BRASS: Broadband Radiometric Axion SearcheS




Attempts to detect the elusive dark matter particle are now gradually shifting their focus of attention towards potential light-mass, weakly interacting sub-eV particles (WISP), including axions and hidden photons with masses well below one electron-Volt (eV).1 Recent studies have singled out the range of 10—104 meV (2.4 GHz—2.4 THz) where axions can both reproduce the dark matter and satisfy the critical constraints from cosmology, fundamental physics, and astrophysical observations. 2,3 

BRASS aims at creating a single experimental framework that can be employed to search for axion dark matter over the entire 10—104 meV range of axion mass. 

BRASS will employ novel experimental approaches and synergies between particle physics experiments and state-of-the-art broadband detection techniques developed in radio astronomy, with the University of Hamburg, the Max-Planck Institute for Radioastronomy in Bonn, the Technical University of Hamburg, and DESY Hamburg working on different aspects of the research.

During its first phase, BRASS will focus on three specific ranges of mass covering the “sweet spots” expected for the axion dark matter. Subsequent measurements will cover the entire range of mass of interest.








Experimental foundation: BRASS will use the Primakoff process in which axions can be converted into photons in presence of magnetic field. This approach was typically realized in narrowband, resonant cavity experiments such as ADMX4,5. In contrast to such experiments BRASS relies on a novel, broadband concept in which the strength of resulting photon signal is proportional to the area , A, of the surface over which the conversion occurs and the square of the magnetic field, B, parallel to the surface.6 To provide sufficient detection sensitivity, the product of B||2A ³ 100 T2 m2 has to be achieved.






Conceptual design:
To ensure reaching the necessary sensitivity, BRASS will employ multiple detection chambers, each combining a permanently magnetized surface and a secondary reflector focussing the signal on low-noise detectors located in a cooled detector room. Signals from multiple detectors will be digitized and combined in a correlator, ensuring the sensitivity provided by the sum of the areas of the magnetized surfaces of the individual detection chambers.



Basic design of BRASS


BRASS will consist of two or more detection chambers, each outfitted with a permanently magnetized surface and a secondary reflector. The axion-photon conversion will take place near the permanently magnetized surface and the resulting photonic signal will then be focused by secondary reflectors and detected in the receiver room, using interchangeable receiver modules. Signals from the individual BRASS chambers will be combined in the correlator module, increasing the measurement sensitivity and also enable constraining the flow direction of the hypothetical axion dark matter.




BRASS sensitivity: Alongside the B||2A figure of merit, sensitivity of BRASS measurements will depend on the detection limits of the receivers and only weakly on the measurement time (µ t1/4). BRASS will use state-of-the art heterodyne detectors reaching the noise temperature Tnoise » 4K and the efficiency of ~2hn/kB at frequencies up to 1 THz. The resulting expected sensitivity of BRASS is compared below to other completed and planned axion search experiments.



Sensitivity of BRASS search runs of 100 days in duration, compared to other completed (dark grey shades) and planned (light grey shades) axion search experiments. The regions of parameter space in which the QCD axion and axion-like particles (ALP) can represent the dark matter are marked with the pink band and dashed red line. The “sweet spots” for the axion DM are shown in darker pink shades. The thick dashed line shows the overall benchmark sensitivity limits of BRASS measurements, and the yellow shaded areas mark the particle mass range which will be targeted during the Phase 1 of BRASS experiments.



Components of BRASS: The magnetized surface will be made of small permanent magnets arranged in specific Halbach array7 configurations providing magnetic fields of order of ~1 T. The secondary reflectors will be manufactured to specifications limiting reflection losses to within about 0.5% at 1 THz. The receiver module will comprise low-noise detectors designed and built on the basis of the radioastronomical detectors used for APEX8 and ALMA telescopes. The correlator module will provide instantaneous spectral processing of a 16 GHz bandwidth with a fractional frequency resolution of ≤10-6, based on the digital broadband converters (DBBC3),9 Mark VI recorders, and digital FX (DiFX) correlator10 developed and used for radioastronomical measurements.


Research opportunities: Already at its preparatory stage, BRASS provides ample opportunities for engaging in research projects at the MSc and PhD level. Examples of potential research areas include:

● Two-dimensional Halbach arrays for creating large magnetized surfaces with strong and homogenous magnetic field.

● Optimization and calibration of the optical system necessary for minimizing optical losses and ensuring the required accuracy of BRASS measurements.

● Frontend design of low-noise detectors ensuring the required performance within the Phase 1 bands of BRASS.

● Broadband digitization of the signal using the DBBC3 technology, streamlining the data throughput of the experiment.

● Data processing and signal detection, optimizing the methodology for axion signal detection in broadband spectra recorded by BRASS.




1.        R. Essig, J.A. Jaros, W. Wester et al., Dark Sectors and New, Light, Weakly-Coupled Particles. SLAC: Menlo Park (2013) [arXiv:1311.0029].

2.        G. Ballesteros, J. Redondo, A. Ringwald, C. Tamarit, Phys. Rev. Lett., 118, 071802 (2017) [arXiv:1608.05414].

3.        A. Ringwald, K. Saikawa, Phys. Rev. D, 93, 085031 (2016) [arXiv:1512.06436].

4.        R. Bradley, J. Clark, D. Kinion et al., Rev. Mod. Phys., 75, 777 (2003).

5.        S.J. Asztalos, ADMX Collaboration., Phys. Rev. Lett., 104, 041301 (2010) [arXiv:0910.5914].

6.        D. Horns, J. Jaeckel, A. Lindner et al., JCAP, 4, 016 (2013) [arXiv:1212.2970].

7.        K. Halbach, Nucl.Instrum. and Methods, 169, 1 (1980).

8.        R. Güsten, L. Å. Nyman, P. Schilke et al., A&A, 454, L13 (2006).

9.        L. Vertatschitsch, R. Primiani, A. Young et al., PASP, 127, 1226 (2015).

10.     A.T. Deller, W.F. Brisken, C.J. Phillips et al., PASP, 123, 275 (2011) [arXiv:1101.0885].