Spectroscopy of Magnetic Insulating Transition Metal Dihalides in High Magnetic Fields

1. I ntroducti on The transition metal dihalides have attracted considerable interest for many decades. Originally, attention was drawn to compounds such as iron dichlor­ ide because they represent archetypes of a variety of naterials known as metamagnets. These exhibit, at low temperatures, an antiferromagnetic state which can be ferromagnetically saturated in relatively weak external magnetic fields. This behaviour, first explained by LANDAU [1J, originates in the fact that the crystals have a layered structure in which the intralayer exchange coupling is strong and ferromagnetic, while that between layers is weak and anti ferromagnetic. In the last decade, it has been the quasi-two-dimensional behaviour due to the lamellar structure which renewed the interest in these compounds in connection with new theoretical and experimental developments on phase transitions and critical phenomena. A large set of references can be found in the review paper on magnetic model systems of DE JONGH and r~IEDEMA [2J. The magnetic properties have been studied in great detail by specific heat and susceptibility measurements, elastic and inelastic neutron scattering ex­ periments, magnetic resonance, Fourier transform spectroscopy in the far in­ frared, Raman light scattering experiments, etc ... Within the series of tran­ sition metal dihalides one can find examples of both easy-axis (FeC1 2) and easy-plane systems (CoC1 2, CoBr2, NiC1 2, NiBr2) and examples of large (FeC1 2, CoC1 2, CoBr2) and small (NiC1 2, NiBr2) single ion anisotropies. More recently, it was the appearance of new concepts in the problem of disorder which attracted attention to the alloys of the transition metal di­ halides which are rather simple and well-known systems. It turns out that mixed antiferromagnetic systems are ideal for testing theories of disorder­ ed materials. Dilute magnetic systems are now largely well understood [3J but concentrated systems, for which much less data are available, pose inter­ esting theoretical questions [4J. In general ,the magnetic excitations associated with impurities are closely connected with the magnitude of the magneti c i nteracti ons bet~ieen the impurity and the host spins. In the most usual case, when the magnetic properties of the impurity and host are similar, the presence of the impurities disturbs the host excitation spectrum and it is diffiCult to get experimental evid­ ence of physical features associated with the impurities. On the other hand, if the magnetic interactions between the impurity spin and the host are very different compared with those between the host spins themselves, it is poss­ ible to get well-defined localized impurity modes. Two cases may be observed.