Stephen Elliott

Stephen Elliott has been a Visiting Professor in the Physical and Theoretical Chemistry Laboratory at Oxford since January 2021. Prior to that, he was Professor of Chemical Physics in the Department of Chemistry at the University of Cambridge until he retired in 2019, having been, in turn, a Demonstrator (Assistant Lecturer), Lecturer and Reader there since 1979. He was a Professorial Fellow at Trinity College, Cambridge, until retiring in 2020, having been a Prize (Research) Fellow, and then a Teaching Fellow in Physics and Chemistry, since 1977. He was College Steward from 2014 to 2017, and has been Chairman of the Wine Committee since 2005, and a member of the Committee since 1977. In addition, he was Professor of Physics at the Ecole Polytechnique, Palaiseau, France, from 1998 to 2000. He was an undergraduate at Trinity College, Cambridge, and obtained a bachelor degree in Theoretical Physics from the University of Cambridge, followed by a PhD on theoretical and experimental studies of amorphous solids at the Cavendish Laboratory, Cambridge. His work has been recognised by a number of awards and prizes, including a Nuffield Foundation Science Research Fellowship in 1991, the 1992 W.H. Zachariasen Prize, the 2001 Stanford R. Ovshinsky (inaugural) Award, the Chancellor’s Medal, University of Pardubice, Czech Republic, awarded in 2012, the 2014 George W. Morey Award of the American Ceramic Society, the 2017 Royal Society of Chemistry John B. Goodenough Award, and a Kavli-Winton Fellowship, University of Berkeley (2017).

Abstract Title: “What use is really bad glass?”

Abstract

In conventional glass science and technology, the thermal stability of a glass is often of primary importance. What then, rather counter-intuitively, might be the technological applications, if any, of glasses exhibiting extremely bad thermal-stability characteristics, i.e. with exceedingly fast crystallization kinetics?

In this lecture, I will discuss such glassy materials, principally tellurides alloyed with metalloid elements, such as germanium or antimony, which exhibit crystallization times of a few nanoseconds. In other words, these must be among the worst possible glass formers from a thermal-stability viewpoint. However, the fact that such materials have large opto-electronic property contrasts between metastable crystal (electrically-conducting ‘SET’ state, {1}) and glass (electrically-resistive ‘RESET’ state, {0}) means that they can form the basis of new non-volatile electronic (or optical) solid-state memory devices – so-called ‘phase-change memory’ (PCM). Binary bits of information {0,1} are stored as the structural state of the memory material, rather than as electrons trapped at the floating gate in conventional silicon CMOS flash-memory devices.

I will discuss our recent million-atom, device-scale computer simulations of SET and RESET processes in Ge-Sb-Te PCMs using machine-learned interatomic potentials with density-functional-theory (DFT) levels of accuracy. In addition, I will provide experimental evidence that, intriguingly, the GeTex system exhibits three distinct types of switching/memory behaviour, depending on the composition: i) non-volatile crystal-glass-crystal PCM behaviour for x = 1-2 (GeTe); ii) volatile all-glass electronic-switching ‘Ovshinsky-threshold-switch’ (OTS) behaviour for x = 2-7; and iii) volatile crystal-liquid-crystal ‘phase-change-switch’ (PCS) behaviour for compositions from x = 8 to pure Te.