Anaerobic Treatment of Wastewater with High Organic Content Using a Stirred Tank Reactor Coupled with a Membrane ﬁltration Unit W

Water Research 37 (2003) 902–908

Anaerobic treatment of wastewater with high organic content using a stirred tank reactor coupled with a membrane ﬁltration unit W. Fuchs*, H. Binder, G. Mavrias, R. Braun

IFA-Tulln, Department of Environmental Biotechnology, Konrad Lorenz Strasse 20, 3430 Tulln, Austria

Received 25 October 2001; received in revised form 17 May 2002; accepted 24 May 2002

Abstract

Using a cross-flow membrane bioreactor, high anaerobic conversion rates of three different types of wastewater with varying organic content were achieved. Loading rates obtained were as follows: 20 g COD LÀ1 dÀ1 for artificial wastewater, approximately 8 g COD LÀ1 dÀ1 from vegetable processing industry (sauerkraut brine) and 6–8 g COD LÀ1 dÀ1 for wastewater from an animal slaughterhouse. At stable conditions, COD-removal rates in all three wastewaters were higher than 90%. Methane yields from the treatment of artificial wastewater, sauerkraut brine, and À1 À1 animalslaughterhousewastewater were in the range of 0.17–0.30, 0.20–0.34, and 0 :1220:32Ln g COD fed, respectively. The complete retention of biomass and suspended solids is a unique feature of this treatment process, which combines a high loading capacity and at the same time, high COD removal rates even for complex wastewater containing high concentrations of particulate matter. r 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Anaerobic digestion; Membrane bioreactor; Membrane ﬁltration; Wastewater treatment

1. Introduction SRT is required to accommodate the slower net growth rate [6,7]. On that account, biomass retention is one of Membrane filtration has gained increasing interest as the most important aspects of anaerobic technology. As a mean of biomass retention in wastewater treatment membrane filtration completely retains all microorgan- systems [1–3]. It is an efficient toolfor maintaining a isms, it can significantly improve the process [8,9]. long solids retention time (SRT) at a relatively short Moreover, the production of electric power from the hydraulic retention time (HRT). While a high SRT is generated biogas can cover the energy demand for the desirable for process stability, a short HRT minimizes filtration process. In this paper, data from laboratory the reactor volume and hence, reduces capital costs [4]. scale experiments are presented that show the advan- A rapid development can be observed on the application tages of this system in terms of volumetric loading rates, of membrane filtration in aerobic treatment plants, the effluent quality and process stability. so-called membrane bioreactors (MBRs) [5]. However, very few studies have focused on the application of this design configuration in the field of anaerobic digestion. 2. Materials and methods In anaerobic systems, microorganisms reproduce less rapidly than in aerobic systems and a longer minimum A schematic presentation of the laboratory plant used in this study is shown in Fig. 1. A stirred tank reactor *Corresponding author. Tel.: +43-2272-66280-553; fax: was coupled to an external filtration device consisting of +43-2272-66280-503. a recirculation pump and a ceramic cross-flow mem- 2 E-mail address: [email protected] (W. Fuchs). brane (material: Al2O3, surface area: 1260 cm , pore size:

biogas

wastewater feed

cross-flow filtration unit bioreactor

permeate

recirculation pump

Fig. 1. Schematic setup of the anaerobic membrane bioreactor.

Table 1 Composition of the three types of substrate used in this study

Artiﬁcialwastewater Slaughterhouse efﬂuent Sauerkraut brine

COD (mg LÀ1) 9700 (29,100)a 5800–20,150 40,700–64,600 BOD (mg LÀ1) 6500 (19,500)a 2200–9800 15,800–24,200 SS (g LÀ1) 0 2.4–4.7 0.39–0.58 pH 7.1 5.3–6.8 3.2–4.2 VFA (mg LÀ1) 820 (2460)a 395–4055 6430–10,940 À1 a NH4-N (mg L ) 14 (42) 102–323 158–453 TKN (mg LÀ1) 690 (2070)a 301–460 1120–1580

a For the ﬁnalloadingrate of 20 g COD L À1 dÀ1 three-fold concentrated substrate was used.

0.2 mm). The total volume, including the filtration loop, 0.75 mm were removed by passing through a rotary sieve. was 7 L. The fixed volume in the reactor was maintained It also contained varying amounts of wastewater from by a level sensor, which controlled a valve on the the washing of equipment and premises, which caused a permeate side of the filtration unit. big variation in the concentration of organic matter. Three different types of substrates (artificialwaste- Seed sludge was taken from an anaerobic sludge water, wastewater from vegetable processing industry digester of a municipalwastewater treatment plant. (sauerkraut brine) and chicken slaughterhouse waste- Process temperature was maintained at 301C. During water) were used. Characteristics of the three types of the operation of the plant, degradation of organic wastewater are presented in Table 1. The artificial matter, pH, gas production rate and methane content of wastewater contained peptone, yeast extract, glucose, the biogas were monitored. A wet gas meter equipped acetate and sodium chloride and was freshly prepared with temperature and pressure gauges was used to twice per week. Two types of industrialwastewater were determine the amount of biogas produced. The obtained collected as individual lots from local companies and values were recalculated to standard conditions (01C, stored at 41C untiluse. Sauerkraut brine was the liquid 760 mm Hg). Methane content of the biogas was generated during the fermentation process of the determined by adsorption of CO2 in 10% KOH as shredded and salted cabbages. It was rich in organic described by Braun [10]. Parameters such as COD, components such as lactic acid, sugars and proteins and NH4–N, TKN, and SS were measured according to the contains a high degree of sodium chloride. Wastewater German standard methods for water, wastewater and from a chicken slaughterhouse was a mixture of the sludge analysis [11]. process water from the defeathering line and from the In addition, the concentration of volatile fatty acids cleaning of the guts. Particles with a size bigger than (VFA) in the reactor was also monitored. VFA (C2–C7) 904 W. Fuchs et al. / Water Research 37 (2003) 902–908 were measured using gas chromatographic analysis in 3.2. Sauerkraut brine accordance to the method reported by Chang and Sanders [12]. VFA were measured as the main indicator Figs. 4 and 5 show the response of the system using whether the system adapts to an increased organic load. sauerkraut brine wastewater. Following a short adapta- A peak concentration occurs after raising the loading tion phase, the suspended solids concentration increased rate. If the bacteria adapt to the higher loading rate, constantly from 22 to 38 g LÀ1. After 49 days, the VFA VFA concentration returns to lower values. On the concentration reached a criticalvalueof 1700 mg L À1. other hand, a continuous elevation of VFA concentra- Therefore, the volumetric loading rate was reduced from tion over severaldays indicates overloadingof the 6.3 to 4.2 to stabilize the system. Subsequently, the system. loading rate was increased again to reach a maximum volumetric loading rate of 7.93. At day 111, high concentrations of VFA were measured. As a result, the loading rate was lowered and part of the sludge was 3. Results and discussion removed from the system to keep the suspended solids concentration below 60 g LÀ1. After 160 days, sludge 3.1. Artificial wastewater was removed severaltimes to maintain a suspended solids concentration of approximately 55 g LÀ1. The To determine the response of the system under a final loading rate, which was also the maximum value controlled environment, artificial wastewater was used. achieved, was 8.6 g COD LÀ1 dÀ1. Only small concen- In this test run, volumetric loading rate (VLR) was trations of VFA were detected in the effluent at that increased in a stepwise mode. Seed sludge was added time. A high chloride concentration (6 g LÀ1), which is severaltimes into the reactor to minimize the adaptation characteristic for the sauerkraut brine, was measured in phase. At sludge concentrations of 20–25 g LÀ1,an organic loading rate as high as 20 g COD LÀ1 dÀ1 was achieved (Fig. 2). After adaptation, COD elimination 0.50 100 rates were higher than 90% and the methane yield was ) -1 0.45 90 in the range of 0:2020:30Ln of CH4 per g of COD in the 0.40 80 influent (Fig. 3). 0.35 70 .gCOD The application of membrane bioreactor was ex- 4 0.30 60 CH 0.25 50 n tended to include the anaerobic treatment of industrial 0.20 40 samples. Treatment of sauerkraut brine and chicken 0.15 30 -yield (l 0.10 20 COD-removal (%) slaughterhouse wastewater were tested at increasing 4 0.05 10 feed rates. However, the different COD concentrations CH 0.00 0 of each wastewater caused lot of fluctuations in the 0 5 10 15 20 25 organic loading rate. On the contrary to the previous Loading rate (gCOD.l-1.d-1) experiment, extra sludge was not added to the reactor after primary inoculation. Sludge was only removed Methane-yield COD-removal when the suspended solids (SS) concentration exceeded Fig. 3. Plot of volumetric loading rate against methane yield À1 60 g L and in negligible amounts, when samples were (per COD fed) and COD removalwith artificialwastewater as a taken. substrate.

30 30 2000 2000

) 25 25 -1 1500 1500 .d ) ) -1 -1 -1

20 20 ) -1 15 15 1000 1000

10 10 SS (g.l VFA (mg.l COD (mg.l 500 500

VLR (g COD.l 5 5

0 0 0 0 020406080 020406080 (a) Time (d) (b) Time (d)

Volumetric loading rate SS-reactor COD-effluent VFA-effluent

Fig. 2. Course of (a) volumetric loading rate, SS in the bioreactor and (b) concentrations of COD and volatile fatty acids (VFA) in the efﬂuent during the run with artiﬁcialwastewater. W. Fuchs et al. / Water Research 37 (2003) 902–908 905

14 70 2500 2500

) 12 60 2000 2000 -1 ) ) .d -1

10 50 -1 -1 )

-1 1500 1500 8 40 6 30 1000 1000 SS (g.l VFA (mg.l 4 20 COD (mg.l 500 500 VLR (g COD.l 2 10 0 0 0 0 0 50 100 150 200 0 50 100 150 200 (a) Time (d) (b) Time (d)

Volumetric loading rate SS-reactor COD-effluent VFA-effluent

Fig. 4. Course of (a) volumetric loading rate, SS in the bioreactor and (b) concentrations of COD and volatile fatty acids (VFA) in the efﬂuent during the run with sauerkraut brine.

À1 À1 0.50 100

) of industrialwastewater were 6–8 g COD L d .In -1 0.45 90 both cases, COD removalrates were higher than 97% at 0.40 80 0.35 70 stable operation conditions, with only a minimal .g COD 4 0.30 60 decrease towards higher loading rates. Corresponding

CH 0.25 50

n residualCOD valuesin the effluent were in a range of 0.20 40 À1 0.15 30 100–400 mg L for slaughterhouse wastewater (Fig. 6b) À1 COD-removal (%) -yield (l 0.10 20 and between 500 and 1000 mg COD L when using 4 0.05 10 sauerkraut brine, which contained very high concentra- CH 0.00 0 0481026 tions of organic matter (Fig. 4b). Moreover, high biogas Loading rate (gCOD.l-1.d-1) yields were obtained. The high biogas yield also gives evidence that the suspended organic matter was not Methane yield COD-removal simply retained by the membrane filtration but it was Fig. 5. Plot of volumetric loading rate against methane yield actually biologically degraded in the reactor. The (per COD fed) and COD removalwith wastewater from theoretical value is 0.35 standard liters methane per g vegetable processing industry as a substrate. COD degraded [7]. In both runs with artificialwastewater and sauerkraut brine wastewater, the measured À1 methane yield was between 0.3 and 0:34Ln CH4 gCOD the influent. However, a negative effect on the anaerobic in the influent, which is close to the maximum value digestion process was not detected. (Figs. 3 and 5). A significant deviation was only seen in the run with slaughterhouse effluent, which was also the 3.3. Slaughterhouse wastewater substrate with the highest particulate matter content. The methane yield per g COD removed was around 0.25 and The performance of the run with slaughterhouse 0:30Ln on an average (Fig. 7). The result indicates that wastewater was very stable until day 57 (Fig. 6). At that approximately 80 of the COD removal is based on point, the high loading rate of 7.4 g COD LÀ1 dÀ1 microbialactivity. This is alsosupported by the Nitrogen (hydraulic retention time: 1.2 days) had overloaded the (N)-balance. The N demand for growth of anaerobic system. At day 60 more than 3000 mg LÀ1 VFA was bacteria is almost negligible. If no accumulation of measured. To give the system the time to recover, the organic matter occurs in the bioreactor, the balance feeding of the reactor was turned off completely. After 4 between totalnitrogen flow in and out of the reactor days, a reduction of VFA concentration to approxi- should be even. However, only 85–90% of the incoming mately 500 mg LÀ1 was obtained. Thus, the system was nitrogen was measured in the permeate. The partial restarted with a loading rate of 3 g LÀ1 dÀ1. The reactor accumulation of organic matter in the bioreactor was fully regained stable operation conditions with almost also documented by the course of the suspended solids in complete COD removal, even after a further increase of the bioreactor. After decreasing the loading rate, a the loading rate to 4.3 g LÀ1 dÀ1 (Figs. 6 and 7). reduction in the suspended solids concentration was observed. The result suggests that accumulated organic 3.4. General performance of the system matter was degraded in periods of lower loading rates. The unique feature of the anaerobic membrane bio- The tested system configuration showed an excellent reactor is the complete biomass retention independent of performance. Maximum loading rates for the two types the biomass’ physiological state. However, highly 906 W. Fuchs et al. / Water Research 37 (2003) 902–908

14 30 2500 2500 12 ) 25 2000 2000 -1 ) ) .d 10 -1 -1 -1

20 )

-1 1500 1500 8 15 6 1000 1000 SS (g.l

10 VFA (mg.l 4 COD (mg.l 500 500 VLR (g COD.l 2 5 0 0 0 0 020406080 020406080 (a)Time (d) (b) Time (d)

Volumetric loading rate SS-reactor COD-effluent VFA-effluent

Fig. 6. Course of (a) volumetric loading rate, SS in the bioreactor and (b) concentrations of COD and volatile fatty acids (VFA) in the efﬂuent during the run with wastewater from a slaughterhouse.

0.50 100

) clogging or sludge ﬂotation [19,27,28]. With the MBR, -1 0.45 90 high loading rates in combination with increased COD- 0.40 80 0.35 70 elimination rates were achieved. The complete retention .g COD 4 0.30 60 of suspended solids improved the efﬂuent quality.

CH 0.25 50

n Particulate material of the influent was kept in the 0.20 40 bioreactor until it was susceptible to biological degrada- 0.15 30 COD-removal (%) -yield (l 0.10 20 tion or drawn with the surplus sludge. In addition, the 4 0.05 10 use of a membrane separation unit allows intensive CH 0.00 0 0481026 mixing of the bioreactor content, which facilitates the Loading rate (gCOD.l-1.d-1) breakdown of solid particles. Filtration parameters were not optimized in this Methane yield COD-removal study. Permeate flow was kept at low levels (5– À2 À1 À1 Fig. 7. Plot of volumetric loading rate against methane yield 10 L m h at a cross flow velocity of 2–3 m s )to (per COD fed) and COD removalwith wastewater from a avoid interference from inadequate operation of the slaughterhouse as a substrate. filtration unit. However, energy requirement and long- term performance of the membranes are the main factors determining the economic feasibility in practical efficient anaerobic treatment processes based on fixed aspects. For aerobic membrane bioreactors, filtration bed or sludge bed reactor systems are available for systems with energy requirements as low as 0.15– substrates containing low particulate matter. To illus- 0.5 kW h mÀ3 have been developed [2,29,30]. Unfortu- trate, Lettinga et al. [13] treated sauerkraut brine in an nately, only preliminary experiences concerning mem- upflow anaerobic sludge blanket (UASB) reactor at a brane performance under anaerobic conditions exist volumetric loading rate of 8–9 kg COD mÀ3 dÀ1 with [31,32]. One major hurdle is the increased susceptibility COD removalrates of 88–93%. Typicaldesign loading to membrane fouling due to higher concentration of rates are 10 kg COD mÀ3 dÀ1 for UASB reactors and residualorganic matter in the treated wastewater [33]. 20 kg COD mÀ3 dÀ1 for the expanded granular sludge Biogas production by anaerobic processes yield bed reactor with standard COD removalrates of 1.27 Â 107 JkgÀ1 COD-converted. To transform this 85–90% [14,15]. energy into electricity, 1.05 J Â 107 are required to On the other hand, the distinct advantages of the produce 1 kW h [34]. Therefore, approximately 5–20 kW investigated process are seen for the treatment of hmÀ3 can be generated using the slaughterhouse waste- wastewater with high particulate matter. Raw slaughter- water as a substrate. Even at an extended energy house wastewater contains high concentrations of requirement for anaerobic membrane bioreactors, this insoluble, slowly biodegradable solids often representing is sufficiently enough to cover the energy demand for the over 50% of the polluting charge [16]. Table 2 gives a filtration process. comparison of results from different anaerobic processes used for the treatment of slaughterhouse wastes. In general, the application of high loading rates leads to 4. Conclusion significant drop in the COD reduction rate. Moreover, process failure was reported frequently due to excessive High anaerobic conversion rates of different types of accumulation of substrate material resulting in reactor wastewater were achieved using a membrane bioreactor. W. Fuchs et al. / Water Research 37 (2003) 902–908 907

Table 2 Comparison of results gained for the anaerobic treatment of slaughterhouse wastewater

Type of reactor VLR (g COD LÀ1 dÀ1) HRT (d) SS contenta (%) COD removal(%) Reference

Anaerobic filter 2–18.5 0.5–5 45 30–85 Tritt [17] Anaerobic filter 5 1.5 15–30 63–85 Ruiz et al. [18] Flocculent UASB 3.5 0.3 40–50 70 Sayed et al. [19] Granular UASB 11 0.5–0.6 40–50 55 Sayed et al. [20] Flocculent UASB 6.5 1.2 15–30 60–90 Ruiz et al. [18] UASB 3.5 0.42 55 70 Manjunath et al. [21] Two stage UASB 15 0.2 55 90 Sayed et al. [22] Integrated UASB-AF 5–32 0.1–0.5 10 45–98 Borja et al. [23] Fluidized bed 35 0.1–0.3 4 85 Borja et al. [24] Expanded granular sludge bed 5.3–15.8 0.2 45 54–80 Nunez and Martinez [25] Anaerobic baffled reactor 4.7 0.1 75 Polprasert et al. [26] Sequencing batch reactor 2.0–4.9 2 52 90–96 Masse and Masse [16] MBR 6–8 1.2 45–55 97 Current study

a COD content of suspended solids as percent of total COD in the untreated wastewater.

Once seed materialwas added to the reactor, it was [6] Lettinga G. Anaerobic digestion and wastewater treatment completely retained by the cross flow filtration process. systems. Antonie van Leeuwenhoek 1995;67:3–28. No granulation or immobilization of the sludge was [7] Speece RE. Anaerobic biotechnology for industrial waste- required. Therefore, the MBR showed a shorter and waters. US: Archae Press, 1996. unproblematic startup when compared to other anaero- [8] Anderson G, Saw C, Fernandes M. Application of porous bic reactor systems. Furthermore, the membrane-en- membranes for biomass retention in biological wastewater treatment processes. Process Biochem 1986;21(6):174–82. hanced degradation showed excellent elimination rates [9] Nagano A, Arikawa E, Kobayashi H. The treatment of for COD and provided a high effluent quality. With all liquor wastewater containing high-strength suspended three substrates, the anaerobic digestion process was solids by membrane bioreactor system. Water Sci Technol very stable and could be easily adapted to increasing 1992;26(3–4):887–95. loading rates. After overloading, the membrane bior- [10] Braun R. Biogas - anaerobic conversion of organic wastes eactor quickly regained stable performance. High biogas (Biogas - Methangarung. Organischer Abfalle).. Vienna, yields were obtained, which demonstrated the effective- Austria: Springer, 1982. ness of this process. The membrane bioreactor is [11] DIN - Deutsches Institut fur. Normung e.V. German specifically useful for the treatment of complex waste- standard methods for the examination of water, wastewater rich in slowly degradable solids, which are often water and sludge. Germany: VCH, Weinheim, 1975. treated insufficiently by other standard anaerobic [12] Chang I, Sanders W. Fast GC analysis of free volatile treatment systems. organic acids, fatty acids, and their isomers using HP INNOVAX columns and EPC. Hewlett Packard Applica- tion Note 228–225, US, 1993. [13] Lettinga G, van Velsen A, Hobma S, de Zeeuw W, Klapwijk References A. Use of upflow sludge blanket (USB) reactor concept for biological waste water treatment, especially for anaerobic [1] Buisson H, Cote P, Praderie M, Paillard H. The use of treatment. BiotechnolBioeng 1980;22(4):699–734. immersed membranes for upgrading wastewater treatment [14] Foresti E. Perspectives on anaerobic treatment in devel- plants. Water Sci Technol 1998;37(9):89–95. oping countries. Water Sci Technol2001;44(8):141–8. [2] Cote P, Buisson H, Pound C, Arakaki G. Immersed [15] Frankin R. Full-scale experiences with anaerobic treat- membrane activated sludge for the reuse of municipal ment of industrialwastewater. Water Sci Technol2001; wastewater. Desalination 1997;113:189–96. 44(8):1–6. [3] Urbain V, Benoit R, Manem J. Membrane bioreactor: a [16] Masse D, Masse L. Treatment of slaughterhouse waste- new treatment tool. J AWWA 1996;88(5):75–86. water in anaerobic sequencing batch reactors. Can Agric [4] Hutter. M, Kramer-Schafhalter. A, Mayr B. Integration of Eng 2000;42(3):131–7. membrane technology in communal wastewater treatment: [17] Tritt W. The anaerobic treatment of slaughterhouse operation and cost analysis. Eur Water Manage 2000;3(3): wastewater in fixed-bed reactors. Bioresource Technol 33–42. 1992;41:201–7. [5] Stephenson T, Judd S, Jefferson B, Brindle K. Membrane [18] Ruiz I, Veiga M, de Santiago P, Blazquez R. Treatment of bioreactors for wastewater treatment. London, UK: IWA slaughterhouse wastewater in a UASB reactor and an Publishing, 2000. anaerobic filter. Bioresource Technol 1997;60:251–8. 908 W. Fuchs et al. / Water Research 37 (2003) 902–908

[19] Sayed S, de Zeeuw W, Lettinga G. Anaerobic treatment of [27] Martinez J, Borzacconi L, Mallo M, Galisteo M, Vinas M. slaughterhouse waste using a flocculent sludge UASB Treatment of slaughterhouse wastewater. Water Sci reactor. Agric Wastes 1984;11:197–226. Technol1995;32(12):99–104. [20] Sayed S, van Campen L, Lettinga G. Anaerobic treatment [28] Saxena K, KaulS, Hasan M, Gadkari S, Badrinath S. of slaughterhouse waste using a granular sludge UASB Packed bed anaerobic reactor for treatment of meat reactor. BiolWastes 1987;21:11–28. wastes. Asian Environ 1986;8:20–4. [21] Manjunath N, Mehrotra I, Mathur R. Treatment of [29] van Dijk L, Roncken G. Membrane bioreactors for wastewater from slaughterhouse by DAF-UASB system. wastewater treatment: the state of the art and new Water Res 2000;34(6):1930–6. developments. Water Sci Technol 1997;35(10):35–41. [22] Sayed S, van der SpoelH, Truijen G. A complete [30] Gander M, Jefferson B, Judd S. Aerobic MBRs for treatment of slaughterhouse wastewater combined with domestic wastewater treatment: a review with cost sludge stabilisation using two stage high rate UASB considerations. Sep Purif Technol2000;18:119–30. process. Water Sci Technol1993;27(9):83–90. [31] Beaubien A, Baty M, Jeannot F, Francoeur E, Manem J. [23] Borja R, Banks C, Wang Z. Performance of a hybrid Design and operation of anaerobic membrane bioreactors: anaerobic reactor, combining a sludge blanket and a filter, development of a filtration testing strategy. J Membr Sci treating slaughterhouse wastewater. Appl Microbiol Bio- 1996;109:173–84. technol1995;43(2):351–7. [32] Elmaleh S, Abdelmoumni L. Experimental test to evaluate [24] Borja R, Banks C, Wang Z. Effect of organic loading rate performance of an anaerobic reactor provided with an on anaerobic treatment of slaughterhouse wastewater in a externalmembrane unit. Water Sci Technol1998;38 (8–9): fluidised-bed reactor. Bioresource Technol1995;52:157–62. 385–392. [25] Nunez L, Martinez B. Anaerobic treatment of slaughter- [33] Choo K, Lee C. Membrane fouling mechanisms in the house wastewater in an expanded granular sludge bed membrane-coupled anaerobic bioreactor. Water Res (EGSB) reactor. Water Sci Technol1999;40(8):99–106. 1996;30(8):1771–80. [26] Polprasert C, Kemmadamrong P, Tran F. Anaerobic [34] Lema J, OmilF. Anaerobic treatment: a key technology baffle reactor (ABR) process for treating a slaughterhouse for a sustainable management of wastes in Europe. Water wastewater. Environ Technol1992;13:857–65. Sci Technol2001;44(8):133–40.