1993 Fiscal Year Final Research Report Summary
Layr-by-layr oxidation mechanism of hydrogen-terminated silicon surfaces and their interface structures
| Project/Area Number |
04452177
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
電子材料工学
|
|---|
| Research Institution | HIROSHIMA UNIVERSITY |
|---|
Principal Investigator |
HIROSE Masataka Hiroshima University, Electrical Engineering, Professor, 工学部, 教授 (10034406)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MIYAZAKI Seiichi Hiroshima University, Electrical Engineering, Associated Professor, 工学部, 助教授 (70190759)
|
|---|
| Project Period (FY) |
1992 – 1993
|
| Keywords | Silicon surface / Surface microroughness / Hydrogen-terminated Si / Atomic step structure / Layr-by-layr oxidation / SiO_2 / Si interface / FT-IR-ATR / AFM |
|---|
| Research Abstract |
It was found that the wet chemical treatment in a pH-modified BHF(pH=3.8)solution or a 0.1%HF+1%H_2O_2 solution significantly reduces the surface microroughness of Si(100)wafers as demonstrated by a sharp SiH_2 stretching vibration peak accompanied with the weak SiH and SiH_3 peaks in FT-IR-ATR spectra. The suppression of anisotropic etching of Si with OH ions in a BHF solution or the selective removal of chemically reactive sites on the surfaces by the oxidation with H_2O_2 followed by oxide etching with HF is thought to be effective to produce the flat Si(100) surface. Taking into account the fact that the native oxidation rate of the flat Si(100)surface is significantly suppressed in the early stages of the oxidation compared with the case of rough Si(100)surfaces, it is likely that the native oxidation starts to proceed from the chemically reactive site such as microfacets and step edges on the surface.
The Si(111)surface becomes atomically flat by 40%NH_4F treatment. The bilayr atomic step structure with a terrace width of a few hundreds nm is clearly observed by AFM.The AFM image of the Si(111)surface covered with a 4nm-thick thermal oxide is almost identical to the surface morphology before oxidation and the step structure is still observable on the oxide. Also, the Si surface after the removal of the oxide by dilute HF is similar to the intial surface. This suggests that the oxidation of the Si(111)surfaces proceeds through a layr-by-layr mechanism and this might be the case also for Si(100)surfaces. Furthermore, it was found that the infrared absorption peak due to the LO phonon mode originating from the Si-O-Si stretching vibration shows a considerable red-shift in the thickness range below 25A.This red-shift is well explained by the existence of the compressive stress near the interface.
|
