Conventional fast neutron flux measurements in the radial piercing channel D of the TRIGA Mark II reactor, Pavia.

The TRIGA Mark II nuclear reactor operated by the Laboratory of Applied Nuclear Energy (LENA) of the University of Pavia is a 250 kW light water moderated facility aimed for training, general purpose research and isotope production [1]. One of the research fields conducted at this reactor is focused on the investigation of nuclear reactions induced by fast neutrons. The accessibility to a fast neutron beam allows a broad variety of applications, ranging from the determination of fast neutron cross section data, study of burnup and transmutation in fuel elements and effects of radiation damage in various materials; all these applications might be, in turn, beneficial for what concerns research and development of the upcoming IV generation of fast nuclear reactors. In order to make available a fast neutron beam at the TRIGA reactor, the realization of a new neutron irradiation facility is planned by modifying the so-called channel D, a pre-existing radial piercing channel without reflector material. The channel D will be adapted by introducing filters to remove the neutron thermal flux component and reduce the gamma background, and a beam catcher to assure operator’s safety. Characteristics and dimensions of filtering and shielding materials will be modulated according to the neutron flux spectra simulated by means of Monte Carlo Neutron Particle (MCNP) software. The quality of the simulated data will be assured by validating the software code using experimental data collected in selected positions along the channel. In this study, preliminary measurements of conventional fast neutron flux in four positions of the actual (unmodified) configuration of channel D are reported. The adopted technique, described in [2], consists of activations and ?-countings of monitor elements. The resulting fast neutron flux relies on a conventional value of the monitor fission-neutron averaged cross section, ?. Ni solutions, obtained from a Ni solid standard dissolved in nitric acid, were selected as monitors to exploit the 58Ni(n,p)58Co threshold reaction; equal volume samples of four solutions prepared with increasing Ni concentration were placed at 45 cm, 75 cm, 125 cm and 195 cm from the vertical axis of reactor core. The neutron exposure lasted 90 minutes at 10 kW power; the absence, during the measurement, of most part of the shielding prevented the achievement of the operational 250 kW power. Gamma spectra including the 58Co 810.8 keV ?-peak emission were acquired with a calibrated HyperPure Ge (HPGe) detector by placing the irradiated samples in contact with the HPGe end-cap. The conventional fast flux, ?f, for each measurement position was obtained using eq. (10) of [3]; we also calculated the reaction rate per target nucleus in Ni monitors, R, by multiplying the ?f times ?. The R value is no longer conventional as it is independent from assumptions concerning the fast neutron flux shape. Since the R values can be evaluated by MCNP code, they are used to assess the quality of simulated data. The ?f and R results, scaled at 250 kW reactor power, ranged from 1.33(4) x 10^11 cm^-2 s^-1 and 1.48(3) x 10^-14 s^-1, respectively (at 45 cm), to 1.32(4) x 10^9 cm^-2 s^-1 and 1.46(3) x 10^-16 s^-1, respectively (at 195 cm). Values in parenthesis indicate the standard uncertainty. The two orders of magnitude decrease in flux and reaction rate along the 150 cm horizontal distance was in agreement with previous knowledge of the facility. To sum up, the experimental data collected in this preliminary measurements offer a valuable independent reference to validate the code that will be used for modeling the structural modifications on channel D. References [1] Prata et al; Eur Phys J Plus, 2014; 129. [2] De Corte; habilitation thesis, University of Gent, 1987. [3]Di Luzio et al; Prog Nucl Energy, 2019; 113.