Newton saw gravity as being the force between massive bodies. Einstein saw it as geometry that is determined by masses and energy. His theory of relativity predicted gravitational waves - oscillations in the curvature of space-time. The effects are very subtle but still carry energy. It was thought that the most likely way that we would detect gravitational waves would be through a binary star system. The two objects spin faster as they get closer together and hence produce more gravitational waves. The final few orbits before they finally crash into each other are at 100 hertz.
The first detectors were built around lots of little masses that would be effected by gravitational waves. The effect would be very small and gravitational waves are much more complicated than electromagnetic waves (10 different numbers are required to describe them) The detectors were 1km circumference and moved only 10-21m - that's a 1000th the diameter of a proton. In 1916, Einstein thought that they would be undetectable.
Then came the Michelson Interferometer Detector, which turned the masses into mirrors. The idea being that gravitational waves would cause a distortion and so the mirrors would oscillate. This was actually using some very non-relativistic ideas but it did win a Nobel prize even though it didn't actually detect gravitational waves (this was because they didn't have lasers at the time)
Then in 1972 Rai Weiss was asked to teach general relativity at MIT. He did this by learning it himself and managing to stay one week ahead of the students. While doing this he realised that it would be possible to build a gravitational wave detector using lasers. This was essentially the design for LIGO. Not only did he design the experiment but he correctly predicted most of the noise sources - the thing that could ruin the whole thing. LIGO features freely floating mirrors - masses suspended from wires in a quad pendulum, which was designed in Glasgow.
The experiment uses infra-red light (as an aside, humans have no blink response to IR, you'll be blind before you even realise) The whole thing is 4km long and features 8,000m3 of a vacuum system that costs a quarter of a million dollars to pump all of the air out of. The system is kept at 300oK and they have to be careful to supress harmonics and exclude any feedback. There are two sites, one in Washington and one in Livingston, Louisiana, which is built in a swamp.
On the 14th of September 2015, both detectors saw the same thing. When the signal was compared to the theoretical source for binary black holes, there was only Gaussian noise difference (ie the signal was almost identical to what the model predicted) In fact, the detection was a 5 sigma event, meaning that the scientists have a huge level of confidence in what they found. However, it was actually detected the day before the machines were due to be switched on! The second detection took place on boxing day 2015.
The binary black holes that collided were massive - one was 36 solar masses (36 times the mass of our sun) and the other was 29. The resultant single black hole that was left was 62 solar masses, that means that 3 solar masses were emitted in the form of gravitational waves. For a tenth of a second this collision outshone the rest of the observable universe combined. The event took place a billion light years away, so while these two black holes were colliding, multi-cellular life was just beginning here on Earth.
LIGO is being switched back on at the end of the year after some changes designed to make it more sensitive (they have made the lasers more powerful) They are hoping to see tens of events that create gravitational waves. A new experiment called GOTO is also going to be run in conjunction with LIGO. This is designed to look for optical evidence of events that create gravitational waves - it has a 40 square degree field of vision and is dedicated to follow LIGO triggers. While black holes colliding don't emit much light, it is thought that gamma ray bursts may happen when neutron stars collide.
Gravitational waves lose none of their information as they propagate. Does this mean that we could detect waves from the big bang itself? It's possible but highly unlikely due to there being too much background information so we probably won't detect those primordial waves. We also can't detect binary white dwarves even though there are more of them than binary black holes. This is because LIGO can't detect anything less than 10 hertz. However, we are still in a better place than we were in 1987. Supernovas happen in our galaxy approximately once every thirty years and when one occurred in 1987 every gravitational wave detector on Earth was switched off!
In the words of Dave Reitze, "we did it! We detected gravitational waves" We now have new instrument for us to explore the universe with. The next steps are to observe more events, continue to improve the sensitivity of LIGO and to build more detectors. Then to use GOTO to detect the optical counterpoint and eventually to launch a space-borne experiment (LISA) that can look for gravitational waves. This is due to go live in 2035. In the meantime, keep a lookout for Rai Weiss who is favourite to be given a Nobel prize this year.
SciBar returns to The Vat & Fiddle on Wednesday 26 October when Dr Quentin Hanley from Nottingham Trent University will talk about "A Naive Person's View of Chemical (and Other) Measurement"
SciBar Meetup Page