Non-intrusive method to identify presence of nuclear materials using energetic prompt neutrons from photon-induced fission

Abstract

Methods and systems for non-intrusively detecting existence of fissile materials by measuring energetic prompt neutrons. The unique nature of the prompt neutron energy spectrum from photo-fission arising from emission of neutrons from almost fully accelerated fragments is used to unambiguously identify fissile material. Angular distribution of prompt neutrons from photo-fission and energy distribution correlated to neutron angle relative to the photon beam are used to distinguish odd-even from even-even nuclei undergoing photo-fission. Independence of the neutron yield curve (yield as a function of electron beam energy or photon energy) on neutron energy is used to distinguish photo-fission from other processes such as (, n). Different beam geometries detect localized samples of fissile material and fissile materials dispersed as small fragments or thin sheets over broad regions. These signals from photo-fission are unique, allowing detection of any material in the actinide region of the nuclear periodic table.

Claims

1. A method of detecting a presence of a material comprising an actinide in a container, comprising: a) locating the container such that one or more neutron detectors view the container at a first angle relative to a photon beam; b) for at least two predetermined cutoff photon energies, illuminating at least a portion of the container with the photon beam comprising photons of energies no greater than the predetermined cutoff photon energy; c) for at least two of said predetermined cutoff photon energies, detecting in at least one of said one or more neutron detectors at least some neutrons produced by an interaction of the photon beam with at least a portion of the container; d) for at least two of said predetermined cutoff photon energies, for each of a plurality of said detected neutrons, determining the energy of the detected neutron; e) choosing a higher neutron energy region where neutrons from a (, n) process are not energetically permitted for any of said at least two predetermined cutoff photon energies, and a lower energy region where neutrons from a (, n) process are energetically permitted for all of said at least two predetermined cutoff photon energies; f) for at least two of said predetermined cutoff photon energies, determining a neutron yield in one of said one or more neutron detectors in at least two predetermined neutron energy ranges, wherein at least one predetermined neutron energy range encompasses the higher energy region where neutrons from a (, n) process are not energetically permitted; and wherein at least one other predetermined neutron energy range encompasses the lower energy region where neutrons from a (, n) process are energetically permitted; and g) based upon comparing the determined neutron yield in the one of said one or more neutron detectors, resulting from incident photon beams comprising photons of energies no greater than said different predetermined cutoff energies, confirming that the material comprising the present actinide is present in the container if an increase in the neutron yield between a lower predetermined cutoff photon energy and a higher predetermined cutoff photon energy, in the higher predetermined neutron energy range where neutrons from a (, n) process are not energetically permitted, is less than an increase in neutron yield in the lower predetermined neutron energy range, where neutrons from a (, n) process are energetically permitted.

2. The method of claim 1, wherein the photon beam comprising photons of energies no greater than a said predetermined cutoff photon energy is a bremsstrahlung beam produced by electrons of the predetermined cutoff energy.

3. The method of claim 1, wherein the photon beam comprising photons of energies no greater than a said predetermined cutoff photon energy is a monochromatic photon beam.

4. The method of claim 1, wherein determining the energy of the detected neutron comprises measuring a time of flight of the detected neutron.

5. The method of claim 1, wherein determining the energy of the detected neutron comprises analyzing the energy deposited in at least one of said one or more neutron detectors.

6. The method of claim 1, wherein the container is located such that at least two neutron detectors of the one or more neutron detectors view the container, the at least two neutron detectors viewing the container from different angles relative to the photon beam, and neutrons are detected in the at least two neutron detectors; further comprising: h) for at least one of said predetermined cutoff photon energies, determining a total neutron yield in the at least two neutron detectors in a predetermined neutron energy range; and i) based upon comparing the total neutron yields from the at least two neutron detectors viewing the container from different angles relative to the photon beam, determining that the present actinide is an odd-even isotope if the total yields disclose an isotropic distribution of neutrons as a function of angle relative to the photon beam, and determining that the present actinide is an even-even isotope if the total yields disclose an anisotropic distribution of neutrons as a function of angle relative to the photon beam.

7. The method of claim 1, further comprising: h) for at least one new angle relative to the photon beam, i) moving at least one of said one or more neutron detectors such that the moved neutron detector views the container from the new angle relative to the photon beam; ii) illuminating at least a portion of the container with the photon beam comprising photons of energies no greater than at least one of said predetermined cutoff photon energies; iii) detecting in said at least one moved neutron detector at least some neutrons produced by an interaction of the photon beam with at least a portion of the container; and iv) for each of a plurality of said detected neutrons, determining the energy of the detected neutron; i) for at least some of said neutron detector viewing angles relative to the photon beam, determining a total neutron yield in the at least one moved neutron detector in a predetermined neutron energy range at the angle relative to the photon beam; and j) based upon comparing the total neutron yields, from the at least one moved neutron detector, for at least some of the neutron detector viewing angles relative to the photon beam, determining that the present actinide is an odd-even isotope if the total yields disclose an isotropic distribution of neutrons as a function of angle relative to the photon beam, and determining that the present actinide is an even-even isotope if the total yields disclose an anisotropic distribution of neutrons as a function of angle relative to the photon beam.

8. The method of claim 1, wherein the container is located such that at least two neutron detectors of the one or more neutron detectors view the container, the at least two neutron detectors viewing the container from different angles relative to the photon beam, and neutrons are detected in the at least two neutron detectors; further comprising: h) for at least one of said plurality of predetermined cutoff photon energies, determining a neutron energy distribution in the at least two neutron detectors; and i) based upon comparing the neutron energy distributions from the at least two neutron detectors viewing the container from different angles relative to the photon beam, determining that the present actinide is an odd-even isotope if the energy distributions do not change by more than a predetermined amount as a function of angle relative to the photon beam, and determining that the present actinide is an even-even isotope if the energy distributions change by more than a predetermined amount as a function of angle relative to the photon beam.

9. The method of claim 1, further comprising: h) for at least one new angle relative to the photon beam, i) moving at least one of said one or more neutron detectors such that the moved neutron detector views the container from the new angle relative to the photon beam; ii) illuminating at least a portion of the container with the photon beam comprising photons of energies no greater than at least one of said predetermined cutoff photon energies; iii) detecting in said at least one moved neutron detector at least some neutrons produced by an interaction of the photon beam with at least a portion of the container; and iv) for each of a plurality of said detected neutrons, determining the energy of the detected neutron; i) for at least some of said neutron detector viewing angles relative to the photon beam, determining a neutron energy distribution in the at least one moved neutron detector; and j) based upon comparing the neutron energy distributions, from the at least one moved neutron detector, for at least some of the neutron detector viewing angles relative to the photon beam, determining that the present actinide is an odd-even isotope if the energy distributions do not change by more than a predetermined amount as a function of angle relative to the photon beam, and determining that the present actinide is an even-even isotope if the energy distributions change by more than a predetermined amount as a function of angle relative to the photon beam.

10. A method of detecting a presence of a material comprising an actinide in a container, comprising: a) locating the container such that one or more neutron detectors view the container; b) for at least two predetermined cutoff photon energies, illuminating at least a portion of the container with a photon beam comprising photons of energies no greater than the predetermined cutoff photon energy; c) for at least two of said predetermined cutoff photon energies, detecting in at least one of said one or more neutron detectors at least some neutrons produced by an interaction of the photon beam with at least a portion of the container; d) for at least two of said predetermined cutoff photon energies, for each of a plurality of said detected neutrons, determining the minimum energy of the detected neutron; e) choosing a higher neutron energy region where neutrons from a (, n) process are not energetically permitted for any of the at least two of said predetermined cutoff photon energies, and a lower energy region where neutrons from a (, n) process are energetically permitted for all of the at least two said predetermined cutoff photon energies; f) for at least two of said predetermined cutoff photon energies, determining a neutron yield in one of said one or more neutron detectors in at least two predetermined neutron minimum energy ranges, wherein at least one predetermined neutron minimum energy range encompasses the higher energy region where neutrons from a (, n) process are not energetically permitted; and wherein at least one other predetermined neutron minimum energy range encompasses the lower energy region where neutrons from a (, n) process are energetically permitted; and g) based upon comparing the determined neutron yield in the one of said one or more neutron detectors resulting from the incident photon beams comprising photons of energies no greater than said different predetermined cutoff energies, confirming that the material comprising the present actinide is present in the container if an increase in the neutron yield between a lower predetermined cutoff photon energy and a higher predetermined cutoff photon energy, in the higher predetermined neutron minimum energy range where neutrons from a (, n) process are not energetically permitted, is less than an increase in neutron yield in the lower predetermined neutron minimum energy range, where neutrons from a (, n) process are energetically permitted.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIGS. 1A and 1B show the fission fragment mass yields and associated average neutron multiplicities as a function of fission fragment mass resulting from neutron-induced fission of .sup.235U and .sup.239Pu, respectively;

(2) FIG. 2 presents the time-of-flight (and energy) spectrum of neutrons for the photo-fission of .sup.232Th;

(3) FIG. 3 shows the energy spectra of photo-neutrons produced by the (, n) process from gold for bremsstrahlung beams of 14.3 and 15.8 MeV endpoint energies;

(4) FIGS. 4A and 4B show the photo-fission yield as a function of bremsstrahlung endpoint energy in .sup.235U and .sup.239Pu, respectively;

(5) FIGS. 5A, 5B, 5C, and 5D display the photon induced reaction cross sections for .sup.239Pu for the (, Total), (, n), (, 2n) and (, f) processes, respectively;

(6) FIG. 6 shows a schematic layout of a possible arrangement for one embodiment of a system for detecting fissile materials in a container by analyzing energetic prompt neutrons resulting from photon-induced fission;

(7) FIG. 7 shows the angular distribution of the neutrons emitted from fission fragments from the photo-fission of .sup.232Th and .sup.238U;

(8) FIGS. 8A and 8B are schematics showing two beam positions for detecting neutrons produced by fission fragments from photon-induced fission with the angles for the detectors relative to the beam interchanged

(9) FIG. 9 shows the angular distribution of the fission fragments from the photofission of .sup.232Th and .sup.238U; and

(10) FIG. 10 shows the (n, f) cross section for .sup.238U.

(11) FIG. 11 shows pulse-shape discriminated energy spectra from HEU and Pb targets irradiated with a 9 MeV bremsstrahlung beam, illustrating the separation of photons from the neutron signal, and the separation of prompt photo-fission neutrons from neutrons produced by the (, n) process in Pb.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

(12) 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.

(13) When fission is spontaneous, initiated by low energy neutrons or by the absorption of photons near the threshold for the (, 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 .sup.235U and in similar regions for other fissionable nuclei. The fragments are accelerated by the strong Coulomb repulsion of their charges (Z.sub.1, Z.sub.2) 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.sup.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.

(14) FIGS. 1A and 1B display the analysis by J. Terrell (Neutron Yields from Individual Fission Fragments, Physical Review, Vol. 127, Number 3, Aug. 1, 1962, pages 880-904, and references therein) for the neutron induced fission of .sup.235U and .sup.239Pu. These figures (which correspond to FIGS. 8 and 9 in Terrell) display the asymmetric fragment mass distributions from the neutron-induced fission and the average number of neutrons emitted from the heavy and light fragments, as a function of the mass of the fragments, for .sup.235U and .sup.239Pu. (The symbols , .sub.L and .sub.H in FIGS. 1A and 1B denote the average total number of neutrons emitted, the average neutrons emitted from the light fragment and the average neutrons emitted from the heavy fragment, respectively, as a function of fragment mass.)

(15) Similar results have been obtained by Terrell for neutron-induced fission of .sup.233U and the spontaneous fission of .sup.252Cf, showing the generality of the phenomena.

(16) Many authors have studied the spontaneous fission of .sup.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 .sup.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:

(17) a) The angular distribution (of the neutrons from the spontaneous fission of .sup.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

(18) 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.

(19) 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.

(20) The work of H. W. Schmitt, J. H. Neiler, and F. J. Walter, Fragment Energy Correlation Measurements for .sup.252Cf Spontaneous Fission and .sup.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 186.51.2 MeV for the spontaneous fission of .sup.252Ca and 171.91.4 MeV for neutron induced fission of .sup.235U. The fragments have substantially all the kinetic energy available from the mutual Coulomb repulsion of the fragments.

(21) Both the energy distribution and the angular distribution of the neutrons from fission fragments created by photon-induced fission are relevant. The case of .sup.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 .sup.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:

(22) 1.) The (, n) threshold energy for .sup.232Th is 6.438 MeV. Therefore, the (, 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 (, n) process even at neutron energies considerably lower than 0.31 or 1.31 MeV, respectively; and

(23) 2.) The fission fragments in photo-fission, (, f), are known to have strongly anisotropic angular distributions from .sup.232Th. The distribution is peaked at 90 degrees to the incident photon beam, and the fragment angular distribution is given by I=a+b sin.sup.2(), where 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.

(24) The results of analysis of the neutron energy spectra from .sup.232Th (, f) are consistent with the following conclusions of Sargent et al:

(25) 1.) The fraction of the prompt neutrons that result from emission from other than the fully accelerated fragments is 0.070.09;

(26) 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

(27) 3.) The energy spectrum of the neutrons in the center of mass frame of reference is characterized by an average energy of 1.140.06 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.) FIG. 7 presents angular distributions of prompt neutrons from the fission fragments produced in the (, f) process for incident photon energies near the threshold for the (, f) process, for .sup.232Th and .sup.238U. It is taken from 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), who corroborate the relevant .sup.232Th results of Sargent et al. and also extend the results to the photo-fission of .sup.238U. These angular distributions are measured by detectors which detect the fragments from neutron induced fission of .sup.238U. Therefore, they are an average over all the energies of the neutrons emitted from the photo-fission fragments convoluted with the (n, f) cross section. This emphasizes neutrons above approximately 1 MeV, where the (n, f) cross section becomes large (See FIG. 10, which presents the (n, f) cross section for .sup.238U. FIG. 10 is reproduced from National Nuclear Data Center, Brookhaven National Laboratory, ENDF, Evaluated Nuclear (reaction) Data File).

(28) FIG. 9, also taken from Nair, presents the angular distributions of the fragments from the photo-fission for .sup.232Th and .sup.238U, for the same incident photon energies as FIG. 7. The peaking of the neutrons from the photo-fission fragments in the direction of the motion of the fission fragments is clearly demonstrated by a visual comparison of FIG. 7 with FIG. 9. (The implications of the shape of the neutron angular distribution are discussed below.)

(29) FIG. 2, which is taken from FIG. 4 of the Sargent et al. reference, displays the time-of-flight (energy) spectrum of prompt neutrons from photo-fission of .sup.232Th at 77 degrees with respect to the direction of an incident 7.75 MeV. photon beam. At the top of FIG. 2 is the prompt neutron energy scale.

(30) One outstanding feature of the neutron spectrum in FIG. 2 is the presence of neutrons at very high energy compared to an evaporation (thermal) spectrum with an average energy of approximately 1.14 MeV, as reported by Sargent et. al. from their analysis of the energy and angular distributions of the prompt neutrons from photon induced fission of .sup.232Th. For example, the intensity at 6 MeV is considerable. The presence of a large number of neutrons at high energy results in part from the considerable boost in velocity transferred to the neutrons by the moving fragments. For example, if the velocity of the fragment corresponds to a kinetic energy of 1 MeV/nucleon, then a 1 MeV neutron emitted in the fragment center of mass frame of reference in the direction of fragment motion will have twice the velocity in the laboratory frame of reference and a kinetic energy of 4 MeV. This follows because the neutron velocity in the laboratory frame is the sum of the fragment velocity and the neutron velocity in the fragment frame. Since these are the same for the energies and directions considered in this example, the velocity is doubled. The kinetic energy varies as the square of the velocity. Hence the neutrons with 1 MeV in the fragment frame of reference have 4 MeV in the laboratory frame of reference. More generally, if the fragment velocity is V and the co-directional neutron velocity in the fragment frame is v, then the neutron velocity in the laboratory frame is V+v. The kinetic energy of the neutron in the laboratory frame is E=(m/2)(V.sup.2+2Vv+v.sup.2) or E=E.sub.f(1+2(E.sub.n/E.sub.f).sup.0.5+E.sub.n/E.sub.f) where E.sub.n is the neutron kinetic energy in the fragment frame and E.sub.f is the kinetic energy of one nucleon of the fragment. Thus, in the above example, a neutron emitted in the fragment direction of motion at 2 MeV in the fragment center of mass frame of reference will have a laboratory kinetic energy of 5.8 MeV.

(31) Energy conservation in the direct (, n) neutron production process does not allow the production of neutrons with an energy above E=E.sub.bE.sub.th, where E.sub.b is the bremsstrahlung endpoint energy of the incident photon beam and E.sub.th is the (, 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.

(32) Since the (,n) threshold of .sup.232Th is 6.438 MeV, a neutron energy of 6 MeV will not be possible from (, n) until the bremsstrahlung endpoint reaches 12.438 MeV. Also, even when the bremsstrahlung endpoint reaches that value, neutrons from the (, n) process will be very small in number because they can only be produced by the few photons at the bremsstrahlung endpoint energy.

(33) These energetic considerations apply in a similar manner for all fissionable nuclear materials, in particular for those with Z89, the region of the actinides. In addition, and most importantly, most heavy elements such as Bi, Pb, W, Ta, etc. have isotopes with (, n) thresholds at or above 6.5 MeV. Therefore, finding neutrons with energies above E=E.sub.bE.sub.th where E.sub.th is in the range of 6 MeV constitutes a very definitive test for the presence of fissile material.

(34) Another test to verify that the detected neutrons result from photo-fission is the sensitivity of the yield of neutrons at energies above E=E.sub.bE.sub.th 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.

(35) FIG. 3 displays spectra of (, n) neutrons for gold. (It is FIG. 2 from W. Bertozzi, F. R. Paolini and C. P. Sargent, Time-of-Flight Measurements of Photoneutron Energy Spectra, Physical Review, 119, 790 (1958)). FIG. 3 illustrates how the nature of the (, n) process causes neutrons produced by that process to be concentrated mostly at low energies. The data in FIG. 3 are normalized to yield the same number of neutrons from the (, n) process in a reference target of .sup.2D with neutron energies E.sub.n>1.4 MeV. Because photon and neutron energy are uniquely related in the (, n) process in .sup.2D, this normalization allows the formation of the difference photon spectrum (the difference between the high energy (15.8 MeV) bremsstrahlung spectrum and the low energy (14.3 MeV) bremsstrahlung spectrum), which corresponds to a broad band of photons centered at approximately 14.5 MeV and with approximately a 2 MeV half width. That is, the neutron energy spectrum produced by the difference in the neutron energy spectra at the two energies in FIG. 3 corresponds to photo neutrons produced by photons in the above energy band centered at approximately 14.5 MeV. FIG. 3 confirms that, because neutrons produced by the (, n) process are concentrated mostly at low energies, the contamination of a photo-fission spectrum by neutrons from the (, n) process is expected to be low at higher neutron energies, even when one looks at neutrons at energies below the E=E.sub.bE.sub.th cutoff established by the strict application of energy conservation.

(36) The spectra in FIG. 3 show the rapid, almost exponential decrease of neutrons from the (, n) process with increasing neutron energy, in contrast to the neutron spectrum from the photo-fission (, f) of .sup.232Th at 7.75 MeV bremsstrahlung energy as shown in FIG. 2. For gold the neutron spectrum from (, n) is nonexistent with if the bremsstrahlung spectrum endpoint is 7.75 MeV, since the (, n) threshold, E.sub.th, is above 8 MeV. Even with a 12 MeV bremsstrahlung endpoint, the highest neutron energy from (, n) in gold would be less than 4 MeV, and neutrons in this energy range from (, n) would not be numerous because they would correspond to photons at the endpoint of the bremsstrahlung spectrum. The neutron yield from (, f) in .sup.232Th is very large at 12 MeV bremsstrahlung for neutron energies above 6 MeV.

(37) Table 1 gives the (, f) and the (, n) thresholds (in MeV) for some typical nuclei in the actinide region. The (, f) thresholds are from H. W. Koch, Experimental Photo-Fission Thresholds in .sup.235U, .sup.238U, .sup.233U, .sup.239Pu and .sup.232Th, Physical Review, 77, 329-336 (1950). The (, n) threshold of .sup.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 (, n) process for bremsstrahlung end point energies up to 11 MeV, including for .sup.207Pb. This energy is to be compared to the spectrum in FIG. 3 showing many neutrons with energies in excess of 6 MeV from .sup.232Th photo-fission using bremsstrahlung of only 7.75 MeV. Even with an 11 MeV bremsstrahlung energy there are no neutrons above 5.7 MeV from any nucleus, and no neutrons above 4.26 from .sup.207Pb, and those at or near these energies would be very few in number because they correspond to the photons at or near the end-point energy of the bremsstrahlung spectrum. It should be noted that the (, f) process increases in importance as the bremsstrahlung endpoint energy increases from 6 to 11 MeV because of the increasing cross section with energy and because of the increasing number of photons in the bremsstrahlung spectrum at lower photon energies where the (, f) cross section is sizable. The (, f) thresholds are almost all lower than the (, n) thresholds, and are all significantly lower than the (, n) threshold for .sup.207Pb.

(38) TABLE-US-00001 TABLE 1 Maximum Neutron Energies from (, n) for Selected Bremsstrahlung Energies and Isotopes. Maximum (, n) Neutron Energy (MeV) (, f) Threshold (, n) Threshold Bremsstrahlung Endpoint Energy, E.sub.b (MeV) Element (MeV) (MeV) 6 7 8 9 10 11 .sup.232Th 5.40 0.22 6.438 0.56 1.56 2.56 3.56 4.56 .sup.233U 5.18 0.27 5.759 0.24 1.24 2.24 3.34 4.34 5.34 .sup.235U 5.31 0.27 5.298 0.70 1.70 2.70 3.70 4.70 5.70 .sup.238U 5.08 0.15 6.154 0.85 1.85 2.85 3.85 4.85 .sup.239Pu 5.31 0.25 5.647 0.35 1.35 2.35 3.35 4.35 5.35 .sup.207Pb 6.738 0.26 1.26 2.26 3.26 4.26

(39) 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 .sup.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 FIG. 3, the electron beam energy would have to exceed 12 MeV to provide a neutron yield above 4 MeV.) However, the yield of neutrons above 4 MeV for the actinides would be a strongly increasing function of electron beam energy starting below 6 MeV electron beam energy since the low (, f) threshold allows the photo-fission process to grow rapidly as more and more photons are available for photo-fission, all of them producing a neutron spectrum independent of photon energy and strongly populating the selected region of neutron energy (above 4 MeV for example). The (, n) process in the actinide examples shown in Table 1 or in other heavy metals such as .sup.207Pb would not be a significant component of the total yield until the electron beam energy is well above 10 MeV since the process involves only the photons near the bremsstrahlung endpoint, E.sub.b.

(40) An additional point, which will be discussed further below, is that the photo-fission cross section is larger than the (, n) cross section over most photon energies by a considerable amount, as shown in FIGS. 5B and 5D. The neutrons from (, f) will dominate (, n) in most situations simply on the basis of the cross sections, aside from the other features discussed herein.

(41) 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 (, 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 (, f) cross section. The maximum neutron energy from (, n) dictated by energy conservation considerations for monochromatic incident photons follows just as discussed above.

(42) 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 (, n) process in relevant heavy materials such as .sup.207Pb and therefore any neutrons detected could only originate from the photo-fission process in an actinide. The data in FIG. 2 show that there are many neutrons above 6 MeV from the (, f) process, and hence 6 MeV could be selected as a trigger energy. Other trigger energies are possible also; the choice is dependent on factors such as the speed of detection that is desirable, the false positives that are to be allowed, and the efficiency of detection that is desired.

(43) 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 (, n) thresholds, such as copper, aluminum, steel or oxygen, then a lower trigger could be selected.

(44) Conversely, hydrogenous material that naturally contains a small percentage of deuterium may be of concern because of its low threshold for the (, 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=(E.sub.b2.2)/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

(45) A more important concern may be .sup.9Be. It has a low (, 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)(E.sub.b1.6) MeV is the maximum neutron energy available. For the example of E.sub.b=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 (, n) process in .sup.9Be by taking advantage of the fact that the (, n) process follows the strict rule for conservation of energy, so that E=( 8/9)(E.sub.b1.6) 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, E.sub.b less than approximately 15 MeV. Therefore, neutrons at an energy greater than E=( 8/9)(E.sub.b1.6), where E.sub.b is the photon beam energy or bremsstrahlung endpoint energy, is proof of a fissile material. At E.sub.b=10 MeV, the presence of neutrons above approximately 7.5 MeV would be proof. At E.sub.b=8 MeV, neutrons above 5.7 MeV would be proof. Also, the prompt neutron energy spectrum is independent of the photon energy while the (, n) process in .sup.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 .sup.9Be.

(46) However, if there were concern that this measurement could not be reliably made, further steps could be taken. Operating at E.sub.b=10 MeV, the maximum neutron energy from beryllium (, n) is approximately 7.5 MeV. By reducing the beam energy to 8 MeV, for example, the maximum energy neutron from beryllium (, 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 (, 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.

(47) Fortunately, .sup.9Be is almost unique in this category. There are a few other nuclei with relatively low (, n) thresholds; .sup.6Li, .sup.13C, .sup.17O and .sup.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.

(48) FIGS. 4A and 4B, from H. W. Koch, Experimental Photo-Fission Thresholds in .sup.235U, .sup.238U, .sup.233U, .sup.239Pu and .sup.232Th, Physical Review, 77, 329-336 (1950), FIGS. 4 and 5, display the yield of fission fragments as a function of bremsstrahlung endpoint energy (Peak Spectrum Energies) for two isotopes, .sup.235U (FIG. 4A) and .sup.239Pu (FIG. 4B). These illustrate the rapid increase of the fission yield as a function of the energy of the electron beam used to produce bremsstrahlung. FIG. 4A also shows the dominance of the .sup.235U contribution over the impurities of .sup.238U in the enriched uranium sample. These data are based upon the detection of the actual photo-fission fragments. The yield of prompt neutrons follows approximately the same yield curve, since neutron emission in the photo-fission process is not dependent on the photon energy in the regions of interest below the Giant Electric Dipole Resonance at approximately 12 to 13 MeV photon energy. The emission of neutrons from the fragments is determined by the complex dynamics, discussed earlier, of splitting the fissioning nucleus into two fragments.

(49) 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 (, n) process, which will start at the endpoint energy given by E.sub.b=E.sub.th+E.sub.n, where E.sub.n is the neutron energy that is desired. They are thus displaced from the (, n) threshold energy, E.sub.th, by the neutron energy, in contrast to the yield curves for (, f). This is a powerful signature that the neutrons detected are from photo-fission rather than from (, n).

(50) FIGS. 5A through 5D display the photon induced reaction cross sections for .sup.239Pu. They are taken from FIG. 7 of B. L. Berman, J. T. Caldwell, E. J. Dowdy, S. S. Dietrich, P. Meyer, and R. A. Alvarez, Photofission and photoneutron cross sections and photofission neutron multiplicities for .sup.233U, .sup.234U, .sup.237Np and .sup.239Pu, Physical Review C Volume 34, Number 6, 2201-2214 (1986). FIG. 5A shows the total photon absorption cross section. FIG. 5B shows the partial cross section for (, 1n), single neutron emission. FIG. 5C shows the partial cross section for (, 2n), double neutron emission. FIG. 5D shows the partial cross section for (, f), photo-fission.

(51) The photo-fission cross section (FIG. 5D) is larger than the (, n) cross section (FIG. 5B) over most photon energies by a considerable amount. This displays the feature common to the actinides that photo-fission is a strong and often dominant process from the (, f) threshold throughout much of the Giant Electric Dipole Resonance. Given that the prompt neutron multiplicities from photo-fission range from approximately 2.5 to more than 3, prompt neutrons from the photo-fission process will dominate the incident photon reaction channel by a large factor at all neutron energies. This feature facilitates identifying the presence of actinide fissionable material despite the potential presence of other heavy elements such as Pb, even without considering the energy-conservation constraints on neutron energy. The photon absorption process in most heavier nuclei is dominated by neutron emission, and the total yield is governed by the giant dipole sum rule that the integrated cross section is proportional to NZ/A, which is a slowly varying function. Since the location of the giant dipole resonance in energy also is a slowly varying function of nuclear mass, a yield of prompt neutrons from the photo-fission process that is 2.5 to 3 times the expected neutron yield from (, n) is a signal of photo-fission in that the (, f) neutron yield alone will produce a markedly higher photon absorption cross section than would (, n) for a given quantity of heavy material. That is, measuring the yield of neutrons per heavy nucleus per photon permits identifying photo-fission as present, if the quantity of heavy material present can be determined by measuring localized density by other methods.

(52) 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.

(53) 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 .sup.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 .sup.235U, .sup.237Np and .sup.239Pu. A strongly anisotropic neutron angular distribution would indicate an even-even fissile species such as .sup.232Th and .sup.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.)

(54) 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.

(55) 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: Angular distributions of photofission fragments relative to the photon beam have been measured as a function of maximum bremsstrahlung energy in the range 6-20 Mev. The nuclides U-233, U-235, Np-237, Pu-239 and Am-241 give an isotropic distribution at all energies studied. The nuclides Th-232, U-234, U-236, U-238, and Pu-240 give anisotropic distributions which can be described by an equation of the form W()=1+ sin.sup.2, where is the angle between fragment and beam. The degree of anisotropy is large at low energy and falls rapidly as the energy is increased. At a given energy Th-232 has the greatest degree of anisotropy and Pu-240 the least.

(56) 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:

(57) TABLE-US-00002 TABLE 2 Angular Distributions (from Berg, et al. Table I) Angular distributions Ratio, counts at 90/counts at 0* Nuclide E.sub.0 = 6.0 E.sub.0 = 6.5 E.sub.0 = 8.0 E.sub.0 = 10.0 E.sub.0 = 20.0 U-233 1.048 0.07 1.032 0.04 0.994 0.03 U-235 1.024 0.05 Np-237 1.024 0.10 Pu-239 1.034 0.927 1.002 0.06 1.013 0.05 0.952 0.03 0.26 0.12 Am-241 0.958 0.07 *The ratio is the number of counts observed at 90 per unit X-ray dose divided by the number observed at 0 for the same dose. E.sub.0 is the maximum energy in million electron volts of the bremsstrahlung spectrum. The 45/0 ratio at E.sub.0 = 6.5 Mev was 1.09 0.23.

(58) TABLE-US-00003 TABLE 3 Corrected Values of (from Berg, et al. Table VI) Corrected values of in W() = 1 + sin.sup.2 E.sub.0 Th-232 U-238 U-236 U-234 Pu-240 6.0 6.6 2 6.0 2.3 6.3 6.7 1.1 6.5 >25 4.4 1.0 2.1 0.4 2.3 0.6 0.65 0.20 7.0 11.0 0.8 2.05 0.24 1.33 0.17 0.90 0.16 0.49 0.12 7.5 10.3 1.6 8.0 4.9 0.6 1.3 0.1 0.79 0.09 0.44 0.08 0.29 0.07 9.0 2.8 0.4 0.51 0.07 9.4 0.44 0.04 10.0 1.61 0.12 0.41 0.05 0.32 0.06 0.17 0.07 14.0 0.46 0.09 0.09 0.04 0.04 0.03 15.0 0.02 0.04* 0.01 0.03* 20.0 0.14 0.06 0.05 0.03 *These values, which do not differ from zero, have not been corrected for isotopic composition.

(59) 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, .sup.235U, .sup.239Pu and .sup.237Np. These will undergo isotropic photo-fission, in contrast to .sup.238U, .sup.232Th and the other even-even isotopes that were measured.

(60) FIG. 9, which displays the fission fragment angular distributions from photo-fission of an even-even nucleus, and FIG. 7, which displays the angular distributions of the prompt neutrons emitted from those fragments, demonstrate the general peaking of the fragment and neutron angular distributions at 90 degrees relative to the photon beam. The neutron yield at 150 degrees is approximately 20% less than that at 90 degrees and the shape of the distribution is approximately symmetric about 90 degrees. The fragment angular distributions show a larger anisotropy as expected because the neutron distributions are produced by folding the isotropic angular distributions in the fragment center of mass with the fragment distributions in angle. In contrast to these distributions, the isotopes .sup.235U, .sup.239Pu and .sup.237Np (not shown), which may be used in the manufacture of weapons, display for the most part isotropic angular distributions of the photo-fission fragments as discussed above and the resulting angular distributions of the prompt neutrons also are isotropic.

(61) One embodiment of a detector system to carry out the methods described herein requires a source of photons with energy capable of exceeding the (, 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.

(62) 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 (, n) process in materials other than actinides, as discussed above.

(63) Because the contamination of non-actinide (, 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.

(64) 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.

(65) 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.

(66) FIG. 11 demonstrates the energy separation of prompt neutrons from photon induced fission from neutrons produced from the (, n) reaction, when incident photon energies below 9 MeV are used. That Figure was obtained using 9 MeV bremsstrahlung beams produced at the CW S-DALINAC at the Technical University of Darmstadt. The spectra from Pb and highly enriched uranium (HEU) targets shown in FIG. 11 were obtained using the technique proposed for the CAARS PNPF module. An organic liquid scintillator was used to determine the energy deposited in the detector and to separate photon and neutron events. The lighter data points represent events from Pb (which as discussed above has a (, n) threshold of approximately 6.5 MeV for the .sup.207Pb isotope) and the darker represent events from HEU (which has a photo-fission threshold of 5.5 MeV). The neutron events are grouped in the lower portion of the figure and are clearly differentiated from the photon events above. Within the neutron events, the box shows where the neutrons from prompt photo-fission in HEU appear unambiguously. A neutron signal in this region is an unambiguous signal of an actinide photo-fission event.

(67) 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 FIG. 6. An electron beam 602 of energy E.sub.b is generated by an electron accelerator 601. The electron beam 602 makes bremsstrahlung radiation photons when it strikes a bremsstrahlung target 603 (also called the radiator). The electron accelerator 601 and radiator 603 optionally may be replaced by a source of monochromatic or nearly monochromatic photons. The optional collimator 604 collimates the bremsstrahlung radiation. A shield 605 may enclose the bremsstrahlung target 603 and electron accelerator 601. The photon beam 607 is directed onto a container 606 which is to be analyzed and which may contain fissile material 608. The distances of the fissile material 608 from (for example) three of the walls of container 606 are designated as x, y and z. A photon detector 609 placed after the container 606 optionally may be used to monitor the transmitted photon flux of photon beam 607. Detectors 610, 611, 612, and 613 may be placed at locations around the container 606 at approximately 90 degrees and at convenient back angles with respect to the collimated photon beam 607. The number and location of the detectors may be varied from that shown in FIG. 6 according to the principles and methods discussed above. In the illustrated embodiment, the detectors 610 and 611 may be placed at known distances L610 and L611 from the container 606 walls. The detectors 610, 611, 612, and 613 optionally may be surrounded by shielding (not shown) and by anti-coincidence counting systems (not shown) if desired. The detectors 610, 611, 612, and 613 themselves may be sensitive to neutron energy or they may be part of a system (such as one utilizing time-of-flight) that will provide a neutron energy for each detected neutron event. A beam dump 614 may be used to absorb the remaining photon flux after the photon beam 607 passes through the container 606 and its contents. The beam dump 614 and optional transmitted flux monitor, detector 609, may be shielded from direct view of the detectors as required. Signals from the detectors 609, 610, 611, 612, and 613 are connected by way of connections 615 to signal processing electronics and/or computer 616, which process the detector signals and optionally may relay them and/or processed information by way of connections 617 to a central control and data analysis and storage system (not shown.) Alternatively, the detector signals may be passed directly to the central system for processing and analysis.

(68) 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.

(69) 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.

(70) 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).sup.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=mv.sup.2/2) and records the data and also transfers it to a central control and analysis system (not shown).

(71) 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.

(72) 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 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.

(73) 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 (, n) processes by energy selection.

(74) Other specific embodiments are possible and some are mentioned herein as further illustrations of methods to articulate the concepts and methods described earlier.

(75) The detectors 610, 611, 612, and 613 in FIG. 6 can be any that are capable of unambiguously detecting a neutron. Rather than measuring the neutron time of flight to determine its energy, it would suffice in some applications to only specify that the event is definitely a neutron and that the energy is greater than a defined amount. This would characterize the neutron energy as above a defined quantity. Several such neutron energies may be involved. Together with control of the electron beam energy or photon energy as discussed above, determining the number of neutrons with energies above certain preset quantities will classify the neutrons as from photo-fission. As discussed above, other processes such as (, n) will not be possible at neutron energies greater than E=E.sub.bE.sub.th, where E.sub.b is the bremsstrahlung endpoint or the photon energy and E.sub.th is the threshold for (, n) for relevant non-actinide materials that may be present and need to be distinguished from the suspected actinide.

(76) 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 (, n), especially in the higher regions of neutron energy, as a result of energy conservation and the requirement E=E.sub.bE.sub.th, 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.

(77) 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 (, 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.

(78) 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), (n, nf), (n, np), (n, nd) and (n, n). 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.

(79) 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.

(80) 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.

(81) 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.

(82) 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.

(83) 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.

(84) 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 (, 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 (, n) processes. Alternatively, of course, other of the methods set forth herein for discriminating between photofission and (, n) processes can be employed in the rescan.

(85) 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.

(86) 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.

(87) This process can be generalized for other angles as well. For example, FIGS. 8A and 8B, each show a container 806 with fissile material 808. A first neutron detector 801 is shown in a first location and a second neutron detector 802 is shown in a second location. In FIG. 8A, a photon beam 807A irradiates the container 806 from a first direction (direction 1). In FIG. 8B, a photon beam 807B irradiates the container 806 from a second direction (direction 2) For each of the two photon beam directions, the neutron detectors 801 and 802 interchange angles relative to the photon beam (807A or 807B) direction. For beam direction 1, first neutron detector 801 is at angle .sub.1 and second neutron detector 802 is at angle .sub.2. For beam direction 2, first neutron detector 801 is at angle .sub.2 and second neutron detector 802 is at angle .sub.1. I.sub.1 and I.sub.2 are the photon beam intensities at the target 808 (which may be a fissile material) for photon beam directions 1 and 2 respectively. If S(E,) is the energy spectrum of neutrons produced in direction , the neutrons detected by the two detectors with photon beam direction 1 are described by the measured functions F.sub.i(E, .sub.j); F.sub.1(E,.sub.1)=I.sub.1A.sub.1(E)S(E,.sub.1) for first neutron detector 801; and, F.sub.2(E,.sub.2)=I.sub.1A.sub.2(E)S(E,.sub.2) for second neutron detector 802.

(88) The neutrons detected by the two detectors with beam in direction 2 are: F.sub.1(E,.sub.2)=I.sub.2A.sub.1(E)S(E,.sub.2) for first neutron detector 801; and, F.sub.2(E,.sub.1)=I.sub.2A.sub.2(E)S(E,.sub.1) for second neutron detector 802.

(89) The attenuation factors A.sub.1 and A.sub.2 remain invariant to the beam position and the ratio can be formed to eliminate these factors so that:
{S(E,.sub.1)/S(E,.sub.2)}.sup.2={F.sub.1(E,.sub.1)F.sub.2(E,.sub.1)}/{F.sub.2(E,.sub.2)F.sub.1(E,.sub.2)}. (Equation 1)

(90) Thus, S(E,.sub.1) and S(E,.sub.2) 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.

(91) 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.

(92) 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 methodsit includes but is not limited to cargo and shipping containers and vehicles.

(93) 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.