All chapters
Framework anionsInductive effectLFP voltageOxygen release1-D diffusion

Chapter 4 left iron in an awkward position: the ideal cathode metal on every economic axis, with a redox couple sitting at about 3.0 V — too low to build a product around.

The fix does not touch iron at all. It changes what surrounds the oxygen atoms that surround the iron, using an element that stores no charge and adds dead weight. It is the most elegant piece of periodic-table reasoning in battery materials, and it produced the cathode that now dominates cost-sensitive storage worldwide.

5.1Why oxygen is the framework anion

Oxygen is the second most electronegative element at χ = 3.44, sits in the second period, has a small ionic radius of 140 pm as O²⁻, is divalent, and is the most abundant element in the crust at 461,000 ppm. Each of those facts earns it a job.

  • High electronegativity gives highly ionic metal–oxygen bonds, which hold the metal at a high oxidation state, which gives a high redox potential.
  • Small and light means dense packing and low dead mass.
  • Divalent means it forms robust three-dimensional networks rather than terminating chains.
  • Second period with no d orbitals means no competing bonding modes and predictable geometry.

5.2What changes if you go down to sulfur

Sulfur sits directly below oxygen in group 16, and the group trend does exactly what chapter 1 predicts: it is larger at 184 pm, less electronegative at 2.58, and considerably more polarisable.

The consequence is that the less ionic metal–sulfur bond destabilises the high metal oxidation state, which pulls the redox potential down. Sulfide cathodes give lower voltages, and lithium–sulfur runs near 2.1 V.

Technical framing

Sulfur’s compensating advantage is that it works by conversion rather than intercalation — the whole S₈ ring is consumed, giving 1,675 mAh/g. That is extraordinary capacity, but polysulfide dissolution and shuttling remain commercially unsolved, which is why Li-S has been five years away for twenty years.

5.3The inductive effect — how LFP gets its voltage

The inductive effect is the transmission of electron density along chemical bonds due to electronegativity differences, altering the electron density — and therefore the energy levels — at a distant atom.

In LiFePO₄, iron is octahedrally coordinated to oxygen, as it would be in any oxide. But each of those oxygens is also covalently bonded to a phosphorus atom in a PO₄³⁻ tetrahedron.

Important

Phosphorus at χ = 2.19 is far more electronegative than iron at 1.83. The strongly covalent P–O bond pulls electron density away from the oxygen atoms, which in turn pull less electron density from iron. The Fe–O bond therefore becomes more ionic.

The result: the Fe²⁺/Fe³⁺ couple is lowered in energy, which means it is raised in voltage — from about 3.0 V in a simple oxide to 3.45 V vs Li/Li⁺ in the phosphate.

SIMPLE OXIDE — Fe–OFeOe⁻ density pulled toward OFe²⁺/Fe³⁺ ≈ 3.0 Vtoo low to be commercially usefulOLIVINE — Fe–O–PFeOPP (χ 2.19) pulls harder than Fe (χ 1.83)Fe–O becomes MORE ionic → Fe redox level LOWERED in energyFe²⁺/Fe³⁺ = 3.45 V2.83.6 V vs Li/Li⁺3.03.45+450 mV, contributed entirely by an atom that stores no charge
Figure 5.1This is why LFP exists as a product. Phosphorus contributes no capacity and adds mass; it earns its place purely by shifting iron's redox energy. And it gives a second, larger gift: the strong covalent P–O bond locks oxygen into a rigid tetrahedron, so LFP does not readily release O₂ on overcharge. LFP's celebrated safety is not a property of iron — it is a property of the phosphorus–oxygen bond.

In plain English

Phosphorus contributes no capacity whatsoever and adds mass to every gram of cathode. It earns its place purely by reaching through an oxygen atom and shifting iron’s redox energy — and that shift is the entire reason LFP exists as a product rather than a curiosity.

5.4The second gift — covalent P–O bonds and safety

The phosphate group’s P–O bonds are strong and covalent, with a bond dissociation energy around 600 kJ/mol. Oxygen in LFP is locked into a rigid tetrahedron, not merely coordinated to a metal.

Oxygen release onset — where the difference actually shows up

CathodeApproximate O₂ release onsetConsequence
LiFePO₄~250–270 °CNo self-sustaining oxidiser supply; thermal events stay local
Nickel-rich layered oxide~150–200 °CReleases O₂ that feeds its own combustion

Why this matters

Restate that as a periodic-table sentence: LFP’s celebrated safety is not a property of iron. It is a property of the phosphorus–oxygen covalent bond, which is a consequence of phosphorus sitting in period 3, group 15.

An iron cathode without the phosphate group would be neither safe nor useful. The safety and the voltage came from the same substitution.

LAYERED — LiCoO₂ / NMCR-3m · 2-D DIFFUSIONMO₆ SLABMO₆ SLABMO₆ SLABMO₆ SLABNi²⁺ on a Li site— channel blockedLi moves freely within each planeHighest capacity · 150–220 mAh/gWeak point: cation mixing, O releaseSPINEL — LiMn₂O₄Fd-3m · 3-D DIFFUSIONInterconnected 3-D tunnel networkBest rate capability · 4.1 V · cheapWeak point: Jahn–Teller, Mn dissolutionOLIVINE — LiFePO₄Pnma · 1-D DIFFUSIONone defectkills thewhole channelLi moves only along the b-axis3.45 V · safest · cheapest · 160 mAh/gWeak point: σ = 10⁻⁹ S/cm, 1-D blockageBLUE = TRANSITION-METAL OCTAHEDRON MO₆ · RED TRIANGLE = PHOSPHATE TETRAHEDRON PO₄ · RED DOT / CHANNEL = LITHIUM
Figure 5.2Dimensionality decides the engineering. Layered oxides give lithium a whole plane to move in, so micron-scale particles work — but the same open plane invites nickel into lithium sites. Spinel offers a 3-D tunnel network and the best rate capability, spoiled by manganese chemistry. Olivine offers exactly one direction, which is why commercial LFP is always nano-sized: shorten each channel until a blocking defect becomes statistically unlikely, and shorten the electron path until the carbon coating can compensate for a conductivity of 10⁻⁹ S/cm.
2.53.03.54.00%25%50%75%100%~100 mV ACROSS 70% OF CAPACITYLFPNMCCELL VOLTAGE / V vs DEPTH OF DISCHARGEflat plateau ⇒ voltage cannot report SOC ⇒ coulomb counting required
Figure 5.3Olivine LFP cycles through a two-phase reaction — at any moment the particle contains regions of LiFePO₄ and regions of FePO₄, and a two-phase equilibrium holds the voltage constant. Hence the plateau. Layered NMC cycles as a solid solution — lithium content varies continuously, and so does the voltage. The engineering consequence is direct: LFP's ~100 mV spread over 70% of capacity means voltage cannot report state of charge, so coulomb counting with periodic recalibration at the curve's ends is not optional.

5.5The price LFP pays

Three costs, and the first two trace to the identical structural feature.

5.5.1Terrible intrinsic electronic conductivity

About 10⁻⁹ S/cm, against roughly 10⁻³ S/cm for LiCoO₂ — six orders of magnitude worse. The isolated PO₄ tetrahedra break up the metal–oxygen network that would otherwise allow electron delocalisation. There is no continuous Fe–O–Fe path for electrons to travel along.

The fix is a carbon coating of 1 to 3 weight per cent, a few nanometres thick, applied by pyrolysing a carbon precursor during synthesis — plus nano-sizing to shorten the electron path.

5.5.2One-dimensional lithium diffusion

In the olivine structure, Li⁺ moves only along the b-axis channels. A single blocking defect — an iron atom on a lithium site, or an impurity — shuts down that entire channel for the length of the particle.

The fix is again nano-sized particles, so each channel is short and the probability of a blocking defect per channel is low. This is why LFP is always made as nanoparticles, while LiCoO₂ can be made at 10 µm. The arithmetic behind that is worked through in chapter 9.

5.5.3Lower energy density

Phosphorus and its four oxygens are dead weight. LFP delivers about 160 mAh/g at 3.45 V, roughly 550 Wh/kg at material level, against NMC811 at about 200 mAh/g and 3.8 V, roughly 760 Wh/kg.

Important

The mass penalty is the price of the inductive effect and the safety. They are not three separate trade-offs — they are one substitution, paid for once and collected three times.

Quick check: test yourself

1.Someone tells you LFP is safe "because iron is stable." What is wrong with that?

Show answer
Iron is not doing the work. The safety comes from the ~600 kJ/mol covalent P–O bond locking oxygen into a rigid tetrahedron, which is why the O₂ release onset is 250–270 °C rather than 150–200. A simple iron oxide cathode would be neither safe nor high enough in voltage to use.

2.Phosphorus stores no charge and adds mass. Why include it?

Show answer
Because the inductive effect it exerts through oxygen lifts the Fe²⁺/Fe³⁺ couple from about 3.0 V to 3.45 V. Without that shift the chemistry is not commercially interesting, so the mass penalty buys the voltage that makes the product viable — and the oxygen locking as a bonus.

3.Why can LiCoO₂ be made at 10 µm while LFP cannot?

Show answer
Two structural penalties from the polyanion framework: electronic conductivity six orders of magnitude worse because isolated PO₄ tetrahedra break the Fe–O–Fe network, and one-dimensional b-axis diffusion where a single defect blocks a whole channel. Nano-sizing shortens both paths.

Chapter summary

Frequently asked questions

What is the inductive effect and how does it give LFP its voltage?+

It is the transmission of electron density along chemical bonds due to electronegativity differences, changing the energy levels at a distant atom. In LiFePO₄ each oxygen bonded to iron is also covalently bonded to phosphorus in a PO₄³⁻ tetrahedron. Phosphorus at χ = 2.19 is far more electronegative than iron at 1.83, so the P–O bond pulls electron density away from oxygen, which in turn pulls less from iron, making the Fe–O bond more ionic. That lowers the Fe²⁺/Fe³⁺ couple in energy, which raises it in voltage — from about 3.0 V in a simple oxide to 3.45 V in the phosphate. That shift is the entire reason LFP exists as a product.

Why is LFP safer than NMC?+

Because of the phosphorus–oxygen bond, not because of iron. P–O bonds are strong and covalent at roughly 600 kJ/mol, so oxygen in LFP is locked in a rigid tetrahedron rather than merely coordinated to a metal. LFP therefore does not readily release O₂ on overcharge or heating: the oxygen release onset is around 250 to 270 °C against roughly 150 to 200 °C for nickel-rich layered oxides, and there is no self-sustaining oxidiser supply. Restated as periodic chemistry, LFP’s safety is a consequence of phosphorus sitting in period 3, group 15.

Why does LFP have to be made as nanoparticles?+

Because of two structural penalties that come with the same polyanion framework. Its intrinsic electronic conductivity is about 10⁻⁹ S/cm against 10⁻³ for LiCoO₂, since isolated PO₄ tetrahedra break up the continuous Fe–O–Fe network that would allow electron delocalisation. And lithium diffuses only along one-dimensional b-axis channels in the olivine structure, so a single blocking defect shuts down an entire channel. Nano-sizing shortens both the electron path and each diffusion channel; carbon coating supplies the missing electronic conduction.

Why are sulfide cathodes lower voltage than oxides?+

Sulfur sits below oxygen in group 16, so it is larger at 184 pm against 140, less electronegative at 2.58 against 3.44, and more polarisable. The less ionic metal–sulfur bond destabilises the high metal oxidation state, which pulls the redox potential down — Li-S runs near 2.1 V. Sulfur’s compensating advantage is that it works by conversion rather than intercalation, consuming the whole S₈ ring for 1,675 mAh/g, but polysulfide dissolution and shuttling remain commercially unsolved.

Reviewed by

SG

Sahil Goyal

Co-founder, Wingzman

LinkedIn
SG

Sourabh Goyal

Co-founder, Wingzman

LinkedIn

The Periodic Table of the EV is an original educational series on the materials science of electric vehicles. All values are standard-condition literature figures for representative materials, not measured data from a specific product, and sources differ on several of them. Always verify against the specific material datasheet in use before making design, purchasing or certification decisions.