Experimental Investigation of Shielding of a Dynamic Gravitational Field

With many advances in gravitational wave research (The LIGO Scientific Collaboration et al., 2015), fully controlled laboratory experiments on dynamic gravitational fields become more and more important (Hirakawa et al., 1980; Astone et al., 1998; Ross et al., 2021, Brack et al. 2022). For static fields, it has not been possible to measure any generally accepted shielding effects for gravitational fields (Majorana, 1920; Braginsky et al., 1963; Unnikrishnan et al., 2000). To the authors’ knowledge, no corresponding experiments are known for dynamic fields, with attenuation expected to be proportional to frequency squared from theory (Weber, 1966).

Usually, dynamic experiments consist of a transmitter system, i.e. a periodically moving mass distribution, and a detector system. Two such systems have been described recently (Brack et al., 2022, 2023). In the latter, the transmitter system consists of two rotating tungsten bars. The detector consists of a high Q (10⁴), 42 Hz resonant bending beam of 1m length made of titanium. Its motion is analyzed using three calibrated laser Doppler vibrometers and multichannel lock-in amplifiers. Of paramount importance is the passive and active vibration isolation of the detector from ambient noise and crosstalk from the transmitter. The interaction is quantitatively modelled with high precision and excellent agreement between theory and experiment. The transmitters' homogeneity is characterized using neutron tomography. Here, we present a new feature of the setup, where a ~400kg metal shield (with dimensions 1.4m x 0.25m x d, where d is its thickness) is periodically lowered in between transmitter and detector. Two methods are presented to analyze the signals for the two shield positions (with/without shield): Measuring the frequency response of the detector and fitting a single degree of freedom response function or a continuous near resonance excitation with fixed frequency. Currently an upper bound for the relative change of the response signal of about 10^-3 has been established for a brass shield at a frequency of more than four orders of magnitude higher than previous work.