Page 1699 – Physics World

Home » Page 1699

Polariton laser reaches room temperature

Polaritons are “quasiparticles” that arise from the coupling of light with an electric dipole in a semiconductor material. More precisely, they consist of a photon and an exciton, which is itself a dipolar quasiparticle comprising a bound electron-hole pair. Polaritons are part light and part matter — and possess new properties not seen in either photons or excitons.

Light is emitted from a polariton laser in a process that involves the scattering of pairs of polaritons. The scattering is stimulated by a light from a separate pump laser. However, unlike conventional solid-state lasers — which consume a great deal of energy “pumping” the majority of valence electrons into the conduction band – very little energy is required to operate a polariton laser and therefore some physicists believe that polaritons offer a way to create lasers with very low power requirements.

While polariton lasers have already been built, they only worked when cooled below 200 K. Now, Jeremy Baumberg and colleagues at Southampton University along with co-workers at Ecole Polytechnique Federale de Lausanne, have built a polariton laser that operates at room temperature (300 K).

The laser is a microcavity structure in which a layer of the semiconductor gallium nitride (GaN) several hundred nanometres thick is sandwiched between two layers of reflecting material. The size of the cavity is chosen to resonate with ultraviolet light of a certain wavelength and polaritons are created in the GaN as the light is reflected back and forth across the cavity.

Baumberg told Physics Web that GaN was chosen because the binding energy of excitons in this material is known to be very high, and therefore it is an ideal candidate for a polariton laser. Unfortunately, GaN is a very difficult material to work with and as a result Baumberg said that it took the group five years to get the lasers to work. However, he is hopeful that the performance of the lasers can be further improved as GaN processing technology matures. Indeed, Baumberg said the group has already reduced the original energy requirements of the laser by a further factor of ten and that more improvements are possible.

Baumberg also believes that the polaritons in GaN microcavities could form a Bose-Einstein condensate (BEC) at room temperature — something that has already been observed at much lower temperatures in microcavities made of other semiconductor materials. A BEC occurs when a significant number of the polaritons (which are bosons) condense into the lowest energy state, forming a macroscopic coherent quantum state. Baumberg believes that polaritons could someday form the basis of a “BEC-on-a-chip” that could be used as an interferometer.

Probe seeks changes in fine-structure constant

Most measurements of the fundamental constants of nature have been made on Earth over the past one hundred years. However, it is possible that these quantities are different when measured elsewhere in the universe or at other times. Indeed, the ability of fundamental constants to vary over space and time plays an important role in some theories that attempt to unify gravity, electromagnetism, and the strong and weak nuclear forces.

Some physicists believe that the value of the fine-structure constant (α) has been growing since the universe was formed in the Big Bang some 13.5 billion years ago. Observations of light from distant quasars suggest that α may have been one part in 10⁵ smaller some 11 billion years ago than it is today. Closer to home, measurements of α derived from studying the decay of radioactive isotopes on Earth suggest that the constant may have changed by one part in 10⁷ over the past 4.6 billion years.

Now Benjamin Wandelt and Rishi Khatri of the University of Illinois at Urbana-Champaign have proposed a way of measuring α as it was 10-100 million years after the Big Bang – during the so-called “dark ages”, when the universe was cool enough for neutral hydrogen atoms to exist, but before stars and galaxies formed. During this period, the hydrogen atoms absorbed cosmic microwave background (CMB) radiation at a wavelength of about 21 cm, which corresponds to a transition between two atomic energy states. The result is an absorption line in the CMB that endures to this day.

Wandelt and Khatri have shown that the precise wavelength of the transition is very sensitive to changes in α. Since microwave radiation from the dark ages can be detected today, they reckon that the precise location of the 21-cm line and the relative strength of the absorption can be used to determine α as a function of time — after correcting for the redshift caused by the expansion of the universe.

As hydrogen atoms absorbed photons throughout the dark ages, Wandelt and Khatri believe that it should be possible to track changes in α over a period of about 100 million years. Indeed, because hydrogen was ubiquitous during the dark ages, the researchers are confident that their technique could be used to create spatial maps of α, which could prove useful in the search for dark energy.

Unfortunately, the scheme cannot take advantage of the current generation of microwave telescopes — such as the Wilkinson Microwave Anisotropy Probe (WMAP) satellite – which do not focus on the region of the microwave spectrum containing the 21 cm absorption line from the dark ages. However, Wandelt told Physics Web that the measurements could be made using the Long Wavelength Array (LWA) telescope that is currently being built in the US state of New Mexico.

Another complication is that the signal is buried beneath an intense background of radiation that originates from within our own galaxy. However, Wandelt is confident that this background can be subtracted to yield a measurement of α to an accuracy of about 0.1% within a decade.

Magnifying superlenses show more detail than before

No matter how smooth and polished a conventional lens is, there will always be a finite amount of detail it can reproduce. This is because light tends to diffract, and so prevents resolution of features much smaller than its wavelength – what physicists refer to as the “diffraction limit”. However, this limit can be surpassed if one can find a way of collecting the idle “evanescent” waves that exist close to the surface of an object. These waves can resolve surface features much smaller than normal propagating waves, but decay too quickly for conventional lenses to capture.

In 2000, John Pendry of Imperial College in London predicted that the decay of evanescent waves can be offset by amplifying them in a material with a negative refractive index – in other words, one that bends incoming light in the opposite direction to an ordinary material. In theory, such a negative-index “superlens” could take evanescent waves from a surface, carry them, and convert them into propagating waves that travel far enough to be captured by a conventional microscope after they leave. Since Pendry’s prediction, several superlenses have been built that have successfully transmitted evanescent waves. However, none has been able to make the crucial conversion to propagating waves – leaving the evanescent waves with the same fast decay rate as when they started.

Now, two groups have managed to create superlenses that can convert evanescent waves into propagating waves. At the University of Maryland, a team led by Igor Smolyanivov has created a flat superlens consisting of concentric polymer rings deposited onto a thin film of gold (Science 315 1699). Meanwhile, a team led by Xiang Zhang at the University of California in Berkeley has opted for a 3D stack of curved silver and aluminium-oxide layers on a quartz substrate (Science 315 1686).

Both of these designs are classified as “metamaterials” – artificial nanostructures made by physicists because substances with a negative index in the optical range do not occur naturally. Superlenses have been made from metamaterials before, but the cylindrical geometry of these new designs enables evanescent waves emitted from illuminated objects to be guided outward. Because momentum must be conserved, this separation forces the “tangential” or side-to-side momentum of the waves to be compressed, resulting in a magnified image beyond the diffraction limit – one that a conventional microscope can register.

Smolyanivov’s
superlens

Smolyanivov used his flat superlens to image rows of polymer dots deposited near the inner ring with a resolution of 70 nm – seven times better than the diffraction limit of the illuminating laser. Zhang, however, took his 3D superlens (or “hyperlens” as he prefers because of the hyperbolic shape) one step farther and imaged the word “ON” inscribed on the surface, albeit with a slightly lower resolution of 130 nm.

Although both of these superlenses imaged objects that were “built-in” to the material, in practice one would keep an object separate but still close enough so that the evanescent waves can be captured. Even so, Smolyanivov told Physics Web that widespread applications may be some way off. This is because a side effect of the increased resolution is the vastly reduced depth of field, meaning focusing must be much more accurate. “The real challenge will be to locate a sample”, he said. “You won’t be able to see it if it’s out of focus.”

Spins take their time to relax

To store and process information using the spin of an electron, the spin needs to be relatively robust. The most important property that determines a spin’s robustness is its spin-relaxation time. When a spin relaxes – something that can happen when the spin is perturbed by interactions with its environment – the information encoded in the spin is lost. “We therefore want the relaxation time to be as long as possible,” said Virginia team member Supriyo Bandyopadhyay.

The researchers made a 50 nm diameter spin valve – a device whose resistance changes in a magnetic field because of spin-related effects. The valve consists of a thin layer of an organic semiconductor, tris (8-hydroxyquinoline), sandwiched between two ferromagnetic electrodes made of cobalt and nickel. The team was able to determine the spin-relaxation time in the organic nanowire by measuring the change in the resistance.

Typically the spin-relaxation time in most materials is a few nanoseconds to a few microseconds, but the researchers found that the spin relaxation in their organic nanowire spin valves could be as long as a second. Moreover, it was relatively unaffected by temperatures up to 100 K.

According to the team, the spin relaxation in their material is very long because the spin tends to remain relatively isolated from perturbations that cause it to relax. They also found that the principal spin-relaxation mechanism is one where the spin relaxes when the electron collides with another electron, or any other obstacle, when moving though the material. This discovery could allow researchers to find new ways to make the spin-relaxation time even longer.

“The organic spin valves we developed are based on self-assembled structures grown on flexible substrates, which could have a tremendous impact on the rapidly developing field of plastic electronics, such as flexible panel displays,” said Cincinnati team member Marc Cahay. “If the organic compounds can be replaced by biomaterials, this would also open new areas of research for biomedical and bioengineering applications, such as ultra-sensitive sensors for early detection of various diseases.”

The materials could also be suitable for opto-spintronics devices, such as spin-enhanced organic light-emitting diodes, where the spin-relaxation time must exceed the radiation recombination lifetime of excitons.

Physicists control light at the nanoscale

Diffraction effects normally mean that the position of a light beam can only be controlled over distances greater than half the wavelength of the light. However, theory suggests that this limitation could be overcome by taking advantage of the interference of light on very short length scales – so-called near-field optics.

Now, Walter Pfeiffer of Bielefeld University, Germany and colleagues in Germany and Spain, have found a way to harness near-field effects by using carefully controlled laser pulses.

The researchers fitted a source of femtosecond laser-pulses with a polarization pulse shaper, which controls how the polarization state of the pulses changes with time. Pulses were fired at a specially-designed nanostructure consisting of six tiny silver disks on a thin film (see figure). The disks each have a diameter of 180 nm and the entire structure measures 800 nm across. The wavelength of the laser light was 790 nm.

Firing the laser at the nanostructure resulted in near-field interference effects that created a pattern of light and dark regions on the nanostructure. By varying the polarization state of the laser pulses, the researchers were able to illuminate different regions of the nanostructure. The illuminated areas were less than 200 nm across, or about one quarter the wavelength of the laser light. The illumination was measured by photoemission electron microscopy (PEEM), which detects electrons that are emitted when light is absorbed by the surface.

Pfeiffer now wants to push the limits of the technique even further, and the team now intends to tackle the challenge of controlling the laser intensity in both space and time simultaneously. According to the team, the precise control of the laser light in time and space could be used in a range of applications, including new space- and time-resolved spectroscopic methods, steering of nano-mechanical processes in optical traps, control of chemical reactions in large molecular aggregates, and new schemes for quantum computation.

Hospital scanners could control cell-sized medical devices

In the 1966 movie Fantastic Voyage, scientists climb into a submarine, shrink themselves down to the size of a red blood cell and are then injected into a dying man to break a blood clot. 40 years on, and the art of shrinking is still far away in the realms of science fiction – but the remote control of miniature devices to perform surgery in the bloodstream may not be.

In a proof-of-concept experiment, a team led by Sylvain Martel from the Nano Robotics laboratory in the École Polytechnique de Montréal has used a conventional MRI system to navigate a 1.5 mm ferromagnetic bead inserted into the blood vessels of a live pig. By individually controlling the magnetic field produced by the MRI system’s three perpendicular magnets, they could propel the bead in 3D around the blood vessels at speeds greater than 11 cm/s.

However, this ability would have been useless if Martel’s team was unable to see where the bead was going, so they devised an algorithm that rapidly alternated the MRI system’s magnets between “propulsion” and “tracking” modes roughly every 20 ms. While in tracking mode, the MRI system behaved as it would when performing a typical diagnostic scan of a patient – in other words, measuring how both the bead and the surrounding tissue interact differently with the magnetic field. The resultant image was then fed into a computer, which calculated the field required to navigate the bead within an accuracy of almost half a millimetre.

After many tests, the physicists found that navigation in arteries roughly twice the bead’s diameter was relatively easy. However, Martel told Physics Web that narrower vessels would need the bead to become smaller than a red blood cell, and extra magnets would be required to create enough force to direct it.

Currently, the least invasive way to operate inside a patient is through “keyhole” surgery, whereby medical instruments are contained in a flexible rod inserted through a small incision. But Martel thinks that one day MRI systems could control various “untethered” devices that perform surgery without the need for incisions at all. For example, simple devices could be fed into the bloodstream to reopen blocked arteries or target aneurysms. “We are presently developing more complex and much smaller micro-devices for various applications, such as targeted drug delivery and navigable biosensors for diagnostics,” said Martel.

Double-negative metamaterial edges towards the visible

Naturally occurring materials have a positive index of refraction, whereas a negative-index metamaterial (NIM) has a structure that is engineered artificially to have a negative index of refraction. NIMs have a number of desirable properties that do not exist in normal materials including the ability to focus light to a point smaller than its wavelength in a so-called “superlens”, which could allow optical microscopes to view much smaller objects than possible today.

A negative index of refraction can occur when only the permittivity of the material is negative and the permeability, although positive, is different from that in free space. However, the effect is much more pronounced (and more technologically useful) in DN-NIMs, in which both permittivity and permeability are negative.

While NIM metamaterials have been developed with negative permittivities for visible light, negative permeability is much more difficult to achieve because the magnetic interaction between light and a metamaterial is more than 100 times weaker than the electrical interaction.

Speaking at the recent March Meeting of the American Physical Society in Denver, Vladimir Shalaev of Purdue University, Indiana, unveiled a new DN-NIM that is tantalizingly close to the visible range. The metamaterial is a thin sheet comprising two layers of silver separated by alumina. The structure is perforated with a regular array of rectangular-shaped holes to create a “fishnet” pattern. The holes are about 120 nm across and are separated by about 300 nm. The magnetic permeability of the material was found to be negative for light with wavelengths between about 799 and 818 nm, while the permittivity is negative from about 700 to beyond 900 nm.

Shalaev told Physics Web that the fishnet structure could be adapted to create a DN-NIM for visible light – something that he and his colleagues are working on right now. However, he cautioned that fishnet NIMs display negative permeability over a relatively narrow band of wavelengths and therefore it is unlikely that a single structure could be used to create a DN-NIM that works throughout the visible spectrum. Also, some of the light passing through fishnet NIMs is absorbed, which means that it could not be used as a superlens. However, he believes that such metamaterials could be used to achieve sub-wavelength imaging using other schemes including a “hyperlens”, which aims to covert near-field evanescent waves into waves that can be focussed to create a far-field image.

Experiment sets the ultimate test for Newton’s laws

For the past 70 years or so, physicists have been bothered by a nagging question: why do the centres of galaxies rotate too fast for the amount of mass we can see through telescopes? The most popular answer is that most of the mass is hidden in large bands of “dark matter”, a substance that is invisible because it doesn’t interact strongly with light. If it exists, dark matter could account for 95% of the mass in galaxies, and would explain many other aspects of the universe.

However, a lack of evidence for dark matter has led a small camp of physicists to promote an alternative answer: the gravitational force that holds galaxies together decays more gently with distance than presently estimated, meaning that Newton’s law of gravitation is not quite as simple as an inverse-square relationship. The theory, which is known as modified Newtonian dynamics (MOND), proposes adding extra factors to Newton’s 300-year-old equations so that the gravitational behaviour only alters at very low accelerations. Unfortunately the turmoil of gravitational forces produced in the galaxy means that such accelerations are hard to come by, leaving proponents of MOND with no easy way to test their theory.

However, Alex Ignatiev from the Theoretical Physics Research Institute in Melbourne claims to have predicted instances on the Earth where most of these forces will cancel out. Ignatiev first considered how an object at rest in the centre-of-mass of our galaxy would appear to be accelerating when viewed from a laboratory on Earth. This involved listing all the major accelerations such as the Earth’s rotation around the Sun and the Sun’s orbit in our galaxy. He then looked for solutions where all of the accelerations add up to zero.

The solutions indicated that, on either of the two annual equinoxes, there will be two places on the Earth’s surface where the force cancellation occurs. For example, on the equinox of 22 September 2008, one will be in the far north of Greenland and the other will be on the opposite side of the world in Antarctica (see figure: “X marks the spot”). Ignatiev says that if a gravitational wave detector is set-up to monitor a static test object at one of these times and places, it might just be able to glimpse a tiny, 0.2 × 10^-16 m deflection over a period of 0.5 ms – what he calls “SHLEM” (static high-latitude equinox modified inertia). If SHLEM is observed, it would be the first evidence in support of MOND.

“Even if the result were negative it would be a very significant step forward, because an interesting theory would be ruled out,” Ignatiev told Physics Web. “But if the predicted SHLEM effect were observed – well, we’d have to rewrite our most basic theories.”

Nanowires could boost memory density

Magnetic domain walls are narrow boundaries between regions where, for example, the magnetic moments point “up” on one side of the wall and “down” on the other. Domain walls can be moved within a material by applying an external magnetic field or injecting a spin-polarized current.

Some physicists reckon that this motion could be exploited in “racetrack” memories, which could store much more data than today’s RAM devices. In a racetrack memory, data are stored as a sequence of magnetic domains – separated by domain walls — along a nanowire (see figure “Racetrack memory”). Individual bits are stored and retrieved by moving the sequence along the nanowire and across magnetic read and write heads.

If this technology is to succeed, a practical way of using spin-polarized currents to move the domain walls along the nanowire must be found. The key challenge is how to reduce the current density needed to move a domain wall when it is “pinned” by a defect in the wire. At the moment, the current density is too high for use in commercial memory devices.

But now Stuart Parkin and colleagues at IBM’s Almaden Research Centre in the US have found a way to reduce the current density by more than factor of five by exploiting the fact that pinned domain walls have a natural frequency of oscillation. When exposed to a train of current pulses with the right length and separation, the amplitude of oscillation increases until the domain wall breaks free of the defect and moves along the wire.

The required pulses were about one nanosecond in length and Parkin told Physics Web that such pulses could easily be used in a racetrack memory – indeed similar pulses are already used routinely in other memory devices.

Geophysicists offer explanation for Andes formation

All tectonic activity on Earth is driven by subduction zones, where one plate is sucked underneath another into the Earth’s mantle. Now, for the first time, a team led by Wouter Schellart at the Australian National University in Canberra has created the first genuine 3D model of how plates move at subduction zones over time.

The model uses fluid dynamics and solid mechanics techniques to extrapolate modern data from dynamic, seismic and chemical studies as far back as 40 million years. This enabled the geophysicists to see how the lateral width of tectonic plates – in other words, their span across the Earth’s surface – affects the shape and movement of their subduction zones.

Schellart’s team found that plate width controls two fundamental features of subduction zones: the curvature of the trench where two plates meet, and the zone’s tendency to “retreat” through erosion. The narrowest plates between 300 and 1,200 km have a concave geometry at the boundary, and retreat relatively fast. On the other hand, plates with widths greater than 4,000 km develop convex boundary geometry and retreat slowly. However, the geophysicists found that the subduction zones of the largest plates have central areas that actually advance – rather than retreat – over periods of 5 to 10 million years.

Peru-Chile trench

One of these advancing areas is a region near Bolivia, where the Nazca plate in the Pacific Ocean subducts under the South American plate. Using the model, Schellart’s team found that the region supports large compressive stresses, fuelled by the westward movement of the South American plate. This could provide an explanation for how the South American Andes began to form some 200 million years ago.

Browse articles by content type