ARTICLE IN PRESS
Nuclear Instruments and Methods in Physics Research A 516 (2004) 228–236
High-energy photon beam production with laser-Compton backscattering K. Aokia, K. Hosonoa,*, T. Hadamea, H. Munenagaa, K. Kinoshitaa, M. Todaa, S. Amanob, S. Miyamotob, T. Mochizukib, M. Aokic, D. Lic a
Graduate School of Engineering, Himeji Institute of Technology, 2167 Shosha Himeji, Himeji, Hyogo 671-2201, Japan b LASTI, Himeji Institute of Technology, Kamigori-Kouto, Hyogo 678-1205, Japan c Institute for Laser Technology, Honmachi, Osaka 550-0004, Japan Received 25 June 2003; received in revised form 12 August 2003; accepted 26 August 2003
Abstract We have produced a beam of high-energy gamma-rays by Compton backscattering of 1064 nm laser photons from 1 GeV electrons circulating in a storage ring. Measuring the energy spectra of the backscattering photons with an HPGe detector, we have found that the maximum energy of 17:6 MeV and the measured energy spectra show agreement with the simulation calculations, while the detected photon yields were measured at about 4 103 ; 2 103 and 3 102 photons s1 mA1 W1 for the 20, 10 and 2 mmf collimators, respectively. The photon energy widths from the collimators correspond to 6.6–17:6 MeV; 12.4–17:6 MeV and 17.3–17:6 MeV; respectively. By using these photons, we measured the total nuclear photoabsorption cross-sections in the E1 giant resonance energy region for 197 Au using the attenuation method and we have demonstrated that the photon attenuation method will be a useful tool for studying photonuclear reactions. r 2003 Elsevier B.V. All rights reserved. PACS: 29.20.Dh; 29.30.Kv; 25.20.Dc; 25:20: x; 27.80.+w Keywords: High-energy photons; Laser-Compton backscattering; Photon absorption coefficient; Photonuclear reaction; Giant resonance
1. Introduction Compton backscattering of laser light from relativistic electrons is a promising method of producing useful yields of high-energy monochromatic polarized photons. The photon beam (g-ray *Corresponding author. Tel.: +81-792-67-4921; fax: +81792-67-4921. E-mail address:
[email protected] (K. Hosono).
beam) is thought to have properties that will open up a range of new possibilities for basic research and applications. The first real gamma-ray beam for nuclear physics research, based on the laser backscattering technique, was developed at the 1:5 GeV ADONE storage ring [1], but there are now a number of facilities which produce polarized photons for nuclear physics studies. In order to carry out photonuclear experiments on the electromagnetic interactons in the energy range of several MeV to tens of MeV, a laser-Compton
0168-9002/$ - see front matter r 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.nima.2003.08.153
ARTICLE IN PRESS K. Aoki et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 228–236
backscattering facility has been installed at the storage ring, NewSUBARU [2] at Laboratory of Advanced Science and Technology for Industry (LASTI) at Himeji Institute of Technology. The angle-energy correlation method is used in that system. Measurements of the resonantly scattered photons represent a unique tool for investigating elementary excitations with low multipolarity. These g-rays can be also used for electron beam diagnoses at the storage ring, calibration of detectors and nondestructive inspection. In this paper, we describe the facility’s present status, including its application to nuclear physics. The head-on collision of relativistic electrons and laser photons creates a pencil-like beam of g-rays, and the energy of the produced g-ray ðEg Þ depends on the backscattering angle (y) of the laser photon and is given by the well-known formula Eg ¼
EL ð1 b cos yL Þ 1 b cos y þ EL ð1 cosðyL yÞÞ=Ee
ð1Þ
where EL is the energy of the laser photon, Ee is the electron beam energy, yL is the incident angle of the laser photon, and b is ue =c: For the case of head-on collision, yL ¼ p; y51 and g ¼ Ee =me c2 b1; the gamma energy is given approximately by Eg ¼
4g2 EL 1 þ ðgyÞ2 þ 4gEL =mc2
ð2Þ
where me is the rest-mass of the electron. Clearly, we obtain the maximum energy of g-ray at y ¼ 0 : With a sufficiently narrow backscattering angle, the gamma-rays are nearly monochromatic. For an effective length of the colliding region LðcmÞ; a laser energy EL ðeVÞ; a CW laser power PL ðWÞ; a stored electron current Ie ðampsÞ and an effective overlap area Aðcm2 Þ of the laser photon and electron bunch at a colliding region, the expected gamma-ray flux Yg is obtained [3,4] as Yg ¼
2:604Ie ðampÞPL ðWÞsðmbÞLðcmÞ : EL ðeVÞAðcm2 Þ
ð3Þ
Here, sðmbÞ is the laboratory cross-section backscattered into a cone of angle ðyc Þ: If the electron beam has a Gaussian distribution and the laser beam has a Gaussian power distribution, A is
given [3,4] as qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiqffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi A ¼ 2p s2L þ s2X s2L þ s2Y
229
ð4Þ
where sL is the RMS radius of the laser beam, sX is the RMS width of the electron beam, and sY is the RMS height of the electron beam.
2. Laser-Compton backscattering facility A plan of the NewSUBARU facility is shown in Fig. 1. The racetrack-shaped ring can operate at energies from 1.0 up to 1:5 GeV and has the ability to store an average current of 500 mA at an injection energy of 1 GeV: The horizontal and the vertical emittances of the stored beam are about 3:8 108 and 2:0 109 m rad; respectively. An 11-m long straight line (BL-1), was used for the laser-Compton backscattering experiment. Fig. 2 shows a schematic overview of the laser-Compton backscattering facility. The energies of the scattered photons as a function of the scattered angles, the energy spectrum and the angular distribution [5] were calculated for backscattering of 1:168 eV laser photons from 1 GeV electrons. The calculated results are shown in Fig. 3. The laser source used is a CW Nd : YVO4 laser (l ¼ 1064 nm; EL ¼ 1:168 eV) and both the horizontal and the vertical waist diameters of the laser source are 0:42 mm and both the full divergences are 3:4 mrad: The laser photons are injected into the interaction area with six mirrors and a lens; the distance between the laser source and the lens is about 7 m and that between the lens and the center of the interaction region is about 15 m: The laser beam waist is located at the center of the interaction area. Fig. 4 shows the calculated overlap areas between the laser beam and the electron beam at several points along the straight line. We estimated the photon flux produced by this facility. In the calculations, we used the following parameters: (1) The stored electron current ðIe Þ is 10 mA: (2) The power ðPL Þ of the 1:168 eV laser is 0:74 W: (3) the cross-section (s) can be obtained from the angular distribution shown in Fig. 3. (4)
ARTICLE IN PRESS 230
K. Aoki et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 228–236
Fig. 1. Plan of the NewSUBARU facility.
Fig. 2. Schematic overview of laser Compton backscattering facility.
The interaction length ðLÞ is 14 m: (5) The effective overlap areas given by Eq. (4) were calculated by taking the variation of the overlap area with the position along the straight line as shown in Fig. 4. Also included is the attenuation of the photons by several media between the interaction
area and the detector. The calculated results are shown in Fig. 5, where the horizontal axis shows the collimation angle of the scattered photons. These calculations do not include the effects of the angular divergence of both the electron and the laser beams.
ARTICLE IN PRESS K. Aoki et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 228–236
231
Fig. 3. Energies of the scattered photons, the angular distribution and the energy distribution calculated for backscattering of 1:168 eV (1064 nm) laser photons from 1:0 GeV electrons.
3. Production of high-energy photon beam We conducted experiments at an electron energy of 1:0 GeV; a current of 5–20 mA and a laser power of 0.5–1:0 W: The backscattered photon beam was collimated in one of 0:67 mrad ð20 mmfÞ; 0:335 mrad ð10 mmfÞ or 0:067 mrad ð2 mmfÞ 10 cm long lead collimators. The photon energy widths correspond to 6.6–17:6 MeV; 12.4– 17:6 MeV and 17.3–17:6 MeV for 20, 10 and 2 mmf; collimators, respectively. The collimator was placed on the gamma-ray beam axis approximately 14 m from the center of the interaction region. Fine-tuning of the laser-beam axis and the collimator alignment were accomplished by maximizing the count rate in the detector. We monitored the produced photon flux with a 180 cm3 coaxial-type HPGe detector. The detector, shielded by lead blocks was placed on the photon beam axis approximately 20 m from the
center of the collision area and about 1 m behind the lead collimator. The g-ray energies were calibrated with the 1:461 MeV g-rays from the 40 K and several standard g-ray sources. There are two potential background sources of photons, namely synchrotron radiation and Bremsstrahlung g-rays which originate from high-energy electrons interacting with residual atoms in the evacuated beam line. The synchrotron radiation energies were all less than 1 keV: The Bremsstrahlung photons were measured with an electron energy of 1 GeV and with a current of 10 mA: The vacuum pressure in the beamline was less than 1 107 Pa: An example of the energy spectra of the Bremsstrahlung g-rays is shown in Fig. 6. The geometry of the present system requires the backscattered photons to penetrate several media between the interaction area and the detector, all of which are background sources. The produced gamma-rays penetrate a 7 mm thick silicon dioxide mirror for the injection of laser
ARTICLE IN PRESS 232
K. Aoki et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 228–236
Fig. 4. Calculated overlap areas between the laser beam and the electron beam at several points along the interaction line. The electron beam is shown in black and the laser beam is shown in gray.
ARTICLE IN PRESS K. Aoki et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 228–236
Fig. 5. Estimated photon flux as a function of collimation angle.
233
The energy spectra of the backscattered photons were measured at an electron energy of 1 GeV: In this measurement, the stored electron current was 10 mA and the laser power was 0:5 W: The solid lines in Fig. 7 show the energy spectra of the backscattering photons measured with the various collimators. The Bremsstrahlung photons have been subtracted from the original spectra, and the maximum measured photon energy shows good agreement with the calculated energy. When using the 2 mmf collimator, the photo peak and the single escape peak can be seen as shown in Fig. 7, though the peaks are very small due to the detector’volume. We compared the obtained energy spectra with the simulated model calculations using the MonteCarlo electron-gamma shower simulation code, EGS4 [6]. We calculated the simulated spectra by assuming an ideal energy distribution at each collision point and by taking into account the geometry including the detector size and attenuation by various media such as a mirror or the glass window, etc. The calculated energy spectra are shown with the dotted lines in Fig. 7. The shapes of the calculated energy spectra reproduce the measured ones. The photon yield can be estimated from the normalization between the measured and the calculated spectra. The detected photons are about 4 103 ; 2 103 and 3 102 s1 mA1 W1 for the 20, 10 and 2 mmf collimators, respectively. These values are consistent with the rough estimated values as shown in Fig. 5.
4. Measurement of photoabsorption cross-section of 197 Au
Fig. 6. An example of the measured energy spectra of the Bremsstrahlung g-rays.
photons and a 5 mm thick optical-glass window which separates the high vacuum from the external atmosphere. The photons will interact with the lead collimator and the scattering may occur in any of these media.
The measurement of the nuclear photoabsorption cross-sections in the E1 giant resonance energy region has been carried out using the attenuation method [7,8]. A beam of photons with an incident intensity I0 ; penetrating a target material with the thickness of d (cm) and the density of rðg=cm3 Þ; emerges with intensity I given by the exponential attenuation law I=I0 ¼ exp½mtot d :
ð5Þ
ARTICLE IN PRESS 234
K. Aoki et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 228–236
Fig. 7. Energy spectra (solid line) of the backscattering photons measured with various collimators and the energy spectra (dotted line) calculated using EGS4 code.
This equation can be rewritten as mtot ðcm1 Þ ¼ d 1 lnðI0 =IÞ
ð6Þ
from which mtot can be obtained from measured values of I0 ; I and d; and mtot ¼ mA þ mN
ð7Þ
where mA and mN are the atomic and the nuclear absorption coefficients, respectively. The parameter mA can be written as the sum over contributions from the principal photon interactions [9], mA ¼ mpe þ mcoh þ mincoh þ mpair þ mtrip
ð8Þ
where mpe is the absorption coefficient of the atomic photoeffect, mcoh and mincoh are the coherent and the incoherent (Compton), respectively, while mpair and mtrip are those for electron–positron production in the fields of the nucleus and of the atomic electrons, respectively. The mA can be calculated with the EGS4 code. The total photo-
nuclear cross-section per atom can be related to mN according to sN ¼ ðmN =rÞA 1:6605402 ðbarn=atomÞ
ð9Þ
where A is the atomic mass of the target material. We measured the absorption coefficients for 197 Au using the backscattered photons, using a 20 mm thick and 25 mm diameter 197 Au block in this experiment. The photon beam was collimated at 0:66 mrad; which covers the photon energy width of 6.5–17:6 MeV: Fig. 8 shows the measured transmitted photon energy spectra for the Au target and for the blank target. The simulated results calculated using the EGS4 code are also shown by solid lines in Fig. 8, while the normalization was carried out at 8:5 MeV near the threshold energy of the ðg; nÞ reaction measured previously [10]. The total absorption coefficients mtot can be derived from Eq. (6). The experimental results (mtot ) and the calculated photoatomic absorption
ARTICLE IN PRESS K. Aoki et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 228–236
235
Fig. 9. Experimental total absorption coefficients (mtot ) and the photoatomic absorption coefficients (mA ) calculated using EGS4 code. The error bars indicate the statistical errors. The dotted line represents the line smoothed by the least-square fit to the calculated absorption coefficients.
Fig. 8. Transmitted photon energy spectra (circles) for the blank target and the 197 Au target. The solid lines are the energy spectra calculated with the EGS4 code. The error bars indicate the statistical errors. The normalization was carried out at 8:5 MeV:
coefficients (mA ) using the EGS4 simulation code are shown in Fig. 9, while Fig. 9 shows the calculated absorption coefficients obtained from the calculated energy spectra. In Fig. 9, the dotted line represents the line smoothed by the leastsquare fits to the calculated absorption coefficients, because mA is well known to have smooth energy dependence; the nuclear absorption coefficient can be obtained from Eq. (7) by using that smoothed line. Fig. 10 shows the total photo-
Fig. 10. Total photonuclear absorption cross-section obtained for 197 Au in the E1 giant dipole resonance energy region and the dotted line is a Lorentz line fit. The error bars indicate the statistical errors.
nuclear absorption cross-section obtained for Au in the E1 giant dipole resonance energy region, whose maximum total cross-section value 197
ARTICLE IN PRESS K. Aoki et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 228–236
236
is 540 mb and the energy is 13:0 MeV: The large fluctuations of the obtained nuclear cross-sections are caused mainly by the measured total absorption coefficients derived from Eq. (6) because we used the smoothed atomic absorption coefficients (mA ) in Eq. (7). It is necessary to increase the transmitted photons from a thick target to improve the results. In the 17.0–17:5 MeV energy region, the cross-section varies also greatly because there are small differences in shapes of the energy spectra of the 197 Au target and the blank target. This may be caused by background subtraction. Under the assumption that there exists a complete overlap of the single levels in the giant dipole resonance, the absorption cross-section for medium and heavy nuclei can be approximated by the following Lorentz shape sðEÞ ¼ s0
ðE12
E 2 G21 E 2 Þ2 þ E 2 G21
ð10Þ
where s0 is the maximum value of the total crosssection for E ¼ E1 and G1 is an arbitrary parameter. We produced a Lorentz line that fits the absorption cross-sections of 197 Au: The dotted line in Fig. 10 shows the Lorentz line of the parameters (s0 ¼ 540 mb; E1 ¼ 13:0 MeV and G1 ¼ 4:75 MeV) for 197 Au: Our experimental results are in agreement with the line, although the E1 value is different from the obtained value from the photoneutron data of Veyssiere et al. [10].
5. Conclusions We have generated a high-energy photon beam using Compton backscattering of the laser beam on the relativistic electron beam in storage ring NewSUBARU. Measuring the backscattering photon’s energy spectra with an HPGe detector, we found that the maximum energy of the backscattered photons is 17:6 MeV; which is in agreement with the calculated value. The shapes of the measured energy spectra also show agreement with the shapes of the simulation calculations. From comparisons between the measured and the calculated energy spectra, the detected
photon yields are about 4 103 ; 2 103 and 3 102 s1 mA1 W1 for the 20, 10 and 2 mmf collimators, respectively. The values agree closely with the estimated ones. The produced photon beam was first applied to the measurement of the total photoabsorption cross-section of 197 Au in the giant resonance region by using the photon attenuation method. The results are consistent with the photoneutron cross-sections of the (g; xn) reaction measured previously [10], showing that the photon attenuation method will be a useful tool for studying photonuclear reactions.
Acknowledgements This work was supported by the Japan Society for the Promotion of Science under the program Grant-in-Aid for Science Research (C) and by the Himeji Institute Technology Promotion Foundation.
References [1] L. Federici, G. Giordano, G. Matone, G. Pasquariello, P.G. Picozza, R. Caloi, L. Casano, M.P. de Pascale, M. Mattioli, E. Poldi, C. Schaerf, M. Vanni, P. Pelfer, D. Prosperi, S. Frullani, B. Girolami, Nuovo Cimento 59B (1980) 247. [2] A. Ando, Y. Fukuda, S. Hashimoto, Nucl. Instr. and Meth. A 481 (2002) 43. [3] A.M. Sandorfi, et al., IEEE Trans. Nucl. Sci. NS-30 (1983) 3083. [4] J. Stepanek, Nucl. Instr. and Meth. A 412 (1998) 174. [5] I.F. Ginzburg, G.L. Kotkin, V.G. Serbo, I. Telnov, Nucl. Instr. and Meth. 205 (1983) 47. [6] W.R. Nelson, H. Hirayama, D.W.O. Rogers, the EGS4 code system, Stanford Linear Accelerator Center Report SLAC-265, Stanford University, 1985. [7] G.M. Gurevich, L.E. Lazareva, V.M. Mazur, S.Yu. Merkulov, G.V. Solodukhov, V.A. Tyutin, Nucl. Phys. A 351 (1981) 257. [8] H. Harada, Y. Sigetome, J. Nucl. Sci. Technol. 32 (1995) 1189. [9] J.H. Hubbell, Proceedings of the 51st National Meeting of the Japanese Society of Radiological Technology, Nagoya, Japan, 1995. [10] A. Veyssiere, H. Beil, R. Bergere, P. Carlos, A. Lepretre, Nucl. Phys. A 159 (1970) 561.
