OBSERVATION OF ANTIFERROMAGNETISM BY NEUTRON DIFFRACTION METHODS

The two necessary conditions for the existence of ferromagnetism are: 1) the atoms of the ferromagnet must possess a magnetic moment due to an unfilled electronic \(d\)- or \(f\)-shell; 2) the exchange integral associated with the exchange of electrons between two neighboring atoms must be positive. When these two conditions are fulfilled, the stable state is a parallel arrangement of the magnetic moments of the atoms in certain regions (domains) in the absence of an external magnetic field.

Some compounds of the transition elements possess an interesting magnetic property: for them the stable state is an antiparallel arrangement of neighboring atomic magnetic moments. In this case condition (1) is satisfied, but the exchange integral for two nearest neighbors is negative. Such substances have been called antiferromagnets. At \(0^\circ\mathrm{K}\), every atomic magnetic moment of such a substance is surrounded by oppositely directed magnetic moments

moments. As the temperature is raised, this order begins to be destroyed, and, upon reaching a certain temperature (the Curie temperature), thermal motion completely destroys the regions of spontaneous antiferromagnetism and order remains only in very small regions, between which there is no correlation; above the Curie temperature the substance exhibits typical paramagnetic properties. At the Curie temperature anomalies of certain properties of antiferromagnets (for example, their heat capacity) must be observed, since this is the temperature of a second-order phase transition; the presence of remnants of order in very small regions—short-range order—makes the anomalies diffuse. Experiments have shown the absence of any structural changes in this temperature interval; therefore the anomalous behavior of the properties of antiferromagnets could not be explained by anything else. It was by the discovery of these anomalies that the presence of antiferromagnetism was formerly judged, and this method was the only one. Naturally, antiferromagnetism cannot be detected by magnetic methods, for the magnetic moments of the atoms completely compensate one another. Neutronography makes possible a direct method for recording the presence of antiferromagnetism.

In antiferromagnets below the Curie point a rigid magnetic lattice is formed. It was shown theoretically and confirmed experimentally1 that the magnetic and nuclear amplitudes of neutron scattering (i.e., the amplitudes of scattering of slow neutrons by the magnetic and nuclear forces of one order) are comparable. Consequently, the interaction of the magnetic moment of the neutron with the magnetic moment of the atom can be determined from the coherent scattering of neutrons by antiferromagnets.

The magnetic scattering of neutrons was studied experimentally for four substances2: MnO, MnF2, MnSO4, Fe2O3. Experiments at room temperature showed: 1) the magnetic scattering is diffuse for MnF2 and MnSO4 (there is no pairing of magnetic moments); 2) the magnetic scattering for MnO is similar to scattering in the liquid phase (an order of antiparallel arrangement of magnetic moments is observed in very small regions, not coordinated with one another—short-range order); 3) in the case of Fe2O3 the presence of strong maxima of coherent scattering is found at positions not allowed from the standpoint of the chemical structure. The last two results are in complete agreement with ideas about antiferromagnetism, since for MnO and α-Fe2O3 the Curie temperatures are 122°K and 950°K, respectively.

The figure presents neutronograms obtained from samples of powdered MnO at room temperature and at 80°K. In the neutronogram of MnO at room temperature there are normal diffraction maxima of coherent nuclear scattering by a regular face-centered cubic lattice and a diffuse background of magnetic scattering at small angles. It should be noted that, although MnO has a lattice of the NaCl type, their neutronograms are opposite (with respect to the intensities of the diffraction maxima (111) and (200)). This is explained by the fact that MnO and O have neutron-scattering amplitudes of different signs. The neutronogram of MnO at low temperature has the same maxima of coherent nuclear scattering, since in this temperature region there are no structural changes. In addition, the neutronogram shows the presence of a strong diffraction maximum at a position where it should not occur from the standpoint of the chemical structure of the lattice. The appearance of this maximum can be explained with the aid of the magnetic interaction of the neutron with the atoms of the lattice.

The diffraction maximum may be assigned the index (111), if the elementary magnetic cell is taken to be twice as large as the chemical one, which was to be expected, since the oxygen atom has no permanent magnetic moment.

In conclusion, it should be noted that the neutron-diffraction method will undoubtedly yield much that is new in the study of the magnetic structure and magnetic transformations of the crystal lattice.

R. P. Ozerov

REFERENCES CITED

1. O. Halpern, M. H. Johnson, Phys. Rev. 55, 898 (1939),
2. C. G. Shull, J. S. Smart, Phys. Rev. 76, 1256 (1949).