Published by the DTI Oil & Gas Directorate for the reservoir engineering and IOR community in the UK.
Send comments on this issue and contributions for next issue to iornewsletter@senergyltd.com by 15 December 2002.

Depressurisation of Waterflooded Reservoirs – Results From Oil-Wet and Mixed-Wettability Micromodels

The Heriot-Watt Institute of Petroleum Engineering at Heriot-Watt University has been studying the depressurisation processes, using experimental (micromodels) and theoretical (network modelling) methods for the past eight years. Phase 1 of the micromodel studies covered a series of high-pressure (up to 5300 psig) experiments using only water-wet micromodels. In the current phase the process is being studied in oil-wet and mixed-wettability models.  Dabir Tehrani (Dabir.Tehrani@pet.hw.ac.uk) presents some of the latest experimental results (co-authors A Danesh, G. Henderson, K. Shahabi and M. Sohrabi).

Many oil reservoirs are produced by waterflooding. When the water-cut is too high, for further production to be economical, fields are usually abandoned. At this stage a significant quantity of oil and gas remain. Some reservoirs contain gas cap and/or high GOR oil and it is likely that by the process of depressurisation significant quantity of gas and some of the residual oil can be produced. The Brent field in the North Sea is one such reservoir [link to article on Brent in issue 2]. By stopping water injection and continuing to produce fluids from such reservoirs, the gas-cap gas is expanded and some gas is also liberated from the residual oil. It may then be possible to economically produce extra quantities of oil and gas. Unfortunately, the underlying physics of such complex three-phase flow is not well understood to allow reliable predictions to be made for economic evaluation. Simulating this recovery method will involve a good knowledge of critical gas saturation (S_gc) and three-phase relative permeability and capillary pressure functions that are appropriate for the depressurisation process, in the reservoir of interest. It is highly impractical to develop such functions for all pertinent conditions occurring in realistic reservoir situations, by core tests. The approach adopted at Heriot-Watt is to develop a pore scale mathematical network simulator, which covers all of the significant physical flow processes involved in this recovery method. But to gain confidence that such a simulator can indeed reflect the physics of the flow realistically, we need to test it against the actual physical micromodel observations.

If the predictions of the network simulator agree with those observed in the micromodel, we can then operate the simulator with real reservoir fluid data and basic rock properties in 3-D space, to calculate the required values of critical gas saturation and pore-scale relative permeability functions. Phase 1 of the studies was an extensive research involving AEA Technology (for core flow studies), Imperial College of London (for low-pressure micromodel work) and Heriot-Watt University for high-pressure micromodel experiments and mathematical network modelling. Only water-wet systems were studied at Heriot-Watt in Phase 1 [1, 2 &3].

Observations made in the micromodel studies of Phase 1 of the project have considerably improved our understanding of the underlying physical principles. Understanding the wettability effects on the process of depressurisation is one of the most important objectives of Phase 2 experiments. In Phase 2 we study the process in oil-wet and mixed-wet models, to see the effect of wettability. We have also included two new types of experiment in the scope of this research. These are carrying out experiments:

• starting with different values of water saturation. This will help identifying the effect of initial water saturation, as in reality the starting value will be different in different parts of a waterflooded reservoir, and
• with pressure cycling, i.e., repressurising the model by water injection, at the end of some of the depressurization experiments, to examine the effect of gas resolution, and subsequent depresssurisation. This should show the effect of redistribution of the fluids in the initial depressurisation period on the secondary depressurisation that will follow the repressurisation of the model by injecting water.

Strongly oil-wet micromodel
To obtain an oil-wet micromodel we changed its wettability using a chemical treatment process [5].  In this method the micromodel is exposed to the vapour of a silane chemical [bis  (dimethylamino) dimethy-lsilane] at 275ºC for about three hours. Figure 1 presents a section of an oil-wet micromodel. The wettability was observed to remain oil-wet for more than three months, while performing some experiments.

Figure 1 – A section of an oil-wet micromodel. Water is colourless and oil is pink. The interface curvatures clearly show that the model is oil-wet. (click image for larger view)

Mixed-wet micromodel
A mixed-wet micromodel is one in which some pore surfaces are water-wet, some oil-wet and some of neutral wettability. In fact, a range of contact angles between 0 and 180º may exist in the model. Starting with an oil-wet micromodel, produced by the chemical treatment noted above, distilled dead water was injected and aged for 48 hours. This aging resulted in a mixed-wet model. The dead water was then displaced with undersaturated live water at 5200 psig (which had been previously brought into equilibrium with methane and normal decane at P_b=5054 psig and T_b=100ºF). Live oil (liquid of C1/nC10, at above-mentioned P_b & T_b in equilibrium with water) was then injected. Residual oil saturation was then established by injection of live water at a very low (capillary dominated) rate. Subsequent examination of the water-oil interfaces, both before and after completion of the depressurisation showed that the wettability had stayed almost the same as that of initial conditions. Figure 2 presents a section of a mixed-wet micromodel prior to depressurisation.

Figure 2 – A section of a mixed-wet micromodel. Water is light blue and oil is pink.  The interface curvatures show a variety of contact angles. (click image for larger view)

Experimental results
After the initial residual oil saturation to waterflood (S_orw) was established, as described above, the strongly oil-wet model was depressurised at the desired depletion rate. Experiments were conducted at S_orw values of 22 to 50% and three depletion rates of 400, 40 and 3.6 psi/h.  A series of experiments was carried out with mixed-wet models of different degrees of oil-wetness at depletion rates of 400, 40 and 3.6 psi/h.

Figures 3 and 4 present sections of the oil-wet and mixed-wet micromodels, depresssurised at a low rate of 3.6 psi/h.

Figure 3 - Depressurisation of an oil-wet micromodel at slow depletion rate (2.6 psi/h) Sorw=33% P_b=5054 psig. (click image for larger view)

Figure 4 - Depressurisation of a mixed-wet micromodel, slow depletion rate (3.6 psi/h), Sorw =37% P_b=5054 psig (click image for larger view)

Conclusions
Very useful pictures and video records of the gas nucleation, bubble growth and flow, under the process of depressurisation, were obtained for oil-wet and mixed wet micromodels, at different depletion rates. Many of these were analysed and a number of graphs of the trends of saturation vs pressure, the number of bubbles created as a function of various parameters were prepared. These provide excellent information for checking the integrity of a network model simulator.

In medium and slow rate of depressurisation the process of gas nucleation was a progressive one, whereas in the fast depletion rate it appeared to be almost instantaneous.

Nuclei appeared to form randomly in the micromodel, independent of the size of the oil ganglia.

As expected, nuclei appeared in the oil, although the oil and water were both saturated at 5054 psig and 100oF. This is attributed to the lower IFT between gas and oil than the IFT between gas and water, near the bubble point pressure.

Acknowledgements
The Depressurisation project at Heriot-Watt U. is equally sponsored by: The UK Department of Trade and Industry, BP Exploration Operating Company Ltd., Marathon International  (GB) Ltd, Norsk Hydro a.s, PDVSA – Intevep S.A., Shell UK Exploration and Production and TotalFinaElf Exploration UK, which is gratefully acknowledged.

References

1. Tehrani D. H., Danesh A., Henderson G.  and Mackay E., "Engineered Depressurisation of Waterflooded Reservoirs – High Pressure Micromodel Studies", Technical Report No. 10, Final Report, February 1997.
2. Sorbie, K. S., McDougal S. R., "Engineered Depressurisation of Waterflooded Reservoirs – Network Modelling of Depressurisation Processes", Technical Report No. 12, Final Report, June 1997.
3. Mackay E. J., Henderson G.D. and Tehrani, D.H. "High Pressure Micromodel Visualisation Studies", DTI's Improved Oil Recovery Seminar 18 June 1997.
4. Mackay, E.J., Henderson G.D., Tehrani D.H. and Danesh A. “The Importance of Interfacial Tension on Fluid Distribution During Depressurisation”, SPE Reservoir Evaluation & Engineering, October 1998, pp.  408-15.
5. Takach, N. E., Bennett, L. B., Douglas, C. B. and Thomas, D. C. “Generation of Oil-Wet Model Sandstone Surfaces, SPE 18465,  SPE International Symposium on Oil Field Chemistry, Houston Texas, February 8-10, 1989.

Disclaimer: The material available on this website is designed to provide general information only. Whilst every effort has been made to ensure that the information provided is accurate, it does not constitute legal or other professional advice.
Please note: The Department of Trade and Industry cannot be held responsible for the contents of any pages referenced by an external link.