sign in sign up
<<
Home , Sewage treatment

Anaerobic bioprocessing of , focusing on degradation of linear alkylbenzene sulfonates (LAS)

Angelidaki I., Torang L., Waul C.M. and Schmidt J.E. Environment 8 Resources DTU, Technical University of Denmark, DK-2800 Lyngby, Denmark; e-mail: [email protected]; TL: +45 45251429; FAX: +45 45932850

Abstract. Anaerobic degradation of sludge amended with linear alkylbenzene sulfonates (LAS) was tested in one stage continuous stirred tank reactor (CSTR) and a two stages reactor system consisting by a CSTR as first step and upflow anaerobic sludge bed (UASB) reactor in the second step. Anaerobic removal of LAS was only observed at the second step but not at the first step. Removal of LAS in the UASB reactors was approx. 80% where half was due to absorption and the other half was apparently due to biological removal as shown from the LAS mass balance. At the end of the experiment the reactors were spiked with I4C-LASwhich resulted in 5.6% ''C02 in the produced gas. Total mass balance of the radioactivity was however not achieved. In batch experiments it was found that LAS at concentrations higher than 50 mg/l is inhibitory for the most microbial groups of the anaerobic process. Therefore, low initial LAS concentration is a prerequisite for successful LAS degradation. The results from the present study suggest that anaerobic degradation of LAS is possible in UASB reactors when the concentration of LAS is low enough to avoid inhibition of active in the anaerobic process. Key words: Sludge, linear alkylbenzene sulfonates (LAS), continuous stirred tank reactors (CSTR), upflow anaerobic sludge bed reactor (UASB), 2-step process, inhibition.

Introduction Anaerobic treatment of sludge is widely used but it must be investigated with respect to its ability to remove certain organic contaminants which may be toxic for many organisms at certain concentrations. Presently linear alkylbenzenes sulfonates (LAS) are the most important group of used in detergents and cleaning product formulations for household and (Beth et al., 1989). In the literature their behavior is well documented under aerobic treatment and they are known to biodegrade rapidly under these conditions (Schober, 1989; Connell, 1997; IPCE, 1996; Bema, et al., (1991, 1995); Romano and Ranzani, 1992; Giger et al., 1989, Prats et al., 1997, Holt et al., 1995 and Greiner and Six 1997). However there are uncertainties about their under anaerobic conditions. Furthermore at treatment plants at least 20% of the mass load on LAS entering the plant will be present on the suspended solids. In most treatment plants, the suspended solids are removed via primary settling and are directed to an anaerobic digester. There are generally large differences in the quantity of LAS in different kinds of . In general sludges exposed to aerobic conditions contain far less LAS than primary sludge or anaerobically digested sludge (Giger et al., 1989). This is Merargument to suggest that LAS is enriched in anaerobic treatment plants and does not really undergo biodegradation (Giger et al., 1989). The huge range of values and variations from the different authors can be attributed to the different measuring techniques used to quantify LAS (Painter and Zabel, 1989). Table 1 shows typical concentration of LAS in various sludge and sludge-amended soils. To date published experimental evidence of LAS being degraded under strict anaerobic conditions is rare. Recent publications have indicated that some removallprimary degradation of LAS in anaerobic treatment using UASB reactors is possible (Haagensen et al. 2002, Mogensen and Ahring, 2002; Sanz et al., 1999). It was also reported that LAS was degraded using NO-3 as the electron acceptor in the acidogenic step of a 2 stage UASB reactor setup (Almendanz et al., 2001). Anaerobic degradation of LAS in batch tests using inoculum from lake sediments, and activated sludge from aeobic environments as also been reported (Angelidaki et al., 2000). By increasing the bioavailable fraction of LAS in CSTR reactors Haagensen et al., (2002) found increased anaerobic removal of LAS. Denger and Cook, (1999) indicated that anaerobes were

161 isolated which were able to assimilate sulphonates as their source of sulfur. In fact they discovered that the facultative strain RZLAS was able to use LAS as a source of sulphur under anoxic conditions in a glucose-salts media. However no mechanisms or pathways for the process were proposed. None of these studies have been conducted with radiolabeled LAS to determine a radiochemical mass balance, identify LAS metabolites and unequivocally demonstrate anaerobic biodegradation of LAS. I Table 1 Typical concentration of LAS in various sludge and soils Type of Sludge Concentration (gjkg dry weight) Primary 5.34-6.31@’ Activated sludge 0.09-0.86”) Anaerobically digested sludge 5-15‘a) 5.2-30.2”) 2- 1 o@’ 3- 1 2‘d’ Aerobically digested 2.1-2.9@) 0.1-0.5(a) Air -dried digested sludge 0.15-0.16”) Agricultural soils amended 0.0002-0.02(8)+ with anaerobic digested sludge Sources: a: Jensen, (1999); b: Painter and Zabel, (1989); c: Giger et al., (1989); d IPCS, (1996); + Soil testing done 4-6 months after last application

So far the few recent observations of LAS being degraded under anaerobic condition does seem to conflict with other publications which indicate no degradation. However the experiments and the scenarios investigated by the different authors did not have many similarities. Lab tests conducted under anaerobic conditions especially with low easily degradable organic matter do need meticulous attention to achieve full anaerobic conditions. Methods available for determination of LAS especially in solid matrices are also complicated and may give misleading results. We will present results showing possibility for bioprocessing by using different reactor systems. Furthermore, batch experiments are made in order to investigate inhibition levels of LAS on different microbial groups of the anaerobic process. Finally we will focus on anaerobic degradation of LAS in CSTR and UASB reactors, including the results of an initial experiment with I4C-LAS.

Material and methods Experimental method for the batch experiments Toxicity tests were performed in batch experiments at 55OC with LAS spiked to initial concentrations of 0, 50, 141, 310, and 384 mg/L to test the inhibitory effects of LAS on different trophic groups of the process. Five different substrates were tested including glucose (5.6 M), butyrate (10 mM), propionate (10 mM), acetate (20 mM) and a mixture of HdCO2 (80:20) at 1 atm. Tests were set up in triplicate flasks (1 16 mL) and inoculated with 10 mL digested manure (Vegger plant) and added different concentrations of LAS, substrate and made up to a total volume of 40mL with mineral medium (Angelidaki et al. 1990). Background samples and autoclaved controls were also included. Removal of volatile fatty acids (VFA) were measured contemporary with methane production in the tests with glucose, propionate, and butyrate.

162 Experimental method for the continues feed experiments Experiments for determining the anaerobic degradability of LAS in sewage sludge under different reactor configurations were conducted using 2 CSTR reactors, and 1 UASB reactor (Fig.1). The first CSTR reactor (one of 4.5 1 total volume and 3 1 liquid volume) was operated as one stage fully mixed reactor with a hydraulic retention time (HRT) of 15 days. The other two reactors were used as a two-stages digestion system, with a CSTR reactor (RzsWes)(total volume of 1.1 L and a liquid volume of 80OmL) as an acidogenic reactor operated at a HRT of 2 days and the UASB reactor (RUASB)(total volume of 0.2 L) operated at 1 days HRT and treating the centrifuged from RzStages.The effluent from R2-stags was centrifiged before pumped into RUASBin order to avoid clogging. The content of the reactors was kept at constant of 35+1"C by the use of warm circulating through the outer jacket of the reactors.

CSTR + UASB

Figure 1. Reactor configurations used in continuous experiments

In the later stages during operation, 2 of the CSTR reactors and the UASB reactor were used to conduct more detail studies on the anaerobic biodegradability of LAS using I4C-labelled LAS.

Substrate characteristics The sludges used in the experiments were obtained fiom the Lundtofte plant, Denmark. The analysis carried out for characterising the sludge, are shown in table 2. The sludge were collected in plastic bottles, then later stored in a freezer at -18°C. Prior to feeding to the reactors the sludge were removed from the freezer and placed in a room at 4°C. Sludge fed to the CSTR reactors was diluted (50%) by a mineral medium (Angelidaki et al. 1990). Sludge was then amended with LAS-12 at a concentration of 100 mgL LAS C12 was the only compound tested among LAS with an alkyl chain length of 9 to 13 units. The phenyl positional isomers of LAS were: 6-Cl2 23.9%, 5-C12 18.8%, 4-C12 16.5%, 3-C12 17.3%, and 2-C12 21.0%, as reported by the manufacturer (Petresa, Spain). The feed solution for the 2 CSTR reactors were prepared in 1L batches each time the feed bottles was to be refilled. The holding time of feed in the feeding bottles was no more than 5 days except the R1-stageand only 2% days in the feed bottle for R2-stages.Prior to refilling of the feed bottles it was always ensured that the content already present was totally emptied, this ensured that the feed was

163 kept fresh, ensuring very little changes in the feed characteristics after being placed in the feeding bottles. Precautions were taken in order to entirely exclude contact of the feed with air. Gas bags containing N2:C02 (80:20) were attached to the -space of the influent substrate flasks, to ensure no degradation of LAS under aerobic conditions in the substrate flasks.

Table 2. Characteristics of the sludge collected at the Lundtofte wastewater treatment plant, Denmark and feed to the CSTR reactors Sludge Feed to the CSlX Feed to the after dilution with UASB reactor basic medium PH 6.53 6.97 6.71 TS (g/L) 19-25 11-15 2.66 (0.28) vs (a) 63% of TS 56% of TS TSS (a) 18-20 COD(g/L) 30-40 1.6-3.0 VFA(mmoVL) 29-30 Total N (g/L) 0.4 (0.01) 0.30 0.04 (0.01) LAS (mg/L) 13.45 90-1 10 13-22 LAS#(mg/g TS) 0.61 Note: values in brackets represent standard deviations. #:Calculatedbased on the mean TS values l4c-~sexperiment At day 83 the experimental setup was modified and I4C-labeled LAS (I4C-LAS: ring labeled 2- C12, specific activity: 0.294 mCi/mmol, purity > 99.7%, Riser National Laboratory, Denmark) was spiked to the influent of the UASB reactor and the total amount of added radioactivity was 40,000 Bq (z1.4 mg) over a period of 5 days. The use of radio labeled LAS made it possible to look for the mineralization products 14CH4 and 14 COZ. dioxide were absorbed in a series of two external base traps with 30 mL 1M NaOH after acidifying @H<2) the effluent in the waste bottle. Radioactivity was quantified by scintillation counting where 5 mL of liquid from the base was carehlly mixed with 10 mL scintillation liquid (Hisafe 3, Wallac). Methane was collected in gas bags after the carbon dioxide traps. The gas bags were after the 5 days added excess and oxidized in a specially designed furnace at 800°C in the presence of a copper oxide catalyst. All the produced carbon dioxide was collected in two traps, each containing 5mL Carbo-Sorb (Packard) and was afterwards added lOmL of Permafluor E" (Packard) prior to the liquid scintillation counting.

Analytical methods LAS CZ2: Samples of 10 mL. for LAS C12 quantifications were dried at 105°C followed by extraction with 5 mL for 48 h at 250 rpm, and then centrihged at 10,000 rpm for 10 min. Centrifuged sample of 2 ml was then dried at 55 "C and re-dissolved in 2 ml of eluent and transferred to 700pL HPLC vials and stored at -20°C prior to analysis. LAS quantification was conducted on a Perkin-Elmer LC 235, equipped with a C18 column (Phenomenex Luna 5 pm, 250~4.6mm) and a fluorescence detector (lex:232nm, lem:290nm), and a Perkin-Elmer Binary LC . Eluent consisted of 88% (v/v) MeOH and 12% Milli-Q water (Millipore Corporation, MA, USA) with 12 gLof NaC104. A flow rate of 1 mL/min was used. The detection limits were about 0.5 mg/L for LAS C12. Volatile fatty acids (VFA), i.e. acetate, propionate, isobutyrate, and butyrate were measured by gas chromatography equipped with a FID-detector (Hendriksen et al., 1996). Temperature and pH

164 were measured in the influent and effluent immediately after sampling using a PHM210 pH meter from MeterLab. Total solids (TS), Volatile Solids (VS), and Suspended Solids (SS) were determinate in accordance with the standard methods (American Association 1985). Dissolved organic carbon (DOC) was quantified on a Model 700 TOC-analyzer. Content of methane, carbon dioxide, and were measured on a GC 82-22 from Mikrolab , Denmark equipped with a dual column (1.1 m 3/16" molsieve and 0.7m x 1/41) chromosorb from Mikrolab Aarhus), and helium was used as carrier gas. Gas production was measured by a sequential bottle system, where the produced gas resulted in displacement of water which was used to quantify the gas produced. To avoid absorption of COZ in the water in the gas measuring bottles the pH was decreased to 2.

Results and discussion Toxicity tests The experimental protocol was designed to examine the effect of the anionic (LAS) concentration on the activity of the anaerobic microorganisms under thermophilic conditions. The methane production from butyrate is shown as an example on figure 2. Butyrate concentration was measured contemporary throughout the experiments and is shown as total VFA in Figure 3. The results from the batch test are summarized in Table 3 and indicate that the butyrate and propionate- utilising bacteria are more sensitive to the presence of LAS than the acetoclastic methanogens. This in accordance with reported fiom Gavala and Ahring (2002). Garcia-Morales et al. (2001) have reported an EC50 (methanogenic) = 6.3 mg/L and EC50 (acidogenic) = 18.9 mgL.

Table 3. Inhibition effects of LAS on different tropic groups of the an+aerobic process. LOEC = Lowest Observed Effect Concentration. Measured by the methane production. Measured by the reduction of the concentration of the corresponding volatile fatty acid (VFA). Substrate LOEC 100% inhibiton mglL mg/L H~lC02 141 384

Acetate 50 310

Propionate + 50 50 Buterate + 50 141 Glucose '* 141 310

40 20

$ 30 f 15 t - 2 20

0 0 0 5 10 15 0 5 10 15 Time (days) Time (days)

t0 mglL 4 50 mglL 8141 mglL -A- 310 mglL +- 384 mglL Figure 2 Methane production from butyrate versus Figure 3 Consumption of butyrate versus the concentration of LAS the concentration of LAS

Continuously fed reactor experiments R1.,,,, and RUASBwere functioning as methanogenic reactors and could successfully convert organic matter to methane and carbon dioxide (Fig. 4). Practically no methane production was

165 observed in Rz-~~~~~proving that this reactor was operating as true acidogenic reactor (hydrolysis) (Fig. 4).

2500 I 1 p 2000 -u -E 1500 2 1000 z5 500

0 1 2 3 4 5 6 7 Time (Weeks)

Figure 4 Methane production in the R1-mg,and R2-stagesand Rvnss

LAS concentration was monitored during the whole experimental period. LAS concentration in Rl-,,,, increased gradually from approx. 10 mgiL which was corresponds to the level of LAS present in the inoculum sludge used to start the reactors up. After approx. 1 month the concentration in the reactor became similar to the incoming LAS concentration. LAS removal was not observed in the RI-~~~reactor (data not shown). In the 2-stages reactor system was LAS not biodegraded in the first stage. (Fig. 5). LAS reactor concentration became similar to influent concentration after the initial start-up of the process. (The reactor start-up period of approx. 1 month is not illustrated in Figure). The inability to remove LAS in R2-staga indicates that separation of acidogenesis and methanogenesis was not promoting LAS biodegradation in contrast to the previous report by Almendariz et al., 2001. However, the first stage in the 2-stages system applied in the experiments of Almendariz et al., 2001 was not truly acidogenic, as methanogenesis was taking place. In our experiments no significant methane production was taking place in the first stage. However, effective removal of LAS was observed in at the second stage of the reactor system. Influent concentration to the RUA~Bwas approx. 20 mg/L (ranging from 13 to 23mg/L), while LAS concentration in reactor was approx. 5 mgiL (ranging fiom 2 to 9 ma)(Fig. 6). The results indicate that more than of 80% of the initial LAS present in the feed was removed. After the end of the experiment granules were extracted for LAS and 148 mg LAS was found absorbed in granules. This account only for half of the LAS fed in RUASB,which indicates that the other half was apparently biologically converted, i.e. 40% of the inlet LAS concentration to the RUASBwas apparently biodegraded .

166 160 25 - $, 140 i- p 20- E. 120 - c C .P 100 15- E; 80 -E ' 60 8 10~ 40 8 v) * 5- 4 20 4

0 0, I , m 0 10 20 30 40 50 60 70 80 90 0 102030405060708090 Time (days) Time (days) +reactor -feed +effluent bottle --t reactor -IC- feed

Figure 5 LAS concentration in the Figure 6 LAS concentration in the reactor reactor, feed and effluent bottle of R2-stages and feed of Runse

Initial mass balances were done on the UASB reactor for the period when 14C LAS was fed; the results showed that approximately 5.6 % of the total mass of I4C LAS fed for the duration of the experimental period was converted to 14C02. The quantity of 14C& was only 0.01%; this can be assumed to be zero for practical purposes. Furthermore there is a generally higher uncertainty of the scintillation counter when measuring samples with low Bq values, as in the measurements obtained for I4C& in the gasbags. If the production of 14C02 can be confirmed in further experiments, it would indicate that a small part of the LAS was totally mineralized. Other authors have reported LAS removaYprimaq degradation in UASB reactors; this would be the first time however that complete mineralization has been reported based on the formation of I4CO2 from 14C LAS.

Conclusions These preliminary experiments suggest that is fossible to degrade LAS in UASB reactors, leading to the formation of C02. The quantity of C02 produced was only approximately 5.6%. Further experiments with I4C 'LAS are in progress to confirm these results. No LAS removal was observed during acidogenic step. LAS was inhibitory for all the steps of the anaerobic degradation process for concentrations higher than 50 mg/L. Acetogenic bacteria (propionate and butyrate degraders) were more sensitive to LAS inhibition compared to methanogens.

Aknowledgements We thank Patricia Cristina Benito for her valuable help. This study was partly funded by the EU 5&Frame Program (BIOWASTE).

References Almendariz, F.J., Me&, M., Sobekn, G. and Monroy, 0. (2001) Degradation of lineal allcylbenzene sulphonate &AS) in an acidogenic reactor bioaugmented with a Pseudomonas aemginosa (M113) strain. Water Science and Technology, 44, 183-188. American Public Health Association. (1985) Standard methods for the examination of wafer and wastewater., Washington, D.C. AngelidakiJ., Petersen,S.P., and Ahring,B.K. (1990) Effects of Lipids on Thermophilic Anaerobic-Digestion and Reduction of Lipid Inhibition Upon Addition of Bentonite. Applied Microbiologv and Biotechnology 33 (4), 469- 472

167 Angelidaki, I., Mogensen, A.S. and Ahring, B.K. (2000): Degradation of organic contaminants found in organic . Biodegradation 11,377-383 BemaJ.L., Moreno,A., and Ferrer,J. (1991) The Behavior of las in the Environment. Journal of Chemical Technology and Biotechnology 50 (3), 387-398. Bema,J.L., Cavalli,L., and Renta,C. (1995) A Life-Cycle Inventory for the Production of Linear Akylbenzene Sulfonates in Europe. Tenside Sufactants Detergents 32 (2), 122-132 Connel, D.W. (1997). Soaps and Detergents. In: Basic Concepts of , 221-242. Lewis Publishers, Boca Raton, New York. Denger, K. and Cook, A.M. (1999): Note: Linear alkylbenzenesulphonate (LAS) bioavailable to anaerobic bacteria as a source of sulphur. Journal ojAppliedMicrobiology 86,165-168. Garcia-Morales,J.L., Nebot,E., Romer0,L.I. & Sales,D. Comparison between acidogenic and methanogenic inhibition caused by linear alkylbenzene-sulfonate (LAS). Chemical And biochemical Quarterly 15, 13-19 (2001). Gavala,H.N. & Ahring,B.K. Inhibition of the anaerobic digestion process by linm alkylbenzene sulfonates. Biodegradation 13,201-209 (2002). Giger , W., Alder, A.C., Bmer, P.H., Marcomini and Siegrist, H. (1989): Behavior of LAS in sewage and sludge and sludge treatment and in sludge-treated soil. Tenside Surfactants Detergents, 26,95-100. Greiner, P. and Six, E. (1997): Evaluation of the results of the LAS- monitoring in Germany.Tenside Sufactants Detergents, 34(4), 250-255. HaggenseqF., Mogensen,A.S., AngelidakiJ., and Ahring,B.K. (2002) Anaerobic treatment of sludge: focusing on reduction of LAS concentration in sludge. Water Science and Technology 46 (lo), 159-165. Hendriksen,K.V. and Ahring,B.K. (1996) Combined removal of and carbon in granular sludge: Substrate competition and activities. Antonie Van Leeuwenhoek International Journal of General and Molecular Microbiology 69 (I), 33-39 Holt, M.S., Walters, J., Comber, M.H.I., Armitage, R., Moms, G. and Newbery, C. (1995): AIS/CESIO Environmental surfactant monitoring programme. SDIA pilot study on linear alkylbenzene sulphonates (LAS). Water Resources, 29(9), 2063- 2070. International Programme on Chemical Safety (IPCS) (1996): Linear alkylbenzene sulphonates and related compound. critereia 169. WHO.Online documentation available at http://www.inchem.org Jensen, J. (1999): Fate and effects of linear alkylbenzene sulphonates (LAS) in the terrestrial environment. The Science of the Total Environment, 226,93-111. Mogensen,A.S. and Ahring,B.K. (2002) Formation of metabolites during biodegradation of linear alkylbenzene sulfonate in an upflow anaerobic sludge bed reactor under thermophilic conditions. Biotechnology and Bioengineering 77 (5), 483-488 Painter, H.A. and Zabel, T. (1989): The behaviour of LAS in sewage treatment. Tenside Surfactants Detergents, 26(2), 108-1 14. Prats, D., Ruiz, F., Vhquez and Rodriguez-Pastor, M. (1997): Removal of anionic and non-ionic surfactants in a wastewater treatment plant with anaerobic digestion. A comparative study. Water Resources, 31(8), 1925-1930. Romano, P and Ranzani, M. (1992) Anionic surfactants removal and biodegradation in a large treatment plant Wafer Science and Technology, 26(9-1 l), 2547-2550 Sam, J.L., Rodriguez, M., Amils, R., Bema, J.L., Femr, J and Moreno, A. (1999). Anaerobic biodegradation of LAS (Linear alkylbenzene sulfonate): Inhibition of the methanogenic process. La Rivista Italiana Delle Sostanze Grasse. Vol. 76,307-3 11 Schoberl, P. (1989): Basic Principles of LAS Biodegradation. Tenside Sufactants Detergents, 26(2), 86-94.

168 Treatment technologies for land use - hygiene