PHENOMENOLOGICAL ELECTRONIC STOPPING-POWER MODEL FOR MOLECULAR-DYNAMICS AND MONTE-CARLO SIMULATION OF ION-IMPLANTATION INTO SILICON

Citation
D. Cai et al., PHENOMENOLOGICAL ELECTRONIC STOPPING-POWER MODEL FOR MOLECULAR-DYNAMICS AND MONTE-CARLO SIMULATION OF ION-IMPLANTATION INTO SILICON, Physical review. B, Condensed matter, 54(23), 1996, pp. 17147-17157
Citations number
49
Categorie Soggetti
Physics, Condensed Matter
ISSN journal
01631829
Volume
54
Issue
23
Year of publication
1996
Pages
17147 - 17157
Database
ISI
SICI code
0163-1829(1996)54:23<17147:PESMFM>2.0.ZU;2-4
Abstract
It is crucial to have a good phenomenological model of electronic stop ping power for modeling the physics of ion implantation into crystalli ne silicon. In the spirit of the Brandt-Kitagawa effective charge theo ry, we develop a model for electronic stopping power for an ion, which can be factorized into (i) a globally averaged effective charge takin g into account effects of close and distant collisions by target elect rons with the ion, and (ii) a local charge density dependent electroni c stopping power for a proton. This phenomenological model is implemen ted into both molecular dynamics and Monte Carlo simulations. There is only one free parameter in the model, namely, the one electron radius r(s)(O) for unbound electrons. By fine tuning this parameter, it is s hown that the model can work successfully for both boron and arsenic i mplants. We report that the results of the dopant profile simulation f or both species are in excellent agreement with the experimental profi les measured by secondary-ion mass spectrometry (SIMS) over a wide ran ge of energies and with different incident directions. We point out th at the model has wide applicability, for it captures the correct physi cs of electronic stopping in ion implantation. This model also provide s a good physically based damping mechanism for molecular dynamics sim ulations in the electronic stopping power regime, as evidenced by the striking agreement of dopant profiles calculated in our molecular dyna mics simulations with the SIMS data.