Different response of transport and magnetic properties of ${\mathbf{BaIrO}}_3$ to chemical and physical pressure

A combination of x-ray absorption, x-ray-diffraction, and transport measurements at high pressure is used to investigate the interplay between the electronic properties of Ir $5d$ states and lattice degrees of freedom in the weakly ferromagnetic insulator ${\mathrm{BaIrO}}_3$. Although the Ir $5d$ local magnetic moment is highly stable against lattice compression, remaining nearly unperturbed to at least 30 GPa, the weak ferromagnetism (net ordered moment) is quickly quenched by 4.5 GPa (3% volume reduction). Under chemical pressure, where Sr is substituted for the larger Ba in ${\mathrm{BaIrO}}_3$, the local magnetic moment on Ir remains stable, but the weak ferromagnetism is quenched after only 1.7% volume reduction. The magnetic ordering temperature $T_m$ is also more strongly suppressed by chemical pressure compared to physical pressure. In addition, under $\sim 23\text{−}\mathrm{at}.\%$ Sr doping, ${\mathrm{BaIrO}}_3$ undergoes a transition to a paramagnetic metallic state. Resistivity measurements indicate that ${\mathrm{BaIrO}}_3$ remains an electrical insulator to at least 9 GPa, a much higher pressure than required to quench the weak ferromagnetism $(\sim 4.5\mathrm{GPa})$. Such a disparate response of transport and magnetic properties to chemical and physical pressure is likely rooted in the different compression rates of the $(a,c)$ lattice parameters with Sr doping and applied pressure and the effect of related lattice distortions on electronic bandwidth and exchange interactions in this strongly spin-orbit-coupled system.

Figure 1

Schematic showing the nonequivalent Ir and Ba positions Ba1 (red), Ba2 (magenta), Ba3 (light pink), Ir1 (green), Ir2 (light green), Ir3 (cyan), and Ir4 (dark green). There are also six nonequivalent O positions, but they all have the same color for the sake of clarity. Note that several polytypes or polymorphs of ${\mathrm{BaIrO}}_3$ may be formed, including multiphased samples containing a mixture of polytypes, depending on the synthesis conditions [17]. The crystal structure of our ${\mathrm{BaIrO}}_3$ sample is similar to that of the nine-layered ${\mathrm{BaRuO}}_3$, a 9R-layer polytype, but with a monoclinic distortion.

Figure 2

Ir $L_{2,3}-$edge XANES spectrum of ${\mathrm{BaIrO}}_3$ recorded at room temperature and several external pressures up to $\sim 79\mathrm{GPa}$. Bottom: corresponding branching ratio.

Figure 3

Intensity of the Ir $L_3-$edge XMCD spectrum of ${\mathrm{BaIrO}}_3$ at $T=15\mathrm{K},H=0.4\mathrm{T}$ (after field cooling) as a function of the applied pressure. Inset: Ir $L_3-$edge XMCD spectrum of ${\mathrm{BaIrO}}_3$ at $T=15\mathrm{K},H=0.4\mathrm{T}$ (after field cooling) as a function of the applied pressure. Only some of the spectra have been plotted for the sake of clarity. The XMCD signal recorded after releasing the gas out of the DAC membrane (3.8-GPa remanent pressure, shown in the inset as a dashed line) does not recover its initial intensity but stays at zero. This suggests that the effect of the applied pressure on the magnetic behavior is not completely reversible.

Figure 4

(Top) Intensity of the Ir $L_3-$edge XMCD spectrum of ${\mathrm{BaIrO}}_3$ at $H=0.4\mathrm{T}$ (after field cooling) as a function of temperature for a few applied pressures. (Bottom) Magnetometry data at various pressures below 1 GPa with the inset showing the rate of $T_m$ suppression with pressure.

Figure 5

Pressure dependence of resistance in ${\mathrm{BaIrO}}_3$. Only a small decrease in resistance is observed up to 8.8 GPa, indicating the persistence of the insulating state across the suppression of the XMCD signal $(\sim 4.5\mathrm{GPa})$. Such an insulating state is further demonstrated by the negative slope of the temperature-dependent resistance observed at 2.3 and 7.9 GPa (inset).

Figure 6

Observed (dots, black) and calculated (red solid line) XRD Rietveld profiles for ${\mathrm{BaIrO}}_3$ at RT and 1.5 GPa. The first series of Bragg reflections correspond to the ${\mathrm{BaIrO}}_3$ phase, and the second series correspond to the Ag powder used for calibration. Peaks marked with asterisks correspond to the DAC and do not change with pressure. The intensities of some of the peaks are not perfectly matched in the fit, likely due to a limited number of powder grains in the focused x-ray beam resulting in less than ideal powder averaging. The inset shows the shift of one of the reflections with increasing pressure.

Figure 7

The lattice parameters $a$ and $b$ and the $c$ axis (left axis) and volume (right axis) of the unit cell as a function of the applied pressure for ${\mathrm{BaIrO}}_3$ at room temperature. The lattice parameters after releasing the pressure (data at 6.5 GPa) are also included. The values of $a,b$, and $c$ after releasing the gas in the cell indicate that there is no irreversible (plastic) structural change.

Figure 8

(Top) Doping (Sr) dependence of the Fourier-transformed EXAFS data. (Middle) feff8 simulations of the effect of doping Sr at Ba sites on the EXAFS signal arising from Ir-Ba/Sr scattering. $x=0$ (black lines), $x=0.06$ (red lines), and $x=0.12$ (green lines). (Bottom) feff8 simulations of the effect of the changes in the Ir-O-Ir buckling angle on the Fourier transform for $\theta =0^{\circ }$ (black lines), ${10}^{\circ }$ (red lines), ${15}^{\circ }$ (green lines), ${19}^{\circ }$ (blue lines), and ${25}^{\circ }$ (orange lines).

Figure 9

(Left panel) Intensity of the Ir $L_3-$edge XMCD spectrum of ${\mathrm{BaIrO}}_3$ as a function of the applied pressure $(T=15\mathrm{K}$ and $H=0.4\mathrm{T}$ after field cooling) and $\left(\mathrm{Ba},\mathrm{Sr}\right){\mathrm{IrO}}_3$ as a function of the Sr doping $(T=5\mathrm{K}$ and $H=0.4\mathrm{T}$ after field cooling). (Right panel) Modification of the $T_m$ with applied pressure (circles) and Sr doping (squares). Red and black data were obtained from XMCD measurements, gray data and magenta data correspond to superconducting quantum interference device measurements (this paper and Ref. [7], respectively). Dotted lines are guides for the eye.