An electric vehicle (EV) battery comes from one of the most spread-out supply chains in modern industry. Before it ever powers a car, it uses minerals mined on one continent, refined on another, and turned into cells somewhere else again. Following that road is one of the best ways to see why the real problem is concentration and chokepoints, not running out of the minerals themselves.
What is actually in the battery
A modern EV battery depends on a short list of minerals, each with its own story.
Lithium is in almost every battery type, and graphite forms the anode in nearly every cell. The bigger differences come from the cathode, the side of the cell that largely sets how much energy it holds. The cathode chemistry that dominated for years is nickel manganese cobalt (NMC), valued for range.
None of these are truly rare in the ground. The problem is where they are concentrated, and even more, where they are processed.
Mining is only half the story
It is easy to focus on mines, but the tighter point usually sits one step later, in refining and processing.
Raw ore has to be refined into battery-grade material, then turned into the cathode and anode material that goes into a cell. This middle stage is far more concentrated than mining, with China holding a dominant position in refining most battery minerals and in making cathode and anode material.
The recent move toward lithium iron phosphate (LFP) batteries makes this sharper. LFP has grown from a small part of the EV market to about half, because it is cheaper and has improved. But the LFP supply chain is even more concentrated, with most LFP cathode material and cells made in a single country. A chemistry that lowers reliance on cobalt can, at the same time, raise reliance on one place.
A road full of chokepoints
Follow the physical path and the risk becomes clear. Minerals move from mine to refinery to material maker to cell maker to vehicle assembly, often crossing several continents and oceans on the way.
That long route passes through chokepoints, narrow passages like major canals and straits that a large share of world trade has to go through. When one of them is blocked, whether by drought, conflict or congestion, ships take longer routes, costs go up, and delivery times stretch. For an industry that runs on tight, just-in-time delivery, one chokepoint can send shocks a long way down the line.
The deeper problem is how connected everything is. Because the network is so concentrated, a decision made in one country does not stay in that country. Cobalt is the clearest recent example. The Democratic Republic of the Congo supplies close to two thirds of the world’s cobalt. When it announced a temporary export ban in early 2025, later replaced by export quotas, prices roughly doubled within months. Concentration and long distance together make the chain fragile.
This is also why so much effort today goes not into finding more ore, but into rebuilding the middle of the chain: new refining capacity, regional processing, recycling, and alternative routes that reduce reliance on any one chokepoint.
The decisions this chain forces
If a chokepoint closes, do you reroute and accept the longer transit, or carry more buffer stock upstream so the delay never reaches the plant? If a refining source becomes unavailable, do you qualify a second supplier now at a higher price, or wait and hope? If demand for one chemistry jumps, do you follow it and deepen your reliance on one region?
None of these have an obvious answer, because each one trades cost against risk against service, and the numbers are hard to hold in your head. Extra transit days, extra inventory, extra cost per unit, all moving at once across a network that spans several continents. And what counts as a good trade-off is not universal. It depends on the priorities and constraints of your own company and supply chain.
That is what makes this supply chain worth building as a model. When you lay the network out, run it, and then close a chokepoint or cut off a source, you do not just see the disruption. You have to choose a response, and then watch what that choice actually costs across the rest of the supply chain.
It is also why we are turning it into a supply chain case study, built on SCM Globe, where students take charge of an EV battery mineral network and steer it through exactly these decisions.
We are building it alongside a second one on the AI chip supply chain, which runs into its own version of the same problem. Both should be ready soon.