Adrian Thompson has publications about Evolutionary Electronics (here). He tries to answer: What can evolutionary design do that conventional methods can't? He is faculty of the School of Cognitive & Computing Sciences: COGS.
He is associated with the Evolutionary & Adaptive Systems group (here), the Cente for Computational Neuroscience & Robotics (here), and the Centre for The Study of Evolution (here).
All this is going on at University of Sussex. And somewhere else I found this paper: Design of single-electron systems through artificial evolution
by Adrian Thompson and Christoph Wasshuber
Abstract We show how evolutionary methods can help in the design of single-electronic circuits with an example of evolving a simple NOR gate. Evolutionary algorithms, capturing the bare essentials of Darwinian evolution, work differently from conventional design methods, and have the potential to explore new territory. Our preliminary evolved circuit is far from an ideal NOR gate, but has interesting properties. It was evolved to work at a temperature of 340mK, and its performance deteriorates if the temperature is lowered, as well as if it is increased. This is contrary to the usual behaviour of single-electronic circuits, which generally improve with decreasing temperature. We hypothesise that the circuit exploits or relies upon the simulated effects of the particular thermal energies of the electrons at around 340mK.

Keywords: Nanoelectronic circuit design, evolutionary algorithms, design automation, single-electron circuit simulation, physics of computation.

Labels: Economics