Effects of Platinum on NiPtSiGe/n-SiGe and NiPtSi/n-Si Schottky Contacts

Solid phase reaction of NiPt/Si and NiPt/SiGe is one of the key issues for silicide (germanosilicide) technology. Especially, the NiPtSiGe, in which four elements are involved, is a very complex system. As a result, a detailed study is necessary for the interfacial reaction between NiPt alloy film and SiGe substrate. Besides using traditional material characterization techniques, characterization of Schottky diode is a good measure to detect the interface imperfections or defects, which are not easy to be found on large area blanket samples. The I-V characteristics of 10nm Ni(Pt=0, 5, 10 at.%) germanosilicides/n-Si₀/₇Ge₀.₃ and silicides/n-Si contact annealed at 400 and 500°C were studied. For Schottky contact on n-Si, with the addition of Pt in the Ni(Pt) alloy, the Schottky barrier height (SBH) increases greatly. With the inclusion of a 10% Pt, SBH increases ~0.13 eV. However, for the Schottky contacts on SiGe, with the addition of 10% Pt, the increase of SBH is only ~0.04eV. This is explained by pinning of the Fermi level. The forward I-V characteristics of 10nm Ni(Pt=0, 5, 10 at.%)SiGe/SiGe contacts annealed at 400°C were investigated in the temperature range from 93 to 300K. At higher temperature (>253K) and larger bias at low temperature (<253K), the I-V curves can be well explained by a thermionic emission model. At lower temperature, excess currents at lower forward bias region occur, which can be explained by recombination/generation or patches due to inhomogenity of SBH with pinch-off model or a combination of the above mechanisms.

Validating Co-Training Models for Web Image Classification

Co-training is a semi-supervised learning method that is designed to take advantage of the redundancy that is present when the object to be identified has multiple descriptions. Co-training is known to work well when the multiple descriptions are conditional independent given the class of the object. The presence of multiple descriptions of objects in the form of text, images, audio and video in multimedia applications appears to provide redundancy in the form that may be suitable for co-training. In this paper, we investigate the suitability of utilizing text and image data from the Web for co-training. We perform measurements to find indications of conditional independence in the texts and images obtained from the Web. Our measurements suggest that conditional independence is likely to be present in the data. Our experiments, within a relevance feedback framework to test whether a method that exploits the conditional independence outperforms methods that do not, also indicate that better performance can indeed be obtained by designing algorithms that exploit this form of the redundancy when it is present.