PhD - Leuven | Just now
The limited scalability of SRAM has become a significant barrier to meeting power and performance targets in modern compute systems. This challenge has driven the design community to explore a range of solutions, from hardware-level innovations to rethinking layout strategies and transistor architectures.
Recent advancements in CMOS technology, particularly backside technology, have opened new avenues for SRAM scaling. These innovations offer the potential to overcome the traditional Power-Performance-Area (PPA) trade-offs that have historically constrained SRAM design.
This PhD research aims to:
This work will leverage imec’s predictive PDK (A14 node and beyond) for backside technology to explore a broad spectrum of SRAM design optimizations, including:
Complete SRAM memory circuits will be designed for PPA benchmarking. Selected key innovations will be taken to tape-out, with fabrication conducted via industry partnerships and/or imec’s internal technology platforms.
Required background: Electrical engineering, Engineering Science, Engineering Technology or equivalent
Type of work: 40% modeling, 50% circuit design, 10% literature
Supervisor: Wim Dehaene
Co-supervisor: Pieter Weckx
Daily advisor: Pieter Weckx
The reference code for this position is 2026-051. Mention this reference code on your application form.