Mathematical properties of one-dimensional semiconductors

From MaRDI portal





The fundamental equations of semiconductor devices depend after a suitable scaling on a very small parameter \(\lambda\). \(\lambda\) is about \(10^{-3} - 10^{-6}\) when the doping function \(D(x)\) is scaled to order 1. Consequently the solutions which are analytic functions in any open subset where \(D(x)\) is analytic, have very steep gradients. This leads to major difficulties from a computational point of view. In the particular case of one-dimensional reverse-biased semiconductor devices and in the stationary case, the authors study the limit of the solutions when \(\lambda\) goes to zero in order to see what the limit problem is and what norms of the solutions remain bounded. They prove that the electric potential approaches the solution of a simple double obstacle problem while each of electron and hole densities tend to a step function. They also prove a uniqueness result for \(\lambda\) small enough.











This page was built for publication: Mathematical properties of one-dimensional semiconductors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3767709)