Scholarly article on topic 'Experimental Investigation of Supercritical CO2 Trapping Mechanisms at the Intermediate Laboratory Scale in Well-defined Heterogeneous Porous Media'

Experimental Investigation of Supercritical CO2 Trapping Mechanisms at the Intermediate Laboratory Scale in Well-defined Heterogeneous Porous Media Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Energy Procedia
Keywords
{"Capillary trapping" / "immiscible displacement experiments" / "surrogate fluids" / "x-ray attenuation method" / "heterogeneous porous media."}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Luca Trevisan, Ronny Pini, Abdullah Cihan, Jens T. Birkholzer, Quanlin Zhou, et al.

Abstract The heterogeneous nature of typical sedimentary formations can play a major role in the propagation of the CO2 plume, eventually dampening the accumulation of mobile phase underneath the caprock. From core flooding experiments, it is also known that contrasts in capillary threshold pressure due to different pore size can affect the flow paths of the invading and displaced fluids and consequently influence the build- up of non-wetting phase (NWP) at interfaces between geological facies. The full characterization of the geologic variability at all relevant scales and the ability to make observations on the spatial and temporal distribution of the migration and trapping of supercritical CO2 is not feasible from a practical perspective. To provide insight into the impact of well-defined heterogeneous systems on the flow dynamics and trapping efficiency of supercritical CO2 under drainage and imbibition conditions, we present an experimental investigation at the meter scale conducted in synthetic sand reservoirs packed in a quasi-two-dimensional flow-cell. Two immiscible displacement experiments have been performed to observe the preferential entrapment of NWP in simple heterogeneous porous media. The experiments consisted of an injection, a fluid redistribution, and a forced imbibition stages conducted in an uncorrelated permeability field and a homogeneous base case scenario. We adopted x-ray attenuation analysis as a non-destructive technique that allows a precise measurement of phase saturations throughout the entire flow domain. By comparing a homogeneous and a heterogeneous scenario we have identified some important effects that can be attributed to capillary barriers, such as dampened plume advancement, higher non-wetting phase saturations, larger contact area between the injected and displaced phases, and a larger range of non-wetting phase saturations.

Academic research paper on topic "Experimental Investigation of Supercritical CO2 Trapping Mechanisms at the Intermediate Laboratory Scale in Well-defined Heterogeneous Porous Media"

<®

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Energy Procedía 63 (2014) 5646 - 5653

GHGT-12

Experimental investigation of supercritical CO2 trapping mechanisms at the intermediate laboratory scale in well-defined

heterogeneous porous media

Luca Trevisan*,a, Ronny Pinib, Abdullah Cihanc, Jens T. Birkholzerc, Quanlin Zhouc,

Tissa H. Illangasekarea

aCenter for Experimental Study of Subsurface Environmental Processes (CESEP), Colorado School of Mines, 1500 Illinois Street, Golden, CO,

80401, USA

bPetroleum Engineering Department, Colorado School of Mines, 1500 Illinois Street, Golden, CO, 80401, USA cEarth Sciences Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA, 94720, USA

Abstract

The heterogeneous nature of typical sedimentary formations can play a major role in the propagation of the CO2 plume, eventually dampening the accumulation of mobile phase underneath the caprock. From core flooding experiments, it is also known that contrasts in capillary threshold pressure due to different pore size can affect the flow paths of the invading and displaced fluids and consequently influence the build- up of non-wetting phase (NWP) at interfaces between geological facies. The full characterization of the geologic variability at all relevant scales and the ability to make observations on the spatial and temporal distribution of the migration and trapping of supercritical CO2 is not feasible from a practical perspective. To provide insight into the impact of well-defined heterogeneous systems on the flow dynamics and trapping efficiency of supercritical CO2 under drainage and imbibition conditions, we present an experimental investigation at the meter scale conducted in synthetic sand reservoirs packed in a quasi-two-dimensional flow-cell. Two immiscible displacement experiments have been performed to observe the preferential entrapment of NWP in simple heterogeneous porous media. The experiments consisted of an injection, a fluid redistribution, and a forced imbibition stages conducted in an uncorrelated permeability field and a homogeneous base case scenario. We adopted x-ray attenuation analysis as a non-destructive technique that allows a precise measurement of phase saturations throughout the entire flow domain. By comparing a homogeneous and a heterogeneous scenario we have identified some important effects that can be attributed to capillary barriers, such as dampened plume advancement, higher non-wetting phase saturations, larger contact area between the injected and displaced phases, and a larger range of non-wetting phase saturations.

* Corresponding author. Tel.: +1-970-690-6591. E-mail address: luca.trevisan@gmail.com

1876-6102 © 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of the Organizing Committee of GHGT-12

doi: 10.1016/j.egypro.2014.11.597

© 2014TheAuthors. Publishedby Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of the Organizing Committee of GHGT-12

Keywords: Capillary trapping; immiscible displacement experiments; surrogate fluids; x-ray attenuation method; heterogeneous porous media.

1. Introduction

Geological storage of carbon dioxide needs to be globally deployed at the industrial scale in order to significantly reduce the anthropogenic emissions of this greenhouse gas to the atmosphere [1]. Even though several pilot tests have been successfully conducted in the field, as well as a number of commercial projects, there is still uncertainty related to the amount of gas that can be permanently stored in most geological repositories due to the scarcity of field data for testing the validity of numerical models.

It is known from the literature that capillary barriers inherent to most reservoir rocks can enhance plume immobilization and inhibit its upward migration [2, 3]; however, laboratory experiments showing the effect of capillary heterogeneity on plume propagation and entrapment at the intermediate scale are still limited. In this investigation we develop two-dimensional experiments conducted in intermediate-scale (centimeter to meter) flow cells aimed to represent a suitable intermediary for mimicking flow processes occurring between one-dimensional column tests and field-scale [4] and to study the effect of flow dimensionality on the distribution of residual saturation in controlled laboratory settings [5]. Eventually, these experiments will provide data sets to test the applicability of numerical multiphase flow models for prediction of long-term CO2 migration and trapping under field conditions [6]. Furthermore, the dependence on the saturation history of the capillary pressure and relative permeability has been shown to have an effect on the amount of residually trapped supercritical CO2 (scCO2) and on the long-term evolution of geological sequestration projects. However, the scale at which capillary trapping phenomena has been limited so far to the reservoir [7, 8] and the sub-core scale [9, 10]. In the immiscible displacement experiments presented here, we inject a non-wetting phase (NWP) in a gently-sloping sand reservoir initially saturated with the wetting phase; then the two fluids are allowed to redistribute until no further change in NWP is observed (days to weeks). Experiments are concluded by imposing a wetting phase flow (i.e. forced imbibition) to observe displacement of any mobile NWP and subsequent residual trapping of the plume.

Nomenclature

scCO2 Supercritical CO2

NWP Non-wetting phase

pnw/w Density (non-wetting/wetting phase)

^nw/w Dynamic viscosity (non-wetting/wetting phase)

IFT Interfacial tension

Oavg Average porosity

d50 Mean particle size

d60/d10 Uniformity coefficient

Snwr,max Maximum residual non-wetting phase saturation

2. Materials and Methods

2.1. Experimental setup

Following the research presented in a previous paper [5], where we conducted immiscible displacement experiments through homogenous sandpacks with variable permeabilities, here we compare two experiments with the goal of understanding the effects of heterogeneity and the influence of capillary barriers on migration and entrapment of a scCO2 plume through a brine-saturated reservoir. The experimental apparatus used to carry out the displacement experiments is illustrated in Fig. 1a, and consists of a rectangular flow cell with internal dimensions (L

x W x H) of 91.4 x 5.6 x 61 cm3. The details of the procedure to prepare the synthetic reservoir and constrain the boundary conditions, as well as the x-ray attenuation method, which was used to non-destructively measure the spatial and temporal variations of NWP saturation, are described in Trevisan et al. [5]. Each experiment consisted of a 5.5-hour injection of NWP, followed by a fluid redistribution period characterized by spontaneous imbibition at the trailing edge of the plume and slow drainage at the front edge facilitated by the 2° inclination of the reservoir. Subsequently, in order to maximize the contact area between the plume and the reservoir by displacing any remaining mobile NWP to regions that were previously un-invaded, a forced imbibition event was created by injecting the denser wetting phase into the reservoir. During both experiments, equivalent volumes of NWP were injected at comparable flow rates, while the fluid redistribution stages were prolonged until no further change in NWP saturation was observed.

2.2. Surrogate fluids

In order to avoid experimental complexity associated to high-pressure systems at the meter scale, we conducted the experiments with surrogate fluids at ambient conditions. We selected Soltrol 220 (dyed with Sudan IV and spiked with 10 %wt. 1-Iodoheptane) and an aqueous solution of glycerol (80 %wt.) to represent the injected (nonwetting) and displaced (wetting) phases, respectively, with the goal of matching the density and viscosity contrasts that are observed for the actual phases, i.e. scCO2 and brine, at reservoir conditions (Table 1).

Table 1. Summary of density and viscosity for surrogates at ambient conditions and for actual fluids at reservoir conditions.

phase p (kg m3) ^ (mPa- s) Hnw^Hw pnw^ pw IFT (mN m-1)

Soltrol 220 Glycerol-water 860 1210 4.9 61 0.08 0.71 15

scCO2 266-733a 0.023-0.061a 0.026-0.20a 0.22-0.75a 19.8b

Brine 945-1230a 0.195-1.58a

a estimates from Nordbotten et al. [11], T = 35-155°C, P = 10.5-31.5 MPa b measurement from Bennion and Bachu [12], T = 43°C, P = 20 MPa, brine salinity = 2.7% wt.

2.3. Heterogeneous system

For comparison purposes, we present experimental results for a spatially uncorrelated permeability field experiment (Fig. 1b) and a homogeneous base case scenario with identical domain extents. The heterogeneous sandpack configuration was generated with three uniform silica sands of differing sieve sizes having the following volumetric proportions: 14% of #30/40 (coarse), 69% of #40/50 (medium), and 17% of #50/70 (fine). The packing grid was defined by 224 cells (28 columns by 8 rows) and each grid block had dimensions of 2.5 x 5.6 x 2 cm3. The distribution of the material zones was chosen with the goal of maximizing the number of permeability contrasts encountered by the NWP plume. Furthermore, the pressure potential driving the injection was maintained lower than the capillary entry pressure of the fine sand, in order to enhance the saturation contrast observed in the porous medium. For the homogeneous scenario, the reservoir was packed with the medium sand, which represented the background material for the heterogeneous experiment. The properties of the sands are listed in Table 2.

Table 2. Physical properties of the silica sand grades used in the displacement experiments.

Material sieve size k (m2) ^av2 (-) d5o (mm) d6o/dio (-) Q r,max

1 (coarse) #30/40 1.14 x 10-10 a 0.354 0.45b 1.22b 0.26

2 (medium) #40/50 6.42 x 10-11 a 0.390 0.28b 1.21b 0.22

3 (fine) #50/70 3.46 x 10-11 a 0.379 0.23b 1.28b 0.25

a from Zhang et al. [13].

_ — 1 "3 - y "3 - r ~"3 1 3 1 1 3 1 1 1 3 1 T ~~ 3 1 1- 1

dl ,3 1 3 1 3 1 1 3 1 1 3 3 1 3 3 1 3 1 3 1 3 3 3 1 3 3 1 3 1 3 1 3 1 3

1 3 1 3 3 1 3 1 1 3

__ 1 3 1 3 1 3 _ — — — —' L —

— — —

0 50 100 150 200 t

Fig. 1. (a) Sketch of the experimental apparatus with dimensions of the synthetic reservoir; (b) Spatial distribution of the coarse (no. 1), and fine (no.3) sand zones. Background sand is no.2 with intermediate permeability. Scale bar in mm.

3. Results

Due to the limitation of the x-ray device to capture the migration of the plume at early times, photographic images are convenient to visually monitor its spatial evolution during the injection stage. Fig. 2 shows the extent of the injected plume for the two sand configurations at 5.5 hours, once the injection of NWP has stopped by disconnecting the constant pressure reservoir from the well. Even though from a qualitative perspective, it can be observed that the abrupt sand interfaces in the heterogeneous reservoir have a slowdown effect on the overall distance traveled by the plume, whereas some coarse sand zones retain larger amounts of NWP. After injection in the homogeneous experiment, a large portion of the plume stopped spreading, while the front edge continued propagating leaving behind a pinned interface. Similar results have been observed in immiscible lock-exchange experiments using glass beads [14], suggesting a possible mechanism for immobilizing scCO2 plumes.

-800 -700 -600 -500 -100 -300 -200 -100 0 -800 -700 -600 -500 -400 -300 -200 -100 0

X (mm) X (mm)

Fig. 2. Photographic images of the injected plume at the end of the injection stage (i.e., t = 5.5 hours): (a) homogeneous experiment; (b) heterogeneous experiment.

Once the injection stopped, the propagation of the plume slowed down significantly, allowing for an accurate quantification of NWP saturations by means of x-ray attenuation analysis. As a result, Fig. 3 illustrates the saturation distribution of NWP before (left column) and after (right column) the forced imbibition event for the homogeneous (top row) and heterogeneous (bottom row) reservoirs. As seen in Fig. 2a, during the homogeneous experiment the plume was able to reach the right boundary through a very thin tongue propagating underneath the fine sand layer, leading to an almost complete immobilization of the plume after 2 weeks of fluid redistribution (Fig. 3a); consequently, the forced imbibition event did not have a noticeable effect on NWP trapping (Fig. 3b). On the other hand, during the fluid redistribution of the heterogeneous scenario, the plume was retained completely inside the reservoir, building up at saturations higher than residual behind capillary barriers (Fig. 3c), a result that has been shown to have a large storage potential by numerical studies [3]. Once completed the forced imbibition, some regions of the plume had been trapped at saturations slightly above Snwr,max, leading to a larger trapped volume with respect to the homogeneous case.

(a) (b)

Fig. 3. NWP saturation contour maps gathered via x-ray attenuation method at different times: (a, b) homogeneous experiment before and after forced imbibition; (c, d) heterogeneous experiment before and after forced imbibition.

Fig. 4 illustrates the distribution of NWP saturations measured before and after forced imbibition in the two experiments. In the heterogeneous case the NWP plume occurs over a wide range of saturations prior to the forced imbibition event, while after the bulk of the plume has been swept out of the reservoir, most of the NWP is immobilized at lower saturations. However, compared to the homogeneous scenario, a large portion of the plume still exists at saturations higher than the maximum residual value for these sands (0.22-0.26).

Fig. 4 NWP saturation distribution before and after the forced imbibition event for (a) homogeneous and (b) heterogeneous experiments. The dotted line highlights the cut-off value separating disconnected and mobile NWP.

Fig. 5 shows the temporal evolution of NWP saturation measured at six representative points. The position of the observation points was selected to help visualizing the amount of trapped NWP as a result of the combined effects of capillary barriers and forced imbibition. As point A corresponds to a low permeability zone, this location is unaffected by the NWP plume until the capillary pressure exerted by the forced imbibition event overcomes the entry pressure of #50/70 sand. Point B corresponds to a coarse sand zone and shows a rapid increase followed by stabilization of NWP saturation, which is due to the presence of finer sand downstream. At this location, the forced imbibition event displaces the mobile NWP away, leaving behind a residual NWP saturation lower than 0.2. Despite their proximity and location within #40/50 sand, points C and D show different trends: the first point is positioned upstream of a fine sand zone, which is responsible of the accumulation of NWP saturation higher than residual, whereas the second point is not affected by any capillary barrier, exhibiting low NWP saturation until the pores at this location get invaded by the "wave" of bulk NWP displaced by the forced imbibition event. This effect is even more pronounced at farther locations from the well. Points E and F are located in medium and coarse sand zones, respectively, and are both followed downstream by fine sand; following the forced imbibition event, the maximum NWP saturation reached at these locations is higher than the maximum value achieved at points closer to the well, leading to higher trapped saturations.

TO 0.6

LO CL 0.4

° A(fine sand)

* B (coarse sand)

0 100 200 300 400

Time (hours)

Fig. 5 NWP saturation evolution at selected observation points for the heterogeneous experiment; dotted line highlights the onset of forced

imbibition event.

4. Concluding remarks

The present study consists of an experimentally driven analysis of the influence of well-defined heterogeneities on migration and entrapment under capillary-dominated flow conditions of a scCO2-surrogate fluid in an unconsolidated sand reservoir initially saturated with a brine-surrogate fluid. Although we are aware of the important role played by small-scale heterogeneity in controlling multiphase flow through reservoir rocks, this study has focused on the effects of larger scale features that eventually will also contribute to the final distribution of a scCO2 plume in a brine-saturated reservoir. By comparing two immiscible displacement experiments performed in homogeneous and heterogeneous sand configurations the following effects can be attributed to capillary barriers: (a) a dampened plume advancement, (b) higher NWP saturations, (c) a larger contact area of the plume with the resident wetting phase, (d) and a larger range of NWP saturations. We have also observed and quantified the effects on NWP saturation of forced imbibition representing a "chase brine" event in combination with capillary barriers. These mechanisms are certainly relevant at the field scale, since they enhance storage efficiency and may eventually promote dissolution trapping of CO2 into formation brine.

Acknowledgements

Funding for this research is provided by the U.S. Department of Energy through the National Energy Technology Laboratory's CO2 sequestration R&D Program under grant DE-FE0004630 and National Science Foundation Award#: EAR-1045282 through the Hydrologic Sciences Program.

References

[1] Pacala, S. and R. Socolow, Stabilization wedges: Solving the climate problem for the next 50 years with current technologies. Science, 2004. 305(5686): p. 968-972.

[2] Krevor, S.C.M., et al., Capillary heterogeneity trapping of CO2 in a sandstone rock at reservoir conditions. Geophysical Research Letters, 2011. 38.

[3] Saadatpoor, E., S.L. Bryant, and K. Sepehrnoori, New Trapping Mechanism in Carbon Sequestration. Transport in Porous Media, 2010. 82(1): p. 3-17.

[4] Oostrom, M., J.H. Dane, and T.W. Wietsma, A review of multidimensional, multifluid, intermediate-scale experiments: Flow behavior, saturation imaging, and tracer detection and quantification. Vadose Zone Journal, 2007. 6(3): p. 610-637.

[5] Trevisan, L., et al., Investigation of mechanisms of supercritical CO2 trapping in deep saline reservoirs using surrogate fluids at ambient laboratory conditions. International Journal of Greenhouse Gas Control, 2014. 29(C): p. 35-49.

[6] Cihan, A., et al., A connectivity-based approach for modeling hysteresis in macroscopic two-phase flow properties, in 12th International Conference on Greenhouse Gas Control Technologies. 2014: Austin, TX.

[7] Flett, M., R. Gurton, and I. Taggart, The Function of Gas-Water Relative Permeability Hysteresis in the Sequestration of Carbon Dioxide in Saline Formations, in SPE Asia Pacific Oil and Gas Conference and Exhibition. 2004, Society of Petroleum Engineers Perth, Australia.

[8] Kumar, A., et al., Reservoir Simulation of CO2 Storage in Deep Saline Aquifers. SPE Journal, 2005. 10(3): p. 336-348.

[9] Perrin, J.C. and S. Benson, An Experimental Study on the Influence of Sub-Core Scale Heterogeneities on CO2 Distribution in Reservoir Rocks. Transport in Porous Media, 2010. 82(1): p. 93-109.

[10] Andrew, M., B. Bijeljic, and M.J. Blunt, Pore-scale imaging of trapped supercritical carbon dioxide in sandstones and carbonates. International Journal of Greenhouse Gas Control, 2014. 22: p. 1-14.

[11] Nordbotten, J.M., M.A. Celia, and S. Bachu, Injection and storage of CO2 in deep saline aquifers: Analytical solution for CO2 plume evolution during injection. Transport in Porous Media, 2005. 58(3): p. 339-360.

[12] Bennion, D.B. and S. Bachu, Dependence on Temperature, Pressure, and Salinity of the IFT and Relative Permeability Displacement Characteristics of CO2 Injected in Deep Saline Aquifers, in 2006 SPE Annual Technical Conference and Exhibition. 2006, Society of Petroleum Engineers: San Antonio, Texas, U.S.A.

[13] Zhang, C.Y., C.J. Werth, and A.G. Webb, Investigation of surfactant-enhanced mass removal and flux reduction in 3D correlated permeability fields using magnetic resonance imaging. Journal of Contaminant Hydrology, 2008. 100(3-4): p. 116-126.

[14] Zhao, B.Z., et al., Interface pinning of immiscible gravity-exchange flows in porous media. Physical Review E, 2013. 87(2).