Gas-Surface Chemistry (Catalysis)


When a high-temperature gas comes in contact with a solid surface, in addition to the convective/conductive heat flux from the gas to the surface, surface reactions may release further chemical energy to the surface and possibly degrade the surface material. Understanding gas-surface chemistry is important for the design of thermal protection systems (TPS) for hypersonic vehicles and spacecraft. High gas temperatures in the shock layer generated by a hypersonic vehicle dissociate molecules in the gas and the reactive atomic species can diffuse through the boundary layer and react with the TPS surface. As an example, the energy released by the heterogeneous recombination (surface catalysis) of atomic oxygen into molecular oxygen, can contribute significantly to the overall heat flux and heat load experienced by a hypersonic vehicle. CFD surface reaction models (surface boundary conditions) have advanced to a point where elementary gas-surface reaction mechanisms are specified along with steric factors and activation energies for each reaction. Thus, the parameters required for CFD models (and DSMC models) are directly linked to fundamental chemistry. Most often, experiments are unable to determine precise reaction mechanisms and activation energies, and so our research attempts to use computational chemistry (molecular dynamics) to determine or refine these model inputs.

Unlike gas-phase collisions (involving on a few atoms per collision), gas-surface interactions require the simulation of large atomic systems; hundreds to thousands of atoms at least. For many of our problems we use the ReaxFF interatomic potential and we collaborate closely with Prof. Adri van Duin at Penn State University. ReaxFF is a state-of-the-art potential that is fit to quantum chemistry single-point energy data (for a specific problem). The functional form of the fit enables natural bond-breaking and bond-formation (chemical reactions) with a computationally efficient empirical potential. This allows accurate molecular simulation of large gas-surface systems. Our research focuses on validating (and in most cases improving) the potential for a specific gas-surface interaction of interest, predicting realistic surface morphology (specifically surface defect chemistries), and finally predicting gas-surface interactions on such realistic surfaces. Our research started with oxygen-platinum systems where a significant amount of high-quality experimental data was available. Research now focuses on oxygen-silica systems (ceramics) important for a large number of real TPS materials used on hypersonic vehicles.

Results and Publications

Oxygen - Platinum Surface Coverage and Catalysis:

    Valentini, P., Schwartzentruber, T.E., and Cozmuta, I., “ReaxFF Grand Canonical Monte Carlo

    simulation of adsorption and dissociation of oxygen on platinum (111)”, Surface Science, 605    

    (2011), pp. 1941-1950.

   Valentini, P., Schwartzentruber, T.E., and Cozmuta, I., “Molecular dynamics simulation of O2

    sticking on Pt(111) using the ab initio based ReaxFF reactive force field”, Journal of Chemical

    Physics, 133 (2010) 084703.

We first validated a ReaxFF potential parameterized for oxygen-platinum interactions with experimental data from molecular beam experiments. Our results investigated the sticking probability of molecular oxygen on platinum (111) as well as scattering angle distributions compared to molecular beam data. In a second study, we investigated the surface coverage of platinum under exposure to dissociated oxygen under a range of temperatures and pressures. Such surface coverage and surface morphology develops over time scales too large to simulate with molecular dynamics. We developed a Grand Canonical Monte Carlo (GCMC) technique that embedded a reactive potential (ReaxFF) to simulate this problem. Example results are shown below along with an animation of the Monte Carlo simulation process. At low temperatures and pressures our simulations predicted an ordered “2x2” coverage of oxygen atoms, whereas at higher temperatures and pressures our simulations predicted displacement of some of the platinum atoms out of the crystal due to much higher oxygen coverages. Both results were in nice agreement with experimental measurements.

Oxygen - Silica Catalysis:

    Norman, P., Schwartzentruber, T.E., Leverentz, H., Luo, A., Paneda, R.M., Paukku, Y., and    

    Truhlar, D.G., “The Structure of Silica Surfaces Exposed to Atomic Oxygen”, J. Phys. Chem. C

    (2013), 117, pp. 9311-9321.

    Kulkarni, A.D., Truhlar, D.G., Srinivasan, S.G., van Duin, A.C.T., Norman, P., and    

    Schwartzentruber, T.E., “Oxygen Interactions with Silica Surfaces: Coupled Cluster and Density

    Functional Investigation and the Development of a New ReaxFF Potential”, J. Phys. Chem. C

    (2013), 117, pp. 258-269.

    Sorensen, C., Valentini, P., and Schwartzentruber, T.E., “Uncertainty Analysis of Reaction Rates

    in a Finite Rate Surface Catalysis Model”, Journal of Thermophysics and Heat Transfer, Vol. 26,

    No. 3 (2012), pp. 407-416.

In this work we study silicon-dioxide surfaces, both crystalline (quartz and beta-crystobalite) and amorphous (a-SiO2) structures. While experiments have been performed on quartz, it is likely that oxide-layers that form on silica-based TPS materials during hypersonic flight are amorphous. We place a heavy emphasis on first determining the chemical structure of realistic silica surfaces that are representative of actual flight conditions. Only then do we proceed to study gas-surface collisions on these realistic atomic surface morphologies. In general, our approach is to validate the ReaxFF potential for bulk structural properties of SiO2 polymorphs, then through annealing we create and validate SiO2 surfaces, and finally we simulate exposure to high-temperature dissociated oxygen and predict the SiO2 surface structure and chemical defect sites present under flight conditions. The surface structures predicted by ReaxFF are then compared with a large number of targeted density functional theory (DFT) calculations to ensure accuracy. This is an extremely challenging project that requires a collaborative effort between our group, Adri van Duin’s group at Penn State, and Don Truhlar’s group in the department of Chemistry at the University of Minnesota. Through this collaborative effort, we have developed a new ReaxFF potential specific to oxygen-silica interactions.

Our main conclusions for oxygen-silica interactions thus far are that: (1) catalytic reactions only occur on surface defects, and not on crystalline surfaces, (2) we have identified the most dominant defects that participate in catalytic reactions, (3) we have determined that the dominant recombination reaction pathways are non-activated.

We have implemented a finite-rate catalytic boundary condition for CFD codes that could use this data and research is now focused on vibrational energy accommodation and reactions involving excited electronic states.