The contents of an entire library could be carried on a mobile phone within ten years. Keith Nuthall meets a man thinking small about huge volumes of data
Universities are set to benefit from a quantum leap in data storage that could allow the pages of 2 million books to be copied on to electronic chips as small as a postage stamp.
The downsizing of digital memory and related computers, mobile telephones, MP3 players and other consumer durables has seemed like an almost inevitable progression. But that is not necessarily the case, says Chris Binns, professor of nanoscale physics at Leicester University. "The smaller the piece of magnetic material (used to store data), the more unstable it is." Indeed, today's personal computer hard disks may be reliable for only 40 years, and smaller devices for less time.
To overcome this, Binns is co-ordinating the "nanospin project", an initiative funded by the European Union's Sixth Framework Programme that aims to enable the manufacture of data devices on the nanoscale.
His team's first task is to write one data-bit onto a nanoparticle of 5 nanometres in diameter. Once that has been achieved, other project partners - who include academics at Reading and Surrey universities and colleagues in Greece, Italy, Spain, Russia and Ukraine - will try to arrange a series of these particles on a medium in a pattern that allows data to be written electronically.
Binns says another project, which could take as long as five years, might be needed to deal with technical difficulties that arise from this process, but nanodata recording devices could be available for sale in a decade.
Such equipment could give academics immense data portability power, with capacity to store the contents of a library on a mobile phone or a datastick. "You will be able to carry an enormous amount of information and always have access to it," Binns says.
In addition to benefiting academics working in the field or travelling, there would also be advantages in the laboratory. Many machines have excellent artificial intelligence but not much memory, which forces scientists to break off from tasks to input settings or data. Nanodata storage would eliminate this difficulty. "It would supplement its built-in intelligence," Binns says.
Nanodata equipment would allow libraries to give users large amounts of data that are now stored on tapes. And in data-mining exercises, where terabytes of information are being sifted, nanostorage would vastly increase a library's ability to store memory-hungry video and audio records, Binns says.
So how does this all work? This is where the "spin" comes in, and the answer lies in magnetism. In small data devices, the magnetic charges run in one direction. In essence, this magnetic power is finite and runs out eventually; and the smaller the recording chip, the faster this happens.
What Binns and his colleagues are attempting is the creation of a nanoparticle whose magnetic charges are in some way contradictory: by tying themselves in electronic knots, their magnetism will not be able to leach out.
To do this, Binns and researchers at Leicester are building a machine that will coat an iron or cobalt-core particle with layers of manganese, chromium, platinum or various alloys in a so-called nano-onion. In this way, manufactured nanoparticles get conflicting magnetic charges, or "spin", that can be altered to write data; and they will keep their power longer than conventional devices.
Will it work? "We are confident that we will be able to demonstrate nanoparticles with magnetisation that are useable at room temperature over a long time," Binns says.
With that in the bag, he is optimistic that commercial applications will not be far behind.