3 Summary and conclusions

Sammensætning af olie og benzin

The purpose of this project was to develop a method that enabled to perform risk assessment of fuel contamination in accordance with the guidelines from the Danish Environmental Protection Agency.

This purpose is fulfilled by:

A collection of the knowledge regarding the content of compounds in the different petroleum products (old lead containing gasoline, 92, 95, and 98 octane lead free gasoline, diesel fuel, and light fuel oil).
A collection of the knowledge of the analysis of total petroleum hydrocarbons in water and soil.
A creation of profiles for each of the petroleum products.
A selection of compounds that represent the groups in the profiles when performing risk assessment calculations.
An assessment of whether petroleum products contain organic compounds, which are not included in the standard analysis that possesses a risk for people.

Even though a lot of information regarding the content of individual compounds in gasoline, diesel fuel and light fuel oil was collected, most of it originated from USA, Canada, and Sweden, and information from Denmark was rare. The oil companies in Denmark control their production based on chemical group parameters or physical parameters and consequently they do not obtain information regarding individual compounds. Thus the profiles developed in this project for use in Denmark are based on the composition of petroleum products in foreign countries. It remains uncertain whether the Danish petroleum products differ considerably from the American products on which the main part of the data has been obtained. This is not suspected to have major influence on the profiles though, since a comparison of e.g. American and Swedish products for the components possible does not differ substantially. The amounts of data that have been obtained are sufficient for creation of reliable profiles for the individual types of gasoline whereas it is not sufficient for diesel fuel and light fuel oil. However profiles have been constructed for diesel fuel and light fuel oil, but they are more uncertain than the profiles for the different types of gasoline.

The profiles are constructed by dividing the compounds of the petroleum products in 7 groups by their boiling points. This grouping is done in the same way as when samples are analysed for total petroleum hydrocarbons by gas chromatography. Five of the groups are further subdivided into parafines and olefines (group A) or aromatics and NSO-compounds (group B). The profiles are based on the amount of organic compounds in each of the groups.

An indicator compound is chosen for each group, choosing a compound with little adsorption, small K_OW (relatively in the group) and high vaporisation (high Henry’s constant). The biodegradability of the compounds has not been taken into account and neither has the toxicity. It has been important that the compound was typical for the fraction it described. The indicator compounds have been chosen to be similar for all types of oil products, even though their relative abundance varies between products.

When risk assessments are performed by means of JAGG, each group is treated separately, represented by the chosen model compound. The contributions from each group are added and compared to the different quality criteria for total petroleum hydrocarbons in air or groundwater. The profiles are constructed for 3 different situations:

Measurement of total petroleum hydrocarbon concentration.
Measurement of total petroleum hydrocarbon concentration divided into 3 fractions (C6-C10, C10-C25, and C25-C35).
Measurement of BTEX (benzene, toluene, ethyl-benzene, and xylenes) in addition to the total petroleum hydrocarbon concentration divided into 3 fractions (C6-C10, C10-C25, and C25-C35).

When analysing for total petroleum hydrocarbons a cut off at C6 and C35 is often applied in Denmark. This lower limit (C5) is due to pentane (C5) being used most often for the extraction of the hydrocarbons from water and soil, and the upper limit (C35) is due to the fact that petroleum products rather seldom contain hydrocarbons with boiling points higher than C35. This report reveals that the hydrocarbons with boiling point less than C6 and thus not included in the analysis constitute 28-34 % (weight/weight) of the total amount of compounds in gasoline. It is not recommended to change the analytical technique but it is important to notice that a large amount of the hydrocarbons in gasoline is not included in the analysis. However, they have been taken into consideration in the constructed profiles by including a special group for these compounds. A very detailed evaluation of each compound belonging to this group is required in order to assess, whether the compounds are so volatile that they only constitute a problem at the service station itself, or whether they also will be part of a contamination. In this group the only compound that is known to be carcinogenic is 1,3-butadiene. However the list of compounds will never be complete.

The cut off at C35 is not that important quantitatively because these compounds only constitute a minor part (in gasoline 0.00013-0.00044 % (weight/weight), in diesel fuel 0.0012 % (weight/weight), and in light fuel oil 0.0014 % (weight/weight) in relation to the total amounts of hydrocarbons found in the fuel products).

The study of the petroleum products revealed that group 1 (less than C6) consisted exclusively of parafines and olefines while only aromatics and NSO-compounds were found in group 6 (C20-C25) and group 7 (above C25). Whether this is a coincidence or due to the fact that gasoline/diesel fuel/light fuel never contains parafines and olefines with boiling points higher than C20 for example is still uncertain.

This report revealed that among the hydrocarbons in the petroleum products 1,3-butadiene was found in gasoline at a concentration of approximately 0.01 % (weight/weight). This compound is very volatile and carcinogenic and, as mentioned previously, too volatile to be included in the normal analytical techniques. However, the compound is probably too volatile to constitute a risk in relation to groundwater or soil contamination.

The method developed in this project to perform risk assessment based on the analysis of total petroleum hydrocarbons at sites contaminated with gasoline, diesel fuel, or light fuel oil has been applied in 2 examples. In one of the examples the potential of a groundwater contamination from a diesel fuel contaminated site is estimated, and in the other the potential of an air contamination from a gasoline contaminated site is estimated.

The results of these risk assessments have been compared to the measurements from the two examples of gasoline contamination. There was a good accordance between the measured and the estimated values when estimating groundwater contamination, while this was not true for the estimation of indoor climate concentrations, which was greatly overestimated. This is thought to be caused by the fact that the calculations are based on the composition of a fresh oil product. It is thus not taken into account that vaporisation of volatile compounds soon after the spill will reduce the presence of these compounds rapidly. Furthermore, the calculation method does not address that the group of most volatile components, if present, will reduce the actual fractions of the less volatile groups in the pore air. Theoretically, it is possible to include this, but since the presence and variation of the group of most volatile components (group 1) is not well documented in relation to soil contamination it is difficult to estimate this properly without further investigations.

It can in general be assumed that the method is quite conservative when calculating indoor climate concentrations. Improvements can be made by measuring relevant parameters on the specific site as suggested in chapter 16. In this connection, it should be remembered that the most volatile components normally are not included in the measured concentrations in pore air either, which will influence the validity if the comparison of calculated and measured concentrations.