A 30-year memory problem solved?

Heat and repeat. If Royal Philips Electronics researchers are right, this is how data will be stored in the future. Photo: Measuring Memory

Michael Kanellos Staff Writer, CNET News.com
Michael Kanellos is editor at large at CNET News.com, where he covers hardware, research and development, start-ups and the tech industry overseas.
Michael Kanellos
2 min read
Researchers at Royal Philips Electronics say they have devised a material that could replace existing types of computer memory through a technique that has been studied for years.

The material, a thin layer of antimony and tellurium, could be used to create phase change memory, which largely works the same way CD and DVD discs do. A laser is directed to a microscopic point on a layer of the material to obtain a reflection. The reflection will differ, depending on whether the molecules at that particular point are amorphous or arranged in a crystal. The two types of reflections then become the ones and zeros of data.

measuring memory

If the material can be used commercially, chips made of it could potentially replace DRAM (the memory used inside computers to store data on a temporary basis), flash memory (employed in cell phones for temporary and permanent storage) or hard drives, depending on how the final device works and gets incorporated into computers.

The challenge, however, has been in changing an amorphous bit to a crystalline one. Typically, phase change materials need to be heated to several hundred degrees Celsius in a few nanoseconds--an operation that requires a lot of electrical energy--while preventing any excess energy from changing an adjacent memory cell. IBM's Millipede, for example, writes data by heating thousands of microscopic spikes to 300 degrees Celsius.

Elpida Memory and Intel, among others, have been experimenting with Ovonic Unified Memory, another type of phase change memory, for years.

Back in 1970, Gordon Moore, the man behind Moore's Law, predicted a strong future for the technology, but no one has commercially mass-produced Ovonic memory yet.

Philips says its doped antimony-tellurium material can switch phases with 0.7 volts, a fairly low voltage, compared with silicon. Phase changes occur extremely quickly, typically within 30 nanoseconds on the prototype.

Researchers around the globe are searching for materials and structures that will enable their companies to get off the hamster wheel of Moore's Law. By switching from making chips out of silicon, companies hope to reduce their manufacturing costs while devising chips that will be faster, consume less energy and fit into smaller spaces.

Chips made of these new types of materials are expected to come out over the next two decades; still, there is no guarantee that any of them will work in mass production.

Further details will be published in the April issue of Nature Materials.