HP's memristor the field leader for new memory technology?

Atomic force microscope image of 17 HP experimental memristors.

HP scientists announced in a paper published in the April 30 issue of the international weekly science journal Nature that they had demonstrated elements in the lab with all the characteristics of the theoretically predicted 'memristor'.

By looking at the equations that describe the behaviour of the three well known circuit elements; resistors, capacitors and inductors, in 1971 Leon Chua, now a Professor at the University of California Berkeley, predicted that a fourth circuit element should exist. Physical proof has been elusive and HP's success is attributed to the behaviour of materials at the nano-scale.

R. Stanley Williams, an HP Senior Fellow and director of the Information and Quantum Systems Lab (IQSL) said "To find something new and yet so fundamental in the very mature field of electrical engineering is a big surprise," – "This opens up a whole new door in thinking about how chips could be designed and operated,".

Advertisement

The resistance of a memristor depends on the amount of charge that has flowed through it. The HP devices could be written to by passing a direct current through them in one direction, erased by reversing the current direction and read using an alternating current.

Duke University image of 'swarms' of carbon nanotubes

The memristor is not the only candidate for a new form of memory; There is a molecule sized switch being developed at Glasgow University, carbon nanotubes at Duke University, a nano scale transistor using graphene at Manchester University and IBM's racetrack magnetic domain memory. The memristor however would seem to have the advantage over all of these because it is potentially a simple device that can be fabricated at very high memory cell densities using existing manufacturing methods. It has been fast-tracked by HP and practical devices could be with us in just a few years time.

Since the Nature article was published there has been controversy over whether the HP teams work really constitutes proof of the memristor, or even if the concept of a fourth circuit element is correct. It has also been said that the HP device is simply another form of RRAM.

(Terry Relph-Knight)

See also:

• IBM: Racetrack memory instead of MRAM or hard disks
• Glasgow University – molecule sized switch
• Manchester University – graphene transistors
• Duke University – carbon nanotubes

(trk)