Yakov Galperin, Ph.D. and Henry Camp

Identification of the product-type in the petroleum-contaminated samples is likely the most common task of an environmental forensic investigation. It is also one of the most important, as it frequently forms the foundation upon which many of the relevant conclusions are derived. Among the analytical methods used to identify a petroleum type are those that focus on specific hydrocarbon classes, such as alkanes, isoprenoids, polynuclear aromatic hydrocarbons (PAH), and polycyclic alkanes (e.g., sterane and terpane biomarkers). These methods, while specialized, are routinely applied by laboratories conducting environmental forensic investigations and the basic principles and application of the techniques are well documented.

Straight-chain alkanes (normal alkanes or “n-alkanes”) are abundant in most crude oils and in a variety of refined petroleum products and intermediates. Traditionally, this class of hydrocarbons is analyzed utilizing gas chromatography with flame ionization detection (GC/FID). An example of a familiar “picket fence” pattern of n-alkanes in crude oil is shown in Figure 1. For petroleum products that have not undergone extensive weathering, the n-alkane pattern can often provide data adequate to identify a fuel-type in the environmental samples and to perform initial evaluation of their source relationship.

The most serious limitations for the use of n-alkane patterns arise from the fact that upon release to the environment, the petroleum is subjected to weathering processes, such as evaporation, water washing and biodegradation. These processes cause the chemical make-up of a product to change, in some cases quickly and drastically. Because n-alkanes are among the most biodegradable hydrocarbons, they are readily broken down and preferentially depleted from environmental samples. A resulting disappearance of the n-alkane pattern renders this approach inadequate and requires consideration of the more refractory classes of hydrocarbons.

Because it utilizes a non-specific detector, the GC/FID technique is less useful for identification of other hydrocarbon classes. For this reason, analytical techniques with greater sensitivity and specificity are employed for the fuel-type identification in environmental samples. An example is the use of gas chromatography-mass spectrometry (GC/MS) in a selected ion monitoring mode (SIM) to analyze for PAH and biomarkers – hydrocarbons that occur at relatively low levels in petroleum products. Whereas these two classes of hydrocarbons are commonly applied in forensic investigations of crude oil and heavy refined products, they are proved to be only marginally useful for identification of light (naphtha) and middle distillate (kerosene-diesel) fuels. The limitations develop because some of PAH and most of biomarkers are beyond the boiling range of common fuels and are excluded from the finished material in the refining process.

The latter point can be illustrated by considering basic principles of petroleum refining. Crude oil is comprised of a wide range of hydrocarbons from light gases to heavy residues. In a simplified description of petroleum refining, crude oil is separated by distillation into three broad fractions: naphtha (boiling range 210-390F), middle distillate (boiling range 300-750F) and residual oil (600-1000F). The naphtha fraction is mainly used for gasoline after further processing for octane improvement. The light-end middle distillate is used for kerosene and kerosene-range products such as specialty solvents (mineral spirits, stoddard solvent, etc), certain jet fuels, and light diesel fuel (diesel #1). Diesel-range products such as diesel fuel #2, heating oils, and some jet fuels are made from the heavy-end middle distillate fraction. The composition of the refined products is thus defined by their boiling ranges so that heavier PAH and biomarkers are mostly excluded.

Fuel-Specific Distribution of Alkylcyclohexanes

The need to determine liability for releasing, and consequently, for the cleanup of petroleum contaminants has resulted in development of the more advanced methods for identification of petroleum products. In particular, the research has focused on hydrocarbons that are more recalcitrant than n-alkanes and therefore, can be useful for evaluation of the weathered environmental samples. The evaluation of the chemical composition of different petroleum products revealed other suitable hydrocarbon classes. One of these classes, the alkylcyclohexanes, was found to be the most useful for identification of fugitive light and middle distillate fuels.

Alkylcyclohexanes belong to a class of naphthenes or cycloparaffins – the most common molecular structures in petroleum. In average, crude oil contains about 50% naphthenes that are formed by joining the carbon atoms in a ring. The homologous series considered in this article are based on the six-membered cyclohexane ring (Figure 2) with a single n-alkane side chain (R) attached to the ring. In typical crude oil, most of these compounds concentrate in the range between methylcyclohexane (R = CH₃), a hydrocarbon with seven carbon atoms, and pentadecylcyclohexane (R = C₁₅H₃₁), a hydrocarbon with twenty one carbon atoms – the range that spans light and middle distillate fuels of interest.

A systematic GC/FID and GC/MS analysis of a wide variety of commercial and military fuels demonstrated that in addition to a well-known alkane pattern, cyclohexane homologous series also exhibits a characteristic distribution pattern for each fuel type. Examples of alkane and alkylcyclohexane patterns for three common fuels are provided in Figure 3. This figure shows a relative content of each member of the series as measured utilizing GC/MS technique. For gasoline fuel, the distribution exhibits an asymmetric rapidly decreasing pattern from methylcyclohexane to heptylcyclohexane. The jet propulsion fuel Jet-A is characterized by a distribution pattern in the range from methylcyclohexane to decylcyclohexane with the maximum at butylcyclohexane. Diesel fuel exhibits alkylcyclohexane pattern from methylcyclohexane to tridecylcyclohexane with maximum at pentylcyclohexane.

The range of hydrocarbons in each petroleum product is determined by its boiling range, whereas the internal distribution pattern (composition) reflects its application-specific formulation. Since the composition of modern fuels is controlled by stringent manufacturing specifications, the range and distribution pattern of each fuel should vary only slightly. The variations are generally related to the crude oil feedstock and the refining practices used in fuel manufacture.

Fuel Identification in Weathered Samples

The examples shown demonstrate that alkylcyclohexanes distribution patterns are as fuel-specific as the alkane distributions. The main advantage of utilizing alkylcyclohexane patterns is that naphthenes are more resistant to environmental alteration and could be detected even when most of the alkanes are degraded. In addition, both the alkane and alkylcyclohexane distribution patterns (as well as other fuel-specific compounds) can be obtained from the single sample analysis by GC/MS. Two case studies illustrate this application.

Case 1. Analysis of a free product sample reveals that it has undergone a fair degree of weathering in the subsurface environment. This is evident in the pattern revealed in the GC/MS extracted ion profile (Figure 4). The largest peaks in this plot are the more recalcitrant isoalkanes of which the two most abundant, pristane and phytane, are identified. The distribution of the hydrocarbons in this plot suggests the presence of a diesel-range product, most closely resembling diesel fuel. Of concern is the possibility that the sample contains another middle distillate product. This is not immediately apparent because of the depletion of the n-alkanes. However, evaluation of the alkylcyclohexane pattern in the figure confirms that the contaminant entirely represented diesel fuel. This conclusion is consistent with other site-specific chemical and historic data.

Case 2. Multiple free product samples collected from different location of the bulk petroleum storage site are evaluated in order to determine their relationship. The alkane distribution patterns of the two representative samples are shown in Figure 5. The plot indicates that Sample A has lost nearly all n-alkanes and would thus appear to represent a severely weathered product. This is in contrast to Sample B where the high abundance of n-alkanes attests to its relatively unaltered nature. This weathering difference confounds the evaluation of the source relationship. However, a comparison of the alkylcyclohexane distributions (Figure 6) clearly indicates that both samples represent the same middle distillate product. Combined with the other site-specific information, this evaluation allows the source of the site contamination to be established. Further, the substantial differences in the degree of weathering and the relative location of the sampling points suggests that multiple releases of the same product have impacted the subject site.

Summary

The systematic evaluation of chemical composition of different hydrocarbon fuels revealed that cyclohexane homologous series exhibit fuel-specific distribution patterns, which can be used for fuel-type evaluation. The major advantage of this approach is that even for moderately weathered environmental samples, when most of n-alkanes are depleted, the alkylcyclohexane distribution patterns provide valuable fingerprinting information. Application of this methodology has provided critical evidence in resolving numerous legal disputes related to the source of contaminants and possible responsible parties.

Yakov Galperin, Ph.D., a Manager, and Henry Camp, a Principal, are members of Arthur D. Little’s Environmental Chemistry and Forensics group in Cambridge, Massachusetts.

Top