On 20 February, dignitaries will descend on Virgo, Europe’s premier gravitational wave detector near Pisa, Italy, for a dedication ceremony to celebrate a 5-year, €24 million upgrade. But the pomp will belie nagging problems that are likely to keep Virgo from joining its U.S. counterpart, the Laser Interferometer Gravitational-Wave Observatory (LIGO), in a hunt for gravitational wave sources that was meant to start next month. What has hobbled the 3-kilometer-long observatory: glass threads just 0.4 millimeters thick, which have proved unexpectedly fragile. The delay, which could last a year, is “very frustrating for everyone,” says LIGO team member Bruce Allen, director of the Max Planck Institute for Gravitational Physics in Hannover, Germany.
A year ago, LIGO confirmed a prediction made by Albert Einstein a century earlier: that violent cosmic events, like the merger of two black holes, would wrench the fabric of spacetime and emit ripples. But LIGO, with two instruments in Livingston, Louisiana, and Hanford, Washington, cannot pinpoint the sources of the waves, which would let astronomers train other telescopes on them. Triangulating on the sources requires a third detector: Virgo.
The detectors all rely on optical devices called interferometers: two straight arms, several kilometers long, positioned at right angles. Inside each arm a laser beam bounces back and forth between mirrors at each end of a vacuum tube, resonating like sound in an organ pipe. The laser light is combined where the two arms meet so that the peak from one laser wave meets the trough of the other and they cancel each other out. But if something, such as a gravitational wave, stretches space and changes the length of the two arms by different amounts, the waves will no longer match up and the cancellation will be incomplete. Some light will pass through an exit known as the dark port and into a detector.
The tiniest vibrations—earth tremors, the rumble of trains, even surf crashing on distant beaches—can swamp the signal of gravitational waves. So engineers must painstakingly isolate the detectors from noise. At Virgo, for example, the mirrors are suspended at the end of a chain of seven pendulums. For the upgrade, steel wires connecting the mirror to the weight above it were replaced with pure glass fibers to reduce thermal and mechanical noise.
But a year ago, the glass threads began shattering, sometimes days or weeks after the 40-kilogram mirrors were suspended from them. After months of investigation, the team found the culprit: microscopic particles of debris from the pumps of the upgraded vacuum system. When these particles settled on the glass fibers they created microcracks, which widened over days and weeks until the fibers failed. “The fibers are very robust until something touches their surface,” says Giovanni Losurdo, Advanced Virgo project leader at Italy’s National Institute for Nuclear Physics in Pisa.
During the investigation, the team temporarily replaced the glass fibers with steel wires—as in the original Virgo—and pressed ahead. But other problems compounded the delays. An examination of small steel triangles that act as vibration-damping springs revealed that 13 out of 350 were cracked or broken. Why remains a mystery, but the team replaced those that showed any sign of damage—40% of the total. Given the complexity of the detector, “it’s not surprising some things don’t work as expected,” says Virgo team member Benoit Mours of France’s National Institute of Nuclear and Particle Physics in Annecy.
The Virgo team has now achieved “lock” in the two detector arms, meaning that light is stably resonating. They soon hope to bring in the central optics and combine the beams. Then they must hunt down and eliminate remaining sources of noise to see what level of sensitivity they can achieve with the steel wires still in place.
For Virgo to make a useful contribution, it needs to be at least one-quarter as sensitive as LIGO. Researchers define sensitivity as the distance to which a detector could spot the merger of two neutron stars with masses 1.4 times the sun’s. LIGO’s Livingston detector can currently sense such an event out to about 80 megaparsecs (260 million light-years). In theory, with all of the upgrades but the mirrors still suspended with steel wires, Virgo should be able to reach 50 megaparsecs, Losurdo says. “As soon as Advanced Virgo reaches the sensitivity to join, we will start.”
Frustratingly for the Virgo team, the steel wires are expected to have the most impact on sensitivity to gravitational waves with lower frequencies than neutron star mergers, such as those from the mergers of black holes. And black hole mergers are precisely the events that LIGO detected last year.
The task of eliminating noise sources is sure to take several months, says Lisa Barsotti of the Massachusetts Institute of Technology in Cambridge, co-chair of the LIGO-Virgo Joint Run Planning Committee. That makes joining LIGO as planned in March a virtual impossibility. The current LIGO run, which began on 30 November 2016, was expected to continue for about 6 months, until late May 2017, but even that may be a stretch for Virgo. Barsotti says LIGO could extend its run by a month or two, “to give Virgo a chance to join.”
Whether Virgo manages to take part in the current run, the team should be able to reinstall the glass fibers and root out other sources of noise by the spring of 2018, when LIGO will start a new observing run. Soon after, a fourth detector is poised to join the hunt: the Kamioka Gravitational Wave Detector, or KAGRA, near Hida City in Japan, which plans to begin operations in 2019. KAGRA’s arms will be underground, below 200 meters of rock, and its mirrors chilled to 20 K—two tricks that should reduce noise and boost sensitivity.