Use of STARS to Model Alkali-Surfactant-Polymer (ASP) Flooding

David Hicks (david.hicks@cmgl.ca), Account Manager, European Region for Computer Modelling Group Ltd (http://www.cmgl.ca) discusses CMG's STARS simulator and presents the results of applying it  to model a field-wide Alkali-Surfactant-Polymer process .

Chemical flooding has typically been associated with low rate, land based projects. One of the primary reasons for this is due to the chemical concentration requirements for a successful flood. Even very low concentrations of chemical, in a high rate offshore injection project, can require huge volumes that require storage and transport to the platform site. So even if theoretically economic, logistics can effectively prevent implementation.

A method, which has shown a lot of success in the North American onshore environment, is Alkali-Surfactant-Polymer (ASP) flooding.

There have been around 30 Alkali-Surfactant-Polymer (ASP) projects, or variations of, conducted to date. The ASP Project List, supplied by Myron Kulman of MK Tech Solutions, provides a reference to these projects.

ASP flooding relies on reducing interfacial tension between oil and water by the introduction of surface-active chemicals to increase recovery rates. The chemical has an oil-loving part (hydrophobe) and a water-soluble part (hydrophile). This results in the interfacial tension being lowered when the chemical (surfactant or soap) concentrates at the oil and water interface. Low (0.05% to 0.5%) surfactant concentrations only, are required for effective implementation.

The addition of alkaline chemicals (Na₂CO₃, NaOH) further enhances the process. The alkaline chemicals convert acids in the oil to soaps to create surfactants in situ.  These, in turn, reduce the interfacial tension and capillary forces. The alkali also helps to reduce surfactant absorption onto the rock surface.

In order to enhance the mobility ratio of the AS slug, and to enhance subsequent water flood displacement, polymer is also often used. The ASP process can be applied at any time during a waterflood, even at the end in a tertiary mode. Incremental oil recoveries of 20% to 30% OOIP have commonly been reported, with typical chemical costs of between $3 and $6 per barrel of oil recovered.

The Computer Modelling Group (CMG), based in Calgary Canada, provides a simulator STARS that effectively models this process and its many physical characteristics. The STARS simulator provides the ability to model any chemically based process in a 3D, multi-component, thermal environment. It has been used for a variety of such field studies over the last 20 years. One such study conducted in CMG’s Calgary office investigated an onshore field consisting of a fairly clean, thin sandstone reservoir interval. The reservoir had a low (27%) residual oil saturation to the existing waterflood, and was underlain by a moderately active aquifer. Producing initially on primary production since the mid-1960s followed by subsequent water injection, the reservoir was coming to the end of its life and the ASP process was considered to try and rejuvenate the field. Figure 1 shows a water saturation profile of the reservoir prior to IOR implementation, with the darker blue representing the highest water saturations. It can be seen that there was still a fair amount of oil that could be extracted given the right development strategy.

Figure 1: Water saturation distribution prior to IOR (Click image for larger view)

Laboratory work on core specimens provided a residual to ASP of ~22%, a 5% improvement on straightforward waterflooding. Initial optimisation of the chemicals was performed in the lab and the resultant core flood data matched using STARS.  The field history of primary production, followed by waterflood was matched with STARS using a reasonably coarse grid to provide rapid control on the general reservoir characteristics. A finer grid model was then created for ASP modelling, the match was confirmed, and then this finer grid model was used for all forecasting.

In order to effectively model the ASP process the simulator must be capable of adequately describing the controlling process mechanisms. Important items which should be addressed by the simulator are:

STARS does this through the following simulator features:

As can be seen in the production profile plot below (Base Case (W/F)), the base case recovery in a do-nothing scenario resulted in a 47% recovery factor. Scenarios with additional injectors both in vertical or horizontal configurations led to only marginal improvement of around 1% and would not have been economic. Applying chemical to the vertical well environment showed that injectivity was poor unless an unreasonable number of wells were used (Vertical (ASP)). The resultant flood sweep of this arrangement was also poor leading to a non-optimal overall performance. A further complication was also observed keeping the chemical and subsequent waterflood in the oil saturated upper portion of the reservoir. Gravity, in combination with a high vertical permeability, caused the water to slump into the lower, watered out sections. In order to counter these issues, horizontal wells were considered the best option (Horizontal (ASP)). They showed improved sweep with low well numbers and good injectivity, and more importantly, the chemical kept mainly to the upper oil saturated sections. This resulted in a final incremental recovery of 10% over continued waterflood, with oil rates improving from 50bbl/day to around 350bbl/day for an extended period, providing a highly economic prospect.

Figure 2: ASP flooding following waterflood (entire field). Oil production rate and oil recovery factor for various well scenarios (Click image for larger view)

As a further reference, Myron Kulman, of MK Tech Solutions provides a useful practical example of the economic benefits of the ASP process ASP_Training.ppt. He compares a practical field example with the STARS simulation model and uses the results to show the economic gain for a typical US pattern water flood.