Arxiv:2104.12784V1 [Hep-Ph] 26 Apr 2021 Beta Decays of Tritium, the Signal Can Be Overwhelmed by Spectrum of Ultra High-Energy Neutrinos [6, 7], Utilising Background

IPPP/20/99

Observing relic neutrinos with an accelerator experiment

Martin Bauer and Jack D. Shergold Institute for Particle Physics Phenomenology, Department of Physics, Durham University, Durham, DH1 3LE, UK

We present a new technique for observing low energy neutrinos with the aim of detecting the cosmic neutrino background using ion storage rings. Utilising high energy targets exploits the quadratic increase in the neutrino capture cross section with beam energy, and with sufficient beam energy, enables neutrino capture through inverse-beta decay processes from a stable initial state. We also show that there exist ion systems admitting resonant neutrino capture, capable of achieving larger capture cross sections at lower beam energies than their non-resonant counterparts. We calculate the neutrino capture rate and the optimal experimental runtime for a range of different resonant processes and target ions and we demonstrate that the resonant capture experiment can be performed with beam energies as low as O(10 TeV) per target nucleon. Unfortunately, none of the ion systems discussed here can provide sufficient statistics to discover the cosmic neutrino background with current technology. We address the challenges associated with realising this experiment in the future, taking into account the uncertainty in the beam and neutrino momentum distributions, synchrotron radiation, as well as the beam stability.

I. INTRODUCTION

Neutrinos travel through the cosmos with minimal interactions with their environment, which makes them clean messengers of solar, astrophysical and cosmic phenomena. The weak interaction responsible for this unique property is also the leading cause of difficulties when attempting to directly detect neutrinos in a lab. Low- energy neutrinos are particularly challenging in this respect, making a future detection of the cosmic neutrino background (CνB) the holy grail of neutrino physics. Es- tablished detection techniques typically have a threshold of ∼ 300 keV [1], as illustrated in Figure 1 and are therefore incapable of detecting neutrinos from the CνB, 0 9 which have mean kinetic energy Ek,ν = 1.8 × 10− keV. The most discussed proposal to detect the CνB was originally suggested by Weinberg [2], which looks for an excess of events beyond the tritium beta decay electron FIG. 1. Fluxes of low-energy electron neutrinos, alongside the endpoint energy, arising from neutrino capture process thresholds for Borexino and radiochemical detection experiments. The dashed lines have units cm−2 s−1 and correspond 3H + ν → 3He+ + e . Currently, KATRIN [3] is the e − to (from left to right): 7Be decay to 7Li∗ and 7Li, and pep most advanced experiment implementing this technique neutrinos. A full discussion of the solar fluxes is given in and is expected to set the best constraint on CνB over- Appendix B. densities. In the future, a dedicated experiment such as PTOLEMY [4] or MARE [5] could reach sensitivity to the CνB. Alternative proposals to detect the CνB include cantly enhanced event rates. However, due to the natural the observation of the Z-resonance absorption dip in the arXiv:2104.12784v1 [hep-ph] 26 Apr 2021 beta decays of tritium, the signal can be overwhelmed by spectrum of ultra high-energy neutrinos [6, 7], utilising background. To overcome this, we may instead consider the neutrino wind using a torsion balance [8] or laser in- inverse beta decay induced by neutrino capture, e.g. terferometry experiments [9], as well as the observation of coherent scattering of CνB neutrinos at accelerators [10]. 3 ++ 3 + + He +ν ¯e → H + e . (2) We explore the sensitivity of an ion storage ring exploiting resonant, neutrino-induced beta decays and elec- Because of the threshold of QHe = 18.59 keV in this ex- tron captures to detect cosmic neutrinos. Such an exper- ample, this process can only occur if an electron neutrino iment could perform various measurements. The same with sufficient energy is absorbed. For neutrinos of mass exothermic process on which PTOLEMY is based can be 0 mν Tν at rest and ions of mass M accelerated to en- observed in accelerated tritium ions [10] ergy E in the lab frame, this threshold corresponds to a 3 + 3 ++ H + νe → He + e− . (1) neutrino energy in the rest-frame of the beam At high energies, the neutrino capture cross section scales mν Eeν = E ≥ QHe . (3) quadratically with the beam energy, allowing for signifi- M 2

1 Here and in the remainder of the paper we denote quan- a factor fc,i ' 100 − 200 for mν = 1 eV [14–16]. In 3 tities in the beam frame with a tilde. For He, (3) trans- very optimistic scenarios enhancements of up to fc,i ' 3 6 lates into E ≥ 56 (eV/mν ) TeV, putting it in reach of a 10 − 10 are considered based on the local baryon over- next-generation 100 TeV collider. As the cross sections density [17, 18]. 2 for both processes (1) and (2) scale as GF , they are very Following the argument of [12], the CνB consists exclu- small unless Eeν significantly exceeds the electron mass. sively of left helicity states in the absence of significant In addition the charge radius of the final state is differ- gravitational interactions, whilst for strongly clustered ent from the initial state, causing it to be ejected from neutrinos the density is be shared equally between both the beam. We argue that both issues can be resolved by helicity states. For the remainder of this paper we leave instead considering resonant 2 → 1 processes for which fc,i, and consequently the left and right helicity relic 2 the interaction rate scales like GF , e.g. neutrino densities nνL,i and nνR,i as free parameters , unless specified otherwise. Finally, we note that in the 3 ++ 3 + He +ν ¯e + e−(bound) → H , (4) rest frame of a highly relativistic ion beam the neutrino number density is further enhanced by the beam Lorentz and the cross section is maximised at the threshold. factor γ = E/M due to length contraction, however the At resonance, the cross section for the process in (4) total number of neutrino captures is a frame independent can exceed the at rest neutrino capture cross section by quantity owing to the competing time dilation effect. many orders of magnitude. For example, in the case of The exact distribution function of relic neutrinos today resonant capture of neutrinos produced in beta beams, is unknown. However, in the absence of significant inter- the neutrino capture cross section for tritium exceeds actions since decoupling it is reasonable to suggest that it the PTOLEMY cross section by seven orders of mag- will be the same as at decoupling, redshifted. Assuming nitude [11]. This technique can be used to capture low- a Fermi-Dirac distribution at decoupling, this leads to a energy neutrinos using accelerated ions instead. The re- distribution per degree of freedom [19] mainder of this paper is structured as follows. We de- dn 2n p2 scribe the sources of low-energy neutrinos in Section II i νi ν = 3 pν , (6) and explore non-resonant neutrino capture on accelerated dpν 3ζ(3)Teff e Teff + 1 ions in Section III. In Section IV we show how utilising when normalised to n . Here p is the magnitude of resonant neutrino capture leads to larger cross sections νi ν the neutrino three momentum and T is the effective compared to the non-resonant case and present novel 3- eff neutrino temperature, which we take to be constant with state systems with a clean, stable final state. In Sec- T ' T 0. tion V we calculate the beam conversion rates for a num- eff ν We propose to detect relic neutrinos by capturing ber of ion systems and address the challenges of realising them on a high-energy ion beam. Summing over the a future experiment in Section VI. We conclude in Sec- mass eigenstates and multiplying with the relative beam- tion VII. neutrino velocity vrel, we arrive at the electron neutrino flux incident on the beam

II. LOW ENERGY NEUTRINOS dφe X 2 dni dni = vrel |Uei| = vrel , (7) dpν dpν dpν i=1,2,3 The spectrum of low-energy neutrinos with an energy below the threshold of existing experiments is dominated where the Uei are the relevant entries of the PMNS neu- by two sources, solar neutrinos and relic neutrinos of the trino mixing matrix. For the remainder of this paper we CνB. While both can be targets of neutrino capture ex- set vrel = 1 for a highly relativistic ion beam. We plot the periments we focus on the CνB. electron neutrino flux for this case in Figure 2. To model Relic neutrinos decoupled from the SM thermal bath at the width, we will treat this distribution as Gaussian cen- dec tred on pν with a standard deviation ∆ν ' 0.291 meV for a temperature Tν ∼ 1 MeV, and have since redshifted 0 4 the remainder of the paper. In addition to relic neutrinos, to Tν ' 1.68 × 10− eV. This corresponds to a mean 0 we briefly discuss the possibility of observing solar neu- momentum of pν = 0.529 meV, which we consider negligible compared with the neutrino mass [12]. The local, trinos below the energy threshold of existing experiments − present day number density of relic neutrinos per degree such as Borexino (20 200 keV [20]). The distribution of freedom can then be written as [13] of low-energy solar neutrinos is depicted by the yellow contour in Figure 1. While solar neutrinos are more en- 3 ergetic, the solar neutrino flux is smaller by several orders nνi = fc,i n0 ' fc,i 56 cm− , (5) where i = e, µ, τ and fc.i is a neutrino mass-dependent enhancement factor capturing local overdensities from 1 The exact value depends mostly on the density profile of the Dark gravitational clustering of non-relativistic neutrinos. Re- Matter halo. Gravitational interactions between neutrinos and of sults from N-1-body simulations find enhancements of neutrinos on Dark Matter are neglected in N-1-body simulations. 2 These naturally satisfy nν + nν = nν . fc,i ' 1.1 − 1.2 for neutrino masses mν < 0.1eV up to L,i R,i i 3

final state electron Qe = −1 and the atomic number of 1 final state isotope Z, whilst ρ ' 1.2A 3 fm is the nuclear radius. Summing over mass eigenstates and exploiting the unitarity of the PMNS matrix, we find that the total neutrino capture cross section on tritium at rest for non- relativistic neutrinos βeν = veν /c 1 is

45 2 σNR = σβeν = 3.85 × 10− cm , (11) Eee=E0

where E0 = me + |QH| + mν is the energy carried by the final state electron, and |QH| = QHe is the energy released by the decay of tritium. This gives the total rest rate for the capture of Dirac neutrinos on NT tritium targets at rest per year rest 1 Rrest = φeNT σNR ' 4fc,i y− (12) where we have used N rest = 2 × 1025 in line with the FIG. 2. Electron neutrino flux from the CνB. T PTOLEMY proposal [4], corresponding to 100 g of tritium. The capture rate for Majorana neutrinos is twice of magnitude compared to the expected flux of relic neu- as large, as there will also be a flux of right-helicity neutrinos. For our calculations of the solar neutrino flux trinos. we use the standard solar model given in [21], which we If instead tritium ions are accelerated on a beam, the discuss further in Appendix B. larger centre of mass energy results in an enhanced neutrino capture cross section. Neglecting final state nuclear recoil, the electron energy and momentum in the tritium III. NEUTRINO CAPTURE ON A BEAM rest frame for accelerated tritium are given by

Eee = me + |QH| + Eeν , (13) The idea to capture relic neutrinos using tritium was q ﬁrst proposed by Weinberg, and the rate has been esti- pee = (Eeν + |QH|)(Eeν + 2me + |QH|) (14) mated based on our current best understanding of the

CνB [2, 22]. In the following we explore the energy de- such that for Eeν > 2me and Eeν > |QH| the cross sec- pendence of the neutrino capture cross section and show tion increases quadratically with the neutrino energy how it can be exploited for accelerated tritium or helium- 2 σ0 ∝ Eeν . For an ion beam with energy E and ion mass 3 to increase the event rate. M, the energy of the neutrinos in the ion rest frame can The cross section for neutrino capture on tritium for neu- then be written as Eeν ' Emν /M, and the quadratic trinos with degenerate masses and helicity seν is given enhancement of the cross section becomes effective once by [12], the beam energy satisfies E > 2meM/mν . Exploiting this quadratic dependence is thus extremely challenging σvν = (1 − 2 sν vν )σ0(Eee), (8) e e e for the capture of cosmic neutrinos on tritium, as it requires a beam energy of where veν is the neutrino velocity in the beam rest frame and the energy-dependent cross section reads eV E & 3 PeV , (15) 2 2 2 m3He mν σ0(Eee) = GF |Vud| F (2, Eee)C(q ) Eeepee . (9) m3H for M = m3H ' 3 GeV. The situation becomes more tenable when considering low energy solar neutrinos. For Here Eee and pee are the energy and the magnitude of the final state electron three momentum, Vud is the CKM Eν = 1 keV the quadratic enhancement begins at a beam matrix element and C(q2) contains details of nuclear energy of E & 3 TeV for the same choice of target, well 2 structure. We will use the value C(q mπ) ' 0.874 within reach of modern accelerators. However, unless throughout. The Fermi function encodes the enhance- the thermal solar neutrino flux significantly exceeds the ments due to electromagnetic effects between the final predicted value shown in Figure 1, the capture rate would state nucleus and outgoing electron, and reads be too low to observe at that energy scale. For an accelerated beam of tritium ions, the helicity 2(1 + S) 2S 2 πη| − |2 dependent terms in (8) can become important. Setting F (Z, Eee) = 2 (2peeρ) − e− Γ(S iη) , − Γ(1 + 2S) veν ' vrel = 1, we find (10) ( 1 p 2 2 σ0(Eee), sν = − , where η = QeZα/ve and S = 1 − (αZ) are given in σβ = e 2 (16) e eν 1 terms of the fine-structure constant α, the charge of the 0, seν = + 2 , 4 such that the neutrino capture cross section for right- helicity states vanishes identically, whilst the left-helicity capture cross section doubles. One would then naively expect the total neutrino capture rate to double with respect to the capture rate at rest (12). However, due to the combined effects of Lorentz contraction and the velocity of the beam relative to the neutrinos, the left- helicity neutrino density for any neutrino flavor in the beam frame reads nνL = γ nνL + P (nνR − nνL ) , (17) e − where P is the probability that a relic neutrino will flip helicity when− boosted into the beam frame  1 1 v¯ν  − arcsin , v¯ sin θ¯ν > v¯ν , P = 2 π v¯ sin θ¯ν − 0, v¯ sin θ¯ν ≤ v¯ν , FIG. 3. Event rate enhancement for a PTOLEMY-on-a-beam (18) type experiment (blue) as a function of the beam energy E andv ¯ andv ¯ν are the average beam and neutrino velocities using mν = 0.1 eV, alongside the rate for the inverse process on a beam (orange), assuming N beam = N rest. Also shown in the lab frame, respectively, and θ¯ν ∈ [0, π] is the aver- T T age angle between the neutrinos and the axis perpendicu- by the dotted line is the decrease in the tritium decay rate as a function of the beam energy. lar to the beam plane. We first consider Dirac neutrinos. In the case of a relic neutrino background close to the prediction by standard cosmology fc,i ' 1, they are ran- ¯ domly oriented in the lab frame, θν = π/2, and nνR = 0. eV. The quadratic enhancement becomes apparent at Since the neutrinos are non-relativistic in the lab frame, E 30 PeV, which far exceeds that of any existing or ¯ & v¯ν v¯ sin θν , it follows that P ' 1/2 and nνL = γnνi /2. − e proposed accelerators. For large overdensities fc,i 1, the strong gravitational The reconstruction of signal events is additionally com- clustering leads to nνL = nνR = nνi /2, and we arrive at 3 ++ the same result. For Majorana neutrinos, regardless of plicated by the fact that the final state He is produced without interacting with background neutrinos by clustering, nνL = nνR = nνi and the number density of 3 + 3 ++ left-helicity neutrinos is twice as large as that for Dirac the natural beta decay of tritium H → ν¯e + He +e− neutrinos in the cases discussed above [12]. We consider with lifetime τ0 = 17.8 y, producing an irreducible background to the signal process. This background is huge, Dirac neutrinos with nνL = nνi /2 for the remainder of the paper. We then arrive at the total capture rate for a even if one takes into account the relativistically in- beam creased lifetime of the beam ions τbeam = γτ0. In Figure 3 number of target ions NT on a beam, we show the decay rate 1/τbeam as a function of E with the dotted line. φe beam Rbeam = N σR(Eee), (19) 2 T This background can be avoided if the beam energy is large enough to enable the inverse process where the relativistic cross section can be written as 3 ++ 3 + + He +ν ¯e → H + e which has a threshold QHe = 18.59 keV. In this case the production of tri- 2Eeepee F (2, Eee) σR(Eee) = p − σNR , (20) tium ions in the final state provides a clean signal state, E E2 − m2 F (2,E0) 0 0 e − because the the final state can only be obtained by the anti-neutrino capture process, at the cost of a decaying and the ratio of Fermi functions is ∼ 1. In order for a signal. The cross section for the inverse process reads PTOLEMY-on-a-beam type experiment to improve the event rate of the experiment at rest one would therefore need inv 2 2 2 m3H σ0 (Eee) = GF |Vud| F+(1, Eee)C(q ) Eeepee, (22) m3He R E 2 m2 N beam beam ' ν T > 1 , (21) R M p 3 2 · 1025 rest 2|QH|me assuming the same form factors as the tritium process rest and defining F+(Z, Eee) as (10) for the positron with where we assume (15) holds and NT = 100NA/3 with 23 1 charge Qe = +1. the Avogadro number NA ' 6 × 10 mol− for the target at rest as proposed in [23]. In Figure 3 we show In the beam frame, the anti-neutrino energy required beam rest the ratio (21) for NT = NT tritium ions as a to overcome the threshold for the inverse process is Eeν > function of the beam energy in blue, using mν = 0.1 QHe, and the electron energy and momentum are given 5 by most energetic neutrino events seen today have an en- 15 ergy of Eν ' 6 · 10 eV [28]. Here we propose to use a Eee = me − QHe + Eeν , (23) beam of accelerated ions to capture neutrinos at nuclear q β-resonances, resulting in a cross section that is indepen- − − pee = (Eeν QHe)(Eeν + 2me QHe) . (24) dent of GF . We consider two resonant processes: the resonant bound beta decay process (RBβ) For m3H ' m3He, and neglecting changes in the form factors, the enhancement for the relativistic cross section fol- A A Z P + νe → Z+1D + e−(bound), (26) lows (20) with the replacement F (2, Eee) → F+(1, Eee). − and the resonant electron capture process (REC) In the high-energy limit Eeν¯ 2me, the enhancement over PTOLEMY is then given by (21) to a very good A A Z P + e−(bound) +ν ¯e → Z 1D, (27) approximation. This behaviour is reflected in Figure 3 − by the orange contour asymptotically tracing the blue for parent and daughter nuclei P and D, respectively. contour for large E. The cross section for these processes involving relativis- We chose the 3H−3He system to illustrate the neutrino tic beam frame neutrinos with energy Eeν and resonance capture processes in this section. In principle the cross threshold Q can be described by a Breit-Wigner function, section (9) can be enhanced if one considers ions with 2 smaller energy gaps, but the scaling with the initial state 2π 2JD + 1 Γ /4 3 3 σRes = BDP , (28) ion mass M in (21) favours the H− He system over any 2 2 2 Ee 2JP + 1 (Eeν − Q) + Γ /4 other choice. For an experiment with accelerated ions ν to realise a substantial enhancement over the neutrino where Γ is the total width of the resonant state D, JP capture rate at an experiment at rest like PTOLEMY and JD denote the total angular momentum of the par- requires beam energies orders of magnitude above what ent and daughter nuclei, and BDP = Br(D → P νe) or can be achieved by current accelerator technology. How- BDP = Br(D → P ν¯ee−(bound)) are the branching ra- ever, with sufficient energy, accelerated ion beams can tios of the daughter D to decay back to the parent state capture cosmic neutrinos even if the capture process has P . For a beam tuned to Eeν ' peν = Q the cross section a threshold. This leads to a clean final state consisting of is resonantly enhanced, with a peak cross section ions that cannot be produced in an experiment at rest. 2 Then provided one can overcome the threshold Eeν > Q, 15 2JD + 1 1 keV 2 σpeak = 2.5 · 10− BDP cm , (29) the inverse process is always superior to the thresholdless 2JP + 1 Q process on a beam. 2 which is suppressed by Q instead of GF . Expressed in terms of the neutral atom masses M(Z,A), the thresh- IV. RESONANT PROCESSES olds for the resonant processes involving nuclei with degree of ionisation I are given by The main limitation identified in the last section for cosmic neutrino capture on a beam is the large energy QREC = M(Z − 1,A) + BI 1(Z − 1) − M(Z,A) required to achieve an enhancement over the same ex- − periment at rest. It is not surprising that the neutrino − BI (Z) + BEC(Z − 1) (30) capture cross section for accelerated tritium ions scales + E∗(Z − 1,A) − E∗(Z,A), quadratically with the beam energy, because the cross QRBβ = M(Z + 1,A) + BI+1(Z + 1) − M(Z,A) section is suppressed by G2 . This G2 suppression is a F F − B (Z) − B (Z + 1) (31) universal problem in proposals to detect the CνB [24]. I Bβ Coherent scattering effects cannot overcome this prob- + E∗(Z + 1,A) − E∗(Z,A), lem, as was established in a no-go theorem prohibiting where B (Z) is the binding energy of the I outermost cross sections scaling linearly in G [25, 26]. The single I F electrons taken away from the neutral atom to produce exception is the Stodolsky effect which can only be mea- the ionised state, while B (Z) > 0 and B (Z) > 0 are sured for a net helicity asymmetry of the cosmic neutrino EC Bβ the binding energies of the missing and captured final background [27]. An alternative method of avoiding G2 F state electrons, respectively. The remaining parameter, suppression is to use a resonance. Very energetic neutri- E (Z), denotes the excitation energy of any nuclear iso- nos can scatter off the cosmic neutrino background at the ∗ mers that may be chosen as the initial or resonant state, Z-resonance with a peak cross section of order G [6], F which can significantly affect the value of Q. In this sec- 12π tion we will consider fully ionised RBβ initial states and → ∝ σZ = 2 Br(Z ν¯ανα) GF . (25) REC initial states with a single 1s shell electron, such MZ that we can neglect electron de-excitation contributions This process has not been observed yet, but would lead to BDP in the resonance state. to a dip in the high energy cosmic neutrino spectrum at The use of a state with at most 1s shell electrons is 22 Eν ' 10 eV for a neutrino mass mν = 1 eV, and the what allows for a cross section that is independent of 6

GF , as the only allowed decays are weak such that BDP A. Resonant 2-state systems is O(1). As a result, the cross section (29) looks very promising, but the total width Γ ∝ GF is narrow. The The most straightforward realisation of resonant neu- lab frame momentum distribution of relic neutrinos has a trino capture on a beam is a 2-state system consisting of finite width, whilst the cross section to capture neutrinos a parent ion P and a resonant daughter state D which 2 away from resonance is suppressed by a factor Γ /4Eν . decays back into the initial parent particle. For reso- Therefore the resonant process will only capture the small nant bound beta decay, the daughter ion will decay back fraction of neutrinos whose energy in the beam frame through electron capture lies on a narrow interval about the resonance peak, a C fact which is not helped by additional broadening due REC P RB P D to a non-monochromatic ion beam. For a very narrow P D . (37) B EC resonance the neutrino capture rate R per target ion NT EC F can be written as Here and in the following processes contributing to the C Z∞ RB F R dφe production of signalREC states are indicated by a solid ar- = dEeν σRes(Eeν ) P D row. DashedP arrows indicateD processes thatP reduce the NT dEeν C EC Q signal or the numberB of targets available to capture neu- B F trinos, e.g. by decays of the daughter ion back into the π dφe ' σpeakΓ parent state. In the case of resonant electron capture the 2 dEeν Eeν =Q daughter decays back into the parent ion by bound beta C + r 3 REC P π 2JD + 1 Γ BDP decay. A fraction of theRB daughter C ionsP will also decay P D = 2 φe , (32) + 2 2JP + 1 Q through continuousP beta decay intoD a the P ion, which B ∆e E REC EC F will be ejectedP from the beamECD eventually because of its where in the last step we have assumed that the neutrino B charge-mass ratio. We can writeEC this process as energy distribution in the beam frame is Gaussian with F C + RB F a standard deviation ∆e E Γ. REC P D P + P D It is then instructive to attempt to estimate ∆E. Given C . (38) EC B B F that the uncertainty in the lab frame neutrino momentum C F is ∆ and the beam momentum is subject to fluctuations RB ν In the case of resonant beta decay (37), the daughter of scale ∆b, we find that P D ion can decay intoEC different nuclear isomers of the parent s B 2 2 ion. Only decays into the parentC statePF isomer+ contribute P ∂Eeν ∂Eeν to the branching ratio that appears in the resonant cross ∆e E = ∆ν + ∆b . (33) REC ∂pν ∂p section (28).P We parameteriseD the fraction of decays into EC the parent isomerB over decays into any other isomer by For non-relativistic CνB neutrinos pν mν and p M χ ≤ 1. Here, χ accounts for theEC fact thatF some daughter RB this results in ions might decay into excited parent states. In the case of P ⇤ D q electron capture the continuous beta decay of the daugh- EC 1 p 2 2 2 2 ∆e E ' (E∆ν ) + (mν ∆b) = Q δ + δ , (34) + M ν b ter ion doesn’t contribute to theC resonantF cross section either. We can thenRB write for branching ratio where we have defined the fractional uncertainties δb = P D ∆ /E and δ = ∆ /E . This leads to the event rate for EC b ν ν ν χ, B RBβ, ∆e E Γ BDP = χ F (39) , REC. 1 + K(Z,Q) r 3 R π 2JD + 1 Γ BDP = φe. (35) 3 p 2 2 Here K(Z,Q) denotes the ratio of the width for the NT 2 2JP + 1 Q δν + δb daughter ion to bound beta decay into the n = 1 shell However, it is more convenient to define the quality factor with nf = 2 vacant s-orbitals, over the width for contin- as the ratio of the rate for producing signal events over uous beta decay, defined in appendix A. the rate of signal loss For any given experiment we are interested in the number of signal states ND on the beam after a runtime t. In γ R Rτ ≡ (36) order to compare ions with vastly different lifetimes we Γ NT introduce the dimensionless variable x = t/γτD, where 2 17 2JD + 1 BDP fc,i 0.1 eV 1 keV τD is the lifetime of the daughter ion at rest and γ the = 1.7 · 10− . p 2 2 Lorentz factor of the beam. For an initial number of 2JP + 1 δν + δb mν Q parent ions N0 on the beam, the expected number of In the following we will use this dimensionless quantity resonant daughter states D on the beam is given by, to measure how effective a given system of ions is at pro- x 2 ducing and retaining signal states on the beam. ND(x) = N0Rτ (1 − e− ) + O(Rτ ) (40) C + REC P RB P D P D B EC EC F

C + RB F REC P D P + P D C EC B B F

C + P P 7 REC P D EC in the table of nuclidesB as illustrated in Figure 4. EC The analogous decay chain of the neutrino-inducedF res- RB C + onant bound beta decay (37) for the 3-state system is P ⇤ REC D P RB P EC D P D + C B EC EC F F RB P D , (41) 106 C + Cd EC B RB F REC F P D P + P D 106 C EC Ag where continuous betaB decay into the ionised ﬁnal state, B F F +, is also allowed. Similarly, the 3-state resonant elec- 106Pd tron capture process decay chain reads

C P + P FIG. 4. Isobar triplets with “stable-unstable-stable” config- REC urations are candidates for 3-state systems for resonant neu- P D . (42) EC trino capture. Graphic taken in part from [29] B EC F RB P ⇤ D EC where the quality factor defined in (36) and is considered The presence of a third state modifies the branching ra- + to be Rτ 1. For completeness we give the full expres- tios for the decay back to the initialC state,F sions for this and other beam population equations in RB appendix C. Determining the rate of resonant neutrino P χBEC, D RBβ, EC capture in 2-state systems can therefore be achieved by BDP = χ(1 − BEC)B (43)  , RECF , solving (40) for Rτ with a measured value of ND. There 1 + K(Z,Q) are diminishing returns in running the experiment beyond x > xmax ' 1, as ND(x & xmax) ' N0Rτ . This where BEC is the branching ratio for the resonance to corresponds to a dynamic equilibrium, where D particles decay by electron capture, widely available in nuclear are produced by neutrino capture at same rate that they data tables [29]. Further, the branching ratios for the decay back into the initial state P . In the cases where daughter ion to decay into the final state are B < χ, no true equilibrium can ever be reached, as the DP  total number of ions on the beam is depleted by decays 1 − BEC +  , RBβ, D → P . However, this effect is negligible for small Rτ . BDF = 1 + K(Z + 2,QDF ) (44) The large x limit therefore represents the theoretical BEC, REC , maximum number of resonances that can be on the beam at any one time, and suggests using short lifetime states where QDF is the Q-value for the D → F transition and such that ND reaches its maximum with a shorter exper- one has to evaluate the ratio of continuous and bound imental runtime. Unfortunately, this presents an experi- beta decay widths K(Z,Qf ) for nf = 2 in (43) and nf = mental challenge, as short lifetime states will inevitably 1 in (44). See appendix A for details. For small quality decay if the beam needs to be slowed to make a mea- factors Rτ 1, we find for the number of final state ions surement of ND. on the beam,

x 2 NF (x) = N0Rτ BDF (x + e− − 1) + O(Rτ ) (45) B. Resonant 3-state systems whilst the number of resonance states on the beam evolves according to (40). The full expression can be The conundrum of 2-state systems for resonant neu- found in appendix C. trino capture can be overcome using systems with a final The advantages of using a 3-state system become ap- state F different from both the parent and daughter ions. parent here, as (45) is a monotonically increasing func- We such an arrangement a 3-state system. In 3-state sys- tion with x. The full expression (C6) satisfies tems the resonant daughter ion can retain a large decay branching ratio into the parent ion and therefore a large N0BDF χ resonance can be achieved, but a third, stable state can lim NF (x) = N0Rτ , (46) x χ − BDP be produced through a different decay of the daughter →∞ ion. If this stable state has the same charge-mass ratio such that unlike 2-state systems, there is no equilibrium as both the P and D states, then it can remain on the at x ' 1 and given an infinite amount of time, a signifi- beam indefinitely and can be detected after the run is fin- cant fraction of the initial state ions will be converted to ished. Candidate 3-state systems correspond to “stable- a stable final state. The remainder will end up in either unstable-stable” triplets of isobars that can be identified the F + or P + states, which are ejected from the beam. 8

The inclusion of the third state typically leads to a The excited state P ∗ is accelerated on a beam and can reduction in BDP and subsequently the quality Rτ is re- either produce the signal state D through bound beta duced compared to a 2-state system with the same Q- decay or decay back into its ground state P by emission value. In contrast to 2-state systems, there are no clear of photon. The daughter state D can capture an electron disadvantages in using processes with daughter states and decay back into the initial state P ∗ or its ground that have as short a lifetime as possible and an exper- state P . iment with a 3-state system therefore retains a fraction Alternatively, analogous to (38), the neutrino can be of the signal even if the beam is slowed down. Addition- captured in an resonant electron capture process by an ally, in the large x limit for 3-state systems the lack of excited state, an equilibrium value allows NF to far exceed ND for an equivalent 2-state system. This leadsP to a remarkable P result: we can achieve a resonantly enhanced neutrino cross sections with a stable, clean signal.EC B , (48) The goal is then twofold; to maximise the quality Rτ such that the number of signal statesRB on the beam in- REC D P + creases as quickly as possibleP ⇤ and to find resonanceD states P ⇤ C EC B with the smallest possible lifetime γτD, such that a larger x can be achieved with minimal runtime. Both of these 2 where the daughter ion can bound beta decay into both can be achieved by minimising the Q-value, as Rτ ∝ Q− 1 the initial state P ∗ or its ground state P , or continuous and γ ∝ Q− . We further note that the width for bound beta decay to P +. Since the beam energy needs to be beta decay is larger for smaller Q-values, leading to an tuned for resonant neutrino capture with P ∗, the larger enhanced branching ratio BDP for resonant electron cap- P Q-value of P rendersP it unable to capture relic neutrinos ture, see appendix A for details. Since the Q value also at the same beam energy. determines the beam energy required for the experiment, EC Following the procedureB outlined in appendix , we can it is the most important parameter for both 2 and 3-state derive the number of signal state particles D on the beam systems. C F + C + RB for the system (47).REC In the approximationP Rτ 1 we P ⇤ D find in termsP ⇤ of the lifetimeD ratio of the daughter and EC B B F excited state, η = τD/τP ∗ , EC F C. Excited states N0Rτ x(η 1) x 2 ND(x) = (1 − e− − )e− + O(R ) (49) η − 1 τ The bottleneck for a realistic experiment with reso- ( N0Rτ ηx x nant states remains the Q-value, which has to be suffi- (1 − e− )e− , η 1 = η . (50) ciently low for an accelerator experiment to probe neu- x N0Rτ (1 − e− ), η 1 trino masses at the eV scale or below. In principle, a lower Q-value can be achieved for any given target by This equation reproduces (40) in the long excited state considering excited nuclear states. Exciting the initial lifetime limit η 1, and in this case the excited state sys- state increases its effective mass, reducing the effective tem retains the same properties previously discussed in Q-value for neutrino capture in (30) and (31). We do Section IV A. However, it has drastically different prop- not speculate about the experimental challenges related erties in the short parent lifetime, η 1, regime. For to exciting and accelerating the ions here, but show how large η, the maximum population of signal states on the transitions with excited states could lead to higher rates beam is obtained for a runtime than resonant neutrino capture experiments with ground state ions. Exciting the initial states P ∗ comes at the cost 1 xmax ' ln η , η 1 , (51) of making them unstable, with a lifetime τP ∗ . We discuss η resonant neutrino capture processes with excited states using both 2- and 3-state systems here. First, we con- which is significantly smaller than for the ground state sider a 2-state system analogous to (37) with an excited system where xmax ' 1. A very short lifetime of P ∗ state P ∗ that can undergo resonant bound beta decay by rapidly depletes the beam of excited initial states P ∗, capturing a neutrino, leaving a beam which consists entirely of ground states P and daughter states D. For x > xmax, the number of resonance states start to decay according to P P N0Rτ x ND(Bx > xmax) ' e− , (52) EC . (47) η

RB such that forRECx < x < 1, the number of signal states D max D P P ⇤ D P ⇤ C EC remains closeB to its maximal value, and the signal states only begin to rapidly decay for a runtime x & 1.

P P

EC B C F C RB REC P P ⇤ D P ⇤ D EC B B F EC F 9

The maximal number of signal states in an experiment The number of stable signal states F on the beam at time with an excited initial state can then be compared to the x is expected to follow maximal signal state population in an experiment de- x ηx signed to capture neutrinos with its ground state isomer N0BDF Rτ ηe− − e− 2 NF (x) = 1 + + O(R ) as described in Section IV A. The excited state system η 1 − η τ is superior to the corresponding ground state system if (58) xmax < x < 1 and their respective quality factors fulfil ( N0 DF Rτ x B (1 − e− ), η 1, ' η (59) 1 x Rτ, > Rτ, , (53) N0BDF Rτ (x + e− − 1), η 1, η ES GS where the small η limit reproduces (45) as expected. where GS and ES refer to the ground state and excited In contrast to the ground state system however, the state system quantities respectively. Neglecting different limiting value for the number of signal states on the beam branching ratios and spin factors, this translates into a is given by condition on the Q-values, √ N0BDF Rτ lim NF (x) = . (60) η QES . QGS , (54) x − B →∞ η + Rτ (1 DP ) which implies that excited state systems only present a For short excited state lifetimes with large values of η, the better choice for very small Q-values or η. If it is possible number of signal states saturates at x ' 1 with NF (x > to re-excite P states on the beam, multiple runs with the 1) N0. Therefore by using excited P states, we have excited initial states for a runtime xmax each would have lost the ability for 3-state systems to convert a significant the potential to produce a sizable population of D states fraction of the beam at large x, although we retain a in the same time needed to reach the maximum signal clean, stable signal. state population for a ground state system at x ' 1. In For a repeated excitation of the initial state P → P ∗ this case the experiment with the excited state system is the condition for which the excited 3-state system superior to the equivalent experiment performed with a produces more signal states than an equivalent ground ground state system if state system in a given time x is given by (55).

1 p In Figure 5 we compare the expected fraction of a beam Rτ,ES > Rτ,PGS =⇒ ln η QES . QGS , (55) P ln η of ions that is converted into the signal state as a function EC of x. The differentB contours correspond to the resonant where we set x = 1 and neglected different branching electron capture process in the 2-state system (38), shown ratios and spin factors.RB As in the case of 2-state sys- in blue,REC the 3-state system (42), shown in orange, and the tems with a ground state parent ion, the signal state is D P + P ⇤ D P ⇤ analogous transitionC with excited initial states shown in unstable and will decay.EC greenB in the case of the 2-state system in (48) and in As before, one can produce stable signal states in a red for the 3-state system in (57). The resonant bound 3-state system with an excited initial state. For reso- beta decay process leads to slightly different conversion nant bound beta decay, the excited 3-state system can rates, due to different contributions from continuous beta be written in analogy to (41) as decay to the total decay width. Since we generally expect P P smaller Q-values for excited states we have fixed Q = 100 P keVP for the ground state systems in Figure 5 and Q = 0.1 EC B keV for the excited states. EC FocusingB first on the ground state systems, Figure 5 RB REC + C F + shows that the 2-stateC systemP + reaches its maximal popu- D P ⇤ RB D P REC P ⇤ C lation of the signal state at x ' 1. Beyond that point the EC P ⇤ B D P ⇤ D EC , (56) beamB has reached equilibrium between the initial and the B F resonant state. Instead,EC forF the 3-state system the final state is stable and in principle gets populated until the and in the case of resonant electron capture in analogy whole beam is either converted or ejected. As a result, 3- to (42) as state systems can achieve substantially larger beam conversion rates, although this process becomes less efficient P P for large x. The point at which the 3-state system beam conversion fraction exceeds the 2-state system is visible EC B at x ' 1. C + C + RB F REC P The main advantage of the systems with excited states is that it can achieve equilibrium at x ∼ (1/η) 1, P ⇤ D P ⇤ D . (57) EC B much faster than ground state systems. Multiple runs of B F EC F excited 2-state systems can then have cumulative beam 10

demonstrate that such isotopes exist in nature, but our list is not extensive. It is likely that better systems using different ions or more esoteric systems like ionised molecules can be identified. In Table I we list a number of ion systems and the number of neutrinos captured per target ion after a 1 year experimental runtime. The first two rows are examples of excited initial states, for which we use the last two columns to denote the ratio of lifetimes between the initial and resonant states as well as the event rate that could be achieved if the initial state is re-excited x/xmax times, respectively. We find that in general the beam conversion fractions are limited by x 1, resulting from the relatively long weak decay lifetimes that are extended by large Lorentz factors γ = Q/mν . Additionally, many of the systems listed suffer from prohibitively large beam energy require- FIG. 5. Fraction of parent ions converted to signal (D or F ments, of order (E/A) > 1 PeV per nucleon. states) for different resonant systems, all with χ = 1, JD = JP , mν = 0.1 eV, Z = 50, A = 100, δb = 0 and BEC = Despite this, there are several targets that show some 0.5 where appropriate. For the ground state (GS) systems, promise. In particular, an experiment with the excited Q = 100 keV, whilst the excited state (ES) systems use Q = 174 RBβ 174 state system Yb∗(1318 keV) −−−→ Lu could be per- 0.1 keV and η = 109. Finally, for the 3-state systems we plot formed with a beam energy of as little as 9.78 TeV per NF using a solid line, and NF + ND using a dotted line. nucleon, well within reach of an LHC-sized experiment. However, due to the incredibly short lifetime of the conversion factors that are orders of magnitude larger 1318 keV initial excited state, the beam would need to be compared to ground state systems. The experimental constantly replenished or re-excited in order to achieve realisation of these systems represents an extraordinary reasonable statistics for a runtime of 1 year. For higher RBβ challenge. It requires to keep repopulating a beam with energies we can use the 193Ir −−−→ 193Pt system, which 174 short-lived excited states and read out the signal before can achieve comparable statistics to the Yb∗ system it is lost at x ' 1 − 10. A 3-state system with an excited without any the limitations of the excited state. This sys- state overcomes this problem by populating a stable final tem would be out of reach of an LHC-sized experiment, state. In contrast to the 3-state system with an initial but would be accessible by a future 100 TeV collider, such ground state, the experiment can likely operate at sub- as the proposed Future Circular Collider (FCC) [39]. stantially smaller values of Q. As shown in Figure 5 the A more significant challenge posed by this experiment maximum number of conversions is smaller compared to is getting sufficiently many ions on the beam to observe ground state systems, because the excited states have the an event with a runtime of 1 year. In the aforementioned additional route to decay back into the ground state. RBβ 193Ir −−−→ 193Pt system an event rate of 1/year would The results shown in Figure 5 have to be compared require 4.24 × 1022 ions on the beam, corresponding to a with the expected rate at PTOLEMY (12), which corre- total beam mass of 13.6 g. This number far exceeds the sponds to a conversion factor of 2 · 10 25 of the target − number of ions on the LHC beam [31], however, as we no tritium atoms per year. For comparison, for a runtime longer wish to collide the beam, it is difficult to compare of t = 10γτ , the conversion rate for the example 2-state D our experiment to a traditional collider experiment. We system in Figure 5 would be ∼ 10 19. − speculate on what may be possible in simplified scenarios in VI. V. REAL WORLD EXAMPLES

Whilst promising in theory, the efficacy of a reso- VI. BEAM LIMITATIONS nant neutrino capture experiment will depend strongly on whether or not an ion system with the right properties exists. In the previous section we have shown that Performing a resonant neutrino capture experiment to properties necessary to maximise the conversion rate in- detect the CνB is fraught with difficulties. In particular, dependent of the process are a small Q-value, short reso- such an experiment would require accelerating a macro- nance lifetimes and sizable branching ratios for the reso- scopic number of particles to extremely high energies, nance to decay back to the initial and final states. Here whilst being able to precisely hit the resonance. In this we compile a list of several ion systems with these desired section we attempt to estimate what can realistically be properties in Table I. We emphasise that these examples achieved in some simplified scenarios. 11

BDP ND(x)/N0 System Q (keV) E/A (TeV) x (t = 1 y) log η x ND (xmax) 10 xmax N0 BDF NF (x)/N0

RBβ 174Yb∗(1318 keV) −−−→ 174Lu 1.05 9.78 0.052 1.99 · 10−5 8.10 · 10−34 17.31 8.25 · 10−23 −34 RBβ Bβ 0.002 2.62 · 10 104Ru∗(988 keV) −−−→ 104Rh −−→ 104Pd 129.06 1.20 · 103 0.401 12.73 2.88 · 10−23 6.39 · 10−5 8.27 · 10−39 RBβ 193Ir −−−→ 193Pt 5.55 51.69 1 2.50 · 10−7 2.36 · 10−23 - - 0.003 8.20 · 10−24 106Cd −−−→REC 106Ag −−→EC 106Pd 213.23 1.98 · 103 7.14 · 10−3 - - 0.995 2.92 · 10−26 RBβ 165Ho −−−→ 165Er 327.60 3.05 · 103 1 1.79 · 10−4 7.29 · 10−24 - - RBβ 157Gd −−−→ 157Tb 15.92 148.22 0.999 6.13 · 10−8 1.41 · 10−24 - - RBβ 71Ga −−−→ 71Ge 224.17 2.09 · 103 1 9.91 · 10−6 5.74 · 10−25 - - 121Sb −−−→REC 121Sn 428.17 3.99 · 103 0.344 5.25 · 10−5 3.83 · 10−25 - - 3He −−−→REC 3H 18.58 174.00 0.031 3.03 · 10−7 1.58 · 10−25 - - 0.029 1.20 · 10−25 64Zn −−−→REC 64Cu −−→EC 64Ni 587.50 5.47 · 103 8.14 · 10−5 - - 0.615 2.99 · 10−30 171Yb −−−→REC 171Tm 149.91 1.40 · 103 0.911 2.41 · 10−7 5.69 · 10−26 - - 63Cu −−−→REC 63Ni 73.92 687.79 0.610 9.29 · 10−9 3.02 · 10−27 - - 107Ag −−−→REC 107Pd 55.03 512.18 0.963 1.94 · 10−13 1.08 · 10−30 - -

TABLE I. Example targets for resonantly capture cosmic neutrinos. Note that the RBβ initial states have 0 electrons, whilst the 2 and 3-state REC systems have 1 and 2 initial electrons respectively. When calculating ND and NF we use fc,i = 1, δb = 0 0 and δν = 0.00291, corresponding to mν = 0.1 eV and Tν = Tν . Input values are taken from [29] and [32–35].

A. Neutrino mass uncertainty is fixed by nature and δm depends on the progress of neutrino mass experiments, this can only be achieved by As CνB neutrinos are non-relativistic, we require prior modifying the beam momentum spread δb. In particular, knowledge of the neutrino mass to precisely hit the res- (62) has a maximum when onance. However, any direct measurement of the neu- ( trino mass will have some uncertainty which, if left un- 0, |δm| < δν δb = (63) accounted for, will lead to the ion beam being tuned to p 2 2 δm − δν , |δm| > δν the incorrect energy. Supposing that the experiment is run with the assumption of a neutrino mass mν,pred that where we have included the dependence of Rτ on δν and differs from the true neutrino mass, mν,true, the event δb. We show the effect of the neutrino mass uncertainty rate is modified to on the effective quality factor Rτ,eff in Figure 6. We also 2 show the contour |δm| = δν using dotted lines, and we δm Reff R 2 2 − 2(δν +δ ) | | = (1 − δm)e b , (61) show the optimal value of δb for δm > δν and a given NT NT δm using solid white contours. As a consequence of δ being a signed quantity there where we continue to assume a Gaussian distribution for m is a slight asymmetry of (62) in Figure 6. This results Eeν and define the signed fractional uncertainty on the mν,true mν,pred in Rτ,pred having a maximum at δm < 0, suggesting that neutrino mass δm = − , whilst R/NT is given mν,true it is better to underestimate the neutrino mass in this by (35). This results in an effective quality factor given experiment, as the higher beam energy will allow for more by, neutrinos in the lower energy tail of the distribution to

2 cross the interaction threshold. δm 2 − 2(δ2 +δ2) Rτ,eﬀ = Rτ (1 − δm) e ν b , (62) B. Number of targets where Rτ takes the form in (36) with mν = mν,true. Then in order to avoid exponential suppression of the event rate, one needs to ensure that the exponent appearing A large number of highly charged ions on an ener- in both (61) and (62) is less than unity. Given that δν getic beam can damage the magnets and the beampipe 12

be controlled the large number of ions necessary to observe resonant neutrino capture events emit a significant amount of synchrotron radiation, which may also cause damage to the experiment. This is due to the heating of the beam pipe through absorption and the emission of molecules and electrons from the beam pipe walls. Here we put a very conservative upper limit on the number of target ions that can be accelerated before the heat induced by the synchrotron radiation melts the beam pipe, the so-called “death beam limit”. We consider a simplified scenario in which the current in the beampipe is constant, allowing us to neglect ohmic heating. Such an experiment would have a beam of constant number density, as opposed to bunches of particles. The heating of the beampipe walls will then be exclusively due to incident synchrotron radiation, which is emitted with total power FIG. 6. Effect of the neutrino mass uncertainty on Rτ , with 2 4 respect to its maximum value. For simplicity, we set R = 2αI γ τ P = NT (66) p 2 2 tot 2 1/ δν + δb . As before, we use δν = 0.00291, corresponding 3Rc 0 to mν,true = 0.1 eV and Tν = Tν . for an ion accelerator ring with radius Rc and target ions with degree of ionisation I. The synchrotron radiation is ' through the loss of beam particles and synchrotron radi- focused in a cone with opening angle θ 2/γ, such that ation. Damage from beam loss is reduced by the colli- it strikes the wall of the beampipe with power per unit mator system, which prevents quenching of the magnets area by removing ions from the beam halo. The performance γ αI2γ5 Parea ' √ 3 √ Ptot = √ 7 √ NT , (67) of the collimators can be measured by the local cleaning 2 2 4 2πRc r 6 2πRc r inefficiency ηC , which is the fraction of the beam ions leaked to sensitive components of the experiment over where r denotes the beampipe radius. the number of ions absorbed by the collimators per unit We next assume that after some time the inner wall of length [36]. The cleaning inefficiency needs to be low the beampipe reaches an equilibrium temperature T , ∞ enough such that a beam with NT ions and energy E which should be chosen to avoid damage to the experi- does not exceed the quench limit, ment, whilst the outer walls of the beampipe are in con- tact with some coolant held at temperature Tc. Using 3 2 8 1 1 7 TeV Fourier’s law and considering only radial heat flow, the Rq(E) = 7.8 · 10 m− s− , (64) E heat conducted away from the beampipe per unit area will then be given by where the reference value has been determined for a pro- dT κcon ton beam in [37] and the scaling with beam energy in [38]. qcon = κcon = (T − Tc), (68) We are interested in the maximum number of ions that dl ∆ ∞ can be put on a beam for a time τb before the magnets for a beampipe of thickness ∆ with constant thermal con- quench, ductivity κcon. The loss of heat per unit area from the beampipe is given by τb NT = Rq(E) (65) 4 4 ηC qrad = εσ(T + Tc ) , (69) ∞ 3 5 1 2 15 h τb i 10− m− 10 TeV where ε and σ are the emissivity of the pipe walls and the ' 2.74 · 10 . Stefan-Boltzmann constant respectively. The last step is 100 h ηC E to solve the equilibrium condition To derive this number we have used the cleaning ineffi- aParea = qcon + qrad ≡ qout(T ,Tc) (70) ciency achieved by the LHC collimator system [36] and ∞ assume a beam lifetime of τb = 100 h [39]. An experi- for NT , where a ∈ [0, 1] is the absorptance that accounts ment designed to observe relic neutrinos is expected to for incomplete absorption of synchrotron radiation by the have fewer beam losses because no beam collisions take beampipe. We find that the maximum number of ions place. In order to significantly exceed the limit (65) how- that can be accelerated for equilibrium temperature T ever, substantial progress on increasing both beam sta- is then given by ∞ bility and cleaning efficiency is required. Additionally, √ 7 √ 2 5 the scaling with the beam energy reinforces the need for 6 2πRc r mν NT = qout(T ,Tc), (71) small Q-values. Even if ion losses from the beam can αI2a Q ∞ 13 where we have used γ = Q/mν . This further emphasises energy and momentum conservation we have that a small Q-value is the most important parameter for the choice of target ions in this experiment. Er = Ef + Eν,f (76) 2 2 2 Using (71) we can derive a crude estimate for an exam- pν,f = p + pf − 2p pf cos(ϕ), (77) ple of a 27 km ion accelerator ring, with beampipe radius r = 5 cm and thickness ∆ = 1 cm, made out of aluminium where ϕ is the angle between the initial and final state 1 1 for which κcon = 240 W m− K− , ε = 0.03 and a = 0.1. ions. We can then use (77) to derive the constraint We further set the coolant temperature to zero, Tc = 0 K and the equilibrium temperature just below the melting p2 + p2 − p2 f ν,f ≤ point of aluminium at T = 900 K. This gives an upper 1, (78) 2p pf limit on the number of ions∞ that can be put on the beam,

where pν,f is obtained from (76). Expanding to leading 2 5 5 17 50 h mν i 1 keV order in small quantities we find NT ' 3.88 · 10 , (72) Z 0.1 eV Q |pf − p| |Rc,P − Rc,F | 2QDF = . , (79) assuming a fully ionised beam such that I = Z. This p Rc,P M example would correspond to a beam energy per nucleon where QDF is the Q-value of the D → F decay. It is E/A ' 10 TeV with mν = 0.1 eV and Q = 1 keV. In order to achieve the much larger numbers necessary for important to note that typically Rc,F < Rc,P . For 3- the real world examples presented in Section V, signifi- state systems, we are therefore limited to using systems cant additional measures such as beam pipe cooling and with QDF satisfying photon stoppers are needed. M r QDF < . (80) 2 Rc,P C. Non-constant charge radius The different charge radii could also serve as a method for extracting F states without stopping the beam. Different members of the isobars in the 2 and 3-state systems considered for resonant neutrino capture can have different charge radii. If the difference in the charge VII. CONCLUSIONS radii is larger than the beampipe radius, the resonance states are eventually lost from the beam due to colli- Attempts at detecting the CνB are fraught with a sions with the beampipe walls. For 3-state systems, we series of seemingly insurmountable challenges, namely have the further constraint that the parent and final state the tiny neutrino interaction cross sections and huge en- charge radii must not differ by more than the beampipe ergy thresholds required for interactions to take place. radius. We can express these conditions as Neutrino capture on beta decaying nuclei has no energy threshold, and represents the most promising method of |Rc,P − Rc,D| < r, 2 and 3-state systems (73) detecting relic neutrinos today. However, the cross section for neutrino capture is suppressed by G2 , and the |Rc,P − Rc,F | < r, 3-state systems (74) F natural beta decay produces an unavoidable background. where P , D and F denote the parent, daughter and final Accelerating targets on a beam can enhance the neu- state ions, respectively. For magnetic field strength B trino capture cross section whilst also allowing for neu- the charge radius for state with 3-momentum magnitude trino capture on radioactively stable targets. We have performed a calculation of the expected neutrino capture pi is given by rates for a number of experimental realisations of this pi idea and discussed the challenges associated with accel- R = √ . (75) c 2BI πα erating a large number of ions. We first explored the possibility of accelerated tritium If we consider the CνB neutrinos to be at rest in the ions, the PTOLEMY-on-a-beam experiment, aiming to lab frame, then by conserving momentum we trivially exploit the quadratic growth of the capture cross section recover that the resonance momentum pr = p such that with energy. While the neutrino capture cross section the charge radii for the parent and daughter states are grows substantially above PeV beam energies, this exper- equal and condition (73) is always satisfied. iment would have a huge beta decay background. Utilis- The scenario for 3-state systems is somewhat more ing the inverse process 3He → 3H on a beam overcomes complicated, as the decaying resonance will impart some this problem, but still requires beam energies of a PeV or fraction of its momentum to the outgoing neutrino. Fur- more to substantially increase the neutrino capture cross ther, the final state momenta in the 1 → 2 decay system section. are not fully determined, and we can only put bounds Our main finding is that resonant neutrino capture on the magnitude of the final state momentum pf . From with an accelerator experiment allows for significantly 14 larger neutrino capture rates that are linear in GF , and provides a distinct final state. For a neutrino mass of mν & 0.1 eV, we have shown that an appropriate choice of targets can resonantly capture neutrinos using beam energies not significantly higher than those currently at- tainable at the LHC. Moreover, the neutrino capture cross sections for these resonant processes scale quadratically with the inverse of the energy threshold of the process, favouring experiments at accelerators with lower beam energies. The resonance states are unstable, but can be observed through decays into a stable final state different from the initial state. We show how to find isobar triplets with the required properties to be 3-state system candidates. We 1have calculated the expected population of resonance and final states on the beam for both 2 and 3-state systems as a function of the experimental runtime and provide explicit examples. Even though the experiment can be performed in prin- FIG. 7. Dependence of the bound beta decay branching ratio ciple, significant experimental challenges remain. The on K(Q, Z). We choose A = 2Z, nf = 2 and set QCβ = biggest challenge is to find an ion with threshold as small QBβ = Q. as possible Q . 10 keV. Because of the associated synchroton radiation, it is difficult to achieve the large number of ions on the beam required to achieve reasonable Appendix A: Bound beta decay rates statistics, which far exceeds the number on any existing accelerator experiment. Systems with lower energy For the ratio of continuous to bound beta decay, we use thresholds are again favoured, as the total power emit- the expression in [11] for an electron captured in a shell ted by synchroton radiation scales like the fourth power with principal quantum number n with nf free orbitals, of the energy threshold. In contrast to a collider exper- 3 2 iment, the ions need not collide, which may allow for a ΓCβ n 1 f(Z,QCβ) me = 2.87+(6.2 10−3)Z , (A1) modified design with more ions on the beam. ΓBβ α nf π Z · QBβ As precisely hitting the resonance requires knowledge of the neutrino mass, a future measurement with low un- where QCβ is the Q-value for continuous beta decay and certainty would allow us to considerably refine the predic- we define the function tions made in this work. Further exploration of systems Q capable of resonantly capturing neutrinos may also yield 1 Z − 2 f(Z,Q) = 5 dEek,e(Q Eek,e) (Eek,e + me) (A2) promising results. me We have not identified a system capable of discovering 0 q the CνB with current technology. However, the systems 2 × Eek,e + meEek,eF (Z, Eek,e + me). proposed for resonant neutrino capture here are not an − extensive list and it is highly likely that a much bet- As we only consider beta-decaying states that are either ter system can be found. Ongoing experiments like KA- fully ionised or have one electron, the branching ratios in TRIN will provide further input within the next 5 years Section IV depend on the ratio which could make resonant neutrino capture a promising method of discovering the cosmic neutrino background. ΓCβ K(Z,Q) = . (A3) The resonant neutrino capture processes we discussed ΓBβ n=1 could be utilised with exotic systems or an entirely different experimental setup. Additionally, this technology We show the dependence of the bound beta decay branch- has the potential to detect other sources of low energy ing ratios on K(Q, Z) in Figure 7, which displays a strong neutrinos, such as those from thermal solar processes or dependence on the Q-value and the atomic number Z. new physics such as majoron decays [40].

Appendix B: Generation of the solar neutrino ﬂux

VIII. ACKNOWLEDGEMENTS For our calculations of the solar neutrino ﬂuxes given in Figure 1, we assume the solar chemical composition Martin Bauer wants to thank Yuval Grossmann for of [21], which we refer to as the low-metallicity (LZ) encouraging discussions on this topic. We further thank standard solar model herein. The LZ model is able to Stewart Boogert for inventing the “death beam limit”. accurately reproduce spectroscopic data, however, it falls 15

−2 −1 −2 −1 Source Ek,max (keV) φLZ (cm s ) φHZ (cm s )

pp 420 6.03 × 1010 5.98 × 1010 7Be 384 (es), 862 (gs) 4.50 × 109 4.93 × 109 13N 1198 2.04 × 108 2.78 × 108 15O 1732 1.44 × 108 2.05 × 108 pep 1440 1.46 × 108 1.44 × 108 8B 1.70 × 104 4.50 × 106 5.46 × 106 17F 1738 3.26 × 106 5.29 × 106 Hep 1.88 × 104 8.25 × 103 7.98 × 103

TABLE II. Non-thermal solar neutrino ﬂuxes at Earth. The abbreviations gs and es refer to the decays of 7Be to the ground state and excited state of 7Li respectively.

FIG. 8. Temperature and plasma frequency profiles of the sun. into tension with helioseismology [41, 42]. On the con- trary, the model given in [43] uses a higher solar metallicity (HZ) and is in better agreement with helioseismology. from transverse (T) plasmon decay reaching Earth is As such, we show the predicted neutrino fluxes from both P 4 3 2 2 the LZ and HZ models in Table II for completeness. Im- dφe i |Uei| R CV GF portantly, we note that as the predictions are all of the = 3 2 (B2) dEk,ν 64π α (1AU) same order of magnitude, the conclusions of this work T 1 are unaffected by the choice of model. Z Z∞ `2ω (`)6 (ω − 2E )2 × p k,ν d` dω ω/T (`) 1 + 2 2 , Considering the three body decay of an unstable ion in e − 1 ω − ωp(`) 15 15 + the sun (e.g O → N + e + νe) where the final state 0 ω0(x) nuclear recoil and small neutrino mass are neglected, the associated electron neutrino flux in the lab frame is where the 1AU denotes one astronomical unit, R is the solar radius, and T (`) is the solar temperature. The lower bound of the integral can be written in terms of

dφ 2 the plasma frequency ωp(`) as 'N C(Ee)F (Z,Ee)E (Q − Ek,ν + me) dE + k,ν k,ν (B1) 2 q ωp(`) 2 2 ω0(`) = Ek,ν + . (B3) × (Q − Ek,ν + me) − me, 4Ek,ν where the positron energy is given by Ee ' me+Q−Ek,ν , Similarly, one finds for longitudinally polarised (L) plas- C(Ee) is the nuclear shape factor and N is a normalisa- mons tion factor, calculated numerically such that the inte- 1 P 4 3 2 2 Z 2 7 grated fluxes agree with those in Table II. For allowed dφe i |Uei| R CV GF x ωp(`) P (`) = d` , and superallowed transitions the shape factor is approx- 3 2 ωp(`)/T (`) dEk,ν L 64π α (1AU) e − 1 imately constant with energy, and so can be safely ab- 0 sorbed into the normalisation factor [44–46]. (B4) In addition to those neutrinos generated by decays in where the sun, we also consider the flux of thermal solar neu- 2 + 3y(`)2 − 6y(`)4 + y(`)6 P (`) = + y(`)2 log |y(`)| , trinos, which dominate the spectrum in the eV energy 12 range. The largest such contribution comes from plas- (B5) mon decays γ → νν¯ , for which the flux differs depend- ing on the polarisation of the plasmon. A full treat- y(`) = (2Ek,ν −ωp(`))/ωp(`), and the condition 0 < ων < 2 ment of thermal sources of neutrinos should also consider ωp(`) is also enforced. In both cases, CV ' 0.9263 is the both the bremsstrahlung e → eνν¯ and photo-production effective vector coupling between electrons and electron eγ → eνν¯ processes [47], however, as the neutrino fluxes neutrinos, and we have neglected the small contribution from these sources are mostly subdominant to the plas- from the heavy lepton neutrino flavours. The tempera- mon decay flux we do not consider them here. ture and plasma frequency profiles used are shown in Fig- Following the formalism of [47] and introducing the ure 8 and were calculated using the BS05(AGS,OP) solar dimensionless variable ` = r/R in terms of the distance model outlined in [48], for which the elemental abun- from the centre of the sun r, the flux of electron neutrinos dances are similar to those of the LZ model. 16

Appendix C: Number of ions on a beam by

dND To find the number of each ion state on the beam at a = Rτ NP (x) − ND(x), (C3) given time, we need to take into account both the effects dx dNF of neutrino captures and the decays of unstable states, = BDF ND(x). (C4) which act to populate and empty states respectively. We dx neglect the universal decay of the beam due to continuous Using the initial conditions NP (0) = N0, ND(0) = 0 and beam losses. NF (0) = 0 yields the number of D states on the beam at We calculate the beam populations for a 3-state system a given value of x, with an excited initial state, from which all other systems x can be derived by taking the appropriate limits. The 2N0Rτ κx (1+Rτ +η) ND(x) = sinh e− 2 , (C5) number of P states on the beam at any one time can be κ 2 found by solving the differential equation p 2 where κ = (1 − Rτ − η) + 4BDP Rτ . The number of dNP R 1 BDP F states is given by = − γ + NP (t˜) + ND(t˜), (C1) dt˜ NT τP ∗ τD B x N0 DF Rτ (1+Rτ +η) NF (x) = 1 − e− 2 (C6) in terms of the beam rest frame time t˜= t/γ. The factor η + Rτ (1 − BDP ) of γ appearing before the term proportional to R is due to 1 + Rτ + η κx κx × sinh + cosh . the increased neutrino flux in the beam rest frame. This κ 2 2 can be re-expressed in terms of dimensionless parameters, the quality factor Rτ , the parent-daughter lifetime ratio To recover the 2-state system equations, one simply sets η and x as BDF = 0. The ground state systems can be obtained from (C5) and (C6) by taking the limit η → 0 and re- dNP placing BDP with BDP /χ. For ground state systems we = − (Rτ + η) NP (x) + BDP ND(x) , (C2) dx divide the branching ratio BDP by the probability to decay into the initial state isomer χ as excited state isomers where we have used Γ = 1/τD. The analagous differential produced by resonance decays on the beam will quickly equations for the number of D and F states on the beam decay to the ground state isomer, becoming available to in terms of the same dimensionless parameters are given capture neutrinos again.

[1] J. Formaggio and G. Zeller, Rev. Mod. Phys. 84, 1307- [13] J. Bernstein, L. S. Brown and G. Feinberg, Rev. Mod. 1341 (2012) [arXiv:1305.7513 [hep-ex]]. Phys. 61, 25 (1989) [2] S. Weinberg, Phys. Rev. 128, 1457 (1962). [14] A. Ringwald and Y. Y. Y. Wong, JCAP 12, 005 (2004) [3] A. Osipowicz et al. [KATRIN], [arXiv:hep-ex/0109033 [arXiv:hep-ph/0408241 [hep-ph]]. [hep-ex]]. [15] P. F. de Salas, S. Gariazzo, J. Lesgourgues and S. Pastor, [4] E. Baracchini et al. [PTOLEMY Collaboration], JCAP 09, 034 (2017) [arXiv:1706.09850 [astro-ph.CO]]. arXiv:1808.01892 [physics.ins-det]. [16] P. Mertsch, G. Parimbelli, P. F. de Salas, S. Gariazzo, [5] E. Andreotti, C. Arnaboldi, P. De Bernardis, J. Beyer, J. Lesgourgues and S. Pastor, JCAP 01, 015 (2020) C. Broﬀerio, M. Calvo, S. Capelli, O. Cremonesi, C. Enss [arXiv:1910.13388 [astro-ph.CO]]. and E. Fiorini, et al. Nucl. Instrum. Meth. A 572 (2007), [17] R. Lazauskas, P. Vogel and C. Volpe, J. Phys. G 35, 208-210 025001 (2008) [arXiv:0710.5312 [astro-ph]]. [6] D. Fargion, B. Mele and A. Salis, Astrophys. J. 517 [18] A. Faessler, R. Hodak, S. Kovalenko and F. Simkovic, (1999), 725-733 [arXiv:astro-ph/9710029 [astro-ph]]. EPJ Web Conf. 71, 00044 (2014) [7] B. Eberle, A. Ringwald, L. Song and T. J. Weiler, Phys. [19] N. Y. Gnedin and O. Y. Gnedin, Astrophys. J. 509, 11-15 Rev. D 70 (2004), 023007 [arXiv:hep-ph/0401203 [hep- (1998) [arXiv:astro-ph/9712199 [astro-ph]]. ph]]. [20] G. Alimonti et al. [Borexino], Nucl. Instrum. Meth. A [8] C. Hagmann, AIP Conf. Proc. 478 (1999) no.1, 460-463 600, 568-593 (2009) [arXiv:0806.2400 [physics.ins-det]]. [arXiv:astro-ph/9902102 [astro-ph]]. [21] Asplund, Martin and Grevesse, Nicolas and Sauval, [9] V. Domcke and M. Spinrath, JCAP 06 (2017), 055 A. Jacques and Scott, Pat, Annual Review of Astronomy [arXiv:1703.08629 [astro-ph.CO]]. and Astrophysics,47, 1 (2009) [10] A. Ringwald and Y. Y. Y. Wong, [arXiv:hep-ph/0412256 [22] K. Akita, S. Hurwitz and M. Yamaguchi, [hep-ph]]. [arXiv:2010.04454 [hep-ph]]. [11] R. Oldeman, M. Meloni and B. Saitta, Eur. Phys. J. C [23] M. G. Betti et al. [PTOLEMY], JCAP 07, 047 (2019) 65, 81-87 (2010) [arXiv:0905.1029 [hep-ph]]. [arXiv:1902.05508 [astro-ph.CO]]. [12] A. J. Long, C. Lunardini and E. Sabancilar, JCAP 08, [24] C. Yanagisawa, Frontiers in Physics, 2, 30 (2014) 038 (2014) [arXiv:1405.7654 [hep-ph]]. [25] N. Cabibbo and L. Maiani, Phys. Lett. B 114, 115-117 17

(1982) [37] J. B. Jeanneret, D. Leroy, L. Oberli and T. Trenkler, [26] P. Langacker, J. P. Leveille and J. Sheiman, Phys. Rev. CERN-LHC-PROJECT-REPORT-044. D 27, 1228 (1983) [38] C. Bracco, CERN-THESIS-2009-031. [27] L. Stodolsky, Phys. Rev. Lett. 34, 110 (1975) [erratum: [39] A. Abada et al. [FCC], Eur. Phys. J. ST 228 (2019) no.4, Phys. Rev. Lett. 34, 508 (1975)] 755-1107 [28] M. G. Aartsen et al. [IceCube], Nature 591, no.7849, [40] Z. Chacko, P. Du and M. Geller, Phys. Rev. D 100, no.1, 220-224 (2021) 015050 (2019) [arXiv:1812.11154 [hep-ph]]. [29] Live Chart of Nuclides [41] N. Vinyoles, A. M. Serenelli, F. L. Villante, S. Basu, https://nds.iaea.org/relnsd/vcharthtml/ J. Bergström, M. C. Gonzalez-Garcia, M. Maltoni, VChartHTML.html C. Penea-Garay and N. Song, Astrophys. J. 835, no.2, [30] A. Abada et al. [FCC], Eur. Phys. J. ST 228, no.4, 755- 202 (2017) [arXiv:1611.09867 [astro-ph.SR]]. 1107 (2019) [42] M. Agostini et al. [BOREXINO], Nature 562, no.7728, [31] J. Jowett, Proceedings, 9th International Particle Accel- 505-510 (2018) erator Conference (IPAC 2018): Vancouver, BC Canada [43] N. Grevesse and A. J. Sauval, Space Sci. Rev. 85, 161-174 (2018). (1998) [32] I. Ahmad, F. T. Porter, M. S. Freedman, R. F. Barnes, [44] A. G. Cocco, G. Mangano and M. Messina, JCAP 0706, R. K. Sjoblom, F. Wagner, J. Milsted and 015 (2007) [hep-ph/0703075]. P. R. Fields, Phys. Rev. C 3 (1971), 390-407 [45] A. G. Cocco, G. Mangano and M. Messina, J. Phys. Conf. doi:10.1103/PhysRevC.3.390 Ser. 110, no.8, 082014 (2008) [33] M. Cardona and L. Ley, Photoemission in Solids I: Gen- [46] H. Behrens and W. Bühring,Nucl. Phys. A 162, 111-144 eral Principles, 1978 (1971) [34] John C. Fuggle, Nils M˚artensson,Core-level binding en- [47] E. Vitagliano, J. Redondo and G. Raffelt, JCAP 12, 010 ergies in metals, Journal of Electron Spectroscopy and (2017) [arXiv:1708.02248 [hep-ph]]. Related Phenomena, Volume 21, Issue 3, 1980, Pages [48] J. N. Bahcall, A. M. Serenelli and S. Basu, Astrophys. 275-281, J. Lett. 621 (2005), L85-L88 [arXiv:astro-ph/0412440 [35] Gwyn Williams, https://userweb.jlab.org/~gwyn/ [astro-ph]]. ebindene.html [36] S. Redaelli, [arXiv:1608.03159 [physics.acc-ph]].