Growth and characterization of 4H-SiC by horizontal hot-wall CVD

4H-SiC layers have been homoepitaxially grown at 1500 degrees C with the use of a horizontal hot-wall chemical vapor deposition (CVD) system, which was built in the author's group. The typical growth rate was 2 mu m/h at a pressure of 40 Torr. The background donor concentration has been reduced to 2.3 x 10(15) cm(-3) during a prolonged growth run. It confirmed the idea that the high background concentration of thin films was caused by the impurities inside the susceptor and thermal insulator The FWHM of x-ray co-rocking curves show 9 similar to 15 aresecs in five different areas of a 32-mu m-thick 4H-SiC epilayer The free exciton peaks dominated in the near-band-edge low-temperature photoluminescence spectrum (LTPL), indicating high crystal quality.

Controlled growth of III-V compound semiconductor nano-structures and their application in quantum-devices

Various techniques on the growth of self-assembled compound semiconductor nano-structures (quantum dots, QDs) have been tried to enhance the controlling on size, density, emitting wavelength, uniformity in size and ordering in location of the QDs. Optimized growth conditions have been used in the application of the QD materials in opto-electronic devices. High-power long-lifetime quantum-dot laser-diodes (QD-LDs) emitting near 1 mu m, QD-LDs emitting in red-light range, 1.3 mu m QD-LDs on GaAs substrate and quantum-dot super-luminescent diodes (QD-SLDs) have successfully been achieved.

Molecular-beam epitaxial growth of position controlled InAs islands on cleaved edge of InGaAs/GaAs superlattice

A promising approach for positioning of InAs islands on (110)GaAs is demonstrated. By combining self-assembly of quantum dots with solid source molecular beam epitaxy (MBE) on cleaved edge of InGaAs/GaAs superlattice (SL), linear alignment of InAs islands on the InGaAs strain layers have been fabricated The cleaved edge of InGaAs/GaAs SL acts as strain nanopattern for InAs selective growth. Indium atoms incident on the surface will preferentially migrate to InGaAs regions where favorable bonding sites are available. The strain nanopattern's effect is studied by the different indium fraction and thickness of InxGa1-xAs/GaAs SL. The ordering of the InAs islands is found to depend on the properties of the underlying InGaAs strain layers.

Effect of noncoherent islands on the optical properties of the 1.3 mu m InAs/GaAs quantum dots during rapid thermal annealing

The effect of thermal annealing of InAs/GaAs quantum dots (QDs) with emission wavelength at 1.3 mu m have been investigated by photoluminescence (PL) and transmission electron microscopy (TEM measurements. There is a dramatic change in the A spectra when the annealing temperature is raised up to 800 degrees C: an accelerated blushifit of the main emission peak of QDs together with an inhomogeneous broadening of the linewidth. The TEM images shows that the lateral size of normal QDs decreases as the annealing temperature is increased, while the noncoherent islands increase their size and densit. A small fraction of the relative large QDs contain dislocations when the annealing temperature increases up to 800 degrees C. The latter leads to the strong decrease of the PL intensity.

Micro-fabricated Al0.3Ga0.7As pyramids for potential SPM applications

A novel technique of manufacturing Al0.3Ga0.7As pyramids by liquid phase epitaxy (LPE) for scanning probe microscopy (SPM) sensors is reported Four meticulously designed conditions-partial oxidation, deficient solute, air quenching and germanium doping result in defect-free homogeneous nucleation and subsequent pyramid formation. Micrometer-sized frustums and pyramids are detected by scanning electron microscopy (SEM). The sharp end of the microtip has a radius of curvature smaller than 50 nm. It is believed that such accomplishments would contribute not only to crystal growth theory, but also to miniature fabrication technology.

Raman scattering study on diluted magnetic semiconductor Ga1-xMnxAs prepared by Mn-ion implantation

Raman scattering measurements have been performed in diluted magnetic semiconductor Ga1-xMnxAs prepared by Mn-ion implantation, deposition, and post-annealing technique. It is found that the Raman spectrum measured from the implanted surface of the sample shows some new weak modes in addition to the GaAs-like modes which are observed from the unimplanted surface. The new vibrational modes observed are assigned to MnAs-like modes. The coupled LO-phonon plasmon mode, and Mn and As related vibrational modes caused by Mn-ion implantation, deposition, and post-annealing are also observed. Furthermore, the GaAs-like modes are found to be shifted by approximately 4 cm(-1) in the lower frequency side, compared with those observed from the unimplanted surface.

Edge dislocation of b=[001]/2 in the InAs nanostructure on InP(001)

The self-organized InAs/In0.52Al0.48As nanostructure were grown on InP (001) using molecular beam epitaxy (MBE). The nanostructure has been studied using transmission electron microscopy (TEM) and high resolution transmission electron microscopy (HRTEM). The edge dislocations with the Burgers vector b = ([001]/2) and extending along the [$(110) over bar $] direction are observed. The results show that in the region near an edge dislocation, no InAs wires were formed, while in the regions free of dislocation, wire-like nanostructures were formed. The mechanisms for the formation of the [001]/2 edge dislocations were discussed.

Continuous speech recognition based on ICA and geometrical learning

We investigate the use of independent component analysis (ICA) for speech feature extraction in digits speech recognition systems. We observe that this may be true for recognition tasks based on Geometrical Learning with little training data. In contrast to image processing, phase information is not essential for digits speech recognition. We therefore propose a new scheme that shows how the phase sensitivity can be removed by using an analytical description of the ICA-adapted basis functions. Furthermore, since the basis functions are not shift invariant, we extend the method to include a frequency-based ICA stage that removes redundant time shift information. The digits speech recognition results show promising accuracy. Experiments show that the method based on ICA and Geometrical Learning outperforms HMM in a different number of training samples.

Research on multi-degree-of-freedom neurons with weighted graphs

In this paper, we redefine the sample points set in the feature space from the point of view of weighted graph and propose a new covering model - Multi-Degree-of-Freedorn Neurons (MDFN). Base on this model, we describe a geometric learning algorithm with 3-degree-of-freedom neurons. It identifies the sample points secs topological character in the feature space, which is different from the traditional "separation" method. Experiment results demonstrates the general superiority of this algorithm over the traditional PCA+NN algorithm in terms of efficiency and accuracy.

A novel image restoration algorithm based on high-dimensional space geometry

A novel image restoration approach based on high-dimensional space geometry is proposed, which is quite different from the existing traditional image restoration techniques. It is based on the homeomorphisms and "Principle of Homology Continuity" (PHC), an image is mapped to a point in high-dimensional space. Begin with the original blurred image, we get two further blurred images, then the restored image can be obtained through the regressive curve derived from the three points which are mapped form the images. Experiments have proved the availability of this "blurred-blurred-restored" algorithm, and the comparison with the classical Wiener Filter approach is presented in final.

Continuous speech research based on HyperSausage Neuron

In this paper, we presents HyperSausage Neuron based on the High-Dimension Space(HDS), and proposes a new algorithm for speaker independent continuous digit speech recognition. At last, compared to HMM-based method, the recognition rate of HyperSausage Neuron method is higher than that of in HMM-based method.

Measurement of GaN/Ge(001) Heterojunction Valence Band Offset by X-Ray Photoelectron Spectroscopy

X-ray photoelectron spectroscopy has been used to measure the valence band offset (VBO) at the GaN/Ge heterostructure interface. The VBO is directly determined to be 1.13 +/- 0.19 eV, according to the relationship between the conduction band offset Delta E-C and the valence band offset Delta E-V : Delta E-C = E-g(GaN) - E-g(Ge) - Delta E-V, and taking the room-temperature band-gaps as 3.4 and 0.67 eV for GaN and Ge, respectively. The conduction band offset is deduced to be 1.6 +/- 0.19 eV, which indicates a type-I band alignment for GaN/Ge. Accurate determination of the valence and conduction band offsets is important for the use of GaN/Ge based devices.

Multi-wafer 3C-SiC heteroepitaxial growth on Si(100) substrates

Epitaxial growth of semiconductor films in multiple-wafer mode is under vigorous development in order to improve yield output to meet the industry increasing demands. Here we report on results of the heteroepitaxial growth of multi-wafer 3C-SiC films on Si(100) substrates by employing a home-made horizontal hot wall low pressure chemical vapour deposition (HWLPCVD) system which was designed to be have a high-throughput, multi-wafer (3x2-inch) capacity. 3C-SiC film properties of the intra-wafer and the wafer-to-wafer including crystalline morphologies, structures and electronics are characterized systematically. The undoped and the moderate NH3 doped n-type 3C-SiC films with specular surface are grown in the HWLPCVD, thereafter uniformities of intra-wafer thickness and sheet resistance of the 3C-SiC films are obtained to be 6%similar to 7% and 6.7%similar to 8%, respectively, and within a run, the deviations of wafer-to-wafer thickness and sheet resistance are less than 1% and 0.8%, respectively.