Production and Monitoring of Neutron Flux by Activation Detectors

The neutron generation technique was tested on the microtron M-10 with an output electron beam of 8.7 MeV. Given the low energy that the microtron can provide to electrons, the bremsstrahlung induced photonuclear reaction 9Be (γ, n), which has a low threshold, was chosen for neutron generation. Cobalt and indium targets were tested as activation detectors to estimate the neutron flux density. In the cobalt target, the isomeric state of 60mCo with an energy of 58.6 keV and a half-life of 10.5 minutes is well activated. Two well-known additional gamma lines of standard cobalt source permit to clarify the absolute value of the neutron flux. The activated indium target has four gamma lines bound to the 116mIn isomer βdecaying with the half-life of 54.4 minutes, what is convenient for measurement of gamma spectrum. Despite the low energy of the output electron beam, at a beam intensity of 5 μA it is possible to obtain an almost isotropic neutron flux of 107 n/(s∙cm2).


I. INTRODUCTION
N the microtron M-10 of Uzhgorod National University there are carried out experimental investigations of photonuclear reactions [1] as well as applied studies of irradiation influence on the properties of new technological materials and electronical components [2]. For irradiation it is possible to use electrons of energies 4-10 MeV and bremsstrahlung. But it was demonstrated [3] that low energy electron accelerator microtron MT-25 with 22 MeV output electron beam can serve as neutron source, which is interesting to compare with the generation of neutrons by a proton accelerator at the same beam energies [4].
Neutrons are unique particles that are of interest from both a fundamental and an applied point of view. Of the unsolved experimental fundamental problems, we note the problem of neutron lifetime [5], the question of the presence of the electric dipole moment of the neutron [6], and the search for neutron clusters [7]. In addition, neutron registration still represents an uneasy problem.
Among the current applied research, we note the study of different types of structures in the physics of condensed matter, the study of the effect of neutron fluxes on living tissues and organisms, neutron diffraction analysis [8], neutron activation analysis. Of great practical importance for materials science is the availability of accurate data on the cross-sections of the interaction of neutrons with nuclei. For energy needs, namely for reactors controlled by accelerators, experimental data on neutron generation processes are required.
In this paper, we present the results of neutron generation on the M-10 microtron of Uzhgorod National University. The experiment consist of few stages -1) output electron beam generate bremsstrahlung spectrum in reaction − + → + − + , 2) neutrons creation in photonuclear reaction (γ,n), 3) activation reaction (n,γ) and 4) measuring and analysis of gamma spectra of activated sample.

A. Schematic view
The Fig. 1 shows a schematic of the experiment. The output electron beam falls on the braking tungsten target after which the bremsstrahlung beam hits the beryllium target. Neutrons are created on a beryllium target in reaction γ + 9 → n + 2 , (1) starting from the threshold Eγ=1.57 MeV. The intensity of generated neutrons is registered by the activation sample (target) due to (n, γ) reaction. Analysis of gamma spectra of the activated sample (detector) gives the value of the neutron flux. The main parameters of our experiment were as follows: the energy of the induced electron beam was 8.7 MeV and a plate of natural tungsten of size 93 x 55 x 2 mm (thickness 2 mm) served as a braking target. The size of the beryllium target is Ø10 x 14 cm, and its weight is 2 kg.
To measure neutrons flux we use 59 Co as activation detectors. The cobalt detector (boxed powder CoCO3*Co(OH)2*nH2O, mass 31 g) was irradiated during 10 min by neutrons Cobalt

Production and Monitoring of Neutron Flux by
Activation Detectors detector is activated by neutrons to both 60 Co ground state and isomeric level 60m Co with half-life 10.47 minutes and following isomeric transition Eγ = 58.6 keV to the ground state. After βdecay (half-time 5.27 years) the ground state gives well-known two lines Eγ = 1173 keV and Eγ = 1332 keV [9]. As a neutron indicator, we also used an aluminum plate coated with indium powder.

B. Gamma Spectrometer
A NaI(Tl) crystal scintillation detector with size Ø63 x 63 mm looking through by photomultiplier 19-M and with the SBS-40 amplitude analyzer board served as a gamma spectrometer (Fig. 8). The accumulation of spectra occurs with the help of a specialized computer program AkWin. The number of channels of the analyzer is variable and can take values from 256 to 8192. The measured energy resolution (FWHM) of the spectrometer for calibration lines and the simulated total peak efficiency ε for the detected gamma lines of the neutronactivated sample 59Co are shown in Table I.

III. BREMSSTRAHLUNG SPECTRUM
The gamma-ray spectrum created on a 2 mm thick tungsten target in the forward direction by a 5 MeV electron beam is showed at the Fig. 2. The spectrum is simulated by the FLUKA code [10]. The bremsstrahlung spectrum is a wide peak in the interval of energy Eγ=0.1-5 MeV. In the energy region, less than 0.1 MeV the characteristic X-ray lines of tungsten atom are present, also the annihilation line of 0.511 MeV is evident. That is, in a tungsten conversion target with a thickness of 2 mm, there is a significant probability of a three-stage process, namely, 1) the braking process on the nucleus e -+Z →γ + e -+Z; 2) formation in the field of the nucleus of electron-positron pairs γ + Z → e + + e -+ Z; 3) the formation of the positronium atom e + + e -→ Ps; 4) annihilation of positronium Ps → γ + γ. The first two processes take place with the participation of the atomic nucleus, and the last two are the atomic processes. Fig. 3. The simulated braked gamma spectrum of 5 MeV electron beam in 2 mm tungsten converter. In Fig. 3 is shown a comparison of the bremsstrahlung spectra simulated for electron beam of energies 4 MeV, 5 MeV, 8 MeV, and 9 MeV [11]. It is seen that the shape of the peak is preserved, and with increasing electrons energy the yield of gamma quanta is increased. In the study of nuclear reactions, only a part of the gamma spectrum with energy greater than the reaction threshold "works". In our case of neutrons generation on a beryllium target, it is gamma quanta with energies greater than 1.6 MeV. But at a set of the absorbed dose, it is necessary to consider the whole bremsstrahlung spectrum. It should be noted that the bremsstrahlung spectrum depends on both the electron energy and the thickness of the braking target and its atomic composition. The penetration depth of a monoenergetic electron beam with energies 3 < Ee <20 MeV in aluminum is well described by phenomenological formula = 0.53 • − 0.106 , where R is the depth of penetration in g/cm 2 , Ee is the electron energy in MeV [12]. For other matter the penetration depth can be estimate by correction where (Z/A)x is the charge to mass ratio of the corresponding element. The estimated penetration depth for an electron beam of 8.7 MeV in tungsten is 2.8 mm.

IV. NEUTRON PRODUCTION
The measured and evaluated data of the energy dependence of the photonuclear reaction cross-section σ(γ, abs) for beryllium, is as follows Fig. 4 in the ground or excited states (the three lowest well-separated levels of 8 Be are 3.04, 11.4, and 16.6 MeV). The thresholds of these channels are indicated in Table I. So for electron beam 8.7 MeV the three channels of reaction 9 Be(γ,n) are opened. Analysis of the kinematics of the 9 Be(γ, n) 8 Be reaction shows that the kinetic energy of a neutron is weakly dependent on its emission angle. Fig. 5 given the dependence of the neutron kinetic energy distribution versus the angle of the outgoing neutron relative to the gamma beam direction for energies of gamma quanta 8, 6, 4, and 2 MeV. However, the shape of the angular distribution can be deformed by the dynamics (matrix element) of the nuclear reaction. Fig. 3 shows that the intensities of the corresponding energies of gamma quanta differ by an order of magnitude. That is, one can conclude that the reaction (γ, n) forming an almost isotropic distribution of neutrons in the whole energy region from zero up to 6 MeV. But in the beryllium target itself, neutrons are moderated by elastic multiple scattering on beryllium nuclei. Therefore, the spectrum of neutrons outside the beryllium target will be much softer. Fig. 6. The angle dependence of kinetic energy of outgoing neutrons for reaction γ + 9 Be → n + 8 Be at energy Eγ=8, 6, 4 and 2 MeV (violet, red, yellow, and green line respectively).

V. ACTIVATION DETECTORS
We chose samples of 59 Co and natural indium as activation detectors which have satisfactory rate activation and disintegration constants. The activation sample was located close to the side surface of the beryllium cylinder.

A. Cobalt sample
The effective cross-section of the reaction 59 Co(n,γ) is shown in Fig. 6. In the region from thermals up to 6 MeV neutrons, the cross-section belongs to interval 2-40 barns [14]. Neutron activation of 59 Co leads to the formation of 60g Co in the ground state and in the isomeric state 60m Co. The decay schemes of cobalt from the ground (T1/2=5.27 years) and isomer (T1/2=10.47 minutes) states [9] are shown in Fig. 7. Isomer state has two decay modes -99.76% isomeric gamma transition with Eγ=58.59 keV and 0.24% βdecay. The ground state is a wellknown standard cobalt-source which after βdecay gives two

VI. RESULTS AND CONCLUSIONS
A beryllium target and activation samples were irradiated for 10 minutes with a braking beam of gamma quanta obtained on a tungsten conversion target by an electron beam with an energy of 8.7 MeV and an intensity of 5 μA. Fig. 9 shows the gamma spectra from samples of indium, cobalt, and background. The isomeric cobalt line is not shown because it is located near channel 20 where the background is very high. The preliminary result of processing the gamma spectra of indium is given in [16], and the processing of cobalt spectra confirmed it, namely, at an electron beam current of 5 μA at an energy of 8.7 MeV and selected bremsstrahlung and neutron producing targets on the M-10 microtron we obtain a neutron flux 2×10 7 n/(cm 2 •s).
The questions of neutron energy distribution and the ratio of neutron and gamma fields remain open [17,18].