At 4,326 meters above sea level, inside a tent on the summit of Mount Blue Sky, a portable ytterbium atomic clock sat humming beside a pair of space heaters while researchers swatted away the invasive moths threatening to land on the laser optics. Down in Boulder, Colorado, an identical clock ticked along in the climate-controlled quiet of the National Institute of Standards and Technology. The gap between them, measured not in distance but in time itself, was the whole point. General relativity predicts that time runs faster at higher altitude, and this experiment was built to confirm that prediction to a depth no one had attempted before, or to find the crack that might break open a new understanding of the universe.
Why Einstein’s equations might be running out of road
The experiment is not just a victory lap for a theory that has already passed every test thrown at it since 1915. It is a search for where that theory quietly fails. The conflict sits at a specific and uncomfortable place in physics: quantum mechanics tells us a particle with mass can exist in two places at once, in superposition. Einstein’s general relativity tells us that a particle with mass bends spacetime in a precise way depending on its location. Put those two ideas together and a question emerges that neither theory can answer. ‘Einstein’s theory of general relativity is silent on the question of what happens to spacetime when a particle is in superposition,’ explained the NIST team. That silence is where the experiment is aimed.
To probe it, the researchers needed a clock accurate to 18 decimal places. Caesium fountain clocks, which have defined the official second since 1967 by counting exactly 9,192,631,770 oscillations of caesium 133, are extraordinary instruments, losing only about one second across tens to hundreds of millions of years. But even that is not fine enough. The team turned to ytterbium 171, which resonates in the optical band at 518 trillion oscillations per second, roughly 56,000 times higher than caesium. A clock built on that frequency will not gain or lose a second across more than the current age of the universe.
Getting ytterbium to behave requires a sequence of steps that reads like controlled punishment. The metal, which resembles fool’s gold in its solid form, is heated to around 500 degrees Celsius in a small oven until it vaporizes and streams out through a narrow aperture. Those atoms are moving at about 300 meters per second, fast enough to cause a Doppler shift that would corrupt the frequency reading at the sixth decimal place when 18-digit accuracy is the goal. So the team fires blue and then green lasers, tuned just below the atom’s absorption frequency, at the moving cloud in three dimensions, slowing the atoms to within a millionth of a degree of absolute zero. The atoms are then trapped inside an optical lattice, a standing wave pattern created by a laser reflected back on itself, holding each atom in a pocket just 300 nanometers wide where its movement can no longer disturb the measurement.
An 80-kilometer laser beam aimed at an eight-micron target
The mountain clock’s signal has to reach the lab clock in Boulder to make the comparison meaningful. A fiber optic cable was ruled out because local wildlife would likely chew through it and the infrastructure simply does not exist on the mountain. Instead, the signal travels from the tent through optical fiber to a beam expander that sends it out as a 10-centimeter beam. Turbulence in the atmosphere spreads that beam to about a meter wide by the time it crosses the roughly 80 kilometers to a rooftop near Denver. On the receiving end waits an eight-micron core fiber optic cable. As one researcher put it: ‘You’re aiming at a one-meter spotlight, 80 kilometers… it’s a very small target.’
Reading the signal once it arrives posed its own decades-long obstacle. No electronic transistor can switch fast enough to count 518 trillion oscillations per second. The solution came from a 1994 moment when German physicist Theodor Hansch saw a titanium sapphire laser at a trade show and recognized that its ultrashort pulses contained a vast range of perfectly evenly spaced frequencies. That optical frequency comb, developed into a practical measurement tool by John Hall’s team in Boulder by 1999, gave scientists a ruler that spans from the microwave frequencies electronics can count all the way up to the optical frequencies the clocks produce. Hansch and Hall shared the 2005 Nobel Prize for the work.
With the link established and the frequency comb reading the signal, the data began to accumulate. General relativity predicts the two clocks should differ by 2.885 times 10 to the power of negative 13 seconds every second across the vertical separation between Boulder and the summit. Weeks after the mountain run, the team confirmed that Einstein is still right: there is no evidence yet of a deviation. But the experiment is probing at close to 10 parts per million precision, and the team is still analyzing the finest digits, digits they kept deliberately hidden from the group calculating the expected gravitational field until both sides were ready to compare results.
The moth that nearly landed on the lasers
Inside the summit tent, the thin walls keeping out the altitude cold also kept out a particular species of moth that had a habit of finding its way into the observatory and settling on the optical equipment.
The moths did not get in. The data did come out. Whether the numbers at the deepest decimal places match Einstein’s slope exactly, or tilt away from it by the tiniest measurable fraction, the answer is still being computed somewhere in Boulder.
Wake up to Only Happy News!
