H. G. Andresen

THE MAMI SOURCE OF POLARIZED ELECTRONS

The present work describes the source of polarized electrons that is run at the 855 MeV race track microtron MAMI at the Johannes Gutenberg Universitat in Mainz. The source is based on photoelectron emission from (III–V)-semiconductors. Presently strained layer InGaP- or GaAsP-cathodes are used, which are processed to negative electron affinity by coverage of the surface with a submonolayer of caesium and oxygen. Electron beams spin-polarized up to a degree of P = 55% at a quantum efficiency of QE = 2% or P = 75% at QE = 0.4%, respectively, are obtained. The well-known but hitherto unsolved problem of limited cathode lifetime has been sidestepped by the attachment of an UHV load lock system to the source electron gun. It allows quick replacement of cathodes without breaking the gun vacuum. Availabilities in excess of 85% are obtained regularly in beamtimes longer than 100h. The source was mainly applied in measurements of nucleon form factors via the reactions 1H(ē, e′ p), 2D(ē, e′ p), 2D(ē, e′ n), and 3He(ē, e′ n). More than 1600 h beamtime have been accomplished in physics experiments with polarized electron beams at MAMI up to now.

PICOSECOND POLARIZED ELECTRON BUNCHES FROM A STRAINED LAYER GAASP PHOTOCATHODE

Abstract At the Mainz electron accelerator MAMI a facility has been set up that allows the measurement of the bunchlength, the energy spread and the averaged and phase resolved polarization of picosecond electron bunches from a source of polarized electrons for MAMI. Electron bunches of typically 10 to 30 ps duration are generated by illuminating a strained layer GaAsP photocathode with femtosecond laser pulses.

Photoemission of spinpolarized electrons from strained GaAsP

Strained layer GaAs.95P.05 photo cathodes are presented, which emit electron beams spinpolarized to a degree of P = 75% typically. Quantum yields around QE = 0.4% are observed routinely. The figure of merit P2 × QE = 2.3 × 10−3 is comparable to that of the best strained layer cathodes reported in literature. The optimum wavelength of irradiating light around 830 nm is in convenient reach of Ti:sapphire lasers or diode lasers respectively. The cathodes are produced using MOCVD-techniques. A GaAs.55P.45-GaAs.85P.15 superlattice structure prevents the migration of dislocations from the substrate and bottom layers to the strained overlayer. The surface is protected by an arsenic layer so that no chemical cleaning is necessary before installation into vacuum. The source of polarized electrons attached to the Mainz race track microtron MAMI works with such cathodes now. More than 1000 hours beamtime have been performed successfully.

Hydrogen maser frequency shifts due to coherently excited Δ m F =±1 transitions between F =1 levels of the atomic hydrogen ground state

Hydrogen maser frequency shifts, caused by the multiple quantum transition nonlinearities of a resonant multiple frequency excitation of the atomic hydrogen four level ground state system have been investigated. The oscillation characteristics of hydrogen maser operation with simultaneously excited, low frequencyΔmF=±1 transitions between theF=1 states of the atomic hydrogen ground state have been analysed theoretically and explicit formulas for hydrogen maser frequency shifts and amplitude response have been derived for arbitrary maser oscillation amplitude and a small signal approximation for theΔmF=±1 “Zeeman” transitions. The comparison with experimentally observed hydrogen maser frequency shifts was specialized to small magnetic fields, for which the difference between the resonance frequencies of the two low frequency,ΔmF=±1 Zeeman transitions is small compared to the linewidth. Special emphasis was placed on the evaluation of frequency pulling effects for a Zeeman transition excitation at off-resonance conditions. For this case the theoretical formulation of frequency pulling effects becomes insensitive against simplifying assumptions about the radiation damping phenomena and a particular good agreement between experiment and theory can therefore be expected. Experimental conditions have been specified, for which the uncertainty of hydrogen maser frequency due to Zeeman transition induced frequency shifts does not restrict the present frequency stability of a hydrogen maser frequency standard.

Determination of the neutron electric formfactor in quasielastic collisions of polarized electrons with 3He and 2D. Collaboration A3 at MAMI

The determination of the neutron electric formfactor from quasielastic reactions 3H↘e(e↘,e′n) and D(e↘,e′,n↘) respectively is one of the present goals of experiments with polarized electrons at the Mainz race track microtron MAMI. A GaAsP‐photoelectron source is used at MAMI to get an 855 MeV electron beam spinpolarized to a degree of 35% at a current of 10 μA. Polarized 3He‐nuclei are produced by optical pumping metastable 3He. Scattered electrons are detected in coincidence with the recoil neutrons, the transverse spinpolarization of the neutrons may be analyzed by neutron‐proton scattering in a double wall plastic scintillator detector. A subset of the final detector set‐up has been tested successfully now by investigating the polarization transfer to the proton in reactions H(e↘,e′p↘) and D(e↘,e′p↘) and to the neutron in D(e↘,e′n↘) at a 4‐momentum transfer with −Q2=8fm−2. First data from the exclusive quasielastic collision 3H↘e(e↘,e′n) indicate a value of the neutron electric formfactor of GnE=0.035±...

Hydrogen maser atomic beam resonances

The resonance dynamics due to the simultaneous excitation of twoΔmF=±1 transitions within the three magnetic sublevels of theF=1 state in the atomic hydrogen ground state have been investigated for resonance excitation within the magnetic six pole state selected atomic beam of a hydrogen maser system. Both transitions were excited with equal strength by means of a single rf-excitation and by the Ramsey technique with two separated rf-fields. The associated changes in the atomic beam energy level population probabilities were measured by monitoring the hydrogen maser oscillation amplitude. Sufficiently strong excitations of bothΔmF=± 1 transitions result in a state selected, but unpolarized atomic beam, which offers the advantage to reduce considerably the sensitivity of hydrogen maser frequency on coherently excitedΔmF=±1 transitions within the maser interaction region. The theoretical analysis showed, that in contrast to other atomic beam preparation methods for hydrogen maser operation with unpolarized atoms, the saturation of atomic beams resonances influences only the maser amplitude but not the maser frequency.

Atomic beam preparation for hydrogen maser operation with unpolarized atoms

The conventional atomic beam preparation technique for hydrogen maser operation consists of a six pole magnetic field focussing in combination with adiabatically changing magnetic fields in the transition region between focusser and maser cavity, which results in a state selected, polarized ensemble of hydrogen atoms. Hydrogen maser operation with state selected, but unpolarized atoms has the advantage to reduce considerably the sensitivity of hydrogen maser frequency on coherently excited, low frequency,ΔmF=±1 Zeeman transitions. The influence of the magnetic field pattern in the transition region on the energy level population probabilities for the focussed hydrogen beam has been analysed. Atomic beam preparation techniques using non-adiabatically changing magnetic field configurations in the transition region are described. The dependence of hydrogen maser oscillation on the energy level population within the atomic beam is theoretically investigated and the results are used for the analysis of the experimentally achieved atomic beam populations. It has been shown that a pulsed magnetic spin guidance field in the transition region, which switches between adiabatically changing magnetic field configurations and a sudden magnetic field reversal, results in an ensemble of hydrogen atoms with vanishing time averaged polarization, which does not deteriorate the conventionally achieved maser oscillation amplitude.

First measurement of the electric formfactor of the neutron in the exclusive quasielastic scattering of polarized electrons from polarized 3He

A first measurement of the asymmetry in quasielastic scattering of longitudinally polarized electrons from a polarized 3He gas target in coincidence with the knocked out neutron is reported. This measurement was made feasible by the cw beam of the 855 meV Mainz Microtron MAMI. It allows a determination of the electric formfactor of the neutron GnE independent of binding effects to first order. At Q2=0.31 (GeV/c)2 two asymmetries A∥(S↘He∥q↘) and A⊥(S↘He⊥q↘) have been measured giving A∥=(−7.40±0.73%) and A⊥=(0.89±0.30)%. The ratio A⊥/A∥ is independent of the absolute value of the electron and target polarization and yields GnE=0.035±0.012±0.005.

Determination of the neutron electric form factor GEn in double polarized electron scattering from 3He

The neutron electric form factor GEn has been measured at the cw electron accelerator MAMI in the double polarized exclusive reactions 3He(e,e′n) and D(e,e′n) in quasi elastic kinematics with one common detector system. Experimental set-up and analysis of a 3He(e,e′n)-measurement in the range of momentum transfer Q2=0.27−0.5 (GeV/c)2 are discussed in this article. For events with low missing momentum (|Pm|⩽100 MeV/c) the PWIA analysis yields GEn=0.0352±0.0033±0.0024 with statistical and systematical error, corresponding to a quadratically added total error of 11.6% relativ. This result will be compared with preliminary results obtained from the D(e,e′n)-reaction and with data from elastic D(e,e′) scattering.

Emission of polarized ps-electron bunches from III–V semiconductor cathodes

At the pulsed electron gun testfacility at MAMI we have measured electron bunches from GaAs samples with different layer thicknesses. We observe an increasing dc-polarization with decreasing layer thickness. From the pulse measurements this behaviour may be explained in terms of bunchlength and spin relaxation time.