Slashdot stories can be listened to in audio form via an RSS feed, as read by our own robotic overlord.

 



Forgot your password?
typodupeerror
Science Technology

Quantum Logic Gate Created Using Excitons 146

Posted by Hemos
from the pr0n-named-physics dept.
Roland Piquepaille writes "In this article, PhysicsWeb reports that researchers in the U.S. "have taken another important step towards making a quantum computer. [They] have created a logic gate using two electron-hole pairs -- also known as "excitons" -- in a quantum dot." According to Wikipedia, "an exciton is a combination of an electron and a hole in a semiconductor or insulator in an excited state These physicists from the University of Michigan and other labs made a quantum dot by using a thin gallium arsenide layer stuck between two aluminium gallium arsenide barriers. And electrons trapped in the middle layer were excited by light to create a quantum logical gate with four states. The group says this could be useful "in other approaches to quantum computing based on the optical control of electron-spin qubits in quantum dots.." This summary contains more details."
This discussion has been archived. No new comments can be posted.

Quantum Logic Gate Created Using Excitons

Comments Filter:
  • by Anonymous Coward on Monday August 11, 2003 @07:03AM (#6664357)
    an exciton is a combination of an electron and a hole in a semiconductor or insulator in an excited state

    I love it when physicists talk dirty to me.
  • by worst_name_ever (633374) on Monday August 11, 2003 @07:04AM (#6664361)
    an exciton is a combination of an electron and a hole in a semiconductor or insulator in an excited state

    Please be advised that by posting this with my quantum computer, I am able to make all possible lewd comments about the above sentence simultaneously.

  • Using Excitons? Wow, exciting! :-)

    This joke goes out to Niels Bohr, my fellow countryman.

    z
    • by Anonymous Coward
      Bohring joke...
    • Apparently so. To quote a major scientist involved:
      "She's [the exciton] giving me excitations. I'm picking up good vibrations!"

      Another nearby scientist, after having accidentally dropped nearly thirty pounts of excitons directly on his crotch added this in high falsetto:
      "Ooo, ooh, ooh! Good Vibrations!"

  • by Suhas (232056) on Monday August 11, 2003 @07:07AM (#6664376)
    Physicists in the US have taken another important step towards making a quantum computer. Duncan Steel of the University of Michigan and co-workers have created a logic gate using two electron-hole pairs - also known as "excitons" - in a quantum dot (X Li et al. 2003 Science 301 809).

    Classical computers deal with binary logic and the bits being processed must be either "0" or "1". Quantum computers, on the other hand, exploit the ability of quantum particles to be in two or more states at the same time. A quantum bit or "qubit" can therefore be "0" or "1" or any combination of the two. This means that a quantum computer could, in principle, outperform a classical computer for certain tasks. However, all the quantum computers demonstrated so far have only contained a handful of qubits.

    Although qubits have been made with trapped photons, atoms and ions, it is generally thought that it should be easier to build working devices with solid-state systems. Several teams have made significant progress with the superconducting approach to solid-state quantum computing. Now Steel and co-workers at Michigan, Michigan State, the Naval Research Laboratory and the University of California at San Diego have demonstrated the first all-optical quantum gate in a semiconductor quantum dot.

    Exciton transitions

    Steel and co-workers grew a thin gallium arsenide layer 4.2 nm thick between two 25 nm aluminium gallium arsenide barriers to make a quantum dot. Electrons are trapped in the dot because the gallium arsenide layer has a smaller energy band-gap than the surrounding material. When excited by light, electrons from the valence band in the dot move to higher energy levels. The excited electron and the 'hole' it leaves behind combine to form an exciton. The system has four states: a ground state containing two unexcited electrons; two states containing one exciton; and a state containing two excitons (see figure). The two single-exciton states can be distinguished from each other because the excitons have different polarizations.

    The researchers showed that they can drive Rabi oscillations between the ground state and the one-exciton states, and also between the one-exciton states and the biexciton state, with lasers. In particular they showed that the quantum-dot system behaves like a controlled-NOT gate in which the value of one qubit is reversed (the NOT operation) if - and only if - the value of the other qubit is 1.

    Although it will not be possible to scale up the system, the group says that many of the ideas and techniques they have developed could be useful in other approaches to quantum computing based on the optical control of electron-spin qubits in quantum dots.

  • Tech problems (Score:5, Interesting)

    by brejc8 (223089) * on Monday August 11, 2003 @07:07AM (#6664380) Homepage Journal
    Many of todays (hoge comparetively) processes suffer from metal migration and huge static power dissapation. If the molecule sized transistors are going to take off they have to solve there problems first or these products will have a lifetime of a few hours.
  • What about.. (Score:5, Interesting)

    by CausticWindow (632215) on Monday August 11, 2003 @07:12AM (#6664398)

    those guys who managed to factor "14" into 7 and 2 with Shors algorithm on an actual quantum computer implementation?

    Heard anything more from them? I googled, but couldn't find anything.

  • by Urkki (668283) on Monday August 11, 2003 @07:13AM (#6664406)
    So they have a quantum gate. But is it just a transistor replacement, which you still would use to build traditional computers, chomping through processor instructions, processing binary (or base 4 or whatever) numbers? Or is for quantum computers working on a completely different paradigm?
    • Check the section called: `The Potential and Power of Quantum Computing' from the page: About quatum computing [caltech.edu]
    • It's still a qbit... (Score:5, Informative)

      by Kjella (173770) on Monday August 11, 2003 @10:05AM (#6665652) Homepage
      Rrom what I understand it, a qbit is in all states simultaniously, even if it is only a single bit (0 and 1 instead of 0 or 1). Tou can emulate it using normal bits, but the way I've understood it's still "opposite", and so it's a completely different paradigm:

      Classic:
      Is 0 the answer? FALSE
      Is 1 the answer? TRUE
      Quantum:
      Qbit x = TestFor(answer) (test all states)
      Read x = 1

      Classic:
      Is 0000000000000000000000000000000000000000000 the answer? FALSE
      ... (about 2^43 lines skipped)
      Is 1111111111111111111111111111111111111111111 the answer? FALSE
      Quantum:
      Qbit xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx = TestFor(answer) (test all states)
      Read xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx = 1010010100010101010010010101001001000101011

      However, noone has been able to get a large number of quantum bits operating. And for few qbits, you'd do faster by simply doing a classic search. A computer using low-qbit "transistors" wouldn't be operating like a base 4 classic computer, but it wouldn't be this wonderful supercomputer either. A cluster of qbit transistors would as I understand simply scale linearly. Two 10qbit transistors would have twice the power of one 10qbit transistor. While on the other hand one 20 qbit transistor would have the power of 2^10 10qbit transistors.

      Kjella
  • by Anonymous Coward
    Damn, does this mean it's the physicists who will get jobs and the electrical engineers who will be unemployed?

    Oh wait, we're both unemployed right now.
  • Not quite yet (Score:5, Insightful)

    by cybermace5 (446439) <g.ryan@macetech.com> on Monday August 11, 2003 @07:17AM (#6664432) Homepage Journal
    It's nicely done, but not the breakthrough that means quantum computers for all. It is, after all, only a NOT gate. I barely consider NOT as a logic function...more like half a function.
    • by HiQ (159108) on Monday August 11, 2003 @07:31AM (#6664511)
      Can a NOT gate half not be?
    • No, it's not. Err... it's not NOT... errr... The gate was a "controlled NOT gate", also known as XOR.
      • Re:XOR (Score:5, Informative)

        by Karhgath (312043) on Monday August 11, 2003 @10:10AM (#6665718)
        Just to educate the masses further, a quantum CNOT(controlled NOT)/XOR gate is not equivalent to a classical XOR gate. A CNOT gate includes a control bit, and retain the information of this bit at the end of the gate.

        What is fun in Quantum Computing is that you do not need a lot of basic gates(AND, OR, XOR, NOT, etc.), you only need a small number of basic gates to make up the Universal gate.

        Furthermore, ALL the elementary gates in QC are reversible!! Unlike classical gates, like XOR, the quantum CNOT, for example, is fully reversible
        • Just to educate the masses further, a quantum CNOT(controlled NOT)/XOR gate is not equivalent to a classical XOR gate. A CNOT gate includes a control bit, and retain the information of this bit at the end of the gate.

          Of course, in a classical system, you're generally feeding more than one particle into each gate lead, which means you can return that "control bit" simply by tapping the appropriate gate input.

          What is fun in Quantum Computing is that you do not need a lot of basic gates(AND, OR, XOR, NOT,
          • Re:XOR (Score:3, Informative)

            by zCyl (14362)
            Um... correct me if I'm wrong... but this would imply that it is impossible to construct an AND gate, right? Because it is impossible to extend a conventional AND gate (or OR, or NOR, or NAND) so that it is reversible. That is, you can't make a reversible gate that takes n inputs to n outputs, such that one of the outputs is the AND of all the inputs.

            All this means for quantum computing, is that in order to emulate an irreversible gate like AND, you have to also keep around enough extra information in the
        • What is fun in Quantum Computing is that you do not need a lot of basic gates(AND, OR, XOR, NOT, etc.), you only need a small number of basic gates to make up the Universal gate.

          And what's fun in standard computing is that you also do not need a lot of basic gates -- you can use just one, such as a NAND, or NOR if you're so inclined.
          • What I meant is that in Quantum Computing, with only the U and XOR gate, you can build all the needed gates, which mean that you only need 2 essential gates instead of about 5-6 in classical science(AND, NOT, OR, XOR, etc.). So quantum gates are a LOT simplier than classical gates... well, it still depends how you look at it =)
    • Shouldn't they be going for a NAND [gsu.edu] gate? Or a NOR gate?
    • I didn't actually want this to go to +5! It's not like I thought about it at all. It just seemed that the term "logic" was a little sensational considering it's only a NOT gate.
    • Re:Not quite yet (Score:3, Interesting)

      by gfody (514448) *
      you only need one instruction, everything else can be built from that. sure you could google up some interesting stuff if your interested [google.com]
    • I barely consider NOT as a logic function...more like half a function.

      But the article said:

      And electrons trapped in the middle layer were excited by light to create a quantum logical gate with four states.

      Like the infamous amplifier in This is Spinal Tap, This is a 4-state Not gate! That's at least as good as two half-nots, and like a double negative, that should make a half a not not a not. Got it?

      I swear, the first one to start into the woodchuck tongue twister...

  • by MosesJones (55544) on Monday August 11, 2003 @07:18AM (#6664435) Homepage

    Was I the only person who read the line

    The group says this could be useful "in other approaches to quantum computing based on the optical control of electron-spin qubits in quantum dots..

    and thought "I'm sure I've seen that on a Powerpoint presentation somewhere". This is clearly uber-smart stuff by uber-smart people, but they are beginning to sound like clueless PHBs dressing things up in techno-babble.

    Maybe this is the fundamental essence of quantum computers, something maybe smart or idiotic depending on the reader, the actual quality of work is only resolved when viewed by multiple individuals.

    I hearby copyright the phrase Quantum-Powerpoint, and the resolution process of determining presentation value which I shall call "De-spinning Qubits"
  • But.... (Score:2, Funny)

    by Anonymous Coward
    Does it run Linux?
    • Re:But.... (Score:1, Funny)

      by Anonymous Coward
      How much is my SCO license going to cost on it?
    • Look, it's only one QBIT NOT gate. So I'm afraid I'm going to have to say potentially. Get it up to a few million QBIT gates of all sorts, and I'm sure we can arrange a port so that it will virtually run Linux. Much more interesting, however, will be if they can port Windows to it, and then write a program that will tell you "34% BSOD" so that you can tell just how potentially unstable your system is, just by thinking about running different programs. (Ummm... MSIE, with gator? 99% dead. Ooops. Bette
  • I think half the fun scientists have these days must be getting to name new particles and "excited states" and all that...

    Lucky guys, I think it /would/ be fun actually...
  • In the future! (Score:2, Interesting)

    by Phishpin (640483)
    I know nothing about quantum computing except that it looks really hard, and I'm betting a lot of other people are in a similar situation. When traditional silicon (or other semiconductor) components have gotten as small and fast as feasibly possible, will quantum computers be anywhere near as usable as the desktops of today are? Meaning, will I be able to use a fast computer without a PhD?
    • Re:In the future! (Score:2, Informative)

      by MindStalker (22827)
      Thats why people make Operating Systems so you don't actually have to use the computer. Anyways the first quantum computers will probably be co-processors as the normal logic based programs wouldn't run on them.
    • Re:In the future! (Score:5, Informative)

      by DrWho520 (655973) on Monday August 11, 2003 @07:58AM (#6664681) Journal
      How much about a standard computer must you know to run anything today? If you are coding in assembly, admittedly you must have an understanding, but do you need to know the base logic of the computer? Do you really have to understand how a transistor or a JK-flip-flop works to write C, C++ or Java? You certainly do not to build a machine. Its just a bunch of black boxes.

      The true power of quantum computing is the idea of a mixed state, the shades of gray if you will, that will be possible with quantum elements. While logic gates take strictly binary inputs (bits), quantum gates will take superpositions of the 1 and 0 states (qubits). Ask a simple question, is it cloudy outside? A bit either says yes or know depending upon a threshold of some sort. Who sets the threshold, does everyone agree on the threshold, and how accurately is the threshold mesured? A qubit can give you a mixture of yes and know, relaxing the systems. Its very similar to fuzzy sets, as elements are not strictly in or out of a set.

      There will be a learning curve. Unfortunately, until there are a large number of gates of a specific type, a deffinitive logic process (fuzzy logic, if you will) cannot be decided upon. (Maybe there will be serveral types, and Intel works with one type of qubit logic and AMD works with a different.) But the logic system is what you will need to understand, that is what people understand now. Is it really a simple process to break down everything into yes or know? You don't need a PhD. for that. I think the fuzziness of a quantum system is much closer to reality than that of binary.

      I picked up my PhD. (Posthole Digger) at the hardware store.
      • Re:In the future! (Score:1, Insightful)

        by ikkonoishi (674762)

        Maybe there will be serveral types, and Intel works with one type of qubit logic and AMD works with a different.

        Grr..
        Non standardized interfaces make Hulk angry!
        Hulk smash puny op-codes

        Seriouly though I think we should have some people theorizing the needed logic models at the same time as they develop the actual hardware components. Otherwise we will end up with >10e30 different implementations which set computing back decades.

        Just my 1.29 cents.

      • Let's see if i got it right:

        The power of quantum computers is the ability to "consider" all combinations of qbit status at once. The more qbits you have the more combinations (quantum status) you can have for the whole system. Reading the status of the qbits will (with a certain probability) make the whole system settle in the most stable status.

        To solve a problem in a quantum computer you have to set up the initial status of all qbits so as to maximize the probability that by feeding energy into the syst
      • The great thing about today's machines is that you don't have to know how they work to use them. But I remember a story where some scientists were able to factor 15 into 3 and 5 using a (7 qubit? Can't remember) quantum machine that looked rather complicated.

        I know one day quantum computers will shrink, and even my watch may be powered by a tiny quantum chip, or green goo, whatever it is.

        My main concern is that by the time standard semiconductors reach their limit the quantum machines are still in the sta
      • Was I the only one annoyed by the use of "know" instead of "no"?
  • Quantum logical states and governators to rule them all.
  • Excitons.. (Score:4, Funny)

    by mumblestheclown (569987) on Monday August 11, 2003 @07:40AM (#6664562)

    Excitons - the particle formed by the collision of Porntons and GNUtrons.

  • Although it will not be possible to scale up the system, the group says that many of the ideas and techniques they have developed could be useful in other approaches to quantum computing based on the optical control of electron-spin qubits in quantum dots.

    So ... maybe this is useful ... maybe not. Not even the researchers know. Don't read too much into this "discovery". :)
  • If they can harness the amazing power of the Gravitron [italintl.com], then I and Joe Sixpack will stand up and take notice.
  • by GedLandsEnd (537573) on Monday August 11, 2003 @07:47AM (#6664601)
    ...were excited by light to create a quantum logical gate with four states.

    Quantum computer? Great, it's in all four states at once. Gotta love a computer that gives you infinitely different results depending on what universe you're in.

    (First application: generating airfares.)
  • GigaQuads! (Score:2, Funny)

    by AnotherBrian (319405)
    And electrons trapped in the middle layer were excited by light to create a quantum logical gate with four states.

    Sweet! Score one for the Star Trek universe. Suck it Star Wars.
  • There is no gain. This is not a transistor in a conventional understanding (nice inverter curve). When you need to make a memory chip out of this, you better hope it can drive more than just one more transistor. This is the problem with all QC proposed now.
  • by Savatte (111615) on Monday August 11, 2003 @07:52AM (#6664645) Homepage Journal
    Excitons to be Bohring.

    ok, i'll kick my own ass for that one.
  • Not Gate? (Score:2, Insightful)

    by NoMercy (105420)
    Call me when they make NAND gates with a way to couple them together, then I'll get excited.
    • I guess one could make a bistable element by cross-connecting two of these. Still, it would be a lot more interesting if they can at least come up with some kind of 2-input gate. Then they probably would have to worry about fan-out and fan-in.

      Somehow, I don't think they're there yet.

  • License? (Score:4, Funny)

    by borgdows (599861) on Monday August 11, 2003 @08:12AM (#6664752)
    Since a Quantum computer can process millions of operations *simultaneously*, how much do I need to pay a million-CPU licence from SCO if I want to use Linux ?
  • From the article:
    "The researchers showed that they can drive Rabi oscillations ..."

    Sounds like a bad joke:
    How do you drive a Rabbi into excited oscillations?

    Oh, wait... that's 'Rabi' not 'Rabbi'.

  • Man, I can't WAIT to put together a beowulf cluster of these!

    please dont hurt me!

  • This summary contains more details.

    No thanks, this will last me a while.

  • Same reason I hope the speed of light isn't broken. There should be room for cleverness in the universe. Being able to compute anything and be anywhere at once just makes things boring.
    • You're nuts. Quantum computing wont let you 'compute anything' just a lot more, possibly. It will open up doors in research areas that have been closed due to lack of computing power. Better models could be created in the medical, environmental, energy... etc. By your logic, we should stop the progress on CPU development outright.
      • If it can be done, you have to do it. Otherwise you'll get killed in war. I'm just sort of hoping that the universe won't allow it. Better computing is a good thing, but I hear the most ardent supporters and technophiles arguing that all encryption could be broken. Just hoping those types are wrong.
    • The speed of light 'c' is quite safe, I wouldn't worry about that just yet. I've taken an introductory (4th year) quantum computing course, and one thing the profs kept mentioning was possible algorithms (based on quantum teleportation) to communicate faster than c, and how all these algorithms failed. There are certain effects in quantum mechanics that seem faster than c, but there is no way to transmit information faster than c.

      There is definately information transmited when a person moves from one place

  • by Sherloqq (577391) on Monday August 11, 2003 @09:14AM (#6665173)
    After all, they say they're using "two electron-hole pairs", so doesn't that translate into 2 two-bit "registers" instead of a single four-bit one? End result is the same, mind you, but the method is different.
  • Wow, I never knew quantum computers would help keep flooded networks afloat!



    See, qubit~=cubit.

    See, like Noah's ark.

    Oh, never mind.
  • choices, choices (Score:1, Flamebait)

    by Darth_brooks (180756)
    These physicists from the University of Michigan and other labs made a quantum dot by using a thin gallium arsenide layer stuck between two aluminium gallium arsenide barriers.

    A major advance in quantum computing is nice and all. But, would it kill U of M to get another college football (real / american) national championship? I mean, Ohio State's got a shiney new one. Ours is from 1997.

    Sigh. Such is the mentality. (thankfully it's not a prevalent at U of M as it is at other universities)
  • I thought the benefit of Quantum computing was that you could utilize Quantum interactions to make calculations and deciphering instantaneous. if all they are doing is replacing silicone with particles, does this achieve the goal?
  • Just a thought here... if you use Quantum Bits to make a standard NOT switch, then it seems to me that the NOT switch is going to behave as a NOT switch, and nothing is going to be any better than a CMOS NOT gate.

    To me, the advantage of going quantum is when you multiplex zillions of gates together to set up (if you will) a programmable analog computer that can calculate potential possibilities based upon your model.

    To do that, it seems to me that we shouldn't be thinking "...duh, how do I program a not g
    • What they've done is not a NOT gate but rather a CNOT (Controlled-Not) gate (in fact they haven't donw that either, they performed a CROT, a controlled rotation but you can show that one is as useful as the other). The interesting thing about the quantum gates is that once you are able to construct a few of them (such as a CNOT and single qubit operations) you can approximate ANY quantum operation to arbitrary precision in principle (and polynomially in the number of gates if I recall correctly) by just cha
  • by SyFryer (173279)
    I'll have to let Megatron know that the decepticons plans have been thwarted again!
  • by Vinnie_333 (575483) on Monday August 11, 2003 @11:31AM (#6666583)
    That's great. I suppose I'll hold off on getting that 3.0 GHz machine I was looking at, then. How big is the hard drive gonna be?
  • i liked the dinobots the best

    and the insecticons

    maybe we should let the department of energy know about enerjon cubes they seem to be renewable and made out of everything easily much better than fuel cells...

  • "And electrons trapped in the middle layer were excited by light to create a quantum logical gate with four states."

    Yeah, and those four states are Mississippi, Utah, New Jersey and North Dakota. Which make this story weird, because I can't even see an electron getting excited about being in one of those four states.

    (Yeah, I know, someone had to say it)
  • Does this mean that the spammers are going to be able to send out e'mail to every computer on the planet at the same time?? ....I'm gonna need more bandwith....
  • Hi, my name is Sam Beckett. You guys should have been at this point years ago. I wonder if... oh boy!
  • You realize they're still going to try to use the gigahertz rating right?

Chemist who falls in acid is absorbed in work.

Working...