New nanoscale experiment reveals why dendrites cripple lithium batteries

A nanoscale experiment has now produced a remarkable answer.

Whether it’s a smartphone, an e-bike or an electric car, lithium batteries are now built into almost everything. Yet cells often lose capacity far sooner than brochures suggest - and in the worst cases, a short circuit and fire can follow. A US research group has now watched a crucial mechanism in action and, in doing so, exposed a decades-old misunderstanding in battery research.

Tiny metal needles can disable entire batteries

Inside every lithium-ion battery, an unseen troublemaker can emerge: so-called dendrites. These are ultra-fine metallic structures that form on the lithium anode during charging. They push out into the electrolyte like miniature needles or branches - as delicate as spider silk, but with outsized consequences.

These dendrites are about a hundred times thinner than a human hair. Even so, they can cause serious damage: if they grow far enough, they pierce the separator - the thin layer that keeps the battery’s positive and negative sides apart.

When that happens, an internal short circuit forms. Electrons then stop flowing through the intended device and instead pass directly from one electrode to the other. The results range from noticeable capacity loss and intense heating to thermal runaway - the feared scenario in which a cell catches fire.

Researchers describe “metallic micro-needles” that silently drill through a battery and, in the worst case, puncture it from the inside.

Until now, many specialists assumed dendrites were relatively soft - roughly as soft as the bulk lithium they are made from. As a result, efforts often focused on trying to “push them back” or cushion them. That underlying assumption is now being seriously challenged.

New experiment shows dendrites are hard as glass

A team from the New Jersey Institute of Technology and Rice University has, for the first time, deliberately put dendrites under mechanical load while observing them in an electron microscope. They did this under high vacuum so the structures would not immediately react or oxidise. The aim was to see not only what dendrites look like, but how they actually behave under pressure.

What they observed is unexpected: dendrites do not bend, and they do not buckle elastically - they snap. The researchers liken the behaviour to dry spaghetti: apply a little force and there is a clean break, with no visible deformation beforehand.

Measurements show these tiny needles can withstand mechanical stresses of around 150 megapascals. Bulk lithium, by contrast, reaches only about 0.6 megapascals. In other words, dendrites are about 250 times stronger than the same material in its “normal” form.

The reason lies in an unassuming surface layer: within fractions of a second, a reaction or oxide film only a few nanometres thick forms on the dendrites’ exterior. This shell turns an inherently soft metal into a brittle, hard structure - effectively a micro-nail with a ceramic-style armour.

A wafer-thin reaction layer turns soft battery metal into a brittle micro-harpoon spear that cannot be pushed aside and instead stabs straight through the separator.

This insight forces a rethink of many existing safety strategies. Designing for a soft adversary is very different from dealing with a stiff, brittle material that can fracture under load into sharp shards.

“Dead” lithium eats away capacity cycle by cycle

Dendrites’ brittleness creates a second issue that looks less dramatic, yet quietly ruins batteries over time. When one of these needles snaps under stress, small pieces of lithium are left behind in the electrolyte. These fragments are electrically isolated and no longer take part in charging and discharging.

In technical terms, this material is called “dead lithium”. With every charging cycle, more of these inactive islands can form. For users, it typically shows up like this:

• a phone lasts only half a day after a year,
• an electric car’s range drops noticeably even though everything else seems fine,
• solar storage systems deliver less energy than originally specified.

The amount of active lithium in the cell shrinks even though, externally, nothing appears to have changed. That means the cell reaches its practical end of life far earlier than simple calculations would suggest.

Why lithium-metal batteries have struggled so far

The findings are particularly relevant for the next generation of cells: lithium-metal batteries. In labs and development programmes, they are widely seen as a major hope for the future. Instead of a graphite anode, they use near-pure lithium - promising a dramatically higher energy density.

Put simply: where today’s electric cars might, at best, manage 300 to 400 kilometres per charge, lithium-metal cells could make 800 to 900 kilometres conceivable - without making the battery pack enormous. That is precisely the target of multi-billion investments by car makers and battery start-ups.

The catch is that dendrites form especially easily in these cells. And in that context, their brittleness becomes fully destructive. Even supposedly robust solid electrolytes - often promoted as a “miracle cure” - can be perforated by these hard needles because their mechanical strength was underestimated.

The new measurement shows that many concepts for safe solid-state batteries simply underestimate how strong dendrites really are.

This helps explain why prototypes can post impressive results in the lab but then fail abruptly in long-term tests. The bottleneck is less about chemistry and more about mechanics on the nanoscale.

Three material strategies to tame dendrites

The researchers propose three concrete materials-based routes to at least mitigate dendrite formation:

• New lithium alloys: Pure lithium reacts extremely quickly at its surface. By alloying it with other metals, it may be possible to influence how the brittle surface layer forms. The aim would be dendrites that are less hard - and therefore less destructive.
• Smarter separators: Instead of relying on thin plastic films alone, batteries need mechanically adaptable layers that absorb stress before a dendrite tip can punch right through. Multi-layer separators combining soft and hard zones are one possible approach.
• Targeted electrolyte additives: Special additives in the electrolyte could alter the crystal structure of dendrites as they grow. That would allow control over whether they spread more laterally rather than developing into sharp needle-like forms.

Taken together, these approaches offer a realistic chance of boosting the range of future electric vehicles substantially, without constantly battling fire risks or rapid capacity fade.

What this means for electric cars and the energy transition

For the automotive industry, the result is both awkward and encouraging. It is awkward because many development programmes were built on assumptions that now look wrong. It is encouraging because a clear diagnosis is often the first step towards solutions that work outside the lab.

If dendrites can be controlled mechanically, high-energy-density batteries could ease several problems at once: smaller battery packs, lower raw-material demand, cheaper vehicles and longer service life. That would make electric cars more appealing not only for high-mileage drivers, but also support large-scale storage expansion for wind and solar power.

For consumers, the takeaway is very practical: battery longevity depends not only on chemistry, but heavily on internal mechanics. Regular extreme fast charging puts cells under higher stress - which favours dendrite growth. More conservative charging habits, such as slower overnight charging and avoiding a constant 100% state of charge, can slow the process even if they cannot stop it entirely.

How a mistaken assumption held back research for decades

The new work also illustrates how risky it can be when basic assumptions go untested for too long. Because dendrites look so similar to bulk lithium, it seemed obvious to treat their strength as comparable - and hardly anyone directly measured it. Only precise nanoscale observation has now cleared up that misconception.

Studies at this level of detail will become even more important in the coming years. As battery technology and electrification accelerate, whole industries become increasingly sensitive to small modelling errors. A misunderstood mechanism on the micro- or nanoscale can ultimately determine the range, cost and safety of entire generations of vehicles.