Rare Events Detected with a Bulk Acoustic Wave High Frequency Gravitational Wave Antenna

This work describes the operation of a high frequency gravitational wave detector based on a cryogenic bulk acoustic wave cavity and reports observation of rare events during 153 days of operation over two separate experimental runs (run 1 and run 2). In both run 1 and run 2, two modes were simultaneously monitored. Across both runs, the third overtone of the fast shear mode (3B) operating at 5.506 MHz was monitored; whereas in run 1, the second mode was chosen to be the fifth overtone of the slow shear mode (5C) operating at 8.392 MHz. However, in run 2, the second mode was selected to be closer in frequency to the first mode; and it was chosen to be the third overtone of the slow shear mode (3C) operating at 4.993 MHz. Two strong events were observed as transients responding to energy deposition within acoustic modes of the cavity. The first event occurred during run 1 on 12 May 2019 (UTC), and it was observed in the 5.506 MHz mode; whereas the second mode at 8.392 MHz observed no event. During run 2, a second event occurred on 27 November 2019 (UTC) and was observed by both modes. Timings of the events were checked against available environmental observations as well as data from other detectors. Various possibilities explaining the origins of the events are discussed.

Figure 1

Experimental setup showing BAW cavity connected to SQUID amplifier and shielding arrangement. Note that 4 and 50 K shields as well as stainless still vacuum chamber not shown.

Figure 2

Timeline of described experiment as well as histogram of total data collection at the detector output. Blue and green lines on timeline show separate data acquisition periods for two runs. Arrows point to the dates of two observed events.

Figure 3

Top figure displays averaged amplitude spectral density (ASD) of each output channel of lock-ins for longest continuous data taking run; here, each mode has been demodulated from the carrier. Bottom figure shows corresponding spectral strain sensitivity determined for each trace, as well as current best sensitivity in region given by Holometer experiment [6], which uses the cross spectral density (CSD) of two identical interferometers to search for HFGWs.

Figure 4

Time series traces for two event signals detected by system. Each plot shows two quadratures for each mode. Also shown are histograms of output magnitude samples from only the 3B mode from both the entire corresponding run (gray) and just 10 s of data around the event (black). It is clear from this plot that the overwhelming majority of non-Gaussian outliers is due to these signals.