Magnetic distorted spiral structure of Nd at 10 K

Nd has been suggested to have the unusual 2q sinusoidal structure below TN = 19.9 K. However, here will be presented an analysis of neutron scattering data, which shows that a flat distorted spiral with aspect ratio 1:3 at TN/2 = 10 K provides a better fit and a physically more appealing model. It turns into a striped (sinusoidal) structure only at T2 = 19.1 K, very close to TN, where the non-magnetic stripes are formed.

A simple mean field theory demonstrates that this transition may be driven by a minute anisotropy. A model of randomly placed singlet pairs á la Anderson resonance-valence-bond (RVB) phase is discussed. The observed splitting up of satellite peaks below T2 is discussed in terms of an epitaxial rotation model.

The results are discussed in relation to previous models.