Why Batteries Still Need Cobalt, Even a Little

Every material on this site could be described purely by where it comes from and how much of it exists. Cobalt is a good place to show why that is not enough: its actual atomic-scale behaviour, both inside its own metallic lattice and inside a battery cathode, is the reason it has proven so hard to fully remove from lithium-ion batteries, despite years of real commercial and ethical pressure to do exactly that.

One element, two crystal structures

Pure cobalt metal is allotropic: it exists in two different crystal structures depending on temperature, a genuinely unusual property among structural metals (iron does the same thing, which is part of why steel heat treatment works at all). Below about 422°C, the stable form is hexagonal close-packed (HCP), sometimes labelled ε- or α-cobalt. Above that temperature, it transforms into face-centred cubic (FCC), labelled β-cobalt. Both are real, physical arrangements of the same cobalt atoms, just packed differently, and the drag-to-rotate models below are both built from the same standard crystallography used to describe them, not illustrations.

Why the “stable” form is not always the one you get

If cobalt simply followed thermodynamics, every piece of room-temperature cobalt would be HCP. In practice, metastable FCC cobalt is common at room temperature, because the HCP-to-FCC transformation is governed by how fast atoms can rearrange, not only by which structure has lower energy. That rearrangement rate follows the same general form used across materials science for any thermally activated process:

At and below room temperature, $RT$ is small relative to the activation energy for this particular rearrangement, so the rate k is tiny: the transformation essentially freezes in place, leaving whichever structure existed when the metal last cooled through 422°C, or formed directly from solution or vapour, largely untouched. This is also why cobalt has an unusually low stacking-fault energy: the HCP and FCC structures differ only in the stacking sequence of their close-packed atomic layers (ABAB for HCP, ABCABC for FCC), so a partial, localised slip of one layer is enough to locally convert one structure into a thin sliver of the other. Real cobalt samples, including the cobalt in WC-Co cutting tool binders discussed below, routinely contain both structures mixed together at the scale of individual grains, a direct, visible consequence of this kinetic trap, not a flaw in the material.

The role inside a cathode

In layered oxide cathodes like LiCoO₂ and the nickel-manganese-cobalt (NMC) family, lithium ions shuttle in and out of a layered crystal structure during charging and discharging, while the transition metal layer (cobalt, nickel, manganese, or a mix) holds that structure together. Cobalt specifically helps stabilise the layered structure and suppresses a failure mode called cation mixing, where nickel ions migrate into lithium’s layer and gradually degrade the structure over repeated cycling. Cobalt also helps limit oxygen release from the cathode at high states of charge, a real safety-relevant mechanism connected to thermal runaway risk, not a marginal effect.

Why “just remove it” is a real engineering problem, not a choice

Manufacturers have genuinely reduced cobalt content over the past decade, moving from cathodes with roughly a third cobalt (NMC 111) toward nickel-rich formulations with far less (NMC 811 and beyond), and lithium iron phosphate (LFP), which contains no cobalt at all, has grown sharply in market share for exactly this reason. But nickel-rich, low-cobalt cathodes are measurably less thermally stable and degrade faster at high nickel content, a real materials trade-off between cost and ethics on one side and safety and cycle life on the other, not a case of cobalt being included out of habit. This is why cobalt has not disappeared from high-energy-density cathodes even as its overall share of the battery market has fallen.

Cobalt as a binder, not just a cathode metal

Batteries are not cobalt’s only materials-science job. In cemented tungsten carbide (WC-Co), the material behind most metal-cutting tool inserts and mining drill bits, cobalt is the ductile metallic binder phase holding a skeleton of hard, brittle WC grains together. Because cobalt metal can exist in either structure described above, and real WC-Co binders often retain a metastable FCC-rich character at room temperature (the same kinetic trap, not a separate phenomenon), the binder stays tough enough to absorb impact energy that the WC grains alone could not. Cobalt-based superalloys (Haynes- and Stellite-type alloys) use the same FCC cobalt matrix, the high-temperature-stable structure above, deliberately held stable at service temperature by alloying, combined with solid-solution strengthening and carbide precipitates, to resist creep and wear in gas turbine and jet engine components. Both applications exist because of cobalt’s specific crystallography, not despite it.

Sources: lattice parameters and the HCP-FCC transition temperature are standard values from metallurgical references (ASM Handbook, Vol. 2). Cathode degradation mechanisms and NMC/LFP market-share trends draw on standard battery materials literature; production context references the production data elsewhere on this site.