Professor Gerard Milburn: turning quantum sin into a virtue (Photograph: Don Thompson)

Guest writer: Professor Paul Davies has been described as Australia's best-known scientist and was nominated as one of Australia's 10 most creative people by The Bulletin in 1996. He currently holds the positions of Honorary Professor in The University of Queensland's School of Physical Sciences and Visiting Professor at Imperial College, London. Professor Davies has published more than 100 research papers in specialist journals, in the fields of cosmology, gravitation, and quantum-field theory, with particular emphasis on black holes and the origin of the Universe. Well-known as an author, broadcaster and public lecturer, he has written more than 25 books including The Fifth Miracle about the origin of life. His latest book is the self-explanatory How To Build A Time Machine. Research team: Professor Gerard Milburn (Node Director), Associate Professor Michael Nielsen (Program Manager, quantum information theory), Dr His-Sheng Goan (Research Associate, condensed matter theory), Dr Mark Fernee (experimental quantum dot spectroscopy) Other UQ participants: Professor Halina Rubinsztein-Dunlop (Program Manager, experimental laser physics, HOD physics), Dr Tim Ralph (Program Manager, quantum optics ARC QEII ), Dr Andrew White (experimental quantum optics, lecturer UQ), Dr He Bi Sun (theoretical condensed matter physics, ARC APD), Dr Ross McKenzie (theoretical condensed matter physics, lecturer, UQ), Dr Alexi Gilchrist (theoretical quantum optics, postdoctoral fellow) plus 15 PhD students in the centre at UQ. External collaborators: Dr John Dowling (Quantum Technology Group, NASA Jet Propulsion Laboratory, Pasadena USA), Dr Daniel James (Los Alamos National Laboratory, USA), Professor Michael Roukes (California Institute of Technology, Pasadena, USA), Dr Crispin Barnes (Cavendish Laboratory, University of Cambridge, UK) Funding: Total funding to all three nodes of the SRC is about $20 million over three years. For UQ alone: 2001 ARC ($200,000 from $1.3 million to the SRC) 2001?2002 Federal Government accelerated support ($1.6 million to UQ from $9 million) University of Queensland ($300,000 per year) US Government (National Security Agency) ($2 million for three years) Email/Web link: milburn@physics.uq.edu.au www.physics.uq.edu.au

People who grew up amid the information revolution often take it for granted that computing power will go on accelerating. However, the days of the incredible shrinking microchip are numbered.

There are fundamental limitations set by the laws of physics on how much circuitry can be packed into a dwindling volume of material.

At some stage the components will get so small that atomic effects start to intrude.

At this point, the weird and wonderful world of quantum physics is unleashed, with all its awesome consequences.

Quantum mechanics is founded on the celebrated uncertainty principle of Werner Heisenberg.

It states that there is an unavoidable vagueness, or indeterminism, in the behaviour of matter on the micro-scale. For example, today an atom in a certain state may do such-and-such, tomorrow an identical atom could do something completely different.

According to the uncertainty principle, it is generally impossible to know in advance what will actually happen ? only the betting odds of the various alternatives can be given.

Essentially, Nature is reduced to a game of chance.

Since reliability is the name of the game when it comes to computation, quantum uncertainty would seem to pose a major threat to the continued miniaturisation of electronic components.

A computer that gives an answer x today and y tomorrow is pretty useless. So does this spell the end of the information bonanza?

Not so, claim a handful of physicists dedicated to turning a quantum sin into a virtue.

Far from stymieing information-processing, they say, quantum physics might actually be harnessed to improve it.

Thus emerges the promise of the quantum computer. The concept of using quantum effects to compute dates from a speculation made in the early 1980s by the late Professor Richard Feynman of the California Institute of Technology, and subsequently elaborated by David Deutsch of the University of Oxford.

It hinges on a central property of all quantum systems known as superposition. In everyday life, physical objects exist in well-defined states, even when there is an element of uncertainty in the outcome.

For example, a tossed coin will come down either heads or tails, though we may not be sure of which ahead of time.

A quantum coin does not behave like this.

It can exist in a curious hybrid, or "superposition", a state in which both heads and tails are somehow present together as ghostly half-realities. How does this relate to computation?

Imagine a row of coins, each of which can be in one of two states: either heads or tails facing up.

The coins could be used to represent a binary number, with 0 for heads and 1 for tails.

Two coins can exist in four possible states: heads?heads, heads?tails, tails?heads and tails?tails, corresponding to the numbers 00, 01, 10 and 11.

Similarly three coins can have eight configurations, four can have 16 and so on. Notice how the number of combinations escalates as more coins are considered.

Now imagine that instead of coins we have electrons, each of which can exist in one of two states.

This is close to the truth, because when electrons are placed in a magnetic field they do indeed adopt only two configurations: parallel or anti-parallel to the field.

Quantum mechanics allows a superposition of all possible such "heads/tails" (or "up/down") alternatives.

With even a handful of electrons, the number of alternative combinations making up the superposition is very large. We now reach the key step in the argument.

Each up-or-down configuration for an electron can be used to encode a bit of information. By letting the state of the system evolve, this information can be processed ? a computation can be performed.

But because all possible up-or-down combinations can co-exist in a quantum superposition, they can all be evolved together.

In effect, a quantum superposition allows for massive parallel computation, resulting in an exponential increase in processing power.

A quantum computer with only 300 electrons, for example, would have more components in its superposition than all the atoms in the observable Universe!

Unfortunately, most superpositions are rapidly destroyed by the influence of the environment, a process known as "de-coherence".

So far physicists have been able to attain stable quantum computational states involving only three or four particles at a time, but researchers in several countries are hastily devising subtle ways to improve on this. Fascination with quantum computation is motivated by more than a curiosity to see if the idea works.

If we had such a machine at our disposal, it could perform tasks that no ordinary computer could accomplish.

A key example is cryptography.

Many government departments, military institutions and businesses encode their messages by multiplying prime numbers (a prime number is one that cannot be divided by any other whole number except one).

Multiplying two primes is relatively easy. Most people could quickly work out that, say, 137 x 291 = 39,867.

But going backwards is much harder.

Given 39,867 and asked to find the prime factors, it could take a lot of trial and error before you hit on 137 and 291.

Even a computer finds the reverse process hard, and if the two prime numbers have 100 digits, the task is effectively impossible even for a supercomputer.

But in 1995, Dr Peter Shor, now at AT&T Labs in New Jersey, demonstrated that a quantum computer could make short work of the arduous task of factorising large prime numbers.

Build one, and you could crack top-secret codes immediately.

It is too soon to know how effectively quantum computers will be able to short-circuit these sorts of search problems in general, but the expectation is that they will lead to a breathtaking increase in speed.

At least some problems that would take a conventional super-computer longer than the age of the Universe should be solvable on a quantum computer in next to no time.

The practical consequences of this awesome computational power have scarcely been glimpsed, but some commentators predict that quantum computers herald a revolution at least as profound as that triggered by the inception of the original electronic computer.

The University of Queensland is at the forefront of this exciting and potentially Earth-shaking development, with the establishment of a node of the new Australian Research Council (ARC) Special Research Centre for Quantum Computation. The node is situated in UQ's School of Physical Sciences.

The other two nodes are at the University of New South Wales (UNSW) and the University of Melbourne.

Total funding for the centre is $20 million over three years including the $4 million ARC grant. The centre's Deputy Director, and Head of the UQ node, Professor Gerard Milburn, is internationally recognised for his work on quantum technology in general, and quantum optics in particular.

Indeed, the best-selling science fiction author Michael Crichton recently drew upon Professor Milburn's ideas for his latest blockbuster novel Timeline.

UQ is collaborating with the Universities of New South Wales and Melbourne in the race to build the first fully functioning machine.

One design that shows special promise is based on embedding phosphorus atoms in a silicon crystal, using the orientation of the phosphorus nuclei as the quantum equivalent of heads and tails.

The centre's Director, Professor Robert Clark of UNSW, is busy fabricating this structure, while Professor Milburn and his colleagues carry out the theoretical analyses.

The Centre for Quantum Computation has a broader brief, however, than simply building a prototype quantum computer.

Deep conceptual issues have emerged, concerning such things as the very nature of information, the theory of computation and even the relationship between quantum physics and life.

Information-processing at the quantum level amounts to more than a faster way of moving bits around; it goes to the very heart of what we mean by information and computation, and their relationship with physical objects such as atoms.

The centre has a total of 60 staff and 20 students tackling a broad range of conceptual and practical problems in this burgeoning field.

And they have already achieved an early success with a "bottom-up" strategy for building an array of phosphorous in silicon with atomic precision.

The UQ node also recently established an experimental program to use linear optics and single photonics, based on a discovery made by Dr Manny Knill and Professor Raymond Laflamme at Los Alamos National Laboratory in collaboration with Professor Milburn, and recently published in Nature.

This approach to quantum computation would neatly complement current developments in optical communication systems.

Quantum computation promises to be one of the hottest topics in fundamental physical science, with the potential for far-reaching practical applications.

It is one in which The University of Queensland is a major stake-holder, and the Centre for Quantum Computation has acquired an international reputation for research in this exciting field.

On this site

Go to top