Final Report: Gas chromatography-isotope ratio mass spectrometry-A novel approach for monitoring the origin and fate of hydrocarbon contaminants in the environment

EPA Grant Number: R826178
Title: Gas chromatography-isotope ratio mass spectrometry-A novel approach for monitoring the origin and fate of hydrocarbon contaminants in the environment
Investigators: Philp, R. Paul , Kuder, T. , Smallwood, B.
Institution: University of Oklahoma
EPA Project Officer: Lasat, Mitch
Project Period: October 1, 1997 through September 30, 2000
Project Amount: $313,743
RFA: Exploratory Research - Environmental Chemistry (1997) RFA Text |  Recipients Lists
Research Category: Water , Land and Waste Management , Air , Engineering and Environmental Chemistry


There were several major objectives of this research project as it evolved over the 3-year period. The first objective, was to demonstrate that the isotopic compositions of individual compounds, as determined by gas chromatography isotope ratio mass spectrometry (GCIRMS), may be used to correlate weathered and unweathered samples and monitor movement of pollutant species, even when excessive biodegradation, water washing, or other weathering processes may render the more conventional techniques, such as GC and GCMS, highly impractical or useless. This includes correlation of hydrocarbons, which have accumulated in various wildlife species affected by oil spills.

The second objective, was to determine whether or not it is possible to use isotopic compositions of individual compounds in different gasoline samples to determine the relative proportions of the gasolines in a mixture of the same. If so, this could become an extremely useful technique for determination of relative proportions of gasoline in a groundwater plume of hydrocarbons derived from leaking underground storage tanks. If successful, attempts would be made to extend the same approach to mixtures of 3 or more gasoline samples.

Oxygenated additives have been used in a few cases to discriminate between gasoline samples from different sources. In this study, it was proposed to determine the isotopic composition of these additives in gasolines from different suppliers to determine whether this might be another viable method for differentiating possible origins for gasolines in groundwater samples.

Additional project goals included GCIRMS analysis of a time series of gasoline sampled daily for a period of one month from one service station, to assess the isotopic variability of gasoline at one site; to optimize HPLC fractionation of the aromatic fraction from a crude oil sample to develop a viable method for correlation of weathered crude oil samples; and to study the isotopic fractionation occurring during biodegradation of methyl tertiary butyl ether (MTBE). If significant isotopic fractionation occurs, it could provide an excellent means of monitoring the extent of remediation.

Summary/Accomplishments (Outputs/Outcomes):

The majority of the above mentioned objectives have been realized. We have demonstrated that carbon isotopes are a powerful tool for discriminating oils or refined products from different sources. This demonstration will be particularly important, as work continues for the study of refined products where the standard techniques using biomarkers will not work due to the absence of biomarkers in the refined products. The most exciting aspect of this work for investigators was the opportunity to study with MTBE, which have developed in collaboration with BP and the Kerr EPA Laboratory in Ada, OK. The investigators initially observed no significant variation in the source isotopic values of MTBE, making it unlikely that such an approach would be useful in differentiating sources. However, further observation concluded that with increasing biodegradation, the isotopic composition of the residual MTBE became significantly enriched in the heavier isotope. The point here is the possibility of using this enrichment in 2 ways: (1) to demonstrate the onset of natural attenuation of MTBE; and (2) to develop the possibility of this enrichment into an age-related diagnostic tool.

A summary of the major findings from this project is provided below. Major results could be divided into the categories of development/optimization of analytical techniques and characterization of the isotopic effects of contaminant's source and weathering.

Gasoline Analysis - Correlation of Sources and Effects of Weathering

Neat Samples: 19 gasoline samples from various locations throughout the U.S. (Table 1) have been analyzed by GCIRMS. Due to the wide ranging concentrations of individual hydrocarbon components within the gasoline samples, each sample had to be analyzed 4 times, twice with a small amount of sample (0.1 µL), and twice with a larger amount (0.6 µL). This procedure ensured that isotopic numbers for the most concentrated compounds and the least concentrated compounds were duplicated. In addition, to determine reproducibility for this procedure, one sample (DTX) was analyzed 14 times at various concentrations (0.1-0.6µL). Reproducibility ranged from 0.03 - 0.57 percent for 46, apparently fully resolved, compounds within the gasoline sample, and was 0.51 and 0.36 for the deuterated standards n-nonane and n-decane, respectively. Comparisons for all other gasolines were made on the basis of these data, although in some instances, due to compounds being absent from some of the gasolines, the total number of components actually used in statistical testing was less than 46.

Ultimately, 16 compounds that were ubiquitous to each gasoline were selected for more detailed statistical analysis. Each compound showed a wide range of variability, the greatest being for benzene, 2,3-dimethylpentane, ethyl benzene, 1-methyl-2-ethylbenzene, and 1,2,3-trimethylbenzene, which have standard deviations greater than 2.40 from the mean for all samples. In addition to this analysis, a statistical test was carried out that compared the isotope numbers for each compound from each gasoline with every other gasoline. This test confirmed that of the 171 possible combinations of gasoline samples, 169 were significantly different from each other. The isotopic fingerprints of 2 pairs of samples were remarkably similar. It has been concluded that these gasolines, even though sampled from different gasoline companies, used the same supplier. The variability observed for the compound specific carbon isotopic distribution of gasolines within this sample group indicate the potential of using GCIRMS as a means to differentiate between fresh gasolines on a local and national scale.

MTBE: A survey of the isotopic composition of "neat" MTBE directly from suppliers throughout the U.S. proved to be impossible, mainly due to the manufacturers' unwillingness to send samples of their product. Bulk isotopes of 2 samples that were purchased gave numbers of -30.74 percent and -30.52 percent. 10 gasolines contained MTBE, and gave reproducible carbon isotope data with a range of numbers from -28.29 percent (DTX) to -30.35 percent (CTX). By comparing the isotopic composition of MTBE of every gasoline, in a similar statistical method to that used above, it was found that 26 of the 45 combinations were isotopically distinct.

Analysis of Gasoline-Range Hydrocarbons in Water: A Purge and Trap interfaced to the GCIRMS enabled the measurement of the isotopic composition of gasoline-range contaminants that are present in extremely low concentrations (e.g., MTBE at 15 ppb, more recently down to 1 ppb) in both groundwater and soil samples. A preliminary investigation of the use of purge and trap verses heated headspace has shown that isotopic numbers are reproducible to standard deviations of 0.4 vs. 1.2 percent for the same compound (MTBE). Gasoline contaminated water and soil samples also have been analyzed using this method, and reproducibility is less than 0.5 percent.

Evaporation and Water Washing: 3 gasoline samples (EOK, FOK, and GOK) were evaporated for a period of 168 hours, with aliquots taken after 24, 48, 120, and 168 hours. Gas chromatograms of sample EOK are shown in Figure 3. The gasoline after 168 hours clearly is different from the fresh gasoline sample and would, therefore, be impossible to use for spill to source correlations. The carbon isotopic composition for those compounds (naphthalene, 1-methylnaphthalene, and 2-methylnaphthalene) remaining after 168 hours are similar to the initial sample and are deemed isotopically stable. A number of compounds are, however, not stable after a period of 24 hours, and these probably cannot be used for spill to source correlations; however, they can be used for remediation studies.

One gasoline sample (DTX) was washed with water to simulate gasoline in contact with groundwater. Analysis of the headspace of the water after a period of one week determined that the only present compound was MTBE, and the isotopic composition of MTBE was similar to that of the fresh gasoline sample (standard deviation 0.40 percent). 12 compounds present in the gasoline after water washing were isotopically stable, 9 compounds were significantly different, and 3 of these were compounds that also were different after the evaporation study. These results suggest that a number of compounds can be used to discern the source of gasoline contamination, while others can be used to determine how long the contamination has been present.

Sub-Fractionation of Aromatic Compounds

Determination of the isotopic composition of aromatics in a crude oil is not feasible by analysis of the total aromatic fraction alone, due to the complexity of the fraction, and consequently, the aromatics have to be fractionated further. A standard mixture of aromatics has been prepared and separated according to ring number using preparative HPLC with a Spherisorb amino column. While various aromatic fractions have been characterized by GCIRMS, this work is still in progress.

Developing Techniques for GCIRMS of Groundwater Contaminants

Purge and trap-GCIRMS and SPME-GCIRMS were applied to a number of environmental samples, including groundwater contaminated with MTBE, and another suite of samples affected by coal tar leachate. Both techniques provided reproducible isotopic data, allowing sample correlation/discrimination. Very promising sensitivity of SPME (< 1ppb for MTBE if no interfering compounds are present, approximately 0.3 ppb for BTEX, with a potential for further optimization), makes it applicable to most water contamination issues.

Using GCIRMS to Determine the Extent of Remediation of MTBE

Spilled product to source correlation based on stable isotope data can be made more difficult if the original isotopic signature is distorted by any fractionation process. In other words, it is crucial for the isotopic signal to be as inert as possible to facilitate correlation. Preliminary data (FY 1999) suggested that this is the case at least for some gasoline components, especially naphthalenes, remaining isotopically stable during water washing and in gasoline evaporation residues. On the other hand, if the original isotopic composition is known and/or, if a series of samples from a plume gradient is analyzed, the extent of fractionation can be informative on the extent of intrinsic remediation taking place after the spill. Results showed that MTBE becomes isotopically heavier during aerobic biodegradation. Fractionation due to inorganic weathering (e.g., evaporation) also can be significant. A number of light-end gasoline products were shown to fractionate upon evaporation, precluding their possible use as a reliable source of information for spill to source correlation. On the other hand, medium-range gasoline hydrocarbons, such as p-xylene and naphthalene, were confirmed to be isotopically stable upon significant mass loss due to weathering, allowing their use for correlation purposes. A distinct Rayleigh-type isotopic fractionation was registered for the more volatile compounds, e.g., benzene, suggesting a good possibility that a stable isotopic proxy for remediation could be developed.

The continued study of the carbon isotopic composition of MTBE samples collected throughout the country was expanded by the measurement of the hydrogen isotope (H/D ratio) composition for 4 MTBE samples. The H/D ratio showed significantly more variation than carbon, even where the samples had similar 13C signatures, adding another variable to distinguish between different sources of the product.

Journal Articles on this Report : 2 Displayed | Download in RIS Format

Other project views: All 13 publications 3 publications in selected types All 2 journal articles
Type Citation Project Document Sources
Journal Article Smallwood B, Philp RP, Burgoyne TW, Allen J. The use of stable isotopes to differentiate specific source markers for MTBE. Environmental Forensics 2000. R826178 (2000)
R826178 (Final)
not available
Journal Article Smallwood BJ, Philp RP, Allen JD. Stable carbon isotopic composition of gasolines determined by isotope ratio monitoring gas chromatography mass spectrometry. Organic Geochemistry 2002;33(2):149-159. R826178 (Final)
not available

Supplemental Keywords:

groundwater, soil, chemicals, VOC, organics, NAPL, remediation, bioremediation, environmental chemistry, analytical methods, petroleum, refined products,, Scientific Discipline, Air, Waste, Ecosystem Protection/Environmental Exposure & Risk, Ecology, Environmental Chemistry, Chemistry, Fate & Transport, Ecological Risk Assessment, Biology, Engineering, Chemistry, & Physics, fate and transport, gasoline, hydrocarbon, mass spectrometry, MTBE, isotope ratio, gas chromatography, chemical kinetics, oil spills, isotope ratio mass spectrometry, weathering, wildlife

Progress and Final Reports:

Original Abstract
  • 1998
  • 1999 Progress Report