Community Research and Development Information Service - CORDIS

FP7

MASTIC Result In Brief

Project ID: 293783
Funded under: FP7-PEOPLE
Country: Spain

Monte Carlo simulations of defects in alloys

Using sophisticated numerical models, EU-funded scientists have been able to track the generation and propagation of defects in the microstructure of transistor semiconductor materials and alloys for nuclear reactors.
Monte Carlo simulations of defects in alloys
Following the famous Moore's law, the semiconductor industry has continued without hindrance the downscaling of transistor dimensions to add more components in integrated circuits and more functionalities to electronic devices. To further reduce the size of transistors, it is necessary to incorporate ionised dopants into semiconductor wafers.

In MASTIC (Multi atomistic Monte Carlo simulation of technologically important crystals), scientists used Monte Carlo techniques to simulate the potential generation of defects as a result of semiconductor doping. Accurate predictions of dopant distribution drastically reduce the time needed to design processes to obtain the desired electronic properties.

Scientists used atomistic object and lattice kinetic Monte Carlo (OKMC and LKMC, respectively) techniques to achieve the timescales needed to study defect evolution in silicon and silicon-germanium alloys. Several atomic configurations were included in the simulator to study amorphous transitions of the silicon-based materials and the formation of defective crystals during these transitions.

A similar approach was adopted to simulate damage of structural materials in nuclear fusion reactors when exposed to high radiation levels. Until now, there have been no experimental facilities to reproduce the same conditions that the materials withstand. Furthermore, whereas modelling dopant effects on bulk crystals requires just a few minutes, modelling fusion materials may take many years.

However, OKMC and LKMC techniques enabled the study of defect evolution in silicon carbide, iron and ferrochrome at different timescales. Specifically, Monte Carlo simulations accurately reproduced experimental measures of point defect diffusion in complex binary systems as well as the alloy's behaviour in the miscibility gap.

MASTIC models, software code and input parameters have been made available for download here. Microelectronic companies have already shown interest in the new tools to customise semiconductor materials and meet the required performance levels for electronic components.

Related information

Keywords

Monte Carlo simulations, alloys, semiconductor, nuclear reactors, dopants, MASTIC
Record Number: 175031 / Last updated on: 2016-07-25
Domain: Energy