Orbital current order of the d-density-wave type (DDW) has been postulated to explain the pseudogap in the high temperature superconductors. We have performed neutron scattering experiments to search for this order and show here the results obtained on an YBa 2 Cu 3 O 6.45 sample using the best neutron spectrometers available. We argue that the data are consistent with a small, largely c-axis-directed moment, found below about 200K.PACS numbers: 72.15. Gd, 61.12.Ld, 71.30.+h 2 The microscopic origin of the pseudogap is one of the most puzzling attributes of the cuprate superconductors. A review of the various experimental techniques used to determine the pseudogap is given by Timusk and Statt. 1 It is generally considered to stem from pairing above the superconducting transition temperature T C 2,3 or result from another ground-state differentiated from superconductivity by a quantum critical point. 4,5 Chakravarty et al. 6 but a competing phase judged to stem from impurities made the identification of the DDW state difficult. The moments from the bond currents result in peaks at the (h/2, k/2, l) superlattice positions of the reciprocal lattice. This is the position that magnetism is found in the parent compound, YBa 2 Cu 3 O 6.15 , which is an insulator with a Néel transition temperature well above room temperature and has been studied previously. 10,11 The defining attribute of the DDW state is that the moment should be largely along the c-axis. Small moments in the sample crystal originating from the parent compound should be located in the a-b plane as is found for this material. Other magnetic impurity phases could be possible, but to our knowledge no such phases 3 are known to have high temperature (above 50K) magnetic order along the c-axis for the YBa 2 Cu 3 O 6+x system.In order to determine the direction of the moments for the antiferromagnetic state it is necessary to use polarized neutrons, as outlined by Moon et al. in 1969. 12 The crystal used in the experiments is twinned so that a* could not be differentiated from b*. We thus consider the crystal to be tetragonal so that the (1, 1, 0) reflection showed no peak indicating that the higher order scattering from the monochromator was well removed. Figure 1b shows the result of a HF SF measurement through the (1/2, 1/2, 1) position at 20 K. The scan range on the high Q side of the scan is limited by intense NSF scattering that is difficult to correct for with the rather small flipping ratio as seen in Fig 3a. Nevertheless, a peak is observed at (0.5, 0.5, 1) and a Gaussian fit results in a height of 11±2 counts with a width of 0.01±0.002 FWHM at the position 0.5±0.001 r.l.u. The peak is resolution limited, giving a correlation length of about 380Å. Figure 1c shows no evidence for a peak at a temperature of 150K. Multiple runs were averaged to obtain the observed error bars.
5These scans took over two days each and are difficult to improve in a reasonable time period.Also, since the magnetic signal is so small, a higher flipping ratio is des...