Introduction

The use of GC-IRMS as an exploration tool has developed rapidly since it was first introduced in the late 80's. Today GC-IRMS is used in the early exploration phase before drilling, i.e. surface geochemistry, and during the drilling phase, e.g. monitoring of the stable carbon isotope composition of headspace gases from cutting samples and/or test samples, through to analysis of condensates/oils and source rock extracts for correlation purposes.

Surface geochemistry

Measurement of the isotope composition of methane has traditionally been one of the most used techniques to determine whether the methane present in surface samples was of biogenic or thermogenic origin. Measurements of d¹³C on methane alone could, however, be quite problematic since slight mixing of biogenic methane with petrogenic methane can distort the data to give a conclusion of biogenic origin. In the same manner, oxidation of biogenic methane can change isotope values towards those normally found for thermogenic methane. The introduction of the GC-IRMS with analysis of C₁ - C₄ in any of the gas fractions from surface geochemical samples, i.e. headspace gas, occluded gas or adsorbed gas, will reduce the possibilities of misinterpretation drastically, since a biogenic input will only affect the methane isotope composition. The GC-IRMS technique has been used on more than 10 offshore surface geochemical surveys during the last 4 years. Examples from the different surveys will be discussed in detail and will clearly show how GC-IRMS data have helped to distinguish between biogenic and petrogenic sources for the gases found in the samples.

Gas profile in a well

Determination of the gas profile in a well, i.e. measurement of the amount and type of gaseous hydrocarbons in the mudstream, on well-site by mudloggers and in geochemical laboratories has been standard practice for decades. Lately there has been a significant interest in determining wheher the cutting gas is of petrogenic or biogenic origin. Traditionally carbon analysis of the methane in the cutting gas was used, but with the surface geochemical samples, slight mixing with biogenic gas often resulted in the wrong conclusion. Lately the GC-IRMS technique has been used for analysis of the C₁ - C₄ hydrocarbons to give a better understanding of the origin of the cutting gas. The carbon isotope data of the C₁ - C₄ hydrocarbons, can be conclusive in determining where there is pure petrogenic gas in the cuttings.

It is also simple to determine where there has been mixing with biogenic gas and the extent of such mixing A number of examples from the Norwegian Continental Shelf of the use of this technique in well analysis will be presented.

Condensates/oil correlation and correlation to source rock extracts

GC-IRMS analysis of n alkanes, iso alkanes, cyclic alkanes and aromatics was introduced eight years ago, and was used for correlation of various samples. It has also been used to evaluate the origin of oils/condensates, including distinction of organic facies of the source rock generating the oils, maturity of oils etc. The technique has also been used on solvent extracts of source rocks. The problem with using data from GC-IRMS analysis of solvent extracts for correlation purposes is that a large proportion of the useful isotopic information is in the C₁₅- fraction which is lost during sample work-up. This problem has been overcome with the introduction of the thermal extraction GC-IRMS instrument, which allows direct comparison of the isotopic composition of the lighter components in source rocks as well as oils. The introduction of MSSV-GC-IRMS and also to a certain degree, open vessel pyrolysis-GC-IRMS has made it possible to artificially mature source rocks and analyse the products. These techniques also allow direct comparison of the GC-IRMS data on the complete range of hydrocarbons in source rocks as well as oils/condensates. A number of cases from Russia and the North Sea where these techniques have been used in oil-oil and oil-source rock correlation will be shown. In many instances, for example with condensates, where the biomarkers normally used for correlation purposes are not present, this technique was the only technique giving useful correlation data.