Purchasing new hardware? Read our latest product comparisons

World's most precise clock only a second out every five billion years


January 22, 2014

The blue cloud of strontium at the heart of the world's most precise and stable clock (Photo: The Ye group and Brad Baxley, JILA)

The blue cloud of strontium at the heart of the world's most precise and stable clock (Photo: The Ye group and Brad Baxley, JILA)

Image Gallery (3 images)

Not satisfied with the accuracy of the "quantum logic clock" (which only gains or loses one second every 3.7 billion years), the National Institute of Standards and Technology (NIST) and JILA have unveiled an even more precise timekeeper. The strontium lattice clock sets new standards for precision and stability, only gaining or losing one second about every five billion years.

Atomic clock performance can be measured by two key metrics. The first is stability, which is how much its speed varies and therefore is a determining factor in how long the clock needs to run to achieve its best performance through continual averaging. The second is precision, which is how closely the clock approaches the true resonant frequency at which its reference atoms oscillate between two energy levels.

NIST says the stability of the new strontium lattice clock is roughly the same as NIST's ytterbium atomic clock unveiled last year. Both are so stable that it takes only a few seconds of averaging to outperform other types of atomic clocks that have been averaged for hours or days. However, NIST says the new clock is about 50 percent more precise than the previous record holder in terms of precision, its quantum logic clock, making it the first clock to hold both the stability and precision records since cesium fountain atomic clocks were introduced in the 1990s.

Although the new strontium clock is more precise, by definition, no clocks can be more accurate than cesium-based clocks. This is because they are still used for the international definition of units of time – the NIST-F1 cesium fountain clock, for example, is the current US civilian time standard clock. However, NIST says the strontium lattice and other experimental clocks may one day assume the mantle of timekeeping standard.

The JILA strontium lattice clock is an optical atomic clock that holds a few thousand neutral strontium atoms in an optical lattice of laser beams. The "ticks" of the strontium atoms, which occur 430 trillion times a second, are detected by bathing them in red laser light at a precise frequency that triggers a switch between energy levels.

Unlike atomic clocks that have only a single ion that flips back and forth between quantum states, the quantum fluctuations of the thousands of atoms in the strontium clock can be averaged out during a single measurement of clock ticks, therefore offering the same accuracy and reproducibility in less time.

For the time being, the strontium lattice clock lays claim to being the most precise and stable timekeeper on the planet, but the clock is no doubt ticking on the appearance of a clock that will knock it off its perch.

Sources: NIST, JILA

About the Author
Darren Quick Darren's love of technology started in primary school with a Nintendo Game & Watch Donkey Kong (still functioning) and a Commodore VIC 20 computer (not still functioning). In high school he upgraded to a 286 PC, and he's been following Moore's law ever since. This love of technology continued through a number of university courses and crappy jobs until 2008, when his interests found a home at Gizmag. All articles by Darren Quick

1 sec every 5 billion years! Why so much?


This is absurd ... for two reasons ...

1) How would they know the error, if there didnt exist a more-accurate clock to compare?

2) the passing of time is not constant: velocity, gravity, both modify it ... In fact , in our GPS satellite systems (as well as inside our submarines stationed in deep-sea) time is always varying ... Orbital speeds are not constant, and being in a sub, hence closer to earth's center, does slow the sub's clocks enough as to require re-targeting data modifications,

If another clock that we believed was more accurate was brought near this lil 5sec/5x10^9 year device, how would we be sure the 'comparator' click didnt experience some fluctuation or another, to account for any discrepancy it might otherwise reveal?



@tkj, I would imagine this is covered by the averaging. Given that the reference clocks are not orbiting, they are all attached to the earth so would experience the same fluctuations to the rate of passage of time...

They know the accuracy because they can calculate it based on the accuracy of the measuring device counting the 430 trillion ticks per second, and the variability of the rate of oscillation or ticks inherent in the setup.

Ian McIntosh

I concur with TJK. Turns out that the day was only 6 hours 50 million years ago as the moon slowly pulls away and other aspects effect spin. That being said, if time is not relative to earth's 24 hour day, then this is worth saving


All very well to claim this clock is accurate to even 1 million years, but will any ancilliaries (lasers etc) last long enough to make it close to that good? Estimations of the length of Earth's day variations due to the distances between sun, moon and Earth make this accuracy fairly useless if the clock stays true but time changes around it. One could postulate the idea that the impact of a tiny asteroid or comet just big enough to shake the Earth's crust (anybody see dinosaurs around lately) could alter the length of the day. Even scientists long, long into the future - assuming this or similar clocks keep continuous power supply - will write this off as "no longer valid" for many reasons. Sorry if I used too many big words.

The Skud

Those clocks mesure time in the UTC time standard, you are talking about the stellar day. Anyway, Mr. ADVENTUREMUFFIN, the day wasn't 6 hours 50 million years ago, it was the Eocene epoch, but the length of one day has been estimated as 21.9 hours 620 million years ago from rhythmites (alternating layers in sandstone).

Leo Parreiras

Is the frequency oscillation of an Quantum particle, (be it a Quark, a boson, or a quantum leaper) change when near the event horizon of a black hole, or is its frequency oscillation only slower in relation to other similar particles further away from the black hole? As mentioned above, time changes in relation to the time piece's proximity to a mass. The worlds most precise clock may only be that precise in one position in space- yet that position is ever changing. Confusing. This leads me to wonder if there is a ( literally) "Universal -Now- moment" everywhere regardless of light speed. Entanglement suggests that there is. Spooky action at a distance does exist. If so, then the other side of the universe is always right next to us, and Time is an illusion. Perhaps the most accurate clock in the world should have only one hand that does not move, pointing to "now".

James Simpson

Don't look at the big picture, look at the small picture! Most people don't care about 5 billion years away, because they will be dead. But if a clock is that accurate at 5 billion years, it is accurate enough to proof many things in short time.

How do you measure the difference between speeds of 2 different atoms if you don't have a clock accurate enough to proof the time difference? This clock can do wonders to proof the theories, and continue research with the correct results instead of staying in the unknown,


Physicists assume they understand time but just look at their definition. The second is defined against the meter and the speed of light. The speed of light is defined against the second and the meter. The meter is defined against the speed of light and the second.

The speed of light was arbitrarily set to a constant in deference to Einstein.

This is not science and this clock has an unknown accuracy.

Post a Comment

Login with your Gizmag account:

Related Articles
Looking for something? Search our articles