Global Field Dossier — Atmospheric Chemistry
Tracing the world's most powerful invisible gases — from a cleanroom exhaust vent to a spectrometer above Boulder, Colorado.
Somewhere above Boulder, Colorado, an air sample is drawn into a gas chromatograph every few dozen minutes, run through a chain of detectors, and reduced to a number in parts per trillion. That number is one of thousands collected daily by a loose confederation of laboratories that never agreed on a name, but effectively function as humanity's nervous system for gases nobody can see.
Fluorinated gases — HFCs, PFCs, SF₆, NF₃, and their relatives — don't occur meaningfully in nature. Every molecule in the atmosphere was manufactured, mostly as a byproduct or working fluid of industries that need extreme chemical stability: refrigeration, aluminium smelting, electrical insulation, and — increasingly — the plasma chambers where semiconductors are etched and cleaned. That same stability, which makes them useful, is what makes them dangerous. Several persist in the atmosphere for centuries or millennia, trapping heat thousands of times more effectively than CO₂ the whole time they're up there.
A molecule of SF₆ released today will still be warming the atmosphere long after the machine that leaked it has been recycled twice over. Field note — atmospheric chemistry briefing
No single body owns F-gas research. It's held together by measurement labs, industry-adjacent engineering institutes, and policy groups that rarely share a building, coordinated only by the data they all publish.
Daejeon, South Korea · 36.35°N 127.38°E
Semiconductor-manufacturing sustainability
Korea builds a fifth of the world's memory chips, and KAIST's environmental engineering groups work through the industry's least glamorous problem: the perfluorinated and hydrofluorocarbon gases used to etch and clean plasma chambers. Their findings feed directly into Samsung and SK Hynix supply-chain emissions targets.
Tokyo, Japan · 35.68°N 139.65°E
Process-emissions research, linked to JEITA
Japan supplies much of the world's semiconductor-grade process gas and photoresist chemistry, which puts these researchers close to the source rather than downstream of it. Their work runs through JEITA — the national electronics industry association — into Japan's F-gas working group, one of the few bodies with binding voluntary reduction targets.
Boulder, Colorado, USA · 40.01°N 105.27°W
Long-running atmospheric F-gas measurement
This is the cross-check. NOAA's Global Monitoring Laboratory and CU Boulder's atmospheric chemists run continuous sampling across a global station network, comparing what's reported against what's actually accumulating overhead — the gap between "top-down" measurement and "bottom-up" inventory has, more than once, been how the world found out an emissions estimate was wrong.
Munich, Germany · 48.14°N 11.58°E
Process-gas abatement engineering
Germany's applied-research powerhouse builds the hardware that closes the gap once it's found: plasma abatement units on fab exhaust lines, on-site fluorine generation that replaces NF₃ delivery entirely, and gas-recycling loops that catch what used to vent straight to atmosphere.
London, UK · 51.50°N 0.17°W
Climate policy and reporting standards
Measurement and engineering only matter if they change what regulators require. Grantham's climate policy researchers work at that translation layer — turning atmospheric data and abatement performance into the metrics and inventory methods that sit underneath national climate pledges and corporate carbon accounts.
Networks like AGAGE (Advanced Global Atmospheric Gases Experiment) and NOAA's HATS program sample air at stations from Mauna Loa to Cape Grim, converting global air into a running, independent ledger of what's actually up there — the only reliable check on national self-reported inventories.
Where the emissions originate, the fix is mechanical: capture, destroy, or replace. Fabs are moving toward point-of-use abatement, gas recycling, and lower-GWP alternatives to SF₆ and NF₃, developed and stress-tested through applied-research bodies like Fraunhofer.
The 2016 Kigali Amendment to the Montreal Protocol — the same treaty that closed the ozone hole — added a global phase-down schedule for HFCs, aiming to avoid up to 0.4°C of warming by 2100. Turning that treaty into national law and corporate accounting is what policy researchers spend most of their time on.
None of these three move alone. A molecule caught by a spectrometer in Boulder becomes a discrepancy flagged by researchers, becomes a design brief for an abatement engineer in Munich, becomes a line item in a reporting standard drafted in London. The solution to F-gas, so far, looks less like a single invention and more like this: a slow, worldwide effort to make an invisible problem impossible to ignore.
| Gas | Primary source | GWP, 100-yr (approx.) | Atmospheric lifetime |
|---|---|---|---|
| HFC-23 (CHF₃) | Byproduct of HCFC-22 production | ~12,400× | ~228 years |
| SF₆ | Electrical insulation, chip fabrication | ~23,500× | ~3,200 years |
| NF₃ | Semiconductor & flat-panel etching | ~17,200× | ~500 years |
| CF₄ (PFC-14) | Aluminium smelting, chip fabrication | ~7,380× | ~50,000 years |
Figures are approximate, drawn from successive IPCC assessment reports; exact values shift slightly between AR5 and AR6 as measurement improves.