Detector for measuring neutron beam spectra and computer implemented method using such detector

Abstract

A passive neutron detector comprising a core and a moderator body housing said core, wherein the core comprises at least a first and second isotope having different respective responses after exposure to a neutron beam and wherein said responses provide at least first and a second resonance for respective first and second neutron energy and wherein first resonance is in a first neutron energy logarithmic decade and the second resonance is in the next or previous logarithmic decade with respect to the first neutron energy logarithmic decade.

Description

[0001] Detector for measuring neutron beam spectra and computer implemented method using such detector

[0002] FIELD OF INVENTION

[0003] The present invention refers to a detector of a neutron beam spectra, preferably an epithermal neutron beam. PRIOR ART

[0004] Precision, compactness and timeliness of methods to detect neutron beam spectra e.g. generated up to 1.5 MeV e.g. during a Boron Neutron Capture Therapy (BNCT) are improvable. For example, a method based on Bonner Sphere Spectrometer (BSS) requires relatively long time of exposure and a relatively large amount of space required by the measuring device.

[0005] Summary and scope of the invention

[0006] The scope of the present invention is achieved by a neutron detector comprising a core and a moderator body housing said core, wherein the core comprises at least a first and second isotope having different respective responses after exposure to a neutron beam and wherein said responses provide at least first and a second resonance for respective first and second neutron energy and wherein first resonance is in a first neutron energy logarithmic decade and the second resonance is in the next or previous logarithmic decade with respect to the first neutron energy logarithmic decade.

[0007] According to this approach, it is possible to collect with a single detector a sufficient amount of data to provide a more precise estimation of the neutron beam energy spectrum during an unfolding process. Furthermore, the moderator body is such to provide an isotropic behavior also in conditions where the neutron beam is anisotropic.

[0008] In a preferred embodiment, said responses provide a first and a second resonance for respective first and second neutron energy defining a neutron energy interval between the first and second resonance and wherein the core is surrounded by a material shielding the core from neutrons having an energy outside of said energy interval.

[0009] By providing the detector with a shield for unwanted neutrons, e.g. neutrons having an energy outside the interval, the subsequent unfolding process and estimation of the neutron beam spectrum is more precise.

[0010] In a preferred embodiment but non limiting embodiment, the moderator body has a first and a second part detachable from one another so as to extract the core from the detector and the at least two isotopes are detachable from one another so as to be extracted from the core one at a time and so as detect the response of each isotope after exposure to a neutron beam, e.g. by a gamma ray sensor.

[0011] As a matter of fact, each isotope is exposed to the very same neutron beam when housed inside the moderator body. However, the response of each isotope after exposure is detected individually and, to do this, each isotope shall be extracted from the moderator body and, if present, from the shield. The latter shall also therefore be openable so as to house isotopes before exposure to a neutron beam and to extract isotopes after exposure to the neutron beam.

[0012] In a preferred embodiment, the core comprises at least 2 isotopes selected from the following group: 115In, 193Ir, 185Re, 197Au, 152Sm, 165Ho, 121Sb, 175Lu, 179Hf, 186W, 108Pd, 187Re, 158Gd, 1271, 79Br, 139La, 1920s, 196Hg, 75As, 81Br, 71Ga, 164Dy, 98Mo, 141Pr, 96Zr, 87Rb, 176Yb, 122Sn, 55Mn, 104Ru, 76Ge, 68Zn, lOOMo, 96Ru, 86Sr, 63Cu, 142Ce, 30Si, 41K, 23Na, 74Ge, 89Y, 64Ni, 138Ba 51V, 37C1, 175Lu, 141Pr, 48Ca for detecting neutron beams in the range of O.4eV-l.3MeV Such list provides elements with a good compromise of at least one meaningful, e.g. clearly detectable, resonance and sufficient time of activity after illumination by a neutron beam, in general up to 1.3 MeV and preferably within the epithermal range 0.4eV-0. IMeV.

[0013] Preferably, the list is reduced to 115In, 185Re, 197Au, 186W, 187Re, 55Mn, 63Cu, 23Na, 51V, 37C1, 193Ir, 175Lu, 1271, 141Pr, lOOMo, 142Ce, 48Ca to include a good compromise also considering other aspects of the isotope e.g. purity, machining capability, cost, availability. According to the invention, the core may comprise three or more isotopes depending on the span of the energy interval to be detected.

[0014] In a preferred embodiment, the mass of each isotope is 3gr or less and / or the largest dimension of the detector, preferably in the shape of a sphere, a cylinder or spheroid or another revolution shape or central symmetry shape, is no greater than 18 cm, preferably no greater than 12 cm.

[0015] This ensures compact dimensions and, at the same time, collection of a relatively large amount of data for the unfolding process.

[0016] In a preferred embodiment especially designed to detect neutron beams from a Bom Neutron Capture Therapy, the shield comprises Cadmium and / or the core comprises at least five among following elements: 115In, 197Au, 55Mn, 63Cu, 23Na, 51V, 185Re, 187Re, 186W.

[0017] Brief description of the drawings

[0018] The present invention will be now described according to one or more non-limiting embodiments whose description is supported by the following drawings:

[0019] Fig. 1 : schematic steps required to estimate an energy spectrum of a neutron beam;

[0020] Fig. 2: shows a plot of a neutron cross section of selective neutron shielding material such as Cadmium

[0021] Fig. 3 shows respective simulations of response of 197Au (a) and 55Mn (b) Fig. 4 is a table of isotopes useful within a detector according to the present invention

[0022] Fig. 5 is an enlarged perspective view of a core in a detector according to the present invention

[0023] Fig. 6 is a neutron reaction capture scheme

[0024] Detailed description of the invention

[0025] Figure 1 shows a schematic flow of the main steps required to estimate the energy spectrum of a neutron beam. A neutron beam source, e.g. a BNCT device, illuminates a detector 1 comprising, according to the invention, at least two isotopes having at least first and a second response resonance for respective first and second neutron energy and wherein first resonance is in a first neutron energy logarithmic decade and the second resonance is in the next or previous logarithmic decade with respect to the first neutron energy logarithmic decade; after illumination by the neutron beam, the activation by each isotope is separately detected via a gamma ray detector; response curves showing resonances are simulated e.g. via Monte Carlo simulations e.g. plotted and, based on such responses and on gamma ray data, according to a process known to the skilled man in the art, a guess spectrum curve is provided in order to start an unfolding computer implemented process. In particular, a response represents the counts obtained by each isotope, for unit neutron fluence, as a function of the energy and is estimated by known computer implemented simulators e.g. Monte Carlo simulators.

[0026] The detector

[0027] One of the scopes of the present invention is to increase the sensitivity of the detector over a given neutron energy range, e.g. the epithermal neutron energy range. Therefore, special attention has been given to making this instrument insensitive to thermal neutrons. This consideration is due in part to the fact that, in most cases, the neutron capture cross-section in the thermal range assumes values that are equal to or even higher than resonance integrals. Moreover, the thermal fluence rate can be easily measured with a standard sub-Cd subtraction technique with passive detectors or using a calibrated active detector. Without proper shielding of the instrument from thermal neutrons, it would have been even more complicated to extract the data of interest, as the final activity would have been generated by both thermal and epithermal neutrons simultaneously. To mitigate this problem, according to a non-limiting example, a Cadmium shield would be placed around the elements / isotopes of the detector core. The capture cross-section of cadmium in the thermal range is such that it is possible to consider this energy component almost completely suppressed. An energy value of 0.4 eV is typically used as the threshold to discriminate between the thermal and epithermal ranges, corresponding to the drop in the neutron absorption crosssection of cadmium. For illustrative purposes, Fig. 2 shows the capture cross-section of 113Cd, which represents 13.47% of the natural isotopic abundance and contributes significantly to the absorption cross-section with a massive value of o0=20615(400) b. In case a different neutron energy range shall be detected, other shield materials can be used e.g. Aluminum and / or Polytetrafluoroethylene (PTFE).

[0028] To evaluate the induced activity in the selected elements, the following study using Monte Carlo modeling has been chosen.

[0029] For this purpose, known isotopes were simulated using a known software e.g. the FLUKA simulation code by CERN. In particular, all simulations were implemented as follows:

[0030] • Squared plates of different isotopes with dimensions: 10 mm x 10 mm x 0.1 mm;

[0031] • 0.5 mm of natCd shield covering the foils;

[0032] • Monochromatic neutron sources ranging from 1.5 meV up to 1.16 MeV; • The capture (n,y) reaction rate inside the foils has been studied in function of the produced neutron energy;

[0033] The output of these simulations was helpful to determine whether the values of the effective resonance energies were correct or if there were other structures in the cross-section that could potentially pose issues for the ultimate goal. In particular, the aim is to avoid using elements with noticeable continuous structures in the epithermal region to be as sensitive as possible to a narrow energy range for each individual element.

[0034] Figures 3a, 3b show the results for two among the simulated isotopes. The x-axis is in logarithmic scale / decades, since the plots are covering nine order of magnitude in energy. On the y-axes is shown the quantity R(E) that corresponds to the number of capture reaction normalized for source particle. It should be noted that this quantity is the "response" in fact, it represents how the isotope responds to neutron energy.

[0035] The single plot will be also called "response curve" and the ensemble of more plots will be called "response curves" or "response matrix". Such nomenclature is quite often used in neutron physics. The output of these simulations was helpful to determine whether the values of the effective resonance energies were correct or if there were other structures in the cross-section that could potentially pose issues for the ultimate goal. In particular, the aim is to avoid using elements with noticeable continuous structures in the epithermal region to be as sensitive as possible to a narrow energy range for each individual element.

[0036] As can be clearly seen, the resonance structure is clearly visible as a massive peak in the response curves. On the other hand, the absolute capture rate reaction is very different between Au and Mn (both of them are mono-isotopic still in the natural abundance). The response for the gold plate at its peak is around 0.45 • 10-3 (A.U.), while for the manganese foil is around 0.45 • 10-5 (A.U.). That means that, assuming a flat epithermal neutron spectrum, to obtain the same activity for manganese a thickness 100 times higher is required. In this case a 0.1 mm thick gold plate is more or less equivalent to a 1 cm thick manganese plate.

[0037] One of the scopes of the invention is to obtain a system able to work in single neutron irradiation to reduce the total time required. A significant difference in the plate thicknesses or in general in the isotope mass, would not allow to fulfill this requirement while maintaining compact dimensions for the final geometry. Fortunately, it is not necessary to have exactly the same activities among the different isotopes. However, with the aim of being able to perform a correct and short measurement with the gamma spectrometer / detector, it would be enormously advantageous to have activity values that are similar between each others. In fact, high activity results in an increase in the detector dead time, causing problems during data analysis. Therefore, significantly different activity values would in someway compromise the success of the measurement. Attempting to increase the activity of elements with a lower response would result in excessively high activity in elements with a higher response, creating a hardly manageable disparity.

[0038] In order to perform a first selection os suitable isotopes, two parameters were mainly taken into account:

[0039] • Effective resonance energy.

[0040] • Expected activity values after irradiation.

[0041] In particular, the second criterion was estimated knowing that after an irradiation time tin the activity of the foil can be written as:

[0042] Where X = 1 / r is the decay constant of the activated element. In the condition of a long irradiation (tin » T ), the saturation condition is reached, the number of active nuclides produced in the unit of time is equal to the decays number, A=Ar, and the equilibrium is achieved.

[0043] As already shown in the epithermal resonance integral I defined as:

[0044] Where sigma is the cross section and activation rate Arin an epithermal field given by (suitable adjustments are possible for any energy range of interest):

[0045] Being <|)' epi the epithermal fluence rate and N the number of target nuclei, the activation rate Ar can be calculated knowing the fluence rate distribution and the cross section in energy.

[0046] In this case, instead of the cross section, the response curve, normalized respect the atomic density, was used. The equation can be rewritten as:

[0047] Where i represents the i-th energy bin and the summation over i is performed from 1.5 meV up to 1.16 MeV, where the response curves are defined.

[0048] The fluence rate Q (E) used for this calculation refers to a known functioning parameter of the neutron beam source e.g. a reconditioned Elekta Precise 18 MV accelerator such the one located at the filing date of the present application at the Physics Department of the University.

[0049] As the irradiation time may not consistently meet the condition tin » T , the value Arholds the role of upper limit for numerous elements. The actual activity, as indicated above necessitates an additional adjustment to account for the time interval elapsed between the conclusion of irradiation and the starting of measurement using the gamma detector (tWait)- The modified formula become:

[0050] In order to obtain the final estimate of activity, equations A and B were combined. The results of this combination can be seen in Fig. 4 under the Af column. It is essential to note that this table provides an estimate for an irradiation time of tin = 1800 seconds and a subsequent waiting time of twait=600.

[0051] As previously mentioned, the simulated isotopes in the form of plates were characterized by dimensions of 10 mm x 10mm x 0.1 mm. The final values achievable in the detector may vary significantly depending on the plate thicknesses and the specific geometry employed, making these values preliminary estimates.

[0052] With a complete list of elements and relevant parameters, it is possible to select the ideal candidates, which were categorized into two groups based on their characteristics. This further classification involved applying cost and machinability criteria.

[0053] Tables I and II show the final selection of the elements. Two elements, V and Ca, have been included in the final selections, although they were not considered in the initial steps. This was because candidates with an effective resonance energy beyond 1 keV exhibited less-than-optimal characteristics for the intended application. Therefore, V and Ca were reintroduced into the list, despite their earlier exclusion due to their short half-lives (less than 10 minutes). In particular, in the case of vanadium, its inclusion was also based on its affordability, isotopic purity, and ready availability.

[0054] Table I

[0055] T

[0056] As a matter of fact, depending on the span of the neutron energy range to be detected, isotopes may be exchanged between Table 1 and II provided that selected isotopes have a corresponding response with at least a corresponding resonance so that each logarithmic decade of the neutron energy interval of interest has one of said corresponding resonances.

[0057] A putative starting geometry of a core of the detector is a multilayer of plates housed inside an openable container of a selective neutron screen material such as Cadmium. Preferably, the screen container is a cylinder with a wall thickness of 0.5 mm (Fig. 4).

[0058] The initial analysis involved recalculating response curves for the scenario in which all the plates were placed together. In this configuration, the real isotopic abundances were incorporated, as none of the selected isotopes required enrichment for practical use. Moreover, the plates dimensions in the simulations matched those of the plates purchased, as detailed in Table III. The indicated thickness values were evaluated and chosen considering not only the plate thickness but also factors related to the real isotopic abundances and the circular shape of the plates used.

[0059] Isotope Diameter (cm) thickness (pm) Mass (g) No of plates

[0060] In 0.635 127 -0.1249 1

[0061] Au 0.635 51 -0.123 2

[0062] Mn 0.635 51 -0.037 4

[0063] Cu 0.635 127-254 -0.144-0.288 8-4

[0064] Na 0.635 1905 -0.482 2

[0065] V 0.500 500 -0.249 1-2

[0066] Re 0.600 1000 -2.619 1

[0067] W 0.625 1000 -2.416 1

[0068] Table III

[0069] The formula for correcting the values shown in Table III is:

[0070] Where FjSis the isotopic fraction, V* / V represents the correction due to the different volume and A*r / Arrepresents the correction due to the different activation rate (extracted by the Monte Carlo simulations). A(tirr, twait) is the activity value as defined by Fig. 6 and by:

[0071] In order to achieve an isotropic behavior i.e. the response of the core having a negligible dependence from the angle of incidence of the neutron beam with respect to a direction perpendicular to the plates, the core is housed within a body of a moderator e.g. HDPE or another suitable hydrogenous-rich material such as plastics or polymers. Neutrons subject to elastic scattering experience random directional changes, modifying the angular distribution of incident neutrons on the moderator. For sufficiently thick moderators, scattering phenomenon becomes dominant, rendering the angular distribution of incident neutrons largely insignificant. Under these conditions, the new angular distribution can be approximated as isotropic, since neutron come from all the possible directions, without any preferences.

[0072] Figure 7 shows an embodiment of the present invention wherein the moderator body is a revolution body or e central symmetry body, in order to favor the isotropic response curve where isotopes of the core come in the form of plates. Accordingly, plates as well have either an axis or a point of symmetry. In view of simulations performed using e.g. FLUKA software, a maximum dimension of the moderator body e.g. a diameter in case of a spherical shape, is preferably 16 cm or less, more preferably 12 cm or less, even more preferably 10 cm.

[0073] Furthermore, the moderator body comprise a first and a second portion detachable from one another so that the core can be extracted from the body. Preferably, the first and second portions define respective cavities each of which houses by interference the core so that the moderator fully surrounds the core when the portions are pressed one against the other fitting on the core and, when the core shall be extracted, the portions are pulled apart.

[0074] The gamma detection

[0075] After being illuminated by a neutron beam, the core emits gamma rays to be detected by e.g. High Purity Germanium (HPGe) detector. It is also possible to use a portable scintillation detector comprising e.g. a crystalline structure comprising cerium-doped lanthanum bromide. In both cases, it is possible to detect the activation of the core as a whole or the activation of each isotope, one at a time. The output of the detection provides data to compute activation of isotopes after illumination by the neutron beam, the activation being an estimation of neutron count per second (cps), preferably assuming that each isotope works as a point-like source.

[0076] The unfolding computer implemented process

[0077] While taking measurements there is always an alteration caused by the detector itself, which biases the experiment. This aspect is massively relevant for neutron application.

[0078] Such bias may lay in the fact that the instrument doesn’t have a 100% efficiency, or it is due to the finite resolution of the electronics or even other sources. All these problems cause the experiment to drift from the ideal measurement expected.

[0079] In order to find the quantity that has been measured, the operator must correct such alterations. The deconvolution of the data is performed by creating a matrix that contains the information about the detector’s smearing of the desired true quantity. This deconvolution is also known as "unfolding".

[0080] In neutron physics, while aiming at the determination of dosimetric quantities that are strongly dependent on the neutron energy, the neutron spectra must be measured. Measuring the neutron spectrum has the difficulty of being present at a wide range of energies (around 10 orders of magnitude spread) making it particularly hard to determine.

[0081] Since every isotope of the core is particularly sensible to a certain range of energy, but not exclusively sensible to said range, the eventual neutron spectra lies hidden in a sum of different measurements from each sphere at each energy. Such superposition can be calculated using what is called the Fredholm Integral Equation of the First Kind:

[0082] Where G are the counts measured by the ith sphere, Rt(E) is the response function of the ith sphere at a given energy E, and is the neutron spectra in place. A myriad of solutions are known to the skilled man to solve the Fredholm equation in its discrete version. Despite such widespread variation of offered solutions, one common ground on each of them is the need of a "guess spectrum". Due to the fact that the phase space composing all the possible solutions is massive (consider a case of 100 bins on the spectra, with a precision of 10-4 on each bin, it would imply up to 10400 mathematically possible solutions) unless the numerical routine is given a hint on the region in which to find the solution, it would take a near-infinite amount of time to check every possible case, not to mention the fact that there may be more than one possible solution that meets the stopping requirements of the applied algorithm. In order to give the computer implemented algorithm t on where to find the solution, the typical approach is to offer a "Guess spectrum", which is essentially the result of a Monte Carlo simulation or a diffusion model calculation, which leads to an expected result on the measurement. Such spectra is given to the algorithm in order to have a starting point and a region around which to look for the optimal result.

[0083] The most relevant methods that will consequently be elaborated can be grouped in 4 families: Iterative (e.g. SPUNIT algorithm, GRAVEL algorithm), Parametric (e.g. MITOM algorithm), Heuristic (e.g. MAXED algorithm), Artificial Intelligence (e.g. Artificial Neural Network).

Claims

CLAIMS1. A passive neutron detector comprising a core and a moderator body housing said core, wherein the core comprises at least a first and second isotope having different respective responses after exposure to a neutron beam and wherein said responses provide at least first and a second resonance for respective first and second neutron energy and wherein first resonance is in a first neutron energy logarithmic decade and the second resonance is in the next or previous logarithmic decade with respect to the first neutron energy logarithmic decade.

2. The detector according to claim 1, wherein said responses provide a first and a second resonance for respective first and second neutron energy defining a neutron energy interval between the first and second resonance and wherein the core is surrounded by a material shielding the core from neutrons having an energy outside of said energy interval.

3. The detector according to any of the previous claims, wherein the moderator body has a first and a second portions detachable from one another so as to extract the core from the body.

4. The detector according to claim 3, wherein the first and second portion, when contacting one another e.g. during exposure to a neutron beam, are both interference fitted on the core.

5. The detector according to any of the previous claims, wherein the at least two isotopes are detachable from one another so as to be extracted from the core one at a time and so as detect the activation of each isotope after exposure to a neutron beam, e.g. by a gamma ray sensor.

6. The detector according to any of the previous claims, wherein the isotopes are in the shape of a plate having an axis of symmetry.

7. The detector according to any of the previous claims, wherein the core comprises at least 2 isotopes selected from the following group: 115In, 193Ir, 185Re, 197Au, 152Sm, 165Ho, 121 Sb, 175Lu, 179Hf, 186W, 108Pd, 187Re, 158Gd, 1271, 79Br, 139La, 1920s, 196Hg, 75As, 81Br, 71Ga, 164Dy, 98Mo, 141Pr, 96Zr, 87Rb, 176Yb, 122Sn, 55Mn, 104Ru, 76Ge, 68Zn, lOOMo, 96Ru, 86Sr, 63Cu, 142Ce, 30Si, 41K, 23Na, 74Ge, 89Y, 64N1, 138Ba 51V, 37C1, 175Lu, 141Pr, 48Ca for detecting neutron beams in the range of 0.4eV-1.3MeV.

8. The detector according to claim 7, wherein the group is 115In, 185Re, 197Au, 186W, 187Re, 55Mn, 63Cu, 23Na, 51V, 37C1, 193Ir, 175Lu, 1271, 141Pr, lOOMo, 142Ce, 48Ca.

9. The detector according to any of the previous claims, wherein the mass of each isotope is 3gr or less and / or the largest dimension of the detector, having a revolution shape or central symmetry shape, is no grater than 18 cm, preferably no grater than 12 cm.

10. Method of estimating a neutron beam energy spectrum comprising the step of illuminating by the neutron beam a detector according to any of the previous claims and detecting in situ the activity of the detector via a portable gamma ray sensor.

11. Method according to claim 10, wherein the illumination step is operated by a Bom Neutron Capture Therapy device, the shield comprises Cadmium and / or the core comprises at least five among following elements: 115In, 197Au, 55Mn, 63Cu, 23Na, 51V, 185Re, 187Re,

Images