Figure 1

Upper: examples of $\mu \mathrm{S}\mathrm{R}$ spectra recorded in zero and longitudinal fields at $T=52.59(2)\mathrm{K}$ [above $T_{\mathrm{C}}=52.49(2)\mathrm{K}$] for ${\mathbf{S}}_{\mu }\perp \mathbf{a}$. The solid lines are fits assuming a squared-Lorentzian distribution for $B_{\mathrm{l}\mathrm{o}\mathrm{c}}$. The dashed line, which is the result of the fit of the zero-field spectrum with an exponential relaxation function, cannot be distinguished from the solid line except above $\sim 11\mu \mathrm{s}$. Lower: the comparison of the 1.0 mT spectrum with the prediction of an exponential fit shows that this model is not valid in longitudinal fields. The field dependence at small field of $P_Z(t)$ proves that the field distribution at the muon site is quasistatic.Reuse & Permissions

Figure 2

Temperature dependence of ${\lambda }_Z$ measured in zero field for ${\mathbf{S}}_{\mu }\perp \mathbf{a}$ and ${\mathbf{S}}_{\mu }\parallel \mathbf{a}$ in the upper and lower parts, respectively. The insets display ${\lambda }_Z(T)$ near $T_{\mathrm{C}}$. The solid and dashed lines are the results of fits for a dipolar Heisenberg ferromagnet as explained in the main text. Since ${\mathbf{B}}_{\mathrm{l}\mathrm{o}\mathrm{c}}\parallel \mathbf{a}$, we cannot observe a spin-lattice relaxation process for ${\mathbf{S}}_{\mu }\perp \mathbf{a}$ in the ordered state. The point for ${\mathbf{S}}_{\mu }\perp \mathbf{a}$ at $(T-T_{\mathrm{C}})/T_{\mathrm{C}}=0.05$ does not fit the critical description, pointing out that it was recorded outside the critical region for that geometrical configuration.Reuse & Permissions