Theoretical and Experimental Studies of Sandstone Acidizing

The matrix acidization of sandstone by a hydrochloric/hydrofluoric acid mixture is described through use of a capillary model. The model was solved first in linear coordinates so that it could be compared with the results of coreflood experiments performed on Berea sandstone. The model predictions showed reasonable agreement with the experimental data and yielded specific information about the reaction characteristics of the sandstone/HCl/HF system. The acidization model then was applied in radial coordinates to generate design curves for a matrix acidization treatment. While these curves strictly apply only to those sandstones having similar mineral compositions, the approach is general. It is based on matching the location of the HF reaction front to the depth of a damaged zone. This method introduces the concept of an optimum injection rate and, in this regard, differs from other design methods reported in the literature. Introduction The matrix acidization of sandstone by an HCl/HF acid mixture is an often-employed oilwell stimulation technique designed to increase permeability in a zone around the wellbore. The acid mixture flowing into the porous medium reacts with the various mineral species present, thus effecting an increase in the matrix porosity and, it is hoped, the permeability. Clearly, one of the factors controlling the depth of acid penetration is the chemical composition of the minerals which the acid contacts. Smith and Hendrickson,1 Gatewood,2 and Lund et al.3–5 have shown that the reaction with calcite is more rapid than with silicate minerals (clay or feldspar), which is, in turn, more rapid than the reaction with silica. Several papers describing the distance of penetration have been published. Smith and Hendrickson1 and Smith et al.6 first suggested the use of linear core tests to predict radial penetration. Farley et al.7 reported tests similar to those conducted by Smith and Hendrickson but measured many additional parameters including the effluent acid concentration, which is quite useful since the effluent concentrations may yield information about reaction characteristics. Experiments conducted in linear systems are difficult to translate in terms of penetration in a radial system, since the fluid velocity varies inversely with radial distance. The obvious approach has been to develop a mathematical model that can be calibrated based on linear flow data and then applied to a radial system. Gatewood2 proposed that the acid penetration distance be determined by assuming that the reaction of HF with the silicate minerals is much faster than with the silica. The distance of penetration is determined in this model by the formation composition and by the stoichiometry of the reactions. Lund et al.5,8 and Fogler and McCune9 developed a model which neglects the reaction of HF with silica but does consider the reactions with the silicate minerals. The advantage of these approaches is that the penetration depth can be predicted based on the formation composition. However, the reaction with silica cannot be neglected in determining the depth of penetration, as will be seen.