Searching for Ultralight Dark Matter with Optical Cavities

We discuss the use of optical cavities as tools to search for dark matter (DM) composed of virialized ultralight fields (VULFs). Such fields could lead to oscillating fundamental constants, resulting in oscillations of the length of rigid bodies. We propose searching for these effects via differential strain measurement of rigid and suspended-mirror cavities. We estimate that more than 2 orders of magnitude of unexplored phase space for VULF DM couplings can be probed at VULF Compton frequencies in the audible range of 0.1–10 kHz.

Figure 1

Experimental setup. Light from a common laser is directed into two Fabry-Perot resonators, one with suspended mirror substrates and one with a rigid cavity spacer. Modulation in, e.g., the electron mass due to dilatonic DM at the 0.1–10 kHz frequency range results in periodic length changes in the rigid cavity, while the DM-induced length changes in the suspended interferometer are suppressed by the low frequency mechanical suspension.

Figure 2

Expected noise sources for the proposed experimental geometry, with parameters of Table 1. Upper panel: cavity with rigid spacer. Lower panel: cavity with suspended mirror substrates.

Figure 3

Strain sensitivity of optical cavity limited by thermal noise for three cavity lengths as a function of VULF Compton frequency. Here we assume a total integration time of $10^7\mathrm{s}$, with an improvement scaling with the averaging time $\tau$ as ${\tau }^{-1/2}$ up to the coherence time of the DM field ($\sim {10}^6$ oscillations), and improving as ${\tau }^{-1/4}$ thereafter. Bounds from equivalence principle tests are shown as shaded yellow region [31, 32]. Strains in the shaded green region are natural for an electron Yukawa modulus with a 10 TeV cutoff [19].

Figure 4

Search reach for an optical cavity of length 10, 30, and 100 cm assuming thermal-noise limited sensitivity along with current experimental bounds from equivalence principle (EP) and fifth force tests [31, 32] as well as the limits derived from the narrow-band AURIGA gravitational wave detector [19, 33]. Also shown are natural theoretical expectations for $d_{m_e}$ for a 10 TeV cutoff [19].