Rotation of the polarization plane of obliquely incident light reflected from a polished ferromagnetic surface is successfully employed to reveal the antiparallel domains lying in the surface of a large single crystal of silicon iron. This Kerr magneto-optic effect is the basis of two methods of domain observation which are described in detail: (1) the optical probe method in which a small, focused beam of plane polarized light scans the crystal surface before being reflected into a photomultiplier tube located behind the Nicol prism analyzer, and (2) the photographic method in which the crystal surface is broadly illuminated with polarized light and the reflected beam is photographically recorded after passage through the analyzer. The truly saturated character of the domains is demonstrated, and the directions of the domains are assigned without ambiguity. Domain wall movement and Barkhausen discontinuities are revealed in photographs taken as the crystal is subjected to a changing magnetizing field. Possible applications and limitations of the method are discussed. DOMAINS BY LONGITUDINALKERR EFFECT 55

U SE of the normal Kerr magneto-optic effect to observe domain patterns in ferromagnetic substances having free surface poles has been described by Williams, Foster, and Wood. 1 Since plane polarized light reflected normally from a polished magnetic surface is not affected by magnetization in the plane of the surface, oblique reflection using the longitudinal Kerr magneto-optic effect 2 has been investigated as a means of observing domains in a single crystal of silicon iron where the magnetization is in the plane of the crystal surface. The plane of polarization of the reflected light suffers a positive or negative rotation depending upon the direction of magnetization. For magnetic saturation this rotation amounts to about 4 minutes at an angle of incidence of 60°. Ambiguity with the transverse Kerr effect is avoided by polarizing the light at right angles to the plane of incidence, for which direction the transverse Kerr effect is zero. 3 Two methods have been used to observe domain structure on the (100) surface of the silicon iron crystal. Observations of Kerr effects over local portions of the surface have been made by illuminating the surface with a small optical probe, passing the reflected beam through a nicol analyzer set about 2° from extinc-DISTANCE-MMS FIG. 1. Percentage increase in transmitted light with magnetization of silicon iron crystal along a line through the center of the crystal in the [001 ] direction. The broken line indicates the domain configuration suggested and substantiated in (b) of Fig. 2.tion, and measuring the intensity of the transmitted light with a sensitive multiplier phototube circuit. A traverse of the crystal in an unmagnetized state with a 0.5-mm probe of light, when compared with a similar traverse after application of a small external field, indicates those regions in which there has been a change in surface magnetization. Figure 1 is the result of such analysis for traverses in the [001] direction across the central portion of the (100) surface.Since the contrast between regions magnetized in antiparallel directions is about 10 percent, photographic observations are possible with a suitable optical system. Figure 2 shows oblique photographs of the crystal with the same domain arrangement as that scanned photoelectrically. (a) and (b) differ 180° in the position of the crystal, resulting in the reversal of the relative intensities of adjacent domains, (c) is the result of magnetizing the crystal along the direction of the original domains, namely the [010] direction. This is the long dimension of the Crystal although it appears otherwise because of the angle at which the picture was taken. While the crystal was electro-polished to a mirror surface, oblique photography and the need for increasing the contrast in development and printing have magnified surface imperfections and resulted in its rough appearance.It is interesting to note that a particular domain pattern is generally quite stable and persists indefinitely under normal conditions of handling. Magnetizatio...

No abstract

LETTERS TO X=0.78 mm FIG. 2. Cathode-ray display of two of the three hyperfine components of the 7 = 0->1 transition of HI at 0.778-mm wavelength. The line to the left is the F = 5/2 -• > 7/2 component, and the one to the right is the F = 5/2 -> 3/2 component. These two components observed on the same scope trace are 163.3 Mc/sec apart. The klystron mode would not cover all three components in a single sweep.

No abstract