Deutsche Tagung f ¨ur Forschung mit ... - SNI-Portal

Magnetismus Poster: Do., 13:00–15:30 D-P246
Where are the limits of non-exponential relaxation ?
Catherine Pappas 1 , Robert Cywinski 2 , Adrian Hillier 3 , Pascal Manuel 3 , Ian
Campbell 4 , Ferenc Mezei 1
1 Hahn Meitner Institut Berlin, Germany – 2 Physics and Astronomy Department,
Leeds, UK – 3 ISIS, Didcot UK – 4 Université de Montpellier 2, France
The fundamental question of the universality of the precursor features of a low temperature
disordered phase cannot be easily answered. Non-exponential relaxation is
found in glasses and spin glasses but not in disordered ferromagnets and antiferromagnets.
According to an Griffithss argument, however, the paramagnetic phase above
disordered ferromagnets should be anomalous showing non-exponential relaxation. We
searched for this so-called Griffithss phase in the disordered system Au1−xFex, by combining
Neutron Spin Echo (NSE) spectroscopy and Muon Spin Relaxation (µSR). This
system is spin glass for x0.155. At the spin
glass side, for x=0.14, the analysis of the strongly non-exponential relaxation leads to
an excellent agreement between NSE and µSR, an unambiguous proof of the homogeneous
character of the spin glass relaxation. A slight increase of the Fe concentration
leads to the disordered ferromagnetic phase (x=0.16, 18) and an abrupt change of the
relaxation seen by NSE, which becomes exponential. The µSR spectra on the other
side remain non-exponential and at high temperatures they are very similar to those of
the spin glass side of the phase diagram, as expected by the Griffithss argument. The
puzzling discrepancy between NSE and µSR in the disordered ferromagnetic phase explains
why neutron scattering and macroscopic magnetization measurements failed up
to now to identify the Griffithss phase in disordered magnets. The observation of the
Griffiths phase by muons and of the disordered ferromagnetic transition with neutrons
provides a deeper link between theory and experiment and gives a natural explanation
of the marked influence of strong disorder on the ferromagnetic and antiferromagetic
second order phase transitions.