UC Berkeley

From astronomical observations and theories of standard cosmological models, we infer the existence of a large unobserved component of mass in the universe that is not luminous but evident through its gravitational interactions with ordinary matter. We call this unobserved component "dark matter". Various theoretical solutions to the nature of dark matter come from beyond-Standard Model theories. One possible solution to its composition is a particle called the axion. As with all candidates, direct detection of the dark matter particle will be required to confirm the theoretical hypothesis. For the axion, the principle experiments in this mass range with sensitivity to the standard axion are ADMX (formerly at LLNL, now the University of Washington), CAPP (Korea) and HAYSTAC (at Yale). These experiments, operating in the GHz range, are based on the principle that the axion may resonantly convert to a single photon in a microwave cavity permeated by a magnetic field. This modality is convenient for the (0.1-10) GHz range, roughly corresponding to the (0.4-40) eV mass range, but the axion might be much higher in mass. There is a practical limitation to searching for the axion by this technique at much higher masses, owing to the fact that much higher frequency resonators are intrinsically much smaller, and thus the conversion power would become undetectably small. A solution to achieving a usably large volume (and thus higher signal power) with a resonator of a desired high frequency (e.g., 10-100 GHz) was proposed by the Wilczek theory group at Stockholm University in 2019. This technique involves a 3D array of very thin wires, a metamaterial that mimics a plasma, the plasma frequency of which can be designed to correspond to very high frequencies. Our Berkeley group has entered a collaboration with the Stockholm group to provide experimental support to validate the fundamental theory of plasmonic resonators, to explore the most practical

ways by which such a metamaterial may be tuned in frequency, and ultimately to provide a pre-conceptual design for a full-scale experiment. This thesis primarily focused on the first two goals. Towards the first goal, a series of wire frames were produced and strung with gold-coated tungsten wires of 50m diameter, spaced by 5.38 mm, and stacked together. The microwave transmission through this stack (i.e., S21) was measured as a function of frequency to determine the plasma frequency p, for various configurations of the wire planes, and compared with the theories of wire-array metamaterials; the agreement was very good. Changing the interplane spacing was found to be the most effective way of tuning the array, and a large dynamic range, nearly a factor of 2, could be achieved by doubling the plane spacing. For the second goal, alternative configurations of wire planes were explored for which there is no currently worked-out theory (rotation of alternate planes relative to the microwave polarization axis; changing the unit cell of the wire array from rectangular to parallelogram but without changing the average density); these produced no useful effect. However, a practical design was conceived by which the spacing of a large number of planes could be accurately and smoothly accomplished, although a prototype was not attainable due to the restricted work conditions of the pandemic. A larger-scale engineering implementation was envisioned, but further work, both theoretical and experimental, remains in how to improve the quality factor of the array, as well as how it could be feasibly read out before such a design could be realized.