Yamichi Ohmura

Electrical properties of n-type Si layers doped with proton bombardment induced shallow donors

Abstract Sheet Hall coefficients and resistivities of n-type Si layers doped with shallow donors produced by proton bombardment have been measured between 35 and 300°K. The donor ionization energy is (26 ± 1) meV. The donor concentration profile has been determined.

Solid-Phase Lateral Epitaxial Growth onto Adjacent SiO2 Film from Amorphous Silicon Deposited on Single-Crystal Silicon Substrate

Single-crystal silicon films (200 nm thick) have been grown laterally by thermal annealing from amorphous silicon evaporated on a single-crystal (100) silicon substrate and implanted with ~1016/cm2 Si, onto an adjacent SiO2 film by solid-phase epitaxy. The key requirements for this kind of lateral epitaxy appear to be a high-dose Si ion implanatation and low temperature (575°C) annealing for a long period in order to suppress the nucleation of randomly-oriented crystals on the SiO2 film.

Determination of the spatial distribution of deep centers from capacitance measurements of pn junctions

Effects of a spatial distribution of deep centers on the junction capacitance are reported for the first time. It is shown that the spatial distribution of deep centers can be calculated from capacitance data measured by both the capacitance‐voltage method and the Copeland method, if their energy level is known or assumed. The theory is examined with the experiment on silicon p+n junctions irradiated with 300‐kV protons. The analysis yields a value 0.40 eV below conduction band for the energy level of radiation‐induced defects, as well as a distribution of the defects with depth.

n-Channel MOS ring oscillators fabricated in electron-beam recrystallized silicon-on-insulator

Both 25-stage n-MOS enhancement driver/enhancement load (E/E) and enhancement driver/depletion load (E/D) ring oscillators with a fan out of one composed of 4-µm channel length transistors have been successfully fabricated in CW electron-beam recrystallized polysilicon/Si<inf>3</inf>N<inf>4</inf>/SiO<inf>2</inf>/

Raman spectra of Si-implanted silicon on sapphire

Abstract Raman scattering from Si-implantation-amorphized and subsequently thermally recrystallized silicon on sapphire (SOS) shows that optical phonon frequencies of silicon well correlate with the recrystallization temperature rather than the epitaxial temperature. This implies that the strain in as-epitaxial SOS has been relieved and replaced by the one which depends upon the recrystallization temperature.

Spatially varied activation of ion-implanted As during the regrowth of amorphous layers in Si

He backscattering technique and sheet‐resistivity measurements show that the electrical activation of As implanted in (100) and (111) Si and annealed at relatively low temperatures (500–600 °C) takes place from the vicinity of the crystalline substrate Si in the As profile. This is accompanied by the regrowth of the implantation‐induced amorphous layer. The annealing temperature required for both the activation and the regrowth is dose dependent. No defect‐induced diffusion of As has been observed in these annealing processes.

A Raman study of Si‐implanted silicon on sapphire

It has been demonstrated by Raman scattering spectroscopy that an Si implantation technique to improve the crystalline quality of silicon on sapphire (SOS) also changes the strain state of SOS. Optical phonon (OP) frequencies from the surface crystalline region in as‐implanted SOS shift larger than those for heavily implanted bulk silicon, which may be due to a superposition of implantation‐induced compressive stress and thermal expansion coefficient difference‐induced compressive strain. These highly shifted OP frequencies (∼10 cm−1 from the bulk) decrease relatively abruptly as the recrystallization proceeds, and the surface implantation‐induced damage is reduced. After the recrystallization is completed, the OP frequencies are close to those which are determined by the expansion coefficient difference‐induced strain between annealing and room temperatures, as though the silicon film were grown at the annealing temperature. The newly formed low strain at 600 °C, however, again becomes high by subsequent...

Effect of atmospheres on arsenic diffusion into silicon from the doped oxide layer

The backscattering method is employed to investigate the effect of atmospheres on arsenic diffusion into silicon from the doped oxide layer. An effect of the coating of undoped silicon oxide and silicon nitride upon the surface of the doped oxide layer is also investigated. Arsenic atoms diffused more efficiently into silicon in oxygen atmosphere than in nitrogen atmosphere. An additional oxide layer was formed by oxidation of the silicon surface during the thermal diffusion in oxygen atmosphere and this layer exists between the doped oxide layer and the silicon substrate. It is considered that such an additional oxide layer may play an important role in the thermal diffusion of arsenic atoms into the silicon substrate.

Evidence for electronic stopping in ion implantation: Shallower profile of lighter isotope 10B in Si

It has been observed by C−V measurements that randomly implanted 10B has shallower profiles in Si than 11B for the acceleration energy range from 50 to 200 keV. This effect can be explained only in terms of electronic stopping in LSS’s theory.

Third-Order Piezoresistance Coefficients in p-Type Si

By tensor property and crystallographic symmetry considerations, it has been shown that there are 20 independent nonzero components of the third-order piezoresistance (TOPR) coefficients (π3) for cubic semiconductors which crystallize in diamond ( Oh) or zincblende ( Td) structures. Piezoresistances for p-type Si obtained from conductivities calculated for sixfold degenerate valence bands with and without stress are highly nonlinear so that they are more suitably expressed with up to the fifth-order (π5) stress terms than merely with up to the π3 term for the odd-order stress terms. For several current and uniaxial stress directions, TOPRs are given in terms of 20 components of TOPR coefficients, and calculated π3s which are obtained using up to the third-order stress term correspond in the order of magnitude with experimental ones.