Article pubs.acs.org/est
CO2 Exsolution from CO2 Saturated Water: Core-Scale Experiments and Focus on Impacts of Pressure Variations Ruina Xu, Rong Li, Jin Ma, and Peixue Jiang* Key Laboratory for CO2 Utilization and Reduction Technology of Beijing, Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, and Department of Thermal Engineering, Tsinghua University, Beijing 100084, China S Supporting Information *
ABSTRACT: For CO2 sequestration and utilization in the shallow reservoirs, reservoir pressure changes are due to the injection rate changing, a leakage event, and brine withdrawal for reservoir pressure balance. The amounts of exsolved CO2 which are influenced by the pressure reduction and the subsequent secondary imbibition process have a significant effect on the stability and capacity of CO2 sequestration and utilization. In this study, exsolution behavior of the CO2 has been studied experimentally using a core flooding system in combination with NMR/MRI equipment. Three series of pressure variation profiles, including depletion followed by imbibitions without or with repressurization and repetitive depletion and repressurization/imbibition cycles, were designed to investigate the exsolution responses for these complex pressure variation profiles. We found that the exsolved CO2 phase preferentially occupies the larger pores and exhibits a uniform spatial distribution. The mobility of CO2 is low during the imbibition process, and the residual trapping ratio is extraordinarily high. During the cyclic pressure variation process, the first cycle has the largest contribution to the amount of exsolved CO2. The low CO2 mobility implies a certain degree of self-sealing during a possible reservoir depletion.
■
drive in earlier studies. The solution gas drive,16 referring to the exsolution of lighter hydrocarbons from crude oil, has been investigated systematically in the literature based on theoretical, numerical, and experimental approaches.17−24 These earlier studies mainly focused on the mechanisms of the solution gas drive, i.e., instantaneous nucleation and mass transfer controlled growth of a bubble or droplet, expansion results from compressibility. However, the significantly lower liquid phase viscosity and higher interfacial tension of the CO2−H2O fluid pair25 compared to that of the typical gas-oil fluid pair could possibly result in different quantitative exsolution behavior. Those studies specifically focusing on exsolution phenomena of the CO2−H2O fluid pair, at both low and high pressure conditions, caused by pressure variations or by thermal perturbations, were given more attention in recent years. Enouy et al. conducted a column experiment and relevant numerical modeling on the CO2 supersaturated water injection process, but the pressure applied in their study was slightly higher than the ambient pressure.26 Luhmann et al. and Tutolo et al. conducted high pressure experiments of CO2 exsolution caused by the increasing temperature.27,28 The CO2 trapping
INTRODUCTION The technology used for Carbon Capture, Utilization, and Storage (CCUS) was proposed for greenhouse gas mitigation since climate change is considered a challenging problem.1,2 CO2 can be sequestered in deep or shallow aquifers3−8,13 and depleted oil or gas reservoirs9 and be used to enhance coal beds methane recovery.10The depth, pressure, and temperature of target reservoirs at various CO2 utilization and storage processes are in the range from 380 to 2500 m, 2 to 28 MPa, and 20 to 100 °C, respectively, as listed in Table S1. Accordingly, the state of CO2 covers the gas state to supercritical state. When CO2 is injected into the reservoirs, reservoir pressure depletion can be caused by various events, i.e., stopping the injection, leakage due to the failure of the caprock or well integrity,11,12or withdrawing too much brine, which was intended to balance the reservoir pressure.6−8,13,14 Due to the dependence on pressure of CO2 solubility in the aqueous (H2O rich) phase, pressure depletion could cause the exsolution of CO2 from the aqueous phase.15 The amounts of exsolved CO2 and the migration of the aqueous and exsolved CO2 phases, which are influenced by the pressure reduction and the subsequent secondary imbibition process, have a significant effect on the stability and capacity of CO2 sequestration. The mechanisms of the CO2 exsolution process in the aqueous phase are considered to be relevant to the solution gas © 2015 American Chemical Society
Received: Revised: Accepted: Published: 14696
August 7, 2015 October 27, 2015 October 28, 2015 October 28, 2015 DOI: 10.1021/acs.est.5b03826 Environ. Sci. Technol. 2015, 49, 14696−14703
Article
Environmental Science & Technology
distribution when the spatial resolution of the imaging is not fine enough to capture it. This study specifically focuses on the behavior of CO2 exsolution from CO2 saturated water and the two phase migration during an imbibition process followed by exsolution, in which the pressure varies nonmonotonically or oscillates cyclically. The pressure range in this study is spanned from 6.3 to 2 MPa, which is relevant to the background of CO2 storage or utilization in shallow reservoirs.6−8 We designed various pressure changing profiles to investigate the influence of the fluid mobility and distribution on the complex pressure variations at the core-scale using online NMR/MRI equipment combined with a high pressure core flooding system.
mechanisms and relative permeability reductions caused by pressure-induced and temperature-induced phenomena are similar. Perrin et al. conducted an experiment of the CO2 exsolution process from an aqueous phase, due to monotonic depression of pressure at CO 2 sequestration reservoir conditions, using the X-ray CT equipment.29 Following their first report, a measurement of the relative permeability during the drainage process,30which occurred simultaneously with the depletion induced exsolution, and a relevant pore scale visualization experiment using a 2D micromodel were conducted by Zuo et al.31 The experimental results indicated that,29−32in terms of mobility and the size of the CO2 ganglia, the exsolved CO2 phase behaves very differently from the CO2 phase directly injected into the core33−37or micromodel.38,39The above-mentioned exsolution studies were the result of the monotonic pressure (or temperature) variation profiles. However, in a practical CO2 sequestration project, nonmonotonic change or cyclic oscillation of local pressure can be observed in various cases, i.e., restarting a stopped injection well or a cyclic change in the injecting flow rate. Understanding the influence of the different pressure variation profiles on the behavior of CO2 exsolution requires a more comprehensive investigation. An imbibition process with or without repressurization or cyclic pressurization/depressurization processes could possibly happen after the depressurization induced CO2 exsolution. The quantitative characterization of the two phase migration of CO2 during these processes has significant importance for large scale modeling to predict reservoir response to pressure variation. The behavior of the secondary imbibition process, after primary drainage by CO2 at the core-scale, has been investigated extensively in previous studies in order to estimate the quantitative amount of residual trapping (also called capillary trapping).40−44Nevertheless, the imbibition after exsolution process is quite different from those processes that occur after primary drainage. The residual CO2 saturation of an imbibition process following an exsolution process has been investigated by Zuo et al., and an extraordinarily high residual trapping fraction has been observed.45However, to the best of our knowledge, none of those previous studies characterized the fluid distribution and mobility after the repressurization or cyclic pressurization-depressurization followed by a first step exsolution process. Moreover, extensive studies have suggested that core-scale information needs to be measured during core flooding tests, including pore-size distribution,50 local and average porosity, 51 and distribution of aqueous phase saturation.52 Investigating the fluid flow and spatial distribution in saturated or unsaturated porous media using the Nuclear Magnetic Resonance (NMR), Magnetic Resonance Imaging (MRI), and X-ray CT techniques have been proposed over several decades.46−49 The high pressure core flooding experiment and online NMR/MRI or X-ray CT visualization techniques have been combined to visualize the fluid saturation profile and to obtain a more comprehensive understanding than the regular core flooding experiments provide.29,30,33,34,37,53,54 The NMR/MRI and X-ray CT have their advantages, respectively. The CT equipment has robustness to the magnetic sensitive elements in the rock sample or aqueous solutions and faster imaging speed than MRI. On the other hand, NMR/MRI equipment can obtain the signal relaxation properties which are related to the fluid surface volume ratio and thus provide more information on pore scale fluid
■
EXPERIMENTAL SYSTEM AND PROCEDURE System Setup. The diagram of the experimental system is presented in Figure S1 in the Supporting Information (SI), which is part of the experimental system shown in our previous publication.37The core sample was placed horizontally in a core holder, which is designed for the NMR scanner using PEEK material. The core sample was wrapped and flexed in a thermal shrink plastic, with negligible NMR signal response, to prevent flow between the pore space and the confining space. The effective confining pressure is controlled by a high pressure manual syringe pump. The value of effective confining pressure is adjusted as 2 MPa in this study. The fluid was equilibrated at 6.3 MPa and 27 °C in the equilibrium cell. A high pressure syringe pump (SSI/ Laballiance Series 1500) supplied pre-equilibrated CO 2 saturated water or DI water to the core sample at the designed flow rate. The pore pressure is controlled by the backpressure regulator (Jasco BP-2080-M) with a digital controller and is measured at the inlet and outlet of the core with two manometers (EJA440A, span: 6−16 MPa, accuracy: ±0.12% of span). The flow rate and the pressure reduction across the core sample can be measured by the mass flowmeter (Bronkhorst M12, 0−200 g/h, 0.5% full-scale accuracy) and the differential pressure transducer (Honeywell, STD924-F1A00000-MB, 100 kPa, 0.065% full-scale accuracy), respectively. The experimental temperature was measured using the Platinum resistance transducer (Pt100, 0.1 K accuracy). Rock and Fluids. A homogeneous Berea sandstone core, with a 24.58 ± 0.02 mm diameter and a 50.03 ± 0.02 mm length, was chosen as the test section. The water permeability (690 mD, measured under initial pressure, 6.3 MPa) and porosity (21.1%) were measured using the steady state method and the weighing method, respectively. The CO2 saturated water was prepared using 99.99% purity CO2 and DI water, with total ion concentration lower than 1 ppm, since certain types of ions can alter the NMR signal significantly. The FC40 (3 M Fluorinert) was selected as the confining fluid, to avoid additional signal influence on the NMR scanning. To eliminate the background NMR signal, the FC40 is tested and is proven to be generating a NMR negligible signal. NMR Measurement and MR Image. A nuclear magnetic resonance imaging scanner (Niumag, MesoMR23-060H-I, 21.3 MHz, 0.5 ± 0.05 T) has been set up for online visualization images and quantitative measurement. A CPMG (Carr− Purcell−Meiboom−Gill)55 impulse sequence has been applied to characterize the average core saturation and pore-water distribution. According to the NMR theory, the longer T2 relaxation time of the signal represents the smaller surface 14697
DOI: 10.1021/acs.est.5b03826 Environ. Sci. Technol. 2015, 49, 14696−14703
Article
Environmental Science & Technology Table 1. Procedure and Information for the Experimental Cases
Figure 1. MR imaging of the local aqueous phase saturation at the various depleted pressures of Series A. (Flow direction is from left to right, and the length scale of the image is 50.0 mm × 24.6 mm.)
volume ratio of the water. Therefore, the T2 curve calculated can capture the information on pore size distribution of porous media in the water saturated case and the amount of the aqueous phase remaining inside the corresponding pore sizes in the unsaturated case.56 The additional description of T2 relaxation theory, the calculation process of fluid saturation, and the calibration of NMR signal can be found in the SI. In this study, two MR imaging sequences, namely T2 mapping and a Multislice Spin Echo Imaging Sequence (MSE), were applied to obtain the spatially resolved T2 relaxation time and the 1H nuclei density imaging, respectively. The images were taken of a 24.6 mm height and 50.0 mm length area, and the slice thickness excited selectively by the NMR impulse was 5.0 mm. The spatial resolution of the NMR images is 0.44 mm (horizontal direction) × 0.47 mm (vertical direction) per pixel. T2 mapping represents an image of the spatially resolved T2 relaxation time.57The T2 mapping of the Berea sandstone in this study is provided in Figure S2 in the SI. The homogeneous T2 mapping implies good homogeneity of the core sample. The MSE impulse sequence has been applied to image local aqueous phase content, which is proportional to the intensity of the 1H
NMR signal. The image of the aqueous phase saturation distribution can be obtained by dividing the intensity value of the unsaturated image with the intensity value of the 100% saturated image at each pixel. Experimental Procedure and Cases. Three series of experiments have been proposed in this study (Table 1). The Series A experiments were conducted using the following steps: (1) inject the aqueous phase equilibrated to 6.3 MPa for at least 20 pore volumes; (2) close the inlet valve and adjust the back pressure controller to the target depletion pressure; and (3) inject the aqueous phase equilibrated to the depleted pressure with a 3 mL/min flow rate and the low Ca number (=μu/σ, ranges from 1.5 × 10−6 to 3 × 10−6 depending on the imbibition pressure) which ensures that the imbibition is capillary dominated. In the case of the Series B experiments, the only difference compared with Series A can be found in step (3): inject aqueous phase equilibrated to 6.3 MPa. For both Series A and B, four target depletion pressures (5.0, 4.0, 3.0, and 2.0 MPa) were applied. The Series C experiments are the repetitive operation of the Series B experiments but with a 5 MPa target depletion 14698
DOI: 10.1021/acs.est.5b03826 Environ. Sci. Technol. 2015, 49, 14696−14703
Article
Environmental Science & Technology pressure. Since there were no resaturating and restarting operations, each cycle was initiated at the condition that the previous cycle ended at. A quantitative CPMG scan and an imaging scan were obtained at the end of each step in the various cases. The permeability measurements were conducted at the end of the imbibition processes, which is depicted as the red line shown in the procedure diagrams in Table 1. In the case of the depletion processes, the steady states were determined by the negligible variation of the NMR measured aqueous phase saturation (less than 1%) and the negligible gas flow rate at the outlet. Similarly, the steady states of the aqueous phase injection (imbibition) steps were determined by the variation of the aqueous phase saturation and the differential pressure (less than 1% and 2% respectively), and the injection was lasted to at least 20 pore volumes.
■
EXPERIMENTAL RESULTS AND DISCUSSION The Aqueous Phase Saturation Distribution after the Exsolution Processes. The NMR images of the local aqueous phase saturation distribution, after the exsolution process at the depleted pressure, 4 and 2 MPa, are presented in Figure 1. The results show the lower depleted pressure and the lower final aqueous phase saturation. The saturation image shows significant spatial uniformity of the exsolved CO2 phase distribution along the whole core sample regardless of the depletion pressure in the experimental cases. This confirms that the CO2 phase exsolves uniformly in a relatively homogeneous porous media. In previous studies on the pure hydrological displacing process of the CO2−H2O binary system at the corescale, a significant saturation gradient along the flow direction can be observed.33,34 This result suggests that the CO2 phase nucleated and grew uniformly in the multiple nucleation sites homogeneously distributed along the core. The Pore Size Occupancy after the Exsolution Processes. The exsolved CO2 phase preferentially occupies the larger pores and gradually invades the finer pores as the depleted pressure decreases (Figure 2). According to a theoretical study, the bubble can grow only if the bubble size is greater than a certain minimum radius, which is inversely proportional to the aqueous phase supersaturation.17 The size of the pore containing the growing bubble also has to be greater than the minimum growth radius. Therefore, the bubble prefers to grow in the less confined, larger pores. On the other hand, during the depressurization process, the growth and expansion of CO2 will cause the capillary dominated drainage. Similar to the hydrological drainage process with the low Ca number in previous core flooding37 and micromodel experiments,38 in cases of hydrophilic samples, the CO2 phase occupies the larger pores preferentially. Since the exsolution behavior is influenced by both the nucleation/growth and the expansion/drainage mechanisms, the occupancy of the CO2 phase in larger pores is dominated by the common action of the bubble size selective growth and the lower capillary pressure. It is worth mentioning that the T2 distribution can also be affected by the possible change in pore geometry. Luhmann et al.27 and Tutolo et al.28 presented the permeability (and thus pore geometry) change during the CO2 exsolution process in both unconsolidated and consolidated porous media. However, in this study, the negligible difference of T2 curves measured at the aqueous phase saturated condition in the first and the last run of the experiment is observed as shown in Figure S4. This
Figure 2. T2 distribution of the remaining H2O NMR signal after the exsolution process in Series A. The larger T2 relaxation time represents the coarser pore size in general. The signal intensity corresponding to the T2 relaxation time longer than 150 ms shows significant reduction for depletion from the initial pressure to 5 or 4 MPa, but the intensity corresponding to the shorter T2 shows very little variation. As the depleted pressure continues to decrease to 3 or 2 MPa, the intensity corresponding to the T2 relaxation time, which ranges from 50 to 150 ms, also starts to decrease. The signal intensity of the T2 shorter than 50 ms does not show significant variation.
indicated that the change in the pore geometry is negligible in cases of this study. Fluid Saturation and Effective Aqueous Phase Permeability. The theoretical maxima of the CO2 saturation after depletion for Series A/B and C were calculated as eqs 1 and 2, respectively ρa, P XCO2, P0 − ρa, P XCO2, Pf 0 f Smax = ρCO , P (1) 2 f
Si + 1,max = Si ,max + (ρCO
2 ,6.3MPa
XCO2,6.3MPa − ρCO
2 ,5.0MPa
⎛ XCO2,5.0MPa)⎜1 − ⎝
ρCO ,5.0MPa ⎞ 2 S ⎟ ρCO ,5.0MPa i ,max ⎠ 2
ρCO
2 ,5.0MPa
(2)
where X represents the solubility at P0 or Pf, ρCO2 and ρa represent the density of CO2 and the aqueous phase at the relevant pressure, and i represents the number of exsolution cycles. For estimating the CO2 saturation after repressurization to 6.3 MPa in Series B and C, we assume that the repressurization process is a simple mass conservative density change process of the CO2 phase. Then, the estimated CO2 saturation after pressurization considering the compression effect, Scomp, turns out to be Scomp =
ρCO ,dep Sdep 2
ρCO , P
2 0
(3)
During the depletion step, the exsolved CO2 saturation increases as the depleted final pressures decrease (Figure 3). The exsolved CO2 saturation, Sdep, is 14% at Pdepleted = 5 MPa and 53−57% at Pdepleted = 2.0 MPa in Series A/B and increased from 14% to 28% as the number of pressure oscillation cycles 14699
DOI: 10.1021/acs.est.5b03826 Environ. Sci. Technol. 2015, 49, 14696−14703
Article
Environmental Science & Technology
increased from 1 to 5 in Series C. The CO2 saturation at the end of the imbibition process, Simb, is 14% at Pimb = 5 MPa and 47% at Pimb = 2.0 MPa in Series A, lower than 14% in Series B, and 11−21% in Series C. The differences between the experimental results, Sdep and Simb, are very small in Series A and become much larger in Series B and C. This is because the exsolved CO2 was compressed by the high imbibition pressure in Series B and C. At the end of the imbibition process, the aqueous phase effective permeability ranges from 47 mD to 366 mD, and, accordingly, the value of the relative permeability ranges from 0.07 to 0.53 (Figure 4). The aqueous phase saturation ranges from 0.5 to 0.9 depending on the exsolved CO2 saturation and whether or not the CO2 was recompressed.
Figure 4. Effective aqueous phase permeability at the end of the imbibition process in Series A, B, and C.
The Influence of Repressurization on CO2 Residual Saturation and Aqueous Phase Relative Permeability. Residual CO2 trapping often happens during the secondary imbibition process, which occurs after a primary drainage process. Accordingly, initial saturation and residual saturation correspond to the CO2 saturation after the primary drainage and the secondary imbibition. In this study, the residual trapping of CO2 can also be observed during the imbibition process, which happens after the exsolution process. In this case, the initial saturation is defined as the CO2 saturation at the end of the exsolution process. In Series A, the imbibed aqueous phase is equilibrated at the depleted pressure to eliminate the influence of exsolution/ dissolution or compression. At least 82% of the exsolved CO2 remained in the core after the imbibition process (Figure 5). Comparing with the fractions of residual trapping observed in the drainage/imbibition cases,34,37,40−43 the residual trapping fraction of the exsolved CO2 phase after imbibition is extraordinarily high. This is consistent with the observation of Zuo et al.45 The residual saturation in Series B is significantly lower than the initial saturation (Figure 5). This volumetric reduction possibly resulted from the CO2 mass loss during imbibition or the density variation due to compression. The initial saturation was corrected using eq 3 and was depicted as the open symbols shown in Figure 5. The ratio of residual saturation to the effective initial saturation also represents the mass ratio of the CO2 phase before and after the compression/imbibition process. The small difference between the residual CO2
Figure 3. CO2 saturation in Series A, B, and C. The Sdep at the depleted pressure Pdepleted = 5 MPa are measured independently in Series A and B and the first cycle of Series C, and the values are all 14%. This suggests the acceptable reproducibility of the independent experiments. The calculated theoretical maximum saturation at depleted pressure 2 MPa in Series A and B exceeded 1. 14700
DOI: 10.1021/acs.est.5b03826 Environ. Sci. Technol. 2015, 49, 14696−14703
Article
Environmental Science & Technology
variation profile does not change the low mobility of the exsolved CO2 phase. On the other hand, each individual cycle can also be considered as an independent experimental case of the exsolution and imbibition process with different initial CO2 saturations. The results show that the accumulative CO2 exsolution effect is significant when the initial CO2 saturation is low; but when the initial CO2 saturation increases, the lower aqueous phase saturation will result in a lower potential amount of CO2 that can be exsolved. The higher initial CO2 saturation will cause the pre-existing and newly exsolved CO2 to be more easily connected and mobilized during the depletion process. These mechanisms limit the CO2 phase saturation increase during the depletion process when the initial CO2 saturation increases.
■
ENVIRONMENTAL IMPLICATIONS AND POTENTIAL LIMITATIONS The high CO2 residual trapping ratio, which represented a low mobility of the exsolved CO2 phase, was found in the experiments after the imbibition process following the depressurization process. The increase of residual trapping and the low mobility of CO2 are beneficial for CO2 storage safety. The cyclic pressure variation and even more complicated pressure variation profiles are possible in the applications since the CO2 injection or brine withdraw rate need to be controlled artificially. The quantitative evaluation in our research indicated the importance of the first cycle in the presence of cyclic pressure variation with CO2 exsolution involved. Therefore, more careful monitoring is needed in the first cycle. Practically, during the CO2 deep saline aquifer storage, the states of the CO2 rich phase are more likely to be supercritical or liquid instead of gas. Our research is helpful for understanding the CO2 exsolution behaviors and mechanisms during reservoir pressure variations. However, CO2 solubility in the aqueous phase, the interfacial tension, density, viscosity, and compressibility of CO2 are different at various reservoir pressures; these may limit the extrapolation of the experimental findings to the supercritical condition in terms of quantitative prediction. The possible factors in practical reservoirs, including the brine salinity, the presence of noncondensable impurities in the aqueous phase, and the existence of the continuous background flow during the depressurization/repressurization processes, potentially make the exsolution behavior more complicated than the cases presented in this study. Moreover, the quantitative relationships between Sdep and Pdepleted, between relative permeability and fluid saturation, and between residual saturation and initial saturation were obtained using a homogeneous sample in our research. These relationships are possibly not suitable for upscaling to the larger-scale related to the heterogeneity.58 Further work needs to be done to understand whether these relationships measured in the homogeneous sample are representatives of the next largerscale related to heterogeneity and, if not, how to effectively include the heterogeneity into the upscaling strategy.
Figure 5. Residual trapping of the exsolved CO2 phase.
saturation and the corrected initial saturation implies that the CO2 residual trapping retains a high ratio when the exsolved CO2 is repressurized. The repressurization effect does not significantly increase the CO2 mobility, and the reduction in the volumetric residual CO2 saturation mainly results from the compressibility. Series B shows a higher aqueous phase relative permeability than Series A when the Pdepleted are equal. Since the residual CO 2 phase saturation is lower due to the recompression effect, the general positive correlation of the aqueous phase relative permeability and the aqueous phase saturation results cause the difference in the relative permeability of Series A and B. The CO2 phase expansion/ compression effect makes a significant contribution to the aqueous phase relative permeability and mobility during the pressure variation. Cyclic Pressure Oscillation. In experiment Series C, the exsolved CO2 saturation at the first cycle is 14.3%, and, after that, 13.5% additional CO2 saturation was accumulated from the second cycle to the fifth cycle (Figure 3(c)). The first cycle contributed to over 50% of the CO2 exsolution. Similar to the exsolved CO2 saturation, the CO2 saturation at the end of the repressurization and imbibition process also contributed approximately 50% in the first cycle and kept accumulating from the second to the fifth cycle. In the pressure cyclic oscillating case, the first cycle contributes more than the following cycles to the exsolved CO2 saturation. Starting from the second cycle, the deviation from the theoretical maximum increased with the increment of the cycle number. The possible explanation is that the theoretical maximum of the second saturation cycle is much greater than the critical CO2 saturation, which represents the saturation corresponding to the appearance of a sample spanning flow path for the exsolved phase. A certain amount of the CO2 clusters grew to the length scale comparable with the length of the core sample, and then they were connected to the outlet of the core. Thus, from the second cycle, a significant amount of CO2 was expelled out of the core during the drainage process. The aqueous phase relative permeability at the end of imbibition gradually reduces from 1 to 0.29 during the 5 cycles as the exsolved CO2 phase saturation increases. Similar to Series A and B, Series C also shows a high residual trapping fraction (Figure 5). This implies that the complex pressure
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b03826. Table S1, typical reservoir conditions in previous CO2 sequestration studies; Figure S1, diagram of experimental 14701
DOI: 10.1021/acs.est.5b03826 Environ. Sci. Technol. 2015, 49, 14696−14703
Article
Environmental Science & Technology
■
(10) Lu, M.; Connell, L. D. Transient, thermal wellbore flow of multispecies carbon dioxide mixtures with phase transition during geological storage. Int. J. Multiphase Flow 2014, 63, 82−92. (11) Viswanathan, H. S.; Pawar, R. J.; Stauffer, P. H.; Kaszuba, J. P.; Carey, J. W.; Olsen, S. C.; Keating, G. N.; Kavetski, D.; Guthrie, G. D. Development of a hybrid process and system model for the assessment of wellbore leakage at a geologic CO2 sequestration site. Environ. Sci. Technol. 2008, 42 (19), 7280−7286. (12) Shukla, R.; Ranjith, P.; Haque, A.; Choi, X. A review of studies on CO2 sequestration and caprock integrity. Fuel 2010, 89 (10), 2651−2664. (13) Li, Q.; Wei, Y. N.; Liu, G.; Lin, Q. Combination of CO2 geological storage with deep saline water recovery in western China: insights from numerical analyses. Appl. Energy 2014, 116, 101−110. (14) Birkholzer, J. T.; Cihan, A.; Zhou, Q. Impact-driven pressure management via targeted brine extractionConceptual studies of CO2 storage in saline formations. Int. J. Greenhouse Gas Control 2012, 7, 168−180. (15) Spycher, N.; Pruess, K.; Ennis-King, J. CO2-H2O mixtures in the geological sequestration of CO2. I. Assessment and calculation of mutual solubilities from 12 to 100 °C and up to 600 bar. Geochim. Cosmochim. Acta 2003, 67 (16), 3015−3031. (16) Handy, L. L. A Laboratory Study of Oil Recovery by Solution Gas Drive. Petroleum Transactions, AIME 1958, 213, 310−315. (17) Bauget, F., Lenormand, R. Mechanisms of bubble formation by pressure decline in porous media: a critical review. SPE Annual Technical Conference and Exhibition. Society of Petroleum Engineers, 2002. (18) Satik, C.; Li, X.; Yortsos, Y. C. Scaling of single-bubble growth in a porous medium. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1995, 51 (4), 3286. (19) Li, X.; Yortsos, Y. C. Theory of multiple bubble growth in porous media by solute diffusion. Chem. Eng. Sci. 1995, 50 (8), 1247− 1271. (20) Bravo, M. C.; Araujo, M.; Lago, M. E. Pore network modeling of two-phase flow in a liquid-(disconnected) gas system. Phys. A 2007, 375 (1), 1−17. (21) Li, X.; Yortsos, Y. C. Visualization and numerical studies of bubble growth during pressure depletion. SPE 1991, 22589, 6−9. (22) Arora, P.; Kovscek, A. R. A mechanistic modeling and experimental study of solution gas drive. Transp. Porous Media 2003, 51 (3), 237−265. (23) Bora, R.; Maini, B. B.; Chakma, A. Flow visualization studies of solution gas drive process in heavy oil reservoirs using a glass micromodel. SPE Reserv. Eval. Eng. 2000, 3 (03), 224−229. (24) George, D. S.; Hayat, O.; Kovscek, A. R. A microvisual study of solution-gas-drive mechanisms in viscous oils. J. Pet. Sci. Eng. 2005, 46 (1), 101−119. (25) Bachu, S.; Bennion, D. B. Interfacial tension between CO2, freshwater, and brine in the range of pressure from (2 to 27) MPa, temperature from (20 to 125) C, and water salinity from (0 to 334 000) mg· L. J. Chem. Eng. Data 2009, 54 (3), 765−775. (26) Enouy, R.; Li, M.; Ioannidis, M. A.; Unger, A. J. A. Gas exsolution and flow during supersaturated water injection in porous media: II. Column experiments and continuum modeling. Adv. Water Resour. 2011, 34 (1), 15−25. (27) Luhmann, A. J.; Kong, X. Z.; Tutolo, B. M.; Ding, K.; Saar, M. O.; Seyfried, W. E. Permeability reduction produced by grain reorganization and accumulation of exsolved CO2 during geologic carbon sequestration: A new CO2 trapping mechanism. Environ. Sci. Technol. 2013, 47 (1), 242−251. (28) Tutolo, B. M.; Luhmann, A. J.; Kong, X. Z.; Saar, M. O.; Seyfried, W. E., Jr Experimental observation of permeability changes in dolomite at CO2 sequestration conditions. Environ. Sci. Technol. 2014, 48 (4), 2445−2452. (29) Perrin, J. C.; Falta, R. W.; Krevor, S.; Lin, Z.; Ellison, K.; Benson, S. M. Laboratory experiments on core-scale behavior of CO2 exolved from CO2-saturated brine. Energy Procedia 2011, 4 (22), 3210−3215.
system; Figure S2, T2 mapping of the Berea sandstone core sample; uncertainty estimation of the effective permeability; supporting information on NMR measurement and MR image; Figure S3, NMR signal intensity calibration curve; Figure S4, T2 curves of saturated conditions at the first and last experiment runs; Figure S5, relative permeability at the end of imbibition process in Series A, B, and C; Figure S6, MR imaging of the local aqueous phase saturation at the various depleted pressures of Series A (PDF)
AUTHOR INFORMATION
Corresponding Author
*Phone: 8610 62772661. Fax: 8610 62770209. E-mail:
[email protected]. Corresponding address: Department of Thermal Engineering, Tsinghua University, Beijing 100084, China. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This project was supported by the National Natural Science Foundation of China (No. 51376104), the Research Project of Chinese Ministry of Education (No. 113008A), and the international S&T cooperation project of China (No. 2013DFB60140).
■
REFERENCES
(1) IPCC. Underground Geological Storage. In IPCC Special Report on Carbon Dioxide Capture and Storage, Prepared by Working Group III of the Intergovernmental Panel on Climate Change; Metz, B., Davidson, O., de Coninck, H. C., Loos, M., Meyer, L. A., Eds.; Cambridge University Press: Cambridge, United Kingdom and New York, 2005; pp 195−276. (2) Herzog, H. J. Peer reviewed: what future for carbon capture and sequestration? Environ. Sci. Technol. 2001, 35 (7), 148A−153A. (3) Zhang, K.; Moridis, G.; Pruess, K. TOUGH+CO2: A multiphase fluid-flow simulator for CO2 geologic sequestration in saline aquifers. Comput. Geosci. 2011, 37 (6), 714−723. (4) Tutolo, B. M.; Kong, X. Z.; Seyfried, W. E.; Saar, M. O. High performance reactive transport simulations examining the effects of thermal, hydraulic, and chemical (THC) gradients on fluid injectivity at carbonate CCUS reservoir scales. Int. J. Greenhouse Gas Control 2015, 39, 285−301. (5) Jiang, P. X.; Li, X. L.; Xu, R. N.; Wang, Y. S.; Chen, M. S.; Wang, H. M.; Ruan, B. L. Thermal modeling of CO2 in the injection well and reservoir at the Ordos CCS demonstration project, China. Int. J. Greenhouse Gas Control 2014, 23, 135−146. (6) Ahmed, T. K.; Nasrabadi, H. Case study on combined CO2 sequestration and low-salinity water production potential in a shallow saline aquifer in Qatar. J. Environ. Manage. 2012, 109, 27−32. (7) Tiamiyu, O. M.; Nygaard, R.; Bai, B. Effect of Aquifer Heterogeneity, Brine Withdrawal, and Well-Completion Strategy on CO2 Injectivity in Shallow Saline Aquifer. SPE International Conference on CO2 Capture, Storage, and Utilization, New Orleans, Louisiana, USA, 2010. (8) Yang, F.; Bai, B.; Dunn-Norman, S.; Nygaard, R.; Eckert, A. Factors affecting CO2 storage capacity and efficiency with water withdrawal in shallow saline aquifers. Environ. Earth Sci. 2014, 71, 267−275. (9) Mathias, S. A.; Gluyas, J. G.; Oldenburg, C. M.; Tsang, C. Analytical solution for Joule−Thomson cooling during CO2 geosequestration in depleted oil and gas reservoirs. Int. J. Greenhouse Gas Control 2010, 4 (5), 806−810. 14702
DOI: 10.1021/acs.est.5b03826 Environ. Sci. Technol. 2015, 49, 14696−14703
Article
Environmental Science & Technology (30) Zuo, L.; Krevor, S.; Falta, R. W.; Benson, S. M. An experimental study of CO2 exsolution and relative permeability measurements during CO2 saturated water depressurization. Transp. Porous Media 2012, 91, 459−478. (31) Zuo, L.; Zhang, C.; Falta, R. W.; Benson, S. M. Micromodel investigations of CO 2 exsolution from carbonated water in sedimentary rocks. Adv. Water Resour. 2013, 53 (1), 188−197. (32) Falta, R. W.; Zuo, L.; Benson, S. M. Migration of exsolved CO2 following depressurization of saturated brines. Greenhouse Gases: Sci. Technol. 2013, 3 (6), 503−515. (33) Fordham, E. J.; Hall, L. D.; Ramakrishnan, T. S.; Sharpe, M. R.; Hall, C. Saturation Gradients in Drainage of Porous Media: NMR Imaging Measurements. AIChE J. 1993, 39 (9), 1431−1443. (34) Krevor, S.; Pini, R.; Zuo, L.; Benson, S. Relative permeability and trapping of CO2 and water in sandstone rocks at reservoir conditions. Water Resour. Res. 2012, 48, W02532. (35) Bennion, B.; Bachu, S. Drainage and imbibition relative permeability relationships for supercritical CO2/brine and H2S/brine systems in intergranular sandstone, carbonate, shale, and anhydrite rocks. SPE Reserv. Eval. Eng. 2008, 11 (03), 487−496. (36) Levine, J. S.; Goldberg, D. S.; Lackner, K. S.; Matter, J. M.; Supp, M. G.; Ramakrishnan, T. S. (2013). Relative permeability experiments of carbon dioxide displacing brine and their implications for carbon sequestration. Environ. Sci. Technol. 2014, 48 (1), 811−818. (37) Manceau, J. C.; Ma, J.; Li, R.; Audigane, P.; Jiang, P. X.; Xu, R. N.; Tremosa, J.; Lerouge, C. Two - phase flow properties of a sandstone rock for the CO2/water system: Core - flooding experiments, and focus on impacts of mineralogical changes. Water Resour. Res. 2015, 51 (4), 2885−2900. (38) Kim, Y.; Wan, J.; Kneafsey, T. J.; Tokunaga, T. K. Dewetting of silica surfaces upon reactions with supercritical CO2 and brine: porescale studies in micromodels. Environ. Sci. Technol. 2012, 46 (7), 4228−4235. (39) Wang, Y.; Zhang, C.; Wei, N.; Oostrom, M.; Wietsma, T. W.; Li, X.; Bonneville, A. Experimental study of crossover from capillary to viscous fingering for supercritical CO2−water displacement in a homogeneous pore network. Environ. Sci. Technol. 2013, 47 (1), 212− 218. (40) Suekane, T.; Nobuso, T.; Hirai, S.; Kiyota, M. Geological storage of carbon dioxide by residual gas and solubility trapping. Int. J. Greenhouse Gas Control 2008, 2 (1), 58−64. (41) Spiteri, E. J.; Juanes, R.; Blunt, M. J.; Orr, F. M. A new model of trapping and relative permeability hysteresis for all wettability characteristics. SPE J. 2008, 13 (3), 277−288. (42) Akbarabadi, M.; Piri, M. Relative permeability hysteresis and capillary trapping characteristics of supercritical CO2/brine systems: An experimental study at reservoir conditions. Adv. Water Resour. 2013, 52, 190−206. (43) Pentland, C. H.; El-Maghraby, R.; Iglauer, S.; Blunt, M. J. Measurements of the capillary trapping of super-critical carbon dioxide in Berea sandstone. Geophys. Res. Lett. 2011, 38, L06401. (44) El-Maghraby, R. M.; Blunt, M. J. Residual CO2 trapping in Indiana limestone. Environ. Sci. Technol. 2013, 47 (1), 227−233. (45) Zuo, L.; Benson, S. M. Process - dependent residual trapping of CO2 in sandstone. Geophys. Res. Lett. 2014, 41 (8), 2820−2826. (46) Mitchell, J.; Chandrasekera, T. C.; Holland, D. J.; Gladden, L. F.; Fordham, E. J. Magnetic resonance imaging in laboratory petrophysical core analysis. Phys. Rep. 2013, 526 (3), 165−225. (47) Kleinberg, R. L.; Jackson, J. A. An introduction to the history of NMR well logging, Concept. Concepts Magn. Reson. 2001, 13, 340− 342. (48) Coates, G. R.; Xiao, L.; Prammer, M. G. NMR Logging Principles and Applications; Halliburton Energy Services: Houston, 1999. (49) Watson, A. T.; Chang, C. T. P. Characterizing porous media with NMR methods. Prog. Nucl. Magn. Reson. Spectrosc. 1997, 31 (4), 343−386. (50) Kleinberg, R. L. Pore size distributions, pore coupling, and transverse relaxation spectra of porous rocks. Magn. Reson. Imaging 1994, 12 (2), 271−274.
(51) Marica, F.; Chen, Q.; Hamilton, A.; Hall, C.; Al, T.; Balcom, B. J. Spatially resolved measurement of rock core porosity. J. Magn. Reson. 2006, 178 (1), 136−141. (52) Freedman, R.; Heaton, N.; Flaum, M.; Hirasaki, G. J.; Flaum, C.; Hürlimann, M. Wettability, saturation, and viscosity from NMR measurements. SPE J. 2003, 8 (4), 317−327. (53) Suekane, T.; Furukawa, N.; Tsushima, S.; Hirai, S.; Kiyota, M. Application of MRI in the measurement of two-phase flow of supercritical CO2 and water in porous rocks. J. Porous Media 2009, 12 (2), 143−154. (54) Zhao, Y.; Song, Y.; Liu, Y.; Liang, H.; Dou, B. Visualization and measurement of CO2 flooding in porous media using MRI. Ind. Eng. Chem. Res. 2011, 50 (8), 4707−4715. (55) Meiboom, S.; Gill, D. Modified spin-echo method for measuring nuclear relaxation times. Rev. Sci. Instrum. 1958, 29, 688−691. (56) Kleinberg, R. L.; Kenyon, W. E.; Mitra, P. P. Mechanism of NMR relaxation of fluids in rock. J. Magn. Reson., Ser. A 1994, 108 (2), 206−214. (57) Roberts, S. P.; McDonald, P. J.; Pritchard, T. A bulk and spatially resolved NMR relaxation study of sandstone rock plugs. J. Magn. Reson., Ser. A 1995, 116 (2), 189−195. (58) Ott, H.; Pentland, C. H.; Oedai, S. CO2−brine displacement in heterogeneous carbonates. Int. J. Greenhouse Gas Control 2015, 33, 135−144.
14703
DOI: 10.1021/acs.est.5b03826 Environ. Sci. Technol. 2015, 49, 14696−14703
