You know microchip technology is coming down in size when scientists can count the number of atoms involved in a circuit. It is a challenge to build things at this scale and then being able to check they can actually do something useful.
The Nanotechnology Group within the Physics Department at Trinity College Dublin is involved with other research teams in an EU-funded project trying to tackle these problems. It has built a £1 million (€1,269,738) piece of equipment to assist this work and has already begun to patent findings coming from the activity.
Dr Igor Shvets came to Trinity in 1991 and became involved in nanotechnology research there. He now heads a team of nine researchers and the group wins research income worth about £300,000 per year. "Most of this comes from research contracts," Dr Shvets said. "We have a number of research projects, some fundamental and some applied. These projects are related to imaging on a very small scale."
The group has joined with the University of Kasel, Austrian firm TMS in Vienna and the German company that leads the project, Omicron of Frankfurt, in one EUfunded research initiative trying to find new ways to build and manipulate incredibly small circuitry.
"The aim of the project is to develop the technology to build nanostructures in silicon," Dr Shvets said. Circuit designers are already building devices down to 300 or 400 nanometres across (a nanometre is a billionth of a metre) but the aim is to make even smaller devices down to 100 nm. Many of these devices are built using light as a tool to "cut" the silicon, but light-based systems have a physical restriction known as the "diffraction limit" related to the wavelength of the light source. Put simply, light can't be focused finely enough to build smaller devices. This was why there was now "an intensive research activity" worldwide to find replacements for light-based devices, Dr Shvets said.
Ironically, the technologies allowing scientists to visualise these tiny circuits have continued to improve, so much so that the atoms making up the surface of a device can actually be "seen" and counted. These machines can be used to capture and assemble atoms in an ultra-small circuit, but this takes hours and would not support the mass production required by the information technology industry.
The research group is taking a new approach, building devices using a jet of chemically-active plasma ejected under pressure through a nozzle just 50 nm or about 500 atoms across. "You bring the nozzle very close to the silicon surface and then etch the surface. You can think of it like a gas torch, but the torch is only 50 nm wide," Dr Shvets explained.
Trinity is responsible for building the nozzles and the instruments that will allow the nozzle to track over the surface without actually touching it. It is more difficult to keep the nanonozzle from "crashing" into the silicon surface. The so-called "flying height" required with these devices is between 30 nm to 50 nm.
The nanotorch would be used to cut a "stencil mask" which does what any stencil does, blocking some parts of a surface and leaving others exposed. The stencils would be used in conjunction with a "focused ion beam" to mass produce silicon devices for the information technology industry. This light-free method would allow the construction of circuits much smaller than the current generation.
The Nanotechnology Group is applying this expertise in another EU-funded project involving the French firm Thompson, Nanosensors of Germany and the University of Stuttgart. The object is to develop a system that can produce an "image" of a tiny electronic circuit by studying the very weak electromagnetic signal coming from its components.
A typical integrated circuit might be 10 mm square, but its active elements might only measure one or two millimetres square. This tiny space might in turn have 100 to 200 individual devices packed into it, all of them positioned in a very precise way.
The project involves developing a "microantenna" that can read the electromagnetic signal coming from these devices. The antenna is in turn controlled by a very sensitive tracking system that allows it to pass over the circuits with a flying height of just 30 nm to 50 nm. "It is essential that the probe never touches the surface," Dr Shvets said, nor should it be able to influence or interfere with what the circuit is actually doing.
Such a device would not produce a visual image. Instead it would provide an "image of the EM radiation", a map that would allow researchers to assess whether the circuit was performing as it should. This should help circuit designers build better devices and help spot problems early on in the building of a new circuit, long before it goes into production.
