Page 1840 – Physics World

Home » Page 1840

Hawking scoops book prize

The award was announced last night at the Science Museum in London by the chair of the panel of judges, Raj Persaud, a consultant psychiatrist at the Maudsley Hospital in London. “This book made a real effort to enliven the subject through readable text and clear illustration,” he said. “It has the production values of a real 21st century book. Even if you don’t understand the entire book, you will still gain so much by a story told by an extraordinary mind.”

The prize confirms that Hawking has succeeded in writing about theoretical physics in layman’s terms in The Universe in a Nutshell. In contrast, A Brief History of Time – which sold over 25 million copies worldwide – was widely considered to be too difficult for a general readership.

But some people remain unconvinced by Hawking’s latest book, which is published by Bantam. Writing in The Guardian, critic Mark Lawson said that he “understood even less of it than I did A Brief History of Time.”

Hawking has been Lucasian Professor of Mathematics at the University of Cambridge since 1979, the post that was held by Newton and Dirac. He celebrated his 60th birthday in January with a meeting organized by the university to reflect on Hawking’s contributions to cosmology and theoretical physics.

Buy the book
The Universe in a Nutshell: Amazon UK/Amazon US

KEK names new leader

The research programme at the KEK lab – which dates back to 1955 – focuses on particle physics, nuclear physics and materials science. Its chief experimental facility is the Belle B-factory, in which electrons and positrons are collided to establish the difference between matter and antimatter.

Another major project at KEK is the K2K experiment in which a neutrino beam is sent 250 km to the SuperKamiokande detector at the Kamioka Observatory. This enormous underground experiment can also detect solar and atmospheric neutrinos. Totsuka pledged last year to rebuild the detector following an accident that destroyed most of its photomultiplier tubes.

KEK is also competing with the DESY lab in Germany and Fermilab in the US to host the next big machine in particle physics – a 500-GeV linear collider. This experiment would collide electrons and positrons at higher energies than ever before in order to explore physics beyond the Standard Model.

Silver foils the diffraction limit

According to the theory of diffraction, only a tiny amount of light can pass through a hole that is narrower than the wavelength of the light. Also, the light that is transmitted is diffracted in all directions. These effects limit the minimum size of many optical devices and techniques, such as the creation of features on semiconductors by optical lithography and the efficient coupling of light into optical fibres.

Now the Strasbourg-led team has found a way to shine more light through a tiny aperture, and to channel it into a collimated beam. Lezec and co-workers created a sub-wavelength aperture in a thin silver film and etched a periodic pattern of grooves around it using a focused ion beam. This corrugated metal surface supports the excitation of surface waves known as plasmons that soak up the incident light. Previous theoretical studies have suggested that these plasmons squeeze through the hole and are converted back into light on the other side. This enhances the optical transmission of the film.

The researchers found that the wavelength of the transmitted light depends on the spacing of the grooves in the film. By patterning the reverse side of the film, they also discovered that the light emerges from the hole as a tightly focused beam that can propagate with very little divergence. The team then found that the direction of the transmitted light could be controlled by changing the symmetry of the periodic pattern.

The technique may be useful in a variety of nanoelectronics applications, including optimizing near-field devices for microscopy or data storage, and improving optical devices such as light-emitting diodes (LEDs) and semiconductor lasers.

Ebbesen told PhysicsWeb that the group now plans to understand the subtleties of the physics of the thin films, but remains tight-lipped about prototype devices under investigation.

Magnetic spins to store quantum information

Physicists have long known that lithium holmium fluoride is a ferromagnet, in which the atomic spins are permanently aligned even when there is no external magnetic field. But if the holmium ions are gradually replaced with yttrium ions, this ferromagnetism is suppressed and eventually disappears. This happens because the material becomes increasingly disordered, and eventually becomes a ‘spin glass’ in which the alignment of the spins is random.

But previous studies have shown that – in contrast with theory – the addition of further yttrium ions destroys this ‘glassy’ state, and that the material becomes progressively more ordered, especially if its temperature is reduced to near absolute zero. This unusual state is dubbed an ‘anti-glass’.

Rosenbaum and colleagues set out to investigate the magnetic properties of this anti-glass state by cooling a centimetre-sized crystal of lithium holmium yttrium fluoride to temperatures of just tens of millikelvins. Then they switched on an oscillating magnetic field and measured the magnetization of the sample for a range of oscillation frequencies, and at several temperatures.

In most disordered magnets, the magnetic susceptibility falls as the temperature drops. But Rosenbaum and colleagues found that the susceptibility of their crystal increased, indicating that the atomic spins had become more closely aligned, or coherent. The shape of the spectra also suggested that the spins in small clusters of atoms had aligned to form ‘oscillators’, which could take on either an ‘up’ or ’down’ collective spin.

These oscillators – each of which contained about 260 atoms – kept the same spin for up to ten seconds. The researchers think that this alignment arises because the oscillators can flip between the two possible spin states by ‘tunnelling’ through the potential barrier that separates them. This could be evidence for quantum behaviour because it cannot be explained by classical theories of magnetism.

Rosenbaum and co-workers now suggest that it could be possible to encode bits of information onto these two-state oscillators using a magnetic field, and that the states could be ‘entangled’ and used in quantum information processing.

Laser lithography makes cheaper chips

The components on a microchip are made by carving patterns into layers of doped and undoped silicon. In the standard technique, light is shone through a stencil onto a silicon wafer that is coated with a light-sensitive polymer known as a resist. Chemical etching then removes the regions of silicon coated with either the unexposed or the exposed polymer, until the desired structure is achieved. Finally, the remaining polymer is washed off.

But such ‘photolithography’ is expensive and complex, and the resolution of the technique is fast approaching the diffraction limit. This means that it will not be able to make features much smaller than the current minimum size of about 130 nm – and that the semiconductor industry could soon violate one of its guiding principles, known as Moore’s Law. Coined in 1965, this law described how the density of components on a chip doubled every 18 months, and was soon adopted by the semiconductor industry as a target.

Now Moore’s Law could be back on track. Chou and co-workers say that their technique – known as laser-assisted direct imprint – can create features as small as 10 nm on silicon wafers. The new process also eliminates the need for the resist and washing steps.

The team inscribed the pattern into a quartz mould using ‘impact lithography’, which is much cheaper and easier than the photolithography of silicon. The mould was then placed on top of a silicon wafer. A pulse of light from a helium-neon laser – with a wavelength of 633 nm – was then fired through the mould to melt the top layer of silicon. The researchers then pressed the mould into this liquid silicon and removed it after the silicon had solidified, leaving the pattern.

Since silicon reflects more light as a liquid than it does as a solid, the team could tell when the mould needed to be removed by measuring how much of the laser pulse was reflected. The researchers also say that the mould can be used several times.

The Princeton team now hopes that its technique will be taken up by the semiconductor industry. “A prototype machine could be available in three years,” Chou told PhysicsWeb. The team also believes it should work on silicon wafers up to eight inches in diameter, and could be adapted for other materials and processes. “There are some technical issues in scaling up, but there are no fundamental issues,” he says.

Jets unravel nebula puzzle

When a Sun-like star reaches the end of its life, its outer layer blows off into space to form a gas shell around its core, which contracts into a white dwarf. Ultraviolet radiation emitted by the white dwarf then excites the gas shell so that it lights up to become a visible planetary nebula. Astronomers have studied many of these nebulas but have long been stumped by the fact that they are rarely spherical, although the explosions that produce them are thought to be perfectly symmetrical.

To shed light on the puzzle, Imai and colleagues used America’s Very Long Baseline Array of telescopes to study the region around a star known as W43A. The star is about 8500 light years from Earth in the constellation Aquila, and astronomers believe it is about to produce a nebula.

The astronomers studied the motion of the gas by measuring the microwave radiation emitted by the water vapour that it contains. An energy transition between two rotational states of water molecules corresponds to microwaves with a wavelength of 22 GHz, which enables the molecules to amplify such radiation. Such clouds of water vapour are dubbed ‘masers’, because they act as microwave lasers.

By combining the microwave signals gathered by the ten telescopes in the array, Imai and co-workers were able to calculate the speed and direction of the clouds of water vapour. They found that two streams of water vapour are moving outwards from the star in opposite directions, at about 600 000 kilometres per hour.

“The path of the jets is curved like a corkscrew, as if whatever is squirting them out is slowly rotating,” says team member Philip Diamond of Jodrell Bank Observatory in the UK. The researchers believe that this precession could explain why many nebulas have irregular shapes. They are unsure why the jets are precessing, but speculate that a companion star or strong magnetic fields could be the cause.

The researchers say they are lucky to have caught W43A during this brief transition period. According to Diamond, the water vapour would be destroyed by the radiation emitted when the star collapses to become a white dwarf.

Incidentally, planetary nebulas have nothing to do with planets – the term is a misnomer that dates back to their discovery in 1764.

Single photons to soak up data

Previous demonstrations of quantum communications systems based on photons used their spin angular momentum states to store information. There are two of these states, corresponding to the vertical and horizontal polarization of the spin, so a photon can take on a logical value of “1” or “0” to act as a quantum bit of information, or ‘qubit’. The photon can also exist in a ‘superposition’ of both states at the same time.

But the orbital angular momentum of a photon can take on an infinite number of values. Since a photon can also exist in a superposition of these states, it could – in principle – be encoded with an infinite amount of information. But so far physicists have failed to measure the orbital angular momentum of individual photons, and this has hindered attempts to use the state for data storage.

Now Padgett’s team has used a sequence of interferometers to sort single photons into four different orbital angular momentum states. They split a beam of light into two, rotated one of the resulting beams through 90° with respect to the other, and recombined them to produce an interference pattern.

If the original beam had an even value of orbital angular momentum, the beams interfered constructively, but if it had an odd value, they interfered destructively. This enabled the team to separate the photons with even and odd values of orbital angular momentum into two different detectors.

By passing both of these beams through the process again, the researchers separated photons into four different values of orbital angular momentum. Using filters, they also reduced the intensity of the beams so that on average, there was only one photon at a time present in each interferometer.

According to the researchers, this technique could – in theory – be extended to detect any number of different orbital angular momentum states, although they concede it could quickly become cumbersome. The Glasgow team now plans to refine its technique by using single-photon emitters and detectors.

Magnets open the gate to nanoscale logic

Logic operations – such as NOT, AND, OR and XOR – are carried out in existing electronic circuits by semiconductor devices including diodes and transistors. But the density of electrons flowing through a semiconductor is limited partly by the separation of dopant atoms, and this restricts how small these devices can be made.

The electron density of a metal, however, is higher than it is in a semiconductor, so a metallic logic gate could be made smaller than a semiconductor logic gate. Although magnetic materials are widely used to store data, they have not so far been used to perform operations on it.

Now the Durham group has taken a step towards a magnetic logic system with the development of its NOT gate – a simple logic device that gives an output of “0” for an input of “1” and vice versa.

The device was built using a ferromagnetic wire, in which the spins on the electrons are aligned even when there is no external magnetic field. This alignment persists over microscopic regions known as domains, which are separated by ‘walls’ that are typically hundreds of nanometres thick. In these walls, the spins gradually change from the orientation of one domain to that of the neighbouring one.

Previous experiments have shown that a domain wall can be propelled along a nanoscale wire by a magnetic field that rotates in the plane of the wire. But Cowburn’s team realized that if they put a hairpin bend in the wire, the wall would only be able to move around the bend in one direction for a given direction of rotating field. Such one-way signal flow is a key requirement of any logic system.

To test their idea, they made a wire 200 nm wide and 5 nm thick from a ferromagnetic alloy of nickel and iron, and bent it. The wire contained one domain wall near one of its ends, and the researchers assigned the logical values of “1” and “0” to the magnetization states either side of it. With one revolution of a magnetic field rotating in the anti-clockwise direction, the wall travelled around the bend. This inverted the magnetization state at the bend and therefore the logical value assigned to it.

“We have achieved digital logic without the equivalent of a transistor,” Cowburn told PhysicsWeb. “We’re pretty excited about being able to do logic by such radically different means.”

The team then linked together eleven of the magnetic NOT gates to make a simple 13-bit shift register, in which a bit of information is passed from one gate to the next each time the rotating magnetic field completes one cycle. Cowburn adds that the data stored in such devices would be stable even without a power source, which could make it ideal for mobile applications like smart cards and phones.

Tiny transistors nudge the nanoscale

Transistors are ubiquitous in today’s silicon-based microelectronics industry. The three-terminal devices act as switches because the current flowing from the source to the drain electrodes can be turned on and off by a voltage applied to the gate electrode. Transistors are usually made from layers of different semiconductors, and the current that flows through them consists of around a billion electrons. But in the new devices, electrons can hop one at a time from the source to the drain electrodes across a ‘bridge’ made from the single atom or molecule.

Paul McEuen and co-workers at Cornell University and the University of California at Berkeley based their transistor on a single cobalt atom suspended in an organic compound (J Park et al 2002 Nature 417 722). They deposited a gold wire about 10 nm thick onto a silicon substrate and then coated it with the complex. The team then made a gap in the wire about 1 nm wide, so that the exposed tips formed source and drain electrodes. Molecules of the organic complex then slipped into this gap, taking a cobalt atom with them. A layer of silicon dioxide insulated the gap region from the silicon substrate, which acted as the gate electrode.

In a similar way, Hongkun Park and colleagues at Harvard University and the University of California at Berkeley trapped divanadium molecules between gold electrodes – also 1 nm apart – and used aluminium oxide to isolate the gap region from their silicon gate electrode (W Liang et al 2002 Nature 417 725).

Both teams then measured the current flowing from the source to the drain as they varied the voltage between these electrodes for a range of gate voltages. For each device, the researchers found that the current flowed only at certain gate voltages, which were related to the energy that an electron needs to hop on – or off – the bridging atom or molecule.

This behaviour is the hallmark of a single-electron transistor, but both groups also found that more electrons took part in the current flow when they placed their devices in a magnetic field. The researchers believe that this phenomenon arises from the ‘Kondo effect’, in which a single magnetic entity – such as a cobalt atom or a divanadium molecule – interacts with the electrons in the non-magnetic material around it.

McEuen’s and Park’s teams also found that the exact electronic properties of their transistors could be tuned by changing the chemical make-up of the organic compounds. This determines the quality of the electrical connection between the atom or molecule and the electrodes.

Both groups are now investigating these effects in transistors based on other atoms and molecules. “We are already working on half a dozen different single-molecule transistors that incorporate different molecules, including single-molecule magnets,” Park told PhysicsWeb.

Lasers propel paper aeroplanes

When a laser beam hits an object, atoms can be ejected from its surface in a process known as laser ablation. These atoms depart with a certain momentum, and impart an equal and opposite momentum to the object. This led scientists to suggest that a laser on Earth could be used to propel a spacecraft with an ‘ablation target’ attached to it. In recent years, the potential of the technique has been proved by the launch of several miniature rockets, each weighing tens of grams.

Now Yabe’s team has shown that the technique could also be used to fly small aircraft. They successfully propelled two small aeroplanes – which were several centimetres long and weighed 0.1 and 0.2 grams – up to speeds of 1.4 m s^-1.

The ablation targets on the aircraft were both made of a piece of aluminium foil 0.1 mm thick and with an area of several square millimetres. One target was coated with clear acrylic, and the other had a layer of water placed on it. Load cells on each plane recorded the push produced when a pulse from an yttrium-aluminium-garnet laser was focused onto the targets. Each pulse supplied 590 mJ of energy and lasted 5 ns.

According to the team, the two-layer structure of their targets is the key to the success of the aeroplanes. In simulations, they showed that in single-layer targets – like those used in used in earlier attempts – most of the laser energy is absorbed by gas atoms just above the surface of the target. This leaves just a few percent of the energy to drive the target, and only a small impulse is generated.

But in the layered targets, the simulations showed that the gas atoms ejected from the target get trapped between the layers and impart a much greater momentum to the target as they are subsequently forced out.

Yabe and co-workers found that their aeroplanes received impulses three times greater those received by aeroplanes with single-layered targets. But they say that they need to find a way to continually refresh the target – or make the whole aircraft from the ablation material – before the aeroplanes can be developed into a practical device. They even speculate that a laser could be used to periodically deform a shape-memory alloy on the aircraft, without the need for ablation.

Browse articles by content type