Reversible structural evolution of sodium-rich rhombohedral Prussian blue for sodium-ion batteries

The study addresses the need for low-cost, scalable cathode materials for sodium-ion batteries (SIBs), motivated by sodium’s abundance and suitability for grid-scale storage. Prussian blue analogs (PBAs) are attractive due to open framework structures enabling fast Na+ transport and synthesis without high-temperature calcination, lowering costs. However, performance strongly depends on precipitation-derived microstructure, Na content, vacancies, and coordinated water. Prior Mn-based PBAs suffer from Jahn–Teller distortion and poor conductivity, while Fe-based sodium-rich Prussian white shows excellent performance but was often prepared hydrothermally with low yield and undesirable byproducts. Co-precipitation routes typically produce Na-poor cubic phases with many defects and coordinated water, leading to poor reversibility and cycling. The research aims to control the precipitation/nucleation-growth process to obtain sodium-rich rhombohedral Na2−xFe[Fe(CN)6] with high crystallinity, to elucidate its reversible structural evolution during Na+ (de)intercalation, and to demonstrate scalable synthesis and practical cell performance.

Previous work established PBAs as promising SIB cathodes; Mn-based PBAs can transform from monoclinic to rhombohedral after removing interstitial water, improving stability but still limited by Mn3+ Jahn–Teller effects. Fe-based Prussian white (Na1.92FeFe(CN)6) with rhombohedral symmetry exhibits excellent rate and cycling, yet traditional hydrothermal synthesis yields low throughput and toxic byproducts (NaCN). Early co-precipitation efforts produced Na-poor cubic phases with nanoparticles, abundant defects/coordination water, and unstable cycling (<200 cycles). In situ PXRD showed irreversible unit-cell shrinkage upon Na+ extraction in Na-poor cubic Na2−xFeFe(CN)6. Strategies like chemical inhibition and gradient substitution improved crystallinity/performance but scalable synthesis of sodium-rich rhombohedral Na2−xFeFe(CN)6 via co-precipitation remained challenging due to rapid nucleation and oxidation of Fe2+. This work builds on those insights, focusing on controlling chelation, atmosphere, and sodium source to slow nucleation, increase Na content, and achieve rhombohedral symmetry.

Synthesis: Sodium-rich Na2−xFe[Fe(CN)6] samples (PB-S1 to PB-S4) were synthesized at 25 °C via modified co-precipitation using sodium citrate as both chelating agent and sodium source. Under N2 protection (to prevent Fe2+ oxidation), solution A (FeSO4·7H2O + sodium citrate) was stirred for 3 h, then added dropwise into solution B (Na3Fe(CN)6·10H2O + sodium citrate) under vigorous stirring (800 rpm) for 6 h, aged, washed (water/ethanol), and vacuum-dried at 120 °C. Sodium citrate content was varied (2.81–15 g in each solution) to regulate nucleation/growth; additional control experiments probed effects of atmosphere (no N2), chelating time (30 min), removing citrate from solution B, and replacing citrate with equimolar NaCl. Characterization: Phase identification and structural refinement employed laboratory PXRD (Cu Kα) and synchrotron PXRD (λ ≈ 0.689 Å). Indexing, Le Bail, and Rietveld refinements were performed in TOPAS 5, with Na positions from difference maps; water positions were not resolved due to disorder. ICP-OES quantified Na/Fe content; TGA/DSC measured water content; XPS assessed Fe valence; SEM and STEM-SAED examined morphology and crystallinity; EDS mapped elemental distributions. In situ synchrotron PXRD monitored structural evolution during cycling. Electrochemistry (coin cells): Cathodes comprised 70 wt% active material, 20 wt% Super P/C45, 10 wt% PVDF in NMP, cast on Al, dried, pressed, and punched. Sodium metal served as counter electrode; electrolyte was 1 M NaClO4 in EC:PC (1:1) with 3% FEC; glass fiber separator; assembly in Ar glovebox. Galvanostatic cycling between 2–4 V; rate tests with stepwise current increases; CV at 0.1 mV s−1; GITT after four formation cycles using 10 min current pulses with 60 min rests to extract Na+ diffusion coefficients. Scale-up and pouch cell: A 100 L reactor produced ~5 kg sodium-rich prussian white (based on PB-S3 recipe) showing rhombohedral structure and microcube morphology. Pouch full cells used the PBA cathode and commercial hard carbon anode (anode capacity 20% excess). Electrodes for pouch cells: 90:5:5 (active:carbon:PVDF); electrolyte 1 M NaPF6 in EC:DEC (1:1) with 2% FEC; polypropylene separator. Long-term cycling was performed between 1.0–3.2 V at 1 C.

- Controlled co-precipitation yielded sodium-rich, highly crystalline rhombohedral Na2−xFe[Fe(CN)6] microcubes. Composition estimates from ICP/TGA: PB-S1 Na1.53Fe[Fe(CN)6]·4.2H2O (cubic, Fm-3m, a=10.3711 Å); PB-S3 Na1.73Fe[Fe(CN)6]·3.8H2O (rhombohedral, R-3, a=b=7.43079 Å, c=17.6133 Å). - Morphology evolution from irregular nanoparticles to aggregated sub-microcubes to uniform single microcubes was achieved by tuning sodium citrate concentration, chelating time, sodium source, and maintaining N2 atmosphere. - Electrochemistry (coin cells): Both PB-S1 and PB-S3 delivered initial discharge ~116 mAh g−1. Initial Coulombic efficiency (ICE): PB-S3 97.4% (near-ideal); PB-S1 120% (indicative of Na insertion from anode due to Na-poor cathode and/or oxidation effects). Rate capability: PB-S3 retained 83% capacity at 500 mA g−1 relative to 10 mA g−1, and ~70 mAh g−1 at 2000 mA g−1; PB-S1 retained 58% at 500 mA g−1. Cycling: PB-S3 retained 71% after 500 cycles at 100 mA g−1; PB-S1 retained 66% after 200 cycles. - CV indicated Fe2+/Fe3+ redox peaks: 2.8–3.1 V (high-spin Fe–N) and ~3.8–4.0 V (low-spin Fe–C); PB-S1 showed more pronounced high-voltage oxidation, consistent with Na extraction from C-site Fe due to Na deficiency. - GITT: Na+ diffusion coefficients of PB-S1 and PB-S3 were comparable; PB-S3 slightly higher, attributed to the stable Na-rich rhombohedral framework. - In situ synchrotron PXRD: PB-S1 exhibited irreversible peak shifts (lattice shrinkage) during charge and incomplete recovery on discharge. PB-S3 displayed highly reversible three-phase transitions: rhombohedral → cubic (~≤3.2 V) → tetragonal (P4/mmm, >3.2 V up to 4.0 V) on charge, and the reverse on discharge; primitive cell volume changed by ~4% during single charge or discharge, recovering fully, evidencing excellent structural reversibility. - Scale-up and pouch cell: 100 L synthesis yielded ~5 kg of rhombohedral prussian white microcubes. Pouch full cells (Na2−xFe[Fe(CN)6] cathode vs hard carbon anode) cycled between 1.0–3.2 V at 1 C showed a discharge plateau ~2.9 V and 78% capacity retention over 1000 cycles.

The work demonstrates that precise control of the precipitation/nucleation process—via sodium citrate chelation/supply, N2 protection, adequate chelation time, and appropriate sodium source—enables formation of sodium-rich rhombohedral Na2−xFe[Fe(CN)6] with large, highly crystalline microcube morphology. This structural state minimizes vacancies and coordinated water relative to fast-precipitated cubic phases, resulting in superior electrochemical characteristics: near-ideal initial Coulombic efficiency, enhanced rate capability, and improved cycle life. In situ synchrotron PXRD reveals that the rhombohedral framework accommodates Na+ (de)intercalation through fully reversible transitions among rhombohedral, cubic, and tetragonal phases with small volume changes (~4%), explaining the mechanical and electrochemical stability and the dominance of a low-voltage (~<3.2 V) plateau linked to high-spin Fe redox. In contrast, Na-poor cubic material exhibits irreversible lattice contraction and poorer durability. The successful 100 L scale synthesis and long-life pouch cell performance substantiate the method’s practicality and potential for commercialization, addressing longstanding barriers of yield, quality, and consistency in PBA cathodes for SIBs.

A scalable, controllable co-precipitation strategy produced sodium-rich rhombohedral Na2−xFe[Fe(CN)6] microcubes with highly reversible three-phase evolution (rhombohedral ↔ cubic ↔ tetragonal) during Na+ cycling. The optimized material delivered near-ideal initial Coulombic efficiency (97.4%), strong rate performance (~70 mAh g−1 at 2000 mA g−1), and extended cycling stability (71% retention after 500 cycles). A 100 L synthesis yielded kilogram-scale product, and pouch full cells achieved 78% capacity retention over 1000 cycles, underscoring application potential. Future work should target further reducing defects and coordinated water to approach theoretical capacity, suppressing Fe2+ oxidation during processing, quantifying/controlling vacancy populations, and extending the controlled precipitation paradigm to other PBAs and multi-cation chemistries for high-energy, long-life SIBs.

- The practical capacities remain below theoretical values, attributed to structural defects, coordinated water, and slight oxidation of Fe2+ during washing/drying. - Vacancy concentrations could not be directly quantified (ICP limitations), and water positions were unresolved due to disorder and lack of periodicity in diffraction. - Partial Fe2+ oxidation occurred despite N2 protection, challenging complete preservation of the Na-rich state. - Electron-beam sensitivity limited detailed TEM analyses to SAED. - While scale-up to 100 L and 5 kg was demonstrated, broader manufacturability aspects (e.g., continuous processing, batch-to-batch variability, electrode loading at industrial areal capacities) were not detailed.