"There’s so much we don’t understand, even about the current technology we’re using," she said. "A lot of times people do something because it works; they don’t know why, but it does. Then suddenly when it doesn’t work, we have to understand what is going on. We’re always coming up with new understanding to explain why."

Nearly a decade ago, Rosenbaum joined the Department of Electrical and Computer Engineering faculty, drawn in part by the research capacity she saw in the Coordinated Science and Micro and Nanotechnology laboratories. Taking a pragmatic approach, she established a program to address problems of devices and integrated circuits.

"In my laboratory, we do extremely practical research. The research isn’t as risky as that of people who are focused on inventing new fields of inquiry," she said, "but it’s still challenging, interesting, and fun. Our work can help companies to use new technologies, make more reliable products, and lower the cost of technology–which can reduce costs for the manufacturer and the consumer."

Initially, her research in integrated circuit reliability was in the area of reliability physics–the study of the mechanisms by which individual transistors wear out or fail. More recently, the focus has shifted to assessment of overall circuit reliability and ways to improve reliability. For example, when an integrated circuit is handled, whether at the factory or by the person who picks it up to plug it into an electronics board, the circuit can pick up static charge, Rosenbaum explained. That same chip will later come in contact with a grounded conductor, and current will flow.

"The current can be enormous, from 1 to 10 amperes," Rosenbaum said, "and if that current travels through the integrated circuit, it goes through tiny devices that today have dimensions much smaller than a speck of dust. We cannot allow this current to go through the transistors in a modern integrated circuit. We need to provide another path for the safe dissipation of this current."

One solution is placing ESD (electrostatic discharge) protection circuits on the chips to provide paths for dissipation of the static charge. Rosenbaum’s group is designing new structures or devices that can do that. These devices are off until a large current pulse is detected, and then they switch on. The challenges are to design them to turn on quickly and to be strongly off the rest of the time. In addition, her research group is designing ways to insert these structures into the integrated circuit without adversely affecting overall operations.

Another critical problem Rosenbaum’s research group is tackling is noise that affects circuit operations and reliability. Research to make silicon-on-insulator integrated circuits immune to the noise that can cause loss of data at circuit nodes is especially promising. Other projects focus on noise in mixed-signal circuits where digital circuits inject current carriers into the common silicon substrate and thus affect analog circuits.

"People have design guidelines–rules of thumb–that are intended to reduce the coupling between these two: put them far away and put insulators between them," Rosenbaum said. "But things that look like good insulators at low frequencies don’t look like good insulators at high frequencies. We’re developing analysis tools so that people can simulate their circuits and see if noise couplings in a particular design are acceptable or not."

The work is valued by industry, and Rosenbaum’s research attracts industry support. Her students also attract industry attention.

"We’re multidisciplinary, so students know device physics, circuit design, and simulation–they receive a broad training and have good skills," she said. "They’ll be set for a career in what they’ve done their dissertation in, and they’ll have the broad background to change fields in engineering if they want to do so.

"It’s fun to work with these students; they keep me young," she added. During the semester terms, Rosenbaum teaches classes on circuit design, modeling, and characterization. She also directs a half-dozen or more graduate students conducting research in her laboratory.

Rosenbaum was able to immerse herself in lab work recently during a seven-month sabbatical leave spent in Belgium. There, she worked on electrostatic discharge problems in the government-funded Interuniversity Microelectronics Research Laboratory with colleagues she had met at conferences over the years. Her research was in the field of ESD protection for RF (radio frequency) integrated circuits, a new area of ESD research for her. She said it was fun to be back in the laboratory, making measurements and carrying out other tasks she usually delegates. More important was the opportunity to learn something new.

"It happens often that new things are developed that you need to learn, or your own interests change, and that causes you to study something new or something you haven’t seen in 20 years," she said. "That is what makes this work intellectually satisfying. We go to the lab, and we see new things, and then we figure out what we’re seeing–it’s exciting."

For information about Elyse Rosenbaum’s research, explore the Department of Electrical and Computer Engineering annual report in the Summary of Engineering Research Web collection.