You are here : Home > Ecodesign: a paradigm shift for more reliable, longer-lasting ICs

News

Ecodesign: a paradigm shift for more reliable, longer-lasting ICs


​CEA-Leti has leveraged machine learning and other techniques to develop component aging models. The goal? To extend integrated circuit lifespans.

Published on 31 May 2022

​Manufacturing integrated circuits that last longer is one way to curb the components' environmental impacts. CEA-Leti is supporting the circular economy by giving engineers component aging simulation tools to help them predict failures and extend product lifespans. Because each transistor is "aged" in its unique operating environment, the simulator generates valuable insights engineers can use to make changes to a circuit's design or, for more critical devices, make the operating conditions less stressful.

The model developed has several unique features. First, unlike with iterative aging models, the intermediate values of a parameter do not have to be calculated to determine the state of a component at a given point in time. Mathematical writing is used instead, and the number of computations that have to be completed is reduced. This advance makes the model faster, of course. But it also means that it could ultimately be integrated into a process design kit (PDK).

Another unique feature is the use of machine learning to extract electrical characterization data under accelerated aging. The values of the model's parameters can be determined based on experimental measurements so that the state of a component in a given circuit at a given point in time can be modelled.

The model, useful for any application with high reliability or performance requirements, garnered attention at the International Reliability Physics Symposium (IRPS). It could help reduce digital technology's environmental footprint.

Top page