Comment on “No evidence for orbital loop currents in charge-ordered ${\mathrm{YBa}}_2{\mathrm{Cu}}_3{\mathrm{O}}_{6+x}$ from polarized neutron diffraction”

Intra-unit cell magnetic order has been observed in four different families of high-temperature superconductors from polarized neutron diffraction experiments and supported by several other techniques. That order, which does not break translation symmetry, is consistent with the predicted orbital moments generated by two microscopic loop currents in each ${\mathrm{CuO}}_2$ cell. Recently, using polarized neutron diffraction, Croft et al. [Phys. Rev. B 96, 214504 (2017)] claim to find no evidence for such orbital loop currents in charge ordered ${\mathrm{YBa}}_2{\mathrm{Cu}}_3{\mathrm{O}}_{6+x}$. Their experiment is done with detwinned samples at least 100 times smaller than in previous experiments without counting much longer. We show by a detailed quantitative analysis of their data that contrary to their conclusion, the magnetic signal falls below their threshold of detection. None of the data reported by Croft et al. challenge the universality of the intra-unit cell order in cuprates.

Figure 1

Inverse of the measured flipping ratio $R_{\mathrm{meas}}^{-1}$ for two detwinned underdoped YBCO samples at two Bragg reflections (a) $Q=\left(010\right)$ and (b) $Q=\left(011\right)$. The black full symbols corresponds to the data of Croft et al. [14] (sample ${\mathrm{YBCO}}_{6.54}$ with $p=0.104)$ and the open red squares ones are those for sample C ${\mathrm{YBCO}}_{6.6}\left(p=0.115\right)$, of Fauqué et al. (Ref. [5] for the (010) reflection and Refs. [1, 12] for (011)). For the open symbols, the error bar is lower than the size of the point. All blue lines correspond to incorrect estimates reported by Croft et al. [14] of the data of Fauqué et al. [1] (see [15] for details).

Figure 2

$1/R=I_{\text{SF}}/I_{\text{NSF}}$ data for sample $\text{H}1\left(p=0.104\right)$ reproduced from Croft et al. [14]. The data points are exactly the same as those reported in Fig. 8 in [14] (reproduced as well in Fig. 1) but here we zoom in on the data. The dashed lines are calculated for the three different scenarios considered in the text. The top right panel represents A3 rocking scans convoluted by the resolution function (blue) for scenario B (red) for scenario C.