Lead-free La₂Ti₂O₇ dielectric ceramics with ultra-high energy storage density and electric field durability through layered ferroelectric layers

Teng Sui^a, Tao Wang^a, Qin Feng*^ab, Changlai Yuan^b, He Qi^c, Nengneng Luo*^a, Xiyong Chen^a, Zhenyong Cen^a and Jiwei Zhai*^d
aState Key Laboratory of Featured Metal Materials and Life-Cycle Safety for Composite Structures, School of Resources, Environment and Materials, Guangxi University, Nanning 530004, China. E-mail: fengqin307@163.com; luonn1234@163.com
^bGuangxi Key Laboratory of Information Materials, School of Material Science and Engineering, Guilin University of Electronic Technology, Guilin 541004, PR China
^cBeijing Advanced Innovation Center for Materials Genome Engineering, Department of Physical Chemistry, University of Science and Technology Beijing, Beijing 100083, P. R. China
^dKey Laboratory of Advanced Civil Engineering Materials of Ministry of Education, Functional Materials Research Laboratory, School of Materials Science & Engineering, Tongji University, No. 4800 Caoan Highway, Shanghai 201804, China. E-mail: apzhai@tongji.edu.cn

First published on 19th August 2025

---

---

1 Introduction

Energy storage ceramic materials find critical applications in military impulse systems, aerospace engineering, and automotive manufacturing due to their rapid charge/discharge kinetics, superior stability, and exceptional power density. These materials are extensively employed in advanced power systems for transportation vehicles, precision electronic circuits, LiDAR technology, and directed-energy weapons.^1–3 The recent paradigm shift in pulse technology applications particularly in electromagnetic armaments, geophysical exploration, medical devices, and electric tank cannon (ETC) systems has imposed increasingly stringent performance requirements on ceramic energy storage materials. Key technical demands now include compact dimensions, enhanced voltage tolerance, ultrafast energy cycling capabilities, and optimized power conversion efficiency.^4,5 Consequently, the development of next-generation ceramic energy storage materials must prioritize simultaneous achievement of miniaturization, high-voltage resilience, rapid energy transfer rates, and ultrahigh power density.^6–8

The effective energy storage density W_rec, the total energy storage density W_tot, and the energy storage efficiency η of energy storage ceramics can be calculated using the following equations.^9,10

	(1)

	(2)

	(3)

In this formulation, P denotes the polarization intensity, where P_r represents the residual polarization, P_max corresponds to the maximum polarization intensity, and E signifies the applied electric field strength. The energy storage density (W_rec) can be optimized through simultaneous maximization of P_max, minimization of P_r, and elevation of the breakdown field strength (E_b) defined as the maximum sustainable electric field prior to dielectric failure. This relationship demonstrates that substrates with superior E_b enable the fabrication of energy storage ceramics exhibiting enhanced dielectric strength and elevated energy storage capacity. Current mainstream ceramic systems, including Bi_0.5Na_0.5TiO₃-based, BaTiO₃-based, BiFeO₃-based, and NaNbO₃-based compositions, are generally constrained by relatively low E_b values (100–500 kV cm⁻¹), fundamentally limiting their energy storage performance. Furthermore, the prevailing strategy of designing complex multicomponent ceramics to induce strong relaxor characteristics inevitably compromises manufacturing scalability.

The intrinsic breakdown field strength (E_b) of dielectric materials is fundamentally governed by their bandgap energy (E_g). A larger E_g establishes a higher potential barrier for electron transitions between valence and conduction bands, thereby enhancing E_b. Three additional microstructural factors critically influence E_b: reduced dielectric permittivity, refined grain dimensions, and highly dense microstructures with uniform phase distribution. Conventional perovskite ceramics exhibit intrinsically limited E_b values due to their characteristically narrow bandgaps and elevated dielectric constants, as extensively documented in prior studies.^11–13 This fundamental limitation has driven the exploration of alternative ferroelectric systems combining wide bandgaps with low-loss characteristics, including tungsten bronze-type architectures and bismuth-layered structured materials.^14,15

Our comprehensive literature investigation identified La₂Ti₂O₇ (LTO) as a compositionally simple ferroelectric material exhibiting low dielectric loss (tanδ) and wide bandgap energy (E_g).^16,17 The substantial E_g (∼3.81 eV) suggests intrinsic potential for enhanced breakdown strength (E_b), enabling superior voltage tolerance and broader operational applicability. A review of the literature reveals that the structure of LTO is a layered perovskite structure (Fig. 1(a)), with the cells arranged in a 2 × 2 × 2 stacked configuration. The crystal structure comprises corner-sharing TiO₆ octahedra coordinated with La³⁺ cations. Each structural layer contains four interconnected TiO₆ units bridged by apical oxygen atoms (red spheres), with La cations occupying distinct crystallographic sites: interlayer positions (green spheres) between perovskite slabs and interstitial sites (white spheres) within octahedral frameworks. This stratified configuration enhances dielectric insulation through restricted interlayer charge transport, thereby achieving exceptional E_b. At the same time, the layered topology allows for a high intrinsic polarization of LTO ceramics, where a high intrinsic polarization strength (P_s) is the basis for obtaining a high P_m and a strong electric field is necessary to overcome various energy barriers, including those for 90° domain wall shifts, to achieve a P_m close to P_s. Furthermore, the inherent TiO₆ octahedral rotational distortion explains the ferroelectric behavior observed in LTO ceramics.

Contemporary research on La₂Ti₂O₇ (LTO) has predominantly targeted luminescence enhancement via rare-earth ion doping (Pr³⁺/Tb³⁺).¹⁸ Intriguingly, LTO has emerged as a promising room-temperature multiferroic material, demonstrating concurrent ferroelectric and weak ferromagnetic ordering.¹⁹ Dielectric property optimization studies have focused on Bi³⁺ substitution strategies to enhance permittivity while suppressing dielectric loss (tanδ).²⁰ Paradoxically, LTO's photocatalytic applications are constrained by its wide bandgap (E_g ≈ 3.81 eV), limiting optical activation to UV wavelengths. This inherent limitation has driven intensive bandgap engineering efforts through nitrogen doping and composite formation to extend visible-light responsiveness and improve photocatalytic efficiency.²¹ Crucially for energy storage applications, LTO's characteristically large E_g traditionally viewed as a photocatalytic disadvantage provides unique dielectric advantages. This fundamental material property enables the development of ceramics with exceptional intrinsic breakdown strength (E_b), forming the cornerstone of our innovative approach.

To the best of our knowledge, no systematic investigations have been reported on La₂Ti₂O₇ (LTO) ceramics as novel lead-free dielectrics for high-field energy storage applications. While preliminary studies have incidentally observed energy storage characteristics in LTO-based flexible freestanding thin films,²² these findings remain complementary rather than contradictory to our core scientific objectives. This pioneering investigation establishes three critical breakthroughs: (1) experimental verification of LTO's exceptional breakdown strength (E_b > 1100 kV cm⁻¹) enabled by its wide bandgap (E_g ≈ 3.81 eV); (2) development of phase-pure LTO ceramics via cost-effective solid-state reaction, achieving optimized microstructure through sintering protocol refinement (1250 °C/4 h); (3) as shown in Fig. 1, demonstration of record-high energy storage performance (W_rec = 8.83 J cm⁻³) with minimal polarization loss (P_m = 19.98 μC cm⁻²). And it exceeds the performance of some other classical systems. Microstructural analysis reveals some synergistic enhancement mechanisms (Fig. 1(e)–(g)): submicron-grained microstructure with uniform density distribution minimizes electric field concentration; stable Ti⁴⁺ valence state and wide E_g collectively suppress charge carrier migration. These combined attributes yield E_b values 2–3 times superior to conventional simple-composition ferroelectrics (BNT/BT/BF/NN systems), positioning LTO as a transformative candidate for next-generation high-voltage capacitor technologies.

2 Experimental section

2.1 Sample preparation

LTO ceramics were produced through the traditional solid-state reaction technique. Appropriate amounts of La₂O₃ (99.99%) and TiO₂ (98.00%) (Sinopharm Holding Chemical Reagent Co., Ltd, China) were dried and utilized as raw materials concurrently. Anhydrous ethanol was mixed with the powder as a dispersant and ground in a planetary ball mill at 300 rpm for 12 hours. After drying in an oven, the mixed powder was heated from room temperature to 850 °C at a rate of 5 °C min⁻¹, followed by calcination at 850 °C for 4 hours. The powder was then subjected to ball milling and drying under the same conditions. Polyvinyl alcohol (PVA) was incorporated into the dried powder to assist with granulation and grinding. The resulting mixture was then subjected to a pressure of 10 MPa for 90 seconds, forming a circular block with dimensions of 8 mm in diameter and 1 mm in thickness. The small block was subsequently heated to 600 °C for 2.5 hours (with a heating rate of 3 °C min⁻¹ from room temperature to 600 °C) and sintered at 1100–1300 °C for 3.5 hours in an air atmosphere. The sintering protocol employed a staged heating approach, with temperature elevation rates carefully modulated to 5 °C min⁻¹ until reaching 1000 °C, followed by a reduced ramp of 3 °C min⁻¹ beyond this threshold.

2.2 Structural characteristics

The prepared samples were ground into a fine powder and passed through a 200-mesh screen to characterize their phase structure using an X-ray powder diffractometer (Rigaku D/MAX 2500 V, Rigaku, Japan) with Cu Kα radiation. The obtained Raman spectra were analyzed using a laser Raman spectrometer (LabRAM HR Evolution). To observe the surface morphology of the samples, a field emission scanning electron microscope (FEI Quattro S) was employed; both sides of the samples were polished, heat-etched at 1200 °C for 30 minutes, and then coated with gold electrodes. Micro-morphological observations and selective area electron diffraction (SAED) imaging were conducted using a 300 kV field emission transmission electron microscope (FEI TECNAI G2 F30). The energy band structure of LTO was calculated using first principles, employing the HSE06 density functional theory. The elemental composition and chemical state of the samples were analyzed using X-ray photoelectron spectroscopy (XPS) (Thermo ESCALAB 250Xi+, USA).

2.3 Electrical characteristics

The polished specimens were subjected to symmetrical silver electrode deposition via screen-printing technique. Dielectric characterization was performed using a precision impedance analyzer (E4990A, Keysight Technologies, Santa Clara, CA, USA) in the frequency range of 1–500 kHz. For ferroelectric evaluation, specimens were mechanically thinned to 70 ± 5 μm thickness through sequential polishing. A patterned electrode configuration was implemented, featuring a circular silver contact (Φ = 1 mm, effective area = 0.785 mm²) on the obverse surface and full-area metallization on the reverse surface. Polarization–electric field (P–E) hysteresis loops and current–electric field (I–E) characteristics were acquired using a ferroelectric test system (TF Analyzer 3000, aixACCT, Aachen, Germany) at 10 Hz excitation frequency. Transient discharge behavior was quantified through overdamped pulsed discharge measurements utilizing a specialized charge–discharge analyzer (PK-CPR1801-10015, PolyK Technology, PA, USA). Optical bandgap determination was achieved via UV-vis-NIR spectrophotometry (UV-3600Plus, Shimadzu) following mechanical grinding into fine powders with particle size < 100 μm.

3 Results and discussion

X-ray diffraction analysis was systematically conducted on air-sintered samples across varying temperatures (1250–1400 °C), as shown in Fig. 2(a) and S1a. All diffractograms display well-defined peaks corresponding to the layered perovskite structure (JCPDS 28-0517), confirming phase purity and structural stability throughout the sintering regime. Rietveld structural refinement using GSAS-II software (Fig. 2(b), S1(b) and (c)) revealed a monoclinic phase (space group P2₁) with refined lattice parameters: a = 7.812 Å, b = 13.010 Å, c = 5.545 Å. The excellent agreement between experimental and calculated patterns is quantified by convergence factors (R_wp < 10%, χ² < 5), with detailed refinement parameters provided in SI Table S1. This rigorous structural analysis validates the phase homogeneity and crystallographic integrity of the synthesized ceramics.

Room-temperature Raman spectroscopy (100–900 cm⁻¹) of LTO ceramics was systematically acquired (Fig. 2(c)). The spectrum exhibits broadened and overlapping phonon modes characteristic of complex crystalline environments. Thirteen distinct Raman-active phonon modes were resolved at 115 cm⁻¹, 155 cm⁻¹, 184 cm⁻¹, 241 cm⁻¹, 346 cm⁻¹, 405 cm⁻¹, 430 cm⁻¹, 451 cm⁻¹, 520 cm⁻¹, 542 cm⁻¹, 560 cm⁻¹, 610 cm⁻¹, and 796 cm⁻¹, corresponding to the monoclinic P2₁ symmetry. Three distinct vibrational regimes were identified: low-frequency modes (<200 cm⁻¹): dominated by interlayer La³⁺ cation displacements (115, 155, 184 cm⁻¹); intermediate modes (200–400 cm⁻¹): attributed to La–O bond bending vibrations (241, 346, 405 cm⁻¹); high-frequency modes (>400 cm⁻¹): originating from symmetric stretching vibrations within the TiO₆ octahedral framework along O–Ti–O chains (430–796 cm⁻¹).

This vibrational hierarchy reflects the layered perovskite architecture, where low-energy modes arise from interlayer interactions while high-energy modes correlate with intra-octahedral distortions.

The temperature-dependent dielectric properties of LTO ceramics were systematically investigated across wide frequency (1 kHz to 1 MHz) and temperature ranges (25–300 °C), as shown in Fig. S2. Both dielectric permittivity (ε_r) and loss tangent (tanδ) demonstrate exceptional thermal stability at low temperatures (<150 °C), exhibiting minimal temperature/frequency dependence with ε_r = 104–130 and tanδ < 0.05. This ultrastable dielectric behavior maintaining ultralow loss coupled with moderate permittivity persists throughout the operational temperature range, fulfilling critical requirements for high-field energy storage applications. The material's unique combination of temperature-insensitive dielectric response and ultralow dissipation positions it as an ideal dielectric for capacitive energy storage systems.²³

To elucidate the microstructure–property relationships in LTO ceramics, the samples were subjected to transmission electron microscopy (TEM) analysis. The grain boundaries appear well-defined and densely packed (Fig. 2(d)). A closer examination of the grain boundaries is presented (Fig. 2(e)), revealing the presence of atomic-level bonding, indicating a high level of density. Furthermore, the observation of Moire fringes (Fig. 2(f)) suggests an interference pattern resulting from the overlap of two different lattice orientations, further confirming the strong and tight bonding at the grain boundaries. Additionally, this lattice interaction leads to local polar lattice distortion, facilitating the formation of ferroelectric domains. The LTO samples have good crystalline properties, indicating lattice spacing and crystal orientation (Fig. 2(g)). The selective area electron diffraction (SAED) images obtained along various crystallographic axes (Fig. 2(h)–(k)) confirm the monoclinic and layered structure of LTO, as previously deduced from X-ray diffraction analysis. Furthermore, the distinctive morphology of ferroelectric domains is depicted (Fig. 2(l)–(o)). It is notable that LTO exhibits relatively large long-range ferroelectric stripe domains, alongside 90° striped domains. In general, 90° ferroelectric domains are less active and less likely to produce high polarization, but at high electric fields, 90° ferroelectric domains still have the potential to produce high P_m. The localized magnification (Fig. 2(n)) highlights the presence of domain walls. These unique orientations of ferroelectric domains significantly contribute to the ferroelectric response of LTO, consequently yielding a high P_m and low P_r under an external electric field. This attribute is beneficial for achieving high energy storage properties.

Systematic microstructural evolution of LTO ceramics was investigated through SEM analysis combined with thermal etching simulations across sintering temperatures (1100–1300 °C), as shown in Fig. S3. The samples sintered at 1150 °C showed submicron grains with high porosity, which seriously affected the dielectric reliability. In contrast, optimal sintering at 1250 °C yields fully dense microstructures featuring uniform grain growth, sharp grain boundaries, and negligible porosity. This defect-engineered microstructure enables record-breaking breakdown field strength (E_b = 1100 kV cm⁻¹) through effective mitigation of electric field concentration hotspots.

To establish microstructure–property correlations in LTO ceramics, multiphysics finite element analysis (FEA) was employed to simulate the evolution of electric trees, dielectric constants, potential, and local electric field distribution of LTO samples sintered at different temperatures. The outcomes demonstrate that this method effectively integrates considerations of grain, grain boundary, and intrinsic dielectric constants to realistically simulate the experimental findings (Fig. 3 and S4, SI).

In addition, the grain size and dielectric constant of LTO ceramics at different sintering temperatures were recorded, which showed that the permittivity (ε_r) of the grains was obviously greater than that of the grain boundaries, indicating the resistance of the grain boundaries to the growth of electric trees (Table S2, SI). Further examination (Fig. 3(a)–(c)) illustrates that over time, electric tree growth predominantly occurs within the grains, where the electric field is lower, facilitating controlled growth and mitigating potential disruptions. Conversely, the grain boundaries serve as focal points for high electric field concentration, filtering the electric potential and impeding electric tree expansion. A comparative analysis with samples sintered at other temperatures (Fig. S4, SI) underscores that samples sintered at 1250 °C exhibit slow and restrained electric tree growth, characterized by well-defined, compact grains that rely on dense grain boundaries to inhibit the electric tree growth. Moreover, these samples manifest fewer regions of high potential and high electric field at the grain boundaries, resulting in a more uniform potential and electric field distribution, ultimately enhancing the breakdown field (E_b) of LTO ceramics and augmenting their energy storage performance.

The ultraviolet-visible (UV-vis) absorption spectra and corresponding bandgap energy (E_g) values of LTO ceramics were analyzed, as depicted in Fig. 4(a) and (b). The E_g value is indicative of the energy barrier for electron transition from the valence band to the conduction band. A higher E_g signifies a greater energy requirement for such transitions, thus hindering electron mobility and enhancing the ceramic's breakdown electric field strength. Deng recently demonstrated an inverse correlation between dopant concentration and bandgap in Co/Nb-modified Bi₀.₅Na₀.₅TiO₃ ceramics, accompanied by substantially elevated leakage currents (>10³ times higher than undoped counterparts).²⁴ The E_g values were calculated using the Tauc equation in conjunction with the UV absorption spectrum.^16,24

(αhν)m = A(hν − Eg)	(4)

In the Tauc equation, A denotes the absorbance of UV-visible diffuse reflection, α represents the absorption coefficient, and hν is the photon energy. For direct semiconductors like LTO, the value of m is 2, whereas for indirect semiconductors, m is 1/2. LTO ceramics exhibit an E_g of 3.81 eV, exceeding those observed in other well-studied systems such as BNT.^24,25 This finding highlights the exceptional breakdown field strength and favorable energy storage properties of LTO ceramics.

To further elucidate the high E_g of LTO ceramics, the energy band structure was computed using a hybrid functional approach. The calculations employed spin-polarized density-functional theory (DFT) to ensure accuracy.^26,27 The projected augmented wave method, based on plane wave basis sets, was implemented in the Vienna ab initio simulation package (VASP).^28,29 Exchange–correlation potentials were determined using a combination of PBE and the generalized gradient approximation (GGA),³⁰ while the DFT-D3 model of Grimme was applied with van der Waals correction.³¹ A cutoff energy of 600 eV was used, and a Γ-centered Monkhorst–Pack grid with dimensions 3 × 2 × 5 was selected to sample the Brillouin zone integrals.³² To maintain a maximum force below 0.01 eV Å⁻¹ for each atom, a fully relaxed crystal structure was utilized. The energy convergence criterion was set at 10⁻⁶ eV. The energy band structure was calculated using the Heyd–Scuseria–Ernzerhof (HSE06) method.³³ The energy band structure and density of states are depicted in Fig. 4(c). The calculated E_g of LTO was 4.44 eV, which is higher than the experimental value. This discrepancy is attributed to the high-precision generalized HSE06 method's tendency to occasionally overestimate the bandgap. Although there is a deviation from the actual measured value. However, this analysis also confirms the high bandgap of LTO. Nevertheless, this analysis confirms the high bandgap of LTO. Examination of the figure reveals that the valence band maximum (VBM) and conduction band minimum (CBM) are closely located at the same wavevector k. For direct semiconductors like LTO, the k offset is minimal, indicating that electrons in the valence band are unlikely to transition to the conduction band. This property enables LTO to withstand a higher breakdown field strength.

To evaluate the breakdown electric field (E_b) of LTO ceramics, the experimental datasets were analyzed by implementing the Weibull probability model (Fig. 4(g)). This statistical framework facilitated the derivation of E_b values, aligning with established methodologies for interpreting material failure thresholds under electrical stress. The computational procedure involved systematic parameter optimization to ensure congruence between theoretical distributions and empirical observations, as documented in the graphical representation provided:^10,34

Xi = lnEi	(5)

Yi = ln(ln(1/1 − Pi))	(6)

Pi = i/(n − 1)	(7)

In this context, n signifies the complete count of samples, while i indicates the specific sample number. E_i refers to the breakdown field (E_b) of the sample, arranged in increasing order according to the sample number. Furthermore, P_i denotes the likelihood of the sample being deemed invalid. Applying this formula, the calculated value for E_b is 1130 V, which closely approximates the actual value. Furthermore, the Weibull modulus β = 17.18 indicates a high level of confidence in the E_b data from the samples.

Impedance analysis of LTO ceramics was carried out in order to study in depth the effect of resistance on the ceramic characteristics. Impedance diagrams were generated at high temperatures between 480 °C and 540 °C, with frequencies spanning from 100 Hz to 1 MHz (Fig. 4(d)). Within these diagrams, Z′ denotes the real part of the impedance, while Z′′ signifies the imaginary component. The Nyquist plot clearly illustrates that the impedance curve of LTO ceramics exhibits a substantial radius. This characteristic suggests significant resistance and robust insulating qualities, advantageous for functioning at elevated breakdown field strengths. The curves form approximately semicircular arcs, and the inset illustrates equivalent circuit diagrams fitted using ZView software, based on the EIS data, indicating that the high resistance of LTO is primarily due to the grains and grain boundaries of the substrate. Additionally, LTO exhibits small grain size with dense and well-defined grain boundaries, underscoring the significant role of grain boundaries and grains in achieving high breakdown field strength (Fig. 3, S3 and S4, SI). Grain boundaries offer substantial resistance to the propagation of electrical trees, with the most significant contribution arising from these boundaries. As temperature increases, the arc radius of LTO ceramics gradually decreases, and the curve's center shifts towards lower frequencies, indicating thermally-induced relaxation phenomena and increased dielectric loss. Moreover, the variation of Z′′ with frequency displays a single Debye peak, which broadens and shifts to higher frequencies as temperature rises (Fig. 4(e)). This behavior suggests that LTO ceramics exhibit thermally-activated relaxation phenomena, potentially associated with oxygen loss in the ceramic samples and an increased presence of oxygen vacancies. The resistance R of the grains and grain boundaries of LTO follows the thermally-activated Arrhenius law:³⁵

	(8)

The resistance R, the prefinger factor R₀, the activation energy of conductivity E_a, the absolute temperature T, and the Boltzmann constant k_B = 8.62 × 10⁻⁵ eV K⁻¹ are related by the thermally-activated Arrhenius' law. The Arrhenius plots of the grains and grain boundaries and the activation energies computed by linear fitting of the LTO samples are presented (Fig. 4(f)). The activation energy reflects the energy required for the potential barrier jump in the sample, and a higher activation energy is more beneficial for achieving a large breakdown field strength. From the figure, it can be observed that the activation energies of the grains and grain boundaries are 1.08 eV and 1.10 eV, respectively. The observation that the activation energy of the grains is lower than that of the grain boundaries confirms the viewpoint that grain boundaries contribute more to the high breakdown field strength than grains. Furthermore, stronger insulation at the grain boundaries correlates with higher activation energy. By leveraging the disparity in conductivity activation energy between grains and grain boundaries, ΔE_a can be employed to elucidate charge transport across the grain boundary laminae. The observed difference in activation energy signifies a substantial variation in charge transport properties between grain boundaries and grains. This disparity leads to charge accumulation at the grain boundaries, which exhibit high resistivity, resulting in interface polarization and enhanced polarization strength in LTO.

The elemental composition and oxidation states of LTO ceramics were analyzed via temperature-dependent X-ray photoelectron spectroscopy (XPS), employing the C 1s peak for binding energy calibration. This methodological approach enabled precise characterization of valence configurations under varying thermal conditions, adhering to standardized protocols for surface-sensitive chemical analysis. The total elemental XPS spectra of LTO ceramics from 25 to 400 °C and the high-resolution XPS spectra of Ti 2p are shown (Fig. 4(h) and (i)). The sample is composed solely of three elements: La, Ti, and O, with no impurities present (C was included for calibration). This result is consistent with the XRD pattern (Fig. 2(a)), confirming the successful synthesis of LTO ceramics. The Ti 2p XPS spectra display well-defined symmetrical peaks for Ti 2p_1/2 and Ti 2p_3/2 at room temperature, with binding energies of 463.96 eV and 458.16 eV, respectively. The separation between Ti 2p_1/2 and Ti 2p_3/2 peaks is approximately 5.80 eV, indicating the presence of the Ti⁴⁺ state.^36,37 As the temperature rises, two new peaks emerge at 456.01 eV and 461.06 eV at 200 °C, corresponding to the Ti³⁺ state.³⁷ This change is attributed to the loss of oxygen atoms from the TiO₆ octahedrons at elevated temperatures, leading to oxygen vacancies and a reduction in the valence state of some Ti ions. At 400 °C, further oxygen loss occurs, resulting in the appearance of a new Ti valence state. Two additional peaks at approximately 455.50 eV and 460.55 eV are observed, indicating the presence of the Ti²⁺ state.^38,39 Additionally, the Ti 2p peak shifts towards higher binding energies with increasing temperature (Fig. 4(i)). This shift is due to oxygen loss in the LTO system, the formation of oxygen vacancies, and changes in the chemical states of Ti and O. It confirms the presence of Ti³⁺ and Ti²⁺ valence states and corroborates the changes observed in the impedance diagram. A summary of the changes in Ti element valence states with temperature is provided (Table S3, SI). The data indicate that the Ti element exhibits excellent temperature stability at low temperatures, maintaining its valence without excessive oxygen vacancy generation within a specific temperature range. This characteristic imparts the substrate with good insulating properties and high breakdown field strength.

The unipolar P–E characteristics of LTO ceramics were evaluated under various electric fields at 20 °C and 10 Hz (Fig. 5(a)), presenting a standard hysteresis loop diagram typical of ferroelectrics. Fig. 5(b) illustrates the corresponding P_m–P_r plot. LTO ceramics demonstrate a high breakdown field (E_b) of approximately 1100 kV cm⁻¹, a high maximum polarization (P_m) of around 19.98 μC cm⁻², and a low remnant polarization (P_r) of about 1.31 μC cm⁻², indicating favorable energy storage properties. Under an electric field of 100 kV cm⁻¹, the P_r of LTO ceramics is remarkably low at only 0.01 μC cm⁻², while P_m measures 1.11 μC cm⁻². As the applied electric field increases, P_m significantly rises to 19.98 μC cm⁻², while P_r increases to 1.31 μC cm⁻² but remains relatively low, resulting in decreased energy storage efficiency. The monopole I–E curve (Fig. 5(c)) under different electric fields at 20 °C and 10 Hz shows minor changes only when the applied electric field exceeds 500 kV cm⁻¹. The P–E curve displays approximate linearity, influenced by the presence of small ferroelectric domains within the substrate. These domains can rapidly respond to the applied electric field, leading to quick domain flipping and a low P_r. However, during this period, polarization intensity growth slows. With further increase in the applied electric field, a prominent current peak gradually emerges, and the P–E loop begins to exhibit hysteresis, indicating the expansion of tiny ferroelectric domains under strong electric fields. The response to the electric field becomes somewhat hysteretic compared to earlier, resulting in an increase in P_r, substantial growth in polarization intensity, and a decrease in energy storage efficiency (η). However, the newly formed long-range ferroelectric domains are not yet fully developed, allowing P_r to remain relatively low. The energy storage of the ceramic shows a gradual increment with rising electric field strength, highlighting its strong dependence on high electric fields (Fig. S5(a), SI). Ultimately, at a high electric field of 1100 kV cm⁻¹, the recoverable energy density (W_rec) reaches 8.83 J cm⁻³, while efficiency decreases from an initial 97% to 71%, indicating a reduction in efficiency but still maintaining satisfactory levels. Leakage current tests on LTO ceramics were also conducted (Fig. S5(b), SI). As the applied electric field increases, the leakage current of LTO ceramics only rises from 8.44 × 10⁻⁹ A cm⁻² to 5.67 × 10⁻⁸ A cm⁻². The extremely low leakage current density indicates excellent insulation performance, beneficial for achieving higher breakdown field strength.

The recently reported relationship between the recoverable energy density (W_rec) in energy storage ceramics and the number of cations in the system is illustrated in Fig. 5(d). It is evident that only two systems achieve W_rec values exceeding 10 J cm⁻³; however, their chemical compositions are exceedingly complex, and the preparation of such materials is both challenging and costly, making them unsuitable for large-scale industrial production. The electric field breakdown (E_b) values of various systems were calculated and compared, including those based on BF,^40–43 BNT-based,^44–47 BT-based,^48–51 and NN-based^52–55 (Fig. S5(c), SI). In contrast to the current scenario, where high energy storage properties require highly intricate chemical compositions, we have demonstrated that LTO ceramics can achieve both high W_rec and elevated E_b values using a simple chemical composition without the need for doping with other elements. Consequently, LTO ceramics exhibit significant potential for applications that demand simple chemical compositions to attain high values of W_rec and E_b.

Another advantage of dielectric capacitors is their exceptionally rapid charging and discharging rates, which result in significantly reduced charging and discharging times. The pulsed discharge characteristics of lithium titanate (LTO) ceramics were examined across multiple electric field intensities using a charge–discharge measurement system, as illustrated in Fig. 6(a). A fixed resistive load (R_L) of 10260 Ω was integrated into the experimental configuration to analyze current behavior during non-oscillatory energy release processes. According to the equation (W_d = W_tol(R_L + ESR)/R_L), it is evident that when the applied load resistance is much greater than the equivalent sample resistance (ESR), it facilitates efficient energy storage aggregation within the load resistance. However, an excessively large load resistor may hinder the discharge rate. Therefore, a moderate resistance of 10260 Ω was selected. Additionally, the variation curves of energy density (W_d) over time under different applied electric fields were plotted (Fig. 6(b)). The value of (W_d) can be obtained using eqn (9).²⁴

	(9)

Here, R_L = 10260 Ω, and V represents the volume of the sample tested. t_0.9 represents the time it takes to reach 90% of the maximum value on the discharge energy density curve under a line load, providing a visual representation of the discharge rate.

With the increase of the applied electric field from 100 kV cm⁻¹ to 600 kV cm⁻¹, it can be observed that t_0.9 values stay below 1.5 μs, indicating an ultra-high instantaneous discharge rate of the sample (Fig. 6(b) and S6(a), SI). Concurrently with the augmentation of the externally applied electric field strength, the observed energy density (W_d) demonstrates a significant increase, progressing from an initial value of 0.05 J cm⁻³ to a peak measurement of 1.95 J cm⁻³. The measured W_d values under different applied electric fields align closely with the W_rec values calculated from the P–E loop at the identical electric field. The slight discrepancies arise due to the energy storage loss resulting from the equivalent resistance of the sample within the line, which is an unavoidable and negligible loss.

To investigate the stability of LTO ceramics' energy storage characteristics under conditions beyond room temperature, several properties of LTO were tested at different temperatures. The XRD curves at different temperatures (Fig. 6(c)) indicate that LTO exhibits high structural stability without any structural changes. W_rec, W_tot, η (Fig. 6(d)), and P_m, P_r (Fig. 6(e)) remain unaffected by temperature fluctuations.

Under an applied electric field intensity of 600 kV cm⁻¹ and a cyclic frequency of 10 Hz, polarization–electric field hysteresis loops (P–E curves) were systematically evaluated across a broad spectrum of thermal conditions. It is noteworthy that the P–E curves of LTO ceramics within the range of 20–100 °C remain almost unchanged, indicating robust stability (Fig. 6(f)). Several properties of LTO at different frequencies were also tested. W_tot, P_m, and P_r exhibit a slight decrease with increasing frequency, while η increases from 74% to 84% with the rise in frequency. W_tot, P_m, P_r, and η all stabilize when reaching above 150 Hz (Fig. S6(b) and (c), SI), with W_rec maintaining high stability. The P–E curves over a wider frequency range were tested at 20 °C, and the curves exhibit minimal changes with increasing frequency (Fig. S6(d), SI). This demonstrates that LTO ceramics exhibit excellent temperature and frequency stability of their energy storage characteristics over a broader temperature and frequency range.

4 Conclusion

This study introduces and explores a novel lead-free layered perovskite, La₂Ti₂O₇ (LTO), as an energy storage ceramic for the first time. This material features a simple chemical composition, outstanding energy storage capabilities, and exceptionally high breakdown field strength. High-density, pure LTO ceramics were successfully synthesized through air sintering. Ferroelectric performance tests confirmed that under an extremely high electric field of 1100 kV cm⁻¹, LTO maintained a polarization (P_m) of 19.98 μC cm⁻² and achieved a recoverable energy density (W_rec) of 8.83 J cm⁻³ with an efficiency (η) of 71%. The high polarization strength of LTO ceramics is attributed to the oriented arrangement of ferroelectric domains and interface polarization under high electric fields. Additionally, the uniform grain distribution, high bandgap (∼3.81 eV), and stable valence state of Ti elements contribute to the high breakdown field strength. The remarkable energy storage characteristics are dependent on the extremely high electric field. Furthermore, during the over-damped pulse discharge test, LTO exhibited an exceptionally fast discharge rate (t_0.9 < 1.5 μs). The energy storage properties also demonstrated stability over a broad temperature range (20–100 °C) and a wide frequency range (10–200 Hz). The comprehensive energy storage characteristics of LTO ceramics surpass those of other energy storage ceramics with simple chemical compositions, underscoring the significant potential of LTO ceramics in applications such as easy production, miniaturization, high electric fields, high power density, and energy storage, including commercial defibrillators, microwave communication, and pulse power applications. Moreover, the temperature stability of LTO ceramics is relatively high, suggesting that future research could explore their properties and performance at higher temperatures, potentially extending applications to high-temperature fields such as the aerospace industry and deep well drilling. It is also feasible to enhance the polarization strength and energy storage efficiency of LTO ceramics under high electric fields by doping beneficial elements and employing strategies like high-entropy and domain engineering, with the aim of achieving superior energy storage performance. This study lays the groundwork for future scientific research on layered perovskite structural systems.

Author contributions

Teng Sui conducted most of the experiments, processed experimental data, and wrote the manuscript. Tao Wang participated in writing the manuscript as well as embellishing it. Qin Feng conceived the initial concept, supervised the project, and participated in manuscript writing. Changlai Yuan provided comments for manuscript writing. He Qi participated in measuring TEM, electrical tree simulation, and variable-temperature XRD, and provided comments for manuscript writing. Nengneng Luo assisted in measuring electrical properties and supervised the project. Xiyong Chen assisted in measuring electrical properties. Zhenyong Cen assisted in measuring electrical properties. Jiwei Zhai supervised the project, and participated in manuscript writing. All authors participated in the discussion and editing of the paper.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The data are available from the corresponding author on reasonable request.

Supplementary information is available. See DOI: https://doi.org/10.1039/d5ta05234c.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (grant no. 52262016) and the Guangxi Major Special Project Foundation (grant no. Guike AA24263030, Guike AA24263039).

References

1. L. Yang, X. Kong, F. Li, H. Hao, Z. Cheng, H. Liu, J.-F. Li and S. Zhang, Prog. Mater. Sci., 2019, 102, 72 CrossRef .
2. Z. Yang, H. Du, L. Jin and D. Poelman, J. Mater. Chem. A, 2021, 9, 18026 RSC .
3. G. Wang, Z. Lu, Y. Li, L. Li, H. Ji, A. Feteira, D. Zhou, D. Wang, S. Zhang and I. M. Reaney, Chem. Rev., 2021, 121, 6124 CrossRef PubMed .
4. I. R. McNab, IEEE Trans. Magn., 1997, 33, 453 CrossRef CAS .
5. H. G. Wisken, F. Podeyn and H. G. G. Weise, IEEE Trans. Magn., 2001, 37, 332 CrossRef .
6. X. K. Wei, N. Domingo, Y. Sun, N. Balke, R. E. Dunin-Borkowski and J. Mayer, Adv. Energy Mater., 2022, 12, 2201199 CrossRef CAS .
7. D. Hu, Z. Pan, X. Tan, F. Yang, J. Ding, X. Zhang, P. Li, J. Liu, J. Zhai and H. Pan, Chem. Eng. J., 2021, 409, 127375 CrossRef CAS .
8. J. Li, F. Li, Z. Xu and S. Zhang, Adv. Mater., 2018, 30, 1802155 CrossRef PubMed .
9. Z. Lu, W. Bao, G. Wang, S.-K. Sun, L. Li, J. Li, H. Yang, H. Ji, A. Feteira, D. Li, F. Xu, A. K. Kleppe, D. Wang, S.-Y. Liu and I. M. Reaney, Nano Energy, 2021, 79, 105423 CrossRef CAS .
10. Z. Che, L. Ma, G. Luo, C. Xu, Z. Cen, Q. Feng, X. Chen, K. Ren and N. Luo, Nano Energy, 2022, 100, 107484 CrossRef CAS .
11. J. He, Q. Feng, H. Huang, T. Sui, Z. Cen, X. Chen, T. Fujita, H. You, C. Yuan and N. Luo, Ceram. Int., 2023, 50, 3334 CrossRef .
12. C. Luo, Q. Feng, N. Luo, C. Yuan, C. Zhou, Y. Wei, T. Fujita, J. Xu and G. Chen, Chem. Eng. J., 2021, 420, 129861 CrossRef CAS .
13. J. Wu, Q. Feng, C. Luo, T. Shi, Z. Cen, X. Chen, C. Yuan, T. Fujita and N. Luo, Ceram. Int., 2023, 49, 18210 CrossRef CAS .
14. D. Jiang, Z. Zhou, R. Liang and X. Dong, J. Eur. Ceram. Soc., 2021, 41, 1244 CrossRef CAS .
15. X. Zhang, P. Zheng, L. Li, F. Wen, W. Bai, J. Zhang, L. Zheng and Y. Zhang, Scr. Mater., 2022, 211, 114514 CrossRef CAS .
16. Y. Li, L. Jiang, Q. Chen and J. Zhu, J. Mater. Sci.: Mater. Electron., 2019, 31, 52 CrossRef .
17. G. Herrera, J. Jiménez-Mier and E. Chavira, Mater. Charact., 2014, 89, 13 CrossRef CAS .
18. M. Li, X. Y. Su, Y. M. Yang, L. Liu, F. Yu, X. Li, B. Han, J. Wu and H. J. Seo, Sci. Adv. Mater., 2017, 9, 587 CrossRef CAS .
19. L. Sun, J. Hu, F. Gao, X. Kong, H. Qin and M. Jiang, J. Alloys Compd., 2010, 502, 176 CrossRef CAS .
20. C. Li, H. Xiang, J. Chen and L. Fang, Ceram. Int., 2016, 42, 11453 CrossRef CAS .
21. J. Wang, Y. Asakura, T. Hasegawa and S. Yin, Chem. Eng. J., 2021, 423, 130220 CrossRef CAS .
22. Y. Cao, P. Tang, Y. Han and W. Qiu, J. Alloys Compd., 2020, 842, 155581 CrossRef CAS .
23. W. Liu, B. Fu, J. Zhang, X. Ma, Y. Mao, Q. Zong, Z. Zhu, H. Yuan, Y. Zhou and W. Bai, Chem. Eng. J., 2025, 512, 162477 CrossRef CAS .
24. J. M. Deng, M. C. Huang, S. Y. Wu, B. Zhang, Y. F. Han, L. J. Liu and W. P. Gong, J. Alloys Compd., 2024, 986, 174110 CrossRef CAS .
25. C. Luo, Y. Wei, Q. Feng, M. Wang, N. Luo, C. Yuan, C. Zhou, T. Fujita and J. Xu, Chem. Eng. J., 2022, 429, 132165 CrossRef CAS .
26. P. Hohenberg and W. Kohn, Phys. Rev. [Sect.] B, 1964, 136, B864 CrossRef .
27. W. Kohn and L. J. Sham, Phys. Rev. [Sect.] A, 1965, 140, A1133 CrossRef .
28. P. E. Blöchl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953 CrossRef PubMed .
29. G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169 CrossRef CAS PubMed .
30. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865 CrossRef CAS .
31. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104 CrossRef .
32. H. J. Monkhorst and J. D. Pack, Phys. Rev. B: Solid State, 1976, 13, 5188 CrossRef .
33. J. Heyd, G. E. Scuseria and M. Ernzerhof, J. Chem. Phys., 2003, 118, 8207 CrossRef CAS .
34. T. Shi, Q. Feng, J. Wu, Z. Cen, X. Chen, N. Luo, Y. Wei, X. Liu, J. Xu and C. Yuan, Chem. Eng. J., 2023, 470, 144205 CrossRef CAS .
35. X. Zhang, Y. Pu, Y. Ning, L. Zhang, Y. Shang, B. Wang and Z. Chen, Ceram. Int., 2023, 49, 27589 CrossRef .
36. L. Lv, H.-D. Yang, Q.-W. Chen, H. Fan and J.-P. Zhou, Environ. Res., 2023, 223, 115441 CrossRef PubMed .
37. B. Qiao, Y. Jiang, T. Yao, A. Tao, X. Yan, C. Gao, X. Li, H. Ohta, C. Chen, X.-L. Ma and H. Ye, Appl. Surf. Sci., 2022, 587, 151599 CrossRef .
38. D. Gonbeau, C. Guimon, G. Pfister-Guillouzo, A. Levasseur, G. Meunier and R. Dormoy, Surf. Sci., 1991, 254, 81 CrossRef .
39. D. Simon, C. Perrin and J. Bardolle, J. Microsc. Spectrosc. Electron., 1976, 1, 175 Search PubMed .
40. S. Liu, W. Feng, J. Li, C. Zhao, C. Hu, B. He, Z. Bao and X. Luan, Chem. Eng. J., 2023, 451, 138916 CrossRef .
41. Y. Yan, X. Zeng, M. Tang, S. Song, F. Chen, Y. Wang, L. Yang, L. Jin and G. Liu, Ceram. Int., 2022, 48, 33553 CrossRef .
42. X. Wang, Y. Fan, Z. Bin, A. Mostaed, L. Li, A. Feteira, D. Wang, D. C. Sinclair, G. Wang and I. M. Reaney, J. Eur. Ceram. Soc., 2022, 42, 7381 CrossRef .
43. S. Liu, W. Feng, J. Li, B. He, M. Liu, Z. Bao, D. Luo and C. Zhao, J. Eur. Ceram. Soc., 2022, 42, 7430 CrossRef .
44. Q. Zhou, F. Zhang, Q. Chai, Y. He, Z. Peng, D. Wu, P. Liang, L. Wei, J. Wang, X. Chao and Z. Yang, J. Alloys Compd., 2023, 935, 168096 CrossRef .
45. W. Cao, R. Lin, P. Chen, F. Li, B. Ge, D. Song, J. Zhang, Z. Cheng and C. Wang, ACS Appl. Mater. Interfaces, 2022, 14, 54051 CrossRef PubMed .
46. R. Kang, Z. Wang, W. Yang, Y. Zhao, L. Zhang and X. Lou, Chem. Eng. J., 2023, 455, 140924 CrossRef .
47. W. Wang, L. Zhang, W. Shi, Y. Yang, D. Alikin, V. Shur, Z. Lou, D. Wang, A. Zhang, J. Gao, X. Wei, H. Du, F. Gao and L. Jin, ACS Appl. Mater. Interfaces, 2023, 15, 6990 CrossRef PubMed .
48. J. Shi, R. Dong, J. He, D. Wu, W. Tian and X. Liu, J. Alloys Compd., 2023, 933, 167746 CrossRef .
49. Y. Li, M. Y. Tang, Z. G. Zhang, Q. Li, J.-L. Li, Z. Xu, G. Liu and F. Li, Rare Met., 2023, 42, 1261 CrossRef .
50. M. Sun, X. Wang, P. Li, J. Du, P. Fu, J. Hao, W. Li and J. Zhai, J. Eur. Ceram. Soc., 2023, 43, 974 CrossRef .
51. C. Long, W. Zhou, L. Liu, H. Song, H. Wu, K. Zheng, W. Ren and X. Ding, Chem. Eng. J., 2023, 459, 141490 CrossRef .
52. Z. Xu, Z. Liu, K. Dai, T. Lu, Z. Lv, Z. Hu, Y. Liu and G. Wang, J. Mater. Chem. A, 2022, 10, 13907 RSC .
53. A. Xie, J. Fu, R. Zuo, X. Jiang, T. Li, Z. Fu, Y. Yin, X. Li and S. Zhang, Adv. Mater., 2022, 34, 2204356 CrossRef CAS .
54. F. Chen, M. Chen, J. Zhang, W. Liu, H. Du, Q. Zong, H. Yu, Y. Zhang, J. Hao, J. Wang and J. Zhai, ACS Nano, 2025, 19, 1809 CrossRef CAS .
55. Z. Sun, J. Zhang, H. Luo, Y. Yao, N. Wang, L. Chen, T. Li, C. Hu, H. Qi, S. Deng, L. C. Gallington, Y. Zhang, J. C. Neuefeind, H. Liu and J. Chen, J. Am. Chem. Soc., 2023, 145, 6194 CrossRef CAS PubMed .

This journal is © The Royal Society of Chemistry 2025