Susan Sinnott

Dr. Sinnott’s research program uses computational atomistic methods to design and investigate materials. This area has seen tremendous growth in the last two decades because of a combination of factors, including the increasing availability and low cost of fast computers, the refinement of atomistic methods, the shrinking of device dimensions, and the improved ability of experimentalists to study materials at the nanometer scale. It approaches well-established continuum level modeling (such as finite element analysis) and fluid dynamics at high length scales (100s-1000s nanometers), and overlaps with traditional physics and chemistry at small length scales (1-10 nanometers). The specific materials examined in my group include polymers, ceramics, metals, and electronic materials.

Research in the Sinnott Group is focused on the application of computational methods at the electronic-structure and atomic scales to (1) examine the chemical modification of polymer and composite surfaces; (2) investigate the influence of grain boundaries, point defects, and heterogeneous interfaces on material properties; (3) design materials using a combination of computational methods, experiment, and data informatics within an interdisciplinary research team; and (4) determine the physical, chemical, optical and electrical properties of surfaces, nanostructures, and doped materials.

A major area of emphasis is the development of inventive methods to enable the modeling of new material systems at the atomic level. This includes extending a very popular reactive atomic-scale method to model hydrocarbon and carbon-based systems to include fluorine, oxygen, and sulfur. Reactive methods allow for bond breaking and new bond formation to occur during a simulation and are thus distinguishable from the force fields used in biological studies that are primarily used to optimize molecular geometries. A many-body, reactive method has also been developed to model molybdenum disulfide, a material of interest as a solid-state lubricant and of increasing interest as a graphene-like lamellar material. Current efforts are focused on development and extension of a reactive method that allows for the modeling of heterogeneous systems that include materials with covalent, metallic, and ionic bonding within the same unit cell. This approach, the charge optimized many-body (COMB) potentials for the atomic-scale modeling of materials, has been incorporated into the open-source massively parallel molecular dynamics software developed at Sandia National Laboratory to make them available to the scientific and engineering communities after rigorous testing.