Waste Management 21 (2001) 643650

www.elsevier.nl/locate/wasman

Anaerobic biotransformation and methane generation potential of cheese whey in batch and UASB reactors
T.H. Erguder, U. Tezel, E. Guven, G.N. Demirer *
Department of Environmental Engineering, Middle East Technical University, Ino Bulvari, 06531, Ankara, Turkey nu Received 7 June 2000; accepted 22 October 2000

Abstract Anaerobic treatability and methane generation potential of cheese whey were determined in batch reactors. Furthermore, the eect of nutrient and trace metal supplementation on the batch anaerobic treatment, and the high-rate anaerobic treatability of cheese whey in upow anaerobic sludge blanket (UASB) reactors were investigated. To this purpose biochemical methane potential experiments were conducted and single- and two-stage UASB reactors with granular cultures were operated. In UASB experiments signicance of process staging, operational parameters such as hydraulic retention time (HRT), inuent chemical oxygen demand (COD) concentration and loading rate were also investigated. The results revealed that nutrient and trace metal supplementation is vital for the anaerobic treatment of cheese whey; the anaerobic methane generation for the cheese whey studied was found to be 424 ml CH4/g COD (23.4 l CH4/l cheese whey); undiluted cheese whey could be treated anaerobically at relatively short HRT values (2.064.95 days) without any signicant stability problems; HRT values as low as 23 days can be used for the anaerobic treatment of cheese whey, with a COD removal eciency of 9597% at inuent COD concentration of 42 70014155 100283 mg/l. # 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Anaerobic; Cheese whey; BMP; UASB; Nutrient and trace metal supplementation

1. Introduction The current trend towards the adoption of anaerobic biotechnology for the treatment of a wide range of industrial and agro-industrial wastewaters can be largely attributed to the development of reactor designs which permit eective biological solids retention. In contrast to the earlier, completely mixed systems in which the solids retention time (SRT) was approximately the same as the hydraulic retention time (HRT), advanced anaerobic reactor designs achieve SRT values which can be up to 10100 times longer than the HRT. The resultant signicant decrease in required reactor volumes and associated capital costs has been an important factor in the recent switch from conventional aerobic to anaerobic treatment systems. Cheese whey is a protein and lactose rich byproduct of the cheese industry. It is very biodegradable ($99%) with very high organic content [up to 70 g chemical oxygen demand (COD)/l] and low alkalinity (50 meq/l
* Corresponding author. Tel.: +90-312-210-58-67; fax: +90-312210-12-60. E-mail address: goksel@metu.edu.tr (G.N. Demirer).

or 2500 mg/l as CaCO3) content [1]. It can be reprocessed into valuable products such as protein and lactose in the dairy industry, or directly used as animal feed. Extremely high organic content of whey renders the application of conventional aerobic biological treatment mainly due to the cost of oxygen supplementation. Anaerobic treatment does not require any oxygen supplementation and generates signicant amounts of energy in the form of methane gas. However, Malaspina et al. [2] stated that raw whey is a quite problematic substrate to treat anaerobically because of the lack of alkalinity, the high COD concentration, the tendency to acidify very rapidly, the diculty to obtain granulation and the tendency to produce an excess of viscous exopolymeric materials of probable bacterial origin that severely reduces sludge settleability and can be a cause of biomass washout. There are many laboratory and pilot-scale studies in the literature on the anaerobic treatment of cheese whey [218]. The literature data on anaerobic treatment of cheese whey is summarized in Table 1. As is clear from Table 1, most of the studies on anaerobic treatment of cheese whey dealt with diluted (or deproteinated) whey, which is much simpler to treat. As Kalyuzhnyi et al. [18] stated, when undiluted cheese

0956-053X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0956-053X(00)00114-8

644

T.H. Ergu der et al. / Waste Management 21 (2001) 643650

Table 1 Data on anaerobic treatment of cheese whey Reactor typea UFFLR DSFFR FBR FBR AAFEB SDFA UASB UASB DUHR AP TSUAD UASB UASB Hydraulic retention time (days) 5 5 0.4 0.10.4 0.60.7 1.5 5 7 8 10 20 2.311.6 5.46.8 Inuent chemical oxygen demand conc. (g/l) 79 13 7 0.810 515 69.8 11 528.7 68 4.4 69.6 69.6 577 4755 Organic loading rate (OLB) (g COD/l day1) 14 2.6 7.7 640 8.222 16.1 7.1 0.96 10 0.55 7 3.5 128.5 79.5 Removal Eciency (RE) (%) 95 88 90 6387 6192 99 94 9799 97 63 32.5 39.5 9599 9094 Reference 6 5 3 9 4 7 10 11 2 15 16 16 18 18

a UFFLR, upow xed-lm loop reactor; DSFFR, downow stationary xed-bed reactor; FBR, uidized-bed reactor; AAFEB, anaerobic attached-lm expanded-bed reactor; SDFA, semicontinuous digester with occulant addition; UASB, upow anaerobic sludge blanket reactor; DUHR, downow-upow hybrid reactor; AP, anaerobic pond; TSUAD, two-stage unmixed anaerobic digester.

whey is directly treated in anaerobic reactors, stability problems arise. Therefore, long HRT values (520 days) were used in studies on anaerobic treatment of undiluted cheese whey (Table 1) to overcome the potential instability problems. However, this may reduce one of the major attractions of anaerobic systems in terms of lower HRT values and corresponding smaller reactor volumes and lower capital costs. Finally even though the superiority of two-stage anaerobic processes over conventional single-stage treatment, in terms of reduced euent Volatile Fatty Acids (VFA) concentrations, achievement of higher organic loadings, etc., was reported by several authors [1922], a two-stage conguration was used for the anaerobic treatment of cheese whey only in one study up to date (Table 1). Therefore, the aim of this study was to determine the anaerobic treatability and methane generation potential of cheese whey in batch reactors, as well as examining the nutrient and trace metal supplementation on the batch anaerobic treatment of cheese whey; and highrate anaerobic treatability of cheese whey in upow anaerobic sludge blanket (UASB) reactors. To this purpose biochemical methane potential (BMP) experiments were conducted and single- and two-stage UASB reactors with granular cultures were operated. In UASB experiments, signicance of process staging, operational parameters such as HRT, inuent COD concentration and loading rate were also investigated. 2. Materials and methods 2.1. Characterization of cheese whey The characterization of the cheese whey used in the BMP and UASB experiments, which was obtained from

a chese production factory located near the city of Samsun, is tabulated in Table 2. 2.2. Basal medium The composition of the basal medium (BM) used in the experiments is as follows (concentrations of the constituents are given in parentheses as mg/l): NH4Cl (1200), MgSO4.7H2O (400), KCl (400), Na2S.9H2O (300), CaCl2.2H2O (50), (NH4)2HPO4 (80), FeCl2.4H2O (40), CoCl2.6H2O (10), KI (10), MnCl2.4H2O (0.5), CuCl2.2H2O (0.5), ZnCl2 (0.5), AlCl3.6H2O (0.5), NaMoO4.2H2O (0.5), H3BO3 (0.5), NiCl2.6H2O (0.5), NaWO4.2H2O (0.5), Na2SeO3 (0.5), cysteine (10), NaHCO3 (6000). This BM contained all the necessary micro and macro nutrients required for an optimum anaerobic microbial growth [21]. 2.3. BMP Experiments In order to determine the anaerobic biodegradability and biogas production of the cheese whey studied, the
Table 2 Characterization of the cheese whey used in biochemical methane potential (BMP) and upow anaerobic sludge blanket (UASB) experiments Parameter Cheese whey used in BMP experiments 3.44 55 2501344 mg/l Cheese whey used in UASB experiments 3.92 74 500424 mg/l 124 mg/l 145.6 mg/l 9380453 mg/l 8280396 mg/l

PH Chemical oxygen demand Total phosphorus Total Khjeldahl nitrogen Suspended solids Volatile suspended solids

T.H. Ergu der et al. / Waste Management 21 (2001) 643650

645

BMP experiments [23] were performed in 125-ml serum bottles capped with natural rubber sleeve stoppers. The liquid volume of the serum bottles was 50 ml and they were maintained at 352 C in a temperature controlled room. Serum bottles were seeded with mixed anaerobic cultures obtained from anaerobic sludge digesters at the Ankara wastewater treatment plant, which has a SRT of 14 days. The cultures were thoroughly mixed and ltered through a screen with a pore size of 1 mm. In order to compare the supplementation of nutrient and trace metals on gas production, one out of two sets of serum bottles received BM (Table 3) for the initial COD concentrations of 0, 5525, 11 050, and 22 100 mg/l. The serum bottles were purged with a 25% CO2 and 75% N2 gas mixture for 34 min to maintain the proper pH and anaerobic conditions. Gas produced in each serum bottle was measured daily for 68 days. Control serum bottles were also run in all experiments to determine the background gas production. Before the experiments, serum bottles were operated until the variation in daily gas production was less than 15% for at least 7 consecutive days. Serum bottles for one COD concentration out of four were run as duplicates. 2.4. UASB experiments In order to achieve the nutrient and trace metal concentrations given in the BM in the inuent of the UASB reactor, the cheese whey described in Table 1 was diluted by the ratio of 4/5 with the ve times concentrated BM. This mixture was used as the feed solution in the UASB experiments. In other words, each liter of feed solution consisted of 0.8 l of cheese whey described in Table 1 and 0.2 l of ve times concentrated BM. Thus, the feed solution had an average COD of 59 600 mg/l. The granular cultures used in the experiments that had Mixed Liquor Suspended Solids (MLSS) and Mixed Liquor Volatile Suspended Solids (MLVSS) values of 107 650.519 974 and 99 930.519 715 mg/l, respectively, were obtained from the UASB reactor of Tekel

Pasabahce Liquor Factory located in Istanbul which has an organic loading rate of 30 kg COD/m3 day1. The UASB reactors were continuously fed. The dimensions, eective volumes, owrates, HRTs and the biomass inventories of the UASB reactors used are given in Table 4.
Table 4 The dimensions, eective volumes, owrates, hydraulic retention times (HRTs) and the biomass inventories of the upow anaerobic sludge blanket (UASB) reactors used Operation Diameter and the height of the UASB (cm) 5.2100 5.2100 Eective volume of the UASB (l) 0.743 0.510 HRT (h) Biomass inventory in the system (g) 74.25 50.96

First-stage Second-stage

2.064.95 1.463.50

Fig. 1. The gas production values of the serum bottles used in the biochemical methane potential experiments.

Table 3 Experimental set-up of biochemical methane potential (BMP) experiments Reactor no. 1 2 3 4a 5 6 7 8a
a

Basal medium supplementation + + + +

Chemical oxygen demand (mg/l) 0 5525 11 050 22 100 0 5525 11 050 22 100

Run as duplicates.

Fig. 2. Comparison of the theoretical and experimental gas production for biochemical methane potential experiments.

646

T.H. Ergu der et al. / Waste Management 21 (2001) 643650

2.5. Analytical methods COD values were measured by Hach spectrophotometer (Model: P/N 45600-02) and vials for COD 01500 ppm. pH measurements were taken with a pH meter (Model 2906, Jenway Ltd.) and a pH probe (G05992-55, Cole Parmer Instrument Co.). Suspended solids and volatile suspended solids concentrations were measured by following standard methods (2540 E; Standard Methods, 1995). Total phosphorus and total

Khjeldahl nitrogen concentrations were also determined by standard methods (4500-P-E and 4500-NorgB, respectively) [24]. Gas produced in each serum bottle was daily measured by a gas displacement device. The percent of methane in biogas was determined using the serum bottles used in Experiment 1. The headspace gas produced was syringed out for injection into an entirely closed serum bottle containing a concentrated KOH stock solution of 20 g/l at atmospheric pressure. The serum

Fig. 3. Anaerobic bioconversion of cheese whey in the rst-stage upow anaerobic sludge blanket reactor: (a) hydraulic retention times of the system; (b) chemical oxygen demand (COD) loading rate; (c) inuent and euent COD concentrations; (d) inuent and euent pH; (e) inuent and euent bicarbonate alkalinity; and (f) euent VFA concentration.

T.H. Ergu der et al. / Waste Management 21 (2001) 643650

647

bottles were shaken manually for 34 min. The remaining gas in the bottle indicated the approximate CH4 gas produced as a result of anaerobic degradation of the wastes and CH4 was syringed out to determine the methane percent of the biogas produced simply by measuring its volume. The result of ve determinations indicated that the average approximate methane content of biogas was 775%. 3. Results and discussion 3.1. BMP experiments The gas production values of the serum bottles used in the BMP experiments for 68 days are depicted in Fig. 1. It is clear from Fig. 1 that the serum bottles with BM resulted in higher rates and amounts of gas production compared to that without BM. For the inuent COD concentration of 5525 mg/l, the cultures with BM exerted more than 90% (157 ml) of the total gas production (176 ml) in the rst 16 days without any indication of inhibition. For the cultures with no BM supplementation for the same COD concentration of 5525 mg/l, the gas production rate was notably lower than that of with BM, and 90% (139 ml) of the total gas production (154 ml) was produced only after 35 days. Similarly, cultures with BM for the inuent COD concentration of 11 050 mg/l exerted more than 90% of (292 ml) total gas production (325 ml) in 24 days. The cultures with BM for the inuent COD concentrations of 11 050 mg/l exerted more than 90% of (274 ml) total gas production (304 ml) in 59 days. For the highest inuent COD concentration of 22 100 mg/l, the cultures with and without BM produced 598 and 125 ml of gas, respectively. This is an indication of an inhibition in the cultures with inuent COD concentration of 22 100 mg/ l. In order to determine whether this inhibition was due to absence of BM or not, the theoretical gas production values were calculated and compared with experimental (observed) COD values (Fig. 2). The amount of gas

produced by the cultures with the inuent COD concentration of 22 100 mg/l with BM supplementation (598 ml) was in the range of theoretical gas production (532.3606.3 ml). Thus, it can be concluded that the lower gas production for the same COD concentration by the cultures with no BM was solely due to the nutrient, trace metal and/or alkalinity deciency. This indicated the signicance of nutrient, trace metal and/or alkalinity supplementation in the anaerobic treatment of cheese whey. Cheese whey samples with BM addition produced 135.5, 250.2 and 460.5 ml of methane gas for the initial COD concentrations of 5525, 11 050, and 22 100 mg/l, respectively. These values indicated that the anaerobic methane generation for the cheese whey studied is 424 ml CH4/g COD or 23.4 l CH4/l cheese whey. 3.2. UASB experiments After the batch (BMP) experiments, the anaerobic treatability of the cheese whey was studied in UASB reactors. The corresponding HRT, COD loading rate, inuent and euent COD concentrations, pH and bicarbonate alkalinity values and euent VFA concentrations are depicted in Fig. 3. In order to minimize the reactor volume requirement, the HRT of the UASB reactor was kept at very low values of 2.064.95 days. Thus, the reactor was operated at a high rate which is also obvious if this range is compared with the HRT values used in the anaerobic reactors for the treatment of cheese whey in the literature (Table 1). Table 5 summarizes the HRT, inuent and euent COD concentrations and loading rates applied and the average COD removal eciency observed in the rststage UASB reactor. As can be seen from Table 5, considerably higher COD removal eciencies (91.997.0) were observed, in spite of the low HRT values applied. Thus, this experiment indicates that HRT values as low as 2 days can be used for the anaerobic treatment of cheese whey in a one-stage UASB reactor, provided that sucient alkalinity and nutrients are supplied. Another

Table 5 Hydraulic retention times (HRT), inuent and euent chemical oxygen demand (COD) concentrations and loading rate applied and the average COD removal eciency observed in the rst-stage upow anaerobic sludge blanket reactor Days HRT (days) COD concentration (mg/l) Inuent 016 1719 2026 2728 2934 3543 4451 2.062.46 3.663.82 2.692.90 3.70 3.37 2.483.10 3.544.95 55 650134458 350495 42 700141 47 350134452 9008910 47 3501344 50 500127353 7501909 50 500127355 100283 51 600120252 7502475 Euent 16691627463 2218131 25751203043180 3465269 1978882695148 2065276443874 22303612803187 COD loading rate (g/l day1) 22.624.6 11.7 16.319.2 12.8 15.015.9 16.322.2 10.414.6 Average COD removal eciency (%) 95.397.0 94.8 94.294.6 92.6 95.096.0 91.995.9 94.795.7

648

T.H. Ergu der et al. / Waste Management 21 (2001) 643650

Fig. 4. Anaerobic bioconversion of cheese whey in the second-stage upow anaerobic sludge blanket reactor: (a) hydraulic reaction times of the system; (b) chemical oxygen demand (COD) loading rate; (c) inuent and euent COD concentrations; (d) inuent and euent pH; (e) inuent and euent bicarbonate alkalinity; and (f) euent VFA concentration. Table 6 Hydraulic retention times (HRT), inuent and euent chemical oxygen demand (COD) concentrations and loading rate applied and the average COD removal eciency observed in the second-stage upow anaerobic sludge blanket reactor Days HRT (days) COD concentration (mg/l) Inuent 06 712 1315 1620 2.132.31 1.761.85 2.433.40 3.193.50 1978882695148 3265127443874 2803187291853 2713742830361 Euent 142841795184 19584 197549194374 COD loading rate (g/l day1) 0.861.17 1.782.52 0.820.99 0.85 Average COD removal eciency 56.359.6 30.132.8 27.231.3

T.H. Ergu der et al. / Waste Management 21 (2001) 643650

649

signicant observation from Fig. 3 and Table 4 is that a notable reduction in COD loading rate did not manifest itself in the improved COD removal eciencies. For example, COD loading rates of 22.624.6 and 10.414.6 g/l day1 yielded very similar COD removal eciencies of 95.397.0 and 94.795.7%, respectively. On the other hand, because the euent pH of the reactor gradually rose from 7.5 to 9.0 in the rst 14 days of operation (Fig. 3d), the inuent alkalinity was reduced from 6000 to 5000 mg/l (Fig. 3e). This change resulted in an gradual decrease in the euent pH of the system, which was stabilized around 7.88.1, as seen in Fig. 3e. The euent VFA concentration was always less than 300 mg/l and below the detection limit of 50 mg/l, except on Days 4044 (Fig. 3f). 3.2.1. Eect of staging In spite of the high inuent COD concentrations of 42 70014158 350495 mg/l and low HRT values of 2.064.95 days, the one-stage UASB achieved 91.997% COD removal. However, the euent COD concentration obtained with the one-stage UASB was between 166916 and 443874 mg/l (Table 5). This range of euent COD values is well over the discharge limits for the receiving environments and must further be treated. Therefore, in order to examine the potential role of a second-stage UASB reactor on achieving a better euent COD concentration, a second UASB was connected in series to the rst one on Day 34. The dimensions, eective volume and the biomass inventory of the second-stage UASB reactor used are given in Table 4. The euent from the rst-stage UASB was fed to the second-stage UASB reactor, and the euent from the second reactor was monitored for 20 days (Fig. 4). Table 6 summarizes the HRT, inuent and euent COD concentrations and loading rate applied and the average COD removal eciency observed in the secondstage UASB reactor. As is clear from Fig. 4 and Table 6, any further COD removal achieved in the second-stage UASB is fairly low (27.259.6%). This is thought to be mainly due to the low inuent COD concentrations (19784438 mg/l) and HRT values (1.763.50 days) of the system. Because the COD concentrations in the inuent of the secondstage UASB reactor is signicantly lower than that of the rst-stage UASB reactor, longer HRT values were needed to achieve a better euent quality in the secondstage reactor. Important operational parameters of pH, bicarbonate alkalinity and VFA were at acceptable levels for the second-stage UASB reactor as well (Fig. 4df). When the performance of the two-stage UASB reactor is considered, a total COD removal eciency of 96 97% and euent COD levels of 14281975 mg/l were obtained. Even though euent quality is still too poor to be directly discharged to a receiving environment, an

aerobic polishing step could easily lower the euent COD concentration below the discharge limits. 4. Conclusions This study indicated that the high rate anaerobic treatment of undiluted cheese whey in UASB reactors is a very ecient and cost-eective method. Furthermore, the following conclusions can be drawn based on the experimental results of this study. 1. Nutrient, trace metal and/or alkalinity supplementation is vital for the anaerobic treatment of cheese whey. This is depicted by the higher gas production rates and amounts in the serum bottles with BM addition, compared to that without BM addition (Figs. 1 and 2). 2. Anaerobic bioconversion of cheese whey yielded biogas which has a high methane content (775%). 3. The anaerobic methane generation from the cheese whey studied was found to be 424 ml CH4/g COD or 23.4 l CH4/l cheese whey. 4. Unlike most of the studies in the literature on the subject (Table 1), undiluted cheese whey could be treated anaerobically at relatively short HRT values (2.064.95 days) without any signicant stability problems. 5. HRT values as low as 23 days can be used for the anaerobic treatment of cheese whey, with a COD removal eciency of 9597% at inuent COD concentration of 42 70014155 100283 mg/l. 6. A second-stage UASB achieved only slight additional COD removal eciency of 27.259.6% (Table 5). However, this eciency could possibly be increased if longer HRT values were used and/ or euent recycle was applied. Acknowledgements This study was funded by the State Planning Organization of the Republic of Turkey. References
[1] Mawson AJ. Bioconversion for whey utilization and waste abatement. Biores Technol 1994;47:195203. [2] Malaspina F, Stante L, Cellamare CM, Tilche A. Cheese whey and cheese factory wastewater treatment with a biological anaerobic-aerobic process. Water Sci Technol 1995;32:5972. [3] Boening PH, Larsen VF. Anaerobic uidized bed whey treatment. Biotechnol Bioengng 1982;14:253956. [4] Switzenbaum MS, Danskin SC. Anaerobic expanded bed treatment of whey. Agric Waste 1982;4:41126. [5] De Haast J, Britz TJ, Novello JC, Verwey EW. Anaerobic

650

T.H. Ergu der et al. / Waste Management 21 (2001) 643650 digestion of deproteinated cheese whey. J Dairy Res 1985;52: 457467. Wildenauer FX, Winter J. Anaerobic digestion of high strength acidic whey in a pH-controlled upow xed-lm loop reactor. Appl Microbiol Biotechnol 1985;22:36772. Barford JP, Cali RG, Callander IJ, Floyd EJ. Anaerobic digestion of high-strength cheese whey utilizing semicontinuous digesters and chemical occulant addition. Biotechnol Bioengng 1986;28:16017. De Haast J, Britz TJ, Novello JC. Eect of dierent neutralizing treatments on the eciency of an anaerobic digester fed with deproteinated cheese whey. J Dairy Res 1986;53:46776. Denac M, Dunn IJ. Packed- and uidized-bed biolm reactor performance for anaerobic wastewater treatment. Biotechnol Bioengng 1988;32:15973. Schroder EW, De Haast J. Anaerobic digestion of deproteinated cheese whey in an upow sludge blanket reactor. J Dairy Res 1989;56:12939. Yan JQ, Lo KV, Liao PH. Anaerobic digestion of cheese whey using upow anaerobic sludge blanket reactor. Biol Waste 1989;27:289305. Venkataraman J, Kaul SN, Satyanarayan S. Determination of kinetic constants for a two-stage anaerobic upow packed-bed reactor for dairy wastewater. Biores Technol 1992;40:25361. Yan JQ, Lo KV, Pinder KL. Instability caused by high strength of cheese whey in a UASB reactor. Biotechnol Bioengng 1993;41:7006. Cohen A, Thiele JH, Zeikus JG. Pilot-scale anaerobic treatment of cheese whey by the substrate shuttle process. Water Sci Technol 1994;30:43342. Monroy H, Vazquez M, Derramadero JC. Anaerobic-aerobic treatment of cheese wastewater with national technology in Mexico: the case of ``El Sauz''. Water Sci Technol 1995;32:149 56. Ghaly AE. A comparative study of anaerobic digestion of acid cheese whey and dairy manure in a two-stage reactor. Biores Technol 1996;58:6172. Malaspina F, Cellamare CM, Stante L, Tilche A. Anaerobic treatment of cheese whey with a downow-upow hybrid reactor. Biores Technol 1996;55:1319. Kalyuzhnyi SV, Martinez EP, Martinez JR. Anaerobic treatment of high-strength cheese-whey wastewaters in laboratory and pilot UASB reactors. Biores Technol 1997;60:5965. Wiegant WM, Hennink M, Lettinga G. Separation of the propionate degradation to improve the eciency of thermophilic anaerobic treatment of acidied wastewaters. Water Res 1986;20: 517524. Speece RE, Duran M, Demirer GN, Zhang H, DiStefano T. The role of process conguration in the performance of anaerobic systems. Water Sci and Technol 1997;36:53947. Demirer GN, Speece RE. Anaerobic biotransformation of four 3carbon compounds (acrolein, acrylic acid, allyl alcohol and npropanol) in UASB reactors. Water Res 1998;32:74759. Demirer GN, Speece RE. Comparison of anaerobic acrylic acid biotransformation in single- and two-staged pH-stat CSTR systems. Water Env Res 2000;72:849. Owen WF, Stuckey DC, Healy JB, Young LY, McCarty PL. Bioassay for monitoring biochemical methane potential and anaerobic toxicity. Water Res 1979;13:48592. Standard Methods for the Examination of Water and Wastewater, 19th ed. Washington, DC: American Public Health Association, 1995.

[6] [7]

[16] [17] [18] [19]

[8] [9] [10] [11] [12] [13] [14] [15]

[20] [21] [22] [23] [24]