Researchers Develop Atomic-Scale Hard Drive That Writes Information Atom By Atom (techcrunch.com)

An anonymous reader quotes a report from TechCrunch: Researchers in the Netherlands have created a microscopic storage system that encodes every bit with a single atom -- allowing them to fit a kilobyte in a space under 100 nanometers across. That translates to a storage density of about 500 terabits per square inch. For comparison, those 4-terabyte hard drives you can buy today are about 1 terabit per square inch. That's because, unlike this new system, they use hundreds or thousands of atoms to store a single bit. "Every bit consists of two positions on a surface of copper atoms, and one chlorine atom that we can slide back and forth between these two positions," explained Sander Otte, lead scientist at Delft University of Technology, in a news release. Because chlorine on copper forms into a perfectly square grid, it's easy (relatively, anyway) to position and read them. If the chlorine atom is up top, that's a 1; if it's at the bottom, that's a 0. Put 8 chlorine atoms in a row and they form a byte. The data the researchers chose to demonstrate this was a fragment of a Feynman lecture, "There's plenty of room at the bottom" (PDF) -- fittingly, about storing data at extremely small scales. (You can see a high-resolution image of the array here.) The chlorine-copper array is only stable in a clean vacuum and at 77 kelvin -- about the temperature of liquid nitrogen. Anything past that and heat will disrupt the organization of the atoms. The research was published today in the journal Nature Nanotechnology.

Comments

Precisely placing atoms is not new.

[wierd_w]

I have seen AFM images of xenon atoms spelling out IBM on a graphite sheet as old as the 90s.

This smacks of "gimme fundingz plz!".

This work is not terribly novel. If they could dynamically change the state of the arrangement withat applied electric or magnetic fields, that would be worth reporting. This however is not, imho.

Re:Precisely placing atoms is not new.

[Anonymous Coward]

What did YOU invent this week?

I'm still waiting for someone to invent something that somehow does NOT end up being a decade-long litigation between themselves and the fucking patent hoarders.

Good fucking luck with that shit.

Re: Precisely placing atoms is not new.

[Anonymous Coward]

Given that his (GP) point was NOT what they were claiming, here: A kilobyte rewritable atomic memoryM

FT(real)A
A clear challenge now is the controlled integration of these individual functional atoms into extended, scalable atomic circuits. Here, we present a robust digital atomic-scale memory of up to 1kilobyte (8,000bits) using an array of individual surface vacancies in a chlorine-terminated Cu(100) surface. The memory can be read and rewritten automatically by means of atomic-scale markers and offers an areal density of 502terabits per square inch, outperforming state-of-the-art hard disk drives by three orders of magnitude

They are talking about manipulating them in a manner to store and retrieve data with error checking, etc. The real article is pretty interesting, the write-up here and the article that they link to is utter crap

Re:Precisely placing atoms is not new.

[ElectricHellKnight]

This however is not, imho.

Yeah who wants tech and science related news on Slashdot, anyway? I want more black lives matter and election coverage! (Before you fools get on my ass, yes I'm aware of the irony of my comment in relation to my current sig.)

Re: Precisely placing atoms is not new.

[hawguy]

Either way, because this requires an extremely cold temperature, it will likely never even end up in even the most state-of-the-art datacenter, nevermind your PC. This is one of those neat yet 100% impractical things that come around every so often. Could a derivative of this technology some day become practical? Maybe, but not with the copper/chlorine combination seen here. I think DNA based storage would probably come sooner.

If they could figure out how to read and write to this quickly, don't lose too much space to ECC and get good durability, the temperature requirement won't keep it out of datacenters - liquid nitrogen is (relatively) cheap (10 - 20 cents/liter), and a 500X increase in density would make it very attractive - replacing 500 racks of disks with one rack of these would pay for a lot of liquid nitrogen.

Re:Precisely placing atoms is not new.

[hawguy]

look, this would be worthwhile if it had any sort of implication of it being feasible for some useful purpose some day. it does not.

it needs "clean vacuum" and a low temperature. it's less feasible,

the _only_ reason they arranged them in 8 bits and put that out in the pr was to get press time for something that otherwise would not have gotten any. they could have gone with "you can write the bible on so and so small thing" approach too.

bubble memory or whatever is more feasible for use than this, that's saying a lot.

Wouldn't a hard drive technician from 40 years ago say the same thing if you told him of your plan to build a helium filled hard drive the size of your fist that would have a million times more capacity than the washing machine sized drives that were state of the art?

Re:Precisely placing atoms is not new.

[mwvdlee]

No. It should mean that, but it doesn't.

Re: Precisely placing atoms is not new.

[Salgak1]

Ahem. It's a proof-of-concept, technology demonstrator. Some of us have been in IT long enough to remember 10 Megabyte disk units for mainframes, that were the size of washing machines (1970s tech). My first PC had a 20MB full-height hard drive: that was 1987. My current box has 2x256 GB SSDs: their combined size is roughly that of a pack of cigarettes (Mind you, I also have several TB of magnetic disk storage. . .).

Technology evolves. This will as well. . .

So?

[meglon]

The chlorine-copper array is only stable in a clean vacuum and at 77 kelvin -- about the temperature of liquid nitrogen. Anything past that and heat will disrupt the organization of the atoms.

As someone who's been using dos/windows for the past 30 years or so.... THIS is the only problem you've got? Meh.

Re:Not Hard

[penguinoid]

In this case I imagine "hard" refers to "storing data atom by atom at 77 Kelvin (-321 F) in a vacuum". Be sure to keep backups in case it heats up to arctic temperatures.

Re:Perfect

[fahrbot-bot]

Now we have a stable product to install on the dark side of the Moon, just need to figure the network end and we'll be in the money!

The a good network is key here, 'cause in space, no one can hear you stream.

Re:Perfect

[Applehu Akbar]

Now we have a stable product to install on the dark side of the Moon, just need to figure the network end and we'll be in the money!

Except that the Moon doesn't have a dark side. It has a hidden side.

What you're getting at is being able to keep an atomic drive (how cool is that just to say!) array in a naturally cold place where minimal refrigeration will be needed to maintain function. There happens to be a crater at the lunar south pole deep enough that the sun never shines into it, keeping it cold as Hillary's heart.

500 is ~ 2^9

[nyet]

So according to moore's law, we have about 18 years of storage density progress left.

Re:500 is ~ 2^9

[CSMoran]

So according to moore's law, we have about 18 years of storage density progress left.

Moore's law refers to the number of transistors (or other electrical components) in ICs, not Cu and Cl atoms in storage devices. Kryder's law would be more appropriate, but still not on the spot. I agree that 500 is ~2^9 though :).

How lavish and extravagant

[JustAnotherOldGuy]

"...they use hundreds or thousands of atoms to store a single bit."

Those wasteful bastards!