Measurement of the Soret Coefficient in Crude Oil

Abstract

Abstract Very recent works show that thermal diffusion (Soret effect) can have a significant impact on the composition grading of thick oil reservoir. So far, neither satisfactory theoretical models of thermal diffusion nor reliable measurements of thermal diffusion coefficients is available. Lack of knowledge of the magnitude of thermal diffusion may lead to an incorrect estimate of GOC location and to an inaccurate estimation of oil and gas in place. In this paper, we illustrate the potential impact of thermal diffusion on the compositional grading of a real field, outline the objectives the SCCO project and detail the technical achievements realised in this frame. The key objectives of the SCCO project are the following: To measure accurately the Soret coefficients of crude oil samples held in reservoir initial conditions (p 350 bar, T 60 C), to develop appropriate compositional modelling algorithms taking into account thermal diffusion and using reliable values of the measured Soret coefficients. To provide undoubted values of Soret coefficients for testing existing and future theories or correlation laws on the thermal diffusion phenomenon. Practically, Soret coefficients in multicomponent mixtures cannot be measured on ground because thermal and compositional gradients unavoidably induce convective mixing. In the SCCO project, measurements of the greatest accuracy will be obtained performing the experiment in microgravify where buoyancy driven convection is absent. In the present configuration of the hardware, three different complex hydrocarbon samples will be processed. Experimental constraints imposed by the high pressure of the samples and by the microgravity environment require the development of original technical solutions. We describe the miniaturised thermal diffusion cells as well as the dedicated sampling interface with the GC for post flight analysis. P. 57