Extending Ground-Based Gravitational-Wave Sensitivity to 5 Hz

Abstract

Extending the sensitivity of terrestrial gravitational-wave detectors below 20 Hz is a long-standing challenge, limited by ground motion and inertial sensing noise. In this letter, we demonstrate ultra-high-vacuum compatible inertial isolation and position sensing technologies that achieve active platform stabilization down to 10 mHz. Our laser position sensors reach a sub-pm/$\sqrt{\rm Hz}$ sensitivity above 10 mHz, independent of the input light polarization, representing a 100-fold improvement over the current LIGO position sensors. In addition, our inertial sensors provide at least a factor of 5 improvement in low-frequency sensitivity compared to state-of-the-art commercial seismometers. We integrate these technologies into a LIGO-like interferometer model and predict a low-frequency sensitivity improvement of up to an order of magnitude at 10 Hz, with enhanced linearity and calibration stability. This extension increases the detection horizon for intermediate-mass black hole binaries of mass $10^3 M_\odot$ by a factor of 3. Our results provide the first experimental demonstration of a practical pathway to sub-10 Hz operation of terrestrial gravitational-wave detectors and establish key technologies for next-generation observatories such as Cosmic Explorer and Einstein Telescope.

Figures (5)

• Figure 1: Detector strain sensitivity comparison for the A+ design and LIGO-5 Hz design LIGO-LF. The technical noise with the current inertial isolation schemes is shown in violet. The estimated improvement in the technical noises using our technologies (black) will enable the A+ design sensitivity to be reached, and demonstrates a clear pathway for terrestrial GW detection at 5 Hz.
• Figure 2: Photos of the three technologies and their respective noise budgets below. (a) Measured noise for the LPS (black). Subtraction of two sensors enables common-mode rejection of the DC-coupled laser frequency noise in the difference channel. Below 100 mHz the noise is dominated by temperature couplings (dark red), whereas the high-frequency noise floor is dominated by down conversions of laser frequency noise from kHz demodulation (blue). Residual seismic couplings are present at the microseism and above 7 Hz. (b) Noise budget for the BIS. Low frequency sensitivity is limited by direct temperature couplings. In the mid-frequencies, readout and thermal noises limit the device, with readout noise dominating above 0.1 Hz. The total noise is shown in solid blue. The T-240 (dark green) and GS-13 (violet) noises are shown for reference. (c) Noise budget for the tilt degree of freedom for C-6D. Classical temperature noise limits the sensitivity below 10 mHz. Above 30 mHz we sit on the horizontal motion used to subtract the low frequency tilt motion (RX - Y/g) to access the noise floor (black). The T-240 (dark green) and GS-13 (violet) noise are shown for reference.
• Figure 3: Active platform stabilization at LIGO's MIT facility. Red traces are the measured ground motion using the out-of-vacuum Streckeisen STS-2 seismometer. The blue (BISs in-loop) and black (C-6D in-loop) traces show the measured platform motion using a T-240 seismometer as a witness sensor. (a) Z degree of freedom. (b) X degree of freedom. (c) X degree of freedom.
• Figure 4: Comparison of different simulated optic motion when using the BISs (blue) and C-6D (black) in-loop to stabilize the relevant active platforms, and LPSs for suspension damping. The violet traces represent the current motion using the current control scheme with commercial seismometers and optical shadow sensors. The dotted lines represent the RMS motion. (a) Signal-recycling cavity telescope optic (SR3) longitudinal motion. (b) SR3 pitch motion. (c) Test mass pitch motion for an individual optic. (d) Beamsplitter longitudinal motion.
• Figure 5: (a) Left: fractional error on the detector-frame masses as a function of redshift for the A+ configuration with the current controls noise. Right: comparison between the A+ configuration with the current control noise (red circles) and A+ design sensitivity when the proposed upgrade is implemented (colored circles). (b) Parameter estimation comparison for an IMBH merger with source-frame masses of $300\,M_{\odot}$ and $200\,M_{\odot}$ and dimensionless spin magnitudes of 0.9 and 0.8 respectively. The binary is placed at a redshift of $2$, near the horizon of A+ detector with the developed technologies.