Numerical simulations of pulsed power electronic components radiation hardness

An increasing degree of both semiconductor components' miniaturization and electromagnetic contamination of the environment brings into question the reliability of modern electronic devices. This is particularly emphasized in the functioning conditions of electronic devices related to: nuclear power facilities, medical devices based on nuclear particle and x-ray radiation and the impulse power device. At the same time, non-resistance of semiconductor components to the effects of particle and electromagnetic radiation represents the basic limitation of the expected development of nanoelectronic components. Since the most commonly used material for manufacturing semiconductor components is monocrystalline silicon, it is of great significance to be aware of the influence of nuclear particle and electromagnetic radiation on its relevant. Based on the well-known effects of nuclear (particle) interactions and electromagnetic radiation, the biggest influence on the electrical characteristics of monocrystalline silicon should be expected from neutron (n), x-ray (x) and gamma (γ) radiation. The evaluation of such influence is often carried out (for economic and security reasons) by simulating the effect of radiation interaction with the material, applying the Monte Carlo method (numerical experiments). The aim of this paper is to evaluate the application of the Monte Carlo method for this class of problems. For this purpose, the effect of n, x and γ radiation on the electrical characteristics of monocrystalline silicon will be determined, under the well-controlled laboratory conditions. The parameters of the experiment are energy and radiation dose. Electric characteristics of monocrystalline silicon are determined using the four-point method and the thermal method. The corresponding numerical experiments will be conducted simultaneously with these laboratory experiments. On the basis of already obtained statistical samples (laboratory and numerical experiments), the testing of both correlation and regression between them will be carried out. The results will provide an answer to the question concerning the feasibility of the Monte Carlo method for the described class of problems.