Researchers at the chemicals company ICI in Everberg, Belgium, are tweaking
the molecular structure of polyurethane foam to make it absorb specific
frequencies of sound from the main sources of noise in a vehicle.

To predict which noises the polyurethane material will absorb, the
researchers are using a technique known as computerised acoustic mapping. This
shows how changes in the molecular structure of a foam will affect its resonant
frequency, and so the frequency of sound waves that are absorbed.

The researchers modify the polyurethane by controlling the molecular weight
of the polymer chains, and the number of bonds joining them. Cutting short the
polymerisation process makes shorter chains and a stiffer foam, and adding other
chemicals to the polymer can promote or break up bonding between chains.

The stiffer the foam, the higher the frequency of sound it muffles. Using
these techniques, researchers can “tune” foams to absorb sounds with frequencies
of between 100 and 1000 hertz—the frequencies of the loudest and most
irritating noises cars generate.

ICI gives its flexible polyurethane foam an adhesive surface, allowing it to
bond directly to the chassis. This removes the need for glue, reducing
manufacturing costs, and also ensures a better fit. Any gaps between the
insulating foam and the car body will allow noise to enter the passenger
compartment.

The tunable foams are adaptable enough to be used inventively. The
characteristic whine of a sports car’s engine, for instance, is one of its
selling points. So the engine noise could be left alone, while foam absorbing
different frequencies could be applied elsewhere in the car. One day it may even
be possible to choose a car by acoustics as well as colour.

Most gravity sensors, or gravimeters, use combinations of masses, balances
and accelerometers to pick up the tiny changes in gravity caused by the
variations in the densities of the rock below. But these instruments cannot
distinguish between acceleration and gravity, making it difficult to use them on
the move. A more promising approach is to measure how the strength of the
gravitational field changes with distance
(“Seeing with gravity,” New ĐÓ°ÉÔ­´´, 14 September 1996, p 24),
using a sensitive instrument known as a gradiometer.

Alexey Veryaskin, a physicist working in New Zealand, has developed a
gradiometer that is built round a superconducting string of atoms kept supercool
by liquid helium. Changes in the gravitational gradient along the string, which
is suspended vertically, cause tiny deformations. The deformations show up as
tiny changes in the string’s magnetic field, which are detected by an
exquisitely sensitive magnetic detector known as a SQUID—a superconducting
quantum interference device. To eliminate any electromagnetic interference, the
string is contained within a superconducting tube.

The detector in Veryaskin’s gradiometer is incredibly sensitive, picking up
displacements of 10^-13 metres—far smaller than the width of an atom. It
is also easier to isolate the string from the effects of motion, which can
confuse other types of gradiometer. Veryaskin says the sensor can separate
changes in gravity from perturbations caused by acceleration because the string
is measuring gradients along its length.

The first major use of the sensor will be for geological surveys. A Veryaskin
gradiometer lowered down a borehole would determine the density of the
surrounding rock, indicating what mineral deposits are present. As Veryaskin’s
device is only 20 centimetres long and largely immune to inertial forces, it can
be used on the move without having to be constantly recalibrated. An airborne
survey could map the Earth’s gravitational field over large areas, and indicate
oil or mineral deposits far more quickly and cheaply than seismic methods.

Veryaskin’s company, Gravitech Instruments, expects to have a prototype of
the gradiometer within six months. “Once you’ve got a tool this sensitive,” says
Simon Frasier of Gravitech, “there are any number of things you can do with
ľ±łŮ.”

The sensor could, for example, be used to pinpoint a submarine’s location by
taking a precise gravity reading and comparing this to a map of the Earth’s
field, or to detect buried land mines, or search for archaeological remains.

Turbines are more usually found in power stations and jet engines, and
previously only found exotic uses in cars—such as the Proteus engine that
drove Bluebird to the world speed record in 1964. In 1996 Chrysler demonstrated
the Patriot, an experimental vehicle that had a flywheel in its engine.

Rosen Motors picked the turbine because of its high fuel efficiency and used
a flywheel to overcome problems posed by the high power required for
acceleration. The flywheel gets the car going, while the turbine keeps it
cruising. An electronic controller selects the best blend of turbine and
flywheel power to supply to two electric drive motors.

The carbon-fibre flywheel stores up to one kilowatt-hour of energy, which can
be converted to electricity at a maximum of 150 kilowatts. The flywheel spins
constantly even when the car engine is turned off. At these times it spins at 52
500 rpm. When the car is being driven it spins at speeds of between 28 000 and
60 000 rpm. It rotates in a vacuum and takes six weeks to come to a stop.

If the flywheel does stop after a long period off the road, the turbine can
be used to restart it to normal speed in two minutes. Energy to keep it spinning
is normally regained during braking or when the car is going downhill.

A surprising benefit comes in performance. The electric drive motors deliver
full torque from a standing start, and have pushed a Mercedes-Benz test vehicle
from 0 to 60 miles per hour in 6 seconds.

The high energy capacity that makes a flywheel attractive also makes it
difficult to handle. The flywheel must be allowed to swivel and turn to prevent
gyroscopic forces from affecting the handling of the vehicle.

In addition, the whole unit has to be securely contained in the event of a
crash. Ben Rosen, chairman of the company, says this latter problem has made
many people sceptical about the engine’s commercial prospects. “There’s an
appropriate amount of scepticism to keep us fired up,” he says.

Rosen Motors is now seeking car makers prepared to use its engines, and the
first cars “Powered by Rosen” could be on the road in six years’ time. To make
it easier to adopt the hybrid it is the same size as existing engines.

Rosen is also chairman of Compaq Computers, and by the end of this year will
have personally invested $20 million in the engine company.

A spokesman for the Motor Industry Research Association in Nuneaton,
Warwickshire, says he has not encountered Rosen Motors’ hybrid engine before.
But he says that previous attempts to use flywheels in car engines have
foundered because the energy in the wheel dissipates very quickly. He adds that
measures to contain it often add weight and affect performance.

Karen Reeson and colleagues in the University’s School of Electronic
Engineering and Information Technology make their devices by bombarding a
silicon chip with iron atoms. This creates small islands of iron disilicide in
the boundary layer between the p-type and n-type conductor materials in the
chip.

When an electric field is applied across the junction, the iron disilicide
behaves like a light-emitting diode, converting electrical energy to light at a
wavelength of 1.5 micrometres—the wavelength used in optical
communications.

Researchers have long hoped to use light instead of electrons to ferry data
within and between chips. A speed increase of a factor of 10 is described by the
Surrey team as “conservative”. The use of light in chips would also make them
less sensitive to heat and radiation.

The researchers are improving the implantation process and the amount of
electrical energy converted to light. The prototype converts 0.1 per cent of the
energy, and the team hopes to reach a commercially viable 2 per cent by
1999.

If the researchers can crack these problems, the semiconductor industry will
find it easy to adopt the new technique because the iron is implanted in the
same way that layers of silicon chips are built up. “If it’s as good as it
seems, it could go into every chip,” says Reeson.

For the past 20 years different Serendip—the Search for
Extraterrestrial Radio Emissions from Nearby Developed Intelligent
Populations—instruments have kept a close eye on the data gathered by
radio telescopes, looking for signals from other intelligent races. Earlier this
month the University of California, Berkeley, SETI programme installed an
instrument called Serendip IV on the Arecibo radio telescope in Puerto Rico.
This piggybacks on the telescope and examines 168 million frequency channels
every 1.7 seconds.

But Serendip IV produces a huge amount of data that the SETI programme does
not have the resources to analyse. So any Internet surfers who have computer
power to spare are being recruited to help analyse data collected by the Arecibo
telescope.

The group of astronomers and computer scientists behind Serendip plan to
separate the Arecibo data into chunks that will be sent over the Internet to
participating computers. Each volunteer will be supplied with an analysis
program that automatically processes the data for a few hours or days, before
returning it to the server. Serendip has developed software to search for
signals at 4 million combinations of frequency, bandwidth and chirp (the drift
of frequency with time). By the time 50 000 PCs are involved, the search will
rival all current SETI projects, without the need for expensive
supercomputers.

The program comes with a screensaver illustrating the search process, and the
faint but tantalising possibility of making a vital discovery. The Internet
project is called SETI@home and is due to launch in spring next year.

Now researchers at the University of St Andrews in Scotland are working to
make such accidents a thing of the past. Their combat identification system
(CID) is somewhat bizarre: it lets friendly forces recognise each other without
revealing themselves to the enemy by using electromagnetic radiation that makes
tanks “look” like hot rocks and trees.

For years, NATO has been looking for a cheap and reliable way to let its
troops recognise each other. Existing systems often rely on vehicles constantly
emitting radio signals or interrogating each other before an attack to find out
if a potential target is friend or foe. But the brief radio transmissions used
to exchange messages also advertise their presence to the enemy. This means that
tanks and personnel carriers often have to switch off their identifiers before
entering hostile territory.

St Andrews has been working with Britain’s Defence Research Agency to develop
the system, which the electronics giant GEC built in time for NATO to include
last month in tests of competing CIDs now being developed in Britain, France,
Germany and the US.

The competition rules were strict: the identifying signal should only be seen
with an appropriate receiver, captured receivers should become useless, and it
must be cheap enough to fit on all gun-bearing vehicles. The best performing
system is likely to be adopted by the armed forces of the competing nations.

The French, German and American systems rely on radio to identify vehicles.
The British system was believed to be the favourite because of its unique
approach. Although the trials were held in May, NATO has not yet revealed the
winning system.

The St Andrews system works by mimicking the electromagnetic radiation
naturally given out and reflected by warm objects like trees and rocks. It
broadcasts similar random emissions that contain hidden information.

The system generates a random sequence of numbers with an algorithm and uses
these numbers to adjust the frequency or amplitude of a broadcast. Tanks or
other army vehicles would be fitted with a millimetre-wave emitter coupled to
the random number generator to produce the signal. Millimetre waves fall between
infrared and microwaves in the electromagnetic spectrum, and are emitted by
anything warm that contains water.

The random number generator used is similar to those found in
computers. But instead of producing a statistically random string of numbers, an
identically “random” sequence is generated each time the program is run. If you
know the original algorithm, it is easy to recognise the pattern of numbers. But
without this “key”, says physicist Jim Lesurf of St Andrews, the broadcast
appears to have no pattern at all. An enemy will see only random noise which
looks just like that emitted by warm rocks or trees.

Each friendly vehicle will possess a receiver that knows the key. If a
receiver is captured, friendly forces can easily change the pattern. “You can
change the key as often as you like,” says Lesurf. “The minute you change the
key, having a receiver becomes useless.”

As the identifying signal is effectively camouflaged, it can be
left switched on all the time without drawing unwanted attention. “It doesn’t
attempt to look like one particular thing,” says Lesurf. “It attempts to look
similar to the kind of variations we see in the natural world.”

With this in mind, Leonardo Seccia and colleagues at the University of
Bologna are creating an image processing tool kit that can draw up a detailed
three-dimensional picture of the state of a mosaic before the restorers begin
their task. The technique can reveal how badly damaged the artwork is, what harm
previous restoration attempts may have done and even what materials were used in
the original.

Seccia and his colleagues are refining the image processing system while
restoring the richly coloured mosaics at the 6th-century church of San
Apollinaire Nuovo and a Neonian baptistry in the Italian city of Ravenna. The
mosaics are composed of thousands of tiny tesserae—coloured glassy
blocks—and depict scenes from the Byzantine era and the New Testament.

The team shines lamps onto the mosaics, and then uses charge-coupled device
(CCD) cameras to capture the reflected light. The researchers are interested in
wavelengths extending through the visible range to the near ultraviolet and near
infrared. CCDs convert light to digital signals that are fed to Matlab image
processing software to create digital images at various wavelengths across the
range. This allows the restorers to record the exact position of each tiny
tile.

The different degrees to which materials in the mosaic reflect heat and light
reveal the inner structure of the artwork. This tells the restorers where layers
of the plaster holding the mosaic have become detached, and where cement has
been used in previous restoration attempts.

Earlier studies of the church and baptistry revealed the materials used to
create the mosaics. Seccia and his colleagues are using this data to verify the
information their imaging system is producing.

Because the process draws on previous conservation work it should provide a
comprehensive and reliable digital database for future restoration work, Seccia
says. The data being collected will also be used to reconstruct the church and
baptistry on the Internet.