Biodegradation of Coproducts of Commercial Linear Alkylbenzene Sulfonate
Allen M. Nielsen, Larry N. Britton, Charles E. Beall, Timothy P. McCormick, and Geoffrey L. Russell
CONDEA Vista Company, Austin, Texas (1997)
Dialkyltetralin sulfonate (OATS) and single methyl-branched isomers of linear alkylbenzene sulfonate (iso-LAS) are coproducts that together can range from 1 to 10% of commercial LAS depending on the manufacturing process. Biodegradation studies using radiolabeled DATS and iso- LAS showed mineralization by indigenous microbial populations in laboratory simulations of aquatic and soil environments. Half-lives ranged from 2 to 20 days, which is rapid enough to suggest that accumulation would not occur in these environments. Upon exposure to laboratory activated sludge treatment, most model iso-LAS compounds showed greater than 98% parent compound removal, extensive mineralization (>50%), and 79-90% ultimate biodegradation (mineralization plus conversion to biomass). Activated sludge treatment of OATS and one of the iso-LAS isomers (methyl group attached to the benzylic carbon of the alkyl chain) resulted in >98% removal, 3-12% ultimate biodegradation and apparent formation of carboxylated biodegradation intermediates that accounted for 88-97% of the original material. These OATS and iso-LAS biodegradation intermediates continued to mineralize in simulated receiving water and soil environments at rates similar to that of sulfophenyl carboxylate (SPC) intermediates of a standard LAS.
Introduction Commercial linear alkylbenzene sulfonate (LAS) is a widely used synthetic surfactant in detergents and household cleaning products. For example, consumption of LAS in Europe, North America, and Japan was approximately 950,000 metric tons in 1994 (1). Numerous studies on the environmental fate and effects of LAS have been published (2). Extensive field monitoring studies in Europe and North America as well as evaluations by several governmental agencies have been completed (3-6). These studies confirm earlier conclusions that LAS is completely biodegradable; environmental concentrations of LAS are low, and the use of LAS at current levels is protective of biological populations in receiving environments (2,7). LAS is often represented as a linear alkyl chain attached to a sulfonated benzene ring, as illustrated in Figure 1. Commercial LAS, however, is a blend of LAS molecules that vary in terms of alkyl chain length, position of the benzene ring along the alkyl chain, and concentrations of coproducts called dialkyltetralin sulfonates and iso-LAS.
Dialkyltetralin sulfonates (DATS) and LAS with single methyl branching on the alkyl chains (iso- LAS) are minor components in commercial LAS. Concentrations range from <1 to 8% for DATS and <1 to 6% for iso-LAS depending on the manufacturing process used. Because of the widespread and high volume use of commercial LAS, significant quantities of LAS coproducts reach the environment. Consequently, several workers developed appropriate analytical methods and monitored DATS and DATS biodegradation intermediates (DATSI). DiCorcia et al. (8) detected and Trehy et al. (9,10) measured DATS and DATSI upstream and downstream of domestic wastewater treatment plants. Field and co-workers (11,12) conducted monitoring of these compounds in a sewage contaminated groundwater, and Tabor and Barber (13) found DATS in some bottom sediments of the Mississippi River. Cavalli et al. (14) demonstrated the complete biode gradability of model iso-LAS compounds in OECD screening and continuous flow activated sludge tests. Recently, Kölbener and his colleagues (15-17) have investigated the soluble organic carbon fraction which remained after biological treatment of commercial LAS in a laboratory, flow-through, test system using immobilized activated sludge. They concluded that this fraction was “refractory” and was composed primarily of biodegradation intermediates of DATS and iso-LAS. In order to better understand the fate of commercial LAS coproducts in the environment, 14C-ring-labeled DATS and iso-LAS model compounds were synthesized and exposed to simulated activated sludge, soil, and receiving water environments. Furthermore, the effluents coming from activated sludge treatment, which contained biodegradation intermediates, were exposed to simulated receiving water environments, and the fates of the radiolabeled chemicals were measured.
Materials and Methods
14C-uniformly-labeled D-glucose (98% radiochemical purity) was purchased from New England Nuclear, Boston, MA. [14C]benzene ring-labeled C12 linear alkylbenzene sulfonate (LAS) was synthesized by New England Nuclear and was shown by autoradiography of thin layer chromatograms (TLC) to be 97.5% radiochemically pure. The Procter and Gamble Company provided the following [14C]benzene ring-labeled single methyl branched alkylbenzene sulfonates (Figure 1): iso-LAS type IA (sodium 5-methyl-2-undecyl [14C]benzenesulfonate), iso-LAS type lB (sodium 10-methyl-2-undecyl [14C]benzenesulfonate), iso- LAS type IIA (sodium 2-methyl-2-undecyl [14C]benzenesulfonate),and iso-LAS type uB (sodium 5- methyl-5-undecyl [14C] benzene-sulfonate). The radiochemical purities of the iso-LAS compounds were determined by high-performance liquid chromatography (HPLC) radiochemcical analysis. Purities were iso-LAS type IA, 97.8%; iso-LAS type lB, 77.6%; iso-LAS type IIA, 94.7%; and iso- LAS type IIB, 97.5%. Two [14C] benzene ring-labeled C8 DATS samples were synthesized and used in this study (Figure 1). Huntsman Corporation, Houston, TX, provided the first sample which was shown by HPLC radiochemcial analysis to be 97.3% chemically pure and 92.7% radiochemically pure. The second sample was synthesized by Wizard Laboratories, Davis, CA, under the direction of CONDEA Vista Company. The radiolabeled dialkyltetralin (DAT) sample was analyzed at 98.9% chemical purity by gas chromatography (GC) and GC-mass spectrometry (MS). Following sulfonation, the radiolabled DAT sulfonic acid was shown by TLC auto radiography to be 99.6% radiochemically pure. The DAT sulfonic acid was neutralized before use.
CONDEA Vista Company synthesized a C12-LAB which was determined by GC analysis to be >98% chemically pure. It was sulfonated and neutralized before use. A type I C12 iso-LAS (i.e., methyl branching on the nonbenzylic carbon atoms at various positions on the aliphatic chain) was prepared by the alkylation of benzene with an iso-alcohol in an excess of A1C13 This iso-alcohol was a mixture with hydroxyl groups alpha to methyl groups and with each vicinal OR and CH3 pair at different positions along the carbon chain. In this process, some benzylic methyl-branched iso-LABs (type II) are inevitably formed. Analysis by GC, GC-MS, HPLC, and 1R and 13C nuclear magnetic resonance (NMR) showed that the majority of the sample (70.5%) was type I iso-LAB; approximately one-fourth (23.3%) was iso-LAB type II and 6.2% was DAT. It was referred to as the type I iso-LAS following sulfonation and neutralization. A second iso- LAB was synthesized and analyzed (GC and GCMS) by CONDEA Vista Company. Since it was shown to contain 90.0% type II iso-LABs, 5.6% C12-LAB, and a 4.4% mixture of DAT and indanes, it was referred to as the type II iso-LAS sample after sulfonation and neutralization. A C8 DAT sample, 89.9% chemical purity by GC analysis, was also synthesized by CONDEA Vista Co. and designated as the C8 DATS sample after sulfonation and neutralization.
Sample Collection/Handling and Test System Setup
Surface soils were collected at the following locations: (1) “pristine” soil from McKinney Falls State Park, Austin, TX; (2) sludge-amended soil from a cornfield adjacent to the Hornsby Bend Wastewater Treatment Plant, Austin, TX; and (3) gray water contaminated soil from the top of a percolation bed that receives surface applications of laundry water from a laundromat at Summit Lake, WI (18). The samples were collected, sealed in”zipper” -seal, plastic bags, stored at 4 0C, and used within 20 days of collection. Samples were screened through no. 4 (4.7 mm opening) and no. 14 (1.4 mm opening) sieves to remove vegetation, rocks, and debris. The moisture content was maintained between 59 and 94% of water field capacity, and samples (90 g) were mixed with 50% by volume perlite to promote aeration and to minimize clumping and compaction.
Each soil sample was mixed thoroughly with 10 mL of a mineral salts medium (19) containing the test surfactant and then placed in a 500 mL Gledhill flask (Ace Glass Inc., Vineland, NJ). Internal C02 traps received 5 mL of 0.5 N KOH with 0.0002% thymolphthalein pH indicator. Units were sealed, flushed with 70% 02/30% N2 for 3 mm at one L/min, and incubated at 23-25 0C. The KOR trapping solution was periodically replaced, and trapped 14C02 determined by liquid scintillation counting (LSC) in Ultima Gold AR scintillation fluid, Packard Instruments, Meriden, CT, in a Packard Instruments Model 2550 TR analyzer.
Sediment samples were collected from the upper inch of a small stream (Lake Creek, Austin, TX), which received effluent from a domestic wastewater treatment plant. Sediments were transported and held in plastic buckets at 4 0C and were used within 24 h of collection. Sediment samples (50 or 98 g) with 100 mL of creekwater were placed in duplicate 500 mL Gledhill flasks and spiked with two mL of the test surfactant in the mineral salts medium. The testing was done as described above except that incubation was with shaking at 70 rpm on a gyrotary shaker.
Periphyton samples were collected as rocks (approximately 1 inch diameter) coated with heavy growth from the same stream locations as were the water and sediment samples. The rocks were carefully placed in “zipper”-seal bags to minimize disturbance of the periphyton layer, transported at 4 0C and used the same day. Four to five rocks were placed in duplicate flasks, covered with 100 mL of overlying water, and spiked with the test compound in mineral salts medium. Incubations and sampling were the same as for the sediment/ water samples. Each test system consisted of duplicate test flasks and a control flask which was identical except for the addition of HgCl2 at 1.0g/flask in waters or 2.5 g/flask in soils. Tests lasted at least 30 days. Radiochemical recoveries of 14C in solids, liquids, and C02 (including 14C032- and H14CO3-) were done at the end of each test and ranged from 65 to 107%; most were >90%.
Porous Pot Biodegradation Test System. The porous pot method for assessing biodegradation of the test compound in a simulated wastewater activated sludge was a modification of ASTM test method E1798-96. The porous pot system consisted of cylindrical glass containers with porous (>65 µ pore size) high-density polyethylene (HDPE) candles inserted into the glass cylinders (Figure 2). The candles held 282 mL of activated sludge, and a glass aeration tube to the bottom of the candle provided mixing and aeration at a 0.5 standard ft3 /h flow rate. Sewage feed was pumped into the units from a 10 L carboy container through Teflon tubing by means of Model QG6 laboratory metering pumps (F.M. I., Inc., Oyster Bay NY). Test chemicals were fed via Teflon tubing with Model 22 syringe pumps manufactured by Harvard Apparatus (South Natick, MA). Effluents from the porous pot units were collected in 2 L vacuum flasks which were manifolded to the laboratory vacuum system. The C02 trapping system was a series of three 500 mL gas collection bottles, each with 300 mL 2 N KOH. A comparison between the performance of the laboratory scale porous pot biodegradation test system and conventional activated sludge treatment in the U.S. is shown in Table 1.
A 21 day acclimation phase was initiated by feeding settled domestic sewage [adjusted to a chemical oxygen demand (COD) of 400 mg/LI and the nonradioactive analogs of the test compounds. The units were operated with the following parameters: nonradioactive chemical feed concentrations: C12 LAS, 5 mg/L; C8 DATS, 500 µg/L; iso-LAS, 250 g/L; hydraulic retention time (HRT), 0.25 days; sludge retention titne (SRT), 10 days (DATS units were 20 days); activated sludge concentration, ~2000 mg/L volatile suspended solids (VSS). Steady state conditions, which were defined as a period of 7 days, in which COD removal was >90% and HRT daily variation was <5%, were obtained and maintained over the final week of acclimation.
During the 15 day test phase, the radioactivities in C02, liquids and solids, and effluent total suspended solids (TSS) and COD were determined each day, and the removal of test compound from the feed by activated sludge treatment was measured once. During the last 10 days, remaining effluents were collected and frozen until use in the die-away tests. Appropriate amounts of sludge were wasted each day in order to maintain the designated sludge retention time (SRT). Radiochemical recoveries for the porous pot test were calculated by adding the total radioactivity recovered in C02, solids, and liquids during the test. Corrections for radiolabeled carbonates and bicarbonates were included. Total recoveries ranged from 95-102%.
Die-Away Tests with Porous Effluents. The combined effluents from individual units were tested for mineralization of radiolabeled parent and intermediate compounds in 2 L Gledhill flasks as described above except that 1 L of combined porous pot effluent and 10-15 periphyton covered rocks or 250 g of sediment were used. In the soil die-away with DATS and LAS, 400 g of sludge amended soil, and 27 mL of effluent were used. All tests were run at least 30 days, and an abiological control for each duplicate was prepared by adding 1 g of HgCl2. The radioactivities at the end of each test were measured in the solids, liquids, and C02 (including corrections for radioactive bicarbonate and carbonates) and were combined to give the radiochemical recoveries. Total recoveries ranged from 67.5 to 100%.
Analysis of Porous Pot Effluents and Porous Pot Die-Away Samples(bold). Approximately 1 L of the remaining water column samples of the porous pot effluents and porous pot effluent die-aways were filtered through 0.22µ Millipore filters. These filtrates were concentrated from 35 to 93-fold in a RotoVap system at 90 0C. Pretests demonstrated that no radioactivity was lost in the overhead fraction. A chromatographic HPLC-ultraviolet fluorescence (UVF) analysis was done on the concentrates to increase sensitivity and improve chromatographic peak shapes over the ion-pair method used to analyze the purity of the test materials. The column used was a DionexASl 1,4 x 250 mm, P/N 44076. Eluent flow rate was I mL/min, and 100 µL was injected. Eluent components were A, 0.2 N NaOH in H20; B, 95/5 acetonitrile/H20; and C, distilled water. The gradients were initial A, 5%; B, 10%; C, 85%; 0 to 10 mm linear gradient to A, 40%; B, 10%; C, 50%;10 to 20 mm hold at A, 40%; B, 10%; C, 50%; 20 to 30 mm linear gradient to A, 40%; B, 60%; and C, 0%; 30 to 40 mm hold at A, 40%; B, 60%; C, 0%; step change to A, 5%; B, 10%; C, 85% then hold to 55 mm. UVF excitation was 225 nm, and emission was at 295 nm. HPLC effluent fractions were collected every minute for the first 40 mm alter injection except that with DATS only 10 fractions were collected. Radioactivity in the each fraction was determined by liquid scintillation counting (LSC). Radiochemical recoveries were the ratios of total radioactivity of all fractions in a sample to the unfractionated sample radioactivity. Recoveries ranged from 83.3 to 108.2%; most were >95%.
Data for C02 evolution were analyzed using nonlinear regression models, a technique which has been successfully applied to several chemicals which manifest first order kinetics (21). The half-life and asymptote were calculated using Statgraphics Plus 5.2 software. These were obtained by minimizing the residual sum of squares using a search procedure suggested by Marquardt (22). All C02 data were corrected for C02 released in the HgCl2 controls before use in kinetic analysis.
Results and Discussion
Mineralization of Model Compounds in Aquatic and Soil Environments. Since commercial LAS is primarily discarded down-the-drain with or without sewage treatment, the initial objective was to evaluate the potential ultimate biodegradation of DATS and iso-LAS in receiving environmental compartments. A typical receiving water (Lake Creek) was the first environmental compartment tested. Long lag times (>10 days) and very slow rates were observed which did not fit first-order kinetics. Therefore, no attempt was made to calculate rate constants for DATS, iso-LAS, and standard LAS in Lake Creek water. This was probably due to the very low level of biodegradable organic compounds (BOD5< 2 mg/L) which resulted in a low population of bacteria (<15 000 cfu/ mL) in the water. By adding biomass to the creekwater with either periphyton-covered rocks or sediment, rapid microbial mineralization occurred (Table 2)
Table 2 Mineralization of Radiolabeled Compounds in Aquatic Environments rate (a) extent(a) [t1/2 (days)] [%T(14) CO(2)(30days)](b) Creek Water and Periphyton-Covered Rocks LAS 3.4; 4.6 60; 76 DATS 14.2; 18.2 19; 19 iso-Las type IA 3.1 71 IB 3.2 70 IIA 4.0 63 IIB 8.6 47 glucose 3.7; 2.6 61; 72 Creek Water and Sediments LAS 4.2; 12.4 56; 63 DATS 10.8; 21.0 27; 30 iso-Las type IA 3.6 50 IB 5.1 52 IIA 5.4 50 IIB 12.1 39 glucose 5.4; 3.6 52; 54 (a) More than one value indicates results of separate experiments (b) Percent theoretical carbon dioxide production from radiolabeled substrate at 30 days
The rates for LAS and most of the iso-LAS isomers were biphasic. The first-order initial rates were at least twice as fast as the apparent zero-order final rates. Explanations on the mechanism or significance of the final rate are lacking. The initial mineralization rate may correspond to both mineralization and incorporation of the radiolabeled substrate carbon into cell components. The slower, final rate may reflect the turnover of the incorporated carbon, or it may represent the formation of more biologically stable catabolites. Static conditions of the experimental system resulting in depletion of cosubstrates, essential nutrients, etc., could also be an explanation. Since the radiolabel is in the benzene ring and assuming that, like LAS, the benzene ring is the last point of catabolic attack, the evolution of C02, regardless of amounts, during breakdown of the DATS and iso-LAS intermediates, is evidence that stable catabolites probably are not responsible for the slower final rate. Only the initial rates and the extents of mineralization alter 30 days are reported in Table 2. In all cases, the mineralization after 30 days had reached a final, slow rate. Mineralization continued (as illustrated by typical curves shown in Figure 3) until termination of the experiments, which for some incubations was 120 days. The rate and extent data are consistent with the interpretation that accumulation would not occur in environments that receive treated wastewater. The large-scale application of wastewater effluents and sludges to soils has led to concerns about the potential accumulation of surfactants in soil environments. Since a large portion of sewage sludges is applied to land in Europe and North America, soil biodegradation should be an important process for removing surfactant residues from the terrestrial environment (25). Gray water contaminated soils and sludge-amended soils offer an opportunity to study biodegradation of DATS and iso-LAS in soils which have been exposed to commercial LAS and should have microbial populations that are acclimated to these coproducts. The laundromat site at Summit Lake, WI, offers an ideal location for such samples and has been extensively characterized and evaluated (18, 24). Surface samples were obtained from the top of the percolation bed which was exposed directly to washing machine effluent. Both DATS and LAS mineralization began immediately and continued in typical curvilinear fashion until the test was terminated after 45 days (Table 3).
Table 3 Mineralization of Radiolabeled Compounds in Soils rate (a) extent(a) [t1/2 (days)] [%T(14) CO(2)(30days)](b) Gray Water Contaminated Soil; Summit Lake, WI LAS 1.8 50 DATS 7.2 33 glucose 1.0 31 Sludge Amended Soil; Austin TX LAS 1.8; 2.3 53; 57 DATS 9.8; 17.7 30; 42 iso-Las type IA 1.6 48 IB 1.6 56 IIA 2.8 60 IIB 5.0 64 glucose 0.8; 3.2 37; 41 (a) More than one vlue indicates results of separate experiments (b) Percent theoretical carbon dioxide production from radiolabeled substrate at 30 days
The mineralization rates and extents of both DATS and iso-LAS in sludge-amended soil are also shown in this table.
The mineralization of iso-LAS (and standard LAS for comparison) was immediate and rapid, whereas DATS mineralization proceeded at a slower rate. Even though DATS mineralization rates are slower than LAS rates in this side-by-side comparison, the rates for DATS are within the range of those reported for LAS in other sludge-amended soils (26).
Although not shown in Table 3, the mineralization of DATS (over 40% within 30 days) was observed in a pristine soil. Because of no known agricultural or industrial activity or recent habitation at this site, the microbial population presumably has never been exposed to commercial LAS. These results suggest that the microbial capacity to mineralize DATS is ubiquitous in soils as has been reported for LAS (23). In this pristine soil, as well as occasionally observed in other soils, the extent of mineralization of LAS and the iso-LAS and DATS coproducts exceeded that of the reference glucose. Unexpectedly low mineralization of radiolabeled glucose in soil has also been observed by Sharabi and Bartha (27). They attribute this to unavailability due to adsorption or alternatively to the possibility that nongrowing steady-state microbial communities may turn over glucose very differently from a community that is growing in response to significant substrate addition. The latter implies that the type and concentrations of test substrates, relative to soil organic matter, may affect mineralization. In any event the data from Table 3 should not be interpreted as indicating that the mineralization of LAS and coproducts is greater than glucose.
Simulated Activated Sludge Treatment and Die-Away Tests
An important consideration of any down-the-drain chemical is its fate in domestic wastewater treatment systems. Therefore, the fate of the model compounds was examined in the porous pot biodegradation system which simulates the most common treatment system-activated sludge. Activated sludge treatment (Table 4) effectively removes the coproducts from wastewater as indicated by biological removal values approaching 99%.
Table 4 Fate of LAS Coproducts in Porous Pot Activated Sludge Treatment(a) parent mineral- percent ultimate residual removal ization in biodegrad. liquids (%) (%14CO(2)) biomass (%) (%) LAS 98.4 57.5 28.6 86.1 13.9 DATS 98.6 .8 1.8 2.6 97.4 iso-LAB IA 99.7 53.0 26.2 79.2 20.8 IB 99.3 58.2 31.5 89.7 10.3 IIA 99.5 51.6 27.8 79.4 20.6 IIB 99.5 7.6 4.2 11.8 88.2 (a) Balues normalized to 100%; actual 14C recoveries wree 95-102%. Ultimate biodegradation = mineralization + radioactivitiy in biomass.
Like LAS, most of the iso-LAS isomers underwent extensive mineralization (>50%) and ultimate biodegradation (79-90%) but released some (10-20%) of their carbon as water soluble intermediates. Activated sludge treatment of DATS and iso-LAS type IIB results in nearly complete removal (>98%) of the test surfactant, some ultimate biodegradation (3-12%) and a concurrent release of apparent carboxylated biodegradation intermediates (88-97%). These intermediates could be characterized as (1) having intact aromatic rings which demonstrate IIV fluorescence; (2) having increased polarity compared to the parent compounds; and (3) eluting on the HPLC-UVF chromatograms in the same region as with SPC biodegradation intermediates of LAS. Confirmatory evidence comes from the recent work of Cavalli et al. (14) and Kölbener et al. (16,17) and the monitoring work of Trehy et al. (10) in which carboxylated intermediates were found to be the major intermediates remaining after biological treatment of commercial LAS. Since the rate-limiting and last steps in SPC biodegradation are desulfonation, ring cleavage, and further oxidation of the ring-cleavage products (28), it is reasonable to assume that analagous biochemical pathways occur during the complete biological destruction of DATS and iso-LAS. Therefore, the complete biodegradation of these coproducts can be assessed by following the oxidation of the ring-labeled compounds in the porous pot effluents as summarized in Table 5.
Table 5 Fate of Biodegration Intermediates in Aquatic and Soil Environments rate (a) extent(a) [t1/2 (days)] [%T(14) CO(2)(30days)](b) Creek Water and Sediment LAS 5.3; 12.1 14; 32 DATS 6.8 9 iso-Las type IA 9.3 53 IB 15.0 35 IIA 34.4 22 IIB 23.2 33 Creek Water and Periphyton-Covered Rocks LAS 3.9; 11.9 23; 34 DATS 2.4 18 iso-Las type IA 4.0 44 IB 9.4 31 IIA 13.7 28 IIB 27.1 30 Sludge-amended Soil LAS SPC 7.8 22 DATS 6.2 34 (a) More than one value indicates results of separate experiments.
No kinetics were calculated for the mineralization of porous pot effluents in the creekwater alone. However, significant mineralization, which followed first-order kinetics, occurred if biomass in the form of periphyton-covered rocks, sediments, or sludge amended soils were added. Furthermore, the mineralization rates for the DATS and iso-LAS carboxylated intermediates were similar to those for the LAS SPCs. Table 6 presents the radiochemical mass balances at the temination of the die-away experiments with creekwater plus periphyton and creekwater plus sediments. Total C02 production was significant for all the intermediates during the 77 or 37 day incubations except for the DATS catabolic intermediates (DATSI) in the water/sediment incubations. In contrast, the total C02 from DATSI incubations in soil was 35% during the 37 day incubation (data not shown).
Table 6 Radiochemical Mass Balances for Porous Pot Effluent Die-Aways(a) water/periphyton actual recovery% CO2 soluble(b) Solids(b) LAS 91.4 48.1 43.1 8.8 DATS 67.5 29.2 62.5 8.3 iso-LAS IA 90.2 59.5 11.2 29.3 IB 92.6 45.3 24.9 29.8 IIA 89.2 51.1 32.6 16.3 IIB 88.4 60.8 31.7 7.5 water/sediment LAS 100.3 46.9 23.6 19.5 DATS 89 11 75.8 13.1 iso-LAS IA 93.7 70.3 5.2 24.5 IB 87.9 57.1 17.2 25.7 IIA 94.3 56.4 7.1 36.5 IIB 96.7 50.3 24.1 25.6 (a) Die-away incub ations in creek water plus periphyton or creek water plus sediments were 77 days except for DATS which was 37 days. (b) Percentage recoveries of CO2, soluble fraction and solids fraction are normalized to 100%. Actual recoveries are given.
Therefore, it appears that mineralization of DATSI can vary with the environmental conditions and presumably the microbial populations. The transient nature of these carboxylated biodegradation intermediates was demonstrated by HPLC-UVF/radiochemical detection. Water soluble fractions were analyzed of the porous pot feeds, effluents, effluent/sediment, and effluent/ periphyton die-aways, and the results are displayed in Figures 4, 5, and 6. The typical chromatographic pattern of C12-LAS is shown in Figure 4. The parent compound is almost completely removed during porous pot activated sludge treatment, and about 14% of it is released into the effluent as many water-soluble biodegradation intermediates. During the effluent/sediment die-away, 47% of the remaining radioactivity was mineralized, and only 3.3% remained in solution alter 77 days of exposure. Only the results for LAS are shown since the HPLC- UVF/radiochemical patterns for iso-LAS types IA, lB, and IIA were the same as those for LAS. Furthermore, very little (6% or less) of these starting radio-activities remained after the die-away. The biodegradation pattern for iso-LAS type IIB is shown in Figure 5. It is clear that one major and several minor water-soluble SPCs are formed during activated sludge treatment. Cavalli (14) also found that one major and several minor SPCs were formed during activated sludge treatment of another iso-LAS type II isomer of the same homolog. This isomer, however, showed a higher level (~87%) of ultimate biodegradation than the type IIB during activated sludge treatment. Even though these SPC intermediates formed by activated sludge treatment of iso-LAS type IIB make up most (88%) of the starting radioactivity, they are not recalcitrant since exposure to biomass on sediments during the 77 day die-away mineralized all the minor SPCs and over 50% of the major intermediate. A very similar pattern was observed for the DATS biodegradation as illustrated in Figure 6, except that almost all (97%) of the intact DATS were converted to water soluble intermediates (DATSI) in the porous pot treatment. These DATSI were metabolized by the presumably different microbiological populations found in the sediments and periphyton. The end result of the DATSI die-aways in water/ periphyton was removal of 56% of the soluble radioactivity (97% or original radiolabel at beginning minus 41% remaining) during the 37 days. Kölbener and co-workers have concluded from their studies on the biodegradation of commercial LAS in a laboratory “trickling filter” test system using immobilized activated sludge that from 3.2 to 13.6% of the commercial LAS carbon is refractory (16). These researchers have extended their studies and have reported that major components of this refractory portion are carboxylated biodeg radation intermediates of DATS and iso-LAS (17). Their conclusion was that many of the coproducts of commercial LAS are recalcitrant. The porous pot results of this paper and field monitoring results of Trehy et al. (10) are consistent with the observations of Kölbener et al. (15-17). It is clear from these studies, as well as other unpublished observations, that the microbial populations of domestic and industrial activated sludge and trickling filter wastewater treatment plants are not capable of the complete mineralization of all the DATS and iso-LAS homologs and isomers found in commercial LAS even though nearly complete primary biodegradation of these coproducts occurs. The new information provided in this paper is that these DATS and iso-LAS biodegradation intermediates produced during wastewater treatment could mineralize when exposed to receiving water environments or soils that contain a more diverse population of degraders.
We thank John Lin (CONDEA Vista Company), James Innis (The Procter and Gamble Company), and Paul Sieving (Wizard Laboratories, Inc.) for synthesis of test compounds; Bruce Leach (formerly of CONDEA Vista Company) for assistance with kinetic calculations; and Paul Filler (CONDEA Vista Company) for technical assistance and John Heinze (Council for LAB/LAS Environmental Research) for helpful discussions.
(1) SRI International, Chemical Economics Handbook, 610.5000 E, 1995.
(2) Environmental and Human Safety of Major Surfactants. Volume 1. Anionic Surfactants. Part 1. Linear Alkylbenzene Sulfonates; Final report from A. D. Little, Inc. to the Soap and Detergent Association: New York, 1991; pp 111-1-V-SB.
(3) Waters, I., Feijtel, T. C. J. Chemosphere 1995. 30 (10), 1939-1956.
(4) McAvoy, D. C., Eckhoff, W. S., Rapaport, R. A. Environ. Toxicol Chem. 1993, 12, 977-987.
(5) Second Report of the Technical Committee on Detergents and the Environment; United Kingdom Department of the Environment, 1994; pp 53-60.
(6) Feijtel, T. C. J., van de Plassche, E. J. Environmental Risk Characterization of 4 Major Surfactants used in the Netherlands; Report No. 67901-025; National Institute of Public Health and Environmental Protection; Bilthoven, The Netherlands and the Dutch Soap Association: Zeist, The Netherlands, 1995.
(7) Kimerle, R. A. Tenside, Surfactants, Deterg. 1989, 26 (2), 169-176.
(8) Di Corcia, A., Samperi, R., Marcomini, A. Environ. Sci. Technol. 1994, 28, 850-858.
(9) Trehy, M. L., Gledhill, W. E., Orth, R. G. Anal. Chem. 1990,62, 2581-2586.
(10) Trehy, M. L., Gledhill,W. F., Mieure, J. P., Adamove, J. E., Nielsen. A. M., Perkins, H. O., Eckhoff, W. S. Environ. Toxicol. Chem. 1996, 15 (3), 233-240.
(11) Field, J. A., Barber, L. B., II, Thurman, M. E., Moore, B. L., Lawrence, D. L., Peake, D. A. Environ. Sci. Technol. 1992,26, 1140-1148.
(12) Field, J. A., Leenheer, J. A., Thorn, K. A., Barber, L. B., II, Rostad, C., Macalady, D. L, Daniel, S. R.J. Contam. Hydrol. 1992, 55-78.
(13) Tabor, C. F., Barber, L. B., II. Environ. Sci. Technol. 1996, 30, 161-171.
(14) Cavalli, L., Cassani, G., Lazzarin, M., Maraschin, C., Nucci, G., Berna, J.L., Bravo, J., Ferrer, J.; Moreno. A. Toxicol. Environ. Chem. 1996, 54, 167-186.
(15) Kölbener, P., Baumann, U., Leisinget, T., Cook, A. M. Environ. Toxicol. Chem. 1995, 14 (4), 561-569.
(16) Kölbener, P., Baumann, U., Leisinger, T., Cook, A. M. Environ Toxicol. Chem. 1995,14 (4), 571-579.
(17) Kölbener, P., Ritter, A., Cotradini, F., Baumann, U., Cook, A. M Tenside Surfactants. Deterg. 1996,33 (2), 149-157.
(18) Larson, R. I., Federle, T. W., Shimp, R. I., Ventullo, R. M. Tenside, Surfactants, Deterg. 1989, 26 (2),116-121.
(19) Dworkin, M., Foster, I. J. Bacteriol. 1958, 75,592-603.
(20) Tchobanoglous, G. Wastewater Engineering: Treatment Disposal Reuse; McGraw-Hill Book Co.: New York, 1979; pp 44-45.
(21) Larson, R. I., Payne, A. G. Appl. Environ. Microbiol. 1981,41(3),621-627.
(22) Marquardt, D. WI. Soc. Ind. Appl. Math. 1963, 2,431-441.
(23) Knaebel, D. B., Federle, T. W., Vestal, J. R. Environ. Toxicol. Chem. 1990, 9, 981-988.
(24) Federle, T. W., Ventullo, R. M. Appl. Environ. Microbiot 1990, 56 (2), 333-339.
(25) Rapaport, R. A., Larson, R. J., McAvoy, D. C., Nielsen, A. M., Trehy, M. 3rd CESIO World Congress; 1992, Section E, pp 78-87.
(26) Berna, I. L., Moreno, A., Ferrer, J. J. Chem. Technol. Biotechnol. 1991, 50, 387-398.
(27) Sharabi, N. E., Bartha, R. Appl. Environ. Microbiol. 1993, 59, 1201-1205.
(28) Schoberl, P. Tenside, Surfactants, Deterg. 1989, 26 (2), 86-94.
Received for review January 10, 1997. Revised manuscript received August 11,1997 Accepted September 13, 1997®
Abstract published in Advance ACS Abstracts, October 15,1997.