Molecular Dynamics Simulations of CO 2 /Water/Quartz Interfacial Properties: Impact of CO 2 Dissolution in Water Gina Javanbakht, Mohammad Sedghi, William Welch, and Lamia Goual* Department of Chemical and Petroleum Engineering, University of Wyoming, 1000 E. University Avenue, Laramie, Wyoming 82071, United States ABSTRACT: The safe trapping of carbon dioxide (CO 2 ) in deep saline aquifers is one of the major concerns of CO 2 sequestration. The amount of capillary trapping is dominated by the capillary pressure of water and CO 2 inside the reservoir, which in turn is controlled by the interfacial tension (IFT) and the contact angle (CA) of CO 2 /water/rock systems. The measurement of IFT and CA could be very challenging at reservoir conditions, especially in the presence of toxic cocontaminants. Thus, the ability to accurately predict these interfacial properties at reservoir conditions is very advantageous. Although the majority of existing molecular dynamics (MD) studies of CO 2 /water/mineral systems were able to capture the trends in IFT and CA variations with pressure and temperature, their predictions often deviated from experimental data, possibly due to erroneous models and/or overlooked chemical reactions. The objective of this study was to improve the MD predictions of IFT and CA of CO 2 / water/quartz systems at various pressure and temperature conditions by (i) considering the chemical reactions between CO 2 and water and (ii) using a new molecular model for α-quartz surface. The results showed that the presence of carbonic acid at the CO 2 /water interface improved the predictions of IFT, especially at low temperature and high pressure where more CO 2 dissolution occurs. On the other hand, the effect on CA was minor. The slight decrease in CA observed across the pressure range investigated could be attributed to an increase in the total number of H-bonds between fluid molecules and quartz surface. 1. INTRODUCTION The successful sequestration of carbon dioxide (CO 2 ) in deep geological formations requires a minimum leakage of CO 2 through the cap rock. The amount of leakage could be estimated from the threshold capillary pressure, which in turn is affected by the interfacial tension (IFT) between CO 2 and the present fluids, as well as the contact angle (CA) between fluids and the solid surface, according to the following equation 1 γ θ = αβ αβδ P r [2 cos( )]/ c , , , (1) where γ α,β is the interfacial tension between phases α and β and θ α,β,δ is the contact angle between phases α and β and solid surface δ. Although this relationship applies to perfect capillary tubes, it is more complicated in pore spaces as pore geometry affects the capillary pressure. 2,3 In addition to CO 2 sequestration, the prediction of IFT in CO 2 enhanced oil recovery (EOR) processes is important since IFT (or γ) controls the spreading coefficient of one fluid over the others 4 according to γ γ γ = − − S gas,brine oil,gas oil,brine (2) The measurement of IFT and CA at reservoir conditions is usually very challenging, especially in the presence of cocontaminants such as SO 2 and NOx. Therefore, the ability to accurately predict these properties at reservoir conditions can be very useful. Several research groups have used molecular dynamics (MD) modeling to estimate the IFT and CA of CO 2 / water/quartz systems under various conditions. 5-12 In 2007, Vega and Miguel calculated the surface tension of water in order to compare different water models. 13 Their study showed that among molecular models for water, TIP4P/2005 14 could produce the closest match to experimental data. Other researchers have reached the same conclusion by using a dummy particle to carry the negative charge of oxygen atom in various water models. 14,15 Older water models (e.g., TIP3P, 16 SPC, 17 TIP4P, 14 and SPC/E 18 ) were not as accurate as TIP4P/ 2005 in predicting the behavior of water molecules during MD simulations. Several models were also suggested for CO 2 such as MSM, EPM2, TraPPE, Errington, etc. 5 In order to find the best model for CO 2 /water systems at high pressure (P) and temperature (T), Liu et al. examined several existing models for water (SPC, TIP4P, TIP4P2005) and CO 2 (EPM2, TraPPE) and found that TraPPE/TIP4P2005 and EPM2/SPC combi- nations provided the most accurate predictions of CO 2 solubility and IFT variations with P and T. 6,16 Kvamme et al. used the SPC/E model for water and a simple EMP model for CO 2 to study the IFT between CO 2 and water at 278-335 K temperatures and 0.1-20 MPa pressures. Their results were in good agreement with experimental data and revealed a decrease in IFT with increasing pressure. 7 More recently, Li et al. studied the IFT of CO 2 /brine systems at higher P and T using TIP4P and SPC/E models for water and EPM2 model for CO 2 . 8 They showed that the SPC/E water model gives a better match between experimental and simulation data. However, the simulation errors were higher at elevated pressures. In addition to IFT, the CA between two fluid phases and a solid phase controls the capillary pressure of the system. Using Received: February 3, 2015 Revised: April 14, 2015 Published: May 12, 2015 Article pubs.acs.org/Langmuir © 2015 American Chemical Society 5812 DOI: 10.1021/acs.langmuir.5b00445 Langmuir 2015, 31, 5812-5819