View the related article [J. Phys. Soc. Jpn. 77 (2008) 013705]

Gigantic Optical Absorption Changes for a Magnetic Field Change of ±300 Oe in CuB2O4 [January 16, 2008]
Katsuaki Sato
(Graduate School of Engineering, Tokyo University of Agriculture and Technology)

   Recently T. Arima and coworkers discovered a gigantic optical absorption change in an antiferromagnetic compound CuB2O4 for a small change of magnetic field [1]. As shown by closed circles in Fig. 1, the value of the absorption coefficient α at a photon energy of 1.405 eV is approximately tripled when the magnetic field changes from −300 to 300 Oe. The field-dependence of optical absorption is not proportional to the magnetic hysteresis, which is plotted by open circles in Fig. 1. The photographs (b) and (c) of transmitted light images for ±300 Oe provide direct evidence of the large absorption change.

Fig. 1 (a) Magnetic field dependence of α at h=1.405 eV and magnetization of CuB2O4 at 15K measured after zero-field cooling. The inset shows the α spectra at hν=1.405 eV under magnetic fields of −500 Oe (orange) and 500 Oe (green). (b), (c) Photographs of transmitted light images recorded by a CCD camera under magnetic fields of −300 and 300 Oe, respectively. The brightness and contrast conditions for the images in (b) and (c) are the same. (Figures are taken from ref. 1)

  It is well known that the optical transmission can be varied with the applied magnetic field using a system consisting of a polarizer, a magneto-optical (MO) material like Y3Fe5O12 (YIG) and an analyzer, in which the polarization direction changes with the magnetic field by Faraday rotation in the MO material leading to a change in the light intensity after the analyzer. In the case of conventional MO materials, a direct change of the light intensity in the absence of an analyzer is negligible. In this newly discovered phenomenon, there is no polarization rotation but a direct change in the absorption changes with the magnetic field.

  One question that arises is what kind of the physical phenomenon is involved in this large change? The authors ascribe this phenomenon to the optical magneto-electric (OME) effect. The OME effect can be briefly described as follows. There are series of materials that possess neither space-inversion symmetry nor time-inversion symmetry, for which the linear magneto-electric effect is expected to occur based on symmetry considerations [2]. Among these materials, those having both spontaneous magnetization Ms and spontaneous electric polarization Ps are expected to exhibit a nonreciprocal optical effect when light propagates along the Ps × Ms direction [3, 4]. The OME effect is defined as the change in the optical constants based on whether the propagation vector k of the light is parallel or antiparallel to the Ps × Ms direction. Such an effect has already been reported in GaFeO3 by the authors’ group, although the effect was as low as 0.2 % [3-5].

  The drastic OME effect observed in CuB2O4 may be characteristic of the peculiar crystallographic and magnetic structures in this material. There are two Cu2+ sites in this crystal; one is the A-site where the Cu2+ ion is surrounded by four oxygen atoms in square planar coordination and the other is the B-site, where Cu2+ ions occupy distorted octahedral sites [6]. The compound undergoes two successive magnetic transitions at 21 K and 10 K, between which the magnetic moments at the A-site are antiferromagnetic with a canted ferromagnetic moment (so-called weak ferromagnetism), while those at the B-site are disordered. The OME effect is observed only in this temperature region. The spin structures are easily changed by the application of a weak magnetic field: all the moments at the A-sites are flipped from the spin state shown in Fig. 2(b) to those shown in Fig. 2(c) with the reversal of the magnetic field along the [110] direction. The experimental result indicates that the transmitted light intensity for the configuration shown in Fig. 2(c) is three times stronger than that for the one shown in Fig. 2(b). Since the configuration shown in Fig. 2(c) is equivalent to that shown in Fig. 2(d), this result implies that the light intensity of light depends on the direction of propagation, or in other words, it implies that the sample has directional dichroism.

Fig. 2 (a) The (001) plane projection of the crystal structure of CuB2O4. The red and green spheres represent oxygen and boron, respectively. The Cu atoms (blue) occupy two nonequivalent positions. The Cu atoms at the A sites coordinated by four oxygen atoms are shown as light blue squares. (b), (c) Alignments of magnetic moments of Cu2+ ions at the A sites under a magnetic field along the (b) [110] and (c) [−1−10] axes at 12 K. The magnetic moments indicated by arrows lie almost in the ab plane. (d) The same alignment as (c) rotated by 180 degrees. ((a)-(c) are taken from ref. 1)

  
Three sharp peaks (1.405, 1.667, and 1.913 eV) are observed in the absorption spectrum. These three peaks are assigned to the zero-phonon lines in the 3d9 state of Cu2+ [7]. The peak at 1.667 eV is assigned to an electric dipole (E1) transition from dx2-y2 ground state to the doubly degenerate dyz and dzx excited state, while the peak at 1.405 eV is assigned to a magnetic dipole (M1) transition from dx2-y2 ground state to dxy excited state. These final states are hybridized by spin orbit coupling. Consequently, the OME effect in CuB2O4 is attributed to the cross-term between the E1 transition and the M1 transition [3, 4]. The M1 transition is generally considered to be much weaker than the E1 transition and the cross term tends to be neglected. However in this compound, the E1 transition is almost forbidden due to the local symmetry and the elements of the E1 transition matrix have opposite sign due to the different boron coordination of the two sublattices. Therefore the cross terms of all the A-site Cu2+ moments always have the same sign. The application of a weak magnetic field causes all the A-site moments to flip, thereby reversing the sign of the cross term, which in turn leads to the gigantic OME effect.
  The directional dichroism effect is considered to be useful for practical application to an optical isolator if a suitable material that exhibits a similar phenomenon at room temperature can be found. The present work provides the fundamental basis for such future studies.

References
[1] M. Saito et al.: J. Phys. Soc. Jpn. 77 (2008) 013705.
[2] G. T Rado and V. J. Folen: Phys. Rev. Lett. 7 (1961) 310; G. T. Rado: Phys. Rev. Lett. 13 (1964) 335.
[3] J. H. Jung et al.: Phys. Rev. Lett. 93 (2004) 037403.
[4] M. Kubota et al.: Phys. Rev. Lett. 92 (2004) 137401.
[5] T. Arima et al.: Phys. Rev. B 70 (2004) 064426.
[6] M. Martinez-Ripoll et al.: Acta Crystallogr., Sect. B 27 (1971) 677.
[7] R. V. Pisarev et al.: Phys. Rev. Lett. 93 (2004) 037204.

The above article should be referred as “K. Sato: JPSJ Online-News and Comments [Jan. 16, 2008]” when citing.

Copyright © the Physical Society of Japan.