Download Molecules in Slow Motion - Max-Planck

FIRSTHAND KNOWLEDGE
MOLECULAR PHYSICS
Molecules in
Slow Motion
In the microworld, slow means cold. Research on decelerated, ultracold atoms has recently
yielded spectacular results – such as Bose-Einstein condensates, which have already become
indispensable for quantum physics. Cold molecules should also have similar potential,
but unfortunately, they are very difficult to produce. The group working with GERARD MEIJER,
Director at the FRITZ HABER INSTITUTE
OF THE
MAX PLANCK SOCIETY
in Berlin, has
High-tech with sophisticated design: The Stark decelerator in
Berlin (opposite page) slows molecules down. Afterwards, they reach
the trap that Bas van de Meerakker is adjusting (above).
P
hysicists seem to have a penchant for operations at full
throttle: they use huge particle accelerators to spin components of all
kinds of matter to the fastest speeds
possible to make them collide. The
subatomic traces of such collisions
have actually been providing valuable information about the building
blocks of our cosmos for decades.
50
MA
X
P
L A N C K
R
E S E A R C H
But many physicists are also fascinated by extreme slowness. At the
Fritz Haber Institute of the Max
Planck Society, for example, we are
working on techniques for decelerating molecules efficiently.
The molecules we use come from a
jet of highly diluted gas. In a “normal”
warm gas, most molecules zip around
so quickly that only a supersonic jet
1/2007
could keep up. Our device allows us
to slow them down and actually get a
proper grip on them. Technologically
speaking, this is no mean feat. But the
effort is worth it, because we can conduct some fascinating basic research
on the slow molecules.
To slow them down, we use methods developed in accelerator physics.
We apply them with reverse effect, so
P HOTOS : N ORBERT M ICHALKE
developed an efficient system for slowing down molecules, and reports on it in this article.
we transform an accelerator into a
trap. The slow molecules can be
stored and studied very closely, looking, for instance, at the quantum
properties of their complex electron
shells. Or we use them to conduct
further experiments, such as slowmotion soft collisions.
In physics, as soon as small particles are involved, slowness is noth-
ing other than cold. High temperatures occur when the atoms or molecules in gases or liquids move quickly – or oscillate back and forth rapidly in solid matter. For example, at
room temperature, nitrogen molecules whiz through the air at an average speed of 500 meters per second,
and are thus faster than sound, which
travels 340 meters per second. So
heat is kinetic energy. If we take it
out of these molecules by “slamming
on the brakes,” then we automatically cool it down.
In the case of atoms in dilute gases,
physicists have since become very
adept at such braking techniques. In
the lab, they routinely achieve extremely low temperatures that would
have still been pure science fiction
20 years ago. They are just a few millionths of a degree above absolute
zero, which is minus 273.15 degrees
Celsius. The consequences of this
progress for science are very exciting, and sometimes completely unforeseen.
To create ultracold atom clouds,
the experimenters combine the brake
effect that laser light can exert on atoms with magnetic fields that confine the cold atoms. Such magnetooptical traps have been used successfully for a good 10 years to produce,
for example, the famous Bose-Einstein condensates. A Bose-Einstein
condensate is a collective quantum
state in which a cloud is “condensed”
from certain types of atoms when
they are cooled to extremely low
temperatures. What fascinates physicists about this trapped quantum
collective is the property that it is
completely freely accessible and manipulable from practically all sides.
GOLD RUSH
IN QUANTUMLAND
For this reason, it is much easier to
research than other multi-particle
quantum states: supraconductivity,
for example, hides in the interior of
metals, which is very difficult to access. That is why the first successful
1/2007 MA
X
PL
A N C K
R
E S E A R C H
51
MOLECULAR PHYSICS
PHYSICAL BRAKEBOXING
ONLY WORKS WITH ATOMS
First, the molecule hits the
electric field of a braking
element (top). Then it must
climb a potential hill and
slows down until it reaches
the peak (center). Going downhill would cause it to accelerate
again, but that is prevented
by turning off the electric
field in due time (bottom).
The trap arrests the decelerated molecule (gray cloud)
with its electric field like a
pillow (left). Now the molecule
is confined, with the field
acting like a funnel (right).
52
MA
X
P
L A N C K
R
E S E A R C H
So it is not surprising that physicists
no longer want to limit this exciting
game to atoms. Ultracold molecules
may promise even more thrilling
discoveries. But it is much more difficult to cool molecules down to extremely low temperatures than it is
with atoms. Unfortunately, the laser
cooling that has proven so effective
with atoms usually fails. To put it
simply, the atoms, with their electron shells, swallow the light quanta
(photons) from the approaching laser beam. Later, they emit these photons again, usually in any direction
at random. On the whole, the laser
beam slows down the atom like a
boxer deters a charging opponent
with targeted punches, while the latter – the atom – merely flails about
with no effect.
Unfortunately, this physical brakeboxing doesn’t usually work with molecules. They are made up of multiple
atoms and have far more complex
electron shells. Inside these, they convert the pulses of the impinging laser
photons differently than atoms do, and
they hardly slow down at all.
So we had to take a different route
to develop an effective molecule
brake. At present, only a few techniques are suitable for this. In Berlin,
we use a small particle accelerator
that we operate in reverse as a particle decelerator. We are now able to
use it to cool molecules down to a
few thousandths of a degree above
absolute zero. Our device is called the
“Stark decelerator,” in memory of Johannes Stark. The German Nobel
laureate in physics in 1919 discovered the effect that our anti-accelerator exploits to elegantly solve a further problem: accelerators work only
on electrically charged particles. This
1/2007
also applies, in principle, to the decelerators derived from them. But we
want to slow down electrically neutral molecules because they are of
greater interest to us. And this is
where the Stark effect helps.
Normally, physicists feed their accelerators with atoms, whose incomplete electron configuration creates
an electric charge, or with electrically non-neutral elementary particles.
Due to their charge, they sense the
accelerator’s electric field. They surf
along on this field as if on a wave,
continuously gaining speed in the
vacuum tube – or losing it in the decelerator. However, the Stark effect
allows us to also draw in neutral
molecules with strong electric fields.
This is because, in many molecules,
the spatial distribution of the electric
charges is somewhat irregular. These
“polar” molecules behave like tiny
antennas that react to the electric
field of the Stark decelerator.
A RIDE ON THE
NANO-CANNONBALL
However, their interaction with the
electric field is much weaker than in
the case of electrically charged particles. Depending on the molecule, it
can be just one ten-billionth of the
normal value. Nevertheless, with our
Stark decelerator, we are able to get
the molecules completely under control. The actual molecule brake here
is only slightly more than one meter
long. So compared with the usually
enormous
particle
accelerators,
which, however, work at a much
higher energy level, our experiment
is very manageable. We developed it
at the FOM Institute for Plasma Physics Rijnhuizen in Nieuwegein, The
Netherlands. Since late 2003, we
have continued this work at the Fritz
Haber Institute, which offers ideal research conditions.
To better understand how the Stark
decelerator works, let’s conduct a little thought experiment. We pick out
a molecule from the gas stream and
sit on it, much like Baron Münchhausen on his cannonball. We now jet
around on our nano-projectile at a
D IAGRAMS : F RITZ H ABER I NSTITUTE
creation of Bose-Einstein condensates in 1995 practically triggered a
“gold rush.” Pioneers Eric Cornell,
Carl Wieman and Wolfgang Ketterle
were even awarded the Physics Nobel
Prize for it in 2001 (MaxPlanckResearch 1/2002, page 62 ff.).
few hundred meters per second toward the entrance of the Stark decelerator. We see it from the front like a
kind of metallic hash mark into
whose central opening we shoot. In
doing so, an increasing force gently
brakes our molecular cannonball.
Inside the hash mark, we see that
its metal rods are arranged one behind another. At that moment, the
braking action suddenly ceases. Our
cannonball speedometer tells us that
we have lost approximately 1 percent
of our speed. The next metal cross is
racing toward us, where the process
repeats itself. In the end, after a good
one hundred such crosses, we are
nearly standing still.
To catch the molecule at just the
right moment when flying through
the next deceleration element, the
controls of the Stark decelerator send
short pulses of high-voltage electricity through all of the elements in succession. The pulses jump along this
cascade like a running light that
grows continuously slower. Now, if a
polar molecule hits a brake element at
the right moment, then the briefly increasing electric field acts like a small
hill of pure energy. The molecule becomes slower – as if it had to roll up
a virtual hill to reach the “energy
peak” at the center of the element.
However, when flying out, it would
roll down the other side of the virtual
hill. It would be just as fast at the end
as it was at the beginning, and nothing would be gained. To prevent this,
the field shuts down as soon as the
molecule reaches the energy peak.
The hill disappears right out from under the molecule’s feet, as it were, and
it can’t accelerate again.
Now the question remains how we
get a molecule to accurately hit a
pulse applied at the entrance of the
Stark decelerator. Very easily: our
gas beam is likewise pulsed. It consists of molecule packets whose arrival we coordinate with the decelerator’s voltage pulses. Consequently, at the end of a braking cycle, we
end up with not just one molecule,
but an entire group of molecules at
the desired speed. We can continu-
P HOTO : N ORBERT M ICHALKE
FIRSTHAND KNOWLEDGE
Molecular physicists in action: While Steven Hoekstra, Bas van de Meerakker (right
front) and Joop Gilijamse (center) work on the experiment, Ludwig Scharfenberg (left)
is busy building the two new, intersecting Stark decelerators.
ously adjust this final speed between
about 700 and just a few meters per
second.
The lowest speed currently attainable corresponds to a temperature of
about ten thousandths of a degree
above absolute zero. Our technique
has helped us successfully decelerate
various types of molecules: carbon
monoxide (CO), diatomic hydroxyl
(OH) and ND3, which consists of nitrogen (N) and three heavy hydrogen
atoms (deuterium, D).
TRAPPED IN THE
ENERGY FUNNEL
To experiment further with the slow
molecules, our lab has various apparatuses that we can connect behind
the Stark decelerator. One is a small
storage ring – another device borrowed from particle physics. We can
park the prepared molecules in an orbit in this ring before we use them in
further experiments. Unlike in particle physics, our storage ring is not
kilometers in size, but a mere 80 centimeters around.
Unfortunately, the molecules in a
packet never have exactly the same
speed. Thus, with every revolution
around the ring, the packet drifts further and further apart. To increase
the “parking time,” we built into the
ring a segment that uses specially
formed fields to push the molecule
packet back together with each revolution. This works so well that we
can now also store multiple packets
in succession in the ring. However,
the molecules still require a certain,
not-too-low residual speed to remain
in orbit. For ND3, for instance, this is
around 100 meters per second.
So if we want to store very slow,
cold molecules, we need another
technique. To this end, we use a special molecular trap. Using an electric
field, it captures the decelerated molecule as if in a pillow. As soon as the
molecule is properly confined, the
trap switches its field to take on the
form of an energy funnel. The molecule is then trapped in this funnel
and can only wobble back and forth
at a few meters per second. When
this happens, its temperature can
drop to about one fifty-thousandth
of a degree above absolute zero.
To date, we have caught ND3 and
the OH radical with the trap. The OH
is of particular interest to us: in principle, it is a water molecule (H2O)
that has lost a hydrogen atom (H).
This makes it highly chemically reactive. OH plays an important role in
1/2007 MA
X
PL
A N C K
R
E S E A R C H
53
54
MA
X
P
L A N C K
R
E S E A R C H
DR. BAS VAN DE
MEERAKKER, 29,
is head of a research group at
the Fritz Haber
Institute of the
Max Planck
Society in Berlin.
He researches
cold free radicals and experiments
with Stark decelerated molecular
beams. He completed his doctoral
studies at the University of Nijmegen in The Netherlands and was a
visiting scientist at Sandia National
Laboratories in Livermore, USA.
1/2007
P HOTO : P RIVATE
PROF. GERARD MEIJER, 44,
is one of the pioneers of
the Stark deceleration
technique. He became a
professor in Nijmegen in
1995 and Director of the
FOM Institute for Plasma
Physics Rijnhuizen in
Nieuwegein in 2000.
There, he launched the research program
on cold molecules and conducted physical
experiments on molecules in the gas phase
with the FELIX free electron laser. He has
been Director of the Department of Molecular Physics at the Fritz Haber Institute
of the Max Planck Society since 2003.
P HOTO : P RIVATE
P HOTO : P RIVATE
only high-energy molecular collisions had
previously been accessible for their experiments. Such brutal collisions conceal more
subtle processes that
occur while the molecules are approaching
one another.
With our decelerated
molecules, we could
now play with the finesse of a master pool
player. During our tests,
the xenon atoms always
maintained the same
speed. At the same time,
Steven Hoekstra makes some adjustments to the first part of
we used the Stark decelthe vacuum machine that produces the beam of OH molecules.
erator to vary the speed
of the OH molecules between 30 and 700 meters
per
second.
We then observed
nature, for example in cosmic gas
that, at certain threshold values, the
clouds and in the chemistry of the
OH molecules reacted differently to
Earth’s atmosphere, as well as in
the collisions with the xenon atoms:
combustion processes.
they no longer bounced off elastically
OH also recently helped us conduct
like billiard balls, but rather absorbed
another fundamental experiment
a portion of the impact energy.
(J. Gilijamse et al., Science 2006, vol.
313, 1617). For this, we intersected
MOLECULAR BILLIARDS
the path of the decelerated OH molWITH GREAT FINESSE
ecules after they left the Stark decelerator with a beam of xenon atoms.
That is what theoretical chemist
The OH molecules look roughly like
Gerrit Groenenboom from Nijmedumbbells with a list. They can theregen had predicted. According to his
fore rotate at various speeds. This roquantum theoretical calculations,
tation speed can change only in inat these threshold values, the total
crements, as it is quantized according
energy of the collision is just
to the laws of quantum physics. Alenough for the striking xenon to
though this is old hat for physicists,
stimulate the next higher energy
quantum of the OH rotation. We
obtained excellent confirmation of
Groenenboom’s predictions. We
clearly understand the physics of
such impacts quite well.
However, in this collision experiment, one of the partners in the impact was still an atom, and a hot
one at that. In the future, we want
to have two molecules that are both
truly cold collide gently. To do this,
we must also send the intersecting
molecular beam through a second
Stark decelerator. We hope that
these slow-motion molecular collisions will give us finely detailed information about the quantum structure of their electron shells. From
this we hope to learn, for example,
how stable molecular complexes,
which play an important role in nature, are created.
With two Stark decelerators, we
can also study how chemical reactions between the cold molecules
proceed. We know that their low energy will then just barely suffice for a
reaction. Physical chemistry predicts
that, in this case, quantum effects
should dominate the behavior of the
reaction partners. But no one can yet
predict precisely what will happen.
Perhaps there will be a reward for the
long years of development work –
one that is as spectacular as the discoveries made with ultracold atomic
gases. But we do not yet know where
the discovery of molecular slowness
will take us. We are just setting off
into this unknown world.
●
ROLAND WENGEN45, is a
physicist, science
journalist and
editor of the
magazine PHYSIK
IN UNSERER ZEIT.
For him, research
is an adventure
and the most important driver of
modern culture. He believes that
as many people as possible should
be able to participate in it through
popular-scientific texts. To this
end, the illustrator in him also likes
to resort to pen and paper.
MAYR,
P HOTOS : N ORBERT M ICHALKE
FIRSTHAND KNOWLEDGE