Still, more work remains before sci-
entists can be certain of the role that
Orai 1 plays. It is possible, for example,
that the newly discovered protein is not
a component of the channel itself but
w w w. s c i a m . c o m
S C I E N T I F I C A M E R I C A N 23
COPYRIGHT 2006 SCIENTIFIC AMERICAN, INC.
news
S
ICS
SCAN
Y
PHI I A Hint of Axions
AN EXPERIMENT MAY HAVE SEEN AN ELUSIVE NEW PARTICLE BY GRAHAM P. COLLINS
Named after a laundry detergent and the mass of the putative axions and how
originally proposed to clean up a strongly they interact. Puzzlingly, however,
problem with particle physics, axions
the results contradict other observations
are curious critters. Axions produced dur-
and do not fit with constraints deduced
ing the big bang could be lurking all around
from astrophysics. In particular, the CERN
us, contributing to the dark matter that con- Axion Solar Telescope (CAST) ran for six
stitutes 22 percent of the universe. Other months in 2003 and failed to detect any ax-
axions, freshly formed inside the sun, could
ions arriving from the sun. That result would
be streaming through us. And according to
seem to rule out a large swath of possible
a paper published in March, laboratory-
masses and interaction strengths, including
made axions might have been detected for the values seen by PVLAS. Furthermore, if
the first time by an experiment in Italy axions interact as strongly as PVLAS indi-
known as PVLAS ( p olarization of the v ac-
cates, they should be produced in large
uum with a las er).
quantities in stars, causing stars to grow old
much faster than they are known to.
Wall
Such considerations “put the bar pretty
Bright
high before one can accept the PVLAS re-
Very dim
spot of
spot of
sults,” says axion expert Pierre Sikivie of
Laser beam
light
light
the University of Florida and CERN. On
the other hand, he adds, “these people are
Laser
Axions
Magnetic field
Photons
very competent, and they have worked on it
created
created
a long time.” By all accounts, the PVLAS
researchers have been careful to exclude ef-
fects that could be confounding the data;
LIGHT BE AM experiment that would confi rm the existence of axions passes a laser beam
moreover, in work that is not yet published,
through a strong magnetic fi eld, converting some photons to axions ( green beam). The
axions penetrate a wall before passing through another magnetic fi eld that converts
the group has also obtained consistent re-
some of the particles back to photons, which form an extremely faint spot on the far wall.
sults with a different laser. Some theorists
have already proposed ways to reconcile the
Axions are posited to have exceedingly PVLAS results with those of CAST and oth-
low mass—less than a millionth that of an
er astrophysical limits.
SAVING A
electron—and are electrically neutral. They
Only further experiments will determine
SYMMETRY
interact only very weakly with other parti-
the truth. If the PVLAS results are correct,
Physicists originally proposed the
cles, making detection diffi cult. But physi-
then axions should appear in an experiment
axion as part of a scheme to
cists predict that a tiny fraction of any pho-
known as “shining a light through a wall.”
explain why the strong nuclear
force preserves so-called CP
tons passing through a magnetic fi eld will The idea is this: A laser beam is sent through
symmetry, which relates the
change into axions. (That is how the sun is
a strong magnetic fi eld at an opaque wall.
properties of particles and
predicted to produce them.) Indeed, the Ital-
Some of the photons in the beam are con-
antiparticles. Calculations using
ian experiment, based at the National Labo-
verted to axions, which pass through the
the Standard Model of particle
ratories of Legnaro and led by Emilio Zavat-
wall. On the other side, another magnetic
physics showed that the strong
force could preserve CP only if a
tini and Giovanni Cantatore of the INFN fi eld induces a small fraction of the axions to
certain parameter in the theory
Trieste, saw evidence for axions in the behav-
convert back to photons, which can be de-