In search of a better semiconductor chip
A University of Texas at Arlington materials science and engineering researcher is working on a project to determine when failure happens in electronic device circuits. The research ultimately will help manufacturers design better semiconductor chips.
Choong-Un Kim, professor in the Department of Materials Science and Engineering, was awarded a $285,0000 grant from the Semiconductor Research Corporation (SRC) for the project “Enabling Electromigration Solver for Solder Joint With Various Packaging Structures and Alloys.” This is the latest in a series of grants he has received from SRC that aims to answer the demand for improved device reliability.
The SRC is a high-tech research consortium and a leading semiconductor research association comprising 25 companies and government agencies, along with more than 100 universities. The SRC is committed to expediting research that improves performance and reliability of electronic devices.
“Semiconductor chips are becoming smaller and smaller with higher component density,” Kim said. “Eventually you run out of mass to withstand an operational load such as electrical current, making devices prone to failure. The resulting limit in reliability is one of the biggest challenges in device development.”
Kim said his team is trying to understand the current-induced failure mechanisms and develop a method of predicting the reliability limit. With that information, they could suggest design rules to microelectronic industries that guide safe operating current for a wide variety of applications.
“Right now, there is no unified theory or database that I can use to predict that this design and material won’t work for that application,” Kim said. “Each situation requires expensive and lengthy independent reliability assessments. Variations include size, materials, complexity of packaging configurations, and operating temperature.
“For example, a chip in a car might operate at 150 degrees Celsius, but in a cellphone that chip might operate at 80 degrees Celsius and allow higher current.”
Those variables add to the difficulty in predicting a failure point, Kim said.
“Now, chips are asked to work in ever more challenging conditions with greater reliability,” he said. “The semiconductor is aiming for perfect reliability with predictable life. One reason is because airplanes, robots, unmanned aerial vehicles and personal medical devices rely on predictable device performance. This project aims to help in that effort.”
Stathis Meletis, professor and chair of the Department of Materials Science and Engineering, said Kim’s work could enable more rapid, reliable product development in the semiconductor industry.
“Finding that critical point of reliability in semiconductors is essential to continue improving at the current rate,” Meletis said. “There are so many industries that could be affected because every sector uses semiconductor devices.”