This position is part of the Netherlands Initiative for Energy-Efficient Computing
NL-ECO. NL-ECO is inspired by the need for a radical improvement in the energy efficiency of information and communication technology (ICT). This energy consumption is problematically high and is growing to unsustainable levels. The NL-ECO program, part of the Dutch National Science Agenda (NWA), aims to develop new concepts and associated materials for this purpose. In NL-ECO, 33 academic, industrial, and societal organisations join forces on one of the major social challenges; how can the rapidly increasing consumption of energy in ICT be curbed?
In 2020 our group in Twente introduced the concept of dopant network processing units (DNPUs, Nature
577, 341-345 (2020)). These are highly nonlinear silicon-based electronic devices that can be integrated in novel architectures for highly efficient computing. In particular, we see application potential for edge AI, where low power consumption and low latency are key. In this PhD project you will work on the fabrication and characterization of a new generation of dopant network processing units (DNPUs), that can be operated at room temperature and that are suitable for real-life static and time-dependent tasks.