Cross section measurements of low-energy charged particle induced reactions using moderated neutron counter arrays

. A high-and-ﬂat e ﬃ ciency moderated neutron detection array of 3 He counters has been recently developed for photoneutron cross section measurements at the future ELI-NP gamma-ray beam source. We have designed a three rings geometry of 31 counters with a ∼ 37% e ﬃ ciency ﬂat within 5% in the 10 keV to 5 MeV neutron energy interval. A commissioning experiment was performed using proton beams delivered by the 9 MV Tandem accelerator of IFIN-HH. We measured the neutron production cross sections for proton induced reactions on nat Cu and 27 Al. The ( p , xn ) reactions on nat Cu were investigated in the 4.5 MeV to 14 MeV proton energy range with 250 keV steps, using pulsed and continuous proton beams. The low energy measurements below 10 MeV served to validate the detection e ﬃ ciency calibration against well-known nat Cu ( p , n ) cross sections. The 27 Al( p , n ) reaction was investigated in the 5.8 MeV to 6.75 MeV energy range with 5 keV steps. Preliminary cross section results are here compared with preceding data.


Introduction
In response to the call launched by the IAEA Coordinated Research Project on Photonuclear Data and Photon Strength Functions (Code F41032; Duration 2016-2019) [1, 2] to perform new photonuclear reactions measurements, two novel moderated neutron detection arrays have been designed and constructed. Their main characteristic is the high-and-flat detection efficiency dedicated to neutron multiplicity sorting measurements of competitive (γ, n), (γ, 2n), . . . photoneutron and (γ, f xn) photofission reactions. One of the detectors has been installed at the NewSUBARU-Japan Laser Compton scattering (LCS) γray beam line [3][4][5] and has been used for an extensive campaign of photoneutron cross section measurements in the Giant Dipole Resonance region [2,6,7]. The measurements made use of neutron-multiplicity sorting procedures developed for low- [8] and high- [9] reaction rate conditions associated to the novel flat-efficiency detection system. ELIGANT-TN [10] is the second high-and-flat efficiency moderated neutron detection array developed by our group, originally dedicated to photoneutron cross section measurements at a γ-ray beam source planned to be constructed at ELI-NP [11]. The array consists of three * e-mail: cristina.clisu@nipne.ro concentric rings of 3 He tubes embedded in a polyethylene moderator. The ELIGANT-TN array has been calibrated and commissioned using proton beams delivered by the 9 MV Tandem accelerator of IFIN-HH. The wellknown proton induced reactions on nat Cu have been used for the neutron detection efficiency calibrations. The 27 Al(p, n) reaction was also investigated, for an additional validation of the neutron detection efficiency curve for monochromatic neutron emission. We here present preliminary calibration results and experimental results for the nat Cu(p, xn) and 27 Al(p, n) measurements in comparison with preceding data.

ELIGANT-TN detector
The ELIGANT-TN detector is composed of three concentric rings of 4, 8 and 16 3 He proportional counters placed at 5.9, 13 and respectively 15.5 cm distance from the beam axis, which coincides with the longitudinal detector axis. The proportional counters are identical cylinders of 25.4 mm diameter and 500 mm active length filled with 12 atm pressure 3 He gas. The counters are embedded in a high density polyethylene block of 46 cm width, 46 cm height and 54 cm length. Cadmium sheets of 0.5 mm thickness and additional 12 cm thick polyethylene plates are  placed on the back, front and side of the moderator block for background neutron suppression. Figure 1 shows the ELIGANT-TN mounted on the beam-line 2 at the IFIN-HH 9 MV Tandem accelerator. Figure 2 shows the ELIGANT-TN neutron detection efficiency for the inner (red), middle (green) and outer (blue) rings, as well as the total efficiency (black) for both monochromatic (dotted lines) and evaporation (full lines) neutron spectra. The simulations are compared with an experimental measurement performed with a 239 PuBe source of (2.27 ± 0.23) × 10 5 neutrons/second activity. The experimental point is represented in Fig. 2 at the mean 239 PuBe neutron energy value retrieved from Ref. [12]. The present comparison serves as a preliminary check of the simulations results, followed by validation against monitor nat Cu(p,n) reaction cross section measurements described in Section 3.1.
The simulations were performed with the eliLaBr code [13,14] implemented using the Geant4 [15][16][17] framework. Particular effort has been made to model the neutron moderation process.
For neutrons with energies smaller than 20 MeV down to thermal energies, we used the following Geant4 physics models: G4NeutronHPElastic for elastic scatterings, G4NeutronHPCapture for neutron capture reactions, G4NeutronHPFission for fission and G4NeutronHPInelastic for all the other types of neutron induced nuclear reactions. The high precision packages mentioned use the tabulated cross sections of the Geant4 data library, which was updated with the ENDF-B/VII database. For neutrons with energies less than 4 eV, the Thermal Scattering package [18,19] was used, which accurately takes into account the structure of the molecules/material. A total detection efficiency of ∼37% flat within 5% in the 10 keV to 5 MeV neutron energy interval is obtained. The total efficiency averaged over neutron energies up to 2, 5 and 10 MeV is 37.8±0.2, 36.4±1.5 and 33.6±3.2% for evaporation spectra neutron sources and 38.1±0.2, 37.0±1.4 and 32.9±4.6% for monochromatic neutron sources.
The flat efficiency dependence makes the measured cross sections insensitive to the neutron emission spectra. However, knowing the energy of the reaction neutrons not only increases the precision of the experimentally determined cross section through using dedicated neutron detection efficiencies, but also offers information on the decay mechanism of the compound nuclear states populated in the reaction. Thus, the ring-ratio method [9] is used to determine the average energy of the neutron emission spectra. Figure 3 shows the experimental ring-ratio values well reproduced by Geant4 simulations.

Commissioning experiment
The ELIGANT-TN array has been calibrated and commissioned using proton beams delivered by the 9 MV Tandem accelerator of IFIN-HH. The proton beams irradiated metallic foils of (0.925 ±0.005) µm thick nat Cu and (1.24±0.01) µm thick 27 Al placed in the center of the array. The target thickness was measured using α-particle transmission method [20].
The signals from each of the 3 He counters were split and transmitted to two parallel data acquisition systems (DAQ). One DAQ system run in triggerless mode and was based on scalers which counted the 3 He counter signals passing a leading edge n/γ discrimination threshold. The second DAQ system based on VME ADCs and TDCs was run on a general neutron OR trigger and simultaneously provided the amplitude of the signals and the time relative to the trigger signal. The information provided by the scaler DAQ was used in the data analysis, while the VME data were used for monitoring the stability of the signal amplitudes.
During irradiations, the time variation of the incident proton beam current was continuously monitored using four charge integrator units coupled to the electrically insulated Faraday cup (FC), rear collimator (C2), reaction chamber (beam line) and target, as shown in Fig. 4. The beam current on the front collimator (C1) was monitored    (p, n). The cyan area represents the neutron emission energies considering the upper limit for forward-and the lower limit for backward-neutron emission. The average neutron emission energy for isotropic emission in the center of mass system is represented by the cyan continuous line.
by the accelerator staff. The charge collection was validated against separate activation measurements on nat Cu. The relative time dependence of the FC, target, C2 and beam line for the nat Cu(p, xn) reactions in the 4.5 MeV to 14 MeV proton energy range is shown in figure 5. We note that the current collected on the beam line is negative, as it mostly consists of electrons scattered from the target. We also observe that the ratio of charge collected in the Faraday Cup increases with the increase in the proton energy.

nat Cu(p, xn) reactions
The (p, xn) reactions on nat Cu were investigated in the 4.5 MeV to 14 MeV proton energy range with 250 keV steps. The present preliminary cross sections are shown in Fig. 6 in comparison with the summed IAEA recommended values for nat Cu(p, x) 62,63,65 Zn [21,22] and ENDF evaluations for nat Cu(p, x) 59,61 Ni, 62,64 Cu and 64 Zn [23]. The experimental error bars account for the statistical uncertainty, which is negligible, as well as 3% for the neutron detection efficiency, 3% for the proton beam charge collection and less than 1% for the target thickness.
The low energy measurements below 10 MeV served to validate the detection efficiency calibration against wellknown nat Cu(p, n) cross sections. The measurements above the 65 Cu(p, 2n) reaction threshold at 10 MeV were performed in both continuous and pulsed proton beam conditions. Neutron multiplicity sorting techniques are currently applied to the combined pulsed and continuous proton beam data for extracting the nat Cu(p, n) and nat Cu(p, 2n) components.

27 Al(p, n) reaction
The 27 Al(p, n) reaction was investigated in the 5.8 MeV to 6.75 MeV energy range with 5 keV steps. Proton beam energy spread values of less than 10 keV have been estimated based on TRIM simulations [24]. The present preliminary results are shown in figure 7 in comparison with the preceding experimental data of Sekharan [25] and Bubb [26]. The new cross sections are in good agreement with the ones reported by Sekharan [25], which are systematically higher than the results of Bubb [26].
The ring-ratio deduced energies for the neutrons emitted in the 27 Al(p, n) reaction are shown in figure 8 in comparison with calculated values obtained from kinematics by taking into account that, due to the low incident proton energies, only the ground state of the residual nucleus is populated in the neutron emission. We show the maximum and minimum neutron emission energies as the upper and lower limits of the cyan band. The central line represents the average neutron emission energy computed considering isotropic angular emission in the center of mass system. The difference between the calculated and the experimental values is given by the fact that the neutron detection efficiency curve and corresponding ring-ratio functions shown in figures 2 and respectively 3 are obtained by considering isotropic emission in the laboratory reference system. We plan to update the simulations based on experimentally determined angular distributions for groundstate neutron emission in the 27 Al(p, n) reaction [27].