Physicists in Japan have developed a new kind of compact gravitational-wave detector that works by measuring the tiny rotations of two suspended blocks of aluminium. A far cheaper alternative to the more conventional interferometer-based devices, this "torsion-bar antenna" could plug a gap in the gravitational-wave spectrum – between the high-frequency waves observable today from the ground and the lower-frequency radiation potentially detectable in space – so expanding the range of very massive objects that astronomers can study.

Gravitational waves are ripples in the fabric of space–time predicted by Albert Einstein in 1916 and detected directly for the first time last September by the Laser Interferometer Gravitational-wave Observatory (LIGO) in the US. Each of LIGO's two detectors is a laser interferometer with two 4 km-long arms at right angles to each other. A passing gravitational wave can stretch one arm by a miniscule amount while compressing the other – and these changes can be measured with very high precision.

The LIGO detectors are shielded from terrestrial vibrations by suspending the interferometer mirrors – turning each mirror into a pendulum. As a result, LIGO cannot detect gravitational waves with frequencies below about 1 Hz, which is the resonant frequency of the mirror pendulums. Since heavier astronomical objects emit gravitational waves at lower frequencies, LIGO is only able to study relatively small objects – its first signal having been produced by the merger of two black holes weighing in at about 30 times the mass of the Sun.

Detectors in space

To observe gravitational waves at lower frequencies, many scientists are instead looking to vibration-free space. Among proposed missions is the European Space Agency's evolved Laser Interferometry Space Antenna (eLISA), which is due to be launched in the early 2030s. This would fire laser beams between free-floating test masses arranged in a triangular formation with arms a million kilometres long, and would target waves with frequencies between about 0.1–100 mHz.

In contrast, Masaki Ando of the University of Tokyo and colleagues aim to detect low-frequency gravitational waves on the ground – at a cost of just a few million dollars. Their detection process involves monitoring the effect of passing gravitational waves on two bar-shaped test masses positioned at right angles to one another and which rotate around a common axis of suspension. Rather than recording a length change, the Japanese group instead measures a tiny relative rotation – the waves would cause one test mass to move in a clockwise direction while sending the other anticlockwise.

It is probably the best concept for a roughly 1 Hz ground-based detector proposed to date
Hartmut Grote, Albert Einstein Institute

In this set-up the resonant frequency is not fixed by the strength of gravity and the length of the suspension – as for a pendulum – but instead by the suspension's tensional strength, diameter and length, as well as the bar's moment of inertia. Putting forward their idea in 2010, Ando and co-workers calculated that 10 m-long bars suspended by very narrow, soft wires would have resonant frequencies as low as a few millihertz and could turn through angles as small as 10–17 of a degree. This, say the researchers, would enable them to detect significant numbers of merging intermediate-mass black holes, which can weigh in at up to about a million solar masses.

Prototype built

The researchers have now built a small prototype detector comprising two bars, each 24 cm long. The detector is shielded from vibrations and the researchers used a laser interferometer to achieve angular sensitivities of up to 10–8 of a degree.

The team also showed that its antenna would be able to obtain three independent measurements from each passing gravitational wave – the average of the two bars' horizontal rotation and both vertical rotations. According to Ando's colleague Ayaka Shoda of Japan's National Astronomical Observatory, this increases the chances of detecting a wave in the first place (given that its direction would be unknown) and also provides more information about the wave's source, such as its location and rate of spin.

Shoda says that the biggest technical challenge in building the full-scale version of the antenna will be developing the cryogenics needed to reduce vibrations in the bars and wire, pointing out that the cryopump, which will be connected to the bars, will itself vibrate. She estimates that reaching design sensitivity could take anywhere between 10 and 20 years, but says that as an intermediate step, they first plan to demonstrate an angular sensitivity of some 10–13 of a degree. At this point, their device could pick up fluctuations in the local gravitational field due, for example, to seismic waves or atmospheric sound waves. Indeed, the researchers say that their technology might one day be used to generate earthquake alerts, given that gravitational effects travel at the speed of light while seismic waves typically travel at just a few times the speed of sound.

Despite the work that still needs to be done on the torsion-bar technology, Hartmut Grote of the Albert Einstein Institute in Hannover, Germany, believes the concept is worth pursuing. "It will take quite a while, plus uncertainties of funding, to get to an astrophysically interesting sensitivity," he says. "But it is probably the best concept for a roughly 1 Hz ground-based detector proposed to date."

Also enthusiastic is Jan Harms of the University of Urbino in Italy. He underlines how difficult it will be to remove gravitational noise from observations, noting that ideas for carrying out such screening remain unproven. But he says it is "important to close the frequency gap" between ground-based interferometers and LISA, and believes that the torsion-bar antenna is "one of the most promising concepts" for doing so.

The research is reported on arXiv.