Glassy spin freezing and NMR wipeout effect in the high-$T_c$ superconductor ${\mathrm{La}}_{1.90}{\mathrm{Sr}}_{0.10}{\mathrm{CuO}}_4:$ Critical discussion of the role of stripes

We report on ${}^{139}\mathrm{La}$ and ${}^{63}\mathrm{Cu}$ NMR/NQR measurements in the high-$T_c$ superconductor ${\mathrm{La}}_{1.90}{\mathrm{Sr}}_{0.10}{\mathrm{CuO}}_4,$ with $T_c=26.5\mathrm{}$ K. Spin fluctuations probed by ${}^{139}\mathrm{La}$ spin-lattice relaxation ${(T}_1)$ continuously slow down on cooling through $T_c.$ We argue that spin freezing and superconductivity are bulk effects in this sample. Thus both phenomena have to coexist microscopically. The distribution of ${}^{139}\mathrm{La}$ $T_1$ values at low temperature reveals a wide spread of spin fluctuation frequencies in ${\mathrm{CuO}}_2$ planes. A simple estimate shows that ${}^{63}\mathrm{Cu}$ nuclei at sites where electronic fluctuations are the slowest are not observable (wipeout effect) because relaxation times are too short. This means that the ${}^{63}\mathrm{Cu}$ NQR wipeout, observed in this sample, can be explained primarily by slow magnetic, rather than charge, fluctuations. The magnetic origin of the wipeout is still compatible with a connection between wipeout and charge order [as proposed by Hunt et al., Phys. Rev. Lett. 82, 4300 (1999)], but this connection is indirect. On the other hand, since the wipeout fraction is not an intensive quantity it cannot define a proper order parameter and cannot be used by itself as a criterion for the existence of a stripe phase.