Mapping Thickness Dependent Thermal Conductivity of GaN

2016 ◽  
Vol 138 (2) ◽  
Author(s):  
Elbara Ziade ◽  
Jia Yang ◽  
Gordie Brummer ◽  
Denis Nothern ◽  
Theodore Moustaks ◽  
...  

Frequency domain thermoreflectance (FDTR) is used to create quantitative maps of thermal conductivity and thickness for a thinning gallium nitride (GaN) film on silicon carbide (SiC). GaN was grown by molecular beam epitaxy on a 4H-SiC substrate with a gradient in the film thickness found near the edge of the chip. The sample was then coated with a 5 nm nickel adhesion layer and a 85 nm gold transducer layer for the FDTR measurement. A piezo stage raster scans the sample to create phase images at different frequencies. For each pixel, a periodically modulated continuous-wave laser (the red pump beam) is focused to a Gaussian spot, less than 2 um in diameter, to locally heat the sample, while a second beam (the green probe beam) monitors the surface temperature through a proportional change in the reflectivity of gold. The pump beam is modulated simultaneously at six frequencies and the thermal conductivity and thickness of the GaN film are extracted by minimizing the error between the measured probe phase lag at each frequency and an analytical solution to the heat diffusion equation in a multilayer stack of materials. A scanning electron microscope image verifies the thinning GaN. We mark the imaged area with a red box. A schematic of the GaN sample in our measurement system is shown in the top right corner, along with the two fitting properties highlighted with a red box. We show the six phase images and the two obtained property maps: thickness and thermal conductivity of the GaN. Our results indicate a thickness dependent thermal conductivity of GaN, which has implications of thermal management in GaN-based high electron mobility transistors.

2015 ◽  
Vol 137 (2) ◽  
Author(s):  
Jia Yang ◽  
Toshiyuki Sato ◽  
Paul Czubarow ◽  
Aaron Schmidt

Frequency domain thermoreflectance (FDTR) imaging is used to create quantitative thermal conductivity maps of porous Aluminum Nitride (AlN) particles embedded in epoxy. The AlN-epoxy composite is polished and coated with a metal layer. A piezo stage is used to move the sample for imaging with our FDTR system. For each pixel, a periodically modulated continuous-wave laser (the red pump beam) is focused to a Gaussian spot, less than 2 um in diameter, to locally heat the sample, while a second beam (the green probe beam) monitors the surface temperature through a proportional change in the metals' reflectivity. The pump beam is modulated simultaneously at six frequencies and the thermal properties of the AlN composite are extracted by minimizing the error between the measured probe phase lag at each frequency and an analytical solution to the heat diffusion equation in a multilayer stack of materials. A schematic of the AlN sample in our measurement system and an optical image of the polished surface of the AlN-epoxy composite before coating with metal is shown in a. Two scanning electron microscope images of the porous AlN particles prior to embedding in epoxy are shown in b. One of the six simultaneously collected phase images of the probe laser is shown in c. The dark blue regions in the phase image are pits on the sample surface. We fit the six phase images to our thermal model and obtain thermal conductivity maps. The conductivity maps of four particles are shown in d. A log color bar is used to highlight the contrast of thermal conductivity in a single particle. The thermal conductivity of the AlN particles ranges from 80W/mK in the dense regions to 5W/mK in the porous regions.


2015 ◽  
Vol 137 (2) ◽  
Author(s):  
Elbara Ziade ◽  
Aaron Schmidt

Frequency domain thermoreflectance (FDTR) imaging is used to create quantitative maps of both in-plane thermal conductance and cross-plane thermal boundary conductance (TBC) for graphene multilayers encased between titanium and silicon dioxide. A graphene flake is encased between a metal layer and a thermally oxidized p-type silicon wafer and a piezo stage is used to raster scan the sample for imaging. For each image pixel, a periodically modulated continuous-wave laser (the red pump beam) is focused to a Gaussian spot, less than 2 um in diameter, that locally heats the sample, while a second beam (the green probe beam) monitors the surface temperature through a proportional change in reflectivity. The pump beam is modulated simultaneously at six frequencies and the thermal properties of the graphene flake are extracted by minimizing the error between the measured probe phase lag at each frequency and an analytical solution to the heat diffusion equation in a multilayer stack of materials. Phase images at six frequencies for the sample are shown in b. Different layers of the graphene flake are clearly shown in 9.9 MHz and 11.3 MHz images. The six phase data points at every pixel are then fitted to our thermal model to generate two thermal property maps of the graphene flake: in-plane thermal conductance and TBC, shown in c. The in-plane thermal conductance map shows an increased conduction of heat in graphene with the number of layers, while the TBC map indicates a constant cross-plane conduction along the flake. Our imaging technique can be used to study thermal transport in graphene and has implications for thermal management in graphene based electronic devices.


2011 ◽  
Vol 3 (3) ◽  
pp. 301-309 ◽  
Author(s):  
Olivier Jardel ◽  
Guillaume Callet ◽  
Jérémy Dufraisse ◽  
Michele Piazza ◽  
Nicolas Sarazin ◽  
...  

A study of the electrical performances of AlInN/GaN High Electron Mobility Transistors (HEMTs) on SiC substrates is presented in this paper. Four different wafers with different technological and epitaxial processes were characterized. Thanks to intensive characterizations as pulsed-IV, [S]-parameters, and load-pull measurements from S to Ku bands, it is demonstrated here that AlInN/GaN HEMTs show excellent power performances and constitute a particularly interesting alternative to AlGaN/GaN HEMTs, especially for high-frequency applications beyond the X band. The measured transistors with 250 nm gate lengths from different wafers delivered in continuous wave (cw): 10.8 W/mm with 60% associated power added efficiency (PAE) at 3,5 GHz, 6.6 W/mm with 39% associated PAE at 10.24 GHz, and 4.2 W/mm with 43% associated PAE at 18 GHz.


Author(s):  
  Жулиан Берже ◽  
  Денис Дутых

The fidelity of a model relies both on its accuracy to predict the physical phenomena and its capability to estimate unknown parameters using observations. This article focuses on this second aspect by analyzing the reliability of two mathematical models proposed in the literature for the simulation of heat losses through building walls. The first one, named DF, is the classical heat diffusion equation combined with the DuFort-Frankel numerical scheme. The second is the so-called RC lumped approach, based on a simple ordinary differential equation to compute the temperature within the wall. The reliability is evaluated following a two stages method. First, samples of observations are generated using a pseudo-spectral numerical model for the heat diffusion equation with known input parameters. The results are then modified by adding a noise to simulate experimental measurements. Then, for each sample of observation, the parameter estimation problem is solved using one of the two mathematical models. The reliability is assessed based on the accuracy of the approach to recover the unknown parameter. Three case studies are considered for the estimation of ( i ) the heat capacity, ( ii ) the thermal conductivity or ( iii ) the heat transfer coefficient at the interface between the wall and the ambient air. For all cases, the DF mathematical model has a very satisfactory reliability to estimate the unknown parameters without any bias. However, the RC model lacks of fidelity and reliability. The error on the estimated parameter can reach 40% for the heat capacity, 80% for the thermal conductivity and 450% for the heat transfer coefficient.


2011 ◽  
Vol 28 (2) ◽  
pp. 34-37
Author(s):  
W.J. Luo ◽  
X.J. Chen ◽  
C.Y. Yang ◽  
Y.K. Zheng ◽  
K. Wei ◽  
...  

PurposeThe purpose of this paper is to report on the stabilization network optimization of internally matched GaN high electron mobility transistors (HEMTs).Design/methodology/approachThe effects of the two stabilization networks on the characteristics of the device are discussed, such as the stability, power gain and output power.FindingsWith the optimized stabilization network, the internally matched GaN HEMTs with 16‐mm gate width exhibited good stability and delivers a 46 dBm output power with 6.1 dB power gain under the continuous wave condition at 8 GHz. By using the optimized stabilization network, the package process of the large‐scale microwave power device of GaN HEMTs can be simplified.Originality/valueThis paper provides useful information for the internally matched GaN HEMTs.


2018 ◽  
Vol 15 (29) ◽  
pp. 257-265
Author(s):  
R. HECHAVARRÍA ◽  
O. DELGADO ◽  
A. HIDALGO ◽  
S. ESPÍN ◽  
J. GUAMANQUISPE

Nanofluids have become nowadays of special importance because of their different uses in industry, therefore, to propose methods to calculate their thermal properties would be useful. In this work, a new variant for the calculation of thermal conductivity and diffusivity of nanofluids is proposed; the possibilities and limitations of this non-stationary method, which uses light radiation as the heat source, are studied. Here, the light is homogenously incident on one of the end surfaces of a cylinder that has a thermally insulated side surface, setting the temperature at the other end to a constant value, then the temperature distribution is obtained as a function of the coordinate and time; adjusting the theoretical model, parabolic heat diffusion equation, to the experimental data obtained. The conditions of validity of the method to measure thermal diffusivity and thermal conductivity of fluids are analyzed; as well as, the way in which it could be used to verify the validity of the Hamilton and Crosser (HC) model in the case of nanofluids. Currently, nanofluids are used to exchange heat, as they have been found to exceed the potential of conventional refrigerants; however, the calculation of thermal properties still does not offer definitive values.


2007 ◽  
Vol 17 (03) ◽  
pp. 509-520
Author(s):  
EIICHI SANO ◽  
TAIICHI OTSUJI

The terahertz region is one of the unexplored bands. This paper first reviews the present status of conventional high-speed devices, especially InP-based high electron mobility transistors (HEMTs), and addresses the technological problems facing the goal of terahertz operation. As an alternative approach to solve these problems, we developed a plasmon-resonant photomixer for realizing a coherent terahertz continuous-wave source. Preliminary results on electromagnetic response to impulsive photoexcitation at room temperature are reported briefly.


Author(s):  
Dipta Sarkar ◽  
Samuel W. Oxandale ◽  
Tyler J. Hieber ◽  
M. G. Baboly ◽  
Zayd C. Leseman

Abstract Thermoreflectance is a common technique to measure thermal properties of micro/nano devices. Most thermoreflectance techniques use a pump-probe scheme with lasers to heat the sample and analyze the temperature. The limiting characteristics of most of these techniques are that they can probe the temperature at only one spot on the sample, assume a value for either heat capacity or thermal conductivity to find the other, and require a semi-infinite substrate. In this paper, a new technique is described, the Suspended ThermoReflectance (STR) technique, which allows measurement of thermal conductivity by probing multiple points along the length of a suspended micro/nano-scale sample. This technique involves a pump laser for heating the tip of a suspended μ-cantilever Si beam and a probe laser to scan the temperature along the μ-cantilever’s length. Thermal conductivity is obtained by applying the heat diffusion equation for the temperature gradient along the beam length. 2.9 μm thick Si μ-cantilever samples are tested over a range of temperatures from 20–300K. It is found that thermal conductivity of the silicon varies from 28 W/mK to 80 W/mK.


Sign in / Sign up

Export Citation Format

Share Document