- Email: [email protected]
The KVI Lamb-shift polarimeter
ARTICLE IN PRESS
Nuclear Instruments and Methods in Physics Research A 516 (2004) 209–211
The KVI Lamb-shift polarimeter H.R. Kremersa,*, J.P.M. Beijersa, N. Kalantar-Nayestanakia, T.B. Cleggb,c a
Kernfysisch Versneller Instituut (KVI), Zernikelaan 25, 9747 AA Groningen, The Netherlands b Triangle Universities Laboratory (TUNL), NC, USA c Department of Physics and Astronomy, University of North Carolina, Chapel Hill, NC 27599-3255, USA Received 26 June 2003; accepted 8 July 2003
Abstract We have designed, built and implemented a compact polarimeter for spin-polarized proton and deuteron beams based on the Lamb-shift principle. We briefly describe some design considerations and present first results obtained with this polarimeter. r 2003 Elsevier B.V. All rights reserved. PACS: 07.60.F; 29.40.M; 25.45; 13.75.C Keywords: Polarization; Lamb-shift polarimeter; Proton beams; Deuteron beams
At KVI, we have operated for the past few years a polarized-ion source of the atomic beam type called POLIS. This ion source is used as an injector for the superconducting K ¼ 600 AGOR cyclotron and produces highly polarized proton or deuteron beams up to an energy of 35 keV [1]. Until recently the only method at KVI to determine the polarization degree was by means of in-beam polarimetry behind the cyclotron, i.e. on the accelerated beam. The KVI In-Beam Polarimeter (IBP) is based on elastic ~ p þ p or d~ þ p reactions and its construction and performance are described in Ref. [2]. Although the IBP works reliably and is routinely used, it does have a few disadvantages: (i) setting up and operating the IBP is time consuming, especially when performing absolute polarization measurements, (ii) expensive cyclotron time has to be used for optimizing and tuning POLIS, and (iii) the analyzing power of the *Corresponding author.
d þ p reaction is not known at all energies available from AGOR. Therefore, we decided to build a low-energy polarimeter to be installed directly behind POLIS. The new polarimeter should be able to determine the polarization degree of proton and deuteron beams within 1 min and with an accuracy better than 1%. The TUNL group has shown that these demands can be met with a so-called Lamb-Shift Polarimeter (LSP) [3,4]. In this letter we discuss the construction of an LSP based on their design and present some first results. The application of an LSP to the online determination of the polarization of a hydrogen target in a storage ring is discussed in Ref. [5]. For LSP theory the reader is referred to the relevant literature [6,7]. The LSP, of which a schematic diagram is shown in Fig. 1, basically consists of five parts: (i) a decelerating lens system, (ii) cesium oven, (iii) rf cavity, (iv) metastable-atom detection system and (v) a long solenoid [4]. Polarized ions entering the
0168-9002/$ - see front matter r 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.nima.2003.07.002
ARTICLE IN PRESS 210
H.R. Kremers et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 209–211
Fig. 1. Schematic layout of the Lamb-Shift Polarimeter (LSP): (1) decelerating lens system, (2) cesium oven, (3) rf cavity, (4) metastable-atom detection system and (5) the solenoid.
LSP from the left are neutralized by chargeexchange collisions with cesium atoms in an oven. The main difference with existing devices is that in our polarimeter the incoming ions are fast and consequently have a very small neutralization probability. Therefore, in order to get a good neutralization efficiency, the ions are decelerated by a two-element electrostatic lens system to E1 keV before entering the cesium oven [8]. We estimate that more than 20% of the incoming ions leave the Cs oven in the metastable 2s1=2 state. Cesium oven, rf cavity and detection system are all mounted inside a large 50 cm long solenoid with a diameter of 23 cm: The solenoid produces a very homogeneous, axially symmetric magnetic field which can be swept in a few seconds over the a–b– e resonances (i.e. from 52 to 62 mT; see below and Ref. [7]) while maintaining homogeneity. For the correct operation of the polarimeter it is mandatory that the magnetic field over the axial region beyond the cesium oven be strong enough to prevent the hyperfine interaction from depolarizing the nuclear spin. This condition is easily met in the magnetic-field range mentioned above. After neutralization, the atoms pass through a specially designed cylindrical rf resonator oscillating in the TM010 mode at 1609 MHz [9]. The rf
electric field component oscillating in the longitudinal direction induces transitions between the 2s1=2 a component (electron-spin up) and the short-lived 2p1=2 e component (also electron-spin up) for magnetic fields in the 52 and 62 mT range. At the same time a transverse static electrical field inside the resonator couples the 2s1=2 b component (electron-spin down) with the 2p1=2 e state. The net effect of the oscillating and static electric fields is that both the a and b metastable components are effectively quenched inside the resonator. However, when the magnetic field of the solenoid is exactly at the b–e crossing point, the metastable atoms survive the resonator and travel downstream to be detected in the metastable detector. This three-state a–b–e resonance occurs at different magnetic field values for hydrogen atoms with nuclear-spin up or down, i.e. at 53:5 mT for proton-spin up and at 60:5 mT for proton-spin down. Deuterium atoms show three resonance peaks corresponding to the three orientations of the deuteron spin, i.e. at 56:5 mT for mI ¼ 1; 57:5 mT for mI ¼ 0 and at 58:5 mT for mI ¼ 1: Metastable atoms surviving the spin filter are detected downstream by quenching them in a strong electric field and measuring the resulting Lyman-a photons with a photomultiplier.
ARTICLE IN PRESS H.R. Kremers et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 209–211
211
during transport and acceleration of the beam. The reason for the depolarization is still under study. Since its completion, the LSP is used routinely for tuning and optimizing the polarized-ion source POLIS. It has met all its design specifications and shown its usefulness in measuring the polarization of proton and deuteron beams.
Fig. 2. LSP spectra for protons (top panel) and deuterons (bottom panel).
By scanning the magnetic field over the resonance peaks we immediately obtain the nuclear vector polarization for hydrogen and the nuclear vector and tensor polarizations for deuterium. Typical examples of such spectra are shown in Fig. 2 for polarized proton and deuteron beams, respectively. These polarization measurements take E60 s and have a statistical error of less than 1%. Surprisingly, the measured polarization degrees with the LSP are consistently 10–15% higher than the IBP values. Up to now we could not find any sources of systematic error which indicates that some depolarization is taking place
We are grateful to the TUNL group for kindly sharing with us their knowledge and experience on Lamb-Shift Polarimeters. Their hospitality during the stay of R.K. and N.K. is very much appreciated. This work is part of the research programme of the ‘‘Stichting voor Fundamenteel Onderzoek der Materie’’ (FOM) with financial support from the ‘‘Nederlandse Organisatie voor Wetenschappelijk Onderzoek’’ (NWO).
References [1] H.R. Kremers, A.G. Drentje, in: R.J. Holt, M.A. Miller (Eds.), AIP Conference Proceedings, Vol. 421, Workshop Polarized Gas Targets and Polarized Beams, Urbana, 1997, p. 507. [2] R. Bieber, et al., Nucl. Instr. and Meth. A 457 (2001) 12. [3] S.K. Lemieux, et al., Nucl. Instr. and Meth. A 333 (1993) 434. [4] A.J. Mendez, et al., Rev. Sci. Instrum. 67 (1996) 3073. [5] R. Engels, et al., Rev. Sci. Instrum. 74 (2003) 4607. [6] W.E. Lamb Jr., R.C. Retherford, Phys. Rev. 81 (1951) 222. [7] G.O. Ohlsen, J.L. McKibben, Los Alamos Scientific Laboratory Report, LA-3725, 1967. [8] P. Pradel, et al., Phys. Rev. A 10 (1974) 797. [9] J.L. McKibben, et al., Phys. Rev. Lett. 20 (1968) 1180.
Nuclear Instruments and Methods in Physics Research A 516 (2004) 209–211
The KVI Lamb-shift polarimeter H.R. Kremersa,*, J.P.M. Beijersa, N. Kalantar-Nayestanakia, T.B. Cleggb,c a
Kernfysisch Versneller Instituut (KVI), Zernikelaan 25, 9747 AA Groningen, The Netherlands b Triangle Universities Laboratory (TUNL), NC, USA c Department of Physics and Astronomy, University of North Carolina, Chapel Hill, NC 27599-3255, USA Received 26 June 2003; accepted 8 July 2003
Abstract We have designed, built and implemented a compact polarimeter for spin-polarized proton and deuteron beams based on the Lamb-shift principle. We briefly describe some design considerations and present first results obtained with this polarimeter. r 2003 Elsevier B.V. All rights reserved. PACS: 07.60.F; 29.40.M; 25.45; 13.75.C Keywords: Polarization; Lamb-shift polarimeter; Proton beams; Deuteron beams
At KVI, we have operated for the past few years a polarized-ion source of the atomic beam type called POLIS. This ion source is used as an injector for the superconducting K ¼ 600 AGOR cyclotron and produces highly polarized proton or deuteron beams up to an energy of 35 keV [1]. Until recently the only method at KVI to determine the polarization degree was by means of in-beam polarimetry behind the cyclotron, i.e. on the accelerated beam. The KVI In-Beam Polarimeter (IBP) is based on elastic ~ p þ p or d~ þ p reactions and its construction and performance are described in Ref. [2]. Although the IBP works reliably and is routinely used, it does have a few disadvantages: (i) setting up and operating the IBP is time consuming, especially when performing absolute polarization measurements, (ii) expensive cyclotron time has to be used for optimizing and tuning POLIS, and (iii) the analyzing power of the *Corresponding author.
d þ p reaction is not known at all energies available from AGOR. Therefore, we decided to build a low-energy polarimeter to be installed directly behind POLIS. The new polarimeter should be able to determine the polarization degree of proton and deuteron beams within 1 min and with an accuracy better than 1%. The TUNL group has shown that these demands can be met with a so-called Lamb-Shift Polarimeter (LSP) [3,4]. In this letter we discuss the construction of an LSP based on their design and present some first results. The application of an LSP to the online determination of the polarization of a hydrogen target in a storage ring is discussed in Ref. [5]. For LSP theory the reader is referred to the relevant literature [6,7]. The LSP, of which a schematic diagram is shown in Fig. 1, basically consists of five parts: (i) a decelerating lens system, (ii) cesium oven, (iii) rf cavity, (iv) metastable-atom detection system and (v) a long solenoid [4]. Polarized ions entering the
0168-9002/$ - see front matter r 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.nima.2003.07.002
ARTICLE IN PRESS 210
H.R. Kremers et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 209–211
Fig. 1. Schematic layout of the Lamb-Shift Polarimeter (LSP): (1) decelerating lens system, (2) cesium oven, (3) rf cavity, (4) metastable-atom detection system and (5) the solenoid.
LSP from the left are neutralized by chargeexchange collisions with cesium atoms in an oven. The main difference with existing devices is that in our polarimeter the incoming ions are fast and consequently have a very small neutralization probability. Therefore, in order to get a good neutralization efficiency, the ions are decelerated by a two-element electrostatic lens system to E1 keV before entering the cesium oven [8]. We estimate that more than 20% of the incoming ions leave the Cs oven in the metastable 2s1=2 state. Cesium oven, rf cavity and detection system are all mounted inside a large 50 cm long solenoid with a diameter of 23 cm: The solenoid produces a very homogeneous, axially symmetric magnetic field which can be swept in a few seconds over the a–b– e resonances (i.e. from 52 to 62 mT; see below and Ref. [7]) while maintaining homogeneity. For the correct operation of the polarimeter it is mandatory that the magnetic field over the axial region beyond the cesium oven be strong enough to prevent the hyperfine interaction from depolarizing the nuclear spin. This condition is easily met in the magnetic-field range mentioned above. After neutralization, the atoms pass through a specially designed cylindrical rf resonator oscillating in the TM010 mode at 1609 MHz [9]. The rf
electric field component oscillating in the longitudinal direction induces transitions between the 2s1=2 a component (electron-spin up) and the short-lived 2p1=2 e component (also electron-spin up) for magnetic fields in the 52 and 62 mT range. At the same time a transverse static electrical field inside the resonator couples the 2s1=2 b component (electron-spin down) with the 2p1=2 e state. The net effect of the oscillating and static electric fields is that both the a and b metastable components are effectively quenched inside the resonator. However, when the magnetic field of the solenoid is exactly at the b–e crossing point, the metastable atoms survive the resonator and travel downstream to be detected in the metastable detector. This three-state a–b–e resonance occurs at different magnetic field values for hydrogen atoms with nuclear-spin up or down, i.e. at 53:5 mT for proton-spin up and at 60:5 mT for proton-spin down. Deuterium atoms show three resonance peaks corresponding to the three orientations of the deuteron spin, i.e. at 56:5 mT for mI ¼ 1; 57:5 mT for mI ¼ 0 and at 58:5 mT for mI ¼ 1: Metastable atoms surviving the spin filter are detected downstream by quenching them in a strong electric field and measuring the resulting Lyman-a photons with a photomultiplier.
ARTICLE IN PRESS H.R. Kremers et al. / Nuclear Instruments and Methods in Physics Research A 516 (2004) 209–211
211
during transport and acceleration of the beam. The reason for the depolarization is still under study. Since its completion, the LSP is used routinely for tuning and optimizing the polarized-ion source POLIS. It has met all its design specifications and shown its usefulness in measuring the polarization of proton and deuteron beams.
Fig. 2. LSP spectra for protons (top panel) and deuterons (bottom panel).
By scanning the magnetic field over the resonance peaks we immediately obtain the nuclear vector polarization for hydrogen and the nuclear vector and tensor polarizations for deuterium. Typical examples of such spectra are shown in Fig. 2 for polarized proton and deuteron beams, respectively. These polarization measurements take E60 s and have a statistical error of less than 1%. Surprisingly, the measured polarization degrees with the LSP are consistently 10–15% higher than the IBP values. Up to now we could not find any sources of systematic error which indicates that some depolarization is taking place
We are grateful to the TUNL group for kindly sharing with us their knowledge and experience on Lamb-Shift Polarimeters. Their hospitality during the stay of R.K. and N.K. is very much appreciated. This work is part of the research programme of the ‘‘Stichting voor Fundamenteel Onderzoek der Materie’’ (FOM) with financial support from the ‘‘Nederlandse Organisatie voor Wetenschappelijk Onderzoek’’ (NWO).
References [1] H.R. Kremers, A.G. Drentje, in: R.J. Holt, M.A. Miller (Eds.), AIP Conference Proceedings, Vol. 421, Workshop Polarized Gas Targets and Polarized Beams, Urbana, 1997, p. 507. [2] R. Bieber, et al., Nucl. Instr. and Meth. A 457 (2001) 12. [3] S.K. Lemieux, et al., Nucl. Instr. and Meth. A 333 (1993) 434. [4] A.J. Mendez, et al., Rev. Sci. Instrum. 67 (1996) 3073. [5] R. Engels, et al., Rev. Sci. Instrum. 74 (2003) 4607. [6] W.E. Lamb Jr., R.C. Retherford, Phys. Rev. 81 (1951) 222. [7] G.O. Ohlsen, J.L. McKibben, Los Alamos Scientific Laboratory Report, LA-3725, 1967. [8] P. Pradel, et al., Phys. Rev. A 10 (1974) 797. [9] J.L. McKibben, et al., Phys. Rev. Lett. 20 (1968) 1180.