MI-TH-1934

PASSAT: Particle Accelerator helioScopes for Slim Axion-like-particle deTection

Walter M. Bonivento,1, ∗ Doojin Kim,2, 3, † and Kuver Sinha4, ‡ 1Istituto Nazionale di Fisica Nucleare, Sezione di Cagliari, Cittadella Universitaria di Monserrato, Monserrato (CA), Italy 2Department of Physics, University of Arizona, Tucson, AZ 85721, USA 3Mitchell Institute for Fundamental Physics and Astronomy, Department of Physics and Astronomy, Texas A&M University, College Station, TX 77843, USA 4Department of Physics and Astronomy, University of Oklahoma, Norman, OK 73019, USA We propose a novel method to search for axion-like particles (ALPs) at particle accelerator ex- periments. ALPs produced at the target via the Primakoff effect subsequently enter a region with a magnetic field, where they are converted to photons that are then detected. Dubbed Particle Accel- erator helioScopes for Slim Axion-like-particle deTection (PASSAT), our proposal uses the principle of the axion helioscope but replaces ALPs produced in the Sun with those produced in a target material. Since we rely on ALP-photon conversions, our proposal probes light (slim) ALPs that are otherwise inaccessible to laboratory-based experiments which rely on ALP decay, and complements astrophysical probes that are more model-dependent. As a first application, we reinterpret existing data from the NOMAD experiment in light of PASSAT, and constrain the parameter space for ALPs lighter than ∼ 100 eV and ALP-photon coupling larger than ∼ 10−4 GeV−1. As benchmarks of feasible low-cost experiments improving over the NOMAD limits, we study the possibility of re- using the magnets of the CAST and the proposed BabyIAXO experiments and placing them at the proposed BDF facility at CERN, together with some new detectors. We find that these realizations of PASSAT allow for a direct probe of the parameter space for ALPs lighter than ∼ 100 eV and ALP-photon coupling larger than ∼ 4 × 10−6 GeV−1, which are regions that have not been probed yet by experiments with laboratory-produced ALPs. In contrast to other proposals aiming at de- tecting single or two-photon only events in hadronic beam dump environments, that rely heavily on Monte Carlo simulations, the background in our proposal can be directly measured in-situ, its sup- pression optimized, and the irreducible background statistically subtracted. Sensitivity evaluations with other beams will be the subject of a future paper. The measurements suggested in this paper represent an additional physics case for the BDF at CERN beyond those already proposed.

I. INTRODUCTION AND MOTIVATIONS photons via ALP-photon conversion. We refer to Ref. [10] for a recent review of these topics. The QCD axion [1–3] and more general pseudo-scalar In contrast, particle accelerator-based experiments axion-like-particles (ALPs), which are ubiquitous in have so far been proposed to search for heavy (i.e., string theory [4, 5], are a major focus of searches for MeV – GeV-scale) ALPs, since they are relatively short- physics beyond the Standard Model (SM). The experi- lived and their decay point is not much displaced from mental ecosystem investigating these particles is vast and their production point. Examples include NA62 [11], rich. The techniques often rely on the ALP-photon cou- SHiP [12], FASER [13, 14], and SeaQuest [15]. In such ex- pling: periments, an ALP is first produced in the target/dump material and subsequently decays to photons in the de- 1 µν cay volume. As the ALP mass is lowered, it exits the Lint ⊃ − gaγγ aF F˜µν , (1) 4 decay volume without decaying and these experiments lose their sensitivity. We refer to [16–19] for recent theo- where g is the coupling between ALP (henceforth de- aγγ retical studies on ALP searches. arXiv:1909.03071v2 [hep-ph] 19 Feb 2020 noted by a) and the SM photon and where F (F˜ ) is µν µν The purpose of this paper is to point out that if the the usual field (dual field) strength tensor of the photon. ALP enters a region with a transverse magnetic field after ALPs can convert to photons and vice versa in the pres- being produced at the target, then a beam dump becomes ence of an external magnetic field, leading to experiments sensitive to very light ALPs (see Figure 1). This is be- based on the axion haloscope and the axion helioscope [6– cause the ALP is no longer required to decay; rather, the 8]. On the other hand, photon regeneration (light shining ALP converts to a photon which can be detected. If the through wall, LSW) experiments [9] attempt to actively length traversed by the ALP is shorter than the associ- produce ALPs with a high-intensity laser beam applied ated oscillation length, the conversion is coherent and a in a magnetic field, followed by detecting the produced net probability of conversion can be obtained as a func- tion of the ALP-photon coupling. The predicted photon signal is a product of the ALP production cross section ∗ [email protected] at the target material and its conversion probability as † [email protected] it subsequently traverses the magnetic field. ‡ [email protected] Dubbed Particle Accelerator helioScopes for Slim 2

II. ALP PRODUCTION AND CONVERSION Beam Target Magnetic field 푎 훾 훾 In this section, we calculate the expected number of events Nex from ALP production via the Primakoff pro- 푁 푁 cess at the target followed by its conversion to a photon in the magnetic field. The quantity Nex is simply given as

Z 2 FIG. 1. Schematic description of the PASSAT search strat- 1 d σa egy under consideration Nex = NPOT · dEadθa · Pa→γ · Psurv , σγ→all dEadθa (2) where NPOT is the total number of protons on target (POT) and σγ→all is the cross section for photon-nucleus Axion-like-particle deTection (PASSAT), our proposal scattering which is dominated by photon conversion to thus combines the principle of the axion helioscope with an electron-positron pair in the nuclear fields. For the traditional ALP production in beam dumps. From the photon energies of our interest, we have σγ→all ≈ 1.2 × 4 2 perspective of a traditional helioscope, the Sun is re- 10 mb in molybdenum and σγ→all ≈ 1.4 × 10 mb in placed by the target material as the source of ALPs; from beryllium [29]. the perspective of a traditional beam dump experiment, A couple of probabilities are involved in this signal the ALP decay process is replaced by the ALP conversion rate calculation. First, Pa→γ stands for the probability of process. ALP-to-photon conversion when an ALP travels distance L in a magnetic field B: While its expected reach cannot rival that of the CAST helioscope, PASSAT outperforms other searches  2  2   gaγγ BL 2 2 qL for laboratory-produced ALPs. We stress that it is im- Pa→γ = sin , (3) portant to pursue laboratory-based experiments to probe 2 qL 2 the ALP parameter space, even in regimes that are con- where the product of the second and third factors is the strained by helioscopes and astrophysical sources. As form factor reflecting the coherence of the conversion. In pointed out in the aftermath of the PVLAS signal [20], the relativistic limit and in vacuum, q is given by there are many assumptions that lie behind constraints given by solar helioscopes (we choose [21–23] among the s  2 2  2 many papers that make this point). The environmental ma 1 q = 2 + gaγγ B . (4) conditions for the production of ALPs inside the Sun or 4Ea 2 in stars are very different from those in laboratories. The coupling gaγγ or the mass ma can depend on a host of en- The other one Psurv describes the survival probability vironmental parameters, such as the temperature, matter that an ALP reaches the detector before decaying into a density, or plasma frequency, as well as the momentum photon pair. The usual decay law suggests transfer at the ALP-photon vertex. Thus, laboratory- ! based searches, apart from being complementary to as- L˜ trophysical searches, are also more conservative. Psurv = exp − lab , (5) La We begin our work by discussing ALP production and conversion at PASSAT, and then apply the resulting for- where L˜ is the distance between the target and the detec- lab malism to several implementations in past, current, and tor and La is the laboratory-frame mean decay length future experimental facilities to obtain an initial estimate of the ALP. In terms of the Lorentz boost factor of ALP lab of the expected performance. We first reinterpret exist- γa and the decay width of ALP Γa, La is given by ing data from the NOMAD experiment [24] in light of p 2 PASSAT. We then discuss the possible future implemen- lab γa − 1 c tation of PASSAT at CERN, combining two components: La = , (6) Γa (i) a permanent fixed target complex composed of the beam line and target infrastructure from the proposed where c is the speed of light and Γa associated with the Beam Dump Facility (BDF) [25] and the muon shield diphoton mode is component from the SHiP detector [26] (ii) followed by the CAST [27] or BabyIAXO [28] magnets, together with g2 m3 Γ = aγγ a . (7) some new detectors. In each case, we provide the experi- a 64π mental parameters and the projected sensitivities, includ- ing the future potential of upgrading LHC-like magnets The integrand in Eq. (2) describes the differential ALP up to 20 T, as projected by studies related to the FCC-hh production cross section via the Primakoff process con- collider. voluted with a differential photon number density profile 3

2 max nγ , in ALP energy Ea and its outgoing angle θa from the Exp. B [Tesla] L [m] A [cm ] θa [mrad] beam axis. This is given by NOMAD [33] 0.4 7.5 3.8 × 104 2.1 CAST [27] 8.4 9.26 14.5 0.36 2 Z 3 d σa 2 2 dσγN BabyIAXO [28] 2 10 7.7 × 10 8.3 = dpT dφ nγ (Ea, pT ) , (8) dEadθa dθa TABLE I. Experimental parameters for the magnetic field where φ denotes the angle in the transverse plane between area of the benchmark experiments. The strength of B for the incoming photon and the outgoing ALP and σγN is BabyIAXO is the claimed average value. The maximum an- the cross section for the Primakoff process. We note that max gular acceptance θa is calculated with respect to the full nγ is a function over the photon transverse momentum distance between the target and the photon detector. See the pT and we have substituted the Eγ dependence in nγ text for details. with Ea in the collinear limit, i.e., Ea ≈ Eγ . In the massless (or ultra relativistic) ALP limit, the spectrum describes the differential π0 number density σγN is approximately of the form [16, 19] ± dNπ/dEπ. Our parameter choices (cα, cβ, cγ ) for π are 2 4 dσγN 1 2 2 2 (4Eat + ma) (3.45, 1.57, 0.517) [31]. We further assume that Nπ is ≈ − αgaγγ Z F (|t|) 2 θa , (9) dθa 16 t normalized to 4 for Ebeam = 400 GeV considering the measurement data in Ref. [32] and the simulation study where α and Z are the fine structure constant and in Ref. [17]. The pion energy Eπ and the photon energy the atomic number of target material, respectively, and Eγ in the laboratory frame are related by the following 2 where t = −(pγ − pa) is given by Lorentz transformation, 4 ma 2 q 2 2 ∗ p 2 ∗ 2 Eγ = Eγ (γπ + γπ − 1 cos θγ ) , (13) t = − 2 − pT + 2Ea pT θa cos φ − Eaθa . (10) 4Ea ∗ ∗ where E (= mπ/2) and θ are the photon energy and We follow Ref. [16] and choose the Helm form factor as γ γ p the photon emission angle in the pion rest frame, respec- F (|t|) which is assumed to be vanishing for |t|R1 > tively, and where the pion boost factor contains Eπ de- 4.49: pendence such that γπ = Eπ/mπ. Since typical pions are p 2 p  2  highly boosted, γ − 1 ≈ γπ so that we have 3j1( |t|R1) |t|s p π F (|t|) = p exp − Θ(4.49 − |t|R1) , |t|R1 2 mπ Eπ ≈ q Eγ . (14) (11) E∗ + E∗2 − p2 where s = 0.9 fm, Θ(x) is the Heaviside step function, γ γ T and j is the spherical Bessel function of the first kind. 1 The density of p2 , w(p2 ), can be calculated from the Here R is parametrized according to Ref. [30], i.e., R = T T 1 1 density of cos θ∗, w(cos θ∗), that is, p(1.23A1/3 − 0.6)2 + 2.18 fm with A being the atomic γ γ mass number of target material. ∗ 2 d cos θγ ∗ 1 We finally turn to the photon number density profile w(pT ) = 2 w(cos θγ ) = q . (15) dpT ∗ ∗2 2 nγ . In principle, the precise determination of nγ requires 4Eγ Eγ − pT a full detector-level simulation. However, in this first ∗ study, we opt to take a (semi-)analytic approach, based Here w(cos θγ ) = 1/2 as only the forward-moving photon on an empirical model. This enables us to perform rapid contributes to the final estimate out of the two decay estimates, essentially as a proof of principle. products. Taking the collinear limit Ea ≈ Eγ again, Since the ALPs under consideration are sufficiently therefore, we find light, the dominant photon source is the decay of mesons (e.g., π0, η). Production of mesons through high-energy 2 dEπ dNπ 2 nγ (Ea, pT ) = w(pT ) , (16) proton beams on target is elegantly parametrized by the dEγ dEπ so-called BMPT model [31] whose fits were tuned with p- beryllium target collision data. In the limit of negligible where the Jacobian factor can be readily computed from transverse momenta of mesons, we find that the differen- Eq. (14). tial production cross section of say, a charged pion has the form of III. REINTERPRETATION OF PAST  cα   dσ Eπ Eπ EXPERIMENTS ∝ Eπ 1 − 1 + cβ dEπ Ebeam Ebeam  −cγ We first discuss the NOMAD experiment since it has Eπ × , (12) experimental data and possesses the main features of E beam PASSAT, a fact that was not appreciated in the past. where Ebeam denotes the particle beam energy. We The ALP search at NOMAD [24] assumes that a frac- assume that the functional behavior of the above tion of the photons produced by a proton beam on the Active Muon Veto Area CAST/BabyIAXO p-beam module 훾-ray detector

50 m L

4

CAST/BabyIAXO 훾-ray beryllium target enters the horn region where they are p-beam Target converted to ALPs. These ALPs are subsequently re- Module detector Active Muon converted back to photons in the NOMAD spectrometer. Shield Area The underlying principle is the LSW class of experiments, with the laser replaced by a beam dump. Like a LSW experiment, it requires two conversion stages in two sep- arate regions with magnetic fields: first from photons to L ALPs, and then ALPs back to photons. 50 m Motivated by the idea of PASSAT applied to NOMAD, BDF SHiP Component we instead focus on the ALPs directly created in the tar- get, jettisoning the first phase of photon to ALP con- FIG. 2. Conceptual design of PASSAT with the CAST or version. ALPs directly produced in the target by the BabyIAXO magnets. The components from the proton beam Primakoff process will subsequently be converted back through the active muon shield area are from the BDF-SHiP to photons in the magnetic field region of the NOMAD complex proposed at CERN. The X-ray-sensitive detector of spectrometer. The distance between the target and the CAST/BabyIAXO is replaced by a γ-ray detector. Some ad- detector is 835 m, and in the conversion region a 0.4 T ditional veto detectors for background suppression may be magnetic field is applied for a length L = 7.5 m. The needed as well. width×height of the magnetic field region is 3.5×3.5 m2. We take a circular area of radius 1.75 m instead of the square cross section for convenience of calculation. The photons in the magnetic field in the shield area, but such produced ALPs should reach the photon detector with- photons will be quickly absorbed to the material. We out being absorbed in other parts of experimental facility. thus neglect any contribution to the final photon count This imposes a maximum allowed value of θa, given ap- arising from conversions in the muon shield area. A frac- proximately by the ratio of radius of cross-sectional area tion ∼ 10−7 of incident ALPs are lost due to conversion A to the entire distance between the target and the (γ- in the muon shield in this manner. ray) detector (see also Table I). Thus, the maximum an- We make a few comments on the input parameters gular acceptance for ALPs in NOMAD-PASSAT is given required to calculate the final expected number of pho- max 20 simply by θa ≈ 1.75/(835 + 7.5) ≈ 2.1 mrad. The val- tons. Firstly, POT is expected to be 2 × 10 in five ues of other key parameters for NOMAD-PASSAT are years. Secondly, when calculating the maximum angular summarized in Table I. acceptances, the length of the muon shield area should be max taken into account; for example, θa for the BabyIAXO case is 0.5/(50 + 10) ≈ 8.3 mrad. The values of the mag- IV. POSSIBLE NEW EXPERIMENTS netic field and lengths of the different conversion mod- ules are tabulated in Table I. Finally, we note that the photon detectors in the CAST/BabyIAXO experiments As far as future experiments are concerned, we propose are designed to be sensitive to X-ray. For our purpose, to recycle the magnets from CAST or BabyIAXO exper- the detector is replaced by a γ-ray photon detector (e.g., iments, after they are decommissioned, and locate them calorimeter). at the BDF complex, possibly after its first use with the SHiP experiment. The BDF project, which is currently in its planning phase, will be housed in the North Area of CERNs Prevessin site and utilize the 400 GeV Super V. EXPERIMENTAL SENSITIVITIES Proton Synchrotron (SPS) proton beam. This compo- nent will provide the beam line and target infrastructure. From Eq. (2), the expected number of events can be It is to this component that the muon shield from SHiP obtained by integrating the differential cross section in 2 will be added, forming the BDF-SHiP complex. Eq. (8) over Ea, θa, φ and pT . Firstly, in all our sen- The conceptual design of the experimental setup sitivity calculations, we impose the requirement that that we are envisioning is depicted in Figure 2. The Ea > 50 GeV. Given the fact that the produced mesons BDF/SHiP module includes the proton beam, target are forward-directed and not much transverse [31], this complex, and active muon shield. In our study we as- implies that the momenta of incoming photons and in sume that the beam energy is 400 GeV and the target in turn their parent mesons along the beam axis dominate the core of the shower is molybdenum (which is contained so significantly that one can neglect their transverse mo- in TZM alloy) like BDF for purposes of illustration. The mentum. In other words, we restrict our initial estimates produced ALPs traverse the muon shield area and enter to the phase space where the negligible transverse mo- the bore of CAST/BabyIAXO where the ALP-photon mentum approximation and the limit of ultra relativistic conversion occurs. mesons are sufficiently valid. The unconsidered phase We note that the iron-filled active muon shield area it- space can provide an additional contribution to the sig- self comes with a 50 m-long transverse magnetic field re- nal sensitivities of interest, and therefore, our estimate gion of approximately 1.8 T. Some ALPs may convert to here may be understood as being rather conservative. 5

background 272 ± 18 events [24] which can occur in the CAST-PASSAT preshower region or in the upstream region. We consider LEP 0.01 BabyIAXO-PASSAT Ea ≈ Eγ ranging 50 GeV to 140 GeV, while conserva- U®inv.+Γ tively assuming that all expected background events are NOMAD-PASSAT + - e e ®inv.+Γ relevant to this energy range. The limit is computed NOMAD -4

D 10 at 90% C.L. for a given background assumption and its 1 Beam - NOMAD 1 Dump statistical uncertainty. The result is shown by the black-

GeV solid curve in Figure 3. The red-shaded region denotes @ 10-6 LSW ΓΓ the bounds published by the NOMAD Collaboration, as- a g SN1987a suming a LSW interpretation. We see that NOMAD- 10-8 CAST-SUMICO PASSAT constrains a wider range of parameter space, not only covering the existing NOMAD bound but ex- −4 −1 HB stars ploring up to gaγγ ∼ 10 GeV for ma . 0.1 keV. 10-10 Cosmology When it comes to the other possible experiments, we 10-6 0.001 1 1000 106 109 assume that 400 GeV of CERN SPS proton beam is in- cident on a molybdenum target with specifications sim- ma eV ilar to those of the target adopted in the BDF project, FIG. 3. Expected experimental sensitivity that can be as mentioned earlier. We include the contribution from achieved by PASSAT with CAST@ (blueD lines) and BabyI- the η meson decay, simply assuming that the expected 0 AXO (red lines) magnets, in the plane of ALP mass ma and number of photons is roughly 1/10 of the π meson the associated photon coupling gaγγ . The NOMAD ALP case [17, 32], although the experimental sensitivities are search in [24] is reinterpreted in terms of PASSAT as the not much affected by this inclusion. Again, the limits experiment possesses the main features of PASSAT, and the are calculated by 90% C.L. with 2 × 1020 POT that BDF black solid line shows the resulting limit. Dashed lines are aims to achieve [25]. the corresponding sensitivities with prospective B of 20 T for CAST-PASSAT and BabyIAXO-PASSAT. The current bounds summarized in e.g., Refs. [18, 35] are from existing VI. CONSIDERATIONS ON BACKGROUND ALP searches. The prospective limits are estimated by 90% C.L. under the assumptions of 10 (negligible) background events for BabyIAXO-PASSAT (CAST-PASSAT) and 2×1020 At the expected proton beam intensities at the BDF, POT. The NOMAD case is calculated with the expected num- there will still be a large flux of muons coming out of the ber of background events reported in [24], being normalized active muon shield. Therefore, to suppress background 19 to 1.08 × 10 POT. from charged particles and possible neutral particles at the exit point of the shield, a set of properly optimized veto detectors will be needed. To avoid interactions in We will perform a dedicated study including full experi- the bore region of the CAST/BabyIAXO module some mental setup details and the (significant) cascade factor level of vacuum could also be needed. of showering in future work [34]. There will be an irreducible background arising from The photons from ALP conversion in the magnetic field νe elastic scattering (ES) events, mostly coming from the should reach the photon detector without being absorbed decays of charmed particles in the beam dump, inside the in other parts of experimental facility, giving a maximum photon detector. We attempt an order of magnitude es- angular acceptance θa for each benchmark experiment, timate here based on previous related studies that can be summarized in Table I. φ simply ranges from 0 to π. found in [26] and subsequent documents for the 9,600 kg 2 2 neutrino detector of the SHiP experiment. The number Finally, pT spans 0 to mπ/4. Since we are restricted to Ea > 50 GeV, the full pT range is within their angular of ES found there can be scaled to what is expected in acceptance. However, the angular acceptance for CAST PASSAT from the detector mass ratio. We assume for 2 2 2 PASSAT a calorimeter made of (i) an electromagnetic is somewhat limited, so we consider pT ∈ (0, 36 ) MeV in the corresponding calculation. section with appropriate longitudinal segmentation for The curves in Figure 3 display our estimate of selecting interactions in the first four radiation lengths, in the experimental sensitivity and existing bounds from order to keep high efficiency on signal photons and good various laboratory-produced ALP search experiments angular resolution and (ii) a hadronic section aiming at (LSW, NOMAD, Υ → inv. + γ, e+e− → inv. + γ, Beam Dump, and LEP) and astrophysical/cosmological searches (CAST-SUMICO, HB stars, SN1987a, and Cos- 1 mology) in the plane of m and g . We first estimate Since the beryllium target in NOMAD is as thin as 100 mm, a aγγ some fraction of incident protons may not scatter in the target the exclusion limit that the NOMAD experiment would but traverse towards the downstream dump area. We here take a reach with existing data. With a 450 GeV proton beam rather simple analysis scheme, not distinguishing ALPs produced on the beryllium target and 1.08 × 1019 POT, no signifi- at the target with those produced at the downstream complex, cant excess has been observed over the expected neutrino which is beyond the scope of this study. 6 rejecting deep inelastic scattering events. A possible im- our numerical scan shows that it takes place around the plementation of such a calorimeter was first described beam dump limits. We further show the expected exper- in [36] and is now the baseline option for the proposed imental sensitivities with a prospective higher magnetic SHiP experiment. field of 20 T,3 by the dashed but same color-coded lines. Assuming that the neutrino flux is uniformly incident BabyIAXO (CAST) can accomplish sizable (mild) im- on the calorimeter, the expected number of ES events provements in the sensitivities as B is increased by an at BabyIAXO-PASSAT would be 17, obtained as fol- order of magnitude (a factor of ∼ 2). lows: 800 (the number of background events for a dark matter search with electron scattering in [26]) × 200 (weight of a 4X0 lead calorimeter of BabyIAXO mag- VIII. CONCLUSIONS AND DISCUSSION net area) / 9, 600 ≈ 17. The requirement of Ea ≈ E > 50 GeV further reduces 80% − 90% of the ES γ We have proposed a novel method to search for ALPs background [26]. In this analysis we conservatively as- at particle accelerator experiments. Our results suggest sume 10 background events for BabyIAXO-PASSAT. On that PASSAT should probe a wide range of parameter the other hand, for CAST-PASSAT we assume negligi- space that none of the laboratory-produced ALP search ble background as the associated cross sectional area A experiments have ever explored. In particular, the ex- is much smaller, implying an even more suppressed neu- pected experimental sensitivity covers regions explored trino flux entering the calorimeter. by the CAST helioscope experiment, providing a con- It should be stressed that, in contrast to other pro- servative and complementary probe. The experimental posed experiments at beam dumps looking for New sensitivity also extends into regions that are currently Physics particles decaying to neutral particles only, such solely constrained by astrophysical observations (e.g., HB as [37] and [38], for PASSAT it is possible to determine stars). the background directly from the data, by running with We make some final comments on the complementarity the magnet current switched off. It is reasonable that the of our work with respect to constraints from astrophysics. data acquisition can include periods with current on and Generally, such constraints depend on certain underly- off for equal duration. This will also allow us to optimize ing astrophysical assumptions: for supernovas, these as- the setup and evaluate its feasibility at the beginning of sumptions involve details of the core-collapse simulation, the experiment. while for stars they depend on standard stellar models and evolutionary stellar timescales. While it is unlikely VII. RESULTS that these astrophysical assumptions would be relaxed to the level where PASSAT would become competitive with astrophysical constraints, we point out that there Our analysis results with CAST and BabyIAXO mag- has been recent work that hints at this possibility, at nets are exhibited by the blue and red solid curves, least for supernovae (for example [39]). A detailed inves- respectively, in Figure 3. We clearly see that all of tigation of the astrophysics is beyond the scope of this the benchmark experiments promoted to PASSAT show paper. Our focus in this paper, instead, has been to en- equally good capabilities and allow for probing a substan- tertain the following possibility: even if one follows stan- tially broader range of parameter space than explored by dard astrophysics and takes results from both SN1987A past laboratory-produced ALP searches. and HB stars, then it is still possible to construct ALP Beyond ma ≈ 0.01−0.1 keV, the associated oscillation models such that stellar or supernova environments are length of the produced ALP becomes shorter than L and blind to the ALP-photon coupling. Thus, the relaxation the conversion mechanism becomes non-coherent, result- of bounds is not due to astrophysics but rather due to ing in a rising sensitivity line. The conversion process hidden sector model building. Models which evade all essentially competes with the decay process. The proba- stellar bounds (including energy loss arguments) have in bility that the produced ALPs decay before reaching the fact been constructed in the literature and were actively detector, Pdecay is given by studied in the wake of the PVLAS anomaly several years ago. Generally, such models proceed by introducing hid- P = 1 − P . (17) decay surv den sector dynamics that switches off the Primakoff pro- Once this probability becomes comparable to the conver- duction in stellar environments, either through appropri- sion probability in Eq. (3), PASSAT starts to lose the sen- ate choices of hidden sector charges and couplings [21], or sitivity and the signal detection via the ALP decay would through the introduction of phase transitions [40], screen- appear competitive.2 For our benchmark experiments, ing mechanisms [23], or form factors [22]. We do not

3 Although high-B magnets are better incorporated to CAST-like 2 It is interesting per se to probe the ALP parameter space through detectors with a small aperture, we perform optimistic estimates, the decay channel under the PASSAT setup. We reserve this for assuming that the same magnet technology will be available for future work. BabyIAXO in the near future. 7 want to judge whether these models are natural; however, form factors are natural would depend on the details of they do motivate the complementary laboratory-based the model-building. Furthermore, since ALPs are pro- approach based on PASSAT that we have pursued. duced in a target material in PASSAT, the Fermi energy The models constructed to evade astrophysical bounds and non-negligible electron density may become impor- typically make the ALP production mechanism depend tant when considering such models. on the ambient plasma mass and high temperature en- Finally, we note that our beam production and back- vironment in the interior of the star. Thus, the proper- ground are in estimation level; a full simulation is being ties of the plasma become important, and the calculation performed for an upcoming paper [34]. We believe that must be performed after taking into account thermal cor- our estimations are conservative and a full treatment will rections. In the case of PASSAT, the production mech- lead to more robust constraints on the ALP parameter anism is simply associated with the usual beam dump space. We also emphasize that the idea of PASSAT can environment where there are no effects from an ambi- be implemented at other facilities and beam lines (e.g., ent plasma. Thus, models created to evade astrophysical electron beam dumps). We reserve these topics for future bounds and account for the PVLAS anomaly by invoking work. plasma/thermal physics would continue to account for a possible anomaly at PASSAT and evade CAST. We note, however, that the Fermi energy and electron density of ACKNOWLEDGMENTS the target material in which ALPs are produced in PAS- SAT may have important effects on such models, com- We would like to thank Babette D¨obrich, Kwang- pared to typical light shining through wall experiments Sik Jeong, and Pierre Sikivie for their careful reading where ALPs are produced in vacuum. of the draft. We would also like to thank Bhaskar In models where the form factor of the ALP-photon Dutta, Ahmed Ismail, Lucien Heurtier, Felix Kling, Jong- vertex is assumed to be momentum-dependent, the strat- Chul Park, Seodong Shin, and Scott Thomas for use- egy to evade astrophysical bounds and account for the ful comments and discussions. D.K. acknowledges the PVLAS anomaly models the form factor to attenuate for hospitality of the Aspen Center for Physics, which is high momentum (∼ keV inside stars) and strengthen at supported by National Science Foundation grant PHY- low momentum (for lasers). In our case, such form factor- 1607611, while completing this work. The work of D.K. dependent models would have to exhibit a different be- was supported in part by the Department of Energy un- havior: they would have to strengthen at high momen- der Grant DE-FG02-13ER41976 (de-sc0009913) and is tum (∼ GeV that is relevant for PASSAT) and attenuate supported in part by the Department of Energy under at low momentum (∼ keV and below, to evade astrophys- Grant de-sc0010813. The work of K.S. is supported by ical and LSW bounds). Whether or not models with such the Department of Energy under Grant DE-SC0009956.

[1] S. Weinberg, Phys. Rev. Lett. 40, 223 (1978). [13] A. Ariga et al. (FASER), (2018), arXiv:1812.09139 [2] F. Wilczek, Phys. Rev. Lett. 40, 279 (1978). [physics.ins-det]. [3] R. D. Peccei and H. R. Quinn, Phys. Rev. Lett. 38, 1440 [14] A. Ariga et al. (FASER), (2019), arXiv:1901.04468 [hep- (1977), [,328(1977)]. ex]. [4] A. Arvanitaki, S. Dimopoulos, S. Dubovsky, N. Kaloper, [15] A. Berlin, S. Gori, P. Schuster, and N. Toro, Phys. Rev. and J. March-Russell, Phys. Rev. D81, 123530 (2010), D98, 035011 (2018), arXiv:1804.00661 [hep-ph]. arXiv:0905.4720 [hep-th]. [16] B. Dobrich, J. Jaeckel, F. Kahlhoefer, A. Ring- [5] M. Cicoli, M. Goodsell, and A. Ringwald, JHEP 10, 146 wald, and K. Schmidt-Hoberg, JHEP 02, 018 (2016), (2012), arXiv:1206.0819 [hep-th]. [JHEP02,018(2016)], arXiv:1512.03069 [hep-ph]. [6] P. Sikivie, Particle physics and cosmology: Dark matter, [17] B. Dobrich, J. Jaeckel, and T. Spadaro, Submitted to: Phys. Rev. Lett. 51, 1415 (1983), [,321(1983)]. J. High Energy Phys. (2019), arXiv:1904.02091 [hep-ph]. [7] P. Sikivie, Phys. Rev. D32, 2988 (1985), [Erratum: Phys. [18] M. Bauer, M. Heiles, M. Neubert, and A. Thamm, Eur. Rev.D36,974(1987)]. Phys. J. C79, 74 (2019), arXiv:1808.10323 [hep-ph]. [8] V. Anastassopoulos et al. (CAST), Nature Phys. 13, 584 [19] J. L. Feng, I. Galon, F. Kling, and S. Trojanowski, Phys. (2017), arXiv:1705.02290 [hep-ex]. Rev. D98, 055021 (2018), arXiv:1806.02348 [hep-ph]. [9] A. Spector (ALPS) (2019) arXiv:1906.09011 [physics.ins- [20] E. Zavattini et al. (PVLAS), Phys. Rev. Lett. 96, 110406 det]. (2006), [Erratum: Phys. Rev. Lett.99,129901(2007)], [10] P. W. Graham, I. G. Irastorza, S. K. Lamoreaux, A. Lind- arXiv:hep-ex/0507107 [hep-ex]. ner, and K. A. van Bibber, Ann. Rev. Nucl. Part. Sci. [21] J. Jaeckel, E. Masso, J. Redondo, A. Ringwald, and 65, 485 (2015), arXiv:1602.00039 [hep-ex]. F. Takahashi, Phys. Rev. D75, 013004 (2007), arXiv:hep- [11] M. B. Brunetti, F. Gonnella, and L. Iacobuzio (NA62), ph/0610203 [hep-ph]. Universe 4, 119 (2018). [22] M. Ahlers, H. Gies, J. Jaeckel, and A. Ringwald, Phys. [12] S. Alekhin et al., Rept. Prog. Phys. 79, 124201 (2016), Rev. D75, 035011 (2007), arXiv:hep-ph/0612098 [hep- arXiv:1504.04855 [hep-ph]. ph]. 8

[23] P. Brax, C. van de Bruck, and A.-C. Davis, Phys. Rev. [32] Aguilar-Ben´ıtez et al., Z. Phys. C 50, 405 (1991). Lett. 99, 121103 (2007), arXiv:hep-ph/0703243 [HEP- [33] F. Vannucci, Adv. High Energy Phys. 2014, 129694 PH]. (2014). [24] P. Astier et al. (NOMAD), Phys. Lett. B479, 371 (2000). [34] W. M. Bonivento, D. Kim, and K. Sinha, Work in [25] K. Kershaw et al., JINST 13, P10011 (2018), progress. arXiv:1806.05920 [physics.ins-det]. [35] J. Jaeckel and M. Spannowsky, Phys. Lett. B753, 482 [26] M. Anelli et al. (SHiP), (2015), arXiv:1504.04956 (2016), arXiv:1509.00476 [hep-ph]. [physics.ins-det]. [36] W. M. Bonivento (SHiP), [27] K. Zioutas et al., Nucl. Instrum. Meth. A425, 480 (1999), Proceedings, CHEF: Lyon, France, October 2-6, 2017, arXiv:astro-ph/9801176 [astro-ph]. JINST 13, C02041 (2018). [28] E. Armengaud et al. (IAXO), JCAP 1906, 047 (2019), [37] G. Magill, R. Plestid, M. Pospelov, and Y.-D. Tsai, arXiv:1904.09155 [hep-ph]. Phys. Rev. D98, 115015 (2018), arXiv:1803.03262 [hep- [29] M. Tanabashi et al. (Particle Data Group), Phys. Rev. ph]. D98, 030001 (2018). [38] L. Harland-Lang, J. Jaeckel, and M. Spannowsky, Phys. [30] J. D. Lewin and P. F. Smith, Astropart. Phys. 6, 87 Lett. B793, 281 (2019), arXiv:1902.04878 [hep-ph]. (1996). [39] N. Bar, K. Blum, and G. D’amico, (2019), [31] M. Bonesini, A. Marchionni, F. Pietropaolo, and arXiv:1907.05020 [hep-ph]. T. Tabarelli de Fatis, Eur. Phys. J. C20, 13 (2001), [40] R. N. Mohapatra and S. Nasri, Phys. Rev. Lett. 98, arXiv:hep-ph/0101163 [hep-ph]. 050402 (2007), arXiv:hep-ph/0610068 [hep-ph].