When phonons are incident on a boundary between two dissimilar materials, only a fraction of the incident phonon energy is transmitted due to interfacial thermal resistance. Although many studies have investigated this phenomenon, there remains no generally accepted and complete theoretical explanation for its occurrence. Nonetheless, as the pending technological revolution progresses towards nanoscale engineering components, the ability to predict and control energy transport across all interfaces becomes increasingly important. Research in this area has primarily focused on a single interface; however, in a complex structure of many boundaries, the interaction among interfaces must be considered.

Molecular dynamics (MD) simulations of three-dimensional Lennard-Jones solids are conducted to investigate the interfacial thermal resistance effects that result in a superlattice constructed of multi-layer films of varying thicknesses. Temperature (T) effects are also considered. Preliminary MD results indicate that for simple dual film structures, interfacial thermal resistance appears to be independent of film thickness and is proportional to T ^-3, as similarly predicted by various theories. In more complex geometries such as the superlattice shown in Figure 1, the story becomes increasingly complex. Figure 2 illustrates the temperature gap that arises in a krypton-argon-krypton structure in which each film is 8 atomic planes thick; a heat flux is imposed across the structure and the steady state time averaged temperature of each atomic plane is recorded. Consequently, a temperature gap exists between the two materials at both interfaces; however, the gap is larger at the lower temperature. For complex superlattices of more than three films, it is expected that the interaction among interfaces will result in a more erratic response than in the simple dual interface case. Film thickness and superlattice period are expected to play an important role.

Molecular dynamics also provides the means for determining the phonon dispersion relation of the superlattice by analyzing the time dependent velocities and positions of the atoms. Preliminary results indicate that MD predicts the additional frequency gap at the wave vector corresponding to the periodicity, d₀, of the superlattice as indicated in Figure 3. Thin film thickness, atomic scale defects (imperfections and dislocations), superlattice periodicity and temperature will influence the shape of the dispersion curve and the size of the frequency gap and thereby affect the heat transport across the structure. Since MD is a versatile tool that offers the capability of easily varying numerous parameters, further computational manipulation of these effects will reveal how they may influence nanoscale heat transfer in multi-interfacial materials. Consequently, phonon filtering capabilities of real engineering nanostructures may be revealed, offering insight into potential design opportunities for improved thermal control in nanodevices.

References

Abramson, A. R. and C. L. Tien, "Interfacial Thermal Resistance in Superlattice Structures," presented at Heat Transfer and Transport Phenomena in Microsystems, Banff, Canada (2000).

Last updated November 4, 2000.
Please send your comments and suggestions to Alexis Abramson