Using a transversely polarized electron beam a relativistic (e, 2e) experiment has been performed to look for a spin up-down asymmetry in the electron-impact ionization process caused by the spinorbit interaction of the continuum electrons in the Coulomb field of the atomic nucleus. An incident energy of 300 keV, coplanar asymmetric kinematics, and the It shell of silver (Z = 47) have been used. We found a distinct spin asymmetry in the recoil peak (up to 16'), whereas in the binary peak the asymmetry is close to zero. This feature is confirmed by theoretical calculations, but quantitatively the agreement is poor. PACS numbers: 34.80.Dp, 34.80.Nz The study of the dynamics of electron-impact ionization of atoms by electron-electron coincidence [or (e, 2e)] experiments has proved to be a valuable method of investigating the physics of ionization processes [1,2]. The application of spin-polarized electron beams allows for an even more severe test of theory, and recently results have been reported for low energies. In these nonrelativistic (e, 2e) experiments spin-dependent asymmetries have been studied which are due to the exchange interaction [3] and to the so-called fine-structure effect [4,5].In this Letter we report on a relativistic (e, 2e) experiment with a transversely polarized electron beam designed to look for a spin asymmetry caused by the spin-orbit interaction of the continuum electrons in the Coulomb field of the atomic nucleus. The spin-orbit interaction arises from the interaction of the magnetic moment of the electrons with the magnetic field felt in the rest frame of the electrons because of their motion in the Coulomb field of the target nucleus (Mott scattering). As a result, a spin up-down asymmetry is to be expected in the triply differential cross section of electron-impact ionization. In particular, we were interested in investigating the angular distribution in view of the question whether the asymmetry is larger within the region of the so-called recoil peak than the asymmetry within the binary peak. Our assumption is based on the following intuitive argument. As the binary peak has a large contribution from a direct binary collision between the incoming electron and the atomic electron with the nucleus in the role of a "spectator, " the spin-orbit interaction will be weak. The recoil peak, however, cannot be explained unless an electron-nucleus interaction is taken into account. Consequently, here a spin-orbit interaction must contribute, and a spin asymmetry is to be expected.A sketch of the experimental arrangement is shown in Fig. 1. The source for the polarized electron beam (described in detail elsewhere [6]) used the photoemission of electrons from a GaAsP crystal irradiated by circularly polarized light of a helium-neon laser. After being deflected by a 90 cylindrical deflector, the extracted electrons are transversely polarized. The spin Hip of the electron beam can be easily realized by reversing the helicity of the laser light. The source is installed in a high voltage termina...

We report here on an (e, 2e) measurement at relativistic electron energy (300 keV) on the K shell of copper (E bind = 9 keV). Using coplanar asymmetric kinematics the angular distribution of the relative triply differential cross section was measured and is compared with a calculation using relativistic distorted-wave Born approximation. We find a distinct structure in the recoil region which is in reasonable agreement with the theoretical data.

No abstract

No abstract