Use of DME as an EOR Agent: Experimental and Modeling Study to Capture Interactions of DME, Brine and Crudes at Reservoir Conditions

Abstract

Abstract Dimethyl ether (DME) is a small, slightly polar component, non-toxic that exhibits nearly ideal behavior with known liquid-state hydrocarbons, and is partially miscible with water/brine. In a multicomponent brine/DME/oil system, it partitions preferentially into the hydrocarbon phase. This unique characteristic of DME makes it a suitable candidate for enhanced waterflood while in essence water can act as the carrier fluid - it has the potential to unlock remaining oil from fields in a mature waterflood state. When DME mixed with brine is injected into the reservoir, it partitions preferentially into the oil phase (hydrocarbon phase) upon contact. As a result, the residual oil swells and its viscosity is reduced. The amount of swelling and reduction in viscosity depend on how much DME partitions from brine into the oil and as well as the compositional characteristics of the oil. In addition, swelling and property changes are function of pressure-temperature conditions, DME concentration in oil as well as the salinity of the aqueous phase. Therefore, the estimation of the partitioning coefficient is very crucial to evaluate and understand the performance of DME enhanced waterflood (DEW) at reservoir or lab/pilot scale, as it controls the mass transfer from aqueous phase to hydrocarbon (oil) phase. In earlier studies, we presented the Equation of State (EOS) modelling and workflows to characterize the phase behavior of DME/brine/hydrocarbon mixtures that is encountered in reservoir processes, to estimate the partitioning coefficient, and calibrated and compared against the in-house experimental measurements. In this study, we extend the earlier work and perform extensive analysis on sensitivity of DME-partitioning coefficients with respect to pressure, temperature and salinities. In particular, we Present the comparison of predicting capability of the PVT models based on the two EOS: cubic plus association (CPA) [1] and Peng-Robinson (PR78) with Huron-Vidal (HV) mixing rules [2] against experimental data from literature or in-house measurements.Show that the model predictions are in agreement with the experimental observations.Present the extensive sensitivity analysis of DME-partitioning coefficient between oil and aqueous phases with respect to molecular weight of oil, temperature, pressure, salinities and DME concentration in oil phase, that can be utilize in quality check of the new data and preliminary screening criteria.Most importantly, develop an observation-based correlation to estimate partitioning coefficient for DME in brine and hydrocarbons, and to generalize the results for future screening studies where none or very limited PVT measurements with DME are available.