Compaction and Dilation Effects on Polymer Flood Performance

Abstract During recent field applications of polymer flooding in unconsolidated reservoirs, questions have arisen concerning the role of compaction and/or dilation on flood performance. The primary goal for this paper was to assess the extent to which the compressible nature of a formation affects polymer flooding. The Tambaredjo field in Suriname (with in-situ oil viscosity ~600 cp) was used as a model for this study. Comparisons were made during simulations where formation compressibility was 5.6 × 10-4 psi-1 versus 1 × 10-6 psi-1. During a simulated 17-year compaction drive with compressibility of 5.6 × 10-4 psi-1, water cut gradually increased to average 20% (consistent with the actual field performance)—compared to 2% if compressibility was 1 × 10-6 psi-1. Oil recovery during this period was 18% OOIP for the highcompressibility case versus 3% OOIP for the low-compressibility case. Subsequent to the above compaction drive, incremental oil recoveries from waterflooding and polymer flooding were significantly less (about half in our case) when compressibility was 5.6 × 10-4 psi-1 than when compressibility was 1 × 10-6 psi-1—simply because the oil recovery target was less. For the many waterflooding and polymer flooding cases, most incremental oil was recovered within five years of starting injection—regardless of formation compressibility. Water cuts rose to high values within five years of injection, regardless of the viscosity of the injected fluid and the compressibility value. Consistent with the actual field application, the response to polymer injection varied greatly from well to well. However, our analysis indicated that these variations were due to existing heterogeneities within the pattern—not to the high compressibility of the formation. During simulation, polymer injection increased porosity by factors up to 1.5 and permeability by factors up to 2.3. Nevertheless, compaction or dilation had a fairly even (proportionate) effect on porosity and permeability throughout the pattern. If polymer injection was stopped after the simulated peak in porosity was reached (after 4-6 years of injection) and compaction was allowed to resume, a modest level of oil recovery resulted from this second compaction period (25%-38% of the incremental oil during polymer injection). However, substantially longer was required for the recovery (15-17 years versus 4-6 years for polymer flooding). Consequently, relying on re-compaction during this period of low oil prices may not be as profitable as one might hope. Our work suggests that there is an optimum rate, viscosity, and pressure for polymer flooding compressible formations. Flooding too rapidly results in pressures that waste much of the injection energy on dilating the formation—thereby detracting from efficient displacement of the oil. These constraints restrict injection rate, viscosity, and pressure to a greater degree for very compressible formations than for incompressible formations.