Every ton
needs a
name on it.

The problem is not that it is hard. The problem is that it cannot disappear — and the volume keeps rising, warming faster per ton than almost anything else we make.

Six hand-offs.
Six owners.
No gaps.

Controlled chemicals work because custody is never ambiguous: someone signs at every hand-off. F-gas needs the same shape — not a tracker on every molecule, but a named owner at every station, with one control and one measurable duty each.

From the point of origin to the person holding the phone.

Six populations of actors already touch this gas. None of them can close the loop alone; each of them can take one move that makes the next station's job possible. Company names below are illustrative of each layer, not an assessment of any individual firm.

What the chain is missing today.

• The atmosphere disagrees

  Independent atmospheric measurements of PFCs have repeatedly come in above the totals reported bottom-up. Somewhere between purchase and disclosure, tonnes lose their owner.

• Default efficiencies

  Abatement is often credited at assumed destruction rates rather than tested ones. Real performance varies with recipe, flow, maintenance and uptime — sometimes a lot.

• Custody ends at the dock

  After a cylinder is sold, no formal record follows the mass. Controlled chemicals never work this way; F-gas mostly does.

• Regulatory blind spot

  HFCs have a phase-down. PFCs, SF₆ and NF₃ are reported in inventories but not governed by a global market-shaping instrument.

• Intensity beats absolutes

  Normalised targets can improve while total tonnes climb, because capacity is climbing faster. Both numbers must be published side by side.

• The build-out

  AI-driven capacity, advanced packaging and new fab construction are adding chambers faster than retrofit programmes are adding abatement.

• The long tail

  Mature nodes, display lines, smaller regional fabs and subcontractors are the least instrumented and the least disclosed part of the chain.

2026 → 2035.
One rung a year.

If this really takes a decade, then let it be a decade with rungs. Each year builds the evidence the next year needs.

• 2026Name the chain

  An open working definition of the six stations, their owners and their data fields — agreed across gas makers, tool makers, fabs and brands. Publish the vocabulary before arguing about the numbers.

• 2027Passport the molecule at sale

  Pilot digital gas passports on real shipments. Producers begin publishing sales by species and region. First voluntary purchase-versus-use reconciliations at a handful of fabs.

• 2028Measure what abatement actually does

  In-situ DRE testing protocol agreed and run under production recipes. Default factors start being replaced by measured ones in public reporting.

• 2029Instrument the biggest stacks

  Continuous species-level monitoring on the highest-emitting subfabs at leading sites; raw data retained for audit. Display lines join the same protocol.

• 2030First reconciled public ledger

  Gas sold, gas used, gas abated, gas emitted — published gas-by-gas and site-by-site for a meaningful share of global capacity, with third-party assurance.

• 2031Check it against the sky

  Top-down atmospheric estimates around major fab clusters published alongside bottom-up totals. Divergence investigated openly, jointly, and without penalty for honesty.

• 2032Move it into contracts

  F-gas becomes a named, verified clause in supplier agreements and a criterion in sourcing decisions — for foundry, for display, for packaging.

• 2033Close the long tail

  Mature nodes, smaller fabs, subcontractors and display suppliers brought into the same reporting floor. Retrofit financing available where capex is the real blocker.

• 2034Substitute where possible

  Lower-GWP chemistries, remote plasma optimisation and recipe redesign scaled where qualification allows — with the emissions benefit verified, not assumed.

• 2035Absolute tonnes falling

  The test is simple and unforgiving: total F-gas emitted by the electronics chain is lower than 2026 in absolute terms, and every ton in the ledger has a name beside it.

The people already watching this gas.

Not endorsements, and no quotes put in anyone's mouth — these are the researchers and institutions whose published work is where a serious study of this question should begin.

Where to read next.

No links here on purpose — links rot, titles don't. Each entry below is named precisely enough to find in one search, from the primary source itself.