In fiberoptics communication systems, detectors are one of the major components. Currently, Ge and GaInAs p-i-n and avalanche photodiodes (APD) are the major detectors for this application. However, an APD with comparable avalanche gain and noise figure to a Si APD is still not available. HgCdTe, because of its controllable low doping, high multiplication gain of over 50, and a high k value (8-12 in contrast of 0.3-2 for Ge and GaInAs), is potentially one of the promising materials for making high performance APD. Phase I of this program investigated the feasibility of using liquid phase epitaxy (LPE) and metal-organic chemical vapor deposition (MOCVD) methods to produce conventional APD and staircase APD. The results, especially the conventional APD produced by LPE, exceeded the program expectation. The characteristics of these APD are not only much better than those recently reported on HgCdTe APD made by other methods, but are also comparable or better than those reported on Ge and GaInAs APD. However, before the HgCdTe APD is used commercially, a number of potential problems have to be solved. Therefore, in this Phase II program, it is proposed to solve these problems and to produce a prototype high quality HgCdTe APD for commercial application. The specific goals include (1) Achieving a low series resistance, (2) Reliability test, (3) Increasing the bandgap of the window layer, (4) Forming a guard ring structure, and (5) Analyzing other device structures. This research addresses the conversion of light beam information into useable electrical signals. This conversion is central to lightwave communication systems. Not only is it important to do this with good efficiency, but it must also be done with as little noise as it possible. This particular research will exploit the demonstrated characteristics of a particular material in which to make such electrical convertors or "detectors" with characteristics superior to those available from other materials.
