Abstract

The development of a procedure, based on the application of synchronous luminescence spectrometry to the problem of differentiating crude oil and oil base drilling mud, is described. Its performance is illustrated by reference to field data from serveral North Sea wells. This ‘fingerprinting’ tool effectively simplifies interpretation of formation samples which are otherwise confused by the presence of oil mud filtrate.

Introduction

The problem of detecting crude oil in borehole rock samples at the well site has been approached by visual inspection of fluorescence in drilled cuttings under ultra-violet (UV) light. Crude oils fluoresce because they have aromatic components in their chemical composition. However, oil base muds, which are becoming increasingly important in exploration drilling, also contain aromatics. Interpretation of formation samples drilled with oil mud can, therefore, be difficult using these traditional method.

Hydrocarbon mixtures may be characterised by their n-alkane distribution using solvent extraction/gas chromatography. However, this has little value at the well site because sample preparation takes a considerable time if high preparation takes a considerable time if high molecular weight alkanes are to be extracted. A technique not previously used in this context is the synchronous excitation of emission spectra, which, by comparison with conventional fluorescence spectrometry, generates spectra with greatly improved resolution. Instead of scanning a large emission wavelength range at a fixed excitation wavelength, as in conventional techniques, the excitation and emission wavelengths are scanned simultaneously with a small fixed wavelength separation between them (typically in the range 10 – 40 nanometers (nm)). This has the effect, at any given instant, of exciting only those aromatic molecules which fluoresce in this reduced (e.g. 10 – 40 nm) part of the spectrum. But by scanning an extensive wavelength range (e.g. 250 – 500 nm) the entire aromatic population of the mixture is sequentially excited. The diagnostic power of chromatography can, therefore, be achieved more rapidly and simply by synchronous luminescence spectrometry. The present account describes a procedure, developed in the field in a number of procedure, developed in the field in a number of North Sea wells, which applies the synchronous technique to the problem of well site evaluation of borehole samples.

MATERIALS and METHODS

There are several commercially available spectromelers capable of monochromating and simultaneously scanning the excitation and emission wavelengths in the UV-visible range. The spectrometer may be part of a computer controlled equipment suite which part of a computer controlled equipment suite which not only allows spectra to be displayed, stored on disc and either printed or plotted, but also presents opportunities for extensive data processing. The results presented here were obtained using the Perkin-Elmer LS5 Luminescence Spectrometer, 3600 Perkin-Elmer LS5 Luminescence Spectrometer, 3600 Data Station, and 660 Printer.

Sample preparation is simple and rapid. Solids are washed and crushed and a standard fluid extract prepared in an appropriate solvent. A number of prepared in an appropriate solvent. A number of fluorescence grade solvents are suitable but the choice of a non-flammable reagent (e.g. dichloromethane) is considered an advantage for well site use. All fluid samples (i.e. rock matrix extracts, mud, crude oil etc.) are diluted to prevent self quenching and self absorption but must be sufficiently concentrated to avoid impairing the signal/noise ratio of the fluorescence response. Optimum concentrations are inversely related to aromatic content of the sample and range from less than for crude oils to 10 for most oil muds and about 100 /1 for base oils with a low aromatic content.

When using a computer it is possible to rapidly subtract solvent fluorescence from each sample (so that all spectra are net of solvent effects), to plot the spectra on a scale expanded to a standard maximum and to superimpose spectra for qualitative comparison.

p. 391

This content is only available via PDF.
You can access this article if you purchase or spend a download.