The National Science Foundation's (NSF) Tokyo Office periodically receives and disseminates reports on research developments in Japan that are related to the Foundation's mission. NSF-sponsored researchers currently working in Japan prepare many of these reports. These reports present information for use by NSF program managers and policy makers; they are not statements of NSF policy.
Mr. Enrique Jaen, a Ph.D. student in the Department of Applied Mathematics and Statistics at the State University of New York at Stony Brook, Stony Brook, New York, prepared the following report. Mr. Jaen was a participant in the 1998 Summer Institute sponsored by NSF/NIH/USDA and the Science and Technology Agency of Japan. Professor Isamu Shimizu of the Tokyo Institute of Technology hosted Mr. Jaen. Mr. Jaen can be reached via email at: ejaen@ipanema.ams.sunysb.edu
Project Description:
During my stay in Japan I had an opportunity to learn first hand about amorphous silicon technology at one of the premier institutions in the field and also to continue my doctoral work involving a theoretical model and computer simulation of crystalline silicon. The model is based on experimentally determined lattice bonding statistics, thermodynamically determined local free energies, and with a Monte Carlo algorithm for bonding dynamics and atomic diffusion. The present focus is on a layer by layer silicon polycrystal growth process developed by Isamu Shimizu and others at the Tokyo Institute of Technology (TIT). The process involves depositing a thin amorphous silicon layer, and then using hydrogen to induce a phase change from amorphous to polycrystalline silicon. My advisor, Charles Fortmann, is currently at TIT where he is a professor of Innovative and Engineered Materials. The TIT group in collaboration with Stanford University is developing device applications for these processes. The development of a descriptive simulation is expected to accelerate the application of this process to the preparation of ultra-scaled microelectronic circuits.
Previous Work:
Before coming to Japan the basic mechanics for the simulation program were in place. The program was able to generate either an amorphous or a crystalline lattice section. However, much of the physics was not yet treated and many aspects of the lattice sections needed adjustment in order to more closely reflect experimental results. In particular, in the amorphous silicon section, the hydrogen content and dangling bond density were too large, and bond angle and bond length distributions needed fine tuning. Also, there were still a great many physical processes to consider such as: nucleation, lattice relaxation, mass transfer, heat transfer, and atomic bombardment.
Summer Work:
By working directly with experimentalists on the cutting edge of silicon technology I was able to gain a greater appreciation for their many accomplishments and some of the limitations imposed by experimental technique. Most significantly, discussions with my host, my advisor and the students in the lab allowed me to identify the previously mentioned shortcomings in the model. By comparing with experimental measurement I was able to ensure the integrity of my theoretical model. I began my summer project by developing a means to test the model's prediction of amorphous structure. I calculated the radial distribution function, RDF, for the generated sections. By comparing my calculated RDF with an experimentally observed RDF I was able to tune the bond angle and bond length distributions to more closely match a real lattice. While considering the bond angle and bond length distributions the issue of connectivity in the amorphous sections was raised. By dealing with the connectivity of the system, improvements have been made in both the hydrogen content and the dangling bond density problems. Although, there is still more work to be done.
I also spent time looking at the behavior of the grain boundary regions in silicon. Of particular interest was the structure of the hydrogenated silicon atoms and the manner in which hydrogen passivates the dangling bonds in the grain boundary region. Much of the program mechanics were in place to simulate a crystal-crystal boundary region when I came to Japan. However, in order to simulate a mixed phase model I first needed to design and implement an algorithm that would allow me to control the seeding and subsequent growth of the lattice sections. I also needed to significantly modify and strengthen the existing object oriented event driven simulation framework to facilitate the modeling of dynamic systems. Accomplishing those tasks occupied most of my summer. The resulting visual images will be posted on my web site ( http://ipanema.ams.sunysb.edu/~ejaen/research2.html). My stay in Japan afforded me a close look at the professional environment of my Japanese counterparts. I believe that the personal friends and professional contacts resulting from this summer's collaborative work will endure a lifetime.
Short Term Plan:
I have begun work on including a thermodynamic description for the energies associated with the bond angle and bond length distortions in the model. With the experimental insight gained through this past summer's discussions I can begin designing an algorithm for a moving boundary region. That is, for growing crystalline silicon in an amorphous lattice section.
Long Term Goals:
As the fundamental behavior of the system becomes clear, I will continue to add more physical processes with the goal of exploring the nanoscale device fabrication possibilities for the layer by layer growth technique.