Coupling Supercritical CO2 and Subcritical (Hot) Water for the

Apr 1, 1998 - Metabolites are combined with Dacthal by placing the disk into the GC autosampler vial containing the SFE extract. The metabolites then ...
0 downloads 10 Views 126KB Size
Anal. Chem. 1998, 70, 1956-1962

Coupling Supercritical CO2 and Subcritical (Hot) Water for the Determination of Dacthal and Its Acid Metabolites in Soil Jennifer A. Field,* Keith Monohan, and Ralph Reed

Department of Agricultural Chemistry, Oregon State University, Corvallis, Oregon 97331

Dacthal and its mono- and diacid metabolites were sequentially extracted from soils by first performing a supercritical carbon dioxide extraction to recover Dacthal, followed by a subcritical (hot) water extraction step to recover metabolites. Dacthal was recovered from soil in 15 min by supercritical carbon dioxide at 150 °C and 400 bar. The mono- and diacid metabolites were extracted from soil in 10 min under the subcritical water conditions of 50 °C and 200 bar. The metabolites were trapped in situ on a strong anion-exchange disk placed over the exit frit of the extraction cell. Metabolites are combined with Dacthal by placing the disk into the GC autosampler vial containing the SFE extract. The metabolites then are simultaneously eluted from the disk and derivatized to their ethyl esters by adding 100 µL of ethyl iodide and heating the vial at 100 °C for 1 h. Using this approach, only a single sample is analyzed, and because the diskcatalyzed alkylation reaction does not transesterify Dacthal, the speciation of Dacthal is maintained. In addition, no sample cleanup steps are required, the use of diazomethane for derivatization is avoided, and the method consumes a total of 5 mL of nonchlorinated organic solvent. Dacthal is a widely used preemergent herbicide that is applied to many crops for the control of annual weeds. Dacthal is typically applied to agricultural soils at 6-14 kg/ha.1 In the soil environment, Dacthal transforms to mono- and diacid metabolites that are more water soluble than the parent herbicide.2-4 In eastern Oregon, where Dacthal is applied to onions, the diacid metabolite is the principal form of Dacthal detected in groundwater obtained from domestic wells.5,6 * Corresponding author. Fax: (541) 737-0497. E-mail: [email protected]. (1) The Pesticide Manual: A World Compendium, 7th ed.; Worthing, C. R., Ed.; British Crop Protection Council: London, UK, 1983. (2) Gershon, H.; McClure, G.W. Contrib. Boyce Thompson Inst. 1966, 23, 291294. (3) Miller, J. H.; Keeley, P. E.; Thullen, R. J.; Carter, C. H. Weed Sci. 1978, 26, 20-26. (4) Ross, L. J.; Nicosia, S.; McChesney, M. M.; Hefner, K. L.; Gonzalez, D. A.; Seiber, J. N. J. Environ. Qual. 1990, 19, 715-722. (5) Monohan, K.; Tinsley, I. J.; Shepherd, S. F.; Field, J. A. J. Agric. Food Chem. 1995, 43, 2418-2423. (6) Field, J. A.; Monohan, K. Anal. Chem. 1995, 67, 3357-3362.

1956 Analytical Chemistry, Vol. 70, No. 9, May 1, 1998

To assess the fate of Dacthal that is applied to soil, both parent and metabolite forms in water and soil should be considered. While rapid methods exist for the determination of Dacthal and its metabolites in water,5,6 quantitative and rapid methods are needed to determine Dacthal and its metabolites in soils, since conventional methods require large volumes of solvent and time to process the extract. For example, the conventional method for extracting Dacthal and its metabolites from soil requires 200 mL of 0.4 M HCl/acetone to extract a 20-g sample and the use of hazardous diazopropane to derivatize the acids to their ester forms.7 Supercritical fluid extraction (SFE) is an attractive analytical technique for recovering organic compounds from soils and sediments. Carbon dioxide (CO2) is currently the fluid of choice, due to its low toxicity and environmental acceptability. The physicochemical properties of supercritical fluids, including low viscosity, variable solvent strength, and high diffusivity, contribute to faster extractions compared to conventional extraction techniques, such as Soxhlet extraction or sonication. Supercritical fluid extraction methods have been successfully developed for nonpolar compounds that exhibit high solubilities in CO2, such as PAHs,8 PCBs,9 dioxins,10 and organochlorine pesticides.11 With the addition of methanol as modifier, supercritical CO2 becomes more amenable to the extraction of moderately polar pesticides, including triazines,12 organophosphate insecticides,13 and sulfonylureas.14 Adding chemical reagents to soil samples prior to SFE has received attention as an alternative for extending supercritical CO2 toward acidic analytes such as chlorophenoxy acid herbicides; however, the recoveries of acid analytes are variable and depend on the sample matrix.15-18 (7) Wettasinghe, A.; Tinsley, I. J. Bull. Environ. Contam. Toxicol. 1993, 50, 226-231. (8) Hawthorne, S. B.; Miller, D. J. Anal. Chem. 1994, 66, 4005-4012. (9) Alexandrou, N.; Pawliszyn, J. Anal. Chem. 1989, 61, 2770-2776. (10) Onuska, F. I.; Terry, K. A. J. High Resolut. Chromatogr. 1991, 14, 829834. (11) Wong, J. M.; Li, Q. X.; Hammock, B. D.; Seiber, J. N. J. Agric. Food Chem. 1991, 39, 1802-1807. (12) Steinheimer, T. R.; Pfeiffer, R. L.; Scoggin, K. D. Anal. Chem. 1994, 66, 645-650. (13) Snyder, J. L.; Grob, R. L.; McNally, M. E.; Oostdyk, T. S. J. Chromatogr. Sci. 1993, 31, 183-191. (14) McNally, M. E.; Wheeler, J. R. J. Chromatogr. 1988, 435, 53-63. (15) Hawthorne, S. B.; Miller, D. J.; Nivens, D. E.; White, D. C. Anal. Chem. 1992, 64, 405-412. S0003-2700(97)01109-8 CCC: $15.00

© 1998 American Chemical Society Published on Web 04/01/1998

The use of subcritical water was reported for the extraction of moderately polar (phenols) and nonpolar (polycyclic aromatic hydrocarbons and polychlorinated biphenyls) organic compounds from solid environmental samples.19-21 By increasing the temperature, the dielectric constant of water is decreased, and hence the polarity of water is decreased. Subcritical water is particularly appealing because it is an environmentally acceptable solvent that does not pose any disposal problems, it is significantly less expensive than analytical grade CO2, and subcritical water conditions are easily achieved with commercial laboratory equipment. Furthermore, water is a good alternative fluid to CO2 since many polar compounds, acids in particular, exhibit limited solubility in supercritical CO2, unlike in supercritical water conditions, which require temperatures greater than 374 °C and pressures greater than 221 bar. Extracting polar pesticide metabolites with subcritical water shifts the problem of concentrating analytes from a solid matrix to extracting the analytes from water. Fortunately, a range of solid-phase techniques can be applied to the problem of concentrating analytes from solution. For example, Hageman et al.21 coupled solid-phase microextraction with subcritical water extraction for the determination of polycyclic aromatic hydrocarbons in soil and air particulate matter. Derivatization is an additional, often cumbersome step in the preparation of acid analytes for gas chromatographic analysis. Fortunately, strong anion-exchange (SAX) extraction disks not only isolate organic acids from water but also act as catalysts in alkylation reactions of acid analytes to volatile esters that are compatible with gas chromatographic analysis.22,23 This approach was reported for the determination of the metabolites of Dacthal6 and chlorophenoxy acids24 in surface water and groundwater and for nonylphenol polyethoxy carboxylates in surface water and in municipal and industrial sewage effluents.25 In this paper, we report the use of supercritical CO2 for extracting Dacthal from soils followed by subcritical water for extraction of the mono- and dicarboxylic acid metabolites of Dacthal. To the best of our knowledge, this is the first reported use of subcritical water for the extraction of very water soluble, acidic compounds from soil. By placing a SAX disk over the exit frit of the extraction cell, the acid metabolites are trapped in situ on the disk during the extraction. Once the disk is removed from the extraction cell, the metabolites are recombined with the SFE extract containing Dacthal. Subsequently, the metabolites are simultaneously eluted from the SAX disk and derivatized to their ethyl esters.6 The combination of supercritical CO2 and subcritical water is demonstrated in a survey of soils collected from eastern Oregon. (16) Lopez-Avila, V.; Dodhiwala, N. S.; Beckert, W. F. J. Agric. Food Chem. 1993, 41, 2038-2044. (17) Rochette, E. A.; Harsh, J. B.; Hill, H. H. Talanta 1993, 40, 147-155. (18) Croft, M. Y.; Murby, E. J.; Wells, R. J. Anal. Chem. 1994, 66, 4459-4465. (19) Hawthorne, S. B.; Yang, Y.; Miller, D. J. Anal. Chem. 1994, 66, 29122920. (20) Yang, Y.; Bowadt, S.; Hawthorne, S. B.; Miller, D. J. Anal. Chem. 1995, 67, 4571-4576. (21) Hageman, K. J.; Mazeas, L.; Grabinski, C. B.; Miller, D. J.; Hawthorne, S. B. Anal. Chem. 1996, 68, 3892-3898. (22) Tang, P. H.; Ho, J. S. J. High Resolut. Chromatogr. 1994, 17, 509-518. (23) Chatfield, S. N.; Croft, M. Y.; Dang, T.; Murby, E. J.; Yu, G. Y. F.; Wells, R. J. Anal. Chem. 1995, 67, 945-951. (24) Field, J. A.; Monohan, K. J. Chromatogr. A 1996, 741, 85-90. (25) Field, J. A.; Reed, R. L. Environ. Sci. Technol. 1996, 30, 3544-3550.

EXPERIMENTAL SECTION Samples. Soil samples were collected from fields in 1993 from the Malheur Experiment Station in Ontario, OR. The soils were taken as cores in 0.3-m increments to 1.2 m below land surface. The soils were air-dried, sieved, homogenized, and placed in glass bottles with aluminum-lined lids for storage at room temperature. The soils analyzed for this study were of pH 8.5-8.6 and had 0.9% organic carbon. Standards and Materials. A standard of Dacthal (dimethyl tetrachloroterephthalate; 95% purity) was obtained from Chem Service (West Chester, PA). Standards of the monoacid metabolite (monomethyl tetrachloroterephthalic acid; 99.7% purity) and the diacid metabolite (tetrachloroterephthalic acid; 100% purity) were obtained from the U.S. EPA (Research Triangle Park, NC) and Ricerca, Inc. (Painesville, OH), respectively. The dipropyl ester of the diacid metabolite was prepared from diazopropane and used as the internal standard for this study.5 Ethyl iodide was purchased from Aldrich Chemical Co. (Milwaukee, WI). Reagent grade acetonitrile was purchased from Baxter (Muskegon, MI). Supercritical Carbon Dioxide and Subcritical Water Extraction. All soil extractions were performed with Isco 260D and 100DX pumps and an Isco SFX 210 extractor (Lincoln, NE). The 260D pump was filled with SFC grade CO2 (Scott Specialty Gases, Plumsteadville, PA) and the 100DX pump was filled with Milli-Q water that had been deaerated by purging with nitrogen for 30 min while stirring. Stainless steel restrictors (Isco, Lincoln, NE) were used to obtain dynamic extraction flow rates of 1-1.5 mL/min as measured at the pump. A 17-mm diameter SAX disk was cut from commercially available 47mm-diameter SAX disks (Varian, Sugarland, TX) and used without any prior pretreatment. The 17-mm disk was placed over the bottom frit of the endcap and secured by screwing the 2.5-mL stainless steel extraction cell body onto the endcap. Soil was then added to the cell, followed by a second endcap (Figure 1). To extract Dacthal and its metabolites from soil, 2 g of dry soil is added to the extraction cell and placed into the 150 °C Isco extractor, plumbed with carbon dioxide. The cell is then pressurized to 400 bar with CO2 and dynamically extracted for 15 min, during which the extract is passed through a heated restrictor at 100 °C and collected in a 10-mL vial containing 5 mL of acetonitrile (Figure 1). After the extraction is complete, the acetonitrile is evaporated to 1 mL under nitrogen and transferred to a 2-mL autosampler vial and set aside. The extraction cell then is removed from the extractor plumbed with CO2 and placed in a second Isco extractor plumbed with deaerated Milli-Q water (Figure 1). The mono- and dicarboxylic acid metabolites then are dynamically extracted with subcritical water at 50 °C and 200 bar for 10 min. Water that passes through the extraction cell and out the restrictor is discarded. Once the extraction is complete, the cell is disassembled, and the soil is discarded. The SAX disk is then removed and dried by placing it on the base of a clean filter holder and drawing air through by vacuum suction for approximately 20 min. The metabolites subsequently are eluted from the disk and derivatized to their ethyl esters using the in-vial disk elution Analytical Chemistry, Vol. 70, No. 9, May 1, 1998

1957

Figure 1. Schematic of process coupling supercritical carbon dioxide followed by subcritical (hot) water extraction for the determination of Dacthal and its mono- and diacid metabolites in soil.

Figure 2. GC/ECD chromatogram for (a) standard containing Dacthal and its monoacid and diacid metabolite and the internal standard (IS) and (b) soil (A-2) containing Dacthal (0.81 µg/g) and monoacid (0.08 µg/g) and diacid (0.98 µg/g) metabolites.

method described by Field and Monohan.6 Briefly, the 17-mmSAX disk is eluted by placing the disk into the 2-mL GC autosampler vial containing the SFE extract, together with 100 µL of ethyl 1958

Analytical Chemistry, Vol. 70, No. 9, May 1, 1998

iodide and 30 µL of the internal standard. The vial is capped and heated to 100 °C for 1 h. Once cool, the SAX disk remains in the autosampler vial, and the vial is placed directly on the GC autosampler for analysis. Gas Chromatography and Quantitation. The GC/ECD conditions used to determine Dacthal and its esterified metabolites are given in ref 6. Quantitation is based on conventional internal standard techniques and is described in ref 6. Because the alkylation with ethyl iodide is a nontransesterfying reaction, a typical chromatogram of a standard (Figure 2a) yields three peaks that correspond to the three forms of tetrachloroteraphthalic acid: the dimethyl ester (Dacthal), the monomethyl monoethyl ester (monoacid metabolite), and the diethyl ester (diacid metabolite). Conventional Acid/Acetone Extraction. Soil samples (20 g) were stirred with 2 × 100 mL of 0.4 M HCl/acetone (20:80). The samples were centrifuged, and the supernatant was recovered, combined, and filtered. The acetone was removed by rotary evaporation. Water (100 mL) was added, and the mixture (pH < 2) was extracted with 3 × 100 mL of diethyl ether. The ether fractions were collected, reduced in volume to 10 mL, and derivatized with 1 mL of diazomethane. Hexane (25 mL) was added, and the mixture was boiled on a steam bath to remove the excess diazomethane. The mixture was transferred to a 9-mmi.d. silica gel column (4.5% deactivated 60-200 mesh grade 62), rinsed with 50 mL of hexane, and eluted with benzene (20 mL). Extracts were analyzed on a Varian model 3700 gas chromatograph fitted with a Supelco SPB-5 wide bore capillary column (30 m × 0.75 mm × 1 µm; Supelco, Bellefonte, PA) and an electron capture detector. The injection port temperature was 260 °C, the oven temperature was maintained at 220 °C, and the detector was operated at 360 °C.

RESULTS AND DISCUSSION Dacthal Extraction from Soil. The first phase of the project was aimed at developing a method to extract Dacthal from soils. Supercritical CO2 was selected as the extraction fluid since Dacthal was expected to exhibit high solubility in supercritical CO2. The first step was to establish that Dacthal could be quantitatively trapped in the collection vial. Quantitative recovery (>95%) was achieved from a 10-mL collection vial containing 5 mL of acetonitrile that was spiked with standard Dacthal and purged for 20 min with CO2 at 150 °C and 400 bar. The next step in development was to establish that Dacthal could be solubilized and swept from the extraction cell by supercritical CO2. For this set of experiments, Dacthal was spiked onto three replicates, each on a 2-3 cm bed of Filter Aid, an inert, high-density glass bead product (3M, St. Paul, MN) and a blank soil, and extracted for 20 min with CO2 at 150 °C and 400 bar. The average recovery of the spiked Dacthal was 100% from the Filter Aid and 99% from the blank soil, indicating that Dacthal is solubilized by supercritical CO2 and quantitatively swept from the extraction cell. Extractions involving Filter Aid and blank soil were not intended to model the recovery of native Dacthal from soil but, rather, were a check on the plumbing of the system. To investigate the extraction of Dacthal from soil, we used soils containing native Dacthal and not soils that had been spiked with Dacthal, since spiked analytes are typically extracted more efficiently and rapidly than native analytes.29 Six replicate samples of a single soil were extracted under two different sets of conditions. Three of the replicate samples were extracted at 50 °C and 200 bar CO2 with 5% (v/v) methanol, which is the set of conditions commonly reported for the recovery of organochlorine pesticides from soil.13,26-28 The second set of three replicates was extracted at 150 °C and 400 bar CO2, as higher temperatures were shown to give faster rates of extraction and higher recoveries for a variety of organic compounds from soil.8,29 Extraction curves were constructed by collecting an extract every 5 min and analyzing each extract separately. The rate of native Dacthal recovery from soil was faster at 150 °C and 400 bar with no additional recovery after 15 min, compared to the rate of extraction obtained using conditions of 50 °C and 200 bar CO2 with 5% (v/ v) methanol, which required 40 min to achieve maximum recovery (Figure 3). The cumulative amount of Dacthal recovered from soil after 60 min of extraction at 50 °C, 200 bar CO2, and 5% (v/v) methanol was 1.27 ( 0.08 µg/g, compared to 1.13 ( 0.08 µg/g, recovered using conditions of 150 °C and 400 bar CO2. Using the t-test and a pooled estimate of the standard deviation, the standard deviations of the two averages were not significantly different at the 95% confidence interval. Because the conditions of 150 °C and 400 bar gave a rapid recovery of Dacthal (15 min) and required only a single pump, conditions of 150 °C and 400 bar CO2 were used for all subsequent extractions. To validate the concentration of native Dacthal in soil obtained by SFE, three replicates (10 g) of the same soil were extracted (26) Dean, J. R.; Barnabas, I. J.; Owen, S. P. Analyst 1996, 121, 465-468. (27) van der Velde, E. G.; de Haan, W.; Liem, A. K. D. J. Chromatogr. 1992, 626, 135-143. (28) van der Velde, E. G.; Dietvorst, M.; Swart, C. P.; Ramlal, M. R.; Kootstra, P. R. J. Chromatogr. A 1994, 683, 167-174. (29) Langenfeld, J. J.; Hawthorne, S. B.; Miller, D. J.; Pawliszyn, J. Anal. Chem. 1995, 67, 1727-1736.

Figure 3. Effect of temperature and modifier on the recovery of native Dacthal from soil using supercritical carbon dioxide.

with 3 × 25 mL of acetone alone. The concentrations of Dacthal were comparable with 1.04 ( 0.06 µg/g obtained by acetone extraction and 1.21 ( 0.15 µg/g obtained by SFE. No metabolites were detected in the acetone extracts, which indicates that HCl is required for the recovery of metabolites by liquid-liquid extraction. Metabolite Extraction by Subcritical (Hot) Water. Because the solubility of the mono- and diacid metabolites of Dacthal in supercritical CO2 was expected be very low, methanol- and acetonitrile-modified CO2 at 150 °C and 400 bar were evaluated for the extraction of metabolites from Filter Aid. Both metabolites were poorly recovered (90%) of the mono- and diacid metabolites spiked onto Filter Aid was achieved in 10 min using 25 and 50 °C water, regardless of whether the SAX disk was inside or outside the extraction cell (Table 1). When the metabolite was trapped on the SAX disk inside the extractor, recoveries of the monoacid metabolite declined from 96% at 25 °C to 8.2% at 150 °C (Table 1). Breakthrough of the monoacid metabolite increased from none at 25 and 50 °C to 21% at 150 °C. Breakthrough of the monoacid on the SAX disk inside the extractor is due to the decreased dielectric constant of water at 150 °C, which is similar to that of methanol.8 Rinsing the SAX disk with methanol prior to elution decreased the recovery of the monoacid metabolite,6 which suggests that hydrophobic interactions with the styrene/ divinylbenzene matrix as well as ion exchange are responsible for retention of the monoacid on the SAX disk. Note that, at 150 °C, the summed recovery of the monoacid (29%), which includes that trapped on the SAX disk and breakthrough, does not equal the 91% recovered at 150 °C when the water extract was collected outside the extractor. Trapping the monoacid metabolite inside the 150 °C extraction cell on the SAX disk may promote monoacid degradation by increasing the retention time of the metabolite under the high-temperature conditions. However, the monoacid does not appear to undergo hydrolysis to the diacid, as there was no corresponding increase in the amount of diacid observed in the extracts obtained at 150 °C. In contrast, the recovery of the diacid metabolite from Filter Aid, when trapped inside the extraction cell on a SAX disk, was unaffected by the temperature of the water with >90% recovery at all temperatures (Table 1). In addition, no breakthrough of the diacid metabolite was observed at any temperature, which suggests a greater selectivity of the SAX disk for the diacid metabolite than for the monoacid metabolite. Although quantitative recovery of the metabolites spiked onto Filter Aid was obtained at 50 °C and 200 bar, the use of these conditions with soil does not necessarily ensure quantitative recovery of either spiked or native residues from soil. For this reason, a number of control experiments first were performed with soils. Blank soils that were extracted at 150 °C for 10 min with CO2, followed by 50 °C water for 10 min, and that showed neither 1960 Analytical Chemistry, Vol. 70, No. 9, May 1, 1998

Dacthal nor any of its metabolites above detection, were spiked to give concentrations of 0.38 µg/g of the monoacid metabolite and 0.23 µg/g of the diacid metabolite. Extractions of the spiked soil gave quantitative recovery (100%) for both metabolites at 50 °C, while none was recovered at 150 °C. In addition, no breakthrough was detected for either metabolite at 150 °C, and a second extraction at 50 °C gave no additional recovery, which suggests that metabolite recovery at 150 °C is not kinetically limited and that metabolites associated with soil may be degraded at 150 °C. Interactions with soil may increase the retention time of metabolites inside the extractor, which, when combined with high-temperature conditions, results in mono- and diacid metabolite degradation. To determine the optimum extraction time for native metabolites, a soil containing native diacid metabolite was extracted (50 °C and 200 bar) in 5-min time intervals. The mass of diacid metabolite recovered over time indicated exhaustive extraction in a total of 10 min. In addition, at 50 °C and 200 bar, no significant difference was observed in the amount of diacid metabolite extracted from soil and trapped on SAX disks inside and outside the extractor, which confirms that SAX disks can be used inside the extraction cell to trap metabolites. Therefore, all subsequent subcritical water soil extractions were 10 min in length and performed with SAX disks inside the extraction cell, because trapping metabolites during the extraction step saves time by reducing the number of subsequent handling steps. Four soil samples containing native Dacthal and its metabolites were extracted for 10 min with 200 bar subcritical water, ranging from 25 to 150 °C. Only the diacid metabolite was detected, and, although Dacthal was present in some of the soils, it was not considered in this set of experiments because subcritical water below 150 °C is ineffective in recovering Dacthal from soil (data not shown). The maximum amount of native diacid metabolite extracted from single samples of each of the four soils was obtained using a water temperature of 50 °C at 200 bar (Figure 4). At 50 °C, the concentration of the diacid metabolite ranged from 1.26 to 2.36 µg/g. Diacid metabolite concentrations declined with increasing temperature, such that the concentrations of the diacid metabolite obtained at 150 °C were 4-63% of that obtained at 50 °C. No breakthrough of the diacid metabolite was

Figure 4. Effect of temperature on the recovery of native diacid metabolite of Dacthal from four soil samples.

detected for any of the soils at any temperature. In addition, no monoacid metabolite was detected at any temperature on the SAX disk inside the extraction cell or in water that had passed through the SAX disk, suggesting that the monoacid is not present in the soil. Blank extractions performed between soil sample extractions indicated no carryover between extractions. Because the objective of the study was to couple supercritical CO2 and subcritical water extraction, a series of control experiments were conducted to determine if a SAX disk in the extraction cell adversely affected Dacthal recovery and if extracting soils first with CO2 was detrimental to metabolite recovery. Supercritical CO2 extractions of soil performed at 150 °C and 400 bar with and without a SAX disk inside the extraction cell gave similar concentrations of Dacthal, indicating that the transfer of Dacthal from the extraction cell to the collection vial is not affected adversely by the SAX disk. Hydrophobic interactions between Dacthal and the styrene/divinylbenzene matrix of the SAX disk apparently are negligible under the conditions of 150 °C and 400 bar CO2. While no monoacid metabolite was detected in the soil, the amount of diacid metabolite obtained from soil samples extracted first with 150 °C and 400 bar CO2, followed by 50 °C and 200 bar subcritical water, was similar to that for samples extracted only with subcritical water. Because metabolite concentrations did not increase as a result of the supercritical CO2 extraction step, hydrolysis of Dacthal to its mono- and diacid metabolites does not appear to occur under supercritical CO2 conditions. Therefore, conducting a supercritical CO2 extraction with a SAX disk in the extraction cell does not adversely effect the recovery of Dacthal or its metabolites and allows for the sequential coupling of supercritical CO2 and subcritical water extractions. Method Detection and Quantitation Limits. The detection and quantitation limits of the combined supercritical CO2 and subcritical water extraction method were determined by spiking a blank soil and extracting the spiked soil for 10 min with 150 °C and 400 bar CO2, followed by a 10-min extraction with 50 °C and 200 bar water with a SAX disk inside the extraction cell. The detection limit of the method, defined as that having a signal-tonoise ratio of 3, is 0.05, 0.06, and 0.08 µg/g for Dacthal, the monoacid metabolite, and diacid metabolite, respectively. The quantitation limit of the method, defined as that having a signalto-noise ratio of 10, is 0.10, 0.12, and 0.15 µg/g for Dacthal, the monoacid metabolite, and the diacid metabolite, respectively. Method Precision. Five replicate samples of two soils were extracted in order to determine the precision of the two-step

extraction process for Dacthal and its metabolites (Table 2). For Dacthal, the precision of the method, as indicated by the relative standard deviation (RSD), was 9.1% and 9.7% for soils B-7 and A-11, respectively, for which five replicate samples were extracted and analyzed. No monoacid metabolite was measured above the detection limit of 0.05 µg/g for soil B-7. The RSD was 11.7% for soil A-11, that had a monoacid metabolite concentration of 0.07 µg/g, which was below the quantitation limit of 0.10 µg/g. The RSD for the diacid metabolite was 14.2% and 13.4% for soil B-7 and A-11, respectively. Application to Environmental Samples. To demonstrate the utility of the method, six soil samples collected from an onion field in eastern Oregon were analyzed for Dacthal and its metabolites. A typical chromatogram indicating the monoacid and diacid metabolites is given in Figure 2b. Dacthal concentrations in the soils ranged from below detection (0.05 µg/g) to 1.39 µg/ g. The concentration of monoacid metabolite typically was below the quantitation limit (0.10 µg/g), and only two samples contained higher concentrations of 0.12 and 0.21 µg/g. Generally, the diacid metabolite was the most abundant form of Dacthal in the soils tested. The concentration of the diacid metabolite ranged from 0.90 to 1.94 µg/g. For the purpose of comparing the amount of Dacthal and its metabolite recovered by sequential supercritical CO2 and subcritical water extraction, the concentrations of the monoacid and diacid metabolites were multiplied by factors of 1.04 and 1.09 in order to report their concentrations as the Dacthal equivalent so that all the residues could be summed and compared to the total Dacthal concentration determined by acid/acetone extraction (Table 2). The total concentration of Dacthal and its metabolites obtained by supercritical CO2 and subcritical water extraction was 75-131% of that obtained by acid/acetone extraction. For half the samples, the concentration of Dacthal and its metabolites was over 100% of that achieved by acid/acetone extraction. Replicate sample analyses were not performed for the acid/acetone extraction, so an RSD was not available. The spent soil from replicate extractions of A-11 soil (75.6% agreement between the two methods, Table 2) was composited and extracted again by the conventional method (0.4 M HCl/acetone). No additional Dacthal was detected in the extract, suggesting that the combined supercritical CO2 and subcritical water extraction process had exhaustively extracted both Dacthal and its metabolites. A greater abundance of the diacid metabolite over both Dacthal, the parent compound, and the monoacid metabolite is consistent with reports that the diacid metabolite is the dominant form of Dacthal in groundwater and soil.4,6 A higher concentration of the diacid metabolite in soil and groundwater compared to those of Dacthal and its monoacid metabolite suggests that the diacid metabolite is more resistant to chemical or biological degradation than Dacthal or the monoacid metabolite, which is consistent with the report by Wettasinghe and Tinsley7 that gave a half-life of 16 days for Dacthal in soil microcosm experiments, with no loss of the diacid metabolite over a period of 300 days. The fact that the monoacid metabolite is subject to both hydrolysis and biological degradation may account for the low to below detection concentrations of the monoacid metabolite in the soils analyzed for this Analytical Chemistry, Vol. 70, No. 9, May 1, 1998

1961

Table 2. Comparison of Native Dacthal and Metabolite Concentrations (Reported as Dacthal) in Six Soil Samples Extracted by the Conventional Acid/Acetone Extraction Method and the Combination of Supercritical Fluid Extraction/Subcritical Water Extraction

soil no.

conventional total Dacthal (µg/g)

no. of samples

B-7

2.37

5

A-11

4.09

5

B-9

1.03

3

A-1

2.30

3

A-2

1.89

3

B-11

2.57

3

SFE and subcritical water extractiona Dacthal (µg/g) monoacidb (µg/g) diacidb (µg/g) 0.11 ( 0.01 (9.1%) 0.90 ( 0.09 (9.7%) nd 1.25 ( 0.12 (9.6%) 0.81 ( 0.04 (4.9%) 1.39 ( 0.29 (20.9%)

ndc 0.07 ( 0.01d (11.7%) 0.12 ( 0.02 (16.7%) 0.22 ( 0.01 (4.5%) 0.08d 0.08d

1.74 ( 0.24 (14.2%) 2.12 ( 0.28 (13.4%) 1.23 ( 0.04 (3.3%) 1.11 ( 0.02 (1.8%) 0.98 ( 0.15 (15.3%) 1.55 ( 0.30 (19.4%)

% of conventional 78.1 75.6 131 112 98.9 118

a Subcritical water conditions include 50 °C and 200 bar water. b Monoacid and diacid concentrations reported as micrograms per gram of Dacthal. Monoacid and diacid concentrations in soil were multiplied by 1.04 and 1.09, respectively, in order facilitate a comparison between total Dacthal obtained by conventional extraction and that achieved by summing the concentration of Dacthal and metabolites obtained by supercritical CO2/subcritical water extraction. c nd, not detected above the detection limit (0.05 µg/g Dacthal, 0.06 µg/g monoacid metabolite, 0.08 µg/g diacid metabolite). d Below the quantitation limit for the monoacid metabolite (0.12 µg/g).

study and others.4,7,31 While the diacid metabolite has a higher water solubility and greater potential for transport through the vadose zone to groundwater, it is less toxic than Dacthal and is considered to be noncarcinogenic.32 CONCLUSIONS Coupling supercritical CO2 and subcritical water allows for the class-selective fractionation of the relatively nonpolar herbicide Dacthal from its very water soluble metabolites in soil. Trapping the metabolites in situ on a strong anion-exchange disk, followed by simultaneous elution from the SAX disk and derivatization, is a rapid alternative to the tedious and time-consuming procedures typically used to determine acid analytes in soil. A total of 25 min is required to complete both extraction steps, and no cleanup (31) Choi, J. S.; Fermanian, T. W.; Wehner, D. J.; Spomer, L. A. Agron. J. 1988, 80, 108-113. (32) Klopman, G.; Fercu, D.; Rosenkranz, H. S. Environ. Toxicol. Chem. 1996, 15, 80-84.

1962 Analytical Chemistry, Vol. 70, No. 9, May 1, 1998

is necessary. The method for extracting Dacthal and its metabolites consumes only 5 mL of acetonitrile and eliminates the use of diazomethane altogether. Moreover, the non-transesterifying derivatization reaction preserves the speciation of Dacthal and its metabolites, which is important, as the form of the pesticide in a soil or water is important for purposes of risk assessment. ACKNOWLEDGMENT This work was funded by a grant from the EPA Office of Exploratory Research (Grant No. R821195). We thank Steve Hawthorne for valuable discussions, Supelco for donation of vacuum manifold, and Lucia Durand for performing the acid/ acetone soil extractions. This is technical report no. 10,751 of the Oregon Agricultural Experiment Station. Received for review October 7, 1997. Accepted February 19, 1998. AC9711091