|
|
||||||||
a Dep. of Soil, Crop and Atmospheric Sciences, Cornell Univ., Ithaca, NY 14853 USA
b Dep. of Environmental Sciences, Univ. of California, Riverside, CA 92521 USA
leiguo{at}mail.ucr.edu
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: AC, activated C BTC, breakthrough curve HPLC, high pressure liquid chromatography
| INTRODUCTION |
|---|
|
|
|---|
In general, sorption is often considered a process that limits pesticide degradation (Dao and Lavy, 1987; Guerin and Boyd, 1992; Ainsworth et al., 1993; Gamerdinger et al., 1997). This can be understood on the basis of the reduced bioavailability of sorbed compounds to soil microorganisms. However, in the field, sorption might increase degradation as a whole by increasing the residence time of pesticides in the root zone where most microbial activity occurs.
The relationship between adsorption and degradation of chemicals obviously is a fundamental one underlying the environmental behavior of pesticides and other chemical contaminants. However, although there are numerous studies focusing on the effects of adsorption on chemical bioavailability (e.g., Smith et al., 1992; Siahpush et al., 1992; Novak et al., 1995; Fu et al., 1996), very few studies have reported the respective degradation rates in soil solution and sorbed phases. Some studies intended to model chemical degradation under different sorption conditions automatically assumed that the degradation rate for sorbed compounds is zero (Mihelcic and Luthy, 1991; Fu et al., 1996; Shelton and Doherty, 1997). Scow et al. (1986) presented a two-compartment model, which explicitly included individual degradation rate constants for the sorbed and dissolved compounds. Unfortunately, the value of these rate constants could not be obtained from the experimental data due to the uncertainty associated with the inverse parameter estimation method adopted.
The purpose of our study was to evaluate quantitatively the effect of adsorption on degradation of the pesticide aldicarb. We present an equilibrium model that relates the degradation kinetics directly to adsorption coefficients. The model was applied to the degradation data of aldicarb obtained from both static incubation reactors and continuous-flow soil columns to estimate the respective degradation rates of the pesticide in the liquid and sorbed phases. In particular, the degradation rate constants calculated from these two sets of experiments were compared to evaluate the influence of water flow on degradation kinetics.
| Materials and methods |
|---|
|
|
|---|
Analytical standard aldicarb (purity 99.8%) was obtained from Rhône-Poulenc Ag Company (Research Triangle Park, NC). The pesticide has a water solubility of 6000 µg mL-1 and is reported to have a low adsorption coefficient (Kd <1 mL g-1; Montgomery, 1993). Aldicarb was selected because of its reported equilibrium transport behavior in a similar soil (Zhong et al., 1986) and its relatively short persistence.
Batch Sorption Experiments
Measurement of aldicarb sorption isotherms and kinetic sorption were carried out in triplicate at 24 ± 1°C. The experiments were performed using 20-mL scintillation vials containing soilwater suspensions. The suspensions were composed of
5 g soil and 10 mL 0.005 M CaSO4 containing 200 µg mL-1 of HgCl2 as biocide. The sorption isotherms were measured using six initial aldicarb concentrations: 5, 10, 20, 40, 60, and 100 µg mL-1 and a shaking period of 72 to 240 h, depending on results of the kinetic experiments.
The kinetic experiments were conducted only with 20 µg mL-1 aldicarb solution. Conditions for the kinetic experiments were identical to the isotherm experiments, except for the shaking time, which lasted 24, 48, 72, 100, or 240 h. Shaking was continuous for up to 48 h and then 12 h d-1 for longer shaking periods.
At the end of shaking, the soilwater suspension was allowed to settle for 10 to 15 min. A 2-mL aliquot of the supernatant was then filtered through a 0.45-µm Nylon-66 membrane filter with polypropylene housing (Fisher Scientific, Pittsburgh, PA). The aldicarb concentration of the filtrate was determined by high pressure liquid chromatography (HPLC). The sorbed concentration was calculated as the difference between the initial concentration and the concentration at equilibrium. Losses to vials and during filtration were determined to be negligible from the blank samples that contained only the pesticide solution.
Incubation Experiments
Two sets of batch incubation experiments were conducted to measure the total soil respiration and aldicarb degradation rate, respectively. The soil respiration experiments were carried out in duplicate in 250-mL sidearm biometers (Bellco Glass, Vineland, NJ) following a modified procedure as described by Bartha and Pramer (1965). These experiments measured total CO2 evolved from the soil, which can be used to indicate overall microbial activity during the incubation period. The purpose of these experiments was to test whether the AC amendment supports microbial growth and thus changes the microbial activity.
Thirty grams of each soil amended with AC was added to the biometer. A 20-mL scintillation vial containing 5 mL of 0.5 M KOH was placed in the sidearm, and the biometer was immediately capped with rubber stops. The CO2 evolved from the incubated soil sample was trapped by the KOH solution and the amount of CO2 was determined by titration once every week. At each sampling, the scintillation vial was removed, and an aliquot (1 mL) of the trapping solution was titrated by 0.1 M HCl to quantify the absorbance of CO2. The endpoint of titration was indicated by 1% (by weight) phenolphthalein.
The incubation experiments for aldicarb degradation were performed in triplicate in 250-mL flasks containing 200 g (dry basis) soil at saturated water content (38.8%). The soil was preconditioned for 4 wk prior to pesticide addition by adding an appropriate amount of distilled water, determined by weighing, to bring the water content to approximately the field capacity (28.8%). Two and one-half milliliters of the aldicarb stock solution (4042 µg mL-1) was then added dropwise evenly across the surface of the soil, yielding an application rate of 50 µg g-1. The soil was mixed carefully with a stainless steel spoon. The water content was then brought to saturation. The flask was left open for the first few days to allow excessive water to evaporate and was then covered with aluminum foil for incubation at the room temperature of 24 ± 2°C. Distilled water was added as needed during the incubation, and soil samples were removed periodically for analysis.
At each sampling event, 10 to 20 g of soil sample was transferred into a 60-mL high density polyethylene bottle, and was extracted following a modified procedure provided by the manufacturer (Rhône-Poulenc Ag Company). The soil sample was first shaken with 30 mL of methanol on a reciprocating shaker for 5 h. An aliquot (10 mL) of the extract was then filtered through a 0.45-µm nylon filter. The filtrate was transferred to a 15-mL graduated glass centrifuge tube, and was further condensed as needed under N2. The aldicarb concentration of the condensed filtrate was determined by HPLC. Due to variation in the weight of soil samples, the water/methanol ratio of the extractant was calculated to be
0.21 to 0.30. Accordingly, the mass recovery of the extraction procedure was measured separately at two water/methanol ratios, 0.25 and 0.40, and was found to be indistinguishable for all five AC soil treatments (1.01 ± 0.028).
Soil Column Leaching Experiments
The soil column leaching experiments were designed to measure pesticide degradation during water flow. For this purpose, a long soil column (26 cm), combined with a low flow velocity (14 cm d-1), was used to increase the residence time of aldicarb in the column. The glass column had an internal diameter of 5 cm, and was connected to a digitized isocratic LC pump (Perkin-Elmer Model 250, Perkin-Elmer Corp., Norwalk, CT) after being packed with each soil at a density of 1.30 g cm-3. At this density, the measured porosity was
43%.
Each column was saturated with 0.005 M CaSO4 from below at a very slow flow velocity. The column was then drained for 2 wk before saturation was resumed. After the flow rate was adjusted to a steady state, a pulse of one pore volume of pesticide solution containing 28 µg mL-1 aldicarb was applied. The pesticide was subsequently displaced by pesticide-free CaSO4 solution. The effluent discharged from the column was collected in distinct fractions and analyzed for aldicarb by HPLC. To facilitate aldicarb degradation through oxidation, the influent reservoir (12-L capacity) was connected to a constant air-flow source in order to maintain a high O2 level in the influent solution. The estimated O2 concentration was
315 µmole L-1 and was calculated to be sufficient to oxidize twice the total aldicarb applied in the column to its primary oxidized product. At the termination of leaching, soil in the column was extruded by applying an air stream from one end. Soil samples were homogenized and an aliquot of
50 g was analyzed for the uneluted pesticide using a similar procedure as described for the samples of incubation experiments.
High Pressure Liquid Chromatography Procedures
The HPLC system was equipped with a Perkin Elmer LC Pump Series 410, an LC-95 UV/Visible Spectrophotometer Detector
(Perkin Elmer, Norwalk, CT), a Hewlett Packard ODS Hypersil column (5 mm packing and 125 x 4 mm; Hewlett Packard Analytical, Palo Alto, CA) and a Rheodyne Model 7125 Syringe Injector with a 100-µl sample loop (Rheodyne, Cotati, CA). The eluting solvent was methanol-water (35:75) containing 4% (v/v) acetic acid. The flow rate of the eluting solvent was set at 1 mL min-1. The retention time for aldicarb under these conditions was
8 min, and the detection limit was 0.2 µg mL-1. The analytical precision of the procedure was ± 2.37% based on standard samples.
| Results and discussion |
|---|
|
|
|---|
|
|
, where s is the sorbed concentration and c is the dissolved concentration, with an estimated Langmuir K of 0.27 and b of 243 µg g-1 (Fig. 2). These results suggest that sorption linearity is not only dependent on solute concentration, but also is affected by the properties of the sorbents. The Kd for AC4 given in Table 1 was that measured based on the linear range of the isotherm.
|
The purpose of using organic amendment was to modify the sorption properties of the soil. We chose AC as the C source because sorption of organic chemicals is known to be very sensitive to AC. The aldicarb Kd increased from 0.36 to 47.4 mL g-1 upon addition of
5000 µg g-1 AC (Table 1). These adsorption coefficients represent 48 to 99% of total aldicarb in the sorbed form.
Soil Respiration
Total soil respiration, as measured by the amount of CO2 evolved from the soil, is plotted in Fig. 3
. The data was calculated assuming that the absorption efficiency of CO2 by KOH was 100%. The CO2 evolution rate, varying between 0.97 to 4.81 µg C g-1d-1, was erratic, but overall rather stable for all soil treatments during the incubation period of
100 d. Based on an F test, there were no significant differences at P = 0.05 among various AC-amended soil treatments, which demonstrates that the activated C added could not support the microbial growth in the soil. This finding is important to the later analysis and conclusions of our study, because the relationship observed between adsorption and degradation would be attributed to totally different causes if microbial activity were changed in the presence of AC.
|
80% during the incubation period of 103 d. However, none of the amended soil yielded metabolites detectable by HPLC with the analytical procedure we used. The negative correlation of aldicarb degradation with AC rate demonstrates the inhibitory effect of adsorption on degradation. Similar results regarding the inhibitory effect of AC on pesticide degradation were also reported for maleic hydrazide by Helweg (1975), for diallate and triallate by Anderson (1981), and for imazaquin and imazethapyr by Cantwell et al. (1989).
|
|
![]() | (1) |
is the soil bulk density, and
is the volumetric water content. Recognizing that for equilibrium adsorption
![]() | (2) |
![]() | (3) |
![]() | (4) |
Equation [3] suggests that if degradation in each phase follows first-order kinetics, the overall decay will also be a first-order reaction. Further characterization of µl and µs can be achieved by nonlinear regression of the observed µKd relationship curve to Eq. [4]. The curve fitting results are also presented in Fig. 5. The estimated µl was 0.1228 d-1, which was 65 times that calculated for µs (0.0019 d-1). These results are consistent with the perception that adsorption of a chemical by soil reduces its availability to microbial uptake and thus reduces its degradation by microorganisms (Ainsworth et al., 1993; Scow, 1993).
A fundamental assumption underlying our analysis and conclusions is equilibrium adsorption. Apparently this assumption is not valid for the soilsolute systems examined in this study, especially in the presence of AC. Our kinetic sorption experiments (Fig. 1) indicated that adsorption of aldicarb required at least 24 h to reach a quasi-equilibrium state, and this time increases with AC rates. So the estimated µl and µs would have an error term which is directly related to the fraction of nonequilibrium adsorption in the total adsorption. The estimation of µl and µs under the assumption of nonequilibrium adsorption is addressed in a forthcoming paper (Guo et al., 1999). Nevertheless, the error is considered to be insignificant considering the time scale during which degradation was assessed and the relatively small fraction of kinetic adsorption at longer times.
Degradation in Continuous Flowing Soil Columns
The degradation rate constant of aldicarb in the continuous-flow soil columns was calculated using the following equation that we developed previously (Guo and Wagenet, 1999):
![]() | (5) |
is the time equal to one-half pulse length (1/2T0), and T is the experimental duration. With all the parameters determined experimentally, the degradation rate constant µ can be calculated by solving Eq. [5] numerically based on the measured breakthrough curve (BTC), which represents f(t). The measured BTCs of aldicarb, along with the experimental parameters and the estimated µ, are presented in Fig. 6 for all soil columns. No BTC was available from the column of AC4, because no detectable aldicarb was found in the effluent samples despite that the length of the column was reduced to 5 cm. A simple calculation using a Kd of 47 (that measured for AC4 in the batch sorption experiments) indicates that about 35 pore volumes of water are needed for aldicarb to break through the column, assuming no degradation.
|
The measured BTCs of aldicarb did not show appreciably elevated tailing (Fig. 6), which suggests that equilibrium sorption prevailed during its transport through the columns. Results of the kinetic sorption experiments indicate that it took
72 h for aldicarb to reach sorption equilibrium in most of the soil treatments. Therefore, sorption achieved during this length of time is still appropriate to be viewed as "instantaneous" with respect to transport at the flow rate used in this study.
The µKd curve obtained for the columns is shown in Fig. 7
. The relationship that was observed for incubation experiments still held under transport conditions: a higher Kd was associated with a slower degradation. This demonstrates again that the sorbed aldicarb was degraded more slowly than the dissolved aldicarb. The curve fitting analysis using Eq. [4] reveals that the accelerated degradation of aldicarb under transport conditions was attributable to both enhanced µl (
in the columns vs. 0.1228 d-1 in the incubation flasks) and µs (0.0055 vs. 0.0019 d-1). However, the increase in µs was larger than in µl. The difference between them reduced from 65-fold in static incubation experiments to 38-fold under transport conditions.
|
| Conclusions |
|---|
|
|
|---|
Due to the difference in the surface properties of the AC that we used in this study and soil organic matter or minerals, it is possible that the degradation rate constants we estimated in this study may deviate somewhat from those in natural soil. The degradation rates for a particular soil would depend on both the nature of the surfaces (most notably hydrophobic vs. hydrophilic surfaces) and the properties of the chemical under consideration.
| ACKNOWLEDGMENTS |
|---|
Received for publication August 5, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Beulke, W. van Beinum, C. D. Brown, M. Mitchell, and A. Walker Evaluation of Simplifying Assumptions on Pesticide Degradation in Soil J. Environ. Qual., October 12, 2005; 34(6): 1933 - 1943. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. L. Lindahl, J. Kreuger, J. Stenstrom, A. I. Gardenas, G. Alavi, S. Roulier, and N. J. Jarvis Stochastic Modeling of Diffuse Pesticide Losses from a Small Agricultural Catchment J. Environ. Qual., June 7, 2005; 34(4): 1174 - 1185. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.-H. Park, Y. Feng, P. Ji, T. C. Voice, and S. A. Boyd Assessment of Bioavailability of Soil-Sorbed Atrazine Appl. Envir. Microbiol., June 1, 2003; 69(6): 3288 - 3298. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||