Espace presseThema

Future Memories

Nanocrystal Memories to Boost European Electronics

The European project NEON aims at building new and higher-performance electronic memories by making use of the nanometric-sized crystals. First results are promising and are likely to be a shot in the arm for the European electronics industry.

Nanocristal de silicium

© D. R.

A silicon nanocrystal (Si) embedded in a thin layer of silicon dioxide and placed between two silicon electrodes. With such a structure, an electric charge can be introduced and stored so that a bit of information can be written, read, or erased at will.

Physicist at the CNRS' Center for Structural Studies and Materials Development (CEMES) and head of the European project NEON (Nanoparticles for Electronics), Alain Claverie is a satisfied man. Since the project was launched three years ago, real progress has been made toward developing nanocrystals for use in the electronic memories of the future. As he would be the first to admit, the funding and collaborations that NEON has attracted have created a very dynamic atmosphere around the project and an accelerated rate of progress.

The electronics industry is the scene of a continual and hard-fought competition to make components smaller, better and cheaper. At the same time, fundamental laboratories like CEMES have in recent years been studying nanoparticles, or tiny bits of matter whose diameter can be measured in nanometers. Now these two pursuits are coming together and one of the points of contact is the hunt for a new type of electronic memory. This is where the project NEON comes in, in a technology named “floating gates” memory which is already revolutionising the components market.

In this technology, the basic building block of memory is a MOS transistor in which the controlling element, the gate electrode sits on a layer of silicon dioxide (SiO2, an insulator) containing a layer of polycrystalline silicon (Si, a conductor), acting as a floating gate. If the silicon layer carries no electrical charge the transistor functions normally and corresponds to a memory state of 1, but if charged the influence of the gate is neutralised and the transistor does not function, which corresponds to a memory state of 0.

The problem arises with the possibility of flaws in either the silicon or the silicon dioxide layer, which would cause the charge to be lost and the memory state with it. The idea behind NEON is that replacing the continuous layer of silicon by a layer of individual nano-crystals not in contact with one another would be the ideal way to make sure charges do not all leak. Therefore, a surefire and efficient means of fabricating and characterising such structures would constitute a major step forward. One of the most challenging steps is to be able to control with nanometric precision the vertical position of the nanocrystals as well as their density. At the moment the most promising technique of the three under study within NEON is to insert the crystals in the silicon matrix by ion implantation (that is, by bombarding it with ionised silicon atoms at low energy – 1keV) followed by thermal annealing. Results have been so promising (at least as good as those obtained after 20 years of optimization of the classical technology) that the project's industrial partners have run several lots which are under evaluation at the industrial level.




Alain Claverie
Centre d'élaboration de matériaux et d'études structurales (CEMES)

View web site

Back to homepageContactcredits