D. Iordache

New Meyer-Neldel relations for the depletion and diffusion dark currents in some CCDs

The temperature dependence of the dark currents in some charged-coupled devices (CCDs) was studied. The obtained results point out: a) the compatibility of the theoretical description of these dark currents by means of depletion and diffusion processes, respectively relative to the experimental data, b) the possibility to evaluate - for each pixel - the difference of energies corresponding to the generation-recombination centers and to the intrinsic Fermi level, respectively, c) some new Meyer-Neldel relations between the exponential pre-factors corresponding to the depletion and diffusion currents, and the energy gap E/sub g/ of studied semiconductor materials.

Time scaling in the parallel processing simulation of diffusion processes

A local interaction simulation approach, based on parallel processing, allows us to treat complex diffusion problems in very large lattices or complex media. In order to study the time evolution, e.g., of microscopic systems, we present here a time scaling method, which can reduce the number of time steps (and therefore the computer time) of several orders of magnitude, up to a manageable number (e.g., 104 or 105). Several examples under different assumptions demonstrate the applicability and reliability of the proposed method.

Study of the temperature dependence of the dark currents non-uniformity for some video-camera chips

In order to explain the experimentally found non-uniformity of the dark currents corresponding to different pixels of a video-camera chip, there were accomplished: a) a numerical study of the temperature dependence of the dark currents, in order to find the most convenient theoretical model which can describe sufficiently accurate the experimental data, b) the evaluation of the basic parameters of the studied pixels, corresponding to the chosen theoretical model, c) the study of the correlation between these parameters and their implications on the temperature dependence of the dark currents non-uniformity for different pixels of the studied chip.

A renormalization group approach to the simulation of diffusion processes by parallel computers

A space scaling procedure is proposed for the treatment of diffusion problems in large lattices. In conjunction with a previously proposed time scaling procedure, space scaling yields a renormalization group technique, which can reduce by an almost arbitrarily large factor both the computer time and the number of lattice sites for a diffusion problem with or without drift. Several numerical examples demonstrate the efficiency and reliability of the method.

Study of Some Numerical Phenomena Intervening in the Computer Simulations of Some Physical Processes

The mechanisms of the computing schemes leading to the main numerical phenomena: instabilities, divergence or pseudo‐convergence, distortions, as well as of the stability and convergence radii, and of the optimization possibilities of the computer simulations of some physical processes were studied by this work.