![]() |
MIT Electrical Engineering and Computer Science
EECS Event |
Tuesday, November 20, 2001
4:00 PM (reception 3:30)
Edgerton Hall, Room 34-101
MTL VLSI Seminar and Boston Area MEMS Seminar
Abstract
Fluidic Self Assembly is a novel technique for accurately assembling large numbers of very small devices. The small size, planarity, and accuracy of the assembly also result in very low parasitic interconnects, comparable to on die traces. This massively parallel assembly process combines the capability and flexibility of assembly with the cost effectiveness of integration.
In the Fluidic Self Assembly process, individual parts are made in large numbers, separated into a random mass in a fluid, and are transported to sites where they orient themselves and assemble. The randomness of the process usually makes physicists and electrical engineers uncomfortable, but chemists and biologists are used to throwing parts in a beaker and having them self assemble! The statistics can be made so much in favor of assembly that the number of incorrect assemblies can be many orders of magnitude less than the number of correct assemblies.
The ultimate goal is to bring the advantages of very low cost and low electronic parasitics to a wide variety of advanced technologies and devices. The flexibility of this approach greatly exceeds that of integration strategies, a very distinct advantage of assembly, while achieving the low cost of integration, and similar electrical parasitics and size.