Waving Back
Of the forces in nature, gravity is by far the weakest. For the proton and electron in a hydrogen atom, the force of gravity is about 1039 times smaller than the electrostatic force, which is what holds atoms together and determines their chemical behavior.
Yet the electrostatic force has no observable effect on the motion of planets, stars, and galaxies. That's because atoms and molecules are electrically neutral, so the electrostatic force from the electrons is canceled by an equal and opposite force from the protons. Gravity, on the other hand, is always attractive, so there is no cancellation, and gravity is the force that shapes the universe.
Gravitational waves emitted by a system of two collapsed stars spiraling into each other (credit: Ligo Laboratory)
Schematic of a laser interferometer gravitational wave detector; the mirror where the arms meet is half-silvered; note that each arm has two mirrors, because in this detector two interferometers, with different lengths, operate simultaneously. (image credit: LIGO Laboratory)
Let's continue the comparison of gravity and electromagnetism. Shake a charge and it gives off electromagnetic waves, or, to say it a bit more succinctly, accelerated charges radiate. But there was no counterpart in gravity, until Einstein in his general theory of relativity predicted in 1916 that gravitational waves would be emitted by accelerated masses (see Gravitational Waves). According to Einstein, as a gravitational wave passes, the very fabric of space stretches slightly, alternately along the direction of the wave and transverse to it. Many physicists were skeptical, and Sir Arthur Eddington, who verified general relativity's prediction that starlight would be bent by the sun, is reported to have said derisively that "gravitational waves propagate at the speed of thought."
As for sources of gravitational waves, physicists concluded that the best candidates would be the rare double star composed of two neutron stars or a neutron star and a black hole. In these exotic systems, the two objects, both already collapsed to diameters of a few tens of kilometers, could approach each other very closely and exert huge gravitational forces on each other, producing correspondingly large accelerations and the emission of gravitational waves (see drawing). As the two objects merge, the frequency of rotation would increase rapidly, giving the emitted waves a characteristic amplitude and frequency signature.
Given the weakness of the gravitational force, and the fact that the spreading of the wave reduces its energy correspondingly, the amplitude observed on Earth would be incredibly small. As the wave swept by, distances would change by a factor of about 10-21, corresponding to roughly a thousandth of a proton over three kilometers.
In the 1960s and 1970s, a US physicist mounted a substantial research effort to detect gravitational waves by measuring the deformation of large aluminum cylinders. He claimed to have found them, and moreover said that they came from the center of the galaxy. As it turned out, his results were called into question, but he inspired the next generation of researchers to tackle this extraordinarily difficult experimental problem.
In the 1980s, physicists began to contemplate how new optical technologies might make it possible for a very large Michelson interferometer to measure the effect of a passing gravitational wave. A Michelson interferometer is an optical device, as shown in the drawing, with two arms and a half-silvered mirror in the center that splits the light beam into two parts, one in each arm. These beams are then reflected back, and when they recombine, they interfere. If the beams cancel—if they're 180 degrees out of phase—there is no light in the recombined beam. Should the distance one of the beams travels change, light will be detected in the combined beam, and this effect is the basis of a gravitational wave interferometer.
How large a change must be detected? Since gravitational waves distort space by a factor of about 10-21, an interferometer with arms four kilometers long—about as long as could be constructed on Earth—would produce a change in length in each arm of about 4 x 10-21 km or 4 x 10-18 km, rough a thousandth the diameter of a proton, an unimaginably small distance to measure.
