Coal-fired power stations produce 65 per cent of all the sulphur dioxide
emitted into the atmosphere in the US. Chang Yul Cha of the University of
Wyoming in Laramie has made a laboratory prototype of his microwave process
that not only rids the exhaust of harmful gases but produces two by-products
that could be sold.

Cha’s research was supported by the US Department of Energy. According
to Soung Kim, of the DOE’s Pittsburgh Energy Technology Center, conventional
scrubbers spray the flue gases from power plants with a limestone slurry
which reacts with the sulphur dioxide to leave gypsum. ‘These scrubbers
meet current sulphur dioxide regulations, but they are very expensive and
create large amounts of sludge, which is causing a disposal problem,’ she
says. Another process, selective catalytic reduction, is commonly used to
remove nitrogen oxides, but the catalysts and equipment are expensive, Kim
says.

Kim believes that the process will not only be more effective – removing
98 per cent of gases, compared with up to 95 per cent of sulphur dioxide
and 85 per cent of nitrogen oxides for conventional scrubbers – but should
also be cheaper. She estimates that it costs a power plant between $250
and $400 per tonne to remove sulphur dioxide using the slurry, while the
Cha process may cost as little as $120 per tonne.

Cha’s process has three stages. First the flue gases are piped into
a reactor full of particles of ‘char’, a form of coal which has been partially
burned in the absence of air. This creates a fluidised bed, a swirling mixture
of gases and char particles, and the sulphur dioxide is adsorbed onto the
char.

The gases, now stripped of their sulphur dioxide, are piped into a decomposition
reactor, also full of char. A microwave generator fills this reactor with
microwaves, which decompose the nitrogen oxides to nitrogen and oxygen.
These gases, and the now cleaned flue gas, are then released through a chimney.

Meanwhile, the char from the fluidised bed in the first reactor becomes
saturated with sulphur dioxide. The saturated char is removed from the fluidised
bed, mixed with coal and fed into a second decomposition reactor. Microwaves
are again used to heat the mixture, decomposing the sulphur dioxide and
driving off the resulting sulphur and oxygen together with all the volatile
constituents of the coal as gases.

This mixture of sulphur, carbon monoxide and carbon dioxide and other
hydrocarbons goes up to a ‘sparger’ where it is sprayed with cool water
that washes out the sulphur. The sulphur and water mixture is sent to an
evaporator and then a centrifuge, leaving powdered sulphur as a by-product.
The remaining gases are sent back to the power station’s boiler to burn.
The sulphur can be used to make sulphuric acid or fertiliser. The second
decomposition reactor creates more char as it works, and this material can
be used as a filter to clean waste water and toxic gases.

The microwave process should cost no more than present scrubbers, Kim
says, and generating the microwaves ought to take less than 1 per cent of
the plant’s electricity output.

A major question remains over how to scale the system up for commercial
operation. In the two generations of microwave scrubbers Cha has built so
far, the reactors are measured in centimetres. They will need to be an order
of magnitude bigger for commercial applications. ‘The distribution of the
microwave energy in the reactor is critical,’ Cha says.

Gilbert De Mey, an electronics engineer at the University of Ghent in
Belgium, and Andre Vandierendonck, a psychologist, collaborated on the ‘thermal
image generator’. De Mey was inspired by the fact that computers are fitted
with fans to dissipate the heat their chips generate ‘I had the idea of
using this heat to create tiny warm spots which blind people could sense
with their fingertips.’

Erwin De Baetselier, an engineer in De Mey’s lab, helped construct a
prototype which mimics printed Braille. ‘Instead of raised dots of paper,
we use resistance elements, printed as thick films, on a ceramic substrate.’
The dots can be as small as a quarter of a millimetre across, and they are
heated in brief pulses to keep the ceramic itself from heating up.

The dots, arranged in the traditional Braille grid of two columns of
three, are controlled by a computer. They are heated to between 30 °C
and 50 °C for reading, and stay warm as long as a finger is resting
on them. De Mey’s device produces just one character, but the elements are
cheap, so he expects to build a display producing 80 characters – one line
of a computer screen – within a year or two, ‘depending on support from
¾±²Ô»å³Ü²õ³Ù°ù²â’.

The real advantage of the thermal display, De Baetselier says, will
be its ability to display graphics and text at the same time. ‘With our
technology, we could reproduce the entire screen, both text and graphic
symbols, making it possible for blind people to read icons and use a mouse.’

People whose fingertips are heavily calloused by extensive reading of
Braille may not be able to sense the thermal images. However, one person
who has tested the system could read the thermal output better than printed
Braille.

The thermal Braille screen should also be cheaper than existing devices.
An 80-character, one-line device using magnetic pins costs around £7000,
while a thermal display, able to reproduce an entire screen, ought to cost
£350 at most, its developers say. Reliability ought also be better,
too, as there are no mechanical components.

Ozone is an unstable form of oxygen gas with molecules made up of three
oxygen atoms rather than the usual two. It reacts violently with organic
materials, destroying itself and oxidising the organic matter. John Gallo
of O₃ Tech, the company based in Fort Pierce, Florida, that developed
the machine, explains that in typical household laundry, 85 per cent of
stains can be oxidised. The rest are either permanent or must be treated
in other ways.

Most of the company’s customers are hotels and prisons, and the machine
will remove 95 per cent of their stains. ‘In only ten minutes, in cold water,
the ozone in our washers gets rid of all the organic soil, and the dirt
is then filtered out of the wash water, so the water can be reused,’ Gallo
says. Non-oxidisable stains, such as saturated fats and grease, are rewashed
in alkali.

Gallo says that the prisons using the machine benefit because time is
saved by not having to rinse the laundry to remove detergent. ‘Our machines
wash three times faster than conventional laundries,’ he says. In addition,
Gallo says, the prisons’ sheets and uniforms last much longer because they
are not subjected to the heat and harsh chemicals of a traditional wash.

The company says its machines have been made practical by refinements
to generators that produce ozone. The gas is produced by splitting up oxygen
molecules with an electric discharge. Some of the atoms then recombine as
ozone. Previous systems required pure oxygen or dried air as their raw material,
but O₃ Tech has developed a generator which can use ordinary
air, making ozone cheaper.

A chip that stores electrical charges by letting them ‘surf’ on sound
waves promises much faster processing of analogue signals. The chip acts
as a delay line which breaks up analogue signals into packets of charge
of different sizes and carries them along on an acoustic wave moving through
the chip. This stores the analogue information very briefly until it is
needed, something that is very difficult to do in conventional chips.

Analogue signals are waves: sound, television and radio signals are
all analogue. Computers have trouble storing analogue information because
of the wide range of values it can take. Instead, computers normally break
down the signals into digital code: at regular intervals they measure the
height of the analogue wave and store it as a digital number. Computers
are far more comfortable dealing with digital data but they then need to
be very powerful to deal with the large volumes of data produced by converting
an analogue signal into a digital one.

The new chip, developed by Comlinear of Urbana, Illinois, uses a technique
called acoustic charge transport (ACT) to store an analogue signal briefly.
According to Daniel Fleisch of Comlinear, the ability to delay a signal
for a fraction of a second, so that it can be compared to a second signal,
lies at the heart of signal processing.

This will be important in fields such as pattern recognition and neural
networks, which are currently limited by processor speeds. In terms of the
speed of processing signals, the technique is ‘three or four orders of magnitude
ahead of digital technology,’ Fleisch says.

The ACT chip is made of gallium arsenide, a semiconductor, which contains
impurities along its top surface to improve its conductivity. At one end
of the chip a surface acoustic wave is generated which moves through the
material at a speed of 2.9 kilometres per second.

Gallium arsenide is also a piezoelectric material which means that when
it is put under stress it produces an electric field. The stresses caused
by the compression and decompression of the acoustic wave create waves of
electric field that move along with the acoustic wave. The depressions of
this electric field wave act as ‘buckets’ that can hold electric charge.

The analogue signal is dropped into these buckets from a contact on
the surface of the chip. As each bucket passes the contact, a quantity of
electrons are injected into it which represent the height of the analogue
signal at that time. This charge is then carried away down the chip.

Further along the chip are several hundred ‘tap’ electrodes which measure
the content of each bucket without changing it. The delay in the signal
at any tap electrode depends on how far it is from the input contact. The
longest possible delay comes from the end contact which retrieves the signal
from the buckets. The buckets pass under the tap electrodes at a rate of
360 million per second.

Using this type of simple analogue circuitry, Fleisch says, ‘signal
correlation can be performed as much as one thou-sand times faster than
on today’s fastest digital signal processors, and at a much lower energy
³¦´Ç²õ³Ù.’

Comlinear expects its first products using ACT chips will be hard disc
drives for computers with capacities as large as 1 gigabyte. Fleisch expects
them to be on the market later this year.

Her system incorporates a robot arm that positions 100 optical fibres
at the focus of a large optical telescope. Each fibre gathers light from
a separate source and channels it to a spectrograph 40 metres below. ‘With
big telescopes being far oversubscribed, this instrument quite literally
allows you to do kinds of astronomy that you can’t even think about tackling
if you have to look at one source at a time,’ Brodie says.

Brodie is now testing her multi-object spectrograph on Lick’s telescope
with a 1-metre wide mirror. First she calculates the positions of the faint,
distant galaxies that she wants to observe. A computer programme then determines
how to position the 100 optical fibres on a metal plate at the primary focus
of the telescope.

The end of each fibre is held in place on the plate with a magnet 3
millimetres across. A prism on each magnet bends the incoming light through
90 degrees and feeds it into the fibre.

The robot arm can reposition all 100 fibres with an error of less than
10 micrometres in about five minutes. This manoeuvre can be carried out
while the telescope is being moved from one position to another. When Brodie
moves her instrument to Lick’s 3-metre telescope, she will be able to get
spectra from hundreds of galaxies in a single night’s observing time.

According to Dan Fabricant, an astronomer from the Harvard-Smithsonian
Center for Astrophysics, similar robotic fibre systems are now being used
in Australia, the US, Chile and elsewhere. But Brodie’s system, which was
entirely custom-built, is among the most accurate in existence.

Bev Oke, at the California Institute of Technology in Pasadena, has
built a system with 176 fibres for the 5-metre Mount Palomar telescope in
California. Unlike Brodie, he is working at the secondary focus of the telescope
– after it has been reflected off two mirrors.

He has put his entire apparatus, weighing 680 kilograms, including the
spectrograph, on the telescope. This means that his fibres are only a few
centimetres long. Also, Oke’s positioning system can afford to be less accurate
because the secondary focus has a much wider focal plane. This means the
images of the galaxies will be further apart.

The first astronomer to build a fibre-robot system, John Hill of the
University of Arizona in 1986, chose a very different approach. Hill’s device
has 32 separate robotic arms, each with its own two-motor drive, and each
with three fibres at its tip. Hill says that his instrument can change the
position of every fibre in just 45 seconds. But an instrument with hundreds
of fibres would involve so many arms and motors that it would become unacceptably
heavy and cumbersome.

According to Thomas Wdowiak, an astrophysicist at the University of
Alabama at Birmingham, the theories that invoke asteroid impacts to explain
the extinction of the dinosaurs 65 million years ago do not adequately tackle
the question of how dinosaurs actually perished. He believes that the soot
injected into the atmosphere by a disintegrated asteroid was laden with
nickel, which was soluble in water and very poisonous to plants.

The hypothesis about asteroids that is accepted by most researchers
proposes a ‘double-barrelled’ effect, Wdowiak says. The initial impact produced
a giant explosion and earthquakes across the planet. The theories say that
the dinosaurs were dealt a final blow by the debris and soot which was blasted
into the atmosphere and blocked sunlight for years. This would have changed
the climate drastically, affecting both flora and fauna. Wdowiak believes
that nickel poisoning of plants provides a more satisfactory explanation
for the animals’ deaths.

Wdowiak credits several people for the theory, including an 18-year-old
student. Stewart Davenport, of Altamont School in Birmingham, came to Wdowiak
two years ago and asked for help in obtaining chemicals for a science project.
He needed the chemicals in order to create a solution for growing plants
hydroponically.

At the time, Wdowiak was analysing some meteorite fragments for NASA.
He suggested that Davenport should use some of the material for his hydroponic
solution, recalling that previous work had shown that plants grow faster
in lunar ‘soil’, which is similar to meteoritic material.

Davenport made up his solution and planted radishes in it. But to the
surprise of Davenport and Wdowiak, the plants did not do well at all. ‘The
seeds would sprout,’ says Wdowiak, ‘but instead of coming out as little
green plants, they would be brownish yellow because they lacked chlorophyll.’

He found that nickel in the solution prevented the radishes from making
chlorophyll.

‘Everybody knows meteorites have a lot of nickel,’ says Wdowiak, ‘but
what was not appreciated is that most of the kind of nickel in this meteorite
is water-soluble.’ This means that it can get into plants very easily, he
says.

At this point, Wdowiak brought in Dan Jones, chairman of biology at
the University of Alabama. Jones, a botanist, confirmed their finding that
meteorite material can poison plants. But it was Wdowiak’s wife, Patricia,
who made the connection between nickel poisoning and mass extinctions. One
evening at dinner, when Wdowiak was talking about the discovery, she said:
‘Maybe that’s what killed the dinosaurs?’

‘There’s a difference between mass killing and mass extinction,’ Wdowiak
contends. ‘I call our theory the nickel coup de grace.’ There was mass killing
at the eruption of Mount St Helens, he says, but no species became extinct,
because plants sprouted out of the ash the following year. ‘But nickel poisoning
would ensure extinction,’ he says. ‘It would be like a biblical salting
of the earth.’

The body that wiped out the dinosaurs would have been about 10 kilometres
in diameter. According to Wdowiak, such a body would have a density of 3000
kilograms per cubic metre. Spread over the Earth’s surface, he says, this
would amount to 3 kilograms per square metre, corresponding to a layer a
few millimetres thick.

According to Wdowiak, this fallout would contain between 130 and 1300
parts per million (ppm) of nickel. ‘The usual concentration in soils is
15 ppm,’ he says, ‘and 40 ppm is toxic.’

Wdowiak’s work will be published later this year by the Geological Society
of America as part of Special paper No 247, Global Catastrophes Proceedings.

ICE that builds up on aircraft wings and helicopter rotors is one of
the most serious hazards for winter flying. But an engineer from NASA has
invented a new kind of de-icer that works faster than existing systems while
consuming less power. The invention comes from Leonard Haslim, a former
pilot from the US Navy who now works at NASA’s Ames Research Center in California.

The heart of the new de-icer, known as an electro-expulsive separation
system, is two conductive strips embedded in a ribbon of polyurethane plastic
about 2.5 centimetres wide and a millimetre thick. The EESS is attached
to the leading edge of an aircraft’s wing, where ice forms. Once the device
is switched on the two strips, 0.1 millimetres apart, are charged so that
electromagnetic forces cause them to repel each other. This causes the plastic
strip to bulge, cracking any ice that has formed on it.

A 1000-ampere pulse produces a force equal to 2240 pounds per linear
foot (3 kilonewtons per metre), Haslim reports. The pulses pass through
the strips thousands of times per second, so the strip vibrates and the
ice disintegrates.

NASA claims that the new de-icer has several advantages over others.
So-called pneumatic boots, which are pumped full of air to crack the ice
above them, work only with comparatively thick layers of ice, and the large
pieces that are cracked off can damage engines and wing surfaces. And electrothermal
de-icers, which melt ice off the plane’s wings, need 25 000 watts per square
metre of their surface.

The EESS will remove any ice, ranging from a coating of frost to a layer
2.5 centimetres thick, and will reduce the ice to tiny fragments. Since
it uses a strong pulsed current, rather than a constant one, the EESS needs
only 50 watts per square metre.

Since the active part of the EESS is just a thin, flexible ribbon, engineers
can apply it to wings, tails and even the leading edges of helicopter rotors,
a position where pneumatic and thermal boots are too large to fit. And its
reduced power consumption can also make a dramatic difference to the aircraft’s
load. For instance, says Haslim, the US Defense Department’s new V-22 tilt-rotor
plane, which takes off like a helicopter then flies like a plane, carries
91 kilograms of generators to power its thermal de-icers: ‘The electro-expulsive
system would reduce their power requirements for de-icing from 50 kilowatts
to 500 watts, and save at least 180 pounds (80 kilograms) in the process,’
he says.

Haslim foresees a variety of other applications for electro-expulsive
devices. With their low demand for power and their relative cheapness, he
says, they could be used to remove ice from ships’ superstructures and from
the surface of motorway bridges. He believes they could also be adapted
to shake cakes out of tins at commercial bakeries.

The fire ant is noted for its fierce sting – hence its name – and for
the aggressive manner in which it defends its nest. It was accidentally
introduced into North America from South America in the 1930s. For decades,
the US Department of Agriculture has been fighting the ant. According to
its figures, the ant infests about 1 million hectares throughout 10 states,
as well as Puerto Rico.

The nest of a fire ant is a hard mound between 20 and 30 centimetres
tall. Any insect predator that attempts to enter the nest is usually attacked
and killed. Orasema, however, has developed a method of entering the nest,
eating the ant larvae, and escaping unharmed.

According to Donald Jouvenaz, an entomologist at the USDA in Gainsville,
Florida, adult wasps lay their eggs on the leaves of plants, and wasp larvae
– smaller than the head of a pin – emerge within a day. The larvae then
attach themselves to unsuspecting fire ants while the ants are foraging
for food. The ants unwittingly carry them back to the nest.

In the nest, fire ants come into close contact frequently. This means
that the larvae are transferred quickly. ‘It’s like getting off one horse
and on another,’ says Jouvenaz. According to Robert Vander Meer, a chemical
ecologist working with Jouvenaz, it’s during this time that the wasp ‘passively
acquires’ the odour of the colony or nest.

Researchers at Vander Meer’s laboratory have used the technique of gas
chromatography to analyse the odour-bearing chemicals on the cuticle, or
skin, of the wasps and ants. Vander Meer likens the nest odour to a fingerprint.
He says that no two colonies of ants will generate the same chemical signature.

The scientists found that ‘fingerprints’ of both the ant colony and
the wasp larvae inside it are identical. A wasp larva does not start out
‘smelling’ the same, so they conclude that it must acquire the colony odour.

Eventually, the wasp larva is transferred to the developing ants, known
as the brood. It burrows into an ant larva, and goes dormant. ‘It seemed
to be waiting,’ says entomologist Daniel Wojcik. ‘But for what, we asked
´Ç³Ü°ù²õ±ð±ô±¹±ð²õ.’

Wojcik believes that the wasps are waiting to sense certain ant hormones.
‘Those hormones tell the larva, ‘OK, now I have an ant host that’s big enough
to allow me to fully develop’,’ he says. When it senses this, it begins
to feed on the ant brood. Once it has consumed an ant pupa, the wasp leaves
the nest to mate and lay eggs.

The damage the wasp does to a fire ant colony is not significant enough
to make it an effective control. But, according to Vander Meer, any parasite
or predator that puts an added burden on the ants is worth examining. Fire
ants have thrived and spread, despite decades of massive programmes to spray
insecticide.

The USDA is also testing poison baits, fungal pathogens and lures based
on fire ant pheremones to fight the insects.