Two-junction ballistic switch in quantum network model

Searching of optimal parameters of nanoelectronic devices is a primal problem in their modeling. We solve this problem on example of the electron ballistic switch in quantum network model. For this purpose, we use a computing scheme in which closed channels are taking into account. It allows calculating correctly a scattering matrix of the switch and, consequently, the electric currents flowing through it. Without losing generality, we consider model of two-junction switch at room temperature. Its character is localization of the controlling electric field in the domain before branching. We optimize switch parameters using a genetic algorithm. At the expense of it for InP, GaAs and GaSb switch efficiency reached 77-78%. It is established that, for the considered materials, volt-ampere characteristics of the device are close to the linear ones at bias voltages 0-50 mV. It allowed describing with a good accuracy electron transport in the switch by means of $3\times 3$ matrix of approximate conductivity. Finally, based on the performed parameters optimization of two-junction switch we formulate the general scheme of modeling nanoelectronic devices in the framework of quantum network formalism.