Experimental and Mathematical Studies of Cyclic Solvent Injection To Enhance Heavy Oil Recovery

Du, Zhongwei
Journal Title
Journal ISSN
Volume Title
Faculty of Graduate Studies and Research, University of Regina

It has been suggested that Cyclic Solvent Injection (CSI) is a highly promising technique to recover heavy oil resources at which other recovery techniques are not economically or efficiently available. Therefore, it is necessary to conduct experimental and mathematical studies on the CSI for effective heavy oil recovery purposes. Experimental and data regression studies have been conducted to investigate the effects of the wormhole on the CSI. Nine tests were completed using three sand-pack physical models with different dimensions. Experimental results suggest that the oil production can be divided into two phases: solvent chamber rising phase and solvent chamber spreading phase. The average production rate in the solvent chamber rising phase is proportional to the wormhole length, while the average production rate in the solvent chamber spreading phase does not change much with the wormhole length. In addition, a relation of the oil production rate to the drainage height is obtained by regression analysis and verified with a different experiment. It is suggested that for a rectangular model, the oil production rate in the chamber rising phase is proportional to h1.1667. Experimental study of effects of pressure decline rate on the CSI has been performed. Twelve tests with ten decline rates through linear and non-linear pressure-drawdown methods were conducted. Results of optimizing the pressure decline rate indicated that the pressure decline rate plays a primary role in the CSI by affecting solvent chamber growth, foamy oil flow performance in Phase 1, and pressure drop force in Phase 2. Different driving mechanisms in different phases lead to different optimum pressure decline rates. Comparison of two pressure-drawdown methods suggests that the main difference between the non-linear pressure-drawdown method and linear pressuredrawdown method is that the former cannot provide a continuous driving force for diluted oil as well as the latter. The effect of the intermittent driving force is much more significant in Phase 2. Therefore, production performance of tests of pressure-drawdown linearly was generally better than that of tests of pressure-drawdown non-linearly. A linear material balance equation of CSI has been proposed to obtain the recovery factor of diluted oil. It is successfully used to obtain the recovery factor of diluted oil of each cycle for a well-designed CSI test in a rectangular physical model (80×40×20 cm3). The relation of the solution-gas oil ratio and the diluted oil formation volume factor with pressure under non-equilibrium state are obtained through linear regression based the material balance equation. They are successfully verified through experimental data of a CSI test in a cylindrical model. Results indicate that the efficiency of oil dilution increases from 4.75% to 10.70% before the Cycle 10. Then it slightly varies from 10% to 16.25% till the Cycle 25. It is dramatically decreased 10.11% in the last five cycles. For first three cycles, the diluted oil recovery factor is up to 40% due to extended production time. Then it almost keeps around 32% till the Cycle 25. Three mixture solvent with the decline rate of 12.5 kPa/min ,5 kPa/min and 1 kPa/min have been conducted. Knowledge of production performance of mixture solvent CSI tests is obtained through the comparison of mixture solvent tests with different decline rate and the comparison between mixture and pure solvent tests. Pure solvent tests had larger recovery factor and average oil production per cycle than mixture solvent tests. The asphaltene precipitation and production time significantly impacted the recovery factor of diluted oil in mixture solvent tests.

A Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Petroleum Systems Engineering, University of Regina. xx, 257 p.