Ferromagnetism

FERROMAGNETISM Comparatively few substances have magnetic properties similar to those of iron. Among these are nickel and cobalt, alloys of these ferromagnetic metals, some compounds of iron, and some compounds and alloys of manganese. Ferromagnetics are char acterized by the high value of the magnetization which may be acquired even in small mag netic fields, and by the fact that in general they may be "permanently" magnetized. In detail, the magnetic properties are of a most varied character, and they are markedly influ enced by mechanical and ther mal treatment. In the field of ferromagnetism there is an enor mous range of phenomena of scientific interest. There is an additional incentive to inves tigation, in that, from a purely technical point of view, a knowl edge of the magnetic char acteristics of the ferromagnetic materials used in the construction of electrical machinery and instruments is of the utmost importance. It is not therefore sur prising that extensive observational data, whose interpretation is by no means clear, have been accumulated.. As yet there is no theory of ferromagnetism which is not open to serious objections, but, amid the welter of empirical details, a theory which cor relates some of the main facts is of great value, even if the theory does not adequately explain them. The theory of ferro magnetism which has proved most successful, and which will be considered here, is due, in essentials, to P. Weiss (1907). To Ewing (189o) is due the credit of realizing the important part played by the mutual magnetic action of the elementary molecular magnets, to which hysteresis effects were related; his theory has been of great service in giving a crude picture of what might be supposed to occur, in co-ordinating a number of observations and explaining them qualitatively, but quantitatively it breaks down. Mutual action of the magnetic "carriers" certainly plays an important part in the theory of Weiss, but the general treat ment is entirely different.

The Theory of Ferromagnetism.

Weiss's treatment of paramagnetism on the basis of the "molecular field" theory has already been considered. It is assumed that, in addition to the external field acting on the magnetic "carriers" (a convenient general term for the molecules, atoms or ions which constitute the elementary magnets), there is also a molecular field pro portional to the magnetization acquired, so that, for the resultant field, Below the critical temperature the substance will be sponta neously magnetized, to an extent which increases with decreasing temperature, reaching saturation in the absolute Langevin sense, only when the temperature is indefinitely reduced. The way in which varies with T/O may be calculated, or found graphi cally, in the way outlined above. The curve giving the variation is shown in fig. 34.

Weiss's theory of ferromagnetism rests on precisely the same assumption as his generalized theory of there is a molecular field proportional to the intensity of mag netization. The development of this assumption leads to the

conclusion that above a certain critical temperature substances may behave as normal paramag netics following a Weiss law (X = as to the depend T — 0 ence of susceptibility on tempera ture; and below it that they may have a "spontaneous magnetiza tion" increasing with decreasing temperature. The relation may be written the function f being that shown in the curve of fig. 34. If the ratio of the spontaneous magnetization to the saturation intensity (acquired at absolute zero) is plotted against the ratio of the temperature to the critical Curie temperature, the curve obtained should be the same for all ferromagnetics.

It is a matter of common observation that a piece of iron, for example, may exist normally in an apparently stable unmagnetized state, so that the spontaneous magnetization cannot be supposed necessarily to extend uniformly throughout a large mass of fer romagnetic material. The spontaneous magnetization can only be supposed to be unidirectional throughout small "domains"; these domains, however, cannot be definitely identified with the ele mentary constituent micro-crystals of the solid. It is supposed that each of the domains is spontaneously magnetized to the de gree appropriate to the temperature, but that the directions of magnetization may be uniformly distributed so that the body as a whole appears unmagnetized. The problem then arises as to what will occur when a magnetic field is applied. Weiss assumes that each of the "domains" will behave in a somewhat similar manner to the elementary crystals of the ferromagnetic mineral pyrrhotite (of approximate composition FeS), which occurs as crystals in the form of hexagonal prisms. The magnetic proper ties of the natural crystals may be accounted for by supposing that they are built up of smaller e l e m en t a r y crystals whose behaviour is particu larly simple. They have a "magnetic plane" (parallel to the base of the hexagonal prism), in which there is a direction of easy magnetization. In the absence of a field the mag netization acquires a saturation value in this direction ; applica tion of the field in the direction of the magnetization does not increase its intensity; if the field is reversed the magnetization remains constant until the coercive field (about 15 gauss) is passed, when the intensity of magnetization changes in sign but not in magnitude. The hysteresis "curves" for a field in the direc tion of easy magnetization are thus simple rectangles, saturation being attained in a field of about 15 gauss. (fig. 35) With a field in the magnetic plane at right angles to this direc tion, saturation is only attained with a field of 7,300 gauss. At