Preview: 1997 MRS Spring Meeting
- PDF / 161,191 Bytes
- 1 Pages / 576 x 777.6 pts Page_size
- 66 Downloads / 225 Views
Preview: 1997 MRS Spring Meeting San Francisco, California • March 31-April 4,1997 Meeting Chairs: David J. Eaglesham Bell Laboratories, Lucent Technologies Linda Griffith-Cima Massachusetts Institute of Technology Alexander H. King State University of New York, Stony Brook
The 1997 Spring Meeting, with 26 technical symposia, reflects the immense breadth of the materials community covering a wide range of topics from polymers to semiconductors to superconductors and from polycrystalline thin films to epitaxial growth mechanisms. The meeting, located at the San Francisco Marriott, runs from March 31 to April 4. A number of symposia on electronic materials will take place at the meeting. In mainstream Si processing, integration problems (Symposium K) have become ever more critical, overlapping reliability issues (Symposium J) and surface preparation (Symposium P). Low-dielectricconstant materials (Symposium N) offer more challenges, but these materials are strongly needed by industry. Many manufacturers are looking to implement them imminently. It is therefore a critical time for their development. A new symposium in the silicon area is Symposium O, which, along with the associated tutorial, will address materials problem-solving techniques and the identification of failure- and yield-loss mechanisms for Si ultralarge-scale-integration processing. It will highlight new inspection and diagnostic methods for understanding effects of microcontamination, identifying problems associated with new dielectrics and electrodes, and understanding other issues associated with devices relying on high-purity materials and nanometer-scale features. Also new is Symposium E, reflecting the need to understand the atomistics of diffusion in Si. The integrated-circuit manufacturer needs to model the details of atomic diffusion in order to predict device performance from process parameters. This need is pushing understanding of Si diffusion processes to the next level. While materials issues still dominate the new nitrides (Symposium D), processing problems (Symposium C) are the focus for more traditional compound semiconductors. Another class of materials combines versatile mechanical properties of polymers and electrical properties of semiconductors to produce light-emitting diodes
and field-effect transistors. These organic materials and issues associated with their use will be considered in Symposium H. For amorphous silicon (Symposium A) and flat-panel displays (Symposium G), the symposium titles remain the same as ones run previously by MRS, but devices are moving into a new era. The relationship of magnetic properties and device performance to structure at the atomic, nanometer, and submicron length scales will be the focus of Symposium M. In the area of optical materials, Symposium S will look at materials that change their transmission properties when exposed to varying light intensities (optical limiting materials). Now that many components are computer-designed, the process of rapid prototyping and solid freeform manufacture
Data Loading...