One obstacle to teleportation is at the atomic level; the Heisenberg
uncertainty principle says it’s impossible to measure all the properties of an
atom exactly—so you cannot replicate it elsewhere. But in 1992, Charles
Bennett of IBM Research in New York and his colleagues conceived of a way of
getting round this using “entangled particles” (This Week, 3 April 1993, p 12).
These are pairs of particles that carry information about each other because
they have interacted in the past.

The idea was that before teleportation, pairs of entangled particles would be
separated and sent to each location. Suppose a sender, Alice, then wanted to
teleport a particle to Bob. She would take an entangled particle and join it to
the particle to be teleported. Alice could scan the state of the joined
particles, and radio the information to Bob.

Bob can then use that information to convert his entangled particle into the
original state of the particle that Alice wanted to teleport. So Bob is left
with the only authentic copy. The Heisenberg uncertainty principle would still
not be violated, because none of the measurements reveals the exact state of a
particle.

Now, Michel Brune at the Ecole Normale Supérieure in Paris and his
colleagues have created entangled atoms for the first time (Physical Review
Letters, vol 79, p 1). To do this they used an array of lasers to tune a
beam of rubidium atoms so that each had a specially chosen speed and state of
excitation. As one atom reached a cavity consisting of two niobium mirrors,
another with a slightly different energy would follow in hot pursuit.

In the cavity, each pair of atoms exchanged a quantum of energy in the form
of a photon of light and became entangled. Brune’s team confirmed this by
measuring the energy states of the emerging atoms. By measuring one, they could
predict that of the second.

Until recently, researchers had only managed to entangle elementary particles
and photons of light. “By demonstrating quantum entanglement with atoms, the
French team has opened the way to much more interesting kinds of interaction,”
says Artur Ekert, a quantum physicist at Oxford University.

A real system for teleporting atoms, let alone large objects, is still far
away, however. “Our system only applies to a limited subsystem of an atom
because the atoms were in just one of two different states—I don’t know of
any system for teleporting the complete quantum state of an atom,” admits Brune.
“But we certainly hope to demonstrate some form of teleportation soon.”

The findings, presented at an international physics conference in Capri this
week, could also throw some light on another of cosmology’s greatest
puzzles—the nature of the Universe’s dark matter.

The results are based on an analysis of six years’ data from the Soudan 2
neutrino detector located in a disused iron mine in Northern Minnesota. The
detector, which consists of 1000 tonnes of corrugated iron, measured neutrinos
produced in the atmosphere by cosmic ray bombardments.

Neutrinos come in three types, all of which belong to the lepton family of
elementary particles. The electron, the muon and the tau are charged and the
neutrinos are their uncharged partners. Theory suggests that cosmic ray showers
should produce twice as many muon-type neutrinos as electron neutrinos. At
Soudan 2, however, researchers measured a ratio closer to 1 to
1—supporting Japanese and American experiments which measured neutrinos
using large volumes of water instead of iron.

Speaking on behalf of the fifty-strong team involved in the project, Hugh
Gallagher of Oxford University said that the results meant scientists could have
much greater confidence that the missing muon neutrinos amounted to a genuine
anomaly and were not an experimental artefact.

The best explanation for the anomaly is the idea of neutrino oscillations. If
each type of neutrino had a tiny but non-zero mass, quantum mechanics suggests
they could oscillate between different types
(
see Diagram). “Each neutrino would
consist of a mixture of matter waves,” says Wade Allison, head of the Oxford
team involved in Soudan 2. “Any difference in rest mass would give rise to
`beats’ between them,” he says, like the beats that occur when two waves closely
spaced in frequency combine to produce a noise that pulsates between loud and
soft. “The result would be that the nature of the mixture would change as the
waves propagate.” Depending on the precise masses, muon neutrinos could turn
into electron neutrinos as they travelled towards detectors on the ground. A
similar mechanism could also explain the embarrassing shortage of electron
neutrinos seen coming from the Sun.FIG-20882301.gif

“Once you have eliminated instrumental problems, neutrino oscillations
certainly become the favoured explanation,” says David Wark, an Oxford physicist
working on the solar neutrino problem.

Neutrino oscillations could also help explain the dark matter problem.
Theorists have long speculated that the invisible dark matter that makes up most
of the mass of the Universe could consist of neutrinos. There are at least 500
million neutrinos in every cubic metre of the Universe, so even if one type has
mass that is a small fraction of that of an electron, they could make a big
contribution.

DAVID DEUTSCH is renowned for his brilliance and for his eccentricities. A
virtual recluse, he prefers not to travel, sleeps by day and works through the
night. Yet face to face, he is one of the most engaging people you could hope to
meet. So much so that one scientist recently told me that a one-hour
conversation with Deutsch had changed his life. Reading this book might just
change yours.

The Fabric of Reality is a hugely ambitious book. Its enormous scope
is hinted at by the intriguing inscription: “Dedicated to the memory of Karl
Popper, Hugh Everett and Alan Turing and to Richard Dawkins. This book takes
their ideas seriously.” It explores so many of the big ideas in science and
philosophy that you might think Deutsch began by compiling a list of
eye-catching topics and then found a cunning way to weave them all together. The
theory of everything, quantum mechanics, virtual reality, scientific method,
evolution, the significance of life, quantum computation, the nature of
mathematics, time travel, the end of our Universe: all these and much else find
their place. What is remarkable is that all these subjects emerge naturally from
Deutsch’s arguments, though often with a surprising twist.

Deutsch is a theoretical physicist and a member of the Quantum Computation
and Cryptography Research Group at Oxford University. In 1985 he wrote a seminal
paper about quantum computers —a new breed of machine that could, in
principle, exploit the quantum mechanical properties of atoms to conjure up
hitherto undreamt of computational powers. His work eclipsed even the efforts of
the late Richard Feynman, and helped to establish quantum computing as an
important new research discipline. Since then he and his Oxford colleagues have
continued to make influential contributions to the field.

The driving force behind this book comes from his unshakeable belief in the
existence of parallel universes, an idea first proposed in 1957 by Everett of
Princeton University to explain the mysteries of quantum mechanics. The
many-universes idea appears to have waxed and waned in popularity over the years
but it still holds sway with a significant number of physicists. According to
this view, our Universe is embedded in an infinitely larger and more complex
structure called the multiverse, which to a good approximation can be regarded
as a system of parallel universes. There are endless copies of our Universe in
the multiverse, some slightly different, some completely different. Furthermore,
every time there is an event at the quantum level—a radioactive atom
decaying or a particle of light impinging on your retina—the Universe is
supposed to “split” or differentiate into different universes.

Not surprisingly, many people find the idea bewildering at first blush.
Deutsch’s account of why we should take the multiverse seriously appears in the
second chapter of this book. Simply titled “Shadows”, it is nothing less than a
masterstroke. Deutsch manages to show how the presence of nothing more than a
few shadows on a laboratory screen leads you inexorably to deduce the existence
of parallel universes. The argument is so clearly and economically
presented—and you don’t need to know anything about quantum theory to
understand it—that it is hard not to be convinced by the end of it. If you
are, you will begin to appreciate how the multiverse offers a completely new way
of looking at almost everything.

Take evolution, for example. According to the multiverse view, whenever an
organism’s DNA is struck by a cosmic ray and mutates, a large range of different
mutations appear in different universes. Deutsch invites us to imagine a magic
microscope that could see into other universes. How would our DNA look under
such a microscope?

According to Deutsch, so-called junk DNA would look completely random in
different universes whereas the conserved regions, those that serve a biological
function, would stand out as “cultivated fields stand out from a jungle in an
aerial photograph” or as “crystals that have precipitated from solution”. It’s
not that the useful genetic material would look the same in all universes, it is
just that the useful stuff would appear the same in closely related
universes.

Turning to information theory, Deutsch points out that it is very hard to
produce a satisfactory definition of knowledge that distinguishes it from random
noise. Unless, of course, you bring in our friend the multiverse. Using the
fictitious magic microscope, knowledge encoded in the form of bits of data in a
memory chip would stand out from random data like a crystal in all the
universes.

In quantum computing Deutsch offers the example of a machine that could
factorise a random 250-digit number that is the product of two large primes. At
the moment there is no prospect of any computer or supercomputer ever being able
to factorise such large numbers by classical means because it would take too
long. Yet a quantum computer could manage the task in an afternoon.

How so? As Deutsch argues, the multiverse offers by far the most intuitive
answer: the quantum calculation relies on the services of a vast number of
parallel universes. How else could one explain the phenomenon given that a
250-digit number would require around 10⁵⁰⁰ universes, a number far larger than
the number of atoms in the visible universe, which is 10⁸⁰? If the visible
universe were the extent of physical reality, says Deutsch, the resources
available would be nowhere near equal to the task.

The multiverse view can even be applied to down-to-earth things such as
weather forecasting. The conventional view of the weather is that it is a
chaotic system and is inherently unpredictable for periods of more than several
days because small errors in our knowledge blow up exponentially with time. Not
so, says Deutsch. The “parallel-universe multiplicity is the real reason for the
unpredictability of the weather”.

Deutsch seems to have an answer for everything. Time travel, for example,
probably will be possible one day in the future, he says, but you need not worry
about people going back and undoing the past. In the multiverse travelling back
in time takes you to a different universe so causing havoc there will not
disturb the universe you left behind.

There is so much else I could mention. But the central thesis of this book is
that we need to take into account four different strands of scientific knowledge
to understand physical reality: the theory of evolution, the theory of
knowledge, the theory of computation and quantum theory.

As Deutsch argues: “The four of them taken together form a coherent
explanatory structure that is so far-reaching, and has come to encompass so much
of our understanding of the world, that in my view it may already properly be
called the first real Theory of Everything.”

Deutsch is very persuasive in his arguments and his writing crackles with
originality. There will be many people, however, who find it hard to accept the
starting point of his conception: the multiverse in all its multifarious
magnificence. After reading this book, I found myself walking down a run-down
high street feeling dazed. Looking at the dreary faces around me, I wondered if
there could there be zillions of copies of these people doing slightly different
things in different universes. Is nature that extravagant? After all, that is
what the multiverse demands.

So what in the final analysis are we to make of the spectacle Deutsch holds
before us? It is almost as if he offers us poor blinkered flat-Earthers the
vision of a round world, but then adds that the interior is made of cheese. Part
of it makes sense, part seems crazy. Either way, this is an awesome book.

These researchers are not the first to have suggested ways of using the laws
of quantum mechanics to encode a digital message. But Klaus Mattle, Harald
Weinfurter and Anton Zeilinger of the University of Innsbruck, working
with Paul
Kwiat of the Los Alamos National Laboratory in New Mexico, have won the race to
bring these ideas to life.

The researchers encoded a message by manipulating the polarisation of light.
In classical physics, each photon of light can carry only one bit of
information
in its polarisation. To send a message, a physicist can send a beam of light
through a polarising filter which can be flipped between its horizontal and
vertical settings. The recipient of the message then passes the arriving light
through another filter that lets through only horizontally polarised
light. When
a photon emerges, the recipient registers a 1. Each gap in the signal is
registered in the form of a 0.

Mattle and his colleagues squeezed an extra level of information out of each
photon by passing ultraviolet light through a crystal with “nonlinear” optical
properties. This crystal converted every photon of light travelling through it
into a pair of photons, each with half the energy of the original photon.

The crystal also polarised the photons, but due to the uncertainties of the
quantum world, each photon emerged with an element of both horizontal and
vertical polarisation. Because the photons were produced in pairs, the
polarisation states of the two photons within each pair were linked: in quantum
terms, they were “entangled”.

To send a message, the physicists sent one photon within each pair directly
to an observer, while the other one reached the observer after passing
through a
device that could delay its horizontally polarised component to varying
extents.
As a result, each altered photon received by the observer was in one of four
possible states relative to its unaltered partner. In practice, however, two of
the altered photon states were indistinguishable, which meant that each altered
photon effectively encoded one of three states—the 0, 1 or 2 of a
trit.

To prove the point, the experimenters sent a message consisting of three
characters, “K”, “M” and “°”. In the ASCII format used by today’s
computers,
these characters are represented by strings of binary code containing
eight bits
of information each, giving a total of 24 bits. Using the quantum system,
however, the message needed only 15 trits (Physical Review
Letters, vol
76, p 4656). “It’s the first experiment which demonstrates a communication
scheme using pure quantum states,” says Weinfurter.

The researchers’ next plan is to refine their system to achieve quantum
“teleportation”. This scheme uses pairs of entangled photons to transmit
information about the state of a third photon. This information can then
be used
by an observer to recreate a photon in exactly the same state (This Week, 3
April 1993, p 12).

The work could also lead to the development of powerful computers that
exploit the uncertainties of quantum mechanics to conduct many calculations
simultaneously.

Carey may not be a scientist but he is a doyen of the literary world, a professor of English at the University of Oxford no less. So I guess that explains why this book has found its way into so many of the reviews sections of newspapers. It appears in columns that would not normally give the time of day to books about science.

The Faber Book of Science has been widely fĂȘted as bridging the gap between the two cultures. And so it does, up to a point. Despite Carey’s lack of scientific pretensions, his opening chapter on scientific writing is illuminating and enticing. What follows is his selection of science writings that move chronologically from the Renaissance to the present.

Beginning with Leonardo da Vinci, we encounter a veritable galaxy of scientific stars including Galileo, Newton, Priestley, Malthus, Faraday, Darwin, Huxley, Einstein, Feynman, Dawkins and Wolpert. And there are also contributions from literary luminaries such as Mark Twain, Primo Levi and John Updike, and from leading scientific popularisers such as Martin Gardner and Isaac Asimov. It is, without doubt, an impressive cast list.

On the downside, we are, inevitably, limited to snapshots of each of the author’s writings. In some cases, Carey has grouped different contributors thematically, to give the reader a glimpse of scientific developments, such as relativity and quantum theory, from several viewpoints. This works well as an aid to understanding but it puts further limits on our consumption of any particular writer.

The result is that some of the selections can seem disappointingly bitty. Albert Einstein’s contributions on relativity, for example, amount to little more than a few paragraphs or sentences broken up with Carey’s commentary. Richard Feynman, who Carey rightly singles out for particular praise, gets two uninterrupted pages but, although Carey has chosen a gem from Feynman’s autobiographical recollections, it ends in a slightly strange place. If I were reading Feynman for the first time here, I doubt whether I would have barely glimpsed his true brilliance as a storyteller and scientific populariser.

Nevertheless, the book does a valuable service in encompassing so many ideas and the people who wrote about them. As a book to pick up and dip into from time to time, this will be a compelling volume to artists and scientists alike. And it will make an excellent present.

Penrose’s thesis was that it was logically impossible for computers to
think like humans. Engaging as the book was, it was hard to believe that
Penrose had not overstepped the mark when he suggested that the human brain
somehow made use of quantum gravity to comprehend the world. The scales on
which quantum gravity operate are so incredibly small that it was difficult to
see how quantum gravitational effects could have any relevance to the
brain.

Sometime after The Emperor’s New Mind was published, I heard Penrose argue
his case at the British Association’s annual science festival. He was followed
by John Taylor, a mathematician and neural network specialist at King’s
College, London, who presented a much more compelling scenario for
understanding the brain in terms of neural circuitry. Taylor, it appeared,
demolished the very foundations of Penrose’s argument.

So now Penrose is back with Shadows of the Mind in which he has revised and
extended some of his original ideas and attempted to answer many of the
criticisms raised against him. Once again, this is a technically demanding
book involving quite a lot of mathematics.

The hardest and perhaps most off-putting material comes in the first half
of the book in which Penrose re-examines his original argument as to whether
computers could ever reason like mathematicians. It is Penrose’s claim that
the way mathematicians think is “non-algorithmic”, that is, their insights
could never be mimicked by a computer. He has a kind of formal proof that
involves a modified form of Go¹del’s famous theorem that there are
mathematical statements that cannot be proved or disproved. Using this,
Penrose shows that there are mathematical statements that cannot be proved and
yet we human beings can see they must be true.

This single remarkable fact is the foundation on which Penrose builds his
thesis. If human beings can see “truth” in a way that cannot be expressed
algorithmically, then we seem forced to accept that the brain is more than an
enormously complicated computing machine. When I first encountered this idea
in The Emperor’s New Mind, I was immediately struck by the following thought.
If it is so easy for us to see the truth of the mathematical statement (and it
is when you follow through the steps in Penrose’s argument), why cannot the
logic behind that insight be incorporated in a computer algorithm?

In Shadows of the Mind, Penrose addresses this point by saying that you
could indeed do just that. However, this would in turn change the nature of
the logical system that generated the unprovable mathematical statement. The
upshot is that you get a different mathematical statement which once again is
beyond the logical jurisdiction of the computer. This kind of argument, in
which one seems to be endlessly chasing one’s tail, leaves me feeling a little
dizzy. But what seems especially odd is that the way we humans rise above the
computer for this problem always seems to involve the same trick. Having
learnt it the first time round, why can’t the stupid machine learn to apply it
in all cases, god damn it?

Although Penrose deals with some twenty different objections and counter-
arguments to his Gošdel-like proof, I could not find an answer to a
possibly more damaging counter-argument from the philosopher Daniel Dennett.
Penrose’s claim that the human mind is noncomputational appears to be
comparable to the argument concerning a highly advanced chess-playing machine:
X is superbly capable of achieving checkmate; there is no algorithm guaranteed
to achieve checkmate; therefore, X does not owe its power to achieve checkmate
to an algorithm.

The argument is clearly fallacious, as demonstrated by the fact that there
is such an X which does use algorithms – one only has to think of the computer
that recently beat the world chess champion, Gary Kasparov. By analogy then,
even if there is no algorithm guaranteed to discover mathematical truth, it
seems dubious to conclude that mathematicians cannot be executing algorithms
in the brain to make their discoveries.

While I have reservations about the first half of Shadows of the Mind, I
was completely won over by the second half. In this, Penrose explores and
dissects in intricate detail some of the extraordinary features of quantum
theory. He classifies some of its central mysteries into different types: ones
that we have to accept as brute facts and others that make it clear that
something is amiss in the state of quantum theory. In particular, Penrose
elucidates why all attempts at explaining away the “collapse of the wave
function”, the sudden change that quantum systems undergo when subject to an
observation, have proved unsatisfactory.

This aspect of quantum physics has troubled many other physicists but in
this book Penrose has gone further than anyone in unravelling its
implications. This ultimately leads Penrose into exploring the connections
between quantum theory and the mind. But whereas in The Emperor’s New Mind,
the role of quantum gravity remained obscure, in this book Penrose is putting
his money on microtubules – mysterious hollow protein structures that make up
the cytoskeletons of cells. Following controversial ideas from people such as
Stuart Hameroff at the University of Arizona, it is Penrose’s belief that
cytoskeletons act as minicomputers within each cell and that these
minicomputers rely on quantum properties to work. The evidence for these ideas
is, to say the least, shaky but research into microtubules is at an early
stage.

If there is any truth in the microtubule idea, it could mean that neural
networkers and neurobiologists have so far missed a huge part of the machinery
of the brain. It would also imply that the brain is vastly more complicated
than anyone has previously thought. According to Penrose’s calculations the
brain would be capable of processing 10²⁴ operations per second,
some 10 orders of magnitude faster than that estimated from a simple nerve
cell count. That’s a rather depressing conclusion because it would put the
prospect of building an artificial brain with powers comparable to a human
brain back by many decades, if not centuries.

But if we are to accept that the brain is non-algorithmic, we need more
than sheer complexity to explain its powers. It is here, Penrose believes,
that quantum theory or quantum gravity must ultimately fill the gap. Quite
how, he still does not know though he does have some ideas. Highly speculative
though it is, I think people who enjoyed reading The Emperor’s New Mind will
find Shadows of the Mind a must. It takes the original arguments further and
is packed with fresh material including many fascinating historical
diversions. For others, though, be warned: this book could strain your
synapses – or should that be microtubules?

This loosely sums up a question that has to be answered every year by
the British government. The aliens turn out not to be little green monsters
in space garb but particle physicists in suits. Look at it from the government’s
point of view. It is spending too much money to balance its books and there
seem to be a bewildering number of politically important causes on which
it could easily spend more: the health service, unemployment, crime, education,
and so on. And here are particle physicists with a begging bowl claiming
to have access to arcane secrets. About ÂŁ100 million a year will
do nicely, they say. Show them the door.

Take the case of the Superconducting Supercollider project in the US.
In his book Dreams of a Final Theory, Steven Weinberg eloquently argued
the case for building the SSC because of the glimpse it might give us of
the ultimate theory of the Universe. That Weinberg was also aware of the
political realities was clear from a revealing passage in his book: ‘When
I testified in favour of the SSC in a Senate committee in 1987, one of the
senators remarked that there were then almost 100 senators in favour of
the SSC but that after the site was announced there would be just two.’

Severe blow

Weinberg’s view turned out to be pessimistic and the project went ahead
– even though it was expected to cost $6 billion. After a year, when costs
had spiralled to $11 billion, Congress cancelled it. The decision, a severe
blow for American physicists, has been a double-edged sword for scientists
at Europe’s particle physics centre, CERN. It considerably strengthened
the case for building the Large Hadron Collider, which will explore similar
physics for only ÂŁ1 billion. But it also made the researchers more
wary of justifying their research solely on intellectual grounds.

In Britain, this change of sentiment became clear last year when the
government published its White Paper on science, Realising Our Potential,
which led to the restructuring of the research councils. The councils were
to be much more business-minded. Each one is now run by a chief executive
instead of a chairman and the ‘mission statements’ all emphasise the enhancement
of Britain’s industrial competitiveness and quality of life. Admittedly,
this has been toned down in the case of the Particle Physics and Astronomy
Research Council (PPARC), but the requirement to take grubby commercial
realities ‘into account’ is still there. But what applications are there
for particle physics?

At the turn of the century, the subject was cheap and amazingly productive.
When Wilhelm Roentgen discovered X-rays in 1895 he used little more than
an evacuated glass tube with a couple of electrodes. Around the same time,
J. J. Thomson was using similar equipment to discover the electron. It
is very unlikely that anyone then could have guessed the impact these discoveries
would have on society. The applications have been many: electronics, TV
tubes, medical X-ray imaging, airport scanners, X-ray crystallography, and
so on.

But in a classic case of the law of diminishing returns: the study of
particle physics since the 1890s has become much harder and a lot more expensive.
Has it become less productive? A few years ago when the SSC was first coming
under intensive scrutiny from politicians, particle physicists around the
world formed a club called the International Committee for Future Accelerators
(ICFA) in order to further their common aims and objectives.

To defend large accelerator projects such as the SSC and LHC, the ICFA
compiled a list of spin-offs from recent research in particle physics. The
results were distributed earlier this year to research institutions. Here
was evidence to be used when the cause of particle physics needed defending.
Three applications may show how useful modern particle physics can be.

The first and perhaps most significant area is in medical imaging and
radiation therapy. One of the most prominent spin-offs, positron emission
tomography (PET) scanning, is a powerful technique for detecting cancers
and anomalies in blood flow, oxygen usage and glucose metabolism. It was
made possible by the production of positron-emitting isotopes in particle
accelerators. The isotopes are injected into patients prior to being scanned
and it is important that the isotopes give a clear signal of their position
in the body without exposing the patient to too much radioactivity. The
positron-emitting isotopes are uniquely effective in this respect.

Another area is to be found in the construction of the next generation
of microchips. Currently, the biggest memory chips are 64-megabit devices.
They are produced using a photolithographic process in which light is used
to project a master ‘mask’ onto the surface of a silicon wafer coated with
a photoresist. Using visible light limits the resolution of the mask to
around 0.2 micrometres, preventing further increase in memory density. To
go to 256 megabit and 1 gigabit chips will require a process with higher
resolution. One of the most promising techniques is to use X-ray lithography,
in which beams of X-rays are generated from a synchrotron particle accelerator.

Future chips

Online computing has also benefited from particle physics. Experiments
at CERN and elsewhere produce vast amounts of data that scientists all over
the world want to access without travelling to Geneva. This has presented
software teams at CERN with a huge problem but, out of it, has emerged one
of the most significant developments yet in computer networking: the World-Wide
Web. For the thousands of people in Britain alone who log onto the Internet
each month, the World-Wide Web is one of the best methods of accessing information.

These are three examples of genuine spin-offs from particle physics.
But how much direct benefit is there for the countries that invest in the
research? There are two aspects to this question. The first concerns contracts
produced as a result of building and running large accelerators. Britain,
for example, contributes 13 per cent of the CERN budget while British industry
gets around 7 per cent of the contracts. The lion’s share go to the host
nations, Switzerland and France. This issue caused Germany and Britain to
delay approval of the Large Hadron Collider earlier in July. It seems France
and Switzerland will now have to pay more for the privilege of being host
nations.

The second aspect – quantifying the economic benefits from the direct
spin-offs of research – is extremely difficult to measure. Again, take Britain.
From the examples cited, it looks rather limited. Oxford Instruments, a
maker of advanced magnets and other specialised scientific equipment,
is the one notable success story. In X-ray lithography, for example, it
produced the world’s first compact synchrotron which went to IBM in the
US. That was three years ago, but the company is hoping for another order
soon. Altogether, though, 10 cyclotrons for fabricating advanced microchips
are being built around the world – six in Japan not a country with a history
of investment in particle physics.

Oxford Instruments also make ‘desktop’ cyclotrons – compact machines
that generate the proton beams needed to produce the radio isotopes for
PET scanning. Because the half-lives of these products are so short, they
have to be made at the hospital where the scans are to be performed. If
PET scanning is to become widely available, many more medical centres will
need their own cyclotrons. So Oxford Instruments clearly has an excellent
opportunity here. However, the two main PET scanning centres in Britain
– the Hammersmith Hospital and Addenbrookes in Cambridge – bought their
machines abroad.

The brilliant idea behind the World-Wide Web, which was invented at
CERN, was made available to anyone who wanted it. The intention was not
to make money but to encourage the spread and development of a system that
could make life easier for Internet users. It is doubtful if the World-Wide
Web would have thrived if it had developed as a commercial service.

Ultimate properties of matter

Nevertheless, David Saxon, the chairman of PPARC, is confident that
Britain will reap greater benefits in the future from CERN. He cites a new
scheme which will award research grants for university and industrial collaborations.
The scheme, called the Physics and Industry Programme Support Scheme, is
meant to encourage researchers in particle physics and space research to
find ways to help British industry.

The first grants will probably assist only British industries to compete
for contracts at CERN. The electronics company EEV, for example, makes gallium
arsenide solar cells which are used in space satellites. Teams from the
universities of Sheffield and Leicester have been awarded a grant to help
EEV adapt their gallium arsenide technology to the demands of building radiation
detectors at CERN. But according to a spokesperson at EEV, it is unlikely
that this technology will help them build better solar cells for their satellite
business.

Many physicists and engineers in universities and industry still believe
that CERN exists for one main reason – to investigate the ultimate properties
of matter. The LHC is now the only hope left for probing sufficiently high
energies to answer questions about the origins of mass and the nature of
dark matter in the Universe. Spending ÂŁ1 billion on the LHC is obviously
a much better deal that the SSC. Also, the energies achieved at CERN have
increased by a factor of a thousand over the past thirty years even though
budgets have stayed roughly constant in real terms. A thousandfold increase
in productivity in 30 years is no mean achievement.

So, what would you be prepared to pay to the visiting aliens? The British
government’s offer is around a few pounds per year for each man, woman and
child in the country. Ultimate truth is not as expensive as it may first
look. Let’s just hope that we really can ‘realise its potential’

Julian Brown is a science writer.

TV screens that roll up like blinds, smart cards that are really flexible and electronic displays that can be embedded in car windscreens could flow from the invention by French scientists of the flexible transistor. Francis Garnier and colleagues at the CNRS Laboratory for Molecular Materials near Paris have developed ways to replace all the structures in a transistor with polymers. What is more, they are able to lay down these components without the complex high-temperature, high-vacuum techniques needed in silicon chip foundries.

In 1990, Garnier’s team unveiled a transistor made mostly from plastics; but this still needed some metallic components, such as gold and silver for the electrodes (Technology, 15 December 1990). Even these have now been dispensed with, replaced by conducting contacts made of graphite-based polymer ink. The contacts are deposited either side of a thin layer of insulating polyester. The whole structure is supported by adhesive tape, which lies beneath the bottom electrode, or gate (see Figure).

With the help of a precision-made mask, Garnier’s team lays down a 40-nanometre film of a polymer semiconductor between the two top contacts, the source and drain. The plastic is a modified form of a sulphur-based polymer, sexithiophene, which has semiconducting properties similar to those of silicon.

The device works in a similar way to a field-effect transistor. The field produced by a voltage applied between a top electrode and gate modifies any current flowing between the source and drain. The device, which is described in last week’s issue of Science, can work as a simple switch or an analogue amplifier.

Garnier’s device is about 50 micrometres in size, more than ten times larger than conventional transistors that are etched onto silicon chips. So it is unlikely that manufacturers will see the polymer transistor as a cheap way to make complex devices such as microprocessors. However, the great advantage of the new devices is that they are completely flexible and can be made virtually transparent.

Their size is still small enough to make them suitable for large flat-panel screens. They could also help to solve a major problem with electronic smart cards – when the cards bend, their chips can fly off. By ‘printing’ plastic transistors onto the card instead, banks could issue cards that are much more robust.

If made transparent, the transistors could be ideal for head-up displays in cars. By incorporating the devices with liquid crystal displays into glass windscreens, Garnier predicts that it will be possible to put visual displays directly in a driver’s line of vision.