Prediction of rocks thermal conductivity from elastic wave velocities, mineralogy and microstructure

Abstract

While knowledge on Thermal Conductivity (TC) of rocks is of interest in many fields, determining this property remains challenging. In this paper, a modelling approach for TC prediction from Elastic Wave Velocity (EWV) measurements is reported. To this end, a new effective TC model for a typical sedimentary rock is introduced that explicitly accounts for the presence of pores, pressure-sensitive microcracks (or grain contacts) and formation fluids. A model of effective elasticity is also devised for this same rock that links its microstructural characteristics to the velocity of elastic waves. The two models are based on the same effective medium approach and involve the same microstructural parameters. A workflow based on this explicit modelling approach is devised that allows for the prediction of the TC of a reservoir rock using (i) the elastic waves velocities, (ii) the dominant mineral content and (iii) the bulk porosity. This workflow is validated using experimental data reported in the literature for dry and water-saturated Fontainebleau and Berea sandstones. The datasets include measurements of TC and EWV as a function of effective pressure. In addition, it is shown that the dependence of TC on the rock microstructure is formally and practically similar to that of EWV. It is also demonstrated that the accuracy of TC predictions from EWV increases with effective pressure (burial depth). The underlying assumptions and limitations of the present approach together with the effect of burial are discussed.