Purge-and-Trap Extraction of Geosmin and 2-Methylisoborneol from Soil

Although a number of techniques are available for estimating the biomass of filamentous microorganisms in soil, determination of their activity is much more difficult. We report on preliminary studies evaluating the potential use of the volatile microbial secondary metabolites geosmin and methylisoborneol as indicators of activity of filamentous microorganisms. The purge-and-trap technique for analysis of low concentrations of volatile organic compounds in air and water was tested for extracting the earthy-musty odor compounds geosmin and 2-methylisoborneol from soil. Two variations of this method were tried, one in which soil samples were placed in water for purging and another in which samples were purged with no added water. Volatile organic compounds purged from soil were collected on a porous polymer sorbent, Tenax TA. Naturally occurring geosmin and 2-methylisoborneol and added standards were successfully extracted from soil by both methods. Recovery efficiencies, however, were low. Results indicated that wet purging was the most efficacious extraction method yielding maximum recovery rates of added geosmin and 2-methylisoborneol standards of 15 and 24%, respectively. G and 2-methylisoborneol (Fig. 1) are volatile organic compounds produced as metabolites of microorganisms and are important contributors to the earthy-musty odor of soil (Gerber and Lechevalier, 1965; Buttery and Garibaldi, 1976; Stotzky and Schenck, 1976). Both compounds are produced by certain actinomycetes (Gerber and Lechevalier, 1965; Gerber, 1979), cyanobacteria (Saiferman et al., 1967; Tabachek and Yurkowski, 1976), and fungi (Mattheis and Roberts, 1992; Karahadian et al., 1985). Geosmin production has also been reported by amoeba (Hayes et al., 1991). However, the biomass of actinomycetes and fungi is far greater in most soils than that of cyanobacteria and amoebae, leading us to hypothesize that filamentous microorganisms produce the large majority of geosmin and 2-methylisoborneol in soil. Because these metabolites are readily degraded by some common soil bacteria (Narayan and Nunez, 1974) and, thus, probably do not accumulate in soil, they may be of value as indicators of activity of the organisms that produce them. The purge-and-trap technique (Swinnerton and Linnenbom, 1967) is an efficient and commonly used method for the determination of low concentrations of volatile organic compounds in air and water. This type of system is commonly used for analysis of volatile organic compounds in the headspace of cultures of microorganisms. A modification of the purge-and-trap method, the closedloop stripping system (Grob, 1973; Grob and Zurcher, 1976), has been used to determine amounts of geosmin and 2-methylisoborneol in water samples at levels below 1.0 ng L" (McGuire et al., 1981; Krasner et al., 1983). The purpose of this study was to investigate the use of purge-and-trap methods for the extraction of geosmin USDA/ARS National Soil Tilth Lab., 2150 Pammel Drive, Ames, IA 50011. Received 11 Oct. 1993. *Corresponding author. Published in Soil Sci. Soc. Am. J. 58:1163-1167 (1994). and 2-methylisoborneol from soil as part of an effort to determine their usefulness as indicators of activity of filamentous microorganisms. The specific objectives addressed in this study were: (i) extract added standards of geosmin and 2-methylisoborneol from soil to determine the efficiency of the purge-and-trap method, and (ii) attempt to extract naturally occurring geosmin and 2-methylisoborneol from soils. MATERIALS AND METHODS Two soils were used in this study, a cropped soil and a relatively undisturbed woodland soil. The cropland soil is a Clarion loam (fine-loamy, mixed, mesic Typic Hapludoll) from the Iowa State University Ankeny Research Farm in Polk County. The woodland soil is a Lester loam (fine-loamy, mixed, mesic Mollic Hapludalf) collected from McFarland Park Nature Center in Story County. Standards of geosmin and 2-methylisoborneol were obtained from Wako Chemicals (Richmond, VA). Both compounds were diluted in denatured ethyl alcohol before addition to soil or used as calibration standards. Purge-and-trap extractions were conducted on soils to which standards were added to make final concentrations ranging from 10.0 to 0.01 mg kg" soil. Standards were added to soil samples by first placing the required amount of geosmin, 2-methylisoborneol, or both in 5 mL of denatured ethanol. The 5-mL diluted aliquot was then slowly added to 120 g of soil in a 500-mL beaker and mixed with a glass stirring rod. The beaker was then covered with aluminum foil and the sample was shaken vigorously for 30 s to thoroughly incorporate the added standards into the soil. Treated soil was then allowed to sit, covered, for 45 min before being divided into 60-g portions for purging. Although the escape of some of the standards could be detected by smell at this time (the human nose is sensitive to these compounds in air at concentrations in the low parts-per-billion range), we believe the loss was insignificant. Two methods for purge-and-trap extraction of geosmin and 2-methylisoborneol from soils were tested: (i) wet purging in which soil samples were placed in water for extraction, and (ii) dry purging where samples were placed in a chamber with no added water and purged. The apparatus used (Fig. 2) consisted of a sample chamber connected to a vacuum manifold. The sample chamber used for wet purging consisted of a tall form gas washing bottle (Krasner et al., 1983) and for dry purging a long glass tube (1 cm i.d. by 50 cm) was used (Fig. 2). Geosmin and 2-methylisoborneol purged from soil samples were collected in two in-line traps consisting of silani/ed glass tubes (11.5 cm by 4 mm i.d.) each loosely packed with Tenax TA (a porous polymer sorbent, Alltech Associates, Deerfield, IL) (Mattheis and Roberts, 1992), attached directly to the outlet of the sample chamber. The vacuum line was connected to the outlet end of the trap to create an air flow through the system, flushing volatiles from the sample and drawing them into the trap (tests of the efficiency of Tenax traps at capturing geosmin and 2-methylisoborneol are described below). Air entering the system was passed through an activated charcoal filter to prevent contamination of the sample. For wet purging, 500 mL of H2O and a magnetic stir bar were placed in a clean tall form gas washing bottle. Sixty grams of soil were then added to the water, the top of the