Memristors ready for prime time

PORTLAND, Ore. — Memristors, the fourth passive circuit in electronic circuit theory, have moved a step closer to prototyping with the harnessing of a substrate material that could yield a new memory device by 2009.

In April, Hewlett Packard Laboratories researchers claimed to have "discovered" memristors, which joined resistors, capacitors and inductors as the fourth passive circuit postulated by University of California at Berkeley professor Leon Chua in a 1971 paper.

Now, HP Labs (Palo Alto, Calif.) said it has demonstrated how to control its memristor material, which changes resistance in response to current flowing through it. The advance promises to speed development of commercial prototype chips for its RRAM (resistive random-access memory) by next year.

"We now have experimental proof that our memristor behaves the way our theory predicts," said Duncan Stewart, principle investigator for memristors at HP Labs. "Plus, we have now demonstrated engineering control over the memristor device structures, which will enable us to build real chips very soon."

An atomic force microscope image shows 17 memristors sandwiched between a single bottom wire'that makes contact with one side of the device and a top wire that contacts the'opposite side.'The wires here are 50-nm wide. (Image courtesy of Jianhua Yang, HP Labs).

HP Labs' memristor is a two-terminal, two-layer semiconductor constructed from layers of titanium oxide sandwiched between two metal electrodes in a crossbar architecture. One layer of titanium oxide is doped with oxygen vacancies, making it a semiconductor; the adjacent layer is undoped, leaving it in its natural state as an insulator. By sensing the resistance between the two electrodes at the crossbar, the "on" or "off" state of the RRAM can be determined.

With one layer of titanium oxide in its natural state as an insulator, the memory switch is in its "off" state. By applying a voltage bias across the crossbar junction, oxygen vacancies drift from the doped layer of titianium dioxide to the undoped layer, causing it to begin conducting, thereby turning "on" the memory switch. Likewise, by changing current direction, oxygen vacancies can be made to migrate back into the doped layer, thus turning the memory switch "off."

The main advantage of the memristor is that its resistance changes are nonvolatile, and remain until a reversed bias voltage is applied. That allows oxygen vacancies to migrate back into the doped layer. Switching times today are about 50 nanoseconds.

"People have been working on materials that exhibit resistance switching similar to our memristors for quite some time, but there have been a wide variety of explanations as to why they work," said Stewart. "Our demonstration puts the mechanism on very solid footing. We know just what is happening: Oxygen vacancies are changing the characteristics of the metal-oxide interface."

Still, knowing that oxygen vacancies changed the resistance of titanium oxide was not enough to exert engineering control over the material. HP researchers also needed to characterize the material with detailed measurements. They initially assumed that oxygen vacancies were affecting the bulk properties of the metal-oxide material. HP now claims that nanoscale changes at the interface between the oxide and the metal electrode, rather than the bulk characteristics of the material, affect the memristor's change in resistance.

"We have now established experimentally that oxygen vacancies are changing the metal-oxide interface's electronic barrier," said Stewart.

The researchers also claim that the memristor material works by thinning the Schottky barrier--the electronic barrier at the interface between metals and semiconductors--rather than by changing the bulk characteristics of the titanium oxide.