At the National University of Singapore, one charged lutetium atom held in a trap now keeps time with a fractional uncertainty of 1 in 10 to the 19th power. Researchers at the university’s Centre for Quantum Technologies report that this is the lowest uncertainty ever recorded for an optical atomic clock, and Live Science puts the gain at 41 percent over the previous record holder.
The work was published in Nature on September 23, and it lands while the world’s timekeeping bodies are weighing a new definition of the second.
Lutetium-176 and a 19-digit result
The clock at the center of the paper uses a single lutetium-176 ion whose electron transition is matched to a laser at a wavelength of 848 nanometers, according to SciTechDaily’s summary. The team was led by Associate Professor Murray Barrett, a principal investigator at the Centre for Quantum Technologies, and the university’s own announcement names Dr. Kyle Arnold and PhD student Michael Lee as joint first authors.
The figure needs a careful reading. Nineteen decimal places is shorthand for an uncertainty of 1 x 10^-19: the clock’s frequency is pinned down so tightly that, were it run for the age of the universe, the error would be a small fraction of a second. It describes how well the transition frequency is controlled, which is a different thing from how the clock performs as an everyday timepiece. To test the claim, the group built two lutetium clocks and compared them over 200 hours using correlation spectroscopy, and the two agreed to within 5.7 x 10^-19, which NUS describes as the most precise clock comparison ever made. Agreement between two independent copies matters because it shows the uncertainty budget is real and not an artifact of a single apparatus: if the systematic shifts were mis-estimated, two clocks built the same way would drift apart.
Barrett was blunt about what he thinks the result means. In the university’s release he said he is confident that the clock now in his laboratory is the most accurate in the world.
An endurance athlete among atoms
Lutetium’s selling point is toughness. Competing optical clocks built on ytterbium, strontium or aluminum ions must be shielded carefully from stray heat and magnetic fields, whereas the lutetium transition is far less sensitive to both. Live Science quotes Barrett calling the element “a natural born endurance athlete,” able to keep performing through temperature swings of up to 5 degrees. The team also developed a technique called hyperfine averaging that suppresses gravitational-field effects and other environmental disturbances.
The Singapore group had been heading here for years. An earlier paper from the same laboratory, posted to arXiv in 2022 by Zhang Zhiqiang, Kyle Arnold, R. Kaewuam and Barrett, showed two lutetium references agreeing at the 10^-18 level without magnetic shielding, active field stabilization or extreme temperature control. The new result pushes the same approach another order of magnitude deeper.
The sensitivity has a physical side effect. OpenGov Asia reports that the comparison could detect a height difference of about 5 millimeters between two clocks sitting on the same table, because gravity slows time slightly with altitude. That makes the device a candidate gravity and geophysics sensor, and the team plans to shrink the apparatus into a transportable system for field comparisons, which would let it be carried to other laboratories rather than waiting for their clocks to travel to Singapore.
The second is still defined by caesium
The current second rests on a microwave transition in the caesium-133 atom. The International Bureau of Weights and Measures says on its redefinition FAQ that the best caesium fountain clocks have already been surpassed by several optical frequency standards, which is the reason a change is on the table, with ratification expected at the 2030 General Conference on Weights and Measures. Caesium standards would remain in use but as secondary references.
The Singapore result therefore feeds a decision that has not been made. NUS says the findings support a possible redefinition in 2030 or later, and Live Science describes lutetium as currently appearing to be the leading candidate to replace the caesium-based definition set in 1968. That ranking comes from the Live Science report alone; the BIPM page does not name a preferred atom, and ytterbium, strontium and aluminum clocks have posted their own records.
The 1968 definition, adopted when caesium clocks were the best available, ties the second to exactly 9,192,631,770 cycles of the caesium-133 transition, and any optical replacement would fix a different, far higher frequency in the same way. Swapping definitions does not change how long a second is in daily life; it changes how finely the unit can be realized and compared worldwide.
What remains open is whether other laboratories can reproduce an uncertainty near 10^-19 in lutetium, since international agreement on a new second depends on independent clocks agreeing with one another, not on one group’s best figure.
This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.
More from Morning Overview
