Environ. Sci. Technol. 2007, 41, 4965-4971
Enantioselective Degradation of Organochlorine Pesticides in Background Soils: Variability in Field and Laboratory Studies PERIHAN BINNUR KURT-KARAKUS,† JACQUELINE L. STROUD,‡ T E R R Y B I D L E M A N , * ,† K I R K T . S E M P L E , ‡ LIISA JANTUNEN,† AND KEVIN C. JONES‡ Centre for Atmospheric Research Experiments, Science and Technology Branch, Environment Canada, 6248 Eighth Line, Egbert, Ontario, L0L 1N0, Canada, and Department of Environmental Science and Centre for Chemical Management, Lancaster Environment Centre, Lancaster University, Lancaster, LA1 4YQ, United Kingdom
Variability in the enantioselective degradation of chiral organochlorine pesticides (R-HCH, cis- and trans-chlordane (CC and TC), and o,p′-DDT) in the field and laboratory was investigated. Background soils presumably receive the same EF signature of a compound via atmospheric deposition and then degrade that compound in a way that can vary over small spatial areas. Background soils from woodland and grassland areas were sampled to compare chiral signatures and determine the spatial variability within a few square meters. The enantiomer fractions, EF ) areas of the (+)/[(+)+(-)]-enantiomers, showed variability between and within ecosystems. For example, the EF of CC varied between 0.272 -and 0.558 in nine samples taken over a few square meters, and a range of 0.4310.506 was found within depths of a few centimeters. Woodland and grassland soils were spiked with R-HCH, TC, CC, and o,p′-DDT, and one portion was placed in the field to monitor changes in EF under in situ conditions and the other taken to the laboratory. In general, the enantiomer degradation preferences in the laboratory paralleled those in the field, with some differences. Soil organic matter content and pH exerted a minor influence on this variability. The results of this study have implications for the use of chiral compounds to make inferences about airsoil exchange and for the mechanisms of biodegradation/ biotransformation of anthropogenic compounds in soils.
Introduction Chiral compounds are of interest to environmental chemists because they can provide information about chemical sources, environmental cycling, and processing. Microbially mediated processes in soils and sediments can be enantioselective, resulting in non-racemic mixtures in the environment over time. Several studies have used profiles of enantiomers to assess how fresh or aged a particular chemical is in soil and water (1-7). Air above soil may contain * Corresponding author phone: (705)458-3322; fax: (705)458-3301; e-mail:
[email protected]. † Environment Canada. ‡ Lancaster University. 10.1021/es0620787 CCC: $37.00 Published on Web 06/07/2007
2007 American Chemical Society
organochlorine pesticides (OCPs) that have chiral signatures similar to the soil residues, giving evidence for soil-air exchange and the influence of old, weathered residues versus fresh use. Several organochlorine compounds have been the focus of such work, namely, R-hexachlorocyclohexane (RHCH), chlordanes (enantiomers of cis- and trans-chlordane and MC5), o,p′-DDT, and chiral PCBs (e.g., refs 3 and 7-12) To use chiral compounds as markers of air-soil exchange, it is necessary to understand the variability in proportions of enantiomers and the factors that influence the variability, including microbial processing. Surprisingly few studies have addressed this issue, although the variability in microbial community structure and activity in soils and sediments is well-established (13-19). Background and agricultural soil types can differ in levels of OCP, nutrients and soil organic matter (SOM), microbial communities, and degree of soil mixing. This may partly explain the range of chiral signatures observed in agricultural (20) and background (6) soils. In the latter study, enantiomers of trans- and cis-chlordane (TC and CC), chlordane MC5, R-HCH, and o,p′-DDT were investigated in 65 background soils (untreated soils that are believed to have only received POPs via atmospheric deposition) from 32 countries. The study showed a wide variation in chiral signatures and reversals of enantioselective degradation between soils sampled only a few hundred meters apart. The objectives of this study were to investigate the spatial variability of chiral pesticide signatures in field soils and the changes of chiral signatures in spiked soils over time under field and lab conditions.
Materials and Methods Field Sampling and Experimental Design. Soils were collected from a field location on the Kintyre Peninsula in western Scotland, a site that was included in the global background soil survey (6). This area was of interest because coniferous woodland and grassland soils collected only a few hundred meters apart contained pesticide residues with different chiral signatures (6), expressed as the enantiomer fraction, EF ) areas of the (+)/[(+)+(-)]-enantiomers. In the global background soil study, the woodland and grassland soils had EFs of 0.375 and 0.577 and 0.478 and 0.535 for TC and R-HCH, respectively. The site was visited again on June 2, 2004 to collect soils and to re-assess the chiral signatures. At this time, single core soil samples from the grassland and woodland areas were taken for analysis. On July 24, 2004, two plots were prepared at the woodland site (W1 and W2) and one at the grassland site (G1). At W1 and W2, the water table was very close to the soil surface and thus waterlogged at most times. Different studies were conducted using the soils, as is described next. Study 1: Variability in EFs of Native OCP Residues in Soil and Air. Soils were sampled from nine locations 1 m apart on 2 m × 2 m plots split into 3 × 3 rows, with point 1 being in the top left of a grid and point 9 in the bottom right; point 5 was at the center. One core (0-5 cm) was taken at each of the nine points. At point 5, a second 10 cm core was taken, and the core was cut in half to obtain the surface (0-5 cm) and bottom (5-10 cm) samples. Samples were wrapped in cleaned aluminum foil, sealed in plastic bags, and stored at -20 °C until analysis. Passive air samples were taken at the grassland and woodland sites. Three polyurethane foam (PUF) disks (14 cm diameter, 1.35 cm thick; density 21.3 mg cm-3) were deployed at a height of 150 cm, two at the woodland site and VOL. 41, NO. 14, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4965
4966
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 14, 2007
nd
Not available due to non-detection or interference problems. b 5-Surface and 5-bottom refer to a separate core, 0-5 cm, and 5-10 cm sections. c Composite EFs are from pooled samples collected on July 24, 2004 and a second sample collected on October 12, 2005. 5-bottom refers to a separate core, subsurface section, 5-10 cm
4.7 5.0 4.9 4.7 4.9 4.7 4.6 5.2 5.0 5.1 5.1 55.8 62.0 62.0 43.7 30.5 52.9 39.4 29.9 31.4 30.9 39.4 22.3 21.1 32.4 68.2 33.8 67.7 29.6 27.6 33.5 23.4 29.6 12.9 12.8 11.3 13.7 19.7 19.8 19.5 6.7 12.7 10.5 9.5 nd nd nd nd nd nd nd nd nd nd nd
nd 0.668 0.454 nd 0.474 nd 0.494 nd nd nd nd 0.454-0.668 0.523 0.098 nd nd 0.440 nd 0.447 0.406 0.506 0.431 0.425 0.485 nd nd 0.415-0.506 0.449 0.035 0.467 0.379 0.398 0.393 nd 0.419 0.487 0.358 0.416 nd 0.403 0.452 0.358-0.487 0.412 0.039 0.439 nda 0.479 0.481 0.491 0.507 0.426 0.511 0.483 0.470 0.463 0.498 0.426-0.504 0.481 0.025 0.503
nd 0.452 0.440 0.462 0.473 0.443 0.474 0.461 0.442 0.435 0.405 0.405-0.474 0.449 0.021 0.453
0.443 0.406 0.496
0.390
nd nd nd nd 0.528 0.455 nd 0.511 nd nd nd 0.455-0.528 0.498 0.038 nd July 24, 2004, gridded samples 0.549 nd 0.522 0.272 0.558 0.296 nd 0.434 0.548 0.461 nd 0.494 0.549 0.494 0.541 0.434 nd 0.483 0.540 0.474 0.537 0.430 0.522-0.558 0.272-0.494 0.543 0.427 0.011 0.079 0.542 0.456
0.478 0.474 June 2, 2004, single core sample 0.546 0.370 na
1 2 3 4 5 5-surfaceb 5-bottomb 6 7 8 9 range mean s.d. compositec
a
5.4 4.9 5.4 5.4 5.5 6.2 5.7 5.7 5.5 5.7 6.1 6.2 5.7 6.2 5.3 6.1 6.0 6.2 6.2 4.2 6.4 5.5
W2 pH
W1 G1 W2 W1
% SOM
G1 W2 W1 MC5 G1 W2 CC
W1 G1 W2
TC
W1 G1
TABLE 1. EF Data for Chlordanes in Grassland (G1) and Woodland (W1 and W2) Soils
one at the grassland site (21, 22). Samplers were left exposed from February 2005 to October 2005, to provide integrated atmospheric signatures for chiral analysis. Study 2: Time Trends in EFs of OCPs Spiked in the Field. At W1, W2, and G1, a second series of nine soil cores (0-5 cm) was collected on each of the same grids, and the cores from each grid were combined to obtain a composite sample and mixed thoroughly. Homogenized composites were prepared at each location and divided into four subsamples, of which three were spiked separately for this field experiment and the fourth portion was transferred to the laboratory, to be used for the experiment described next. Enough composite was prepared so that three aluminum containers (tins) with an 11 cm diameter and 13 cm depth could be filled with soil, after the soil had been spiked. The spiking solution was prepared in acetone containing 0.1 ng/µL TC, 0.27 ng/µL CC, 0.68 ng/µL R-HCH, and 3.78 ng/µL o,p′-DDT. Volumes of 185, 462, and 555 µL were used for G1, W1, and W2, respectively, to bring the spike concentrations (on a dry weight basis) about 3-30 times higher than the background concentrations of each sampling point. The spiking solution was mixed with 300 mL of distilled water in the field. Both ends of the aluminum cans were removed, and a 2 mm mesh aluminum sieve was placed at the bottom of the can, to keep the spiked soil isolated from the surrounding soil environment, while allowing infiltration of water. Pesticide spiked water (300 mL) was stirred into the soil, and the soil was mixed thoroughly for several minutes. Following spiking, a sub-sample (t ) 0) was taken for analysis, and then the rest of the spiked soil was transferred into the tins and placed back in situ. The top of each tin was covered with moss, and leveled with the surrounding soil, to recreate undisturbed conditions. Three tins were set up at each experimental area. At W1, tins were in grid locations 2, 7, and 9; at W2, tins were in grid locations 2, 6, and 8; and at G1, tins were in grid locations 1, 2, and 9. The sites were visited again, and the soil was sampled from the tins after 209 and 445 days. Study 3: Laboratory Study of Enantioselective Degradation. Approximately 50 g of soil from the woodland sites was spiked with 10.5 ng of TC, CC, and R-HCH. A portion of the soil (10 g) was first spiked with 30 µL of acetone and stirred for 5 min. The spiked soil was then allowed to air-dry for 5 min, before the remaining soil (40 g) was added and mixed thoroughly for another 10 min (23). The spiked soil was divided into 10 jars; three jars were replicates at each sampling time of 0 and 12 months. At each sampling, the fourth jar was a sterile control, spiked but subjected to γ-irradiation (32.7 kGy, Isotron PLC), and a fifth jar contained irradiated unspiked soil. A glass fiber filter (45 mm diameter, EPM2000, Whatman International Ltd.) was placed on the top of each jar and kept moist with a daily application of distilled water. The lids of the jars were left loose, to ensure air exchange during the experiment. After sampling, the soil was placed in the freezer until analysis. Laboratory Methods and Chemical Analysis. Sample extraction and clean up and microbiological parameters were performed at Lancaster University. Chiral analysis was performed at Environment Canada. Methods for OCPs are described in Kurt-Karakus et al. (6). The soil organic matter content (% SOM) was determined by loss on ignition at 450 °C, and the soil pH was measured in water (10 g of soil in 25 mL). Microbial Parameters. Microbial respiration was measured by placing 10 ( 0.2 g of soil into a respirometer, as described by Reid et al. (24). The optimum glucose concentration was determined with a 4 mM solution for W1 and a 2 mM solution for W2. Soils were amended with the optimum 12C-glucose and 100 Bq g-1 of [1-14C] glucose, and the evolution of 14CO2 was measured. Additionally, analytical
FIGURE 1. Time trends for EFs in the spiked field soils in W1. Asterisks indicate EFs significantly different from those of racemic standards. nd ) not detected. blanks were run with non 14C-spiked respirometers (n ) 3). The respirometers were sealed and placed into an incubator shaker at 21 ( 2 °C, 100 rpm for 5 days. Each CO2 trap (1 mL of 1 M NaOH) was replaced every 24 h, and 5 mL of liquid scintillation fluid (Goldstar) was added to the sampled vial. Following overnight storage, the level of radioactivity was determined by liquid scintillation counting (Canberra Packard Tri-Carb 2250 Ca). Following the 5 day incubation, the uptake of the 14C into the soil microbial biomass was determined by a chloroform-biomass extraction, using a standard procedure of chloroform fumigation as described by Boucard (25) and Vance et al. (26). Estimation of soil bacterial numbers (colony forming units (CFUs)g-1 dry wt soil was also performed. At each time point, 1.0 ( 0.2 g of soil was weighed into a Teflon centrifuge tube. Sterile Ringers solution (10 mL of 1/4 strength) was added, and the tube was sealed and shaken at 21 ( 2 °C, 100 rpm, for 1 h; this was carried out in triplicate for each condition. The tube was then removed and left to stand for 1 h prior to sampling, after which a 1 mL portion of the supernatant was serially diluted and plated out on a plate count agar (Oxoid). The plates were incubated at 21 ( 2 °C for 5 days prior to determination of CFUs. Quality Control. Racemic standards were injected repeatedly for every five samples, to determine the reproducibility in measuring EFs. Average EF values of the standards were 0.499 ( 0.004 for TC; 0.500 ( 0.005 for CC; 0.501 ( 0.004 for MC5; 0.500 ( 0.006 for R-HCH; and 0.498 ( 0.004 for o,p′-DDT. Judgments concerning racemic or non-racemic residues in soils were based on whether the EFs of residues fell within or outside the mean EF ( 2 standard deviation (SD) for standards. Extracts of unspiked soil samples from woodland and grassland plots were injected in duplicate or triplicate to determine chromatographic reproducibility for TC, CC, R-HCH, and o,p′-DDT at low concentration levels. Replicate injections of the same unspiked soil extract gave 1.3-2.4% RSD (n ) 22-33, including samples from KurtKarakus et al. (6)). Duplicate analysis of six spiked soils including extraction and clean up steps and three injections of each extract gave 1.1-2.6% RSD for the compounds. Isotope ratios of the target/qualifier ions (IRs) for each enantiomer peak were required to fall within the 95%
confidence interval for standards for a satisfactory analysis or the result was rejected (6). Blanks were run using unexposed PUF disks and Soxhlet thimbles filled with sodium sulfate. None of the chiral OCPs was detected in the blanks. Analytical blanks were run with 12C spiked respirometers during measurement of microbial parameters. These gave baseline values that were subtracted from respirometry results.
Results Variability in the Enantiomer Composition of the Field Soil and Air. Results from the June 2, 2004 survey (first trip to sampling site after global survey (6)) of soils from W1, W2, and G1 are given in Table 1. Soils from the three different locations had different EFs, confirming the findings of the previous study (6). Particularly noteworthy is the reversal of the enantiomer degradation preference for CC between W1 and W2, even though these sites are only a few tens of meters apart. In addition, there is a wide range in EFs for TC among the three sites (0.358-0.506). EFs in air samples from passive air samplers were TC ) 0.476 ( 0.002, CC ) 0.511 ( 0.001, and R-HCH ) 0.480 ( 0.001, with no difference between grassland and woodland sites. The EFs are in the midrange of those measured in the soils and are typical of those reported for regional background air (e.g., 0.454-0.463 and 0.5110.512 for TC and CC (27) and 0.475-0.500 for R-HCH (12)) in southern Sweden. This was expected since the EFs in the passive air samples are affected by air masses originating from a wide area and integrated over 8 months (21, 22). KurtKarakus et al. (6) pointed out that the soils often showed different EFs from those in ambient air (for chlordanes and R-HCH), indicating that post-depositional degradation had taken place. The results of the current investigation confirm this. The variability in EFs of chlordanes is given in Table 1 for the nine grid points taken at G1, W1, and W2 on July 24, 2004. Although the EFs of R-HCH and o,p′-DDT were measurable in the single cores from the June survey, the July study showed that these two compounds were detectable in only a few of the grids. In Table 1, it can be seen that there is considerable variability and reversal of EFs over the spatial range of meters (among the grids) for chlordanes. This was most marked for CC in W2 (0.272-0.494), TC in W1 (0.358VOL. 41, NO. 14, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4967
0.487), and MC5 in W1 (0.454-0.668) counting the surface and deep samples. This indicates the high variability in field samples over small areas. The composite samples, obtained by pooling cores from the nine grid points in each plot, gave EFs that were close to the mean values of individual grid points in some cases but showed greater differences in others (i.e., EF of TC in G1). Depth samples of 0-5 cm (surface) and 5-10 cm (bottom) at the same grid point (grid 5 at each experimental area) showed different patterns for some compounds. For example, in G1, the EF of TC was 0.426 in the surface layer and 0.511 in the bottom layer. The surface layer had a CC EF of 0.506, with 0.431 in the bottom layer. Relationships to SOM and pH. EF data for the study locations were compared statistically with data on % SOM and pH using independent samples t tests in the 95 and 99% confidence intervals (CI), using Statistical Package for the Social Sciences (SPSS) software (version 13.0). Results showed that EFs of TC in G1 were different from those in W1 and W2 at 99% but that EFs in W1 were not different from those in W2 at 95%. EFs of CC were different at 99% in G1 versus W1, and in W1 versus W2, but EFs in G1 were not different from those in W2 at 95%. Significant correlations (p ) 0.021, r2 ) 0.55) of EFs of TC with % SOM in W1 and EFs of CC (p ) 0.013, r2 ) 0.56) with % SOM in W2 were observed. In the global soils survey, the correlations between deviation of EFs from racemic and % SOM were significant but weak (r2 ) 0.16, p ) 0.0031 for CC and r2 ) 0.16, p ) 0.0022 for TC (6)). CC showed a statistically significant but weak correlation with pH in soils from W2 (p ) 0.021, r2 ) 0.50) but no significant correlation in soils from W1 and G1. This provides some support for the assertion that soil properties can influence chiral signatures in soils, by influencing the activity of the soil microbial community (28). The EF data for TC, CC, R-HCH, and o,p′-DDT over time for the composite soils from W1, W2, and the G1 are shown in Figure 1and Figures S1a and S1b (Supporting Information), respectively. The samples were taken at t ) 0 (i.e., when the composition of the spike was racemic, EF ) 0.500), t ) 209 days, and t ) 445 days. The unspiked day 0 situation represents the EFs of the chemicals that were native to the soil and have been undergoing processing since their atmospheric deposition. Differences in EFs were evident among the three soils, and among the three tins deployed at a site, reinforcing the results of the grid study that showed short range variability. A change in the EF value for spiked samples was observed for R-HCH, which showed an EF < 0.45 in the W1 soil and 0.05). Native TC residues were significantly non-racemic in W1 and W2 (but not G1) soils. No significant deviations from racemic were found for spiked CC in the three soils, despite the fact that native CC residues were significantly non-racemic in G1, W1, and W2 soils. Changes in EFs in the Laboratory Study. Measurements of the microbial biomass, CFUs, and respirometry indicated that the soils remained microbially active throughout the experiment. The initial extent of 14CO2 evolved was 59 and 49% for the W1 and W2 soils, respectively. All subsequent time points were lower but constant (45.2 ( 3.4 and 43.5 (
FIGURE 3. Ratios of chiral OCPs to native trans-nonachlor (TN) in field (A and B) and laboratory (C and D) experiments. Field values are normalized to OCP/TN ratios in spiked soils at day 0; laboratory values are normalized to ratios in spiked sterilized soils at day 0. Bars in field experiments indicate standard deviations for three tins in G1 and two tins in W1. 2.7% for W1 and W2, respectively). The 14C-biomass was 5.5 ( 0.26% at t ) 0 and 7.9 ( 1.6% 14C on day 387 in soil W1 with CFUs of 1.5 × 104 and 1.2 × 104, respectively. For the W2 soils, the 14C-biomass was 7.0 ( 0.04 and 6.6 ( 0.84% on days 0 and 387, respectively, with CFUs of 9 × 104 and 1.1 × 104, respectively. Microbiological analyses were also carried out on the sterilized spiked soils, and it was shown that these soils were microbially inactive throughout the experiment. Results for the laboratory experiment with W1 and W2 soils are presented in Figures 2 and S2, respectively. EFs for the sterile spiked treatments did not vary from racemic over time, suggesting that they remained microbially inactive throughout the experiment. As in the field experiments, the most apparent enantioselective degradation was for R-HCH, which showed an EF drop in the unsterile spiked soils from racemic at day 0 to 0.456 in W1 and 0.462 in W2 at day 387. Changes in EFs were also observed for TC and CC in the unsterile spiked W1 soil (EFs 0.477 and 0.516 at day 387). At day 387, the EF drop for TC was less in unsterile spiked W2 soil (0.484), while CC remained racemic (EF 0.496). Lower EFs of the native TC and CC residues in the sterile unspiked W1 and W2 soils were found at day 387 as compared to day 0. This is unexplainable since no enantioselective degradation was seen in the sterile spiked soils. Perhaps the sterilization for these two samples was not completely effective. Another explanation could be the spatial variability of EFs in the pooled soil, although we tried to homogenize the soil prior to setup of the experiment. Relative Dissipation Rates from Field and Laboratory Spiked Soils. Further insight to dissipation processes was provided by estimating the relative loss rates of TC, CC, and R-HCH with respect to trans-nonachlor (TN) (6). TN is native to the soil, most likely due to atmospheric deposition, and is the most recalcitrant of the three chlordane compounds. Ratios of the chiral OCPs (sum of both enantiomers) to TN were calculated from peak areas, normalized to their ECNI
response factors (6). This was done for the three tins in G1 and two tins in W1. Relative rates were not obtainable for the other tins due to anomalous compound ratios, which appear to have been caused by inhomogeneity in the soil spiking. Results are exemplified in Figure 3 for the G1 (Figure 3A) field experiment, where OCP/TN ratios, normalized to day 0, are shown for unspiked soil and spiked soil. Results show that R-HCH was dissipated quickly, with about 40% of the original spike relative to TN remaining at day 209 and 4-9% remaining at day 445. Chlordanes were dissipated from the G1 soil more slowly; 51% of the TC and 73% of the CC spikes remained at day 445. Most of the loss occurred by day 209, with little further change at day 445. Since dissipation occurred under field conditions, it is not possible to say that degradation was entirely responsible. Indeed, there was no evidence for enantioselective degradation of R-HCH in the G1 soil (Figure S1a). Other processes, such as volatilization, non-enantioselective degradation, and leaching, may have contributed to the total loss. Field results for W1 (Figure 3B) also showed dissipation of residues over time. Of the R-HCH spiked, 20-32% remained after 209 days and 4-7% after 445 days. For the two chlordanes, 47-76% of the spiked amounts was present after 209 days and 46-73% after 445 days. In the case of R-HCH, the loss was accompanied by significant enantioselective degradation (Figures 1 and S1b). However, enantioselective degradation for the chlordanes was significant only for TC in W1. From these results, we conclude that the loss of spiked R-HCH in the Scotland soils occurred by enantioselective and non-enantioselective processes, while dissipation of the spiked chlordanes was largely nonenantioselective. Similar to the field spiking experiments, the ratios of chiral OCPs/TN were determined in the laboratory studies for soils W1 and W2 (Figure 3C,D). As in the field experiments, extensive loss of R-HCH from unsterile spiked soils occurred over time, with only 4-6% remaining after 387 days. Leaching was ruled out, and volatilization was expected to be minor VOL. 41, NO. 14, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4969
TABLE 2. Summary of Preferential Enantiomer Degradation of Spiked Chiral Pesticides Under Field (Day 445) and Laboratory (Day 387) Conditions field experiment G1 TC CC R-HCH o,p′-DDT
racemic racemic racemic (-), racemicb
W1 racemica
(+), racemic (+) (-)c
laboratory experiment W2
W1
racemic racemic (+) nad
(+) (-) (+) nse
W2 (+) racemic (+) ns
a Two tins showed degradation of the (+)-enantiomer; the EF in the third tin was racemic. b Two tins showed degradation of the (-)-enantiomer; the EF in the third tin was racemic. c Two tins showed degradation of the (-)-enantiomer; o,p′-DDT was not detectable in the third tin. d Not available due to chromatographic interferences. e Not spiked in the laboratory experiments.
during the laboratory experiments, so the observed losses probably occurred by microbial degradation, which suggests a similar behavior in the field study. Little change occurred in the TC/TN or CC/TN ratios over time, for either the sterile or the unsterile soils. The fact that changes in EFs were measureable for the chlordanes in the unsterile soils points out the value of chiral analysis for such studies. The high precision of chiral analysis (
