Michael J. Platek

Stabilization of Indium Tin Oxide Films to Very High Temperatures

Thin film strain gages based on indium-tin-oxide (ITO) are being developed to measure to static and dynamic strain at temperatures approaching 1500°C. These ceramic strain gages exhibit excellent oxidation resistance and high temperature stability, surviving more than 25 hours of testing in air at 1470°C. Electron spectroscopy for chemical analysis (ESCA) studies indicated that interfacial reactions between ITO and alumina can increase the stability of ITO at elevated temperature. Solid state diffusion of aluminum into the ITO at these temperatures can produce a very stable ITO/Al 2 O 3 solid solution [1, 2]. To determine the nature of the interfacial reaction product, ITO films were deposited onto both Al 2 O 3 and AlN surfaces and thermally cycled to 1500°C. AlN films were used to reduce/eliminate oxygen transport to the interface, so that aluminum-indium interactions alone could be studied. ITO films were deposited onto Al 2 O 3 and AlN films, which were rf sputtered on platinum-coated alumina substrates. The resulting ESCA depth files showed that an interfacial reaction had occurred between the ITO and the Al 2 O 3 and AlN. The presence of two new indium-indium peaks at 448.85 and 456.40eV, corresponding to the indium 3d5 and 3d3 binding energies were observed in both cases; i.e. the AlN and the Al 2 O 3 . These binding energies are significantly higher than those associated with stoichiometric indium oxide. In addition, aluminum doped ITO films were formed by co-sputtering from multiple targets and electrical stability of these films was compared to undoped ITO films over the same temperature range (25–1500°C) [1–4].

Analysis of the performance of optimized distributed erbium-doped fiber amplifiers

With theoretical modeling, we optimize the fiber design for distributed erbium-doped fiber amplifiers which maintain a constant signal level across 10 - 100 km lengths. A constant signal level is maintained by fully inverting an amplifier with very low doping concentrations in the 10 - 50 ppb range. Included in the analysis are the pump wavelength, pumping configuration, erbium doping concentration, noise, temperature effects, macrobending loss and chromatic dispersion. We found that the optimum design would consist of a V value of 1.775 with a corresponding doping concentration of approximately 22 ppb.

A compilation of cold cases using scanning electron microscopy at the University of Rhode Island

Scanning electron microscopy combined with microchemical analysis has evolved into one of the most widely used instruments in forensic science today. In particular, the environmental scanning electron microscope (SEM) in conjunction with energy dispersive spectroscopy (EDS), has created unique opportunities in forensic science in regard to the examination of trace evidence; i.e. the examination of evidence without altering the evidence with conductive coatings, thereby enabling criminalists to solve cases that were previously considered unsolvable. Two cold cases were solved at URI using a JEOL 5900 LV SEM in conjunction with EDS. A cold case murder and a cold missing person case will be presented from the viewpoint of the microscopist and will include sample preparation, as well as image and chemical analysis of the trace evidence using electron microscopy and optical microscopy.

Performance characteristics of an optimized distributed erbium-doped fiber amplifier

Abstract In our previous work, we reported the optimum design for 10 km and 50 km length fibers for a distributed erbiumdoped fiber amplifier(DEDFA), based on a comprehensive computer model. This design contained an erbiumdoping concentration of 20 - 22 parts per billion, V of 1.775 with a signal wavelength of 1555 nm using low signalpowers, -30 dBm, and bidirectionally pumping at 1480 nm. This paper investigates the effects of operating aDEDFA at the optimum doping concentration with large input signal power in which the ASE has a greater effecton the overall performance. The final issue addressed is the noise figure under similar circumstances. 1 Introduction Much work has been performed in the area of distributed erbium doped fiber amplifiers, which differ fromlumped amplifiers in that the gain is distributed over the entire fiber span and compensates for the intrinsic fiberloss{1-24]. Ofprimary importance for DEDFA design is to have the amplifier gain per unit length be compensatedfor the intrinsic loss( 0.2-0.4 dB/krn) over the entire fiber length. Due to technological difficulties in controllingthe erbium concentrations in DEDFAs, much of the recent work use DEDFA designs which provide much gainin some sections of fiber and loss in other sections, resulting in a net gain of unity. It should be clear that sucha system cannot perform as well as a truely distributed amplifier if noise is considered. Not only will a properlydesigned DEDFA have better noise characteristics , but it will have the added advantage of being length andUIT1 power insensitive. The only requirement is that sufficient pump power be used to fully invert the amplifierfor the chosen length. The basic requirement for achieving truely distributed gain is to use sufficiently low dopingconcentrations so that the fully inverted amplifier will have a gain per unit length equal to the intrinsic loss of thefiber. In this paper, the key fiber parameters for optimizing the design for DEDFA are addressed. The secondsection of this paper concentrates on applying the optimum design to applications in DEDFA, namely, higherinput signal powers to determine gain compression effects and to determine the corresponding noise figures.