Estimating the Date of a Hydrocarbon Fuel Release

Determining the time of a fuel or crude oil release is site dependent and requires consideration of a wide set of environmental parameters. Among these parameters are:

  • fuel composition
  • selected soil characteristics
  • site hydrology
  • temperature
  • moisture content
  • permeability
  • oxygen and nutrient availability

In some environments, these parameters will favor fuel preservation. In others, local environmental conditions will strongly alter the composition of the fuel and cause it to degrade completely in a short period of time (Atlas, 1981). Several approaches can be taken to determine the approximate time of a release.

6.1 Composition at the Time of Production

Composition of a fuel depends on when it was produced. The first method is to use the knowledge of when composition of a fuel, including it’s additives, was changed, as a markers to identify the time of production and release. Certain formulations were only used during restricted time periods.

Benzene and sulfur. An example is the demand for reformulation of fuels, which has been imposed in the USA during the early 1990's. For example, new regulations imposed by the California Air Resource Board (CARB), reduced the most volatile hydrocarbon components of gasoline, especially benzene (which was decreased to ~0.5% of total gasoline weight), and lowered the sulfur content of diesel No. 2 fuel to 500 ppmw.

Lead alkyls. Among the gasoline additives, lead alkyls and lead scavengers are among the most important for fingerprinting purposes and as time markers. Leaded gasoline was first marketed in 1923, and until 1960 tetraethyl lead (TEL) was used as the only antiknock agent. Since 1960, when Chevron (then Standard Oil Co.) introduced another antiknock agent, tetramethyl lead (TML), different combinations of these two additives, as well as redistribution reaction mixtures of TEL and TML were used (Gibbs, 1990). A typical commercial redistribution reaction mixture resulting from the use of equimolar amounts of TML and TEL consists of 3.8% TML, 23.4% TMEL, 42.4% DMDEL, 25.6% MTEL and 4.8% TEL.

In addition, the total amount of lead added to gasoline as lead alkyls has changed with time. The recommended maximum content of lead at 3.17 grams of lead per U.S. gallon of gasoline (g Pb/gal) was introduced in 1926. The maximum permitted level peaked in 1959 at 4.23g Pb/gal. Due to Government regulations on the use of lead, maximum levels gradually decreased to 1.0g Pb/gal in 1980, 0.5 g Pb/gal in 1985 and 0.1 g Pb/gal in 1988 (Gibbs, 1993).

By the end of 1992 California completely eliminated the manufacture of leaded gasoline, whereas other states had already done so in previous years. However, some states still have a waiver on the use of leaded gasoline. A plot of the average historic concentration of lead in U.S. gasolines compiled by EPA is shown in Figure 18 (Caldwell, 1994). During the phase-out of lead additives, TEL became the most abundant lead-alkyl additive with decreasing amounts of TML usage until about 1985-1987.

Fig. 18. Historical trend in lead usage

Lead scavengers. To reduce adverse effects of lead oxide formation in the engine after fuel combustion, lead scavengers ethylene dibromide (EDB) and ethylene dichloride (EDC, also commonly known as 1, 2-DCA) were introduced in 1928 (Nickerson, 1954). The ratio between the two has changed over the years. A typical motor mix for automotive gasoline additives in the 1980's consists of about 62% TEL (or a redistribution reaction mixture of lead alkyls), 18% EDB, 18% EDC, and 2% of other inactive ingredients, such as dye, antioxidants, petroleum solvent and stability improvers.

For aviation piston engines, TEL is still used as an antiknock additive and the scavenger consists entirely of EDB. The chronology provided allows us to relate composition and concentration of a lead additive in a gasoline NAPL to the time-period of its manufacture. Because EDB and EDC are moderately soluble, they are dissolved out of the product and transferred into the groundwater. The presence of EDB and especially EDC in a monitoring well at the site is an indication that leaded gasoline has been released at or near the site. Lead alkyls are strongly adsorbed to soil and hydrolyzed by water. Hence, in the absence of NAPL, the presence of lead scavengers may be the only proof for the release of a leaded gasoline at the site.

Oxygenates. With Government-mandated phase-out of lead additives, oxygenate compounds such as ethers and alcohols have been increasingly blended with gasolines to maintain high octane rating and to reduce vehicle emissions of carbon monoxide. The most common oxygenate, methyl tertiary-butyl ether (MTBE), has been blended with gasoline since its first commercial manufacture in 1979. Its documented use on the East Coast of the USA was in 1982 and in California in the late 1980's (Squillace, 1996). The rapid growth of MTBE usage was further accelerated by the Clean Air Amendments of 1990.

Modern reformulated gasoline in the USA contains as much as 11% by volume of MTBE. Methyl-, ethyl- and tertiary-butyl alcohol as well as ethyl tertiary-butyl ether (ETBE), tertiary-amyl methyl ether (TAME) and diisopropylether (DIPE) have all been blended with gasoline by different refineries. Some states, such as Alaska or Washington, either exclusively or primarily use alcohols as oxygenate blending agents. These oxygenates are much more soluble than hydrocarbons and will usually be present at the leading edge of a dissolved gasoline plume.

6.2 Rate of Change in the Environment of Certain Constituents after Release

Weathering. Another approach to resolve time of release of a fuel or crude oil, is to determine the rate of change of certain constituents in different hydrocarbon classes. For light hydrocarbon fuels such as gasoline, evaporation and dissolution are the most important weathering processes. Several parameters control the rate of fuel volatilization, the most important of which is the vapor pressure of its hydrocarbon constituents. According to empirical vapor pressure data (Lyman et al, 1992), it may require approximately one month to remove n-butane by volatilization from a silty soil saturated with gasoline from a depth of 12 feet below soil surface, whereas n-octane may still be detected in the soil after more than 10 years.

Dissolution or water washing rates reflect the substantial differences in aqueous water solubility of gasoline hydrocarbons (McAuliffe, 1966; Sutton and Calder, 1975). The process alters the composition of a fuel by preferential removal of the more soluble components (e.g. benzene: 1780 mg/L, toluene: 515 mg/L, ethylbenzene: 152 mg/L, o-xylene: 170 mg/L, m-xylene: 146 mg/L and p-xylene: 56 mg/L), which can be measured by monitoring changes in concentration ratios of individual gasoline hydrocarbons, e.g. BTEX (Odermatt, 1994; Yang et al., 1995).

Site hydrology. This can also provide additional means for estimating time of a release. When a hydrocarbon fuel is dissolved in groundwater it becomes part of a continuous, mobile phase. Mobility of different fuel hydrocarbons in this phase will reflect their relative water-soil partitioning characteristics, which are conventionally measured in terms of solute transport retardation (Fetter, 1992). Based on field studies conducted in unconfined, sandy aquifers, benzene was estimated to be transported at about 90% of the groundwater velocity, toluene at 75%, ethylbenzene and xylene isomers at about 67%, while MTBE migrated with groundwater flow without any retardation (API, 1994).

Retardation effects are found to be closely related to organic carbon content and physical properties of the soil. By monitoring changes in selected hydrocarbon and oxygenate concentrations downgradient from the source of a fuel release and using available fuel component mobility data and contaminant transport modeling (Walton, 1988), it is often possible to estimate the time of a release.

6.3 The (B+T)/(E+X) Ratio as a Time Indicator for Gasoline Release

Due to the difference in mobility, concentration ratios of the individual BTEX compounds (e.g., B/E, T/X, etc.) in gasoline-contaminated groundwater change uniformly with time. However, because relative content of BTEX compounds in manufactured gasoline has varied historically, initial ratio values are often unavailable, which limits application of these ratios as time indicators.

From numerous site studies, it has been determined that the cumulative BTEX ratio,

R b = (B+T)/(E+X)

is a more useful time estimate parameter. Use of this ratio minimizes uncertainties related to historical variations in BTEX composition of manufactured gasoline and provides a valuable tool for monitoring environmental changes in a dissolved gasoline plume with time (Kaplan et.al, 1996).

Multiple case investigations and laboratory BTEX partitioning studies performed by Kaplan et.al demonstrate that near the source and immediately after a gasoline spill, R b reaches values between 1.5 and 6, depending on the amount of gasoline in contact with groundwater. When R b measured in a groundwater monitoring well near the source falls in this range, it indicates that a recent release (typically less than 5 years) has occurred.

In the absence of NAPL, R b in a dissolved gasoline plume is close to that in the original gasoline (0.8 - 1.1). The ratio then decreases as a function of time, and values below 0.5 usually indicate gasoline residence time longer than 10 years (Kaplan et al., 1996). The accuracy of time estimates may be improved by using a best fit regression on historical site data collected over an extended time interval.

Based on the modified solution of the advective-dispersion equation for solute transport in a homogeneous isotropic media (Walton, 1988), reduction of R b with time near the source of an instantaneous gasoline release may be described by an exponential function. An exponential approximation of changes in R b with time also seems reasonable for a dissolved hydrocarbon plume near a floating gasoline layer (Grifoll and Cohen, 1996). An example of an exponential approximation of the BTEX monitored data from an actual case study is shown in Figure 19.

Fig. 19. Time derivation based on historic BTEX data and exponentiol approximation. R b = 6.0 exp(-0.308*T)

The high R b value at the time of initial measurement is indicative of a NAPL, and is validated by a report of that 1 m thick free product layer was observed in the monitoring well shortly after the accidental gasoline release which occurred approximately 4 years prior to the initial measurement. In this particular example, extrapolation of the curve shows that at T=0 (the time of gasoline release), R b = 6, confirming the presence of a thick free product layer. A twofold decrease of R b occurs in 2.3 years.

6.4 The n-C17/pristane Ratio as Time Indicator for Diesel No. 2 Biodegradation

Biodegradation is the major weathering process for middle distillates and residual products released into the environment. Most commonly used parameters for measuring degree of biodegradation are the ratios n-C 17/Pr or n-C 18/Ph (Glazer, 1991).

Following the examination of several northern European sites containing diesel fuel No. 2 in soil, where time of fuel releases had been documented, Christensen and Larsen (1993) concluded that it was possible to estimate the length of time the diesel fuel had been in the environment by using the numerical values of the n-C 17/Pr ratios. They obtained a linear relationship extending over a range of about 20 years, which can be approximated by the following equation:

In this study the average initial n-C17/Pr ratio in dispensed diesel fuel was determined by the authors as 2.3, which is similar to ratios for this parameter measured in USA diesel fuel No. 2. The environmental constraints on obtaining realistic estimates of release time are discussed below. Despite the constraints described, the above equation is very useful in establishing the approximate residence time for a mid-range distillate in a soil profile.

6.5 Constraints on Estimating Release Time

The problems associated with determining time of a hydrocarbon fuel release are compounded by a lack of reliable analytical data for sites with documented release dates. Described below are "extreme" situations where estimation of a fuel release time is very difficult because of the soil composition and hydrologic controls.

Bulk Liquid Phase Fuel Saturated Soil. Bulk liquid hydrocarbons which coat soil particles are much less subject to environmental alteration, including biodegradation, than hydrocarbons at lower concentration uniformly dispersed in soil or dissolved in water. This is because physical alteration processes mainly affect the interface and not the body of the bulk product. Biodegradation inside the body of a free product is extremely slow, because of the limitation of oxygen, water and nutrients (Hoad and Marley, 1986). It is likely that a hydrocarbon fuel in such conservative environments can survive for a very long time (over 20 years in some cases) with only minor changes in chemical composition.

Free Floating Product.When a thick layer (greater than 15 cm) of a free phase floating product develops on the groundwater surface, its rate of alteration is slower than for a thin layer. The reason for this, is that the hydrocarbon fluid is almost immiscible with the groundwater and alteration occurs only at the interface of the fluid and the water below, or the fluid and the moist soil above the product. At the center of the free product layer, interactions with the ambient environment is likely to be exceedingly slow, unless there is a process of turbulent mixing (Lyman et al, 1992). Hence, if the free product fluid is rapidly transported to the water table or to an impermeable clay layer (aquatard) and does not undergo substantial alteration during its vertical migration in the soil, it may remain in a relatively unaltered state for a period of time as long as several decades, depending on the thickness of the fluid and the dynamics of the hydrologic system. The most likely alterations to occur will be evaporation of the most volatile hydrocarbons and dissolution of the most soluble components such as oxygenates, monoaromatic compounds and lead scavengers. Hence, a careful sampling technique must be employed for a relatively thick NAPL plume to estimate an age of release based on its chemical composition.

Soil Structure. Fine-grained sediments are loosely described as clays, which frequently have very low porosity and permeability to fuels, water or nutrients. Rates of hydrocarbon alteration in clay-rich soils have been shown to be very slow. In contrast, the presence of highly aerated, wet and nutrient-rich environments in silty-sandy soil favors rapid degradation of a hydrocarbon fuel, which is the conceptual basis for bioremediation technology (Riser-Roberts, 1992). Such conditions promote rapid changes in the composition of a fuel, which in some cases can result in almost complete hydrocarbon removal within a few months.

Next: Appendix

Interested in Environmental Services?

We'd love to tell you more!

Contact Us Today