Chaotic algorithms: A numerical exploration of the dynamics of a stiff photoconductor model
- Universidad de los Andes, Merida (Venezuela)
The photoconducting property of semiconductors leads, in general, to a very complex kinetics for the charge carriers due to the non-equilibrium processes involved. In a semi-conductor with one type of trap, the dynamics of the photo-conducting process are described by a set of ordinary coupled non-linear differential equations given by where n and p are the free electron and hole densities, and m the trapped electron density at time t. So far, there is no known closed solution for these set of non-linear differential equations, and therefore, numerical integration techniques have to be employed, as, for example, the standard procedure of the Runge-Kutta (RK) method. Now then, each one of the mechanisms of generation, recombination, and trapping has its own lifetime, which means that different time constants are to be expected in the time dependent behavior of the photocurrent. Thus, depending on the parameters of the model, the system may become stiff if the time scales between n, m, and p separate considerably. This situation may impose a considerable stress upon a fixed step numerical algorithm as the RK, which may produce then unreliable results, and other methods have to be considered. Therefore, the purpose of this note is to examine, for a critical range of parameters, the results of the numerical integration of the stiff system obtained by standard numerical schemes, such as the single-step fourth-order Runge-Kutta method and the multistep Gear method, the latter being appropriate for a rigid system of equations. 7 refs., 2 figs.
- OSTI ID:
- 535420
- Journal Information:
- Journal of Computational Physics, Vol. 132, Issue 2; Other Information: PBD: Apr 1997
- Country of Publication:
- United States
- Language:
- English
Similar Records
Strong Stability Preserving Integrating Factor Two-Step Runge-Kutta Methods
Entropy–Preserving and Entropy–Stable Relaxation IMEX and Multirate Time–Stepping Methods