SQUID-Based Microwave Cavity Search for Dark-Matter Axions

Axions in the $\mu \mathrm{eV}$ mass range are a plausible cold dark-matter candidate and may be detected by their conversion into microwave photons in a resonant cavity immersed in a static magnetic field. We report the first result from such an axion search using a superconducting first-stage amplifier (SQUID) replacing a conventional GaAs field-effect transistor amplifier. This experiment excludes KSVZ dark-matter axions with masses between $3.3\mu \mathrm{eV}$ and $3.53\mu \mathrm{eV}$ and sets the stage for a definitive axion search utilizing near quantum-limited SQUID amplifiers.

Figure 1

Schematic of the ADMX experiment. Photons, created in the cavity by the conversion of axions, are picked up by the antenna and amplified by the SQUID and HFETs. The signal is mixed in two stages, with band-limiting filtering between, to audio frequencies. The audio spectrum is measured and stored to disk. Sweep oscillator $A$ provides a reflection measurement to enable adjustment of antenna coupling; oscillator $B$, weakly coupled, allows measurement of the cavity resonant frequency.

Figure 2

Schematic of a microstrip SQUID amplifier.

Figure 3

Noise temperature of two representative SQUID amplifiers (with resonant frequency $f$) as a function of physical temperature. Dashed line indicates $T_Q$, the quantum noise temperature at $\approx 700\mathrm{MHz}$. Dotted line has a unity slope, indicating that $T_A\propto T$ in the classical regime.

Figure 4

Dark-matter axion signals simulated with Monte Carlo calculations and imposed on real data for two dark-matter axion distribution models. Example masses chosen to appear at 842.017 MHz and 842.023 MHz.

Figure 5

Axion-photon coupling excluded at the 90% confidence level assuming a local dark-matter density of $0.45\mathrm{GeV}/{\mathrm{cm}}^3$ for two dark-matter distribution models. The shaded region corresponds to the range of the axion-photon coupling models discussed in 28.