Search

The continuous strive for more efficient and reliable nuclear power plants is leading to increasing complexity in the design and operations of reactors. As a consequence, comprehensive analyses are required to assess the reactor's safety during normal and accidental conditions. Validated simulation tools with high-fidelity capabilities, i.e., at the pin or sub-pin scale, are necessary to predict local neutronics effects in the core. However, the scarcity of high-resolution in-core experimental data imposes limitations to the validation of such high-fidelity pin-resolved neutronics codes. The challenges in performing online high-resolution in-core experiments are numerous: from the accessibility of in-core locations to the availability of neutron detection technologies with adequate dimensions and online capabilities. In the present thesis, the development of a novel miniature and minimalistic (MiMi) neutron detection technology is presented, allowing unprecedented spatial resolution for the online study of the neutron flux in zero-power research reactors. The MiMi neutron detectors are tested in the EPFL zero-power reactor CROCUS and used to build a data set of high-resolution neutronics experiments in CROCUS, including measurements of local gradients and directionality of neutron flux, and the local impact of a fuel rod displacement. As the next level of development, a three-dimensional (3D) full-core mapping system named SAFFRON, consisting of 149 MiMi neutron detectors distributed in-core, is designed and installed in CROCUS. Static thermal neutron flux maps are measured in absolute terms and between different core configurations, e.g., water level vs. control rod operation at criticality. The obtained results open up the investigation of a variety of space-dependent neutronics phenomena in CROCUS.

Highly localized in-core measurements are necessary for the validation of neutron transport calculations with high spatial resolution. In the present work, a miniature neutron detector developed at EPFL in collaboration with PSI was used to carry out a set of thermal neutrons counting measurements in the zero-power CROCUS reactor core within a spatial range in order of mm. The miniature detector, positioned close to the core reflector, shows a gradient of +(4.29 ± 0.10)% in the count rate profile in the radial direction within 1.3 cm, with higher values pointing towards the core reflector because of the higher share of neutrons in the thermal range. On the contrary, in a control rod guide tube the count rate gradient is -(4.37 ± 0.10)% and it is directed towards the core center. The measured values are compared with the azimuthal trend of the normalized 6Li reaction rate calculated with an iterative three-steps method performed with the Monte Carlo code Serpent 2. These measurements proved the feasibility of resolving spatial effects in the mm-range and they represent a basis for further investigating highly spatially-resolved phenomena in the CROCUS core.

This article presents the development, full-characterization, and in-core testing of a miniature neutron detector for the investigation of highly localized in-core neutronics effects in the zero-power reactor CROCUS, operated at the Ecole Polytechnique federale de Lausanne (EPFL), Switzerland. A ZnS:(LiF)-Li-6 screen, mixing inorganic scintillator and neutron converter, with a surface of 1 mm(2) and a thickness of 0.2 mm, was coupled with a silicon photomultiplier (SiPM) via a 10-m optical fiber. Analog readout electronics were adopted in this prototype version to process the SiPM output signal. A first campaign was carried out to assess the detection system capabilities in terms of neutron detection and counting in mixed neutron-gamma radiation fields. The miniature detector was thoroughly characterized with a Pu-Be source installed in the CARROUSEL facility and was subsequently tested inside a control rod guide tube of the CROCUS reactor in different reactor conditions, i.e., at shutdown, startup, and critical. The detector shows a linear response with a reactor power increase up to 6.5 W (i.e., around 108 cm(-2) . s(-1) total neutron flux) and excellent neutron counting capabilities, when compared to other localized detection systems available in CROCUS, such as the miniature fission chambers and single-crystal vapor deposited (sCVD) diamond detectors. In addition, the exposure of the miniature detector to a Co-60 source with an activity of similar to 250 GBq, combined with the computation of the gamma flux in CROCUS with the Serpent 2 Monte Carlo code, confirmed the gamma insensitivity of the system in a mixed neutron-gamma field.

The goal of this work was the improvement of the capabilities of a miniature neutron detector, for easing its use, and extending its range of applicability. Firstly, the Red Pitaya STEMlab 125-14 was the object of an exploratory study in order to propose an alternative method to the two current acquisition electronics based on photon counting, using analog and digital means. The study allowed to establish a new methodology based on the direct use of the SiPM pre-amplifier output, requiring the hypothesis that neutron events can be discriminated directly by pulse amplitude. It was first tested in the low neutron flux of the CARROUSEL facility, and the results were compared to the analog electronics. The agreement in count rate, spectrum, and linearity demonstrates the validity of this new methodology. Moreover, its application in CROCUS allowed demonstrating that it is working in reactor condition up to a flux of 1.13∗108 s−1cm−2. Secondly, an array of six miniature neutron detectors with different sensitivities was designed and tested. The qualification study in CARROUSEL allowed to show that all detectors behave normally, and that the sensitivity of the detectors increases as a function of the scintillator’s volume at low neutron flux, as expected. The conclusion is limited by the knowledge in effective dimensions of each scintillator. In a first experiment in CROCUS using the same setup, including digital threshold (50), a loss of linearity was observed simultaneously for all detectors at a power ofabout 4W(i.e. 7∗107s−1cm−2). As the cause was expected to be a higher gamma background, the experiment was repeated with a higher threshold (120): consistently with the CARROUSEL test, count rate and loss of linearity depends on the scintillation’s volume. Further studies are needed to conclude on the phenomenon at hand, but since the main parameter for gamma discrimination proved efficient, the gamma contribution increased at higher powers seems a good candidate. In conclusion, the results are globally satisfying for the new methodology with Red Pitaya STEMlab 125-14. However, the experimental campaign of this work was only a feasibility study. The results have to be confirmed with further studies. Moreover, this system could be improved. In particular, it can be adapted firstly for both RF inputs of Red Pitaya and secondly for dynamical measurements. The results of the multi-sized array of miniature neutron detector were also encouraging. However, the behaviour of certain detectors prevent to deduce with certainty some expected trends. Further studies have to be carried out by multiplying the number of detectors of each dimension.

The Laboratory ofReactor Physics and System Behaviour (LRS) at EPFL is finalising the preparation of a detection system for dynamic studies on the CROCUS experimental reactor. In order to model the temporal behaviour of the reactor within the framework of these experiments, it is necessary to study a new numerical method, based on fission matrices. The objective of the internship is to investigate the fission matrix method recently implemented in the Monte Carlo calculation code Serpent 2, then to develop analysis tools and finally to apply them on the CROCUS reactor. The study is limited for the moment to static cases. The methodology is first studied on a simple system (Pressurised Water Reactor fuel rod), the results of which can be compared with previous studies and with classical criticality calculations. In order to be able to analyse the results quantitatively, a novel method for estimating uncertainties in the fission matrix spectrum was developed and compared to a reference method for estimating uncertainties in Monte Carlo codes. The analysis tools were coded in Python, using the Numpy and Matplotlib libraries. The difference obtained between the eigenvalue of the fundamental mode of the fission matrix on the one hand, and the multiplication factor estimated by criticality calculations on the other hand, is of the order of 10−5. The convergence of the fission matrix spectrum with the number of simulated neutrons, as well as with the number of cells used for the spatial discretization, has been shown on the PWR rod. The developed uncertainty estimation method overestimates the standard deviation for the fundamental mode by about 30%, but is found to be 100 times faster than the reference method. This method is therefore a fast way to estimate the uncertainties on the fission matrix spectrum. Finally, the application of this methodology on the CROCUS reactor has made it possible to calculate several hundred of its eigenvectors in three dimensions, as well as their associated uncertainties, something impossible with the usual criticality calculations. The analysis and uncertainty estimation tools developed during this internship make it possible to obtain the eigenvalues and eigenvectors, as well as their associated uncertainties, of a fission matrix calculated with the Serpent 2 code. This study must be continued by taking into account the time dependence of the fission matrices, in order to be able to model dynamic experiments.