Potential room-temperature multiferroicity in cupric oxide under high pressure

CuO, known to be multiferroic (MF) from $T_L=213\mathrm{K}$ to $T_N=230\mathrm{K}$ at ambient pressure, has been the subject of debates about its ability to exhibit multiferroicity at room temperature (RT) under high hydrostatic pressure. Here we address this question based on theoretical and experimental investigations. The influence of hydrostatic pressure on $T_L$ and $T_N$ has been estimated from ab initio calculations combined with classical Monte-Carlo simulations and a quasi-1D antiferromagnetic analytical model. From the experimental side, electric permittivity anomalies related to ferroelectric transitions have been followed with dielectric measurements on single crystals up to 6.1 GPa. We show that the temperature $T_N$ below which the MF state forms increases with pressure linearly to higher pressure that hitherto supposed, and indeed based on our calculations, should exceed RT above about 20 GPa.

Figure 1

Magnetic exchange couplings are considered in this study. The different interactions are highlighted by green arrows. Blue and red spheres indicate copper and oxygen atoms, respectively. Square planar environments are represented in blue.

Figure 2

Magnetic exchange interactions vs hydrostatic pressure. Positive and negative signs correspond to AFM and FM interactions, respectively.

Figure 3

(a) Diagonal and (b) top view of the dispersion of the magnetic ground state within the (${\mathit{q}}_{\mathit{x}},{\mathit{q}}_{\mathit{z}}$) plane at 0 GPa. The red color highlights highest energies, while dark blue regions evidence energy minima.

Figure 4

CMC results. (a) Electric polarization vs temperature for different pressure values. The electric polarization was estimated by $\mathit{P}\propto 〈{\mathit{e}}_{\mathit{ij}}\times ({\mathit{S}}_{\mathit{i}}\times {\mathit{S}}_{\mathit{j}})〉$. The gray dashed line is the reference arbitrarily chosen to estimate the MF phase transitions. (b) MF phase diagram. $T_L$ and $T_N$ variations are represented by red and blue lines, respectively, and normalized at the estimated CMC $T_N$ value at 0 GPa.

Figure 5

Dielectric constant curves of single crystal CuO for $\mathit{E}$//$\mathit{b}$ from high-pressure capacitance measurements with increasing temperature. The inset shows the region of the two magnetic phase transitions for 1.4 GPa, indicated by the arrows. The upper transition ($T_N$) appears as a sharp step reproducible on warming and cooling. The lower transition ($T_L$) appears as a much smaller anomaly. The region between $T_L$ and $T_N$ shows marked hysteresis between the cooling and the warming cycles, with irreproducible behavior on cooling.

Figure 6

Zoom of the dielectric constant curves shown in Fig. 5 after subtraction of an arbitrary polynomial background to reveal $T_N$ and $T_L$ transitions evidenced by arrows. Curves have been shifted for clarity.

Figure 7

(a) Temperature-pressure phase diagram of CuO. (a) Experimental values extracted from our dielectric measurements (red plain circles) and from Chatterji et al. [8] (red open diamonds). Our theoretical estimated values from CMC and the analytical expression [Eq. (4)] are represented by black lines and blue dashed lines, respectively. Our CMC simulations have been adjusted to fit the experimental Néel temperature at 0 GPa. (b) Zoom in the pressure range from 0 to 7 GPa to emphasize the experimental/theoretical deviations.