Magnetism of /sup 57/Fe spins close to the /sup 57/Fe/sub 3/O/sub 4//CoO interface

Summary form only given. CoO/Fe/sub 3/O/sub 4/ bilayers and multilayers show exchange-biasing i.e. the hysteresis loop of the Fe/sub 3/O/sub 4/ layers is shifted by the presence of anti ferromagnetic CoO layers. Recent neutron diffraction measurements clarified to a great extent the temperature and magnetic field dependent behavior of the CoO part of these superlattices. A magnetic domain wall in the CoO layer, which is normally assumed in models of exchange biasing, was not observed. The possible formation of a domain wall in Fe/sub 3/O/sub 4/ was suggested. A surprising result is the enhanced Neel temperature, T/sub N/, of the CoO layer (with respect to the bulk value) and the decreased blocking temperature, TB, Of the system for CoO thickness below 10 nm because it is widely believed that TB follows TN in exchange biased systems. In our work we studied CoO/Fe/sub 3/O/sub 4/ bilayers by nuclear resonance scattering at grazing incidence. It is a powerful technique for studying layered structures based on the combination of time-domain hyperfine nuclear spectroscopy with grazing incidence X-ray scattering. We were looking at the full /sup 57/Fe/sub 3/O/sub 4/ layer and at the CoO/Fe/sub 3/O/sub 4/ interface by means of a probe Fe/sub 3/O/sub 4/ layer, at different temperatures from 130 K up to 340 K and in the presence of an external magnetic field up to 5 T at 130 K, where the system is exchange biased via field cooling. It appears that the presence of an intermediate CoO layer leads to a decrease of the room-temperature hyperfine fields in the Fe/sub 3/O/sub 4/ overlayer. Just as in the case of Fe/sub 3/O/sub 4/ on MgO[100], the spins are not oriented in the plane. Interesting differences are found in the spin dynamics. Whereas in ultrathin (1.6 nm) /sup 57/Fe/sub 3/O/sub 4/ films on MgO[100] the /sup 57/Fe spins start to fluctuate already at 80K on the time scale of the measurement (10 ns), this is not the case when the films are in contact with an AF CoO layer, at least up to 200 K. However, fluctuations are clearly present at 280 K, 11 degrees below the bulk Neel temperature of CoO. We also observed that /sup 57/Fe spins in the interface region show essentially static behavior at least up to 310 /spl deg/C if they are part of a thicker (10 nm) Fe/sub 3/O/sub 4/ layer. We also tried to observe the direct effect of exchange biasing on the interface /sup 57/Fe spins, by measuring NRS spectra at various points at the hysteresis loop of the /sup 57/Fe/sub 3/O/sub 4//CoO/MgO[100] system. So far we have no evidence of domain wall formation in the Fe/sub 3/O/sub 4/ layer during the field cycling.