Army Research Laboratory Organization, Mission, and Research:
Combining Design with Materials Science for Next-Generation Materials Development
Dr. Mark A. Tschopp, Dr. Efraín Hernández-Rivera
Lightweight and Specialty Metals Branch, Weapons and Materials Research Directorate,
U.S. Army Research Laboratory, Adelphi (MD)
The mission of the U.S. Army Research Laboratory (ARL) is to discover, innovate, and transition science and technology to ensure dominant strategic land power. In order to accomplish this mission, ARL is implementing a materials-by-design approach to discover and develop materials that will enable us to address the conflicts of the future.
One such program is based on understanding and engineering the relationship between interfaces and material properties in boron-based armor ceramics. Interfaces play a commanding role in the bulk properties of polycrystalline metals, interacting with dislocations or cracks, absorbing defects and solute atoms, and evolving with stress and/or temperature. Suffice it to say that understanding the structure-property relationships of interfaces in metals is critical to designing new material systems. To access these important relationships, we need material models (interatomic potentials) that can describe and predict atomic interactions with high fidelity. Interatomic potentials serve as a bridge between quantum mechanics and molecular dynamics that enables simulations that can explore phenomena and mechanisms at larger length-/time-scale simulations by reducing the degrees of freedom associated with the material model.
In the present work, we discuss a framework for quantifying the sensitivity and uncertainty in properties of interatomic potentials using statistical sampling approaches combined with mathematical models (i.e., metamodels, surrogate models) to describe the relationship between potential parameters and potential properties. In short, this presentation will detail (1) the U.S. ARL’s motivation for studying how interfaces influence micro-structural properties and how these can be tailored to enhance bulk-material properties; (2) the exploration of properties in the hyper-dimensional potential parameter space (and the reduction of this space, if necessary); (3) the selection of appropriate metamodels using various criteria; and (4) the validation and use of these models to obtain the Pareto frontier of optimal interatomic potentials. This approach has been applied to MEAM and ReaxFF potentials in the Fe-He system, C-H system, and B-C system.