Take on a podcast episode from Interchange Recharged, originally published Tue, 24 Fe. Listen: <>

TL;DR

  • Hyperscaler power demand has flipped carbon capture from a regulation-driven to a buyer-led market in ~18 months. Plausible, and consistent with the Meta/Hyperion-type deals.
  • ION claims natural gas + capture + storage lands at ~31 kgCO2/MWh — competitive with solar+storage on a lifecycle basis. Aggressive; rests on a 0.75% methane leakage assumption that is below US field averages.
  • Cost adder: ~$20–25/MWh for retrofits, $16–18/MWh for new builds, inclusive of transport and storage. Useful number if it holds.
  • ION’s pitch on its amine solvent: doesn’t degrade in oxygen, so 99% capture is achievable without the usual exponential energy penalty. Worth probing.
  • Execution risk — getting power purchase agreements to final investment decision — is now the binding constraint, not policy or tech. 45Q stayed at $85/t under the current administration.

Bridget van Dorsten hosts Tim Vail, CEO of ION Clean Energy, for a fairly technical hour on post-combustion capture at natural gas plants serving AI data centers. The framing is unapologetically pro-gas-plus-capture as a “clean firm” option alongside nuclear and geothermal — if you want a skeptical take on that premise, this isn’t it, but if you want the seller’s most coherent version of the argument with actual numbers, it’s here.

What’s worth knowing. The load-bearing claim is the lifecycle carbon intensity comparison: ION says gas+CCS comes in around 31 kgCO2/MWh versus ~32 for wind/solar+storage, when you count battery manufacture, blade disposal, etc. on the renewables side and full wellhead-to-busbar on the gas side. The methodology assumes 0.75% upstream methane leakage — Vail and van Dorsten both acknowledge real-world Permian basin numbers run 0.5% to 4%, with outliers to 9%. That’s the soft spot. The whole CI parity argument collapses if you plug in a basin-average leak rate instead of a best-in-class verified one. Vail’s answer is that Williams’ “next gen” certified gas (verified via Context Labs, blockchain-based MRV) makes the low-leakage assumption contractually real for ION’s projects. Fair, but it means the parity claim applies to a subset of available gas, not the marginal molecule.

The cost stack is the other useful artifact. Vail quotes $20–25/MWh adder for retrofits and $16–18/MWh for greenfield integrated builds, including transport and sequestration — not just capture. He explicitly notes hyperscalers don’t want $/ton numbers, they want $/MWh delivered. That maps a gas+CCS PPA to roughly $60–80/MWh all-in, which he benchmarks against Microsoft’s reported $80–100/MWh Constellation nuclear restart deal. If those numbers survive contact with an actual signed contract, gas+CCS is genuinely competitive with the other 24/7 clean firm options. The 99% capture rate claim (vs. the industry’s begrudging move from 90% to 95%) is the technology differentiation pitch — Vail attributes it to a non-degrading amine solvent that avoids the oxygen-driven breakdown plaguing legacy MHI/Shell Cansolv chemistries.

Context. This sits alongside the broader hyperscaler clean-firm scramble — Google’s recent gas+CCS deal in Illinois (with a competing technology), the UK’s Teesside cluster reaching FID, and growing buyer interest in Frontier Climate-style durable removals as a complementary, not substitute, pathway. Worth noting: this episode is about avoided emissions on power generation, not durable CDR. The bioenergy with carbon capture and storage (BECCS) angle Vail mentions briefly is where this technology stack overlaps with the removal credit market.

Worth the hour if you’re sizing the gas+CCS pathway as a competitor for hyperscaler clean-firm budgets, or mode