The spontaneous magnetization of alloys I. Copper nickel alloys

Abstract
The spontaneous magnetization is of fundamental importance in the theory of ferromagnetism but only meagre experimental data are at present available. The paper describes the initial stage of a project to determine the temperature variation of spontaneous magnetization for a number of ferromagnetics. It is impossible to deduce the spontaneous magnetization unequivocally over the complete temperature range from purely magnetic measurements so that a determination of the magnetocaloric effect must also be made. An alloy from the system copper-nickel (23.9 atomic % copper, rest nickel) was chosen for the first investigation because the system occupies an important position in the electron theory of ferromagnetism, whilst it also has the additional advantage of freedom from metallurgical complications. The experimental work is in two parts. An apparatus was constructed which enables the temperature rise in the magnetocaloric effect (ΔT) to be measured accurately over a wider temperature range than hitherto. This incorporates a galvanometer amplifier and the overall sensitivity is 1000 mm/°C. The specific intensity of magnetization (σ) was obtained by using the balance devised by Sucksmith (1939). Both the sets of measurements were made at closely spaced intervals up to the Curie point and, in the case of the magnetocaloric effect, considerably above this. In addition, the susceptibility above the Curie point was measured from 170 to 775° C and was found to obey the Curie-Weiss law x = C/(T — θ) over most of this range with C = 3.90 x 10-3 and θ = 111° C. The three methods of deducing the spontaneous magnetization from observations are critically discussed in general, and then in relation to the particular results obtained, which are similar to those obtained for other materials except that the dependence of the temperature rise on σ2 above the Curie point is not linear. This is a new feature which it is considered may have important theoretical consequences. The values of the spontaneous magnetization deduced from purely magnetic measurements are in agreement, but these differ from the values obtained by using the magnetocaloric effect. The spontaneous magnetization-temperature curve deduced from the magnetocaloric effect lies below that obtained from magnetic measurements in the immediate neighbourhood of the Curie point, and is notable for the absence of any ‘ tail '. The departure of the reduced σ against T curve from the law of corresponding states is firmly established by the present work in which the purely magnetic measurements are in fair agreement with earlier investigations.
Keywords