In a recent study [J. Liu et al., J. Chem. Phys. 149, 224104 (2018)], we developed a general mixed quantum-classical framework for studying heat transport through molecular junctions, in which the junction molecule is treated quantum mechanically and the thermal reservoirs to which the molecule is coupled are treated classically. This framework yields expressions for the transferred heat and steady-state heat current, which could be calculated using a variety of mixed quantum-classical dynamics methods. In this work, we use the recently developed “Deterministic Evolution of Coordinates with Initial Decoupled Equations” (DECIDE) method for calculating the steady-state heat current in the nonequilibrium spin-boson model in a variety of parameter regimes. Our results are compared and contrasted with those obtained using the numerically exact multilayer multiconfiguration time-dependent Hartree approach, and using approximate methods, including mean field theory, Redfield theory, and adiabatic mixed quantum-classical dynamics. Despite some quantitative differences, the DECIDE method performs quite well, is capable of capturing the expected trends in the steady-state heat current, and, overall, outperforms the approximate methods. These results hold promise for DECIDE simulations of nonequilibrium heat transport in realistic models of nanoscale systems.

Quantum–classical dynamics simulations enable the study of nonequilibrium heat transport in realistic models of molecules coupled to thermal baths. In these simulations, the initial conditions of the bath degrees of freedom are typically sampled from classical distributions. Herein, we investigate the effects of sampling the initial conditions of the thermal baths from quantum and classical distributions on the steady-state heat current in the nonequilibrium spin-boson model—a prototypical model of a single-molecule junction—in different parameter regimes. For a broad range of parameter regimes considered, we find that the steady-state heat currents are ∼1.3–4.5 times larger with the classical bath sampling than with the quantum bath sampling. Using both types of sampling, the steady-state heat currents exhibit turnovers as a function of the bath reorganization energy, with sharper turnovers in the classical case than in the quantum case and different temperature dependencies of the turnover maxima. As the temperature gap between the hot and cold baths increases, we observe an increasing difference in the steady-state heat currents obtained with the classical and quantum bath sampling. In general, as the bath temperatures are increased, the differences between the results of the classical and quantum bath sampling decrease but remain non-negligible at the high bath temperatures. The differences are attributed to the more pronounced temperature dependence of the classical distribution compared to the quantum one. Moreover, we find that the steady-state fluctuation theorem only holds for this model in the Markovian regime when quantum bath sampling is used. Altogether, our results highlight the importance of quantum bath sampling in quantum–classical dynamics simulations of quantum heat transport.

The kinetic energy is the center of a controversy between two opposite points of view about its role in the formation of a chemical bond. One school states that a lowering of the kinetic energy associated with electron delocalization is the key stabilization mechanism of covalent bonding. In contrast, the opposite school holds that a chemical bond is formed by a decrease in the potential energy due to a concentration of electron density within the binding region. In this work, a topographic analysis of the Hamiltonian Kinetic Energy Density (KED) and its laplacian is presented to gain more insight into the role of the kinetic energy within chemical interactions. This study is focused on atoms, diatomic and organic molecules, along with their dimers. In addition, it is shown that the laplacian of the Hamiltonian KED exhibits a shell structure in atoms and that their outermost shell merge when a molecule is formed. A covalent bond is characterized by a concentration of kinetic energy, potential energy and electron densities along the internuclear axis, whereas a charge-shift bond is characterized by a fusion of external concentration shells and a depletion in the bonding region. In the case of weak intermolecular interactions, the external shell of the molecules merge into each other resulting in an intermolecular surface comparable to that obtained by the Non-covalent interaction (NCI) analysis.

No abstract