Crossbar latches take on transistors

US scientists have made nano-scale devices they claim could one day replace current transistor technology. The tiny devices, 'crossbar latches', are made up of a combination of crossed-over platinum wires with steric acid molecules set at their junctions.

The Hewlett-Packard researchers said they could potentially do a better job than present transistors, improving the performance of computers.

The research from the scientists at HP Labs' Quantum Science Research Unit offers crossbar latches as an alternative to conventional transistors.

The crossbar latch provides a key element needed for building a computer using nanometre-sized devices that are relatively inexpensive and easy to build.

The wire grid that makes up crossbar latches is criss-crossed - where two wires cross, there is a switch.

Like a transistor, the structure can manipulate an electrical signal that passes through the crossbar latch.

The tiniest features of transistors that are based on silicon now are about 90 nanometres in size.

That is about 100,000 smaller than the width of a human hair. But the crossbar latch method works in a space of about 2-3 nanometres.

This means it can work at much smaller scales and perform more functions.

The sticking points for the technology which have to be worked out are the lifetime of the devices and their switching speed.

Currently, the devices only work for hundreds of computing cycles, and switching speed is many thousands of times slower than silicon.

The current method for making faster silicon chips is expected to reach a technical dead-end in about a decade.

This process relies on laser light to etch a circuit layout onto silicon wafers. Chemical washes then cut out spaces according to this template, making channels that will be filled to make the circuit features - transistors, and the like.

But this lithographic approach will find it increasingly difficult to focus the laser light at ever smaller small scales.

At the end of 2003, IBM demonstrated its alternative method for microchip fabrication which used polymer molecules that naturally arranged themselves into hexagonal patterns.

Other firms and research laboratories worldwide have also been looking at novel ways to fit more features onto computer chips.

The size barrier can be overcome through molecular electronics. The HP researchers said that their technology is still some way off in terms of widespread use and would not be commercially viable until about 2012.