The third-year Ph.D. candidate pioneering work on the “first switch effect” will help researchers understand the reliability of ferroelectric field effect transistors.

Artificial intelligence (AI) and machine learning (ML) model parameters have been growing exponentially in the past decade, driving the need for advanced hardware to keep pace with rising demands in energy consumption, speed, and efficiency.

Ferroelectric field effect transistors (FEFET) have emerged as a promising solution for logic and memory applications in next-generation edge computing, artificial intelligence, and enterprise cloud applications. The non-volatile memory technology harnesses electric polarization to store data in binary, and has been renowned for its speed, density, and ultra-low energy consumption.

Their widespread adoption, however, hinges on addressing critical reliability challenges.

Georgia Tech School of Electrical and Computer Engineering (ECE) third-year Ph.D. candidate Priyankka Ravikumar undertook the challenge of bringing clarity to one major reliability concerns in her paper, “First write pulse-induced interface damage in ferroelectric field-effect transistors.”

The research, completed in ECE associated professor Asif Khan’s lab, recently won the Best Student Paper Award at the 2024 IEEE International Integrated Reliability Workshop.

The research reported on a new phenomenon in FEFET, known as the first switch effect and explains the phenomenon’s origin. The findings reveal that the very first write operation on a FEFET contributes to nearly 50% of the total degradation seen over the device's lifetime.

The work is an important step towards understanding the fundamental drivers of degradation in FEFETs, laying the groundwork for developing innovative strategies to mitigate degradation, improve device performance, and facilitate the widespread adoption of the technology.

Ravikumar will be presenting this work at an invited talk next year at the 2025 IEEE International Reliability Physics Symposium.