[time-nuts] Shrinking Atomic Clocks

Thomas A. Frank ka2cdk at cox.net
Sun Mar 15 17:46:54 EDT 2009

Time to shrink the atomic clock

14 March 2009 by Anil Ananthaswamy
Magazine issue 2699. Subscribe and get 4 free issues.
ATOMIC clocks, currently the size of fridges, could shrink to the  
microscale thanks to a new way of measuring the second. The technique  
could also see aluminium displace caesium as the standard of time.

The world's most accurate atomic clocks are at the National Institute  
of Standards and Technology (NIST) at Boulder, Colorado. Known as  
fountain clocks, they send clouds of caesium atoms through a vacuum  
chamber in a magnetic field. Large atoms like caesium and aluminium  
have multiple energy levels that are so close together they appear  
indistinguishable. The magnetic field separates these levels into two  
"hyperfine" states.

The chamber is also filled with microwaves, which excite the atoms.  
They then emit light as they drop to the lower hyperfine state. The  
microwave frequency that maximises this fluorescence is used to  
define the length of a second, currently the time it takes for  
9,192,631,770 cycles of microwave radiation.

All this takes place in a large vacuum chamber and so fountain clocks  
are big devices, about a cubic metre in size. That makes it hard to  
keep the magnetic field and the device's temperature uniform over the  
whole area, which can lead to errors of measurement.

That's why Andrei Derevianko and Kyle Beloy of the University of  
Nevada in Reno and colleagues have come up with the idea of trapping  
the atoms in place using lasers. This means their energy states could  
be monitored in an area only a few micrometres across, potentially  
leading to more accurate measurements. This is difficult to get  
right, though, because the lasers distort an atom's energy levels in  
a complex way, making it impossible to define a jump that equates to  
a second.

Derevianko's team overcome this problem by finding a laser frequency  
that alters both hyperfine states by exactly the same amount - a  
trick that works in aluminium and gallium but not as well in caesium  
(www.arxiv.org/abs/0808.2821). "Then, the energy difference between  
the levels is the same as if the atoms are in vacuum," says Derevianko.

Using this method, the team has calculated the second to be 1506  
million cycles of microwaves for aluminium-27 and 2678 million cycles  
for gallium-69.

Although the atoms can be trapped in an area only a few micrometres  
across, the lasers, and cooling and computing equipment will add to  
the bulk. Nevertheless, the team say the clocks may be portable and  
could be used in space-based experiments that require extremely  
accurate timekeeping, such as those for detecting gravitational waves  
or for testing Einstein's theories.

Tom Heavner, who works on fountain clocks at NIST, describes the  
proposal as forward-thinking and original. "It is a really clever way  
to meld together the old-style clocks with new laser technology," he  


More information about the time-nuts mailing list