Experimental Realization of Spin-1/2 Triangular-Lattice Heisenberg Antiferromagnet
H. Tanaka and K. Kindo
It is one of the main subjects of condensed matter physics that revealing the ground state of a frustrated quantum magnet. A two-dimensional spin-1/2 Triangular-Lattice Heisenberg Antiferromagnet (TLHAF), which is one of the well-known frustrated quantum magnets, has an ordered ground state of the 120° spin structure. Although the ground state at zero-field is the same as that for the classical spin, the ground state of a small spin TLHAF in a magnetic field cannot be defined with the classical spin model. For example, when an up-up-down spin state in the TLHAF is stabilized in a magnetic field, one-third magnetization plateau must appear in the magnetization process as a quantum effect.
Here, we report the result of magnetization on Ba3CoSb2O9 and present verification that this substance is almost the ideal spin-1/2 TLHAF [1]. Ba3CoSb2O9 crystallizes in a highly symmetric hexagonal structure P63/mmc, and is composed of a single CoO6 octahedron and a face-sharing Sb2O9 double octahedron. Magnetic Co2+ ions frame regular triangular lattice layers on the ab plane and are separated by the non-magnetic layer of the Sb2O9 double octahedron and Ba2+ ions. Thus, Ba3CoSb2O9 is expected to be categorized as a good two-dimensional magnet.
Figure 1 (a) shows the magnetization process (M) and the derivative magnetization (dM/dH) for Ba3CoSb2O9 measured at 1.3 K. The magnetic field was applied up to 53 T and the whole magnetization process on Ba3CoSb2O9 was observed. The saturation of the Co2+ spin occurs at the saturation field (HS) of 31.9 T. The increase in magnetization above HS arises from the large temperature-independent Van Vleck paramagnetism characteristic of Co2+ in the octahedral environment [2]. Figure 1 (b) shows the magnetization curves corrected for the Van Vleck paramagnetism. The quantum magnetization plateau is clearly observed at MS/3. Thick dashed and solid lines denote fits by the higher order coupled cluster method [3] and exact diagonalization for a 39-site rhombic cluster [4], respectively. Both theoretical curves include the one-third magnetization plateau but the classical magnetization curves (thin dotted lines) show no one-third magnetization plateau. Although the experimental magnetization curve is smeared around the critical fields due to the finite temperature effect and the small anisotropies of the g factor and the interaction, the agreement between the experimental and theoretical results is excellent. This work verifies recent theory on the magnetization process for the spin-1/2 TLHAF
References
- [1] Y. Shirata, H. Tanaka, A. Matsuo, and K. Kindo, Phys. Rev. Lett. 108, 057205 (2012).
- [2] M. E. Lines, Phys. Rev. 131, 546 (1963).
- [3] D. J. J. Farnell, R. Zinke, J. Schulenburg, and J. Richter, J. Phys. Condens.: Matter 21, 406002 (2009).
- [4] T. Sakai and H. Nakano, Phys. Rev. B 83, 100405(R) (2011).