Segmented Solenoids Lower the Cost of GUT-Scale Axion Searches

Axion searches below the microelectronvolt scale face an engineering problem as much as a physics one. The QCD axion models motivated by grand-unified-scale symmetry breaking predict very small axion-photon couplings, so conventional resonant searches push toward large magnetic volumes, high fields, and high stored energy. DMRadio-Core proposes a different magnet-and-pickup geometry: keep the high-field region narrow, segment the solenoid, and collect the axion-induced signal with resonant pickups that place much of their area in a low-field region outside the magnet coils.

The paper frames DMRadio-Core as a near-term search from 30–200 MHz, corresponding to 120–830 neV/$c^2$, and as a pathfinder for a larger DMRadio-GUT instrument covering 100 kHz–30 MHz, or 0.4–120 neV/$c^2$. The central claim is not that a new readout principle changes the axion signal model. It is that a different electromagnetic geometry can keep useful signal coupling while reducing the magnetic stored energy that has made this mass range expensive to instrument.

Core Contribution

The main idea is to decouple signal collection from the largest high-field volume. In a standard intuition, increasing axion sensitivity suggests filling a large volume with strong dc magnetic field. DMRadio-Core instead uses a segmented, narrow-bore solenoid and places most of each LC pickup in a lower-field region. The narrow pickup sections pass through the high-field bore, where the axion effective current is sourced, while the larger loop area sits outside the magnet coils and couples the signal into tunable resonators.

That trade-off matters because stored magnetic energy scales harshly with field and volume. By using soft iron to guide returning field lines and by shaping the pickup around the field geometry, the design aims to reduce stored energy without giving up the electromagnetic overlap that sets scan sensitivity.

Technical Approach

The proposed DMRadio-Core magnet reaches a 5 T peak dc field at the center of a segmented solenoid. The design uses nine coil segments, soft iron cross sections, and multiple pickups with different outer diameters. The paper reports 120 cm outer diameter pickups at the top and bottom and 146 cm pickups in the central region. The pickup set is frequency-dependent: at frequencies where a given pickup has an unfavorable scan rate, that pickup can be omitted, allowing the instrument to maintain continuous sensitivity across the target band.

Figure 4 captures the mechanism behind this geometry. The field returns through soft iron and at larger radii, creating a low-field pickup region outside the solenoid coils. The simulated axion effective current follows the dc magnetic field lines, while the induced physical current runs on the pickup walls; the induced current grows at larger radii inside the bore because each field line contributes additional current flowing outward through the structure.

The same design logic scales to DMRadio-GUT. The paper sketches an 18 T peak-field segmented solenoid with a 90 cm bore and pickups of radius 2.9 m. This is a much larger instrument, but the geometry keeps much of the resonant pickup away from the high-field region and uses soft iron to reduce the dc magnetic field where readout and tuning hardware sit.

Results and Analysis

The paper’s quantitative target is coverage rather than a completed exclusion result. DMRadio-Core is designed for 30–200 MHz, or 120–830 neV/$c^2$, while the staged DMRadio-GUT concept extends the program down to 100 kHz–30 MHz, or 0.4–120 neV/$c^2$. Those ranges matter because they overlap axion masses associated with pre-inflationary symmetry breaking and GUT-scale decay constants, where couplings are small enough that magnet cost and stored energy become limiting design constraints.

The evidence is therefore strongest at the design and simulation level. COMSOL field simulations support the magnetic-field shaping and pickup-current picture, and the paper gives concrete magnet parameters: 5 T for Core, 18 T for GUT, nine coil segments in the Core design, 90 cm bore for the GUT concept, and meter-scale pickup dimensions. The authors also specify a staged path from a near-term 30–200 MHz experiment to the broader 100 kHz–30 MHz instrument.

The practical reading is that DMRadio-Core is a serious engineering proposal for a difficult frequency range, not yet a measured dark-matter result. Its value is in showing how the sensitivity problem can be reframed as an impedance, pickup-geometry, and magnetic-energy optimization problem. The main open question is whether the simulated scan-rate advantages survive construction details: losses in large tunable resonators, tolerances around the slit and pickup surfaces, magnetic-field imperfections, vibration, cryogenic integration, and readout noise over the full operating band. If those are controlled, the design offers a credible route to testing GUT-scale axion parameter space with less stored magnetic energy than a straightforward large-bore high-field magnet would require.