What a Battery Actually Is

Open a lithium-ion cell — carefully, and not at home — and you will find no moving parts, no reservoir, nothing that looks like a source of power. What you will find is a long sandwich rolled up like a carpet: a sheet of copper foil coated in black powder, a sheet of aluminium foil coated in a different black powder, and between them a thin porous film, the whole thing soaked in a volatile organic liquid.

That is the entire machine. Its operation consists of moving one kind of atom from the left-hand powder to the right-hand powder, and then moving it back.

The shuttle

The word most people get wrong is storage. A battery does not store electricity. It stores a chemical arrangement that will produce electricity when permitted to relax, and in this particular chemistry, the arrangement is positional.

The negative electrode is graphite: carbon in sheets stacked like pages. The positive is a metal oxide whose crystal structure contains regular vacancies. Both are, structurally, car parks. Lithium ions occupy spaces within them without altering their frameworks — intercalation, and the reason these cells last thousands of cycles where earlier chemistries that plated metal onto a surface did not.

Charging drives lithium ions out of the oxide, through the liquid electrolyte, through the porous separator, and into the graphite. The electrons cannot follow through the liquid, which is an insulator, so they travel the long way round through the charger. Discharge reverses it: the ions return to the oxide and the electrons do useful work on the way.

Nothing is consumed. The same lithium shuttles back and forth for the life of the cell, which is why a battery at zero per cent contains exactly as much lithium as one at a hundred. Recycling an electric vehicle battery does not recover a spent fuel. It recovers materials that are chemically intact and merely in the wrong place, and this is the single most important fact about the industry's long-term resource position.

Why cells degrade anyway

If nothing is consumed, why does a battery wear out?

Chiefly because the organic electrolyte is not stable at the voltage graphite operates at. On the first charge it decomposes at the electrode surface into a thin passivating film. The film is not a defect — a cell will not function without it, and the formation step is among the most time-consuming stages of manufacture. But it keeps growing slowly for the cell's whole life, consuming a little lithium each time it thickens, and it grows faster with heat and with high states of charge. A battery held at a hundred per cent in a hot climate ages several times faster than one cycled between thirty and eighty in a temperate one. Most of what owners call degradation is this film, quietly thickening.

The second mechanism is mechanical. Electrode particles expand and contract as ions enter and leave — modestly for graphite, dramatically for silicon, which is why silicon offers far higher capacity and has taken two decades to incorporate in meaningful quantity. Repeated swelling cracks particles, exposes fresh surface, and grows more film.

The chemistry choice

Cells are named for their positive electrode, and the choice is a genuine trade-off rather than a ranking.

Nickel-rich oxides deliver the highest energy per kilogram and are used where mass matters most. They require cobalt for structural stability, tolerate abuse poorly, and demand careful thermal management.

Iron phosphate delivers twenty to thirty per cent less energy per kilogram and is, in exchange, markedly more stable when hot, cheaper, longer-lived, and free of both nickel and cobalt. Because it tolerates full charging without accelerated ageing, it recovers some of that disadvantage in practice. Its share of global production has risen sharply — the clearest evidence that the industry values cost and durability over specification-sheet density for most applications.

Sodium-ion works by exactly the same intercalation principle using a more abundant element, at lower energy density, and has begun to appear where mass is not the binding constraint — stationary storage, and short-range vehicles in cold climates, where it performs comparatively well.

Where the material comes from

The supply chain is the part of this industry that most commentary misunderstands, because the bottleneck is not extraction.

Lithium is obtained by two routes. Continental brine, pumped from beneath salt flats, is concentrated in evaporation ponds over twelve to eighteen months — a process driven by sunshine, cheap in energy and slow to scale, since responding to a price signal means starting a pond and waiting more than a year. Hard rock ore is mined conventionally, crushed, and concentrated into a mineral product that is then shipped, often across the world, for chemical conversion.

That conversion step is the actual constraint. Battery-grade lithium hydroxide or carbonate must meet purity specifications in the parts-per-million range, and the plants that produce it are expensive, slow to build, and geographically concentrated to a degree far exceeding the concentration of the mines themselves. The same is true downstream: cathode powder, anode-grade graphite and separator film are each produced by a small number of qualified suppliers, and qualification of a new source by a cell maker typically takes one to two years of testing before a single commercial cell is built.

The implication is that resource abundance and supply security are separate questions. Lithium is not geologically scarce. Refining capacity, qualified precursor supply and trained process engineers are, and none of them can be produced in the timeframe that a shortage creates.

The manufacturing problem

A cell factory is a coating and drying operation of extreme precision, closer to newspaper printing than to metalworking. Electrode slurry is coated onto moving foil at high speed, dried, compressed and slit. Tolerances are in micrometres; a contaminant particle of a few tens of micrometres can eventually pierce the separator and short the cell.

Consequently, yield governs economics. A new plant may begin at yields low enough that a substantial fraction of output is scrapped, and the ramp to acceptable yield routinely takes longer than the construction. Firms have announced capacity and shipped nothing for years, not because the design was wrong but because the process was not yet stable — and process stability is accumulated experience, which cannot be bought or accelerated by capital.

The trajectory

Against all of this, the price record is remarkable. Cell costs have fallen by roughly an order of magnitude since 2010, at a fairly consistent rate per doubling of cumulative production, driven not by a breakthrough but by thousands of incremental improvements in yield, format, packaging and chemistry.

That is the ordinary way industrial technologies get cheap, and it is worth remembering when reading about the next battery that will change everything. The chemistry that wins is rarely the one with the best laboratory numbers. It is the one that can be made, in millions of units, at consistent quality, by people who have made it before.