Pressure tuning of structure, magnetic frustration, and carrier conduction in the Kitaev spin liquid candidate Cu2IrO3

The layered honeycomb lattice iridate ${\mathrm{Cu}}₂{\mathrm{IrO}}₃$ is the closest realization of the Kitaev quantum spin liquid, primarily due to the enhanced interlayer separation and nearly ideal honeycomb lattice. We report pressure-induced structural evolution of ${\mathrm{Cu}}₂{\mathrm{IrO}}₃$ by powder x-ray diffraction (PXRD) up to $\ensuremath{\sim}17$ GPa and Raman scattering measurements up to $\ensuremath{\sim}25$ GPa. A structural phase transition (monoclinic $C2/c\phantom{\rule{0.222222em}{0ex}}\ensuremath{\rightarrow}$ triclinic $P\overline{1}$) is observed with a broad mixed phase pressure range ($\ensuremath{\sim}4$ to 15 GPa). The triclinic phase consists of heavily distorted honeycomb lattice with Ir-Ir dimer formation and a collapsed interlayer separation. In the stability range of the low-pressure monoclinic phase, structural evolution maintains the Kitaev configuration up to 4 GPa. This is supported by the observed enhanced magnetic frustration in dc susceptibility without emergence of any magnetic ordering and an enhanced dynamic Raman susceptibility. High-pressure resistance measurements up to 25 GPa in the temperature range 1.4--300 K show resilient nonmetallic $R(T)$ behavior with significantly reduced resistivity in the high-pressure phase. The Mott 3D variable-range-hopping conduction with much reduced characteristic energy scale ${T}₀$ suggests that the high-pressure phase is at the boundary of localized-itinerant crossover. First-principles density functional theoretical (DFT) analysis shows that monoclinic $P{2}₁/c$ phase of ${\mathrm{Cu}}₂{\mathrm{IrO}}₃$ is energetically lower than its $C2/c$ phase at ambient pressure and both the structures are consistent with experimental XRD pattern. Our analysis reveals structural transition from $P{2}₁/c$ to $P\overline{1}$ structure at 7 GPa in agreement with experiment and uncovers the interplay between oxidation states, spin, Ir bond dimerization and their relevance to electronic band gap.