Optimized quantum drift diffusion model for a resonant tunneling diode

The main aim of this work is to optimize a Quantum Drift Diffusion model (QDD) (V. Romano, M. Torrisi, and R. Tracina, "Approximate solutions to the quantum drift-diffusion model of semiconductors," J. Math. Phys., vol. 48, p. 023501, 2007; A. El Ayyadi and A. Jungel, "Semiconductor simulations using a coupled quantum drift-diffusion schrodinger-Poisson model," SIAM J. Appl. Math., vol. 66, no. 2, pp. 554-572, 2005; L. Barletti and C. Cintolesi, "Derivation of isothermal quantum fluid equations with Fermi-Dirac and bose-einstein statistics," J. Stat. Phys., vol. 148, pp. 353-386, 2012) by comparing it with the Boltzmann-Wigner Transport Equation (BWTE) (O. Muscato, "Wigner ensemble Monte Carlo simulation without splitting error of a GaAs resonant tunneling diode," J. Comput. Electron., vol. 20, pp. 2062-2069, 2021) solved using a signed Monte Carlo method (M. Nedjalkov, H. Kosina, S. Selberherr, C. Ringhofer, and D. K. Ferry, "Unified particle approach to Wigner-Boltzmann transport in small semiconductor devices," Phys. Rev. B, vol. 70, pp. 115-319, 2004). A situation of high non equilibrium regime is investigated: electron transport in a Resonant Tunneling Diode (RTD) made of GaAs with two potential barriers in GaAlAs. The range of the suitable voltage bias applied to the RTD is analyzed. We find an acceptable agreement between QDD model and BWTE when the applied bias is low or moderate with a threshold of about 0.225 V over a length of 150 nm; it is found out that the use of a field dependent mobility is crucial for getting a good description of the negative differential conductivity in such a range. At higher bias voltages, we expect that QDD model loses accuracy.