NIRT: Merged CMOS/Molecular Integrated Circuit (Mol-MOS) Fabrication, Analysis and Design

University Of Virginia Main Campus, Charlottesville VA

Investigators

Abstract

This proposal was received in response to Nanoscale Science and Engineering initiative, NSF 01-157, category NIRT. This program addresses the issues of fabrication and integration of molecular devices with conventional electronic technology. The research team will address nanoelectronics from a comprehensive viewpoint by considering how nano-devices and nano-circuits can be assembled, modeled and designed with the requirements of large-scale integration and manufacturability in mind. For one, we will examine new combinations of electrode and active materials that are more compatible with conventional device and processing technologies. Modern integrated circuits use a variety of metals (Al, Cu, Ta, W, Ti, etc.) and insulators (SiO2, SiN, etc.) in multilevel metalization schemes that result in a complex 2.5 dimensional arrangement of contacts and wires. In contrast, most molecular devices use planar arrangements of Au electrodes that are not compatible with MOS-based electronics. This problem might be overcome by designing organic molecules with end groups engineered to attach only to Cu electrodes. These would ultimately be mated with metal electrode structures formed using state-of-the-art film deposition/growth techniques and electron beam lithography. These structures would incorporate sacrificial insulating spacer layers. Near the end of the process sequence, the insulating layers would be etched away leaving pockets that by their shape, size and Cu endpoints, provide an ideal "home" for the target molecule. The completed template could then be rinsed in a solution containing these molecules, adding them to the structure. Since the volatile organics would not be subject to high temperature processing, this would provide a viable means of adding molecular electronic devices to underling microelectronic circuits. We will also develop black-box models of molecular devices. These models are essential if one is to anticipate novel computing architectures where molecular devices may function far differently from modern transistors, and where optimized circuit design my entail radically different patterns of device interconnection. Presently, such models do not exist but are absolutely required to allow design with nanodevices at higher levels of abstraction. A major part of our effort will be dedicated to education and outreach. The subject of nano-electronics is highly interdisciplinary and does not fall within the normal pedagogical bounds of engineering and scientific disciplines. We will thus offer a compelling 3D animation-based website (building on our existing expertise) and a graduate level web-published "Frontiers of Nanoscience" course emphasizing the fundamentals of nano-device operation and their giga-scale integration.