Excitation spectrum and low-temperature magnetism in disordered defect-fluorite Ho2Zr2O7

TL;DR

This study resolves the crystal-field scheme and low-temperature magnetism of the highly disordered Ho2Zr2O7 defect-fluorite. By combining bulk thermomagnetic measurements with inelastic neutron scattering and developing two CEF models—S with $O_h$ symmetry and E incorporating disorder-induced lower symmetry—the work shows a small $0.9~\mathrm{meV}$ gap to the first excited state and a ground state whose character depends on the model, with both predicting zero magnetic moment for the ground state. INS reveals a broad $~60~\mathrm{meV}$ CEF feature broadened by disorder, while bulk data indicate local antiferromagnetic correlations and slow spin dynamics without long-range order down to $\sim150$–$200~\mathrm{mK}$, with Arrhenius parameters $E_b\approx26~\mathrm{K}$ and $\tau_0\approx8\times10^{-13}~\mathrm{s}$. The results suggest that structural disorder enables mixing of low-lying levels and sustains magnetism in a frustrated lattice, challenging simple Ising explanations and offering a framework to model disorder in non-Kramers CEF systems. Overall, the work highlights disorder as a tuning parameter that can promote correlated magnetic states in defect-fluorite oxides and provides a practical methodology for incorporating disorder into crystal-field analyses.

Abstract

In this work, we report on the thermomagnetic characterization and crystalline-electric field (CEF) energy scheme of the disordered defect-fluorite Ho2Zr2O7. This structural phase is distinguished by the coexistence of magnetic frustration and extensive disorder, with Ho3+ and Zr4+ sharing randomly the same 4a site with even 50% occupancy, and an average 1/8 oxygen vacancy per unit cell. AC magnetic susceptibility measurements performed on powder samples down to 0.5 K revealed signs of slowing spin dynamics without glassy behavior, including a frequency dependent peak at 1 K. Yet, no evidence for long-range magnetic order is found down to 150 mK in specific heat. Inelastic neutron scattering measurements show a weak, low-lying CEF excitation around 2 meV, accompanied by a broad level centered at 60 meV. To fit our observations, we propose an approach to account for structural disorder in the crystal-field splitting of the non-Kramers Ho3+. Our model provides an explanation to the broadening of the high-energy, single-ion excitations and suggests that the zirconate ground-state wave function has zero magnetic moment. However, structural disorder acts as guarantor of the magnetism in Ho2Zr2O7, allowing the mixing of low lying states at finite temperatures. Finally, we show that this scenario is in good agreement with the bulk properties reported in this work.

Figures (8)

• Figure 1: X-ray diffraction pattern of sol-gel synthesized Ho2Zr2O7 sample (black circles), showed along with the refined defect-fluorite phase (red line). Difference between data and refinement is shown at the bottom of the figure (blue line).
• Figure 2: Inelastic neutron scattering measured with $E_\mathrm{i}=120$ meV at 7 K (top), 100 K (middle) and 200 K (bottom). A broad CEF excitation is observed around 60 meV at all temperatures.
• Figure 3: CEF excitation at around 60meV with the calculated neutron cross-section for each model at 7K. The dotted curves within the peak help to visualize the position of the transitions for (a) the standard model and (b) the effective model.
• Figure 4: Magnetization versus the applied magnetic field at 2K. The dashed lines are the single-ion magnetizations calculated using the CEF steven parameters obtained by the standard (S) and effective (E) point-charge models. Inset: temperature dependence of the inverse magnetic susceptibility, the Curie-Weiss temperature, and effective magnetic moment determined by the Curie-Weiss fitting (green line).
• Figure 5: Temperature dependence of the real part of the AC Magnetic Susceptibility. Inset: dependence of the frequency versus the inverse of the temperature of the maximum in the magnetic susceptibility, with the Arrhenius law fit.