Alkaline flooding appears to be a promising process for enhancing heavy oil recovery after primary or secondary production from those reservoirs where thermal processes are not applicable. However, due to the heterogeneity of the reservoirs, the injected chemical solutions may flow to the producing wells through the high permeable channels and bypass the residual oil, resulting in poor volumetric sweep efficiency. Thus, evaluating the chemical blends and the injection strategies through laboratory tests with proper heterogeneous physical models can substantially reduce the risks of field application of this process. In this study, a new experimental method, channelled-sandpack flooding test, is designed to simulate the channeling phenomenon of actual heavy oil reservoirs during the chemical flooding process. The channelled-sandpacks are used to re-evaluate the effectiveness of chemical blends screened through the homogeneous sandpack flood tests. It is found that the effectiveness of the best blend obtained through homogeneous sandpack flood tests is significantly reduced in the channelled-sandpack flood tests. This demonstrates that using heterogeneous physical model, rather than the conventional homogeneous sandpack flood tests, to screen the chemical blends is necessary for designing the chemical flooding process in heterogeneous heavy oil reservoirs. With further experiments carried out for the modified blends with the channelled-sandpack flood tests, an optimal chemical blend can be determined for certain heterogeneous reservoirs. The optimal chemical blend can selectively block the high permeability zones and improve the sweep efficiency for a particular heterogeneous sandpack, thereby significantly enhancing the tertiary oil recovery. The new developed channelled-sandpack flood test should facilitate the optimization, design and implementation of chemical flooding and other enhanced heavy oil recovery processes. (c) 2006 Published by Elsevier B.V.
