April 2010 Archives

A collaboration led by Pascal Martin of the Curie Institute in Paris yoked live and virtual cells to tackle the question, How does the ear amplify faint signals by factors of up to 1000? Hearing relies on converting mechanical vibrations to electrical impulses. The transduction takes place in the cochlea and is carried out by micron-scale hairs that sprout from specialized hair cells. To the tiny hairs, the watery liquid that surrounds them is viscous, just like honey would be to a tuning fork. An active feedback mechanism in the hair cells not only overcomes the viscosity, but also amplifies faint signals. But the feedback can't provide all, or even most, of the ear's prodigious amplification. Two years ago, theorists proposed that elastic coupling among groups of hair cells could make up the shortfall. Coupling, they argued, lowers the detection threshold by averaging out each cell's noise fluctuations. Hair cells in frogs, humans, and other terrestrial vertebrates are indeed coupled to each other by various flexible structures. To prove that coupling boosts amplification, Martin's team extracted a vibration-sensing organ from a bullfrog and attached a flexible whisker to one of the hair cells (see schematic). Through sophisticated computer control, the whisker was made to vibrate in a way that mimicked the net force on the live cell (green) exerted by two neighboring hair cells (gray). The expected enhancement to the amplification showed up. (J. Barral et al., Proc. Natl. Acad. Sci. USA, in press.)—Charles Day

In the presence of a suitable nucleating agent, a liquid in a metastable state below its thermodynamically defined melting point freezes. That’s what happens when atmospheric aerosol particles cause supercooled water droplets in clouds to form snowflakes. Researchers have suspected that the atomic surface structure of such seeding particles acts as a template, inducing local order in the disordered liquid and catalyzing its crystallization. Conversely, a solid with a different structure can inhibit crystallization, as has now been observed at the European Synchrotron Radiation Facility in Grenoble, France, by Tobias Schülli and his colleagues. The researchers coupled x-ray scattering data with molecular-dynamics simulations to study supercooled gold-silicon droplets on a silicon substrate, a system that is used to grow Si nanowires. Surprising results emerged when they heated the AuSi alloy above 676 K: As it cooled, the Si atoms leached onto the substrate and, as the figure shows, rearranged its surface atoms into pentagonal clusters. The alloy’s atoms near the interface mimicked the substrate’s surface structure (see inset), but the resulting local order did not promote crystallization in the droplets, which froze at 513 K, about 120 K below the freezing temperature for the AuSi alloy. Apparently, the pentagonal geometry inhibits freezing because it is not conducive to crystal packing. That finding suggests that substrates with such atomic structures offer a simpler method of maintaining and observing the supercooling process than such techniques as magnetically levitating or otherwise suspending the liquid droplet. (T. Schülli et al., Nature 464, 1174, 2010.)—Jermey N. A. Matthews

The outer layer of a pollen-grain wall generally includes apertures through which the grain can gain or lose water. When in an arid environment, pollen grains avoid becoming dangerously dry by undergoing a process called harmomegathy—the grain’s apertures are effectively sealed until the pollen lands in a wetter location. For more than a century, scientists have known that wall structure helps determine the form that a pollen grain assumes after harmomegathy. Now Harvard University’s Jacques Dumais, former Harvard student Eleni Katifori, and colleagues have presented the first quantitative model of the process and confirmed it with electron micrographs such as shown here (the scale bars represent 20 µm). The model incorporates the classic result that stretching a surface costs a lot of energy; instead of stretching, the grain surface bends as the wall folds onto itself to avoid further desiccation. The lily grain in panel a, for example, has an elongated aperture that allows harmomegathy to proceed somewhat like the way in which one makes a cone by connecting the edges of a disk that has had a slice removed. Strictly followed, that process yields vertices with high concentrations of bending energy; in reality the lily grain stretches a little at the vertices and ends up looking like a US football. The other grains illustrated in the figure have built on the same simple physics—avoid stretching and kinks—to achieve more intricate but equally effective harmomegathic responses. (E. Katifori et al., Proc. Natl. Acad. Sci. USA, in press, doi:10.1073/pnas.0911223107.) —Steven K. Blau

Carved into many Martian dunes are narrow, sinuous channels. Studies of their morphology and laboratory simulations suggest that the likely origin of the gullies is surface or near-surface water ice that melts and forms a flowing slurry of sandy debris. Dennis Reiss and colleagues at the University of Münster's Institute for Planetology report new evidence for such transient liquid water. Their observations rely on data from the High Resolution Imaging Science Experiment, a camera on the Mars Reconnaissance Orbiter that provides nearly 10 times the resolution of the camera aboard the earlier Mars Global Surveyor. Armed with 23 sequential HiRISE images of the Russell crater dune field in Mars’s southern hemisphere over two successive Martian years (spanning November 2006–May 2009), the researchers uncovered signs of multiple flow events that, as seen in this image, in turn deepen and widen the channels. They also observed gullies lengthening over the course of the early Martian spring. Factoring in near-IR reflectance data, which tracked the melting of frozen carbon dioxide, and calculations of daily springtime temperature profiles at the surface, they discount dry flows and CO₂ flow mechanisms; they instead conclude that the gully changes are best explained by the seasonal melting of small amounts of water ice. (D. Reiss et al., Geophys. Res. Lett. 37, L06203, 2010, doi:10.1029/2009GL042192.)—Richard J. Fitzgerald

When chemically modified with water in a process called hydration, cement morphs into the durable binder that holds gravel, sand, and other additives together to form concrete—the most used manmade material in the world. The main constituent of hydrated cement is CaO-SiO₂-H₂O (called C-S-H) in the form of nanoscale colloidal aggregates, the size, shape, and packing of which are crucial to the ultimate strength and stability of concrete. The solid C-S-H nanoparticles are generally thought to be analogous to the claylike minerals tobermorite and jennite, mixed with calcium hydroxide. But new neutron-scattering studies by Jeffrey Thomas and Hamlin Jennings of Northwestern University and Andrew Allen of NIST in Gaithersburg, Maryland, show that C-S-H has a higher-than-expected atomic packing density. The mass density of solid C-S-H is roughly 10% higher than that of a mixture of its widely used mineral analogues with the same composition. The result has important implications for the modeling of cement paste. (See, for example, Physics Today, November 2009, page 23, where that model's starting point is dry tobermorite.) The researchers also investigated the composition and density of C-S-H cured at elevated temperatures and with various additives. In particular, they found that curing the cement at 80 °C led to a lower atomic packing density. Such atomic packing variations suggest the possibility to control chemical shrinkage and the associated cracking of concrete. (J. J. Thomas, H. M. Jennings, A. J. Allen, J. Phys. Chem. C, in press. —Stephen G. Benka

Several approaches have been devised for suppressing the rolling motion of ships at sea. Active systems that position a movable mass to provide a countering torque can rapidly damp the rolling, but the inherent friction not only is noisy but also produces constant wear. Researchers from the Korea Institute of Machinery and Materials recently demonstrated that the technology employed to reduce friction in magnetically levitated trains can do the same for ships. Indeed, their antirolling device is essentially a maglev train car on a short track that runs side to side across the ship's midline. For the demonstration, the team built a 118-kg, small-scale model with the cross-sectional shape of a twin-hulled catamaran. A C-shaped, 4-kg mass—the "train car"—wrapped around a 1-m-long rail along the top of the model; electromagnets at the bottom of the mass faced the rail from underneath and provided the levitating force. Mounted in the rail was a linear motor, like those used to propel maglev trains (and some roller coasters). Whereas the more common rotary motors produce torque, linear motors produce linear forces. In the researchers' setup, as the motor controller received signals from a tilt sensor on the ship, it quickly moved the mass to the appropriate position to damp the rolling motion. In tests, the model ship's free oscillations were suppressed within 4 seconds. Though it is more expensive than conventional alternatives, the researchers think maglev antirolling technology could have broad potential, from pleasure boats to oceanographic research vessels. (C. H. Park et al., Rev. Sci. Instrum., in press.)—Richard J. Fitzgerald

When an electric current flows in a material, the coupling of the electrons’ spin and orbital angular momentum causes spin-up electrons to bend in one direction and spin-down electrons in the other, both transverse to the charge current. That phenomenon is known as the spin Hall effect (see Physics Today, February 2005, page 17). Eiji Saitoh (Tohoku University), Sadamichi Maekawa (now at the Japan Atomic Energy Agency), and their colleagues have used that effect and its inverse—the creation of a charge current from a transverse spin current—as the basis for a counterintuitive demonstration: the transmission of a DC electrical signal a macroscopic distance through an insulator. Although the bandgap of a magnetic insulator prevents charge conduction, it doesn’t prevent the excitation of a spin wave, the collective precession of localized magnetic moments. For their demonstration, the researchers used a heterostructure composed of a micron-thick slab of yttrium iron garnet on which sits two thin rectangular platinum electrodes a millimeter apart, as pictured here. An electric current J_C applied to one electrode is first converted to a spin current that flows downward. Provided it’s large enough, that current, in turn, induces in Y₃Fe₅O₁₂ a spin wave J_S, which propagates to the second electrode. There the wave is reconverted to a spin current of conduction electrons and, thanks to the inverse spin Hall effect, is detectable as a small voltage signal. The work sets the stage for exploiting a rich class of materials—ferromagnetic insulators—in spintronics research. (Y. Kajiwara et al., Nature 464, 262, 2010.)—R. Mark Wilson

Type Ia supernovae—the powerful explosions of compact white dwarfs—look remarkably uniform. That uniformity, which makes them useful to cosmologists as standard candles for gauging distances across the universe, has been attributed to a common mechanism: When a white dwarf gaining mass from its binary companion nears the Chandrasekhar limit of 1.4 solar masses, it must contract and then explode. Several years ago researchers with the Supernova Legacy Survey spotted a supernova, SN 2003fg, that bore all the marks of being type Ia but appeared far too luminous to have had a Chandrasekhar-mass progenitor. Now three research teams—from the Nearby Supernova Factory (SNfactory), the University of California, Berkeley, and the universities of Tokyo and Hiroshima in Japan—have gathered more detailed data on two more type Ia supernovae, SN 2007if and SN 2009dc, that were at least as luminous as SN 2003fg and shared some of its other unusual features. Comparing their observations with published theoretical models, the Berkeley and SNfactory researchers found that each of their supernovae was likely the result of a merging pair of white dwarfs whose combined mass exceeded the Chandrasekhar limit. The too-bright supernovae are rare enough that they themselves probably won’t have much of an effect on cosmological measurements. But they do suggest that better understanding the physics of type Ia supernovae—and how to categorize them by their observed properties and explosion mechanisms—could yield a better standard candle. (R. A. Scalzo et al., Astrophys. J. 713, 1073, 2010; J. M. Silverman et al., http://arxiv.org/abs/1003.2417; M. Yamanaka et al., Astrophys. J. 707, L118, 2009.) —Johanna Miller

It has generally been assumed that the most energetic extragalactic cosmic rays (CRs)—with energies ranging from about 10¹⁸ to 10²¹ eV—are mostly protons accelerated in distant active galaxies. But a new paper from the Pierre Auger Observatory challenges that assumption and raises intriguing issues. The 3000-km² observatory in the high plains of western Argentina is studded with particle detectors and fluorescence telescopes that record the showers of secondary particles engendered by ultraenergetic CR primaries hitting the atmosphere. A clue to a high-energy primary’s identity is how far a shower penetrates into the atmosphere before attaining its maximum development. A proton-induced shower penetrates deeper before reaching that maximum than does a shower induced by a heavier nucleus of the same energy. From the shower-maximum depths of several thousand well-measured ultrahigh-energy events, the Auger collaboration finds that iron nuclei appear to become increasingly dominant over protons in the cosmic-ray flux above 10¹⁹ eV. Alternatively, the highest-energy protons might unexpectedly behave more and more like heavy nuclei at collision energies far above anything yet measured in the laboratory. Such behavior would signal new physics beyond particle theory’s standard model. But if the highest-energy cosmic rays really are Fe²⁶⁺ ions from distant galaxies, their arrival directions should be severely scrambled by intervening magnetic fields. That’s hard to reconcile with the apparent correlations between the anisotropic distribution of matter within a few hundred million light-years and the arrival directions of the most energetic CRs reported earlier by the Auger collaboration. Perhaps the iron sources are just a few active galaxies very close by. Stay tuned! (J. Abraham et al. [Auger collaboration], Phys. Rev. Lett. 104, 091101, 2010.)—Bertram Schwarzschild