Miljøprojekt, 905, Teknologiudviklingsprogrammet for jord- og grundvandsforurening – Vurdering af metoder til analyse af olie i jord

Evaluation of Methods to Analyse Mineral Oil in Soil

Summary and conclusions

This report presents the testing and optimisation of a new method for determination of the content of mineral oil in soil, and a method to calculate the content of hydrocarbons in soil samples. The report is intended as a guideline for evaluating the extent of mineral oil contamination in soil samples.

The report is divided into two sections; one concerning the testing, optimisation and validation of the ISO/DIS 16703 method "Soil Quality – Determination of Mineral Oil Content by Gas Chromatography", Draft International Standard, and the second section concerning a method to determine whether the content of hydrocarbons originates from mineral oil or from natural organic matter.

Part 1:

The method ISO/DIS 16703 is tested together with the most commonly used methods to determine the oil content in soil, for example extraction with pentane and 0.05M Sodium pyrophosphate /2/, which is at present the Danish reference method.

The ISO/DIS 16703 has been optimised during the development and testing, so that the method is improved with respect to Danish soil quality requirements.

This new method requires a greatly enhanced performance for the gas chromatographs, compared to the previous reference method. Therefore, it has been necessary to optimise the performance and analytical method of the gas chromatograph. This report shows an example of how to set up a method for the gas chromatograph for split-less injection.

ISO/DIS 16703 uses a mixture of acetone and n-heptane for extraction. In this report it is concluded that the ISO/DIS 16703 method is the most efficient method for extraction of hydrocarbons from soil. It is shown that the ½ hour extraction time, which is used in ISO/DIS 16703 is not sufficient for complete extraction. Equilibrium of the extraction is not reached until 12 - 16 hours, and in this report, this is recommended as the minimum time of extraction. Heptane is replaced by n-pentane, which gives the opportunity to quantify benzene, toluene, and the xylenes from the same extract as used for determination of mineral oil in soil.

The quantification method in ISO/DIS 16703 is performed with a calibration mixture of two distinct mineral oil products. This method of quantification has been shown to give relatively large interlaboratory deviations /3/, but when using specific n-alkane mixtures, the precision is increased. It is shown that n-alkanes can be used for calibration and calculation of the content of oil in soil samples.

It was the project's intention to quantify the mineral oil by GC-MS, but this procedure was not successful.

In the ISO/DIS 16703, an analytical step with washing of the extract and purification of the extract with florisil is recommended. It is shown in this report that part of both the mineral oil and the natural hydrocarbons is lost during these analytical steps. It is therefore recommended not to perform any cleaning of the extract before analysis by gas chromatograph. In special cases with problematic samples the cleaning can be performed, and this is described in the report.

Furthermore, the method is validated;:Total Hydrocarbons (Benzen – C10, >C10 – C25, >C25 – C35, >C35 – C40), BTEX and PAH (16 EPA-PAH), and the analytical procedure is described in section 10 and Appendix A.

Part 2:

This part of the report concerns biomarkers in mineral oil and in natural organic matter (hydrocarbons).

A selected group of hydrocarbons, which are easy to identify, and which are present in mineral oil, were chosen, and the content is quantified in some mineral oils.

Some indexes, which can be used in evaluating the origin of the oil are mentioned. CPI-index, isoprenoid content and unresolved complex matter are calculated for a number of samples, and a method for evaluating the results is included.

Special PAH's: naphthalenes and phenanthrenes, are investigated, and some of these can be used as biomarkers in the evaluation of the origin of the hydrocarbons.

Certain compounds, which are found in the higher boiling range of mineral oil products, the hopanes and stearanes, are mentioned in this report. These are true indicators for hydrocarbons of petrogen origin. Examples of the pattern of hopanes by GC-FID and GC-MS techniques are shown.

These compounds are very useful as indicators to determine if the content of hydrocarbons is derived from mineral oil or is of natural origin, especially for the fraction of n-alkanes C25-C40.

To map the content of natural hydrocarbons that may be present in "clean" soil samples, and to determine the extent to which these are degraded, samples were taken from selected clean areas, where the content of hydrocarbons from natural sources is expected to be high, for example in woods etc. The samples were collected in a depth profile to map the changes due to degradation of the natural hydrocarbons. Samples were also taken in a smaller fraction (scrape) of about 2 cm. These samples were all analysed by GC-MS in scan-mode, and every peak was, if possible, identified by the mass spectrum and boiling point range. A few selected reference compounds were also analysed.

Some general features were seen at the "clean" locations. A relatively large amount of phytosterols, n-alkanes and fatty acids were detected. Phytosterols are mainly plant specific compounds, but some phytosterols are seen in all the analysed samples, for example, Sitosterol (beta- or gamma-). This compound can be used as a biogenic (natural hydrocarbon) biomarker in soil samples.

Based on experience with these samples a GC-MS method was developed for fast quantification of the odd n-alkanes and phytosterols, and the calculated amount can be deducted from the total hydrocarbons found by the GC-FID method.

As a separate chapter, the Swedish reference method for analysis of hydrocarbons in soil is described. This standard method uses GC-MS for quantification of hydrocarbons. It is very different from the Danish method of quantification of hydrocarbons, and the authorities also have different requirements. The Swedish method has a patented method of deduction of the phytosterols from the total hydrocarbon content in the soil samples.

The experience achieved during this project has lead to the development of a guideline by which the content of hydrocarbons can be evaluated in soil samples.

Finally, 49 samples were analysed by the new method, the content of hydrocarbons was calculated and evaluated according to the guideline.