AMOEBA Polarizable Force Field for Molecular Dynamics Simulations of Glyme Solvents.

Classical molecular dynamics (MD) simulations of electrolyte systems are important to gain insight into the atom-scale properties that determine the battery-relevant performance. The recent Tinker-HP software release enables efficient and accurate MD simulations with the AMOEBA polarizable force field. In this work, we developed a procedure to construct a universal AMOEBA model for the solvent family of glymes (glycol methyl ethers), which involves a refinement scheme for valence parameters by fitting the AMOEBA-derived atomic forces to those computed at the DFT level. The refined AMOEBA model provides a good description of both local and nonlocal properties in terms of the spectroscopic response of glyme molecules, as well as the liquid glyme density and dielectric constant. In addition, the complexation energies of alkali and alkaline-earth metal cations with tetraglyme molecules obtained from AMOEBA calculations are in good agreement with DFT results, demonstrating the suitability of the developed AMOEBA model for an accurate simulation of glyme-based battery electrolytes. We also expect the procedure to be transferable to the development of AMOEBA models for other battery electrolyte systems.

Combining Electromagnetic Spectroscopy and Ground-Penetrating Radar for the Detection of Anti-Personnel Landmines.

Dual mode detectors combining metal detection and ground-penetrating radar are increasingly being used during humanitarian demining operations because of their ability to discriminate metal clutter. There are many reports in the academic literature studying metal detector and ground-penetrating radar systems individually. However, the combination of these techniques has received much less attention. This paper describes the development of a novel dual modality landmine detector, which integrates spectroscopic metal detection with ground-penetrating radar. This paper presents a feature-level sensor fusion strategy based on three features extracted from the two sensors. This paper shows how the data from the two components can be fused together to enrich the feedback to the operator. The algorithms presented in this paper are targeted at automating the location of buried, visibly obscured objects; however, the system described is also capable of collecting information which could also be used for the potential classification of such items.