Estimation of losses in a 300 m filter cavity and quantum noise reduction in the KAGRA gravitational-wave detector

The sensitivity of the gravitational-wave detector KAGRA, presently under construction, will be limited by quantum noise in a large fraction of its spectrum. The most promising technique to increase the detector sensitivity is the injection of squeezed states of light, where the squeezing angle is dynamically rotated by a Fabry-Pérot filter cavity. One of the main issues in the filter cavity design and realization is the optical losses due to the mirror surface imperfections. In this work we present a study of the specifications for the mirrors to be used in a 300 m filter cavity for the KAGRA detector. A prototype of the cavity will be constructed at the National Astronomical Observatory of Japan, inside the infrastructure of the former TAMA interferometer. We also discuss the potential improvement of the KAGRA sensitivity, based on a model of various realistic sources of losses and their influence on the squeezing amplitude.

Figure 1

Initial Virgo map (top) and Advanced Virgo map (bottom), along with mirror maps (left) and PSD (right).

Figure 2

Round trip losses as a function of the radius of curvature. Peaks correspond to values of RoC for which the cavity is quasidegenerate.

Figure 3

Roughness maps for the Advanced Virgo configuration, mirror map (left), and PSD (right).

Figure 4

Round trip losses as a function of RoC for different values of the mirror diameter. Peaks that correspond to cavity degeneracies are reduced for smaller mirrors while the floor of round trip losses increases.

Figure 5

Amplitude of the field scattered out of the cavity by high spatial frequency defects. The field is observed after being reflected from a mirror with a real map and then propagated for 300 m.

Figure 6

Squeezing degradation budget. Quantum noise relative to coherent vacuum in the signal quadrature for an ideal system (blue curve) is compared with that obtained taking into account degradation mechanisms (one by one).

Figure 7

Quantum noise for KAGRA without squeezing, compared with the quantum noise using 9 dB frequency-dependent squeezing, both in the ideal system and when all the degradation mechanisms are taken into account. We remark that quantum noise without squeezing is relative to official KAGRA sensitivity using a configuration with a homodyne detection angle of 121.8°, while quantum noise in the presence of squeezing uses a standard homodyne angle of 90°.

Figure 8

Improvement in KAGRA sensitivity using 9 dB frequency-dependent squeezing, considering lossy cavity and other degradation mechanisms.

Figure 9

The quantum noise in the presence of 9 dB squeezing and lossy system (filter cavity RTL 80 ppm) is compared with the other noise sources. Note that below 100 Hz, the contribution from thermal noise would prevent an improvement in the total sensitivity even in the case of a further reduction of quantum noise.