HEAVY OIL Effects of temperature on heavy oil-water relative permeability of sand BRIJ B. MAINI Petroleum Recovery Institute and T. OKAZAWA Esso Resources Canada Limited ABSTRACT Although a number of studies have reported significant effects of temperature on relative permeabilities no consensus has emerged on the generality of such effects nor on possible mechanisms causing such effects. Some of the recent studies have found relative permeability to be independent of temperature and have suggested that most of the reported temperature effects can be attributed to artifacts of the unsteady-state technique. The objective of this study was to critically examine the use of the unsteady-state technique for measuring relative permeability in heavy oil systems and to ex-perimentally determine the effect of temperature on relative permeability curves for a clean silica sand/heavy crude oil deionized water system. Unsteady-state measurements were carried out in a 45 cm tong, 5.6 cm diameter sand core at five different temperatures ranging from room temperature to 200'C using a heavy crude oil and deionized water. It was found that the unsteady-state technique when employed in heavy oil systems is more suscep-tible to experimental artifacts (compared to its use in light oil systems), however, a careful analysis of the displacement data can provide meaningful relative permeability curves in spite of the inevitable artifacts.The relative permeability curves derived from production and pressure drop histories of the displacements at different temperatures showed that, in this system, relative permeability curves vary with temperature. The endpoint water permeability as well as the effective water permeability at intermediate satura-tions increased with increasing temperatures. The endpoint oil permeability was found to be independent of temperature-While the shape of the oil relative permeability curve displayed a complex dependence on temperature, its significance remains uncertain due to the presence of several artifacts. Introduction Production of oil from petroleum reservoirs usually involves simultaneous flow of two or more immiscible fluids through a porous rock. Multiphase flow in porous media is a complex process that depends on a number of factors including the Keywords: Heavy oil, Relative permeability, Temperature, Flowin porous media. absolute permeability, pressure drop, capillary pressure, fluid viscosities, and relative permeabilities of each phase. Of these, the relative permeability is probably the most important para-meter in determining reservoir performance. For modelling thermal recovery processes for heavy oil recovery, one needs to know not only the relative permeabilities at the original reser-voir conditions but also the effect of increasing temperature on the relative permeability curves.A number of studies have discussed the temperature effects on oil/water relative permeabilities. Unfortunately, no con-sensus has emerged on the generality of reported temperature effects nor on possible mechanisms ...

Summary Some heavy oil reservoirs in western Canada and Venezuela show anomalously high primary recovery under solution gas drive process. The pressure decline rate in these reservoirs is low compared to that expected under solution gas drive in conventional oil reservoirs. There is now increasing evidence that gas mobility is extremely low in these reservoirs. The objective of this study is to conduct solution gas drive experiments in a sandpack saturated with a heavy oil and examine the effect of depletion rate. Depletion rate was varied by more than two orders of magnitude. The results showed that gas mobility was a function of depletion rate and decreased with increasing depletion rate. Other notable observations were that supersaturation increased with depletion rate and that critical gas saturation was 3 to 4%, slightly increasing with increasing depletion rate. Interpretation of the results confirmed that gas mobility is quite low. Representation of the low mobilities using relative permeability required low values of the order of 10-5-10-4, which decreased with increasing depletion rate. Introduction Some of the heavy oil reservoirs in western Canada1-2 and Venezuela3 show anomalous behavior under solution gas drive. Once below the bubblepoint pressure, the producing GOR does not increase sharply, and the rate of pressure drop is low. Relatively high primary recovery factors in excess of 10% have been reported from some of these reservoirs. Similar behavior is being reported in other heavy oil reservoirs in China and Albania.4,4 To explain the anomalously high primary recovery under solution gas drive, several theories were initially proposed.1,3,6-9 Most researchers now agree that the gas mobility in this process is extremely low10,11 and leads to effective oil recovery. In this paper, we present the data of carefully conducted depletion experiments in heavy oil and investigate the effect of depletion rate on the solution gas drive process. Effect of depletion rate in solution gas drive in light oils has received previous attention. 12-14 Here, using a heavy oil, we present the pressure and recovery data when depletion rate is varied by more than two orders of magnitude. The data are then interpreted using a simplified method to show the effect of depletion rate on gas relative permeability. Literature Review The behavior of heavy oil reservoirs under solution gas drive has intrigued the oil industry for over a decade now.1,2 However, research on solution gas drive dates to much earlier years. In the early '50s, Stewart et al.12 examined solution gas drive in heterogeneous limestones. They showed that the relative permeability of external gas drive is different from that under solution gas drive. Furthermore, they found that higher rate of depletion leads to higher oil recovery. The authors attributed this to a larger number of gas bubbles at higher depletion rates, which in turn leads to lower gas/oil relative permeability ratio. Later, Handy13 used two oils with dead oil viscosities of 1.8 and 25 cp and confirmed the same conclusions for solution gas drive in a sandstone core. Dumore14 conducted solution gas drive experiments in two high permeability sandpacks of 15 and 350 darcy. He suggested that conditions that led to more gas dispersion led to higher recovery. The author showed that higher rate of pressure drop and higher permeability lead to more gas dispersion. All of the above authors were interested in behavior of solution gas drive in light oils. Extensive research in solution gas drive in heavy oils was initiated following Smith's1 publication reporting high oil recovery and production rate in some heavy oil reservoirs under primary depletion. He suggested that in these reservoirs, gas flows in the form of tiny bubbles in heavy oil. He further stated that these gas bubbles do not coalesce to form a continuous gas phase. Maini et al.6 suggested that a discontinuous gas phase is dispersed within the continuous oil phase and used the term "foamy oil flow" to describe the flow. Later, Bora et al.15 studied the effect of rate of depletion on bubble nucleation and the foamy oil flow in a micromodel. They concluded that the higher rate of pressure drop results in the nucleation of more bubbles and more dispersed flow. To explain the favorable performance of solution gas drive in heavy oils, a geomechanical effect (i.e., formation of "wormholes," which are essentially high-permeability channels), has also been proposed.8 The present study investigates the behavior of heavy oil reservoirs in the absence of geomechanical effects. Shen and Batycky9 suggested increased oil mobility because of lubrication caused by nucleation of bubbles at pore walls as a possible mechanism leading to enhanced primary recovery. Some authors have suggested that high critical gas saturation may explain the high recoveries observed.3,6,16 In a detailed study, Li and Yortsos17 developed a network model and studied the effect of depletion rate on critical gas saturation. The authors concluded that for sequential nucleation, critical gas saturation increases with depletion rate. In more recent works,7,18 the gas phase mobility under solution gas drive experiments in heavy oil was determined. It was found that gas mobility in heavy oil is much lower than that in light oil.7 The low gas mobility was suggested to lead to improved oil recovery. It is not, however, known what factors affect gas mobility. Various observations from the literature suggest that solution gas drive in heavy oil depends on depletion rate, similar to what has been observed in light oils.12–14 However, no systematic study has been reported yet. The objective of this paper is to investigate the effect of depletion rate on solution gas drive in heavy oil. Flow rate, pressure, and pressure drop information will be analyzed to infer relative permeability functions. Earlier12 as well as more recent studies19 suggest that gas relative permeability under internal drive is different from that under external drive. There are only few published data7,18 of gas relative permeability in the presence of heavy oil, using depletion techniques. These studies have used the core depletion set-up to represent the solution gas drive process. In this study, a similar apparatus was set up to conduct the desired experiments. Several improvements were made to perform the experiment under highly controlled conditions. Pressure was measured at different points along the length of the sandpack. The core-holder was rotated to negate gravity affects. Overburden and axial pressures were applied during the experiments. A connate water saturation was established in the sandpack, and more accurate pressure transducers were used. The experimental setup and procedure is explained in the following. This is followed by presentation of the experimental results and their analysis.

The VAPEX analytical model is extended to cover situations when diffusion coefficients are dependent on concentration due to the extreme viscosity reduction with solvent dissolution into bitumen. The new analytical model covers such situations along with the cases in which the diffusion coefficient and viscosity relate to each other under the Stokes-Einstein law. In the process, a new concept of the 'average flow fraction of bitumen' in the flowing mixtures is introduced. The modelled result on overall functionality of the drainage rate of bitumen has confirmed the square-root relationships to most of the key reservoir parameters as the previous theories indicate. However, its dependence to the inverse of kinematic viscosity at the interface is closer to linear rather than square-root correlation with the concentration-dependent diffusion coefficient. The theoretical relationships are confirmed by the correlations of the unified bitumen rates to kinematic viscosity at the interface using the existing VAPEX and SAGD experimental data in the literature. This finding indicates that VAPEX process in heavy oils with lower native viscosity can be more effective than originally recognized. Introduction Dunn et al.(1) developed the theoretical model of the gravity drainage process for bitumen recovery known as VAPEX based on the model of the steam-assisted gravity drainage (SAGD) process by Butler et al.(2) This model assumed that the diffusion coefficients of solvent-bitumen systems are constant similar to the case of thermal diffusivity. Thus, the steady-state profiles of solvent concentration ahead of the solvent-bitumen interface is the smooth exponential decay towards an infinite distance. In reality, the diffusion coefficients of both solvent and bitumen are strongly dependent upon compositions due to the extreme viscosity contrasts between the solutes and solvents. As a result, the observed concentration profiles in diffusion experiments exhibit the abrupt front-end profiles(3). The theoretical endeavour here is to understand the impact of the non-exponential concentration profiles on the VAPEX drainage and bitumen rates. Governing Mechanisms The most fundamental mechanism of the process is the gravity drainage caused by the density difference between the liquid-bitumen phase and the injected vapour phase. The drainage flow of the bitumen phase occurs only from viscosity reduction due to the impact of the injected solvent (or heat in the case of SAGD) of the otherwise semi-solid bitumen. Therefore, how the injected solvent penetrates into the bitumen phase in the reservoir is of primary importance to the process. According to Fick's law, a material balance across a differential distance, dx, in this situation can be expressed as a continuity equation for the change of concentration (volume fraction is chosen), C, with time, t, by using the diffusion coefficient, D: Equations (available in full paper)

Osmotic pressures of moderately concentrated solutions of polystyrene in toluene are examined with the object of evaluating the free energy parameter B which describes the long-range interactions between segments. The B value evaluated through an analysis of the data is used to estimate the excluded-volume parameter, z =0.330B (6 (s' >0/M)-3/'M1I·, for polystyrene solutions in which solvent is toluene. The behavior of a. 3 = [1] J/[1] Jo over a wide range of z (0 < z < 6) is compared to the theoretical predictions. It is found that ai is not a linear function of z. The dependence of a. 3 on z seems to be described better by the recent Fixman relation than the Stockrnayer-Fixrnan relation.