February 16, 2024
Journal Article

Quantum bath augmented stochastic nonequilibrium atomistic simulations for molecular heat conduction

Abstract

Classical molecular dynamics (MD) has been shown to be effective in simulating heat conduction in certain molecular junctions since it inherently takes into account some essential methodological components which are lacking with quantum Landauer-type transport model, such as many-body full force-field interactions, anharmonicity effects and nonlinear responses for large temperature biases. However, the classical mechanics reaches its limit in the environments where the quantum effects are significant (e.g. with low-temperatures substrates, presence of extremely high frequency molecular modes). Here, we present an atomistic simulation methodology for molecular heat conduction that incorporates the quantum Bose-Einstein statistics into an ``effective temperature'' in the form of modified Langevin equation. We show that the results from such a quasi-classical effective temperature (QCET) MD method deviates drastically when the baths temperature approaches zero from classical MD simulations and the results converge to the classical ones when the bath approaches the high-temperature limit, which makes the method suitable for full temperature range. In addition, we show that our quasi-classical thermal transport method can be used to model the conducting substrate layout and molecular composition (e.g. anharmonicities, high-frequency modes). Anharmonic models are explicitly simulated via the Morse potential and compared to pure harmonic interactions, to show the effects of anharmonicities under quantum colored bath setups. Finally, the chain length dependence of heat conduction is examined for one-dimensional polymer chains placed in between quantum augmented baths.

Published: February 16, 2024

Citation

Chen R., M. Dinpajooh, and A. Nitzan. 2023. Quantum bath augmented stochastic nonequilibrium atomistic simulations for molecular heat conduction. Journal of Chemical Physics 159, no. 13:Art. No. 134110. PNNL-SA-187803. doi:10.1063/5.0168117

Research topics