Where IBM managed 25 years ago to painstakingly write their name in xenon-atoms, a team of the Kavli institute for Nanophysics at the Dutch university of Delft has managed to write a complete speech of Richard Feynman on an invisibly small surface using 60000 chloride atoms.
In 1959 Feynman said, in a famous 'There's plenty of room at the bottom' speech that "if we can use atoms, we have all the space in the world"
https://en.wikipedia.org/wiki/There%27s_Plenty_of_Room_at_the_Bottom.
Researcher Sander Otte say "this is our tribute to Feynman".
The technique developed promises to be a stepping stone towards a new way of data storage.
The 'atom memory' of the Delft team has room for about 1 kilobyte of text, using a technique that looks like some kind of braille, and works by shifting some of the chloride atoms in a matrix. This isn't much in our age of terabytes, but what is remarkable, is that it uses 500 times less surface area to store the 1kb than current memory chips use.
"What our research shows is, that we can now really control matter at the atom scale".
His team has been researching the transition from particle physics to effects in materials.
"The transition can be surprisingly abrupt", Otte says. "Sometimes you add one or two magnetic atoms, and suddenly you have a true microscopic magnet."
Arranging atoms used to be an extreme laboratory challenge, with scanning tunneling microscopes playing the main part. These are complex machines which have microscopic needles that can be positioned extremely close to a surface, using computer guidance. The needles can then pick up individual atoms and move them to a new location.
Otte's team uses chloride atoms on a copper surface, because that forms weak, but stable bonds. This allows the tunneling microscope's needle to drag the atoms to any location of choice using a weak current.
The team wrote a 'programming language' which allows them to write any text on their chip, fully automated.
The Feynman text took a full week of shuffling atoms.
Otte acknowledges that it's not a very time-efficient way of storing data.
"I don't say that we should start storing our data like this. But instead of using existing techniques to make smaller and smaller chips, we decided to work on a new approach from the bottom up. It is a new approach, and it works."
The study has been published in today's editon of Nature Nanotechnology. In the same edition, US scientist Steven Erwin writes in a commentary that he is impressed. "This is a factor 100 to a thousand times as compact as a flash drive or hard disk. That is - to put it mildly - remarkable"
http://www.volkskrant.nl/wetenschap/delftse-chlooratomen-vormen-supergeheugen~a4341738/