Datensatz

Zusammenfassung

The QCD Axion is an excellent dark matter candidate, while originally introduced to explain CP conservation in strong interactions. Axions could be detected using their conversion to photons at boundaries between materials of different dielectric constants in a strong magnetic field. Combin- ing many such surfaces (booster), one can enhance this conversion significantly using constructive interference and resonances (boost factor). The proposed ‘MAgnetized Disk and Mirror Axion eXperiment’ (Madmax) containing approximately 80 high dielectric disks with 1.25 m diameter in a 9 T magnetic field could probe the well-motivated mass range of 40–400 μeV, a range which is at present inaccessible by existing cavity searches. Previous studies rely on an idealized lossless 1D model of the setup and lack experimental verification. This work hence validates the concept under more realistic boundary conditions using simulations along with an experimental setup and thereby derives important implications from systematic uncertainties on the boost factor. To this end, we upgrade the previous 1D calculations to a numerically efficient three dimen- sional description of the system. We investigate diffraction, near fields as well as axion-velcity effects. We derive central implications for experimental design. We find that for the considered benchmark system the disks need a dielectric loss below tan δ ∼ 2 × 10 −4 , be tilted less than 100 μm over their diameter and have a flatness better than 2 μm root-mean-square. Moreover, crucial parameters for antenna and magnet design are deduced. We also make suggestions on how to optimize the design with respect to 3D effects to maximize sensitivity in the final setup. In addition, we explore the calibration and optimization process of the boost factor using reflectivity measurements. In order to validate the concept experimentally, we present a proof of principle booster consisting of a copper mirror and up to five sapphire disks. The mechanical accuracy, calibration of unwanted reflections and the repeatability of a basic tuning algorithm are investigated. The electromagnetic response in terms of the group delay predicted by the models is sufficiently realized in our setup. The boost factor frequency and amplitude repeatability of the tuning is at the percent level, and would have negligible impact on sensitivity. Besides, we discuss how axion haloscopes can be made directionally sensitive to the axion velocity distribution. This would not only lead strong evidence for its nature as dark matter, but also enable a new era of ‘axion astronomy’. In summary, this work lays out one of the first coherent studies of systematic effects for Madmax and design requirements therein. The presented results form the basis for a full under- standing of the Madmax booster and its corresponding systematic uncertainties, which is crucial for a successful run of the experiment. This work is also applicable to other haloscope experiments ranging from dish antennas, other dielectric haloscopes to even neutron star observations.