Slashdot Mirror
Nano-Scale Robot Arm Moves Atoms With 100% Accuracy
destinyland writes "A New York professor has built a two-armed nanorobotic device with the ability to place specific atoms and molecules where scientists want them. The nano-scopic device is just 150 x 50 x 8 nanometers in size — over a million could fit inside a single red blood cell. But because of its size, it's able to build nanoscale structures and machines — including a nanoscale walking biped and even sequence-dependent molecular switch arrays!"
If we go to the referenced Nature article abstract we see that the development "yields programmed targets in all cases."
The correct terminology then would be "100% Success Rate" not "100% Accuracy".
P.S. Presumably "success" is defined by something like "90% Accuracy", to put an ironic spin on it. But it makes no sense to speak of accuracy in terms of percentage without a reference, such as "a single atom". So the criteria was probably something like X nanometers accuracy.
Can they make gold?
This device manipulates atoms and molecules, not individual protons and neutrons within the nucleus of an atom. So no, it can't make gold out of another element. You can do that with nuclear reactions if you want to live the alchemists' dream.
It's still really amazing. I wish Feynman had lived to see it.
"...always new atoms but always doing the same dance, remembering what the dance was yesterday." -Richard Feynman
Actually, the article is about using a DNA strand to place individual atoms where you want with a 100% success rate. Basically, its using the DNA strand as a robotic arm, in that it does exactly what you would expect a robotic arm to do.
i don't know why this is rated funny, but it's true. even your table has wave-like properties, and theoretically it could be passed through a diffraction grid, and you'd get cool positive/negative interference of the table with itself if you put a wall on the other side of the grid. the only problem is that the table would have to move very slowly...
weinersmith
It's still a chemical reaction, it's just a very precisly controlled one. You would still have to add energy to break the bond in a molecule of CO2. I suspect that if someone goes through all the trouble to do that, they'll have it produce diamonds instead of pencil lead, since at least then you can sell the result and maybe make a bit of profit off of it (though not for long, what with economies of scale and everything. If this is really possible in large scale diamond will be cheaper than glass someday).
no, they can't make gold, because they don't move elementary particles, they move atoms. gold is an atom, hence they'd need gold to make gold, which isn't a very impressive feat. what would be cool is if they could take simple graphite (pencil lead), and assemble it into diamonds, and make the whole process significantly cheaper than diamonds are today. it could be a real game-changer, and i'd really enjoy seeing diamonds that now cost millions of dollars lose almost all their value, thus screwing over anyone who has made large investments into diamond jewelery. something like this happened with aluminium - it used to be a very expensive metal, because it was difficult to extract it from the ore, so there was a lot of aluminium jewelery. then some guy came up with a new way to extract it, and it became the cheap-ass metal we all know and love today.
weinersmith
h >= dp * dx / 2 * pi
Modded informative? This equation is backwards.
h is a lower bound on certainty, not an upper bound.
Ack!! Thank you!
h <= dp * dx / 2 * pi
is of course the correct equation. Note that the text was correct; I just fat-fingered the inequality.