The experimental tweezers, whose working end is a pair of electrically
controlled carbon nanotubes, were developed by Philip Kim and Charles Lieber
from Harvard University. Using the tool, they have successfully grasped clusters
of polystyrene spheres of around 500 nanometres in diameter鈥攖he same scale
as cellular substructures. They were also able to remove a semiconductor wire 20
nanometres wide from a mass of entangled wires.

The team says that because carbon nanotubes are not only very strong but also
good conductors of electricity, the tweezers could also be used to measure the
electrical characteristics, such as resistance, of nanostructures within their
grasp. They could analyse microcircuits or even the components of future
nanorobots that might perform surgery inside the body.

The tweezers were made by depositing gold electrodes on either side of a
tapered glass micropipette 100 nanometres across
(see Diagram). The tweezer鈥檚
arms were each made of a bundle of multiwalled carbon nanotubes or
buckytubes鈥攃ylindrical versions of the famous buckyball, arranged
concentrically like the layers of an onion. One of these was then glued with a
conductive adhesive to each electrode.

To operate the tweezers, a voltage is applied across the electrodes, causing
one nanotube arm to develop a positive electrostatic charge and the other to
develop a negative charge. The attractive force can be increased or decreased by
varying the applied voltage. The team found that they could completely close the
arms at a voltage of 8.5 volts, while lower voltages would give certain degrees
of grip.

At present, each of the tweezer鈥檚 arms is about 50 nanometres wide and 4
micrometres long. But by growing single-walled nanotubes directly onto the
electrodes, the researchers hope to produce tweezers small enough to grab
individual molecules.

Japanese researchers have already designed a pair of chemical tweezers that
can pick up one molecule at a time
(New 杏吧原创, 17 August 1996, p 22)
but their device could only recognise and clamp tightly to one particular
type of molecule, such as a sugar molecule. They are not designed to be general
purpose, like the Harvard tweezers.

鈥淎 one kilogram wrench dropped from one metre is enough to cause serious
damage to a carbon composite structure,鈥� says Glyn Davies of Imperial College,
London. Unlike metal structures, which get obvious dents, the effects of these
impacts on the carbon composites used in the latest military jets can be
invisible. 鈥淪ixty per cent of the materials鈥� compressive strength could be lost
due to impact damage, but the world can鈥檛 see it. The composite just springs
back to shape, hiding the damage inside,鈥� says Davies.

This makes visual inspections for damage unreliable, so there鈥檚 a clear need
for something that tells you if an aircraft has been hit, and where, he says.
Ultrasound and X-ray techniques are both used, but these are slow and costly.
Getting an aircraft back into the air quickly, or keeping it on the ground if it
is badly damaged, can be vital. So researchers having been trying to
develop smart materials that sense any damage as it happens.

The DERA team, based at centres in Malvern and in Farnborough, has proposed
measuring both the location and the extent of impact damage using
鈥渢riboluminescent鈥� materials that emit light if fractured. This is an effect you
can see when someone crunches mints in a darkened room: tiny flashes of light
can sometimes be glimpsed between the teeth. The flashes occur when positive and
negative charges inside the crystals recombine. Some crystals with no rotational
symmetry and some symmetrical ones with added impurities have proven good
emitters
(New 杏吧原创, 24 May 1997, p 19).

A feasibility study has showed that a variety of highly triboluminescent (TL)
materials, ranging from a derivative of aspirin to metallo-organic compounds,
could be included in carbon composite structures (Smart Materials and
Structures, vol 8, p 504). The tests also show that the worse the damage,
the more light the materials emit, and that light is emitted only if the
crystals are fractured, so there鈥檚 no possibility of false alarms.

The idea is to embed these TL materials in an aircraft鈥檚 structure and
monitor their output using optical fibres linked to a detector
(see Diagram). If
the plane is struck, the number of crystals that flash and the intensity of the
light they emit would give a real-time picture of the extent of the damage.FIG-22042301.jpg

To locate the damage, crystals that emit light at different wavelengths could
be embedded in different places. Alternatively, the colour of the light produced
by crystals in different areas could be altered using conventional fluorescent
laser dyes.

鈥淲e plan to investigate the most efficient ways of coupling the light output
from the sensors to the detector, and to develop a representative
demonstration,鈥� says Grant Bourhill of DERA.

THE next time a stranger at a party asks me what the point of physics is,
I鈥檒l tell them to read Curt Suplee鈥檚 excellent Physics in the 20th
Century. He has managed to achieve that seemingly impossible task:
describing all the major developments in physics in this century without
trivialising or oversimplifying them. When they reach the end, general readers
will have a fair grasp of what Einstein鈥檚 theories of relativity are about, why
chaos theory is important, how nuclear fusion may one day provide clean energy,
and much more.

It鈥檚 a lot of ground to cover. As Suplee says: 鈥淭he world as it is now
understood differs more dramatically from the Victorian view than the science of
Galileo, Copernicus, and Newton differed from that of Aristotle.鈥� But a superb
layout, bite-sized chunks of text and a stunning array of pictures make this
scientific epic an effortless read.

I particularly liked the way the research process itself is presented. For
example, 鈥渁t the turn of the century even such formidable figures as Austrian
physicist Ernst Mach were still insisting that the supposed atom was no more
than a useful fiction鈥�, writes Suplee. He then describes the experiments that
revealed the atom鈥檚 basic structure. The gradual refinement of the technology
led to astounding developments such as electron microscopes with resolutions so
high that you can see individual large atoms, and the magnetic resonance imaging
techniques so vital to medicine. 鈥淪cience is not some set of specific goals to
be achieved once and for all,鈥� concludes Suplee, 鈥渋t is a never-ending,
ever-improving, self-correcting means of expanding human understanding of
苍补迟耻谤别.鈥�

Throughout, he focuses on how physics has influenced our everyday lives, from
the electronic revolution triggered by the transistor to the development of
lasers and radar. Even the chapter on quantum mechanics is refreshingly down to
earth. Instead of straying towards the field鈥檚 wilder hypothetical shores, it
concentrates on observable effects, spin-off products such as scanning
tunnelling microscopes, and the potential for superfast quantum computers that
might be able 鈥渢o compute a number of possibilities simultaneously鈥�.

Suplee knows when to tackle the larger questions, too. In his chapter on
cosmology, he states that the nature of 鈥渄ark matter鈥濃�攚hich, while making
up around 90 per cent of the Universe, is not yet understood鈥攈olds the key
to 鈥渨hether the cosmos will continue expanding indefinitely, remain about the
way it is, or slow down and ultimately collapse back upon itself鈥�.

There are numerous illustrations, and their captions often go into more
detail than the text, which helps to make more complex topics more palatable.
The images themselves are truly amazing and diverse. Here you鈥檒l find one of the
first X-rays, taken by their discoverer Wilhelm R枚ntgen, and a human hair
notched by a laser beam., Or you can gape at the three technicians aboard a raft
inside the Super-Kamiokande neutrino detector in Japan, snapped as they cleaned
the faces of 13 000 photomultipliers while the detector filled with water.

There is a slight American bias鈥攇iven that it was sponsored by the
American Physical Society and the American Institute of Physics, but it鈥檚 a mild
irrititant for internationally minded readers. And in a few places, the text may
puzzle people without a scientific background. But these are minor criticisms to
set against his achievement. Suplee even manages to capture the excitement the
original experimenters must have felt when they made their groundbreaking
discoveries.

While Physics in the 20th Century is intended for nonspecialists, it
reminded me of why I spent so many years studying this remarkable subject. I
urge anyone with an interest in physics to buy this. Believe me, your coffee
table has room for it.

Nanotubes鈥攖he cylindrical version of the buckyball, or
fullerene鈥攁re good electron emitters. A flat array of them could replace
the bulky cathode ray tube in a TV. Instead of a single electron beam scanning
the whole screen, electrons would be fired from the end of each tiny tube.
Unlike today鈥檚 flat-panel displays, which are all based on liquid crystal
technology, nanotubes promise a brighter picture that can be viewed from a
broader range of angles.

However, because nanotubes emit electrons only from their ends, rather than
their sides, they must be aligned perpendicularly to the TV screen. Previous
attempts at growing well-aligned nanotubes on glass worked at only temperatures
high enough to damage the glass. But researchers at the State University of New
York at Buffalo have now succeeded in growing well-aligned tubes at lower
temperatures. In their method, the surface of the glass is coated with a thin
layer of nickel.

Acetylene and ammonia gases are then introduced into a low pressure chamber
containing the glass. Heat breaks the acetylene gas down into carbon and
hydrogen, while the nickel and ammonia act as catalysts for the formation of
nanotubes on the glass.

Using this approach, the researchers were able to obtain nanotubes between
0.1 and 50 micrometres long, and from 20 to 400 nanometres wide. Those
nanotubes with diameters above 50 nanometres were well-aligned (Science, vol
282, p 1105). The next step will be to work out the optimum nanotube size and
density for efficient electron emission, says Zhifeng Ren, a member of the
Buffalo team.

As waves head towards land, they interact with the rising seafloor, creating
tiny鈥攂ut proportional鈥攕eismic waves, which travel inland. In a test
reported in Measurement Science and Technology (vol 9, p 1772), the UWA
team measured these seismic waves on their vibration isolator from 10 kilometres
inland. The results showed a 92 per cent agreement with wave data from a buoy
that was 40 kilometres out to sea.

Vibration isolators are used under delicate equipment in experiments where
readings would be ruined by shaking鈥攚hether caused by footsteps or a
distant earthquake鈥攁nd are designed to keep equipment still. UWA鈥檚
vibration isolator is made from aluminium, and has a 30-centimetre long
horizontal plate suspended from a frame by flexible titanium hinges.

The hinges enable the plate to remain still to about one tenth of a nanometre
when the frame is shaken. The seismic waves cause the frame to move by a few
micrometres, and an optical sensor detects the relative motion between frame and
plate. This difference is directly proportional to the wave height.

UWA鈥檚 David Blair suggests that the devices 鈥渃ould create a system to image
the ocean waves coming in all round the coast鈥�. He also plans a smaller version
for surfers.

In a pilot study reported in Measurement Science and Technology (vol
9, p 1259), rigid lenses from three different manufacturers were measured. They
all showed marked deviations from the shape expected. In a follow-up clinical
study, 鈥減oor edge shape was shown to correlate with unsatisfactory fitting and
low comfort鈥�, says the instrument鈥檚 developer Klaus Ehrmann.

Optical profilometers are not new, but because they rely on detecting
reflected light, they can scan only relatively flat surfaces as curves deflect
the light out of the path of the instruments鈥� fixed sensors. But in the new
profilometer, the CD laser head is mounted on a motorised, rotating stage.
Feedback from the CD head鈥檚 鈥渢racking error鈥� signal鈥攚hich normally stops
the head skipping data tracks鈥攊s used to rotate the head until it is
perpendicular to the lens surface. This ensures that no data is lost, whatever
the surface shape.

The head is scanned across the surface, and at each data point, the angle
through which it must be rotated corresponds to the slope of the lens profile.
The lens can also be rotated, allowing scanning in successive planes.

The device can measure the surface shape of a contact lens to an accuracy of
2 micrometres. And because it uses a laser beam as a probe, it does not damage
the lens. So for the first time, complete three-dimensional representations of
lens shapes can be built up.

Graeme Young, director of Visioncare Research of Farnham, Surrey, which tests
eye products, cites poor comfort as the most common reason for people giving up
rigid contact lenses. 鈥淭his instrument is therefore a welcome tool in helping to
develop more comfortable lenses,鈥� he says.

The most common testing method involves touching a chip with a probe to
measure voltages. This shows up failed components, but is no help working out
what has gone wrong. Probing millions of components is impractical, so usually
only a few points are tested, which limits quality control. Electron-beam
scanning is occasionally used but it cannot distinguish between components made
of different materials. Both techniques draw current from the circuit and
interfere with what they are measuring.

The Sussex technique can detect signals passing through any component using a
device called a scanning electric potential microscope. The SEPM uses a tungsten
probe connected to a novel voltmeter. When mounted a few micrometres above a
chip, the probe and circuit form a capacitor. Any AC signal flowing beneath the
probe causes a displacement current to flow through this capacitor. The value of
this current changes depending on the amplitude and phase of the AC signal,
enabling the signal to be imaged.

Feedback circuitry ensures that the probe draws displacement currents of only
10^-15 amps鈥攖oo low to affect the current flowing in the
circuit鈥攂y giving the probe an extremely high resistance to AC current.
Taking measurements of components 1 micrometre across means the probe needs a
resistance of over 10¹⁷ ohms.

So far the researchers have scanned an operational amplifier 1 micrometre
across (Measurement Science and Technology, vol 9, p 1229). But Terry
Clark and Robert Prance of the Sussex team say the instrument will be able to
scan 10 nanometre components once it is given a high enough resistance. At this
scale, individual electric charges could be detected.

Akhtar Rasul, yield improvement manager at semiconductor maker Motorola,
says: 鈥淚f it鈥檚 going to give you fast diagnostic capability at high resolution,
nondestructively, then the semiconductor industry will welcome such a
诲别惫别濒辞辫尘别苍迟.鈥�

The water leaving the plant to be piped to consumers has to be constantly
monitored for levels of chemicals and contaminants, to make sure it is fit to
drink. One measure is the water鈥檚 clarity. This can be tested by shining a light
through a sample and measuring how much of it is scattered or absorbed.

One difficulty with this technique is that the transparent window through
which the light shines into the water soon becomes covered with a film of
impurities, making nonsense of the readings. Cleaning the window is a slow and
labour-intensive business, which limits testing.

Now Mark Johnson, a researcher at North West Water, has invented a system
that uses a light beam to measure turbidity but keeps the water away from the
window so the glass stays clear (Measurement Science and Technology,
vol 9, p 399). The new technique creates a small air bubble at the bottom of the
sample vessel, and shines a light from there up through the liquid.

The shape of the bubble is maintained by keeping the air pressure inside it
constant. A pressure sensor controls an air pump built from a pair of
8.5-centimetre loudspeaker cones. As the cones move they pump air in or out of
the bubble.

The system allows treatment plants to test their output at least once an
hour. The same technique can be used to test if waste water from sewage
treatment plants is fit to be discharged into rivers. Dennis Peters, manager of
sensors and instruments at the Water Research Centre, a consultancy based in
Buckinghamshire, is impressed with the device, though he says it may not have
all the answers. 鈥淭he ultimate solution is likely to be total noncontact
systems,鈥� he says.

The nose, which was developed at Warwick University, can identify common pathogens that cause infections of the ear, nose and throat. At the moment it takes about three days to get results from ear and throat swabs, and in that time a patient may receive no treatment or the wrong antibiotic. The nose can make its diagnosis in a matter of minutes.

The idea of using the nose clinically grew from its use detecting food and drink that had gone off (Technology, 20 February 1993, p 18). 鈥淚t struck me that what we were really looking at was the growth of bacteria which were responsible for food spoilage,鈥� says Julian Gardner, head of the Warwick team.

The nose contains an array of sensors, tuned to pick up the odours from growing bacteria. Smelly air from above the sample is passed over the sensors, changing their electrical resistance. The pattern of response is then fed through a neural network program, which compares it with patterns from known smells.

The latest system is not only quick but extremely accurate. In tests, reported in Measurement Science and Technology () the nose correctly identified Staphylococcus aureus 100 per cent of the time, and Escherichia coli with 92 per cent reliability.

People with ear, nose and throat infections may not be the only patients to benefit. 鈥淲hat was very exciting to us was that this generic technology could be used to detect infections in different parts of the body,鈥� says Gardner. Diabetics, for example, have acetone on their breath.

Peter Mantle, a microbial biochemist from Imperial College, thinks the immediacy of the system is a real benefit. 鈥淚t gives extra punch to the use of the term nose by enabling the sort of quick response that a human gives to a nasty smell. I can see all sorts of opportunities for it.鈥�

Until now, magnetic images this detailed could only be produced using a
superconducting quantum interference device. A SQUID is expensive to operate
because both the device and the sample being imaged have to be cooled to 4
kelvin. The Sussex device works at room temperature and at higher resolutions
than the SQUID.

Video heads can read magnetically coded information thanks to a tiny
insulation gap cut into the torus of iron-based material they are made from. The
gap allows part of the head鈥檚 magnetic field to leak out so that when a magnetic
source, such as a video tape, is close to it, the head鈥檚 inductance changes.

If these changes are monitored electronically, computerised images of the
magnetic properties of the sample can be produced. To reach resolutions of a few
micrometres, the video heads were machined down in size. The researchers say
that using smaller heads would take resolutions below 1 micrometre.

One of the first applications of the magnetic microscope could be spotting
forged US dollar bills or British cheques. Although it currently takes tens of
minutes to image a dollar bill, Helen Prance, a researcher on the Sussex team,
says arrays of heads could speed scanning.

Although British banknotes are not printed with magnetic inks, Heather Hill,
a document examiner from the Forensic Science Service, thinks the machine may
have a role. 鈥淚t could have a use in detecting the magnetic inks on cheques,鈥�
she says. Current FSS detection methods can only show whether an ink is magnetic
rather than whether it is forged or has been tampered with.

The first magnetic images of dollar bills made with this system appear in the
May edition of Measurement Science and Technology (vol 8, issue 7, p 734).