It takes a very long time for a light to reach us from a very distant star. Do GWs take as long ?

Yes. According to a calculation derived from Einstein's theory of general relativity, the propagation speed of GWs is the same as the speed of light. However, when a supernova occurs, a star's composition turns into gas and fills all of the space surrounding it, so it takes some time for the light to escape. GWs, on the other hand, are able to be propagated through all matter, so GWs are thought to reach the earth before light. Someday we hope to observe the details of the exact moment of explosion.

If neutrinos reach earth before a GW, we have to consider the possibility that the speed of GWs is different than the speed of light. In that case, we may have to make adjustments to the theory of general relativity. If the speed of a GW can be proven to be the same as the speed of light, it would support the theory of general relativity.

So the detection of GWs might result in the need to correct the theory of general relativity, which originally predicted the very existence of GWs. Now, please tell us about the TAMA300 GW detector.

TAMA300 is an interferometric detector that uses laser beams. Within each arm of its L-shaped tubes, a pair of mirrors is suspended at a distance of 300m from each other. In those two opposing mirrors, laser light is accumulated. Fluctuation of the mirror distance is detected from changes to the interference condition.
The longer the tubes are (the greater the distance between the two mirrors), the higher the accuracy of the interference becomes. The largest facilities in the world even have four-kilometer-long tubes.

You say that lasers are "accumulated to be measured." Does that mean that the lasers repeatedly bounce between the mirrors ?

Overview of TAMA300
L-shaped tubes are placed underground

Yes. The method you just mentioned is called "delay line," when lasers are reflected in a zig-zag pattern to attain the same effect as stretching the distance between mirrors.

TAMA300 employs the "Fabry-Perot" method. The reflectance of the front mirror is 98.8%. That means the transmissivity is 1.2% and almost the entire laser is reflected. However, utilizing the optical resonance, lasers can be increasingly accumulated within a Fabry-Perot resonator. In addition, 1.2% of the light that enters the front mirror leaks out, but the rest is accumulated within the resonator.

Summary of Fabry-Perot resonator

Just as with the delay line method, a detector's sensitivity improves when more laser beams accumulate. TAMA300 is able to achieve a practical efficiency 300 times the actual length of the facility.

What part did Nikon play in TAMA300 ?

Nikon mirror suspension system
Most of the optical parts are suspended on the suspension sysytem

Nikon Engineering supplied us with the suspension systems for optical parts such as mirrors. Through extensive basic research, we determined the required specifications, such as the length of wire that suspends the mirrors. However, to design an actual assembly, we had to consider various facts not only performance but also compatibility, installation and maintenance. That process required engineering. We chose Nikon Engineering because they are used to receiving such detailed, customized orders, and we felt comfortable discussing specifications with them. We have 11 suspension systems made by Nikon Engineering in this facility.

To accumulate light within the resonator, incoming and outgoing light has to be resonated, or match the peaks of the phases perfectly. To achieve that, the position of the two mirrors becomes crucial. To cancel internal vibration, caused by the adjustment of the mirrors, and external vibrations, such as seismic ground movement, we also need stabilizing optical parts such as mirrors for long-term observation.

Overview of TAMA300 detector

• Nikon Engineering (in Japanese language only)

Have observations with TAMA300 been proceeding smoothly ?

Image of the observable range by GW detector
Observable range: LCGT (planned at Kamioka mine, Gifu): 200Mpc,
LIGO/VIGO: 20Mpc,
GEO600: 1Mpc
TAMA300: our galaxy
Mpc: Distance unit used in astronomy
1pc (per sec): 3.26 light years (30,857 billion km)
M (Mega) indicates one million times

We began installation in 1997 and started operation in 1999. In 2001, we maintained a continuous observation for 50 days a world first. Now, we have succeeded in increasing sensitivity up to 2 x 10-21 √Hz at around 1kHz and are able to continue observing at a band of 100Hz to 3kHz.

At present sensitivity levels, however, we can only monitor our own galaxy, where the possibility of a GW-scale event is very low. If we increase sensitivity further, more galaxies can be observed and chances for detecting GWs increase. We are trying to raise the sensitivity of TAMA300 to as close as possible to the theoretical limit of 2 x 10-22√Hz.

It is relatively easy to increase sensitivity at the high-frequency range, but there is a limit to increasing sensitivity at the low-frequency range due to the fact that TAMA300 is located in an urban area and is susceptible to seismic noise. To overcome this, we are proposing the LCGT (Large-scale Cryogenic GW Telescope) project, creating a large-scale detector in a mine in Gifu, Japan. The length of tubes in the LCGT will be 3km, 10 times that of TAMA300. By increasing the sensitivity of TAMA300 to its theoretical limit and by applying this know-how to a large-scale detector, we are hoping to get closer to detecting GWs.

TAMA300 observation data
Shows the relation between the frequency and sensitivity
The sensitivity is improved from February 1999 (pink line) to black line.
The gray line at the bottom is the value of the immediate goal