Perovskite to Brownmillerite Topotactic Transformation

• This topotactic transformation is a solid‐state process that converts perovskite (ABO₃) to brownmillerite (ABO₂.₅) by creating ordered oxygen vacancies while preserving the cation framework.
• It is achieved through thermal annealing, chemical reduction, or electric-field gating, with noble-metal capping reducing oxygen migration barriers from ~3.1 eV to below 1 eV.
• The phase change shifts material behavior from metallic ferromagnetism to insulating antiferromagnetism, enabling applications in memristors, fuel cells, and neuromorphic devices.

A perovskite to brownmillerite topotactic phase transformation is a solid-state transition wherein the highly symmetric, oxygen-rich perovskite lattice (ABO₃) converts to an ordered oxygen-deficient brownmillerite phase (ABO₂.₅). This process is mediated by the controlled insertion or removal of oxygen, resulting in dramatic changes to crystal symmetry, electronic structure, magnetic ground state, transport properties, and functional device behavior. Unlike reconstructive phase transitions, topotactic reactions preserve the cation (A/B-site) lattice framework and epitaxial registry, with oxygen ions migrating to create ordered channels of vacancies and concomitant changes in local coordination from octahedral (BO₆) to tetrahedral (BO₄). Such transitions are realized via thermal annealing in controlled atmospheres, electric-field-driven gating, chemical reduction, or even interfacial engineering, offering access to low-temperature, reversible, and spatially heterogeneous phase transformations. These mechanisms are essential in advancing resistive switching, solid-oxide ionics, thermal management, and neuromorphic applications (D'Anna et al., 10 Jan 2026, Jeen et al., 2015, Zhang et al., 2023).

1. Crystallographic and Chemical Basis

The perovskite (ABO₃, typically cubic Pm3̄m) is defined by a corner-sharing network of BO₆ octahedra, with 12-coordinated A-site cations. Ordered oxygen removal produces brownmillerite (ABO₂.₅, commonly Ibm2, Imma, I2bm), where the lattice incorporates alternating BO₆ octahedral and BO₄ tetrahedral layers aligned along the crystallographic c-axis, yielding a unit cell doubling or quadrupling in periodicity. The transformation is strictly topotactic: cation sublattice registry is preserved, oxygen vacancies segregate into one-dimensional channels, and minimal cation diffusion occurs. The canonical chemical equation for Co-based systems is: $$\mathrm{La}_{0.7}\mathrm{Sr}_{0.3}\mathrm{CoO}_{3} \longrightarrow \mathrm{La}_{0.7}\mathrm{Sr}_{0.3}\mathrm{CoO}_{2.5} + \tfrac{1}{2}\mathrm{O}_2$$ (D'Anna et al., 10 Jan 2026, Jeen et al., 2015, Zhang et al., 2023).

The process is accompanied by contraction of the in-plane lattice parameter (a drop of a few tenths of a percent) and expansion of the out-of-plane parameter (to accommodate the vacancy channels), as confirmed by reciprocal space mapping and XRD. In fully converted brownmillerite, oxygen vacancies are ordered in channels, resulting in fractional Bragg peaks such as BM(006), and a loss of symmetry compared to the parent perovskite.

2. Mechanisms, Thermodynamics, and Kinetics

Topotactic PV→BM transition relies on the creation and ordering of oxygen vacancies under reducing environments (thermal vacuum, chemical getters, noble-metal interfaces, or electric fields). The transformation occurs at temperatures much lower than bulk solid-state reaction routes: typically 150–400 °C for Co, Mn, Fe, Ni-based oxides under suitable conditions (Jeen et al., 2015, Meng et al., 2023, Wang et al., 2021, Yin et al., 28 Aug 2025). Activation barriers for oxygen migration can be dramatically reduced by noble-metal capping (Pt, Ag), which introduces interfacial charge transfer and weakens M–O bonds; for SrCoO₃, this reduces the migration barrier from ~3.1 eV to well below 1 eV, enabling room-temperature transitions within hours (Wang et al., 2021).

Thermodynamic driving forces are governed by the chemical potential difference in oxygen between matrix and ambient/reducing phase. For example, Al oxygen-getter annealing achieves pO₂ < 10⁻¹⁶ bar, strongly favoring vacancy formation (Yin et al., 28 Aug 2025). Kinetics are further dictated by vacancy diffusion constants (Arrhenius: D = D₀exp(–Eₐ/k_BT)), with values Eₐ ~0.6–1.1 eV for typical cobaltites and manganites (Zhang et al., 2023, Yin et al., 28 Aug 2025).

The transition is often spatially and temporally heterogeneous. In-situ Bragg XPCS resolves two distinct nanoscale timescales: (i) a constant domain-growth process with timescale τ₁ ≈ 10³ s (domain wall speed v_d ≈ 6 × 10⁻⁴ nm/s), and (ii) a slower, thermally-driven depinning process yielding accelerated rearrangements with aging exponent –2.2 ± 0.5 (D'Anna et al., 10 Jan 2026). These features reflect glassy oxygen diffusion-limited transformation mechanisms.

3. Oxygen Vacancy Ordering and Local Structure

Ordered vacancy formation in BM drives local coordination changes and superlattice formation. The oxygen deficit parameter δ (in ABO₃–δ) varies continuously from δ ≈ 0 (perovskite, only octahedral BO₆) to δ = 0.5 (brownmillerite, equal octahedral/tetrahedral layers) (Zhang et al., 2023, Lannerd et al., 6 Oct 2025, Jeen et al., 2015). The site occupancy framework:

• O_O = 1 (fully occupied O sites)
• O_eq = 1 (equatorial O fully occupied in octahedron)
• O_ap = 0 (apical O vacated in every second layer), produces vacancy channels along [110]_P, lowering symmetry and enabling strain relief.

Crystallographic tables (excerpted for the ABO₃ ↔ ABO₂.₅ transformation in SrCoOₓ):

Phase	Space Group	Key Lattice Parameters	Features
Perovskite	Pm3̄m	a ≈ 3.83–3.90 Å	All BO₆ octahedral
Brownmillerite	Ibm2/I2bm	a ≈ 5.5 Å, b ≈ 15.8 Å, c ≈ 5.6 Å	BO₆/BO₄ layer alternation

The ordered vacancy motif causes abrupt electronic and structural changes: BO₆ layers convert into BO₄ tetrahedral sheets, 1D vacancy channels arise, and local polarization emerges in specific polymorphs (e.g., I2mb SrCoO₂.₅, with Berry-phase polarization P_[1̄10] ≈ 6.4 μC/cm²) (Lim et al., 2018).

4. Electronic, Magnetic, and Transport Properties

The PV→BM transformation drives a transition from metallic, ferromagnetic states to insulating, antiferromagnetic ground states. In Co-based systems:

• PV phase (SrCoO₃): metallic, ferromagnetism below T_C ≈ 220 K, Ms ≈ 0.6 µ_B/Co (Jeen et al., 2015, Lim et al., 2018).
• BM phase (SrCoO₂.₅): insulating, no long-range ferromagnetism, G-type AFM with band gap ≈ 0.5 eV (Lim et al., 2018). The suppression of pdσ hybridization due to increased Co–O–Co distances (from ≈180° to ≈132–156°) kills double-exchange FM and localizes carriers (Lim et al., 2018).

In Mn-based systems (La₀.₇Sr₀.₃MnO₃ ↔ La₀.₇Sr₀.₃MnO₂.₅), FM metal with Curie T_C ≈ 327 K and metallic ρ(300K) ≈ 1.5 × 10⁻² Ω·cm is transformed to AFM insulator with dominant Mn²⁺ at the surface (as evidenced by XAS), ρ(300K) ≈ 1 Ω·cm for BM (Yin et al., 28 Aug 2025, Jin et al., 13 Mar 2025).

Thermal and electrical transport exhibit similarly abrupt transitions. Room-temperature ion-gel gating in LSCO films enables over fivefold tuning of thermal conductivity, from ≈2.4 to ≈0.85 W m⁻¹ K⁻¹, correlating with a metal-insulator transition (Zhang et al., 2023).

5. Dynamic and Nanoscale Heterogeneity

The PV→BM transformation is not spatially uniform. Domain-growth processes (quantified via in-situ XPCS) show stable propagation of domain walls at nm/h scales, whereas depinning events—localized rearrangements mediated by defects and strain—accelerate dynamically as the transformation "ages." The fraction of dynamically active domain wall material increases from ≈18% toward ≈25% over multi-hour timescales (D'Anna et al., 10 Jan 2026). This coexistence of uniform and stochastic processes engenders drift, slow response times, and long-term instability in device properties such as resistance.

Strain is also a decisive variable: thermal strain can drive reversible topotactic transitions in LSCO films at fixed δ, with alternating BM domain orientations scaling with temperature; above a critical temperature (≈360 °C), new vacancy-ordered superstructures (period-3a) nucleate and rapidly grow, demonstrating that non-redox-driven topotactic control (via elastic energy) offers efficient phase manipulation for ionic devices (Inkinen et al., 2019).

6. Synthetic Strategies, Phase Engineering, and Characterization

A wide repertoire of synthetic techniques enables PV→BM transformations:

• Low-temperature hydride reduction (e.g., CaH₂, NaH): enables phase change in Mn-, Ni-, Fe-, Co-based oxides at 200–500 °C (Meng et al., 2023, Wu et al., 2023, Jin et al., 13 Mar 2025).
• Vacuum annealing with oxygen getters (Al foil): drives deep reduction and vacancy ordering at lower temperatures (Yin et al., 28 Aug 2025).
• Noble-metal surface capping (Pt, Ag): reduces oxygen migration barrier, permitting room-temperature transitions (Wang et al., 2021).(Meng et al., 2023)
• Electric field or ion-gel gating: dynamic (nonvolatile) control of δ near ambient conditions (Zhang et al., 2023).
• Electron-beam irradiation: locally triggers PV→BM conversion in STEM (Meng et al., 2023).

Characterization includes XRD (lattice, superlattice peaks), reciprocal space mapping (strain, coherence), in-situ XPCS/STEM (domain and defect dynamics), XAS/RIXS (valence state, hybridization), EELS (vacancy distribution, cation oxidation states), and transport/magnetometry for functional property correlation (D'Anna et al., 10 Jan 2026, Jeen et al., 2015, Jin et al., 13 Mar 2025, Yin et al., 28 Aug 2025).

The fully reversible nature of the topotactic transformation, with preserved crystallinity and phase coherence, makes it uniquely advantageous for cyclic device operation, memristors, solid-oxide cells, and electrochemical sensors, among others.

7. Applications and Future Directions

PV→BM transitions underpin several device concepts:

• Resistive switching and memristors (SrCoOₓ, LaCoOₓ, Fe/Ni analogs): multi-level, nonvolatile conductance, ultra-fast switching (Meng et al., 2023, Jeen et al., 2015).
• Solid-oxide fuel cells: rapid oxygen exchange and mixed conduction channels (Meng et al., 2023).
• Neuromorphic and artificial synapse elements: conductance-based learning (Meng et al., 2023, D'Anna et al., 10 Jan 2026).
• Electrochemical sensors, batteries, and catalysis: robust redox cycling and oxygen vacancy control (Jeen et al., 2015, Meng et al., 2023).
• Dynamic thermal switches: voltage-control over thermal conductivity in LSCO epitaxial films (Zhang et al., 2023).

Contemporary research focuses on atomically resolved oxygen migration tracking, interface and strain engineering to tune transformation kinetics and stabilize intermediate or metastable BM structures, and scaling up low-temperature topotactic patterning for wafer-scale oxide electronics. Nonstoichiometric, protonated, and defect-rich BM phases are of particular interest for next-generation multifunctional materials, though their thermodynamic instability may limit reversibility in some electrochemical regimes (Lannerd et al., 6 Oct 2025).

Overall, the perovskite↔brownmillerite topotactic transition remains a central platform for understanding coupled ion-electron transport, and for the rational design of oxide systems with tunable structure-property relationships and emergent functionalities.