X17 search project with EAR2 neutron beam

. We present the state of the art of the n_TOF Collaboration Working Group activity dedicated to study how to solve the puzzle about the existence of the so called new particle X17, spotted for the ﬁrst time few years ago by a team at ATOMKI in Hungary and since then never conﬁrmed by other independent experimental collaborations but also never refuted. An “ad hoc” detection set-up is under realization for this goal, in order to reach an angular resolution of the two emerging trajectories from the X17 decay and an energy resolution for the invariant mass reconstruction enough to cast light in a deﬁnitive way about this puzzle. To design the present detection setup we work in close contact with the Pisa Nuclear Theory team, that has deeply studied the implication of X17 existence and extracted by the ATOMKI results its eventual nature, kinematics and general properties.


Introduction
Two significant anomalies have been recently observed in the emission of electron-positron pairs in the 7 Li(p,ee + ) 8 Be and 3 H(p,ee + ) 4 He reactions [1], [2]. These anomalies have been deciphered as the signature of the existence of a new boson called X17 of mass M X17 16. 8 MeV that could be a mediator, if it exists, of a fifth force, characterized by a strong coupling suppression of protons compared to neutrons (protophobic force) [3]. The evidence for such a particle was first reported in 2012 at a workshop in Italy by the group led by Attila Krasznahorkay of ATOMKI Institute for Nuclear Research at Debrecen (Hungary): they reported about unexpected results observed in their nuclear physics experiments. In April 2015, the same group of ATOMKI confirmed the existence of the new light boson, giving an evaluation of its mass (nearly 17 MeV): this time the sigma levels were 6.8 excess. According to the team, this was the simplest way to explain the experimental results of anomalies in Beryllium transitions and such excess of sigma levels cannot be related to any statistical fluctuations [1]. Recently the ATOMKI group published new results about 11 B(p,γ) 12 C nuclear reaction (J π =1 -, Γ =1.15 MeV, 12 C excited state) at E p =1.388 MeV incident proton energy, that also suggest the existence of X17 boson [4]. In the first ATOMKI measurements the target was irradiated by protons of energy around 1.03 MeV, thus exciting lithium nuclei up to the 18.15 MeV beryllium state. The excited beryllium nucleus is expected to reach its ground state throughout the transition 0 + → 0by internal pair conversion (IPC), the only possible way according to nuclear spectroscopy rules (angular momentum and parity conservation). The angular distribution between two emerging leptons from IPC has an apparent peak around 10 • -15 • , after this angular interval is going fast to zero, i.e. the probability to measure a relative angle for the ee + pairs over 30 • is negligible. What it has been observed instead is a clear peak (up to 6.8 excess from the e + ebackground) in the angular range of 100 • -130 • , with a maximum around 120 • (see Fig. 1). The importance of such a discovery -if confirmed -is beyond nuclear physics. Every new particles, specially bosons, could be associated with a new force or at least with a new, unknown and unexpected aspect of one of the known forces. X17 could be related to the role of quarks, trapped in the hadronic bag and to the nucleon binding energy inside nucleus. Preliminary theoretical studies [3] has pointed out the hypothesis of different behaviour for up and down quarks unveiled by a much stronger cross section when neutron transitions are observed with respect to the ones of protons (protophobic scenario). This innovative scenario could provide new ideas to explain, at least partially, the long-standing anomaly on the muon magnetic moment. More in general, the possible existence of a new particle is of paramount importance in particle physics and in cosmology (dark matter). Therefore, the ATOMKI claim [1], [2], [4], [5] clearly calls for new experimental studies.

Physics case
Presently we are engaged to carry on a first series of measures at n_TOF, where the excited levels of 4 He, 8 Be can be populated via the conjugated 3 He(n,ee + ) 4 He and 7 Be(n,ee + ) 8 Be reactions using the spallation neutron beam EAR2 at CERN. This approach has two relevant advantages: (i) for the first time X17 existence is investigated through neutron induced reactions exploiting the world unique properties of EAR2 beam [6] and (ii) the experimental setup is completely different with respect to the one used by the ATOMKI group.
Our efforts aim to realize a suited detection setup for the determination of particle kinematics with a strong capability to discriminate the reaction ejectiles (assumed to be ee + pairs) from background in a wide energy range. If the existence of X17 is confirmed, we would be able to establish quantum numbers and mass of the X17 boson, and to shed light on the so-called protophobic nature of a fifth force. In fact, state-of-the-art "ab-initio" calculations are in good agreement with present literature data (in particular for the "few body" 4 He nucleus) and would provide quantitative predictions to establish the X17 nature, e.g. if it is a scalar, pseudoscalar, vector or axial boson [8] and to get information on the interaction of the X17 boson with quarks and gluons [3]. EAR2 station of the n_TOF facility at CERN provides a pulsed neutron beam in a wide energy range, which broadly covers the region of interest for this experiment, i.e. 10 3 eV < E n < 10 7 eV [6], [7]. Counting rate estimations show that the neutron intensity at EAR2 is high enough to carry on a conclusive experiment within approximately one measurement month. Working at a spallation neutron beam line means  4 He reaction at E p =900 keV. The red points are the experimental data showing the ee + pairs excess around 115 • . The blue is the expected angular distribution of ee + pairs based on standard IPC physics, while the green line is the best data fit assuming M X17 =16.87 MeV to have to tackle the problem of beam induced background. It is well-known that the dominant outgoing channel in the neutron-induced reaction is the proton final state, in the case of 3 He(n,p) 3 H with a reaction Q-value = 764 keV. The sub-dominant process 3 He(n,γ) 4 He produces single photons with E γ > 20 MeV. Finally, virtual photons can produce ee + pairs through the internal pair conversion (IPC) channel 3 He(n,ee + ) 4 He. For this reaction, with incident neutrons of E n ≥ 21 MeV, it would be possible to excite the 0 -→ 0 + level of 4 He with Γ ∼ 0.84 MeV, the 2 -→ 0 + 4 He excited level with Γ ∼ 2.01 MeV can be reached at E n ≥ 21.8 MeV. The IPC pairs and the ee + pairs generated by the interaction of real photons with the material surrounding the 3 He target represent a relevant background for the process of interest. However, the amount of ee + pairs produced by (virtual or real) photons rapidly decreases while increasing their emission relative angle. Instead, the 4 He * → 4 He + X17 process and successive X17 → ee + decay produces pairs with a large relative angle, determined by the X17 mass and energy. As a consequence, the excess of pairs at large relative angles (see Fig. 1) represents a clear signature of new physics. It is worth pointing out that a possible existence of a vector X17 boson would alleviate the present tension between data and calculations concerning the muon magnetic moment (see for instance the recent [9]). The prediction for the value of the (electron) muon anomalous magnetic moment includes three parts: 165 918 04(51) The first two components that represent the photon and lepton loops (a QED µ ) and the W boson, Higgs boson and Z boson loops (a EW µ ) can be both calculated precisely. The third term (a hadron µ ) represents hadron loops and it cannot be calculated accurately from theory alone, it is estimated from experimental measurements of the ratio of hadronic to muonic cross sections in electron-positron (e + e -) collisions [10]. As of July 2020, the a S M µ measurements disagree with the Standard Model by about 3.5 standard deviations, suggesting physics beyond the Standard Model may be having an effect (or that the theoretical/experimental uncertainties are not completely under control). This is one of the long-standing discrepancies between the Standard Model and experiments. If X17 exists and has a hadron nature, the hadronic term of the anomalous magnetic moment must be increased taking into account this new contribution that it could be evaluated roughly speaking from 10 -6 up to 10 -5 .

Experimental
To measure the particle kinematics observables, to reach a high level of particle identification in a wide energy range and to optimize the signal-to-noise ratio, thus disentangling the different contributions eventually leading to e + epairs, we have studied a detection setup based upon two trackers of rectangular shape placed at the side of the target at around 5 cm far, one in front of the other. Back to each tracker it is placed an outer electromagnetic calorimeter made of two planes of EJ-200 slabs, 50 cm long, 1 cm wide and 0.6 cm thick. The two EJ-200 slab planes are mounted in a perpendicular way, one along x-axis and the other along y in order to strength the incident particle trajectory reconstruction that starts into the tracker. Back to the EJ-200 slabs the volume is fulfilled by 16 cubes of EJ-200 scientillators, 10 cm side, able to stop and to absorbe all the incident leptons energy. The detection setup is the same on both side of the target. In the following all the items that jointly cooperate to build up the experimental apparatus are discussed and analyzed.

Targets
The first data collection campaign will be dedicated to measure 3 He(n, ee + ) 4 He nuclear reaction. The 3 He target would be a cylinder 3 cm diameter, 3.5 cm long of carbon fiber 0.6 mm thick. According to preliminary tests this target could hold up to 350 atm of 3 He, we plan to fill the target up to 300 atm to keep a safety margin. The carbon fiber cylinder is placed at close contact inside another cylinder of about the same dimensions made of Al 1.2 µm thick, to avoid or at least to minimize any 3 He leaks throughout the very thin carbon fiber. As far as 7 Be(n, e + e -) 8 Be reaction is concerned, we plan to have an enriched 7 Be solid target 4-5 mg/cm 2 thick evaporated onto a 4-5 µm thick Ta backing. The experimental apparatus is designed to minimize the dead layers that the emerging leptons have to pass through with the aim to reduce multiple scattering effect. Multiple scattering is a source of systematic error in the mutual e + eangle measures, it could affect the trajectory reconstruction and dead layers could also spoil the overall energy resolution.

Tracker
In order to dissolve all the doubts about the existence or not of X17, the most of kinematic observables of the induced reactions must be measured at low uncertainties. Then the core of the detection set-up must be a tracker, in Fig. 2 (see caption) a block diagram is shown. The outer blue sector represents an electromagnetic calorimeter dedicated to the total energy of the emerging leptons and then allowing a precise measure of X17 mass.
Each tracker is realized by TPC that could deliver direct 3D track information for pattern recognition even in high multiplicity events and particle identification over a wide momentum range. It is even suited to work, if needed, inside a magnetic field that deflects the tracks enabling the measure of the particle momentum. The drift time of each track segment could be measured to provide the vertical coordinate Z. The position in the XY projection is obtained by recording the induced charge profiles on the segmented readout plane after amplification, the recorded charge provides dE/dx useful for particle identification (PID). The TPCs have rectangular shape, 50 cm long 30 cm wide, with a radial electric field, read out by Micro Pattern Gaseous Detectors (MPGD) like µRwell. The very low material budget of a TPC with a helium-based gas mixture and an anode of 3 mm Cu minimizes the multiple scattering experienced by the ejectiles, while the radial field, compared to a more traditional longitudinal configuration, reduce the diffusion experienced by the ionization electrons drifting toward the anode, so improving the space resolution of the detector. These features are expected to produce a substantial improvement in the measurement of the e + erelative angle. The active path available for tracking is of 3 cm. For the segmented readout the use of bulk-µRwell technology (a well-known technology with limited ion feedback) shows the following advantages: i) all-in-one detector: blind areas minimized (including edges and corners), ii) simple design, cheap & robust, iii) good uniformity of performances.
The performances of the prototype have been investigated with and without the target on the beam. The results of the test provide us with information on the response to the γ-flash and helps to evaluate the highest neutron energy that can be reached (around few MeV), together with a first measure of the background rate induced by the neutron beam in the target.

Electromagnetic calorimeter
The use of a dedicated segmented calorimeter composed by two arrays of scintillating slabs placed perpendicular to each other and back to the two slab planes, in the outer position, 16 cubic tiles 10x10x10 cm 3 of EJ-200 (equal situation on both target sides), together with the action of the tracker, allows a trajectories reconstruction of ∆Θ = ∆φ = 5 deg. or better. The cubic tiles are thick enough to fully absorb all the electron energy and stop them. According to theoretical model [8] the leptons emerging from X17 share an energy fraction between 2 (minimum) and 15 (maximum) MeV. By GEANT4 simulation, electrons of up of 20 MeV energy, that could be considered a safe upper limit for our measures, will have a range slightly over 11 cm in EJ-200 scintillator and each electron has to get across two slabs of 0.6 cm thick and a cube of 10 cm side, in the minimum path when they are emitted perpendicular to the electromagnetic calorimeter (without considering the tracker). The slab readout is provided by SiPM array placed at each ends and related electronics. The EJ-200 cubes will be read by fast PMTs with active divider in order to ensure constant gain up to the maximum allowable current: this is necessary because of the impact of γ flash on the detection system, the main source of beam induced background. All the electron energy would be fully contained inside the calorimeter and the energy loss inside the slabs and the cubes will be summed up: we expect to reach an energy resolution for X17 around 10%. The electromagnetic calorimeter provides complementary measurements with respect to the tracker detector. It is worth mentioning that the electromagnetic calorimeter would make possible to study in detail the two nuclear reactions 3 He(n,γ) 4 He and 3 He(n,e + e -) 4 He differential cross section: these measurements are very important to firmly establish the structure of excited levels of 4 He, providing a robust experimental footing for the ab-initio calculations. Moreover, in case it is confirmed the scenario in which X17 decays into two unknown particles (and with unknown charge), it is possible to observe the different response of the calorimeter with respect to pairs at small relative angle, uniquely due to e + epairs from IPC, providing an independent check concerning particle identification. Of course, the scintillator is also sensitive to the protons produced by the 3 He(n,p) 3 H channel, that have approximately the same energy of the interacting neutrons (Q=764 keV), but we are confident to identify the protons with the joint action of tracker, electromagnetic calorimeter and TOF measure of the pulsed neutron beam. In fact, neutrons with E n > 15 MeV reach the EAR2 experimental area not later than 350 ns after the spill time, while events generated by neutrons with E n < 10 MeV are detected not before 400 ns

Conclusion
X17 is offering an exciting challenge for experimental nuclear physicists, with the unique opportunity of a real fundamental discovery or at least to contribute to cast light in a promising new physics field. Even if the existence of X17 would be not confirmed by the new experimental efforts, the data taken in the efforts to solve this puzzle would be useful to investigate effects that for too long have been neglected, like the IPC phenomenon on light nuclei and testing experimentally the validity of the ab-initio nuclear theories, thanks to a new enthusiastic collaboration from nuclear theorists and experimentalists. Our experimental approach to X17 hunting is focussed on leptons trajectory reconstruction, high precision e + eangle (we plan to reach an overall uncertainty ≤ 3 deg.) and X17 mass measurements. From the phenomenological side we would investigate the possible protophobic effect that could be seen comparing the X17 cross section production (if any) by proton and neutron induced reactions (neutron cross section is expected to be around 30% higher than proton one [3], [8]): this is the main motivation in our choice to start the measurements at n_TOF neutron spallation beam (at the moment there are no data on neutron induced reactions). In the optimistic side we could claim that these efforts triggered by X17 hunting, independent from the success or not of the chase, would start a new era for the study of strong interaction and nuclear structure.