Black mass and lithium-ion stock control: a practical primer for recyclers

The stock line that quietly eats your margin

Most yards that start taking batteries do it the same way. Someone turns up with a mixed load, the scale house has no code for it, and the ticket gets written against whatever is closest: "Battery Scrap". Six months later that one line holds lead-acid units, loose lithium-ion cells, e-rickshaw packs and a few drums of powder from the separation line, and nobody can say what any of it cost or what any of it is worth.

On a mixed steel line you survive that. On a battery line you do not, because the value per kilogram is an order of magnitude higher and the hazard profile is different at every stage.

What black mass actually is

Black mass is what is left after end-of-life lithium-ion batteries are discharged, dismantled or shredded, and mechanically separated. Casings, current collectors, wiring and plastics come out. What remains is a fine dark powder made up of the cathode and anode active material.

That powder is where the money is. Depending on the chemistry of the feed it carries lithium, nickel, cobalt and manganese along with graphite. A hydrometallurgical refiner buys it to pull those metals back out and return them to cell manufacturing.

Two things follow, and both matter operationally:

• Black mass is not a grade of battery. It is a manufactured output. It has a production step behind it, a yield, and an input batch it came from. Treating it as just another purchased material breaks the link back to the feed.
• Its value depends on the chemistry that went in. Black mass from nickel and cobalt rich cells is worth substantially more per tonne than black mass from lithium iron phosphate cells, because there is less payable metal in the LFP fraction. Your buyer pays on their own assay.

Put NMC packs and LFP packs into the same hopper and you have permanently averaged a high value stream into a low value one. No accounting recovers that afterwards.

Three materials, three stock lines

Lead-acid, lithium-ion and black mass have almost nothing in common except the word "battery". They belong in separate categories, with separate costing, from day one.

The failure mode from lumping them is not just reporting noise. If inventory carries one blended average cost across all three, every issue out of that line is costed wrong, every margin report is wrong, and a shipment of the good stuff looks identical on paper to a shipment of the cheap stuff. Weighted-average costing is only honest inside a genuinely homogeneous line.

State of charge is a stock-control problem too

The most useful single attribute you can record about incoming lithium-ion is whether it has been discharged.

A cell arriving at the gate is a stored energy device. It does not become inert because it is now scrap. A cell holding meaningful charge that is crushed, punctured, shorted or heated can go into thermal runaway, and runaway is self-sustaining: it generates its own heat, so it does not behave like an ordinary fire and it propagates to the cells around it.

None of this needs a laboratory. At an operational level:

• Discharge before processing, and record the discharge as an operation rather than assuming it happened.
• Isolate terminals. Tape or cap anything not yet discharged. Most avoidable incidents start as a short between two pieces of metal in a bin.
• Quarantine damaged, swollen or water-affected units separately. Different risk class, different bay.
• Store in non-combustible containers, with spacing and real distance from buildings, fuel and other stock.
• Watch heat. A pack that is warm, hissing or giving off a sweet smell has already started. The response is isolation and distance, not investigation.
• Do not charge anything on site, and do not let staff test packs already booked as scrap.

The stock-control link is direct. If condition and discharge state are attributes of the lot, a supervisor can pull a list of live, undischarged and damaged lots on site at any moment. If they live in someone's head, that list does not exist.

Why small unexplained differences cost so much here

Every yard runs a tolerance on stock differences. A percent or two of unexplained loss on a ferrous line is annoying but survivable. Apply the same tolerance to a battery line and the arithmetic changes shape.

Consider a yard holding 10 tonnes of black mass. Suppose, purely for illustration, that mixed ferrous realises 30,000 per tonne and this black mass realises 3,00,000 per tonne. Use your own numbers; the ratio is the point.

Same tolerance, same weight, ten times the money. And unlike a ferrous pile, black mass differences are rarely one big event. They accumulate quietly: moisture pickup and drying, dust loss on transfer, unticketed sweepings, drums moved between bays with no transfer entry, a residual nobody reconciles after a shipment.

So a high value line needs a tighter tolerance and a shorter count cycle than the rest of the yard. Counting black mass drum by drum weekly is cheap next to what a quarterly surprise costs.

Structuring categories and grades so recovery is measurable

The reason to be fussy about the category tree is not tidiness. It is that recovery per chemistry is only calculable if input and output are classified the same way.

Three axes, not one

Use a category for what it is, a grade for its condition, and attributes for the rest.

Keep a mixed or unsegregated category and use it honestly. The temptation is to guess a chemistry so the report looks clean. A visible mixed line you are actively trying to shrink is worth more than a tidy report built on guesses.

Then the yield falls out on its own

Once feed and output share a classification, yield per chemistry is a subtraction rather than an investigation. Illustratively:

Output always sits well below feed, because casings, current collectors, plastics, electrolyte and moisture leave by other routes. That is normal. What you are watching is the same category drifting over time, and the gap between a segregated batch and a mixed one. That third row is the argument for segregation at the gate, stated in kilograms instead of opinions.

Two habits make it work: book the production step on the day, feed consumed against output produced under a batch reference, because if output is booked and feed is not your yield history is fiction. And do not reclassify after the fact. If a batch was fed as mixed, its output is mixed derived, even if you later believe it was mostly NMC.

The compliance side of the same records

In India, battery waste sits under a dedicated set of rules with extended producer responsibility at the centre of them. Recyclers are expected to register with the relevant pollution control authority, to store and handle the material in line with hazardous waste requirements, to report quantities received and processed, and to meet recovery obligations that tighten over time. Movement carries its own documentation expectations. Check the current text and your own registration conditions rather than a figure somebody quoted you, because thresholds and phase-in dates have moved more than once.

The operational point holds regardless: the credits and certificates that flow from recovery are calculated on quantities recovered, evidenced by records. A yard whose battery stock is one lumped line cannot evidence recovery by chemistry, so it cannot substantiate what it claims. The discipline that protects your margin is the discipline that survives an audit.

Where a system helps

On paper or in a spreadsheet, two things break first: the link from feed batch to output batch, and the count cycle on high value drums.

In Scraplytics the pieces that carry the load are ordinary ones used carefully. Scale tickets with weighbridge capture and photos at intake, so condition is evidenced rather than remembered. Inventory by category and grade with weighted-average costing held per line, so expensive material never averages into cheap. Multi-location stock, so a quarantine bay is a real location with its own balance. Documents against the lot for manifests and registrations. Reports showing yield and stock difference per category rather than in aggregate.

None of that is battery specific magic. It is the discipline you already apply to copper, applied to a stream where being sloppy costs ten times more.

The short version

Black mass is a produced output with a yield, not a purchased grade. Lead-acid, lithium-ion and black mass are three businesses sharing a word, and they need three stock lines. Record state of charge; it is a safety attribute and an inventory attribute at once. Tighten the tolerance in proportion to value per kilogram. And classify feed and output the same way, because the moment you do, recovery per chemistry stops being an argument and becomes a number.