Micro/Nano Scale Heat Transfer Lab
Aerospace and Mechanical Engineering
865 Asp Avenue, Room 217
University of Oklahoma
Norman, OK 73019
Ph: 405 325 5232
Ph.D Massachusetts Institute of Technology, 2011
Our group focuses on design of advanced composite materials for thermal management and energy conversion applications. Experimental research involves material synthesis and characterization of thermal properties. Simultaneously we use first-principles atomistic simulations to elucidate the mechanisms of thermal transport processes in a wide range of materials including bulk semiconductors, nanostructured materials such as superlattices, disordered materials like alloys, and polymer composites.
High thermal conductivity (k) aligned polymer-graphene nanocomposites
The goal of this research is to achieve polymer-graphene nanocomposites with two orders of magnitude higher k compared to pristine polymers by realizing the full potential of ultra-high k of graphene, first by aligning its high k planar surface with the direction of heat transfer and second by engineering vastly superior thermal interaction between graphene and polymer through the idea of edge-bonding (discussed below). Unstrained polymeric systems contain ordered-crystalline regions called lamellae which can have thermal conductivities exceeding those of metals. Random orientation of polymer lamellae, however, leads to an overall low thermal conductivity. Mechanical strain aligns these high thermal conductivity regions along one direction, enhancing polymer k. Similarly, graphene nanoplatelets (GnPs) have very high in-plane k (~1000-2000 W/mK) but low out-of-plane k (~10-20 W/mK). Aligning the planar surface of GnPs along the direction of heat transfer realizes full advantage of their high in-plane k. Through simultaneous alignment of polymer lamellae and dispersed GnPs, we recently achieved 1100% enhancement in composite k using just 6 volume% GnPs. These promising results have been published in the journal of Nanoscale, DOI: 10.1039/C7NR04686C. [pdf] High thermal conductivity achieved through alignment effects can improve thermal management in wide array of technologies including electronics cooling and automotive and aerospace applications.
Edge-bonding scheme with superior interface thermal conductance
As a second effect we are exploring the large potential of edge-bonding in enabling vastly superior thermal interaction between polymer and graphene. Enhancing interface thermal conductance between polymer and graphene is key to achieving high composite k. Edge-bonding offers the advantage of allowing all sheets of a graphene nanoplatelet (GnP) to be covalently bonded with the surrounding polymer matrix enabling efficient heat conduction from the polymer to the entire nanoplatelet. Bonding on the basal plane only couples outermost layers of the nanoplatelet. Interface thermal conductance between polymer and graphene is also higher at a junction on the edge relative to on the basal plane of graphene. We demonstrated this effect using state-of-the-art nonequilibrium Green’s function method. Interatomic force interactions needed for the analysis are derived ab initio allowing a complete first-principles prediction of interface thermal conductance. Initial results indicate edge bonding to enable almost two-fold higher interfacial thermal conductance through efficient coupling of polymer phonons with the in-plane vibrations of graphene. These results were recently published in Applied Physics Letters, 110, 093112 (2017)[pdf]. Through a combination of above two effects, edge bonded nanoplatelet provides efficient heat transfer through the composite seen in the form of superior thermal mixing in the images on the left. We are experimentally exploring the superior effect of edge-bonding using edge-specific functionalization schemes.
Near-field radiative heat transfer
When the separation of two surfaces approaches sub-nanometre scale, the boundary between the two most fundamental heat transfer modes, heat conduction by phonons and radiation by photons, is blurred. We have developed an atomistic framework based on microscopic Maxwell’s equations and lattice dynamics to describe the convergence of these heat transfer modes and the transition from one to the other. The formalism uses non-equilibrium Green’s function method to compute energy transmission across the gap. Short range force interactions are modeled using empirical nearest neighbor potential. Long range Coulomb forces are computed using the electric fields generated by oscillating ions. For gaps>1 nm, heat transfer occurs through resonant coupling of electric fields with high-frequency optical phonons, that is phonon-polaritons. At these gaps, the predicted conductance is in excellent agreement with Rytov’s continuum theory of fluctuating electrodynamics. However, for sub-nanometre gaps we find the conductance to be enhanced up to four times compared with continuum theory. This enhancement is found to be due to tunneling of low frequency acoustic phonons through the vacuum gap assisted by coupling with evanescent electric fields, providing additional channels for energy transfer. Our results were recently published in Nature Communications 6, 6755 (2015) [pdf] and hold importance for accurate prediction of heat transfer across small gaps in technologies such as heat assisted magnetic recording and nanoporous materials.
First-principles study of thermal transport in semiconductor materials
We are using state-of-the art first-principles methods to gain fundamental understanding of thermal transport processes in wide array of semiconductor materials including bulk, nanostructured and disordered semiconductors. Dominant heat carriers in these materials are lattice vibrations called phonons. We use highly accurate harmonic and anharmonic interatomic force interactions derived from density-functional theory along with an exact solution of phonon Boltzmann transport equation to achieve unprecedented accuracy in the prediction of phononic thermal conductivity in these materials. This is seen by the excellent agreement between measured and predicted k values for natural GaN [pdf], GaAs [pdf] and PbTe [pdf]. We have also used these methods to accurately predict thermal conductivity of alloys, recently presented in Physical Review Letters [pdf], and to demonstrate coherent transport of phonons in superlattices, presented in the journal of Science [pdf].
We are also using the outlined ab initio framework to explore another unique effect in the thermal conductivity of group III-V semiconductors, namely, the suppression of anharmonic phonon scattering through the existence of an energy gap in the phonon dispersions of these materials. In Silicon (with no energy gap) an acoustic phonon can scatter by absorbing another acoustic phonon to convert into an optical phonon. Such an absorption channel satisfies both energy and momentum conservation. Energy gap in the phonon dispersion of BAs, however, forbids such absorption scattering channels since the third phonon mode needed for energy conservation does not exist (required frequency for energy conservation lying in the forbidden energy gap). This dramatically enhances phonon lifetimes by almost two orders of magnitude, thus enhancing thermal conductivity. We demonstrated this effect in ideal short period superlattices, in a Nano Letters publication [pdf] and more recently in GaN [pdf] where the effect was also found to lead to a spectral focusing of thermal conductivity in a very narrow frequency range. We are exploring the role of strain in further enhancing the outlined effect by increasing the energy gap further decoupling low and high energy phonons, leading to even larger enhancement in phonon lifetimes. This can provide new avenues to enhance thermal conductivity.
Ab initio methods are also being used to investigate low thermal conductivity materials with application towards high energy conversion efficiency thermoelectrics.
Barrier heterostructures for higher thermoelectric efficiency
The goal of this research is to better understand the role of highly coupled non-equilibrium transport of both phonons and electrons in enhancing energy conversion efficiency of solid state thermionic devices based on superlattice single-barrier heterostructures. We are using first-principles driven coupled electron-phonon Monte Carlo simulations for predicting the performance as a function of device parameters. Effective Seebeck coefficient is predicted as a function of barrier width. Superlattices also offer the advantage of ultra-low thermal conductivity, further enhancing efficiency. Finally role of sub-band formation in superlattices in suppressing joule heating is being studied.