Patent Application
20090074128
This disclosure relates to systems and methods for detecting the presence of fissionable nuclear materials. The systems and methods make use of the distinctive signals provided by the energy and angular distributions of the prompt neutrons produced in photon induced fission of nuclei. They may be used to detect the presence of actinide nuclei (in particular those with Z greater than or equal to 89, that of actinium). Some of these nuclei are classified as Special Nuclear Materials (SNM) and may be used in weapons of mass destruction such as nuclear explosives and in dirty bombs.
Illicit clandestine shipment of nuclear explosives, materials that can be employed in the fabrication of nuclear explosives, and materials that can be employed in the fabrication of dirty bombs may constitute a major threat to the peace and security of the world. Such materials may be secreted and smuggled in cargo or other shipments in various containers including ordinary luggage, crates, vehicles, cargo containers, etc. by terrorists, potential terrorists, terrorist supporters, or others. Effective and efficient methods and systems are required for the reliable, non-intrusive, detection of such contraband materials in ports and in other cargo and shipping locations in order to reduce the risk of successful illicit shipments, without unduly impeding the worldwide flow of cargo in a manner that is disruptive of normal commerce. Accordingly, it is especially important that the detection methods not produce large numbers of false positive detection events.
Passive detection methods, as for example gamma spectroscopy of natural decay, have not proven universally effective since many of the materials of interest are not highly radioactive and are relatively easily shielded. X-ray techniques do not readily distinguish between fissionable nuclear materials and innocuous high-Z materials like lead or tungsten that may be legitimately present in cargo.
In addition to passive detection, several approaches to detection have been employed, attempted, or proposed using active techniques employing probing beams.
In one such active technique, an external neutron source has been used to detect fissionable nuclear materials by detecting induced fission events by the neutron multiplication effect of the fission events. However, it has been difficult to discriminate between the probing neutrons and the fission induced prompt neutrons, especially when the energy of the probing neutrons is as high as the energy of the more energetic prompt neutrons from fission or when large containers are involved. Alternative techniques have induced fission events in fissionable nuclear materials with pulsed external neutron sources, then detecting delayed emission of neutrons by fission products, using time delay, as a means of distinguishing the detected signal from the probing neutrons. This delayed neutron signal is a much weaker signal, and is subject to signal-to-noise ratio problems.
In other active techniques, gamma ray probe beams have been employed to induce photofission (γ, f) of nuclear materials with detection of neutrons resulting from the fission events. Scattered gamma rays from the probe beam as well as photo-neutrons (direct (γ, n) events resulting from interaction of the gamma probe beam with fissionable and/or non-fissionable nuclei) induced by the probe beam contribute noise to the detection of prompt neutrons from the fission events, contributing to unreliable or ambiguous detection. Photofission also results in delayed neutron production by the fission fragments, but as with neutron-induced fission, the delayed neutron signal is weaker and detection suffers from noise problems.
It is therefore an object of this disclosure to provide improved systems and methods for detecting fissionable nuclear material in an article with reduced error and ambiguity.
It is a further object of this disclosure to provide improved systems and methods for detecting contraband fissionable nuclear materials by improving discrimination of prompt fission neutrons in the presence of noise-contributing factors.
Another object of this disclosure is to provide systems and methods for analyzing the energy or an energy spectrum of prompt fission neutrons to detect the presence of fissionable nuclear materials in an article.
A still further object of this disclosure is to provide systems and methods for detecting an angular distribution of prompt fission neutrons to detect the presence of fissionable nuclear materials in an article.
Yet another object of this disclosure is to provide systems and methods for using an angular distribution of prompt fission neutrons and an energy distribution of prompt fission neutrons to detect the presence of fissionable nuclear materials in an article.
The objects set forth above as well as further and other objects and advantages of the present disclosure are achieved by the embodiments described below.
A prompt neutron is a neutron emitted immediately after the fission process; it is characterized by being emitted from a fission fragment generally after the fragment has reached a significant fraction of its final velocity, and thus may be referred to as a fully accelerated fragment. The final velocity is imparted to the fragment by the strong Coulomb repulsion between the fission fragments. Some neutrons arise from photon induced fission at the point of scission (just as the fragments break apart) but these have been shown to be small in number compared to those emitted by the fragments in flight. There are also delayed neutrons that arise following the beta-decay of some of the fragments, but these are not considered herein since they are only a small percentage of the neutrons emitted promptly and thus have a negligible effect on the practice of the methods disclosed herein. One of the advantages of utilizing prompt neutrons from photo-fission as a detection technique is that they are produced with approximately 200 times the yield of delayed neutrons; this allows for higher probabilities of detection, lower false positive rates, and faster scan times.
The techniques and methods described herein make use of the boost in velocity (and thus energy) of a neutron that arises because the neutron is emitted from a rapidly moving nuclear fragment which has been produced by the (, f) process. This boost places the neutron in an energy range that will allow for the unambiguous determination of the presence of fissionable nuclei; this energy range is not possible from other processes that could occur with other non-fissionable nuclei such as direct neutron production by photons (γ, n). Additional features of interest are the nucleus-dependent angular distribution of the fragments in the photo-fission process and the prompt neutron energy distributions at various angles. Thus the signature of photon-induced fission is unique. Also, by controlling the incident photon energy used to cause the fission, (γ, n) processes from other nuclei may be reduced in importance or eliminated as a background. Since the process of photon-induced fission is ubiquitous with the actinides, these methods will identify fissionable nuclear materials within a container, in particular those which have Z equal to or greater than 89, that of actinium.
This disclosure describes systems and methods for detecting fissile materials by measuring prompt neutron energies and examining prompt neutron energy spectra. The energy spectra of prompt neutrons that originate from photo-fission are readily distinguishable from the energy spectra of neutrons that originate from other processes that may occur in non-fissile materials such as (γ, n). Neutrons at energies greater than E=Eb−Eth, where Eth is the threshold for the (γ, n) process in relevant other heavy non-fissile elements and Eb is the endpoint energy of the incident bremsstrahlung photon beam (or the energy of an incident monochromatic photon beam), indicate with no ambiguity the presence of fissile material in the actinide region. No other photon-induced process can generate neutrons with these energies.
Angular distributions of these neutrons reflect the angular distributions of the fission fragments from which they arise: distributions deviating significantly from isotropy indicate the presence of even-even nuclei while almost isotropic distributions indicate the presence of odd-even or even-odd fissile species. (Hereinafter, in the interests of conciseness, “odd-even” shall denote a nucleus with an odd number of nucleons, whether protons or neutrons, and thus the term hereinafter shall encompass both “odd-even” nuclei and “even-odd” nuclei).
Comparison of the energy distribution of the prompt neutrons at different angles also provides potentially useful information about the species present. If the energy distributions at different angles are nearly identical, the isotopes undergoing fission are odd-even; if the energy distributions differ significantly at different angles, the isotopes undergoing photo-fission are even-even.
Another signature of photo-fission is the fact that the relative yield of prompt neutrons at different neutron energies (i.e., the shape of the yield curve) does not depend upon the incident photon energy. This is in contrast to other processes such as (γ, n) where the relative yield of neutrons at different energies is strongly dependent on incident photon energy, particularly at the highest energies possible.
For a better understanding of the present disclosure, together with other and further objects thereof, reference is made to the accompanying drawings and detailed description.
Fission is a complex process that has been the subject of many theoretical and experimental studies. (See generally Bohr and Mottelson, “Nuclear Structure”, 1998, World Scientific Publishing Co. Pte. Ltd. Singapore, and references therein). However, common empirically established features imply certain general regularities of the process independent of nucleus or initiating particle.
When fission is spontaneous, initiated by low energy neutrons or by the absorption of photons near the threshold for the (γ, f) process, the dominant mode of fission is the breaking apart of the nucleus into two fragments of unequal masses. These unequal masses are in the regions of nucleon numbers 95 and 140 for 235U and in similar regions for other fissionable nuclei. The fragments are accelerated by the strong Coulomb repulsion of their charges (Z1, Z2) and gain kinetic energy ranging approximately from 160 to 180 MeV, depending on the nucleus undergoing fission. Most of this Coulomb energy is gained in approximately 10−22 sec as the fragments separate by several nuclear diameters. The final fragment velocities correspond to kinetic energies of approximately 1 MeV/nucleon for the light fragment and approximately 0.5 MeV/nucleon for the heavy fragment. The rapidly moving fragments are generally excited and emit prompt neutrons, mostly after they have gained most of the kinetic energy available from the Coulomb repulsion.
Similar results have been obtained by Terrell for neutron-induced fission of 233U and the spontaneous fission of 252Cf, showing the generality of the phenomena.
Many authors have studied the spontaneous fission of 252Cf, including Harry R. Bowman, Stanley G. Thompson, J. C. D. Milton and J. Swiatecki: “Velocity and Angular Distributions of Prompt Neutrons from Spontaneous Fission of 252Cf”, Phys. Rev., Volume 126, Number 6, Jun. 15, 1962, page 2120-2136 and references therein. These authors were able to demonstrate by direct measurement that:
a) “The angular distribution (of the neutrons from the spontaneous fission of 252Cf) is strongly peaked in the direction of the fission fragments. The relative intensities in the direction of the light fragment, in the direction of the heavy fragment and at right angles are about 9, 5 and 1 respectively”: and
b) “The broad features of the energy and angular distributions are reproduced by the assumption of isotropic evaporation (in the fragment frame of reference) of the neutrons from fully accelerated fragments.”
While not the only important conclusions of the Terrell and Bowman works, those quoted and discussed here sustain the general description of spontaneous fission or fission at low energies that is important to the discussion herein.
The work of H. W. Schmitt, J. H. Neiler, and F. J. Walter, “Fragment Energy Correlation Measurements for 252Cf Spontaneous Fission and 235U Thermal-Neutron Fission”, Phys. Rev. Volume 141, Number 3, January 1966, Page 1146-1160, provides additional evidence of the features described above. They find that the average total fragment kinetic energies before neutron emission are 186.5±1.2 MeV for the spontaneous fission of 252Ca and 171.9±1.4 MeV for neutron induced fission of 235U. The fragments have substantially all the kinetic energy available from the mutual Coulomb repulsion of the fragments.
Both the energy distribution and the angular distribution of the neutrons from fission fragments created by photon-induced fission are relevant. The case of 232Th reported in C. P. Sargent, W. Bertozzi, P. T. Demos, J. L. Matthews and W. Turchinetz, “Prompt Neutrons from Thorium Photofission”, Physical Review, Volume 137, Number 1B, Jan. 11, 1965, Pages B89-B101 is illustrative. These authors measured the spectra of neutrons from the photo-fission of 232Th at pairs of angles simultaneously, 157 and 77 degrees relative to the photon beam, and 130 and 50 degrees relative to the photon beam. They used bremsstrahlung photons from electrons with kinetic energies of 6.75 and 7.75 MeV. Several subsidiary facts were important in their analysis:
1.) The (γ, n) threshold energy for 232Th is 6.438 MeV. Therefore, the (γ, n) process cannot contribute neutrons of energy greater than 0.31 MeV and 1.31 MeV, respectively at the two energies of the electron beam, 6.75 MeV and 7.75 MeV. Since these neutron energies are achieved only at the end points of the respective bremsstrahlung spectra, there will not be important contributions to the neutron spectra from the (γ, n) process even at neutron energies considerably lower than 0.31 or 1.31 MeV, respectively; and
2.) The fission fragments in photo-fission, (γ, f), are known to have strongly anisotropic angular distributions from 232Th. The distribution is peaked at 90 degrees to the incident photon beam, and the fragment angular distribution is given by I=a+b sin2(θ), where θ is the angle between the incident photon beam direction and the fission fragment direction. The ratio b/a is considerably larger than 1 at the energies discussed herein and remains larger than one even at incident photon energies higher than 9 MeV. (E. J. Winhold, P. T. Demos and I. Halpern, Physical Review, 87, 1139 (1952): and, A. P. Berg, R. M. Bartholomew, F. Brown, L. Katz and S. B. Kowalski, Canadian Journal of Physics, 37, 1418 (1959)). This fragment directionality provides the correlation between neutron angle and neutron energy that results from the velocity boost if the prompt neutrons are emitted from fragments that have their full kinetic energy.
The results of analysis of the neutron energy spectra from 232Th (γ, f) are consistent with the following conclusions of Sargent et al:
1.) The fraction of the prompt neutrons that result from emission from other than the fully accelerated fragments is 0.07±0.09;
2.) The prompt neutron angular distributions and energy distributions are consistent with isotropic neutron evaporation with a thermal-type spectrum in the center of mass frame of reference of the moving fragments, where the fragments are moving with their fully accelerated velocities; and
3.) The energy spectrum of the neutrons in the center of mass frame of reference is characterized by an average energy of 1.14±0.06 MeV. There are no significant components of temperature as high as or higher than this average energy. (That is, the ensuing Maxwellian energy distribution, were it applied to a fragment at rest in the laboratory frame of reference without the kinematic boost from the motion of the photo-fission fragments, would not yield many neutrons at the high energies that result from applying the kinematic boost to neutrons emitted in the fragment frame of reference).
One outstanding feature of the neutron spectrum in
Energy conservation in the direct (γ, n) neutron production process does not allow the production of neutrons with an energy above E=Eb−Eth, where Eb is the bremsstrahlung endpoint energy of the incident photon beam and Eth is the (γ, n) threshold energy for producing neutrons from other relevant heavy elements. Therefore, detecting neutrons with energies above this value is definitive evidence of the presence of fission.
Since the (γ,n) threshold of 232Th is 6.438 MeV, a neutron energy of 6 MeV will not be possible from (γ, n) until the bremsstrahlung endpoint reaches 12.438 MeV. Also, even when the bremsstrahlung endpoint reaches that value, neutrons from the (γ, n) process will be very small in number because they can only be produced by the few photons at the bremsstrahlung endpoint energy.
These energetic considerations apply in a similar manner for all fissionable nuclear materials, in particular for those with Z≧89, the region of the actinides. In addition, and most importantly, most heavy elements such as Bi, Pb, W, Ta, etc. have isotopes with (γ, n) thresholds at or above 6.5 MeV. Therefore, finding neutrons with energies above E=Eb−Eth where Eth is in the range of 6 MeV constitutes a very definitive test for the presence of fissile material.
Another test to verify that the detected neutrons result from photo-fission is the sensitivity of the yield of neutrons at energies above E=Eb−Eth to a modest increase in incident photon energy. In particular, measuring the increase in yield relative to the yield of neutrons below this energy is significant. The increase or relative increase in neutron yield is not substantial when the neutrons are emitted from photo-fission fission fragments because energetic considerations independent of the exact incident photon energy, such as the boost in velocity from fission fragment motion, are most important in determining the yield.
The spectra in
Table 1 gives the (γ, f) and the (γ, n) thresholds (in MeV) for some typical nuclei in the actinide region. The (γ, f) thresholds are from H. W. Koch, “Experimental Photo-Fission Thresholds in 235U, 238U, 233U, 239Pu and 232Th”, Physical Review, 77, 329-336 (1950). The (γ, n) threshold of 207Pb is also listed, as it is a component in natural lead material that may be used as a shield against detection of fissile materials. The table shows the maximum neutron energy available from the (γ, n) process for bremsstrahlung end point energies up to 11 MeV, including for 207Pb. This energy is to be compared to the spectrum in
232Th
233U
235U
238U
239Pu
207Pb
The data in Table 1 indicates how the yield of neutrons above a specified energy would change as the bremsstrahlung endpoint energy is changed. For 207Pb, Table 1 indicates, there would be no neutron yield above 4 MeV until the electron beam energy exceeded approximately 11 MeV. (For gold, as discussed above in connection with
An additional point, which will be discussed further below, is that the photo-fission cross section is larger than the (γ, n) cross section over most photon energies by a considerable amount, as shown in
The data in Table 1 is based upon continuous bremsstrahlung spectra with specific endpoint energies, but a similar discussion applies to monochromatic photon beams. The neutron energy spectra from photo-fission retains the same dependence on neutron energy for different photon energies, but the total yield is modulated for monochromatic photons only by the cross section for (γ, f) at the specific photon energy. In contrast, the total yield for neutron production from a bremsstrahlung beam is modulated by the convolution of the bremsstrahlung spectrum with the (γ, f) cross section. The maximum neutron energy from (γ, n) dictated by energy conservation considerations for monochromatic incident photons follows just as discussed above.
Other energies than 4 MeV could be used as the “trigger” or cutoff for defining the presence of fissionable nuclear material. That is, for any specific electron beam energy, a “trigger” energy can be selected such that the presence of neutrons with an energy above that “trigger” energy will be energetically impossible for the (γ, n) process in relevant heavy materials such as 207Pb and therefore any neutrons detected could only originate from the photo-fission process in an actinide. The data in
In addition, the choice may be dictated by the specific nature of the cargo in a container; if the cargo is made of materials with high (γ, n) thresholds, such as copper, aluminum, steel or oxygen, then a lower trigger could be selected.
Conversely, hydrogenous material that naturally contains a small percentage of deuterium may be of concern because of its low threshold for the (γ, n) process, 2.2 MeV. However, because the energy release is shared almost equally by the neutron and proton, the maximum neutron energy is given by E=(Eb−2.2)/2 MeV and, for the example of an electron beam energy of 10 MeV, the maximum neutron energy is approximately 3.9 MeV and a 9.2 Mev photon results in a neutron energy of 3.5 MeV. Thus, a higher trigger may be appropriate
A more important concern may be 9Be. It has a low (γ, n) threshold of only approximately 1.6 MeV and the energy sharing results in a neutron that has most of the available energy, E=(8/9)(Eb−1.6) MeV is the maximum neutron energy available. For the example of Eb=10 MeV, the maximum neutron energy is approximately 7.5 MeV. This high energy could present a serious background. However, one could distinguish neutrons from actinide photo-fission from neutrons from the (γ, n) process in 9Be by taking advantage of the fact that the (γ, n) process follows the strict rule for conservation of energy, so that E=(8/9)(Eb−1.6) defines the maximum neutron energy possible, while the photo-fission process has a neutron energy spectrum largely independent of the photon energy in the energy region under discussion, Eb less than approximately 15 MeV. Therefore, neutrons at an energy greater than E=(8/9)(Eb−1.6), where Eb is the photon beam energy or bremsstrahlung endpoint energy, is proof of a fissile material. At Eb=10 MeV, the presence of neutrons above approximately 7.5 MeV would be proof. At Eb=8 MeV, neutrons above 5.7 MeV would be proof. Also, the prompt neutron energy spectrum is independent of the photon energy while the (γ, n) process in 9Be produces a neutron spectrum that is strongly dependent on photon energy. This difference also permits distinguishing the presence of a fissionable element from the presence of 9Be.
However, if there were concern that this measurement could not be reliably made, further steps could be taken. Operating at Eb=10 MeV, the maximum neutron energy from beryllium (γ, n) is approximately 7.5 MeV. By reducing the beam energy to 8 MeV, for example, the maximum energy neutron from beryllium (γ, n) would be reduced to 5.6 MeV but the photo-fission neutron energy distribution would be unchanged. If there are neutrons above 5.6 MeV the process is unquestionably photon induced fission. If there remains any doubt that neutrons are from fission, the photon beam energy can be further reduced. For example at 5 MeV photon or bremsstrahlung beam energy there will be little or no photo-fission. But beryllium (γ, n) will produce neutrons of up to approximately 3 MeV at that photon beam energy. The presence of these neutrons will clearly establish the presence of beryllium. From the yield of these neutrons, the contributions from beryllium to higher neutron energies when higher photon energies are used can be calculated, the neutron energy distribution from beryllium removed, and the remaining spectrum analyzed for the presence of actinide neutrons.
Fortunately, 9Be is almost unique in this category. There are a few other nuclei with relatively low (γ, n) thresholds; 6Li, 13C, 17O and 149Sm are notable among these with thresholds of 5.66, 4.95, 4.14 and 5.87 MeV, respectively. The same procedures outlined above can be used to eliminate these sources as contributors masking fissionable nuclei.
FIGS. 4A and 4B, from H. W. Koch, “Experimental Photo-Fission Thresholds in 235U, 238U, 233U, 239Pu and 232Th”, Physical Review, 77, 329-336 (1950),
As a result, the shape of the yield curve of prompt neutrons of a given energy as a function of bremsstrahlung energy will be essentially independent of the neutron energy. That is, the yield curve for 6 MeV neutrons will have the same dependence on bremsstrahlung endpoint energy as the yield curve for 2 MeV, 3 MeV, 4 MeV and etc. neutrons. This is in contrast with the yield curves for neutrons from the (γ, n) process, which will start at the endpoint energy given by Eb=Eth+En, where En is the neutron energy that is desired. They are thus displaced from the (γ, n) threshold energy, Eth, by the neutron energy, in contrast to the yield curves for (γ, f). This is a powerful signature that the neutrons detected are from photo-fission rather than from (γ, n).
The photo-fission cross section (
The angular distribution of the prompt neutrons and the relationship of the neutron energy to the fragment angular distribution also are signatures of fissile material and the photofission process, and can be used in detection schemes.
The fragment angular distributions are not as distinct for odd-even nuclei as for even-even nuclei, in part because of the high population of spin states. Odd-even nuclei angular distributions are almost isotropic as reported by L. P. Geraldo, “Angular Distribution of the Photofission Fragments of 237Np at Threshold Energy”, Journal of Physics G: Nuclear Physics, 12 1423-1431 (1986), which shows angular anisotropy of approximately 10% at 5.6 MeV, 6% at 6.61 MeV and ˜2% at 8.61 MeV. These results are very much in contrast with the large anisotropy for fragments from the photo-fission of even-even nuclei where ground state spins are zero. Thus, once actinide photo-fission is detected, a nearly isotropic neutron angular distribution is an indicator of an odd-even fissile species such as 235U, 237Np and 239Pu. A strongly anisotropic neutron angular distribution would indicate an even-even fissile species such as 232Th and 238U. (See S. Nair, D. B. Gayther, B. H. Patrick and E. M. Bowey, Journal of Physics, G: Nuclear Physics, Vol 3, No. 7, 1977 (pp 1965-1978) and references therein, for example).
The energy distributions of the neutrons at various angles are themselves indicators of the fragment anisotropy, and thus of the type of nucleus. This fact was used in the analysis of the work by Sargent et al, discussed above. If the fragments are strongly anisotropic (even-even fissile species), then the energy spectra of the neutrons will show distinct differences at different directions with respect to the photon beam. As an example, if the fragments are strongly peaked at 90 degrees with respect to the photon beam, then the neutron spectrum at 90 degrees will exhibit to a different degree the boost in velocity due to the velocity of the fragments than the neutron spectrum at angles near 180 degrees or 0 degrees to the photon beam. However, if the fragment angular distribution is nearly isotropic (odd-even fissile species), then the energy distribution of the neutrons will be the same at all angles. In both situations, the higher energies reflect the motion of the fragments, but the contrast in the energy distribution of the neutrons at different angles will reflect the fragment anisotropy with angle.
The fragment angular distributions dominate the neutron angular distributions and the neutron energy distributions as a function of angle. The results of E. J. Winhold, P. T. Demos and I. Halpern, Physical Review, 87, 1139 (1952); E. J. Winhold and I. Halpern, Physical Review, 103, 990-1000 (1956); and, A. P. Berg, R. M. Bartholomew, F. Brown, L. Katz and S. B. Kowalski, Canadian Journal of Physics, 37, 1418 (1959) show the fragment angular distributions for various isotopes. The following abstract from Berg et al. is offered as a summary of the data in that paper:
The result quoted in the abstract is in basic agreement with that of the other papers referred to herein. In addition, some greater detail about the results from Berg et al. is shown in the two tables taken from that reference:
Table 2 (“Angular Distributions . . . ”) shows that the ratio of events at 90 degrees to those at 0 degrees for the photo-fission of the odd-even isotopes shown is approximately equal to 1 over the energy range of the bremsstrahlung endpoints shown in the table. Thus, the value of b/a discussed earlier is 0 and the angular distribution is isotropic. Table 3 (“Corrected values . . . ”) shows the fit to the normalized form of the angular distribution as exhibited in the table also as a function of bremsstrahlung endpoint. The derived angular distributions are clearly anisotropic. From these data, the quoted abstract, and the theoretical basis referred to in the references herein, the generalization is accurate; the odd-even actinides undergo isotropic photo-fission while the even-even actinides undergo anisotropic photo-fission. In particular, the result is experimentally demonstrated for the isotopes most likely to be used for a nuclear weapon, 235U, 239Pu and 237Np. These will undergo isotropic photo-fission, in contrast to 238U, 232Th and the other even-even isotopes that were measured.
One embodiment of a detector system to carry out the methods described herein requires a source of photons with energy capable of exceeding the (γ, f) threshold and a detector for neutrons. The photons may be monochromatic, may be produced by a source capable of variable energy, or may be distributed over a broad range of energy with a good definition of the highest energy possible, such as an electron-generated bremsstrahlung spectrum in accordance with the discussion above. When an accelerator is used to provide the electrons, the electron accelerator may have the capability to vary the energy of the electron beam from below the fission barrier (threshold) to higher energies in order to exploit all the modalities discussed above.
Any neutron detector that is capable of distinguishing neutron energy is appropriate. A detector that takes advantage of energy deposition, such as proton recoil from neutron elastic scattering in a hydrogenous scintillator, is a possible choice. A detector that measures a reaction energy induced by the neutron is another possible choice. A method of measuring neutron energy by time of flight is also an appropriate detection scheme. The energy resolution required for such detection methods will have to be sufficient to eliminate neutrons from the (γ, n) process in materials other than actinides, as discussed above.
Because the contamination of non-actinide (γ, n) can be controlled and rendered small by the choice of incident photon energy (or bremsstrahlung endpoint) and neutron energy measured, the resolution required is well within a number of measurement techniques. Specific resolutions required may depend in detail on the particular situation under consideration, but resolutions of approximately 0.5-0.75 MeV at 4 to 6 MeV neutron energy may be adequate.
A detection method may be required to operate in a possible flux of photons in some embodiments, these photons being produced by scattering from the material under study in the direction of the detectors. Photons may also be produced by natural radioactivity and cosmic rays. Therefore, the neutron detectors may have to be shielded using passive and active shielding techniques.
In addition, as a consequence of the above, a neutron detector may be required to distinguish between photons and neutrons. This can be accommodated by the reaction process used, the time of flight of the photons compared to neutrons and by the ability of the detector to distinguish between the deposition of energy by heavy particles (e.g., neutrons) compared to electrons. Organic and inorganic scintillators that have different decay times according to the density of ionization produced by the passage of a charged particle may be suitable. Separation of photons from neutrons may be achieved in such scintillators utilizing signal processing techniques that exploit these different charged particle responses.
One exemplary embodiment of a system 600 for detecting fissile materials in a container by analyzing energetic prompt neutrons resulting from photon-induced fission is illustrated in
As an alternative to determining neutron energy directly in the neutron detector, a low duty cycle LINAC (e.g. Varian linatron) or other suitable electron accelerator may be pulsed to permit a time of flight (TOF) technique. Compared to other detection techniques, such as pulse shape discrimination using a continuous incident photon beam, the TOF method is expected to have a higher efficiency for collecting high energy neutrons, reduced environmental background, and a higher likelihood of determining angular distributions. The TOF method may use a shortened pulse structure (10 ns) and gated detectors to reject gamma flash. The advantages inherent in the TOF method, combined with the modified LINAC and detectors, may partially compensate for the reduced duty cycle of commonly deployed pulsed accelerators.
In a time-of-flight (TOF) embodiment, the electron accelerator 601 or other source may be pulsed to produce electron beam 602 (pulsed on) for a time period T and turned off for a time long enough to have all the detectable neutrons (resulting from interactions of the photon beam 607 with the container 606 and its contents) pass through the detector(s). Then the electron beam 602 may be pulsed on again for a time period T. This sequence may be repeated until the desired detection data is obtained.
The electron accelerator 601 or some subsidiary target (not shown) near the bremsstrahlung target 603 or in the bremsstrahlung or photon beam 607 may provide a fiducial signal that informs the signal processing electronics and/or computer 616 when the photon beam 607 was generated. Neutrons generated by photofission in the fissile sample 608 travel to a detector in the time L/v where L is the distance from the fissile sample 608 to the detector in question and v is the neutron velocity. For detector 611, for example, which is opposite the fissile sample 608 at a right angle to the incident photon beam 607 in the embodiment shown, L=L611+y, the distance from the fissile sample 608 to the corresponding wall of the container 606 nearest detector 611. The velocity of the neutrons is given by v=(2E/m)1/2, where E is the neutron kinetic energy and m is the neutron mass. The signal from detector 611 goes to the signal processing electronics and/or computer 616, which converts the difference between the fiducial signal arrival time and the detector 611 signal arrival time into the time-of-flight (TOF) of the neutron to the detector. Using the relation TOF=(L611+y)/v, the signal processing electronics and/or computer 616 calculates the neutron velocity and therefore its energy (E=mv2/2) and records the data and also transfers it to a central control and analysis system (not shown).
The energy resolution of the detection system will depend on the TOF of the neutrons, T, L and the dispersion of the flight distance to different portions of the detectors. Those experienced in the art will recognize that these parameters, including the electron beam pulse width T, and the geometry of the system can be adjusted to achieve energy resolution adequate for the purposes of this disclosure.
The (narrow) photon beam 607 may be scanned across the container 606 sequentially to illuminate discrete columns where the fissile sample 608 may be located. This serves to better localize the position of any fissionable material and will reduce backgrounds from other neutron producing materials in a container. Alternatively, the photon beam 607 may be a wide fan-like beam encompassing a greater region of the container 606 with the fan opening out in the direction toward the detectors at 90 degrees, for example. This allows a broad scan region of the container but limited in the narrow direction. Such an embodiment would facilitate scanning the container in shorter times for fissile materials. It would detect fissile materials distributed over the dimensions of the fan beam. In this geometry x and y will not be known but they may be inferred from a comparison of the neutron energy spectra on both sides of the container since they should be very close to identical, especially at the highest energies. Starting with any assumption for “a”, such as ½ the width of the container (x=y), the resulting spectra can be adjusted by varying “a” until the spectra are made to have the same high-energy shape.
The technology for short duration electron beam pulses is a well-known art, and pulses of a few nanoseconds are readily generated for high energy electron beams. Time of flight for a 1 MeV neutron over 1 m is 72 nanoseconds. Thus, flight distances of a few meters result in flight times (˜71 nanoseconds for 6 MeV neutrons over a distance of 3 meters, for example) that allow beam pulse duration times of 10 to 20 nanoseconds to separate photo-fission neutrons from those from (γ, n) processes by energy selection.
Other specific embodiments are possible and some are mentioned herein as further illustrations of methods to articulate the concepts and methods described earlier.
The detectors 610, 611, 612, and 613 in
As discussed above, the energy distribution of neutrons from photo-fission is very independent of the energy of the photons used to induce photo-fission in the photon energy regions discussed herein, in or below the Giant Electric Dipole Resonance. Another embodiment uses this fact to determine whether the neutrons originate from photo-fission. Varying the photon energy or the bremsstrahlung endpoint energy will not substantially alter the energy distribution of the neutrons from photo-fission. However, this is not true for other processes such as (γ, n), especially in the higher regions of neutron energy, as a result of energy conservation and the requirement E=Eb−Eth discussed earlier. Therefore, measuring the energy distribution of the neutrons for different photon energies, and comparing the results, can identify actinide photo-fission. Alternatively, measuring and comparing the number of neutrons above a certain energy as the photon energy is changed can achieve the same result.
Another embodiment would measure the neutron yield at a given neutron energy, as the photon energy is varied, and would do this for several neutron energies. This would generate yield curves for neutrons of the given energies as a function of photon energy. Because the neutron energy spectra from photon-induced fission is independent of the incident photon energy, the same yield curve as a function of photon energy would result for all neutron energies if the spectrum is dominated by photo-fission. However, if the neutron spectrum originates from (γ, n) for relevant non-actinide materials, each neutron energy has a yield curve as a function of photon energy displaced in photon energy by that explicit neutron separation energy, in particular for the neutrons at the highest energy possible. Once again this follows from energy conservation.
Neutron detection can be based on reaction energies between the neutrons and the component materials in the detector. Detectors of such a nature may sometimes but not always be called “threshold detectors” because a reaction will occur only if the neutron energy is greater than a certain amount. Examples of such reactions include but are not limited to (n, n′γ), (n, n′f), (n, n′p), (n, n′d) and (n, n′α). Detection of the event may be based on, but not limited to, the detection of: a scintillation event and measuring the deposited energy; the charge created by ionization in a material and measuring the total charge; and, the detection of radioactive nuclei, wherein the radioactivity would be induced only if the neutron energy (energies) were greater than a certain value (or values). All such methods are included in the embodiments described in this disclosure.
As discussed above, some commercially available plastic and liquid scintillators can identify neutrons unambiguously using suitable signal processing techniques. Such detectors also have fast enough time response to qualify for the purposes herein and these will be known to those skilled in the art. Such detectors operate in part as proton recoil detectors, based on the energy imparted to protons by the elastic scattering of neutrons from the protons in the hydrogenous material. Therefore, in part, they can function as “threshold detectors” as discussed above, as well as providing the time for an event in a detector and identifying the event as a neutron. Such detection methods are part of the embodiments described herein.
Delayed neutrons following beta decay can also be detected by the methods discussed herein and serve as a method of detecting fissile materials. They will be less abundant than prompt neutrons by a very large factor, as discussed above. In most cases their presence can be used as a further detection method to augment the embodiments discussed herein. They can be distinguished from prompt neutrons by several techniques. Using TOF with a pulsed beam set to measure prompt neutrons, delayed neutrons appear as a uniform distribution in time that builds up with exposure time or the number of pulses in the TOF embodiment discussed above. The time for buildup of the delayed neutron signal is characteristic of beta-decay lifetimes. If the beam is turned off they will diminish in times characteristic of beta-decay lifetimes. The presence of the delayed neutrons may be neglected in many situations as a minor contribution. In some cases they may be used as an aid to the detection of fissile material. In all situations, the presence of delayed neutrons may be accounted for and the results corrected accordingly if the correction is required by these embodiments.
The photon beams may be of the pulsed variety described above in discussing TOF embodiments, or they may be of continuous character as from continuous duty radiofrequency accelerators, DC accelerators or similarly functioning photon sources of a monochromatic or nearly monochromatic nature.
Another scan embodiment would employ a very broad beam geometry in all directions transverse to the beam direction with collimation so as to limit the beam size to that of the container width in its largest manifestation. This embodiment would be very effective in the detection of fissile materials dispersed in small samples over a large volume, such as thin sheets broadly distributed over a large region of the container or small pellets broadly distributed.
Many beam geometries are possible, each with specific advantages for certain situations as will be recognized by those skilled in the art, and they are all included in this disclosure.
In order to carry out scanning of containers as rapidly as possible, it may be preferable to carry out an initial scan with a low threshold or trigger neutron detection energy, in order to maximize the signal from photofission, even at the cost of obtaining a signal from (γ, n) processes. If no events are recorded from the container or a portion thereof in an appropriate interval, or no events above an acceptable background, the scan can be continued to a further portion of the container, or the container can be passed on if th entire container has been scanned. If events are detected, the threshold or trigger neutron detection energy can be increased, and the container or portion thereof rescanned, using the higher neutron threshold or trigger detection energy to reduce or eliminate the contamination from the competing (γ, n) processes. Alternatively, of course, other of the methods set forth herein for discriminating between photofission and (γ, n) processes can be employed in the rescan.
Because angular distributions may be difficult to measure given the differential absorption and scattering of different cargo loadings, it is important to recognize that, as discussed above, if the energy distribution of the prompt neutrons is independent of angle relative to the photon beam, then the fragments are emitted isotropically and the fissile material is an odd-even isotope: however, if the prompt neutrons have a spectrum with greater population at the higher energies at 90 degrees to the photon beam relative to the prompt neutron spectrum at large angles near 180 degrees, then the fragments have an angular distribution peaking at 90 degrees and the fissile material is an even-even isotope. Therefore, measuring the neutron energy distribution at two angles will enable this determination to be made.
Another embodiment removes the uncertainty in the energy distribution and angular dependencies of the prompt neutrons caused by the differential absorption along different paths that neutrons take in traversing a container to the different detectors. This embodiment directs the photon beam into the container in different directions. For example, in one arrangement the photon beam may enter the container from the top and the neutron detectors view the neutrons at 100 degrees to the beam and at 170 degrees from the beam. By altering the photon beam direction to enter from the side of the container the detectors change roles. That one previously at 100 degrees is now at 170 degrees and the one previously at 170 degrees is now at 100 degrees. However, the differential aspects of neutron absorption remain exactly the same. The two measurements now provide a clear indication of the influence on the neutron energy distribution of the angle of emission of the neutron relative to the photon direction as well as the angular distribution of the neutrons relative to the photon beam direction. As one particular feature, if the photo-fission process is isotropic the relative neutron yields in the detectors will not change. A change indicates anisotropy in the original photo-fission process.
This process can be generalized for other angles as well. For example,
F
1(E,θ1)=I1×A1(E)×S(E,θ1) for first neutron detector 801; and,
F
2(E,θ2)=I1×A2(E)×S(E,θ2) for second neutron detector 802.
The neutrons detected by the two detectors with beam in direction 2 are:
F
1(E,θ2)=I2×A1(E)×S(E,θ2) for first neutron detector 801; and,
F
2(E,θ1)=I2×A2(E)×S(E,θ1) for second neutron detector 802.
The attenuation factors A1 and A2 remain invariant to the beam position and the ratio can be formed to eliminate these factors so that:
{S(E,θ1)/S(E,θ2)}2={F1(E,θ1)×F2(E,θ1)}/{F2(E,θ2)×F1(E,θ2)}. (Equation 1)
Thus, S(E,θ1) and S(E,θ2) are related via measured quantities and can be compared directly. A person skilled in the art will be able generalize this technique to more than two detectors and this embodiment is intended to contain all these variations.
Unless otherwise specified, the illustrative embodiments can be understood as providing exemplary features of varying detail of certain embodiments, and therefore, unless otherwise specified, features, components, modules, and/or aspects of the embodiments can be otherwise combined, specified, interchanged, and/or rearranged without departing from the disclosed devices or methods. Additionally, the shapes and sizes of components are also exemplary, and unless otherwise specified, can be altered without affecting the disclosed devices or methods. Other specific embodiments are possible and some are mentioned herein as further illustrations of methods to articulate the concepts and methods described earlier.
Although the terms “nuclear material”, “fissionable nuclear material”, “fissile material”, and “fissionable material” have been variously used in this disclosure, the intent of the inventors is that these terms are used interchangeably and are all intended to designate those materials that can be induced to fission by the effect of a gamma ray or by a thermal neutron or fast neutron. These terms are not intended to mean materials that emit neutrons in response to gamma or neutron irradiation, unless such materials also may be induced to fission by the effect of a gamma ray or by a thermal neutron or a fast neutron. The term “container” as used herein is intended to include any enclosure or partial enclosure that may enclose or partially enclose a fissionable material so as to hide or partly hide it or shield it or partly shield it from conventional detection methods—it includes but is not limited to cargo and shipping containers and vehicles.
While the systems and methods disclosed herein have been particularly shown and described with references to exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the disclosure. It should be realized this disclosure is also capable of a wide variety of further and other embodiments within the spirit of the disclosure. Those skilled in the art will recognize or be able to ascertain using no more than routine experimentation, many equivalents to the exemplary embodiments described specifically herein. Such equivalents are intended to be encompassed in the scope of the present disclosure.
The present application claims the benefit of: (1) U.S. Provisional Patent Application Ser. No. 60/944,009, entitled “Non-Intrusive Method To Identify Presence Of Nuclear Materials Using Energetic Prompt Neutrons From Photon Induced Fission” which was filed on Jun. 14, 2007 by Robert J. Ledoux and William Bertozzi, and is hereby incorporated by reference; and (2) U.S. Provisional Patent Application Ser. No. 60/971,638, entitled “Non-Intrusive Method To Identify Presence Of Nuclear Materials Using Energetic Prompt Neutrons From Photon Induced Fission And Neutron-Induced Fission” which was filed on Sep. 12, 2007 by Robert J. Ledoux and William Bertozzi, and is also hereby incorporated by reference.
This invention was made with government support under Contract No. N66001-07-D-0025/Delivery Order No. 0001 awarded by the U.S. Navy. The government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 60944009 | Jun 2007 | US | |
| 60971638 | Sep 2007 | US |
