EP1810057A1

EP1810057A1 - Method and device for probing material by photofission

Info

Publication number: EP1810057A1
Application number: EP05777236A
Authority: EP
Inventors: Henri Safa
Original assignee: Commissariat a lEnergie Atomique CEA
Current assignee: Commissariat a lEnergie Atomique et aux Energies Alternatives CEA
Priority date: 2004-06-21
Filing date: 2005-06-14
Publication date: 2007-07-25
Also published as: US20090065693A1; FR2871896A1; WO2006008360A1; FR2871896B1; US9297908B2; JP2008503742A

Abstract

The invention relates to a method and system for probing an object, which contains potentially radioemitting elements, by photofission. The object is bombarded by a beam of electrons of sufficient power and the conversion thereof into photons by Bremsstrahlung is effected directly inside the object to be probed, which is likely to contain photofissile material. No target is added to the packet for said conversion. Preferably, the object that is to be probed is a radioactive waste packet having a thick absorbent concrete container.

Description

"Method and apparatus for probing nuclear material by photofission."

The present invention relates to a method and a device for probing an object containing nuclear material. The latter is irradiated by means of a particle beam whose energy is sufficient to cause photofission of actinide elements contained in this nuclear material.

Photofission is a nuclear fission, that is to say the separation of the nucleus of a heavy atom with release of a certain energy, caused by the incidence of energetic photons, notably on actinides such as uranium and plutonium. . The de-excitation of an impacted atom gives rise to the emission of fast neutrons, and to a much smaller quantity of delayed neutrons. In particular, a neutron detector makes it possible to detect the actinides present. The present invention finds a particularly interesting application in the field of the characterization of nuclear waste packages, in particular to determine their storage location. The term "package" means drums of concrete waste with a diameter of the order of a meter or more, or metal containers whose volume can reach several cubic meters. It is also important, when storing radioactive waste, to know the nature and activity of radio transmitters and in particular actinides. Document US Pat. No. 4,617,169 (Brodzinski) is known which proposes a passive method for the detection of nuclear material in a waste package. This method does not use probe particles. Using both gamma and neutron detectors placed outside the package, one tries to extract either gamma lines characteristic of a radioactive nucleus present, or the neutron emission if necessary. This technique is also commonly used to characterize nuclear waste packages. However, it does not make it possible to determine the alpha emitters, radiations that are entirely stopped in the package and does not allow to probe large packages containing small amounts of radioactive material.

Active methods are also known which are more sensitive and allow a much lower detection limit. In these methods, the waste is subjected to a flux of neutrons and / or photons, itself obtained by electron bombardment of a target placed in the immediate vicinity of the waste drum.

The probe particles used are either photons, neutrons or both.

Photons, usually of the gamma type, can be produced either by radioactive decay or by

Bremsstrahlung. In this case, a beam of electrons of sufficient energy (several MeV) strikes a target generally in tungsten, and the passage of the electrons in the neighborhood very close to the atoms of the target gives rise to the "braking radiation" characterized by a photon gamma, of energy at most equal to the energy of the incident electron.

Neutrons, which constitute the other alternative, can be produced either directly by a neutron source, or from nuclear reactions, by bombardment of a Beryllium target constituting a converter. Thus US 4 497768 (Caldwell) proposes to evaluate only the quantities of fissile or fertile material in a sample from a pulsed irradiation of gamma photons and neutrons. Document US Pat. No. 5,495,106 (Matsny) proposes an assembly for measuring possible contaminations in the ground from pulsed X-ray irradiation. US 5,838,759

(Armistead) proposes a tool for preventing nuclear material from being smuggled by X-ray or neutron bombardment on mobile containers with respect to the X-ray emitter. The document FR 2 726090 (US equivalent 5)

828069) (Umiastowski) describes a counting chamber of neutrons with layers of neutron moderators and layers of cadmium for shielding.

Finally, document FR 2 764383 (equivalent US6452992) (Umiastowski) is known which will be described below in more detail. In general, the interaction of a photon of a few MeV with a nucleus can, if the excitation is sufficient, supply enough energy to the nucleons to cause nuclear reactions. The nucleus is de-excited by generally emitting either one (or more) photons, or one (or more) neutrons, or, more rarely, charged particles (p, α, ...). In the case of actinides, the excited nucleus can also disintegrate by fissioning. There is an energy domain for which the photon capturing cross section by the nucleus is relatively high. This is the domain of Giant Dipolar Resonance (GDR). For intermediate energies (typically ranging from 25 MeV for light cores to 12 MeV for heavier ones), the nucleus is then excited in a collective mode, reflecting a mode of electrical resonance due to the relative movement of protons relative to neutrons. The integral of the total cross section is given by the dipole sum rule

Iσ.dE≈60 NZ) mbarn.MeV

where N, Z and A conventionally refer to the total number of neutrons, protons and nucleons.

It can be seen that the GDR can induce cross sections of a barn fraction, of the same order of magnitude as the standard nuclear reactions (strong interaction), for example neutronics, although it is of electromagnetic origin. It is therefore, at these photon energies, a physical phenomenon of importance.

For minor actinides, the major de-excitation pathways are neutron emission, mainly (γ, n) and (γ, 2n), and photofission (fission disintegration). - A- It turns out that these three paths are practically equiprobable. The fission thus contributes for practically (1/3) of the total and integrated photonuclear effective cross section. Document FR 2 764 383 (US6452992 equivalent) describes a system as illustrated in FIG. 1. A converter 1 is used to generate gamma photons from electrons supplied by a particle accelerator. When the electrons come very close to the nucleus of the atoms of the target constituting the converter 1, they generate a braking radiation and lose energy. Several such interactions take place successively, until the electrons have lost most of their energy. The photons 2 pass through the air to be sent on the package 3 to be probed. It is therefore the photons that are used as a probe. The detection is done through neutron detectors 4 placed around the package. Due to the importance of the "gamma flash" when sending the electron pack, the detectors 4 are blinded in two ways. On the one hand, the huge amount of gamma photons produced by Bremsstrahlung from the electron beam can create ionizations in neutron detectors. On the other hand, the conversion target having a high atomic number produces, under the effect of photon irradiation, parasitic neutrons in a quantity greater than the fast neutrons resulting from photofission.

According to the prior art taught by the documents

US4497768 and US6452992 using a photon probe, the

"Gamma flash" induces background noise several orders of magnitude higher than useful signals. A solution proposed by the document US4497768 would consist of firstly thermalizing the neutrons and then measuring the fast neutrons resulting from fission reactions induced by neutrons, that is to say about 0.5 ms to 2.5 ms after the flash, by surrounding the cadmium detectors to absorb slow thermalized neutrons. The immediate consequence is the generation of a very important neutron environment

(neutron noise) drowning the useful signal.

Another document, US5838759, uses a radioactive source and a Beryllium converter to generate neutrons as a probe particle, and then measures the gamma lines characteristic of elements that captured a neutron using gamma detectors. However, even by this technique a neutron flash blinds the detectors and forces to make measurements after this flash.

Finally, a US5495106 document proposes to use X-rays (actually gamma rays produced by Bremsstrahlung electrons impacting a target) to create photofission in the contaminants present in the soil: uranium, plutonium and Beryllium . Then he proposes to measure, up to 100 ms after the photon source, the neutrons that have been produced by these contaminants. It is therefore necessary to inhibit the detectors 4 during the "gamma flash", which limits this system to the measurement of delayed neutrons, arriving well after the flash. This limitation results in the impossibility of identifying the actinides present in the package, which constitutes a first limitation of this method.

In addition, generally, the package is probed globally because once created, the photons continue their journey in a straight line, without being able to be directed to a specific spot of the package. Fine localization of the waste inside the package is not possible in this system according to the prior art.

Moreover, the spectrum of these Bremsstrahlung photons ranges from zero to the maximum energy of the incident electron (rare case where the electron stops and radiates all its energy). This spectrum is for example illustrated in Figure 2 where a 50 MeV electron strikes a 5 mm thick tungsten converter. The density of photons decreases exponentially with energy, the majority of photons being emitted at low energy. In this example, each electron emits on average about forty photons of which only 0.6 are in the "useful" energy range for the GDR, that is to say between 10 and 20 MeV.

If the emitted photons interact with an actinide nucleus present in the target to be probed, those of them having the good energy will be able to excite a GDR dipolar giant resonance and possibly to fission the nucleus.

During the fission, neutrons are emitted during the de-excitation: they are the fast neutrons. But the fragments of fission are generally themselves radioactive, and in excited states. Some emit neutrons with a delay time with respect to the fission itself (thus depending on the lifetime of said fragment): these are the delayed neutrons.

The detection of the amount of actinides is based on the measurement of the number of these delayed neutrons, after a burst of electrons has been sent by the accelerator to the target, producing a gamma flash.

The amount of delayed neutrons produced by the photofission is proportional to the mass of actinides present, according to the linear relationship:

nfs ∞ m _[g] I _[μA]

The number of delayed neutrons emitted per second is proportional to the mass of fissionable elements (by photofission) and to the average current of electrons. The coefficient of proportionality depends on the photon efficiency, the transport and the attenuation of these photons and the cross sections of the various elements considered. It should be noted that the method according to document FR 2,764,383 (equivalent US6452992) makes it possible, by measuring the number of delayed neutrons, to deduce the total mass of the actinides in a target. It is therefore a very interesting measure because fast (from a few minutes to a few hours, once the package is in place) and non-destructive. However, it does not allow the direct identification of the different actinides.

A second limitation of this method, according to the document FR 2,764,383 (equivalent US6452992), lies in the fact that it measures only the mass of actinides and not their activity. Indeed, the delayed neutrons are emitted inside the package by number n * proportional to the intensity of the electron beam and the mass m ¹ _g actinides present in the package. After emission, they undergo repeated shocks and slowdowns, then broadcast to the outside of the package and through the ambient air before reaching a detector. This process causes the number of delayed neutrons detected to be a small fraction of the neutrons emitted, mainly because of the absorption in the package. The proportion of delayed neutrons detected will be all the more important as the distance to be traveled inside the waste package will be small. It therefore strongly depends on the position of the actinide inside this package. If d denotes the distance separating the position of the actinide from the outside of the package, the number n * of neutrons detected will therefore be related to the number of pulses S per second of the detector by the approximate relation:

^S [co, _Ψ s / s) ^{∞ β} "

The coefficient of proportionality integrates the geometry of the device (solid angle) and the efficiency of the detector used (number of shots per neutron arriving on the detector). Strictly speaking, the coefficient α and the coefficient of proportionality depend on the energy of the neutron. But in practice we can admit that the spectrum in delayed neutron energy is identical to that of the classical (neutron-induced) fission of average energy around 450 keV. This eliminates the impact of the spectral distribution which may differ depending on the type of actinide present in the waste package. Therefore, there are only two unknowns left: the quantity m of radioactive material, and its location inside the package that is expressed by its distance d to a given detector. To determine these two unknowns, one uses either several detectors or several measures by for example turning the package around its central axis. In order to improve the accuracy of the measurements, both can be made together.

The direct relationship between the number of hits measured on the detector and the total mass of actinides:

^S [shot _S / _S] can then be written by considering A as a parameter dependent on prior measurements (calibration). It establishes well that the number S of pulses per second measured by the detectors is proportional to the mass m of actinides present in the package.

Each detector will therefore be calibrated to have the corresponding exact value of its coefficient of proportionality A. Since the thickness of material traversed d is unknown, it is possible by using several detectors (theoretically two situated at 180 ° on either side of the package should be sufficient), deduce the two unknowns d and m. Of course, in the very likely case where several actinides located in several places emit, one would have to sum up their different contribution.

A third limitation of this method, according to the document FR 2 764 383 (equivalent US6452992), resides in the noise which limits the sensitivity of the system to a value too high to allow to classify the package into a category. storage and determine where it should be stored.

The experiment shows that this limit of detection is about 5 g / t for 15 MeV electrons. Of course, if the actinide is not at the center of the package, or if the package is smaller in size, the limit of its detection will be even lower (exponential variation).

But this limit of 5 g / t is far too high for this technique to be used for the orientation of the outlet of the packages with a view to their eventual decommissioning of waste B to A. In fact, let us take for example a package containing nuclear waste coated with medium density concrete 1.74. The total weight of the package is 4.7 tons for dimensions of 1.5x1,5x1,2 m- Suppose also that this package contains mainly ²³⁹ Pu.

The maximum admissible activity (for the isotope considered) for the storage of a TFA package is 100 Bq / g.

The radioactivity of ²³⁹ Pu is 2.284 GBq / g. It follows that the TFA classification limit for this type of package must be such that

For this package which is about 4.7 tons, it will be necessary to detect masses of less than 200 mg. We see that it will be necessary to obtain detection sensitivities about 100 times lower than those that can be achieved by the system of FR 2,764,383 (equivalent US6452992).

In other words, the main disadvantages of the systems of the prior art and in particular of document FR 2 764383 (equivalent US6452992) are the following:

On the one hand, the neutrons inside the waste drum have a poorly controlled path, which does not allow to reach all the zones of this drum, and on the other hand the enormous The amount of neutrons (produced by the impact of the photons on the target) blinds all the detectors, to the point of making it impossible to measure fast neutrons. Only delayed neutrons can be measured effectively, and the sensitivity of measurements is greatly affected. Finally, the efficiency of the process is extremely low, a very small portion of emitted photons having sufficient energy to trigger a photofission reaction.

In addition, when the waste stream is no longer subject to neutrons but to highly energetic photons (produced by the impact of electrons on a target placed in the immediate vicinity of the waste drum), the distribution throughout the useful section is a priori uniform, which is an important advantage. However, the penetration depth of the photons is low, and the vast majority of them, insufficiently energetic, stop in the very first centimeters, which does not allow to know the composition of the drum in the whole of its thickness. Finally, if the photons produced by the impact of electrons go in a straight line (according to the direction of incidence of the electron that created them), the entire photon beam is nevertheless affected by some divergence, sufficient to blind the neutron detector. More generally, the method according to FR 2,764,383 (equivalent US6452992) has the following three disadvantages: firstly a measure of the mass of the actinides and not their activity, secondly the impossibility of identifying the actinides present finally, the impossibility of lowering the sensitivity limit sufficiently to determine the category in which the waste must be stored.

The present invention aims to improve the systems of the prior art by providing a device capable of effectively and simply detect the prompt and delayed neutrons. It is known that the ratio between the fast neutrons and the delayed neutrons varies greatly with the type of actinide considered. The present invention allows the measurement of fast neutrons in addition to the measurement of delayed neutrons, allows to determine this report is therefore to determine the type of actinides present in the package.

The present invention also aims an active method having a high sensitivity of detection, and able to focus this sensitivity on a specific area inside the package. It also aims to accurately measure any amount of actinides present in a large object.

At least one of the above objectives is achieved with a novel method for probing an object containing potentially radio-transmitter elements. This object is irradiated by means of a particle beam whose energy is sufficient to cause photofission of the radio-transmitter elements of the object. The neutrons produced by photofission are measured by means of at least one neutron detector. According to the invention, the particles that probe the object by photofission are electrons directly irradiating the object, and the conversion of the electrons into photons capable of generating the photofission is performed directly by the object to be probed. In contrast to the prior art, the electrons are used directly to strike the object to be probed, and no heavy metal target, such as tungsten, is added to the interior of this object to convert the electrons in photons able to generate the photofission of the potentially radiating elements.

The object to be probed generally consists of a peripheral container, and a content that occupies the central part. When this object is a package of nuclear material, the container is generally of absorbent concrete, thick, and the content comprises nuclear material. In the case of such a package having a large thickness of absorbent concrete, surrounding nuclear material, the invention is particularly advantageous since it gives rise to brake radiation mainly at the concrete container. However, the constituents of the concrete have a low atomic number, on average around 20 (the heaviest component, the barium of atomic number 56 being in a very small amount), so that the photofission reaction does not create substantially direct neutrons to blind the detectors. More generally, the measurement is favored by arranging around the object to be probed or in its container a material having a sufficiently high atomic number so that the Bremsstrahlung reactions are sufficient in number for the photofission, and sufficiently low to limit the production of neutrons. As an indication, this range can be located, without limitation, between 10 and 50.

Moreover, it is known that at the considered energies, the electron-photon conversion takes place on a high atomic number material, such as tungsten, in a few mm.

When it is carried out in concrete, this conversion takes place in several dm, so that if the thickness of the concrete is less than this distance, there remain unconverted electrons in photons which directly impact the waste contained in the concrete. 'object.

In this case, electrons directly encountering a piece of potentially fissile material, such as uranium, convert into photons in a few mm, then immediately afterwards can give rise to a photonuclear reaction. Thus a sufficiently small thickness of the container makes it possible to benefit from the greatest ease of directing electrons rather than photons. In summary, the invention makes it possible to make full use of the electron directivity to explore the contents of the object by choosing a number material container. the

atomic enough, and giving it a sufficiently small thickness for a sufficient part of the electrons to pass through before being converted into photons. By sufficient part is meant an amount sufficient so that the resulting photofission reactions can be measured with a signal-to-noise ratio compatible with the needs.

The above features combine to locate the photofission reactions as close as possible to the nuclear materials of the package, and, if desired, to distribute them relatively homogeneously over the entire volume of the nuclear material to be probed.

The process according to the invention has many advantages.

A first advantage is that the electrons can be manipulated more easily than photons: we can vary the size of the electron beam so as to vary the accuracy of the irradiation, we can focus the electron beam on a part of the nuclear material, we can transport the electrons, deflect them, carry out a well targeted or delimited scan, etc ... We can thus make sure that the electron beam sweeps all or part of the nuclear material, or even the totality of the object to be probed, which may be a cask for example. The electron beam is then moving relative to the object. In the same way, the object can be moved in translation and / or rotation relative to the electron beam so that the electron beam radiates all or part of the contents of the package.

Preferably, the electron beam that irradiates the package is pulsed, and the at least one neutron detector detects the delayed neutrons, generated by the photofission, after the resulting pulsed flash of the electron beam. Said at least one neutron detector can also detect the prompt neutrons emitted during the electron beam pulse.

Advantageously, insofar as the electrons very rapidly lose their energy in the air and disperse easily, the nuclear material can be irradiated by transporting the electron beam via a vacuum line.

According to one embodiment of the invention, by scanning the object by means of the electron beam, particles emitted by the scanned parts are simultaneously detected so as to locate the nuclear material within the object.

The second advantage is that the electrons have a penetrating power in the material. As a result, they can get as close as possible to the actinides present in the package, and thus increase the chances of a nuclear reaction. This results in a considerable improvement in energy efficiency and greater measurement sensitivity. The third advantage is that the direct use of electrons makes it possible to minimize the noise induced by gammas during the first impact (Bremsstrahlung gamma flash) which blinds the neutron detectors. Thus, when the electron beam irradiates the nuclear material in the form of at least one pulse, it is possible to detect and measure fast neutrons emitted during the pulse. It is also possible to detect and measure the delayed neutrons emitted after the pulse. The present invention thus allows both the measurement of the fast neutron flux and the measurement of the delayed neutron flux. The second measurement makes it possible to determine the mass of actinides as taught in the documents FR 2 764383 (equivalent US6452992) or US4497768, but with a much better resolution (absence of noise, better distribution of the photon-probe). Advantageously, the ratio of these two measurements makes it possible to further determine the isotopic composition of actinides present in the material (elements radio transmitters). Indeed, as already seen, the various actinides have a fast neutron report on different delayed neutrons. The experimental measurement of this ratio will therefore be a characteristic of the isotope considered. The total amount of actinides present in the radio - transmitter elements from the fast neutrons and the measured delayed neutrons can also be determined.

According to another aspect of the invention, there is provided a device for probing an object containing radio-transmitter elements. This device comprises means for irradiating said object by means of a particle beam whose energy is sufficient to cause a photofission of the radio-transmitter elements of the object, and at least one neutron detector for measuring the neutrons produced by photofission. According to the invention, said particles are electrons directly irradiating the nuclear material, the conversion of the electrons into photons able to generate the photofission being carried out directly by the object to be probed.

Preferably, the electron beam is transported via a vacuum line whose exit window is brought into contact with the object. Advantageously, the means for detecting fast neutrons are arranged behind the object when the electron beam arrives from the front of this object.

Other advantages and characteristics of the invention will appear on examining the detailed description of an embodiment which is in no way limitative, and the appended drawings, in which: FIG. 1 is a simplified schematic view of a device characterizing nuclear waste according to the prior art; FIG. 2 is a graph illustrating the photon spectrum emitted by Bremsstrahlung for a 50 MeV energy electron on a 5 mm thick tungsten target, according to the prior art; and FIG. 3 is a simplified schematic view of a nuclear material characterization device according to the present invention.

Although the invention is not limited thereto, an embodiment will now be described allowing the characterization of a small waste within a large package.

In Figure 1, there is a large object to be probed 6 containing a very small amount of nuclear material 7, an actinide. This material 7 is located at a precise point within the object 6. There is an electron accelerator 8 emitting, in pulsed form, an electron beam 13 intended to strike the nuclear material 7. This beam 13 is transported by a vacuum line 9 which is a vacuum tube connecting the output of the accelerator 8 to the object to be probed 6. However, in order to be able to scan the electron beam 13, a scanning device 10 is installed. magnetic deflection elements (techniques well known in the field of accelerators) which is controlled by a processing unit (not shown) for a fine localization of the nuclear material 7 in the object 6 or to test a very large surface, possibly the entire object in one go. An exit window is also provided at the end of the vacuum line 9. This window may come into contact, for example with the scanning device 10, with the object 6 so as to avoid the passage of the electron beam through the air.

The present invention makes it possible to measure the prompt neutrons emitted during the electron pulse, while continuing to measure the delayed neutrons coming from the object to be probed, thus from a few ms to 100 ms, or even more after the pulse. For delayed neutrons, the devices are identical to those described in documents FR 2 764383 (equivalent US6452992) or US4497768.

To do this, detectors 11 of delayed neutrons are arranged around the object 6. On the other hand, for neutrons Promptly, the use of electrons as probe beam makes it possible to overcome the background noise of the "gamma flash". Indeed, the electrons striking only a small part of the surface of the object, will end (as well as all the various particles produced by the successive reactions in the object) either by being backscattered (towards the back), or to be absorbed into the object itself. It is thus possible to measure the prompt neutrons emitted during pulsing with fast neutron detectors 12. It is shown that the optimal position of these detectors from the point of view of the signal-to-noise ratio is to place them behind the object to be probed. The detected fast neutron signal (much more intense than that of the delayed neutrons mentioned above) is used to deduce the amount of nuclear material present in the package. The delayed neutron signal is also measured between two pulses. Moreover, it is the conjunction of two simultaneous information (delayed signal and delayed signal) that allows, by performing their ratio, to go back to the isotope of the actinide tested. A movement of the object (vertical and / or in rotation around the vertical axis) makes it possible to precisely locate the place where the material 7 is in the object 6. The precision is determined by the size of the beam of 13. However, and precisely thanks to the direct electron-probe method according to the invention, the size of the electron beam can be as small as desired (in the limit of the emittance of the beam), simply placing a proper focus (not shown) in the vacuum transport line 9. Another feature of the present invention is the pulse width of the probe beam. While with photons or neutrons, the pulses used are generally very intense (up to 200 mA) and very short (generally less than 4 μs), it is shown that in the present case, long pulses for example 4 ms, which is a thousand times higher, could as well be used with intensities that can be much lower, some microamperes are enough.

Of course, the invention is not limited to the examples which have just been described and many arrangements can be made to these examples without departing from the scope of the invention.

Claims

the CLAIMS

A method for probing an object (6) containing potentially radiating elements, wherein said object is irradiated by means of a beam (13) of particles whose energy is sufficient to cause photofission of the radio-transmitter elements of the object , and the neutrons produced by photofission are measured by means of at least one neutron detector, said method being characterized in that said particles (13) are electrons directly irradiating the object, and in that the conversion of the electrons into photons capable of generating the photofission is performed directly by the object (6) to be probed.

2. Method according to claim 1, characterized in that the object to be probed is a package of radioactive waste containing nuclear material.

3. Method according to claim 1 or 2, characterized in that the object to be probed comprises a container made of concrete, and a content comprising the nuclear material.

4. Method according to any one of claims 1 to 3, characterized in that the electron beam which irradiates the package is pulsed, and that the at least one neutron detector detects the delayed neutrons, generated by the photofission, after the resultant pulse flash of the electron beam.

5. Method according to claim 4, characterized in that the at least one neutron detector also detects the fast neutrons emitted during the electron beam pulse.

6. Process according to claim 5, characterized in that the isotopic composition of actinides is determined. present in the radio-transmitter elements from the ratio of the fast neutrons to the measured delayed neutrons.

7. Process according to claims 4 and 5 or 6, characterized in that the total amount of actinides present in the radio-transmitter elements from the fast neutrons and the measured delayed neutrons is determined.

8. Method according to any one of the preceding claims, characterized in that irradiating the object to probe by transporting the electron beam to the object by a vacuum line (9).

9. Method according to any one of the preceding claims, characterized in that the electron beam scans all or part of the object to be probed.

10. Method according to any one of the preceding claims, characterized in that the object (6) to probe in motion relative to the electron beam so that the electron beam radiates all or part of the object to be probed.

11. Method according to any one of the preceding claims, characterized in that the size of the electron beam is varied so as to vary the accuracy of the irradiation.

12. Method according to any one of claims 3 to 11, characterized in that by scanning the object by means of the electron beam, particles emitted by the scanned parts are simultaneously detected so as to locate the radio-transmitter elements at the same time. within the object.

13. Device for probing an object (6) containing radio-transmitter elements (7), this device comprising means for irradiating said object by means of a beam (13) of particles whose energy is sufficient to cause photofission of the radio-transmitter elements of the object, and at least one neutron detector for measuring the neutrons produced by photofission, characterized in that said particles (13) are electrons directly irradiating the nuclear material, the conversion of the electrons into photons capable of generating the photofission being carried out directly by the object (6) to be probed.

14. Device according to claim 13, characterized in that it comprises means for emitting the electron beam in the form of at least one pulse, and means (12) for detecting and measuring fast neutrons emitted during the impulse.

15. Device according to claim 14, characterized in that the detection means (12) are arranged behind the object when the electron beam arrives from the front of the object.

16. Device according to any one of claims 13 to 15, characterized in that it comprises means (11) for detecting and measuring delayed neutrons.

17. Device according to claims 14 and 16, characterized in that it comprises a processing unit for determining the isotopic composition and the total amount of actinides present in the radio-transmitter elements from the prompt neutrons and measured delayed neutrons.

18. Device according to any one of claims 13 to 17, characterized in that it comprises a vacuum line (9) for transporting the electron beam.

19. The method of claim 18, characterized in that the end of the vacuum line comprises an exit window which is brought into contact with the object.

20. Device according to any one of claims 13 to 19, characterized in that it comprises scanning means (10) adapted to direct the beam of ^one electron.

21. Device according to any one of claims 13 to 20, characterized in that it comprises means for moving the object relative to the electron beam.

22. Device according to any one of claims 13 to 21, characterized in that it comprises focusing means for focusing the electron beam on a portion of the nuclear material.