INTRODUCTION

Measurement of fuels and other petroleum products in soils is an important environmental activity. Field screening methods for this purpose are especially useful for defining boundaries of contamination and providing information on where samples should be taken for more expensive and time consuming laboratory analyses. Field screening methods can also result in cost savings for site excavations by minimizing their duration and the incidence and costs of unnecessary removal of uncontaminated soils.

The quality of data generated by field and laboratory methods for analyzing soils for hydrocarbon contamination is dependent on the availability of information concerning the contaminant and the availability of a portion of the specific contaminant for standardization (Rhodes et al. 1996). Since the contaminant fuel is rarely available for calibration, the analytical results are dependant on the method and calibration material used. Thus, the various methods rarely provide comparable or truly accurate results. The common laboratory methods, which involve gas chromatographic separations, usually disregard the presence of materials heavier than diesel-range contaminants. Thus, used motor oil and heavy fuel oil contaminants are rarely reported. No two fuel analysis methods, laboratory or field, can be expected to give the same answer, because of the tremendous variations in fuels, the different principles of measurement, and the variation in soil samples, even if they are collocated. Typically, a laboratory reference method is agreed upon by the regulatory agency, the site owner, and the contractor. A field screening method is often used to select the points from which samples are collected for the laboratory analysis. It is important that the field method used does not provide false negative results. In addition, the manner in which the selected field screening method relates to the laboratory method should be understood. In planning remediation activities, field and laboratory method selection is very important.

Volatile petroleum products in soil, such as gasoline, can be screened in the field by simple headspace photoionization or flame ionization organic vapor analyzers (OVAs). However, mid-range distillates, such as diesel fuel, and heavier products are not effectively measured by these techniques and require a screening procedure that is sensitive to diesel-range and heavier petroleum products. ASTM Method D 5831, Standard Test Method for Screening Fuels in Soils, provides such a procedure (ASTM 2000). This screening method has been tested in the laboratory and shown to recover mid-range and heavier petroleum products from soil (Schabron et al. 1995). Fuels containing aromatic compounds, such as diesel fuel, as well as other aromatic-containing hydrocarbon materials, such as motor oil, crude oil, and coal oil can be determined using the procedure. This screening method focuses on aromatic components in the contaminant, which are generally considered to be the more toxic components of petroleum hydrocarbon products with regard to both human health and the environment. This aspect of the method is particularly useful because as diesel fuel in soil is subjected to bacterial degradation and weathering processes, the remaining fuel is more aromatic than the starting material and less volatile (Douglas et al. 1992).

The ASTM screening method uses low-toxicity chemicals and can be used to screen organic-rich soils. It is fast, taking about 10 minutes per sample; easy; and inexpensive to perform. The procedure calls for mixing a five-gram soil sample with approximately five grams of calcium oxide, which prevents interferences from moisture and humic materials in the soil (Schabron et al. 1995). The mixture is extracted with 50 milliliters of isopropyl alcohol (IPA) for three minutes. The resulting extract is filtered, and the ultraviolet (UV) absorbance of the extract is measured at 254 nm. If the contaminant fuel is available for calibration, the approximate concentration of fuel in the soil can be determined; if the fuel type is known, but a sample of the contaminant fuel is not available for calibration, an estimated concentration of fuel contamination in the soil can be calculated using an average response factor, which is given in the method; and if the nature of the contaminant fuel is not known, the UV absorbance value is used to indicate the presence or absence of contamination.

The Diesel Dog® soil test kit has been developed for field application of ASTM Method D 5831. The ASTM method can easily be performed in the field using the soil test kit. A 5-gram soil sample is weighed using a portable balance. After addition of calcium oxide, which is provided in moisture-proof, pre-weighed packets, and IPA, the mixture is stirred using a mechanical 12V extractor. Extraction is not performed manually, as is common for most field analysis methods, because manual agitation has been shown to be inefficient and non-repeatable (Schabron et al. 1995). The mechanical extractor has only glass, Teflon®, and stainless steel wetted parts and operates with a portable battery pack. A disposable syringe and filter are used to transfer the extract to a quartz cuvette, which is then inserted into a 12V portable photometer. The absorbance of the extract is measured at 254 nm, which provides direct measurement of the aromatic components. The portable photometer also operates with the battery pack. The portable battery can be charged before being taken to the field and also has a cigarette lighter socket adapter for convenient use of the kit on the tailgate of a pickup truck.

The Diesel Dog soil test kit and ASTM Method D 5831 have been used at several sites to successfully screen fuel contamination in a variety of soils (Butler et al. 1997, Schabron et al. 1997, Sorini and Schabron 1996). In addition to these uses, the soil test kit and ASTM method have recently been used by ENSR Corporation and Wyoming Department of Environmental Quality (WYDEQ) at a number of field sites. These include an aged diesel-pond site in the U.S. Virgin Islands, a diesel-contaminated site in Georgia, a decades-old diesel and road tar site near Jackson, Wyoming, and a filling station site in Sheridan, Wyoming. This article describes use of the Diesel Dog soil test kit and ASTM Method D 5831 at these sites, as well as their use at a private residence where a diesel spill caused the risk for potential contamination of a drinking water well. For each of these case studies, information is provided on how the method and soil test kit performed in the various applications.

CASE STUDIES

Aged Diesel-Pond Site

ENSR Corporation of Acton, Massachusetts used a Diesel Dog soil test kit and ASTM Method D 5831 at a field site in the U.S. Virgin Islands. The site was a one-year old pond impacted by diesel fuel. The soil in the area was very wet and sandy. Soil samples were screened in the field using the soil test kit and ASTM method. Soil samples were also analyzed in the laboratory using the Massachusetts Department of Environmental Protection extractable petroleum hydrocarbon (MADEP EPH) method (MADEP 1995). This method involves methylene chloride extraction of the soil in a Soxhlet apparatus. The solvent is removed from the soil extract, and the extract is re-dissolved in hexane prior to separation into aliphatic and aromatic fractions. The resulting extracts are analyzed using gas chromatography with flame ionization detection (GC-FID).

As discussed, the results generated by the ASTM and MADEP EPH methods can be expected to vary because of the differences in measurement techniques. The concentration values for ASTM Method D 5831 are estimated values that were calculated using the response factor for diesel fuel that is given in the ASTM method. The ASTM and MADEP EPH methods detected hydrocarbon contamination in all of the samples that were analyzed using both procedures. ASTM Method D 5831 results are higher than the MADEP EPH method results for five of these eight samples. For these samples, there may have been some heavier hydrocarbon materials (>C22) present, which the GC method would not detect. As is required of a screening method, no false negative results were generated by ASTM Method D 5831.

Diesel-Contaminated Site

ENSR Corporation also used a Diesel Dog soil test kit and ASTM Method D 5831 at a site in Georgia contaminated with diesel fuel. The soil test kit and ASTM method were used to screen soil samples collected at drilling points. Soil samples were also analyzed in the laboratory using EPA Method 8015B (US EPA 1996), which involves analysis of methylene chloride extracts of the soil using GC-FID.

As discussed, differences in the results generated by the two methods can be expected because of the differences in measurement techniques. The concentration values for ASTM Method D 5831 are estimated values that were calculated using the response factor for diesel fuel that is given in the ASTM method.

The approximate quantitation limit (LOQ) of the ASTM method for diesel is 75 mg/Kg. The LOQ of EPA Method 8015B for analysis of the samples in this study was 12 mg/Kg. Of the 18 samples analyzed by both methods, five of the samples (6, 8, 11, 12, and 14) were determined to have contaminant concentrations less than the LOQs for both methods. Five other samples (2, 3, 9, 10, and 15) were determined to have estimated diesel concentrations slightly above the method LOQ using the soil test kit and below or just above the method LOQ using Method 8015B. For samples 1, 5, 13, and 18, EPA Method 8015B determined <12 mg/Kg of contaminant to be present, while ASTM Method D 5831 estimated concentrations of 130, 930, 1,700, and 120 mg/Kg, respectively, to be present in the samples. The laboratory method missed contamination in these four samples. Any weathering and bacterial degradation that occurred at the site may have degraded aliphatic portions of the contaminant leaving aromatic structures, which are tightly adsorbed to the soil matrix. IPA is a more powerful chromatographic solvent for displacing these adsorbed species than methylene chloride used in the laboratory method, resulting in a greater extraction efficiency (Snyder 1968 and Schabron et al. 1995). Three samples (4, 7, and 17) showed significant contamination by both methods, and for two of these, the ASTM method results were higher, which may also be due to the extraction efficiency of IPA.

The Diesel Dog soil test kit and ASTM Method D 5831 did not fail to detect contamination in any of the samples when compared to the data generated using EPA Method 8015B. The results from this study show that if ASTM Method D 5831 and the Diesel Dog soil test kit were used to guide an excavation at this site, the user could be confident that laboratory data generated by EPA Method 8015B would show that the cleanup had been performed completely and successfully.

Decades-Old Diesel and Road Tar Site

WYDEQ used ASTM Method D 5831 and a Diesel Dog soil test kit to guide excavation at a decades-old, fuel-contaminated site near Jackson, Wyoming. The site was contaminated with diesel fuel and road oil from prior transportation department activities. The soil was a heavy wet clay. Under oversight by the WYDEQ, the engineering firm of Dames and Moore excavated about 6,000 cubic yards of soil, including about 2,000 cubic yards of overburden. Photoionization detector-based OVAs could not detect the contamination because the fuel had been weathered severely. A Diesel Dog soil test kit was used by a chemical engineer and a civil engineer to perform ASTM Method D 5831 on the tailgate of a pickup truck to provide rapid field data. According to the engineers, the soil test kit provided data within minutes. Soil samples were also analyzed in the laboratory using EPA Method 8015B. The laboratory data were obtained by a purge and trap sampling method for total volatile petroleum hydrocarbons (TVPH, C6 - C10) and by solvent extraction for total extractable petroleum hydrocarbons ( TEPH, C11 - C28).

Two sets of data were generated using ASTM Method D 5831. Estimated concentrations of diesel fuel in the samples were calculated using the response factor for diesel (209 mg/L/AU) that is given in the method. The other contaminant at the site was road tar. The location of the site in Wyoming suggests that the road tar came from the highly aromatic Recluse, Wyoming oil. As a result, the concentration of oil in the samples was estimated using the response factor for coal oil (58.7 mg/L/AU), which corresponds to a highly aromatic oil.

The results of ASTM Method D 5831 for samples 1-6 are significantly higher than those determined by the laboratory method, regardless of whether the response factor for diesel or coal oil is used. This result is not surprising for several reasons. First, the contamination at the site occurred decades ago, and extensive weathering and bacterial degradation have occurred. Aliphatic portions of the contaminant have been degraded by bacterial action, leaving the most persistent portion of the contaminants, the aromatic structures, which are tightly adsorbed to the soil matrix. These can have aromatic structures >C22, which would not be detected by the gas chromatography method. Also, as mentioned, IPA is a more powerful chromatographic solvent for displacing these adsorbed species than methylene chloride, resulting in a greater extraction efficiency. A similar trend was observed in a study in which spiked soils were weathered artificially (Schabron et al. 1995) and in a study involving use of the ASTM method and soil test kit at a railroad site where diesel fuel had been released during railroad maintenance activities for a period spanning approximately 80 years (Schabron et al. 1997).

When samples 7 and 8 were analyzed using the soil test kit in the field, the results showed significant contamination requiring action. The field screening had identified a “hot spot” area within two days before site activities were to be terminated. The highly contaminated area was successfully excavated using ASTM Method D 5831 and the Diesel Dog soil test kit data. Data confirming the “hot spot” of contamination and successful excavation of the area were obtained from laboratory analysis more than a week after the site cleanup effort had ended.

Filling Station Site

WYDEQ conducted an excavation project involving removal and disposal of 9,200 cubic yards of contaminated soil from a filling station site in Sheridan, Wyoming. The contamination was the result of two recent gasoline spills and several smaller older spills. Four soil samples were taken from the excavation, two from the bottom and two from the sidewalls. It was assumed that the side- wall samples were not contaminated. The results from ASTM Method D 5831 analyses of the soil samples from the sidewalls showed no contamination. The samples from the bottom of the excavation contained estimated concentrations of 180 and 220 mg/Kg diesel fuel according to the Diesel Dog soil test kit results. The corresponding laboratory GC analysis results were 32.5 and 66.7 mg/Kg, respectively, for gasoline to diesel-range fuels. The results that were generated using the ASTM method are higher than the laboratory GC results, which as discussed, is to be expected for weathered fuel contamination.

Bus Barn Site

WYDEQ conducted an excavation of 2,670 cubic yards of contaminated soil from a transportation depot facility. The ground was contaminated by leakage from an underground storage tank, which had been removed in 1989. Following the excavation, two soil samples were collected from a sidewall and analyzed at the site using ASTM Method D 5831 and the Diesel Dog soil test kit. No contamination was detected in the samples, and follow up laboratory analysis was deemed to be unnecessary.

Emergency Response for a Diesel Spill

A contractor was hired at a private residence to steam clean the carpets and upholstery. The steam cleaning equipment was in a large trailer, which contained a diesel engine. After the contractor had completed the job, which took several hours, a diesel spill on the driveway and in the soil and gravel along the side of the driveway was discovered. The exact perimeter of the spill was difficult to determine because a lawn sprinkler had wetted the area before the spill was discovered. The homeowner was concerned because the spill was only about 15 feet from his drinking water well. The contractor estimated the spill to be 5 to 10 gallons. Although there was little chance that this amount of diesel could penetrate the soil to the aquifer 150 feet below, there was the potential that the well and aquifer could become contaminated via the well casing. If the outside of the casing and the surrounding bedrock had not been sealed with concrete at the time that the well was drilled, surface water could flow down the outside of the well casing into the aquifer below.

ASTM Method D 5831 and the Diesel Dog soil test kit were used at the residence to determine the concentration of diesel contamination and the extent of the excavation needed to remediate the spill site to proper cleanup standards. The WYDEQ regulatory level of 100 mg/Kg was used as the remediation standard. A semicircle at the edge of the driveway about 5 feet wide and 2 feet deep was excavated. Soil was sampled from three equally spaced locations around the inside perimeter of the excavated area. The soil samples were analyzed using the soil test kit on the tailgate of a Jeep®. The estimated concentrations of diesel in the samples were determined using the response factor for diesel that is given in the ASTM method. Samples 1 and 2 show that the first excavation was sufficient to collect the spilled diesel that absorbed into the soil around the driveway. However, the third sample, which was collected from under the concrete slab of the driveway, contained a very high concentration of diesel. As a result, additional excavation was performed, and when this was completed, two soil samples were collected from under the driveway and analyzed using the soil test kit. The estimated concentrations of diesel in these samples fall below the regulatory level of 100 mg/Kg. With these results, the homeowner and contractor were confident that the spill had been cleaned up to proper levels, and the possibility of aquifer and well contamination had been avoided.

SUMMARY

ASTM Method D 5831, Standard Test Method for Screening Fuels in Soils, provides an easy, fast, and inexpensive screening method for fuel contamination in soil. The Diesel Dog soil test kit was developed for field application of ASTM Method D 5831 for use in site evaluations, cleanup activities, etc. This article describes use of the Diesel Dog soil test kit and ASTM Method D 5831 at a variety of fuel-contaminated sites. For each of the case studies, information is provided on how the method and soil test kit performed. In general, ASTM Method D 5831 and the Diesel Dog soil test kit provided higher values than the laboratory methods (MADEP EPH method and EPA Method 8015B). This is due to the differences in measurement techniques. In many cases, the screening method detected higher contaminant concentrations because of its sensitivity to the more aromatic components in the contaminant and because of the extraction efficiency of IPA. Of particular importance is that the ASTM method and soil test kit did not provide false negative results for any of the samples described in the case studies.

REFERENCES

American Society for Testing and Materials, 2000, ASTM Method D
   5831-96, Standard Test Method for Screening Fuels in Soils. Annual
   Book of ASTM Standards, Vol. 11.04, 319-326.

Butler, E.L., S.H. Frisbie, J.F. Schabron, S.S. Sorini, and A.D. Wait,
   1997, New ASTM TPH Screening Method. Environmental Laboratory,
   February-March, 11-14.

Douglas, G.S., K.J. McCarthy, D.T. Dahlen, J.A. Seavey, W.G.
   Steinhauer, R.C. Prince, and D.L. Elmdorf, 1992, The Use of
   Hydrocarbon Analyses for Environmental Assessment and
   Remediation. in Kostecki, P.T. and E.J. Calabrese, eds.,
   Contaminated Soils - Diesel Fuel Contamination, Lewis Publishers,
   Chelsea, MI, 1-21.

MADEP, 1995, Method for the Determination of Extractable Petroleum
   Hydrocarbons (Public Comment Draft 1.0). Massachusetts
   Department of Environmental Protection, Woburn, MA.

Rhodes, I.A.L., E.M. Hinojosa, D. A. Barker, and R. A. Poole, 1996,
   Conventional TPH Pitfalls. Environmental Lab, December/January
   1995/96, 16-20.

Schabron, J.F., N.D. Niss, B.K. Hart, and S.S. Sorini, 1995, Remote
   Chemical Sensor Development: A New Field Screening Method for
   Soil Fuel Contamination. Laramie, WY, WRI Report WRI-95-R016.

Schabron, J.F., S.S. Sorini, E.L. Butler, and S. Frisbie, 1997, Field
   Evaluation of a Standard Test Method for Screening Fuels in Soils at a
   Railroad Site. Proceedings of the 1997 USEPA/A&WMA International
   Symposium on Field Screening Methods for Hazardous Wastes and
   Toxic Chemicals, Air & Waste Management Association, 238-243.

Snyder, L. R., 1968, Principles of Adsorption Chromatography. Marcel
   Dekker, Inc., New York, NY, 194-195.

Sorini, S.S. and J.F. Schabron, 1996, Evaluation of a Standard Test
   Method for Screening Fuels in Soils. Proceedings of the 1996
   EPA/ACS Waste Testing and Quality Assurance Symposium,
   Washington, D.C., 51-58.

U.S. EPA, 1996, Method 8015B: Nonhalogenated Organics Using
   GC/FID. Test Methods for Evaluating Solid Waste: Physical/Chemical
   Methods (SW-846), Vol. 1B, 3rd Ed, Final Update III.

ACKNOWLEDGEMENTS

Funding for this work was provided by the U.S. Department of Energy, National Energy Technology Laboratory, under Cooperative Agreement DE-FC26-98FT40323. The authors would also like to acknowledge ENSR Corporation, Dames & Moore, and Wyoming Department of Environment Quality for their contributions to the information presented.

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe on privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

Susan S. Sorini is a senior scientist, John F. Schabron, Ph.D. is a principal scientist, and Joseph F. Rovani, Jr. is a senior scientist at Western Research Institute in Laramie, WY.
 

Top