No abstract

The gyromagnetic ratio of the free electron is measured by a method which is an extension of the classical • double-scattering experiment. A magnetic field is interposed between the first and second scattering foils, whose direction is parallel to the path followed by the electrons. The electron spins precess in the magnetic field, resulting in a rotation of the plane of maximum asymmetry, as observed after the second scattering event. In the experiment reported, the rotation is approximately 1800 degrees. In the motion of the electron between the two scatterers the small lateral component of velocity gives rise to a "cyclotron'' motion whose frequency is, theoretically, the same as the spin precession frequency to within about one part in a thousand. The cyclotron motion, therefore, furnishes a convenient reference frequency, but it also introduces problems in that it causes the asymmetries which have their origin in geometrical misalignment, finite aperture, etc., to follow the rotation of the spin asymmetry. By comparing all measurements made with the foils of high atomic number with measurements made with an aluminum foil of equal scattering power, and by further precautionary procedures and cross checks, the spin asymmetry is separated from asymmetries of other origin. The result, for 420-kev electrons and gold scatterers, is g=2.00db0.01. Plans for a more precise measurement are mentioned.

Betatron oscillations in the University of Michigan synchrotron have been investigated by exciting these oscillations with a transverse electric field. The results indicate that the betatron oscillation is composed of a group of component frequencies which are separated by the frequency of synchrotron phase oscillation. The suggested explanation of the observed splitting is the modulation of the betatron oscillation frequency by the synchroton oscillations.

100-kev electrons in 0.1-yusec bunches are sent into a gold foil. The part of the electron bunch which is scattered at right angles, and which, consequently, is partially polarized, is trapped in a magnetic field and held for a measured length of time (up to 300 jusec). The bunch is then released from the trap and allowed to strike a second gold foil. Counters receive the electrons scattered at plus and minus approximately 90°. The cycle is repeated 1000 times per sec. The asymmetry in intensity in the two directions depends upon the final direction of polarization. A plot of the intensity asymmetry vs trapping duration is a cosine curve, whose frequency is the difference between the orbital frequency and the spin precession frequency. This is related to the g factor as follows:coDnioc/Be-a, where g is 2(l+#). Thus the "anomaly/' a, is measured directly. The determination of B presents some difficulty because the field must be slightly nonuniform in order to trap the electrons. The spatial variation in B from the center of the trap to the ends is only 0.3%, and the time average of B which applies to the trapped electrons is evaluated to 0.1%. Measurements made at other electron energies, down to 50 kev, showed a slight dependence of a upon energy. The dependence is attributed to electrostatic charges on the surfaces in the trapping region. The final standard error quoted is, however, purposely made large enough so that the variation of a with energy is bracketted. The result is a=0.0011609±0.0000024.

The Mott asymmetry was observed at 121 kev for gold targets and scattering angles of 0i = 9O° and 02 = 80° to 140°. The cosine dependence of the asymmetry on the azimuthal angle was shown in all cases. A weak magnetic lens (maximum field: 12.5 gauss) was used between the two scatterers, and a low-resolution (8%) electrostatic energy analyzer was used after second scattering. The measured asymmetry amplitudes are compared with the calculations of Sherman (pure Coulomb field) and of Mohr and Tassie (screened Coulomb field). The experimental values for 0 2 = 9O°, 100°, 110° agree well with either theoretical curve but are not accurate enough to distinguish between them. The measured value at 0 2 = 8O° is enough higher (17%) than the pure Coulomb field curve to indicate a screening eifect. The experimental values for 0 2 = 120°, 130°, and 140° fall from 15% to 20% below the theoretical curves. This discrepancy is attributed to plural scattering. Measurements made using the magnetic lens but not the energy analyzer yield asymmetries reduced in amplitude about 33% and shifted in phase by 26° when compared with the measurements discussed above. This is believed due to the presence of an appreciable number of polarized electrons at a considerably lower energy than 121 kev in the first scattered beam.