Modeling of a twisted-Kagome HoAgGe spin ice using Reduced-Configuration-Space Search and Density Functional Theory

Gunnar F. Schwertfeger, Po-Hao Chang, Predrag Nikolić, Igor I. Mazin · 2025

Read Modeling of a twisted-Kagome HoAgGe spin ice using Reduced-Configuration-Space Search and Density Functional Theory by Gunnar F. Schwertfeger, Po-Hao Chang, Predrag Nikolić, Igor I. Mazin (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-04146.txt.

Introduction

Gunnar Schwertfeger and co-authors model a frustrated magnet with a twisted honeycomb-like lattice in a 2025 paper. The material HoAgGe arranges its atoms on a twisted Kagome net of corner-sharing triangles. It behaves as a two-dimensional spin ice with an unusually rich phase diagram.

In a magnetic field the compound's magnetization climbs in steps at simple fractions of saturation. Earlier simulations with fitted parameters explained only part of that behavior. The team combined a reduced-configuration search with density functional calculations to ground the model in first principles. Their approach aims to explain the full plateau pattern and the strong in-plane anisotropy behind it.

Catalog record

CREATORGunnar F. Schwertfeger, Po-Hao Chang, Predrag Nikolić, Igor I. Mazin
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE26 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-04146

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Modeling of a twisted-Kagome HoAgGe spin ice using Reduced-Configuration-Space Search and Density Functional Theory Gunnar F. Schwertfeger, Po-Hao Chang, Predrag Nikolić, Igor I. Mazin Abstract The Kagome lattice is a 2D network of corner sharing triangles found in several rare earth materials resulting in a complicated and often frustrated magnetic system. In the last decades, modifications of the motif, such as breathing Kagome, asymmetric Kagome, and twisted Kagome were brought into the limelight. In particular, the latter has lower symmetry than the original Kagome and thus allows implementations of an “Ising-local” Hamiltonian, leading to a 2D spin ice. One such material implementation, HoAgGe, was recently reported to have an exceptionally rich phase diagram and is a strongly frustrated 2D spin-ice material with a twisted-Kagome geometry. In the presence of an external magnetic field the compound exhibits step-like magnetization plateaus at simple fractions of the saturation magnetization. It is believed that this phenomenon results from strong single-site anisotropy, which in HoAgGe was found to be in-plane and along a high-symmetry direction. Previous Monte Carlo simulations with empirical exchange parameters explain some, but not all experimental observations. In this work we present (a) first-principle calculations of the crucial model parameters and (b) direct energy minimization via a Reduced-Configuration-Space search, as well as Monte-Carlo simulations of the field-dependent phase diagram. We find that for HoAgGe the calculated exchange parameters are very different from the earlier suggested empirical ones, and describe the phase diagram much more accurately. This is likely because the first-principles parameters are, in addition to geometrically, also parametrically frustrated. I Introduction Magnetic frustration has been the source of many exotic phenomena in matter. In 3D rare-earth pyrochlore compounds, for instance, one finds a rather unique phenomenon dubbed “spin ice” Ramirez et al. 1999 . In a new twist, a 2D material, HoAgGe, was recently reported Zhao et al. 2020 to be an example of 2D spin ice. Spin ice is, typically, a highly frustrated magnetic system where each magnetic site has a strong uniaxial anisotropy, with the axis varying from site to site according to the crystal symmetry. The kagome lattice is formed from corner sharing equilateral triangles (Fig. 1 ). As the kagome lattice is the 2D analogue of the 3D pyrochlore lattice, it is natural to look for spin ice behavior there. In the last decades, various interesting modifications of the kagome lattice have also been discussed in the literature, such as breathing kagome Ezawa 2018 , distorted kagome Hering et al. 2022 , square kagome Richter et al. 2009 etc. One of the newest developments is the so-called Huang et al. 2023 twisted Kagome lattice, where the alternating triangles in the lattice are twisted around their centers in the opposite diretions. This operation lowers the symmetry from P6/mmm to P \bar{6} /2m. In particular, the site symmetry is lowered from mmm to m2m, as one can see from Fig. 1 , and the global inversion symmetry is also lost. The recently reported magnetic compound, HoAgGe Zhao et al. 2020 has this twisted kagome structure, with the twisting angle \sim 15.6^{\circ} (again as depicted in Fig. 1 ). Ho 3+ atoms in this compound have 10 f-electrons, a large orbital moment and strong spin-orbit coupling. Therefore, the singe-site anisotropy is large and, generally speaking, can have any easy axis, in or out of plane Lee et al. 2023 . Experimental evidence suggests that the easy axis for Ho is along a high-symmetry direction as indicated by the arrows in Fig. 1 . Our first-principle calculations described below confirm this. In this case, the magnetic Hamiltonian, to a good approximation, is local-Ising, in the sense that each spin can assume only two directions. This results in a ”two-in-one-out” spin-ice rule in plane for the triangular nearest neighbor interactions between Ho atoms in the twisted kagome lattice. This is analogous to the ”two-in-two-out” rule observed in the tetrahedra in pyrochlore spin-ice compounds Ramirez et al. 1999 . Existing experimental data are reasonably well consistent with this picture Zhao et al. 2020 ; Bhandari et al. 2025 (albeit small, but distinct deviations from the infinite-anisotropy model were reported in Ref. Bhandari et al. 2025 ). As the dominant interaction is antiferromagnetic, the ground state in zero field has zero magnetization, and the spins form a \sqrt{3}\times\sqrt{3} supercell, as depicted in Fig. 7 , with the magnetic space group retaining the hexagonal symmetry, P \bar{6}^{\prime} m2 ′ . In the in-plane saturation field, whose value depends on the field direction, all spins assume the direction most parallel to the applied field, Experimentally, the “iceness” of the system is reflected in the multiple magnetization steps in an in-plane magnetic field \mathbf{h} . Indeed, HoAgGe exhibits, in an external magnetic field, several of these unique magnetic steps. These steps are simple fractions of the total magnetization. While some of the larger steps (such as the 1/3 and 2/3 magnetization plateaus observed for \mathbf{h}||y ) have been explained in earlier works Zhao et al. 2020 , many smaller magnetization steps, particularly those observed for \mathbf{h}||x , have not been explained. To this end, we have, first, performed first-principles DFT+U calculations, making sure that the Hund’s rule on Ho is respected Lee et al. 2023 ; Lee et al. 2024 , as discussed below, and extracted six Heisenberg exchange parameters, up to the 5th nearest neighbors (as opposed to three parameters used in the empirical model of Ref. Zhao et al. 2020 ). The resulting parameters are completely different from latter, yet provide a better description of experimentally observed plateaus not only for \mathbf{h}||y , but also for \mathbf{h}||x . Using these parameters, we constructed full exchange Hamiltonians (as discussed below, the Dzyaloshinskii-Moriya and dipole-dipole interactions appear to be not important), and directly minimized the energy by performing a search over a Reduced-Configuration-Space for all possible supercells up to 18 formula units and constructed the zero-temperature phase diagram for both field directions. We found that all major phases are very stable, and secondary phases, corresponding to smaller magnetization plateaus, are on the borderline of stability, but still present. Thus, we identified all 10 experimentally observed phases. In addition, we attempted Monte-Carlo simulations similar to those in Ref. Zhao et al. 2020 , but, as opposed to their model, our calculated Hamiltonian is much more frustrated, so finding the global energy minimum in MC calculations is challenging. Figure 1 : Twisted Kagome Lattice with J_{1} bonds visualized (a 3\times 3 supercell is shown). The arrows indicate the easy axis in HoAgGe. Figure 2 : Possible spin models on a twisted kagome lattice with a strong easy axis. Left: Ising-ice model (same as Fig. 1 ). Center: 4-Potts-ice model. Right: 8-Potts-ice model. The red rods indicate the hypothetical easy axes. II general model II.1 Crystal Lattice Hamiltonian The Hamiltonian for the Spin-Ice crystal HoAgGe is: H[M;J_{\alpha},h]=-\sum_{\alpha}J_{\alpha}\sum_{\left<i,j\right>_{\alpha}}\hat{\bf S}_{i}\cdot\hat{\bf S}_{j}-h\sum_{i}\hat{\bf S}_{i}\cdot\hat{\bf h}\ , (1) where M=\{\hat{\bf S}_{0},\hat{\bf S}_{1},...\hat{\bf S}_{N}\} defines a particular state of the crystal lattice, J_{\alpha} are Heisenberg exchange parameters and h the applied magnetic field along \hat{\bf h} . All spins, \hat{\bf S}_{i} and \hat{\bf h} are unit length vectors in \mathbb{R}^{3} . Here, i,j label lattice sites, and \alpha enumerates the Ho-Ho bonds of different lengths on

Read the complete text