Link to Research
Carbon-Based Energy Systems > CO2 Storage

Multiphase Flow of CO2 and Water in Reservoir Rocks

Start Date: September 2011
PDF version

Investigator

Sally Benson, Department of Energy Resources Engineering, Stanford University

Objective

This research program combines laboratory experiments, numerical methods and analytical solutions to evaluate the fundamental science underpinning sequestration of atmospheric carbon dioxide in saline aquifers and the multiphase flow of CO2, brine and to a lesser degree, oil.  As the research progresses, researchers will assess which, if any, modifications to currently accepted multiphase flow theories are needed and will develop approaches for reliably predicting field-scale performance.

Background

Carbon dioxide capture and sequestration (CCS) in deep geological formations is an important component of the portfolio of options for reducing greenhouse emissions. Saline aquifers have the largest sequestration capacity, as compared to oil and gas reservoirs or deep unminable coal beds. Saline aquifers are also more broadly distributed and thus, closer to more emission sources. However, unlike oil and gas reservoirs with proven seals that have withstood the test of time, saline aquifers must be carefully characterized to assure that CO2 will achieve high retention rates and remain permanently trapped in the subsurface. Improved fundamental understanding of multiphase flow (i.e., the simultaneous flow of more than one fluid phase through a porous medium) and trapping in CO2-brine systems is needed to take advantage of the large storage capacity of saline aquifers.

Approach

Optimizing the design and operation of injection projects will depend on knowledge of injectivity (i.e., the rate and pressure at which fluids can be pumped underground without fracturing the formation), trapping capacity, distribution of CO2 in the subsurface and the extent of the subsurface plume. A combination of multiphase flow experiments, numerical simulations, flow theory and reservoir characterization is used to address fundamental issues as shown in Figure 1.  Interaction and iteration among these four approaches improves the ability to identify and test new phenomena and techniques that can be applied to quantitative models.

Capillary pressure is one of the dominant forces in long-term sequestration, and a better understanding of the capillary heterogeneity and relative permeability of CO2-water systems is essential for accurate simulations ranging from the sub-core to the basin scale.

Figure 1

Figure 1: Illustration of the interrelated components used to study multiphase flow and trapping of CO2.

Recent work on core-scale multiphase flow properties of CO2 and water in sandstone rocks at reservoir conditions focuses on four components related to capillary forces:

(1) In situ measurements of capillary pressure and capillary heterogeneity
A novel method has been developed to measure drainage capillary pressure at the core and sub-core scale.  This improved computational tool enables research on the implications of small-scale capillary heterogeneity on CO2 plume movement, solubility trapping and capillary trapping.

(2) Enhanced trapping due to capillary barriers
This work investigates the impact of natural capillary heterogeneity in sandstone rock on CO2 saturation buildup and trapping.  CO2 core flooding experiments conducted at high pressure and temperature provide evidence for trapping due to capillary heterogeneity and suggest that depositional structures in the rock will impact the movement and steady-state distribution of a CO2 plume.

(3) Influence of capillary pressure treatment on solubility trapping
This work assesses the most appropriate treatment for the capillary entry pressure.  Differences in how capillary pressure curves are modeled can impact entry-pressure representations.

(4) Semi-analytical solutions for predicting CO2 saturations as a function of flowrate, interfacial tension and core geometry
The purpose of this work is to provide a simple approximate semi-analytical solution for predicting the average saturation during core flood experiments, thus avoiding the need for 3-D simulations. A new criterion has been established to identify the viscous-dominated regime at core scale. Variations of interfacial tension, core permeability, length of the core and the effects of buoyancy, capillary and viscous forces are also accounted in the theoretical solutions.