Comment on "Two-spinon and four-spinon continuum in a frustrated ferromagnetic spin-1/2 chain"

S.-L. Drechsler,
S. Nishimoto,
J. Málek,
W. E. A. Lorenz,
J. Richter,
J. van den Brink,
M. Schmitt,
H. Rosner

I. M. Frantsevich Institute for Problems of Materials Science of the NAS of Ukraine, Krzhizhanovsky str., 3, Kyiv, 03142, Ukraine
Phys. Rev. Lett. - Maryland, USA: American Physical Society (APS), 2011, #106


We show that LiVCuO4 should be described by strongly ferromagnetically coupled Heisenberg antiferromagnetic chains (HAC) in sharp contrast with the effective exchange integrals Ji given in Enderle et al., Phys. Rev. Lett. vol. 104, 237207 (2010), and the main issues of that work, namely, (i) LiVCuO4 is well described by two weakly ferromagnetically coupled interpenetrating Heisenberg antiferromagnetic spin-1/2 chains, (ii) the extracted exchange integrals J1, J2 agree with a previous spin-wave description (Enderle et al., Euphys. Lett. vol. 70, 237 (2005)), (iii) the spectral density of inelastic neutron scattering (INS) above 10 meV is ascribed to a 4-spinon continuum. Applying exact diagonalization and DMRG methods to fit their INS and magnetization M(H) data, supported by two independent microscopic methods (5-band Hubbard model and LSDA+U calculations), we demonstrate that LiCuVO4 exhibits strong inchain frustration with alpha =-J2/J1 < 1, i.e. strong coupling of the HAC at odds with (i). An alternative phenomenological set in accord with various experimental results is proposed. In view of the recent possible discovery of quantum-spin nematics and Bose condensation of two-magnon bound states (M. Zhitomirsky et al. arXiv:1003.4096v2 (2010), L. Svistov et al. ibid. 1005.5668v2 (2010)) in LiCuVO4 precise knowledge of the main J-values is of key importance.