Elsevier

Advances in Water Resources

Volume 52, February 2013, Pages 190-206
Advances in Water Resources

Relative permeability hysteresis and capillary trapping characteristics of supercritical CO2/brine systems: An experimental study at reservoir conditions

https://doi.org/10.1016/j.advwatres.2012.06.014Get rights and content

Abstract

We present the results of an experimental study on the effects of hysteresis on capillary trapping and relative permeability of CO2/brine systems at reservoir conditions. We performed thirty unsteady- and steady-state drainage and imbibition full-recirculation flow experiments in three different sandstone rock samples, low- and high-permeability Berea and Nugget sandstones. The experiments were carried out at various flow rates with both supercritical CO2 (scCO2)/brine and gaseous CO2 (gCO2)/brine fluid systems. The unsteady-state experiments were carried out with a wide range of flow rates to establish a broad range of initial brine saturations (Swi). This allowed investigation of the sensitivity of residual trapped CO2 saturation (SCO2r) to changes in Swi. The values were successfully compared with those available in the literature. For a given Swi, the trapped scCO2 saturation was less than that of gCO2 in the same sample. This was attributed to brine being less wetting in the presence of scCO2 than in the presence of gCO2. Post-imbibition dissolution of trapped CO2 and formation of dissolution front was also investigated. During the steady-state experiments, scCO2 and brine were co-injected with monotonically increasing or decreasing fractional flows to perform drainage and imbibition processes. We carried out seven sets of steady-state flow tests with various trajectories generating a comprehensive group of relative permeability hysteresis curves. The scanning curves revealed distinct features with potentially important implications for storage of scCO2 in geological formations. For both series of experiments, the ratio of SCO2r to initial CO2 saturation (1- Swi) was found to be much higher for low initial CO2 saturations. The results indicate that very promising fractions (about 49 to 83%) of the initial CO2 saturation can be stored through capillary trapping.

Highlights

► A major portion of the initial SCO2 can be trapped by brine imbibition. ► Relative permeability hysteresis was characterized using steady-state method. ► A rapid reduction in imbibition relative permeability of scCO2 was observed. ► For a given Sw, imbibition krw was greater than drainage value.

Introduction

Capturing CO2 from stationary sources and injecting it into underground geologic formations for storage purposes is currently receiving significant attention as a potentially viable option for geologic storage and ultimate sequestration of significant quantities of CO2. Among various geologic formations, deep saline aquifers are considered important CO2 sinks because of their potential storage capacity for large volumes of carbon dioxide. It has been reported that saline aquifers have a global storage capacity of 400–10000 Gt of CO2 [1], [2].

In the sequestration process, CO2 captured from commercial or industrial operations is injected into deep salt water aquifers to prevent its emission to the atmosphere. The depth of the aquifers provides the pressurized environment necessary to keep the CO2 at supercritical conditions, namely at pressure and temperature conditions above the critical point of CO2, (Tc = 30.978 °C, Pc = 7.377 MPa [3]). Conditions that support the existence of supercritical CO2 (scCO2) should be present at depths greater than about 750 m. The average depth of the formations targeted for CO2 storage range from about 1000 m, for shallow sites, to 3000 m for deeper cases. Therefore density of scCO2 and brine may vary from 266 to 733 kg/m3 and 945 to 1230 kg/m3, respectively, depending on pressure and temperature conditions [4].

Carbon dioxide injected into a geologic formation may be sequestrated through four mechanisms [5]: (1) structural and stratigraphic trapping: CO2 may stay as a large continuous plume which moves while more of the fluid is injected. The plume gets trapped below low permeability caprocks. At the pore-level, this is equivalent to well-connected clusters of CO2 that reside at the centers of the pores and throats; (2) residual trapping: CO2 may also stay as a stagnant residual phase. This is the CO2 that gets trapped, for instance, after CO2 injection is terminated. It is a consequence of displacement process named imbibition that may take place due to pressure gradients, background velocity of aquifer or chase brine injection. During this process, CO2 gets trapped, in water-wet systems, by displacement mechanisms such as snap-off and pore-body filling. At the pore-level, the trapped CO2 resides in randomly distributed trapped stagnant clusters. The elements of these clusters are CO2 sitting at the centers of larger pores and throats as the non-wetting phase; (3) solubility trapping: CO2 can also dissolve in brine forming an acidic solution. The amount of dissolution is controlled by pressure, temperature, salinity of brine, dispersion, and Darcy velocity of the aqueous phase; and (4) mineral trapping: acidic solution formed owing to dissolution of CO2 in brine may react with the hosting rocks producing secondary minerals that may precipitate indirectly sequestrating carbon. These reactions are relatively slow and therefore this sequestration mechanism becomes important over larger time scales. Carbon sequestrated through this mechanism, however, has the lowest risk of leakage.

From the above-mentioned geologic storage mechanisms, it is evident that stratigraphic and capillary trapping of CO2 are the principal short-term trapping processes. The dynamics of these two mechanisms are directly controlled by their respective relative permeability functions and associated hysteresis as well as residual trapping trends. Therefore, it is essential to carefully investigate these properties to develop an improved understanding of their characteristics under different flow conditions relevant to those seen in deep saline acquires.

Over the last decade several experimental studies of CO2/brine flow systems in various rock samples have been performed. Researchers have used unsteady-state [6], [7], [8], [9], [10], [11], [12], [13] and steady-state [14], [15], [16], [17] core-flooding techniques to measure initial brine and residual CO2 saturations, drainage and imbibition relative permeabilities, and other system properties under different flow, temperature, and pressure conditions. Table 1, Table 2 provide a list of these experimental studies with their experimental techniques and the properties they report.

In this work, we present the results of an experimental study that, for the first time, carefully characterizes relative permeability hysteresis using steady-state method. We also use unsteady-state technique to investigate sensitivity of residual CO2 trapping to variations in initial brine saturation at different flow conditions and fluid properties (scCO2 and gCO2). We study drainage and imbibition flow experiments in three different outcrop sandstone core samples using a reservoir-conditions core-flooding apparatus. All the experiments were performed through vertically-placed cores while a CT scanner is used to measure in-situ saturations.

This paper is structured as follows. First we present a section on the rock samples and fluids as well as experimental conditions, setup, and procedure used in this study. We then include two large categories of results generated using (A) unsteady-state and (B) steady-state core-flooding techniques. In each category we present findings for displacement of brine due to CO2 injection (to establish Swi) and trapping of CO2 due to brine flood. We include results for both supercritical and gaseous CO2 and discuss the trends using pore-level physics and displacement mechanisms. A section on dissolution of trapped scCO2 is also included that shows the formation of dissolution front along the length of the sample at various stages of the process.

Section snippets

Experiments

In this section, we present detailed information regarding the material and experimental conditions, setup, and procedure used in this study.

Results

In this section we present results of the two categories of flow experiments mentioned earlier, i.e., (A) Unsteady-state, and (B) Steady-state tests. Table 4 lists the number of experiments that were performed along with their corresponding experimental conditions.

Conclusions

We used a robust full-recirculation core-flooding system to perform thirty drainage and imbibition CO2/brine flow experiments at reservoir conditions in three different sandstone core samples. The flow test were performed at various flow conditions and fluid properties (gCO2 and scCO2). Both unsteady- and steady-state core-flooding techniques were employed. We used a medical CT scanner to measure in-situ saturations during the experiments and developed a specific technique to maintain an

Acknowledgment

We gratefully acknowledge financial support of the Department of Energy, Encana, and The School of Energy Resources and the Enhanced Oil Recovery Institute at the University of Wyoming.

References (29)

  • IPCC, IPCC Special Report on Carbon Dioxide Capture and Storage. Cambridge: Cambridge University Press;...
  • Bennion B, Bachu S. Relative permeability characteristics for supercritical CO2 displacing water in a variety of...
  • Bachu S, Bennion B. Effects of in-situ conditions on relative permeability characteristics of CO2-brine systems....
  • B. Bennion et al.

    Drainage and imbibition relative permeability relationships for supercritical CO2/brine and H2S/Brine systems in intergranular sandstone, carbonate, shale and anhydrite rocks

    SPE Res Eval Eng

    (2008)
  • Cited by (259)

    View all citing articles on Scopus
    View full text