YbCu$_{1.14}$Se$_2$: an exchange disordered 2D triangular random singlet phase?

Abstract

Quantum spin liquid (QSL) phases exist in theory, but real candidate QSL materials are often extraordinarily sensitive to structural defects which disrupt the ground state. Here, we investigate candidate triangular QSL material YbCu$_{1.14}$Se$_2$ and discover the absence of magnetic order, but also no compelling evidence of a QSL ground state due to significant structural disorder. We instead look at the results through a lens of a 2-dimensional (2D) random singlet phase. YbCu$_{1.14}$Se$_2$ behaves strikingly similar to other disordered triangular lattice materials, suggesting universal behavior of random singlet formation in 2D frustrated systems.

Figures (10)

• Figure 1: (a,b) Crystal structure of YbCu$_{1.14}$Se$_2$ obtained from SC-XRD measurements. The magnetic Yb ions sit at the corners of the unit cell, forming a triangular lattice. The copper position is shown as a partially filled sphere representing its partial occupancy. (c,d) $0kl$ and $hk0$ precession images. There are no deviations from the expected Bragg peak positions for a hexagonal $P$ lattice (yellow circles), indicating that the Cu vacancies are not ordered.
• Figure 2: YbCu$_{1.14}$Se$_2$ magnetic susceptibility as a function of temperature. Panel (a) shows the susceptibility, and panel (b) shows inverse susceptibility with Curie-Weiss fits to high-temperature and low-temperature data. The nonlinearity in $\chi^{-1}$ is due to crystal field effects.
• Figure 3: Zero-field heat capacity of YbCu$_{1.14}$Se$_2$. Panel (a) shows the temperature-dependent heat capacity along with a modeled Schottky anomaly and low-temperature electronic power law. Panel (b) shows $C/T$ with the Schottky upturn subtracted alongside the singlet-distribution model in Appendix \ref{['sec:dimer-model']}. The black line is the model, the dashed line shows the raw singlet-distribution, and the dotted line shows the fitted $T^3$ phonon background. Panel (c) shows the integrated entropy (with the low-temperature Schottky upturn subtracted).
• Figure 4: Dilution refrigerator susceptibility of YbCu$_{1.14}$Se$_2$. Panel (a) shows the zero-field temperature-dependent susceptibility, showing a frequency-dependent symmetric peak indicating a glass-freezing transition around 80 mK. (b) shows the field-dependent susceptibility (offset vertically for clarity) at different temperatures. The absence of a sudden dip in susceptibility suggests YbCu$_{1.14}$Se$_2$ does not have a 1/3 magnetization plateau like $A$Yb$X_2$ compounds.
• Figure 5: Calculated specific heat for distributions of singlets. Panel (a) shows the relative distribution $w(\Delta)$ of singlet-triplet gaps $\Delta$ for a triangular (red), square (blue) and delta function (green) distribution. Panel (b) shows the numerically computed heat capacity via Eq. \ref{['eq:SumSpecificHeat']}, shown overtop the Schottky-subtracted experimental data from Fig. \ref{['fig:heatcapacity']}.