Sodium-Ion Cathode Materials — O3/P2 Layered Oxides, Prussian Blue Analogues, and NASICON Frameworks

Q: What is the initial Coulombic efficiency problem in hard carbon and why does it matter?

Initial Coulombic efficiency (ICE) is the fraction of charge put into a cell on its first charge that can be recovered on the first discharge. In graphite/NMC cells, ICE is approximately 90–93% — 7–10% of the charge goes into irreversible SEI formation rather than into stored capacity. In hard carbon/Na-ion cells, ICE is typically 70–85% — a 15–30% first-cycle irreversibility. The lower ICE is caused by: (1) larger surface area of disordered hard carbon (more surface for SEI formation), (2) sodium clusters in nanopores are partially irreversible (some sodium becomes trapped in pores that close upon first sodium insertion), and (3) the sodium-electrolyte interface chemistry produces thicker SEI than lithium-graphite. The practical consequence: in a full cell (cathode + anode), the cathode must provide excess sodium on first charge to compensate for the hard carbon's first-cycle loss. This is typically addressed by pre-sodiation — adding extra sodium to the cathode or the anode before cell assembly. Pre-sodiation adds manufacturing complexity but is essential for full-cell performance. Improving ICE above 85% in hard carbon is one of the primary research targets for Na-ion cell developers — it directly increases the effective energy density of the full cell.

Layered Transition Metal Oxides: O3 and P2 Structures
Prussian Blue Analogues (PBAs): Low-Cost Open Frameworks
NASICON-Type Frameworks: The Stable Alternative
The Hard Carbon Anode: Disordered Structure, Two-Mechanism Storage
The Initial Coulombic Efficiency Challenge
O3 Structure: High Sodium Content, Phase Instability
P2 Structure: Better Rate Capability, Lower Initial Capacity

Sodium-ion battery development is complicated by an abundance of riches at the cathode — unlike lithium-ion, where decades of competition converged on two dominant cathode families (NMC for energy density, LFP for stability), sodium-ion still has three competing cathode families each with genuine advantages, and none with an unambiguous lead. The cathode choice determines the cell's operating voltage, energy density, cycle life, cost, moisture sensitivity, and manufacturing complexity. Understanding the structural chemistry of each family explains why these trade-offs exist and why no single Na-ion cathode has yet dominated.

The anode side is less diverse but more novel — graphite, the universal lithium-ion anode, is essentially useless for sodium-ion. Hard carbon is the only practical Na-ion anode in commercial production, and its disordered microstructure creates both the storage mechanism and the most important manufacturing challenge: initial Coulombic efficiency.

TL;DRKey Points

Three cathode families compete in Na-ion: layered transition metal oxides (O3/P2 structures — highest energy density, phase transition challenge), Prussian Blue Analogues (PBAs — cheapest synthesis, moisture sensitive), and NASICON-type frameworks (most stable, lower energy density).
O3-type layered oxides (NaNiMnO₂-based) are the choice for commercial high-energy Na-ion cells (CATL Gen 1, Faradion). Phase transitions during cycling (O3→P3→P2) are the primary cycle life challenge.
P2-type structures have better rate capability (prismatic Na sites allow faster ion movement) but start with lower Na content — limiting capacity.
PBAs (iron/manganese hexacyanoferrate) are synthesised at room temperature from cheap precursors but require strict moisture control. HiNa Battery uses PBA cathode in their commercial cells.
Hard carbon anodes store Na via two mechanisms: intercalation into pseudo-graphite nanocrystals (sloping region, 0.1–1.0V) and pore-filling with Na clusters (flat plateau, <0.1V). Low initial Coulombic efficiency (70–85%) is the key anode engineering challenge.

Layered Transition Metal Oxides: O3 and P2 Structures

Layered oxides for Na-ion have the general formula NaxMO₂, where M is one or more transition metals (Mn, Fe, Co, Ni, Cu, Ti) and x is the sodium stoichiometry (0 < x ≤ 1). These are structurally analogous to the lithium layered oxides (LiCoO₂, NMC) that dominate commercial Li-ion — the same layered architecture, different metal chemistry.

O3 Structure: High Sodium Content, Phase Instability

O3 structure has sodium in octahedral coordination, three layers per crystallographic repeat. Starting composition is close to x = 1.0 (fully sodiated). As sodium is extracted during charging (x decreases), the structure undergoes a sequence of phase transitions:

O3 → O3' (monoclinic distortion at x ≈ 0.8) → P3 (x ≈ 0.5–0.6) → P3' → (desodiation limit)

Each phase transition involves a change in the stacking sequence of the oxide layers — the transition metal oxide slabs glide relative to each other to accommodate the new coordination geometry of sodium at lower content. These glide events are partially irreversible: after many cycles, cumulative microstrain from repeated gliding reduces structural integrity and capacity.

O3-NaNiMnO₂ theoretical capacity | 240 | mAh/g

O3-NaNiMnO₂ practical capacity (25°C, 0.1C) | 160–185 | mAh/g

P2-Na₀.₆₇MnO₂ theoretical capacity | ~175 | mAh/g (limited by initial Na content)

P2-Na₀.₆₇MnO₂ practical capacity (0.1C) | 130–165 | mAh/g

PBA (Na₂MnFe(CN)₆) practical capacity | 120–150 | mAh/g

NASICON (Na₃V₂(PO₄)₃) practical capacity | 115–120 | mAh/g

Hard carbon anode capacity (commercial) | 250–320 | mAh/g

Hard carbon initial Coulombic efficiency | 70–85 | %

Graphite initial Coulombic efficiency (in Li-ion) | 90–93 | %

Addressing O3 phase instability: Surface coating (Al₂O₃, TiO₂) applied to O3 cathode particles suppresses direct electrolyte contact and partially buffers the surface structure against phase transition stress. Dopants (substituting a fraction of Mn with Ti, Mg, or Al) stabilise the O3 lattice and reduce the extent of the O3→P3 transition during cycling. CATL's first-generation Na-ion cell uses an O3-type cathode with proprietary composition adjustments that stabilise the high-sodium region and improve cycle life beyond 1,500 cycles.

P2 Structure: Better Rate Capability, Lower Initial Capacity

P2 structure has sodium in trigonal prismatic coordination, two layers per repeat. Starting composition is x ≈ 0.6–0.7 (partially sodiated), which limits theoretical capacity compared to O3 (less sodium available to cycle). However, sodium in prismatic sites has faster diffusion kinetics than in octahedral sites — the prismatic coordination is less confining and the activation energy for sodium hopping between adjacent sites is lower. P2 cathodes can sustain higher discharge rates (better rate capability) and show better capacity retention at fast rates.

P2-O2 transition issue: When P2 materials are charged to full desodiation (x → 0), an irreversible P2 → O2 phase transition can occur. The O2 phase (sodium in octahedral, two-layer repeat) has lower sodium mobility and is difficult to fully reduce back to P2. This limits the upper charge voltage that P2 cathodes can safely access, reducing usable capacity. Strategies: voltage window limitation (charge only to 4.0V rather than 4.2V), multi-element doping to stabilise P2 at low Na content.

◆

Key Insight

The research strategy to overcome O3 instability and P2 capacity limitations has converged on mixed-phase approaches. "O3/P2 composite" cathodes use synthesis conditions that produce materials with intergrown O3 and P2 domains at the nanoscale. The P2 regions provide fast-rate channels, the O3 regions provide high sodium content, and the intergrown interface provides a template that partially suppresses long-range phase transitions during cycling. Samsung SDI, POSCO Chemical, and several Chinese battery companies have published results with O3/P2 composites showing better combined rate capability and cycle life than single-phase materials.

Prussian Blue Analogues (PBAs): Low-Cost Open Frameworks

PBAs have the general formula A₂M[M'(CN)₆] where A = Na (in Na-ion applications), M and M' are transition metals connected by cyanide (CN⁻) bridges. The cubic open-framework structure has large interstitial sites (approximately 8 Å diameter) that can freely host Na⁺ ions. The open-framework structure provides a 3D diffusion network — sodium can move in any direction — which gives PBAs inherently good rate capability.

Synthesis advantage: PBA synthesis is simply mixing aqueous solutions of the two metal salts and sodium cyanide at room temperature — no high-temperature calcination required. This is dramatically simpler and cheaper than oxide cathode synthesis, which requires mixing, grinding, and calcining at 700–900°C. PBA synthesis energy cost is approximately 5–10× lower per kilogram of cathode than layered oxide synthesis.

The moisture problem: PBA crystal structures contain water molecules in interstitial vacancies (coordinated water and zeolitic water). This water is partially trapped during synthesis and extremely difficult to remove without structural damage. Residual water in the PBA cathode reacts with the Na-ion electrolyte (NaPF₆ in carbonates), generating HF (hydrofluoric acid) that attacks both the cathode surface and the hard carbon anode. Managing moisture in PBA cathode synthesis and cell manufacturing is the primary commercial challenge. HiNa Battery (a Chinese Na-ion company partly affiliated with the Institute of Physics, CAS) has developed PBA synthesis with controlled water content and commercialised these cells at 120 Wh/kg cell level energy density.

Fe-Fe vs Mn-Fe PBA:

Na₂FeFe(CN)₆ (FeHCF): operates at ~3.1V average, lower energy density, good cycle life
Na₂MnFe(CN)₆ (MnHCF): operates at ~3.4V average, higher energy density but Mn site can show Jahn-Teller distortion at certain sodium contents, causing structural irregularity
Na₂NiFe(CN)₆ (NiHCF): high voltage (~3.8V), but nickel cost and cycle stability issues

NASICON-Type Frameworks: The Stable Alternative

NASICON (Sodium Super Ionic Conductor) was developed in the 1970s as a solid electrolyte. The structure — a 3D framework of corner-sharing MO₆ octahedra and XO₄ tetrahedra (X = P, S, As) — provides a 3D open channel network for Na⁺ transport. NASICON structures are intrinsically stable because the [M₂(XO₄)₃] framework (the polyanion backbone) is covalently bonded and robust against phase transitions.

Na₃V₂(PO₄)₃ (NVP): The most studied NASICON cathode for Na-ion. Vanadium occupies the M sites, phosphate the X sites. Two sodium per formula unit are electrochemically active (one is structurally fixed). Average voltage ~3.4V, capacity ~115 mAh/g. Outstanding cycle life (>10,000 cycles reported in some studies) due to the rigid framework that shows essentially zero volume change during cycling (NASICON is the sodium equivalent of LFP's structural stability, but with a 3D framework rather than olivine). The limitation is vanadium toxicity — vanadium compounds are toxic and present environmental concerns for mass-market batteries. Vanadium-free NASICON variants (Ti, Fe substitution) sacrifice some electrochemical performance.

Property	O3 Layered Oxide	P2 Layered Oxide	Prussian Blue Analogue	NASICON (NVP)
Practical energy density	High (160–185 mAh/g)	Moderate (130–165 mAh/g)	Moderate (120–150 mAh/g)	Lower (115–120 mAh/g)
Average voltage	3.2–3.4 V	3.3–3.5 V	3.1–3.4 V	3.4 V
Cycle life (25°C)	1,000–2,000 cycles	1,500–3,000 cycles	500–2,000 cycles	>5,000 cycles
Synthesis temperature	700–900°C	700–900°C	Room temperature	700–900°C
Moisture sensitivity	Moderate	Moderate	Very high	Low
Primary concern	Phase transitions	Low initial Na content	Water in structure	V toxicity
Commercial developers	CATL, Faradion	Research stage	HiNa Battery	Aquion, research

The Hard Carbon Anode: Disordered Structure, Two-Mechanism Storage

Graphite does not work for Na-ion — the thermodynamics of sodium intercalation between graphite layers are unfavourable. Hard carbon is the practical replacement, and its storage mechanism is the most distinctive feature of Na-ion cell chemistry.

Hard carbon is produced by pyrolysis of oxygen-containing organic precursors at 1,000–1,400°C in inert atmosphere (argon or nitrogen). Polymers (PAN, PVC), biomass (sucrose, cellulose, coconut shell, rice husk), and phenolic resins are all used as precursors. The pyrolysis process:

Decomposition of organic functional groups, releasing H₂O, CO₂, CO, and H₂
Partial graphitisation: short-range ordered graphene nanodomains (2–5 nm, 2–5 stacked layers) form
Turbostratic disorder: the nanodomains do not align globally — they stack at random angles, preventing long-range graphitic order
Nanopore formation: closed nanopores (0.5–2 nm) between randomly stacked nanodomains

The precursor, pyrolysis temperature, and atmosphere all affect the final microstructure — and therefore the electrochemical properties. Higher pyrolysis temperature (1,200–1,400°C) produces harder carbon with better graphitisation and higher plateau capacity (more nanopores) but lower surface area and better ICE. Lower temperature (800–1,000°C) produces more defective carbon with higher slope capacity but lower plateau capacity and worse ICE.

Note

India has exceptional feedstock availability for hard carbon production. Rice husk (from India's rice industry, annual production ~22 million tonnes of husk) is one of the most studied biomass precursors for hard carbon — its natural silica content is removed by acid washing before pyrolysis, and the resulting hard carbon has properties suitable for Na-ion anodes. Sugarcane bagasse (India is the world's second-largest sugarcane producer) and coconut shells (India is the world's third-largest coconut producer) are also viable precursors. If Reliance or another Indian company establishes domestic hard carbon production from these agricultural byproducts, India could have a complete Na-ion supply chain: domestic soda ash for cathode sodium source + agricultural biomass for anode = zero dependence on lithium, cobalt, or graphite imports for the anode-cathode combination.

The Initial Coulombic Efficiency Challenge

Hard carbon's disordered structure creates significant first-cycle irreversibility. On first charge, sodium ions enter the hard carbon and a significant fraction become trapped — forming SEI on the large hard carbon surface, filling nanopores that close and cannot re-release sodium, and irreversibly reacting with surface functional groups (–OH, –COOH) present on the hard carbon.

Practical ICE values: 70–85% for most commercial hard carbon materials. This means for every 100 sodium ions from the cathode on first charge, 15–30 are permanently lost to irreversible anode reactions.

Pre-sodiation strategies:

Sacrificial anode additive: Na₂C₂O₄ (sodium oxalate), Na₃P (sodium phosphide), or metallic sodium particles added to the anode slurry. On first charge, these sacrifice their sodium to compensate for the anode's irreversibility — but require safe handling of reactive sodium-containing materials.
Cathode pre-sodiation: Extra sodium added to the cathode by co-synthesising with a sodium excess or by adding a Na-rich additive. Allows compensation without adding sodium to the anode side.
Electrochemical pre-sodiation: A separate pre-sodiation step in the cell formation process, using an external sodium source electrode to sodiate the hard carbon before the full cell is assembled.

✦Key Takeaways

Three cathode families compete in Na-ion with different trade-offs: O3/P2 layered oxides (high energy density, phase transitions), PBAs (cheap synthesis, moisture problems), NASICON (excellent stability, vanadium toxicity, lower energy density). O3 layered oxide dominates commercial high-energy Na-ion cells in 2024.
The O3→P3→P2 phase transition sequence during cycling is the primary cycle life challenge for layered oxide cathodes. Surface coating, dopant stabilisation, and O3/P2 composite architectures are the main mitigation strategies.
PBA synthesis at room temperature offers dramatically lower manufacturing energy cost than oxide cathode synthesis. HiNa Battery's commercial PBA Na-ion cells demonstrate this pathway, but moisture control remains the primary manufacturing challenge.
Hard carbon anodes store sodium through two mechanisms: intercalation into graphene nanodomains (sloping voltage region) and nanopore-filling with Na clusters (flat voltage plateau). The practical capacity of 250–320 mAh/g is competitive with graphite for lithium — but initial Coulombic efficiency of 70–85% requires pre-sodiation to achieve full-cell performance.
India has a natural supply chain advantage for Na-ion hard carbon anode production: rice husk, sugarcane bagasse, and coconut shells are abundant, cheap agricultural byproducts that produce high-quality hard carbon. Domestic hard carbon manufacturing from these feedstocks would complete an India-specific Na-ion supply chain with zero lithium, cobalt, or graphite import dependence.

Part of the cell-chemistry Series

← PreviousNMC 811 vs NMC 622 vs NMC 532 — What the Numbers Actually Mean Next →SEI Growth, Lithium Inventory Loss, and Cathode Cracking — The Molecular Mechanisms of Battery Degradation

Written by

Sai Chaitanya Dasari

Battery Systems Engineer · Volvo Eicher Commercial Vehicles

3+ years in commercial EV pack development. Writing about real battery engineering from the bench.

Frequently Asked Questions

Why doesn't sodium intercalate into graphite like lithium does?

Sodium ions do not intercalate efficiently into graphite because of a combination of thermodynamic and geometric factors. The graphite interlayer spacing is approximately 3.35 Å (angstroms). A Li⁺ ion has an ionic radius of 0.76 Å and forms the stage-I compound LiC₆ (one Li per six carbons) with an interlayer expansion to 3.70 Å — a modest 10% expansion that graphite accommodates easily. A Na⁺ ion has an ionic radius of 1.02 Å (34% larger than Li⁺). The formation of NaC₆ would require a larger interlayer expansion, and calculations show the energy of the NaC₆ stage-I compound is thermodynamically unfavourable — sodium does not bind to the graphene layers with sufficient stability to drive intercalation at the relevant electrochemical potentials. In practice, only trace sodium can be stored in graphite (below 35 mAh/g), compared to 372 mAh/g theoretical capacity for lithium in graphite. Hard carbon works because its disordered structure has nanopores (0.5–2 nm diameter) that accommodate sodium through a different mechanism — sodium clusters (pseudo-metallic, quasi-zero-dimensional storage) in the pores rather than forming an ordered intercalation compound between layers.

What is the O3/P2 structural notation for layered oxides and what does it mean physically?

The O3/P2 notation was introduced by Delmas et al. in 1980 to classify layered transition metal oxide structures. The notation has two components: a letter (O or P) and a number (2 or 3). The letter indicates the coordination environment of the alkali ion (Na or Li) in the interlayer space: O = octahedral (the alkali ion is surrounded by six oxygen atoms forming an octahedron), P = prismatic (the alkali ion is surrounded by six oxygen atoms forming a triangular prism). The number indicates how many transition metal oxide layers are stacked before the pattern repeats in the c-axis direction: 2 = two layers per repeat unit (ABBA...), 3 = three layers per repeat unit (ABCABC...). O3-type: sodium in octahedral sites, three-layer repeat — typical of fully sodiated NaxMO₂ materials (x close to 1.0). At high sodium content, O3 is favoured. P2-type: sodium in prismatic sites, two-layer repeat — stable at lower sodium content (x = 0.6–0.7). P2 structures tend to have better rate capability (sodium can move through prismatic sites faster) but start with lower sodium content, limiting theoretical capacity. During cycling, O3 materials often transition through O3 → O3' → P3 → P2 as sodium is extracted — these phase transitions involve Na/vacancy ordering and can be disruptive to the crystal structure, causing capacity fade.

What are Prussian Blue Analogues (PBAs) and what makes them attractive for Na-ion batteries?

Prussian Blue (Fe₄[Fe(CN)₆]₃) is the dark blue pigment that has been manufactured since the early 18th century. Its crystal structure is an open framework of iron atoms connected by cyanide ligands (Fe-CN-Fe) in a cubic arrangement, creating a three-dimensional network of channels and cavities large enough to host alkali metal ions. Prussian Blue Analogues (PBAs) replace one or both iron atoms with other transition metals (Mn, Ni, Co, Cu): general formula A₂M[M'(CN)₆] where A is the alkali ion (Na in Na-ion), M is the nitrogen-coordinated metal, and M' is the carbon-coordinated metal. For Na-ion batteries: Na₂MnFe(CN)₆ (manganese hexacyanoferrate, MnHCF) and Na₂FeFe(CN)₆ (iron hexacyanoferrate, FeHCF) are the most studied. Advantages: synthesis is simple and low-temperature (just mixing aqueous solutions, no high-temperature calcination), iron and manganese are among the cheapest transition metals, the open framework easily accommodates Na⁺ ions with minimal strain (large cubic cavities). Disadvantages: moisture sensitivity (the cyanide ligands exchange water in the crystal vacancies, and water impedes Na⁺ transport), structural vacancies at the hexacyanometallate sites reduce practical capacity, and solvent for synthesis must be carefully controlled for high-performance PBA cells.

What is the 'house-of-cards' microstructure of hard carbon and how does it store sodium?

Hard carbon is not a single material but a family of disordered carbons produced by pyrolysis of organic precursors at 1,000–1,400°C. At these temperatures, the organic material decomposes and partially graphitises, but turbostratic disorder prevents full graphitisation — short-range graphene-like domains (pseudo-graphite nanocrystals, 1–3 nm in size) are stacked at random orientations, creating the 'house-of-cards' structural model. Between these random stacked domains are nanopores (0.5–2 nm diameter) — too small to be detected by normal mercury porosimetry but visible in small-angle X-ray scattering (SAXS) and BET surface area measurements. Sodium storage in hard carbon occurs through two distinct mechanisms operating at different voltages: (1) Intercalation into the pseudo-graphite nanocrystals (similar to graphite intercalation but in short-range ordered domains) — this occurs at 0.1–1.0V vs Na/Na⁺ and produces a sloping voltage profile. (2) Pore-filling (sodium clusters forming in the closed nanopores) — this occurs below 0.1V vs Na/Na⁺ and produces a flat voltage plateau, similar to metallic sodium deposition but inside nanopores rather than at the surface. The total practical capacity is 250–350 mAh/g — competitive with graphite for lithium (372 mAh/g theoretical) and significantly better than graphite for sodium (<35 mAh/g).

What is the initial Coulombic efficiency problem in hard carbon and why does it matter?

Initial Coulombic efficiency (ICE) is the fraction of charge put into a cell on its first charge that can be recovered on the first discharge. In graphite/NMC cells, ICE is approximately 90–93% — 7–10% of the charge goes into irreversible SEI formation rather than into stored capacity. In hard carbon/Na-ion cells, ICE is typically 70–85% — a 15–30% first-cycle irreversibility. The lower ICE is caused by: (1) larger surface area of disordered hard carbon (more surface for SEI formation), (2) sodium clusters in nanopores are partially irreversible (some sodium becomes trapped in pores that close upon first sodium insertion), and (3) the sodium-electrolyte interface chemistry produces thicker SEI than lithium-graphite. The practical consequence: in a full cell (cathode + anode), the cathode must provide excess sodium on first charge to compensate for the hard carbon's first-cycle loss. This is typically addressed by pre-sodiation — adding extra sodium to the cathode or the anode before cell assembly. Pre-sodiation adds manufacturing complexity but is essential for full-cell performance. Improving ICE above 85% in hard carbon is one of the primary research targets for Na-ion cell developers — it directly increases the effective energy density of the full cell.

References

Layered Transition Metal Oxides: O3 and P2 Structures
Prussian Blue Analogues (PBAs): Low-Cost Open Frameworks
NASICON-Type Frameworks: The Stable Alternative
The Hard Carbon Anode: Disordered Structure, Two-Mechanism Storage
The Initial Coulombic Efficiency Challenge
O3 Structure: High Sodium Content, Phase Instability
P2 Structure: Better Rate Capability, Lower Initial Capacity