A toxic solvent, and the plant built to catch it
Coating a battery electrode means spreading a wet slurry, a thick mixture of active material and binder, onto a metal foil, then baking the liquid out in an oven. That liquid is usually N methyl pyrrolidone, called NMP, a solvent the United States Environmental Protection Agency states can carry a risk of fetal death after a single 8 hour exposure at work, along with damage to the liver, kidney, immune system and nerves. NMP cannot simply be vented into the air. A battery plant that uses it needs an entire separate system to catch the fumes and turn them back into liquid solvent, and the agency proposed rule keeps lithium ion battery manufacturing running under new workplace controls, covering roughly 44% of the solvent total production volume.
What switching to water removes
An Oak Ridge National Laboratory and Argonne National Laboratory engineering study prices that recovery system item by item. The single largest piece is a condenser, at 1,380,000 dollars, followed by a distillation column at 1,000,000 dollars and a chiller at 950,000 dollars. The full system for 1 production line comes to 5,285,500 dollars. The same study finds that removing it entirely, by coating with water instead of NMP, saves 3 to 6 million dollars in equipment for a factory making 100,000 battery packs a year, each holding 10 kilowatt hours, the unit that measures how much energy a battery can store, the size used in a plug in hybrid electric car. Water carries no explosion risk, so a plant also saves 75,000 dollars by no longer needing monitors that watch for explosive vapor in the air.
These are the 5 largest of 8 line items in the source own cost table. 3 smaller items, a zeolite wheel, a scrubber and a main blower, are not shown, so the 5 bars do not add up to the full 5,285,500 dollar system total given above.
Show the numbers
| Condenser | 1,380,000 |
| Distillation column | 1,000,000 |
| Chiller | 950,000 |
| NMP storage tank | 560,000 |
| Air to air heat exchanger | 460,000 |
Where the energy saving actually comes from
Water is not cheaper to dry than NMP. Boiling a kilogram of water off a coated electrode takes 2,260 kilojoules of heat, well more than the 510 kilojoules NMP needs for the same job, so a water based dryer alone uses more heat, 823 kilowatts against 650 for NMP, in the same laboratory model. NMP based drying then needs an extra 458 kilowatts, on top of its own 650, to condense the evaporated solvent back into liquid, a step water skips entirely because it can simply be released into the air. Counting both steps together, the water based line uses 25.7% less total energy than the NMP based line, a national laboratory own detailed engineering model finds.
This is the laboratory own model for anode drying, the only stage it models in this much detail. Water needs more heat to dry alone, but solvent based drying needs an extra step water skips entirely, and counting both steps the solvent based line uses more total energy overall, 25.7% more than the water based line in the same model.
Show the numbers
| Water based drying | 823 |
| NMP based drying | 650 |
| NMP solvent condensation | 458 |
Solved for the anode, not yet for the cathode
A lithium ion battery has 2 coated electrodes, the anode and the cathode, the pair lithium moves between as a battery charges and discharges. Water based coating for the anode, made of graphite, is established industry practice, a Karlsruhe Institute of Technology study states. The cathode is different. Water dissolves a measurable amount of metal straight out of nickel rich cathode material, and the same national laboratory measured 9.4 milligrams of lithium leaching into 1 liter of water after a lithium iron phosphate cathode sat in it for just 1 day, rising the longer it soaks. Every one of the 5 sources checked for this article describes water based cathode coating running at laboratory or pilot scale only, coin cells and small test lines, not at the volume a real factory runs. The same national laboratory reports it now runs both electrodes with water as standard practice in its own research line, proof the chemistry works, not proof that a production plant has removed its solvent recovery system yet.