Representing Reservoir Oil Using Recombination Techniques

Abstract The approach to reservoir optimization often begins with the appropriate characterization of the reservoir fluid. Deficiencies in sampling methods may lead to erroneous conclusions regarding the fluids in situ and, therefore, the exploitation strategies considered. Moreover, once sampled, the way in which fluids are recombined is many times inadequate. Frequently, bottom-hole samples are also problematic and must be analyzed correctly in order to infer the appropriate information about the reservoir system. This paper discusses sampling and recombination methods that provide the best representation of reservoir fluids. A number of examples are provided where standard approaches to characterization are inadequate and a protocol of recombination is presented. The benefits of the approach are shown. The impact of characterization is also shown relative to allowable production rates and adherence to regulatory edicts. BACKGROUND Much of the engineering that is involved in the development and exploitation of reservoirs world-wide depends on representative fluid samples. Whether it is the measurement of PVT properties such as density (α), formation volume factor (®o), viscosity (µ), interfacial tension (IFT), gas-oil ratio (GOR) or compressibility (c), or the generation of relative permeability relationships or the assessment of enhanced oil recovery (EOR) strategies, each of these endeavors requires a representative reservoir fluid. Although it may seem that the method for the acquisition of a representative reservoir fluid would be straightforward, it is surprising to see the number of fluid characterizations that are not done correctly. This deficiency then carries through to all of the analyses that are performed and consequently, the results of the engineering may be in error primarily due to the fluid with which the work was commenced; thus the importance in establishing a reliable protocol for the preparation of representative reservoir fluid. Oil and gas reservoir opportunities are distributed amongst a number of different fluid types (Figure 1). The oils may be heavy oil containing very little gas and very heavy, high-density components, more conventional oils containing components that are easily partitioned into the gas phase, or volatile oils where the difference between the gas phase and the liquid phase is much less pronounced. As one moves to the right of the critical point, the classification of fluid moves to very-rich condensates, retrograde condensates, to wet gas and finally to dry gas. The extremes on either side of the critical point are easy fluids to represent. Heavy oil has almost exclusively methane in the solution gas and C20+ in the liquid. There is very little "gray area". Heavy oils do not normally have characterization difficulties since the solution gas is almost always at least 93% methane, in which case, recombination methods do not significantly impact the recombined oil properties. Dry gas has only gas-phase components and therefore these gases are easy to sample and to represent. The problems enter with the fluids that contain components that partition easily into both liquid and gas phases. These fluids can then become very sensitive to pressure influences which are ubiquitous in reservoir engineering.