© GizmagIllustration of the preferred magnetic orientation of an iron atom on a specially prepared copper surface. The ability of an atom to maintain its magnetic orientation can help determine that atom's suitability for storing data.
The rewritable device stores the equivalent of a news story in a space one-thousandth the size of a needle tip. Pound for pound, it can hold around 1000 times as much information as current hard disks and flash drives.
"In theory, this storage density would allow all books ever created to be written on a single post stamp," said team leader Sander Otte of Delft University of Technology in The Netherlands.
The prototype, which is just 100 nanometres wide — roughly one-thousandth the width of a human hair — features a grid of movable chlorine atoms on a tiny sheet of copper. It needs just one atom to store a "bit", the basic information unit of computing. Eight bits grouped together form a byte, which can hold a single typed letter.
Current storage systems use many thousands of atoms to store a bit. In 2012, computing giant IBM announced that it had reached the "atomic limits of magnetic memory" after building a device that stored a bit in 12 atoms.
The Dutch-led team used the same technology, a "scanning tunnelling microscope", which harnesses an extraordinarily sharp needle to manipulate individual atoms. The researchers created a binary storage system by moving the atoms around like the coloured squares on a Rubik's cube, with atoms representing a "0" and gaps between atoms representing a "1".
The team managed to keep the configuration in place for more than 40 hours, suggesting the device can reliably maintain data. On the downside, it requires "very clean vacuum conditions" and temperatures of almost 200C below zero.
"Actual storage of data on an atomic scale is still some way off," Dr Otte acknowledged. "But we have come a big step closer."
The device is also painfully slow, with text blocks taking a few minutes to read and about 10 minutes to write. But American physicist Steven Erwin said this could be improved using available electronics.
"An advance of this size is remarkable," said Dr Erwin, who was not involved in the study.
"every book ever written..." Isn't that the metadata problem the NSA et al are facing? So much noise or crap? Selection seems more to the point of creating a really useful system. Who's gonna be put in charge of that? Allowed to select which data is important and which is not? That gets political, right?