Interface thermal conductance and the thermal conductivity of multilayer thin films

10.1068/htwi9 ◽  
2000 ◽  
Vol 32 (2) ◽  
pp. 135-142 ◽  
Author(s):  
David Cahill ◽  
Andrew Bullen ◽  
Seung-Min Lee
Author(s):  
Jie Zhu ◽  
Dawei Tang ◽  
Wei Wang ◽  
Jun Liu ◽  
Ronggui Yang

The thermal conductivity of thin films and interface thermal conductance of dissimilar materials play a critical role in the functionality and the reliability of micro/nano-materials and devices. The transient thermoreflectance methods, including the time-domain thermoreflectance (TDTR) and the frequency-domain thermoreflectance (FDTR) techniques are excellent approaches for the challenging measurements of interface thermal conductance of dissimilar materials. A theoretical model is introduced to analyze the TDTR and FDTR signals in a tri-layer structure which consists of metal transducer, thin film, and substrate. Such a tri-layer structure represents typical sample geometry in the thermoreflectance measurements for the thermal conductivity and interface thermal conductance of thin films. The sensitivity of TDTR signals to the thermal conductivity of thin films is analyzed to show that the modulation frequency needs to be selected carefully for a high accuracy TDTR measurement. However, such a frequency selection is closely related to the unknown thermal properties and consequently hard to make before the measurement. Fortunately this limitation can be avoided in FDTR. Depending on the modulation frequency, the heat transport in such a tri-layer could be divided into three regimes based on the thickness of the film and the thermal penetration depth, the thermal conductivity of thin films and interface thermal conductance can be subsequently obtained by fitting different frequency regions of one FDTR measurement curve. FDTR measurements are then conducted along with the aforementioned analysis to obtain the thermal conductivity of SiO2 thin films and interface thermal conductance SiO2 and Si. FDTR measurement results agree well with the TDTR measurements, but promises to be a much easier implementation than TDTR measurements.


2010 ◽  
Vol 108 (9) ◽  
pp. 094315 ◽  
Author(s):  
Jie Zhu ◽  
Dawei Tang ◽  
Wei Wang ◽  
Jun Liu ◽  
Kristopher W. Holub ◽  
...  

2020 ◽  
Vol 56 (3) ◽  
pp. 2241-2274
Author(s):  
S. Q. Jia ◽  
F. Yang

Abstract Copper/diamond composites have drawn lots of attention in the last few decades, due to its potential high thermal conductivity and promising applications in high-power electronic devices. However, the bottlenecks for their practical application are high manufacturing/machining cost and uncontrollable thermal performance affected by the interface characteristics, and the interface thermal conductance mechanisms are still unclear. In this paper, we reviewed the recent research works carried out on this topic, and this primarily includes (1) evaluating the commonly acknowledged principles for acquiring high thermal conductivity of copper/diamond composites that are produced by different processing methods; (2) addressing the factors that influence the thermal conductivity of copper/diamond composites; and (3) elaborating the interface thermal conductance problem to increase the understanding of thermal transferring mechanisms in the boundary area and provide necessary guidance for future designing the composite interface structure. The links between the composite’s interface thermal conductance and thermal conductivity, which are built quantitatively via the developed models, were also reviewed in the last part.


2012 ◽  
Vol 1404 ◽  
Author(s):  
Anthony Frachioni ◽  
Bruce White

ABSTRACT1020 Joules of energy are generated by the United States each year; 60% of this energy is lost to waste heat [1]. Thermoelectric based energy scavenging has tremendous potential for the recovery of significant quantities of this waste heat. However, utilization of thermoelectric devices is limited due to relatively low energy conversion efficiency and the utilization of relatively scarce materials. This work focuses on generating sustainable and efficient thermoelectric materials through modifications to the lattice vibrations of materials with excellent thermoelectric electronic properties (Seebeck coefficients larger than 500 μV/K). In particular, Anderson localization of phonons in random multilayer thin films has been explored as a means for reducing lattice thermal conductivity to values approaching that of aerogels (∼10 mW/m-K). Silicon has been a sample of choice due to its high crust abundance and Seebeck coefficient. Reverse non-equilibrium molecular dynamics simulations have been utilized to determine the thermal conductivity of structures of interest. Simulations with pure Lennard-Jones argon solids have been performed to establish a methodology and to characterize the effect of different kinds of disorder prior to the examination of silicon. The simulation results indicate that mass disorder confined to randomly selected planes to be an effective way in which to reduce lattice thermal conductivity with the lattice thermal conductivity decreasing by a factor of thirty (to 4 mW/m-K) in the argon case and a factor of over ten thousand (to 15 mW/m-K) for silicon. Based on models in which the charge carrier mean free path is limited by scattering from the planes with mass disorder, the mobility of silicon is expected to reach values of 10 cm2/V-s. At this mobility the thermoelectric figure of merit, ZT, (utilizing the Wiedeman-Franz law to calculate the electronic thermal conductivity) varies between 4.5 and 11 as the mass ratio of the disordered planes is varied from 4 to 10 in 20% of the lattice planes. These results indicate that the pursuit of nanostructured thermoelectric materials in the form of random multilayers may provide a path to efficient and sustainable thermoelectric materials.


1999 ◽  
Vol 121 (4) ◽  
pp. 945-953 ◽  
Author(s):  
G. Chen

Heat conduction in thin films and superlattices is important for many engineering applications such as thin-film based microelectronic, photonic, thermoelectric, and thermionic devices. Past modeling efforts on the thermal conductivity of thin films were based on solving the Boltzmann transport equation that treats phonons as particles. The effects of phonon interference and tunneling on the heat conduction and the thermal conductivity of thin films and superlattices remain to be explored. In this work, the wave effects on the heat conduction in thin films and superlattices are studied based on the consideration of the acoustic wave propagation in thin film structures and neglecting the internal scattering. A transfer matrix method is used to calculate the phonon transmission and heat conduction through these structures. The effects considered in this work include the phonon interference, tunneling, and confinement. The phonon dispersion is considered by introducing frequency-dependent Lamb constants. A ray-tracing method that treats phonons as particles is also developed for comparison. Sample calculations are performed on double heterojunction structures resembling Ge/Si/Ge and n-period superlattices similar to Ge/Si/n(Si/Ge)/Ge, It is found that phonon confinements caused by the phonon spectra mismatch and by the total internal reflection create a dramatic decrease of the overall thermal conductance of thin films. The phonon interference in a single layer does not have a strong effect on its thermal conductance but for superlattice structures, the stop bands created by the interference effects can further reduce the thermal conductance. Tunneling of phonon waves occurs when the constituent layers are 1–3 monolayer thick and causes a slight recovery in the thermal conductance when compared to thicker layers. The thermal conductance obtained from the ray tracing and the wave methods approaches the same results for a single layer. For superlattices, however, the wave method leads to a finite thermal conductance even for infinitely thick superlattices while the ray tracing method gives a thermal conductance that decreases with increasing number of layers. Implications of these results on explaining the recent thermal conductivity data of superlattices are explored.


Sign in / Sign up

Export Citation Format

Share Document