In 2010, a molecular modeling program was initiated within the Fluids & Energy Group. Pablo Debenedetti, Athanassios Panagiotopoulos, and Jeroen Tromp use molecular simulations to provide new insights into the behavior of CO2 and methane in subsurface environments.

Figure 20: Snapshots depicting the formation of cages during the course of a molecular dynamics simulation of homogeneous nucleation of methane hydrate in water. For clarity, only the water molecules forming the cages are shown (“disordered” water and methane not shown). The cage color code is 512 (green), 51262 (purple), 51263 (orange).


Molecular simulation of hydrate melting and formation

The Debenedetti group is using state-of-the-art molecular modeling tools to gain insights into the mechanisms and rates of melting and formation of carbon dioxide and methane hydrates, as well as on their thermodynamic stability across broad ranges of temperature, pressure and salinity. The studies are relevant to CO2 sequestration (either in the solid form as a hydrate, or as pool of liquid CO2 below a cap of its hydrate), gas storage and transportation, climate change, and ocean stability.

In 2012, the Debenedetti team performed microsecond-long molecular dynamics simulations of homogeneous nucleation of methane hydrate in bulk water, the first time that such nucleation had been simulated in the absence of an interface. The calculations yielded novel insights into the nucleation mechanism, such as the transient appearance of 51263 (twelve pentagons and 3 hexagons per cage) and 51264 cages (twelve pentagons and four hexagons), which are not found in structure I (sI) hydrates; and the aggregation of sub-critical clusters over time scales spanning hundreds of nanoseconds (Figure 20).

The solid phase present at the end of microsecond-long simulations lacks long-range order, however, and is characterized by a ratio of large (51262 )-to-small (512) cages that is significantly smaller than observed experimentally in sI crystals (ca. 1 vs. 3). This indicates that long molecular dynamics simulations, though valuable for providing phenomenological insight into nucleation and melting mechanisms, need to be supplemented by path-sampling techniques in order to yield quantitative information on actual rates of hydrate formation. The implementation of such path-sampling simulations is the focus of our ongoing work, which is being done in collaboration with Athanassios Panagiotopoulos (see next section).

Figure 21: A snapshot from interfacial molecular dynamics simulations of the CO2 -H2 O-NaCl system. Na+ (blue), Cl-(yellow), C (green), H (white) and O (red) atoms are shown explicitly.


Molecular modeling of CO2 capture and storage

In October 2011, a project was initiated within the Fluids & Energy group to develop molecular-based computational tools for predicting fundamental physicochemical characteristics required for understanding and rational design of CO2 separation processes and long-term CO2 storage in geological formations. This ongoing research is a collaboration between Athanassios Panagiotopoulos and Pablo Debenedetti, and Jeroen Tromp.

In the past year, the main focus of the project has been the use of atomistic simulations to obtain the phase behavior and interfacial tension of CO2 -H2O-NaCl mixtures over a broad temperature and pressure range. The researchers demonstrated the applicability of interfacial molecular dynamics methods to the systems and properties of interest. Within the range of the temperature, pressure, salt concentration and system size in our study, they find no NaCl in the CO2 -rich phase at phase coexistence. The work also highlighted the limitations of the existing force fields with fixed-charged, additive pair interactions in predicting phase equilibrium and interfacial properties of the mixtures of interest, suggesting an urgent research need for their improvement.

A PhD student in Chemical and Biological Engineering, Arun Prabhu, was recruited in January of 2012 to work jointly with Professors Debenedetti and Panagiotopoulos in the general area of the CMI project. Arun’s initial studies have focused on hydrate nucleation using bulk and interfacial molecular dynamics simulations, as detailed in Prof. Debenedetti’s section on “Molecular simulation of hydrate melting and formation.”