Ultrahigh energy storage density in lead-free antiferroelectric rare-earth-substituted bismuth ferrite

Dielectric capacitors hold an enormous advantage for energy storage that requires a fast charging/discharging rate; but relatively low energy capacity is a key limitation for conventional dielectric materials. Recently, design strategies by tuning the ferroelectric ${\mathrm{BiFeO}}_3$ (BFO) to an antiferroelectric or relaxor have shown great promise, especially owing to the large polarization at high electric field. Here, using a first-principles-based method, it is predicted that rare-earth substitution of varied elements and composition can systematically tune the stability of the antiferroelectric phase, leading to further enhanced energy densities and high efficiencies, for instance, reaching 239 J ${\mathrm{cm}}^{-3}$ in thulium substituted BFO and generally beyond 80%, respectively. The storage performance is further interpreted based on a simple model, pointing to the importance of transition fields, polarization of the ferroelectric state, and dielectric constant.

Figure 1

Structures of ${\mathrm{Bi}}_{1-x}R{}_x{\mathrm{FeO}}_3$ under zero and applied electric field. (a) The antiferroelectric orthorhombic Pnma phase, characterized by the antipolar distortions along the pseudocubic $\left[110\right]$ direction, and the $a^{-}{a}^{-}{c}^{+}$ octahedral tilting pattern. (b) The ferroelectric tetragonal P4mm phase, characterized by the polarization along the $\left[001\right]$ direction with no octahedral tiltings. (c) The ferroelectric orthorhombic Amm2 phase, characterized by the polarization along the $\left[110\right]$ direction with no octahedral tiltings. (d) The ferroelectric rhombohedral R3c phase, characterized by the polarization and $a^{-}{a}^{-}{a}^{-}$ tiltings along the $\left[111\right]$ direction. The arrows denote atomic displacements.

Figure 2

Calculated unipolar $P\text{−}E$ hysteresis curves of ${\mathrm{Bi}}_{1-x}{\mathrm{Sm}}_x{\mathrm{FeO}}_3$ solid solutions for compositions ranging between 0.2 and 1.0 under four different electric field orientations, viz., (a) [001], (b) [100], (c) [110], (d) [111]. The displayed polarization is the projected component of the total polarization along the direction of the applied $E$ field. The initial state is the antiferroelectric Pnma phase. Only curves with approximately recoverable behaviors are shown.

Figure 3

The calculated energy densities of ${\mathrm{Bi}}_{1-x}{R}_x{\mathrm{FeO}}_3$ (R=La, Nd, Sm, Gd, Dy, or Tm) solid solutions at various compositions ($x$=0.2–1.0) under four different electric field orientations, viz., (a) [001]. (b) [100]. (c) [110]. (d) [111], with the maximum applied $E$ field ($E_{\text{max}}$) being 4.37 MV ${\mathrm{cm}}^{-1}$.

Figure 4

The calculated efficiencies of ${\mathrm{Bi}}_{1-x}{R}_x{\mathrm{FeO}}_3$ (R=La, Nd, Sm, Gd, Dy, or Tm) solid solutions at various compositions ($x$=0.2–1.0) under four different electric field orientations, viz., (a) [001], (b) [100], (c) [110], (d) [111], with the maximum applied $E$-field ($E_{\text{max}}$) being 4.37 MV ${\mathrm{cm}}^{-1}$.

Figure 5

(a) Schematic unipolar double hysteresis $P\text{−}E$ curve of antiferroelectrics. $E_{\text{up}}$ and $E_{\text{down}}$ are the abrupt AFE-to-FE and FE-to-AFE transition fields, respectively. (b) Calculated energy density and efficiency of representative ${\mathrm{Bi}}_{1-x}{R}_x{\mathrm{FeO}}_3$ (R=La, Nd, Sm, Gd, Dy, or Tm) solid solutions as a function of electric field.