In my opinion, the papers of the phase change memory (PCM) session at MRS this year could be divided into two groups, those that were either updates or progress reports, offering little that was really new, while the rest were filling in the detail to account for observed and known effects in conventional PCM device materials. For this latter group, while their work is almost certainly of the highest quality, the following phrase springs to mind to summarize their activities: Filling in the potholes on the road that leads to PCM memory nowhere. There was however one very notable exception.
The new way forward
The paper  by author Semyon D. Savransky was selected for this review because it appeared to offer an entirely new direction and a new type of NV memory based on chalcogenide glasses. It does so by removing the most serious problems that have plagued PCM array progress: the high current reset pulse and, with scaling, the associated high current density and voltage.
There may be a number of secondary conventional PCM related problems that this new approach also removes, including elemental separation on crystallization and the need for an isolation device associated with each memory bit in an array. This new direction does not claim to offer an instant plug-and-play solution to the problems that so far have prevented competitive PCM arrays appearing in production and serious applications. There is still much work still to be done. However, effort in this new direction may offer significant NV memory and NV-RAM rewards.
Some may consider that PCM in the title of this article as a misnomer; no apologies are made or offered in that respect. Readers will have to accept that two different amorphous phases in the same material represent a type of phase change and the materials are from one of the chalcogenide glass families, the similarity is close enough.
Savransky's approach to NV memory uses data states that are defined by different polyamorphic states of chalcogenide glasses that are capable of switching between two disordered states. These glasses are called the polyamorphic chalcogenides (PACs). One example is Ge-Si-As-Te.
While the disorder-to-disorder transition is in itself subtle, it results in a significant and easily detectable change in another electrical characteristic, the threshold switching voltage. It is the different threshold switching voltage that defines the data states and also provides the chosen name switching memory (SM) for this proposed new family of NV memory devices.
The work described has its historical origins in threshold switch device investigations that were carried out in Russia. It is alleged that the PACs were rejected as candidate materials for threshold switches because the threshold voltage jumped between two states and was considered unstable and unsuitable for the intended purpose. Now, with hindsight, it would appear that the apparent instability was the result of the material undergoing polyamorphic changes. The PACs are relaxation semiconductors with electrical characteristics determined by recombination effects, from traps at the Fermi level.
We asked Ron Neale to select a paper from the Phase Change Memory session of the recent MRS that he considered might be able to make a significant contribution to bringing PCM to a position where competitive memory array products can be manufactured. Please make use of the comment section for questions or observations.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.