Measurement and calculation of neutron leakage spectra from slab samples of beryllium , gallium and tungsten irradiated with 14 . 8 MeV neutrons

In order to make benchmark validation of the nuclear data for gallium (Ga), tungsten (W) and beryllium (Be) in existing modern evaluated nuclear data files, neutron leakage spectra in the range from 0.8 to 15 MeV from slab samples were measured by time-of-flight technique with a BC501 scintillation detector. The measurements were performed at China Institute of Atomic Energy (CIAE) using a D-T neutron source. The thicknesses of the slabs were 0.5 to 2.5 mean free path for 14.8 MeV neutrons, and the measured angles were chosen to be 60◦ and 120◦. The measured spectra were compared with those calculated by the continuous energy Monte-Carlo transport code MCNP, using the data from the CENDL-3.1, ENDF/B-VII.1 and JENDL-4.0 nuclear data files, the comparison between the experimental and calculated results show that: The results from all three libraries significantly underestimate the cross section in energy range of 10–13 MeV for Ga; For W, the calculated spectra using data from CENDL-3.1 and JENDL-4.0 libraries show larger discrepancies with the measured ones, especially around 8.5–13.5 MeV; and for Be, all the libraries led to underestimation below 3 MeV at 120◦. Due to a very low melting point (29.78 ◦C) and a very high boiling point (2229 ◦C), Ga becomes an candidate element in Chinese Initiative Accelerator Driven Systems (CIADS) project for liquid–metallic coolant [1]. W is one of the most promising candidate for spallation targets and other structural materials of CIADS project, as well as an important material in fission and fusion devices [2]. Be is another important material in the fission and fusion devices for multiplying neutrons in the core of fission research reactors and in the blankets of Deuteron-Tritium (D-T) fusion reactors [3]. The accuracy of nuclear data files for Ga, W and Be is very important for nuclear device design. The purpose of this work is to validate currently available modern nuclear data files, especially the CENDL-3.1 file, by carrying out benchmark experiment 1. Experiments and calculations

Due to a very low melting point (29.78 • C) and a very high boiling point (2229 • C), Ga becomes an candidate element in Chinese Initiative Accelerator Driven Systems (CIADS) project for liquid-metallic coolant [1].W is one of the most promising candidate for spallation targets and other structural materials of CIADS project, as well as an important material in fission and fusion devices [2].Be is another important material in the fission and fusion devices for multiplying neutrons in the core of fission research reactors and in the blankets of Deuteron-Tritium (D-T) fusion reactors [3].The accuracy of nuclear data files for Ga, W and Be is very important for nuclear device design.The purpose of this work is to validate currently available modern nuclear data files, especially the CENDL-3.1 file, by carrying out benchmark experiment

Experiments
The measurements were performed by using the benchmark experimental facility at China Institute of Atomic Energy (CIAE).A schematic view of the experimental arrangement is shown in Fig. 1, which was described in detail earlier in Reference [4].
The T(d, n) 4 He reaction served as fusion neutron source.The D+ beam was bunched about 1.5 ns in Full Width at Half Maximum (FWHM) and the repetition a e-mail: nieyb@163.comrate was 1.5 MHz.The average beam current and beam energy were about 30 µA and 300 keV, respectively.An air-cooling device was used for cooling the target.For getting the time structure of the pulsed source neutrons, a stilbene scintillation crystal (monitor) of 5.08 cm in diameter and 2.54 cm in length was placed at about 800 cm from the Ti-T target at 0 • .A silicon surface barrier (SSD) detector, positioned at 135 • with respect to the D + beam, was used to monitor the neutron yields by counting the associated 4 He particles.All the samples used for the experiment have purity better than 99.3%.For detecting the leakage neutrons from samples, a BC501A liquid scintillator detector, which was located behind the concrete wall with a collimated hole, with a size of 5.08 cm in diameter and 2.54 cm in length was used.
The leakage neutron spectra were obtained from the results of the sample-in measurements by subtracting those of sample-out measurements, and the results were normalized by the number of source neutrons.The uncertainties combine the statistical and systematic uncertainties.The systematic uncertainties were mainly caused by the neutron detection efficiency (≤3%), the source neutron yield (≤3%) and the scattering angle ambiguity (≤1%).

Monte Carlo calculations
The theoretical calculation was performed using a threedimensional continuous energy Monte Carlo transport code, MCNP-4B.The evaluated nuclear data of Ga, W and   Be were taken from CENDL-3.1 [5], ENDF/B-VII.1 [6] and JENDL-4.0 [7], the data of other materials were all taken from the ENDF/B-VII.1.A point detector estimator was used to tally the leakage neutron time of flight spectra for comparing with the measured ones.The calculations for two angles of about 60 • and 120 • were performed with the model shown in Fig. 2. The left sample was filled with air when the simulation was performed for 60 • , while the right sample was filled with air for 120 • simulation.For the background simulation, both samples were filled with air.The neutron histories adopted were 10 9 and the statistical uncertainties of each time bin were smaller than 1%.
In the MCNP simulations, the detailed experimental parameters were taken into account.These included the neutron energy distribution produced on the thick target, the angle dependent energy distribution of the source neutrons, the neutron detection efficiency, and the time structure of the pulsed beam (TME).Both of the angular distribution and angle dependent energy distribution of the source neutrons were calculated by the TARGET code [8].The value of TME variable was defined by using experimentally measured time of flight spectra of source neutrons by the monitor.

Ga
The neutron leakage spectra for Ga calculated with the evaluated data files comparing with the measured ones at 60 • and 120 • are shown in Fig. 3. From the comparison, we can conclude that: 1.In the 13-15 MeV neutron energy range, a sharp elastic peak is observed.All the calculated results using these three libraries have large discrepancies with the experimental ones.2. In the 10-13 MeV neutron energy range, a small, but clear peak is observed.The contribution originates from discrete level of inelastic scatterings ((n,n )d) and all MCNP simulations using the three libraries do not show a peak as the experimental data do.3.In the 3-10 MeV neutron energy range, the contribution comes from the continuous level of inelastic scattering ((n,n )C).For 60 • , the calculated spectra with the CENDL-3.1 are overestimated about 15%.For 120 • , all the calculated spectra agree well with the experiment.4. In the 0.8-3 MeV neutron energy range, in which most of the contribution comes from the (n, 2n) reaction channel.The calculated spectra with the ENDF/B-VII.1 are overestimated about 20% at 60 • and 10% at 120 • .

W
The neutron leakage spectra from W calculated with the evaluated data files comparing with the measured ones at 60 • and 120 • are shown in Fig. 4. From these results, the following observations are made: 1.In the 12.5-15 MeV neutron energy range, the major contribution comes from the elastic scattering.The calculated elastic scattering peak with the evaluated data of the JENDL-4.0 libraries is higher than the experimental one, while those of the ENDF/B-VII.1 and CENDL-3.1 libraries are underestimated.2. In the 7-12.5 MeV neutron energy range, the calculated spectra with the ENDF/B-VII.1 give agreement with the experimental ones at 60 • and 120 • , but those with the JENDL-4.0 and CENDL-3.1 are largely underestimated.

Figure 1 .
Figure 1.Lay-out of experiment for measuring the neutron leakage spectra from slab sample.

Figure 2 .
Figure 2. Model for the MCNP calculations.

Figure 3 .
Figure 3.Comparison of experimental and calculated neutron leakage spectra from nat Ga slab with 143 keV electron equivalent threshold (6.4 cm thickness).

Figure 4 .
Figure 4. Comparison of experimental and calculated neutron leakage spectra from nat W slab with 143 keV electron equivalent threshold (7.35 cm thickness).