New semiconductor design could completely change system architecture by enabling brain-like computers

'Bio-inspired' computer systems would disrupt the traditional Von Neumann architecture

By combining a specific semiconductor with ferromagnetic cobalt, scientists at the University of Groningen in the Netherlands have taken a step towards the creation of faster, more energy-efficient computers.

Integrating storage, memory and processing onto a single chip is a challenge in computer architecture, but the new unit - made by combining a niobium doped strontium titanate (SrTiO3) semiconductor with ferromagnetic cobalt - represents progress towards it.

The group, led by professor of spintronics of functional materials Tamalika Banerjee, says that the interface between the cobalt and semiconductor is ‘key' to their success: "We have shown that a one-nanometre thick insulating layer of aluminium oxide makes the TAMR [(tunnelling anisotropic magnetoresistance)] effect disappear," she said.

TAMR is a spin-based phenomenon, which the scientists have combined with the memristor effect (the ability to limit or regulate the flow of an electrical current and remember the amount of charge that has previously flowed through it - important to computer memory) of semiconductors in the new device.

The group has added an electroresistance effect by combining the SrTiO3 semiconductor with cobalt. This means that an electrical field can be used to change its state from high resistance to low resistance, and back.

Another new property, which has not previously been seen in other material systems, is the coexistence of both a large change in the value of TAMR (by applying a magnetic field across the cobalt interface) and electroresistance across the same device at room temperature. Banerjee said:

"This means we can store additional information in a non-volatile way in the memristor, thus creating a very simple and elegant integrated spin-memristor device that operates at room temperature."

The same coexistence cannot be realised with silicon: "You need the heavy atoms in SrTiO3 for the spin orbit coupling at the interface that is responsible for the large TAMR effect at room temperature."

Banerjee said, "We are now considering how to create a bio-inspired computer architecture based on our discovery." The new semiconductors could be used like synapses that connect the neurons in an organic brain; the synapse responds to an external stimulus, but the response also depends on its ‘memory' of previous stimuli.

The researchers say that the ability of their new device to operate while producing much lower amounts of heat could be useful for the internet of things.