Battery ageing

Why does keeping a phone charged to full shorten the life of its battery?

A battery is rated in cycles, so the intuition writes itself: a battery wears out by being used, the way a tyre wears out by being driven on. Leave the phone plugged in at a hundred percent and you are using nothing — surely that is the gentlest life available.

Yet the advice from the people who build these cells runs the other way. Do not store a device full. A laptop that lived on its charger for two years can have a worse battery than one carried about daily. What could a battery be doing, sitting still and full, that costs it anything?

Start with what "full" means physically. In a lithium-ion cell, charging drives lithium ions out of the cathode and into the graphite anode, where they sit between the carbon layers. A full battery is one whose anode is stuffed with lithium.

Now ask what that does to the anode's chemistry, not just its contents. A lithiated graphite anode sits at a potential very close to lithium metal — which is to say strongly reducing, chemically aggressive toward whatever it touches. What it touches is the liquid electrolyte, which is not stable at that potential. The electrolyte is attacked, and the products build a thin film on the anode surface: the solid electrolyte interphase, or SEI.

Here is the pivot. That film is not simply damage — the cell cannot work without it. The SEI passivates the anode, letting lithium ions through while blocking further electron transfer to the electrolyte. It is the scab that stops the wound. But building it consumes lithium, and that lithium is locked away for good. Every atom spent on the film is one no longer available to shuttle back and forth, so the cell's capacity drops by exactly that much.

Does the film grow forever? Not at a constant rate — as it thickens it slows itself, because reactants must diffuse through it to reach fresh surface, and growth roughly follows the square root of elapsed time. But notice the two words that answer our question: elapsed time. This clock runs whether or not you touch the phone. That is calendar ageing, and it depends on only two things you control — how hot the cell is, since chemistry accelerates with temperature, and how full it is, since a fuller anode is a more aggressive one. A phone held at a hundred percent in a warm room runs that reaction at its fastest available setting, continuously, for nothing.

So what does cycling actually add? Something mechanical. Graphite swells as lithium enters it and shrinks as it leaves — around a tenth in volume, every cycle. Particles crack, exposing fresh unpassivated surface, which immediately grows new SEI and spends more lithium. And if you charge fast, or charge cold, ions arrive faster than they can slot into the graphite, so some plate out as metallic lithium instead. That lithium is largely lost, and the deposits are the origin of the dendrites that make abused cells dangerous.

One more distinction, because it explains symptoms people misread. Losing lithium reduces how much charge the cell holds: capacity fade, felt as a shorter day. But the same films and cracks also make the cell harder to push current through: impedance rise, felt as a voltage that sags whenever the processor works hard. A phone that dies at thirty percent in the cold is usually not empty — it is resistive. The charge is there; the cell cannot deliver it fast enough to hold its voltage above the cutoff.

"Cycles" is a poor mental model for wear because it hides half the story. A cycle count treats ageing as a tally of uses, when two clocks run at once: a chemical one that never stops, and a mechanical one that ticks only when you use the cell. Hence the paradoxical-sounding advice — a battery kept cool and near the middle of its range does better than one kept untouched and full, because the untouched one still ages, and ages faster.

None of which means charging to full is harmful in itself. Time spent at high charge is what costs, not arriving there. Charging fully before a journey and then draining the cell is unremarkable; sitting at a hundred percent for weeks is the expensive part. That is what the "optimised charging" features in modern phones and cars exploit — holding the cell lower overnight and topping it up shortly before you need it.

An honest caveat: the broad picture here — SEI growth, plating, particle fracture — is well established, but the relative weight of each mechanism varies a great deal with chemistry, cell design and pack management, and quantitative lifetime prediction remains an active research area rather than a settled formula.

• Why lithium iron phosphate cells tolerate being kept full far better than nickel-rich ones.
• Why fast charging in the cold is the single most damaging thing routinely done to electric cars.

Every term the collection defines is gathered in the glossary.