J. Daniel Gezelter of the University of Notre Dame is supported by an award from the Theoretical and Computational Chemistry program to develop and test efficient methods for carrying out molecular dynamics (MD) simulations of solvated metal nanoparticles. These particles are simulated as they interact with capping agents and other species in solution. His group is developing empirical metal-solvent interaction potentials, as well constant-pressure simulation algorithms for non-periodic systems.
The ultimate goal of the research is to understand how metallic nanoparticles can be adapted for use in photothermal therapies, and to provide a detailed picture of how heat is transferred between the nanoparticles and the surrounding solvent. All of the methods and code developed for this grant are to be released with an open source license. Dr. Gezelter also directs "The OpenScience Project", a web-based directory of open source scientific software which highlights excellent examples of scientific tools and research codes which are available to other researchers and to the public.
The work supported by this grant led to the creation of algorithms that allow computers to probe chemical systems in ways that were not previously possible. One area of work was in reverse non-equilibrium molecular dynamics (RNEMD) methods that drive simulations in ways that allows the user to measure how heat and material are transported on a molecular scale. Using the newly-developed algorithms, it is now possible for simulations to predict thermal conductivity at interfaces, shear viscosity (and interfacial friction), diffusion (and permeability through interfaces). The most recent generalization of this algorithm now works in non-rectangular and shape-adapting simulation cells so that the same physical properties can be predicted for the curved interfaces of nanoparticles. The work also included algorithms to simulate large structures like proteins and nanoparticles that are moving through a solvent, but does not require any atoms from the solvent to be present in the simulation. This speeds up the simulations by a large amount. The supported work also applied these new algorithms to materials of interest in the materials science, catalysis and nano-medicine communities. The PI and his group worked extensively on heat flow between metal nanoparticles and their environments, a key aspect of photothermal therapies that employ metal particles that are as small as billionth of a meter. At the surfaces of metal nanoparticles, the chemical bond between the metal and the protecting molecules bound to them introduces an additional avenue for heat to escape. The vibrational motion of the surface molecules is similar to the vibrational motions in the surrounding solvent, and this efficiently moves heat away from the metal particles. This grant helped show that the presence of a partial-layer of surface coverage results in more efficient heat transfer than from the barre particles. The RNEMD methodology was also used to study flow of water over the facets of ice crystals. It was used to discover that water flowing over the basal face of ice (the large flat face of snowflakes) exerts a much larger frictional force than when it flows over the prismatic facets at the edges of a snow flake. One of the applications was to simulate and understand of how the surfaces of metallic catalysts (e.g. the Platinum 557 crystal facet) behave in the presence of reactant gases like carbon monoxide. This grant helped to provide a mechanism for the surface restructuring of metallic catalysts under operating conditions.
