Molecular simulations of CO2 at interfaces: A combined force field and quantum mechanical study of CO2 geologic sequestration in carbonate rocks

TY - BOOK

T1 - Molecular simulations of CO2 at interfaces

T2 - A combined force field and quantum mechanical study of CO2 geologic sequestration in carbonate rocks

AU - Silvestri, Alessandro

PY - 2018

Y1 - 2018

N2 - CO2 anthropogenic emissions into the atmosphere have long been recognized as themain driver for global climate change and ocean acidication. Carbon sequestrationin geologic formations is a promising approach for decreasing the net amount of CO2emitted into the atmosphere. It relies on various trapping mechanisms that act overdierent time scales, where eectiveness is determined by phenomena that occurat the interfaces between CO2, pore uids and the pore surfaces. Solid theoreticalunderstanding of the nanoscale interactions that result from the interplay of intermolecularand surface forces acting at the interface is currently limited. To gain afundamental understanding of the sequestration process and to assess the risks andenvironmental impacts associated with it, we need better insight into the underlyinginteractions.Two physicochemical parameters are of great importance in the context of carbonsequestration because they control the capillary phenomena that can trap CO2in the pores of the storage reservoir: the interfacial tension, IFT, between CO2and water, and the contact angle, , between the gas, liquid and mineral surface.Both IFT and have been characterized for a wide variety of conditions: pressure,temperature, pore solution salinity and various mineral surfaces. However, achievingrepresentative subsurface conditions in experiments is challenging and reporteddata are aected by experimental uncertainties and sometimes are contradictory.Molecular modelling is a valuable tool for complementing experimental studies andit can help us interpret results and gain insight under conditions where experimentsare dicult or impossible to perform. It has already been applied for predictingthe CO2{water IFT and the contact angle on atomically smooth surfaces and theirdependence on pressure, temperature and salinity.In this thesis, I report predictions for the CO2{water interfacial tension for awide variety of pressures, temperatures and salinity. I used two computationalapproaches. One is a quantum chemical approach, that combines density functionaltheory (DFT) with the statistical thermodynamics treatment provided bythe COSMO-RS implicit solvent model. The other is classical molecular dynamics(MD), based on two exible potential models for CO2 and water. Both methodsprovided results that agree well with each other, with experimental results and withprevious simulations. DFT has some advantages over MD in terms of computationalcost but MD allows exploration of larger and more complex systems.Dierent options for geologic storage of CO2 have been proposed. Carbonateminerals are ubiquitous, limestone, chalk and marble constitute a signicant fractionof the sedimentary rock record and the formations are generally porous so theirprobable response to CO2 sequestration needs to be investigated. However, despitethe large number of geologic sequestration publications on water{rock interactionsover the last decade, studies on carbonate reservoirs remain scarce. Carbonatereservoirs are being considered as CO2 storage sites so more information is needed,for providing data for safety assessment models and for fundamental understandingof the relevant processes and their inuence.Theoretical approaches for the carbonate minerals are even more scarce and onlyfew molecular dynamics simulations have been performed for CO2{water{carbonateminerals. The main reason has been the lack of reliable force eld parameters fordescribing the interaction between CO2 and the calcite surface. In my research, Iproduced these parameters and used them to run classical molecular dynamics andmetadynamics simulations for the CO2{water{calcite system. The new calculatedcontact angles provide a base for comparing the few, widely variable experimentaldata that are reported in the literature. A fundamental result of this work is thatCO2 can neither penetrate the ordered water layers at the calcite-water interface,nor adsorb directly on the solid surface. However, a weak anity of CO2 for thesurface of the ordered water layers is observed and this leads to nucleation of a CO2droplet located above two structured water layers that form on the mineral surface.This can have important implications for trapping residual CO2 bubbles, which isone of the main trapping mechanisms in geologic carbon sequestration.

AB - CO2 anthropogenic emissions into the atmosphere have long been recognized as themain driver for global climate change and ocean acidication. Carbon sequestrationin geologic formations is a promising approach for decreasing the net amount of CO2emitted into the atmosphere. It relies on various trapping mechanisms that act overdierent time scales, where eectiveness is determined by phenomena that occurat the interfaces between CO2, pore uids and the pore surfaces. Solid theoreticalunderstanding of the nanoscale interactions that result from the interplay of intermolecularand surface forces acting at the interface is currently limited. To gain afundamental understanding of the sequestration process and to assess the risks andenvironmental impacts associated with it, we need better insight into the underlyinginteractions.Two physicochemical parameters are of great importance in the context of carbonsequestration because they control the capillary phenomena that can trap CO2in the pores of the storage reservoir: the interfacial tension, IFT, between CO2and water, and the contact angle, , between the gas, liquid and mineral surface.Both IFT and have been characterized for a wide variety of conditions: pressure,temperature, pore solution salinity and various mineral surfaces. However, achievingrepresentative subsurface conditions in experiments is challenging and reporteddata are aected by experimental uncertainties and sometimes are contradictory.Molecular modelling is a valuable tool for complementing experimental studies andit can help us interpret results and gain insight under conditions where experimentsare dicult or impossible to perform. It has already been applied for predictingthe CO2{water IFT and the contact angle on atomically smooth surfaces and theirdependence on pressure, temperature and salinity.In this thesis, I report predictions for the CO2{water interfacial tension for awide variety of pressures, temperatures and salinity. I used two computationalapproaches. One is a quantum chemical approach, that combines density functionaltheory (DFT) with the statistical thermodynamics treatment provided bythe COSMO-RS implicit solvent model. The other is classical molecular dynamics(MD), based on two exible potential models for CO2 and water. Both methodsprovided results that agree well with each other, with experimental results and withprevious simulations. DFT has some advantages over MD in terms of computationalcost but MD allows exploration of larger and more complex systems.Dierent options for geologic storage of CO2 have been proposed. Carbonateminerals are ubiquitous, limestone, chalk and marble constitute a signicant fractionof the sedimentary rock record and the formations are generally porous so theirprobable response to CO2 sequestration needs to be investigated. However, despitethe large number of geologic sequestration publications on water{rock interactionsover the last decade, studies on carbonate reservoirs remain scarce. Carbonatereservoirs are being considered as CO2 storage sites so more information is needed,for providing data for safety assessment models and for fundamental understandingof the relevant processes and their inuence.Theoretical approaches for the carbonate minerals are even more scarce and onlyfew molecular dynamics simulations have been performed for CO2{water{carbonateminerals. The main reason has been the lack of reliable force eld parameters fordescribing the interaction between CO2 and the calcite surface. In my research, Iproduced these parameters and used them to run classical molecular dynamics andmetadynamics simulations for the CO2{water{calcite system. The new calculatedcontact angles provide a base for comparing the few, widely variable experimentaldata that are reported in the literature. A fundamental result of this work is thatCO2 can neither penetrate the ordered water layers at the calcite-water interface,nor adsorb directly on the solid surface. However, a weak anity of CO2 for thesurface of the ordered water layers is observed and this leads to nucleation of a CO2droplet located above two structured water layers that form on the mineral surface.This can have important implications for trapping residual CO2 bubbles, which isone of the main trapping mechanisms in geologic carbon sequestration.

UR - https://rex.kb.dk/primo-explore/fulldisplay?docid=KGL01010814109&context=L&vid=NUI&search_scope=KGL&tab=default_tab&lang=da_DK

M3 - Ph.D. thesis

BT - Molecular simulations of CO2 at interfaces

PB - Department of Chemistry, Faculty of Science, University of Copenhagen

ER -