Comparaison de diverses méthodes de dosage des argiles d'un sable de gisement. Dosage des argiles
Comparison of Different Methods of Determining Clays in a Reservoir Sand. Quantitative Analysis of Clays
2 Université Toulouse III
Les argiles d'un sable de gisement, concentrées dans la fraction de diamètre Phi < 40µm, ont été dosées par trois procédés différents : - un ensemble de méthodes dites classiquesretenues par Dejou et al (1977) reposant sur les analyses chimiques et thermopondérales; - une méthode mise au point au laboratoire qui donne la teneur globale en argiles après déduction des proportions de majeurs non argileux : le quartz étant dosé par diffraction X et la jarosite calculée à partir de la teneur en soufre; - une méthode de traitement statistique des analyses chimiques données par la microsonde électronique. La teneur en argiles obtenue par les méthodes classiques (2,53%) est surestimée essentiellement à cause des approximations faites sur les données cristallochimiques de ces minéraux. La méthode proposée (2,08%) et l'analyse statistique (2,04%) donnent des résultats très comparables. La première par sa rapidité et sa simplicité se prête commodément à l'étude de séries d'échantillons; la deuxième offre l'avantage de quantifier l'ensemble des minéraux présents dans la fraction.
Oil, gas and geothermal reservoirs all contain clayey fractions no matter how small they may be. This has been blamed whenever operating or producing problems arise. It may be revealed by phenomena of mechanical resistance, permeability or interfacial properties (ion exchange, adsorption, etc. ). Tests to understand such phenomena then go via the quantitative mineralogical analysis of the clays present. This analysis must also be looked at in terms of methods. It is subjected to constraints of cost, instrumentation, competence or deadlines. This article proposes:(a) A so-called conventional route (Dejou et al, 1977) based on chemical and weighted analyses. (b) An overall assessment method of the clay phase by difference (determination of two nonclay species). (c) A method based on the statistical processing of microanalytic data obtained by an electronic microprobe. The material examined was a quartzose arenite made up mainly of quartz, jarosite, orthoclase, plagioclases, calcite, dolomite, muscovite, kaolinite, illite, montmorillonite, interstratified illite-montmorillionite, iron oxyhydroxides and accessory minerals such as rutile, zircon, garnet, tourmaline and hydroxylapatite. The arenite was subjected to an ultrasonic treatment (Letelier, 1986) to recover pellicular or weakly cemented clays. After this treatment, all the free clays were found in the < 40 m fraction which were used for the measurements. The so-called conventionalmethod is based on the associating of multiple techniques that are normally used for analyzing clays. They include X-ray diffraction, TDA, TGA, selective dissolution, CEC, adsorption of various reagents and gravimetric separations. They have been reviewed by Dejou et al (1977). The results they give depend on the grain size, chrystallochemistry, presence of amorphous elements and especially the typical chemical compositions assigned to the minerals present. The clay content of arenite obtained by this method was 2. 53%. A second method consisted in determining the overall proportion of clays by deducing the quartz and jarosite contents, which alone can be quantitatively retained. Jarosite was determined from the sulfur content. Quartz was determined by X rays according to the generalized addition method of Alegre and Debray (1966), which consists in measuring the intensities of the phase to be determined and the same pure phase. The clay content of arenite thus obtained was 2. 08%. The third method was based on the statistical processing of microprobe data. Mineralogical and principal-components analyses were used to define the actual mineralogical phases or the ones assumed to exist. Calculating was based on the electroneutrality of the minerals and of the completion of the material balances. The associating of statistical analysis (CPA) with the calculation of standards then enables these phases to be determined from microanalytic data. The clay content of arenite found here was 2. 04%. Thus the results obtained by conventional methods were greatly overestimated. The errors were due mainly to the approximations made for the crystallochemical data of clays. These data make use of the statistical processing of analysis performed by the microprobe and have the advantage of quantifying the relative proportions of each of the clay phases. Lastly, the method of determining by difference, because of speed and simplicity of implementation, lends itself very well to analyzing series of samples.
© IFP, 1990