Page 1676 – Physics World

Home » Page 1676

Nanoantenna focuses laser light

Many structures within living cells are less than one micrometre in size and therefore biologists are keen to have a tabletop microscope that can analyse samples at resolutions of hundreds of nanometres or less. However, the spatial resolution of a conventional microscope is always larger than the wavelength of the light used — which is typically several micrometres in a modern Fourier transform infrared microscope.

Now, Federico Capasso and colleagues at Harvard University and Agilent Laboratories have worked out a way to focus infrared light from a quantum cascade laser to a spot 100 nm or less in size. Such a spot could, in principle, be scanned across a sample to generate an image with nanoscale resolution.

Quantum cascade lasers use multiple alternating layers of high- and low-band-gap semiconductors to create a series of electron “traps” or quantum wells. The energy levels of the wells are such that electrons “cascade” through the device, emitting a photon at each step. Unlike conventional semiconductor lasers, quantum cascade lasers can be tuned to a wide range of emission wavelengths, and are ideally suited to operating at the mid-infrared wavelengths favoured for biochemical applications.

The trick according to Capasso, who is co-inventor of the quantum cascade laser, is to focus the light by adding an optical antenna to the laser. Such an antenna consists of two small gold rods that lay end-to-end separated by a tiny gap. The laser light’s electric field causes the electrons in the nanorods to oscillate and accumulate at the gap ends. This creates an intense spot of light in the gap, and this light could be in principle, shone onto a sample.

The team put antennas on two quantum cascade lasers that had emission wavelengths of 7 μm and 5 μm. In each case, sharp optical spots were observed, each having dimensions comparable to the antenna gaps (100 nm and 70 nm respectively). “Typical Fourier transform interferometer microscopes have a spatial resolution of 10-20 μm. We can go two orders of magnitude below that,” said Capasso

Building a microscope from the modified quantum cascade laser will be the next step, Capasso told physicsworld.com. The localization of light in the gap is a “near-field” effect. To use the light for imaging or spectroscopy, samples would need to be brought very close to the gap.

New supersolid signature sighted

Supersolidity was first predicted in 1969 by Russian theorists Alexander Andreev and Ilya Liftshitz. They said that lattice vacancies in solid helium could all collapse into the same quantum state if the helium were cooled to an extremely cold temperature. This Bose-Einstein condensate (BEC) of vacancies would behave as a coherent entity, moving throughout the rest of the solid like a superfluid.

In 2004 Chan and his graduate student Eun-Seong Kim found the first evidence for superfluidity in a torsional oscillator, which consisted of a cylindrical cell filled with high-pressure helium-4. The cell, which was suspended from a rod, was rotated back and forth while being cooled. When the temperature reached about 200 mK, the researchers saw a sudden change in the oscillation period of the cell, which they interpreted as evidence that about 1% of the helium had “decoupled” from the solid helium-4 and was not oscillating.

Although this was taken as a sign of supersolidity, subsequent experiments by Chan and others revealed that the decoupling temperature and the percentage decoupled varied significantly from sample to sample. This led some physicists to suggest that the observed effect was not caused by condensing vacancies, but rather by either the flow of superfluid helium along grain boundaries in polycrystalline samples, or by a transition to a “superglass” phase. Earlier this year Chan and colleagues repeated their torsion experiments on a single-crystal of helium-4 – which had no grain boundaries. This time the decoupling occurred at the lower temperature of 75 mK with only 0.3% of the sample decoupling – the strongest evidence yet that the vacancies became a superfluid.

Now Chan and colleagues have measured the specific heat – the energy required to change the temperature of a material – of several helium-4 samples. They found peaks in the specific heat at about 75 mK, which they concluded to be a “probable” signature of the supersolid phase. “If there is a real phase transition from the normal solid to the supersolid phase, there should be a thermodynamic signature, like a peak in the specific heat of solid helium”, explained Chan. However, the experiment was not designed to measure the simultaneous decoupling of mass, and therefore the team cannot be certain that the peak is associated with supersolidity.

The team also found that the temperature dependence of the specific heat was at odds with that expected if a glassy state was forming in the solid – ruling out the superglass explanation for decoupling.

The peak in the heat capacity suggests that the supersolid state emerges in a second-order or continuous phase transition – just like superfluidity. “There is no theoretical consensus on the nature of the supersolid phase”, said Chan, “[but] based on our knowledge of superfluids, the transition is more likely to be second order, than say first order”.

Chan told physicsworld.com that the team have incorporated new and more sensitive thermometers into their apparatus, which should allow them to get a better idea of the exact shape of the peak. By analyzing the shape, the team could get a better understanding of the nature of the phase transition. Chan is also keen to look for other signatures of supersolidity such as “second sound”, which is a dramatic increase in the thermal conductivity of a material when it becomes a superfluid.

China launches mission to the Moon

The satellite is named Chang’e 1 after the Chinese goddess of the Moon and was launched yesterday from the Xichang centre in the province of Sichuan, south west China.

Weighing over 2300 kg, the spacecraft is carrying a variety of instruments that will help it map the lunar surface including a CCD camera and imaging interferometer. The presence of radioactive isotopes on the surface will be detected by a gamma-ray spectrometer and a high-energy particle detector. Chang’e 1 will operate 200 km above the surface of the Moon in a low circular orbit and will perform measurements for one year.

The launch is the latest move in an Asian space race. Last month Japan launched a lunar probe, which is currently in orbit around the Moon. India plans to launch a satellite early next year with the aim of creating a high resolution 3D map of the Moon’s surface.

New isotopes push out the drip-line

The work gives the first firm experimental evidence for the discovery of the bound Mg isotope, magnesium-40, a so called even-even isotope containing 12 protons and 28 neutrons, and the odd-odd Al isotope aluminuim-42 containing 13 protons and 29 neutrons. The experiments were carried out by researchers at the National Superconducting Cyclotron Laboratory (NSCL) at Michigan State University by firing calcium-48 nuclei into a tungsten target. The resulting fragments could be identified by separating them first according to their momentum-to-charge ratio and then stopped in detectors which measured their energy.

Although magnesium-40 had been predicted to exist from the two best “global” theoretical models, the semi-classical finite range droplet model (FRDM) which uses a semi-classical description of the macroscopic contributions to the nuclear binding energy and the Hartree-Fock-Bogoliubov (HFB) model; a quantum mechanical model. However, both models do not predict the existence of a bound aluminuim-42 isotope and the mechanism of binding a aluminuim-42 isotope is currently not understood. “Seeing [aluminuim-42] is surprising, as usually this close to the drip-line binding occurs for an even number of neutrons”, says Michael Thoennessen, a Associate Director for Nuclear Science at NSCL.

The team also detected hints of an even heavier aluminium-43 isotope. Indeed, a more recent variation of the HFB model, named HFB-9, suggests that an aluminium-48 isotope could exist, with 13 protons and 35 neutrons. However, current experimental facilities may not be able to probe the limits of this modified drip-line towards heavy nuclei such as aluminuim-48, and heavier neutron-rich elements need to be used in the projectile fragmentation method instead of calcium-48, so that the nuclei that break off from the fragmentation contain more neutrons. “We need the next generation accelerators to determine the drip-line for aluminium and beyond to other elements like silicon and phosphorous,” says Thoennessen. Such a chance could come from the Radio Isotope Beam Factory (RIBF) at RIKEN in Japan, which just started initial operations earlier this year.

Electrons timed with attosecond accuracy

The classical Bohr theory of the atom predicts that an electron takes about 150 attoseconds (as) to “orbit” a hydrogen atom. The atomic nuclei move much more slowly, which means that attosecond spectroscopy can be used to study electron behaviour while the atoms are essentially frozen in time.

While attosecond spectroscopy of atomic gases have been possible for some time, similar experiments on solids had been limited to about 10 fs (10^-14s) resolution. Now, Ferenc Krausz and colleagues at the Max Planck Institute for Quantum Optics in Garching, Germany along with physicists at universities in Germany, Austria and Spain have worked out a way to do attosecond spectroscopy on electrons emitted from the surface of a solid.

The technique uses a 300-as pulse of extreme ultraviolet light (XUV), which enters the sample and ejects electrons via the photoelectric effect. At the same time a much longer pulse of infrared light is reflected from the surface of the sample. Once ejected, the electrons are accelerated by the infrared light towards a time-of-flight detector positioned over the sample. This measures the arrival times of the electrons to attosecond accuracy.

The team proved the effectiveness of the technique by studying the time it takes for electrons to be ejected from a tungsten sample after absorbing a XUV photon. They discovered that electrons escaped the material in two distinct groups separated in time by about 110 as. By measuring the kinetic energy of electrons in each group, Krausz and colleagues concluded that first to leave were conduction electrons, followed by electrons that had been in a bound f-state.

According to the team, about 20 as of this delay arises because excited bound electrons can travel further through tungsten than excited conduction electrons, and therefore the bound electrons are likely to have come from deeper in the sample. The remaining 90 as delay corresponds to the expected difference in kinetic energy between bound and conduction electrons that have each absorbed one XUV photon.

The attosecond timescale is the absolute limit on how fast an electronic device could operate. As a result tiny electronic circuits a few atoms in size could, in principle, switch electric currents at petahertz frequencies (10¹⁵), which is nearly a million times faster than processors today. However, very little is currently known about how electrons would actually move through such circuits – which is why Krausz and colleagues believe their new technique will play a valuable role in the development of future electronics technologies.

Unravelling the mysteries of coiling ropes

Neil Ribe at the University of Paris-7 and colleagues in Iran and the Netherlands used a reel powered by an electric motor to feed ordinary rope or thread down through a hole and onto a glass or paper plate below. The rate of descent and the distance between the reel and the plate could be changed, allowing the team to study coiling over a wide range of speeds and drop lengths. A second set of similar experiments looked at the coiling of soft strands of spaghetti.

Ribe told physicsworld.com that the team is the first to perform controlled lab experiments on coiling and their use of different materials allowed them to build up a comprehensive understanding of why some ropes coil and others don’t.

According to Ribe, one surprise result is that the coiling always occurred at several different “frequencies” for fixed values of the feed rate and fall distance. These frequencies correspond to the vibrational modes of the nearly vertical upper part of the falling rope. They discovered that coiling occurs when any of these frequencies matches the angular frequency at which bottom end of the rope whirls into a coil.

The team were also able to describe their observations in a numerical model that treated coiling as a fine balancing act between elastic, gravitational and inertial forces acting on the rope. According to Ribe the model was able to reproduce the observed multi-frequency nature of coiling.

“This is an exciting paper, which details the many different coiling patterns in an elastic rope,” says Herbert Huppert, a geophysicist at the UK’s University of Cambridge who has an interest in such materials. “The agreement [between experiment and numerical model] gives confidence to the detailed and complex nature of this sub-field of highly nonlinear dynamical systems, in contrast to many other situations for which the description is at best qualitative. Many a physicist is going to enjoy playing with his pasta after reading this paper.”

Everyday ropes are the simplest example of elastic ropes — a class of materials that includes encompassing electrical cables, plant vines, DNA and steel rods. Elastic ropes can act as nonlinear dynamical systems, the behaviour of which can be very difficult to understand. Ribe and colleagues hope that their simple experiments will cast light on the complex nature of these common materials.

Semiconductor bends light “wrong” way

Negative-index metamaterials are artificial structures that are engineered to have a negative index of refraction. Light travelling through such a material is therefore bent in the opposite direction as light in a normal material – a property that could in principle be used to create “invisibility cloaks” and “superlenses”.

Most negative-index metamaterials are designed to have both a negative electrical permittivity and a negative magnetic permeability. However, for NIMs operating in the infrared, physicists have only been able to achieve this “double resonance” over very narrow ranges of wavelength. Other drawbacks of double-resonance NIMs are that they absorb much of the light passing through them and their intricate design makes it very difficult to make them thick enough to be of any practical use.

Now, Claire Gmachl and colleagues at Princeton University, Oregon State University and Alcatel-Lucent in New Jersey have created a different kind of NIM that does not rely on double resonance. Instead it has an electrical permittivity that is negative along only one direction in the metamaterial. This gives it a negative index of refraction over a broad range of wavelengths, and with little loss of light.

The metamaterial is made by depositing alternating layers of two semiconductors — indium gallium arsenide and aluminium indium arsenide — onto a substrate using molecular beam epitaxy. Each layer of the metamaterial is about 80 nm thick, which is much smaller than the wavelength of the infrared light.

When infrared light is shone into the material, its wavefront continues moving through in roughly the same direction. However the “Poynting vector” describing the energy flow of the light through the metamaterial is bent away from the direction of the wavefront (see Wrong way). This bending occurs because the permittivity is negative in the direction perpendicular to the layers and positive parallel to the layers.

Such a material could be a boon to those designing infrared optics, explains team member Anthony Hoffman of Princeton University: “Currently, the infrared lens is a massive object,” said Hoffman. “This new material may enable more compact mid-infrared optics because we have an entirely new set of optical parameters in our toolkit.” The team is now attempting to make a superlens from the NIM that could be used to obtain images of structures smaller than the wavelength of infrared light, which is impossible with a normal infrared lens.

John Pendry at Imperial College London, who is a pioneer of negative refraction, said of the work “They have found a way to make anisotropic materials to a high degree of perfection, and that’s quite remarkable.” he said. But he added that this material cannot be termed a negative-index metamaterial in the strictest sense. “Since it is not completely isotropic, its properties and hence the applications may be quite different from the metamaterials that have been built by other teams.”

Light snakes across tiny gaps

When light is radiated between two surfaces separated by less than a wavelength, there is not enough room for the light to cross the gap as a propagating wave. Instead it bridges the gap via “evanescent” waves, which are standing waves that only exist about a wavelength or less from a surface. While physicists are starting to exploit such waves in nano-optical devices such as superlenses, exactly how light behaves in tiny gaps is very difficult to predict using electromagnetic theory.

Now, Zhuomin Zhang and colleagues at the Georgia Institute of Technology have come up with a new way to predict the behaviour of infrared light in such gaps. Instead of focussing on the light itself, they modelled how evanescent waves transfer heat across a 100 nm vacuum gap between two plates of silicon carbide. This involved calculating the “Poynting vector”, which describes the flow of electromagnetic energy.

Their simulations show that instead of travelling in a straight line, the light snakes its way across the gap. The team believe that this new ability to visualize how light travels across tiny gaps should help physicists make important decisions when designing nano-optical devices — such as what shape the plates should be and how far they should be separated.

According to Zhang, the simulations could help researchers create nano-optical structures that are extremely good at absorbing infrared radiation. Such structures could be used to make very efficient thermophotovoltaic (TVP) cells, which convert infrared radiation into electricity. TVP cells could someday be used to generate electricity from the waste heat of industrial processes.

The simulations also provide a vivid illustration of the negative refraction of light at both of the vacuum/silicon carbide interfaces. Negative refraction occurs when light travelling from one medium to another bends in the direction opposite to the direction normally associated with refraction. Physicists have already taken advantage of this effect in “superlenses” made from slabs of silicon carbide that are able to take images of tiny structures that are much smaller than the wavelength of light used.

“Spookytechnology” anyone?

Tahan came up with the name in a nod to Einstein’s famous dismissal of quantum entanglement as “spooky action at a distance”. But while Einstein’s words reflected his unease about some implications of quantum theory, Tahan believes that spookytechnology will actually help quell any public fears of quantum computers and other “spookytechnologies”.

At the heart of Tahan’s proposal is his belief that the quantum-technology community must act now to avoid the identity crisis faced by those working in “nanotechnology”, which quickly became a very broad term covering everything from advanced materials to genetic engineering. As a result, physicists working on quantum dots tend to get lumped together with biologists developing new forms of life, even though they have little in common.

According to Tahan, this unfortunate public misperception has arisen because scientists did not provide an authoritative initial definition of nanotechnology – something he has done for spookytechnology.

The first part of his definition says: “Spookytechnology encompasses all functional devices, systems and materials whose utility relies in whole or in part on higher order quantum properties of matter and energy that have no counterpart in the classical world”.

David Deutsch of the UK’s Oxford University who is one of the founding fathers of quantum computing is not keen on the name. It would, he told physicswold.com “exacerbate the mystical connotations that quantum theory has acquired”. According to Deutsch, the mysticism associated with Einstein’s comment “has been seized upon to justify almost every kind of pseudoscientific anthropocentric nonsense”. Instead, he believes that the public should be encouraged to understand entanglement, rather than being told that it is spooky.

Tahan answers such criticism with a question: “If or when we really understand quantum behaviour beyond just being able to describe it mathematically, will it still be spooky?” He believes the answer is yes, because something like entanglement will always be counterintuitive.

He believes that the public need not be shielded from the bizarre results of quantum theory and one reason for his proposal is to make the public more aware of the exciting work going on in the field. In this endeavour, Tahan has the support of Richard Jones of Sheffield University in the UK, who is a polymer physicist and pioneer of nanotechnology. However, Jones also expressed concern about using spooky to describe the “profoundest mysteries of the universe”.

Tahan is also keen to establish spookytechnology as a friendly term with no “direct environmental or toxicological ramifications”. This, he hopes, would prevent applications of quantum mechanics from being given a bad name as nanotechnology did when US scientist Eric Drexler dubbed the possibility of self-replicating nanorobots as the “grey goo scenario”.

Capacitance sheds light on complex flows

Flowing mixtures of liquids and gases play an important role in the transport and processing of materials as varied as carbonated drinks, pharmaceuticals and petroleum. The complex flow patterns that can occur in such mixtures must be monitored to ensure that, for example, a drug is mixed properly or that oil can flow up through an undersea pipeline.

However, current techniques for measuring such “multiphase” flows have limited efficacy. While a flow can be monitored using high-speed cameras, they provide little quantitative information about the mixture. Probes placed directly in the flow can provide more useful information, but this is limited to the area near to the probe, which is of little use when studying complex multiphase flows. More sophisticated techniques such as gamma-ray tomography can do better, but the added expense and radiation hazards are often not justified by the modest improvement in results.

Now, Uwe Hampel and colleagues at the Forschungszentrum Dresden-Rossendorf (FZD) in Dresden, Germany, have come up with a sensor that promises to change all this. The sensor consists of two planes of parallel wire electrodes stretched across a pipe flange. The planes are perpendicular to each other and separated slightly to form a mesh.

An oscillating voltage is applied to wires in one plane and the resulting electrical fields are detected at the electrodes in the second plane. This allows the sensor to measure the capacitance of materials in the regions where two wires cross.

Different materials have different capacitances, which allow the sensor to differentiate between different phases in the flow. According to Hampel, the measurement can be done very rapidly so that the flow in a typical pipe cross-section can be scanned up to 10,000 times per second at a spatial resolution of 0.5 mm.

“This catches even small bubbles moving at several metres per second through the sensor, giving highly accurate information on the flow structure,” said Hampel. The sensor is similar to an earlier prototype built by the team that measured the electrical conductivity between the wires and therefore could only be used with conducting liquids.

According to multiphase flow expert Barry Azzopardi at the UK’s University of Nottingham, the new sensor shows great promise because it can be used with organic fluids such as petroleum, which are non-conducting.

The sensor will probably never be deployed in a real oil pipeline because it would be destroyed by the solid “pigs” that are sent through the line to clean and inspect the pipe. Instead the sensors could be used on mock pipelines in the lab gain a better understanding of multiphase flow.

The team will be taking the sensor to Nottingham in November, where they will working with Azzopardi to study the flow of gas and oil mixtures in pipelines.

The team now plans to study various complex flows, such as oil-gas transport through pipelines in order to identify a range of applications for the device. “Currently, the sensor exists as a prototype but we hope to think further about commercialization once a broader range of applications have been identified”, said Hampel.

Browse articles by content type