Dark Matter Nuclear Magnetic Resonance is Sensitive to Dark Photons and the Axion-Photon Coupling

Carl Beadle, Sebastian A. R. Ellis, Jacob M. Leedom, Nicholas L. Rodd · 2025

Read Dark Matter Nuclear Magnetic Resonance is Sensitive to Dark Photons and the Axion-Photon Coupling by Carl Beadle, Sebastian A. R. Ellis, Jacob M. Leedom, Nicholas L. Rodd (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-15897.txt.

Introduction

Carl Beadle and three co-authors show that dark matter detectors based on nuclear spins have wider reach than planned in a 2025 paper. One experiment uses a large polarized spin sample to sense axion dark matter through a tiny effective field. The authors note the same setup responds to real magnetic fields too.

A dark photon hiding inside the shielded chamber would generate exactly such a real field. An axion coupled to light would do the same through the lab's own magnet. The paper shows all three signals can be detected and told apart. At full sensitivity, the instrument would probe extremely weak dark photon mixing for free. A search built for one quarry may thus catch two others on the side.

Catalog record

CREATORCarl Beadle, Sebastian A. R. Ellis, Jacob M. Leedom, Nicholas L. Rodd
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE72 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-15897

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Dark Matter Nuclear Magnetic Resonance is Sensitive to Dark Photons and the Axion-Photon Coupling Carl Beadle, Sebastian A. R. Ellis, Jacob M. Leedom, Nicholas L. Rodd Abstract We demonstrate that nuclear magnetic resonance based searches for dark matter (DM) have intrinsic and powerful sensitivity to dark photons and the axion-photon coupling. The reason is conceptually straightforward. An instrument such as CASPEr-Gradient begins with a large sample of nuclear spins polarised in a background magnetic field. In the presence of axion DM coupled to nucleons, the spin ensemble feels an effective magnetic field 𝐁 ∝ ∇ a \mathbf{B}\propto\nabla a that tilts the spins, generating a potentially observable precession. If the magnetic field is real rather than effective, the system responds identically. A real field can be generated by a kinetically mixed dark photon within the shielded region the sample is placed or an axion coupled to photons through its interaction with the background magnetic field. We show that all three signals are detectable and distinguishable. If CASPEr-Gradient were to reach the QCD axion prediction of the axion-nucleon coupling, it would simultaneously be sensitive to kinetic mixings of ϵ ≃ 3 × 10 − 16 \epsilon\simeq 3\times 10^{-16} and axion-photon couplings of g a ​ γ ​ γ ≃ 2 × 10 − 16 ​ GeV − 1 g_{a\gamma\gamma}\simeq 2\times 10^{-16}\,{\rm GeV}^{-1} for m ≃ 1 ​ μ ​ eV m\simeq 1\,\mu{\rm eV} . Particle dark matter (DM) remains the most compelling hypothesis to explain the universe’s missing matter. However, evidence for its existence does not pinpoint a particular mass or strength of coupling to visible matter. Viable particle DM parameter space therefore spans a daunting range of masses. Historically, WIMPs and the electroweak scale have taken preeminence. More recently the community has cast a broader net, exemplified by the efforts to search for ultralight DM ( m\ll 10\,{\rm eV} ), where DM transitions to a classical wave description [ 1 ] . A question less often considered is what is the full range of physics that our nets may catch. Planned experiments aim to cover enormous swathes of unexplored ultralight DM parameter space. Unexpected signals could emerge. For example, these devices can detect relativistic signals such as a cosmic axion background [ 2 , 3 ] or gravitational waves [ 4 , 5 , 6 , 7 , 8 , 9 , 10 ] . With this Letter we instead demonstrate that even DM signals have been overlooked : devices designed to search for one type of wave DM can be strongly sensitive to another. More to the point, ongoing efforts to detect axion DM coupled to nucleons with nuclear magnetic resonance (NMR), for example by CASPEr-Gradient [ 11 , 12 ] , would be highly sensitive to the axion-photon coupling and dark photon DM, as is the case in other torque based searches [ 13 ] . Given our ignorance as to the nature of DM, it stands that these situations should be highlighted whenever possible. To justify our claim we recall the basics of axion NMR. Consider a sample of nuclear spins aligned in a background magnetic field \mathbf{B}_{0} . In conventional NMR, one fires an electromagnetic wave through the sample that generates an interaction according to {\cal H}_{B}=-\gamma\mathbf{B}\cdot\mathbf{S} , with \gamma and \mathbf{S} the nucleon gyromagnetic ratio and spin operator. Consequently, the spins are tipped into the plane transverse to \mathbf{B}_{0} and begin to precess at the Larmor frequency, \omega_{0}=\gamma B_{0} , yielding an oscillatory signal in the transverse plane. CASPEr-Gradient exploits the same physical principle. If the spins are immersed in an axion DM wave, then the axion-nucleon coupling, {\cal L}_{a{\scriptscriptstyle N}}=g_{{\scriptscriptstyle N}}(\partial_{\mu}a)\bar{N}\gamma^{\mu}\gamma_{5}N , generates an interaction {\cal H}_{a{\scriptscriptstyle N}}=-2g_{{\scriptscriptstyle N}}\nabla a\cdot\mathbf{S} . By analogy to {\cal H}_{B} the axion induces an effective magnetic field of the form \mathbf{B}_{a{\scriptscriptstyle N}}=g_{{\scriptscriptstyle N}}(2/\gamma)\nabla a . The axion field has an identical effect: DM tips the spins, which then precess leading to a detectable signal. There is a resonant enhancement when \omega_{0}\simeq m and therefore one performs a resonant search by varying B_{0} . Figure 1: Projected reach of DM NMR to dark photons (left) and axions (right). Both searches can be performed simultaneously along with a search for the axion nucleon coupling. The purple and blue lines correspond to a sample constituted of xenon-129 ( \gamma=12 MHz/T, n=1.3\times 10^{22} cm -3 , H_{\text{max}}=10 T) and helium-3 ( \gamma=32 MHz/T, n=2.8\times 10^{22} cm -3 , H_{\text{max}}=20 T). Solid curves assume T_{2}=100 s, whereas dashed curves assume a line width limited by a part-per-million. A search capable of reaching the QCD axion prediction for g_{{\scriptscriptstyle N}} would have the sensitivity shown in red [ 14 ] . The reach assumes a single TM mode with the sample placed at the optimal radius of r\simeq 0.77R . See the text for further details. We can now explain our core insight: if DM generates a real rather than effective magnetic field, CASPEr-Gradient can detect it. Consider first dark photon DM, \mathbf{A}^{\prime} , that mixes with the visible photon via a kinetic mixing parameter \epsilon [ 15 ] . To mitigate the impact of stray magnetic fields, the spin sample must be placed within a shielded region of geometric size L . The dark photon will penetrate that shield and generate a magnetic field \mathbf{B}_{A^{\prime}}\sim\epsilon m^{2}L\mathbf{A}^{\prime} [ 16 ] . Setting B_{aN}\sim B_{A^{\prime}} , the qualitative dark-photon sensitivity in terms of the axion-nucleon coupling is \epsilon\sim g_{{\scriptscriptstyle N}}v/\gamma mL , with v\sim 10^{-3} the local DM speed. We can rephrase the sensitivity as follows. Writing g_{{\scriptscriptstyle N}}=c_{{\scriptscriptstyle N}}m/m_{\pi}f_{\pi} , which depends on the pion mass and decay constant, we have \displaystyle\epsilon\sim 10^{-16}\,c_{{\scriptscriptstyle N}}\left(\gamma_{\rm He}/\gamma\right)\left(10\,{\rm cm}/L\right)\!, (1) assuming a helium-3 sample. Although an ambitious target, the ultimate goal of all axion experiments is QCD sensitivity, where c_{{\scriptscriptstyle N}}\sim 1 . Note the above scaling is distinct from ideas to repurpose axion-photon searches for the dark photon, as in e.g. Ref. [ 17 ] . In those cases the dark photon sensitivity does not gain from the large magnetic field volume that is maximised for the axion sensitivity. Here, however, the scaling of both the axion and dark photon signal is controlled by the density of spins, or effectively the background magnetization. A similar argument applies to the axion-photon coupling, {\cal L}_{a\gamma\gamma}=-\tfrac{1}{4}g_{a\gamma\gamma}aF\widetilde{F} , where axion DM couples to \mathbf{B}_{0} and induces B_{a\gamma}\sim g_{a\gamma\gamma}B_{0}L(\partial_{t}a) , where now L is the scale over which B_{0} is generated. The expected sensitivity from B_{aN}\sim B_{a\gamma} is g_{a\gamma\gamma}\sim g_{N}v/\gamma B_{0}L , or taking g_{a\gamma\gamma}\sim c_{\gamma}\alpha_{{\scriptscriptstyle{\rm EM}}}m/2\pi m_{\pi}f_{\pi} , \displaystyle c_{\gamma}\sim c_{{\scriptscriptstyle N}}\left(10\,{\rm T}/B_{0}\right)\left(\gamma_{\rm He}/\gamma\right)\left(10\,{\rm cm}/L\right)\!. (2) This result demonstrates that at the highest frequencies – recall B_{0} is tuned to adjust the resonant frequency – an instrument sensitive to the QCD axion value of c_{{\scriptscriptstyle N}} would also be sensitive to the QCD axion prediction of c_{\gamma}\sim 1 . In what follows we refine the qualitative results of Eqs. ( 1 ) and ( 2 ) culminating in the projections shown in Fig. 1 . These results show the reach for several sensitivity estimates for CASPEr-Gradient that exist in the literature, specifically Refs. [ 18 , 14 ] , each shown for two different target nuclei, xenon-129 and helium-3. The results

Read the complete text