Role of heat flow direction, monolayer film thickness, and periodicity in controlling thermal conductivity of a Si–Ge superlattice system

Abstract

Superlattices are considered one of the most promising material systems for nanotechnological applications in fields such as high figure of merit (ZT) thermoelectrics, microelectronics, and optoelectronics owing to the possibility that these materials could be tailored to obtain desired thermal properties. Factors that could be adjusted for tailoring the thermal conductivity of the superlattices include the monolayer film thickness, periodicity, heat flow direction, straining, and temperature of operation. In the presented study, nonequilibrium molecular dynamics (NEMD) simulations are performed to obtain an understanding of the effect of such factors on the thermal conductivity of Si–Ge superlattices at three different temperatures (400, 600, and 800 K). The NEMD simulations are performed using Tersoff bond-order potential. The thermal conductivity is found to increase with an increase in the number of periods as well as with the increase in the period thickness. The dependence of thermal conductivity on the direction of heat flow is found to be sensitive to the extent of acoustic mismatch at the interface (i.e., heat flowing from Si to Ge versus heat flowing from Ge to Si in a single period). Superlattices with Ge–Si interfaces (heat flows from Ge monolayer to Si monolayer in a period) are found to have lower thermal conductivity than superlattices with Si–Ge interfaces (heat flows from Si monolayer to Ge monolayer in a period). The superlattices thermal conduction, therefore, can be considered to have a characteristic somewhat similar to a thermal diode. Both compressive and tensile strains are observed to be an important factor in tailoring the thermal conductivity of the analyzed superlattices. Particularly, straining can help in reducing the thermal conductivity. The influence of straining is found to increase with increasing period thickness and periodicity.

Understanding straining induced changes in thermal properties of tropocollagen-hydroxyapatite interfacial configurations

“…Samvedi and Tomar (2009) have used such an approach for a variety of nanocomposite simulations. The emphasis in this work is on the analysis of the effects of component thickness, interfacial period and levels of straining on the thermal diffusivity and thermal conductivity of hierarchical TC-HAP interfacial systems.…”

“…In addition, the system with less number of interfaces also has thicker monolayer films (since overall length is constant) leading to increase in thermal conductivity. The contribution of cumulative effect is that thermal conductivity remains almost constant as a function of increase in number of interfacial periods, (Samvedi and Tomar, 2009) , Figures 7(d). The change of the specific heat and thermal conductivity discussed above and the relation of the diffusivity, specific heat and conductivity expressed by equation (8) lead to the change of the thermal diffusivity with increasing the number of the interfacial periods, i.e., thermal diffusivity increasing as a function of increase in number of interfacial periods, Figures 7(b).…”

“…Thermal conductivity values also increase as the thickness of the systems increases. This trend can be explained by a drop in the overall temperature gradient owing to increase in the overall length Samvedi and Tomar (2009. The size effect can also be explained based on the expression of the thermal conductivity as given by kinetic theory of gases, equation (9).…”

“…Since the simulations are performed at sub-nanometer length scale, the MFP corresponds to the thin film thickness in this case (Samvedi and Tomar, 2009. With the increasing of the thickness of the system, the phonon group velocity does not change at the same temperature.…”

“…The atomistic simulation has been performed on Si, Ge and other semiconductor materials and proved to be the efficient method to analyse the thermal behaviour of solids under the conditions where the experiment is difficult to perform. Samvedi and Tomar (2009 analysed the thermal behaviour of Si-Ge biomimetic nanocomposite as a function of strain and structural arrangement at a range of temperatures using non-equilibrium molecular dynamics (NEMD) simulations. Jeng et al (2008) have used Monte Carlo simulations to study the phonon transport and thermal conductivity reduction in the biomimetic nanocomposites with Si nanoparticle embedded in Ge host.…”