SIMULATION OF IMPURITY DISTRIBUTION IN THE BASE FOR OPERATIONAL CALCULATION OF THE STRUCTURE OF A DRIFT N-P-N TRANSISTOR

Abstract

Based on modelling the actual distribution of impurities in the base region, the problem of rapid calculation of the structure of drift bipolar n–p–n transistors is considered. It is shown that the parameters and electrical characteristics of drift transistors are closely related to the concentration profile of the doping impurity, which determines the depth of p–n transitions, breakdown voltage, and current gain. Additionally, this concentration profile affects the frequency-pulse characteristics and the current density through the transistor. The paper analyses modern technological processes for forming base regions in silicon structures of drift transis- tors, in particular the influence of boron ion doping doses, temperature and time of thermal diffusion of the impurity in the base region, as well as the role of the segregation coefficient in the interaction of the impurity with the silicon oxide layer. Based on experimental data on the depth of p–n junctions and electrical parameters of drift transistors of various technological series, the results were summarised and a simplified analytical model was determined to describe the actual distribution of boron impurities in the base region. Empirical dependencies were proposed that link the depth of the maximum impurity concentration with the product of the diffusion coefficient and the heat treatment time and with the doping dose. An expression has been obtained for calculating the depth xjе and maximum concentration Nб.max, which make it possible to estimate the breakdown voltage of the emitter p–n junction without conducting complex technological tests. A comparison of experimental and calculated data showed the correctness of the results (the discrepancy does not exceed 0.1%), which confirms the reliability of the proposed model. It has been determined that the optimal electrical characteristics of drift transistors are achieved when the depth of the emitter p–n junction coincides with the depth of the maximum concentration of the base impurity. The proposed model can be used for rapid prediction of drift transistor parameters at the microcircuit design stage and allows reducing the number of experimental stages in the technological cycle.