Intense slow positron production at the 15 MeV LINAC at Argonne National Laboratory
Introduction
Positron annihilation spectroscopy coupled with a slow positron beam has been widely used in polymer, surface, atomic physics, and material science research [1], [2]. The demand for, and interest in, positron beams have been greatly increased in recent years for applications of slow positrons in materials and imaging research [3]. Due to the limitation of the intensity available from radioactive sources, the pair production positron source has formed the basis of a number of intense positron beams using a linear accelerator (LINAC) which have been constructed [3], [4], [5]. Most of these are electron LINAC based, where high-energy electrons bombard a high-Z material target, and Bremsstrahlung photons whose energies are greater than 1.022 MeV interact with the target nuclei and their energy is converted into e+e− pairs. Theoretically, there is no physical limit on the number of pairs produced for a LINAC energy on the order of a few MeV or above. However, the efficiency of conversion and thermal load in the converter may limit the positron yield. Recently, Perez and Rosowsky [6] indicated that a glancing incidence angle for electrons on a thin converter film will increase the positron yield for the same amount of energy deposited in the converter. In this paper, we report a series of computer simulations for a Ta converter with electrons incident at glancing angle by using the program EGS4 [7]. The aim was to optimize the geometry for slow positron production, and to investigate the positron energy and angle distribution, for the LINAC at the Argonne National Laboratory (ANL).
Section snippets
EGS4 simulation geometry
The simulation geometry is shown in Fig. 1. A Ta sheet converter is on the xy-plane and d is the Ta sheet thickness (the z direction). The incident electron is along the xz-plane with an angle γ, which is from the x-axis. The normal incident angle will thus be γ = 90°. The particles coming out from the Ta converter can be e+, e−, and photons. Their positions are determined by x, y, and z coordinates and their directions of motion are determined by the angles α, β, and θ, which are from the x, y,
Results and discussions
Most existing simulations have been made for normal incidence for particular particles and incident energies, depending on the accelerators’ specifications [8], [9], [10], [11]. For a fixed energy, the positron yield increases with the increase of converter thickness. When the thickness increases to a certain value, the positron yield begins to decrease because the positrons will annihilate inside the target. Therefore, there is an optimized thickness for the maximum positron yield for the
Proposed converter design
Because tungsten materials can be used both as a converter and a moderator and any distance between the two will reduce the amount of positrons available for moderation, we proceed to design both converter and moderator in one piece of tungsten. Fig. 6 shows a possible converter/moderator design using tungsten, for the 15 MeV LINAC at ANL.
Tungsten blinds (or vanes) have been used as a converter for many slow positron beams [3], [4], [5]. They can also be used as a moderator. For the possible
Conclusion
We have simulated the positron yield, energy, and angular distributions using the EGS4 program for a Ta converter, and the performance of a combined converter/moderator made of W vanes for different electron incident angles based on the 15 MeV LINAC at ANL. The results show that, if the converter thickness and converter/moderator configurations are chosen carefully, the yield of positron can be much improved. An intense slow positron beam producing 1010 slow positrons per second at the ANL LINAC
Acknowledgements
This research is supported by the National Science Foundation, Department of Energy, and Air Force Research Labs. We appreciate useful discussions with Profs. K.G. Lynn, A.P. Mills, C. Surko and Dr. P. Asoka-Kumar.
References (14)
- et al.
Nucl. Instrum. Methods A
(2004) - et al.
Nucl. Instrum. Methods B
(1993) - et al.
Nucl. Instrum. Methods A
(2000) - et al.
Rev. Mod. Phys.
(1988) - et al.
Appl. Phys.
(1987)