TY - THES

T1 - Sensitivity Analysis of the Trapping Potential of Different Rock Types using Digital Rock Physics

AU - Danicic, Lobel

N1 - no embargo

PY - 2022

Y1 - 2022

N2 - Carbon Capture and Storage (CCS) is one of the few options to store carbon dioxide on a large scale and therefore actively reducing greenhouse gas emissions. Thorough knowledge of reservoir properties and trapping mechanisms is needed to ensure the most proper application while ensuring maximum storage safety. Fluid-fluid and rock-fluid properties are assessed from routine and special core analysis programs (SCAL), from which displacement and trapping efficiencies can be derived. However, SCAL experiments take a considerable amount of time and effort. Advances in computational power and imaging techniques in the field of Digital Rock Physics (DRP) enable the simulation of capillary trapping processes within a reasonable time frame, while varying reservoir properties. Using experimental data as calibration it allows the computation of different fluid-fluid and rock-fluid combinations. This thesis focuses on the simulation of capillary trapping curves for the evaluation of the carbon dioxide trapping potential for different rock types. The used simulation tool is GeoDict in particular its module Satudict, which models drainage and imbibition processes based on the morphological method. The investigation includes six different digital twins of real sandstone rocks to observe the effect that the different pore structures have on the capillary trapping. To compute the trapping potential of a rock type a drainage process is performed, followed by imbibition scanning curves. Subsequently the residual non-wetting phase (NWP) saturation is evaluated. As the imbibition process is of major interest, two approaches are used for its modelling. Whereas the first approach is limited to spontaneous imbibition only, while the second one incorporates the full imbibition branch including the forced imbibition. Obtained capillary pressure results with both imbibition modelling approaches are then analysed with the Land model. The investigation includes the analysis of alternating wetting conditions. The results show that the rock type and wettability have the largest influence on the trapping potential. Furthermore, it was shown that domain size and capillary end effects majorly influence the results, indicating that larger domains are favourable. In addition, the correlation between the rocks petrophysical properties and their trapping potential was investigated. It was shown that the pore size distribution in combination with the pore throat size distribution correlates with the trapping potential.

AB - Carbon Capture and Storage (CCS) is one of the few options to store carbon dioxide on a large scale and therefore actively reducing greenhouse gas emissions. Thorough knowledge of reservoir properties and trapping mechanisms is needed to ensure the most proper application while ensuring maximum storage safety. Fluid-fluid and rock-fluid properties are assessed from routine and special core analysis programs (SCAL), from which displacement and trapping efficiencies can be derived. However, SCAL experiments take a considerable amount of time and effort. Advances in computational power and imaging techniques in the field of Digital Rock Physics (DRP) enable the simulation of capillary trapping processes within a reasonable time frame, while varying reservoir properties. Using experimental data as calibration it allows the computation of different fluid-fluid and rock-fluid combinations. This thesis focuses on the simulation of capillary trapping curves for the evaluation of the carbon dioxide trapping potential for different rock types. The used simulation tool is GeoDict in particular its module Satudict, which models drainage and imbibition processes based on the morphological method. The investigation includes six different digital twins of real sandstone rocks to observe the effect that the different pore structures have on the capillary trapping. To compute the trapping potential of a rock type a drainage process is performed, followed by imbibition scanning curves. Subsequently the residual non-wetting phase (NWP) saturation is evaluated. As the imbibition process is of major interest, two approaches are used for its modelling. Whereas the first approach is limited to spontaneous imbibition only, while the second one incorporates the full imbibition branch including the forced imbibition. Obtained capillary pressure results with both imbibition modelling approaches are then analysed with the Land model. The investigation includes the analysis of alternating wetting conditions. The results show that the rock type and wettability have the largest influence on the trapping potential. Furthermore, it was shown that domain size and capillary end effects majorly influence the results, indicating that larger domains are favourable. In addition, the correlation between the rocks petrophysical properties and their trapping potential was investigated. It was shown that the pore size distribution in combination with the pore throat size distribution correlates with the trapping potential.

KW - Kapillarfang

KW - Digitale Gesteinsphysik

KW - SCAL

KW - Landmodell

KW - Unsicherheitsmodellierung

KW - Capillary trapping

KW - Digital rock physics

KW - SCAL

KW - Land model

KW - Uncertainty modelling

M3 - Master's Thesis

ER -