Scaling of Cocurrent and Countercurrent Capillary Imbibition for Surfactant and Polymer Injection in Naturally Fractured Reservoirs

Abstract

Summary This study tests the existing capillary imbibition scaling formulations in literature and proposes modifications to them for surfactant- and polymer-injection applications. Laboratory tests were performed using Berea Sandstone cores with different shapes, sizes, and boundary conditions. Kerosene, engine oil, mineral oil, and crude oil were selected as the oleic phase. Brine and two different concentrations of surfactant and polymer solutions were the aqueous phases. The capillary imbibition recovery curves were then used to test the existing scaling formulations. It was observed that the scaling groups were applicable with certain modifications to the cocurrent flow of surfactant and polymer solution-oil pairs as well as the brine-oil cases. In case of cocurrent flow, the gravity forces dominate recovery for low-interfacial tension (IFT) experiments, and modification to the gravity-scaling group for the boundary conditions is required. For countercurrent flow, however, the scaling groups were observed to be inapplicable even after the modifications. Introduction When the matrix part of a naturally fractured reservoir (NFR) contains great amounts of oil, enhanced oil recovery methods are aimed to recover this oil effectively. Matrix-fracture interaction is required for matrix oil recovery, and capillary imbibition is the underlying mechanism if the matrix is water-wet and enough water is supplied in the fracture network. The capillary suction of water by matrix, and expelling the oil simultaneously, is a phenomenon governed by numerous parameters. Matrix permeability,1–4 size and shape,1,3-5,7 wettability,8–12 heterogeneity,2,8,13 and boundary conditions3,5-7,14,15 determine the rate of the oil recovery. The properties of imbibing water,16 viscosities of the phases,11,12,17-19 and IFT3,15,20-23 also play a role on the capillary imbibition recovery rate. Mattax and Kyte1 first proposed a scaling equation for capillary imbibition based on the Rapoport's24 scaling laws. Later, modifications to Mattax and Kyte's scaling group were proposed.3,5,6,11,12,23,25-27 Scaling equations for gravity-dominated imbibition recovery were also provided for different purposes.21,26-28 On the other hand, under unfavorable conditions such as high oil viscosity, low matrix permeability, high IFT, and unfavorable matrix boundary conditions, capillary imbibition recovery may be very slow and may also yield high residual oil. Different methods can be applied to overcome these difficulties. Heat injection resulting in the reduction of oil viscosity and IFT,11,12,29,30 injection of surfactant,15,20-23 and polymer solutions18 have been tested in laboratory conditions previously for this purpose. Laboratory-scale experiments showed that these methods yield an increase in ultimate recovery. Heat injection also gives rise to an increase in the production rate. Therefore, the applicability of these methods is proven, but the existing scaling groups should be tested and modified, if necessary, for the field-scale applications of chemical injection into NFRs. Under the tertiary recovery applications, matrix fracture transfer due to capillary imbibition is more complicated than it is with the brine-oil pair. Additionally, the gravity forces could be effective depending on the boundary conditions and the type of injected fluid. Some attempts have been made to scale the capillary imbibition experiments for low IFT, polymer, and heat injection. The scaling of capillary imbibition under thermal effect has been examined in a few studies. Reis29 reviewed the matrix recovery mechanisms for nonisothermal conditions and provided the characteristic recovery times for them. Babadagli12 proposed an approach for scaling the capillary imbibition under temperature effect. However, due to the number of mechanisms involved, there still exist certain difficulties in the scaling associated with the thermal effects, as also pointed out by Briggs et al.30 Difficulties arise due to thermal expansion of oil, change in oil viscosity, IFT, and, therefore, relative permeabilities under temperature. Scaling the capillary imbibition for a low-IFT case is also challenging. Schechter et al.21 studied low-IFT capillary imbibition and observed that gravity may dominate the matrix recovery at low values of IFT. This observation entails an inclusion of the gravity factor into scaling formulation. Al-Lawati and Saleh23 proposed an approach to incorporate both the gravity and capillary forces, but they observed a poor correlation with the experimental data. Employing polymer solution as an aqueous phase in capillary imbibition recovery has also been studied. Ghedan and Poettmann18 observed changes in recovery curves when the polymer solutions were used instead of brine, but they did not study the scaling of the phenomenon. Note that low-IFT solutions as an aqueous phase can be useful for accelerating the capillary imbibition recovery as well as reducing the residual oil in matrix.3,15,20-23 Especially under unfavorable matrix boundary conditions, low IFT can be a remedy for low matrix recovery as shown by Babadagli et al.15 Considering the fact that chemical addition into aqueous phase to reduce IFT or increase its viscosity is a useful technique to increase the recovery rate and amount, scaling these processes up to reservoir conditions is necessary for performance estimation. This study aims to test the validity of available scaling groups for static imbibition when low-IFT and polymer solutions are used as the aqueous phase. Capillary imbibition experiments were conducted for different rock and fluid properties. After testing the scaling groups for different cases, the conditions for which the existing scaling group applies were identified. For other cases, modifications to scaling groups were proposed. Experiments All the experiments were conducted on the same type of rock (unfired Berea sandstone), but different types of oil, aqueous phases, and matrix boundary conditions were used. The sample preparation and experimental procedure are explained below.

Experimental study of some important factors on nonwetting phase recovery by cocurrent spontaneous imbibition

“…Hamon and Vidal (1972) performed four one-dimensional water-oil COCSI tests on synthetic water-wet porous media with different heights to study the effect of block height on imbibition. Babadagli et al (1999) and Babadagli (2001) claimed doing pure COCSI experiments. However, the boundary conditions used in their experiments reveal that those tests were mainly COUCSI with some minor cocurrent productions.…”

A laboratory feasibility study of dilute surfactant injection for the Yibal field, Oman

“…Surfactant solutions are expected to change capillary imbibition performance on sandstone and chalky carbonates (Keijzer and de Vries, 1990;Al-Lawati and Saleh, 1996;Alveskog et al, 1998;Babadagli et al, 1999;Spinler, 2000;Babadagli, 2001Babadagli, , 2002Babadagli, , 2003. This change could be due to lowered IFT or change of the wettability from oil-to water-wet.…”

Fractal Prediction Model of Spontaneous Imbibition Rate

“…It is important to characterize the fluid flow during spontaneous imbibition in order to establish the relationship between production rate and time. Many researchers have studied the spontaneous imbibition in porous media (Aronofsky et al 1958;Du Prey 1978;Hamon and Vidal 1986;Reis and Cil 1993;Cuiec et al 1994;Al-Lawati and Saleh 1996;Zhang et al 1996;Babadagli (2001) ;Høgnesen et al 2004;Tavassoli et al 2005). In this study, we reviewed many models reported in the literature and proposed a fractal model to predict the spontaneous imbibition rate by considering the rock's property represented by fractal dimension.…”