Ultrafast switching of the structural phase of phase-change memory materials

We will apply our expertise in the physics of ultrafast melting to phase-change memory (PCM) materials. In such memories, “0” and “1” are represented by the amorphous and crystalline phase respectively.

We will seek the fastest way to switch phase in the commonly used material Ge2Sb2Te5 (GST).The potentially fastest way to write “0” is through non-thermal amorphization which means that the material can go amorphous within a picosecond without entering the liquid state. This will be investigated by a series of two experiments in one sub-project.

If thus curiosity driven project is successful it will be possible to do this 1000 times faster than at present.To write “1” we must go through a liquid state that recrystalize. We have 20-years of experience of this process in the III-V semiconductor material InSb. GST is more complex, and may undergo liquid-to-liquid phase transitions like GeSb which we recently have studied.

We will investigate the bottlenecks and conditions for crystalizing the GST in three experiments.

This subproject will apart from increasing the write speed in GST also facilitate  studies of liquids not easily available in the laboratory and visualize liquid-liquid phase transitions.The experiments will be carried out by my group at femtosecond X-ray beamlines such as FemtoMAX or free-electron lasers.

My group has been pioneering X-ray methods and instrumentation for visualizing photo-induced structural dynamics for over two decades.