Bringing an Atomic Clock Back to Life

A journey to the heart of time puts NIST’s fountain clock back online.  

When physicist Vladislav Gerginov arrived at the National Institute of Standards and Technology in early 2020, he was handed a daunting task — and one that sits at the heart of the agency mission: Restore NIST’s cesium fountain clock to operation as a primary frequency standard.

Fountain clocks, though few in number, play a critical role in our globally connected society: They help synchronize the billions of regular clocks and networks we all rely on. Without fountain clocks, our global timekeeping system — the one that creates and distributes the time that appears on our phones, computers and smartwatches — would be more wobbly and less reliable. So would our telecommunications and transportation systems, financial trading platforms, GPS and more.

Cesium fountains play a second vital role: They are the primary standards that measure the fundamental unit of time — the second — as it is officially defined by international agreement. They both contribute to and calibrate international atomic time; all other time measurements ultimately trace back to the fountains.

But these are not typical clocks — and not just anyone can open them up and make a broken one tick again. They don’t have springs, gears or quartz crystals. They are atomic clocks — complex, high-precision devices that extract timing pulses out of the hearts of atoms. Fountain clocks don’t even tell the time in a literal sense. Instead, they calibrate frequency, as a tuning fork does.

Fountain clocks are also among the rarest, most bespoke and most technologically sophisticated instruments in the scientific world. Fewer than 30 have ever been built; fewer than 20 are operating today. The number of people who know how to make them tick is probably not much larger than that. And Gerginov is one of them.

Over the past four and a half years, he has taken the lead in restoring NIST to the elite group of labs running cesium fountains and helping provide the ultimate foundation for global timekeeping.

“Vladi was the perfect person to come when we needed help with the fountains,” says Liz Donley, director of NIST’s Time and Frequency Division in Boulder, Colorado. “We couldn’t have gotten a better person.”

A Strange Kind of Clock

Cesium fountain clocks are a type of atomic clock — a complex, high-precision device that extracts timing pulses from atoms. They use a design first proposed by MIT physicist Jerrold Zacharias in the 1950s, demonstrated by Stanford University physicist Steven Chu and colleagues in the 1980s and refined in the 1990s by scientists in France. First, a cloud of tens of thousands of cesium atoms is cooled to near absolute zero using lasers. Then, a pair of laser beams toss the atoms gently upward, after which they fall under their own weight.

During their journey, the atoms pass twice through a small chamber full of microwave radiation. The first time, as the atoms are on their way up, the microwaves put the atoms into a quantum state that cycles in time at a special frequency known as the cesium resonant frequency — an unchanging constant set by the laws of nature.

About one second later, as the atoms fall back down, a second interaction between the microwaves and the atoms reveals how close the clock’s microwave frequency is to the atoms’ natural resonant frequency. This measurement is used to tune the microwave frequency toward the atomic resonance frequency.

A detector then counts 9,192,631,770 wave cycles of the fine-tuned microwaves. The time it takes to count those cycles defines the official international second.

National labs devoted to measurement science, or metrology, started developing fountain clocks in the 1990s, building on advances that had given physicists powerful new tools for cooling and controlling atoms. 

In late December 1999, NIST-Boulder scientists Steve Jefferts and Dawn Meekhof unveiled NIST-F1, NIST’s first fountain clock. The clock was so accurate that it could have, in theory, ticked for 20 million years before gaining or losing one second, putting it among the world’s best atomic clocks at that time. 

NIST, along with SYRTE and PTB, the national metrology labs of France and Germany respectively, became the first to use fountain clocks to help keep official time. Soon, fountain clocks were enshrined as the primary frequency standards that most accurately defined the official unit of time: the second.

NIST-F1 ran for more than a decade and a half and was used to perform regular frequency calibrations. But fountain clocks can be as fragile as they are precise, and after a move to a new building in 2016, the clock had to be restored and carefully tested to operate as a primary frequency standard again.

NIST has continued to keep accurate time by running other atomic clocks and calibrating them to the global time standard produced by the International Bureau of Weights and Measures (often known by its French acronym BIPM), located outside Paris.

But without a cesium fountain clock, the U.S. has had to rely on other countries for the pristine seconds that define the official second and tune international time. And with so few fountain clocks anywhere — only nine other countries operate them, as of this writing — the loss of even one could be felt everywhere.

Getting Back on Track

Gerginov took charge of NIST’s fountain clock in March 2020. It wasn’t his first tour of duty at NIST: During earlier stints, he had worked on smaller atomic clocks and devices to precisely measure magnetic fields.

Almost at the same time, the coronavirus pandemic hit, shutting down much of the world — including much of NIST. Gerginov quickly got special authorization to come to campus, because complex machines like cesium fountain clocks cannot be fixed remotely — and because NIST’s timekeeping duties are considered mission critical. “Vladi worked in the lab through the whole pandemic, every day,” Donley says.

Before coming to NIST, Gerginov had spent eight years at the German national metrology institute PTB, which runs fountain clocks that are as accurate and stable as any in the world. National metrology institutes such as PTB and NIST play a special role in global timekeeping: Their scientists run cesium fountain clocks and other clocks and send their output to BIPM, which averages data from clocks around the world. So at PTB, Gerginov had gained experience working at the heart of the fountain clock world.

His first challenge at NIST was diagnosing the NIST-F1 fountain. Gerginov found that the clock’s ticking rate changed too much when the cesium atoms deviated from a perfectly vertical trajectory. This frequency shift led him to suspect a problem with the microwave cavity — a sophisticated metal chamber the size of a soda can that sits at the heart of the clock, where the atoms are immersed in fine-tuned microwaves that enable pinpoint timekeeping measurements. He and his colleague at NIST, physicist Greg Hoth, now knew what they needed to do: build and rewire a new cavity from scratch.

Gerginov and Hoth based the cavity design on a simple cylindrical geometry that was relatively easy to model and fabricate, with dimensions close to those used in PTB’s fountains. They also tapped into expertise housed just down the road from NIST’s Boulder campus at JILA, a joint institute between NIST and the University of Colorado Boulder, whose machinists can shape parts to exacting specifications.

The specs for the cavity were exacting indeed. Even minute variations in the shape and size could shift the microwaves enough to throw the clock’s measurements off by an unacceptable amount. The JILA machinists needed to achieve tolerances of 5 to 10 microns — roughly one-fifth the width of a human hair. The tolerances had to be verified by cavity resonance measurements with instruments that were, ironically, referenced to a commercial atomic clock.

Gerginov installed the new microwave cavity in December of 2022.

But installing the new microwave cavity was just the first step. The team next had to wrap the clock in aluminum foil and “bake” it to remove residual gases from the system using vacuum pumps. They added and fine-tuned new electric heating coils, magnetic coils, optics and microwave components. The NIST team decided to name the new fountain NIST-F4.

(NIST has built two other fountain clocks, NIST-F2 and NIST-F3, making NIST-F4 the fourth in the series. NIST-F2 was an advanced, cryogenically cooled fountain clock that operated from 2014 to 2016. NIST-F3 serves as a stable frequency reference for the NIST lab that produces official U.S. time but lacks the accuracy to serve as a primary frequency standard. Read more about NIST’s fountain clocks.)

Then it was time to work out all the kinks. More than 10 different factors can upset the clock’s frequency measurements, including pressure fluctuations, stray electric and magnetic fields and collisions between cesium atoms that aren’t properly accounted for. Even the presence of people in the room where the clock is running can change the room temperature and disrupt the clock’s measurements. The NIST team spent more than a year taming the various forces conspiring to upend the fountain clock’s delicate quantum dance.

“You basically have a list with 100 things to be done,” Gerginov says of a typical day. “You generally work on number 1 and number 2, and if you get to 3, it’s a miracle.” 

The ideal clock output is as boring as possible: very steady, very precise, like a healthy heartbeat on an EKG. In early 2024, just when they thought they had achieved that, Gerginov and Hoth noticed that a new source of noise had popped up. For days at a time, their frequency measurements would be all over the place; then, just as mysteriously, the problem would disappear. It took several months of frustrating trial and error to pinpoint the problem, which ultimately required rebuilding the part of the clock used to measure the atoms after their second pass through the microwaves. 

“I would describe NIST-F4 as stubborn,” Gerginov says. Compared to other clocks he’s worked on, “it gave me a lot more headaches.”

“It takes a special person with a lot of grit to not give up,” Donley observes.

“It would be hard to imagine a better person than Vladi,” agrees Stefan Weyers, a physicist at PTB who worked with Gerginov there. “It was a very lucky situation for NIST that Vladi was available.”

Gerginov and Hoth eventually tackled the noise problem and ran the clock for several weekslong periods to amass data that they sent to BIPM, which assembles a team of experts that consider allowing new fountain clocks to become official primary frequency standards and contribute to international atomic time.

The international timekeeping body is understandably “very conservative” when it comes to certifying new fountain clocks, Gerginov says. “You don’t do these things lightly. You don’t accept an instrument that has some question marks — it could lead to problems.”

Ultimately, as the NIST team reported in April in the journal Metrologia, NIST-F4’s frequency measurements were accurate to within 2.2 parts in 10 quadrillion (1 with 16 zeroes after it), making it one of best fountain clocks in the world. Another way to think about that: If a clock that accurate ran for 100 million years — since back when dinosaurs roamed — it would be off by less than one second today.

Such accuracy is obviously far beyond what any of us need to be on time for appointments or tell a microwave oven how long to heat up lunch. But our modern lives now depend on intricate global systems — the internet, the stock market, cellphone networks, GPS — that do require that kind of accuracy. If the timing of a GPS satellite were off by just one millionth of a second, for example, that could lead to a positioning error as large as several city blocks.

The world’s fountain clocks provide the ultimate foundation for the entire edifice of global timekeeping, and by extension, much of the world’s economy. And thanks to Gerginov and his colleagues, NIST can once again provide a cornerstone in that foundation.