Hydrocarbon solvent-assisted processes are thought to beeffective EOR technologies for heavy oil and bitumen production. In order to describe those processes quantitatively, the diffusivity of hydrocarbon solvent in oil sands is a crucial required parameter.
Computer Assisted Tomography (CAT), as a non-intrusive method, is a very useful tool to obtain the physical properties of cores (density, porosity, heterogeneities) and the fluids within the cores (saturation or concentration profiles). However, the concentration profiles obtained from diffusion experiments of solvent in oil sands were fluctuating and cannot be used to evaluate the diffusivity. In order to solve the problem, during the data processing of CAT scanning, porosity heterogeneity of porous media and volume changes on mixing are considered to improve the accuracy of concentration profiles dramatically. Subsequently, the diffusivity of hydrocarbon solvent in heavy oil saturated sands is determined based on an updated mathematical model, which considers porosity heterogeneity of porous media in the major diffusion direction and volume changes on mixing. Thus, the impact of porosity heterogeneity of porous media on diffusivity determination of hydrocarbon solvent in heavy oil saturated sands is investigated.
Recently, some novel processes of combining the benefits of steam and solvents in the recovery of heavy oil and bitumen have been developed, such as: Expanding Solvent SAGD  (ESSAGD), Solvent Aided Process  (SAP), Liquid Addition to Steam for Enhancing Recovery  (LASER), Steam Alternating Solvent process  (SAS). These solvent-assisted processes are aimed at improving oil rates and oil-to-steam ratio, and reducing energy and water consumption. Some of these processes have been field-tested successfully. In order to simulate these solventassisted processes numerically, the diffusivity of solvents in heavy oil saturated reservoir is required.
In porous rocks, molecular diffusion takes place along tortuous paths. Therefore, the effective diffusivity De in the porous medium is smaller than the molecular diffusivity Do in the bulk fluids. For the estimation of effective diffusivity in porous solids, usually two empirical relationships are used. Equation 1  gives a rough approximation to the diffusion coefficient in unconsolidated porous media. Equation (1) (Available in full paper)
The second method involves an analogy between electrical conductivity and diffusion in porous media as equation 2  shows. Equation (2) (Available in full paper)
where F is the formation electrical resistivity factor and is the cementation factor. O assumes different values in different packings. Specifically, for unconsolidated packs, the value of the cementation factor is 1.3 . This equation can be used for cemented rocks as well as unconsolidated packs.
Because the porosity of cores is usually less than 0.4, the results based on these two relationships are very different. Besides, these equations are only approximations for the ratio of effective diffusivity De to molecular diffusivity Do. More accurate methods are needed to investigate the diffusivity of solvent in oil sands.
Diffusivity of solvents in heavy oil or bitumen has been studied extensively.