Why Lithium Won, and Why Nothing Has Replaced It

Magnesium. Aluminium. Sodium. Zinc. Each announcement is technically accurate about the advantage it claims, and each one quietly omits the thing that has kept lithium in place for thirty years.

It is worth understanding what that thing is, because it tells you which announcements to take seriously — and there is one you should.

Lithium is element number three. Only hydrogen and helium are lighter, and both are gases. It sits at the extreme top-left of the periodic table, and almost everything that matters about it follows from that one coordinate.

• •It is the lightest metal — 6.94 g/mol at a density of 0.534, so it floats on water and on cooking oil. Capacity per kilogram has molar mass in the denominator, so being light buys storage for free: lithium metal’s theoretical capacity is 3,862 mAh/g.
• •It is the most electropositive element there is, at −3.045 V. Cell voltage is cathode potential minus anode potential, so lithium extracts the highest possible voltage from whatever cathode you pair it with. The same cathode against zinc gives less than half the voltage, and therefore less than half the energy.
• •Its ion is small and carries only one charge. Li⁺ is 76 picometres across, and that combination — small and singly charged — is unique in the periodic table. It is the property nobody talks about, and the one that actually decides everything.

Lithium has the highest first ionisation energy of all the alkali metals — 520 kJ/mol against caesium’s 376. It holds its outer electron more tightly than any of its group. By that measure it should be the worst reducing agent in the family, not the best.

It isn’t, because electrode potential is not ionisation energy. It is a three-step accounting: turn the solid into gas, strip off the electron, then surround the resulting ion with solvent. Lithium’s tiny ion pulls solvent molecules in so tightly that the energy released in the third step more than repays what the second cost. It loses the middle round and wins the match.

That is a nice piece of chemistry, and it is not why the alternatives have failed.

Here is the fact that catches most people out: magnesium’s ion is smaller than lithium’s. Mg²⁺ is 72 pm against Li⁺’s 76. Aluminium’s is smaller still at 53.5 pm. And magnesium carries two electrons per ion to lithium’s one, aluminium three.

On paper both should win. Aluminium has the highest volumetric capacity of any metal — better than lithium by a factor of four. It is the third most abundant element in the crust, and already an industrial commodity produced by the million tonnes.

So size is not the problem. Charge is.

A cation’s grip on its surroundings scales roughly as charge divided by radius squared. Run that for magnesium and it binds its neighbours about 2.2 times as hard as lithium does. For aluminium, six times as hard.

That grip has to be paid twice on every single hop — once to leave the site the ion is sitting in, once to squeeze through the gap to the next one. In a layered oxide a lithium ion faces a barrier of roughly 0.2 to 0.4 eV. Magnesium faces 0.6 to 1.2 eV.

There is no clever engineering around Coulomb’s law.

Even if you solved the kinetics, magnesium has a separate difficulty that is arguably worse.

Lithium-ion cells work because of a fortunate accident. The electrolyte is not actually stable at lithium potentials — it decomposes on first charge. But its decomposition products happen to be ionically conductive and electronically insulating. Lithium ions can pass through the film; electrons cannot. So the decomposition stops itself after a few nanometres, and what is left is a protective layer the cell then runs on for years. That layer is the SEI, and it is the single most important accident in battery history.

Magnesium in the same electrolyte forms magnesium oxide and magnesium carbonate. Those films are insulating to electrons and blocking to magnesium ions. The electrode passivates against its own working ion, and the cell simply stops.

That — not capacity, not voltage — is why magnesium batteries have been four decades in the laboratory.

Sodium. It is the one credible successor, and it is credible precisely because it fails on none of the above. Same group as lithium, same chemistry family, −2.71 V, forms a working SEI, and a thousand times more abundant in the crust.

Its problems are ordinary engineering problems rather than physics walls. The sodium ion is 34 per cent larger and the atom is 3.3 times heavier, so capacity drops to 1,166 mAh/g. And sodium will not intercalate into graphite in any useful quantity, so sodium cells need a different anode — hard carbon — with a less convenient voltage curve.

That is a cost-versus-energy trade, not a dead end. Which is exactly why sodium-ion is now in commercial production and magnesium is not.

The full argument — with the Born–Haber cycle behind the electrode potential, the Arrhenius arithmetic behind the migration barrier, and a candidate-by-candidate account of hydrogen, potassium, calcium, zinc and beryllium — is in The Periodic Table of the EV.