Episode 7      22 min 59 sec
A Quantum Leap in Computing

Professor David Jamieson discusses Quantum Computing, the new frontier in computer design, with Science host, Dr Shane Huntington.

Guest: Professor David Jamieson, Director of the Melbourne node of the Australian Research Council Centre of Excellence for Quantum Computer Technology

Topic: A Quantum Leap in Computing

"A standard computer memory consisting of strings of ones and zeros can only store one piece of information at a time. A quantum memory with its quantum bits or qbits could store simultaneously a large number of different numbers and then process them all simultaneously using the laws of quantum mechanics." - Professor David Jamieson




           



Professor David Jamieson
Professor David Jamieson

Professor David Jamieson, Director of the Melbourne node of the Australian Research Council Centre of Excellence for Quantum Computer Technology.

Credits

Host: Dr Shane Huntington
Producers: Kelvin Param and Eric Van Bemmel
Audio Engineer: Craig McArthur 
Theme Music performed by Sergio Ercole. Mr Ercole is represented by the Musicians' Agency, Faculty of Music
Voiceover: Paul Richiardi

Series Creators: Eric Van Bemmel and Kelvin Param

Melbourne University Up Close is brought to you by the Marketing and Communications Division in association with Asia Institute, and the Melbourne Research Office.

View Tags  click a tag to find other episodes associated with it.

Download file Download mp3 (22.1 MB)

A Quantum Leap in Computing

VOICEOVER
Welcome to Melbourne University Up Close, a fortnightly podcast of research, personalities and cultural offerings of the University of Melbourne, Australia. Up Close is available on the web at upclose.unimelb.edu.au.au, that!|s upclose.u-n-i-m-e-l-b.edu.au.

SHANE HUNTINGTON
Hello and welcome to Up Close coming to you from Melbourne University, Australia. I!|m Dr Shane Huntington and today!|s topic is quantum computing. I am happily joined by Prof David Jamieson, from the School of Physics, University of Melbourne. Prof Jamieson is Director of the Melbourne node of the Australian Research Council Centre of Excellence for Quantum Computer Technology. Welcome David.

DAVID JAMIESON
Thank you, Shane.

SHANE HUNTINGTON
Now, first of all, let me ask you a little bit about your background. How did you get involved in physics?

DAVID JAMIESON
Well, that goes back a very long way. I!|ve always had a great interest in science. And, my formative years were during the boom in the space program. And it seemed like every year there was a new milestone. A new breakthrough. The Gemini Program and then the Apollo Program. This really developed my interest in physics and science in general. I was always interested in science, but this was just a fantastic time to be growing up. In 1969, getting sent home from school to watch the moon landing !V and of course, I was very young in those days, but, [I was] an authority for my peers for what was going on. This just took off from there. But I always interested in science. I can!|t ever remember a time when I wasn!|t.

SHANE HUNTINGTON
Excellent. Now, we are going to talk to you today about quantum computers, so what we should probably do first is give our listeners an idea of how today!|s computers or, the computers of the last 50 years actually work, so, give us a run down on the current model.

DAVID JAMIESON
Well, actually, we can go back more than 50 years because the current models use principles that were introduced thousands of years ago. In fact, modern Pentium Duo 2 Core chip uses the same principles as a Chinese abacus that has been around for a couple of thousand of years. Principles of classical physics, which is the physics of the everyday world.
The idea, as you know, in an abacus is you!|ve got beads on a wire. And by moving the bead from one side of the frame to the other, you represent a number: zero or one or zero and five, depending on what part of the abacus the bead is located. So, that is basically a binary computer. By having a bead or a voltage level in an integrated circuit, it is the same principle. It is either up or down, zero or one, on or off, it is binary logic, binary mathematics. And that is how all classical computers work: just by representing information in strings of zeros and ones.

SHANE HUNTINGTON
Now that sounds good and I think everyone has the experience of buying a new computer and for a few weeks, it seems to be a little faster, but then we all realise, sooner or later, that the computer we have is pretty much as fast as the one we had 10 years ago. What!|s the problem and the limitation there? What!|s going on?

DAVID JAMIESON
That!|s a strange paradox, isn!|t it? I think our software engineers have got a lot to answer for in that. The phenomenon of code-bloat. So, when you your processor and your memory is getting more and more powerful, people immediately write software to take advantage of that new power and of course, a lot of the fancy graphics and helpful little hints that pop up take up a lot of memory, a lot of processor power. And I think that is why it doesn!|t appear to be any faster. But, I assure you it is. Under the bonnet of your Pentium, under the hood for our North American listeners, of a modern computer you find a very, very fast computer chip indeed. And I have to say, these up-to-date personal computer chips with over a 100 million transistors in them, I don!|t think any human has ever owned a 100 million artificial gadgets before and that is really a remarkable accomplishment that you can make that available to the average person. But those 100 million transistors are really busy, drawing all those fancy graphics on the screen for you and trying to do it as quickly as possible.

SHANE HUNTINGTON
Now, we keep seeing, every year new computers coming out. Can we keep doing this? Is there a limitation to what we can build?

DAVID JAMIESON
Well, that is a very good question. We are not 100 per cent sure whether this remorseless pace of improvement can continue indefinitely. This is codified by something called Moore!|s Law which I am sure most people listening to this will be familiar with. But it is the idea that, the computer power or the memory capacity doubles every 18 months. And this has been going on now for more than 40 years, which is an amazing pace of change [which is] unmatched by any other sort of technology, I should say. But this remarkable pace of improvement, has been accomplished by making the individual components ever smaller. So, the transistors in your modern Pentium computer chip, the individual components that make it work, may be as small as 90 billionths of a meter in size. Now, when you get that small, the ordinary laws of physics that we are familiar with, our common sense, everyday life type laws of physics break down because now we are entering the quantum realm. And these very, very tiny components, which need to be made ever tinier to improve the power and speed of the computer will eventually enter the quantum realm and they!|ll stop working in the way that we expect them to work. They!|ll work in new ways, that we can perhaps exploit, but certainly the conventional way of doing business with beads on your abacus or voltage levels in a circuit being on or off or up or down will not become valid anymore.

SHANE HUNTINGTON
You!|re listening to Melbourne University Up Close. I!|m Dr Shane Huntington and we!|re speaking with Prof David Jamieson about quantum computing.
David, so tell us now a little about the quantum computer. This is the new realm we are moving into. How is this different and how will this work?

DAVID JAMIESON
If you imagine your abacus again, [it is] a lot easier to imagine an abacus than a CMOS flip flop in an integrated circuit in your Pentium chip !V but the principles are the same, as I have explained. Your bead in the abacus can either be left or right or up or down on the wire, but in the quantum world, a quantum bead doesn!|t have to obey those laws. And in fact it doesn!|t obey those laws. So, you can imagine grabbing he wire of your abacus that!|s got the beads on it, pulling it out of the frame and now allowing the wire to occupy any orientation in space it likes. So, the beads could not only be not only up and down, they could be left and right, in or out, backwards or forwards in their orientation in space. Now, that is what a quantum bead could do. But there is one further thing that is very important. The quantum bead can be up and down !V simultaneously. A quantum particle can be in two places at the same time, or in two states at the same time. Now, I can!|t actually give you a clear picture of how this works, because nobody knows why this happens or how this happens, it is just the attribute a quantum particle has. So this give us an opportunity to exploit this very strange quantum attribute to make a very powerful method of processing information. A quantum memory consisting of quantum particles or quantum beads could store an enormous amount of information. A standard computer memory consisting of strings of ones and zeros can only store one piece of information at a time. A quantum memory with its quantum bits or qbits could store simultaneously a large number of different numbers and then process them all simultaneously using the laws of quantum mechanics.

SHANE HUNTINGTON
Amazing. Now, we know what are current computers look like and what we know we need next year !V the faster computer that can run the new software !V what will we need in terms of a quantum computer? Is this the sort of thing that only governments would own or do we all want to have one? What!|s it going to look like?

DAVID JAMIESON
Well, it is not clear what a useful quantum computer is going to look like because that is still a matter of active research. There are a whole lot of different architectures that people are working on at the moment. One thing is for sure that it is not going to look too much like a standard computer, because the quantum computer is not going to be a general purpose computer in the same way that a personal computer chip could run a word processing package or do some advanced mathematical modelling so you can figure out what your tax liability is this year. A quantum computer will do several specialised tasks extremely well. It will do them so quickly !V in a matter of minutes !V yet, a classical computer, a Pentium chip, for example, might take millions of years to do the same task. The quantum computer offers a short cut to some specialised problems using laws of quantum mechanics.
Now, as for what it might look like physically, it might look like a refrigerator. A very good refrigerator, operating at extremely low temperatures !V only a few thousandths of a degree above absolute zero. That!|s about minus 273 degrees Celsius. Because quantum mechanics is actually very delicate and the delicate quantum states that exist in the memory of the quantum computer can be easily destroyed by an external perturbation !V that is to say, a little bit of heat getting in there !V!¢FDwhoosh!| !V wipes your quantum memory. Causes it to collapse into its classical values. That is a bit of a downside on these quantum computers. Using quantum mechanics to process information, you!|ve got to isolate the computer memory and the computer processor from the outside world while it is doing the processing and then only at the end, when it is finished should you look at it to read the output. This is a strange paradox of quantum mechanics: you can only exhibit quantum behaviour, when no one is looking.

SHANE HUNTINGTON
Now, just before we focus in on your particular efforts down in Melbourne, let!|s talk about Australia more globally. What efforts are going on in Australia towards, producing a quantum computer?

DAVID JAMIESON
There!|s a lot of activity in Australia, as the rest of the world on this relatively new field of quantum information and quantum computing. It is a field that has really expanded in the last decade from a very low base. Although I should say that Richard Feynman !V the late, great theoretical physicist, Nobel Laureate, predicted the quantum computing revolution in a very visionary speech that he gave back in the 1950s, where he suggested using quantum states on atoms to process information. But the technology was not available to make it a reality in those days. It only became available in the last decade.
But here in Australia, there is most activity on two different architectures. Two different ways of using quantum states to process information. The first is to use light. Light is a terrific quantum entity. Photons, the quantum particles of light, dance to your command and exhibit all sorts of wonderful quantum phenomena. The only difficulty with light is making individual quantum particles talk to each other. You have to do that through a third medium, usually a lump of matter. But you could use light itself in your quantum processor and there is a big activity in that area, mainly up at the University of Queensland and its collaborators around the country.
The other way of building a quantum computer that we are very keen on here in Australia is to use the old, standard silicon. Silicon, of course, is the basis of our Pentium chips and most computing today !V although there are few chips that use an alternative material called gallium arsenide. But if we could build a quantum computer in silicon, we could immediately adopt all of the technology that has been developed to make these very tiny !V 90 billionth of a meter !V sized transistors that we find in our modern computers. Some of that technology would map across and be useful for building quantum computer chips based in silicon. So it is a very attractive option !V even though it is a very challenging technical problem of how to engineer quantum states into silicon chips. But, we are working on that, here in Melbourne in close collaboration with our colleagues at the University of New South Wales and other laboratories around the country.

SHANE HUNTINGTON
Let focus in now, on Melbourne and your group.

DAVID JAMIESON
Okay, here in Melbourne we have a fairly sizeable group working on two aspects of the problem. The first is the theory. How exactly would we use quantum mechanics to encode and process information? We!|ve got the big picture. We!|ve had that for more than a decade. But we really need to know this down at the fine detail. It is one thing to have a big picture, it!|s another thing to build a working device. So, we!|ve got a large theoretical team, headed up by Professor Lloyd Hollenberg, who are working on the fundamental physics of how you would use quantum mechanics to encode and process information. Quantum mechanics, of course, is one of the most successful physical theories of all time. It accounts for just about everything we see around us. But the quantum computer revolution that is coming has caused us to reappraise these fundamental theories and apply them at the cutting edge of the cold face in quantum computing.
The second major activity, here in Melbourne, involves building the device itself in silicon. In close collaboration with our co-workers at the University of New South Wales, our job here in Melbourne, is to insert single atoms into blocks of ultra pure silicon. And when I say, !¢FDsingle atoms!|, I mean one-point-zero atoms. Not one, plus or minus one !V but exactly one. And verifying that we!|ve got the atom in there and that it is ready to encode quantum states for quantum computing applications.

SHANE HUNTINGTON
This seems like a momentus task, does Australia, and your group, and the colleagues you speak of, have significant international links in order to achieve these goals, or are we going it alone?

DAVID JAMIESON
We do have significant international links. But the problem of doing this in silicon is so difficult, we are one of the few groups who have developed the advanced technology to go down that path. And we!|ve had numerous visits from authorities in the field who have commented that the scale of our operation and the somewhat unusual methods we are using to solve this problem are unique in the world and we have some leadership in this aspect of the technology of engineering silicon with single atoms.

SHANE HUNTINGTON
You!|re listening to Melbourne University Up Close. I!|m Dr Shane Huntington and we!|re speaking with Prof David Jamieson about quantum computing.
David, tell us about some of the interesting goals that have been achieved by your team and your colleagues in the last few years.

DAVID JAMIESON
Okay, we had a major breakthrough a few years ago when we figured out a method of putting the single atoms in as I described. We!|ve done this by adapting a standard technique that has already used for modifying silicon in the silicon chip industry, which is a technique called ion implantation. It is a very simple idea. You have a block of pure silicon and you want to introduce a trace element to give it the desired properties for making a computer or any other sort of silicon chip. You take the impurity atom you want to put in there, you accelerate it up, in a particle accelerator, and you embed it in the silicon. We!|ve developed a way of actually counting the atoms as they arrive, one by one. We do that by actually making the silicon chip itself a very sensitive ion impact detector and we wire that up to a sensitive bank of electronics which gives a click every time an atom arrives at the desired place in the silicon. And in the devices we are making at the moment, we count two clicks, and then we turn everything off. There it is. The device is finished. Two atoms, safely installed. And then it goes back to our colleagues in Sydney at the University of New South Wales for further processing and doing the delicate quantum measurements. Seeking the delicate quantum states that we are going to use to encode the information.

SHANE HUNTINGTON
Sounds like you are making some fabulous progress. Are there alternative mechanisms for working with silicon internationally that we are competing with in Australia?

DAVID JAMIESON
Not using the techniques we are developing here. There are alternative ways of using silicon in possible quantum architectures. There are also ways of using superconducting materials for building quantum chips, quantum computer devices. And there are also, of course, our friends in the gallium arsenide community !V I mentioned gallium arsenide earlier !V they have made some astonishing progress. Particularly, a group at Harvard in using some of the unique properties of gallium arsenide to build devices where electrons are already displaying the required quantum behaviour which is very, very exciting indeed. And, if they can do it in gallium arsenide, then we should be able to do it in silicon.

SHANE HUNTINGTON
Now, has anyone actually managd to make a quantum computer, I guess, even a simple version as yet, anywhere in the world?

DAVID JAMIESON
Well, they certainly have. In fact the IBM laboratory, made a quantum computer with seven quantum bits in it, quite a few years ago now. It didn!|t actually look very much like a computer. It was a big tank of special liquid !V a special hydrocarbon, cooled down to very low temperatures and then put in something very much like a magnetic resonance imaging system in a hospital. And this was able to demonstrate the required quantum behaviour. And they ran a simple algorithm on this computer that found the prime number factors of the number 15. And I!|m pleased to say that it found that they were three and five. But it did it, using the quantum properties of the device and not the ordinary, classical properties. So that showed that it was viable and it does work.

SHANE HUNTINGTON
Now David, this is an incredibly exciting area, some of listeners, no doubt, will be looking to find more information on quantum computing. Do you have any advice on where they should look?

DAVID JAMIESON
Well, that depends on how technical you want to get. There is a very nice book called Introduction to Quantum Computers [actual title: Quantum Computation and Quantum Information] by Michael Nielson and Isaac Chuang which has got all the technical details you could want. The full quantum mechanical treatment. It runs to about 600 pages. Very authoritive overview of the field. But there are also numerous articles in the popular literature. There is a nice book that was published a few years ago, called A Shortcut Through Time by George Johnson !V a journalist, I think !V which gives a nice overview to the field. And of course, one of our colleagues, from the University of Queensland, Gerard Millburn, has also written a nice popular book called The Feynman Processor !V harking back to Feynman prophetic speech of the 1950s with a popular introduction to quantum computing.

SHANE HUNTINGTON
Now, what other interests in your research do you have?

DAVID JAMIESON
Okay, well, that is a good question. I run a fairly large group here in the School of Physics at the University of Melbourne and I have a number of different research programs underway in my laboratory that I!|m very interested in. I!|ve mentioned very major work on silicon, but in parallel with that project, I!|m participating in another project which is almost as large, which involves, diamond. Now that is very interesting. If you recall column four of the periodic table, immediately above silicon, is carbon. So, single crystal carbon has a lot of similar attributes to single crystal silicon, and of course, single crystal silicon, is the basis of the silicon industry. And so, we are looking at ways of using single crystal carbon, which is of course, diamond, as an alternative way of building a quantum processor. Now diamond has a lot of fascinating properties !V it is an extreme material in all respects, but it turns out that if you introduce trace elements of nitrogen and other elements into diamond crystals, they take on very intersting quantum attributes. Diamond is a very difficult material to work with. What would you use to cut it with, for example? You can!|t use a blade. You have to use a laser to burn your way through it because it is the hardest known material. But if we can overcome those processing difficulties it offers some very promising possibilities.

SHANE HUNTINGTON
Prof David Jamieson, School of Physics, University of Melbourne. Thank you very much for joining us today on Up Close.

DAVID JAMIESON
It has been a pleasure, Shane.

SHANE HUNTINGTON
Melbourne University Up Close is brought to you by the Marketing and Communications Division in association with Asia Institute, of the University of Melbourne, Australia. Our producers for this episode were Kelvin Param and Eric van Bemmel. Audio recording by Craig McArthur. Theme music performed by Sergio Ercole. Melbourne University Up Close is created by Eric van Bemmel and Kelvin Param. I!|m Dr Shane Huntington, until next time, thank you for joining us, Goodbye.

VOICEOVER
You!|ve been listening to Melbourne University Up Close, a fortnightly podcast of research, personalities and cultural offerings of the University of Melbourne, Australia. Up Close is available on the web at upclose.unimelb.edu.au.au, that!|s upclose.u-n-i-m-e-l-b.edu.au. Copyright 2007, University of Melbourne.


show transcript | print transcript | download pdf