HYSTERESIS MEASUREMENTS ON RCo5 MICRO-PARTICLES

Magnetic hysteresis loops measured on isolated single particles of a few microns size are shown for RCos compounds with R = Sm, Pr and La, respectively. The critical fields at which walls nucleate and propagate are measured separately, and the influence of ageing is observed. Pinning of domain walls seems to be significant in determining the coercive force. The magnetization process of RCO, compounds (R for rare-earth metal) takes place at field strengths far below the anisotropy field, which indicates that it must be interpreted in terms of nucleation and displacement of Bloch walls. The coercivity depends largely on the field strength H,, at which a reverse domain is nucleated and on the field strength H,, at which an already existing domain wall is propagated. It is interesting to know whether H,, or H, is determining coercivity in SmCo,, how these critical fields are affected by ageing and how they behave in other RCO, compounds. A wall-nucleation model devised by Becker [I] considers the nucleation field H,, as mainly determining the coercivity. However, other measurements make it likely that in certain cases wall pinning can be predominant, leading to a wall-displacement model [2, 31. These discussions are hampered by the fact that hysteresis measurements are made on samples that contain a large number of crystals giving the average of many wall processes. The nucleation and displacement of single walls cannot be observed individually and manifest themselves merely as noise. However, if measurements are made on samples small enough to display one elementary process at a time, more information on the magnetization process might be gathered. Such samples should have a size of a few microns. Hysteresis measurements on these samples are possible with a new type of magnetometer, the ({Vibrating Reed Magnetometer >> 141. We have used this instrument to observe wall processes in a single particle taken from a SmCo, powder having a coercivity of 14 kOe. The hysteresis loop measured along the easy axis of the particle is shown in figure la. The powder was ground in a vibration mill with steel balls under hydrogen. After grinding it was handled under exclusion of air until a few hours before the measurement. The measurement starts at 21.5 kOe, this being the highest field we could make in the magnetometer. The magnetization remains constant until at about zero field a small jump occurs. As this jump is reversible on this field scale it is ascribed to the magnetization reversal of some small amount of iron possibly abraded from the steel balls. It will be ignored in the following discussion. At 11.4 kOe a wall is nucleated, which jumps through a large part of the crystal until it becomes FIG. 1. Hysteresis loops (magnetic moment rn versus applied field H ) of a single SmCos particle of about 5 micron size, measured along the easy axis. a) immediately after grinding the powder ; b) after 12 days exposure to air at room temperature ; c) after 30 minutes at 100 OC in air ; d) after 17 3 hours at 100 OC in air ; e) after 600 hours at 100 OC in air. The vertical scale is in arbitrary units. All figures have the same field scale. stuck somewhere. Upon further increase of the counterfield it creeps slightly until at 14.8 kOe another jump occurs after which the magnetization of about threequarters of the crystal volume is reversed. By creep is meant a continuous but irreversible displacement of the wall due to field variation. No time effects are observed. The remaining part of the magnetic moment is reversed by a creep process, which we have found in all hysteresis loops measured on RCo5 single particles up to now. Evidently the wall is reluctant to leave the crystal. As this portion of the loop is irreversible there must be some concentration of pinning sites preventing the wall from crossing the surface layer. This is in Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19711372