Search for: All records

Thermal transport coefficients, notably the interfacial thermal conductance, were determined in planar and spherical gold interfaces functionalized with CTAB (cetyltrimethylammonium bromide) or MTAB (16-mercapto-hexadecyl-trimethylammonium bromide) using reverse non-equilibrium molecular dynamics (RNEMD) methods. The systems of interest included (111), (110), and (100) planar facets as well as nanospheres (r = 10 Å). The effect of metal polarizability was investigated through the implementation of the density-readjusted embedded atom model (DR-EAM), a polarizable metal potential. We find that conductance is higher in MTAB-capped interfaces, due in large part to the metal-to-ligand coupling provided by the Au-S bond. Alternatively, CTAB does not couple strongly with either the metal or the solvent, and it is largely a barrier to heat transfer, resulting in a much lower interfacial thermal conductance. Through analysis of physical contact between the ligand and the solvent, we find that there is significantly more overlap in the MTAB systems than the CTAB systems, mirroring the trends we observed in the conductance. 

Reverse nonequilibrium molecular dynamics simulations were used to study heat transport in solvated gold interfaces which have been functionalized with a low-molecular weight thiolated polyethylene glycol (PEG). The gold interfaces studied included (111), (110), and (100) facets as well as spherical nanoparticles with radii of 10 and 20 Å. The embedded atom model (EAM) and the polarizable density-readjusted embedded atom model (DR-EAM) were implemented to determine the effect of metal polarizability on heat transport properties. We find that the interfacial thermal conductance values for thiolated PEG-capped interfaces are higher than those for pristine gold interfaces. Hydrogen bonding between the thiolated PEG and solvent differs between planar facets and the nanospheres, suggesting one mechanism for enhanced transfer of energy, while the covalent gold sulfur bond appears to create the largest barrier to thermal conduction. Through analysis of vibrational power spectra, we find an enhanced population at low-frequency heat-carrying modes for the nanospheres, which may also explain the higher mean interfacial thermal conductance (G) value. 

The interfacial thermal conductance from solvated gold nanostructures capped with sodium citrate was determined using reverse nonequilibrium molecular dynamics (RNEMD) methods. The surfaces of spherical nanoparticles and the (111) surfaces of fcc gold slabs were modeled using the density readjusting-embedded atom method (DR-EAM) as well as with the standard embedded atom method (EAM), and the effects of polarizability on the binding preferences of citrate were determined. We find that the binding configurations of citrate depend significantly on gold surface curvature and are not strongly influenced by surface polarizability. The interfacial thermal conductance was also determined for the spherical nanoparticles and (111) surfaces, and we find that applying DR-EAM increases the interfacial thermal conductance for systems with spherical nanoparticles much more sharply than for systems with (111) surfaces. Through analysis of excess charge density near the interface, we find that inclusion of polarizability has a larger impact on image charge creation in nanospheres than it does for the planar (111) interfaces. This effectively increases the interaction strength to polar species in the solvent, yielding larger interfacial thermal conductance estimates for the nanospheres.