Abstract
A primary goal at the interface of theoretical and experimental quantum magnetism is the investigation of exotic spin states, mostly notably quantum spin liquids (QSLs) that realize phenomena including quasiparticle fractionalization, long-ranged entanglement, and topological order. Magnetic rare-earth ions go beyond the straightforward paradigm of geometrical frustration in Heisenberg antiferromagnets by introducing competing energy scales, and in particular their strong spin-orbit coupling creates multiple split crystal electric-field (CEF) levels, leading to anisotropic effective spin models with intrinsic frustration. While rare-earth delafossites have a triangular-lattice geometry and thus have gained recent attention as candidates for hosting spin-1/2 QSL physics, the reliable extraction of effective spin models from the initial many-parameter CEF spectrum is a hard problem. Using the example of CsYbSe2, we demonstrate the unambiguous extraction of the Stevens operators dictating the full CEF spectrum of Yb3+ by translating these into parameters with a direct physical interpretation. Specifically, we combine low-field susceptibility measurements with resonant torsion magnetometry (RTM) experiments in fields up to 60 T to determine a sufficiently large number of physical parameters - effective Zeeman splittings, anisotropic van Vleck coefficients, and magnetotropic coefficients - that the set of Stevens operator coefficients is unique. Our crucial identification of the strong corrections to the Zeeman splitting of Kramers doublets as van Vleck coefficients has direct consequences for the interpretation of all anisotropic magnetic susceptibility measurements. Our results allow us to determine the nature and validity of an effective spin-1/2 model for CsYbSe2, to provide input for theoretical studies of such models on the triangular lattice, and to provide additional materials insight into routes for achieving magnetic frustration and candidate QSL systems in rare-earth compounds.
| Original language | English |
|---|---|
| Article number | 043202 |
| Journal | Physical Review Research |
| Volume | 3 |
| Issue number | 4 |
| DOIs | |
| State | Published - Dec 2021 |
Funding
We thank M. Mourigal, S. Nikitin, and K. Ross for helpful discussions. We are grateful to M. Chan and A. Shekhter for technical assistance with our pulsed-field measurements. Experimental work at the University of Colorado Boulder was supported by Award No. DE-SC0021377 of the U.S. Department of Energy (DOE), Basic Energy Sciences (BES), Materials Sciences and Engineering Division (MSE). Theoretical work at the University of Colorado Boulder was supported by Award No. DE-SC0014415 of the U.S. DOE, BES, MSE. Work at Oak Ridge National Laboratory was supported by the U.S. DOE, BES, MSE. A portion of this work was performed at the National High Magnetic Field Laboratory, in the Pulsed Field Facility at Los Alamos National Laboratory, which is supported by National Science Foundation Cooperative Agreement No. DMR-1644779, the State of Florida, and the U.S. DOE. The publication of this article was funded partially by the University of Colorado Boulder Libraries Open Access Fund.