Answers are grounded in this paper and cite their sources.
A number of different potentials are currently being used in molecular dynamics simulations of semiconductor nanostructures. Confusion can arise if an inappropriate potential is used. To illustrate this point, we performed direct molecular dynamics simulations to predict the room temperature lattice thermal conductivity λ of thin GaAs, InAs and InP nanowires. In each case, simulations performed using the classical Harrison potential afforded values of λ about an order of magnitude smaller than those obtained using more elaborate potentials (an Abell-Tersoff, as parameterized by Hammerschmidt et al for GaAs and InAs, and a potential of Vashishta type for InP). These results will be a warning to those wishing to use computer simulations to orient the development of quasi-one-dimensional systems as heat sinks or thermoelectric devices.
No takes yet. Share an insight, caveat, or question.
Carrete et al. (2011) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: