Data are presented and analyzed for the nuclear spin-echo decay of $^{63,65}\mathrm{Cu}$ and $^{17}\mathrm{O}$ in ${\mathrm{La}}_{1.85}$${\mathrm{Sr}}_{0.15}$${\mathrm{CuO}}_{4}$ at T=100 K. The echo-decay data are compared with calculated waveforms in which time fluctuations of coupled neighbor spins are taken into account. Data for the two copper isotopes are found to show the effects of both ${\mathit{T}}_{1}$ and nuclear spin exchange fluctuations. Experimental decay rates for the $^{63,65}\mathrm{Cu}$ are in good agreement with indirect nuclear spin-spin interactions calculated using values of \ensuremath{\chi}'(q,0) inferred from neutron-scattering data for S(q,\ensuremath{\omega}) for this system. For $^{17}\mathrm{O}$, the echo decay is found to be in good agreement with an interpretation based on Cu-O nuclear dipolar couplings alone, in which the same Cu ${\mathit{T}}_{1}$ and nuclear spin exchange values are employed as for the copper echo decay. Calculated Cu-O indirect couplings of comparable magnitude to the dipole-dipole terms are therefore concluded to be absent. We further conclude that the transferred Cu-O hyperfine coupling is much smaller than that deduced from the measured $^{17}\mathrm{O}$ NMR shift. It follows that the excitations which provide the $^{17}\mathrm{O}$ NMR shift and relaxation are separate and distinct from those associated with spin fluctuations on the copper sites. Similar conclusions were reached in a recent study of $^{17}\mathrm{O}$ nuclear spin-lattice relaxation in this system.
Read full abstract