(623e) Chemical Compositions in Low Salinity Waterflooding of Carbonate | AIChE

(623e) Chemical Compositions in Low Salinity Waterflooding of Carbonate

Authors 

Radke, C. - Presenter, University of California-Berkeley
Yutkin, M., King Abdullah University of Science and Technology (KAUST)
Abstract

Success of low salinity waterflooding (LSW) is attributed to increasing water wettability of the reservoir rock. Consequently, surface chemistry of the crude oil/rock/brine interfaces is a critical component of the process. A large number of laboratory studies, therefore, focus on wettability measurements including, for example, contact angles, surface charge densities, ion exchange, and zeta potentials, in addition to oil-recovery flooding. In carbonate rocks, however, injected chemistry is drastically altered by reaction with reservoir minerals, chnaging injection chemistry substantially. We demonstrate both experimentally and theoretically that carbonate reaction rates demand local equilibrium in the rock. We also present a new flow model of reactive transport that includes complicated carbonate-reaction chemistry, detailed reaction kinetics, ion exchange, and axial dispersion. For the first time, we predict local aqueous-species compositions and concomitantly, in-situ surface-chemistry species during forced displacement. Flow experiments on 6-cm Indiana limestone cores confirm model predictions.

Table 1 reveals the various species involving in carbonate chemistry along with equilibrium constants at ambient temperature [1]. Each of these species, in addition to sodium and chloride ions, must be tracked during flow through the porous reactive rock. This is a daunting task, especially when other reactive reservoir minerals are also present, such as anhydride and dolomite. We assume that the surface area of the matrix rock is not large enough for surface complexing reactions [1] to influence pore-space aqueous compositions. However, high-surfac- area clays may be present to initiate ion exchange between calcium and sodium ions at the rock surface. Accordingly, we write mass conservation for species i in a 1D porous calcite porous medium during frontal displacement as

does not copy (1)

where is species i pore molar concentration, is the surface ion-exchange molar concentration per solids volume, t is time, x is axial distance, is porosity, is superficial velocity (i.e., frontal advance rate), Diis the dispersion coefficient, is rock surface area per solids volume, and is the net production rate of mineral species i per unit solids surface area (mol/m2/s). Only sodium and calcium ions undergo ion exchange which is assumed to obey local equilibrium or

does not copy (2)

where the ion-exchange isotherm is specified by classical mass action. The key to understanding reactive-mineral behavior is through the mineral dissolution-reaction rate, . Here we consider only calcite as the reactive mineral. For small deviations from equilibrium, the dissolution rate of calcite (i.e., the difference between dissolution and precipitation rates) is given by [1]

does not copy (3)

where is the intrinsic dissolution reaction-rate constant (m/s), and in Table 1) is the solubility product of calcium carbonate. Based on the extensive experimental review of Morse and Arvidson [2] and others, we find that Eqn. 3 is valid and that is approximately 10-5 m/s under ambient conditions.

Not all terms in Eqn. 1 apply to each species. For example, chloride ions do not participate in ion exchange or in the species reactions in Table 1. For chloride ions, the ion-exchange and reaction-rate terms are absent in Eqn. 1. This means that chloride ions transport along the core as a salinity wave with no holdup [3]. Similar simplifications occur for other ions.

Solution of Eqn. 1 for all chemical species in Table 1 presents a significant numerical challenge. We adopt the extensively used PHREEQC speciation code of the US Geological Survey

[1,4]. PHREEQC is not constructed to handle flow processes directly. We extend it towards unsteady displacement in a 1D porous medium including chemical-reaction kinetics, ion exchange, equilibrium speciation, and axial dispersion. Predicted species profiles and histories depend on the Damköhler number for reaction, the Damköhler number for mass transfer, the dispersion Péclet number, the sodium-calcium cation exchange capacity and isotherm, and the equilibrium constants listed in Table 1. Upon injection of a low salinity composition, our calculations demonstrate a salinity wave that moves through the porous medium at the frontal advance rate. Species compositions behind the salinity front is influenced by the specifics of the ion-exchange isotherm and the initial and injected species compositions. Lagging the salinity wave is a self-spreading ion-exchange front [3]. Throughout the medium, reaction kinetics dissolves the rock and aqueous species equilibria adjust concentrations. However, for realistic reaction rate constants, net dissolution (precipitation) occurs only at the front face of the core (or reservoir) [1]. Thereafter, no dissolution occurs. We conclude that aqueous equilibration with carbonate rock is essentially instantaneous.

Comparison of the PHREEQC-based reactive transport model is made to simple displacement experiments in a 10-cm long Indiana limestone core at a high velocity of 100 ft/day to lower Damköhler numbers and probe for reaction-controlled dissolution. 1/10th pore-volume samples are analyzed for calcium and sodium by ICP, for total carbon my mass, and for pH by microelectrodes. Agreement between theory and experiment is excellent although the limestone core exhibited significant dispersion.

We conclude that in LSW injected flooding concentration is not that appearing in the reservoir pore fluids. For example, no matter the injected pH, the in-situ pH is close to 8. Given the injected fluid composition, our proposed reactive-transport model accurately estimates pore-fluid compositions actually occurring in the reservoir. This allows more realistic laboratory characterization experiments for evaluating the effectiveness of and mechanisms in LSW.

References

[1]. Maxim P. Yutkin, John Y. Lee, Mishra Himanshu, Tadeusz W. Patzek, and Clayton J.
Radke, Bulk and Surface Aqueous Speciation of Calcite: Implications for Low-Salinity
Waterflooding of Carbonate Reservoirs, SPEJ DOI: 10.2118/182829-PA

[2]. Morse, J. W. and Arvidson, R. S. 2002. The Dissolution Kinetics of Major Sedimentary

Carbonate Minerals. Earth Sci. Rev. 58 (1–2): 51–84.

[3]. Pope, G.A., Lake, L.W., Helferrich, F.G., Cation Exchange in Chemical Flooding: Part I
- Basic Theory without Dispersion, SPEJ, December 418-434 (1978).

[4]. Charlton, S. R. and Parkhurst, D. L. 2011. Modules Based On the Geochemical Model
PHREEQC for Use in Scripting and Programming Languages. Comput. Geosci. 37 (10):
1653–1663. https://doi.org/10.1016/j.cageo.2011.02.005.

Table 1 does not copy