Download Physics News in 1997

Edited by Phillip F. Schewe and Ben P. Stein
Public Information Division
American Institute of Physics
A Supplement to APS News
INTRODUCTION
For 28 yearsAIP has prepared an end-of-year review of physics highlights.
The aim of this publication has been to inform science journalists and others of
important research in a timelyfashion. The texttraditionally consisted of
articles prepared by various APS divisions and by some of AIP’s other member
societies. How ever, in an efforttostreamline our operations we arechanging
the formatfor Physics News. Beginning with this 1997 edition, the text will
consist mostly of selected articles from our weekly one-page newsletter,Physics
News Update; most of the items will haveappeared also in the Physics Update
section of Physics Today magazine (with some further modifications by editor
Stephen Benka).
Physics News Update was originally created to provide science writers with
brief summaries of breaking news of significant physics research, especially in
recent and upcoming issues of Physical Review Letters
. APS members can learn
about physics research news throughout the year by obtaining a free email
subscription to Update. To subscribe, send an email message to
[email protected]. Leave the subject line blank and in the body of the letter
specify “add physnews” to subscribe or “delete physnews” to cancel. Or you
can view the current Update and all past Updates, and have the use of a searchable seven-year archive,by going to this W eb address: www.aip.org/physnews/
update.
Phillip F. Schew e and Ben Stein
American Institute of Physics
[email protected]
TABLE OF CONTENTS
• astrophysics
• atoms, molecules, optics
• biological/medical physics
• condensed matter/materials physics
• earth sciences/solar system
• nanotechnology
• particles, nuclei, plasma
astrophysics
ASTROPHYSICS MAY HELP the solar neutrino problem after all. The widely
accepted standard solar model (SSM) calls for more neutrinos, especially arising from electron capture by beryllium-7 and boron-8, than are actually observed. (See PHYSICS TODAY, July 1996 and April 1995.) This shortfall has
led to a search for new particle physics, such as neutrino oscillations. But what
if the Sun really produces fewer neutrinos than called for? Wick Haxton of the
University of Washington and Andrew Cumming of the University of California, Berkeley, have reexamined the SSM, particularly the role of helium-3 mixing. They looked at various perturbations of the SSM, subject to three constraints: All standard nuclear and atomic microphysics was retained, the known
solar luminosity could not change and the resulting model must be steadystate. An intriguing solution emerged in which 3He was rapidly transported
deep into the Sun’s core, in localized plumes, then returned to the outer core in
a slow, broad, buoyant outflow. The altered balance between 3He’s production
and consumption would then nudge the expected neutrino fluxes much closer
to the observed values. While stressing that this idea is speculative, they point
out that such 3He mixing could have testable consequences for helioseismology,
stellar evolution and 3 He ejection by red giant stars. (A. Cumming, W. C.
Haxton, Phys. Rev. Lett. 77, 4286, 1996.)
ARE STANDARD SOLAR MODELS RELIABLE? Yes, says John Bahcall
of the Institute for Advanced Study. Fewer than expected solar neutrinos register in terrestrial detectors. This solar neutrino problem is usually attributed to
the hypothetical transmutation of neutrinos from one type into another en route
from sun to earth. One alternative is to propose that helium-3 from the cooler
outer layer of the sun sinks to the lower, warmer depths, thus moderating neutrino production (Update 295). This modification in the model of the sun is
undesirable, Bahcall believes (Bahcall et al., Physical Review Letters, 13 Jan.
1997). He cites recent helioseismic measurements of the velocity of sound
waves inside the sun; the velocity in turn depends on the ratio of the temperature to the mean molecular weight at any particular depth inside the sun.
Bahcall’s analysis finds a very good agreement (for all depths in the sun) between the measured values of sound velocity and the values predicted using the
standard solar model, and a much poorer agreement for models including
helium-mixing. (Physics Today, March 1997.)
A BLACK HOLE’S EVENT HORIZON HAS BEEN DETECTED. Ramesh
Narayan and his colleagues at the Harvard-Smithsonian Center for Astrophysics have used the orbiting ASCA x-ray telescope to study x-ray novas, binary
systems in which gas from one star is pulled toward an accretion disk and the
spherical region surrounding a compact companion. These systems occasionally flash prominently at x-ray wavelengths (hence the name x-ray nova), but
Narayan is more interested in what happens during the quiescent intervals between upheavals. His recent theory, called the advection-dominated accretion
flow (ADAF) model, suggests that if the accretion rate is slow enough the
inspiraling gas will refrain from radiating away its accumulating energy. Instead the gas continues to get ever hotter, reaching temperatures as high as
1012 K. Eventually this enormous energy buildup is dealt with in one of two
ways: if the compact object is a neutron star, the gas will fall onto its surface,
where it heats the star, causing it to radiate. In contrast, if the object is a black
hole, there is no surface for the gas to fall upon; instead, like a prisoner being
led to execution, the gas crosses the black hole’s event horizon, never to be seen
again. In effect, 99% of the gas energy disappears from the universe. Because
of this, x-ray binaries containing a black hole should be dimmer than those with
neutron stars. Naryan, speaking at the January 1997 meeting of the American
Astronomical Society in Toronto, reported on 9 binaries which fit the ADAF
pattern of behavior. Four of these were thought to harbor black holes (because
of their higher masses), and indeed these are all dimmer than the five neutron-star
binaries. Narayan judges this dimness, and the binaries’ x-ray spectra, to be the
sign that an event horizon is at work, and that this in turn constitutes the most
direct evidence yet for the existence of black holes.
Harvard-Smithsonian Press Release: http://Cfa-www.harvard.edu/blackhole/
A NEW CELESTIAL CLOUD of positron-annihilation radiation has been found
by NASA’s Compton Gamma Ray Observatory (CGRO). Positrons result from
some radioactive decays and are routinely made on Earth (at many hospitals and
accelerators, for example) and in high-energy phenomena in deep space. For
instance, the heart of the Milky Way has long been known to emit 511 keV gamma
rays from positron annihilation. At a meeting in Williamsburg, Virginia, in late
April 1997, a multi-institutional CGRO scientific team, led by William Purcell
of Northwestern University, presented maps that show a new and surprising 511
keV emission region projecting north asymmetrically 3000 light-years out of the
plane of our galaxy.
An artist’s conception (not to scale) of two binary star systems: the top one containing a black
hole, and the bottom one containing a neutron
star. (Courtesy Harvard-Smithsonian Center for
Astrophysics)
OXYGEN DATING THE MILKY WAY. A
new technique uses stardust to determine the age
of our Galaxy. By looking at the isotopic composition of “primitive” meteorites—those that
formed along with the Solar System some 4.5
billion years ago—scientists can tell whether
certain grains in them came from outside the
Solar System. Such specks of stardust would
also necessarily predate the Solar System. Larry
Nittler of the Carnegie Institution of Washington has sorted oxide grains according to the composition ratio 16O/ 18O. From this huge sample, he isolated 87 grains for which
that ratio was highly unusual. Of those, he thinks that 13 came from low-mass
red giant stars whose lives ended before our Sun’s began. Using theories of
stellar nucleosynthesis and galactic chemical evolution, Nittler and Ramanath
Cowsik of the Indian Institute of Astrophysics estimate that the Milky Way is
14.4 billion years old. They admit that the systematic uncertainties could be
large. (L. R. Nittler, R. Cowsik, Phys. Rev. Lett. 78, 175, 1997.)
IS THE UNIVERSE HONEYCOMB-LIKE? As astronomers measure redshifts, and therefore distances, for an ever-increasing inventory of galaxy superclusters, the three-dimensional architecture of the universe becomes more evident. New redshift surveys benefit from fiber optics and automation to generate
an abundance of high quality data quickly. Now, a fresh analysis of current
redshift catalogs offers some evidence for a regular three-dimensional arrangement of superclusters, separated by voids, on a scale of about 120 megaparsecs
(about 390 million light-years). A cellular distribution of superclusters and voids
has been known for more than a decade, but their apparently regular, honeycomb-like pattern is new. The researchers suggest that some new physics might
be needed to explain the sort of immense regularity they seem to be finding in
the data. (J. Einasto et al., Nature 385, 139, 1997.)
OUR LOCAL GROUP OF GALAXIES is still forming. For decades, astronomers have wondered about the origin of certain fast-moving clouds of atomic
hydrogen in the vicinity of the Milky Way. Some clouds appeared to be plunging into the plane of the Galaxy (at speeds up to 500 km/s), while other clouds
seemed to be moving away from the Milky Way. In either case, they were not
rotating with the Galaxy. A synthesis of new radio telescope measurements,
together with numerical simulations and reevaluated data from COBE and the
Hubble Space Telescope, indicates that the clouds may be orbiting the center of
mass of the Local Group, whose largest shareholders are the Andromeda galaxy
(with 65% of the mass of the group) and our own Milky Way (30%). Thus, the
clouds could be the raw material of galaxy formation. Reporting at the January
1997 American Astronomical Society meeting in Toronto, Leo Blitz of the University of California, Berkeley, and David Spergel of Princeton University said
that the high-velocity clouds will continue to feed the Milky Way (providing
fuel for future star formation) and might even harbor dark matter, which would
account for their continued stability and unexplained large internal velocities.
Spergel said that the new interpretation of the nearby high-velocity clouds might
also apply to larger, more distant hydrogen clouds in the cosmos.
THE HIPPARCOS STAR CATALOG will be published in June, with new measurements of the positions of and distances to a myriad of stars. Operating from
late 1989 until June 1993, the European Space Agency’s Hipparcos satellite recorded 1000 gigabytes of data, which had to be analyzed en masse to determine
the positions of more than 100,000 stars with milliarcsecond accuracy (a 100fold improvement over present catalogs) and the positions of a million more
stars with an accuracy between 20 and 30 milliarcseconds. This greater knowledge of star locations is quickly being put to use, as was revealed on 14 February 1997 at a meeting of the Royal Astronomical Society in London. Of special
interest are the distances to certain variable stars—Cepheid-, RR Lyrae– and
Mira-types—whose luminosity changes can be used to deduce the distances to
various clusters in the Milky Way as well as to faraway galaxies. For example,
very preliminary revisions of stellar ages and distances suggest that globular
cluster stars, thought to be the oldest stars in our Galaxy, may be only 11 (not
15) billion years old. Further, other galaxies may be perhaps 10% farther from
us than earlier studies implied. Meanwhile, resolution of the embarrassing dilemma caused by the oldest stars appearing to be older than the universe itself
still awaits peer review. (More on Hipparcos can be found in Science 275,
1064, 1997; and Science News 151, 101, 1997.)
2
A galactic cloud of antimatter. (Courtesy W. Purcell (Northwestern University) et al., OSSE, Compton Observatory, NASA.)
SPRINGTIME FOR COMET HALE–BOPP. Now long past its prime and
lost in the glow of the setting sun, Hale–Bopp was first spotted three years ago
as far away as seven astronomical units. Astronomers could thus observe the
thawing process from an earlier stage than is usual for comet watches, and
their observations covered the spectrum from ultraviolet to radio wavelengths.
So, what are they learning from the comet’s approach? First of all, the diameter of Hale–Bopp’s nucleus is estimated to be 30–40 km, at least three times
bigger than that of comet Halley. Of the cometary products vaporized on the
inward trip toward the sun, the chief gases were water vapor, carbon monoxide
and carbon dioxide, which are also the main constituents of ices in dense interstellar molecular clouds. The comet’s dust seems to contain silicates, particularly magnesium-rich crystalline olivine. Crystalline silicates are also seen in
dust around some young stars. Gas and dust production in general were more
than 20 and 100 times greater, respectively, for Hale–Bopp than for Halley at
comparable distances from the Sun. Chemical composition analysis is consistent with the comet originating in the Oort Cloud rather than the Kuiper Belt.
(Several articles in Science 275, 1997.)
Comet Hale-Bopp Home Page: (http://www.jpl.nasa.gov/comet/)
Infrared image of Hale-Bopp recorded at the NASA Infrared Telescope Facility (IRTF) in Mauna Kea Observatory on March 8, 1997. (Courtesy Alan
Tokunaga and Roland Meier, Institute for Astronomy, University of Hawaii.)
QUARK MATTER IN NEUTRON STARS could be detectable. A typical
neutron star is primarily composed of normal hadronic matter. In its dense,
high-pressure core, however, the hadrons can undergo a phase transition in
which they become deconfined into quark matter. This is the quark–gluon
plasma state currently being sought at CERN in Geneva and (in the next few
years) at the Relativistic Heavy Ion Collider (RHIC) at Brookhaven National
Laboratory. According to the Lawrence Berkeley National Laboratory’s Norman
Glendenning and his colleagues, because a rapidly spinning neutron star (pulsar) will gradually shed energy and angular momentum in the form of radio
emissions and an electron–positron stellar wind, the star will become more
spherical, with a corresponding rise in the central density, making conditions
more favorable for the creation of quark matter. The era during which the
quark core grows as the star spins down would be marked by a decreasing
moment of inertia and an anomalous value for the “braking index,” which
would differ by an order of magnitude or more from the canonical value of 3.
The deviation can be in either direction—the star might “spin up.” Glendenning
suggests that one in a hundred pulsars is undergoing the baryon-melting phase
transition, so perhaps seven of the presently known pulsars could show the
effect. (N. Glendenning, S. Pei, F. Weber, Phys. Rev. Lett. 79, 1603, 1997.)
As certain neutron stars contract, some neutrons can “melt,” creating an environment in which normally unstable particles such as hyperons and strange
quarks (dark region) can survive indefinitely and can even come to predominate in more and more of the star (a, b, c). (Courtesy AIP)
atoms, molecules, optics
THE QUANTUM/CLASSICAL BORDER has gotten a bit sharper. Ideally,
when a macroscopic measuring device is coupled to a microscopic system like
an atom, the result should be a quantum superposition of the “device+atom”
system. However, that is not observed. Instead, an irreversible reduction process seems to take place in which the macroscopic, coherent entanglement
quickly dissipates. Now, Serge Haroche, Jean-Michel Raimond, Michel Brune
and their coworkers at the Ecole Normale Supérieure in Paris have created
such an entangled system and measured its progressive decoherence. They
sent individual rubidium atoms—each of which was in a superposition of two
Rydberg states—through a cavity containing a microwave field. Each of the
two quantum states shifted the phase of the microwave field by a different
amount—so the field also became a superposition of two states. As the cavity
field exchanged energy with its surroundings, the superposition collapsed into
a single definite state. The researchers monitored the decoherence by measuring correlations between the energy levels of pairs of atoms sent through the
cavity with various time delays between the atoms. (M. Brune et al., Phys.
Rev. Lett. 77, 4887, 1996)
ULTRASHORT HARD X-RAY PULSES have been produced at Lawrence
Berkeley National Laboratory by shooting terawatt, 100-femtosecond bursts of
infrared laser light at right angles across a beam of relativistic electrons. A
highly collimated, 30 keV x-ray pulse (generated by 90 degree Thomson scattering) traveled in the direction of the electrons. The pulse duration—300 fs—
was set by the transit time of the initiating laser pulse across the electron beam.
A more tightly focused electron beam will lead to a shorter burst of x rays.
These pulses are ideal probes for glimpsing ultrafast structural phenomena in
condensed matter and biological systems. For example, the LBNL researchers
are using the x-ray pulses to study melting silicon. (R. W. Schoenlein et al.,
Science 274, 236, 1996.)
SYMPATHETIC COOLING, a process by which particles of one type cool
particles of another type, has been demonstrated for the first time with neutral
atoms. Using a combination of lasers and magnetic fields, Christopher Myatt
and his colleagues at NIST and the University of Colorado trapped a group of
rubidium-87 atoms each having one of two possible values for spin, a quantity
that describes how a particle responds to a magnetic field. Atoms with one spin
value are less tightly bound in the trap’s magnetic fields and can be used to
cool atoms with the other spin value since the weakly confined atoms could
more easily escape the trap and carry away the energy given up by the second
species during collisions. Applying this technique to the two rubidium spin
species, the researchers have created, for the first time, two overlapping clouds
of Bose-Einstein condensates, the new state of matter in which a group of
atoms falls into exactly the same quantum state. They also observed that the
BECs of the two rubidium species repelled each other. Sympathetic cooling
may help enable Bose-Einstein condensation for rare isotopes, and may greatly
facilitate comparative studies between fermions and bosons. (C.J. Myatt et al.,
Phys. Rev. Lett. 78, 586, 1997.)
A RUDIMENTARY ATOM LASER has been created at MIT, promising significant improvements in high precision measurements with atoms and offering the prospect of future nanotechnology applications, such as atom lithography, in which lines are drawn on integrated circuits (by directly depositing
atoms) with greater precision than ever before. In an atom laser the output
beam consists of a single coherent atom wave, just as in a regular laser the
beam consists of a coherent light wave. The working substance for the atom
laser is a Bose-Einstein condensate (BEC) of sodium atoms, cooled and contained within an atom trap by a shaped magnetic field. BEC itself was achieved
for the first time only as recently as 1995. It is a condition in which atoms are
chilled to such low energies that, in a wavelike sense, the atoms begin to overlap and enter into a single quantum state.
Wolfgang Ketterle and his colleagues at MIT make their claim of producing
the first atom laser on the basis of two experimental developments, as reported
in two journals this week. In the first effort (M.-O.Mewes et al., Phys. Rev.
Lett., 78, 582, 1997) a portion of a sodium condensate was successfully extracted under controlled conditions. They achieve “output coupling” by applying radiofrequency radiation to the BEC; this “tips” the atoms’ spins by an
adjustable amount, putting the atoms in a superposition of quantum states.
Thereafter some of the atoms feel the effect of the surrounding magnetic field
in a different way and are able to leave the atom trap. It is these departing
atoms, still enjoying the coherent properties of the BEC state, that constitute
an atom laser beam. Pulled downward by gravity, the beam was observed over
a distance of millimeters, although in principle it could travel further in an
undisturbed vacuum environment.
The second development was to verify that the atom waves are indeed coherent (M.R. Andrews et al., Science 275, 637, 1997). At the time of the original
BEC discovery, many physicists expected the atoms in the condensate to fall
into a single quantum state; some hypothesized that it could take a time equal
to the age of the universe for true coherence to come about. The MIT group
addressed this issue by creating two BEC clouds in a special trap. Turning off
the trap allows the clouds to expand, overlap, and interfere, producing a pattern of light and dark fringes. The observed patterns (viewed with an electronic camera) could only exist if each BEC was an intense coherent wave.
The MIT team determined that the atom wave associated with each BEC had a
wavelength of 30 microns, a million times larger than the wavelengths associated with room-temperature atoms.
In addition to coherence, the atom laser waves are analogous to the light
waves in an optical laser in another respect as well. Just as a laser beam is
more intense than an equivalent stream of light from the Sun, the MIT atom
beam is also more intense (for a given beam spotsize) than ordinary atom beams
(whose atoms possess a variety of energies) since it delivers a powerful, directional stream of atoms in a single quantum state. In other ways, the atom lasers
and light lasers are different. According to Ketterle, “Photons can be created
but not atoms. The number of atoms in an atom laser is not amplified. What is
amplified is the number of atoms in the lowest-energy quantum state, while the
number of atoms in other states decreases.”
MIT Web Site on the atom laser: http://bink.mit.edu/dallin/news.html#atomlaser
Interference of two Bose-Einstein condensates. (Courtesy MIT)
THE CASIMIR FORCE has now been precisely measured. According to
quantum electrodynamics, fleeting electromagnetic waves and particles continually pop in and out of a vacuum. However, when a pair of conducting
surfaces bound the vacuum, only electromagnetic modes with wavelengths
shorter than the distance between the surfaces can appear. According to
Casimir’s 1948 prediction, the exclusion of the longer wavelengths results in a
tiny geometry-dependent force between the conductors. Using a torsion pendulum (a twisting horizontal bar suspended by a tungsten wire), Steve
Lamoreaux of Los Alamos National Laboratory measured the attraction between a gold-plated sphere and a gold plate. He found the expected nonlinear
increase in the force as the plates’ separation decreased, rising to more than
100 microdynes at 0.6 ÿm, and agreeing with theory to within 5%. (S. K.
Lamoreaux, Phys. Rev. Lett. 78, 5, 1997.)
ATOM-INTERFEROMETER GYROSCOPES, though still lacking longterm stability, are now at least as sensitive as the best commercial laser gyroscopes, new experiments have demonstrated. By sending two beams of atom
waves along different paths and then recombining them, an interference pattern of light and dark fringes can be created. Rotating the interferometer while
the atom waves travel freely through the device makes the fringes shift from
their usual positions—known as the Sagnac effect—and enables researchers to
detect rotation rates. In one recent experiment, David Pritchard and his coworkers at MIT passed a beam of sodium atoms through a series of
nanofabricated diffraction gratings and achieved a sensitivity of 3 x 10-6 rad/s.
Meanwhile, Mark Kasevich and his colleagues at Stanford University used laser beams for diffraction gratings to realize a sensitivity of 2 x 10-8 rad/s. For
comparison, the Earth’s rotation rate is 15 degrees per hour, or 7.3 x 10-5 rad/
s. Both experiments sought short-term sensitivity in their one-second observations, not the long-term stability that must still be demonstrated for most realworld applications, such as sensing rotational effects in autos and tanks or
making inertial guidance systems for aircraft. The intrinsic sensitivity of atom
interferometers could surpass that of laser interferometers by a factor of 10 10 :
Atom wavelengths are much shorter than those of visible light, potentially
making them more sensitive to smaller changes, and their speeds are much
slower than that of light so the interferometer has more time to rotate while the
particles travel through the device, producing larger fringe shifts. (A. Lenef et
al., Phys. Rev. Lett. 78, 760, 1997. T. L. Gustavson, P. Bouyer, M. A. Kasevich,
Phys. Rev. Lett. 78, 2046, 1997.)
Lay language paper on Stanford interferometer: http://www.aps.org/
BAPSAPR97/vpr/laya6-2.html
MIT web site on “Optics and Interferometry with Atoms and
Molecules,” http://coffee.mit.edu/pubs/AAMOP/AAMOP.html
Diagram of atom-interferometer gyroscope built at Stanford University. (Courtesy Philippe Bouyer, Stanford.)
EXPLODING ATOM CLUSTERS yield high-energy ions. Femtosecond lasers can be used to convert electromagnetic energy into kinetic energy with
great pyrotechnic effect. For example, they can blow up molecules in a Coulomb explosion, imparting a kinetic energy of up to 100 eV to individual outgoing ions. Aiming femtosecond pulses at a solid can produce ions with keV
energies and substantial bulk heating. Now, in the intermediate size range,
scientists at the University of London’s Imperial College have observed much
higher energy ions (up to 1 MeV) flying away from the miniature fireball
caused by shooting ultrashort (150 fs), high-intensity (2 x 1016 W/cm2) laser
pulses at clusters of more than 100 xenon atoms. The pulse both ionizes the
atoms and rapidly heats the free electrons. The resulting charge separation
then leads inevitably to very rapid expansion of the ions in this miniplasma.
The efficiency of this process is stunning, with up to 90% of the laser energy
being transferred to the ions. Very high charge states were observed, with a
distribution peaking around 20+ but going up to about 40+. The researchers
3
speculate that with a gas of deuterium and tritium clusters tabletop fusion experiments may be possible. (T. Ditmire et al., Nature 386, 54, 1997.)
AN EXCITED ATOMIC STATE WITH A TEN-YEAR lifetime has been
observed in an ytterbium ion, raising hopes for atomic clocks a thousand times
more accurate than now possible. According to the Heisenberg uncertainty
principle, the longer a system can be observed, the smaller is the uncertainty in
its energy. Therefore, it is extremely desirable to tune an atomic clock to a
long-lived excited state because of the precision with which the transition frequency can be known. Researchers at the National Physical Laboratory in the
UK used a laser photon to boost a single cooled and trapped ytterbium ion’s
outermost electron to the long-lived, metastable 2F7/2 state. Rather than waiting ten years, the researchers subsequently boosted the electron to a yet higher
state, from which it then decayed to its ground state. From the characteristics
of the laser and the rate at which it drove the transitions, the researchers determined a 3700-day lifetime for the metastable state. In addition to being the
most durable excited energy state yet detected in an atom, this is the first time
a rare electric octupole transition—in which the electron changes its angular
momentum by a relatively large amount of three units—has been driven. An
atomic clock based on this transition would be very precise but would require
much additional development. (M. Roberts et al., Phys. Rev. Lett. 78, 1876, 1997.)
UNUSUAL PROPERTIES OF CUBANE. First synthesized in 1964, cubane
(C 8H 8) is a molecule with eight carbon atoms arranged at the corners of a
cube plus single hydrogen atoms sprouting symmetrically from each carbon.
Thus the C–C–C bond angle is 90°, rather than the 109.5° customary in other
hydrocarbon molecules. A great deal of strain energy is therefore stored in the
chemical bonds—150 kcal/mole or 6.5 eV per molecule. Replace all the Hs
with NO2 groups and you get a terrific fuel or explosive, with nearly twice the
power of TNT. Replacing the Hs with other chemical groups results in potentially more salubrious derivatives, some of which are currently undergoing tests
for fighting the AIDS virus, bone marrow cancer and Parkinson’s disease. For
all its potential, however, solid cubane is not well understood. Like other molecular solids (such as solid C60), in which molecules rather than atoms make
up the underlying lattice, cubane exhibits a “plastic phase” close to its melting
point in which the molecules start to swivel about one or more of their axes.
Recently, a University of Chicago–NIST collaboration has experimentally
worked out the structure of this plastic phase, using x-ray powder diffraction.
Surprisingly, unlike most other molecular solids, the plastic phase of cubane is
not face-centered cubic. It’s rhombohedral. The researchers’ work shows that
cubane undergoes a large volume expansion of 5.4% at the first-order phase
transition temperature of 394.3 K. (T. Yildirim et al., Phys. Rev. Lett. 78,
4938, 1997.)
Cubane, a novel molecule made of eight carbon atoms (the red spheres) and
eight hydrogen atoms (the blue spheres), may have a use in a variety of compounds. (Image courtesy Taner Yildirim, National Institute of Standards and
Technology.)
STORING THE MAXIMUM AMOUNT of classical information in a photon
or any other quantum particle is possible even in the presence of noise, researchers have concluded. Various properties, such as polarization or energy,
of a photon can be used to store information. Using polarization, for example,
horizontal could represent “zero,” vertical could be “one,” 45° could be “two,”
and so on. Furthermore, one can store many digits simultaneously in a single
photon by putting it into a superposition of many states. However, quantum
mechanics prevents a measuring device from perfectly distinguishing between
all these different states. Previously, physicists discovered that, no matter how
much is “written” on a photon, the maximum amount of information that can
be read can be no greater than the amount of entropy in the ensemble of signals
that were sent. Whether this limit was realizable was an open question. Now,
two investigations have independently shown that this upper limit can be
reached, even in a noisy environment involving mixed states. Several strategies, such as employing only those quantum states that are most distinguishable, were found to be useful. (B. Schumacher, M. D. Westmoreland, Phys.
Rev. A 56, 131, 1997; A. S. Holevo, to be published in IEEE Trans. Inf. Theory,
1997.)
AIP web site on this work: http://www.aip.org/physnews/preview/1997/qinfo/
TAKING THE TEMPERATURE OF DARK-STATE ATOMS. Shortly after the 1997 Physics Nobel Prize, Claude Cohen-Tannoudji and his colleagues
at the Ecole Normale Superieure in Paris have announced a new way to explore the coldest realm in the universe. One problem in this line of research is
that traditional methods of thermometry have failed to measure how truly low
the temperatures are for the large clumps of ultracold atoms in the “dark state.”
Traditionally, physicists have used “time-of-flight” methods: by allowing a cloud
of ultracold atoms to expand freely and measuring how quickly the cloud expands, the researchers estimate the range of velocities in the gas atoms; a nar-
4
rower range corresponds to a colder temperature. But for large clouds of ultracold
atoms, including those in the dark state, the clouds expand too imperceptibly for
researchers to make good temperature measurements. To create the dark state,
researchers first trap and cool helium atoms using a combination of laser beams
and magnetic fields. Then, two laser beams traveling in opposite directions put
each atom into a combination or “superposition” of two low-energy states that
interfere with each other so as to prevent the atoms from absorbing or emitting
laser light. This is important since a helium atom absorbing or emitting a single
photon recoils by 9.2 cm/second, corresponding to a temperature of 4
microkelvins. Oblivious to photons, atoms in dark states can have temperatures
well below this “single photon recoil limit.” To determine these “subrecoil” temperatures more precisely, Cohen-Tannoudji’s group probes the wavelike properties in the group of atoms. Each dark-state atom can be thought of as a superposition of two “wavepackets,” corresponding to the two low-energy states which
interfere to prevent light absorption. Associated with the two wavepackets are
two equal and opposite momentum states characterizing the movement of the
atom as a whole; in effect the atom is moving in two opposite directions at the
same time. As long as the dark state lasers are on, these two wavepackets are
constantly superimposed. But when the researchers turn off the lasers in their
experiment, the two wavepackets fly apart. A subsequent laser pulse applied
after a certain time measures the various degrees of overlap in the wavepacket
pairs that make up the cloud of atoms, allowing the researchers to measure the
momentum (and therefore velocity) distribution of the atoms and thereby the
temperature as well. Applying this technique to subrecoil helium atoms, the researchers have measured a temperature (at least in the one dimension probed by
their laser) of 5 nanokelvins, 1/800 of the recoil limit. This is the lowest fraction of the recoil temperature ever measured for an atom; the lowest absolute
temperature, 3 nanokelvins for much heavier cesium atoms, was measured by
the same group in 1995. (B. Saubamea et al., Phys. Rev. Lett. 79, 3146, 1997.)
biological/medical physics
IMPORTANT PROCESSES IN SINGLE DNA MOLECULES have been
observed for the first time by using the atomic force microscope (AFM), in
which the deflections of a tiny stylus over the contours of a surface can be
turned into molecular-scale images. At the 1997 March APS Meeting in Kansas City, Carlos Bustamante of the University of Oregon and his colleagues
presented movies showing the first stages of DNA replication, in which a protein is seen to slide on DNA like a bead on a string to find the exact site where
it could attach and start the replication process. Binding DNA and RNA
polymerase (the protein that mediates the transcription of DNA into RNA) to a
mica surface, Neil Thomson of UC-Santa Barbara and his colleagues produced
5-nm-resolution movies of the transcription process, in which RNA polymerase
pins down the middle of a single DNA strand and then pulls the strand through
as it starts transcribing the DNA into RNA using RNA-building-blocks called
NTPs (Biochemistry 36, 461, 1997.) Using an AFM, Gil Lee of the Naval
Research Laboratory found that a force of about 600 piconewtons was required
to tear apart two complementary strands of DNA, namely a 20-base-pair-long
strand of polycytosine (a form of single-strand DNA) from single strands of
polyinosine averaging 160 base-pairs long.
A sequence of images, taken with an atomic force microscope (AFM), shows a
single DNA molecule being transcribed into RNA. (Courtesy Neil Thomson/
Biochemistry.)
WHITE BLOOD CELLS SORT THEMSELVES by type in artificial capillaries. While 7 µ red blood cells (which carry oxygen) easily deform to pass
through a 4–5 µm capillary, 10 µ white blood cells (which are essential to the
immune system) must squeeze through the entrances and stick at the exits with
the help of complex elastic, chemical and hydrodynamic forces. Robert Austin
(Princeton University) and his collaborators have built an array of artificial
capillaries—rows of side-by-side 5 x 5 µm polyurethane channels, with lengths
ranging from 20 to 110 µm. Sending fluorescently labeled white blood cells
through the channels, the researchers observed that T-lymphocytes (antibodyproducing cells from the thymus) penetrated farther into the array than did the
granulocytes and monocytes. Modeling showed that the T-cells lowered their
sticking probability in proportion to the local granulocyte cell density, a form
of “heteroavoidance.” The unexpected self-sorting process suggests that the
artificial capillaries could serve as a tool for isolating white blood cell populations and identifying blood cell disorders. (R. H. Carlson et al., Phys. Rev.
Lett. 79, 2149, 1997.)
A thin intermediate layer of gallium arsenide is twistbonded to the gallium arsenide substrate in a technique developed by Cornell scientists. The boundaries are defect-free, making it suitable as a semiconductor. (Courtesy Cornell University)
White blood cells known as T-lymphocytes (orange-red) avoided sticking to regions already occupied by granulocytes and monocytes (both labelled green),
which are two other types of white blood cells. (Courtesy R. H. Carlson et al./
Physical Review Letters)
A GOLD BIOSENSOR has been demonstrated. Single strands of DNA have
a remarkable talent for recognizing and attaching to complementary strands.
If one strand is snipped in half, the intact complementary strand will bring the
“lost” halves back together. This is the principle used by Northwestern University scientists, who glued various “probe” DNA segments onto 13 nm gold
particles suspended in a solution. When an intact “target” strand of DNA in
the solution happens to complement DNA segments already stuck to the particles, the probes and target link up, drawing the nanoingots together into a
polymer-linked network. The optical properties of this web of gold are very
different from those of the original solution—the color changes in a well-defined temperature-dependent way. Should the target have one or two base imperfections (such as base pair mismatches), the color change occurs at a different temperature. The researchers have extended this approach to a simple
solid-state spot test, which under unoptimized conditions can detect 10
femtomoles of single-strand target DNA. (R. Elghanian et al., Science 277,
1078, 1997.)
condensed matter/
materials physics
CAN HYDROGEN BE A HIGH-Tc SUPERCONDUCTOR? That hydrogen
can be metallic was demonstrated last year (PHYSICS TODAY, May 1996),
when it was also established that the H nuclei (protons) remained largely paired.
Now, although theoretical predictions of superconducting transition temperatures have been notoriously difficult to make, two Cornell University physicists, Neil Ashcroft and Clifton Richardson, have used an approach that works
well when applied to conventional metals to predict that both atomic and diatomic hydrogen should have superconducting phases near or above room temperature—but only at megabar pressures. Their calculations for mon-atomic
metallic hydrogen agree with earlier work. For proton-paired metallic hydrogen, however, an interesting difference arises in the direct electron–electron
term that normally works against superconductivity. The theorists found correlated fluctuations between electrons and holes in overlapping bands at very high
pressure, which reduce that term. To date, the trademark zero resistance cannot
be seen directly in a tiny sample within a diamond-anvil cell, because the probes
are easily pinched off, but indirect, inductive methods might succeed. (C. F.
Richardson, N. W. Ashcroft, Phys. Rev. Lett. 78, 118, 1997.)
METAL INCLUSIONS CAN HAVE QUANTIZED SIZES. Whether or not
nature dictates the size and shape of chunks of one element embedded in another solid is important at the microscopic level. For example, the melting point
of some materials can be raised or lowered considerably by burying extremely
small pieces of them inside another material. A Berkeley–Copenhagen–Rio de
Janeiro collaboration has now shown that lead inclusions, a few nanometers
across, in equilibrium within an aluminum matrix, assume only special (“magic”)
sizes. These preferred sizes (and the inclusions’ faceted shapes) are determined
by the material minimizing the residual strain energy imposed by the crystalline
mismatch between the two elements. In time, this magic-size phenomenon might
be useful for tailoring specific thermodynamic, magnetic, electronic, or optical
properties of materials. (U. Dahmen et al., Phys. Rev. Lett. 78, 471, 1997.)
A ONE-SIZE-FITS-ALL SUBSTRATE for semiconductors has been demonstrated, potentially allowing researchers to deposit crystals of many previously
mismatched materials onto a semiconductor surface. If the lattice spacing of a
crystal differs by as little as 1% from that of the surface onto which it is deposited, defects can form that prevent the proper functioning of the material. YuHwa Lo and his colleagues at Cornell University have engineered a “compliant
substrate” by bonding a thin (3–10 nm) film of gallium arsenide to a bulk GaAs
substrate in such a way that the film’s lattice is twisted relative to the bulk. The
resulting lattice structure is more compliant to the lattice of a different crystal
grown on its surface than it would be without the twist-bonded film. So far,
working with crystal growers at Sandia National Laboratories and at the Materials Directorate of Wright-Patterson Air Force Laboratories, the researchers
have successfully deposited crystals of InGaP, GaSb and InSb—the latter two
of which had previously been unachievable. Even with lattice mismatches between crystal and surface as high as 15%, the density of defects was reduced by
a factor of 105 compared to that for regular substrates. Moreover, if gallium
nitride (mismatched by 20%) could be deposited onto this surface, the researchers believe that high-quality blue and ultraviolet semiconductor lasers might
result. (F. E. Ejeckam et al., Appl. Phys. Lett. 70, 1685, 1997.)
SINGLE MAGNETIC ATOMS disrupt superconductivity on an atomic scale. As part of their microscopic study of magnetism, Ali Yazdani and his colleagues at IBM’s Almaden Research Center deposited single manganese and gadolinium atoms, each
of which has a magnetic moment, onto a niobium surface, which is a superconductor at low temperatures.
By measuring the tunneling current that flows from
the surface to the probe of a scanning tunneling microscope (STM), the researchers detected a unique change in the superconductor’s
properties. Nonmagnetic silver atoms had no significant effect. The researchers’ model calculation indicates that the STM actually measures the presence of
an unpaired electronic state localized around the magnetic atom. (A. Yazdani et
al., Science 275, 1767, 1997.)
Simultaneously acquired topographic (top) and spectroscopic (bottom) images
of three gadolinium atoms on top of a superconducting niobium surface. (Courtesy IBM)
A SUPERFLUID GYROSCOPE, designed like an AC SQUID, has been demonstrated by Richard Packard and his colleagues at University of California,
Berkeley. SQUIDs—superconducting quantum interference devices—exploit a
peculiar property of superconductors: The amount of magnetic flux through a
circulating supercurrent must be a multiple of a basic flux unit. At the heart of
such a device is one or more very thin insulating barriers that interrupt the ringshaped superconductor. Electron pairs tunneling through the insulation interfere with each other in a way that depends on the amount of flux threading the
superconducting circuit; thus the quantization of flux can be exploited to measure tiny magnetic fields. In a superfluid, by contrast, the quantized quantity is
fluid circulation, which can be exploited to measure tiny rotations. In the Berkeley experiment, the flow of superfluid helium through a ring-shaped vessel is
interrupted by a barrier containing a submicrometer-sized pinhole. When the
<R>vessel is rotated, the helium must squirt back through the hole to maintain
constant circulation in space, and the squirting can be monitored. With their
proof-of-principle demonstration, the metrologists measured the rotation of Earth
with a precision of 0.5%. (K. Schwab, N. Bruckner, R. E. Packard, Nature 386,
585, 1997.)
DISCRETE ELECTRONIC STATES in a metal quantum dot have been investigated by physicists at Harvard University. In general, reducing an object’s
dimensionality makes its quantum nature more manifest. In a semiconductor,
for example, confining mobile electrons to a two-dimensional plane, a one-dimensional wire or a zero-dimensional dot enforces an ever sharper limit on the
allowed energies, and this effect can be exploited in producing compact and
highly controllable electronic devices. Michael Tinkham, Dan Ralph (now at
Cornell University) and Charles Black (now at IBM’s T. J. Watson Research
Center) succeeded in attaching leads to a 10 nm aluminum particle to make a
tunneling transistor. The spectrum for “electron-in-a-box” energy levels for the
interacting electrons inside the particle was measured from the current–voltage
curves. Unlike a semiconductor dot, a metal nanoparticle can be made superconducting or ferromagnetic, and knowledge of the individual energy levels can
provide a detailed view of the forces that govern these properties. Indeed, in
such a speck of aluminum, the group observed the electron spectrum while an
applied magnetic field broke up the superconducting state by flipping one electron spin at a time. Nonequilibrium excitations were also detected and characterized. (D. C. Ralph, C. T. Black, M. Tinkham, Phys. Rev. Lett. 78, 4087, 1997.)
AN AMORPHOUS SOLID CAN BEHAVE LIKE A CRYSTAL. For a quarter-century, all amorphous solids have been found to damp out low-energy vibrations quickly. Put another way, they all have had comparable, relatively high
values of “internal friction.” By contrast, if you shake a pure silicon crystal,
chilled to cryogenic temperatures, it will ring for an hour or more at various
frequencies. Now a collaboration of physicists from Cornell University and the
National Renewable Energy Laboratory in Golden, Colorado, has incorporated
a small, judicious amount of hydrogen into amorphous silicon, greatly improving the electronic stability of the material by “tying down” some of its dangling
bonds. In the process, the group found something unexpected: The low-energy
vibration modes persisted for an hour, just as in the material’s crystalline counterpart. This as-yet-unexplained property gives the researchers an experimental
tool for exploring the role of hydrogen in these solids and for studying amorphous solids in general. For example, one can observe what happens to these
low-energy excitations as impurities are added to the material. Amorphous silicon is a low-cost rival to crystalline silicon for photovoltaic applications. (X.
Liu et al., Phys. Rev. Lett. 78, 4418, 1997.)
5
A MOLECULAR VOLCANO, perhaps related to episodic gas releases from
comets, has been seen by Bruce Kay and his colleagues at the Pacific Northwest
National Laboratory. They deposited an ultrathin (5.4 monolayers) film of carbon tetrachloride (CCl4) on a cold gold substrate, followed by water vapor. At
the frigid temperature involved (85 K), the H 2O was flash frozen into amorphous solid water (ASW), not crystalline ice. With a thick slab (30 monolayers)
of ASW blanketing the CCl4, they then slowly warmed it all up. They found
that the CCl4, which would normally begin to desorb at 130 K and be completely gone at 142 K, was held firmly in place by the ASW until, at a temperature of 156 K, it all burst forth at once, like magma erupting from beneath Earth’s
crust. They also investigated an overlayer of amorphous solid D 2O and found
abrupt desorption at 160 K. The temperature at which ASW begins to crystallize into something more recognizable as ice is 156 K, and 160 K is the analogous temperature for D2O. Thus, the researchers suggest that the structural
changes—such as cracks, fissures and grain boundaries—associated with this
phase transition provide the necessary escape channels for the CCl4. Previous
suggestions involved diffusion and pressure as the drivers of desorption. This
work was part of a general study of ASW, which is of interest in areas as diverse
as glassy materials, cryobiology and cometary and interstellar ices. (R. S. Smith
et al., Phys. Rev. Lett. 79, 909, 1997.)
ment, 2.7 nm quantum dots start out 1.2 nm apart from each other, at which
distance the film is an insulator and the dots are classically coupled, transferring
energy by mutually inducing charge polarization across the intervening dielectric. As the separation between nanocrystals decreased below 1.0 nm, the researchers observed changes in the optical properties, signaling a transition to
quantum coupling, in which the nanocrystals’ electron clouds overlapped. Below 0.5 nm separation, a sharp transition to a metallic-like film was observed.
All changes are reversible. This work is part of a large effort to create new
materials with “tunable” chemical properties. (C. P. Collier et al., Science 277,
1978, 1997.)
A camcorder still shows that the compressed film has a metallic sheen.
Above a temperature of 140 K, amorphous water begins to form tiny crystalline
grains of ice (b). When enough ice grains have formed, carbon tetrachloride
molecules, trapped under the amorphous water layer, can now escape by percolating up between the ice grains and emerge as a “molecular volcano” (c).
(Courtesy AIP)
DISCRETE ELECTRONIC STATES in a metal quantum dot have been investigated by physicists at Harvard University. In general, reducing an object’s
dimensionality makes its quantum nature more manifest. In a semiconductor,
for example, confining mobile electrons to a two-dimensional plane, a one-dimensional wire or a zero-dimensional dot enforces an ever sharper limit on the
allowed energies, and this effect can be exploited in producing compact and
highly controllable electronic devices. Michael Tinkham, Dan Ralph (now at
Cornell University) and Charles Black (now at IBM’s T. J. Watson Research
Center) succeeded in attaching leads to a 10 nm aluminum particle to make a
tunneling transistor. The spectrum for “electron-in-a-box” energy levels for the
interacting electrons inside the particle was measured from the current–voltage
curves. Unlike a semiconductor dot, a metal nanoparticle can be made superconducting or ferromagnetic, and knowledge of the individual energy levels can
provide a detailed view of the forces that govern these properties. Indeed, in
such a speck of aluminum, the group observed the electron spectrum while an
applied magnetic field broke up the superconducting state by flipping one electron spin at a time. Nonequilibrium excitations were also detected and characterized. (D. C. Ralph, C. T. Black, M. Tinkham, Phys. Rev. Lett. 78, 4087, 1997.)
FRACTIONALLY CHARGED CARRIERS have been detected experimentally. Charge carriers come in a variety of forms, such as electrons in copper
wires, pairs of electrons in superconductors, and even holes (the absence of an
electron) in certain semiconductors and high-temperature superconductors. More
precisely, a hole is a “quasiparticle,” an excitation of a physical system (e.g., a
chunk of silicon) as a whole. Quasiparticles are important (some of the data in
your computer is encoded in the form of holes) but quasiparticles can’t exist
independently of the lattice through which they move; they arise from the collective behavior of many electrons. Such a collective behavior is at the heart of
the quantum Hall effect, a phenomenon in which, at conditions of low temperature and high magnetic field, the electrons at the boundary between two semiconductors form a two-dimensional electron liquid possessing discrete energy
states and exhibiting a quantized electrical resistance. Theorists predicted more
than a decade ago that excitations in some of the collective electron states could
have a charge equal to a fraction of the basic electron charge e, but only now
have scientists been able to confirm this view in the lab. Using the latest techniques for making very small electrical contacts (100-300 nm) and for detecting
minuscule currents, researchers at the Condensed Matter Lab at CEA/Saclay
and the Lab for Microstructures and Microelectronics in Bagneux, France, have
studied the “shot noise” emerging from a tiny GaAs sample. This form of noise
represents the fluctuation in the current owing to the random way (governed by
quantum mechanics) in which carriers tunnel from one side of an electrical junction to the other (reminiscent of the discrete fall of raindrops on a roof). What
the French researchers found in probing the “granularity” of the quasiparticle
carriers in the sample was that their charge equaled e/3, demonstrating that fractional charges could carry the current in a conductor. The French results (L.
Saminadayar et al., Physical Review Letters) were obtained by measuring current fluctuations at kHz frequencies, while a competing group (publishing separately) at the Weizmann Institute in Israel, taking a comparable approach, worked
in the MHz range.(Also please note the work of Goldman and Su, Science, 17
February 1995.)
A REVERSIBLE INSULATOR–METAL TRANSITION has been created, at
room temperature and pressure, in a quantum dot monolayer. University of California researchers, led by James Heath of UCLA, prepared a Langmuir film—a
single layer of nearly identically sized silver nanocrystals, or quantum dots. Each
silver dot was coated with compressible organic molecules, which allowed the
dots to interact via attractive dispersion forces. By increasing the surface pressure on the film, the particles were brought closer together. In a typical experi-
6
TURNING ONIONS INTO DIAMONDS. Graphite material can be made into
diamond the hard way, using very high pressure (above 10 GPa) and temperature (typically above 1600 K), and requiring catalysts to produce good yields.
Now, significant yields of nanodiamonds have been nucleated and grown by
irradiating carbon “onions” (nested shells of graphite) with ion beams. Florian
Banhart’s and Heinz-Dieter Carstanjen’s groups at the Max Planck Institute for
Metals Research, in Stuttgart, used a beam of 3 MeV neon ions to pelt the onions for 30 hours; the process created vacancies and interstitials in the graphite
through knock-on collisions with the carbon nuclei. Diamonds nucleated in the
regions of high curvature at the cores of the onions, which acted like miniature
pressure cells. After nucleation, the diamonds grew without the need of high
pressure. Electron beams were previously used, but with diamond yields that
were smaller by a factor of 10 5–10 6. With an accelerator capable of higher ion
currents than theirs, the researchers believe that macroscopic amounts of irradiation-induced diamond can be produced. Because no catalysts are needed, the
diamonds are very pure. (P. Wesolowski et al., Appl. Phys. Lett. 71, 1948, 1997;
M. Zaiser, F. Banhart, Phys. Rev. Lett. 79, 3680, 1997.)
Using particle beams, a “carbon onion,” a
structure consisting of nested fullerene-like
balls, can be converted into a diamond.
Here a growing diamond can be seen inside
concentric graphitic layers. The diamonds
can assume sizes of up to 100 nanometers.
(Image courtesy of Florian Banhart, Max
Planck Institute in Stuttgart, Germany.)
A NEW WAY TO CALCULATE nuclear
magnetic resonance (NMR) spectra can be
used for real materials. NMR is well known
as an imaging technique, but it is also a valuable spectroscopic tool for deducing
the chemical, electronic and geometric structures around nuclei in different environments. The spectrum of an atom’s nuclear magnetic states depends on the
local geometry, just as an atom’s electron spectrum changes if the atom is suddenly lodged in a crystal. Previously, NMR calculations could produce spectra
only for isolated atoms or clusters. Now, Steven Louie and his collaborators at
the University of California, Berkeley, have devised a method for dealing with
divergent terms, making possible rigorous calculations of NMR spectra of extended systems such as crystals, liquids, polymers or even amorphous or biological materials. They have now used their technique on an industrially important material—synthetic diamond films. Their theoretical NMR spectra for various carbon configurations are in very good agreement with the observed spectra, leading them to a microscopic, physical interpretation of otherwise obtuse
experimental data. (F. Mauri, B. G. Pfrommer, S. G. Louie, Phys. Rev. Lett 79,
2340, 1997.)
earth sciences/solar system
THE EARLY FAINT SUN PARADOX goes as follows: 4 billion years ago
the sun (its fusion fire not yet having worked up to present levels) was 25-30%
cooler than now. Terrestrial temperatures would have been sub-freezing, precluding liquid water. How then did life form in these early eras? Carl Sagan,
in a posthumous paper co-authored by Chris Chyba (Science, 22 May) suggests
a possible scenario. Ultraviolet radiation from the sun, they argue, would combine with existing methane to form solid hydrocarbons in the upper atmosphere.
This in turn would shield ammonia (otherwise broken up by the UV) long
enough for the ammonia to produce a greenhouse warming adequate for liquid
water. Sagan and his interest in life in extreme environments was the subject
of a session at the Spring 1997 meeting of the American Geophysical Union in
Baltimore. According to David Morrison of NASA Ames, there are only two
places on Earth where life has not been found—on the Antarctic ice sheet and
in the upper atmosphere. Everywhere else, whether in hot springs (even above
boiling temperatures) or a kilometer below the surface, life seems to thrive.
One speaker, Todd Stevens of the Pacific Northwest Lab, asserted that some
subsurface “rock-eating” microbes constituted an ecosystem independent of
photosynthesis and that their metabolism (in some cases amounting to a biomass doubling time of millennia) was perhaps the slowest of all life forms.
DETECTING LONG-TERM TRENDS in ultraviolet radiation can be a problem. A controversial 1988 study concluded that the UV radiation reaching the
ground was actually decreasing overall, along with the stratospheric ozone that
absorbs it. Now, that claim has been laid to rest. Betsy Weatherhead (with
NOAA) and her colleagues have published an exhaustive analysis—statistical,
geophysical and instrumental—of data from 14 US sites covering 1974–91.
Within the data, they discovered severe limitations for establishing trends. After
they eliminated several geophysical causes (ozone, clouds, haze, particulates
and temperature) of the apparent UV decrease, they were led to conclude that
the trend was in the network of instruments. The UV meters were originally
intended to measure relative changes in UV at specific locations, not absolute
trends on a continental scale. Thus, over the years, there were changes in
network management, calibration techniques, and even instrument locations,
all attended by poor record keeping. “For one site,” said Weatherhead, “we
found an 11% drop in UV that couldn’t be accounted for. Later, we discovered
that an antenna had been built there, which immediately explained it.” (E. C.
Weatherhead et al., J. Geophys. Res. 102 (D7), 8737, 1997.)
CHAOS CONTROL OF EL NIÑO in a sophisticated computer simulation
has been achieved by an Israel-US team. El Niño is a prolonged warming of the
Pacific Ocean surface near the equator every 3 to 6 years, bringing about storms
and widespread climate effects. Recent theories suggest that El Niño is chaotic: its behavior is unpredictable but sensitive to initial conditions of such
variables as temperature, atmospheric pressure and winds. A Weizmann-Columbia group altered the magnitude of ocean waves reflecting from the western boundary of the Pacific Ocean, in a realistic El Niño prediction model
developed at Lamont-Doherty Earth Observatory in New York. When they did
this, the model produced an El Niño about every 4 years with periodic and
perfectly predictable cycles of temperature, winds, and ocean currents. Although the researchers do not propose to apply chaos control directly to El
Niño, they believe it may help them better understand the crucial factors governing El Niño’s behavior. In addition, the research required improvements in
existing chaos control methods, as the El Niño model has many more variables
and parameters than previously controlled chaotic systems. (Eli Tziperman et
al., Phys. Rev. Lett. 79, 1034, 1997.)
NOAA Site on El Niño: http://www.elnino.noaa.gov/
Actual sea-surface temperature (SST) anomalies recorded on January 20, 1998.
Such anomalies are typical in an El Niño episode. (Courtesy National Oceanic
and Atmospheric Administration)
EARTH’S CRUST AND MANTLE might have slewed around quickly in the
early Cambrian period, like an orange’s skin coming loose from the fruit and
slipping about. This hypothesis, advanced by Caltech geobiologist Joseph
Kirschvink and his colleagues, is based on his study of the pattern of fossil
magnetism in rock strata worldwide. The colossal slippage of the outer part of
the Earth relative to the core, a process called true polar wander, would have
arisen from the differently oriented moments of inertia in the two regions coming more into alignment. This slippage would have produced a rapid (in about
15 million years), global redistribution of surface topology, although the planet’s
spin axis would have remained fixed. During this newly postulated geodynamic
event, the crustal plates, which today float along on top of the mantle at a
stately pace of a few centimeters per year, would have raced along (as one unit
with the mantle) at up to a meter per year. The ensuing grand trek could have
rotated the continental landmasses by as much as 90 degrees. All of this happened in a blink of the geological eye, according to Kirschvink, about 530
million years ago, and was thus simultaneous with, and might have influenced
or caused, the “Cambrian explosion,” the greatest evolutionary proliferation of
diverse living organisms in our planet’s history. (J. L. Kirschvink, R. L.
Ripperdan, D. A. Evans, Science 277, 541, 1997.)
OCEANS’ CIRCULATION COULD BE HALTED by the rapid addition of
atmospheric greenhouse gases and the associated global warming, according
to Thomas Stocker and Andreas Schmittner of the University of Bern in Switzerland. Warm surface currents such as the Gulf Stream transport huge amounts
of heat to northern latitudes, and release it to the atmosphere. The loss of heat
is accompanied by an increase in water density; the denser water sinks and
returns south as a cold, deep flow. The addition of freshwater—less dense than
seawater—slows down the overturning. Beyond a critical threshold of freshwater input (first proposed in 1961), the entire “thermohaline circulation” grinds
to a stop, with severe consequences for global clmate and ocean ecology. Enter
global warming, which is expected to both warm the surface waters and increase high-latitude precipitation, pushing the North Atlantic Ocean closer to
the threshold; this effect has been seen in models before. The Bern researchers
used a simplified coupled ocean–atmosphere climate model—consistent with
the more elaborate, more computer-intensive models—and found, surprisingly,
that the rate of increase of greenhouse gases is as crucial a parameter for reductions in the thermohaline circulation as the actual amount of the gases. If
confirmed by other modelers, the result implies that acting soon to slow down
greenhouse gas emissions could help stabilize ocean circulation for the long
haul. (T. F. Stocker, A. Schmittner, Nature 388, 862, 1997.)
nanotechnology
A NANOTUBE HAS NOW BEEN USED as a microscope probe. Richard
Smalley and his colleagues at Rice University have glued single carbon nanotubes
(several micrometers long but only 5–20 nm wide) to the tips of scanning tunneling microscopes and scanning force microscopes. These long, slender tips
are reproducible, conductive, inexpensive and useful for mapping rugged terrain, even the bottoms of trenches. Unlike a standard, pyramidal probe, a
nanotube-tipped probe easily survives “catastrophic” crashes with the surface—
it simply bends, then springs back to its original straight shape when withdrawn.
(H. Dai et al., Nature 384, 147, 1996.)
COUNTING ON NANOTECHNOLOGY, researchers have demonstrated a
molecular abacus. Scientists at IBM’s research laboratory in Zurich have used a
scanning tunneling microscope probe to reposition carbon-60 molecules along
the “rails” of a stepped copper substrate, making a room-temperature device
capable of storing and manipulating numbers at the single molecule level. The
big, sturdy buckyballs can be pushed back and forth many times. “The device is
slow,” says James Gimzewski, “but we may soon see arrays of hundreds and
even thousands of STM probes for simultaneously imaging, and repositioning,
many atoms and molecules.” (M. T. Cuberes, R. R. Schlittler, J. K. Gimzewski,
Appl. Phys. Lett. 69, 3016, 1996.)
A series of STM images showing the numbers
0 through 10 represented by single carbon-60
molecules (buckyballs) on a copper surface.
(Courtesy IBM Zurich Research Laboratory.)
TINY SILICON BRIDGES that can form the
basis of a new class of charge, particle and
energy sensors have been fabricated by two
Caltech physicists. Andrew Cleland and
Michael Roukes used a combination of photolithography, electron beam lithography and
etching techniques to create minuscule, freestanding single-crystal silicon bridges, each about 0.2 mm thick and suspended
about 0.5 mm above a silicon substrate. A double torsional oscillator structure
is suspended, supported by the fingers around the periphery. A current in a gold
wire around the circumference of the washer, combined with a magnetic field in
the plane of the structure, generate a Lorentz force that causes the structure to
oscillate about its symmetry axis. The particular resonance frequency depends
on the structure’s size; thus far, resonances ranging from 400 kHz to 120 MHz
have been achieved. The nano-engineers are trying to push the technology to a
few GHz, at which point macroscopic quantum effects and interactions with
phonons might be observable. (Resonant bridges are described in A. N. Cleland,
M. L. Roukes, App. Phys. Lett. 69, 2653, 1996.)
A NEW ELECTROLUMINESCENT DEVICE runs on 15–25 V, an order of
magnitude less than the 150–200 V needed by current devices. Head-mounted
displays for automobile, aircraft and microsurgery environments need small currents and voltages to be practical. At the heart of today’s thin film electroluminescent (TFEL) devices is a host material such as zinc sulfide doped with luminescing centers such as manganese atoms. Electrons are supplied by defects at
the interfaces with insulating layers on either side of this material. The high
voltage, applied across the whole sandwich, launches electrons into the ZnS where
they excite manganese atoms, which then emit light. The new TFEL concept,
developed at the Georgia Institute of Technology, employs tailored band offsets
between silicon (as the electron source), the very thin insulating layers and the
luminescing material. The offsets give the electrons their required kinetic energy using much less voltage. The efficiency of the new device is still low and
the cost of growing the crystalline insulating layers is comparatively high, but
the lower voltage requirements, and the smaller circuitry this will permit, may
make the approach worthwhile. (C. J. Summers et al., Phys. Rev. Lett. 70, 234,
1997.)
A SINGLE-ELECTRON MEMORY, operating at room temperature, has been
developed at the University of Minnesota. In electronics, “smaller” usually means
faster response, less power consumption and more components per chip. In the
tiny Minnesota transistor, a bit of information is stored in the form of a single
electron, resident on a dot of silicon (acting as a “floating gate”) and having the
ability to influence the current flow in a silicon channel. The dot is about 7 nm
square by 2 nm thick, and the channel is about 10 nm wide—much less than the
70 nm Debye screening length for a single electron. The device is orders of
magnitude smaller than, but otherwise similar to, the kind of metal oxide semiconductor (MOS) transistor used in conventional computer memories. (L. Guo,
E. Leobandung, S. Y. Chou, Science 275, 649, 1997.)
MAGNETIC RESONANCE FORCE MICROSCOPY (MRFM) is inching
toward detecting the spin of a single electron, which will be a major milestone.
Using a highly sensitive cantilever 230 microns long and a mere 55 nm thick, an
IBM–Stanford team of physicists led by Daniel Rugar has now detected a force
of six attonewtons (6 x 10 -18 N), which should easily detect the expected force
due to a single electron spin, about 80 x 10 -18 N. The force due to a nuclear
spin is three orders of magnitude weaker yet. The team’s result was reported at
the March 1997 meeting of the American Physical Society in Kansas City, Missouri. The ultimate goal of this nascent technology is to locate and chemically
identify individual nuclear spins with 0.1 nm spatial resolution and thus gener-
7
ate three-dimensional images of molecular structures without destroying them.
The marriage of magnetic resonance imaging (which requires at least 10 14 nuclei for detection) and atomic force microscopy (which can image only a sample’s
surface) is being arranged by several groups. The technique uses a nanometerscale magnetic tip on the cantilever to create an inhomogeneous magnetic field
in a sample, causing spins to precess at varying rates. An RF coil is then tuned
to resonate with the spins in a particular value of the magnetic field, and a spin
in that “resonant slice” is detected by measuring a small oscillating force between the tip and the spin. Three-dimensional images can be formed by scanning the magnetic tip over the sample and varying the RF frequency. The IBM–
Stanford group hopes to map the spins of electrons in dispersed defect sites in
silicon dioxide. “The biggest villain is noise,” said John Sidles of the University of Washington’s School of Medicine. Sidles, who originated the concept of
MRFM, stressed that many of the most interesting biological structures are unknown on the nanometer scale, and cannot be crystallized for x-ray studies.
MRFM could be used to determine the binding sites of the HIV virus, for example. Also at the APS meeting, Chris Hammel represented Los Alamos National Laboratory’s collaboration with Michael Roukes of Caltech in applying
MRFM to multilayer electronic devices. They hope eventually to map buried
structures, such as defects at interfaces between layers.
Lay language paper on this topic: http://www.aip.org/physnews/graphics/condensed/1997/mrfm/sidles.htm
In a typical magnetic resonance force microscopy (MRFM) setup, a thin silicon
cantilever is poised above a tiny sample
to be imaged. (Courtesy University of Washington.)
PROTON TRANSISTOR MEMORY.
Electrons do most of the work in electronic
devices; indeed, heavier, mobile, positively
charged ions are usually a nuisance. A new
experiment, however, has made hydrogen ions into the primary carriers of information in a Si–SiO2–Si device. The protons, buried in the central layer of the
semiconductor sandwich, migrate between the interfaces with the outer silicon
layers. Judged as a storage device, this transistor did pretty well: It retained its
state (on or off) for up to 25 hours at 200 degrees C; it successfully underwent
10,000 write–erase cycles, showing that the protons are imprisoned between the
silicon walls; and it could be switched in 50 ms. The chief virtue of this nonvolatile memory device may prove to be its ease of construction. (K. Vanheusden
et al., Nature 386, 587, 1997; K. Vanheusden et al., J. Non-Cryst. Solids, in
press.)
A “PHOTON CONVEYOR BELT” has been developed at the University of
Munich, bringing about a new method for processing and storing light signals
on a chip. In the new device the storage is accomplished by first sending a
pulsed surface acoustic wave (SAW) along a piezoelectric semiconducting quantum-well structure. A laser pulse is then converted into a splash of excitons
(electron–hole pairs), which are caught in the electric fields of the leisurely
propagating SAW. In effect, photons become spatially separated electrons and
holes that surf along on different parts of the guiding acoustic wave, like on a
conveyor belt. Later, and in a different location on the device, the electron–
hole pairs are made to recombine, reproducing the photons, which are then
detected. The signal has essentially been converted from a speed-of-light wave
into a speed-of-sound wave, and back again. According to Achim Wixforth,
the strong lateral electric fields in a SAW can be exploited to combine the large
absorption of a direct-gap semiconductor with the long radiative lifetimes of
an indirect system. Typical excitons live for mere nanoseconds before recombining; in this experiment, they have survived for microseconds. (C. Rocke et
al., Phys. Rev. Lett. 78, 4099, 1997.)
Animation and further details at: http://www.aip.org/physnews/graphics/condensed/1997/conveyor/
HOLLOW-NANOPARTICLE LUBRICANTS have performed well in friction and durability tests, and may be superior to other solid lubricants, which
usually come in powdered form. The lubricant consists of balls of tungsten
disulfide only 100 nm across (much smaller than conventional 4 mm powder
grains) that are flexible, buckyball-like hollow cages. Their elasticity, chemical inertness, small size, low adhesion to substrates and tendency to roll rather
than slide when pushed, make for excellent lubricating properties. The Israeli
researchers who developed the lubricant foresee a bright industrial future for
such nano–ball bearings. (L. Rapoport et al., Nature 387, 791, 1997.)
A NEW MAGNETOELECTRONIC DEVICE can serve as a nonvolatile
memory element or logic gate. Researchers at the Naval Research Laboratory
in Washington, DC, use magnetic fringe fields from the edge of a ferromagnetic thin film to generate a Hall voltage in a semiconducting thin film. By
reversing the magnetization of the ferromagnet (which is then maintained even
without power), the sign of both the fringe field and the output voltage is
switched. The prototype is several micrometers across, but the output parameters should improve as the device shrinks—a minimum feature size of 100 nm
would correspond to 2 giga-bytes/cm 2. For logic gate operation, the next step
is to demonstrate that several of the structures can be coupled together. (M.
Johnson et al., Phys. Rev. Lett. 71, 974, 1997.)
NANOSCOPIC ELECTROMAGNETIC FIELDS can now be directly imaged in real time, using only an electron holographic microscope (EHM). Conventional electron holography is a two-step—and therefore not real-time—process, requiring taking a hologram (step 1), then reconstructing the image (step
2). A previous real-time method combined an EHM and an optical reconstruc-
8
tion system, but proved difficult to implement. Now, researchers in Japan have
passed three electron-wave beams, two of which are reference beams, through
the electromagnetic field of interest (typically emanating from a small object)
to record a “three-wave interference” pattern onto a film or CCD camera. The
fringes alone contain little information, but their intensity modulations reveal
lines of electric equipotential or magnetic flux. Applying their technique to a
0.5 ÿm electrically charged latex particle, the researchers deduced from the
imaged field that it was created by about 400 electrons. They also imaged the
weakening electric field of a charged insulating particle as its temperature was
increased, and the dipole-like magnetic field of a single-domain particle of
barium ferrite. (T. Hirayama et al., J. Appl. Phys. 82, 522, 1997.)
Technique for viewing a submicroscopic object’s electric
and magnetic fields in real
time. Shown is an electrically
charged latex particle (0.5
microns in diameter) and its
associated electric field.
(Courtesy AIP/T. Hirayama et
al.)
TRAPPING A SINGLE
NANOPARTICLE between
two electrodes can now be
done in a straightforward way, offering possibilities such as single-nanoparticle
switches. Researchers in The Netherlands constructed a circuit containing two
platinum electrodes separated by as little as 4 nm—a gap that they narrowed
from about 25 nm by sputtering the Pt. To trap molecules or clusters electrostatically, they then immersed the electrodes in a solution containing the
nanoparticles, and applied a voltage to the circuit. The electric field between
the electrodes polarized the particles and attracted them to the gap. When the
gap was bridged, current flowed through the circuit, and a resistor then sharply
reduced the electric field, discouraging additional nanoparticles from entering
the gap. A single particle is all that’s needed to span a small enough gap. In
principle, this electrostatic-trapping technique can work for any polarizable
nanoparticle; it has been demonstrated for 17 nm palladium clusters, micrometer-long car-bon nanotubes and a 5 nm long chain of thiophene (a conducting
polymer). The researchers have also studied the transport properties as single
electrons tunnel into a Pd nanocluster between the electrodes. (A. Bezryadin, C.
Dekker, G. Schmid, Appl. Phys. Lett. 71, 1273, 1997.)
A cluster of palladium atoms (yellow) is trapped between platinum
electrodes (red). The palladium
cluster is 17 nanometers in diameter,
and the electrodes are separated by
only nanometers. (Image copyright
DIMES institute, Delft University of
Technology, The Netherlands)
A PHOTOREFRACTIVE (PR)
POLYMER with a large optical gain
has been demonstrated by researchers at the University of California,
San Diego. In a PR material, an incident light beam redistributes electrons or
holes so as to produce a spatially varying index of refraction. Inorganic PR
crystals have been known for decades and can have an optical gain (characterizing the energy transferred from one beam to another) of 10 4–105, but they are
expensive and difficult to make. Until now, organic PR polymers at best have
just balanced optical gains against losses. The new material is multilayered;
each layer blends buckyballs (which offer holes), poly(n-vinyl carbazole) molecules (which carry the holes along their backbone) and PDCST molecules (which
form an asymmetric potential for electron motion and thereby make the refractive index electric field-dependent). The effect of the layers was cumulative.
The researchers demonstrated an optical gain of about 400%; this can open the
door for inexpensive and easily made devices, such as “optical transistors” that
can amplify or attenuate a light beam. Self-oscillation based on this PR gain
was also observed—for the first time in an organic material. (A. Grunnet-Jepsen,
C. L. Thompson, W. E. Moerner, Science 277, 549, 1997.)
particles, nuclei, plasma
QUARKS HAVE NO APPARENT STRUCTURE. Last year, based on an
analysis of proton–antiproton collisions, the Collider Detector at Fermilab collaboration reported an excess of events with high-energy jets shooting away
from the interaction at large angles. The measurement was interpreted by some
(although not by the experimentalists themselves) as possible evidence for
subquarks. (See Physics Today, March 1996.) To test the subquark idea, the
same group has now reported a study of the angle of emission of high-energy
jets. They find that quarks are pointlike at the 10–19 m level, that there is no
additional evidence for subquarks, and that the extra high-energy jets may be
more simply explained by extra gluons inside the proton. (F. Abe et al., Phys.
Rev. Lett 77, 5336, 1996.)
THE NAMES OF ELEMENTS 104-109 have finally been accepted by nuclear
scientists and certified by the International Union of Pure and Applied Chemistry. The delay over the names was caused partly by rival claims to priority;
the pertinent experiments rendered mere handfuls of atoms. Physics and chemistry students worldwide will now have to memorize the following additions to
the Periodic Table: Rutherfordium (abbreviated Rf, element 104), Dubnium
(Db, 105), Seaborgium (Sg, 106), Bohrium (Bh, 107), Hassium (Hs, 108), and
Meitnerium (Mt, 109). (The New York Times, 4 March 1997.)
PEEKING AT BARE ELECTRONS. Modern quantum theory holds that the
electromagnetic coupling constant αQED should increase with increasing momentum transfer Q 2, even as the strong coupling constant αstrong decreases.
At high enough energies, they should become equal. According to theory, an
electron is surrounded by a cloud of virtual photons, which themselves can
dissociate into virtual fermion–antifermion pairs within the limits of the uncertainty principle. The “bare” electron is thus screened by attracting the positively charged virtual particles and repelling the negative ones. A collision
with high Q2 penetrates closer to the bare electron, resulting in a higher coupling constant, as now shown by physicists studying e+ e– collisions at the
TRISTAN accelerator in Japan. At a center-of-mass energy of 57.77 GeV, well
below the mass of a Z0, they found 1/αQED = 128.6±1.6, in excellent agreement with the theoretical value of 129.6±0.1, and significantly different from
the canonical low-energy value of 137. (I. Levine et al., Phys. Rev. Lett. 78,
424, 1997.)
expected chemical properties, based on position in the periodic table, are expected for the heaviest elements due to strong relativistic effects. Elements 104
and 105 did indeed show some surprising properties. Now, using both isothermal gas and liquid chromatography on only seven atoms of element 106 (living
for mere seconds) GSI researchers have established that 106 behaves chemically much like tungsten and molybdenum, the elements lying directly above it
in the periodic table. Thus, for 106 at least, the periodic table is restored. (M.
Schädel et al, Nature 388, 55, 1997.)
THE SMALLEST NONZERO BRANCHING RATIO to date has been measured. Calculating and then measuring the relative likelihoods (branching
ratios) of the different routes of particle decay are important diagnostics for the
Standard Model (SM) of particle interactions. A particularly rare form of decay transforms the K+ meson into a π+ meson, a neutrino and an anti-neutrino—a flavor-changing neutral current process that could point toward nonSM physics. This particular decay route is quite sensitive to the coupling between top and down quarks. The large E787 Collaboration, working at
Brookhaven National Laboratory, has examined more than a trillion K + decays
and, after years of applying ever tighter constraints, has finally found one event
with the telltale signature. The event has an estimated background of 0.08
events. The resulting branching ratio for this particular K+ decay is 4.2+9.73.5 x 10 -10, consistent with SM expectations, which are centered at 1 x 10-10.
In the next year or so, additional data should help firm up the numbers. (S.
Adler et al., Phys. Rev. Lett. 79, 2204, 1997.)
A K+ meson decays into three end products: a π+ meson, a neutrino (denoted by
υe) and an anti-neutrino (denoted by a υe
with a horizontal bar on top). This particular decay path for the K + was recently observed for the first time. (Courtesy
Brookhaven)
In this artist’s rendering, a “bare” electron is denoted by the bright spot at the
center of the figure. The wispy white lines represent electric field lines radiating out from the electron. Virtual particle-antiparticle pairs are represented
by blue-gold ellipses; the blue side, corresponding to a positively-charged particle, is nearer to the electron. This polarization effect reduces the effective
charge of the electron that we observe at a large distance. (Courtesy of Purdue
University).
EXOTIC MESON AT BROOKHAVEN? Testing the notable proposition that
quarks always bind in groups of two (quark-antiquark objects called mesons)
or three (baryons), physicists at BNL send 18-GeV pi mesons into a hydrogen
target and then harvest events in which the emergent debris includes eta mesons and pi mesons. In particular, they seek to study the parent particle that
decayed into the eta and pi. Most of the time this is the humdrum a 2 meson,
with a mass of 1320 MeV. But about 3% of the time another particle appears to
be the parent, one with a mass of about 1370 MeV. Two things in combination
hint that something unusual is happening: first, the relative angle between the
outgoing eta’s an pi’s suggests a rivalry or “interference” between the a2 meson and the mysterious particle, strengthening the argument that the particle
exists and is not just a spurious blip in the data. Second, the new particle’s
quantum numbers (its spin, parity, and charge conjugation number) do not
conform to what one would expect for a conventional quark-antiquark meson.
Scientists in the Northwestern-Rensselaer-Massachusetts-Notre
Dame-BNL-Moscow State-IHEP collaboration speculate that the exotic particle might be either an unprecedented 4-quark state (two quarks and two antiquarks) or a quark-antiquark-gluon state; gluons are the particle-like carriers
of the strong nuclear force, but in this case the gluon in question would be an
independent particle on an even footing with the quarks. (D.R. Thompson et
al., Phys Rev Lett 79, 1630, 1997.)
PROTON PAIRS EJECTED FROM NUCLEI have now been studied in
enough detail to begin to probe short-range correlations within the nucleus.
At the National Institute for Subatomic Physics (NIKHEF) in Amsterdam, The
Netherlands, a beam of electrons was directed at oxygen nuclei. For just the
right collision energy, a single electron can knock two protons out of the nucleus.
The experimenters showed that the two protons (which are detected in coincidence with the scattered electron) emerge primarily in an S state, in which
their relative angular momentum is zero. Such protons would have been within
10–15 m of each other—and strongly correlated—in the nucleus just before
being struck by the electron. The NIKHEF experiment was the first to obtain
clean enough spectra to see the ejected S-state proton pairs, thereby leaving the
residual carbon nuclei in the ground, or a low-excited, state. (C. J. G.
Onderwater et al., Phys. Rev. Lett. 78, 4893, 1997.)
REAL PHOTONS CREATE MATTER. Einstein’s equation E = mc2 formulates the idea that matter can be converted into light and vice versa. The
vice-versa part, though, hasn’t been so easy to bring about in the lab. But now
a Stanford-Tennessee-Princeton-Rochester team of physicists at SLAC have
produced electron-positron pairs from the scattering of two “real” photons (as
opposed to the “virtual” photons that mediate the electromagnetic scattering of
charged particles). To begin, light from a terawatt laser is sent into SLAC’s
highly focused beam of 47-GeV electrons. Some of the laser photons are scattered backwards, and in so doing convert into high-energy gamma ray photons. Some of these, in turn, scatter from other laser photons, affording the
first ever creation of matter from light-on-light scattering of real photons in a
lab. (D.L. Burke et al., Phys Rev Lett 79, 1626, 1997.)
HEAVY-ELEMENT CHEMISTRY. Scientists at the Laboratory for Heavy
Ion Research (GSI) in Darmstadt, Germany, not only have made a number of
the heavy elements (up to element 112) in recent years, but have also performed some nimble chemical tests on the short-lived atoms. Deviations in the
RECORD HIGH LEVELS OF FUSION POWER AND ENERGY have been
observed at the Joint European Torus (JET) device in England. A peak power of
16 MW was reached and a power output of 10 MW was sustained for at least a
half second, pretty impressive achievements for a fusion experiment. The burning of the deuterium-tritium fuel inside the chamber, and the production of alpha
particles (which, if they can be contained, aid the heating), went according to
schedule, bolstering expectations that the proposed International Thermonuclear
Experimental Reactor (ITER) would work as planned. (Nature, 6 Nov.) Like all
fusion experiments to date, the recent JET demonstration did not generate as
much power as had been poured into the reactor to start the fusion process. Still,
the ratio of output power to input power was an impressive 0.65, more than
double the previous record.
More information at JET Web Site: http://www.jet.uk/
THE ANAPOLE MOMENT OF A NUCLEUS IS DETECTED. Parity
violation---the differentiation between left and right---was first observed (1957)
in transitions between nuclear states. Later certain transitions in atoms too
were seen to violate the conservation of parity. Now an experiment at the
University of Colorado not only makes the most accurate measurement of this
effect in cesium atoms but also observes, for the first time, the anapole moment
for a nucleus, the internal electromagnetic moment in the nucleus which comes
about because of the weak force. (C.S. Wood et al., Science, 21 March 1997.)
FIRST RESULTS FROM JEFFERSON LAB. This new nuclear physics facility in Newport News, Virginia explores the interface between the physics of
the nucleus (made of protons and neutrons) and the physics of individual protons and neutrons (made of quarks held together by particles known as gluons). The main machine at Jefferson Lab is the Continuous Electron Beam
Accelerator Facility (CEBAF), which accelerates continuous streams of electrons to energies of 4 GeV (with a maximum energy of 8 GeV planned for the
future); the electrons are then diverted to one of three experimental halls where
they collide with fixed targets containing nuclei. At the Spring 1997 APS/
AAPT meeting in Washington, DC, Rolf Ent of Jefferson Lab described how
electron collisions with nuclei are ejecting protons from nuclei at a greater rate
than anticipated by the present theories on the subject. Exploring how gamma
rays break up deuterons (containing a proton and neutron), Haiyan Gao of
Argonne presented measurements showing that the quark substructure inside
the deuteron must be taken into account to properly understand the breakup
process.
The Continuous Electron Beam Accelerator Facility (CEBAF) accelerates continuous streams of electrons to energies of 4 GeV (with a maximum energy of 8
GeV planned for the future); the electrons are then diverted to one of three
experimental halls where they collide with fixed targets containing nuclei. (Thomas Jefferson National Accelerator Facility.)
FEWER J/PSI PARTICLES ARE DETECTED when the energy of heavy-ion
collisions increases in the NA50 experiment at CERN, offering possible evidence for the creation of a quark-gluon plasma, the exotic state of nuclear
matter that last existed naturally when the universe was 10 millionths of a
second old. In today’s universe, all quarks come in bundles of two and threes
held together by other particles called gluons. In contrast, a quark-gluon plasma
is a hot soup of single quarks and gluons. Creating and studying the quark-gluon
plasma is the goal of experiments at CERN and the upcoming Relativistic Heavy
Ion Collider (RHIC) at Brookhaven. In recent experiments at CERN involving
high-energy collisions between lead nuclei and fixed targets, researchers noted
that the number of detected J/Psi particles (objects consisting of a charm and
9
anticharm quark) diminished as the energy of the collisions increased. One possible explanation is that collisions briefly created a quark-gluon plasma, which
dissembled the J/psis into single quarks which became too far separated to recombine when the quark-gluon plasma quickly cooled. However, there are less
dramatic explanations for the observed experimental results--in the “co-moving
hypothesis,” one of the leading alternative explanations, particles called pions
created from the high-energy collisions strike the J/psis, which then disappear in
a spray of particles. (S. Gavin et al., Phys. Rev. Lett. 78, 1006, 1997; H. Sorge et
al., Phys. Rev. Lett., 79, 2775, 1997; J.-P. Blaizot et al., Phys. Rev. Lett., 77,
1703, 1996; also see CERN Courier, November 1997.)
CERN Experiment NA50 Home Page: http://www.cern.ch/NA50/RHIC
Web Site: http://www.rhic.bnl.gov/
other physics topics
SONOLUMINESCING BUBBLES collapse at more than Mach 4, new experiments have shown. Previous experiments could establish only that the bubble
collapsed faster than the speed of sound in the gas within the bubble.
Sonoluminescence (SL) is the still-mysterious process in which sound waves
cause bubbles in a water tank to collapse and generate ultrashort light flashes,
which represent a trillionfold concentration of the original sound energy. (See
PHYSICS TODAY, September 1994; December 1995; November 1996.) Now,
UCLA researchers have determined the speed of bubble collapse—about 1500
m/s—by measuring the amount of light scattered from ultrashort (100 fs) laser
pulses impinging on a bubble at different times during its implosion. The amount
of scattered light is proportional to the square of the bubble radius. The work
provides the best evidence yet for the shock-wave model—the leading explanation for SL. But the experiment does not rule out competing explanations because all experiments to date have probed only the outer surface of the bubble,
not what is happening inside. The UCLA team also found that the SL flash
occurs within 300 ps of when the bubble reaches its minimum radius of less than
1 ÿm; at that time the bubble’s surface is accelerating by at least 1011 g, yet
amazingly remains intact. (K. R. Weninger, B. P. Barber, S. J. Putterman, Phys.
Rev. Lett. 78, 1799, 1997.)
ONE HUNDRED YEARS OF ELECTRONS. On April 30, 1897, at a meeting of the Royal Institution in London, physicist Joseph John (J.J.) Thomson
declared that cathode rays lighting up a fluorescent screen were made of negatively charged particles. Thomson boldly proclaimed that these particles—which
we now know as electrons—could be found in all atoms. The term “electron” as
it applied to electricity actually came about in 1891 to describe the unit of electric charge in a chemical reaction. The electron was the first known subatomic
corpsucle and its discovery marks the advent of particle physics. Michael Riordan
(editor of SLAC Beamline, whose Spring 1997 issue is devoted to the electron
centennial) refers to the electron as a truly “industrial strength” particle, since it
is the workhorse of electronics, including television, telephones, and personal
computers. (Many of these devices organize electrons inside transistors which
were themselves developed exactly half a century ago.) Labor saving devices
aside, electrons are of course the outer constitutents of all atoms and the principal currency of exchange in all chemical reactions.
AIP web site on The Discovery of the Electron: http://www.aip.org/history/electron/
J.J. Thompson in his office. (Courtesy
AIP Center for History of Physics)
NONEXPONENTIAL DECAY of a
quantum system has been observed for
the first time. Unstable systems are
known to decay exponentially in all
fields of science, but quantum mechanics predicts deviations from this law at
both very short and very long times in a
system’s evolution. Physicists at the
University of Texas at Austin have now
devised a scheme to study the tunneling rate of trapped sodium atoms at very
early times. They found a decay rate that was initially flat, followed by one that
was steeper than the final decay rate. After 10–15 ms, the usual exponential
decay took over. The researchers think that tunneling can be suppressed during
the early times. (S. R. Wilkinson et al., Nature 387, 575, 1997.)
10
50 TH ANNIVERSARY OF THE TRANSISTOR. This simple three-terminal
electronic device can act as amplifier or switch by allowing a tiny electrical
(gate) signal to control a much bigger current. (For the history of the transistor,
see Physics Today, December 1997 and the book “Crystal Fire: the Birth of the
Information Age,” by Michael Riordan and Lillian Hoddeson, W.W. Norton &
Company, 1997.) Examples of ongoing research include the development of allpolymer transistors; spin transistors, in which the spin as well as the charge of
electrons is important (Physics Today, July 1995); room-temperature, singleelectron transistors; silicon-carbide transistors for high-temperature applications;
10-nm metal transistors; the development of molecular-scale transistors (New
Scientist, 2 August 1997); and neuron transistors, in which gate signals are supplied by ions from leech neurons (M. Jenkner et al., Phys. Rev. Lett. 79, 4705,
1997.)
GAMMA RAYS FROM A FREE-ELECTRON LASER. Physicists at Duke
University used ultraviolet photons, Compton backscattered from 500 MeV electrons inside a storage-ring free-electron laser (FEL), to produce a beam of 12.2
MeV gamma rays. The emittance and divergence of the electron beam were so
low that by collimating the gamma beam, a nearly monoenergetic beam of gammas (with an energy spread of about 1%) resulted. The intensity of the gamma
beam is expected to be 1000 times greater than that produced with non-FEL
systems. A beam like this will be useful for high-resolution nuclear gamma-ray
spectroscopy and cancer therapy. It can also be used for high-precision gammaray transmission radiography. Other facilities that use Compton backscattering
in FELs to produce gammas are UVSOR and NIJI-IV in Japan and Super-ACO
in France. (V. N. Litvinenko et al., Phys. Rev. Lett. 78 , 4569, 1997.)
A SONOLUMINESCING BUBBLE OF AIR may change to argon. In
sonoluminescence, an air bubble repeatedly collapses, producing ultrashort
flashes of light. Recently, a German–US team proposed that such a collapsing
air bubble, with its imploding shock waves, becomes hot enough to dissociate
the air’s nitrogen and oxygen molecules, which then react with hydrogen and
oxygen radicals from dissociated water vapor. The compounds thus formed are
water soluble and, after a few collapse cycles, only inert gases (mainly argon)
are left to produce the light. At the Acoustical Society of America meeting in
June, William Moss of Lawrence Livermore National Laboratory reported simulations from a rigorous fluid dynamics code (originally developed for imploding
fusion pellets) that includes the physics of partially ionized plasmas. Moss and
his colleagues simulated sonoluminescing bubbles of both pure argon and pure
nitrogen. Their argon spectrum closely matched the experimental spectrum from
a collapsing air bubble, but their nitrogen spectrum did not, lending strong support to the notion of shock-driven chemistry being important. (D. Lohse et al.,
Phys. Rev. Lett. 78, 1359, 1997. W. C. Moss, D. B. Clarke, D. A. Young, Science 276, 1398, 1997.)
Lay language paper on this topic: http://www.acoustics.org/133rd/2apa10.html
The final 50 picoseconds (ps) of the calculated collapse of an argon bubble. The
bubble radius (outermost curve), shock
wave location (inner curve), and emitting
regions [optically thin (shaded), and optically thick (solid)] are shown. P is the
relative emitted power (energy per unit
time) of light in the visible part of the spectrum. (Courtesy William Moss, Livermore.)
THE 1997 NOBEL PRIZE FOR PHYSICS has been won by Steven Chu of
Stanford, Claude Cohen-Tannoudji of the Ecole Normale Superieure in France,
and William Phillips of NIST for their development of laser cooling for neutral
atoms. In this case “cooling” means reducing the relative velocities of atoms.
In these experiments, an array of laser beams converges on a gas of atoms. In the
simplest type of laser cooling, the wavelength of the light is tuned so that just
the fastest atoms moving in a particular direction will absorb a photon head-on,
thus slowing their motion in that direction. The atoms will eventually re-emit a
photon but in random directions. The effect of the laser bombardment is a net
slowing of the atoms. This “optical molasses” can slow millions of atoms to
temperatures just millionths of a degree above absolute zero. Adding magnetic
fields to the laser configuration enables one to trap the atoms and cool them
further. As a result of these techniques, physicists can cool atoms closer to
absolute zero than ever before, to temperatures of nanokelvins in some cases.
Reducing the distracting presence of thermal motion permits the study of atomic
properties with much greater precision. Furthermore, laser cooling serves as the
first stage in reaching the exotic condition known as Bose-Einstein condensation, the new state of matter in which many atoms begin to “overlap,” eventually
assuming a single common quantum state. With his laser setup, Phillips can
create “optical lattices,” crystal-like arrays of atoms held in place by light waves.
Chu has used his laser array to split ultracold atoms into separate waves and
recombine them to form interference patterns that can provide detailed information on the atoms. In a particularly sophisticated form of laser cooling,
Cohen-Tannoudji has put helium atoms into a “dark state,” whereby the coldest
atoms become unable to absorb additional light and fall to temperatures even
lower than previously imagined possible. What else can be done with the chilled
atoms? They may become the basis for extremely precise atomic clocks, accelerometers, and gyroscopes. (Background articles: Scientific American, March
1987, trapping atoms, William Phillips; July 1993, accurate time measurements,
Norman Ramsey; February 1992, Steven Chu, trapping neutral particles; Physics Today, October 1990, Phillips and Cohen-Tannoudji, laser cooling.)
Official 1997 Nobel Prize in Physics site: http://www.nobel.se/laureates/physics-1997.html
(Left to right) Steven Chu, Claude Cohen-Tannoudji, William D. Phillips are
the winners of the 1997 Nobel Prize in Physics. (Copyright Nobel Foundation;
Used by permission)
AT LAST YEAR’S PHYSICS OLYMPIAD, held in Sudbury, Ontario July
13-21, 1997, students from 56 nations took challenging experimental and theoretical exams. The Russian team received 4 gold medals, China earned 3, and
Australia garnered 2. Everyone on the US team brought home a medal; Boris
Zbarsky of Rockville, Maryland won a gold medal.
American Association of Physics Teachers Web Page on 1997 Physics Olympiad: http://www.aapt.org/programs/97res.html
USA Team members for 1997 Physics Olympiad
(from left to right) Mary Mogge(coach), Boris
Zbarsky,
Noah
Bray-Ali,Dwight
Neuenschwander(coach). Chris Hirata, Travis
Hime, and Michael Levin. (Courtesy American
Association of Physics Teachers).
MOST INTENSE MANMADE SOUND.
The production of sound waves with 1600 times more energy per unit volume
than previously achieved has been announced by researchers at a December 1997
meeting of the Acoustical Society of America in San Diego, opening up possible
new uses for sound in science and technology. Sound waves, patterns of compression and expansion in a gas such as air, are often created and studied in
closed or semi-closed containers called cavities. In the past, attempts to make
such sound waves louder (by adding more sound energy into the cavity) would
fail beyond a certain point because additional energy would merely lead to the
formation of a shock wave which would quickly dissipate the energy as heat.
Until the late 1980s, researchers thought shock-wave formation was inevitable.
In a new technique called “resonant macrosonic synthesis,” Tim Lucas and colleagues at MacroSonix Corporation in Virginia have built cavities with special
shapes (horns, bulbs, cones) each tailored to promote certain distinct modes of
sound vibration which combine in such a way as to inhibit the creation of shock
waves, allowing sound waves of unprecedented energy density to build up. Filling the containers with gas, and vibrating them to generate sound waves inside,
the researchers produced sound waves with oscillating pressures up to 500 pounds
per square inch. The first technological application for these powerful sound
waves will be in an “acoustic compressor”which uses sound rather than moving
parts to compress gas inside refrigerators and air conditioners.
Using a specially designed resonator
such as the one shown in this photo, researchers have created sound waves with
1,600 times more energy per unit volume
than any previous human-made sound wave.
(Courtesy MacroSonix Corporation, Virginia.)
11