A NEW TYPE OF SILICON-ON-INSULATOR WITH A PERFECT SURFACE SILICON LAYER

A novel silicon structure consisting of a silicon-on-defect layer (SODL), with enhanced surface Hall mobility in the surface layer on a buried defect layer (DL), has been discovered [J. Li, Nucl. Instr. and Meth. B59/60 (1991) 1053]. SODL material was formed by using proton implantation and subsequent two-step annealing. The implantation was carried out with a Varian 350D ion implanter. Based on the discovery, a standard measurement method (current-voltage curve method) was adopted to measure the true resistivity value of the DL in order to replace the spreading resistivity measurement by which the true resistivity in seriously defective silicon cannot be obtained. By adopting the current-voltage current method, the true resistivity value of the DL is measured to be 4.2 x 10(9) OMEGA cm. The SODL material was proved to be a silicon-on-insulator substrate.

NEW TYPE OF SILICON MATERIAL WITH VERY HIGH-QUALITY SURFACE-LAYER ON INSULATING DEFECT LAYER

A new-type silicon material, silicon on defect layer (SODL) was proved to have a very high quality surface microstructure which is necessary for commercially feasible high-density very large scale integrated circuits (VLSI). The structure of the SODL material was viewed by transmission electron microscopy. The SODL material was also proved to have a buried defect layer with an insulating resistivity of 5.7 x 10(10) OMEGA-cm.

New aspects of K promoted nitridation of the InP(100) surface

The effect of molecular nitrogen exposure on the InP(100) surface modified by the alkali metal K overlayer is investigated by core-level photoemission spectroscopy using synchrotron radiation. The alkali metal covered surface exhibits reasonable nitrogen uptake at room temperature, and results in the formation of a P3N5 nitride complex. Flash annealing at 400 degrees C greatly enhanced the formation of this kind of nitride complex. Above 500 degrees C, the nitride complex dissolved completely. (C) 1997 American Vacuum Society.

A new structure of In-based ohmic contacts to n-type GaAs

Electrical, structural and reaction characteristics of In-based ohmic contacts to n-GaAs were studied. Attempts were made to form a low-band-gap interfacial phase of InGaAs to reduce the barrier height at the metal/semiconductor junction, thus yielding low-resistance, highly reliable contacts. The contacts were fabricated by e-beam sputtering Ni, NiIn and Ge targets on VPE-grown n(+)-GaAs film (approximate to 1 mu m, 2 x 10(18) cm(-3)) in ultrahigh vacuum as the structure of Ni(200 Angstrom)/NiIn(100 Angstrom)/Ge(40 Angstrom)/n(+)-GaAs/SI-GaAs, followed by rapid thermal annealing at various temperatures (500-900 degrees C). In this structure, a very thin layer of Ge was employed to play the role of heavily doping donors and diffusion limiters between In and the GaAs substrate. Indium was deposited by sputtering NiIn alloy instead of pure In in order to ensure In atoms to be distributed uniformly in the substrate; nickel was chosen to consume the excess indium and form a high-temperature alloy of Ni3In. The lowest specific contact resistivity (rho(c)) of (1.5 +/- 0.5)x 10(-6) cm(2) measured by the Transmission Line Method (TLM) was obtained after annealing at 700 degrees C for 10 s. Auger sputtering depth profile and Transmission Electron Microscopy (TEM) were used to analyze the interfacial microstructure. By correlating the interfacial microstructure to the electronical properties, InxGa1-xAs phases with a large fractional area grown epitaxially on GaAs were found to be essential for reduction of the contact resistance.

Biomimetic (topological) pattern recognition - A new model of pattern recognition theory and its application

A new theoretical model of Pattern Recognition principles was proposed, which is based on "matter cognition" instead of "matter classification" in traditional statistical Pattern Recognition. This new model is closer to the function of human being, rather than traditional statistical Pattern Recognition using "optimal separating" as its main principle. So the new model of Pattern Recognition is called the Biomimetic Pattern Recognition (BPR)(1). Its mathematical basis is placed on topological analysis of the sample set in the high dimensional feature space. Therefore, it is also called the Topological Pattern Recognition (TPR). The fundamental idea of this model is based on the fact of the continuity in the feature space of any one of the certain kinds of samples. We experimented with the Biomimetic Pattern Recognition (BPR) by using artificial neural networks, which act through covering the high dimensional geometrical distribution of the sample set in the feature space. Onmidirectionally cognitive tests were done on various kinds of animal and vehicle models of rather similar shapes. For the total 8800 tests, the correct recognition rate is 99.87%. The rejection rate is 0.13% and on the condition of zero error rates, the correct rate of BPR was much better than that of RBF-SVM.

Strain gradient theory based on a new framework of non-local model

A new framework of non-local model for the strain energy density is proposed in this paper. The global strain energy density of the representative volume element is treated as a non-local variable and can be obtained through a special integral of the local strain energy density. The local strain energy density is assumed to be dependent on both the strain and the rotation-gradient. As a result of the non-local model, a new strain gradient theory is derived directly, in which the first and second strain gradients, as well as the triadic and tetradic stress, are introduced in the context of work conjugate. For power law hardening materials, size effects in thin metallic wire torsion and ultra-thin cantilever beam bend are investigated. It is found that the result predicted by the theoretical model is well consistent with the experimental data for the thin wire torsion. On the other hand, the calculation result for the micro-cantilever beam bend clearly shows the size effect.