LFP vs NMC batteries — how do you choose?

Choose LFP for grid storage and NMC where weight matters, such as long-range electric vehicles. NMC has the higher energy density but contains cobalt, enters thermal runaway at around 210°C and lasts 2,000–3,000 cycles. LFP is cobalt-free and cheaper, with a thermal runaway threshold near 270°C and 8,000–12,000 cycles. LFP is expected to take more than 70% of new grid storage additions worldwide by 2030.

Every battery decision is a trade-off among energy density, safety, cost and cycle life. NMC's advantage is energy density: more stored energy per unit of volume or mass, which is decisive for cutting vehicle weight. Its drawbacks are equally clear — cobalt is expensive and comes mainly from the Democratic Republic of the Congo, with serious human rights problems attached; the cathode readily releases oxygen when overheated or overcharged, making thermal runaway more likely than with LFP; and cycle life is typically only 2,000–3,000. LFP has lower energy density but wins on almost every other axis: no cobalt, lower cost, a thermal runaway threshold of about 270°C against NMC's roughly 210°C, and 8,000–12,000 cycles.

For grid storage the ranking metrics are safety, cost and cycle life, not weight — a container bolted to the ground has no reason to diet. That is why LFP is taking over stationary storage, with over 70% of new global grid storage additions expected to be LFP by 2030. Product signals point the same way: Tesla's Megapack line has moved entirely to LFP chemistry, and CATL's 2023 Cell-to-Pack (CTP) design achieves 5 MWh in a single 20-foot container with integrated liquid cooling.

The cost curve makes the choice more lopsided still. Lithium carbonate peaked above RMB 700,000 per tonne in 2022, then collapsed below RMB 100,000 in 2023–2024, a fall of more than 80%. Combined with manufacturing scale, this pushed LFP pack prices below USD 100/kWh for the first time in 2024; by 2025 the BNEF global average pack price fell to about USD 108/kWh, with stationary storage packs reaching as low as USD 70/kWh — the point at which battery storage's lifetime cost undercuts a new gas peaker, often called grid parity for storage. Regional spreads are wide: Chinese-made LFP packs average about USD 84/kWh against roughly USD 121 in North America and USD 168 in Europe.

The safety gap becomes concrete inside the thermal runaway chain. Overcharge, overheating or mechanical damage starts the solid electrolyte interphase (SEI) decomposing exothermically at around 90°C; at 130–150°C the electrolyte reacts with the anode and the separator melts, shorting the electrodes internally; above 200°C the cathode decomposes and releases oxygen, so a sealed cell suddenly contains its own oxidiser, and the heat and flammable gas propagate to neighbouring cells. Modern BESS designs must therefore satisfy NFPA 855 and UL 9540A, whose core criterion is zero propagation: one cell in thermal runaway must not ignite its neighbours. LFP's roughly 60°C higher threshold is the cheapest insurance on that chain. Supply chains also shape the choice: China holds about 58% of global lithium refining and close to 100% of graphite processing capacity, the US Inflation Reduction Act offers a USD 35/kWh credit for North American cell manufacturing, and the EU Battery Regulation requires a Battery Passport tracking carbon footprint and recycling rates.

For grid storage, the metrics that matter are safety, cost and cycle life — not weight.

This question is covered in depth in The Full Spectrum Every Energy Source Explained — A Singapore Perspective，第十三章 13.5–13.6