fouling mechanisms in the membrane-coupled anaerobic bioreactor," Water Research, Vol. 30, No. 8, p. 1771. Choo, K.H. and Lee, CH. (1996b). "Effect of Anaerobic Digestion Broth Composition on Membrane Permeability," Water Science & Technology, Vol. 34, No. 9, p. 173. Chu, L., Yang, F. and Zhang, X. (2005) "Anaerobic treatment of domestic wastewater in a membrane-coupled expended granular sludge bed (EGSB) reactor under moderate to low temperature", Process Biochemistry, Vol. 40, p. 1063-1070. Chung Y.C, Jung J.Y., Ahn D.H. and Kim D.H. (1998). "Development of two phase anaerobic reactor with membrane separation system," Journal of Environmental Science and Health A, Vol. 33, p. 249. Collins, A. G , T. L. Theis, S. Kilambi, L. He, and S. G. Paulostathis (1998). "Anaerobic treatment of low-strength domestic wastewater using an anaerobic expanded bed reactor," Journal Environmental Engineering, Vol. 124, (7), p. 652-655. Converti, A., Borghi, M. and Del Ferraiolo, G. (1992). "Influence of organic loading rate on the anaerobic treatment of high strength semisynthetic waste waters in a biological fluidized bed," Chemical Engineering Journal and Biochemical Engineering Journal, Vol. 52, No. 1,, p. B21-B28. DeFrance, L. and Jaffrin, M. (1999). "Comparison between fdtrations at fixed transmembrane pressure and fixed permeate flux: application to a membrane bioreactor used for wastewater treatment," Journal of Membrane Science, Vol. 152, p. 203. Denac, M., Lee, P.L., Newell, R.B. and Greenfield, P.F. (1990). "Automatic control of effluent quality from a high rate anaerobic treatment system," Water Research, Vol. 24, p. 583-586. Draaijer, H., Maas, J.A.W., Schaapman, J.E. and Khan, A., (1991). "Performance of the 5MLD UASB reactor for sewage treatment at Kanpur, India, " in Proc. 6th Int. Symp. Anaerobic Digestion, May, Sao Paulo, Brazil, p. 115-124. Elefsiniotis, P. and Oldham, W.K. (1994), "Effect of HRT on acidogenic digestion of primary sludge". Journal of Environmental Engineering, Vol. 120, p 645-660. Elmaleh, S. and Abdelmoumni, L. (1997). "Cross-flow filtration of an anaerobic methanogenic suspension," Journal of Membrane Science, Vol. 131, p. 261. Elmaleh, S. and Abdelmoumni, L. (1998). "Experimental test to evaluate performance of an anaerobic reactor provided with an external membrane unit," Water Science & Technology, Vol. 38, No. 8-9, p. 385. Elmitwalli, T.A., Oahn, K.L.T., Zeeman, G. and Lettinga, G. (2002). "Treatment of domestic sewage in a two-step anaerobic filter/anaerobic hybrid system at low temperature," Water Research, Vol. 36, p. 2225. Fakhru'l-Razi, A., (1993). "Performance study of a new membrane anaerobic digester configuration," Proceedings of the International Conference on Environmental Management, Geo-Water & Engineering Aspects, p. 115. Fakhru'l-Razi, A., (1995). "Steady states study in a membrane anaerobic system (MAS) for wastewater treatment," Proceedings of the WEF 68th Annual Conference & Exposition, Vol. 3, p. 669. Fane, A.G, Fell, C.J.D. and Suzuki, A. (1983). "The effect of pH and ionic environment on the ultrafiltration of protein solutions with retentive membranes, " Journal of 116 Membrane Science, Vol. 16, p. 195. Fawehinmi, F., Lens, P., Stephenson, T., Rogalla, F. and Jefferson, B. (2004). "The influence of operating conditions on extracellular polymeric substances (EPS) production and biofouling in anaerobic membrane bioreactors," Proc. Water Environment - Membrane Technology Conference, Seoul, South Korea, p. 469. Fongastitkul. P., Mavinic, D.S. and Lo, K.V. (1994). "Two-phased anaerobic digestion process, concept, process failure and maximum system loading rate," Water Environ Res. Vol. 66, p. 243-254. Foresti, E. (2002). "Anaerobic treatment of domestic sewage; established technologies and perspectives," Water Science & Technology, Vol. 45, No. 10, p. 181. Ghyoot, W.R. and Verstraete, W.H. (1997). "Coupling membrane filtration to anaerobic primary sludge digestion," Environmental Technology, Vol. 18, p. 569. Greenberg, A. E., Clesceri, L. S., and Eaton, A. D., 1998, Standard Methods for Examination of Water and Wastewater. 20th Edition, APHA, AWWA and WEF, Washington, DC. Grethlein, H.E. (1978). "Anaerobic digestion and membrane separation of domestic wastewater," Journal WPCF, p. 754. Grin, P., Roeersma, R. and Lettinga, G (1985). "Anaerobic treatment of raw domestic sewage in UASB-reactors at temperatures from 9-20 °C," Proc. Seminar/Workshop on Anaerobic Treatment of Sewage, M. S. Switzenbaum, ed., Amherst, Massachusetts, University of Massachusetts. Guan, Y.T., Jiang, Z.R, Zhu, W.R. and Jin, P. (200 0). "Two-phase anaerobic membrane biosystem for treatment of papermill wastewater," Chinese Journal of Environmental Science, Vol. 19, No. 6, p. 56 (in Chinese) Gujer, W., and Zehnder, A.J.B. (1983) "Pathway of anaerobic digestion," Water Science & Technology 25, 127'. Hall, E.R. (1992). "Anaerobic treatment of wastewaters in suspended growth and fixed film processes," Design of Anaerobic Processes for the Treatment of Industrial and 117 Municipal Wastes, edited by J.F. Malina Jr. and F.G. Pohland, Technomic Publishing Co. Ltd., Lancaster, PA, p. 41. Harada, H., Momonoi, K., Yamazaki, S. and Takizawa, S. (1994). "Application of anaerobic-UF membrane reactor for treatment of a wastewater containing high strength particulate organics," Water Science & Technology, Vol. 30, No. 12, p. 307. Harper, S.R. and Pohland, F.G. (1986). "Recent trends in hydrogen managemenet during anaerobic biological wastewater treatment," biotech. Bioeng., Vol. 28, p. 585-602 He, Y., Li, C. Wu, Z. and Gu, G. (1999). "Study on molecular w eight cut-off of the anaerobic ultratiltration membrane bioreactor," China Water & Wastewater, Vo. 15, No. 9., p. 10. Hernandez, A.E, Belalcazar, L.C. Rodriguez, M.S. and Giraldo, E. (2002). "Retention of Granular Sludge at High Hydraulic Loading Rates in an Anaerobic Membrane Bioreactor with Immersed Filtration," Water Science & Technology, Vol. 45, No. 10, p. 169. Henze, M. and Harremoes, P. (1983). "Anaerobic treatment of wastewater in fixed film reactors - A literature review," Water Science and Technology, Vol. 15, No. 8-9, p. 1-101. Hickey, R.F., Vanderwielen, J., and Swtzenbaum, M.S., (1987). "The effects of organic toxicants on methane production and hydrogen gas levels during the anaerobic digestion of waste activated sludge," Water Research Vol. 21, No.ll, p. 1417-1427. Hogetsu, A., Ishikawa, M., Yoshikawa, T., Tanabe, T., Yudate, S. and Sawada, J. (1992). "High rate anaerobic digestion of wool scouring wastewater in a digester combined with membrane filter," Water Science & Technology, Vol. 25, No. 7, p. 341. 118 Imasaka, T., Kanekuni, N., So, H. and Yoshino, H. (1989). "Cross-flow of methane fermentation broth by ceramic membrane," Journal Fermentation Bioengineering, Vol. 68, p. 200. Jeris, J.S., and McCarty, P.L., (1965). " The biochemistry of methane fermentation using C M tracers," J. WPCF, 37, p. 178-192 Jones, W.J. (1991). "Diversity and physiology of Methanogens," in Microbial production and Consumption of greenhouse Gases: Methane, Nitrogen Oxides, and Halomethanes, J.E. Rogers and W.B. Williams, eds., American Society of Microbiology, p.39-55. Kang, I.J., Yoon, S.H. and Lee, CH. (2002). "Comparison ofthe filtration characteristics of organic and inorganic membranes in a membrane -coupled anaerobic bioreactor," Water Research, Vol. 36, p. 1803. Kataoka, N., Tokiwa, Y., Tanaka, Y. Fujiki, K. Taroda, H. and Takeda, K. (1992). "Examination of Bacterial Characteristics of Anaerobic Membrane Bioreactors in Thee Pilot-Scale Plants for Treating Low-Strength Wastewater by Application of the Colony-Forming-Curve Analysis Method," Applied and Environmental Microbiology, Vo. 58, No. 9, p. 2751. Kennedy, K.J. and van den Berg, L. (1982). "Stability and performance of anaerobic fixd film reactors during hydraulic overloading at 10-35 °C," Water Research, Vol. 16, p. 1391-1398. Kim, J.S., Lee, CH. and Chang, I.S. (2001). "Effect of pump shear on the performance of a crossflow membrane bioreactor," Water Research, Vol. 36, p. 2137. Kim, J-O., Somiya, I., Shin, E-B., Bae, W., Kim, S-K. and Kim, R-H. (2002). "Application of membrane-coupled anaerobic volatile fatty acids fermentor for dissolved organics recovery from coagulated raw sludge," Water Science & Technology, Vol. 45, No. 12, p. 167. Kiriyama, K., Tanaka, Y. and Mori, I. (1992). "Field test of a composite methane gas production system incorporating a membrane module for municipal sewage," Water Science & Technology, Vol. 25, p. 135. 119 Kirmura, S. (1991). "Japans aqua renaissance '90 project," Water Science & Technology, Vol. 23, No. 7-9, p. 1573. Kitamura, Y., Maekawa, T., Tagawa, A., Hayashi, H. and Farrell-Poe, K.L. (1996). "Treatment of strong organic, nitrogenous wastewater by an anaerobic contact process incorporating ultrafiltration," Applied Engineering in Agriculture, Vol. 12, No. 6, p. 709. Kobayashi, H.A., Stenstrom, M.K., and Mah, R.A. (1983) "Treatment of low strength domestic wastewater using the anaerobic filter", Water research Vol.17, p. 903. Laborie, S., Cabassud, C , Durand-Bourlier, L. and Laine, J.M. (1997). "Flux enhancement by a continuous tangential gas flow in ultra-filtration hollow fibres for drinking water production: effects of slug flow on cake structure", Filtration & Separation, 34, 887-891. Laborie, S., Cabassud, C , Durand-Bourlier, L. and Laine, J.M. (1998). "Fouling control by air sparging inside hollow fibre membranes - effects on energy consumption", Desalination, 118, 189-196. Langenhoff A.A.M., Intrachandra N. and Stuckey D.C. (2000) "Treatment of Dilute Soluble and Colloidal Wastewater in an Anaerobic Baffled Reactor: I nfluence of Hydraulic Retention Time" Water Research, Vol. 34, No. 4, p.1307. Langenhoff A.A.M. and Stuckey D.C. (2000) "Treatment of Dilute Wastewater Using an Anaerobic Baffled Reactor: Effect of Low Temperature" Water Research, Vol. 34, No. 15, p.3867. Lee, S., Burt, A., Rusoti, G. and Buckland, B. (1995). " Microfiltration of recombinant yeast cells using a rotating disk dynamic filtration system, " Biotech. & Bioeng., Vol.48, Iss. 4, p. 386-400. Lee, J., Ahn, W-Y. and Lee, C-H. (2001a) "Comparison of the filtration characteristics between attached and suspended growth microorganisms in a submerged membrane bioreactor," Water Research, Vol. 35, p. 2435. 120 Lee, S.M., Jung, J.Y. and Chung, Y.C. (2001 b) "Novel Method for Enhancing Permeate Flux of Submerged Membrane System in Two-Phase Anaerobic Reactor," Water Research, Vol. 35, No. 2, p. 471. Lesjean, B., Rosenberger, S., Laabs, C , Jekel, M., Gnirss, R. and Amy, G. (2005). "Correlation between membrane fouling and soluble/colloidal organic substances in membrane bioreactors for municipal wastewater treatment," Water Science and Technology, Vol. 51, p. 1-8. Lettinga, G., Rosema, R. and Grin, P. (1983) "Anaerobic treatment of raw domestic sewage at ambient temperatures using a granular bed UASB reactor ", Biotechnology bioengineering Vol. 25, pp 1701. Lew, B., Belavski., Admon, S., Tarre, S. and Green, M. (2003). "Temperature effect on UASB reactor operation for domestic wastewater treatment in temperate climate regions," Water Science & Technology, Vol. 48, No. 3, p.25. Liibbecke, S., Vogelpohl, A. and Dewjanin, W. (1995). "Wastewater Treatment in a Biological High-Performance System with High Biomass Concentration," Water Research, Vol. 29, No. 3 p. 793. Malina Jr., J.F. (1992). "Anaerobic sludge digestion," Design of Anaerobic Processes for the Treatment of Industrial and Municipal Wastes, edited by J.F. Malina Jr. and F.G. Pohland, Technomic Publishing Co. Ltd., Lancaster, PA, p. 41. Mallevialle, J., Odendaal, P.E. and Wiesner, M.R. (editors) (1996) "Water treatment membrane processes," McGraw Hill, New York, NY, USA. Manariatis, l.D. and Grigoropoulos, S.G. (2002). "Low-strength wastewater treatment using an anaerobic baffled reactor," Water Environment Research, Vol. 74, No. 2, p. 170. Masoudi, R., Tohidi, B., Danesh, A. and Todd, A.C. (2004). "A new approach in modeling phase equilibria and gas solubility in electrolyte solutions and its applications to gas hydrates," Fluid Phase Equilih., Vol. 215, p.163-174. Naoaki, K., Tokiwa, Y., Tanaka, Y., Fujili, K., Taroda, H. and Takeda, H. (1992). "Examination of bacterial characteristics for anaerobic membrane bioreactors in 121 three pilot-scale plants for treating low-strength wastewater by application of the colony-forming-curve analysis method,"^ /?/?/. Environ. Microbiol., Vol. 58, No. 9, p. 2751-2757. Ndon, U.J. and Dague, R.R. (1997). "Effects of temperature and hydraulic retention time on anaerobic sequencing batch reactor treatment of low -strength wastewater," Water Res., Vol. 31, p. 2455-2466. Perez-Garcia, M., Romero-Garcia, L.I., Rodriguez-Cano, R. and Sales-Marquez, D. (2005). "Effect of the pH influent conditions in fixed-film reactors for anaerobic thermophilic treatment of wine-distillery wastewater," Water Science and Technology, Vol. 51, No. 1, p. 183-189. Pillay, V.L., Townsend, B. and Buckley, C.A. (1994). "Improving the performance of anaerobic digestors at waste water treatment works: the coupled cross -flow microfiltration/digestor process," Proceeding of the 7th International Symposium on Anaerobic Digestion, Cape Town, South Africa, p. 577. Pohland, EG, (1992). "Anaerobic treatment: fundamental concepts, applications, and new horizons," Design of Anaerobic Processes for the Treatment of Industrial and Municipal Wastes, edited by J.F. Malina Jr. and F.G. Pohland, Technomic Publishing Co. Ltd., Lancaster, PA, p. 1 Powell, G E., and Arches, D.B., (1989). "On-line titration method for monitoring buffer capacity and total fatty acid levels in anaerobic digesters," Biotechnology and Bioengineering, Vol. 31, p. 287-294. Renard, P., Van Breusegem, V, Nguyen, M.T., Naveau, H. and Nyns, E.J., (1991). "Implementation of an adaptive controller for the start-up and steady-state running of a biomethanation process operated in the CSTR mode, " Biotechnology and Bioengineering, Vol. 38, p. 805-812. Ross, W.R. and Strohwald, N.K.H. (1994) "Membranes add edge to old technology", Water quality international WQI No. 4, pp. 18-20 Ross, W.R., Strohwald, N.K.H., Grobler, C.J. and Senetra, J. (1994). "Membrane-assisted anaerobic treatment of industrial effluents: The South African ADUF process," 122 Proceeding of the 7th International Symposium on Anaerobic Digestion, Cape Town, South Africa, p. 550. Ross, W.R., Barnard, J.P., Le Roux, J. and Villiers, H. A. (1990). "Application of ultrafiltration membranes for solids-liquid separation in anaerobic digestion systems: The ADUF process," Water Science & Technology, Vol. 16, p. 85. Sainbayar, A., Kim, J.S., Jung, W.J., Lee, Y.S. and Lee, CH. (2001). "Application of Surface Modified Polypropylene Membranes to an Anaerobic Membrane Bioreactor," Environmental Technology, Vol. 22, p. 1035. Sanz, I. and Fdz-Polanco, F. (1990). "Low temperature treatment of municipal sewage in anaerobic fluidized bed reactors," Water Research, Vol. 24, No. 4, p. 463-469. Saw, C.B., Anderson, G.K., James, A. and Le, M.S. (1986). "A membrane technique for biomass retention in anaerobic waste treatment," Proceedings 40th Industrial Waste Conference, Purdue University, West Lafayette, Indiana, Ann Arbor Science, Ann Arbor, Michigan, p. 805. Schellinkhout, A. and Collazos, C.J., (1991). "Full scale application of the UASB technology for sewage treatment," in Proc. 6th Int. Symp. Anaerobic Digestion, May, Sao Paulo, Brazil, p. 145-152. Schiener P., Nachaiyasit S., and Stuckey D.C (1998). "Production of Soluble Microbial Products (SMP) in an Anaerobic Baffled Reactor: Composition, Biodegradability, and the Effect of Process Parameters," Environmental Technology, Vol. 19, No. 4, p. 391. Seghezzo, L., Guerra, R.G., Gonzalez, S.M., Trupiano, A.P., Figueroa, M.E., Cuevad, CM., Aeeman, G. and Lettinga, G. (2002). "Removal efficiency and methanogenic activity profiles in a pilot-scale UASB reactor treating settled sewage at moderate temperatures," Water Science & Technology, Vol. 45, No. 10, p. 243. Shimizu, U., Rokudai, M., Tohya, S., Kauawake, E., Yazawa, T., Tanaka, H. and Eguchi, K. (1989). "Effect of pore size on the filtration characteristics of ceramic membrane for membrane bioreactors," Kagaku Kogaku Ronbunshu, Vol. 15, p. 322. 123 Singh, K.S. and Viraraghavan, T. (2003). "Impact of temperature on performance, microbiological, and hydrodynamic aspects of UASB reactors treating municipal wastewater," Water Science & Technology, Vol. 47, No. 1, p. 139. Stensel, H.D. and Strand, S.E. (2004). "Evaluation of feasibility of methods to minimize biomass production from biotreatment," Water Environment Reasearch Foundation, Project Report 00-CTS-lO (in publication). Stephenson, T., Judd, S., Jefferson, B., and Brindle, K. (20 00). "Membrane bioreactors for wastewater treatment," IWA Publishing., London. U.K. Stuckey D.C. and Hu A. (2003). "The Submerged Anaerobic Membrane Bioreactor (SAMBR): An Intensification of Anaerobic Wastewater Treatment, " Presented at the IWA Leading Edge Conference on Drinking Water and Wastewater Treatment Technologies, May 26-28, Noordwijk/Amsterdam, The Netherlands. Stuckey D.C. (2003). "The Submerged Anaerobic Membrane Bioreactor (SAMBR): An Intensification of Anaerobic Wastewater Treatment", Abstrac t of Presentation to the Dept. Civil Engineering at the University of Minnesota, September 10, Minneapolis, USA. Sutton, P.M., Li, A., Evans R.R. and Korchin S.R. (1983). "Dorr-Oliver's fixed film and suspended growth anaerobic systems for industrial wast ewater treatment and energy recovery," Proceedings 37th Industrial Waste Conference, Purdue University, West Lafayette, Indiana, Ann Arbor Science, Ann Arbor, Michigan, p. 667. Sutton, P.M. and Evans, R.R. (1983). "Anaerobic system designs for efficient tr eatment of industrial wastewater," Proceedings of the 3rd Symposium on Anaerobic Digestion, IAWQ. Sutton, P.M. (1986a). "Advanced anaerobic processes: scale-up results," Proceedings of BioExpo 86, The American Commercial and Industrial Conference and Exposition in Biotechnology, Boston, Massachusetts. Sutton, P.M. (1986b). "Innovative biological systems for anaerobic treatment of grain and food processing wastewater," Starch/Starke, Vol. 38, No. 9, p.314. 124 Sutton, P.M., Berube, P. and Hall, E.R. (2004). "Membrane Bioreactors for Anaerobic Treatment of Wastewaters - Review of Existing Literature," Water Environment Research Foundation, Project 02-CTS-4. Switzenbaum, M.S. and Jewell, W.J. (1980). "Anaerobic attached-film expanded-bed reactor treatment," J. Water Poll. Control Fed., Vol.52, p. 1953-1965. Tchobanoglous, G., Burton, F. and Stensel, D.H. (2003) Wastewater Engineering: Treatment, Disposal & Reuse. Fourth Edition, McGraw-Hill Book Co., New York, USA. Van Houten, R., Evenblij, H. and Keijmel, M. (2001 ). "Membrane bioreactors hit the big time - ten years of research in the Netherlands," STOW A, Foundation for Applied Water Research 2001, H 20, October, p. 26. Vartax, D.R., Engler, C.R., Ricke, S.C., McFarland, M.J., and Byers, F.M. (1994). "Low temperature anaerobic digestion response to organic loading rate and bioaugmentation," Bioresource Technology, Vol. 47, No. 2, p. 173-176 Vazquez, M.I. and Benavente, J. (2003). "A study of temperature effect on chemical, structural and transport parameters determined for two different regenerated cellulose membranes," JMembr Sci., Vol. 219, p. 59-67. Verberk, J.Q.J.C., Hoogeveneen, P.E., Futselaar, H. and Dijk, J.C. (2002). "Hydraulic distribution of water and air over a membrane module using AirFlush ", Water Acience and Technology: Water Supply Vol.2, No. 2, p. 297-304. Visvanathan, C , Ben Aim, R. and Parameshwaran, K. (2000). "Membrane separation bioreactors for waste treatment," Crit. Rev. Environ. Sci. Technol. 30(1), 1-48. Viraraghavan, T. and Kikkeri, S.R. (1990). "Effect of temperature on anaerobic filter treatment of dairy wastewater," Water Science and Technology, Vol. 22, p.191-198. Wang, B., Wei, T. and Yu, Z. (2005). "Effect of operating temperature on component distribution of west indian cherry juice in a microfiltration system," Food Science and Technology, Vol. 38 p. 683-689. 125 Wen, C , Huang, X. and Qian, Y. (1999). "Domestic Wastewater Treatment Using an Anaerobic Bioreactor Coupled with Membrane Filtration," Process Biochemistry, Vol. 35, p. 335. Winkler, M.A., (1981). Biological Treatment of Wastewater, Ellis Horwood Limited, England, P282 Wisniewski, C , and Grasmick, A. (1998). "Floe Size Distribution in a Membrane Bioreactor and Consequences for Membrane Fouling, " Colloids and Surfaces, A: Physicochemical and Engineering Aspects, Vol. 138, p. 403. Yoon, S.H, Kang, I.J. and Lee, CH. (1999). "Fouling of Inorganic Membrane and Flux Enhancement in Membrane-Coupled Anaerobic Bioreactor," Separation Science and Technology, Vol. 34, No. 5, p. 709. Yu, H. and Anderson, G.K. (1996). "Performance of a combined anaerobic reactor for municipal wastewater treatment at ambient temperature," Resour. Conserv. Recycling, Vol. 17 p. 259-271. Yushina, Y. and Hasegawa, J. (1994). "Process performance comparison of membrane introduced anaerobic digestion using food industry wastewater, " Desalination, Vol.98, No. 1-3, p. 413. Zinder, S.H. (1988). "conversion of acetic acid to methane by thermophiles," Anaerobic Digestion, E.R. Hall and P.N. Hobson eds., Pergamon pre ss, London, p.l -12 Zoh, K.D. and Stenstrom, M.K. (2002). "Application of a membrane bioreactor for treating explosives process wastewater," Water Research, Vol. 36, p. 1018. 126 A P P E N D I X A C A L C U L A T I O N OF C O D R E M O V A L E F F I C I E N C Y COD removal efficiency = (The influent COD - The effluent COD)/ The influent COD Calculation example: Using data taken in Phase 3 (Day 42 to Day 91) as example The influent COD = The average influent COD ± confident interval = 500 ± 35 mg/L The effluent COD = The average effluent COD ± confident interval = The average of three membrane effluents COD ± confident interval = 81 ± 5 mg/L (500 ±35) - (81 ±5) COD removal efficiency = * 100% (500 ±35) = (84 ± 7)% 127 A P P E N D I X B C A L C U L A T I O N S F O R D I S S O L V E D M E T H A N E Methane solubility calculation at 32' C and 25' C At a temperature of 32" C and atmospheric pressure CH4 concentration in liquid/mole fraction = 1.65* 10"5 (Masoudi et al., 2004) CH 4 concentration = 1.65*10"5*16 g/18 g = 1.47*10"5 g CH4/g H 20 = 1.47*10 2 gCH4/L H 20 Density of CH4= 1.48 L/g The solubility of CH 4 (L CH 4 /L H20) = 1.48* 1.47*10"2 = 2.7*10"2 (L CH 4 /LH 2 0) At a temperature of 25' C and atmospheric pressure CH 4 concentration in liquid/mole fraction = 1.85*10 "5 (Masoudi et al., 2004) Calculated with the same method as the above The solubility of CH4= 2.43*10"2 (L CH 4/ L H20) Calculation for revised biogas production in terms of m3/kg COD removed at 32' C Revised biogas production A (L/d) = Measured gas production (L/d)+ 2.7* 10"2* flow rate (L/d) /methane percentage (%) Revised biogas production B (m3/kg COD removed) =A*1000/(influent TCOD-effluent COD)(mg/L)/flow rate (L/d) Calculation for revised methane production in terms of m 3 CBU/kg COD removed Measured methane production C (m3 CH4/kg COD removed) = Measured gas production (L/d)* 1000/(influent TCOD - effluent COD ) (mg/L)/flow rate (L/d)*methane percentage(%) Revised methane production D (m3 CH4/kg COD removed) = C + 0.0217*1000/(influent TCOD - effluent COD) (mg/L) 128 APPENDIX C EXPERIMENTAL DATA Table C-l System operational conditions Date Operational days Influent flow rate (L/day) HRT (hours) OLR (kgCOD/m3«day) Temperature (°C) pH Influent Effluent 31-Aug-04 31.0 7.00 l-Sep-04 27.5 7.40 2-Sep-04 29.0 7.50 3-Sep-04 30.0 7.40 4-Sep-04 28.9 7.36 5-Sep-04 29.0 7.40 6-Sep-04 7-Sep-04 29.0 7.30 8-Sep-04 29.0 7.35 9-Sep-04 10-Sep-04 1 10 61 26.0 7.50 ll-Sep-04 2 10 61 30.5 7.35 12-Sep-04 3 10 61 28.7 • 7.28 13-Sep-04 4 10 61 27.7 7.29 14-Sep-04 5 10 61 0.23 30.1 7.40 15-Sep-04 6 10 61 27.7 7.40 16-Sep-04 7 10 61 30.4 7.30 17-Sep-04 8 10 61 0.17 29.8 7.25 18-Sep-04 9 10 61 30.2 7.10 19-Sep-04 10 10 61 30.5 7.30 20-Sep-04 11 10 61 0.20 21-Sep-04 12 10 61 0.24 29.1 7.30 22-Sep-04 13 10 61 28.4 7.10 23-Sep-04 14 10 61 30.0 7.00 24-Sep-04 15 10 61 0.24 32.3 7.20 25-Sep-04 16 10 61 31.9 7.00 26-Sep-04 17 10 61 32.2 7.00 27-Sep-04 18 25 24 0.54 33.1 6.80 28-Sep-04 19 25 24 36.9 7.10 29-Sep-04 20 25 24 31.7 7.10 30-Sep-04 21 25 24 29.4 129 Date Operational days Influent flow rate (L/day) HRT (hours) OLR (kgCOD/m3«day) Temperature (°C) pH Influent Effluent l-Oct-04 22 25 24 32.3 7.10 2-Oct-04 23 25 24 0.57 32.2 7.10 3-Oct-04 24 25 24 31.9 7.00 4-Oct-04 25 25 24 0.54 30.1 7.10 5-Oct-04 26 25 24 34.0 7.10 6-Oct-04 27 25 24 0.60 33.1 6.80 7-Oct-04 28 25 24 31.5 7.00 8-Oct-04 29 25 24 0.60 33.5 7.00 9-Oct-04 30 25 24 30.2 7.00 10-Oct-04 31 25 24 0.51 33.4 7.00 ll-Oct-04 32 25 24 31.7 7.00 12-Oct-04 33 25 24 0.59 33.0 7.00 13-Oct-04 34 25 24 31.2 7.00 14-Oct-04 35 25 24 31.9 7.20 15-Oct-04 36 25 24 0.51 34.3 7.00 16-Oct-04 37 25 24 0.47 33.3 7.00 17-Oct-04 38 25 24 0.54 33.6 7.00 18-Oct-04 39 25 24 0.55 32.1 7.00 19-Oct-04 40 25 24 0.56 29.0 7.00 20-Oct-04 41 25 24 29.0 7.00 21-Oct-04 42 40 15 33.0 7.00 22-Oct-04 43 40 15 0.74 31.7 7.00 23-Oct-04 44 40 15 0.89 32.5 7.20 24-Oct-04 45 40 15 0.89 31.8 7.10 25-Oct-04 46 40 15 1.03 30.5 7.10 26-Oct-04 47 40 15 1.01 30.6 7.20 27-Oct-04 48 40 15 31.1 7.00 28-Oct-04 49 40 15 29.9 6.90 29-Oct-04 50 40 . 15 0.80 31.4 6.90 30-Oct-04 51 40 15 1.08 30.2 6.90 31-Oct-04 52 40 15 0.82 31.3 6.90 1-Nov-04 53 40 15 0.89 31.7 6.90 2-Nov-04 54 40 15 0.86 30.3 6.90 3-Nov-04 55 40 15 30.7 7.00 4-Nov-04 56 40 15 30.8 7.00 5-Nov-04 57 40 15 0.84 29.3 7.00 6-Nov-04 58 40 15 1.01 29.9 7.10 7-Nov-04 59 40 15 0.89 31.9 7.00 130 Date Operational days Influent flow rate • (L/day) HRT (hours) OLR (kgCOD/m3«day) Temperature (°C) pH Influent Effluent 8-Nov-04 60 40 15 1.16 31.7 7.00 9-Nov-04 61 40 15 0.78 32.9 6.90 lO-Nov-04 62 40 15 ' 0.57 33.8 7.10 ll-Nov-04 63 40 15 . 33.8 7.00 12-Nov-04 64 40 15 0.74 33.2 7.00 13-Nov-04 65 40 15 0.58 30.9 7.00 14-Nov-04 66 40 15 0.79 30.7 7.00 15-Nov-04 67 40 15 0.64 31.6 7.00 16-Nov-04 68 40 15 0.89 32.3 7.00 17-NOV-04 69 40 15 33.1 7.10 18-Nov-04 70 40 15 32.0 7.00 19-Nov-04 71 40 15 0.81 30.4 7.00 20-Nov-04 72 40 15 0.50 30.2 6.80 21-Nov-04 73 40 15 0.84 29.7 7,00 22-Nov-04 74 40 15 0.54 29.3 7.00 23-Nov-04 75 40 15 0.79 29.3 6.80 24-Nov-04 76 40 . 15 0.63 29.5 6.90 25-Nov-04 77 40 15 30.7 7.20 26-Nov-04 78 40 15 32.6 7.30 7.20 27-Nov-04 79 40 15 31.5 7.20 28-Nov-04 80 40 . 15 0.81 29.8 7.60 7.20 29-Nov-04 81 40 15 0.57 29.4 7.10 30-Nov-04 82 40 15 0.78 29.3 7.60 7.20 l-Dec-04 83 40 15 0.61 29.2 7.30 7.20 2-Dec-04 84 40 15 29.4 7.40 7.20 3-Dec-04 85 40 15 0.84 29.4 7.20 7.20 4-Dec-04 . 86 40 15 29.3 7.23 7.15 5-Dec-04 87 40 15 0.67 33.1 7.20 7.20 6-Dec-04 88 40 15 28.9 7.20 7-Dec-04 89 40 15 0.75 29.1 7.10 8-Dec-04 90 40 15 0.67 29.0 7.10 9-Dec-04 91 40 15 29.9 7.30 10-Dec-04 92 40 13 0.88 31.2 7.40 7.20 1 l-Dec-04 93 40 13 33.4 7.30 7.20 12-Dec-04 94 40 13 0.90 33.1 7.40 7.20 13-Dec-04 95 40 13 32.4 7.10 14-Dec-04 96 40 13 0.96 34.7 7.60 7.20 15-Dec-04 97 40 13 0.80 36.4 7.10 131 Date Operational days Influent flow rate (L/day) HRT (hours) OLR (kgCOD/m3«day) Temperature (°C) pH Influent Effluent 16-Dec-04 98 40 13 34.2 7.40 7.20 17-Dec-04 99 40 13 0.76 31.0 7.10 18-Dec-04 100 40 13 32.6 19-Dec-04 101 40 13 1.19 20-Dec-04 102 40 13 1.18 36.6 7.30 7.20 21-Dec-04 103 40 13 0.87 36.7 7.10 22-Dec-04 104 40 13 1.10 23-Dec-04 105 40 13 36.2 7.20 24-Dec-04 106 40 13 36.2 7.10 25-Dec-04 107 35.5 7.20 26-Dec-04 108 50 8 1.30 39.0 7.50 27-Dec-04 109 50 8 39.8 7.00 28-Dec-04 110 50 8 1.50 40.6 7.45 7.30 29-Dec-04 111 50 8 38.7 7.70 7.25 30-Dec-04 112 50 8 1.61 36.9 7.70 7.23 31-Dec-04 113 50 8 1.33 39.0 7.10 l-Jan-05 114 50 8 1.50 39.8 7.60 7.20 2-Jan-05 115 50 8 38.7 7.70 7.10 3-Jan-05 116 50 8 38.1 7.60 7.10 4-Jan-05 117 50 8 1.61 36.3 7.70 7.10 5-Jan-05 118 50 8 36.3 7.70 7.20 6-Jan-05 119 50 8 1.59 38.6 7.30 7.15 7-Jan-05 120 50 8 1.34 40.0 7.40 7.15 8-Jan-05 121 50 8 40.8 7.50 7.21 9-Jan-05 122 50 8 1.41 41.5 7.30 7.10 10-Jan-05 123 50 8 41.3 7.70 7.10 ll-Jan-05 124 50 8 1.60 38.8 7.15 12-Jan-05 125 50 8 1.34 40.2 7.30 13-Jan-05 126 50 8 40.3 7.30 14-Jan-05 127 50 8 40.8 7.30 7.16 15-Jan-05 128 50 8 1.67 38.7 7.30 7.20 16-Jan-05 129 50 8 1.74 35.6 7.20 7.00 17-Jan-05 130 50 8 35.9 7.10 18-Jan-05 131 50 8 1.43 37.8 7.70 7.10 19-Jan-05 132 50 8 1.33 39.4 7.20 7.10 20-Jan-05 133 50 8 40.2 7.00 7.20 21-Jan-05 134 50 8 1.24 41.6 7.10 22-Jan-05 135 50 8 1.74 35.7 7.10 7.20 132 Date Operational days Influent flow rate (L/day) HRT (hours) OLR (kgCOD/m3«day) Temperature (°C) pH Influent Effluent 23-Jan-05 136 50 8 36.5 7.30 24-Jan-05 • 137 50 8 36.5 7.40 7.00 25-Jan-05 138 50 8 1.76 37.3 7.40 7.00 26-Jan-05 139 50 8 36.5 7.50 7.30 27-Jan-05 140 50 8 1.39 7.30 28-Jan-05 141 50 8 1.33 7.30 7.10 29-Jan-05 142 50 8 1.86 7.40 7.25 30-Jan-05 143 50 8 1.52 7.30 31-Jan-05 144 50 8 33.3 7.20 l-Feb-05 145 50 9 1.47 32.4 7.60 7.10 2-Feb-05 146 50 9 1.57 31.9 7.33 7.40 3-Feb-05 147 44 10 1.10 31.4 4-Feb-05 148 37.7 7.20 5-Feb-05 149 44 10 1.31 33.7 7.25 7.10 6-Feb-05 150 44 10 1.07 31.9 7.24 7.10 7-Feb-05 151 44 10 28.0 7.20 8-Feb-05 152 44 10 1.22 28.3 7.30 9-Feb-05 153 44 10 1.12 26.5 10-Feb-05 154 44 9 1.25 31.0 7.30 1l-Feb-05 155 44 9 1.17 29.0 7.21 12-Feb-05 156 44 9 1.28 27.9 7.54 7.15 13-Feb-05 157 44 9 1.06 28.0 7.63 7.17 14-Feb-05 158 44 9 30.2 7.82 7.20 15-Feb-05 159 44 9 28.1 7.28 16-Feb-05 160 44 9 1.19 28.1 7.38 17-Feb-05 161 44 9 1.15 27.9 7.40 7.07 18-Feb-05 162 44 9 0.99 28.1 19-Feb-05 163 45 9 1.46 29.0 7.60 7.12 20-Feb-05 164 43 9 1.11 30.4 7.60 7.16 2l-Feb-05 165 43 9 30.8 22-Feb-05 166 43 9 1.28 30.3 8.21 7.50 23-Feb-05 167 43 9 1.11 29.7 7.40 24-Feb-05 168 42 10 0.89 29.9 7.50 7.27 25-Feb-05 169 41 10 30.6 7.46 7.27 26-Feb-05 170 41 10 31.3 7.53 7.13 27-Feb-05 171 41 10 32.2 7.30 28-Feb-05 172 40 10 29.5 7.98 7.48 l-Mar-05 173 40 10 1.43 30.1 133 Date Operational days Influent flow rate (L/day) HRT (hours) OLR (kgCOD/m3«day) Temperature (°C) pH Influent Effluent 2-Mar-05 174 40 10 1.10 30.1 7.63 7.40 3-Mar-05 175 38 11 1.09 30.4 7.76 7.41 4-Mar-05 176 38 11 0.96 31.2 7.43 7.17 5-Mar-05 177 38 11 1.15 31.3 7.28 6-Mar-05 178 38 11 0.93 31.6 7.87 7.25 7-Mar-05 179 37 11 31.5 7.58 8-Mar-05 180 37 11 1.13 31.8 • 7.78 7.32 9-Mar-05 181 37 11 1.05 32.6 10-Mar-05 182 36 11 0.84 32.1 7.47 7.29 11-Mar-05 183 36 11 1.22 32.4 12-Mar-05 184 30.4 7.00 7.38 13-Mar-05 185 50 8 1.29 32.2 7.78 7.19 14-Mar-05 186 50 8 30.1 7.97 7.26 15-Mar-05 187 50 8 29.6 16-Mar-05 188 50 8 1.30 30.2 7.36 17-Mar-05 189 50 8 30.7 7.47 7.15 18-Mar-05 190 50 8 31.0 19-Mar-05 191 50 1.41 31.0 7.53 7.16 20-Mar-05 192 50 8 1.37 31.3 7.79 7.14 21-Mar-05 193 50 8 31.2 7.59 7.14 22-Mar-05 194 50 8 1.89 30.6 7.14 23-Mar-05 195 50 8 1.72 27.8 7.75 7.25 24-Mar-05 196 50 8 1.61 26.3 7.22 25-Mar-05 197 50 8 1.43 26.2 7.48 7.11 26-Mar-05 198 50 8 1.92 27.9 7.40 7.21 27-Mar-05 199 50 8 1.36 29.8 7.15 7.10 2 8-Mar-05 200 50 8 28.2 7.92 7.23 29-Mar-05 201 50 8 1.61 29.4 7.62 30-Mar-05 202 50 8 1.89 27.7 7.63 7.10 31-Mar-05 203 50 8 1.72 28.6 7.90 7.45 1-Apr-05 204 50 8 25.0 7.50 7.29 2-Apr-05 205 50 8 1.70 25.0 7.60 7.30 3-Apr-05 206 50 8 1.60 25.0 7.37 7.18 4-Apr-05 207 50 8 25.0 7.36 7.00 5-Apr-05 208 50 8 1.39 25.1 7.20 7.17 6-Apr-05 209 50 8 24.9 7.35 7.20 7-Apr-05 210 50 8 1.32 24.9 7.15 7.04 8-Apr-05 211 50 8 1.62 24.9 134 Date Operational Influent HRT OLR Temperature pH days flow rate (L/day) (hours) (kgCOD/m3«day) (°C) Influent Effluent 9-Apr-05 212 50 8 1.36 25.1 7.20 10-Apr-05 213 50 8 1.11 25.0 ll-Apr-05 214 50 8 24.9 7.20 12-Apr-05 215 88 5 25.0 7.00 6.50 13-Apr-05 216 108 4 1.85 24.8 6.70 6.60 14-Apr-05 217 108 4 24.9 7.00 6.70 15-Apr-05 218 108 4 26.7 6.90 16-Apr-05 219 108 4 25.0 6.80 17-Apr-05 220 108 4 1.77 24.4 7.00 6.70 18-Apr-05 221 108 4 24.9 6.90 6.30 19-Apr-05 222 108 4 1.64 25.1 6.90 6.50 20-Apr-05 223 108 4 24.9 21-Apr-05 224 108 4 25.0 6.60 7.20 22-Apr-05 225 108 4 24.9 7.50 6.90 23-Apr-05 226 108 4 25.0 7.00 6.80 24-Apr-05 227 108 4 25.1 7.10 6.70 25-Apr-05 228 108 4 25.4 7.20 6.60 26-Apr-05 229 108 4 34.0 27-Apr-05 230 108 4 28.9 7.00 6.60 28-Apr-05 231 108 4 1.56 26.0 7.20 7.00 29-Apr-05 232 108 4 25.0 7.10 6.80 30-Apr-05 233 108 4 25.0 6.80 l-May-05 234 108 4 1.47 25.0 7.20 6.80 2-May-05 235 108 4 25.0 7.20 6.80 3-May-05 236 108 4 25.0 4-May-05 237 108 4 1.58 24.9 7.10 6.50 5-May-05 238 108 4 25.1 7.10 6.80 6-May-05 239 108 4 1.79 25.0 7.10 7.00 7-May-05 240 108 4 25.0 7.40 6.80 8-May-05 241 108 4 1.87 25.0 7.40 6.80 9-May-05 242 108 4 25.0 7.60 6.80 10-May-05 243 108 4 25.0 1 l-May-05 244 108 4 25.0 7.20 ' 6.80 12-May-05 245 108 4 25.1 7.10 6.60 13-May-05 246 108 4 1.60 25.0 14-May-05 247 143 2 25.0 7.3.0 6.80 15-May-05 248 143 •2 25.0 7.40 6.90 16-May-05 249 143 2 25.1 7.50 6.70 135 Date Operational days Influent flow rate (L/day) HRT (hours) OLR (kgCOD/m3«day) Temperature (°C) pH Influent Effluent 17-May-05 250 143 2 25.1 18-May-05 251 143 2 2.87 25.1 7.00 6.70 19-May-05 252 143 2 25.0 7.50 7.00 20-May-05 253 143 2 2.92 24.9 7.20 6.70 21-May-05 254 143 2 24.9 7.10 6.80 22-May-05 255 143 2 2.82 24.9 7.40 6.80 23-May-05 256 143 2 24.9 24-May-05 257 143 2 25.0 7.40 6.50 25-May-05 258 143 2 25.0 7.30 6.70 26-May-05 259 143 2 25.1 7.90 6.90 27-May-05 260 143 2 2.89 25.0 7.10 6.80 28-May-05 261 143 2 25.1 7.30 7.00 29-May-05 262 143 2 2.74 25.1 30-May-05 263 143 2 26.3 7.50 6.70 31-May-05 264 143 2 26.7 l-Jun-05 265 143 2 2.94 25.2 7.10 6.40 2-Jun-05 266 143 2 25.0 7.10 6.80 3-Jun-05 267 143 2 2.82 25.1 7.10 6.80 4-Jun-05 268 143 2 25.0 5-Jun-05 269 143 2 25.0 7.10 6.80 6-Jun-05 270 143 2 25.0 7-Jun-05 271 143 2 25.0 7.40 6.70 8-Jun-05 272 143 2 3.07 25.0 7.10 6.50 9-Jun-05 273 143 2 25.0 7.00 6.60 10-Jun-05 274 143 2 2.53 ll-Jun-05 275 143 2 25.0 12-Jun-05 276 143 2 2.79 25.1 6.90 6.40 13-Jun-05 277 143 2 14-Jun-05 278 143 2 15-Jun-05 279 143 2 25.0 7.10 6.70 16-Jun-05 280 143 2 25.0 7.10 6.50 136 Table C-2 COD measurement results Date Operational days COD Measurements (mg/L) Influent Effluent Infl. NA Sinf. NA Inf. A Sinfl. A PCI Mem. Koch Ave. COD 31-Aug-04 l-Sep-04 2-Sep-04 ' 3-Sep-04 4-Sep-04 5-Sep-04 6-Sep-04 . 7-Sep-04 8-Sep-04 383 190 608 515 588 480 9-Sep-04 10-Sep-04 1 ll-Sep-04 2 . 12-Sep-04 3 13-Sep-04 4 14-Sep-04 5 264 73 589 398 125 171 148 15-Sep-04 6 16-Sep-04 7 17-Sep-04 8 225 125 423 343 115 161 138 18-Sep-04 9 19-Sep-04 10 20-Sep-04 11 295 188 517 380 133 145 139 21-Sep-04 12 408 210 595 480 70 110 90 22-Sep-04 13 23-Sep-04 14 24-Sep-04 15 328 175 618 385 25-Sep-04 16 65 100 83 26-Sep-04 17 27-Sep-04 18 342 265 550 435 28-Sep-04 19 29-Sep-04 20 83 98 91 30-Sep-04 21 l-Oct-04 22 2-Oct-04 23 300 180 573 418 137 Date Operational days COD Measurements (mg/L) Influent Effluent Infl. NA Sinf. NA Inf. A Sinfl. A PCI Mem. Koch Ave. COD 3-Oct-04 24 73 65 69 4-Oct-04 25 318 233 545 445 5-Oct-04 26 85 68 77 6-Oct-04 27 300 160 603 213 7-Oct-04 28 73 65 69 8-Oct-04 29 485 205 605 428 9-Oct-04 30 10-Oct-04 31 285 130 513 345 ll-Oct-04 32 75 83 79 12-Oct-04 33 325 133 600 253 13-Oct-04 34 75 90 83 14-Oct-04 35 15-Oct-04 36 248 113 518 298 16-Oct-04 37 478 210 83 88 86 17-Oct-04 38 285 150 543 340 18-Oct-04 39 555 220 63 108 86 19-Oct-04 40 358 140 565 310 20-Oct-04 41 203 103 85 94 21-Oct-04 42 22-Oct-04 43 338 115 465 325 23-Oct-04 44 563 285 98 63 81 24-Oct-04 45 360 125 565 343 25-Oct-04 46 648 275 58 88 73 26-Oct-04 47 300 135 635 275 27-Oct-04 48 240 93 85 89 28-Oct-04 49 29-Oct-04 50 298 123 505 350 30-Oct-04 51 680 305 65 90 78 31-Oct-04 52 233 120 515 330 l-Nov-04 53 563 435 55 55 55 2-Nov-04 54 148 540 370 3-Nov-04 55 80 90 85 4-Nov-04 56 5-Nov-04 57 305 128 528 365 6-Nov-04 58 640 283 70 92 81 7-Nov-04 59 305 125 565 383 8-Nov-04 60 730 298 65 65 65 138 Date Operational days COD Measurements (mg/L) Influent Effluent Infl. NA Sinf. NA Inf. A Sinfl. A PCI Mem. Koch Ave. COD 9-Nov-04 61 258 100 493 308 lO-Nov-04 62 363 300 78 98 88 ll-Nov-04 63 12-Nov-04 64 388 113 468 388 13-Nov-04 65 368 398 75 88 82 14-Nov-04 66 250 90 500 338 15-Nov-04 67' 403 255 80 83 82 16-Nov-04 68 290 110 565 308 17-Nov-04 69 65 70 68 18-Nov-04 70 19-Nov-04 71 293 173 513 330 20-Nov-04 72 318 255 83 70 77 21-Nov-04 73 328 163 528 340 22-Nov-04 74 338 250 75 75 75 23-Nov-04 75 323 170 500 328 24-Nov-04 76 398 333 115 88 102 25-Nov-04 77 26-Nov-04 78 27-Nov-04 79 83 88 86 28-Nov-04 80 353 95 510 348 29-Nov-04 81 363 310 83 90 87 30-Nov-04 82 295 175 495 400 l-Dec-04 83 383 340 98 75 87 2-Dec-04 84 3-Dec-04 85 295 175 528 313 4-Dec-04 86 75 78 77 5-Dec-04 87 213 115 423 350 6-Dec-04 88 7-Dec-04 89 243 143 473 370 8-Dec-04 90 420 313 103 103 9-Dec-04 91 10-Dec-04 92 228 145 488 338 1 l-Dec-04 93 12-Dec-04 94 498 260 13-Dec-04 95 14-Dec-04 96 310 162 530 468 15-Dec-04 97 445 328 80 93 87 139 Date Operational days COD Measurements (mg/L) Influent Effluent Infl. NA Sinf. NA Inf. A Sinfl. A PCI Mem. Koch Ave. COD 16-Dec-04 98 17-Dec-04 99 285 130 420 148 83 130 107 18-Dec-04 100 19-Dec-04 101 658 460 2.0-Dec-04 102 368 215 650 368 103 132 118 21-Dec-04 103 310 103 . 480 318 22-Dec-04 104 610 313 73 60 67 23-Dec-04 105 24-Dec-04 106 320 140 75 75 25-Dec-04 107 310 138 538 343 26-Dec-04 108 438 338 65 103 78 82 27-Dec-04 109 28-Dec-04 110 290 145 503 310 29-Dec-04 111 60 60 30-Dec-04 112 310 113 540 348 31-Dec-04 113 448 348 70 78 70 73 l-Jan-05 114 268 113 505 330 2-Jan-05 115 55 83 50 63 3-Jan-05 116 4-Jan-05 117 285 128 540 348 5-Jan-05 118 80 100 70 83 6-Jan-05 119 298 293 535 365 7-Jan-05 120 450 333 55 73 70 66 8-Jan-05 121 305 110 348 9-Jan-05 122 473 368 63 78 68 70 10-Jan-05 123 ll-Jan-05 124 363 150 538 348 12-Jan-05 125 450 345 88 100 98 95 13-Jan-05 126 14-Jan-05 127 15-Jan-05 128 348 143 560 360 16-Jan-05 129 585 395 108 123 105 112 17-Jan-05 130 18-Jan-05 131 443 125 480 338 19-Jan-05 132 448 398 113 120 117 20-Jan-05 133 21-Jan-05 134 415 365 130 123 140 131 140 Date Operational days COD Measurements (mg/L) Influent Effluent Infl. NA Sinf. NA Inf. A Sinfl. A PCI Mem. Koch Ave. COD 22-Jan-05 135 370 115 583 335 23-Jan-05 136 83 128 128 113 24-Jan-05 137 25-Jan-05 138 345 145 590 378 26-Jan-05 139 695 410 85 135 125 115 27-Jan-05 140 420 280 468 348 28-Jan-05 141 445 355 55 70 58 61 29-Jan-05 142 333 120 623 373 30-Jan-05 143 510 305 60 48 54 31-Jan-05 144 l-Feb-05 145 333 115 558 328 2-Feb-05 146 595 330 85 95 90 3-Feb-05 147 340 110 473 333 4-Feb-05 148 515 78 85 82 5-Feb-05 149 333 115 563 355 6-Feb-05 150 463 313 80 80 7-Feb-05 151 8-Feb-05 152 353 528 335 9-Feb-05 153 485 348 95 108 102 10-Feb-05 154 265 103 478 300 1l-Feb-05 155 447 323 78 120 99 12-Feb-05 156 308 100 488 310 13-Feb-05 157. 403 310 14-Feb-05 158 15-Feb-05 159 340 115 530 253 16-Feb-05 160 455 295 128 148 138 17-Feb-05 161 280 98 440 318 18-Feb-05 162 378 303 110 88 99 19-Feb-05 163 328 105 545 328 20-Feb-05 164 435 335 80 83 82 2l-Feb-05 165 22-Feb-05 166 285 105 498 303 23-Feb-05 167 433 328 53 48 51 24-Feb-05 168 160 58 355 268 25-Feb-05 169 43 45 44 26-Feb-05 170 27-Feb-05 171 70 60 65 141 Date Operational days COD Measurements (mg/L) Influent Effluent Infl. NA Sinf. NA Inf. A Sinfl. A PCI Mem. Koch Ave. COD 28-Feb-05 172 l-Mar-05 173 295 105 598 323 2-Mar-05 174 460 333 58 68 63 3-Mar-05 175 288 130 483 325 4-Mar-05 176 423 310 68 63 66 5-Mar-05 177 260 115 508 333 6-Mar-05 178 412 303 85 78 82 7-Mar-05 179 8-Mar-05 180 313 138 513 345 95 70 83 9-Mar-05 181 475 340 10-Mar-05 182 195 77 390 265 70 88 79 1 l-Mar-05 183 570 250 12-Mar-05 184 288 160 493 365 13-Mar-05 185 433 305 14-Mar-05 186 15-Mar-05 187 16-Mar-05 188 230 103 435 413 65 65 65 17-Mar-05 189 18-Mar-05 190 19-Mar-05 191 273 123 473 310 20-Mar-05 192 460 285 63 78 71 2 l-Mar-05 193 22-Mar-05 194 423 130 633 317 23-Mar-05 195 578 388 73 63 68 24-Mar-05 196 430 115 542 25-Mar-05 197 480 330 83 83 26-Mar-05 198 338 108 645 330 27-Mar-05 199 455 360 80 50 65 28-Mar-05 200 29-Mar-05 201 348 120 540 333 70 55 63 30-Mar-05 202 635 470 68 70 69 3 l-Mar-05 203 295 100 578 330 l-Apr-05 204 95 58 7^7 2-Apr-05 205 343 118 570 323 3-Apr-05 206 538 343 83 70 77 4-Apr-05 207 5-Apr-05 208 328 153 465 308 142 Date Operational days COD Measurements (mg/L) Influent Effluent Infl. NA. Sinf. NA Inf. A Sinfl. A PCI Mem. Koch Ave. COD 6-Apr-05 209 83 80 82 7-Apr-05 210 285 85 443 320 8-Apr-05 211 543 315 95 75 85 9-Apr-05 212 208 70 455 350 10-Apr-05 213 373 343 83 83 11-Apr-05 214 12-Apr-05 215 13-Apr-05 216 288 70 87 78.5 14-Apr-05 217 15-Apr-05 218 103 65 90 77.5 16-Apr-05 219 17-Apr-05 220 275 165 68 60 64 18-Apr-05 221 19-Apr-05 222 255 93 60 48 54 20-Apr-05 223 21-Apr-05 224 22-Apr-05 225 23-Apr-05 226 370 105 65 50 58 24-Apr-05 227 25-Apr-05 228 26-Apr-05 229 27-Apr-05 230 28-Apr-05 231 243 128 90 80 85 29-Apr-05 232 30-Apr-05 233 1-May-05 234 228 100 78 68 73 2-May-05 235 3-May-05 236 4-May-05 237 245 110 123 105 114 5-May-05 238 6-May-05 239 278 103 120 110 115 7-May-05 240 8-May-05 241 290 133 113 95 104 9-May-05 242 10-May-05 243 11-May-05 244 355 130 110 100 105 12-May-05 245 143 Date Operational days COD Measurements (mg/L) Influent Effluent Infl. NA Sinf. NA Inf. A Sinfl. A PCI Mem. Koch Ave. COD 13-May-05 246 248 75 133 98 116 14-May-05 247 15-May-05 248 358 133 150 150 16-May-05 249 17-May-05 250 18-May-05 251 293 235 155 138 147 19-May-05 252 20-May-05 253 298 130 138 118 128 2 l-May-05 254 22-May-05 255 288 113 115 90 103 23-May-05 256 24-May-05 257 25-May-05 258 26-May-05 259 27-May-05 260 295 150 135 163 149 28-May-05 261 29-May-05 262 280 193 130 108 119 30-May-05 263 3 l-May-05 264 l-Jun-05 265 300 200 118 120 119 2-Jun-05 266 3-Jun-05 267 288 138 180 125 153 4-Jun-05 268 5-Jun-05 269 383 138 153 153 153 6-Jun-05 270 7-Jun-05 271 8-Jun-05 272 313 155 157 125 141 9-Jun-05 273 10-Jun-05 274 258 105 ll-Jun-05 275 12-Jun-05 276 285 113 118 128 123 13-Jun-05 277 14-Jun-05 278 15-Jun-05 279 143 140 142 144 Table C-2 Biogas production and compositions Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition COj/CFL, % Ratio C0 2 o 2 N 2 CH 4 C0 2 CH 4 31-Aug-04 l-Sep-04 7.10 2-Sep-04 5.15 3-Sep-04 3.74 4-Sep-04 4.08 5-Sep-04 6-Sep-04 7-Sep-04 2.56 8-Sep-04 9-Sep-04 10-Sep-04 1 4.72 ll-Sep-04 2 2.34 12-Sep-04 3 1.84 13-Sep-04 4 14-Sep-04 5 2.79 0.78 0.27 8.1% 1.2% 56.3% 34.4% 19.0% 81.0% 15-Sep-04 6 1.90 16-Sep-04 7 3.53 17-Sep-04 8 1.46 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C02/CH4 % Ratio C0 2 o2 N 2 CH 4 co2 CH 4 18-Sep-04 9 1.21 19-Sep-04 10 20-Sep-04 11 21-Sep-04 12 1.51 22-Sep-04 13 2.04 23-Sep-04 14 1.00 24-Sep-04 15 1.73 25-Sep-04 16 1.98 26-Sep-04 17 0.35 27-Sep-04 18 2.00 4.2% 0.8% 58.1% 36.9% 10.3% 89.7% 28-Sep-04 19 0.84 29-Sep-04 20 0.95 30-Sep-04 21 0.00 l-Oct-04 22 3.91 2-Oct-04 23 2.99 3.2% 0.9% 42.8% 53.1% 5.7% 94.3% 3-Oct-04 24 2.68 4-Oct-04 25 3.07 3.9% 0.7% 31.5% 64.0% 5.7% 94.3% 5-Oct-04 26 3.27 6-Oct-04 27 2.43 3.6% 0.6% 26.6% 69.2% 5.0% 95.0% 7-Oct-04 28 2.77 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 o 2 N 2 CH4 C0 2 CH 4 8-Oct-04 29 2.91 9-Oct-04 30 2.27 10-Oct-04 31 3.60 2.9% 1.9% 21.1% 74.1% 3.8% 96.2% ll-Oct-04 32 2.54 0.30 0.24 3.1% 0.4% 17.1% 79.4% 3.7% 96.3% 12-Oct-04 33 2.96 13-Oct-04 34 3.9% 0.5% 15.3% 80.4% 4.6% 95.4% 14-Oct-04 35 15-Oct-04 36 3.60 16-Oct-04 37 2.91 0.35 0.29 3.7% 0.5% 13.1% 82.7% 4.2% 95.8% 17-Oct-04 38 3.59 18-Oct-04 39 3.24 0.34 0.28 3.4% 0.4% 13.0% 83.2% 3.9% 96.1% 19-Oct-04 40 2.46 20-Oct-04 41 2.34 0.25 0.21 3.0% 0.8% 14.1% 82.1% 3.5% 96.5% 21-Oct-04 42 3.40 22-Oct-04 43 3.82 23-Oct-04 44 4.56 0.32 0.28 3.0% 0.0% 11.1% 85.9% 3.3% 96.7% 24-Oct-04 45 3.96 25-Oct-04 46 3.71 0.22 0.19 2.4% 0.5% 12.4% 84.7% 2.7% 97.3% 26-Oct-04 47 27-Oct-04 48 3.58 0.21 0.18 2.5% 0.3% 13.6% 83.7% 2.9% 97.2% Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition CO2/CH4 % Ratio C0 2 o 2 N 2 CH 4 C0 2 CH 4 28-Oct-04 49 3.72 29-Oct-04 50 4.09 30-Oct-04 51 3.72 0.23 0.19 2.8% 0.3% 13.0% 83.9% 3.2% 96.8% 31-Oct-04 52 4.16 l-Nov-04 53 4.45 0.28 0.24 2.6% 0.5% 12.0% 84.8% 3.0% 97.0% 2-Nov-04 54 4.53 3-Nov-04 55 4.54 0.30 0.27 2.7% 0.3% 9.9% 87.2% 3.0% 97.1% 4-Nov-04 56 3.80 5-Nov-04 57 3.84 6-Nov-04 58 4.60 0.28 0.24 2.4% 0.4% 10.0% 87.1% 2.7% 97.4% 7-Nov-04 59 4.27 8-Nov-04 60 4.04 9-Nov-04 61 4.20 lO-Nov-04 62 4.61 0.41 0.36 3.1% 0.2% 9.0% 87.7% 3.4% 96.6% ll-Nov-04 63 3.80 12-Nov-04 64 4.03 13-Nov-04 65 4.15 0.38 0.33 2.6% 0.3% 9.6% 87.5% 2.9% 97.2% 14-Nov-04 66 4.15 15-Nov-04 67 4.01 0.34 0.29 2.8% 0.4% 9.9% 86.9% 3.1% 96.9% 16-Nov-04 68 4.68 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 o 2 N 2 CH 4 C0 2 CH 4 17-Nov-04 69 5.12 0.31 0.27 3.1% 0.3% 9.0% 87.6% 3.4% 96.6% 18-Nov-04 70 4.06 19-Nov-04 71 3.50 20-Nov-04 72 4.07 0.37 0.33 2.8% 0.3% 9.8% 87.2% 3.1% 96.9% 21-Nov-04 73 3.74 22-Nov-04 74 3.83 0.34 0.29 2.6% 0.6% 11.4% 85.4% 3.0% 97.0% 23-Nov-04 75 3.91 24-Nov-04 76 3.80 0.35 0.30 2.7% 0.5% 10.2% 86.5% 3.0% 97.0% 25-Nov-04 77 4.33 26-Nov-04 78 4.27 27-Nov-04 79 3.95 2.6% 0.3% 10.1% 86.6% 3.4% 96.6% 28-Nov-04 80 5.04 29-Nov-04 81 3.94 30-Nov-04 82 4.14 1-Dec-04 83 4.35 2.7% 1.6% 12.6% 83.2% 3.1% 96.9% 2-Dec-04 84 4.55 3-Dec-04 85 4.27 4-Dec-04 86 4.44 0.30 0.26 2.8% 0.4% 9.5% 87.3% 3.1% 96.9% 5-Dec-04 87 6-Dec-04 88 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 o 2 N 2 CH 4 C0 2 CH 4 7-Dec-04 89 8-Dec-04 90 3.72 0.36 2.9% 0.2% 27.2% 69.6% 4.0% 96.0% 9-Dec-04 91 4.97 10-Dec-04 92 6.88 11-Dec-04 93 0.91 12-Dec-04 94 6.05 13-Dec-04 95 4.30 14-Dec-04 96 4.44 15-Dec-04 97 5.22 0.39 0.34 3.4% 0.2% 9.5% 86.9% 3.8% 96.2% 16-Dec-04 98 3.67 17-Dec-04 99 5.88 18-Dec-04 100 1.46 3.1% 0.2% 13.4% 83.3% 3.5% 96.5% 19-Dec-04 101 20-Dec-04 102 5.00 0.28 0.23 4.0% 0.3% 13.4% 82.3% 4.7% 95.3% 21-Dec-04 103 6.05 22-Dec-04 104 7.90 23-Dec-04 105 4.43 24-Dec-04 106 5.43 3.4% 0.2% 28.4% 68.0% 4.8% 95.3% 25-Dec-04 107 5.41 26-Dec-04 108 5.11 0.34 3.2% 0.2% 33.7% 63.0% 4.8% 95.2% Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 o2 N 2 CH 4 co2 CH 4 27-Dec-04 109 4.99 28-Dec-04 110 5.38 29-Dec-04 111 4.90 0.29 3.5% 0.2% 19.6% 76.7% 4.4% 95.6% 30-Dec-04 112 4.96 3 l-Dec-04 113 5.96 0.35 0.28 3.4% 0.0% 15.0% 81.7% 4.0% 96.0% l-Jan-05 114 5.60 2-Jan-05 115 5.45 0.31 0.26 3.3% 0.2% 12.6% 83.8% 3.8% 96.2% 3-Jan-05 116 6.36 4-Jan-05 117 5.59 5-Jan-05 118 1.96 5.4% 0.0% 10.4% 84.2% 6.0% 94.0% 6-Jan-05 119 7.00 7-Jan-05 120 5.90 0.34 0.28 3.8% 0.0% 14.6% 81.7% 4.4% 95.6% 8-Jan-05 121 6.22 9-Jan-05 122 6.56 0.39 0.34 3.7% 0.3% 9.4% 86.6% 4.1% 95.9% 10-Jan-05 123 5.78 ll-Jan-05 124 4.90 12-Jan-05 125 5.97 0.36 0.31 3.6% 0.3% 9.2% 86.9% 4.0% 96.0% 13-Jan-05 126 6.14 14-Jan-05 127 6.00 4.9% 0.0% 8.8% 86.3% 5.4% 94.6% 15-Jan-05 128 5.29 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 o 2 N 2 CH4 co 2 CH4 16-Jan-05 129 4.55 0.25 0.22 4.1% 0.0% 8.6% 87.3% 4.5% 95.5% 17-Jan-05 130 5.33 18-Jan-05 131 5.62 19-Jan-05 132 6.37 0.44 0.38 3.3% 0.3% 9.0% 87.4% 3.6% 96.4% 20-Jan-05 133 6.08 21-Jan-05 134 5.89 22-Jan-05 135 4.60 23-Jan-05 136 4.57 0.25 0.22 3.0% 0.2% 9.1% 87.7% 3.3% 96.7% 24-Jan-05 137 5.51 25-Jan-05 138 5.22 26-Jan-05 139 7.34 27-Jan-05 140 6.15 28-Jan-05 141 6.95 0.42 0.34 3.2% 0.2% 14.2% 82.5% 3.7% 96.3% 29-Jan-05 142 5.62 30-Jan-05 143 5.37 0.26 0.22 3.5% 0.2% 12.3% 84.0% 4.0% 96.0% 31-Jan-05 144 6.21 l-Feb-05 145 4.37 2-Feb-05 146 4.50 0.24 0.20 2.6% 0.3% 12.1% 84.9% 3.0% 97.0% 3-Feb-05 147 4.52 4-Feb-05 148 7.85 3.3% 0.3% 16.6% 79.8% 4.0% 96.0% Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgC0D removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 o 2 N 2 CH4 C0 2 CH 4 5-Feb-05 149 4.79 6-Feb-05 150 7.19 0.44 0.36 3.1% 0.2% 14.4% 82.3% 3.6% 96.4% 7-Feb-05 151 5.18 8-Feb-05 152 4.95 -9-Feb-05 153 4.26 0.30 0.25 2.5% 0.3% 13.1% 84.1% 2.9% 97.1% 10-Feb-05 154 2.28 1l-Feb-05 155 4.86 0.37 0.32 2.4% 0.3% 12.7% 84.6% 2.8% 97.3% 12-Feb-05 156 4.27 13-Feb-05 157 4.27 2.3% 0.3% 12.6% 84.8% 2.6% 97.4% 14-Feb-05 158 4.80 15-Feb-05 159 4.37 16-Feb-05 160 3.22 17-Feb-05 161 1.41 18-Feb-05 162 3.79 19-Feb-05 163 3.57 20-Feb-05 164 4.75 0.34 0.27 3.5% 0.3% 15.7% 80.6% 4.1% 95.9% 2l-Feb-05 165 5.33 22-Feb-05 166 4.81 23-Feb-05 167 4.40 0.31 0.26 2.7% 0.4% 12.7% 84.2% 3.1% 96.9% 24-Feb-05 168 4.59 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C02/CH4 % Ratio C0 2 o 2 N 2 CH 4 C0 2 CH 4 25-Feb-05 169 5.02 0.48 0.40 3.2% 0.3% 11.9% 84.6% 3.7% 96.4% 26-Feb-05 170 4.91 27-Feb-05 171 1.6% 12.4% 63.1% 23.0% 6.6% 93.4% 28-Feb-05 172 3.86 2.2% 0.3% 77.9% 16.6% 10.0% 90.0% 1-Mar-05 173 3.04 2-Mar-05 174 3.50 0.30 3.0% 0.2% 55.9% 40.9% 6.9% 93.1% 3-Mar-05 175 4.64 4-Mar-05 176 4.24 0.38 3.6% 0.6% 38.1% 57.8% 5.8% 94.2% 5-Mar-05 177 4.59 6-Mar-05 178 4.27 0.38 3.5% 0.3% 29.7% 66.4% 5.0% 95.0% 7-Mar-05 179 4.38 8-Mar-05 180 4.32 9-Mar-05 181 4.79 3.6% 0.4% 22.3% 73.7% 4.6% 95.4% 10-Mar-05 182 4.39 11-Mar-05 183 4.47 12-Mar-05 184 2.79 13-Mar-05 185 5.42 3.5% 0.5% 18.0% 78.0% 4.3% 95.7% 14-Mar-05 186 4.79 15-Mar-05 187 5.12 16-Mar-05 188 4.30 0.31 0.24 3.8% 0.2% 16.3% 79.8% 4.5% 95.5% Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 0 2 N2 CH 4 co 2 CH 4 17-Mar-05 189 5.99 18-Mar-05 190 5.36 3.2% 0.5% 14.7% 81.6% 3.8% 96.2% 19-Mar-05 191 5.10 20-Mar-05 192 5.67 0.35 0.29 3.1% 0.5% 13.9% 82.5% 3.6% 96.4% 2 l-Mar-05 193 5.68 22-Mar-05 194 6.03 23-Mar-05 195 4.18 0.20 0.18 2.8% 0.4% 12.7% 84.1% 3.2% 96.8% 24-Mar-05 196 5.13 25-Mar-05 197 5.39 0.31 0.27 2.6% 0.4% 12.4% 84.6% 3.0% 97.0% 26-Mar-05 198 5.58 27-Mar-05 199 6.09 28-Mar-05 200 5.24 29-Mar-05 201 6.08 30-Mar-05 202 6.11 0.31 0.19 2.8% 0.4% 36.7% 59.9% 4.4% 95.6% 3 l-Mar-05 203 5.68 l-Apr-05 204 4.95 0.36 1.5% 0.3% 71.5% 26.6% 5.3% 94.7% 2-Apr-05 205 4.47 3-Apr-05 206 5.32 1.9% 0.2% 51.2% 46.7% 3.9% 96.1% 4-Apr-05 207 4.43 5-Apr-05 208 5.09 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 o 2 N 2 CH 4 C0 2 CH 4 6-Apr-05 209 4.42 0.32 2.1% 0.4% 36.5% 61.1% 3.3% 96.7% 7-Apr-05 210 5.18 8-Apr-05 211 4.44 0.29 0.21 2.2% 0.3% 28.0% 69.5% 3.1% 96.9% 9-Apr-05 212 5.49 10-Apr-05 213 4.21 0.34 0.26 2.1% 0.4% 24.9% 72.6% 2.8% 97.2% 11-Apr-05 214 4.65 12-Apr-05 215 1.46 13-Apr-05 216 1.54 14-Apr-05 217 15-Apr-05 218 2.10 2.3% 0.5% 33.3% 64.0% 3.4% 96.6% 16-Apr-05 219 1.70 17-Apr-05 220 1.59 0.24 0.16 2.2% 0.6% 37.6% 59.7% 3.5% 96.5% 18-Apr-05 221 2.97 19-Apr-05 222 2.42 0.32 0.18 2.2% 3.1% 43.0% 51.7% 4.1% 95.9% 20-Apr-05 223 2.63 21-Apr-05 224 1.55 22-Apr-05 225 1.06 23-Apr-05 226 1.26 0.15 0.10 2.6% 0.8% 37.4% 59.2% 4.3% 95.8% 24-Apr-05 227 3.92 25-Apr-05 228 1.62 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 0 2 N 2 CH 4 C0 2 CH 4 26-Apr-05 229 6.27 27-Apr-05 230 2.05 28-Apr-05 231 1.72 0.33 0.21 2.0% 0.6% 38.4% 59.1% 3.3% 96.7% 29-Apr-05 232 1.38 30-Apr-05 233 1.44 l-May-05 234 1.04 0.34 0.19 1.7% 1.5% 46.8% 50.0% 3.2% 96.8% 2-May-05 235 1.08 3-May-05 236 1.06 4-May-05 237 2.64 0.50 0.29 2.5% 0.6% 43.5% 53.4% 4.5% 95.6% 5-May-05 238 1.87 6-May-05 239 3.35 0.46 0.24 2.5% 0.8% 47.5% 49.2% 4.8% 95.2% 7-May-05 240 1.60 8-May-05 241 1.76 9-May-05 242 1.23 10-May-05 243 1.69 1 l-May-05 244 1.73 0.23 0.13 2.5% 0.6% 45.8% 51.0% 4.7% 95.3% 12-May-05 245 1.18 13-May-05 246 1.08 0.42 0.22 2.5% 1.1% 48.7% 47.7% 5.0% 95.0% 14-May-05 247 1.18 15-May-05 248 1.32 0.12 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C0 2/CH 4 % Ratio C0 2 0 2 N 2 CH4 C0 2 CH 4 16-May-05 249 1.06 17-May-05 250 1.40 18-May-05 251 0.92 0.35 0.19 2.5% 0.6% 49.0% 48.0% 4.9% 95.1% 19-May-05 252 0.85 20-May-05 253 1.34 0.36 0.17 2.4% 1.5% 54.4% 41.8% 5.4% 94.6% 2 l-May-05 254 0.39 22-May-05 255 0.69 0.28 0.14 2.4% 0.5% 51.3% 45.7% 5.1% 94.9% 23-May-05 256 1.16 24-May-05 257 1.32 25-May-05 258 1.48 26-May-05 259 0.14 27-May-05 260 0.25 0.36 0.17 1.6% 0.0% 55.6% 42.9% 3.6% 96.4% 28-May-05 261 0.82 29-May-05 262 0.82 30-May-05 263 0.53 3 l-May-05 264 0.60 l-Jun-05 265 0.44 2-Jun-05 266 1.10 2.5% 0.0% 61.6% 36.0% 6.4% 93.6% 3-Jun-05 267 0.91 4-Jun-05 268 0.95 Date Operational days Gas production rate (L/d) Revised gas prod'n (m3/kgCOD removed) Revised methane prod'n (m3/kgCOD removed) Gas Analysis Percent Composition C02/CH4 % Ratio C0 2 o 2 N 2 CH 4 C0 2 CH 4 5-Jun-05 269 1.54 0.41 0.12 1.9% 6.1% 66.4% 25.6% 6.9% 93.1% 6-Jun-05 270 1.11 7-Jun-05 271 1.25 8-Jun-05 272 0.70 9-Jun-05 273 1.15 10-Jun-05 274 ll-Jun-05 275 0.68 12-Jun-05 276 1.83 2.5% 0.0% 82.8% 14.7% 14.6% 85.4% 13-Jun-05 277 14-Jun-05 278 15-Jun-05 279 2.6% 0.0% 71.2% 26.2% 9.0% 91.0% Table C-3 The concentrations of VFAs in influent and effluent Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 21-Sep-04 12 202.538 4.170 15.358 3.615 0.000 0.000 0.000 0.000 22-Sep-04 13 23-Sep-04 14 24-Sep-04 15 25-Sep-04 16 26-Sep-04 17 27-Sep-04 18 212.962 5.270 15.326 3.410 0.000 0.000 0.000 0.000 28-Sep-04 19 29-Sep-04 20 30-Sep-04 21 l-Oct-04 22 2-Oct-04 23 95.387 2.643 20.695 5.254 5.487 0.000 4.629 0.000 3-Oct-04 24 4-Oct-04 25 5-Oct-04 26 37.591 9.582 6-Oct-04 27 5.842 0.000 0.000 2.355 0.000 7-Oct-04 28 8-Oct-04 29 338.102 4.792 13.771 4.166 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 9-Oct-04 30 4.250 0.000 5.217 0.000 10-Oct-04 31 ll-Oct-04 32 12-Oct-04 33 142.964 3.978 14.478 2.224 3.314 0.000 1.905 0.000 13-Oct-04 34 14-Oct-04 35 15-Oct-04 36 10.291 1.654 16-Oct-04 37 143.349 5.633 0.000 1.856 0.000 17-Oct-04 38 18-Oct-04 39 19-Oct-04 40 226.897 6.551 20.975 3.193 20-Oct-04 41 97.746 9.660 2.295 1.759 2.491 1.197 21-Oct-04 42 22-Oct-04 43 172.790 2.638 10.914 2.312 23-Oct-04 44 143.003 4.508 5.778 0.000 3.264 0.000 24-Oct-04 45 25-Oct-04 46 26-Oct-04 47 115.162 5.801 10.961 2.222 27-Oct-04 48 69.182 8.655 2.972 1.807 2.239 2.014 28-Oct-04 49 29-Oct-04 50 204.903 2.049 5.531 1.131 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 30-Oct-04 51 186.029 3.874 4.031 0.000 3.026 0.000 31-Oct-04 52 214.271 2.938 13.826 . 2.232 1-Nov-04 53 112.866 2.548 1.072 0.000 0.717 0.000 2-Nov-04 54 271.306 2.736 14.567 2.409 3-Nov-04 55 4.225 0.000 4.712 0.000 4-Nov-04 56 5-Nov-04 57 216.143 3.385 12.252 1.999 6-Nov-04 58 133.327 3.974 2.467 0.000 1.831 0.000 7-Nov-04 59 199.728 3.871 9.772 1.731 8-Nov-04 60 133.681 4.087 3.535 0.000 2.342 0.000 9-Nov-04 61 180.268 1.351 3.748 0.000 lO-Nov-04 62 167.827 2.642 2.926 0.000 3.293 0.000 11-Nov-04 63 12-Nov-04 64 213.500 2.564 10.820 1.838 13-Nov-04 65 177.478 3.137 2.232 0.000 1.655 0.000 14-Nov-04 66 198.877 2.083 2.218 0.000 15-Nov-04 67 110.258 2.228 0.927 0.000 1.503 0.000 16-Nov-04 68 150.177 1.297 0.000 17-Nov-04 69 1.627 0.000 1.158 0.000 18-Nov-04 70 19-Nov-04 71 204.809 1.651 4.828 0.000 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 20-Nov-04 72 132.691 2.197 1.311 0.000 0.000 0.000 21-Nov-04 73 191.420 2.369 7.778 1.388 22-Nov-04 74 130.353 2.841 2.622 1.161 23-Nov-04 75 168.236 2.418 1.903 0.000 24-Nov-04 76 167.076 2.384 3.394 0.000 2.001 0.000 25-Nov-04 77 26-Nov-04 78 27-Nov-04 79 98.124 1.320 7.636 0.000 2.035 0.000 28-Nov-04 80 203.738 2.690 9.882 1.900 29-Nov-04 81 158.503 2.871 2.564 0.000 2.559 0.000 30-Nov-04 82 156.616 1.967 2.048 l-Dec-04 83 179.117 2.130 1.178 0.000 2.409 0.000 2-Dec-04 84 3-Dec-04 85 187.166 1.484 7.514 1.305 4-Dec-04 86 0.820 0.000 5-Dec-04 87 194.068 2.914 9.764 6-Dec-04 88 7-Dec-04 89 203.065 1.908 9.004 1.552 0.695 0.000 8-Dec-04 90 187.889 1.384 2.359 0.000 9-Dec-04 91 10-Dec-04 92 206.619 4.164 11.839 1.144 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 11-Dec-04 93 12-Dec-04 94 124.531 1.346 1.143 0.000 0.650 0.000 13-Dec-04 95 14-Dec-04 96 192.932 1.789 7.594 0.624 15-Dec-04 97 2.018 0.000 5.258 0.000 16-Dec-04 98 17-Dec-04 99 199.213 1.977 8.353 0.000 0.000 0.000 0.000 0.000 18-Dec-04 100 19-Dec-04 101 280.323 2.930 20-Dec-04 102 196.112 4.903 48.581 2.067 3.196 0.000 0.000 0.000 21-Dec-04 103 144.945 2.140 2.188 0.000 22-Dec-04 104 93.687 6.313 0.000 0.000 1.974 0.000 23-Dec-04 105 24-Dec-04 106 11.718 1.665 0.000 0.000 0.000 0.000 1.472 0.000 25-Dec-04 107 190.920 2.245 9.970 1.448 26-Dec-04 108 190.593 4.038 6.129 0.000 1.113 27-Dec-04 109 10.504 1.700 28-Dec-04 110 29-Dec-04 111 199.681 3.197 11.380 1.681 2.202 0.000 5.883 0.000 30-Dec-04 112 219.070 2.300 31-Dec-04 113 204.600 3.050 0.000 0.000 0.000 0.000 4.049 0.000 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic l-Jan-05 114 2-Jan-05 115 208.924 2.630 10.464 1.954 3.900 0.000 15.924 0.000 6.396 0.000 3-Jan-05 116 4-Jan-05 117 225.212 3.400 13.700 1.800 5-Jan-05 118 4.349 0.000 17.666 0.662 8.333 0.000 6-Jan-05 119 7-Jan-05 120 8-Jan-05 121 217.471 3.338 13.700 1.902 9-Jan-05 122 207.492 4.796 0.000 0.000 0.000 0.000 0.000 0.000 10-Jan-05 123 ll-Jan-05 124 177.776 3.887 12-Jan-05 125 188.998 4.635 0.000 0.000 0.000 0.000 0.000 0.000 13-Jan-05 126 14-Jan-05 127 15-Jan-05 128 220.891 3.149 12.958 16-Jan-05 129 11.052 0.000 13.880 0.000 15.184 0.000 17-Jan-05 130 18-Jan-05 131 155.742 2.029 19-Jan-05 132 154.041 3.410 0.000 8.645 0.000 7.732 0.000 20-Jan-05 133 2 l-Jan-05 134 156.617 11.072 2.911 4.323 0.000 11.574 0.000 6.940 0.000 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 22-Jan-05 135 23-Jan-05 136 24-Jan-05 137 25-Jan-05 138 198.104 3.515 11.303 2.284 26-Jan-05 139 188.477 4.081 8.104 0.000 9.375 0.000 9.277 0.000 27-Jan-05 140 28-Jan-05 141 29-Jan-05 142 217.431 4.014 14.432 2.562 30-Jan-05 143 199.499 4.086 10.029 0.000 6.399 0.000 31-Jan-05 144 l-Feb-05 145 172.562 4.219 10.179 2.101 12.445 1.827 15.001 0.000 2-Feb-05 146 210.784 6.053 3-Feb-05 147 199.633 5.685 0.000 0.000 4-Feb-05 148 3.795 0.000 6.298 1.966 7.549 0.000 5-Feb-05 149 189.915 3.014 14.176 2.865 6-Feb-05 150 210.676 4.776 0.000 0.000 10.382 0.000 7-Feb-05 151 8-Feb-05 152 198.172 5.869 15.387 3.652 9-Feb-05 153 209.702 4.381 19.782 0.000 19.383 0.000 10-Feb-05 154 193.621 1.382 1 l-Feb-05 155 208.370 3.612 12.113 0.000 12.993 0.000 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 12-Feb-05 156 202.013 2.555 10.281 1.732 13-Feb-05 157 204.056 1.813 16.611 0.000 14.785 0.000 14_Feb-05 158 15-Feb-05 159 114.891 2.078 16-Feb-05 160 136.464 5.793 6.812 0.000 6.582 0.000 17-Feb-05 161 194.236 1.674 18-Feb-05 162 211.609 3.666 14.250 0.000 13.689 0.000 19-Feb-05 163 209.195 5.199 9.753 1.719 20-Feb-05 164 203.406 2.762 11.356 0.000 7.982 0.000 2l-Feb-05 165 22-Feb-05 166 179.670 1.449 2.517 23-Feb-05 167 227.400 3.873 5.452 0.000 5.107 0.000 24-Feb-05 168 175.853 2.024 25-Feb-05 169 12.237 0.000 0.000 0.000' 26-Feb-05 170 27-Feb-05 171 28-Feb-05 172 l-Mar-05 173 175.906 3.373 5.662 1.773 2-Mar-05 174 206.252 7.121 0.000 0.000 0.000 0.000 3-Mar-05 175 166.133 3.765 7.185 0.000 4-Mar-05 176 173.494 4.732 0.000 0.000 0.000 0.000 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 5-Mar-05 177 201.156 4.115 6.797 0.000 3.366 0.000 0.000 0.000 6-Mar-05 178 169.857 5.066 7-Mar-05 179 8-Mar-05 180 166.782 14.777 3.394 9-Mar-05 181 192.139 7.131 4.086 0.000 0.000 0.000 10-Mar-05 182 185.447 3.415 4.600 0.000 0.000 0.000 11-Mar-05 183 101.586 12-Mar-05 184 195.306 3.509 4.261 0.000 13-Mar-05 185 177.079 6.629 3.563 0.000 2.150 0.000 14-Mar-05 186 15-Mar-05 187 1.908 0.000 16-Mar-05 188 202.819 3.409 13.038 2.876 0.000 0.000 0.000 0.000 17-Mar-05 189 173.713 2.928 15.288 2.761 18-Mar-05 190 165.411 5.563 3.475 0.000 1.963 0.000 19-Mar-05 191 181.160 2.238 11.688 3.017 20-Mar-05 192 131.057 0.309 0.000 21-Mar-05 193 22-Mar-05 194 185.401 5.035 24.834 6.133 23-Mar-05 195 173.474 7.555 4.707 2.400 6.084 0.000 24-Mar-05 196 195.703 4.110 10.395 4.110 25-Mar-05 197 9.628 0.000 17.980 4.800 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 26-Mar-05 198 182.684 5.187 8.645 2.291 27-Mar-05 199 199.440 8.762 11.212 0.000 6.592 0.000 28-Mar-05 200 29-Mar-05 201 196.916 2.970 14.390 2.672 9.884 0.000 8.656 0.000 30-Mar-05 202 175.588 5.192 3.038 0.000 2.797 0.000 3 l-Mar-05 203 218.003 0.686 5.660 0.653 l-Apr-05 204 3.016 0.000 3.645 0.000 2-Apr-05 205 203.404 2.030 8.837 1.840 3-Apr-05 206 204.400 4.016 8.577 1.457 8.461 1.782 4-Apr-05 207 5-Apr-05 208 203.377 3.884 17.253 3.496 6-Apr-05 209 7.136 1.049 9.124 1.762 7-Apr-05 210 198.588 3.694 11.733 2.256 8-Apr-05 211 6.849 0.000 9.554 0.000 9-Apr-05 212 203.097 2.545 10-Apr-05 213 189.147 1.214 4.690 0.000 6.145 0.000 1 l-Apr-05 214 12-Apr-05 215 13-Apr-05 216 14.322 1.449 0.000 0.000 0.000 0.000 14-Apr-05 217 15-Apr-05 218 6.671 0.000 0.000 0.000 0.130 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 16-Apr-05 219 17-Apr-05 220 6.343 0.711 1.534 0.000 1.507 0.427 18-Apr-05 221 19-Apr-05 222 6.015 0.633 3.228 1.309 3.672 1.791 20-Apr-05 223 21-Apr-05 224 22-Apr-05 225 23-Apr-05 226 14.166 0.085 3.628 0.000 24-Apr-05 227 25-Apr-05 228 26-Apr-05 229 27-Apr-05 230 28-Apr-05 231 11.050 2.866 5.838 2.574 6.692 4.489 29-Apr-05 232 30-Apr-05 233 1-May-05 234 11.960 0.775 11.202 2.603 10.501 2.898 2-May-05 235 3-May-05 236 4-May-05 237 15.730 2.297 16.243 4.377 16.828 5.777 5-May-05 238 6-May-05 239 29.464 3.957 26.011 3.501 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 7-May-05 240 8-May-05 241 11.452 1.673 16.027 10.092 18.636 4.808 9-May-05 242 10-May-05 243 1 l-May-05 244 19.539 3.291 25.274 6.049 25.425 5.336 12-May-05 245 13-May-05 246 27.277 5.674 27.000 5.129 14-May-05 247 15-May-05 248 16.612 2.914 26.106 5.908 27.743 5.075 16-May-05 249 17-May-05 250 18-May-05 251 15.858 1.273 36.381 8.266 19-May-05 252 20-May-05 253 12.210 0.000 24.052 5.407 2 l-May-05 254 22-May-05 255 15.594 2.739 31.928 8.152 30.229 7.403 23-May-05 256 24-May-05 257 25-May-05 258 26-May-05 259 27-May-05 260 14.590 1.520 30.654 6.914 30.118 8.331 Date Operational days Influent VFAs (mg/L) Effluent VFAs (mg/L) With acetate No acetate PCI Membralox Koch Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic Acetic Propionic 28-May-05 261 29-May-05 262 17.380 3.910 39.020 8.620 40.720 8.430 30-May-05 263 31-May-05 264 l-Jun-05 265 15.410 3.740 37.210 11.790 33.350 10.350 2-Jun-05 266 3-Jun-05 267 23.760 3.070 34.770 6.980 35.620 6.910 4-Jun-05 268 5-Jun-05 269 17.750 2.460 34.320 8.460 31.890 8.560 6-Jun-05 270 7-Jun-05 271 8-Jun-05 272 21.520 3.960 39.980 9.770 40.820 8.930 9-Jun-05 273 10-Jun-05 274 16.370 2.160 ll-Jun-05 275 12-Jun-05 276 25.640 2.800 35.910 11.440 36.870 8.220 13-Jun-05 277 14-Jun-05 278 15-Jun-05 279 42.310 10.590 42.800 10.910 Table C-5 External membranes' permeate flux and permeability Date Days PCI Membralox Koch2 Koch Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi 25-Sep-04 16 37 1.22 106 3.55 26-Sep-04 17 27-Sep-04 18 38 1.28 106 3.55 27-Sep-04 19 28-Sep-04 20 37 1.22 106 3.55 28-Sep-04 21 29-Sep-04 22 37 1.22 106 3.55 30-Sep-04 23 l-Oct-04 24 37 1.22 106 3.55 2-Oct-04 25 3-Oct-04 26 4-Oct-04 27 38 1.26 106 3.55 5-Oct-04 28 6-Oct-04 29 35 1.16 97 3.23 7-Oct-04 30 8-Oct-04 31 35 1.16 126 4.19 9-Oct-04 32 10-Oct-04 33 ll-Oct-04 34 12-Oct-04 35 13-Oct-04 36 14-Oct-04 37 15-Oct-04 38 30 1.00 97 3.23 16-Oct-04 39 17-Oct-04 40 18-Oct-04 41 37 1.24 87 2.90 19-Oct-04 42 20-Oct-04 43 31 1.04 87 2.90 21-Oct-04 44 22-Oct-04 45 29 0.96 87 2.90 23-Oct-04 46 24-Oct-04 47 25-Oct-04 48 26 0.88 87 2.90 26-Oct-04 49 27-Oct-04 50 25 0.82 87 2.90 28-Oct-04 51 173 Date Days PCI Membralox Koch2 Koch Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi 29-Oct-04 52 28 0.94 87 2.90 30-Oct-04 53 31-Oct-04 54 1-Nov-04 55 22 0.72 2-Nov-04 56 3-Nov-04 57 23 0.76 68 2.26 4-Nov-04 58 5-Nov-04 59 26 0.86 77 2.58 6-Nov-04 60 7-Nov-04 61 8-Nov-04 62 23 0.76 68 2.26 9-Nov-04 63 lO-Nov-04 64 23 0.76 68 2.26 11-Nov-04 65 12-Nov-04 66 21 0.70 68 2.26 13-Nov-04 67 14-Nov-04 68 15-Nov-04 69 23 0.76 77 2.58 16-Nov-04 70 17-Nov-04 71 22 0.72 68 2.26 18-Nov-04 72 19-Nov-04 73 19 0.64 68 2.26 20-Nov-04 74 21-Nov-04 75 22-Nov-04 76 20 0.68 68 2.26 23-Nov-04 77 24-Nov-04 78 20 0.66 68 2.26 25-Nov-04 79 26-Nov-04 80 20 0.68 68 2.26 27-Nov-04 81 28-Nov-04 82 29-Nov-04 83 20 0.66 58 1.94 30-Nov-04 84 1-Dec-04 85 2-Dec-04 86 3-Dec-04 87 4-Dec-04 88 5-Dec-04 89 29 0.98 58 1.94 174 Date Days PCI Membralox Koch2 Koch Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi 6-Dec-04 90 21 0.70 7-Dec-04 91 20 0.66 8-Dec-04 92 16 0.52 9-Dec-04 93 10-Dec-04 94 1 l-Dec-04 95 12-Dec-04 96 13-Dec-04 97 15 0.50 48 1.61 14-Dec-04 98 15-Dec-04 99 38 68 2.26 16-Dec-04 100 16 0.52 68 2.26 17-Dec-04 101 55 1.83 126 4.19 18-Dec-04 102 45 1.50 145 4.84 19-Dec-04 103 25 0.83 135 4.52 20-Dec-04 104 19 0.63 155 5.16 2 l-Dec-04 105 17 0.57 155 5.16 22-Dec-04 106 14 0.35 203 5.08 40 1.00 23-Dec-04 107 14 0.35 184 4.60 34 0.86 24-Dec-04 108 15 0.38 242 6.05 46 1.14 25-Dec-04 109 13 0.33 223 5.56 46 1.14 26-Dec-04 110 11 0.28 213 5.32 33 0.83 27-Dec-04 111 10 0.25 194 4.84 43 1.07 28-Dec-04 112 10 0.25 184 4.60 40 1.00 29-Dec-04 113 10 0.25 194 4.84 46 1.14 30-Dec-04 114 3 l-Dec-04 115 8 0.20 184 4.60 40 1.00 l-Jan-05 116 8 0.20 174 4.35 40 1.00 2-Jan-05 117 8 0.19 174 4.35 40 1.00 3-Jan-05 118 4-Jan-05 119 7 0.18 160 3.99 46 1.14 5-Jan-05 120 7 0.16 155 3.87 43 1.07 6-Jan-05 121 7 0.16 160 3.99 46 1.14 7-Jan-05 122 8-Jan-05 123 7 0.18 150 3.75 49 1.21 9-Jan-05 124 10-Jan-05 125 6 0.15 140 3.51 40 1.00 ll-Jan-05 126 12-Jan-05 127 7 0.18 145 3.63 46 1.14 175 Date Days PCI Membralox Koch2 Koch Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi 13-Jan-05 128 14-Jan-05 129 6 0.15 126 3.15 37 0.93 15-Jan-05 130 16-Jan-05 131 6 0.16 135 3.39 40 1.00 17-Jan-05 132 18-Jan-05 133 7 0.16 121 3.02 40 1.00 19-Jan-05 134 20-Jan-05 135 5 0.13 131 3.27 40 1.00 21-Jan-05 136 22-Jan-05 137 5 0.13 126 3.15 40 1.00 23-Jan-05 138 24-Jan-05 139 5 0.13 126 3.15 37 0.93 25-Jan-05 140 26-Jan-05 141 10 0.25 135 3.39 46 1.14 27-Jan-05 142 2 0.05 28-Jan-05 143 29-Jan-05 144 121 3.02 40 1.00 30-Jan-05 145 31-Jan-05 146 116 2.90 43 1.07 l-Feb-05 147 2-Feb-05 148 169 4.23 46 1.14 3-Feb-05 149 4-Feb-05 150 135 3.39 5-Feb-05 151 6-Feb-05 152 116 2.90 31 0.79 7-Feb-05 153 8-Feb-05 154 126 3.15 34 0.86 9-Feb-05 155 10-Feb-05 156 121 3.02 29 0.71 1 l-Feb-05 157 12-Feb-05 158 121 3.02 34 0.86 13-Feb-05 159 14-Feb-05 160 Ul 2.78 29 0.71 15-Feb-05 161 16-Feb-05 162 111 2.78 49 1.21 17-Feb-05 163 126 3.15 29 0.71 18-Feb-05 164 116 2.90 23 0.57 19-Feb-05 165 176 Date Days PCI Membralox Koch2 Koch Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi 20-Feb-05 166 116 2.90 23 0.57 2 l-Feb-05 167 116 2.90 23 0.57 22-Feb-05 168 116 2.90 23 0.57 23-Feb-05 169 24-Feb-05 170 116 2.90 23 0.57 25-Feb-05 171 26-Feb-05 172 116 2.90 17 0.43 27-Feb-05 173 28-Feb-05 174 111 2.78 17 0.43 l-Mar-05 175 2-Mar-05 176 111 2.78 17 0.43 3-Mar-05 177 4-Mar-05 178 111 2.78 17 0.43 5-Mar-05 179 6-Mar-05 180 106 2.66 17 0.43 7-Mar-05 181 8-Mar-05 182 106 2.66 23 0.57 9-Mar-05 183 10-Mar-05 184 61 2.04 1 l-Mar-05 185 97 2.42 57 1.91 23 0.57 12-Mar-05 186 92 2.30 47 1.56 23 0.57 13-Mar-05 187 97 2.42 45 1.12 23 0.57 14-Mar-05 188 97 2.42 40 1.00 23 0.57 15-Mar-05 189 97 2.42 38 0.94 23 0.57 16-Mar-05 190 102 2.54 37 0.93 23 0.57 17-Mar-05 191 97 2.42 37 0.92 23 0.57 18-Mar-05 192 102 2.54 36 1.19 23 0.57 19-Mar-05 193 97 2.42 37 1.24 23 0.57 20-Mar-05 194 97 2.42 37 1.24 17 0.43 2 l-Mar-05 195 92 2.30 37 1.24 17 0.43 22-Mar-05 196 87 2.18 35 1.17 17 0.43 23-Mar-05 197 87 2.18 37 1.22 17 0.43 24-Mar-05 198 87 2.18 37 1.24 17 0.43 25-Mar-05 199 102 2.54 47 1.56 20 0.50 26-Mar-05 200 102 2.54 47 1.58 20 0.50 27-Mar-05 201 102 2.54 48 1.61 17 0.43 28-Mar-05 202 97 2.42 48 1.61 17 0.43 29-Mar-05 203 97 2.42 50 1.66 17 0.43 177 Date Days PCI Membralox Koch2 Koch Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi 30-Mar-05 204 54 1.79 17 0.43 31-Mar-05 205 92 2.30 52 1.73 17 0.43 1-Apr-05 206 87 2.18 50 1.66 11 0.29 2-Apr-05 207 92 2.30 50 1.66 11 0.29 3-Apr-05 208 11 0.29 4-Apr-05 209 77 1.94 52 1.73 11 0.29 5-Apr-05 210 82 2.06 54 1.79 11 0.29 6-Apr-05 211 82 2.06 53 1.75 7-Apr-05 212 102 2.54 56 1.86 8-Apr-05 213 102 2.54 54 1.81 9-Apr-05 214 97 2.42 50 1.68 10-Apr-05 215 97 2.42 54 1.81 11-Apr-05 216 97 2.42 53 1.77 12-Apr-05 217 121 3.02 59 1.98 13-Apr-05 218 126 3.15 63 2.09 14-Apr-05 219 116 2.90 66 2.19 15-Apr-05 220 121 3.02 63 2.09 16-Apr-05 221 111 2.78 64 2.12 17-Apr-05 222 111 2.78 65 2.16 18-Apr-05 223 19-Apr-05 224 104 2.60 24 0.81 20-Apr-05 225 102 2.54 21-Apr-05 226 106 2.66 49 1.65 22-Apr-05 227 102 2.54 49 1.65 23-Apr-05 228 111 2.78 50 1.66 24-Apr-05 229 25-Apr-05 230 111 2.78 60 2.00 26-Apr-05 231 97 2.42 53 1.77 27-Apr-05 232 97 2.42 52 1.73 28-Apr-05 233 97 2.42 53 1.77 29-Apr-05 234 92 2.30 52 1.72 30-Apr-05 235 92 2.30 50 1.66 1-May-05 236 87 2.18 52 1.73 2-May-05 237 106 2.66 53 1.77 3-May-05 238 111 2.78 55 1.82 4-May-05 239 116 2.90 55 1.82 5-May-05 240 121 3.02 64 2.12 6-May-05 241 116 2.90 53 1.77 178 Date Days PCI Membralox Koch2 Koch Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi 7-May-05 242 106 2.66 50 1.68 8-May-05 243 111 2.78 55 1.82 9-May-05 244 116 2.90 60 2.00 10-May-05 245 116 2.90 54 1.81 11-May-05 246 116 2.90 55 1.84 12-May-05 247 111 2.78 58 1.93 13-May-05 248 111 2.78 57 1.91 14-May-05 249 116 2.90 62 2.05 15-May-05 250 121 3.02 63.2 2.11 16-May-05 251 121 3.02 62.7 2.09 17-May-05 252 131 3.27 64.8 2.16 18-May-05 253 116 2.90 62.1 2.07 19-May-05 254 116 2.90 61.6 2.05 20-May-05 255 121 3.02 62.1 2.07 21-May-05 256 121 3.02 61.1 2.04 22-May-05 257 111 2.78 58.4 1.95 23-May-05 258 116 2.90 56.3 1.88 24-May-05 259 111 2.78 51.0 1.70 25-May-05 260 102 2.54 47.8 1.59 26-May-05 261 121 3.02 51.0 1.70 27-May-05 262 116 2.90 46.2 1.54 28-May-05 263 111 2.78 46.7 1.56 29-May-05 264 111 2.78 45.1 1.50 30-May-05 265 106 2.66 45.7 1.52 31-May-05 266 116 2.90 50.4 1.68 l-Jun-05 267 116 2.90 53.1 1.77 2-Jun-05 268 116 2.90 52.6 1.75 3-Jun-05 269 116 2.90 57.9 1.93 4-Jun-05 270 111 2.78 53.6 1.79 5-Jun-05 271 106 2.66 53.1 1.77 6-Jun-05 272 111 2.78 50.4 1.68 7-Jun-05 273 HI 2.78 49.4 1.65 8-Jun-05 274 66.9 2.23 9-Jun-05 275 116 2.90 58.4 1.95 10-Jun-05 276 111 2.78 57.3 1.91 ll-Jun-05 277 0.00 12-Jun-05 278 13-Jun-05 279 111 2.78 54 1.81 179 Date Days PCI Membralox Koch2 Koch Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi Lmh Jv/psi 14-Jun-05 280 111 2.78 52 1.72 180 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0063245/manifest

Comment

Related Items