Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
THERMAL-NEUTRON DETECTORS NOT MAKING USE OF HE-3, AND METHOD FOR THEIR MANUFACTURING
Document Type and Number:
WIPO Patent Application WO/2014/188458
Kind Code:
A1
Abstract:
A method for producing thermal-neutron detectors that do not make use of 3He, characterized in that it envisages the following steps: I) producing neutron converters by deposition of thin films of 6LiF on various solid substrates, such as glass, plastic, silicon, scintillating materials, such as CsI and the like; II) coupling one or more thin layers of the neutron converter thus obtained to a scintillator; and III) detecting said scintillation photons via one or more photosensors. The invention also regards a thermal-neutron detector, characterized in that it is constituted by a single scintillator bar, deposited on one face of which either directly or via a purposely provided solid substrate is a neutron-converter material, said bar being coated with reflecting material and being coupled at the two ends to photosensors designed to detect the scintillation light produced by the particles that penetrate into the scintillator.

Inventors:
FINOCCHIARO PAOLO (IT)
MARCHETTA CARMELO (IT)
CONSENTINO LUIGI GIOVANNI (IT)
PAPPALARDO ALFINO DOMENICO (IT)
BARBAGALLO MASSIMO (IL)
Application Number:
PCT/IT2014/000116
Publication Date:
November 27, 2014
Filing Date:
April 29, 2014
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
INFN ISTITUTO NAZ DI FISICA NUCLEARE (IT)
International Classes:
G01T3/06
Domestic Patent References:
WO2012146415A22012-11-01
WO2009117477A22009-09-24
Foreign References:
EP0682268A21995-11-15
US20070272874A12007-11-29
US7141799B12006-11-28
EP2256177A12010-12-01
US20120280132A12012-11-08
US20100294943A12010-11-25
US20070057194A12007-03-15
Attorney, Agent or Firm:
SARPI, Maurizio (Via Collina 36, - Rome, IT)
Download PDF:
Claims:
CLAIMS

1) A method for producing thermal-neutron detectors that do not make use of 3He, characterized in that it envisages the following steps:

1) producing neutron converters by deposition of thin films of 6LiF on various solid substrates, such as glass, plastic, silicon, scintillating materials, such as Csl and the like; and

II) coupling one or more thin layers of the neutron converter thus obtained to a scintillator, such as, for example, CsI(Tl), which emits scintillation photons when struck by the triton or alpha particle generated by fission of a 6Li atom;

III) detecting said scintillation photons via one or more photosensors, producing electrical signals corresponding to the detection of tritons and/or alpha particles, which constitute a signature of the impact of a neutron.

2) The method as per Claim 1, characterized in that it envisages depositing the converter film directly on the scintillator via an appropriate evaporator system and a fine control of the operating parameters .

3) The method as per the preceding claims, characterized in that it envisages collecting the scintillation light from two or more sides of said detectors, using a balancing algorithm for deriving the total amount of light produced (irrespective of the point of impact of the neutron) and the position of impact.

4) The method as per the preceding claims, characterized in that it envisages discriminating the gamma rays by introducing an appropriate threshold value in amplitude of the total light signal, which makes it possible to select detection of tritons and/or alpha particles, which constitute the signature of impact of a neutron.

5) The method as per the preceding claims, characterized in that it envisages use of reflecting material or a geometrical reflector for enveloping the scintillator and guiding the light produced towards the photosensors set at the two ends.

6) The method as per the preceding claims, characterized in that, to obtain a configuration of a single-scintillator detector, it envisages the steps of:

depositing a converter layer on one or both faces of a thin scintillator bar preferably made of monocrystalline inorganic material;

coating said bar with reflecting material; and coupling it at the two ends to photosensors by gluing .

7) The method as per Claim 6, characterized in that as scintillation material thallium-doped caesium iodide is used, in that the coating reflector is constituted by white reflecting material or tape, or else by a geometrical reflector, and in that optical coupling to the photosensors is obtained with transparent silicone grease or transparent resin.

8) The method as per the preceding claims, characterized in that, to obtain a multiple- scintillator configuration, it envisages the steps of:

assembling together in strict mutual contact a plurality of thin scintillator bars on which a 6LiF film is deposited; and

optically coupling said assembly to two distinct photosensors set at the opposite ends. 9) The method as per the preceding claims, characterized in that, in order to increase the detection efficiency of a single-scintillator- detector configuration, it envisages the steps of: coupling the scintillator bar at the two ends to photosensors using transparent optical silicone grease or gluing it using transparent resin; and depositing the converter on a thin foil of reflecting material, in particular aluminium, in which the scintillator bar is wrapped, taking care to set the 6LiF deposition in contact with the scintillator, and guiding the light produced towards the photosensors set at the two ends.

10) A thermal-neutron detector, characterized in that it is constituted by a single scintillator bar, on one face of said bar there being deposited, either directly or through a purposely provided solid substrate, a neutron-converter material, said bar being coated with reflecting material and being coupled at the two ends to photosensors designed to detect the scintillation light produced by the particles that penetrate into the scintillator.

11) The thermal-neutron detector as per Claim 10, characterized in that the scintillator bar is constituted by thallium-doped caesium iodide (Csl (Tl) ) , whereas as neutron converter 6LiF is used, i.e., lithium fluoride enriched in the isotope 6 of lithium.

12) The thermal-neutron detector as per Claim 11, characterized in that, as reflecting material, tape, paint, or other white material is used and, for optical coupling to the photosensors, a resin transparent to visible light is used.

13) The thermal-neutron detector as per Claims 10 to 12, characterized in that the photosensors are silicon photomultipliers (SiPMs) .

14) The thermal-neutron detector as per Claims 10 to 13, characterized in that it is constituted by a plurality of 6LiF layers interspersed with a plurality of CsI(Tl) bars assembled together in strict mutual contact and optically coupled to photosensors .

15) The thermal-neutron detector as per Claims 10 and 11, characterized in that it is a sandwich constituted by a plurality of independent layers of 6LiF and Csl (Tl) , where each CsI(Tl) layer is individually wrapped in a thin reflecting foil, preferably of aluminium, on which the 6LiF film is deposited so that each layer of the sandwich has the 6LiF film in contact with both faces.

16) The thermal-neutron detector as per Claims 10 and 11, characterized in that it is made in the form of a sandwich with multiple alternating thin converter and scintillator layers, deposited by evaporation or some other technique for producing thin films.

17) The thermal-neutron detector as per Claim 16, characterized in that the 6LiF films have a thickness of not more than 2 μπ\, preferably 1.8 μιτι, whereas the Csl (Tl) films have a thickness of less than 30 μπι, preferably 10 μπι.

Description:
THERMAL-NEUTRON DETECTORS NOT MAKING USE OF HE-3, AND

METHOD FOR THEIR MANUFACTORING

•k -k -k -k -k

DESCRIPTION FIELD OF THE INVENTION

The present invention relates to high-efficiency thermal-neytron scintillation detectors that do not make use of 3 He, and more in particular regards a method for producing thermal-neutron scintillation detectors, which enables configuration of thermal- neutron detectors that do not make use of 3 He in a flexible way in order to optimize case by case some characteristics thereof according to the application. In particular, it is possible to provide detectors that maximize detection efficiency, which are sensitive to the position with a millimetric or submillimetric precision.

PRIOR ART

Detection of thermal neutrons is mainly based upon the use of detectors used for detection of charged particles with materials referred to as "converters" that react in an efficient way with the neutrons, enabling detection thereof.

Irrespective of the technique or the state of the detector, i.e., whether it is a scintillation detector, a solid-state detector, or a gas detector, a good converter must present the following characteristics:

it must have a large cross section so as to maximize the efficiency of the detector and reduce its spatial dimensions; and

the energy released in the reaction Q must be high so as to have products that leave in the detector signals that are clearly distinct from those coming from the gamma background, which are also associated to the neutrons.

The noble gas 3 He, associated to which is an energy Q of 0.76 MeV and a branching ratio of 100%, has been so far widely used for detection of neutrons through the reaction

3 He+ n→ 2 H + p+ 0.764 MeV

For thermal neutrons, the Q-value of the reaction leads to proton energies of 0.573 MeV and tritium energies of 0.191 MeV. The cross section is relatively large, namely, 5330 barn, and shows a dependence of the 1/v type with respect to the energy.

In the field of the measurements made of fissile material for the purposes of safeguarding, the majority of the detectors used are counters proportional to the amount of 3 He .

In nature 3 He is very rare but, starting from 1955, huge amounts have been obtained from dismantling of U.S. nuclear weaponry. In the past years, more than 200 000 litres of 3 He have been accumulated thanks to the decay of tritium for nuclear warheads of U.S. production. Currently, however, against an availability of approximately 8000 litres per year, there is a demand of several tens of thousands of litres per year of 3 He .

Known from EP0682268 is the use of a neutron converter, chosen from among the existing ones, in which Csl is used as electron-emitting material. Said electrons are then multiplied and detected by means of an avalanche chamber or in vacuum conditions via an array of microchannels (microchannel plate) .

According to the present invention, and unlike EP0682268 where pure Csl is used, which is an electron emitter, there is envisaged the use of CsI(Tl) or of other scintillators for detecting directly the scintillation light.

Also known from US7141799 is a stacked structure for the combined detection of gamma rays, and fast and thermal neutrons. Said structure also thermalizes the neutrons (via various polyethylene layers), hence loosing the capacity to make a clear detection of originally thermal neutrons.

Detection of thermal neutrons in US7141799 is based upon scintillating fibres of 6 Li glass, i.e., scintillating fibres made of glass loaded with 6 Li. Such a glass is known to present various drawbacks: it is rather opaque (it has a light-attenuation length that is very short, and hence its thickness must be small, and is produced in the form of minute fibres having a short length of a few centimetres); it contains 6 Li within its formulation, and this is the main cause of its opacity;

the 6 Li content of this glass is limited and cannot go beyond a small percentage; and

the scintillation yield is poor.

A purpose of the present invention is also to avoid a structure such as the one described in US7141799, and this is the reason why, according to the present invention, scintillators with good performance are used coupled to independent external neutron converters .

The document No. US 2007/272874 describes the use of CsI(Tl) for detecting gamma rays, and not neutrons, which are detected in another part of the device. According to the present invention, CsI(Tl), or some other scintillating material, is used for detecting alpha particles and/or tritium, which are the signature of neutrons after capture in 6 Li . Via the use of a threshold value, the discrimination is then made between gamma rays and neutrons.

The invention disclosed herein proposes as an alternative solution to the use of the 3 He gas for detection of neutrons, providing a detector that does not make use of said gas, is compact, and has a low cost, and can be used in numerous applications both in the nuclear-physics research field and in monitoring radioactive waste and sites for storage and treatment of exhausted nuclear fuel bars, for personal and environmental dosimetry, as likewise in structural investigations of materials that resort to the use of neutron beams.

SUMMARY OF THE INVENTION

A first aspect of the invention is a method for producing neutron detectors that do not use 3 He by means of detection techniques via scintillators of a commercial type, which envisages:

coupling to a scintillator a substrate made of a suitable material deposited on which is (for example, by evaporation) a thin film of 6 LiF neutron converter;

or alternatively,

depositing one or more thin layers of converter directly on the scintillator.

A second aspect of the invention is to implement the method just described to arrive at schemes of configurations of detectors in such a way as to optimize some characteristics thereof on a case-to- case basis according to the application, such as the increase in neutron-detection efficiency and a good gamma rejection, small dimensions, sensitivity to the impact position, and low cost. In particular, recourse is had to the strategy of: collecting, via photosensors, the scintillation light from two or more sides of said detectors, using a balancing algorithm for deriving the total amount of light produced (irrespective of the point of impact of the neutron) and the position of impact; identifying the events due to the neutrons and consequently discriminating the gamma rays, by introducing an appropriate threshold value in amplitude of the total light signal, which makes it possible to select detection of tritons (and/or alpha particles), which constitute the signature of impact of a neutron; and

using the reflecting material or a geometrical reflector for enveloping the scintillator and guiding the light produced towards the photosensors set at the two ends.

In conformance with the above, to obtain a configuration of single-scintillator detector, it is envisaged to:

deposit a converter layer on one face of a thin scintillator bar, which is preferably made of monocrystalline inorganic material, such as for example CsI(Tl), i.e., thallium-doped caesium iodide; coat said bar with white reflecting material or paint (such as, for example, a wrapping of the bar in white reflecting Teflon tape) ; and

couple said bar, at the two ends, to photosensors using techniques currently known in the sector, for example with transparent silicone grease or gluing using transparent resin.

It should be noted that, even though CsI(Tl) is an excellent scintillator, it is not the only one that can be used according to the present invention. Tests have shown that also other scintillators have yielded very interesting results.

To obtain a configuration with a number of scintillators, it is envisaged to:

assemble together in strict mutual contact a plurality of thin scintillator bars on which a 6 LiF film is deposited; and

optically couple them to two distinct photosensors set at the two opposite ends.

And once again to increase beyond a factor of two the detection efficiency of a single-scintillator- detector configuration, it is envisaged to:

couple the scintillator bar at the two ends to photosensors applying known techniques, for example using transparent optical silicone grease or by gluing with transparent resin; and

deposit the converter on a thin foil of reflecting material, in particular aluminium, in which the scintillator bar is wrapped, taking care to set the deposit of 6 LiF in contact with the scintillator, and guide the light produced to the photosensors set at the two ends.

It is on the other hand possible to sandwich a plurality of bars each provided with a reflector such as the one illustrated above to obtain a multiple configuration where the photosensors are able to detect neutrons that interact in any one of the bars constituting the sandwich.

The invention will now be described in greater detail with reference to the attached plates of drawings, which illustrate, purely by way of non- limiting example, some preferred embodiments thereof. In the plates of drawings:

Figure 1 is a scheme of the reaction of neutron capture by 6 Li;

Figure 2 is the scheme of thermal-neutron detection using a converter and a scintillator: the products of reaction are emitted in opposite directions, and hence just one of them can be detected: in a) 3 He (triton) is detected, in b) He (alpha particle) is detected;

Figure 3 is a diagram provided by way of example of assembly for a detector measuring 5 cm x 3 mm x 1 mm;

Figure 4 is a diagram of the complete detector;

Figure 5 is an alternative configuration in which the scintillating crystal is without converter, which is, instead, deposited on a thin aluminium foil in which the crystal itself is wrapped;

Figure 6 is an example of histogram obtained exposing the detector of Figure 4 to a source of neutrons thermalized via polyethylene; appearing on the abscissae and ordinates in arbitrary units are the amplitude of the signals measured at the two ends, while the grey scale indicated on the right represents the frequency; it is to be noted that the source emits also a considerable amount of gamma radiation, but notwithstanding this the detector manages to distinguish the neutrons clearly;

Figure 7 shows the distribution of the amount of light emitted by the scintillator;

Figure 8 is yet another example of histogram obtained by exposing the detector of Figure 4 to a source of neutrons thermalized via polyethylene; appearing on the abscissae in arbitrary units is the position of impact, whereas appearing on the ordinates, once again in arbitrary units, is the amount of light produced; the third co-ordinate (the grey scale on the right) represents the frequency; the precision in position has been estimated as being in the region of 1-2 mm;

Figure 9 shows the steps of production of a multilayer configuration: various scintillator bars on which a 6 LiF film is deposited are assembled together in strict mutual contact (Figure 9, at the top) , and then optically coupled to the photosensors (Figure 9, at the bottom) exactly as in the case of a single scintillator;

Figure 10 is the scheme of detection of both of the products of neutron capture in the case of a sandwich detector;

Figure 11 shows the sandwich configuration with independent layers, where each scintillator layer is individually wrapped in a thin reflecting foil (for example, aluminium) on which the 6 LiF converter film is deposited; and

Figure 12 shows possible multiple planar configurations and multiplane configurations.

DETAILED DESCRIPTION OF THE INVENTION

The present invention uses as neutron converter the material 6 LiF, i.e., lithium fluoride, which is a very stable fluorine salt enriched in the isotope 6 of lithium. The typical enrichment required is 95- 96%, but there may be provided configurations with lower enrichment or with the use of natural lithium, which contains approximately 7% of 6 Li, according to the requirements of the particular application.

6 Li possesses the property of having a large capture cross section for thermal neutrons (960b), with a single decay channel that consists in emission of 4 He (also referred to as "alpha particle") and 3 H (also referred to as "triton") at 180° from one another, with kinetic energies of approximately 2.05 MeV and 2.73 MeV, respectively (Figure 1) . As is well known in the literature, as the energy of the incident neutron increases, the capture cross section decreases in inverse proportion to its velocity, and in any case remains still appreciable up to a few kiloelectronvolts . The capturing reaction in question is

*U + n→ ΙΛ → ' Η (2.73 MeV)+'He. (2.05 MeV)

Detection and identification of one of the two particles, or better still of both of them, constitutes a signature of successful detection of a thermal neutron. 6 LiF has an integrated capture cross section for thermal neutrons of approximately 57 cm -1 ; hence, in a thickness of 100 urn , there is a capture probability of approximately 50%. Obviously, it is not possible to provide detectors with such converter thicknesses, in so far as the particles produced by the reaction would remain trapped in the excessive thickness and hence would not reach the detector .

In a first embodiment of the present invention there is envisaged evaporation at high temperature of thin films of 6 LiF neutron converter on various substrates (glass, plastic, silicon, scintillating materials, such as Csl, etc.). Said film may be deposited over even particularly large surfaces (even up to 15 cm x 15 cm) , for example via a suitable evaporator system and fine control of the operating parameters according to a consolidated technique.

Control of the uniformity and of the thickness is obtained via techniques currently known in the literature (quartz oscillator, loss of energy of alpha particles) .

By coupling a neutron-converter film to a scintillator, or depositing it directly thereon, the inventors have obtained specific thermal-neutron detectors. The detectors in question may be separate from and independent of the converter, and in this case the converter itself, deposited on an appropriate substrate, is to be affixed on the detector in such a way as to cause the particles produced following upon neutron capture to be detected by the detector itself.

By way of example, listed below are some substrates on which a 6 LiF film having a thickness of 1.8 urn has been evaporated:

aluminium foils measuring 5 cm x 5 cm, with a thickness of 16 μιτι;

strips measuring 5 cm x 0.3 cm x 0.1 cm and plates measuring 5 cm x 5 cm x 0.1 cm, made of monocrystalline CsI(Tl) scintillating material;

thin plates of glass measuring 5 cm x 5 cm x 0.1 cm; and

sheets of monocrystalline silicon measuring 3 cm x 3 cm x 0.03 cm.

The detector may also be constituted by a sandwich of various alternating scintillator and converter units .

The scintillating material may be chosen from among a wide range existing on the market. The inventors have preferably used CsI(Tl), which presents excellent optical properties and rather good mechanical properties. The technique can be immediately applied to other scintillating materials, in the case where a greater mechanical strength becomes necessary, or a better precision in the measurement of the times of flight, or again it is desired to use photosensors that require a different wavelength of the light to be detected. An example in this sense may be LYSO, which, even though it is denser than CsI(Tl) and hence more sensitive to gamma rays, has a decay constant approximately 100 times faster .

The thickness of the sensitive part of the scintillator may be very small in such a way as to reduce the sensitivity to gamma radiation considerably. To detect the particles produced by neutron capture in the converter, very small scintillator thicknesses are sufficient (just a few tens of microns), which also makes it possible to obtain the scintillator layer by direct evaporation of its components. It is hence possible to optimize the sensitivity of detection of thermal neutrons by reducing as much as possible contamination by gamma rays. The spectrum of the energy deposited on the detector (i.e., of the amount of scintillation light produced) enables identification of the events due to neutron capture, separating them from those produced by gamma interaction.

In the sequel, by "converter" is meant a thin layer of 6 LiF.

Single-scintillator configuration

The single-scintillator configuration is the basic one. It consists of a scintillating element on which a converter layer is affixed, whether deposited directly or using a purposely provided substrate (for example, aluminium) .

In this configuration the neutron capture will give rise to detection of just one of the two particles produced by decay, as illustrated in Figure 2. The particle that penetrates into the scintillator, alternatively triton ( 3 H) or alpha particle ( 4 He) , will produce light scintillation that will be detected by an appropriate photodetection device.

An example of how to provide the basic configuration is described in what follows. A thin scintillator bar is used, preferably made of monocrystalline inorganic material such as Csl (Tl) (thallium-doped caesium iodide) , deposited on one face, or on both faces, of which is a converter layer. The bar is coated with white reflecting material (for example, white Teflon tape, or a white enamel) and then coupled at the two ends to photosensors using transparent optical silicone grease or glued with transparent resin. After gluing, a further touching-up at the ends using enamel or white tape may prove necessary.

In the example of Figure 3, Teflon tape is used as reflecting material and a bicomponent resin transparent to visible light is used for optical coupling to the photosensors. The photosensors used in the example are silicon photomultipliers (SiPMs). All the components listed (apart from the converter film) are commercially available.

In particular, in Figure 3 the converter is directly deposited on one face of the scintillator, in this case Csl (Tl) . The ensemble is then wrapped in reflecting white Teflon tape, and finally glued to two SiPM photosensors with transparent resin (Figure 4) .

An alternative configuration, which increases the detection efficiency beyond a factor of two, consists in depositing the converter on a thin aluminium foil, having a thickness of about ten microns, in which the scintillator bar is wrapped taking care to set the 6 LiF deposition in contact with the scintillator. In this configuration, represented schematically in Figure 5, a geometrical reflector (aluminium) is hence used for enveloping the scintillator and guiding the light produced towards the photosensors set at the two ends.

It should be noted that the detector can function also with a single photosensor, even though with lower performance.

Operation of the detector

The operating principle of the detector is based upon detection of at least one of the two products of the reaction of neutron capture. The main problem of the existing neutron detectors is gamma discrimination. In fact, practically all existing detectors are also sensitive to gamma radiation, and consequently a good detector must be able to present a good gamma rejection. In particular, in the case in question, a technique has been implemented based upon selection of the signals on the basis of the amplitude .

Using a rather small scintillator thickness, the mean energy deposited by gamma radiation that strikes it may be rendered relatively low (thicknesses of 2 mm, 1 mm, 0.3 mm are used) . Instead, since the particles (both triton and the alpha particles) produced by neutron capture stop within extremely small thicknesses (a few tens of microns) , they will deposit all their kinetic energy in the scintillator. Consequently, it amounts to a comparison between the some hundreds of kiloelectronvolts deposited on average by gamma radiation and the 1500 to 2500 kiloelectronvolts deposited by the particles (the loss in the thin thickness of the converter before penetration into the scintillator has been taken into account) .

The scintillating materials, and among these but not exclusively CsI(Tl) used by the inventors for the first prototypes, when struck by gamma radiation or ionizing particles produce scintillation light in an amount proportional to the energy deposited by the particles themselves. The constant of proportionality for tritons and alpha particles is smaller than the one for gamma radiation, roughly a factor of two. Notwithstanding this reduction, using thin scintillators, light signals are obtained having an amplitude that is clearly distinct between gamma rays and particles, that it is possible to convert into electrical signals using photosensors on which amplitude discrimination can be carried out that enables identification of the signals due to neutrons .

The use of a scintillator bar read at both of its ends using two distinct photosensors enables higher levels of performance to be obtained.

The use of the normal coincidence technique, i.e., the technique of imposing that both of the photosensors produce a signal within a pre-set time interval that can range from a few tens of nanoseconds to a few microseconds according to the scintillator that is used, enables abatement of the contribution due to spurious electronic noise.

Surprisingly, however, the correlation between the amplitude of the two signals measured at the two ends enables the neutron signal to be distinguished from the gamma signal, and the position of impact of the neutron detected to be determined with millimetric precision .

Provided by way of example in Figure 6 is a histogram obtained by exposing a detector (measuring 2.5 cm x 3 mm x 1 mm) to a source of neutrons thermalized via polyethylene, which moreover emits a more than considerable amount of gamma radiation both with low energy (59 keV) that with high energy (4.4 MeV) . Appearing on the abscissae and ordinates is, in arbitrary units, the amplitude of the signals measured at the two ends, whereas the third co ¬ ordinate (the grey scale) represents the frequency. From the histogram it may be inferred how, notwithstanding the presence of a considerable background due to gamma radiation, the events due to neutrons will be clearly distinguishable.

The shape of the histogram is due to the fact that the scintillation light, on account of multiple reflections along the path from the point of production as far as the photosensor, is attenuated. Said attenuation is the greater, the greater the distance traversed by the light, with a typical exponential pattern. Consequently, the signal measured at each end will be smaller as the distance of the point of impact from the ends increases, and since moving away from one end means approach to the other end, the pseudo hyperbolic shape of the histogram is explained.

More precisely, for an event that produces an amount of light (number of photons) Q 0 in a point on the scintillating bar at a distance d from the photosensor, the amount of light Q(d) measured will be

Q(d) = ^ 6 PDE e d/ ^ (1)

2 where ε is the optical-coupling efficiency (≤1), PDE is the photon-detection efficiency, and L at the effective light-attenuation length within of the bar.

The important aspect of Eq. (1) is that, for each event detected, by combining the values measured at the two ends, it is possible to obtain the value of the amount of light produced without any dependence upon the various constants just mentioned.

Q>= Q · Q (2)

If hence, event by event, the calculation of Eq. (2) is made, the two values measured can be reduced just to one, proportional to the amount of light produced in the event irrespective of the point of impact (which is proportional because the analog-to- digital conversion systems used supply values proportional to the quantities measured) .

The histogram of Figure 7 shows the distribution of the amount of light emitted by the scintillator of Figure 4 exposed to the neutron source. Appearing clearly from Figure 4 are the neutrons detected, represented by the peak on the right produced by the tritons. The alpha particles produced unfortunately give rise to a smaller amount of light, appearing in the central area of the histogram where a considerable gamma contribution is present. Consequently, the single-scintillator configuration enables easy identification as neutron events only those in which the triton is emitted in the direction of the scintillator (case a of Figure 2) .

The region on the right corresponds to triton detection, which is the signature of successful interaction of neutrons, at the centre is a region where gamma rays and alpha particles are superimposed, and on the left is the region of below- threshold noise.

Starting from Eq. (1) we may easily note that

&

provides a precise information on the position of impact of the radiation detected (i.e., on the point in which the light scintillation has been produced) . The expression, which is to be computed event by event, may also be approximated in various ways. In any case, after appropriate calibration for example using a small point-like alpha, beta, or gamma radioactive source, it is possible to obtain the measurement of the position of impact of the neutron detected .

Provided in Figure 8 is an example of histogram obtained exposing the detector of Figure 4 to a source of neutrons thermalized via polyethylene. Appearing on the abscissae in arbitrary units is the position of impact, whereas appearing on the ordinates is, once again in arbitrary units, the amount of light produced. The third co-ordinate (the grey scale) represents the frequency. The precision in position for this prototype has been evaluated, via calibration with alpha and gamma radioactive sources, as being in the region of 1-2 mm. The perfectly horizontal pattern of the histogram as the position varies provides further confirmation of the goodness of the algorithm of reconstruction of the total light produced. Maltiple-scintillator (sandwich) configuration

The configuration with a number of scintillator layers is based upon the same operating principle as that of the single-scintillator configuration. The basic difference lies in the neutron-detection efficiency. In fact, the main limitation of the single-scintillator configuration arises from the fact that, to maintain a reasonable power of identification of neutrons and discrimination from gamma rays, it is necessary to use 6 LiF films of a thickness not greater than 2 μιη. The prototypes provided by the present inventors make use of films having a thickness of 1.8 μιτι, which from many points of view represents the optimal thickness. Since the macroscopic cross section for thermal neutrons in 6 LiF is approximately 57 cm -1 , in a thickness of 1.8 μπι there is a capture probability of approximately 1%, hence relatively low. A sandwich with ten layers of 6 LiF enclosed by eleven bars for example of Csl (Tl) having a thickness of 300 μπι affords levels of performance similar to those of the single-scintillator detector, but with a capture probability of approximately 10%, which represents an altogether satisfactory detection efficiency.

A scheme of the multilayer (sandwich) configuration is illustrated in Figure 9. Various thin scintillator bars, on which a 6 LiF film is deposited, are assembled together in strict mutual contact (Figure 9, at the top) , and then optically coupled to the photosensors (Figure 9, at the bottom) exactly as in the case of the single scintillator.

Moreover, the sandwich detector presents a further advantage: since the alpha particles and tritons produced by neutron capture are emitted at 180° with respect to one another, in the majority of cases (around 80%) they will both be detected. Consequently, neutron capture will give rise to a light signal of approximately twice the intensity, enabling an even better separation from the gamma background (Figure 10) .

Obviously possible are sandwich configuration with independent layers, where each scintillator layer is individually wrapped in a thin reflecting foil (e.g., aluminium) on which the converter film is deposited (Figure 11). Said configuration does not enable doubling of the light intensity since one of the two particles enters the scintillator, whereas the other stops (or is slowed down enormously) in the reflecting foil. On the other hand, however, each layer of the sandwich will have the converter on both faces. Hence, for example, a sandwich thus obtained constituted by eleven scintillator bars would have twenty-two layers of 6 LiF film, with a capture probability of approximately 22%, which is a highly satisfactory value, considering the size of the detector .

Other possible configurations

Multiple configurations can hence be provided by assembling together a number of detectors of the same type to obtain planar structures and possibly multiplane structures (Figure 12). Finally, it is also possible to produce sandwiches by depositing multiple alternating thin converter and scintillator layers. It is possible, for example, to deposit on an aluminium (or glass) substrate various 6 LiF films having a thickness of 1-2 μπι alternating with CsI(Tl) films having a thickness of 10 μπι, and in this way detectors can be obtained with a very high capture efficiency .

INNOVATIVE ASPECTS

From what has been described so far the innovative characteristics of the present invention are evident, according to the various solutions listed below:

coupling of a converter film, deposited on a thin substrate, to a commercial scintillator (the best from many points of view is CsI(Tl)); main advantages: practicality and flexibility of assembly, separate production of converter and scintillator; direct deposition of converter on scintillator; main advantages: possibility of providing very compact sandwiches and high thermal-neutron-detection efficiency;

deposition of converter films alternating with scintillator films; main advantages: possibility of obtaining very thin sandwiches with high neutron efficiency and negligible gamma efficiency, hence with a very high gamma-rejection capacity;

production of sandwiches constituted by setting on top of one another various sheets/scintillator bars individually wrapped in a reflecting substrate (e.g., aluminium) deposited on which is a converter film; main advantages: compact thermal-neutron detectors, low cost, high efficiency, good gamma rejection, high sensitivity to impact position;

collection of the scintillation light from the above detectors using silicon photomultipliers (SiPMs, commercially available); main advantages: compact thermal-neutron detectors, low cost, high efficiency;

collection of the scintillation light from two or more sides of said detectors, using a balancing algorithm for deriving the total amount of light produced (irrespective of the point of impact of the neutron) and the impact position; main advantages: compact thermal-neutron detectors, low cost, high efficiency, good gamma rejection, high sensitivity to impact position.

identification of the events due to neutrons, and consequent gamma discrimination, via an appropriate threshold value of amplitude of the total light signal, which makes it possible to select detection of tritons (and/or alpha particles), which are the signature of impact of a neutron; main advantages: compact thermal-neutron detectors, low cost, high efficiency, good gamma rejection, high sensitivity to impact position.

SOME POSSIBLE FIELDS OF APPLICATION

Amongst the numerous possible fields of application, we may cite the following:

research in nuclear physics;

national safety, contrast to possible smuggling of nuclear material;

personal and environmental dosimetry;

monitoring of radioactive waste;

monitoring of sites for storage and treatment of exhausted nuclear fuel bars;

search for possible nuclear material accidentally dispersed amongst scrap metal and/or refuse;

oil industry;

structural investigations on materials (with neutron beams) such as time-of-flight neutron diffraction, Bragg-edge Transmission, neutron resonance capture analysis, neutron radiography.