IEEE Transactions on Electron Devices · 2006 · 38 citations · 7 references
EngineeringElectrical EngineeringMaterials ScienceComputational PhysicsHeat TransferTransport PhenomenaDevice SimulationElectron TransportElectronic Transport ModelingMonte Carlo MethodsMonte Carlo SimulationMonte Carlo Transport MethodsCharge TransportElectron Transport SimulationElectronic Transport SimulationElectrothermal ModelingHeat Diffusion EquationMonte Carlo Device SimulationMonte Carlo Electron TransportMonte Carlo Transport Simulation
The electrothermal simulator developed in this work uses an iterative procedure that self-consistently couples a Monte Carlo electronic trajectory simulation with a fast Fourier series solution of the heat diffusion equation. Results presented in this paper are obtained from the simulation of In <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.15</sub> Ga <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.85</sub> As/Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.28</sub> Ga <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.72</sub> As HEMTs. The negative differential output conductance (thermal droop) is observed in the electrothermal I <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ds</sub> -V <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ds</sub> characteristics of the simulated devices. Temperature profiles across the simulated region corresponding to different heat generation distributions are shown to be nonuniform with peak temperature and temperature range values dependent upon the device bias. The microscopic details of charge transport are studied, and the relationship between the thermal droop and the microscopic velocity properties is analyzed. The reduction in the length of the semiconductor die is shown to affect the peak temperature values without significantly altering the temperature range. The distribution of heat generation across the devices is simulated using a microscopic level count of phonon emission and absorption events and compared with that obtained using the current density-electric field (JmiddotE) dot product. The J middot E calculation was found to overestimate the local heat generation in the most electrically active regions of the device
7
W. Batty, C.E. Christoffersen, A.J. Panks et al. · IEEE Transactions on Components and Packaging Technologies · 2001 · 100 citations
Engineering, Computational Engineering, Electrical Engineering +17
Heat generation in semiconductor devices
U. Lindefelt · Journal of Applied Physics · 1994 · 91 citations
Fabrizio Bonani, Giovanni Ghione · Solid-State Electronics · 1995 · 90 citations