Device and method for determining a neutron absorber in an object

Abstract

Method for determining a quantity (Cj, Cj(x,y,z)) of a neutron absorber in an object (3), the method comprising a) arranging the object between a neutron source (2) and a detector (10), the detector comprising at least one pixel (10i), each pixel comprising a first detection layer (11) formed of a first material, the first material being a scintillator doped with an isotope suitable for capturing a thermal neutron, each pixel being configured to form a pulse following an interaction of a neutron or a γ photon in the first detection layer, the or each pixel being connected to a processing unit (20) configured to discriminate a pulse, formed in said pixel, resulting from an interaction of a neutron from a pulse resulting from an interaction of a γ photon; b) irradiating the object with neutrons emitted by the neutron source;c) during irradiation, determination of at least one counting rate (TCi(n,E')) representative of thermal neutrons detected per unit of time by the or each pixel; d) from each counting rate resulting from c), determination of the quantity of neutron absorber in the object.;

Description

Title of the invention: Device and method for determining a neutron absorber in an object Technical field

[0001] The technical field of the invention is the determination of a content of at least one neutron absorber in an object, and this by means of a non-destructive measurement. PRIOR ART

[0002] Certain isotopes are conducive to the absorption of neutrons. These are, for example, isotopes of cadmium (113Cd), boron (10B), lithium, gadolinium (157Gd, 155Gd), silver (107Ag) or indium (113In or 115In). These isotopes are subsequently referred to as "neutron absorbers". In the nuclear field, neutron absorbers are integrated into neutron-absorbing components, the latter being able to be control rods or screens intended for radiation protection.

[0003] It may be necessary to control the neutron-absorbing effect of these absorbers, by determining the quantity of neutron-absorbing elements they contain. This type of control may take place during the manufacture of these components, for quality control purposes.

[0004] It may also be useful to determine the quantity of neutron absorbers in nuclear waste. Indeed, the implementation of nuclear measurements, intended to estimate the radiological activity of waste, may include a measurement of neutrons emitted by the waste. This type of measurement is for example implemented when the waste is likely to contain certain neutron-emitting radionuclides. The presence of neutron absorbers in the waste may induce an error in the interpretation of the measurements. In order to correct this type of error, the quantity of neutron absorbers present in the waste can be estimated.

[0005] The determination of the quantity of neutron absorbers can be carried out on samples taken and analyzed in the laboratory, in a destructive manner. The question then arises of the statistical representativeness of the sampling. In addition, laboratory measurements require transport of the sample. They also assume a certain delay between the collection of the sample and obtaining the analysis result.

[0006] The publications Basturk M. “Analysis of neutron attenuation in boron-alloyed stainless Steel with neutron radiography and JEN-3 gauge”, Journal of Nuclear Material 341 (2005), 189-200, and Makil H. “Quality control of neutron-absorber materials for the nuclear fuel cycle, Principle of the JEN-3 neutron backscattering gauge”, ICARST 2017, describe a method for determining a quantity of neutron absorbers, in a backscattering configuration, using a so-called "JEN-3" probe. This probe uses 3He neutron counters and a 252Cf neutron source. The use of such a neutron source has certain disadvantages, in particular its short radioactive half-life (2.64 years). Thus, the number of neutrons emitted by the source per unit time varies rapidly over time, which requires frequent updating. Furthermore, the use of 3He counters makes the method expensive. In addition, such a detector can only detect neutrons, without the possibility of obtaining spatially resolved measurements.

[0007] The inventors propose a device for improving the performance of the JEN-3 probe, in particular by enabling multimodal measurement, i.e. addressing different types of particles. The device can also enable spatially resolved measurements to be carried out, so as to establish a map of the quantity of neutron absorber in the characterized object. Statement of the invention

[0008] A first object of the invention is a method for determining a quantity of a neutron absorber in an object, the method comprising: - a) arrangement of the object between a neutron source and a detector, the detector comprising at least one pixel, the or each pixel comprising a first detection layer formed from a first material, the first material being a scintillator doped with an isotope suitable for capturing a thermal neutron, each pixel being configured to form a pulse following an interaction of a neutron or a photon y in the first detection layer, the or each pixel being connected to a processing unit configured to discriminate between: • a pulse, formed in said pixel, resulting from an interaction of a neutron; • and a pulse, formed in said pixel, resulting from an interaction of a photon y; - b) irradiation of the object by neutrons emitted by the neutron source; - c) during irradiation, determination of at least one counting rate representative of thermal neutrons detected per unit of time by the or each pixel; - d) from each counting rate resulting from c), determination of the quantity of neutron absorber in the object.

[0009] The processing unit can be configured to determine a counting rate representative of a quantity of photons detected there, by the or each pixel, per unit of time for different amplitudes of said interactions.

[0010] According to one possibility, - the or each pixel comprises a second detection layer formed from a second material, the second material being a scintillator material different from the first material, the second detection layer being superimposed on the first detection layer, the second detection layer extending between the first detection layer and the object, the second detection layer being thinner than the first detection layer; - the or each pixel is configured to form a pulse following an interaction of a charged particle in the second detection layer; - the processing unit is configured to discriminate, in the or each pixel, a pulse of a charged particle having interacted in the second detection layer from a pulse of a neutron or a photon having interacted in the first detection layer.

[0011] According to one possibility, - wherein the or each pixel comprises a third detection layer formed from a third material, the third material being a scintillator material different from the first material and the second material, the third detection layer being superimposed on the second detection layer, the third detection layer extending between the second detection layer and the object, the third detection layer being thinner than the second detection layer; - the or each pixel is configured to form a pulse following an interaction of a particle a in the third detection layer; - the processing unit is configured to discriminate, in the or each pixel, a pulse from a charged particle having interacted in the second detection layer from a pulse from a particle having interacted in the third detection layer.

[0012] According to one embodiment, - the detector has several pixels adjacent to each other; - the object is virtually segmented into different elementary volumes; - step d) involves determining the quantity of neutron absorber in the different elementary volumes of the object.

[0013] Optionally, the processing unit is configured to discriminate, in the or each pixel, an impulse of a charged particle that has interacted with the first detection layer, an impulse of a neutron or a photon that has interacted therewith in the first detection layer.

[0014] Optionally, a neutron thermalizing material, which is conducive to the thermalization of neutrons, is disposed around a space including the source, the object, and the detector or between the source and the object.

[0015] Step d) may include taking into account a calibration function that establishes a relationship between at least one counting rate resulting from a pixel and the quantity of the neutron absorber.

[0016] A second object of the invention is a device configured to estimate a quantity of a neutron absorber in an object, the device comprising: - a neutron source; - a detector, comprising at least one pixel, the or each pixel comprising a first detection layer formed from a first material, the first material being a scintillator doped with an isotope suitable for capturing a thermal neutron, each pixel being configured to form a pulse following an interaction of a neutron or a photon y in the first detection layer; - a processing unit, configured to discriminate between: • a pulse, formed in said pixel, resulting from an interaction of a neutron; • and a pulse, formed in said pixel, resulting from an interaction of a photon y; - the device being configured to place the object between the neutron source and the detector; - the device being such that: • the processing unit is programmed to determine at least one counting rate representative of a quantity of thermal neutrons detected per unit of time by the or each pixel; • the device comprises a calculation unit configured to estimate a quantity of the neutron absorbing element in the object from the counting rates resulting from the processing unit.

[0017] According to one embodiment: - the detector comprises a second detection layer formed from a second material, the second material being a scintillator material different from the first material, the second detection layer being superimposed on the first detection layer, the second detection layer extending between the first detection layer and the object, the second detection layer being thinner than the first detection layer; - the or each pixel is configured to form a pulse following an interaction of a charged particle in the second detection layer; - the processing unit is configured to discriminate, in the or each pixel, a pulse of a charged particle in the second detection layer, from a pulse of a neutron or a y photon in the first detection layer.

[0018] According to one possibility: - the or each pixel comprises a third detection layer formed from a third material, the third material being a scintillator material different from the first material and the second material, the third detection layer being superimposed on the second detection layer, the third detection layer extending between the second detection layer and the object, the third detection layer being thinner than the second detection layer; - the or each pixel is configured to form a pulse following an interaction of a particle a in the third detection layer; - the processing unit is configured to discriminate, in the or each pixel, a pulse of a charged particle in the second detection layer from a pulse of a particle in the third detection layer.

[0019] The detector may comprise several pixels adjacent to each other.

[0020] The processing unit can be programmed to determine an amount of the neutron absorbing element in different elementary volumes of the object from the counting rates resulting from the processing unit.

[0021] The scintillator material forming the first layer may be an organic scintillator or an inorganic scintillator.

[0022] The invention will be better understood upon reading the description of the exemplary embodiments presented in the remainder of the description, in conjunction with the figures listed below. FIGURES

[0023] [Fig.1A] shows a diagram of a device according to the invention.

[0024] [Fig. 1B] represents a single-layer pixelated detector.

[0025] Figures IC and 1D represent variants of the device described in connection with Figures 1A and 1B.

[0026] [Fig.2] schematizes a pulse resulting from an interaction of a neutron in a scintillator material.

[0027] [Fig.3A] shows a diagram of a pixelated single-layer detector.

[0028] [Fig.3B] shows a pixelated bilayer detector

[0029] [Fig.3C] shows a pixelated three-layer detector.

[0030] Figures 4A to 4D represent neutron capture cross sections as a function of energy for different neutron absorbers, respectively 155Gd, 157Gd, ([Fig.4A]), 113Cd ([Fig.4B]), 10B ([Fig.4C]), 113In and 115In ([Fig.4D]). The abscissa axis corresponds to the energy (MeV) and the ordinate axis corresponds to the cross section (barn).

[0031] [Fig.5] shows schematically the main steps of a process for determining a neutron absorber content.

[0032] Figures 6A and 6B represent a modeled configuration.

[0033] [Fig.7A] shows a calibration function, establishing a relationship between a quantity of thermal neutrons detected, in an organic scintillator and a quantity of 10B in a plate.

[0034] [Fig.7B] shows the calibration function described in [Fig.7A], as well as another calibration function establishing a relationship between a number of particles a reaching a bilayer detector and the amount of 10B in the plate.

[0035] [Fig.8A] shows a spatial distribution of the amount of 10B in different elementary volumes of a plate.

[0036] [Fig.8B] shows a quantity of thermal neutrons detected in the pixels of a pixelated organic scintillator respectively located opposite each elementary volume of the plate.

[0037] [Fig.9A] shows a spatial distribution of the amount of different neutron absorbers in different elementary volumes of a plate.

[0038] [Fig.9B] shows a modeled pixel of a detector.

[0039] [Fig. 10A] represents, for different neutron absorbers (x-axis), the quantities of different types of particles generated by neutron capture during exposure of the absorber to a neutron emitted by an Am / Be type neutron source.

[0040] [Fig. 10B] shows an energy distribution of particles detected by a detector when exposing a boron sample to a neutron source. The x-axis corresponds to a number of particles detected per neutron emitted by the source.

[0041] [Fig.10] shows an energy distribution of y-photons detected by a detector during the exposure of a sample formed of boron to a neutron source. The abscissa axis corresponds to a number of y-photons detected per neutron emitted by the source.

[0042] [Fig.10D] is the equivalent of [Fig.10C] considering a gadolinium sample.

[0043] [Fig.10E] is the equivalent of [Fig.10C] considering a sample of cadmium.

[0044] [Fig. 10F] is the equivalent of [Fig.l0C] considering a sample of indium.

[0045] [Fig.l0G] shows an energy distribution of particles [3 detected by a detector during exposure to a neutron source of a sample formed of boron, or gadolinium, or cadmium, or indium.

[0046] [Fig. 10H] shows an energy distribution of α particles detected by a detector upon exposure to a neutron source of a sample formed of boron, or gadolinium, or cadmium, or indium.

[0047] In Figures 10B to 10H, the abscissa axis corresponds to the energy. The ordinate axis represents a number of particles detected for a neutron emitted by the neutron source. PRESENTATION OF SPECIAL EMBODIMENTS

[0048] [Fig.1A] shows schematically a device 1 allowing an implementation of the invention. The device comprises a neutron source 2, configured to emit neutrons. The neutron source can be a neutron generator or a source comprising a mixture of radionuclides emitting neutrons, for example a 241AmBe source. The use of a 252Cf type source, producing neutrons by spontaneous fission, is possible but has the disadvantage of a short radioactive period, as described in connection with the prior art. It is therefore necessary to take into account the radioactive decay of the source in the processing of the measurements.

[0049] The device is intended to characterize a quantity of at least one neutron absorber in an object 3. In this example, the object 3 is a plate. Other geometries can be envisaged.

[0050] The device 1 is configured so that the object 3 is arranged between the neutron source and a detector 10, according to a so-called transmission configuration. The object 3 can for example be held on a support. Preferably, the distance between the object 3 and the detector 10 is small, for example less than 10 cm. The object 3 can be arranged in contact or in near-contact with the detector. By near-contact, we mean a distance of the order of cm, or less than 1 cm, without contact between the object 3 and the detector 10.

[0051] The device comprises a detector 10, connected to a processing unit 20 and a calculation unit 30.

[0052] The detector comprises at least a first detection layer 11 of a first material, of scintillator type. The first layer 11 is optically coupled to at least one photodetector 15. Under the effect of an interaction of an ionizing particle (a, [3, gamma) or of a neutron, scintillation photons are generated in the first layer, forming a light pulse. The, or each, photodetector 15 is configured to convert the light pulse into an electrical pulse. Each photodetector 15 may be of the silicon photomultiplier (SiPM) type or a tube or photodiode photomultiplier, for example an avalanche photodiode.

[0053] Generally the amplitude of each electrical pulse is correlated with the amount of energy deposited by the ionizing radiation during the interaction in the scintillator material. Thus, under the effect of an interaction with ionizing radiation, in this case a neutron or a gamma photon, the first layer 11 emits a light pulse, the amplitude of which depends on an energy deposited by the neutron or the γ photon during the interaction. At least one photodetector 15 detects the light pulse and generates an electrical pulse the amplitude of which depends on the amplitude of the light pulse, the latter depending on the energy deposited by the neutron or the γ photon during the interaction.

[0054] The relationship between the energy deposited by the particle and the amplitude of the pulse corresponds to an energy calibration function. The latter can be obtained experimentally and / or numerically.

[0055] In the intended application, the first layer 11 is essentially intended for the detection of thermal neutrons. But it can also be used for the detection of gamma photons, and in particular in certain energy bands. The volume of the first layer 11 must be sufficient for the detection sensitivity to be acceptable. Thus, the thickness of the first layer is greater than 1 cm, and preferably greater than 2 cm or 5 cm. The thickness is considered along the axis of propagation of the neutrons emitted by the source and reaching the first layer 11.

[0056] Advantageously, the scintillator material forming the first detection layer is an organic scintillator. An advantage of organic scintillators is that it is possible to obtain high detection volumes, typically greater than 1 dm3 or even 10 dm3 or several tens of dm3. Thus, the first layer 11 can extend, facing the neutron source, over a surface area of ​​a few tens of cm2.

[0057] The detector 10 is preferably pixelated, i.e. segmented into portions, each portion being optically coupled to a photodetector 15. The number of photodetectors 15 determines the number of pixels.

[0058] The detection layer 11 may be homogeneous. However, it is preferable that the first detection layer 11 is segmented into portions separated from each other by a light-tight partition 16, for example a thin partition of the PTFE (polytetrafluoroethylene) type. Preferably, each partition 16 is reflective, so as to increase the amount of scintillation light reaching the photodetector. Each portion is optically coupled to a photodetector 15, which forms a pixel 10;. The index i is an integer corresponding to a rank assigned to a pixel. The number of photodetectors 15 determines the number of pixels.

[0059] The use of an organic scintillator facilitates the manufacture of a pixelated detector, i.e. a detector segmented into portions, each portion being optically coupled to a photodetector 15. The pixelated layer 11 can be obtained by molding: the organic scintillator is poured into a PTFE mold in the liquid state. It is then solidified

[0060] The scintillator can be obtained by machining in the mass, so as to form pixels. This can allow parallelepipeds, preferably cubic, each parallelepiped forming a pixel.

[0061] As is customary in the field of scintillators, the detector 10 is preferably wrapped in a sheet of light-tight material, for example Mylar (registered trademark).

[0062] Alternatively, the first scintillator material forming the first layer 11 may be an inorganic scintillator, for example lithium-doped sodium iodide, usually designated NAIL.

[0063] Preferably, the first scintillator material, forming the first layer 11, has previously undergone an addition of a capture isotope, suitable for neutron capture, resulting in an emission of a charged particle. Such an addition has been described in FR3083622. The neutron capture may in particular be a capture of type (n,a). The capture isotope may for example be 6Li or 10B. The addition of the capture isotope allows detection of thermal neutrons (energy less than 0.5 eV) or epithermal neutrons (energy between 0.5 and 500 keV). The added capture isotope allows an emission of a charged particle, in the first scintillator material, under the effect of the capture of a thermal or epithermal neutron. When the capture isotope is 6Li, neutron capture results in the emission of an a particle and a triton (3H nucleus). The mass fraction of the capture isotope is preferably between 0.05% and 20%.The higher the mass fraction, the more sensitive the detector is to thermal or epithermal neutrons.

[0064] Alternatively to 6Li the capture isotope can be chosen from an isotope of the following elements: Gd, B,, Cd, this list not being limiting.

[0065] Preferably, the processing unit 20 comprises a multi-channel analyzer 22, usually designated by the acronym MCA (Multi-Channel Analyzer) so as to process the pulses resulting from each photodetector. In Figures 1A, 1C and 1D, a multi-channel analyzer is shown connected to each photodetector 15. Alternatively, the same MCA can be connected to several photodetectors 15. In a known manner, the MCA makes it possible to establish an amplitude spectrum of the detected pulses.

[0066]

[0067]

[0068]

[0069]

[0070] during a measurement period. The amplitude spectrum is usually referred to as the energy spectrum Sp, the amplitude of each pulse being correlated with the energy deposited in the detector material by an interaction that generated said pulse. The energy spectrum Sp is usually represented in the form of a histogram, representing a quantity of pulses detected for each amplitude channel, each amplitude channel corresponding to an energy band. The MCA 22 is generally preceded by a circuit 21 for amplification and shaping of the pulses. The processing unit 20 may also comprise a pulse shape analyzer 23, coupled to a discriminator 24, the discriminator performing a classification of the pulses according to their shape. The detector 10 is sensitive to ionizing radiation as well as to neutrons. More particularly, the first layer 11 is likely to be exposed to photons and neutrons. Each detected interaction, generated by a neutron or a photon, gives rise to an electrical pulse, resulting from a photodetector 15. The shape of the pulse depends on the radiation having interacted with the first detector material. Figure 2 shows a pulse detected respectively following an interaction of a photon in the first detector material (solid curve) or following an interaction of a neutron in the first detector material (dotted curve). In the latter case, the pulse is affected by a “tail”, or “trail”, usually designated by the Anglo-Saxon term “tail”. For each detected pulse, a shape parameter can be determined, for example a ratio Qtot-Qshort , where; - Qtot corresponds to the total charge detected, i.e. the integral of the pulse; - Q short corresponds to the proportion of Qtot corresponding to the impulse without the train: see hatched part of [Fig.2]. The measurement of the Qtot and Qtot-Qshort parameters, each interaction allows a Qtot discrimination between gamma photons and neutrons. Gamma photons form a distributed energy zone (variable Q^), with a high Qtot-Qshort ratio. tor Qtot This reflects the fact that the pulses resulting from interactions of y photons in an organic scintillator are relatively symmetrical. They are generally highly distributed in amplitude. The pulses resulting from neutron interactions in the organic scintillator are more asymmetric. The Qtot-Qshort ratio is lower than that of Qtot pulses. resulting from an interaction of a gamma photon.

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077] In order to be able to efficiently separate the pulses due to photons and the pulses due to neutrons, the processing unit 20 comprises a shape analyzer 23, allowing a determination of the total charge Qtot detected by each pulse as well as the part Q t corresponding to the pulse without the tail. The shape analyzer makes it possible to determine the shape parameter Qtot-Qshort for each pulse. On the basis of and Qtot-Qshort the pulse is assigned either Qtot to a photon, or to a neutron by the discriminator 24 previously described. When the organic scintillator is doped with a capture isotope, such as 6Li, the amplitude of the pulses generated following the capture of a thermal or intermediate neutron corresponds to the energy of the ionizing particle, in this case an a particle, emitted following the capture of the neutron by the capture isotope. Thus, an asymmetric pulse whose amplitude corresponds to an energy band E' representative of the a particle emitted, in the scintillator, by the capture, is the signature of an interaction of a thermal neutron in the detector. [Fig.lB] shows a diagram of the detector 10, segmented into pixels 10;. [Fig.lC] shows a configuration in which the source and detector are surrounded by a thermalizing material 5. The thermalizing material can be HDPE (High Density Polyethylene). The thermalizing material allows neutrons to be thermalized between the neutron source 2 and the object 3. This increases the number of thermal neutrons that can be absorbed by the object. The thermalizing material may also be arranged around the neutron source, or between the neutron source and the object, such a configuration being shown in [Fig.lD], In [Fig.3A], a section of a pixelated detector 10 is shown, comprising a single layer 11, called the first layer, as described in connection with Figures 1A to 1D. Such a detector makes it possible to measure thermal neutrons and γ photons of different energies. In [Fig.3B], another embodiment is shown, in which the detector comprises a second layer 12, superimposed on the first layer 11, and optically coupled to the latter. The second layer 12 is formed of a second scintillator material, suitable for the detection of charged particles, of type a or [3-. The second scintillator material is different from the first scintillator material. The second layer 12 may be formed of a thin layer (thickness less than or equal to 1 or 2 mm) of ZnS type scintillator, in which case it is mainly sensitive to particles a. The second layer 12 may be thicker, in which case it is sensitive to particles a and [3 . A portion of the photons of scintillation formed in the second layer propagates through the first layer 11 and are detected by a photodetector 15. Since the first and second scintillator materials are different, the shape of the pulses resulting from the first and second layers is different. The difference may, for example, arise from the pulse decay time. The shape analyzer 23 and the discriminator 24 of the processing unit 20 make it possible to separate the pulses generated in the first layer 11, which are mainly pulses resulting from photons and neutrons, from those generated in the second layer 12, the latter being mainly pulses due to charged particles. The second layer 12 extends between the first layer 11 and the object 3.Within the second layer 12, the discrimination between the pulses respectively generated by a particles and [3 particles can be carried out by amplitude discrimination, the amplitude of the interactions due to the a particles often being greater than the amplitude of the interactions due to the [3- particles. .

[0078] The detection principles of a multi-layer scintillator have been described in WO2022263449.

[0079] In [Fig.3C], another embodiment is shown, in which the detector comprises a third layer 13, formed of a third scintillator material, superimposed on a second layer 12, formed of a second scintillator material, the latter being superimposed on the first layer 11. The third layer 13 is a thin layer, with a thickness less than or equal to 1 or 2 mm. The third scintillator material is suitable for the detection of charged particles, of type a. The third scintillator material is different from the second scintillator material, the latter being different from the first scintillator material. The second layer 12 is intended for the detection of particles [3 having passed through the third layer 13. Its thickness is preferably less than 1 cm.Thus, according to this configuration, the first layer 11 is intended for the detection of neutrons or γ photons, the second layer 12 is intended for the detection of particles [3 and the third layer 13 is intended for the detection of α particles. The three layers 11, 12 and 13 are optically coupled. The scintillation photons created in each layer are detected by a photodetector 15. The pulses resulting from the latter are then processed by the processing unit 20. The three scintillator materials being different from each other, it is possible to discriminate the pulses generated respectively in each of them, as described in connection with [Fig.3B]. .

[0080] Depending on the configuration of the detector 10 and the processing unit 20, the latter makes it possible to determine different counting rates, for different types of particles. For the same particle (for example a, [3, y, fast neutron), the unit processing unit 20 may be configured to establish a count rate per energy band, by using an MCA. The count rates resulting from the processing unit 20 are then used by the computing unit 30 to quantify, from the neutron absorption, the quantity of neutron absorber. The computing unit may be configured to identify one or more neutron absorbers, and determine their respective quantities in the object.

[0081] The processing unit 20 may comprise analog or digital means. The calculation unit 30 preferably comprises a microprocessor, programmed to execute instructions making it possible to estimate a quantity of neutron absorber.

[0082] When the detector is pixelated, as shown diagrammatically in Figures 3A to 3C, the device makes it possible to estimate a quantity of a neutron absorber in different parts of the object. This provides a spatially resolved measurement, which is particularly advantageous. This, however, assumes a certain proximity between the object 3 and the detector 10: the smaller the distance between the object and the detector, the better the spatial resolution.

[0083] Figures 4A to 4D represent neutron capture cross sections as a function of energy for different neutron absorbers, respectively 155Gd, 157Gd, ([Fig.4A]), 113Cd ([Fig.4B]), 10B ([Fig.4C]), 113In and 115In ([Fig.4D]). In each of these figures, the abscissa axis corresponds to the neutron energy (unit MeV) and the ordinate axis corresponds to the capture cross section (unit barn).

[0084] It is observed that the invention can be implemented for the detection of low energy neutrons, i.e. of energy between 10 keV (0.01 MeV) and a few eV or a few tens of meV.

[0085] Table 1 gives examples of nuclear capture reactions, capture cross sections and natural abundance of certain isotopes. It is observed that neutron capture, in the characterized object, can be accompanied by an emission of particles, for example α (10B), β (155Gd), γ photons in predetermined energy bands (for example 0.48 MeV for 10B). These emissions constitute a signature of neutron capture in the object to be characterized, which depends on the isotope absorbing the neutrons. Thus, the detection of particles emitted following neutron capture, in the characterized object, can make it possible to identify the neutron absorber. It is therefore relevant to use a detector that can allow the detection and discrimination of different types of particles: y / neutrons (see [Fig.3A]), a, y and neutrons ([Fig.3B]) and a, |3-4y and neutrons (see [Fig.3C]).Using an MCA allows spectrometric measurements to be obtained, i.e. knowledge of the energy released by the particles in the detector. Photons are emitted there according to discrete energies. Their detection, by the first layer. detection, can be modeled, which makes it possible to assign, to each neutron absorber, one or more energy bands in which photons are detected there.

[0086] The spectrometric function also makes it possible to identify, among the detected neutrons, the neutrons having the energy E' released in the scintillator by a charged particle, during the capture of a thermal neutron by the capture isotope. This allows a quantification of the number of thermal neutrons detected. Isotope Nuclear reaction Cross section (bams) Natural abundance (%) 10B "'B + n 7Li (0.84 MeV) + a (1.47 MeV) + y (0.48 MeV) [94%] "'B + n 7Li (1.02 MeV) + a (1.78 MeV) [6%] 3845 (25.3 meV) 19.9 113Cd 113Cd + n 114Cd (1.02 MeV) + y's (9 M eV) 19970 (25.3 meV) 12.23 155Gd 155Gd + n 156Gd + e + y's (8.536 MeV) 60737 (25.3 meV) 14.80 157Gd 157Gd + n 158Gd + e + y's (7.937 MeV) 252912 (25.3 me V) 15.65 107Ag 107Ag + n 108Ag + y 1770(16.3 eV) 51.83 113In "dn + n "4In + y's 1720(1.80 eV) 4.29 115In 115In + n —> 116In + y's 29350(1.45 eV) 95.72

[0087] [Table 1]

[0088] The notation y's followed by the parenthesis designates the fact that there are several photons y emitted at energies distributed according to an emission spectrum. The energy value in the parentheses corresponds to the average energy of the spectrum.

[0089] [Fig.5] summarizes the main steps of the invention.

[0090] During a step 100, the object to be characterized 3 is interposed between the source 2 and the detector 10.

[0091] During a step 110, the object 3 is irradiated by the source 2 and the processing unit 20 generates counting rates TC; for different types of particles, or even different energies, and this for each pixel 10;. Depending on the configuration of the detector 10, the processing unit 20 can generate the counting rates TC;(a), TCi([3 ), TCi(y), TCi(y,E), TC;(nE).

[0092] TCi(p,E) ​​denotes a count rate generated by pixel i for particle p, possibly in an energy range E. The energy range E is particularly useful when the particle is a photon y or a neutron.

[0093] The thermal neutron counting rate corresponds to the counting rate TCi(nE') in the energy band E'. The latter corresponds to the energy of the charged particle emitted, in the first scintillator material, by the capture isotope.

[0094] During a step 120, from the counting rate(s) generated by the processing unit 20, the calculation unit 30 identifies each neutron absorber according to its spectral signature, and estimates the quantity Q of each identified neutron absorber. The spectral signature of different absorbers has previously been stored. By spectral signature, is meant for example a list of emission energies (value of the spectrum at predetermined energy bands) or a ratio between values ​​of the spectrum at different energy forms or a form of the spectrum. The object can be virtually discretized into different elementary volumes centered around a coordinate (x,y,z). The calculation unit 30 is then parameterized to estimate a quantity Cj(x,y,z) for each elementary volume. The elementary volumes can be distributed, in the object, according to a two-dimensional or three-dimensional mesh.

[0095] Step 120 assumes a calibration step 90, so as to establish a correspondence between each counting rate TC;(p,E) ​​resulting from the processing unit 20 and the quantity Cj(x,y,z) of each neutron absorber j in an elementary volume of the object centered around the coordinate (x,y,z).

[0096] When the detector is not pixelated, the latter is considered to have only one pixel. In this case, an average quantity Cj of the neutron absorber in the object is obtained.

[0097] Figures 6A and 6B illustrate a simulated measurement configuration. A single-layer scintillator detector 10 formed by 144 pixels distributed in a matrix of 12 pixels by 12 pixels was simulated. The modeled scintillator detector was a plastic scintillator doped with less than 1% 6Li. The modeled scintillator detector is described in the publication Frangville C, “Nanoparticles-loaded plastic scintillators for fast / thermal neutrons / gamma discrimination: Simulation and results”, Nuclear Inst. And Methods in Physics Research, A 942 (2019). Each pixel measured 3 mm on each side. The thickness of the scintillator was 4 mm. The simulated object was a 1 mm thick boron steel plate. Different boron concentrations were simulated, varying between 0 and 7500 ppm.

[0098] The device used was as shown in [Fig.lC]: a 12.5 cm thick HDPE shell was arranged around the assembly formed by the source, the object and the detector. The use of an Am-Be type source, emitting neutrons at energies distributed according to an emission spectrum, the average energy being 4.05 MeV. The distance between the source and the plate was 4.5 cm.

[0099] Figures 6A and 6B respectively represent a section of the device perpendicular to the plane of the detector and in the plane of the detector.

[0100] The MCNP (Monte Carlo N particles) particle transport code was used to determine the thermal neutron reaction rate in the scintillator. The reaction rate corresponds to the number of neutron captures (n,t) in the scintillator material per second. The capture is noted (n,t) because, in addition to the particle a, it generates a tritium nucleus noted t. In [Fig.6A], the abscissa axis corresponds to the concentration of 10B in the plate (unit ppm). The ordinate axis corresponds to the reaction rate (n,t) per neutron emitted by the source. It is observed that the higher the concentration of 10B, the more the reaction rate decreases, due to the absorption of thermal neutrons in the plate. [Fig.6A] corresponds to a calibration function, making it possible to establish a relationship between the quantity of neutron absorber (in this case 10B) and the reaction rate per neutron emitted by the source.In this example, the calibration function can be modeled by a second-order polynomial with a correlation factor of 99.98%.

[0101] As described in connection with Table 1, neutron capture by 10B generates α particles (E = 1.47 MeV at 94% - E = 1.78 MeV at 6%). The emission of α particles can be detected provided that the object is placed in contact or near-contact with a multilayer detector, as described in connection with [Fig.3B]: a first layer 11 allows the detection of photons and neutrons, while a second, superficial layer 12 allows the detection of α particles. A first layer 11 was modeled as described in the previous example, covered with a thin layer of inorganic scintillator ZnS(Ag), forming the second layer. The thickness of the second layer 12 typically varies between 10 pm and 500 pm. The number of α particles reaching the second layer 12 per unit of time was determined.

[0102] Fig. B corresponds to Fig. A: same x-axis, same left y-axis. The right y-axis corresponds to the number of α particles reaching the second layer 12 by neutrons emitted by the source (triangle markers). It is observed that as the quantity of 10B increases, the quantity of particles reaching the detector increases, the latter being proportional to the α counting rate resulting from the detector.

[0103] Fig. B shows that the 10B content can be obtained by using two different calibration functions: one applied to the thermal neutron counting rate in the first layer 11, the other applied to the α counting rate in the second layer 12. The counting rate corresponds to a number of particles detected per second. This demonstrates the benefit of having a 10-layer multilayer detector, making it possible to determine, in addition to thermal neutron count rates, count rates relating to the detection of other types of particles.

[0104] Thus, in addition to the counting rate of thermal neutrons detected in the first layer 11, the counting rate detected in the second layer 12 can make it possible to quantify or confirm the 10B content of the object. Too large a difference between the respective counting rates can indicate the presence of another neutron absorber in the object.

[0105] The use of two count rates relating to different particles, using two different calibration functions, makes it possible to improve the confidence in the estimation of the amount of neutron absorber.

[0106] Figures 8A and 8B illustrate the ability of the device to establish a spatial distribution of the content of a neutron absorber, thanks to the pixelation of the detector. In [Fig.8A], a model of a neutron-absorbing plate, 1 mm thick, segmented into eight elementary volumes is shown. The 10B content in each elementary volume (ppm unit) is indicated. In the plane of the plate, each elementary volume extends along a section of 25 cm2.

[0107] The thermal neutron counting rate in the different pixels of a single-layer scintillator similar to that described in connection with [Fig.7A] was modeled, the pixel area being 25 cm2. Each pixel coincided with an elementary volume of the object. [Fig.8B] shows the reaction rate (n,t), corresponding to the number of particles detected per neutron emitted by the source, in each pixel of the scintillator per pixel. The x-axis corresponds to the concentration of 10B in each elementary volume.

[0108] [Fig.9A] shows another modeled sample. It has four different parts, comprising boron, gadolinium, indium and cadmium respectively. These elements are neutron absorbers. A detector as shown in [Fig.9B] was modeled, each pixel of which is arranged in contact with each part of the sample. The thicknesses of layers 11, 12 and 13 were 4 mm, 300 pm and 30 pm respectively.

[0109] [Fig.10A] shows, for each neutron absorber (x-axis), the detection efficiency, the latter corresponding to a number of particles detected for 1 neutron emitted by the neutron source. The detection efficiency corresponds to the y-axis. Detection efficiencies were determined for different types of particles: thermal neutrons nth, fast neutrons nf, y, [3, a. [Fig.10A] shows disparities in the detection of different particles for the different neutron absorbers. It is understood that the comparison of the detection rates counts detected, respectively for different types of particles, allows identification of a neutron absorbing element.

[0110] Thus, for 10B, the neutron absorption reactions are

[0111] 10B + n 7Li (0.84 MeV) + a (1.47 MeV) + y (0.48 MeV) [94%]

[0112] 10B + n 7Li (1.02 MeV) + a (1.78 MeV) [6%]

[0113] This explains the detection of a, in [Fig. 10A], in the part of the figure corresponding to boron.

[0114] The energy distribution of the detected α particles was simulated. See [Fig. 10B]. Peaks are observed around 1.47 MeV and around 1.78 MeV. These peaks are circled in [Fig. 10B]. The energies observed from 1000 keV correspond to α particles having lost part of their energy between the object and the detector.

[0115] The energy distribution of the particles detected therein was simulated. Cf. [Fig.lOC]. Peaks are observed around 308 keV, 86 keV and 32 keV. Note that the particles detected therein have very predominantly interacted in the first detection layer 11, which is made of an organic scintillator. The probability of absorption by the photoelectric effect is low. The predominant interaction is Compton scattering. Thus, the detected energies correspond to the energies released in the first detection layer during a Compton interaction. The modeling makes it possible to identify characteristic energy peaks, which do not necessarily correspond to the emission energies of the particles therein given the composition of the first detection layer.

[0116] The energy distribution of the y particles detected by the pixel placed directly above the part of the sample containing gadolinium was simulated. See [Fig.10D]. Characteristic peaks are observed at energies different from those observed in [Fig.10C]. Thus, determining the energy of peaks in a y spectrum can make it possible to identify a neutron absorber present in the sample. This constitutes part of the spectral signature of the neutron absorber.

[0117] The energy distribution of the particles y detected by the detector placed directly above the part of the sample containing cadmium (see [Fig. 10E]) and indium (see [Fig. 10F]) was simulated. As in Figures 10C and 10D, it is observed that knowledge of the energy of the particles y detected constitutes a spectral signature of the neutron absorber present in the sample.

[0118] In Figures 10D, 10E and 10F, a peak centered on the energy 308 keV is observed. This peak corresponds to a detection of a photon y emitted by the absorption of a neutron by 10B. This peak is clearly visible in [Fig.10C], and appears attenuated in Figures 10D, 10E and 10F. This is a crosstalk peak, corresponding to the detection of a photon y characteristic of the absorption of 10B by adjacent detection pixels of the pixel located opposite the part of the sample containing boron. The different pixels are separated from each other by a thickness of an optically insulating material, for example Teflon, the thickness of which is adjusted so as to minimize crosstalk.

[0119] In Figures 10B to 10F, the y-axis corresponds to a number of photons detected for a neutron emitted by the source. The x-axis is the energy.

[0120] By using an inorganic scintillator instead of the organic scintillator, the detected spectra would be different, with the presence of photoelectric peaks corresponding to the energies of the photons emitted there. The energies of the different photoelectric peaks constitute part of the signature of each neutron absorber present in the sample.

[0121] [Fig.10G] represents energy spectra of the interactions detected in the second detection layer. The abscissa axis corresponds to the energy and the ordinate axis corresponds to the detection efficiency, i.e. the number of particles for 1 neutron emitted by the source. It is recalled that the second detection layer is, given its thickness and its position, essentially sensitive to particles [3. It is observed that the number of particles [3 detected is much greater for the pixel of the detector placed facing the part of the sample containing gadolinium. This is due to the fact that the capture of neutrons by Gd leads to the emission of particles [3 by the capture reactions, as described in connection with Table 1.

[0122] [Fig.10H] shows the spectrum of the energies of the particles detected by the third scintillation layer for each pixel of the detector. It is known that the capture of a neutron by 6Li releases a tritium nucleus t, with energy 2.73 MeV by the reaction 6Li(n,a)t, as well as an a particle with energy 2.055 MeV. The tritium nucleus t, called triton, is a charged particle, just like an a particle. In [Fig.10H], two peaks are observed, corresponding respectively to the energies of the triton (2.73 MeV) and the a particle (2.055 MeV).

[0123] The invention may be implemented for the characterization, after manufacture, of reactor control elements, for example control rods, for quality control purposes. It may also be applied to neutron-absorbing screens, used for radiation protection. It may also be used to characterize waste resulting from dismantling operations, for example borated concrete, steel or aluminum alloys.

Claims

Claims

1. Method for determining a quantity (Q , Cj(x,y,z)) of a neutron absorber in an object (3), the method comprising - a) arranging the object (3) between a neutron source (2) and a detector (10), the detector comprising at least one pixel (10;), the or each pixel comprising a first detection layer (11) formed of a first material, the first material being a scintillator doped with an isotope suitable for capturing a thermal neutron, each pixel being configured to form a pulse following an interaction of a neutron or a photon y in the first detection layer, the or each pixel being connected to a processing unit (20) configured to discriminate between: • a pulse, formed in said pixel, resulting from an interaction of a neutron in the first layer; • and a pulse, formed in said pixel, resulting from an interaction of a photon y in the first layer;- b) irradiation of the object by neutrons emitted by the neutron source; - c) during the irradiation, determination of at least one counting rate (TC;(n,E')) representative of thermal neutrons detected per unit of time by the or each pixel; - d) from each counting rate resulting from c), determination of the quantity of the neutron absorber in the object.;

2. Method according to claim 1, in which the processing unit is configured to determine a counting rate (TC;(y,E)) representative of a quantity of photons y detected, by the or each pixel, per unit of time for different amplitudes of said interactions.

3. A method according to any preceding claim, wherein - the or each pixel comprises a second detection layer (12) formed from a second material, the second material being a scintillator material different from the first material, the second detection layer being superimposed on the first detection layer, the second detection layer extending between the first detection layer and the object, the second detection layer being thinner than the first detection layer; - the or each pixel (10i) is configured to form a pulse following an interaction of a charged particle in the second detection layer; - the processing unit (20) is configured to discriminate, in the or each pixel, a pulse of a charged particle having interacted in the second detection layer from a pulse of a neutron or a photon having interacted in the first detection layer.

4. The method of claim 3, wherein the or each pixel comprises a third detection layer (13) formed of a third material, the third material being a scintillator material different from the first material and the second material, the third detection layer being superimposed on the second detection layer, the third detection layer extending between the second detection layer and the object, the third detection layer being thinner than the second detection layer; - the or each pixel is configured to form a pulse following an interaction of a particle a in the third detection layer; - the processing unit is configured to discriminate, in the or each pixel, a pulse of a charged particle having interacted in the second detection layer from a pulse of a particle having interacted in the third detection layer.

5. A method according to any preceding claim, wherein - the detector comprises several pixels (10i) adjacent to each other; - the object is virtually segmented into different elementary volumes; - step d) comprises a determination of the quantity of neutron absorber (Cj(x,y,z)) in the different elementary volumes of the object.

6. Method according to any one of the preceding claims, in which the processing unit (20) is configured to discriminate, in the or each pixel, a pulse of a charged particle having interacted with the first detection layer, from a pulse of a neutron or a photon having interacted there in the first detection layer.

7. A method according to any preceding claim, wherein a thermalizing material (5) suitable for thermalizing neutrons is arranged around a space comprising the source, the object and the detector or between the source and the object.

8. Method according to any one of the preceding claims, in which step d) comprises taking into account a calibration function, establishing a relationship between at least one counting rate resulting from a pixel and the quantity of the neutron absorber.

9. Device (1), configured to estimate a quantity of a neutron absorber in an object, the device comprising:

10. - a neutron source (2); - a detector (10), comprising at least one pixel (10i), the or each pixel comprising a first detection layer (11) formed from a first material, the first material being a scintillator doped with an isotope suitable for capturing a thermal neutron, each pixel being configured to form a pulse following an interaction of a neutron or a photon y in the first detection layer; - a processing unit (20), configured to discriminate between: • a pulse, formed in said pixel, resulting from an interaction of a neutron in the first layer; • and a pulse, formed in said pixel, resulting from an interaction of a photon y in the first layer; - the device being configured to place the object between the neutron source and the detector; - the device being such that: • the processing unit (20) is programmed to determine at least one counting rate representative of a quantity of thermal neutrons detected per unit of time by the or each pixel; • the device comprises a calculation unit (30) configured to estimate a quantity of the neutron absorbing element in the object from the counting rates resulting from the processing unit. Device according to claim 9, wherein: - the detector comprises a second detection layer (12) formed from a second material, the second material being a scintillator material different from the first material, the second detection layer being superimposed on the first detection layer, the second detection layer extending between the first detection layer and the object, the second detection layer being thinner than the first detection layer; - the or each pixel is configured to form a pulse following an interaction of a charged particle in the second detection layer; - the processing unit is configured to discriminate, in the or each pixel, a pulse of a charged particle in the second detection layer, from a pulse of a neutron or a y photon in the first detection layer.

11. Device according to claim 10, wherein: - the or each pixel comprises a third detection layer (13) formed of a third material, the third material being a scintillator material different from the first material and the second material, the third detection layer being superimposed on the second detection layer, the third detection layer extending between the second detection layer and the object, the third detection layer being thinner than the second detection layer; - the or each pixel is configured to form a pulse following an interaction of a particle a in the third detection layer; - the processing unit is configured to discriminate, in the or each pixel, a pulse of a charged particle in the second detection layer from a pulse of a particle in the third detection layer.

12. Device according to any one of claims 9 to 11, in which the detector comprises several pixels adjacent to each other.

13. Device according to claim 12, in which the processing unit is programmed to determine a quantity of the neutron absorbing element in different elementary volumes of the object from the counting rates resulting from the processing unit.

14. A device according to any one of claims 9 to 13, wherein the scintillator material forming the first layer is an organic scintillator.

15. A device according to any one of claims 9 to 13, wherein the scintillator material forming the first layer is an inorganic scintillator.

Images