354
OPTICS LETTERS / Vol. 35, No. 3 / February 1, 2010
Isotope-specific detection of low-density materials with laser-based monoenergetic gamma-rays F. Albert,* S. G. Anderson, G. A. Anderson, S. M. Betts, D. J. Gibson, C. A. Hagmann, J. Hall, M. S. Johnson, M. J. Messerly, V. A. Semenov, M. Y. Shverdin, A. M. Tremaine, F. V. Hartemann, C. W. Siders, D. P. McNabb, and C. P. J. Barty Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, California 94550, USA *Corresponding author:
[email protected] Received October 7, 2009; accepted November 16, 2009; posted December 23, 2009 (Doc. ID 118125); published January 26, 2010 What we believe to be the first demonstration of isotope-specific detection of a low-Z and low density object shielded by a high-Z and high-density material using monoenergetic gamma rays is reported. The isotopespecific detection of LiH shielded by Pb and Al is accomplished using the nuclear resonance fluorescence line of 7Li at 478 keV. Resonant photons are produced via laser-based Compton scattering. The detection techniques are general, and the confidence level obtained is shown to be superior to that yielded by conventional x-ray and ␥-ray techniques in these situations. © 2010 Optical Society of America OCIS codes: 340.7480, 000.2190.
In this Letter, we report the demonstration of a technique that enables the detection and isotopic identification of low-Z and low-density targets shielded by high-Z and high-density materials. We describe the detection of a LiH target shielded by Pb and Al using nuclear resonance fluorescence (NRF) and the Thomson-radiated extreme x-ray (T-REX) monoenergetic gamma-ray (MEGa-ray) source of Lawrence Livermore National Laboratory (LLNL). The T-REX is a laser-based MEGa-ray source that produces quasi-monochromatic, tunable, polarized ␥ rays via Compton scattering of energetic short duration laser pulses from high-brightness relativistic electron bunches. The detection of low-Z and low-density objects shielded by a high-Z dense material is a longstanding problem that has important applications ranging from homeland security and nonproliferation [1] to advanced biomedical imaging and paleontology. X rays are sensitive to electron density, and x-ray radiography yields poor contrast in these situations. Within this context, NRF offers a unique approach to the so-called inverse density radiography problem. NRF is a process in which nuclei are excited by discrete high-energy (typically mega-electron-volt) photons and subsequently re-emit ␥ rays at discrete energies determined by the structure of the nucleus. Because the resonance structure is determined by the number of neutrons and protons present in the nucleus, NRF provides an isotope-specific detection and imaging capability [2]. NRF transitions, however, are narrowband 共⌬E / E = 10−6兲 and are thus inefficiently excited by the broad bandwidth (near 100% bandwidth) of bremsstrahlung sources. The bremsstrahlung sources also create a considerable background via the elastic Compton scattering and other processes, and the dose accumulated during detection is much higher than for a MEGa-ray source created via laser-based Compton scattering. Such sources can produce extreme peak brilliance in the mega-electron-volt spectral range [3,4], a range that is beyond third-generation synchrotrons. Previously laser-based Compton scattering source development has been motivated by the 0146-9592/10/030354-3/$15.00
desire to produce either subpicosecond pulse duration x rays for dynamic studies [5,6] or tunable radiation in the 10–100 keV range for specific x-ray radiographic applications [7]. Recently, laser-based Compton MEGa-ray sources have also been used for nuclear science and applications purposes. At Duke University the HI␥S 2–60 MeV high-energy ␥-ray facility produces polarized photons, with a photon flux of approximately 105, via intracavity Compton backscattering in a free-electron laser and has been used as a research tool to assign the parity of excited states in nuclei [4]. In Japan ␥ rays with a photon flux of 105 up to 5.7 MeV have been produced by the collision of a Q-switched laser beam and a highenergy electron beam from a storage ring [8] and have been used to detect 208Pb concealed in an iron box. The development of the T-REX MEGa-ray source for NRF-based material detection at LLNL has optimized laser-based Compton scattering to create a record peak brilliance of 1.5 ⫻ 1015 photons/ mm2 / mrad2 / s / 0.1% bandwidth (BW) at 478 keV. The T-REX utilizes an existing 120 MeV S-band linear accelerator (linac) and custom laser systems designed specifically for laser-based Compton scattering x-ray and ␥-ray sources. The accelerator has been upgraded from previous laser Compton experiments [9] to increase the electron beam brightness and energy. The experiment (Fig. 1) was conducted in three different below-ground caves: the outer detector cave, where the interaction laser, producing the near-time-bandwidth-limited, colliding photon beam (150 mJ, 10 ps, and 532 nm) was located; the accelerator cave containing the photoinjec-
Fig. 1. T-REX experimental layout and detection with HPGe detectors. © 2010 Optical Society of America
February 1, 2010 / Vol. 35, No. 3 / OPTICS LETTERS
tor, the accelerator, and the interaction point; and finally the 0° cave, on the other side of a thick concrete wall where the ␥-ray diagnostics, including Ge detectors, were set up. The measured spatial and spectral properties of the source, shown in Fig. 2, are consistent with Compton scattering theory. The MEGa-ray beam is collimated and the on-axis spectrum is sharply peaked and relatively narrowband. The on-axis spectrum is peaked at an energy of 4␥ 2EL, where ␥ is the electron relativistic factor and EL is the laser photon energy. The beam profile has been measured with a 16 bit 1024⫻ 1024 pixels intensified CCD (ICCD) Andor camera coupled to a 3:1 optical fiber reducer and a CsI[Tl] scintillator. The beam profile shown in Fig. 2 indicates a 6.0⫻ 10.4 mrad2 divergence and an integrated x-ray dose of 105 photons/ shot. The divergence is lower along the horizontal direction due to polarization effects [9]. The source spectrum has been measured through a 6 mm Pb collimator indirectly by scattering the ␥ rays off a 3-mm-thick Al plate onto a 50% efficiency, 2.8 keV resolution at a 662 keV high purity germanium (HPGe) detector (with a 6 ⫻ 8 cm cylindrical Ge crystal) operating in a statistical single-photon counting mode (rate of 20%) and placed 150 cm away from the plate. The scattered ␥ rays of the energy E␥⬘ are correlated with the incident ␥ rays of the energy E␥ by the Compton scattering relation E␥⬘ = E␥ / 关1 + E␥ / E0共1 − cos 兲兴, where E0 = 511 keV is the electron rest energy and = 48° is the scattering angle of the ␥ rays on the detector. The spectrum peaks at 365 keV corresponding to an incident photon energy of 478 keV. The shape of the recorded spectrum and the relative bandwidth of 12% FWHM (50 keV) agree with Monte Carlo simulations of the source and detector arrangement based on the MCNP5 code [10]. Both the expected and measured spectra are plotted in Fig. 2. The continuum below 250 keV is due to the incomplete energy absorption (elastic Compton scattering) in the detector. The peak at 110 keV arises from the double Compton scattering off the Al plate and the adjacent wall followed by the photo-absorption in the Ge detector. The peak near 88 keV is from the x-ray fluorescence in the Pb shield surrounding the Ge detector. The re-
Fig. 2. (Color online) Experimentally measured on-axis spectra (dots) and corresponding Monte Carlo simulation (dashed).
355
sults are consistent with the laser-based Compton scattering predictions based on the measured laser and electron beam parameters [3]. All the source parameters described above are summarized by the peak brilliance at 478 keV of 1.5 ⫻ 1015 photons/ mm2 / mrad2 / s / 0.1% BW. This peak brilliance compares with the polychromatic bending radiation produced by large, third-generation synchrotrons and is well suited for the excitation of NRF. T-REX photons have been used to excite NRF in 7 Li and subsequently to detect concealed 7Li. Two methods have been employed. First we directly observe isotope-specific NRF photons scattered from a test object. Second we observe the attenuation of on resonance ␥-rays in the transmitted beam to determine the presence or absence of the target material. In a first experiment, the source has been used to excite the 478 keV line of 7Li 共ZLi= 3兲 in a low density 共LiH= 0.36 g / cm3兲 LiH sample shielded by higher Z elements: 3 mm of Al共ZAl= 13兲 and 8 mm of Pb共ZPb= 82兲. The ratio Pb/ LiH, where is the mass attenuation coefficient, is equal to nearly 60. 225 g of LiH are placed in a light plastic bottle with a diameter of 8 cm and are irradiated by the ␥-ray beam 20 m away from the source. At this location, the beam diameter, limited by apertures along the beam path, is 4 cm. The NRFscattered photons are detected by a HPGe detector positioned at 90° with respect to the incident beam axis and 15 cm away from the center of the beam. With the source parameters and experimental arrangement described above, the NRF photons are expected to be scattered in a dipole radiation pattern at a rate of 16 photons/h. The HPGe detector was positioned in the x-ray polarization plane to maximize the NRF signal from the M1 transition in 7Li [11]. A scattering angle of 90° was chosen in order to minimize the amount of the elastic Compton scattering background from the LiH target [12], since the ␥ rays are linearly polarized. The count rate on the HPGe was close to 10% to avoid pile up. The results are shown in Fig. 3. The raw data (0.2 keV/channel) are binned to a 1 keV resolution. The peak observed at 511 keV results from the annihilation of e + −e − pairs created by the background highenergy bremsstrahlung from the linac. This line is
Fig. 3. Experimental results for NRF detection in 7Li.
356
OPTICS LETTERS / Vol. 35, No. 3 / February 1, 2010
still present when the interaction laser is off. The NRF line from 7Li is observed at 478 keV with a 6 confidence level. For comparison, if one considers the transmission of ␥ rays in the same energy range [13] propagating through 1/8 in. of Al, 8 cm of LiH, 5/8 in. of Pb (shielded LiH), 1/8 in. of Al, and 5/8 in. of Pb (shielding alone), the attenuation of ␥ rays alone, under our experimental conditions, does not differ enough to detect the presence of LiH. This difference (5%) lies within the standard deviation of the background (27%) in our experiment. Moreover, only NRF provides a characteristic energy signature and isotopic sensitivity. In addition, simple calculations, using the known NRF cross section of 7Li and x-ray NIST attenuation data [13], show that NRF becomes more effective as the Pb shielding increases. Having detected NRF from 7Li, we investigated another method to ascertain the presence or absence of a given isotope that has the potential to yield lower false positive and negative rates [1]. In this method, proposed in [14], ␥ rays are transmitted through the material under interrogation to a reference sample containing the isotope of interest. If NRF is detected from the reference, one can conclude that the interrogated material did not contain the isotope. If no NRF is observed, either the resonant photons have been scattered by the isotope in the interrogated sample or the material is optically too thick. Narrowband Compton sources have been proved advantageous for this method [1]. To illustrate this detection method, an identical LiH test sample was placed in the beam path before the reference LiH bottle. The upstream test sample was mounted on a horizontal translation stage so that we could observe the effect on the NRF-scattered spectrum in the reference sample. The data displayed in Fig. 4 represent the scattered spectra obtained by the reference detector when the test sample is in and out of the beam, with background subtraction. In each spectrum the 511 keV line can be observed as in Fig. 3. When the test sample is in the beam, the NRF line is not observed on the reference detector [Fig. 4(a)]. When the test sample is pulled out, the NRF line appears on the reference detector [Fig. 4(b)]. In conclusion, we have shown that a low-Z and low density material 共 7Li兲 can be detected behind higher-Z and higher density materials (Pb and Al) with photons by using NRF excited by a high-energy,
Fig. 4. NRF detection in 7Li with two samples: output of reference detector when the interrogated LiH is (b) out of and (a) in the beam.
high-brightness MEGa-ray source created via laserbased Compton scattering. NRF provides unique isotope sensitivity. The capability to observe optically thin objects embedded in an optically thick material with a deeply penetrating radiation has numerous potential applications ranging from medical imaging to homeland security, nuclear waste assay, and stockpile stewardship. The creation of MEGa-ray Compton sources with ⬎103 higher flux than used in these proof-of-principle experiments is now under construction. This work was performed under the auspices of the U.S. Department of Energy (DOE) by Lawrence Livermore National Laboratory under contract DEAC52-07NA27344. The support from the Domestic Nuclear Detection Organization (DNDO) of the Department of Homeland Security is also acknowledged. References 1. J. Pruet, D. P. McNabb, C. A. Hagmann, F. V. Hartemann, and C. P. J. Barty, J. Appl. Phys. 99, 123102 (2006). 2. U. Kneissl, H. M. Pitz, and A. Zilges, Prog. Part. Nucl. Phys. 37, 349 (1996). 3. F. V. Hartemann, W. J. Brown, D. J. Gibson, S. G. Anderson, A. M. Tremaine, P. T. Springer, A. J. Wootton, E. P. Hartouni, and C. P. J. Barty, Phys. Rev. ST Accel. Beams 8, 100702 (2005). 4. V. N. Litvinenko, Nucl. Instrum. Methods Phys. Res. A 507, 527 (2003). 5. R. Schoenlein, W. Leemans, A. Chin, P. Volfbeyn, T. Glover, P. Balling, M. Zolotorev, M. Kim, S. Chattopadhyay, and C. Shank, Science 274, 236 (1996). 6. D. J. Gibson, S. G. Anderson, C. P. J. Barty, S. M. Betts, R. Booth, W. J. Brown, J. K. Crane, R. R. Cross, D. N. Fittinghoff, F. V. Hartemann, J. Kuba, G. P. Lesage, D. R. Slaughter, A. M. Tremaine, A. J. Wooton, E. P. Hartouni, P. T. Springer, and J. B. Rosenzweig, Phys. Plasmas 11, 2857 (2004). 7. W. D. Andrews, F. E. Carroll, J. W. Waters, C. A. Brau, R. H. Price, D. R. Pickens, P. A. Tompkins, and C. F. Roos, Nucl. Instrum. Methods Phys. Res. A 318, 189 (1992). 8. N. Kikuzawa, R. Hajima, N. Nishimori, E. Minehara, T. Hayakawa, T. Shizuma, H. Toyokawa, and H. Ohgaki, Appl. Phys. Express 2, 036502 (2009). 9. W. J. Brown, S. G. Anderson, C. P. J. Barty, S. M. Betts, R. Booth, J. K. Crane, R. R. Cross, D. N. Fittinghoff, D. J. Gibson, F. V. Hartemann, E. P. Hartouni, J. Kuba, G. P. Le Sage, D. R. Slaughter, A. M. Tremaine, A. J. Wooton, and P. T. Springer, Phys. Rev. ST Accel. Beams 7, 060702 (2004). 10. http://mcnp-green.lanl.gov/, retrieved January 19, 2010. 11. N. Pietralla, Z. Berant, V. N. Litvinenko, S. Hartman, F. F. Mikhailov, I. V. Pinayev, G. Swift, M. W. Ahmed, J. H. Kelley, S. O. Nelson, R. Prior, K. Sabourov, A. P. Tonchev, and H. R. Weller, Phys. Rev. Lett. 88, 012502 (2001). 12. L. W. Fagg and S. S. Hanna, Rev. Mod. Phys. 31, 711 (1959). 13. http://physics.nist.gov/PhysRefData/Xcom/Text/ XCOM.html, retrieved January 19, 2010. 14. W. Bertozzi, S. E. Kobly, R. J. Ledoux, and W. Park, Nucl. Instrum. Methods Phys. Res. B 261, 331 (2007).
