A thorium 229 prototype in Vienna has locked a laser to a nuclear transition, and its real payoff is dark matter and tests of fundamental physics, not better timekeeping.
The most accurate timekeepers in human history tick by the gentle dance of electrons around an atom's outer shell. Physicists have now built the first clock that ticks by something far more stubborn: a transition inside an atom's nucleus, in a prototype device from Thorsten Schumm's group at TU Wien that the team describes as the first working nuclear clock.
The prototype was reported in Nature in 2024 under the title "Frequency ratio of the 229mTh nuclear isomeric transition and the 87Sr atomic clock". It locks a tabletop ultraviolet laser to a transition in thorium-229, an isotope whose nucleus harbors an unusually low-energy excited state. The very title of the paper places the new device against one of today's best strontium optical lattice clocks. The headline result is not a new champion of precision. It is a new class of measurement tool.
Conventional atomic clocks use transitions of electrons between energy levels. The nuclear clock drives a transition of the thorium nucleus itself, at roughly 8.4 electron volts in the deep ultraviolet. Most nuclear transitions sit at energies a million times higher, demanding particle beams rather than lasers. Thorium-229 is the lone exception: an isomer whose low-lying transition is just accessible enough for tabletop physics. As Schumm told New Scientist, the system is "the most simple thing you can imagine."
That simplicity is more apparent than real. The thorium-229 atoms are doped into a calcium fluoride crystal. A frequency-stabilized ultraviolet laser is swept across the nuclear transition, and the system locks when absorption is equal from both sides. The device does not need laser cooling, an ultrahigh vacuum chamber, or a cryogenic environment. It does require a UV source stable enough to interrogate a transition that has taken roughly two decades of effort to pin down precisely. The exact laser frequency was only nailed in 2023. The 2024 device is the first closed-loop clock built on top of that discovery.
The pitch for a nuclear clock has always rested on a single physical fact: the nucleus is shielded from its surrounding electron cloud. Electric and magnetic fields that would shift an electron transition barely perturb a nuclear one. In principle, that means nuclear transitions can be sharper, with quality factors no optical clock can match.
In practice, the TU Wien prototype is not yet sharper than today's best optical clocks. It loses tens of seconds over a billion years. The best strontium and ytterbium lattice clocks at NIST and JILA lose roughly one second over forty billion years. The nuclear clock sits one to two orders of magnitude behind. Independent experts quoted by New Scientist, including Harry Morgan of the University of Manchester and Eric Hudson of UCLA, described the device as a proof of principle, not a precision timekeeper. Schumm's group frames it that way as well.
The payoff lives elsewhere. A transition that physicists can drive with a laser becomes a precision probe. The same 2024 Nature paper used the nuclear transition's high energy to set a constraint on ultralight dark matter coupling to nuclei, comparing the thorium clock to a strontium reference. Hudson told New Scientist he expects orders-of-magnitude improvements in such searches as the technology matures. The instrument is also sensitive to possible drift in fundamental constants, and to relativistic effects on satellite timescales. Those applications have less to do with making a better wristwatch than with asking whether the laws of physics are the same everywhere and everywhen.
The concept is not new. Ekkehard Peik of PTB, the German national metrology institute, and Christian Tamm proposed a thorium-229 nuclear clock in 2003. A 2020 review by Lars von der Wense and Benedict Seiferle catalogued the long search for the transition energy. What the TU Wien team demonstrated in 2024 is the first device in which a laser is actually locked to that nuclear transition. The two-decade gap reflects how hard it is to find an energy level inside an atom that a tabletop laser can reach at all.
There is a practical wrinkle worth flagging. Thorium-229 is radioactive, an alpha emitter with a 7,940-year half-life. A clock built on it is not a casual laboratory tool. It requires radiation handling, isotope production, and a long-term plan for source supply. None of that has stopped metrology institutes from pursuing it, but it is the kind of detail that disappears in headlines about a new era in timekeeping.
The honest summary is narrower than the marketing. A thorium-229 device in Vienna has ticked for the first time by the rules of the nucleus. It is not yet a better clock. It is a new instrument, and the questions it can ask are bigger than time.