Tight-Binding Quantum Chemical Molecular Dynamics Study on Depth Profile Prediction in Low Energy Boron Implantation Process

Abstract

Creation of shallow junction for the future generation LSI is a crucial step in semiconductor industry and low-energy boron implantation process is considered to be a key technology. In this study, we have statistically investigated the effects of orientation of implantation on the dynamic behavior of boron implantation process into hydrogen-terminated Si(001) 2×1 surface by using our original tight-binding quantum chemical molecular dynamics method, which is over 5,000 times faster than conventional first-principle molecular dynamics method. It was found that depth profile of boron implantation can be controlled by orientation of boron implantation and the shallowest implantation depth was obtained in the case of tilt angle equal to 7° among the investigated tilt angles of 0°, 7°, 15°, 22.5°, 30° and 45° at the initial boron energy of 100 eV. At the boron implantation process of over 1 keV energy the tilt angle of 7° has been employed experimentally and the same tilt angle was predicted to be the best even at low-energy region of 100 eV. Furthermore, we investigated the effect of rotation angle on the depth profile and at all the investigated tilt angles the average implantation depth becomes shallower for rotation angle of 45° that is along <011> direction, than for rotation angle of 0° that is along <001>. Hence, the shallowest depth profile was obtained in the case of tilt angle of 7° and rotation angle of 45°, where the distribution of intruded boron atom was more concentrated than for the same tilt angle but rotation angle of 0°. The effect of tilt and rotation angles on the boron implantation process has not been clarified experimentally at low-energy boron implantation process of less than 1 keV and hence we concluded that theoretical optimization of low-energy boron implantation process has been succeeded by means of our original tight-binding quantum chemical molecular dynamics method.