Evaluation of the comparative performance of four service contractors has highlighted often inexplicable inconsistencies and discrepancies in laboratory waterflood data, despite using similar test cores, identical test fluids, and similar test procedures. Depending upon the methods laboratories prefer to establish initial water saturation, and their favoured waterflood rates, the difference between oil initially in place and residual oil saturation varied from 10 to over 60 saturation units. Duplicate tests carried out by the co-ordinating laboratory failed to replicate two laboratories’ results, and showed that both residual oil saturation and relative permeability are rate-dependent, but relative permeability appeared to be viscosity-independent. Although low rate "bump" floods might yield appropriate end-point data for water-wet rocks, virtually all laboratory waterfloods invalidate classical interpretation theory. Analytical models must therefore allow for capillary pressure, wettability, heterogeneity and viscous instability. Recommendations on the key steps in representative laboratory core flood test design include appropriate wettability conditioning, initial water saturation, selection of viscosity ratio and flow rates, and quality control procedures.

You can access this article if you purchase or spend a download.