Frustration-driven unconventional magnetism in the Mn$^{2+}$ ($S=\frac{5}{2}$) based two-dimensional triangular-lattice antiferromagnet Ba$_{3}$MnTa$_{2}$O$_{9}$

TL;DR

BMTO investigates a high-spin ($S=5/2$) TLAF realized in a quasi-2D trigonal triple perovskite. The authors combine dc/ac susceptibility, specific heat, ESR, XRD, and first-principles calculations (DFT+$U$) with atomistic spin dynamics to map magnetic interactions and dynamics. They find no long-range order down to 0.5 K, a dominant intralayer AFM exchange $J_1$ with weaker interlayer couplings, and broad short-range correlations evidenced by a broad $C_{mag}(T)$ peak and entropy $S_m$ not completing $R\ln(6)$; these features, along with ESR linewidth growth, point to frustration-driven unconventional magnetism in a quasi-2D setting. Theoretical results show Mn$^{2+}$ moments of about $4.5\mu_B$ and a $1.9$ eV insulating gap, with a $J_1$-driven frustrated ground state, making BMTO a promising platform to realize exotic magnetic states in high-spin TLAFs.

Abstract

A triple perovskite oxide Ba$_{3}$MnTa$_{2}$O$_{9}$ has been synthesized and its magnetic properties have been investigated through dc and ac magnetization, specific heat, electron spin resonance (ESR) measurements, and density functional theory (DFT) calculations. Mn$^{2+}$ ($S$ = 5/2) ions are the only magnetic species present in the material. These Mn$^{2+}$ ions constitute a quasi-two-dimensional triangular network in the crystallographic $ab$-plane. Magnetization and specific heat measurements reveal the absence of any long-range magnetic order down to 0.5\,K despite the presence of antiferromagnetic correlations between the magnetic ions, suggesting the presence of geometric frustration in the material. The entropy release is lower than the expected theoretical value of $Rln(6)$, further suggesting the presence of frustration. First-principles calculations using density functional theory (DFT) and atomistic spin dynamics (ASD) simulations further support this lack of static magnetic order even at low temperatures and identify the competing magnetic interactions along with the quasi-2D magnetic dimensionality as the underlying origin of such an unconventional magnetic behavior.

Figures (7)

• Figure 1: The room temperature powder x-ray diffraction data (blue circle) along with its double-phase Rietveld refinement (red solid line). Bragg positions for the main phase (BMTO) and the small fraction of non-magnetic secondary phase Ba$_5$Ta$_4$O$_{15}$ are shown as vertical lines in green and orange, respectively. Both phases belong to the same space group ($P\overline{3}m1$). Peaks corresponding to the Ba$_5$Ta$_4$O$_{15}$ phase are marked with ($\ast$). The violet solid line indicates the difference between the observed and calculated intensities. The XRD pattern of the non-magnetic analog BZTO is shown with a solid gray line. This pattern is shifted along the $y$-axis for clarity.
• Figure 2: (a) The idealized crystal structure of BMTO is shown. The solid (black) lines indicate the unit cell. Mn$^{2+}$ ions (only at $1b$ site) and Ta$^{5+}$ ions (only at $2d$ site) are shown with the purple and light blue octahedral oxygen environment, respectively. Three thick double-sided arrows (horizontal, vertical, and slanted) indicate the distance between the first, second, and third nearest neighbor Mn$^{2+}$ ions, respectively. (b) Two edge-shared triangular layers of the Mn$^{2+}$ ions are shown with the exchange interaction between the first ($J_1$), second ($J_2$), and third ($J_3$) nearest neighbor Mn$^{2+}$ ions. (c) MnO$_{6}$ octahedra and the high-spin electronic configuration (3$d^5$) of Mn$^{2+}$ ions are shown.
• Figure 3: (a) Zero-field-cooled (ZFC) magnetic susceptibility measured with $H$ = 10 kOe is shown as a function of temperature on the left $y$-axis. Its fitting with the Curie-Weiss (CW) law in the temperature range of 100 K - 750 K is shown by the solid red line. The inverse susceptibility $(\chi-\chi_0)^{-1}$ along with CW fitting (and its extrapolation) is plotted on the right axis. (b) The low-temperature part of the ZFC and field-cooled (FC) susceptibility measured with 100 Oe and 10 kOe are shown. (c) The ac-susceptibility ($\chi^{'}$) as a function of temperature measured with different frequencies is shown on a semi-log scale. (d) Isothermal magnetization curves at various temperatures measured with PPMS VSM (symbols) are shown. The arrow direction indicates increasing temperature.
• Figure 4: (a) Specific heat $C_{p}$ of BMTO and non-magnetic Ba$_{3}$ZnTa$_{2}$O$_{9}$ (BZTO) are shown as a function of temperature $(T)$ in log-log scale. Data at different applied fields for BMTO are shown by symbols and zero field data for BZTO is shown by the green solid line. The blue solid line indicates the zero field $C_{p}$ data of Ba$_{3}$MnNb$_{2}$O$_{9}$ adopted from Ref. lee2014 (b) The $C_{p}/T$ is plotted as a function of $T^2$ in the low-temperature region both for BMTO and BZTO. The straight line corresponding to BZTO is extrapolated in the range 0.5 K - 1.8 K (see text) and shown as a dotted line. (c) $C_{mag}$ of BMTO for different fields are plotted as a function of temperature. The solid arrow indicates the increasing field (d) The magnetic entropy for various applied fields is shown with an arrow that guides the rising of the fields. The horizontal dashed line indicates the maximum entropy released.
• Figure 5: (a) The normalized ESR spectra at different temperatures are presented. (b) The linewidth as a function of temperature is plotted. Inset shows the Lorentzian fit to the ESR signal at $T$ = 300 K.