CORE CONCEPTS

Amazingly precise optical atomic clocks are morethan timekeepersAdam Mann, Science Writer

In March, physicists unveiled the world’s most accu-rate timepiece. But this was no pocket watch. At itsheart is an atomic pendulum that swings 500 trilliontimes per second and can measure fractions of a sec-ond out to 19 decimal places (1). The clock is so pre-cise that it loses at most a tenth of a second over theentire lifetime of the universe.

And yet, the device, known as an optical atomicclock, will allow physicists to do a lot more than telltime. Such clocks can help investigate fundamentalconstants of the universe with higher precision thanever before, searching for discrepancies in our currenttheories of reality. And because time and space areintimately related, these ultra-accurate clocks can alsoact like measuring tape, mapping the size and shape ofEarth with a resolution of a centimeter or less. This, inturn, can help climate researchers monitor sea level riseand geologists track the movement of tectonic plates.

It may seem startling that precision clocks havesuch ramifications. But any time science gets a newtool to study nature, it throws up surprises, saysphysicist Jun Ye of JILA, a research institute jointlyoperated by the University of Colorado Boulder andthe National Institute of Standards and Technology(NIST). Ye’s team constructed the record-holdingclock announced in March. He thinks major new in-sights are sure to come from the high-precision mea-surement of time. “I think you could argue that it’skind of at the core of everything,” he says.

Time is TellingAlbert Einstein famously defined time as “that thingyou measure using a clock.” Clocks need a movingpart, like a pendulum, to mark the passage of time. In1955, physicists Louis Essen and Jack Parry developedthe first atomic clock at theNational Physical Laboratoryin London by using ions of cesium-133 as tiny sub-atomic pendulums (2). These atoms emit radiation asthey oscillate between two closely spaced energy lev-els. The work led to the modern definition of a secondas the time it takes to complete 9,192,631,770 suchoscillations of the cesium-133 atom.

The cesium-133 radiation is in the microwave portionof the electromagnetic spectrum. But other atoms—such as aluminum, strontium, and ytterbium—have natural

oscillations at optical frequencies, corresponding to visiblelight, which are about 100,000 times faster. Although per-fect for even more precise clocks, these atoms posed aserious challenge. “Optical transitions are just way toofast,” says physicist Mukund Vengalattore of CornellUniversity in Ithaca, NY. No one knew how to countthese lightning-fast ticks.

A breakthrough came in 1999 when physicistsdeveloped what is known as a femtosecond opticalfrequency comb (3, 4). This device uses a laser to gen-erate extremely short pulses of light that appear asspikes at regularly spaced intervals, creating whatlooks like the teeth of a comb. Each pulse is of a dif-ferent color or frequency, and the comb spans theentire visible spectrum. Researchers could line up oneof the teeth of the laser light to the oscillations of anatom radiating in optical frequencies, lock the laser

Trapped atoms of strontium-87 produce a bluish glow within a shielding boxsurrounded by thermal sensors. The synchronized atoms are used to countinfinitesimal ticks for the world’s most precise optical atomic clock. Imagecourtesy of Ye Labs, JILA.

Published under the PNAS license.

www.pnas.org/cgi/doi/10.1073/pnas.1809852115 PNAS | July 17, 2018 | vol. 115 | no. 29 | 7449–7451

CO

RE

CO

NCEPTS

Dow

nloa

ded

by g

uest

on

July

19,

202

0

frequency to that line, and use sophisticated elec-tronics to tally the transitions. This opened the doorfor creating optical atomic clocks.

In 2013, optical atomic clocks were shown to bebetter than cesium atomic clocks, and new records inaccuracy have arrived yearly since then (5). With eachorder of magnitude in improvement, researchers haveimagined new ways to exploit the extreme abilities ofthese clocks. “When you have a new instrument,whatever it is, you always think how can you use it totest the known laws of physics to high precision,” saysphysicist Michael Romalis of Princeton University inNew Jersey.

Reality CheckIn recent years, researchers have been using opticalatomic clocks to investigate why our universe is theway it is. The standard model of particle physics de-scribes all known particles and the forces (other thangravity) that act on them. Physicists would like to ex-plain the values of the fundamental constants in thistheory from first principles. To do so they have tosometimes consider theories in which these constantscan vary from place to place in the universe and overtime. “Tests of fundamental physics using these kinds ofclocks is a far more viable way to go beyond the stan-dard model than the giant colliders,” says Vengalattore.

One such number is the so-called fine-structureconstant, a dimensionless quantity denoted by theGreek letter α that characterizes the strength of theelectromagnetic force. By studying the spectra of lightfrom distant, extremely bright objects called quasars,which existed approximately 10 billion years ago,

some astrophysicists have claimed evidence for aminute variation in the fine-structure constant—notjust in terms of past values compared with today’svalue of α, but even when observations of α from dif-ferent regions of the universe are compared (6, 7)

But astronomical observations have large uncer-tainties. So some teams have taken a different ap-proach to measuring variations in α, trying to see if itvaries over the course of a year on earth. If so, it wouldcause an extremely tiny but noticeable variation in theticks of atomic clocks—speeding them up or slowingthem down.

Several teams have carefully monitored opticalatomic clocks at different times during a year and de-termined that α cannot be changing in today’s universeby more than one part in 1018 (8). Physicist VictorFlambaum of the University of New South Wales Syd-ney says this almost rules out the effect seen in theastrophysical data, but the best tests will require as-yet-unrealized clocks based on a transition in thorium-229,which happens to be an especially sensitive probe ofvariations in α.

The optical atomic clocks are also sensitive to themass ratio of electrons to protons, another dimen-sionless quantity that is fine-tuned in the standardmodel. But so far, searches for any fluctuations in thisratio have turned up empty (9).

Clocks as Measuring TapeIt’s not obvious that keeping time has anything to dowith gravity, but according to Einstein’s general rela-tivity a clock near a strong gravitational field, such asnear the surface of the Earth, will tick more slowly than

A mobile optical atomic clock has the potential to revolutionize geodesy, a field that entails measuring the Earth’sgravitational field with high precision. A team at Germany’s national metrology institute is developing prototypesapproaching the accuracy of the best atomic clocks. Image courtesy of Physikalisch-Technische Bundesanstalt.

7450 | www.pnas.org/cgi/doi/10.1073/pnas.1809852115 Mann

Dow

nloa

ded

by g

uest

on

July

19,

202

0

one farther away, where gravity is weaker. The timeresolution of optical atomic clocks is such that raisingthem a single centimeter in height causes a noticeablechange in their ticking rate. This makes them ideal forgeodesy, which involves measuring the size and shapeof the Earth and its gravity field.

The data collected by geodesists so far is inconsis-tent. “They actually see huge discrepancies betweensome of the data of geodetic surveys from differentcountries,” says physicist Andrew Ludlow of NIST.“And even within a country, using different techniquesfor geodesy actually gets very different answers.”

Optical atomic clocks could help with superiorgeodesy, but they currently remain stuck in precisionlaboratory environments, requiring a tabletop of per-fectly calibrated lasers and mirrors. “You want tominiaturize all that, put it in a car trailer, move itsomewhere and reconstruct it, and make sure it stillworks after 3 weeks,” says physicist Christian Lisdat ofthe Physikalisch-Technische Bundesanstalt, Germany’snational metrology institute in Brunswick.

That’s exactly what Lisdat’s team is developing,though his portable devices are still an order of mag-nitude or so less accurate than the best laboratoryclocks (10). He expects to have mobile optical atomicclocks capable of sensing centimeter-scale changes inthe height of earth’s surface within the next 3 years.“This would be really a revolution for geodesy becauseyou could use a clock tomeasure the Earth’s [gravitational]potential directly,” says geodesist Markus Rothacherat the Swiss Federal Institute of Technology in Zurich.

The gravitational field at a particular location de-pends on the density of mass underfoot—the denserthe mass, the greater its gravitational field. Portableoptical atomic clocks could potentially discoverchanges in the subsurface mass and density and helpidentify deposits of oil and minerals without usingtraditional prospecting techniques (11).

The clocks could also revolutionize earth sciences.Studies of changes in mass and density could helpgeophysicists gain a more fine-grained understandingof the structure of the lithosphere and watch crustal

plates rise, slide, and deform by centimeters peryear. Placed in protective casing and positionedclose to a volcano, the clocks could track the move-ment of magma and provide data that might helppredict eruptions. Oceanographers and atmosphericresearchers could use them to monitor enormousoceanic storms and other large masses of water andair, which contribute to the total gravitational field intheir vicinity.

Optical atomic clocks located aboard GPS satelliteswould be able to determine positions on the Earth withthe accuracy of a centimeter, helping to autonomously

drive cars, land airplanes, or direct military missiles.Geodesists could use such detailed mapping to de-velop an International Height Reference System thatcould be useful for monitoring the effects of climatechange such as sea level rise.

Even as researchers dream up applications foroptical atomic clocks, others continue to improve theclocks themselves. Ye’s record optical atomic clockcontains a cube of 10,000 strontium atoms cooled tofractions of a degree above absolute zero. Sitting in a3D lattice of laser light, each atom is cradled like anegg in an egg carton, stabilized and shielded fromdisturbances by its neighbors, the ticks of all theatoms perfectly synchronized. Ye hopes to add moreatoms to the configuration and believes he willachieve another order of magnitude in accuracy within5 years.

With such techniques in mind, Vengalattore andothers think optical atomic clocks could get evenmoreprecise and spawn yet-unimagined applications. “I thinkin a certain sense,” he says, “we’re just scratching thesurface of accuracy.”

1 Marti GE, et al. (2018) Imaging optical frequencies with 100 μHz precision and 1.1 μm resolution. Phys Rev Lett 120:103201–103207.2 Essen L, Parry J (1955) An atomic standard of frequency and time interval: A cæsium resonator. Nature 176:280–282.3 Reichert J, Holzwarth R, Udem Th, Hänsch TW (1999) Measuring the frequency of light with mode-locked lasers. Opt Commun172:59–68.

4 Diddams SA, Ma LS, Ye J, Hall JL (1999) Broadband optical frequency comb generation with a phase-modulated parametric oscillator.Opt Lett 24:1747–1749.

5 Le Targat R, et al. (2013) Experimental realization of an optical second with strontium lattice clocks. Nat Commun 4:2109–2117.6 Murphy M, et al. (2003) Further evidence for a variable fine-structure constant from Keck/HIRES QSO absorption spectra.Mon Not RAstron Soc 345:609–638.

7 Webb JK, et al. (2011) Indications of a spatial variation of the fine structure constant. Phys Rev Lett 107:191101–191106.8 Huntemann N, Sanner C, Lipphardt B, Tamm C, Peik E (2016) Single-ion atomic clock with 3 × 10−18 systematic uncertainty. Phys RevLett 116:063001–063006.

9 Huntemann N, et al. (2014) Improved limit on a temporal variation ofmp/me from comparisons of Yb+ and Cs atomic clocks. Phys RevLett 113:210802–210807.

10 Grotti J, et al. (2018) Geodesy and metrology with a transportable optical clock. Nat Phys 14:437–441.11 Mehlstäubler TE, Grosche G, Lisdat C, Schmidt PO, Denker H (2018) Atomic clocks for geodesy. Rep Prog Phys 81:064401–064474.

“I think in a certain sense, we’re just scratching thesurface of accuracy.”

—Mukund Vengalattore

Mann PNAS | July 17, 2018 | vol. 115 | no. 29 | 7451

Dow

nloa

ded

by g

uest

on

July

19,

202

0