TY - GEN
T1 - Determination of local diffusion coefficients and directional anisotropy in shale from dynamic micro-CT imaging
AU - Zhang, Yulai
AU - Mostaghimi, Peyman
AU - Fogden, Andrew
AU - Arena, Alessio
AU - Sheppard, Adrian
AU - Middleton, Jill
AU - Armstrong, Ryan T.
N1 - Publisher Copyright:
© 2017, Unconventional Resources Technology Conference (URTeC).
PY - 2017
Y1 - 2017
N2 - Molecular diffusion is an important mass transport mechanism for gas production from shale reservoirs. Characterization of matrix diffusivity is fundamental to understand the recovery from shale gas plays. By micro-computed tomography (micro-CT) imaging of X-ray dense and transparent liquids mixing in a saturated shale sample, 4D dynamic and spatially-resolved monitoring of diffusion experiments has been successfully achieved. This work extends the previously presented method by applying a new mathematical procedure to measure the local, directional Fickian diffusion coefficients matching the observed concentration fields. Small centimeter-sized vertically-cored plugs of shales from the Eagle Ford formation and Permian Basin are imaged over the course of diffusion experiments. The time lapse step and overall duration are designed to minimize artifacts and uncertainties in the local diffusive flux and concentration gradient obtained from image analysis. For the Eagle Ford sample, local diffusion coefficients of the horizontal direction are in the range of 10-14-10-12 m2/s, with an average of 8.4�10-13 m2/s; while for the vertical direction, they are in the range of 10-15-10-13 m2/s, with an average of 2.2�10-14 m2/s. The diffusion process is heavily influenced by fractures. For the Permian Basin sample, the diffusion is dominated by the matrix due to fewer fractures. The horizontal direction local diffusion coefficients are in the range of 10-14-10-12 m2/s, with an average of 3.2�10-13 m2/s. For the vertical direction, they are in the range of 10-15-10-13 m2/s, with an average of 4�10-14 m2/s. Both of the samples exhibit an intermediate level of heterogeneity in terms of the measured Dykstra-Parsons coefficients. The matrix diffusion coefficients are extremely anisotropic with up to 3 orders of magnitude change from parallel to perpendicular to bedding. Combined dynamic micro-CT imaging and local directional diffusion coefficient measurements is a powerful tool to characterize mass transport in shales and provides a benchmark for comparison to flow simulations on static images. New understandings on mass transport properties will be helpful for prediction and optimization of shale gas production.
AB - Molecular diffusion is an important mass transport mechanism for gas production from shale reservoirs. Characterization of matrix diffusivity is fundamental to understand the recovery from shale gas plays. By micro-computed tomography (micro-CT) imaging of X-ray dense and transparent liquids mixing in a saturated shale sample, 4D dynamic and spatially-resolved monitoring of diffusion experiments has been successfully achieved. This work extends the previously presented method by applying a new mathematical procedure to measure the local, directional Fickian diffusion coefficients matching the observed concentration fields. Small centimeter-sized vertically-cored plugs of shales from the Eagle Ford formation and Permian Basin are imaged over the course of diffusion experiments. The time lapse step and overall duration are designed to minimize artifacts and uncertainties in the local diffusive flux and concentration gradient obtained from image analysis. For the Eagle Ford sample, local diffusion coefficients of the horizontal direction are in the range of 10-14-10-12 m2/s, with an average of 8.4�10-13 m2/s; while for the vertical direction, they are in the range of 10-15-10-13 m2/s, with an average of 2.2�10-14 m2/s. The diffusion process is heavily influenced by fractures. For the Permian Basin sample, the diffusion is dominated by the matrix due to fewer fractures. The horizontal direction local diffusion coefficients are in the range of 10-14-10-12 m2/s, with an average of 3.2�10-13 m2/s. For the vertical direction, they are in the range of 10-15-10-13 m2/s, with an average of 4�10-14 m2/s. Both of the samples exhibit an intermediate level of heterogeneity in terms of the measured Dykstra-Parsons coefficients. The matrix diffusion coefficients are extremely anisotropic with up to 3 orders of magnitude change from parallel to perpendicular to bedding. Combined dynamic micro-CT imaging and local directional diffusion coefficient measurements is a powerful tool to characterize mass transport in shales and provides a benchmark for comparison to flow simulations on static images. New understandings on mass transport properties will be helpful for prediction and optimization of shale gas production.
UR - http://www.scopus.com/inward/record.url?scp=85044126473&partnerID=8YFLogxK
U2 - 10.15530/urtec-2017-2695407
DO - 10.15530/urtec-2017-2695407
M3 - Conference contribution
SN - 9781613995433
T3 - SPE/AAPG/SEG Unconventional Resources Technology Conference 2017
BT - SPE/AAPG/SEG Unconventional Resources Technology Conference 2017
PB - Unconventional Resources Technology Conference (URTEC)
T2 - SPE/AAPG/SEG Unconventional Resources Technology Conference 2017
Y2 - 24 July 2017 through 26 July 2017
ER -