betaFeedback

WorksPhysics and Astronomy

Constraints on millicharged particles from nuclear gamma-decays

Abstract

We consider nuclear gamma decays and γγ-emitting reactions that can be an efficient source of hypothetical millicharged particles (χχ). In particular, we revisit the production of millicharged particles in nuclear reactor environment, pointing out that γγ cascades from 239^{239}U is an overlooked yet a powerful source of χχˉχ\barχ pairs. This leads to an increased flux compared to previous studies. We then apply new estimates of the flux to derive novel limits on the value of millicharge, ε=Qχ/e\varepsilon = Q_χ/e, from the electron recoil searched for in a variety of experiments placed in proximity to the reactor cores. The derived limits on ε\varepsilon are the strongest in the interval of masses ∼0.7−2\sim 0.7-2 MeV. We also derive the MCP flux from the Sun and point out potential sensitivity of the low-threshold dark matter search experiments.

Introduction

While Standard Model (SM) of particles and fields has been extremely successful in describing observable world, it cannot be a “final theory”. Notable exceptions include neutrino physics (mixing and oscillations), as well as cold dark matter (DM) that must be associated with new degrees of freedom. Generalization of existing ideas about neutrino masses, as well as particle dark matter leads to the notion of dark sectors (DS). DS constructions are not tied to any specific energy scale, and can be indeed much lighter than the scale of weak interactions. Light DS, by necessity, would have to have states with rather small couplings to the SM states, and their search would typically require high-intensity experiments [1, 2]. In light of this, all possible connections (or portals) between light DS and SM have been described, classified and searched for in ever increasing number of high-intensity experiments. One unique portal to DS is via the so-called millicharged particles (MCP).

In this paper, we will revisit some aspects of searches for the MCP χ\chi that we will assume to be a fermion state χ\chi with an electric charge much smaller than the proton charge, Qχ≡εe≪eQ_{\chi} \equiv\varepsilon e \ll e. We note that while such construction may look totally artificial, there are theoretically natural ways of ascribing the smallness of ε\varepsilon to a small amount of mixing between regular and “dark” photons [3]. There has been a steady interest in phenomenology of millicharged particles, not least because they are capable of inducing a variety of interesting phenomena. MCPs can bring change to the early Universe cosmology [4, 5, 6]; they can be produced and accelerated by supernovae [7, 8, 9]; their population on Earth can be enhanced due to possibly small collision lengths within Earth’s material [10, 11].

Several years ago, MCP dark matter was invoked as a possible explanation of the abnormal hydrogen spin temperature signal observed by the EDGES collaboration [12] that consists in stronger-than-expected 21 cm absorption feature at redshifts of z∼O(15−20)z \sim O(15 - 20). An explanation of such phenomenon may reside in late re-coupling of ordinary baryonic matter and dark matter, and that can naturally occur if DM has a small millicharge [13, 14, 15]. These scenarios, in particular, prefer relatively light, sub-100-MeV MCP particles with ε∼O(10−5)\varepsilon\sim O(10^{-5}).

All existing MCP limits and proposals for future searches can be subdivided into two categories: those that exploit production and detection of MCPs and those that search/constrain DM-style MCP relics. Any signal in the first method would typically scale as the fourth power of ∼ε\sim\varepsilon, while the second class of searches is ∼ε2\sim\varepsilon^{2}, but is more model-dependent3. We will concentrate on the first method that involves the production and detection of the MCPs, and limit our study to relatively light, sub-10-MeV particles. Specifically, the MCPs were searched for in the electron beam dump experiments [17] that provide the most direct limits in the sub-100-MeV range. Similar style limits can be derived from the past proton beam dump/neutrino experiments [18], and new dedicated experiments placed along the line of sight of various particle beams [19, 20, 21, 22, 23, 24, 25]. Moreover, cosmic-ray-initiated production of MCPs can be used to derive equally stringent limits on ε\varepsilon [26, 27, 28, 29].

If masses of DS particles are light, below 10 MeV, one must anticipate that the strongest direct constraints may come from the particle physics experiments in direct proximity of nuclear reactors. This is because commercial nuclear reactors can be viewed as very powerful beam dump experiments, where nuclear fission generates copious amounts of νˉ\bar{\nu}, γ\gamma, nn radiation with O(MeV)O(\mathrm{MeV}) energy spectra. Recent years have seen the proliferation of low-threshold detectors installed in O(30 m)O(30\,\mathrm{m}) distance from cores of nuclear reactors with the main goal of detecting coherent neutrino-nucleus recoil. Low threshold achieved in these experiments also help to set stringent constraints on MCPs, as typical cross section for detection scales inversely proportional with the kinetic energy of the outgoing electron, Erecoil−1E_{\mathrm{recoil}}^{-1}. Several collaborations have placed limits on MCPs using such devices [30, 31]. Previous searches of the neutrino magnetic moments as well as actual measurements of the νˉe\bar{\nu}_{e} scattering can also be used to be recast as limits on the MCPs. The intense photon flux in nuclear reactors also provide an efficient environment to probe other DS particles, such as axion-like particles, as has been done in [32, 33, 34].

Several experimental studies [30, 31] have recognized the potential of nuclear reactors and the enormous photon flux that they generate as an important source of MCPs. They derived stringent limits on such particles already. Our present study is motivated by the observation that existing treatments for MCPs [30, 31] seem to have rather limited reach in terms of mχm_{\chi} stemming from the fact that only e+γ→e+χˉ+χe+\gamma\to e+\bar{\chi}+\chi reactions are included as the source of MCPs. Yet it is clear that any nuclear process resulting in emission of a gamma quantum of energy ω\omega can in principle emit a charged pair [35]. We adopt this production mechanism for nuclear reactors, following recent treatment of pair emission in Ref. [36]. We find that the stringent limits can be extended to higher masses.

While nuclear reactors provide the largest fluxes of MCPs for the closely located detectors, we note that there is also an ambient flux of MCP created by natural radioactivity (mostly caused by the radioactive isotopes of K, U and Th). While fluxes of MCP created this way are going to be much lower than those in proximity of reactor cores, the detection methods that can be employed could involve neutrino detectors, and large self-shielding dark matter detectors (such as those based on xenon dual-TPC technology). These are large deep-underground detectors operating in a maximally background-free environment, capable of achieving much lower counting rates than detectors placed in reactors’ proximity. Similarly, the MCPs can be created as a byproduct of nuclear reactions in the Sun. Here, however, the value of ε\varepsilon must be much smaller to avoid the slow-down and thermalization of MCPs by the solar matter.

Along with the new estimates of fluxes from MCPs, we also calculate the production of massive dark photons. If the mass is above 2me2m_e, the dark photons are unstable against the decays to e+e−e^{+}e^{-}. If kinematically allowed, they can be produced in nuclear reactors, and possibly detected by e.g. experiments studying neutrino-electron scattering.

The rest of our paper is organized as follows: § 2 of this work describes the production of MCPs accompanying nuclear decays. § 3 treats different experiments and derives corresponding limits on mχ−εm_\chi-\varepsilon plane. § 4 estimates the fluxes and detection rates of MCPs from natural radioactivity and solar production. § 5 addresses limits on massive dark photons. We reach our conclusions in § 6.

Pair-production from nuclear de-excitation

Any process that results in the emission of an on-shell γ\gamma of frequency ω\omega may also produce a pair of MCP particles with 2mχ<ω2m_\chi<\omega. For nuclear-initiated emission, the recoil is negligible, and therefore the entire energy of a photon can go into the MCP pair, so that the kinematic limit for mχm_\chi is close to ω/2\omega/2. For photons emitted from the electron lines, in contrast, the kinematic limit is lower due to the electron recoil. In previous analyses [30, 31], the main source of the MCP was assumed to originate from e+γ→e+χ+χˉe+\gamma\to e+\chi+\bar{\chi} reaction. This results in a sharp drop of the χ\chi flux for mχ>0.5m_\chi>0.5 MeV. We note, however, that the spectrum of photons emitted in nuclear fission is extended into a several MeV range, and therefore higher mχm_\chi than those considered in Refs. [30, 31] can be emitted. The photons can be emitted in the de-excitations of the fission products, and in the process of the subsequent capture of free neutrons. As ∼60%\sim60\% of the neutrons produced in fission undergo mostly (n,γ)(n,\gamma) reactions on protons or on heavier elements, there is a plenty of MCP sources that should be added to the production mechanisms discussed in [30, 31].

In this section, we calculate the rate of χχˉ\chi\bar{\chi} pair production from neutron capture in the core region of the nuclear reactor. Each fission in the reactor releases an average of 200 MeV energy [37] and 2.5 neutrons [38], which corresponds to the neutron yield per second as

Nn(s−1)=7.8×1019×Power⁡(GW).(1)N_n\left(\mathrm{s}^{-1}\right)=7.8\times10^{19}\times\operatorname{Power}(\mathrm{GW}). \tag*{(1)}

Following their emission, neutrons undergo relatively rapid thermalization before either contributing to the next fission cycle or being captured by 238U{}^{238}\mathrm{U}, by protons in natural-water reactors, or by other elements. A large fraction of the neutron capture events leads to the emission of γ\gamma-rays, and, in the presence of millicharged particles, off-shell γ∗\gamma^{*} which decays into a MCP pair. Following the procedure in [36], we first calculate the cross section ratio between these two processes

r:=σ(n+N→N′+χ+χˉ)σ(n+N→N′+γ),(2)r:=\frac{\sigma\left(n+N\to N^{\prime}+\chi+\bar{\chi}\right)}{\sigma\left(n+N\to N^{\prime}+\gamma\right)}, \tag*{(2)}

and then use the well-studied γ\gamma-ray emission branching ratios to calculate the MCP pair production. The key assumption here is that the nuclear matrix elements for photon emission do not change in a significant way when the photon is taken off-shell. This is ultimately related to the small size of atomic nuclei compared to wavelengths of emitted photons and MCPs.

After a multipole expansion of the nuclear matrix element, the differential ratios for electric dipole (E1) and magnetic dipole (M1) transitions are given as

drE(B)dEχdEN′=M8π2ωRE(B),(3)\frac{d r_{\mathcal{E}(\mathcal{B})}}{d E_{\chi} d E_{N'}} = \frac{M}{8\pi^{2}\omega}\mathcal{R}_{\mathcal{E}(\mathcal{B})}, \tag*{(3)}
RE=4παε2ω21k4[(Eχ+Eχˉ)2(4mχ2+2k2)−4k2EχEχˉ+k4],(4)\mathcal{R}_{E} = \frac{4\pi\alpha\varepsilon^{2}}{\omega^{2}}\frac{1}{k^{4}}\left[(E_{\chi}+E_{\bar{\chi}})^{2}(4m_{\chi}^{2}+2k^{2})-4k^{2}E_{\chi}E_{\bar{\chi}}+k^{4}\right], \tag*{(4)}
RB=4παε2ω21k4{4mχ2[(Eχ+Eχˉ)2−k2]−k4+2k2(Eχ2+Eχˉ2)}.(5)\mathcal{R}_{B} = \frac{4\pi\alpha\varepsilon^{2}}{\omega^{2}}\frac{1}{k^{4}}\left\{4m_{\chi}^{2}\left[(E_{\chi}+E_{\bar{\chi}})^{2}-k^{2}\right]-k^{4}+2k^{2}\left(E_{\chi}^{2}+E_{\bar{\chi}}^{2}\right)\right\}. \tag*{(5)}

Where MM is the mass of the excited state formed by neutron capture, ω=M−MN′\omega=M-M_{N'} is the available energy, k=p1+p2k=p_{1}+p_{2} is the total momentum of the millicharged particle pair, and RE(B)\mathcal{R}_{E(B)} is the ratio between the angular-averaged pair production and photon emission amplitudes for E1(M1) transition. We assume recoil-less processes, i.e. taking the leading order expansion in 1/M1/M in the above and in the following discussion. The kinematical limits of EN′E_{N'} are determined by having χ\chi and χˉ\bar{\chi} moving either parallel or antiparallel to each other. After integrating over dEN′dE_{N'}, we get the differential ratios as functions of EχE_{\chi}, which is convenient for the discussion of detection rates:

drEdEχ=αε22πω3[2∣pχ∣∣pχˉ∣+2mχ2ω2∣pχ∣∣pχˉ∣(EχEχˉ+mχ2+∣pχ∣∣pχˉ∣)(EχEχˉ+mχ2−∣pχ∣∣pχˉ∣)+(Eχ2+Eχˉ2)ln⁡EχEχˉ+mχ2+∣pχ∣∣pχˉ∣EχEχˉ+mχ2−∣pχ∣∣pχˉ∣],(6)\frac{d r_{E}}{d E_{\chi}}=\frac{\alpha\varepsilon^{2}}{2\pi\omega^{3}}\left[2\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert+\frac{2m_{\chi}^{2}\omega^{2}\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert}{(E_{\chi}E_{\bar{\chi}}+m_{\chi}^{2}+\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert)(E_{\chi}E_{\bar{\chi}}+m_{\chi}^{2}-\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert)}+(E_{\chi}^{2}+E_{\bar{\chi}}^{2})\ln\frac{E_{\chi}E_{\bar{\chi}}+m_{\chi}^{2}+\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert}{E_{\chi}E_{\bar{\chi}}+m_{\chi}^{2}-\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert}\right], \tag*{(6)}
drBdEχ=αε22πω3[−2∣pχ∣∣pχˉ∣+2mχ2ω2∣pχ∣∣pχˉ∣(EχEχˉ+mχ2+∣pχ∣∣pχˉ∣)(EχEχˉ+mχ2−∣pχ∣∣pχˉ∣)+(Eχ2+Eχˉ2−2mχ2)ln⁡EχEχˉ+mχ2+∣pχ∣∣pχˉ∣EχEχˉ+mχ2−∣pχ∣∣pχˉ∣].(7)\frac{d r_{B}}{d E_{\chi}}=\frac{\alpha\varepsilon^{2}}{2\pi\omega^{3}}\left[-2\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert+\frac{2m_{\chi}^{2}\omega^{2}\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert}{(E_{\chi}E_{\bar{\chi}}+m_{\chi}^{2}+\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert)(E_{\chi}E_{\bar{\chi}}+m_{\chi}^{2}-\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert)}+(E_{\chi}^{2}+E_{\bar{\chi}}^{2}-2m_{\chi}^{2})\ln\frac{E_{\chi}E_{\bar{\chi}}+m_{\chi}^{2}+\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert}{E_{\chi}E_{\bar{\chi}}+m_{\chi}^{2}-\lvert p_{\chi}\rvert\lvert p_{\bar{\chi}}\rvert}\right]. \tag*{(7)}

where

Eχˉ=ω−Eχ,∣pχ∣=Eχ2−mχ2,∣pχˉ∣=(ω−Eχ)2−mχ2.(8)E_{\bar{\chi}}=\omega-E_{\chi},\qquad\lvert p_{\chi}\rvert=\sqrt{E_{\chi}^{2}-m_{\chi}^{2}},\qquad\lvert p_{\bar{\chi}}\rvert=\sqrt{(\omega-E_{\chi})^{2}-m_{\chi}^{2}}. \tag*{(8)}

These results are plotted in Figure 1.

Differential ratios for E1 (1a) and M1 (1b) transitions

Figure 1. Differential ratios for E1 (1a) and M1(1b) transitions with ε=1\varepsilon= 1 for different values of mχ/ωm_\chi/\omega. Quantities with a bar on top are normalized by ω\omega for universality: mˉχ=mχ/ω\bar{m}_\chi= m_\chi/\omega, Eˉχ=Eχ/ω\bar{E}_\chi= E_\chi/\omega. Notice that when mχ≪ω/2m_\chi\ll\omega/2, the E1 and M1 yields of MCP are similar, while for heavier MCP (relative to ω/2\omega/2), the M1 yields are more suppressed, as expected.

The type of neutron capture interactions along with their yield per fission are listed in [38]4, and the emission lines resulting from neutron capture are documented in the ENSDF database [39]. The two most important reactions we consider is p(n,γ)2Hp(n,\gamma){}^{2}\mathrm{H} and 238U(n,γ)239U{}^{238}\mathrm{U}(n,\gamma){}^{239}\mathrm{U}. With the goal of producing ∼MeV\sim\mathrm{MeV} MCP pairs, we include the 3.297MeV and 4.060MeV E1 transitions from the decay of the 239U{}^{239}\mathrm{U} state at the threshold of 238U+n{}^{238}\mathrm{U}+n and the 2.223MeV M1 transition from 2H{}^{2}\mathrm{H}. The number of γ\gamma emission events per second for each emission line is given as

Nγ(N,ω)=Nn×Yn(N)∑Yn×Iγ(N,ω)100,(9)N_{\gamma}^{(N,\omega)}=N_{n}\times\frac{Y_{n}^{(N)}}{\sum Y_{n}}\times\frac{I_{\gamma}^{(N,\omega)}}{100}, \tag*{(9)}

where Yn(N)Y_{n}^{(N)} is the neutron capture yield on isotope NN per fission, and Iγ(N,ω)I_{\gamma}^{(N,\omega)} is the number of γ\gamma with energy ω\omega emitted per 100 neutron captures on the isotope NN.

The differential flux of millicharged particles is then found by

dϕχdEχ=24πD2∑N,ωdrE(B)dEχNγ(N,ω),(10)\frac{d\phi_{\chi}}{dE_{\chi}}=\frac{2}{4\pi D^{2}}\sum_{N,\omega}\frac{d r_{E(B)}}{dE_{\chi}}N_{\gamma}^{(N,\omega)}, \tag*{(10)}

where the factor of 2 comes from counting both χ\chi and χˉ\bar{\chi}, and DD is the distance from the reactor core to the detector. The summation runs over the aforementioned transitions, and for both transitions of 239U{}^{239}\mathrm{U} we use rEr_E, while for the 2H{}^{2}\mathrm{H} transition we take rBr_B.

Millicharged particle detection via atomic ionization

The MCPs entering the detector can interact electromagnetically with the detection material. If the energy exchange is comparable to the binding energy of atoms, an electron could be released from the atom, producing a detectable signal. The use of this atomic ionization process in millicharged particle searches as well as methods to calculate the cross section are discussed in [30, 31, 40]. In the asymptotic regime, when the binding energy and the momentum of the initial electron can be neglected, the cross section for a MCP to transfer at least an energy of T0T_0 to the electron is given by the free electron approximation (FEA), σ(T>T0)≃ε2×2.6×10−25 cm2×(MeV/T0)\sigma(T > T_0) \simeq\varepsilon^2 \times2.6 \times10^{-25}\,\mathrm{cm}^2 \times(\mathrm{MeV}/T_0) [18]. For the range of mχm_\chi and EχE_\chi of our interest, the Photo Absorption Ionization model (PAI) is more realistic as it accounts for the asymptotically FEA behavior when the energy transfer is significantly larger than the electron binding energy, as well as for the deviation from FEA when these two energy scales become close. Thus, we adopt PAI model where the differential cross section per atom is given by

dσdT=ε2αβ2πσγ(T)Tln⁡[(1−β2ϵ1)2+β4ϵ22]−1/2+Zε2αβ2πNe(β2−ϵ1∣ϵ∣2)Θ+ε2αβ2πσγ(T)Tln⁡(2meβ2T)+ε2αβ2π1T2∫0Tσγ(T′)dT′,(11)\begin{aligned} \frac{d\sigma}{dT} ={}& \frac{\varepsilon^2\alpha}{\beta^2\pi}\frac{\sigma_\gamma(T)}{T}\ln\left[\left(1-\beta^2\epsilon_1\right)^2+\beta^4\epsilon_2^2\right]^{-1/2} + \frac{Z\varepsilon^2\alpha}{\beta^2\pi N_e}\left(\beta^2-\frac{\epsilon_1}{|\epsilon|^2}\right)\Theta\\ &+ \frac{\varepsilon^2\alpha}{\beta^2\pi}\frac{\sigma_\gamma(T)}{T}\ln\left(\frac{2m_e\beta^2}{T}\right) + \frac{\varepsilon^2\alpha}{\beta^2\pi}\frac{1}{T^2}\int_0^T \sigma_\gamma(T')dT', \tag*{(11)} \end{aligned}

In these expressions, TT is the energy transfer, σγ\sigma_\gamma is the photo-absorption cross section, β\beta is the speed of χ\chi, ϵ1\epsilon_1 and ϵ2\epsilon_2 are the real and imaginary part of the dielectric constant (not to be confused with the value of the millicharge, ε\varepsilon), NeN_e is the electron number density, and Θ=arg⁡(1−ϵ1β2+iϵ2β2)\Theta= \arg\left(1-\epsilon_1\beta^2+i\epsilon_2\beta^2\right). Details of the model can be found in [41]. A comparison between the PAI model and other models is shown in Figure 2.

Atomic ionization differential cross sections for germanium crystal under Photo Absorption Ionization model (PAI), free electron approximation (FEA), and equivalent photon approximation (EPA)

Figure 2. Atomic ionization differential cross sections for germanium crystal under Photo Absorption Ionization model (PAI), free electron approximation (FEA), and equivalent photon approximation (EPA) for mχ=1 keVm_\chi= 1\,\mathrm{keV}, Eχ=1 MeVE_\chi= 1\,\mathrm{MeV} (2a) and mχ=1 MeVm_\chi= 1\,\mathrm{MeV}, Eχ=1.5 MeVE_\chi= 1.5\,\mathrm{MeV} (2b). The photo-absorption cross section is obtained from [42]. When the millicharged particle is ultra-relativistic, PAI agrees with the maximum between FEA and EPA as expected, when EχE_\chi and mχm_\chi are relatively close, EPA ceases to be a good approximation, while PAI continues to capture the enhancement at low energy transfer and the free electron behavior at high energy transfer.

The differential count rate is then determined by the MCP flux and the atomic ionization cross section:

dRdT=ρA∫Eχ,min⁡Eχ,max⁡dσdTdϕχdEχdEχ,(12)\frac{dR}{dT} = \rho_A \int_{E_{\chi,\min}}^{E_{\chi,\max}} \frac{d\sigma}{dT}\frac{d\phi_\chi}{dE_\chi}dE_\chi, \tag*{(12)}

where ρA\rho_A is the number of atoms per unit mass. To set the constraint on the millicharge we compare the differential count rate in this work with the experiment and analysis performed by the TEXONO collaboration [30], which utilizes a PCGe detector with an energy threshold of 300 eV and a mass of 500 g located at 28 m away from a 2.9 GW thermal power nuclear reactor. The limiting value for ε\varepsilon is determined when the counting rate (12) is matched to the experimental rate in [30]. An example of this type of comparison is shown in Figure 3a. Note that the millicharge is proportional to the fourth root of the count rate, therefore a small uncertainty in the comparison will only lead to a tiny change in the constraint. The constraint we get and comparison with other work at the relevant mass range is shown in Figure 3b. We observe that the inclusion of nuclear transition enhances the sensitivity to ε\varepsilon, and indeed in the range of 0.7 − 2 MeV produces the strongest constraint to date.

Comparison of the TEXONO analysis with this work and constraints on MCPs

Figure 3. (3a): An example of the comparison between the analysis from TEXONO (blue dots) and this work (purple line) with mχ=1m_\chi= 1 MeV, ε=2.7×10−5\varepsilon= 2.7 \times10^{-5}. The constraint on the millicharge is obtained by requiring that the differential counting rate does not exceed the 2σ2\sigma region of the TEXONO analysis. Near T=1.4T = 1.4 keV, the energy transfer becomes enough to ionize electrons in the L shell of Ge (as documented in [42]) and thus produces a small peak in the rate. (3b): The constraint we get on MCPs based on comparing the MCP flux generated by nuclear de-excitation process with the TEXONO analysis [30], in comparison to the constraints from TEXONO [30], CONNIE and Atucha-II [31], and SLAC [17]. Our result, based on theoretical re-evaluation of the MCP flux in conjunction with past experimental searches, provides the strongest constraint in the mass range 0.7−20.7 - 2 MeV.

We discuss in passing a number of other near-reactor experiments with different thresholds and counting rates. The CONNIE and Atucha-II experiments [31] utilize Skipper-CCDs, allowing the detection of particles interacting with bulk plasmons at the eV energy scale. For MCP pair-production from photons scattering with electrons, their result shows a better sensitivity compared with TEXONO. On the other hand, for the pair-production from nuclear de-excitation, our calculation shows TEXONO has a slightly better sensitivity, but the details depend on a precise knowledge of the interaction cross-section at low energy transfer. The MUNU experiment [43] tracks electron recoils above 0.7 MeV with a differential counting rate of ∼10−1 MeV−1kg−1day−1\sim10^{-1}\,\mathrm{MeV}^{-1}\mathrm{kg}^{-1}\mathrm{day}^{-1} from the hypothetical neutrino magnetic moment, which could as well be from MCPs. Our estimate shows the sensitivity to ε4\varepsilon^4 from the MUNU experiment on millicharged particles is a factor of a few hundred weaker than the TEXONO experiment, mostly because of 1/T1/T scaling of the cross section. The TEXONO νˉe\bar{\nu}_e-electron scattering experiment [44] could also in principle receive signals from MCPs, although at a threshold of T>3T > 3 MeV both the allowed MCP mass range and the available phase space become limited. Finally, this year, the CONUS+ experiment has achieved very low counting rates and very low thresholds, finally detecting the neutrino-nucleus coherent scattering for reactor neutrinos [45]. These results could potentially be also recast as constraints on MCPs.

MCPs from naturally occurring decays and reactions

Away from the nuclear reactors, excited nuclei are naturally produced from the decay chain of natural radioisotopes. This will result in a geo-background of MCPs detectable by large deep-underground detectors including Borexino [46], XENON [47], LZ [48], and PANDAX [49]. Also, pppp and CNO reaction chains in the Sun produce photons in the 0.50.5 MeV-7.57.5 MeV range which should also contribute to the natural MCP abundance.

The main γ\gamma emission channels from natural terrestrial radioactivity with energy above 11 MeV include a number of emission lines in the de-excitation of 214Po{}^{214}\mathrm{Po} in the 238U{}^{238}\mathrm{U} chain, 208Pb{}^{208}\mathrm{Pb} in the 232Th{}^{232}\mathrm{Th} chain, and 40Ar{}^{40}\mathrm{Ar} in the 40K{}^{40}\mathrm{K} decay. Especially promising are the 2.6152.615 MeV E3 emission line from the de-excitation of 208Pb{}^{208}\mathrm{Pb} occurring at the highest energy, and the 1.4611.461 MeV E2 emission line in 40K{}^{40}\mathrm{K} decay accompanied by the highest component of the geo-neutrino flux [50].5

To estimate the MCP flux, we focus on the XENONnT experiment [47] and normalize the MCP flux on the geo-neutrino flux from the same decay chain measured by the Borexino experiment [51] at the same site. Different from neutrinos which travel through the Earth almost freely, MCPs lose energy as they travel through the Earth as a result of the electromagnetic interaction with the surrounding material, based on the review in [52], the energy loss rate for typical values of βγ\beta\gamma under consideration for MCPs travelling through the Crust is estimated to be

dEdx=−(0.003−0.03) MeV100 km(ε10−5)2(13)\frac{dE}{dx}=-\frac{(0.003-0.03)\,\mathrm{MeV}}{100\,\mathrm{km}}\left(\frac{\varepsilon}{10^{-5}}\right)^2 \tag*{(13)}

therefore, for the values of ε\varepsilon of our interest, it is appropriate to focus on the MCPs produced from the Local Crust (LOC, within ∼300 km\sim300\,\mathrm{km} from the detector, see [50] for details), where the energy loss of MCPs can be neglected. Based on the geo-neutrino measurement available at Borexino, we focus specifically on the MCP pair production accompanying the 2.615 MeV2.615\,\mathrm{MeV} emission line in the 232Th{}^{232}\mathrm{Th} chain. The geo-MCP flux is then determined by

dϕχdEχ=26drE3dEχϕν(232Th,LOC)(14)\frac{d\phi_{\chi}}{dE_{\chi}}=\frac{2}{6}\frac{d r_{\mathrm{E3}}}{dE_{\chi}}\phi_{\nu}({}^{232}\mathrm{Th},\mathrm{LOC}) \tag*{(14)}

where the prefactor 2/62/6 comes from one 208Pb{}^{208}\mathrm{Pb} de-excitation per six neutrino emission in the 232Th{}^{232}\mathrm{Th} chain and two MCPs per pair production; drE3/dEχd r_{\mathrm{E3}}/dE_{\chi} is the ratio between MCP pair production events and γ\gamma emission events for E3 multipole transition, with the explicit form derived in Appendix A. Finally, ϕν(232Th,LOC)\phi_{\nu}({}^{232}\mathrm{Th},\mathrm{LOC}) denotes the neutrino flux from 232Th{}^{232}\mathrm{Th} decay in the Local Crust. Assuming that the 238U/232Th{}^{238}\mathrm{U}/{}^{232}\mathrm{Th} abundance ratio is independent of the material considered, ϕν(232Th,LOC)\phi_{\nu}({}^{232}\mathrm{Th},\mathrm{LOC}) corresponds to approximately ∼10%\sim10\% of the total geo-neutrino flux measured by the Borexino experiment.

Under the above assumption, the constraint we get is ∼ε<10−4\sim\varepsilon< 10^{-4} at mχ∼1 MeVm_{\chi}\sim1\,\mathrm{MeV}, however, at this value of ε\varepsilon the energy loss in the Local Crust becomes non-eligible, therefore, we were not able to set a solid independent bound on ε\varepsilon based on the XENONnT experiment. We also considered the Borexino experiment which has a larger detector volume, but due to the higher detection threshold the MCP slow-down is even more crucial, and the conclusion is similar.

It is well appreciated that the total solar neutrino flux dominates over geo-neutrino νˉe\bar{\nu}_e flux by ∼\sim four orders of magnitude. Using well-understood properties of the solar energy generation cycles, we can easily calculate the MCP production rate. Of course, for the O(10−5)O(10^{-5}) values of the millicharge, the χ\chi particles produced in the core of the Sun will be thermalized and trapped in the solar interior for considerable time. The main processes that will create χ\chi's are listed below:

e++e−→χ+χˉ;mχmax⁡=0.511 MeV(15)e^{+}+e^{-}\rightarrow\chi+\bar{\chi};\quad m_{\chi}^{\max}=0.511\,\mathrm{MeV} \tag*{(15)}
D+p→3He+χ+χˉ;2mχmax⁡=5.5 MeV(16)\mathrm{D}+p\rightarrow{}^{3}\mathrm{He}+\chi+\bar{\chi};\quad2m_{\chi}^{\max}=5.5\,\mathrm{MeV} \tag*{(16)}
3He+3He→6Be+χ+χˉ;2mχmax⁡=11.5 MeV(17){}^{3}\mathrm{He}+{}^{3}\mathrm{He}\rightarrow{}^{6}\mathrm{Be}+\chi+\bar{\chi};\quad2m_{\chi}^{\max}=11.5\,\mathrm{MeV} \tag*{(17)}
12C+p→13N+χ+χˉ;2mχmax⁡=1.9 MeV(18){}^{12}\mathrm{C}+p\rightarrow{}^{13}\mathrm{N}+\chi+\bar{\chi};\quad2m_{\chi}^{\max}=1.9\,\mathrm{MeV} \tag*{(18)}
13C+p→14N+χ+χˉ;2mχmax⁡=7.5 MeV(19){}^{13}\mathrm{C}+p\rightarrow{}^{14}\mathrm{N}+\chi+\bar{\chi};\quad2m_{\chi}^{\max}=7.5\,\mathrm{MeV} \tag*{(19)}
14N+p→15O+χ+χˉ;2mχmax⁡=7.3 MeV.(20){}^{14}\mathrm{N}+p\rightarrow{}^{15}\mathrm{O}+\chi+\bar{\chi};\quad2m_{\chi}^{\max}=7.3\,\mathrm{MeV}. \tag*{(20)}

In addition, there are processes with the de-excitations of 7Be{}^{7}\mathrm{Be} and 7Li{}^{7}\mathrm{Li} that create 0.430.43 and 0.48 MeV0.48\,\mathrm{MeV} energy pairs. The corresponding photon-emitting cross sections for most of the listed reactions are relatively well-known. Before evaluating the rate of MCP production, we would like to make several remarks:

  • One positron is emitted for each unit of pppp cycle. e+e^{+} annihilation is an important source of MCP, because in that case the probability of χ\chi production scales as O(ε2)O(\varepsilon^2), and not O(αε2)O(\alpha\varepsilon^2).

  • Process (17), 3He+3He{}^{3}\mathrm{He}+{}^{3}\mathrm{He} to 6Be{}^{6}\mathrm{Be}, is not very well known as the transition occurs due to E0 multipole and should be suppressed. This reaction is expected to be subdominant as the main 3He+3He{}^{3}\mathrm{He}+{}^{3}\mathrm{He} fusion reaction occurs without photon emission.

  • N(p,γ)O\mathrm{N}(p,\gamma)\mathrm{O} reaction to the ground state of oxygen is suppressed (Br∼6%\mathrm{Br}\sim6\%) and can be neglected. Reaction (19), on the other hand, occurs mostly due to E1 transition to the ground state (see e.g. [53] and TUNL Nuclear Data Project compilations6). Reaction (18) can be neglected on account of small QQ-value and the overall subdominance of the CNO chain.

  • We shall assume that D+p\mathrm{D}+p fusion reaction at solar energies is an equal mix of E1 and M1 transitions with more advanced treatment possible if needed [54]. This assumption is not crucial if 2mχ2m_{\chi} is away from 5.5 MeV threshold.

  • MCPs can also be emitted at order O(α2ε2)O(\alpha^{2}\varepsilon^{2}) from weak decays of e.g. 8B{}^{8}\mathrm{B}. These are even smaller contributions than the rest, and we neglect them.

In light of the above, we use reactions (15), (16) and (19) together with our results for the branching ratios into χχˉ\chi\bar{\chi} to calculate the overall production rate as function of mχm_{\chi}. A convenient way of expressing the answer is as follows:

14π(A.U.)2dNχ+χˉdt≃12 cm−2s−1×(ε10−5)2×F(mχ),(21)\frac{1}{4\pi(\mathrm{A.U.})^{2}}\frac{dN_{\chi+\bar{\chi}}}{dt}\simeq12\ \mathrm{cm}^{-2}\mathrm{s}^{-1}\times\left(\frac{\varepsilon}{10^{-5}}\right)^{2}\times F(m_{\chi}), \tag*{(21)}

where the function F(mχ)F(m_{\chi}), plotted in Figure 4, is dictated by the available phase space, as well as the relative strength of pppp and CNO chains. Notice that the probability to create an MCP pair from the positron annihilation is given by

Figure 4. Relative production of solar MCPs due to different reactions. For mχ<mem_{\chi} < m_e the flux is dominated by the positron annihilation.

⟨σv⟩e+e−→χχˉ⟨σv⟩e+e−→γγ=ε2×(1+r22)1−r2,r=mχ/me.(22)\frac{\langle\sigma v\rangle_{e^{+}e^{-}\to\chi\bar{\chi}}}{\langle\sigma v\rangle_{e^{+}e^{-}\to\gamma\gamma}}=\varepsilon^{2}\times\left(1+\frac{r^{2}}{2}\right)\sqrt{1-r^{2}},\quad r=m_{\chi}/m_{e}. \tag*{(22)}

In the limit of mχ≪mem_{\chi}\ll m_{e}, F=1F=1, and this probability is just ε2\varepsilon^{2}. Therefore the total MCP flux is 2ε22\varepsilon^{2} times the pppp neutrino flux, which explains the prefactor in Eq. (21). The discussion of production of other exotic particles in solar nuclear reactions can be found in recent work [55].

Quantity (21) is not affected by the slow-down of millicharged particles by the solar material, and in the steady state regime will correspond to an outgoing flux at an astronomical unit away from the Sun. The average energy of particles arriving from the Sun is, however, a more difficult question. One expects that for ε>10−6\varepsilon>10^{-6} most of the MCPs will thermalize and leave the Sun with relatively small energies. In that case, their propagation near the Earth’s surface may get affected by the magnetic fields. On the other hand, MCP particles with ε<10−7\varepsilon<10^{-7} are expected to exit the Sun without appreciable energy loss. They can be then very efficiently probed by the planned low electron recoil experiments such as Oscura [56]. We note that the recent results from the DAMIC-M experiment [57] have a potential sensitivity to ε∼0.5×10−6\varepsilon\sim0.5\times10^{-6} via the flux of solar MCPs. However, more in-depth investigation of χ\chi slow-down is needed to fully address this question. Our calculation of the overall flux (21) can be further augmented by the secondary Compton-like production of MCPs e+γ→e+χ+χˉe+\gamma\to e+\chi+\bar{\chi} from all photons produced in nuclear reactions.

Dark photon from nuclear de-excitation

In addition to MCPs the nuclear de-excitation could also produce massive dark photons A′A^{\prime} which effectively interact with ordinary matter through the coupling κQJμAμ′\kappa Q_J\mu A^{\prime}_{\mu}. Following the formalism in [36] and accounting for the dark photon mass, we obtain the ratio between dark photon and regular photon production rate for E1 and M1 transition as

rE(A′)=κ2(1+12mA′2ω2)1−mA′2ω2(23)r_{\mathcal{E}}^{(A^{\prime})} = \kappa^2 \left(1+\frac{1}{2}\frac{m_{A^{\prime}}^2}{\omega^2}\right)\sqrt{1-\frac{m_{A^{\prime}}^2}{\omega^2}} \tag*{(23)}
rB(A′)=κ2(1−mA′2ω2)32(24)r_{\mathcal{B}}^{(A^{\prime})} = \kappa^2\left(1-\frac{m_{A^{\prime}}^2}{\omega^2}\right)^{\frac{3}{2}} \tag*{(24)}

The produced dark photons subsequently decay to an electron-positron pair, allowing the possibility for detection. The decay rate of dark photons is given by

Γ=ακ2mA′3(1+2me2mA′2)1−4me2mA′2(25)\Gamma= \frac{\alpha\kappa^2m_{A^{\prime}}}{3}\left(1+\frac{2m_e^2}{m_{A^{\prime}}^2}\right)\sqrt{1-\frac{4m_e^2}{m_{A^{\prime}}^2}} \tag*{(25)}

For MeV scale dark photon with κ∼10−7\kappa\sim10^{-7}, the typical distance L=γβ/ΓL = \gamma\beta/\Gamma dark photons travel during their lifetime is a few kilometers, and at that distance the geo-dark photons are unlikely to play any role. Therefore we focus solely on detectors that are placed at a distance D≪LD \ll L from the nuclear reactors. Since the loss of dark photons from the source to the detector can be ignored, the dark photon decay event rate inside the detector is given by

R=ϕA′V/L,(26)R = \phi_{A^{\prime}}V/L, \tag*{(26)}

where ϕA′=Nγ(N,ω)rE(B)(A′)/(4πD2)\phi_{A^{\prime}} = N_{\gamma}^{(N,\omega)}r_{\mathcal{E}(\mathcal{B})}^{(A^{\prime})}/(4\pi D^2) is the dark photon flux, and VV is the detector volume. Given the proportionality of the event rate to the detector volume, dark photons are ideally constrained by large size detectors near nuclear reactors. We considered the measurement by the TEXONO Collaboration with CsI(Tl) scintillating crystals [44], that measured νˉe\bar{\nu}_e scattering on electrons. A dark photon decay inside the detector would be classified as a three-hit cosmic-ray unrelated event, and the counting rate that was observed for this class of events was at the level of O(10−2 events/kg/day/MeV)O(10^{-2}\,\mathrm{events/kg/day/MeV}). Somewhat less competitive limits can be derived from Ref. [58], that searched for the appearance of e+e−e^{+}e^{-} pairs from a hypothetical massive neutrino decay. Among the emission lines under consideration, the 4.060 MeV4.060\,\mathrm{MeV} emission line from 238U(n,γ)239U{}^{238}\mathrm{U}(n,\gamma){}^{239}\mathrm{U} provide the strongest sensitivity. The constraint we get by repurposing results of Ref. [44], together with comparison with other constraints at the same mass range, are shown in Figure 5. While our constraint does not exceed existing limits, the use of reactors as the source of dark photon provides an independent confirmation for the exclusion of dark photons at this mass and coupling range.

Excluded region for dark photon from this work and the SLAC E137 experiment

Figure 5. Excluded region for dark photon from this work and the SLAC E137 experiment [59, 60, 61]. The entire plot region is also excluded by the constraint based on supernova [62].

One can also set constraints on visibly decaying dark photons using large neutrino detectors deep underground. Specifically, KamLAND detector measured the solar 8B{}^{8}\mathrm{B} neutrinos, achieving effectively ∼5 MeV\sim5\,\mathrm{MeV} detection threshold [63]. A dark photon produced in nuclear reactors with EA′>5 MeVE_{A'} > 5\,\mathrm{MeV} will appear in that sample, and therefore can be constrained if energetic enough nuclear transitions can be found. The lines considered in this paper up to this point are not suitable as they would be below the threshold. One candidate for more energetic transitions is the neutron capture on iron, 56Fe(n,γ)57Fe{}^{56}\mathrm{Fe}(n,\gamma){}^{57}\mathrm{Fe}, that does result in significant emission of 7.6 MeV7.6\,\mathrm{MeV} gammas (corresponding to two fairly strong M1 transitions [64], discussed recently in connection with axion searches [65]). It is somewhat difficult to estimate exact fraction of neutrons captured on reactor walls without detailed simulations, and as a rough estimate we adopt 1%. With these numbers in mind, and taking and average baseline of 180 km, and the total thermal power of all reactors to be ∼100 GW\sim100\,\mathrm{GW}, we can calculate the expected event rate from A′A' decay and compare it with recorded event rate in KamLAND in the electron energy bin 7.5−8 MeV7.5 - 8\,\mathrm{MeV}. In this energy range, most of the observed signal is due to the solar neutrinos. For simplicity, we quote our result for a specific value of the dark photon mass of 2 MeV.

dNA′ decay/dtdNobserved/dt∣mA′=2 MeV∼4×10−4×(κ3×10−8)4.(27)\left.\frac{dN_{A'\,\mathrm{decay}}/dt}{dN_{\mathrm{observed}}/dt}\right|_{m_{A'}=2\,\mathrm{MeV}} \sim4 \times10^{-4} \times\left(\frac{\kappa}{3 \times10^{-8}}\right)^4. \tag*{(27)}

Notice that the normalization of κ\kappa in this expression is chosen to coincide with the decay distance similar to KamLAND baseline, as larger κ\kappa would lead to a dominant fraction of A′A^\prime decay before the detector. We conclude that the sensitivity of a “far detector” is less stringent than the results of the experiments near the reactor cores. It is possible that in the near future the JUNO experiment [66] could surpass KamLAND and set tighter bounds on visibly decaying A′A^\prime.

We would also like to discuss a special case of mA′<2mem_{A^\prime}<2m_e. In this case, the visible decay mode is a loop-induced A′→3γA^\prime\to3\gamma [67, 68], and it is very suppressed. The main mechanism for detecting A′A^\prime is in fact a Compton-like scattering, e+A′→e+γe+A^\prime\to e+\gamma. Using this process, Ref. [69] derived novel limits on dark photon in this mass range, obtaining κ<10−5\kappa<10^{-5}. Would nuclear reactors be able to improve on this bound? The inverse Compton process will create eγe\gamma in the final state, and it is not obvious what the limits are on this sub-category of events from the existing searches in detectors near the reactor cores. It will definitely have background from similar channels in 208Tl{}^{208}\mathrm{Tl} decays. To avoid this complication, we use KamLAND signal again, and estimate that

dNe+A′→e+γdNobserved/dt∣mA′<2me∼0.1×(κ3×10−5)4.(28)\left.\frac{dN_{e+A^\prime\to e+\gamma}}{dN_{\mathrm{observed}}/dt}\right|_{m_{A^\prime}<2m_e}\sim0.1\times\left(\frac{\kappa}{3\times10^{-5}}\right)^4. \tag*{(28)}

As before, neutron capture on iron nuclei is used as a source of A′A^\prime. We conclude that currently the reactors provide subdominant sensitivity to the LSND limits for dark photons below 2me2m_e mass. JUNO, however, would be able to improve this limit significantly, mostly due to a superb energy resolution and the possibility to look for sharp lines in the spectrum, and from the ∼3.5\sim3.5 times smaller baseline.

Discussion and Conclusions

Experiments at nuclear reactor experiments have been instrumental in determining the properties of neutrinos and constraining new physics such as axions and axion-like particles. Clearly, other MeV-scale particles from dark sectors can be constrained as well, with clear prospects of improving sensitivity to several classes of models (for an axion-like model see e.g. recent works [70, 71], and for very light dark photons [72]). In this paper, we have revisited the constraints on millicharged particles imposed by detectors placed very near reactor cores. We have also calculated the production of MCPs from natural radioactivity and nuclear reactions powering the Sun. The main difference with the previous treatment is that we point out that every photon emission in nuclear γ\gamma decays can be accompanied by the χχˉ\chi\bar{\chi} production provided that 2mχ<ω2m_\chi<\omega. We have derived the corresponding energy spectrum of MCP particles, using previous simulations of the γ\gamma emission, mostly from (n,γ)(n,\gamma) reactions and decays of radioactive fission products. The calculated flux is typically much harder than the Compton-type production of MCP in the eγe\gamma scattering. Consequently, higher mχm_\chi masses can be probed, and we derive novel limits in the 0.7−20.7-2 MeV range, surpassing the sensitivity of the SLAC MCP search. These limits can be further improved, as the MCP scattering cross section scales inversely proportional to the electron recoil energy, and new low-threshold detectors will help to increase sensitivity.

While nuclear reactors provide the largest fluxes of MCPs for the closely located detectors, we note that there is also an ambient flux of MCP created by natural radioactivity (mostly caused by the radioactive isotopes of K, U and Th). While fluxes of MCP created this way are much lower than those in proximity of reactor cores, the detection methods that can be employed could involve neutrino detectors, and large self-shielding dark matter detectors (such as those based on xenon dual-TPC technology). These are large deep-underground detectors operating in a maximally background-free environment, capable of achieving much lower counting rates than detectors placed in reactors’ proximity. While a solid constraint on ϵ\epsilon based on these detectors relies on a detailed knowledge of the composition of the Local Crust, our analysis shows that the constraints derived in this way are likely to fall within the ballpark of ε∼10−4\varepsilon\sim10^{-4}.

Perhaps more promising avenue for the future studies is the solar MCP flux. If mχm_\chi is below the mem_e the MCPs will be very efficiently produced by the positron annihilation. For heavier MCPs, the reaction D(p,γ)3He(p,\gamma){}^{3}\mathrm{He} will provide the dominant contribution up to mχ∼2.5 MeVm_\chi\sim2.5\,\mathrm{MeV}. Future low recoil experiments will be sensitive to the flux even for a very small value of millicharge, ε<10−6\varepsilon< 10^{-6}.

The improved energy reach achieved by considering the (n,γ)(n,\gamma) reactions also allows the on-shell production of dark photons heavy enough to decay to e+e−e^{+}e^{-} pairs. We derived the rate for near-reactor detectors to detect such decay signals and converted the observed signal rate to the constraints on dark photon couplings in the 1.1−4 MeV1.1-4\,\mathrm{MeV} range. While the constraints derived in this way do not supplant the existing limits, we note that this is the first dark photon constraint obtained from near-reactor experiments for the MeV range, and the limits can be further improved with a detailed analysis of the background and the use of larger volume detectors.

Before we close, we would like to point to several items of discussion:

  • We improved the reactor bounds on MCPs because (n,γ)(n,\gamma) reactions give access to higher frequencies and higher mχm_\chi masses compared to bremsstrahlung pair-production of χχˉ\chi\bar{\chi}. It would be also important to investigate whether in the mχ<mem_\chi< m_e range the bounds could be improved even further. A considerable fraction of these hard γ\gamma’s would undergo an e+e−e^{+}e^{-} pair production on heavy nuclei such as uranium. Subsequent annihilation of e+e^{+} with a branching to MCPs would produce χ\chi’s with the O(ε2)O(\varepsilon^{2}) probability, which is much larger than O(αε2)O(\alpha\varepsilon^{2}) in case of the bremsstrahlung production. In order to be able to do this re-analysis, one would have to simulate the relative weight of the pair production and Compton degradation of hard γ\gamma, or in other words, determine the yield of e+e^{+}. This will be best done by experimental collaborations themselves, accounting for photon conversions inside (possibly dominated by pair production) and outside (presumably dominated by Compton process) nuclear fuel rods.

  • In light of the recent success in developing extremely low threshold DM detectors [57], one would need a better quality estimate of the solar MCP flux. More careful calculations will take into account the spatial distribution of χχˉ\chi\bar{\chi} pair-production inside the solar core, and most importantly the subsequent slow-down, and, for large enough ε\varepsilon, eventual thermalization of these particles. Estimates show that ε∼10−6\varepsilon\sim10^{-6} appears to be a critical value of the millicharge, but a better calculation is feasible.

  • The range of masses and couplings explored in this paper is relevant for the possible solutions of EDGES anomaly. One should not, however, loose sight of the fact that such light MCPs are troublesome from the point of the very early cosmology, as they would certainly modify the predictions of the Big Bang Nucleosynthesis (BBN). “Fixing” deuterium and helium abundances will most likely require additional ingredients/actors from the DS at the time of BBN.

  • While (n,γ)(n,\gamma) reactions explored in this paper are necessarily limited in their kinematic reach to 2mχ2m_\chi and/or mA′m_{A'} under 5 MeV, the question remains of how efficiently the process of direct nuclear fission, i.e. nuclear scission itself, may produce exotic particles. While it will be suppressed by some powers of v/cv/c, where vv is the velocity of separating fragments, the potential kinematic reach will be higher than for the (n,γ)(n,\gamma) sources, as up to 180 MeV is available per every single occurrence of fission.

  • It is worth noting that deep underground far detector can also be used for deriving constraints on visibly decaying A′A'. It remains to be seen if the ongoing JUNO program will improve on some

of the discussed limits, once the solar 8B^{8}\mathrm{B} neutrino signal (suitable for deriving such limits) is measured.

Acknowledgments

The authors would like to thank Drs. A Berlin and Z. Liu for useful discussions. This work is supported in part by the DOE grant DE-SC0011842. MP would also like to acknowledge the hospitality of Perimeter Institute for Theoretical Physics in Waterloo, Ontario, Canada.

Pair production from E2 and E3 transitions

In this appendix we derive the ratio between MCP pair production and γ\gamma emission rates in nuclear de-excitation for E2 and E3 transitions under the recoil-less approximation. Up to some normalization for the multipole moments we expand the electric part of the nuclear transition current as

Ji=ωQi(1)+ωQij(2)qj+ωQijk(3)qjqk,(29)J_i = \omega Q_i^{(1)} + \omega Q_{ij}^{(2)}q^j + \omega Q_{ijk}^{(3)}q^j q^k, \tag*{(29)}
J0=qiJi/ω=Qi(1)qi+Qij(2)qiqj+Qijk(3)qiqjqk,(30)J_0 = q_iJ_i/\omega= Q_i^{(1)}q^i + Q_{ij}^{(2)}q^i q^j + Q_{ijk}^{(3)}q^i q^j q^k, \tag*{(30)}

where Qi(1)Q_i^{(1)}, Qij(2)Q_{ij}^{(2)}, Qijk(3)Q_{ijk}^{(3)} are the electric dipole, quadrupole, and octapole moments, and k=(ω,q)k=(\omega,q) is the photon’s (or virtual photon’s) four-momentum. Evaluating the squared amplitude for pair production and γ\gamma emission involves products of the form JμJν∗J_\mu J_\nu^*, when averaged over all directions and spins, such products can only depend on the magnitude of the multipoles rather than their components, therefore the products of the multipoles can be replaced by

Qi(1)Qj(1)∗→13∣Q(1)∣2δji,(31)Q_i^{(1)}Q_j^{(1)*} \to\frac{1}{3}\lvert Q^{(1)}\rvert^2\delta^i_j, \tag*{(31)}
Qij(2)Qkl(2)∗→15∣Q(2)∣2(δk(iδlj)−13δ(ij)δ(kl)),(32)Q_{ij}^{(2)}Q_{kl}^{(2)*} \to\frac{1}{5}\lvert Q^{(2)}\rvert^2\left(\delta_k^{(i}\delta_l^{j)}-\frac{1}{3}\delta^{(ij)}\delta_{(kl)}\right), \tag*{(32)}
Q(3)ijkQlmn(3)∗→17∣Q(3)∣2(δl(iδmjδnk)−35δ(ijδlk)δmn),(33)Q^{(3)ijk}Q_{lmn}^{(3)*} \to\frac{1}{7}\lvert Q^{(3)}\rvert^2\left(\delta_l^{(i}\delta_m^j\delta_n^{k)}-\frac{3}{5}\delta^{(ij}\delta_l^{k)}\delta_{mn}\right), \tag*{(33)}
Q(m)i⋯jQk⋯l(n)∗→0,form≠n,(34)Q^{(m)i\cdots j}Q_{k\cdots l}^{(n)*}\to0,\quad\text{for}\quad m\ne n, \tag*{(34)}

where the coefficients are determined by the requirement Qi⋯j(m)Qi⋯j(m)∗=∣Q(m)∣2Q_{i\cdots j}^{(m)}Q_{i\cdots j}^{(m)*}=\lvert Q^{(m)}\rvert^2 and the traceless condition of the multipole moments. These results then lead to

J0J0∗→13∣Q(1)∣2q2+215∣Q(2)∣2q4+235∣Q(3)∣2q6,(35)J_0J_0^* \to\frac{1}{3}\lvert Q^{(1)}\rvert^2q^2+\frac{2}{15}\lvert Q^{(2)}\rvert^2q^4+\frac{2}{35}\lvert Q^{(3)}\rvert^2q^6, \tag*{(35)}
J0Ji∗→13∣Q(1)∣2ωqi+215∣Q(2)∣2ωq2qi+235∣Q(3)∣2ωq4qi,(36)J_0J_i^* \to\frac{1}{3}\lvert Q^{(1)}\rvert^2\omega q_i+\frac{2}{15}\lvert Q^{(2)}\rvert^2\omega q^2q_i+\frac{2}{35}\lvert Q^{(3)}\rvert^2\omega q^4q_i, \tag*{(36)}
JiJj∗→13∣Q(1)∣2ω2δij+215∣Q(2)∣2ω2(34δijq2+14qiqj)+235∣Q(3)∣2ω2q2(23δijq2+13qiqj).(37)J_iJ_j^* \to\frac{1}{3}\lvert Q^{(1)}\rvert^2\omega^2\delta_{ij}+\frac{2}{15}\lvert Q^{(2)}\rvert^2\omega^2\left(\frac{3}{4}\delta_{ij}q^2+\frac{1}{4}q_iq_j\right)+\frac{2}{35}\lvert Q^{(3)}\rvert^2\omega^2q^2\left(\frac{2}{3}\delta_{ij}q^2+\frac{1}{3}q_iq_j\right). \tag*{(37)}

For γ\gamma emission the squared amplitude is given by

∣M(γ)2∣=−gμνJμJν∗→23∣Q(1)∣2q2+15∣Q(2)∣2q4+8105∣Q(3)∣2q6,(38)\lvert\mathcal{M}_{(\gamma)}^2\rvert=-g^{\mu\nu}J_\mu J_\nu^*\to\frac{2}{3}\lvert Q^{(1)}\rvert^2q^2+\frac{1}{5}\lvert Q^{(2)}\rvert^2q^4+\frac{8}{105}\lvert Q^{(3)}\rvert^2q^6, \tag*{(38)}

for pair production the squared amplitude is given by

∣M∣(χχˉ)2=TμνJμJν∗,Tμν=ε2e2k4(−gμν2k2+4pχμpχˉν+4pχˉμpχν),(39)|\mathcal{M}|^{2}_{(\chi\bar{\chi})}=T^{\mu\nu}J_{\mu}J_{\nu}^{*},\qquad T^{\mu\nu}=\frac{\varepsilon^{2}e^{2}}{k^{4}}\left(-g^{\mu\nu}2k^{2}+4p_{\chi}^{\mu}p_{\bar{\chi}}^{\nu}+4p_{\bar{\chi}}^{\mu}p_{\chi}^{\nu}\right), \tag*{(39)}

after some algebra we arrive at

∣M∣(χχˉ,E1)2=2e2ε23k4∣Q(1)∣2[k4+k2(−4EχEχˉ+2ω2)+4mχ2ω2],(40)|\mathcal{M}|^{2}_{(\chi\bar{\chi},E1)}=\frac{2e^{2}\varepsilon^{2}}{3k^{4}}\left|Q^{(1)}\right|^{2}\left[k^{4}+k^{2}\left(-4E_{\chi}E_{\bar{\chi}}+2\omega^{2}\right)+4m_{\chi}^{2}\omega^{2}\right], \tag*{(40)}
∣M∣(χχˉ,E2)2=e2ε215k4∣Q(2)∣2[−4k6+k4(16EχEχˉ−3ω2)+6k2ω2(−2EχEχˉ−2mχ2+ω2)+12mχ2ω4],(41)|\mathcal{M}|^{2}_{(\chi\bar{\chi},E2)}=\frac{e^{2}\varepsilon^{2}}{15k^{4}}\left|Q^{(2)}\right|^{2}\left[-4k^{6}+k^{4}\left(16E_{\chi}E_{\bar{\chi}}-3\omega^{2}\right)+6k^{2}\omega^{2}\left(-2E_{\chi}E_{\bar{\chi}}-2m_{\chi}^{2}+\omega^{2}\right)+12m_{\chi}^{2}\omega^{4}\right], \tag*{(41)}
∣M∣(χχˉ,E3)2=4e2ε2105k4∣Q(3)∣2[3k8−k6(12EχEχˉ+ω2)+2k4ω2(10EχEχˉ+4mχ2−3ω2)+4k2ω4(−2EχEχˉ−4mχ2+ω2)+8mχ2ω6].(42)|\mathcal{M}|^{2}_{(\chi\bar{\chi},E3)}=\frac{4e^{2}\varepsilon^{2}}{105k^{4}}\left|Q^{(3)}\right|^{2}\left[3k^{8}-k^{6}\left(12E_{\chi}E_{\bar{\chi}}+\omega^{2}\right)+2k^{4}\omega^{2}\left(10E_{\chi}E_{\bar{\chi}}+4m_{\chi}^{2}-3\omega^{2}\right)+4k^{2}\omega^{4}\left(-2E_{\chi}E_{\bar{\chi}}-4m_{\chi}^{2}+\omega^{2}\right)+8m_{\chi}^{2}\omega^{6}\right]. \tag*{(42)}

To obtain the ratio of cross sections, we integrate over the phase space for each process. This gives the differential ratio in the form of Eq. (3). After changing variables using k2=M2+M′2−2MEN′k^{2}=M^{2}+M'^{2}-2ME_{N'}, the problem reduces to calculating the elementary integral

In=∫kmin⁡2kmax⁡2k2ndk2,(43)I_{n}=\int_{k_{\min}^{2}}^{k_{\max}^{2}}k^{2n}\mathrm{d}k^{2}, \tag*{(43)}

the kinematical limits correspond to χ\chi and χˉ\bar{\chi} moving parallel or antiparallel to each other, with

kmin⁡2=2EχEχˉ+2mχ2−2∣pχ∣∣pχˉ∣,kmax⁡2=2EχEχˉ+2mχ2+2∣pχ∣∣pχˉ∣.(44)k_{\min}^{2}=2E_{\chi}E_{\bar{\chi}}+2m_{\chi}^{2}-2\lvert\mathbf{p}_{\chi}\rvert\lvert\mathbf{p}_{\bar{\chi}}\rvert,\qquad k_{\max}^{2}=2E_{\chi}E_{\bar{\chi}}+2m_{\chi}^{2}+2\lvert\mathbf{p}_{\chi}\rvert\lvert\mathbf{p}_{\bar{\chi}}\rvert. \tag*{(44)}

Finally, in terms of InI_n, the differential ratio is given by

drE1dEχ=αε24πω3[I0+(−4EχEχˉ+2ω2)I−1+4mχ2ω2I−2],(45)\frac{\mathrm{d}r_{E1}}{\mathrm{d}E_{\chi}}=\frac{\alpha\varepsilon^{2}}{4\pi\omega^{3}}\left[I_{0}+\left(-4E_{\chi}E_{\bar{\chi}}+2\omega^{2}\right)I_{-1}+4m_{\chi}^{2}\omega^{2}I_{-2}\right], \tag*{(45)}
drE2dEχ=αε212πω5[−4I1+(16EχEχˉ−3ω2)I0+6ω2(−2EχEχˉ−2mχ2+ω2)I−1+12mχ2ω4I−2],(46)\frac{\mathrm{d}r_{E2}}{\mathrm{d}E_{\chi}}=\frac{\alpha\varepsilon^{2}}{12\pi\omega^{5}}\left[-4I_{1}+\left(16E_{\chi}E_{\bar{\chi}}-3\omega^{2}\right)I_{0}+6\omega^{2}\left(-2E_{\chi}E_{\bar{\chi}}-2m_{\chi}^{2}+\omega^{2}\right)I_{-1}+12m_{\chi}^{2}\omega^{4}I_{-2}\right], \tag*{(46)}
drE3dEχ=αε28πω7[3I2−(12EχEχˉ+ω2)I1+2ω2(10EχEχˉ+4mχ2−3ω2)I0+4ω4(−2EχEχˉ−4mχ2+ω2)I−1+8mχ2ω6I−2].(47)\frac{\mathrm{d}r_{E3}}{\mathrm{d}E_{\chi}}=\frac{\alpha\varepsilon^{2}}{8\pi\omega^{7}}\left[3I_{2}-\left(12E_{\chi}E_{\bar{\chi}}+\omega^{2}\right)I_{1}+2\omega^{2}\left(10E_{\chi}E_{\bar{\chi}}+4m_{\chi}^{2}-3\omega^{2}\right)I_{0}+4\omega^{4}\left(-2E_{\chi}E_{\bar{\chi}}-4m_{\chi}^{2}+\omega^{2}\right)I_{-1}+8m_{\chi}^{2}\omega^{6}I_{-2}\right]. \tag*{(47)}

References

  1. [1]G. Lanfranchi, M. Pospelov, and P. Schuster, “The Search for Feebly Interacting Particles,” Ann. Rev. Nucl. Part. Sci. 71 (2021) 279–313, arXiv:2011.02157 [hep-ph].DOI
  2. [2]J. Beacham et al., “Physics Beyond Colliders at CERN: Beyond the Standard Model Working Group Report,” J. Phys. G 47 no. 1, (2020) 01501, arXiv:1901.09966 [hep-ex].DOI
  3. [3]B. Holdom, “Two U(1)’s and Epsilon Charge Shifts,” Phys. Lett. B 166 (1986) 196–198.DOI
  4. [4]S. Davidson, S. Hannestad, and G. Raffelt, “Updated bounds on millicharged particles,” JHEP 05 (2000) 003, arXiv:hep-ph/0001179.
  5. [5]H. Vogel and J. Redondo, “Dark Radiation constraints on millicharged particles in models with a hidden photon,” JCAP 02 (2014) 029, arXiv:1311.2600 [hep-ph].DOI
  6. [6]A. D. Dolgov, S. L. Dubovsky, G. I. Rubtsov, and I. I. Tkachev, “Constraints on millicharged particles from Planck data,” Phys. Rev. D 88 no. 11, (2013) 117701, arXiv:1310.2376 [hep-ph].DOI
  7. [7]J. H. Chang, R. Essig, and S. D. McDermott, “Supernova 1987A Constraints on Sub-GeV Dark Sectors, Millicharged Particles, the QCD Axion, and an Axion-like Particle,” JHEP 09 (2018) 051, arXiv:1803.00993 [hep-ph].DOI
  8. [8]D. Dunsky, L. J. Hall, and K. Harigaya, “CHAMP Cosmic Rays,” JCAP 07 (2019) 015, arXiv:1812.11116 [astro-ph.HE].DOI
  9. [9]D. F. G. Fiorillo and E. Vitagliano, “Self-Interacting Dark Sectors in Supernovae Can Behave as a Relativistic Fluid,” Phys. Rev. Lett. 133 no. 25, (2024) 251004, arXiv:2404.07714 [hep-ph].arxiv.org/abs/2404.07714
  10. [10]M. Pospelov and H. Ramani, “Earth-bound millicharge relics,” Phys. Rev. D 103 no. 11, (2021) 115031, arXiv:2012.03957 [hep-ph].
  11. [11]A. Berlin, H. Liu, M. Pospelov, and H. Ramani, “Terrestrial density of strongly-coupled relics,” Phys. Rev. D 109 no. 7, (2024) 075027, arXiv:2302.06619 [hep-ph].arxiv.org/abs/2302.06619
  12. [12]J. D. Bowman, A. E. E. Rogers, R. A. Monsalve, T. J. Mozdzen, and N. Mahesh, “An absorption profile centred at 78 megahertz in the sky-averaged spectrum,” Nature 555 no. 7694, (2018) 67-70, arXiv:180.05912 [astro-ph.CO].DOI
  13. [13]R. Barkana, N. J. Outmezguine, D. Redigolo, and T. Volansky, “Strong constraints on light dark matter interpretation of the EDGES signal,” Phys. Rev. D 98 no. 10, (2018) 103005, arXiv:1803.03091 [hep-ph].
  14. [14]E. D. Kovetz, V. Poulin, V. Gluscevic, K. K. Boddy, R. Barkana, and M. Kamionkowski, “Tighter limits on dark matter explanations of the anomalous EDGES 21 cm signal,” Phys. Rev. D 98 no. 10, (2018) 103529, arXiv:1807.11482 [astro-ph.CO].DOI
  15. [15]H. Liu, N. J. Outmezguine, D. Redigolo, and T. Volansky, “Reviving Millicharged Dark Matter for 21-cm Cosmology,” Phys. Rev. D 100 no. 12, (2019) 123011, arXiv:1908.06986 [hep-ph].DOI
  16. [16]A. Berlin, N. Blinov, G. Krnjaic, P. Schuster, and N. Toro, “Dark Matter, Millicharges, Axion and Scalar Particles, Gauge Bosons, and Other New Physics with LDMX,” Phys. Rev. D 99 no. 7, (2019) 075001, arXiv:1807.01730 [hep-ph].DOI
  17. [17]A. A. Prinz et al., “Search for millicharged particles at SLAC,” Phys. Rev. Lett. 81 (1998) 1175-1178, arXiv:hep-ex/9804008.DOI
  18. [18]G. Magill, R. Plestid, M. Pospelov, and Y.-D. Tsai, “Millicharged particles in neutrino experiments,” Phys. Rev. Lett. 122 no. 7, (2019) 071801, arXiv:1806.03310 [hep-ph].DOI
  19. [19]K. J. Kelly and Y.-D. Tsai, “Proton fixed-target scintillation experiment to search for millicharged dark matter,” Phys. Rev. D 100 no. 1, (2019) 015043, arXiv:1812.03998 [hep-ph].DOI
  20. [20]R. Harnik, Z. Liu, and O. Palamara, “Millicharged Particles in Liquid Argon Neutrino Experiments,” JHEP 07 (2019) 170, arXiv:1902.03246 [hep-ph].DOI
  21. [21]ArgoNeuT Collaboration, R. Acciarri et al., “Improved Limits on Millicharged Particles Using the ArgoNeuT Experiment at Fermilab,” Phys. Rev. Lett. 124 no. 13, (2020) 131801, arXiv:1911.07996 [hep-ex].
  22. [22]F. Kling, J.-L. Kuo, S. Trojanowski, and Y.-D. Tsai, “FLArE up dark sectors with EM form factors at the LHC forward physics facility,” Nucl. Phys. B 987 (2023) 116103, arXiv:2205.09137 [hep-ph].DOI
  23. [23]SENSEI Collaboration, L. Barak et al., “Search by the SENSEI Experiment for Millicharged Particles Produced in the NuMI Beam,” Phys. Rev. Lett. 133 no. 7, (2024) 071801, arXiv:2305.04964 [hep-ex].
  24. [24]Y.-D. Tsai, I. Hwang, R. Schmitz, M. Citron, K. Gunthoti, J. Steenis, H. Jeong, H. Moon, J. H. Yoo, and M. X. Liu, “LANL SENSEI-mQ: Dedicated search for milli/fractionally charged particles at LANL,” arXiv:2407.07142 [hep-ph].arxiv.org/abs/2407.07142
  25. [25]R. Essig, P. Li, Z. Liu, M. McDuffie, R. Plestid, and H. Xu, “Probing millicharged particles at an electron beam dump with ultralow-threshold sensors,” JHEP 04 (2025) 057, arXiv:2412.09652 [hep-ph].arxiv.org/abs/2412.09652
  26. [26]R. Plestid, V. Takhistov, Y.-D. Tsai, T. Bringmann, A. Kusenko, and M. Pospelov, “New Constraints on Millicharged Particles from Cosmic-ray Production,” Phys. Rev. D 102 (2020) 115032, arXiv:2002.11732 [hep-ph].DOI
  27. [27]R. Harnik, R. Plestid, M. Pospelov, and H. Ramani, “Millicharged cosmic rays and low recoil detectors,” Phys. Rev. D 103 no. 7, (2021) 075029, arXiv:2010.11190 [hep-ph].DOI
  28. [28]M. Du, R. Fang, and Z. Liu, “Millicharged particles from proton bremsstrahlung in the atmosphere,” JHEP 08 (2024) 174, arXiv:2211.11469 [hep-ph].arxiv.org/abs/2211.11469
  29. [29]H. Wu, E. Hardy, and N. Song, “Searching for heavy millicharged particles from the atmosphere,” Phys. Rev. D 110 no. 11, (2024) 115037, arXiv:2406.01668 [hep-ph].arxiv.org/abs/2406.01668
  30. [30]TEXONO Collaboration, L. Singh et al., “Constraints on millicharged particles with low threshold germanium detectors at Kuo-Sheng Reactor Neutrino Laboratory,” Phys. Rev. D 99 no. 3, (2019) 032009, arXiv:1808.02719 [hep-ph].
  31. [31]CONNIE, Atucha-II Collaboration, A. A. Aguilar-Arevalo et al., “Search for reactor-produced millicharged particles with Skipper-CCDs at the CONNIE and Atucha-II experiments,” arXiv:2405.16316 [hep-ex].
  32. [32]TEXONO Collaboration, H. M. Chang et al., “Search of axions at the Kuo-Sheng nuclear power station with a high-purity germanium detector,” Phys. Rev. D 75 (2007) 052004, arXiv:hep-ex/0609001.
  33. [33]J. B. Dent, B. Dutta, D. Kim, S. Liao, R. Mahapatra, K. Sinha, and A. Thompson, “New Directions for Axion Searches via Scattering at Reactor Neutrino Experiments,” Phys. Rev. Lett. 124 no. 21, (2020) 211804, arXiv:1912.05733 [hep-ph].DOI
  34. [34]D. Aristizabal Sierra, V. De Romeri, L. J. Flores, and D. K. Papoulias, “Axionlike particles searches in reactor experiments,” JHEP 03 (2021) 294, arXiv:2010.15712 [hep-ph].arxiv.org/abs/2010.15712
  35. [35]N. M. Kroll and W. Wada, “Internal pair production associated with the emission of high-energy gamma rays,” Phys. Rev. 98 (1955) 1355–1359.DOI
  36. [36]C. Pitrou and M. Pospelov, “QED corrections to Big-Bang nucleosynthesis reaction rates,” Phys. Rev. C 102 no. 1, (2020) 015803, arXiv:1904.07795 [astro-ph.CO].DOI
  37. [37]P. Talou and R. Vogt, Nuclear Fission: Theories, Experiments, and Applications. Springer Nature, 2023.
  38. [38]TEXONO Collaboration, B. Xin et al., “Production of electron neutrinos at nuclear power reactors and the prospects for neutrino physics,” Phys. Rev. D 72 (2005) 011006, arXiv:hep-ex/0502001.
  39. [39]From ENSDF database as of December 14, 2024. Version available at http://www.nndc.bnl.gov/ensarchivals/.
  40. [40]LZ Collaboration, J. Aalbers et al., “First search for atmospheric millicharged particles with the LUX-ZEPLIN experiment,” arXiv:2412.04854 [hep-ex].
  41. [41]W. W. M. Allison and J. H. Cobb, “Relativistic Charged Particle Identification by Energy Loss,” Ann. Rev. Nucl. Part. Sci. 30 (1980) 253–298.DOI
  42. [42]B. L. Henke, E. M. Gullikson, and J. C. Davis, “X-Ray Interactions: Photoabsorption, Scattering, Transmission, and Reflection at E = 50-30,000 eV, Z = 1-92,” Atom. Data Nucl. Data Tabl. 54 no. 2, (1993) 181–342.DOI
  43. [43]MUNU Collaboration, Z. Daraktchieva et al., “Final results on the neutrino magnetic moment from the MUNU experiment,” Phys. Lett. B 615 (2005) 153–159, arXiv:hep-ex/0502037.
  44. [44]TEXONO Collaboration, M. Deniz et al., “Measurement of Nu(e)-bar -Electron Scattering Cross-Section with a CsI(Tl) Scintillating Crystal Array at the Kuo-Sheng Nuclear Power Reactor,” Phys. Rev. D 81 (2010) 072001, arXiv:0911.1597 [hep-ex].
  45. [45]N. Ackermann et al., “First observation of reactor antineutrinos by coherent scattering,” arXiv:2501.05206 [hep-ex].
  46. [46]Borexino Collaboration, M. Agostini et al., “First Simultaneous Precision Spectroscopy of pp, ⁷Be, and pep Solar Neutrinos with Borexino Phase-II,” Phys. Rev. D 100 no. 8, (2019) 082004, arXiv:1707.09279 [hep-ex].
  47. [47]XENON Collaboration, E. Aprile et al., “Search for New Physics in Electronic Recoil Data from XENONnT,” Phys. Rev. Lett. 129 no. 16, (2022) 161805, arXiv:2207.11330 [hep-ex].
  48. [48]LZ Collaboration, J. Aalbers et al., “Search for new physics in low-energy electron recoils from the first LZ exposure,” Phys. Rev. D 108 no. 7, (2023) 072006, arXiv:2307.15753 [hep-ex].
  49. [49]PandaX Collaboration, X. Zeng et al., “Exploring New Physics with PandaX-4T Low Energy Electronic Recoil Data,” Phys. Rev. Lett. 134 no. 4, (2025) 041001, arXiv:2408.07641 [hep-ex].
  50. [50]Y. Huang, V. Chubakov, F. Mantovani, R. L. Rudnick, and W. F. McDonough, “A reference earth model for the heat-producing elements and associated geoneutrino flux,” Geochemistry, Geophysics, Geosystems 14 no. 6, (2013) 2003–2029.arxiv.org/abs/1301.0365
  51. [51]Borexino Collaboration, M. Agostini et al., “Spectroscopy of geoneutrinos from 2056 days of Borexino data,” Phys. Rev. D 92 no. 3, (2015) 031101, arXiv:1506.04610 [hep-ex].
  52. [52]Particle Data Group Collaboration, S. Navas et al., “Review of particle physics,” Phys. Rev. D 110 no. 3, (2024) 030001.
  53. [53]F. Ajzenberg-Selove, “Energy levels of light nuclei A = 13-15,” Nucl. Phys. A 523 (1991) 1–196.
  54. [54]L. E. Marcucci, G. Mangano, A. Kievsky, and M. Viviani, “Implication of the proton-deuteron radiative capture for Big Bang Nucleosynthesis,” Phys. Rev. Lett. 116 no. 10, (2016) 102501, arXiv:1510.07877 [nucl-th]. [Erratum: Phys.Rev.Lett 117, 049901 (2016)].DOI
  55. [55]F. D’Eramo, G. Lucente, N. Nath, and S. Yun, “Terrestrial detection of hidden vectors produced by solar nuclear reactions,” JHEP 12 (2023) 091, arXiv:2305.14420 [hep-ph].
  56. [56]Oscura Collaboration, S. Perez et al., “Searching for millicharged particles with 1 kg of Skipper-CCDs using the NuMI beam at Fermilab,” JHEP 02 (2024) 072, arXiv:2304.08625 [hep-ex].
  57. [57]DAMIC-M Collaboration, K. Aggarwal et al., “Probing Benchmark Models of Hidden-Sector Dark Matter with DAMIC-M,” arXiv:2503.14617 [hep-ex].
  58. [58]C. Hagner, M. Altmann, F. von Feilitzsch, L. Oberauer, Y. Declais, and E. Kajfasz, “Experimental search for the neutrino decay neutrino (3) —> j-neutrino + e+ + e- and limits on neutrino mixing,” Phys. Rev. D 52 (1995) 1343–1352.pubmed.ncbi.nlm.nih.gov/10019356
  59. [59]J. D. Bjorken, S. Ecklund, W. R. Nelson, A. Abashian, C. Church, B. Lu, L. W. Mo, T. A. Nunamaker, and P. Rassmann, “Search for Neutral Metastable Penetrating Particles Produced in the SLAC Beam Dump,” Phys. Rev. D 38 (1988) 3375.
  60. [60]S. Andreas, C. Niebuhr, and A. Ringwald, “New Limits on Hidden Photons from Past Electron Beam Dumps,” Phys. Rev. D 86 (2012) 095019, arXiv:1209.6083 [hep-ph].DOI
  61. [61]L. Marsicano, M. Battaglieri, M. Bondi’, C. D. R. Carvajal, A. Celentano, M. De Napoli, R. De Vita, E. Nardi, M. Raggi, and P. Valente, “Dark photon production through positron annihilation in beam-dump experiments,” Phys. Rev. D 98 no. 1, (2018) 015031, arXiv:1802.03794 [hep-ex].DOI
  62. [62]J. H. Chang, R. Essig, and S. D. McDermott, “Revisiting Supernova 1987A Constraints on Dark Photons,” JHEP 01 (2017) 107, arXiv:1611.03864 [hep-ph].DOI
  63. [63]KamLAND Collaboration, S. Abe et al., “Measurement of the ⁸B Solar Neutrino Flux with the KamLAND Liquid Scintillator Detector,” Phys. Rev. C 84 (2011) 035804, arXiv:1106.0861 [hep-ex].
  64. [64]R. B. Firestone, T. Belgya, M. Krťička, F. Bečvář, L. Szentmiklósi, and I. Tomandl, “Thermal neutron capture cross section for Fe56(n,γ),” Phys. Rev. C 95 no. 1, (2017) 014328.DOI
  65. [65]L. Waites, A. Thompson, A. Bungau, J. M. Conrad, B. Dutta, W.-C. Huang, D. Kim, M. Shaevitz, and J. Spitz, “Axionlike particle production at beam dump experiments with distinct nuclear excitation lines,” Phys. Rev. D 107 no. 9, (2023) 095010, arXiv:2207.13659 [hep-ph].DOI
  66. [66]JUNO Collaboration, F. An et al., “Neutrino Physics with JUNO,” J. Phys. G 43 no. 3, (2016) 030401, arXiv:1507.05613 [physics.ins-det].
  67. [67]M. Pospelov, A. Ritz, and M. B. Voloshin, “Bosonic super-WIMPs as keV-scale dark matter,” Phys. Rev. D 78 (2008) 115012, arXiv:0807.3279 [hep-ph].DOI
  68. [68]S. D. McDermott, H. H. Patel, and H. Ramani, “Dark Photon Decay Beyond The Euler-Heisenberg Limit,” Phys. Rev. D 97 no. 7, (2018) 073005, arXiv:1705.00619 [hep-ph].DOI
  69. [69]M. Pospelov and Y.-D. Tsai, “Light scalars and dark photons in Borexino and LSND experiments,” Phys. Lett. B 785 (2018) 288–295, arXiv:1706.00424 [hep-ph].DOI
  70. [70]V. Brdar, B. Dutta, W. Jang, D. Kim, I. M. Shoemaker, Z. Tabrizi, A. Thompson, and J. Yu, “Axionlike Particles at Future Neutrino Experiments: Closing the Cosmological Triangle,” Phys. Rev. Lett. 126 no. 20, (2021) 201801, arXiv:2011.07054 [hep-ph].DOI
  71. [71]NEON Collaboration, B. J. Park et al., “New Constraints on Axionlike Particles with the NEON Detector at a Nuclear Reactor,” Phys. Rev. Lett. 134 no. 20, (2025) 201002, arXiv:2406.06117 [hep-ex].
  72. [72]M. Danilov, S. Demidov, and D. Gorbunov, “Constraints on hidden photons produced in nuclear reactors,” Phys. Rev. Lett. 122 no. 4, (2019) 041801, arXiv:1804.10777 [hep-ph].DOI

Paper details

vlnc:W_strvbyu9
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents