IN SUMMARY
Most life cycle assessments of batteries assume a single, static impact value for each component material. That hides the reality that the same battery, built to the same chemistry, can carry a very different carbon footprint depending on where and how its raw materials are produced. In this foundational study, Minviro modelled an NMC-811 battery, the most common chemistry in Western EVs, across low, baseline and high-impact supply chain scenarios to show just how much that variability matters at the level of a finished battery pack.
The same battery, nearly double the footprint. An NMC-811 pack ranges from 70 kg CO₂e per kWh with low-impact materials to 138 kg CO₂e with high-impact routes, against an 82 kg CO₂e baseline.
Raw materials, not the factory, drive the difference. As manufacturing electricity decarbonises over time, the impact of producing raw materials stays relatively steady, becoming the dominant and harder-to-cut share of a battery’s footprint.
Nickel and graphite are the swing factors. Nickel sulfate dominates the cathode, while graphite’s contribution rises around ninefold between low and high scenarios to make up roughly a quarter of impacts, confirming it as the hidden impactor.
Why supply chain choices matter more than the chemistry
The energy transition needs vast quantities of raw materials for batteries, motors and magnets, and the supply chains for those materials create very different embodied impacts depending on how they are produced. Most battery LCAs miss this by assuming fixed values for each material, and the underlying data sometimes underestimates certain materials outright. There is also a timing dimension: as electricity grids decarbonise, the use-phase and manufacturing emissions of batteries will fall, but the energy needed to produce the raw materials stays relatively steady, partly because future feedstocks will be lower-grade and less pure. That makes raw material production the part of a battery’s footprint that matters most and is hardest to cut, which is exactly where this study focuses.
How the study was built
Minviro modelled an NMC-811 battery using a bill of materials from the GREET model, with a functional unit of one kilowatt-hour of storage. Three scenarios were built, low, baseline and high impact, varying five inputs that Minviro’s database shows have the widest real-world range: nickel sulfate, cobalt sulfate, manganese sulfate and lithium hydroxide for the cathode, plus graphite for the anode. All other bill-of-materials data came from Ecoinvent 3.7.1, and cell assembly was held at a static 25 kg CO₂e per kWh. By changing only the five variable inputs, the study isolates how much raw material sourcing alone moves the footprint of an otherwise identical battery.