Abstract

Vapex is a process that employs hydrocarbon vapours (usually propane and/or butane) as solvents to dilute and mobilize heavy oil or bitumen in the formation. The diluted oil or in some cases, de-asphalted oil, drains downwards by gravity to a horizontal producer located at the bottom of the reservoir. Because Vapex is a diffusion controlled mass transfer process and is slow, long horizontal wells are required to achieve the economic attractive rates.

In gravity drainage dominated processes, horizontal producers are placed near the base of the reservoir but the location of the injection well can be varied. In Vapex, the proper selection of well configurations is desirable to control the initial displacement, and, more importantly, the profiles of the vapour chambers in the formation. De-asphalting and its effect on the flow of diluted oil could also be related to the determination of well configurations which control the shape and direction of interface movement.

In this study, experiments were carried out in a scaled packed model in which Tangleflags heavy oil was extracted by co-injection of solvent (n-butane) and non-condensible gas (nitrogen) at a pressure above the vapour pressure of the injected solvent. The effect of solvent to non-condensible gas injection ratio on the production performance and on the quality of products was studied. Reservoir heterogeneity was altered in the model and experimental results with injectors located at various positions of the model are presented.

Introduction

For gas displacement processes, the injected gas tends to ride over the top of the reservoir due to the gravity effect and this results in a poor vertical sweep efficiency. For a vertical confined reservoir, oil recovery by gas displacement in the reservoir is determined by the volume of gas fingers or gas invaded zones. In the Vapex process, the stabilized drainage rate by gravity depends on the contact area between the solvent vapour and the oil and drainage height. Because of the nature of the Vapex process in which the flow of vapour is limited in pore spaces, the change of the permeability and capillary pressure would affect the growth and development of the vapour chamber.

Jiang and Butler (1995) reported experimental results with scaled packed models where continuous low-permeability layers and discontinuous low-permeability lenses were studied. It was observed that the presence of low-permeability layers or lenses resulted in a poorer Vapex performance and the locations of the injection well in layered reservoir were very important in achieving communications between layers. Experiments showed that, for layered reservoirs, injection from the top of the reservoir resulted in an improved oil production rate than injection from a lower well that was located right above the production well. In conventional heavy oil reservoirs like those in the Tangleflags field, the proper combination of displacement and gravity drainage mechanisms would result in a higher oil production rate. This can be accomplished by selecting optimal well configurations.

In experiments described in this paper, n-butane was mixed with non-condensible gas (nitrogen) before injection into the models.

P. 877

This content is only available via PDF.
You can access this article if you purchase or spend a download.