If you’ve compared two published “battery carbon footprint” numbers and found them a factor of ten apart, the numbers are probably both right. They’re just answering different questions. This site’s own Carbon Footprint vs Cell Chemistry piece cites a manufacturing range of 10 to 394 kg CO2-eq/kWh across chemistries and regions, and that spread is itself a clue: a figure at the low end almost certainly stops at the factory gate, while a figure at the high end has probably folded in raw material extraction, or is normalised against a different functional unit entirely. Before any two of these numbers are comparable, three things have to match: what’s counted, where the assessment boundary sits, and what unit it’s measured against. For how manufacturing emissions trade off against cycle life once a time axis is added, see Carbon Footprint is a 2-axis Problem.
What GWP Actually Aggregates
Global Warming Potential (GWP) isn’t a direct emissions count. It’s a conversion factor. A cell manufacturing process emits more than CO2: methane from upstream gas supply, N2O from certain thermal processing steps, and refrigerant leakage from plant cooling systems all contribute, each with a different warming effect per kilogram. GWP factors convert each gas into a common “CO2-equivalent” (CO2e) basis using its relative warming effect over a fixed time horizon, which is how a single number ends up representing a mixed basket of emissions. This is the aggregation ISO 14040 and ISO 14044 assume happens during the impact assessment phase, without spelling out how. Those pages cover the LCA framework itself; this one covers what the framework is actually being asked to add up.
Scope 1, 2, and 3 at the Gigafactory
The GHG Protocol splits emissions into three categories, and mapping them onto a cell plant clarifies where a manufacturing carbon footprint actually comes from:
- Scope 1: direct emissions from sources the manufacturer owns or controls on-site, such as combustion for electrode drying, thermal processing during formation, and on-site generators.
- Scope 2: indirect emissions from purchased electricity. For a battery plant this is usually the single biggest lever a manufacturer controls, because cell manufacturing (electrode coating, drying, dry-room conditioning, formation and aging) is electricity-intensive enough that the carbon intensity of the local grid dominates the result before any process improvement does. It’s a large part of why the same NMC chemistry runs roughly 57 kg CO2e/kWh in Europe against 90 to 96 kg CO2e/kWh in China on this site’s own chemistry comparison; a good share of that gap is Scope 2, not chemistry.
- Scope 3: everything else in the value chain, from upstream mining and refining of lithium, nickel, cobalt, and graphite before they reach the plant, to downstream transport, use, and end-of-life after the cell leaves it.
A number described as a “carbon footprint” without saying which of these scopes it includes isn’t yet comparable to anything else.
Four Boundaries, Same Battery
ISO 14040 defines the LCA framework; it doesn’t fix where the system boundary sits. That’s decided project by project. For a battery, the same functional unit (commonly 1 kWh of energy delivered over the battery’s service life, the convention both this site’s ISO 14040 page and the EU’s methodology use) can be assessed against several different boundaries:
- Cradle-to-gate: raw material extraction through to the finished cell or pack leaving the factory. This is what most manufacturing-only kg CO2e/kWh figures represent, including the chemistry comparison on this site.
- Gate-to-gate: just the manufacturing plant itself, excluding upstream material sourcing. Useful for comparing two factories running the same chemistry; not useful for comparing two chemistries.
- Cradle-to-grave: cradle-to-gate plus the use phase and end-of-life. This is the boundary a full “lifecycle” environmental claim should mean, and it’s the one that can let a lower-embodied-carbon, faster-degrading chemistry come out worse overall than a higher-embodied-carbon, longer-lived one.
- Cradle-to-cradle: a cradle-to-grave assessment where end-of-life material recovery is credited back into a new product’s material inputs. This is the boundary that puts a number on recycling’s carbon benefit rather than leaving it as a qualitative claim.
None of these is “more correct” than the others. They answer different questions, and the failure mode is comparing answers to different questions as if they were the same one.
What the EU Battery Regulation Actually Mandates
Article 7 of the Batteries Regulation, Regulation (EU) 2023/1542, requires a carbon footprint declaration for EV, LMT, and rechargeable industrial batteries above 2 kWh, building in stages toward performance classes and then maximum thresholds. The methodology it points to is the EU’s Product Environmental Footprint (PEF) method, developed into battery-specific Product Environmental Footprint Category Rules (PEFCRs), so the boundary and functional unit aren’t left to the manufacturer to choose. The mandated boundary is cradle-to-grave excluding the use phase (raw material acquisition, production, distribution, and end-of-life), against the same 1 kWh-delivered functional unit described above. Primary, company-specific data is required for the production stage; the other stages can draw on approved secondary datasets. Compliance dates, and how the declared figure feeds the Digital Battery Passport, are covered on the Battery Passport page rather than repeated here. They sit under separate delegated acts that are still being finalised.
Reading a Number Before You Trust It
Given a published kg CO2e/kWh figure, four questions decide whether it’s actually comparable to another one. What’s the functional unit: per kWh of rated capacity, or per kWh actually delivered over life? What’s the boundary: gate-to-gate, cradle-to-gate, or cradle-to-grave? Is Scope 3 upstream (mining and refining) included, or does the number quietly stop at the factory gate? And is the grid mix behind Scope 2 disclosed, given how much of the regional spread in published figures is really an electricity-mix difference rather than a chemistry difference?
Needs review: raw material share of Scope 3 is left unquantified here as I couldn’t verify a specific percentage from a source worth citing (paywalled papers blocked automated access). Also worth a sanity check on the EU Battery Regulation compliance dates against the current delegated acts before this goes live, since sources checked during research disagreed slightly and dates are deliberately deferred to the Battery Passport page rather than restated here.
References
- GHG Protocol, Corporate Value Chain (Scope 3) Accounting and Reporting Standard: FAQ. ghgprotocol.org
- Regulation (EU) 2023/1542 of the European Parliament and of the Council (the Batteries Regulation), Article 7 and recitals 27 to 28. EUR-Lex.
See also: Energy used to Manufacture a Cell