New measurements of the magnetisation of a CuMn spin glass (0.25 at.%) are presented in order to test the hypothesis that the field-cooled magnetisation M(H,T) of a spin glass can be treated by the relationships of equilibrium thermodynamics. Comparisons are made between direct measurements of ( delta M/ delta T) H and ( delta 2 M/ delta T 2 ) H and those deduced from recent heat capacity and magnetocaloric experiments performed on similar specimens. The results show that above 0.05 T all three experiments agree closely. At low fields, however, there are small but significant differences that most probably arise from the use of different specimens for the three experiments but might have different causes. It is concluded that the equilibrium hypothesis is consistent with the experimental data at present available but that further experiments at low fields are required to test it thoroughly. The hypothesis has the remarkable implication that just below the glass temperature the entropy of a field-cooled spin glass increases with field in low fields and goes through a maximum as the field is increased (at 0.05 T for our specimen). An analogy is drawn between this behaviour and that of an ordered antiferromagnet for which the entropy is also an increasing function of field.
No takes yet. Share an insight, caveat, or question.
Guy et al. (1983) studied this question.