To assess the long-term behavior of the carbon dioxide disposal and geological storage, it is important to understand the phase behavior of CO2-H2O mixtures over a broad range of thermodynamic conditions, as well as in confined geometries. CMI researchers Pablo Debenedetti, Athanassios Panagiotopoulos, and Jeroen Tromp are using a variety of molecular modeling tools to study CO2 hydrate phase behavior, the kinetics of hydrate formation and dissociation, and phase behavior of CO2/water and CO2/brine mixtures. These studies will provide insights important to geological storage of carbon dioxide as well as for the rational design of hydrate-based CO2 storage systems over broad ranges of temperature and pressure.


Monte Carlo simulations of CO2+H2O+NaCl mixtures in geological environments

During the past year, Pablo Debenedetti and colleagues investigated the phase behavior of CO2+H2O+NaCl mixtures, a system that is commonly seen in geological environments relevant to CO2 capture and storage. The solubility of CO2 in brine was calculated via Gibbs Ensemble Monte Carlo simulations over the broad temperature and pressure ranges of 50 to 250°C and 0 to 600 bar, and the NaCl molality range of 0 to 6 mol/kg. The SPC, EPM2 and SD models were used to represent water, carbon dioxide and sodium chloride, respectively.

The computed CO2 solubility was found to possess the qualitatively correct monotonic dependence on pressure. Reasonable agreement between experimental data and simulation results is achieved at 150°C and at 200°C. At temperatures below 150°C or above 200°C, the simulations underpredict the solubility of CO2 by 10 to 50% percent, depending on the temperature, pressure and NaCl molality. The solubility of CO2 was found to decrease as the NaCl concentration increased in the liquid phase, indicating a strong salting-out effect.

The next step in this project will be to perform an extensive investigation of various intermolecular potentials for water, carbon dioxide and sodium chloride in order to assess their ability to reproduce the experimental phase behavior over even broader temperature and pressure ranges. The ultimate goal of this study is to understand and quantify the influence of NaCl on the phase behavior of CO2+H2O mixtures over a broad range of temperatures and pressures relevant to carbon capture and storage, as well as to CO2-assisted geothermal energy production.


Molecular simulation of hydrate melting and formation

Engineering any technology to sequester CO2 in the solid form as a hydrate, or as pool of liquid CO2 below a cap of its hydrate, requires an understanding of both the thermodynamics and kinetics of CO2 hydrate formation. Having completed a computational study of CO2 hydrate dissociation, attention during the past year focused on the kinetics of hydrate formation. The methane-water system was chosen as a starting point, as the corresponding hydrate forms under milder conditions than for CO2. Several 1-microsecond-long molecular dynamics simulations of supersaturated methane-water solutions at 240 K and 200 bar were performed. A hydrate phase was nucleated in every simulation under these conditions.

This study represents the first example of hydrate nucleation in the absence of an interface. The final structures formed in the course of the simulations comprised elements of sI hydrate structure (i.e., 512 (twelve pentagons) and 51262 cages (twelve pentagons and two hexagons) – see Figure 12), but lacked long-range order. This is consistent with previous studies of hydrate nucleation and indicates that the formation of a crystalline hydrate phase may involve two steps: formation of a hydrate-like cluster and growth of the crystalline hydrate phase from the cluster. In addition to the cages found in the sI hydrate structure, a few 51263 and 51264 cages were also present. In all simulations, the 512 cages appeared first, followed by 51262 cages. Further analysis of the simulations to understand the mechanisms of hydrate nucleation is currently in progress.

Although this study provides important insights into hydrate nucleation, this approach does not yield the statistics needed to compute the rate of hydrate nucleation accurately. To this end, the Debenedetti Group is currently investigating the use of path-sampling techniques such as forward-flux sampling. This should eventually allow the accurate estimation of hydrate nucleation rates under conditions relevant to carbon capture.

Figure 12. Snapshot of a configuration obtained after 1 μs‐long molecular dynamics simulation of the methane‐water system at 240 K and 200 bar. The different water cages formed are shown (cyan: 512, red: 51262; yellow: 51263 ). For clarity, only the water molecules forming cages are shown.


Molecular modeling of CO2 capture and storage

A recently started (Oct. 2011) project, led by Pablo Debenedetti and new CMI researchers Athanassios Panagiotopoulos and Jeroen Tromp, aims to develop molecular-based computational tools for predicting the physical and chemical behavior of systems relevant to CCS.

In the first three months of the project, Debenedetti and Panagiotopoulos have recruited Arun Prabhu, one of the top first-year graduate students in Chemical and Biological Engineering, to work on the CMI project. Arun will be completing his required core courses in the Spring of 2012 and will start simulations related to phase behavior and transport properties of CO2-brine mixtures in the early summer. In order to rapidly optimize parameters for describing the phase behavior, a molecular-based equation-of-state description (based on the Statistical Associating Fluid Theory, or SAFT, approach) will be developed by Dr. Thomas Lafitte, a postdoctoral associate in the Panagiotopoulos Group who has significant experience with SAFT modeling.

Tromp and his group are working to improve CO2 sequestration monitoring by focusing on proper representation of physical properties in porous reservoirs. Specifically, the researchers are investigating the importance of poroelasticity by contrasting poroelastic simulations with elastic and acoustic simulations. Discrepancies highlight a poroelastic signature that cannot be captured using an elastic or an acoustic theory and that may play a role in accurately imaging and quantifying injected CO2.

Future research will also include investigation of fundamental physicochemical characteristics required for understanding and rational design of CO2 separation processes from flue gases using novel solid adsorbents.