Thermal Boundary Conductance: A Materials Science Perspective

Abstract

The thermal boundary conductance (TBC) of materials pairs in atomically intimate contact is reviewed as a practical guide for materials scientists. First, analytical and computational models of TBC are reviewed. Five measurement methods are then compared in terms of their sensitivity to TBC: the 3ω method, frequency-and time-domain thermoreflectance, the cut-bar method, and a composite effective thermal conductivity method. The heart of the review surveys 30 years of TBC measurements around room temperature, highlighting the materials science factors experimentally proven to influence TBC. These factors include the bulk dispersion relations, acoustic contrast, and interfacial chemistry and bonding. The measured TBCs are compared across a wide range of materials systems by using the maximum transmission limit, which with an attenuated transmission coefficient proves to be a good guideline for most clean, strongly bonded interfaces. Finally, opportunities for future research are discussed.

Thermal boundary conductance of two-dimensional MoS2 interfaces

“…In summary, we have described detailed MD simulations examining the thermal boundary conductance of a 2D material (one to five layers of MoS2) with two technologically relevant insulating substrates, SiO2 and AlN. These TBC values are near the lower limit of TBCs known for solid-solid interfaces, 9,10 due to weak van der Waals bonding and PDOS mismatch between the 2D material and 3D substrate. Nevertheless, the TBC could be increased by maintaining clean interfaces, by strengthening the bond with the substrate, by improving the PDOS matching, or by increasing the number of 2D layers.…”

“…In comparison, the low end of known solid-solid interface TBC is ~8 MWm -2 K -1 for Bi-diamond, only a factor of two to three lower, while the upper end of known TBC is ~14 GWm -2 K -1 for Pd-Ir interfaces where electronic heat conduction is significant. 9,10 The TBC of graphene with SiO2 has been measured in the range of 25 to 65 MWm -2 K -1 , weakly proportional to the number of layers. 38,39 From Fig.…”

“…In particular, recent experiments have found that the interface between MoS2 and SiO2 exhibits a high thermal (Kapitza) resistance, equivalent to that of ~90 nm of SiO2, 4 which is among the highest thermal resistances for solid-solid thermal interfaces. 9,10 However, a detailed theoretical understanding of this thermal interface is presently lacking, an important ingredient necessary for further development and optimization of 2D nanoelectronics.…”

Microscopic nonequilibrium energy transfer dynamics in a photoexcited metal/insulator heterostructure

“…The diffuse mismatch model is used widely to describe the thermal boundary or Kapitza resistance [18] with extensions including optical phonons [19], inelastic scattering [20,21], and interface roughness [22]. By now it is clear that the strength of interface bonding as well as phonon dispersion and population are decisive [23]. In case of metal-insulator interfaces an additional electronic contribution to the Kapitza conductance may become important and lead to a considerable increase in the energy transfer due to scattering of electrons with interface vibrational modes [20,24].…”

Phonon Dynamics at an Oxide Layer in Silicon: Heat Flow and Kapitza Resistance

“…The experimental and theoretical studies on Si/SiO x interfaces have focused so far on the interface thermal boundary (or Kapitza) resistance R K . [13] Experimental data have been reported [14][15][16] for various Si/SiO x interfaces, with large uncertainties in the Kapitza resistance: 20 Â 10 À9 [14] to AE4.5 Â 10 À9 [15] and 2.3 Â 10 À9 Km 2 W À1 [16] near room T. These differences could be related to the way the oxide layers were grown, leading to distinct interfaces, as well as differences in the measurement techniques which often involve a second (metallic) interface.…”