Strongly Enhanced Magnetic Fluctuations in a Large-Mass Layered Ruthenate

We have studied the magnetic excitations in ${\mathrm{C}\mathrm{a}}_{2-x}{\mathrm{S}\mathrm{r}}_x{\mathrm{R}\mathrm{u}\mathrm{O}}_4$, $x=0.52$ and 0.62, which exhibit an anomalously high susceptibility and heavy mass Fermi liquid behavior. Our inelastic neutron scattering experiments reveal strongly enhanced magnetic fluctuations around an incommensurate wave-vector (0.22,0,0) pointing to a magnetic instability. The magnetic fluctuations show no correlation in the c direction and also along the ${\mathrm{R}\mathrm{u}\mathrm{O}}_2$ planes the signal is extremely broad, $\Delta q=0.45{Å}^{-1}$. These fluctuations can quantitatively account for the high specific heat coefficient and relate to the high macroscopic susceptibility. The magnetic scattering is attributed to the $d_{xy}$ band, the active band for spin triplet superconductivity in ${\mathrm{S}\mathrm{r}}_2{\mathrm{R}\mathrm{u}\mathrm{O}}_4$.

Figure 1

(a)-(d) Scans across the magnetic scattering near $\mathbf{Q}=(1,0,0)$ in ${\mathrm{C}\mathrm{a}}_{1.38}{\mathrm{S}\mathrm{r}}_{0.62}{\mathrm{R}\mathrm{u}\mathrm{O}}_4$ at $T=12\mathrm{K}$ and at an energy transfer of 4.1 meV obtained with different scattering geometries, the scan directions are illustrated in part e). The map in part f) was obtained by scans of type a) at the same conditions. For clarity the background has been subtracted and the data have been corrected for the magnetic form factor of ${\mathrm{R}\mathrm{u}}^{+}$.

Figure 2

Constant-energy scans performed at different energy transfers around $\mathbf{Q}=(1,0,0)$ along the $[0,1,0]$ direction at $T=1.5\mathrm{K}$. The scattering angle dependent background was subtracted for clarity. The horizontal bars indicate the spectrometer resolution. The solid lines are fits to Gaussian functions. The data were acquired with monitor 1E7, which corresponds to a counting time of approximately 7 min at 4.1 meV.

Figure 3

(a) ${\chi }^{\prime \prime }(q_{inc-xy},\omega )$ as function of energy for different temperatures. The solid lines represent fits with a single relaxor, Eq. (3). The data were converted to absolute units by comparing with data reported for ${\mathrm{S}\mathrm{r}}_2{\mathrm{R}\mathrm{u}\mathrm{O}}_4$ [2]. (b) the same ${\chi }^{\prime \prime }(q_{inc-xy},\omega )$ data plotted against temperature. Temperature dependence of the characteristic energy (c) and amplitude (d) of the scattering at ${\mathbf{q}}_{\mathbf{i}\mathbf{n}\mathbf{c}-\mathbf{x}\mathbf{y}}$; in (d) we also show the temperature dependence of ${\chi }_{\mathrm{m}\mathrm{a}\mathrm{c}}$, see [8].

Figure 4

(a) Comparison of the inelastic signal around $\mathbf{Q}=(1,0,0)$ in ${\mathrm{C}\mathrm{a}}_{2-x}{\mathrm{S}\mathrm{r}}_x{\mathrm{R}\mathrm{u}\mathrm{O}}_4$ for $x=2.0$, 0.62, and 0.52 at an energy transfer of 4.1 meV ($x=2.0$ and 0.62) and 2.1 meV (x=0.52). Data were normalized with a reference phonon scan and background scattering has been subtracted. The data for the pure compound ($x=2.0$) were taken from Ref. 4. Inset : sketch of the calculated Fermi surfaces of the $d_{xy}$ sheet in ${\mathrm{S}\mathrm{r}}_2{\mathrm{R}\mathrm{u}\mathrm{O}}_4$ (light) and ${\mathrm{C}\mathrm{a}}_{1.5}{\mathrm{S}\mathrm{r}}_{0.5}{\mathrm{R}\mathrm{u}\mathrm{O}}_4$ (dark), indicating the influence of the downward shift of the $d_{xy}$ band due to the rotation of the octahedra.