Placeholder Image

字幕列表 影片播放

  • I was done in a couple of other video shown.

  • We're gonna talk about those links between physics and computing on dhe.

  • Again, we're back to the ultimate limits of computation on Dhe.

  • The Brooke I have in Front of me is a book that has caused a massive amount of consternation over the last few decades.

  • It's by a guy called Carrick Drexler.

  • Now, the reason it's perhaps of interest to compete putrefied audiences.

  • 1992 more start standing computer science book.

  • Well, what Drexler is talking about directions.

  • A very polarizing figure is.

  • Can we do computing with individual atoms?

  • Can we think of instead of compiling cold?

  • Could we compile a matter?

  • Or to put it in another way?

  • Drexler didn't put it in these terms, but could we do three D printing with individual atoms?

  • Could you use Those are very, you know, very smallest building block.

  • And if we could, that's a remarkable technology.

  • That means we, you know, because are some of the more extreme proponents of Drexler's vision I had in mind.

  • We could, for example, envisioned some technology, which is practically looks like a microwave.

  • We'd go outside, we'd get some grass we come.

  • We put it into our thing that looks like a microwave oven with press a button and the Dane.

  • 30 seconds later, I pops a stick.

  • This is like Replicators and stuff.

  • It's exactly like Star Trek Replicators.

  • One pan fried catfish.

  • We talked about this on one of previous videos that you guys have manipulated atoms and made a switch with that thes sorts of basic building blocks are being pushed around at the minute they are on dhe.

  • But Drexler's vision is much more to grand.

  • Or that now you're actually claims that we should be able to take any lump of matter unconverted not quite into any other lump of matter, but certainly to break those chemical bonds down on former structure to a predefined blueprint.

  • Importantly, he's not saying that we have, you know, that we're going to disobey the laws of physics.

  • It's not saying that we're operating in some strange parallel universe with the normal laws of physics.

  • Don't work.

  • But here's, for example, talking about systems like Diamond.

  • He spends a lot of time with Diamond, which is a strongly bonded substance.

  • You know, Diamond is fairly hard tight, Covalin bonds holding the atoms in place.

  • And it makes the argument that we should be able to under those circumstances, If we can control the bonding on a single chemical bond by single chemical bomb bases, the atoms will stay in place on.

  • We should be able to build up different structures.

  • So he focused a lot of his time on diamond.

  • Um, I can show you a short video shown.

  • Okay, so we have this microwave object, and on that, the side here with these canisters.

  • What?

  • We have a very simple food.

  • Stocks are very simple stock.

  • That's prehensile.

  • Simple molecules.

  • Ex seal to Water CEO May thin, perhaps molecules that can be broken down into component atoms and then built up into something bigger.

  • And just how much bigger will see in a second on dhe curling, et cetera.

  • Quite how it all works.

  • We don't know.

  • This is science fiction, but like the best science fiction, it makes you think.

  • Is this possible?

  • Could we manipulate matter like this out there in computer file and think about this?

  • Does this disappear into the laws of physics on You know, Rex has been widely castigated, perhaps unfairly, I think, certainly unfairly, in some cases, because these are these are grand ideas, grand visions that are worth worth exploring.

  • Are we anywhere near this?

  • No, we're nowhere near this.

  • But you can see what's happening now is we're zooming in, and now we're down to the 10 nanometer level.

  • So just to give you an idea and atoms a few, there's a fraction off a nanometer across the country.

  • Even animators want to attend nano meters.

  • We're talking about 6 300 out something like that across.

  • So here's these molecules streaming in, so this gets it see to hitch to.

  • What's happening is they're being fed through moving across.

  • It looks very much like the type of machinery were on mills on factories were used to in the macroscopic world, except it's all shrunk down.

  • Each one of these single spears is a single atom, and what's happening is you're getting at, um, album transfer.

  • You're breaking these molecules down.

  • You can see what's happening now transferring them across to these tips.

  • These probes.

  • This is where scanning probably close.

  • It was to get, like myself, get very, very interested in that we have a sharp probe we bring it in close to a surface moving back and forth.

  • This probe is atomic Lee Sharp, so it allows us to see single atoms and single molecules on manipulate single atoms and single monitors.

  • So some scaling for my cross comes At least look at this and think world parts of that we can do problem is very, very small part of it.

  • We can take a move in the video on that's stroll through various different elements.

  • What happens later on as we have different blocks that are transferred across different elements Quite how any of this works, we don't know.

  • Where does it get its energy source?

  • We don't know, and out of the end pops a laptop, the argument being that what's happened is that those very simple molecules at the start by building them up, manipulating the very individual atoms we've compiled matter into this final final form.

  • So this video that you said this is science fiction.

  • This isn't science fiction as far as you have written.

  • This is Drexler.

  • Drexler believes this is this is something that that would be possible in the future.

  • Many of us have many issues surfaces in particular a particular issue in terms of it looks very simple that we just get some blocks and with snap them together.

  • In reality, surface physics is incredibly surface chemistry is incredibly, incredibly challenging.

  • Thio Surmount Put the again.

  • The ideas here are fascinating.

  • It really is.

  • Could we Could we do this even at the level off?

  • You know, how would we get a power source here for that little sorter for the molecules?

  • Forget about a whole laptop.

  • Let's let's reduces all the way down.

  • Can we do information processing with single atoms of single molecules or groups of atoms of groups, groups of molecules?

  • Can we translate all that complexity down on think?

  • Can we, on the basis of individual atoms and molecules, do information processing?

  • And, yes, we've done that.

  • The nano science community has done that.

  • I suspect not many computer file and are familiar with this.

  • It's a beautiful, beautiful piece of work and stoning.

  • The elegant piece of work from, uh, Donna glows Groups of Don's retired now, but his group at IBM all Mardin was searched.

  • Labs was responsible for a lot of the pioneering work in number sense.

  • Indeed, eyeglass group is responsible for manipulating atoms for the very first time on what they spelled out was the IBM local.

  • This is beyond elegant.

  • They don't computing information processing if set up logic.

  • It's with molecules on dhe they don't know really fascinating, where because they've effectively used a domino effect in terms of high molecules, interact with each other to transfer information from imports to, I put on, they've done on the basis of CEO carbon monoxide.

  • Very, very small quantities.

  • You don't have to worry about any poisoning or anything like that.

  • Also, in an ultrahigh vacuum, all sword.

  • Very, very low temperatures four degrees above absolute table.

  • So it's not toxic a tall, but they put them down on a copper surface.

  • When the CEO molecules go down in the copper surface, they absorb in a number of different states they absorb like this where it forms this little dark parts.

  • The red dot represents a sea or molecules, or it forms this type of structure where the two molecules are beside each other, which is a timer to see our molecules together.

  • Or it forms this structure, which is a trauma three molecules together.

  • How did they emit this.

  • How did they see the individual molecules?

  • Will it comes back to that scanning probe technology Sharp probe Closer surface moved back and forth on dhe.

  • If the probe is atomic is sharp, you can see individual atoms and molecules, so what they do is they manipulate the C O molecules.

  • There's a seal molecule so you can operate in a mouldering scan.

  • All you offer something very straightforward.

  • You push the probe towards the surface, and you do that when you move it across the surface so you can manipulate an image just by moving the height of the probe above the surface.

  • So what they do is they set up these arrangements.

  • They set up timers Diamond, diamond, diamond, Diamond, diamond.

  • Then what they do is they bring in another molecules, this one of the green arrow.

  • Bring it close on what that does.

  • Is it triggers?

  • See the red dots of the positions and seal.

  • It triggers this cascade process whereby they all rearrange just like a seven dominoes.

  • They're not falling.

  • The molecules are rearranging themselves.

  • It's not like the falling over, but the rearranging themselves.

  • Now that's clever enough.

  • That's really clever and a little.

  • This is 2002.

  • Although it's 16 years ago, it's still like an incredible, phenomenal people.

  • Just not reversible.

  • Process that No, because dominates on easy on this block.

  • So what?

  • What?

  • We're not talking about it.

  • So this is this is the difference in fundamental science and technological application.

  • First of all, you know, if we were to take this type off approach, I think we're gonna be hard pressed to sell it to anyone because most people are not gonna want to top the computer open liquid helium.

  • That's the first thing pretty, pretty complicated and irritating to keep an ultrahigh vacuum going.

  • Moreover, the bandwidth is about 0.1, or maybe 0.1 of her hurts because you have to set this entire system up and then set it off.

  • It's not reversible the roar of the systems for which it is reversible, but not this.

  • But this is this is exploring the limits.

  • What they do here is they developed different gates by setting up different arrangements of molecules.

  • So here's one example of a non get whereby you got import, ex and input Why here.

  • And so by bringing in molecules either by causing cascade here or causing a cascade here or causing a cascade of both points.

  • Then the output here will change.

  • So basically here if we bring Ah molecule in here to set a one here.

  • But there's zero here.

  • There's no molecule here.

  • Then what happens is there's no change.

  • The aiport remains.

  • However, I'm similar to the opposite way.

  • However, in this case, if you bring in a tripping molecule here and tripping molecule here or control molecule here, then what happens is there is a big chance and you can read that as a change in the airport.

  • So what we have is a logic.

  • It mutated molecules which is is made out of this molecular casket.

  • Okay, it were operates incredibly slowly.

  • Okay, It's under extreme conditions.

  • Borders, a piece of elegant science.

  • It's phenomenal.

  • Why did they need is incredibly low temperatures.

  • Why do they need to be at four degrees above absolute zero?

  • The problems with this system with molecules on metal surfaces, what happens is if you want the temperature of even slightly 10 2030 degrees, something like that.

  • Then Thurman motion kicks in on the molecules, starts to wonder across the surface, so these beautifully engineered structures you've made will fall apart.

  • Just instituted diffusion, effectively grainy in motion is not quite prone in motion with the same type of thing.

  • So what you need to do is find systems whereby this type of diffusion doesn't kick in.

  • And actually Drexler had it right.

  • So what you need to think about is called Vaillant.

  • Bonded systems with the bonds are stronger instead of weak bonds between the molecules on the surface, which is actually quite good because it allows you to do this with these two slave things around with these.

  • But the problem is then, in terms of trying to exploit that in any type of realize herbal technology, it's really difficult.

  • So you need something where the bonds are quite strong on where you can raise the temperature without things hopping around.

  • And they've bean major efforts over the last few years to do this with silicon on.

  • In fact, we're now at the point where this is Michelle Simmons Group in university.

  • New site whales are a number of groups working on this Michelle's group in university site.

  • Where's Bob?

  • Wall calls group in University of Alberta.

  • There's Christian, Christian Jew, Watchem George.

  • Um, sorry, Christian.

  • You know, I always get the pronunciation wrong.

  • CNRS, Who's led to lose who's led a big effort on this, does Stephen Scofield's group in University College, London.

  • We've doubled a bit us.

  • Well, not quite at the level of those other group, for we've doubled a little bit of this in terms off, really manipulating at the single bond level.

  • Michelle's group, however, has may have done important leap whereby they've taken this type of atomic scale engineering, and they made a transistor writer.

  • But this this paper really nears its colors to the mast, a single atom transistor on what they do is they pattern a silicon surface.

  • So we get a video for 60 symbols a number of years ago, which describes just how this works.

  • But basically you take a silicon surface, you crack it open and you form two surfaces.

  • The atoms here and the atoms here are really uncomfortable.

  • They're no longer in the nice environment where they were bonded to the neighbors.

  • Atoms are very gregarious.

  • They like to bond, so when you spit them open when you split that that Krystal open exposed surfaces.

  • It's a very high energy state in the Athens Don't like to be in that state.

  • So what you can do is you can passive it that surface so you can make it less reactive by adding order atoms to it So the bonds are tied up in a really good one is hydrogen, so you for silicon.

  • So you take the silicon crystal in, vacuum the surface and you've bombarded with hydrogen on.

  • What you can do is you can sit above an individual hydrogen atom, bombarded with electrons and remove one single hydrogen atom.

  • So you got one chemical bond.

  • And again, to be fair to Drexler, this is the type of thing he was talking about quite some time ago.

  • Then what you can do, and this is what Michelle's group has done.

  • Is that that one dangling bond?

  • Now you got that one bond.

  • It's literally called dangling born.

  • That's what we call it.

  • So you got this in your its surface with this one dangling bond.

  • Then what happens if you put all the molecules down?

  • In this case, they've they've put off Boss List containing Compliment down is that it would bond just that that point, and that means you can put down of single Foster's autumn at that point.

  • And that's what they've done on phosphorus is what's called a dope int in silicon, but basically, you add impurities to control the electronic properties of Silicon and Dato practice.

  • Every single electronic device and every single semiconductor device relies on the properties of doping in this case were done to the limit of doping with a single atom.

  • And that's what they've based.

  • This this single atom transistor on is they've managed to access and address that single atom to former of a single atom transistor, so that that's really quite phenomenal.

  • Interesting.

  • The one thing we were missing okay, you could create this this dining born, but really you need to be able to correct it and to reverse it to be able to ever correct.

  • For one thing, it's taking a long time to go.

  • It's like a nineties when it was shown that you could put the you remove the hydrogen second, a long time to get enough control over the probe to be able to just connect the probe or push the problem of the surface and transfer that single hydrogen atom to hell out defect.

  • So here's the defect that one single bond and, as you can see, hits function, eyes tip.

  • Then they bring it in, and they connect with directly that diamond bones will tip comes in dining bone.

  • Tip on.

  • At the end of the tip was a single hydrogen atom and the transfers and you heal it up.

  • So we're now at the point where we can make these structures make these standing bond features but also repair them.

  • Which means that when it comes to generating logic, it's in terms off making structures whereby we can have a raise of thes standing bone features like communicate with each other not in the same not quite the same as the molecular cascades, but the same idea where perhaps we transfer instead of a physical transfer of molecules.

  • We have a transfer of electrons, then we really are processing at the atomic limit.

  • We got a long way to go before we get back to Drexler's Nano Systems.

  • A long, long way to go.

  • But then the next ideas well, can we Actually, we got information processing at the single atom limit.

  • We're processing matter of a single atom limit.

  • How far can recourses 13213 There we are.

  • Back up.

  • Nice.

  • 51 three Time