MONITORING HOW CARBONATE CEMENT DISSOLUTION AFFECTS ROCK FRAME PROPERTIES DUE TO CO2 INJECTION Ludmila Adam a , Jackson MacFarlane a , Kasper van Wijk b , Jeffrey Shragge c and Karen Higgs d a

University of Auckland, School of Environment, 3A Symonds Street, Auckland, New Zealand; University of Auckland, Department of Physics, 3A Symonds Street, Auckland, New Zealand; c The University of Western Australia, School of Physics, 35 Stirling Highway, Crawley, Australia; d GNS Science, 1 Fairway Drive, Lower Hutt, New Zealand. Corresponding author e-mail: [email protected] b

Summary Time-lapse seismic signatures can be used to quantify fluid saturation and pressure changes in a reservoir. Examples of this are when seismic surveys are acquired over fields where carbon dioxide is injected for underground storage, or to enhance oil recovery. Either way, the injection of CO2 acidifies the water, which may dissolve and/or precipitate minerals. Understanding the impact on the rock frame from field seismic time-lapse changes is an outstanding challenge. Here, we study the effects of carbonate-CO2 water reactions on the physical properties of rock samples with variable levels of carbonate cementation, and how these effects translate to the elastic wave properties. So far, we have characterized the set of samples (NMR, CT scanning, thin sections) and performed ultrasonic P-wave velocity measurements with a laser ultrasonic setup on the dry samples. The first sample reacted with CO2 shows changes in P-wave velocity, that vary spatially from significant (-20%) to the more subtle (-5%).

Introduction Since 1960, the atmospheric concentration of CO2 has increased by 24% (Jones, 2013). Carbon dioxide geosequestration is a proposed method to reduce the carbon emissions into the atmosphere through the injection of this gas into the subsurface (Hepple and Benson, 2005; Bachu, 2003; Armitage et al., 2013; Dodds et al., 2009; Litynski et al., 2009). In these projects monitoring the movement of the CO2 plume with geochemical, and in most instances, geophysical methods will be crucial to ensure the safe storage of this fluid in the subsurface. As most rocks in the subsurface are water-saturated, the injection of carbon dioxide into a reservoir forms carbonic acid. This complicates the dynamics of these reservoirs as carbonic acid will react with the rock frame, changing the petrophysical and geochemical rock and fluid properties (Ross et al., 1982; Kharaka et al., 2006; McGrail et al., 2006; Pruess et al., 2003). Such rock-fluid interaction can result in mineral dissolution (Armitage et al., 2013; Vialle and Vanorio, 2011; Grombacher et al., 2012; Pimienta et al., 2014), precipitation (Oelkers et al., 2008; Wigand et al., 2008; Vialle and Vanorio, 2011; Adam et al., 2013), or, in some instances, produce no significant rock frame changes (Lebedev, 2013; Hangx et al., 2013). Commonly, these changes result from the dissolution and precipitation of carbonate, evaporite and clay minerals (Noiriel et al., 2004; Luquot and Gouze, 2009; Vialle and Vanorio, 2011; Grombacher et al., 2012; Pimienta et al., 2014). Ultimately, the physical properties of rocks resulting from mineralCO2 interaction, such as porosity and permeability (Noiriel et al., 2004), may be picked up by remotely

1

3rd International Workshop on Rock Physics, Perth, 13th -17th April 2015

sensed seismic waves. Therefore, the changes should be calibrated in the laboratory, in terms of the changes seen in elastic waveforms. In sedimentary basins, time-lapse seismic changes are used to monitor subsurface properties and movement of the CO2 plume (Chadwick et al., 2005; Daley et al., 2008). However, the changes in wave velocities are attributed to either changes in fluid saturation (Wang et al., 2000; Mikhaltsevitch et al., 2014; Alemu et al., 2013) and/or fluid pressures (Landrø, 2001; Tura and Lumley, 1998). Geophysicists have rarely interpreted time-lapse seismic wave signatures as a result of fluid-rock interactions that change the rock frame (Ivandic et al., 2014). The most common carbon capture and sequestration reservoir are sandstone rocks (Bachu, 2003), and in such reservoirs, carbonate cement can be pervasive. Elastic wave velocities are highly dependent on the packing and cement in sandstone reservoirs. It is therefore important to understand and quantify how the injection of CO2 affects the petrophysical and time-lapse seismic properties of carbonate-cemented sandstone cores. Here we study the petrographic and acoustic properties of a set of sandstone core samples from the Taranaki Basin, New Zealand. We performed XRD, NMR, CT scanning, thin section petrography and ultrasound laser measurements before and after reactions for water saturated with CO2 . We will correlate the changes in the rock microstructure to the change in elastic properties to quantify how carbonate cement dissolution affect 4D seismic signatures.

1

Sample descriptions

Nine samples from the Pohokura-03 well located in the offshore section of the Taranaki Basin, west of the North Island of New Zealand (Knox, 1982; Higgs, 2009), were selected based on carbonate cement volume. Thin section photomicrographs of four representative samples are shown in Figure 1, revealing a predominant composition of quartz and feldspar, with differences in grain size, sorting and volume of carbonate cement and clay minerals.. Thin section images show that carbonate cement consists of siderite/ankerite for samples S3, S4, S5, S6; and ferroan-calcite for samples S7 and S8. Samples S1, S2 and S18 have less than 1% carbonates, which, for the purposes of this study are clean in terms of carbonate cement. Table 1 summarizes the porosity and permeability estimated from NMR analysis and the volumetric carbonate cement from thin section point count analysis. An increasing volume of carbonate cement and clays corresponds with lower permeability, but a direct correlation with porosity is not observed. In this study, we will also look into clay minerals. Notably, rock physical property analysis seems to indicate that clays potentially block some of the pore throats for our samples, decreasing rock permeability which we believe could interfere with the rock-fluid interactions. In addition to the above, we have used computerized tomography (CT) imaging, which is highly effective in estimating the volume of carbonate cement due to the significant density contrast between carbonate cement and quartz/feldspar grains. The CT images are also used to identify sample texture and will be used to quantify rock physical changes before and after reactions.

Methods Laser-based ultrasonics can illuminate the rock frame changes before and after reactions with a CO2 water mixture. A high-energy pulsed laser excites ultrasonic waves via thermoelastic expansion of the sample. These waves travel through the sample and are detected using a laser vibrometer, measuring the displacement at the sample surface due to a wave perturbation (for details see Blum et al., 2013). The sample is moved on computerized stages and waveforms measured at 16-bits dynamic range at a sampling rate of 1e7 samples per second. Data acquisition is automatized by PLACE, an open-source python-package (Johnson et al., 2014). Because we are interested in changes in the rock-frame properties we perform the experiments on dry rocks at room condition. Data acquisition consists of a rotational scan, 2

3rd International Workshop on Rock Physics, Perth, 13th -17th April 2015

S1

S3

S4

S6

Figure 1: Microphotographs of four of the samples in the study showing variable carbonate cement. Sample S1 is a clean sandstone (black minerals are pyrite), while samples S3, S4 and S6 show variable carbonate cements, mostly seen as the black/brown minerals in the images. Q = detrital quartz, Qog = quartz overgrowths, Ka = authigenic kaolinite, S = siderite/ankerite, Kf = K-feldspar, Pl = plagioclase feldspar, LF = lithic fragment, AC = grain-replacive authigenic clay, Py = pyrite, B = biotite, MP = macroporosity. a linear scan and a 3D tomography which incorporates both. In this abstract we present the results of a few of the rotational scans, where the source and receiver are aligned on opposite sides of the sample, and the cylindrical sample is on an automated rotational stage (Blum et al., 2013). Next, we will place the samples in a CO2 -water reactor bath to analyse the effect of carbonate cement dissolution on the rock elastic and physical properties. Up to now, only one samples was reacted. Sample S5 was exposed to a 4:1 water-CO2 mixture by volume at 200 psi for 24 hours and then dried. Results Examples of waveforms before reactions acquired in a rotational scan for samples S1 and S3 are plotted in Figure 2. On visual inspection the samples appear isotropic and with moderately sorted grain size. However, S1 has a massive texture, while S3 presents faint lamination, potentially as a result of carbonate cemented bands. This rock texture affects the wave speeds. Wave speeds are fast parallel to the banding (140◦ ) and slow perpendicular to the bands. By performing such high density waveform scanning the effect of rock inhomogeneity and anisotropy are well-defined. The P-wave velocity for all samples before reactions ranges between 2.0 and 3.7 km/s. The porosity of the samples ranges between 7 and 12%. We observe that there is little correlation between P-wave velocity and porosity, as well as for carbonate volume. Laser ultrasonic rotational scans for samples S5 before and after rock-fluid reactions are plotted in Figure 3. First, observe the high repeatability of the waveforms. The most obvious decrease in P-wave velocity (-20%) is measured in the section of the rock between 40◦ and 70◦ , potentially due to carbonate dissolution. Finally, even though other sections of the rotational scan might not show obvious changes, coda wave interferometry was applied to study small changes in velocity. We average P-wave velocity change between 0◦ and 30◦ is -5%. Therefore, combining direct wave and coda wave analysis the P-wave velocity changes due to rock-CO2 reactions can be quantified. We have also performed preliminary estimates for changes in P-wave velocity due to fluid substitution based on laboratory measured dry samples. We use Gassmann’s equation to estimated changes in velocity for a rock 100% water-saturated and 100% CO2 -saturated. On average, the P-wave velocity changes by 3

3rd International Workshop on Rock Physics, Perth, 13th -17th April 2015

Sample

S1 S2 S3 S4 S5 S6 S7 S8 S18

Total porosity (%) 11.3 9.3 7.3 9.4 12.5 6.1 9.2 4.5 11.1

Clay bound water (%) 0.0 0.0 1.6 1.7 2.3 5.3 0.2 0.5 0.0

permeability (mD) 56.7 6.45 0.0097 0.0544 0.103 0.00004 11.1 0.0028 15.4

Carbonate minerals (%vol. of rock) 0.0 0.7 22.7 12.3 23.7 32.1 14.3 21.6 0.3

Clay minerals (%vol. of rock) 6.8 7.2 11.4 14.8 19.2 10.6 7.5 9.3 6.6

Table 1: Total porosity, clay bound water, and permeability as interpreted from NMR analysis. Carbonate and clay volumes with respect to the total rock volume are estimated from point count thin section petrography.

[S1]

[S3]

Figure 2: Laser ultrasonic waveforms for samples S1 and S3 for a rotational scan setup. Data is acquired every 1 degree. The first arriving energy is a direct ultrasonic P-wave. -13%. This value is close to the observed changes due to frame weakening. Therefore, based on our preliminary data, if mineral dissolution is occuring in the field, time-lapse changes due to fluids can be of the same order as does due to rock-fluid reactions. We are currently equilibrating water with sections of the samples to resemble reservoir water conditions, to avoid starting the reactions with distilled water. The remaining eight samples will be reacted and measured with the laser ultrasonic setup. We aim to report whether (and if so, how) the potential carbonate dissolution affects the elastic rock frame and the petrophysical rock properties, from repeat NMR, CT scans, XRD and thin section petrography. Our goal is also to investigate how clay minerals and rock permeability affect the dissolution of the samples.

Conclusions Preliminary experiments suggest that CO2 -water-rock interactions should be considered when interpreting time-lapse seismic signatures. Especially in carbonate-cemented sandstones, we expect that the effect of matrix weakening might be of the same magnitude as the fluid substitution of CO2 for water. We have set the next set of samples for reactions to study rock-fluid reactivity for a range of carbonate volume cement in sandstones.

4

3rd International Workshop on Rock Physics, Perth, 13th -17th April 2015

[a]

[b]

Figure 3: Laser ultrasonic scans for sample S5 before (a) and after (b) reactions with a CO2 -water mixture. Acknowledgements We thanks Randall McDonnell and Brad Field for help selecting the field and core, Jami Johnson for her help with the laser ultrasonic measurements, Cheng Yii Sim for fluid substitution modeling. Andres Arcilla helped in sample preparation. Magritek performed the NMR measurements. The project is funded by the University of Western Australia and the University of Auckland. Core samples were provided by New Zealand Petroleum and Minerals.

References Adam, L.; Wijk, K.; Otheim, T., and Batzle, M. Changes in elastic wave velocity and rock microstructure due to basalt-CO2 -water reactions. Journal of Geophysical Research: Solid Earth, 118(8):4039–4047, 2013. Alemu, Binyam L; Aker, Eyvind; Soldal, Magnus; Johnsen, Øistein, and Aagaard, Per. Effect of subcore scale heterogeneities on acoustic and electrical properties of a reservoir rock: a CO2 flooding experiment of brine saturated sandstone in a computed tomography scanner. Geophysical Prospecting, 61(1):235–250, 2013. Armitage, P.J.; Faulkner, D.R., and Worden, R.H. Caprock corrosion. Nature Geoscience, 6(2):79–80, 2013. Bachu, Stefan. Screening and ranking of sedimentary basins for sequestration of CO2 in geological media in response to climate change. Environmental Geology, 44(3):277–289, 2003. Blum, Thomas E; Adam, Ludmila, and van Wijk, Kasper. Noncontacting benchtop measurements of the elastic properties of shales. Geophysics, 78(3):C25–C31, 2013. Chadwick, RA; Arts, R, and Eiken, O. 4D seismic quantification of a growing CO2 plume at Sleipner, North Sea. In Geological Society, London, Petroleum Geology Conference series, volume 6, pages 1385–1399. Geological Society of London, 2005. Daley, Thomas M; Myer, Larry R; Peterson, JE; Majer, EL, and Hoversten, GM. Time-lapse crosswell seismic and VSP monitoring of injected CO2 in a brine aquifer. Environmental Geology, 54(8):1657– 1665, 2008. Dodds, Kevin; Daley, Tom; Freifeld, B; Urosevic, Milovan; Kepic, Anton, and Sharma, Sandeep. De-

5

3rd International Workshop on Rock Physics, Perth, 13th -17th April 2015

veloping a monitoring and verification plan with referenceto the Australian Otway CO2 pilot project. The Leading Edge, 28(7):812–818, 2009. Grombacher, Denys; Vanorio, Tiziana, and Ebert, Yael. Time-lapse acoustic, transport, and NMR measurements to characterize microstructural changes of carbonate rocks during injection of CO2 -rich water. Geophysics, 77(3):WA169–WA179, 2012. Hangx, Suzanne; van der Linden, Arjan; Marcelis, Fons, and Bauer, Andreas. The effect of CO2 on the mechanical properties of the Captain Sandstone: Geological storage of CO2 at the Goldeneye field (UK). International Journal of Greenhouse Gas Control, 19:609–619, 2013. Hepple, R.P. and Benson, S.M. Geologic storage of carbon dioxide as a climate change mitigation strategy: performance requirements and the implications of surface seepage. Environmental Geology, 47(4):576–585, 2005. Higgs, K. Opportunities for underground geological storage of CO2 in New Zealand-report CCS-08/6onshore Taranaki Paleogene reservoirs. GNS Science Report, 59, 2009. Ivandic, Monika; Juhlin, Christopher; Lueth, Stefan; Bergmann, Peter; Kashubin, Artem; Sopher, Daniel; Ivanova, Alexandra; Baumann, Gunther, and Henninges, Jan. Geophysical monitoring at the Ketzin pilot site for CO2 storage: New insights into the plume evolution. International Journal of Greenhouse Gas Control, 2014. Johnson, Jami L; tom W¨orden, Henrik, and van Wijk, Kasper. PLACE an open-source python Package for Laboratory Automation, Control, and Experimentation. Journal of laboratory automation, page 2211068214553022, 2014. Jones, Nicola. Troubling milestone for CO2 . Nature Geosci, 6(8):589, 2013. Kharaka, Y. K.; Cole, D. R.; Hovorka, S.D.; Gunter, W. D.; Knauss, K. G., and Freifeld, B. M. Gaswater-rock interactions in Frio Formation following CO2 injection: Implications for the storage of greenhouse gases in sedimentary basins. Geology, 34(7):577–580, 2006. Knox, G.J. Taranaki Basin, structural style and tectonic setting. New Zealand journal of geology and geophysics, 25(2):125–140, 1982. Landrø, Martin. Discrimination between pressure and fluid saturation changes from time-lapse seismic data. Geophysics, 66(3):836–844, 2001. Lebedev, M. An experimental study of acoustic responses on the injection of supercritical CO2 into sandstones from the Otway Basin. Geophysics, 78(1):D293–D306, 2013. Litynski, John; Plasynski, Sean; Spangler, Lee; Finley, Robert; Steadman, Edward; Ball, David; Nemeth, Kenneth J; McPherson, Brian, and Myer, Larry. US department of energy’s regional carbon sequestration partnership program: overview. Energy Procedia, 1(1):3959–3967, 2009. Luquot, Linda and Gouze, Philippe. Experimental determination of porosity and permeability changes induced by injection of CO2 into carbonate rocks. Chemical Geology, 265(1):148–159, 2009. McGrail, B Peter; Schaef, H Todd; Ho, Anita M; Chien, Yi-Ju; Dooley, James J, and Davidson, Casie L. Potential for carbon dioxide sequestration in flood basalts. Journal of Geophysical Research, 111 (B12), 2006. Mikhaltsevitch, Vassily; Lebedev, Maxim, and Gurevich, Boris. Measurements of the elastic and anelastic properties of sandstone flooded with supercritical CO2 . Geophysical Prospecting, 62(6):1266– 1277, 2014. Noiriel, Catherine; Gouze, Philippe, and Bernard, Dominique. Investigation of porosity and permeability

6

3rd International Workshop on Rock Physics, Perth, 13th -17th April 2015

effects from microstructure changes during limestone dissolution. Geophysical research letters, 31 (24), 2004. Oelkers, Eric H; Gislason, Sigurdur R, and Matter, Juerg. Mineral carbonation of CO2 . Elements, 4(5): 333–337, 2008. Pimienta, L; Esteban, L; Sarout, J; Liu, K; Delle Piane, C, and Clennell, B. Experimental evidence of calcite dissolution and induced precipitation during supercritical CO2 residence. In 76th EAGE Conference and Exhibition 2014, 2014. Pruess, Karsten; Xu, Tianfu; Apps, John; Garcia, Julio, and others, . Numerical modeling of aquifer disposal of CO2 . SPE Journal, 8(01):49–60, 2003. Ross, Graham D; Todd, Adrain C; Tweedie, John A; Will, Andrew GS, and others, . The dissolution effects of CO2 -brine systems on the permeability of UK and North Sea calcareous sandstones. In SPE Enhanced Oil Recovery Symposium. Society of Petroleum Engineers, 1982. Tura, Ali and Lumley, David E. Subsurface fluid flow properties from time-lapse elastic wave reflection data. In SPIE’s International Symposium on Optical Science, Engineering, and Instrumentation, pages 125–138. International Society for Optics and Photonics, 1998. Vialle, S. and Vanorio, T. Laboratory measurements of elastic properties of carbonate rocks during injection of reactive co2 -saturated water. Geophysical Research Letters, 38(1), 2011. Wang, Z.; Nur, A., and Ebrom, D. A., editors. Seismic and acoustic velocities in reservoir rocks, chapter The Gassmann equation revised: Comparing laboratory data with Gassmann’s predictions by Z. Wang, pages 8–23. Society of Exloration Geophysicists, 2000. M.C.Wigand, JW; Sch¨utt, H.; Spangenberg, E., and Erzinger, J. Geochemical effects of co2 sequestration in sandstones under simulated in situ conditions of deep saline aquifers. Applied Geochemistry, 23(9): 2735–2745, 2008.

7