jo melville
banner
jmelville.science
jo melville
@jmelville.science
climate tech (electrochemistry, industrial decarb, synthetic fuels, carbon removal, deep biogeochemistry, solar radiation management)
formerly: ARPA-E, Georgetown, MIT (PhD), UC Berkeley (BSc)
ask me anything: https://jmelville.science/ask/
🇺🇸→🇸🇬
but it's also not hard to see this as a sunk-cost fallacy: we spent all these billions developing carbon capture systems that barely work at real-world CCGT effluent stream concentrations, so we need to develop new combustion techs that can retroactively justify carbon capture
January 10, 2026 at 2:00 AM
if I'm reading this generously the interpretation is that the CO2-rich effluent streams from methane SOFCs are more amenable for cost-effective CCUS with existing capture technology (it's much like oxy-combustion in this respect)
January 10, 2026 at 2:00 AM
the job of an american congressperson is to (1) raise money and (2) win elections. in between elections is a prime opportunity for a canny operator to raise a little more money, which will help them get elected so they can raise more money later.

also sometimes between elections they vote on stuff.
December 29, 2025 at 6:29 PM
this is what successful CCUS policymaking and monitoring should look like. a failure would be if we deployed EOR around the world at gigaton scale without understanding the consequences, and only discovered this methanogenesis risk after it was too late to do anything about it!
December 23, 2025 at 9:33 PM
With all due respect, Kevin, I know we don't see eye-to-eye on this but my perspective is shaped by reading the literature and talking to experts. I gave a talk on this topic 3 weeks ago. my Zotero sublibrary on this topic is >100 papers. happy to share if any of this would convince you
December 23, 2025 at 9:11 PM
I'm convinced by my arguments because I have been studying this field for the past three years and spoken with dozens of biogeochemists about this issue. the existence of methanogenic microbes is not itself novel but the unintentional stimulation of growth by CO2 CCUS injections absolutely is.
December 23, 2025 at 9:11 PM
relative to a Class VI well, a class II well is both more likely to be methanogenically active (obviously you only do EOR in petroleum geologies with associated microbiota) and more likely to leak any methane that does form (due to the profusion of production in addition to injection wells)
December 23, 2025 at 8:52 PM
class VI wells are for dedicated sequestration. EOR wells are Class II and have even laxer requirements for pre injection characterization and ongoing monitoring post injection. what monitoring there is, is for CO2 leakage and not methane.
December 23, 2025 at 8:46 PM
the EPA Class VI well spec was written in 2010. it doesn't account for downhole microbiota (no CCUS regs anywhere in the world do) because stimulated methanogenesis has only gained salience recently (breakthrough paper I cited is from 2021!)
December 23, 2025 at 8:43 PM
wow, it is almost like there is probably way more methane leakage than we thought, and as our methane detection capabilities advance we consistently find more methane leakage from gas infrastructure

@kevinjkircher.com

www.sciencedirect.com/science/arti...
Accounting for methane from natural gas infrastructure in United States greenhouse gas emission estimates
The Environmental Protection Agency (EPA) estimates that United States net greenhouse gas emissions have declined over the last two decades and are no…
www.sciencedirect.com
December 23, 2025 at 8:40 PM
some amount of it presumably adds to associated gas content during EOR, but ongoing methanogenesis continues for decades~centuries after the well is eventually unprofitable and shut down. oftentimes these old wells leak CO2 — if CH4 forms it probably leaks too

www.sciencedirect.com/science/arti...
Fate of injected CO2 in the Wilcox Group, Louisiana, Gulf Coast Basin: Chemical and isotopic tracers of microbial–brine–rock–CO2 interactions
The “2800’ sandstone” of the Olla oil field is an oil and gas-producing reservoir in a coal-bearing interval of the Paleocene–Eocene Wilcox Group in n…
www.sciencedirect.com
December 23, 2025 at 8:19 PM
it's not a matter of opinion. structural/stratigraphic trapping like you describe is considered to be the least secure form of geologic carbon storage. long-term sequestration requires residual trapping, dissolution, and mineralization — all of which methane is much worse at than CO2:
December 23, 2025 at 8:16 PM
Reposted by jo melville
Larger point: 45Q is arguably the most perverse energy subsidy from a climate pespective. Govt paying corporations to capture and store CO2, instead of actually reducing emissions, means incentivizing more extraction and combustion of fossil fuels because more emissions means more public $.
December 23, 2025 at 5:12 PM
the problem with academia was that it was too difficult to skim a paper while making fundamental errors in comprehension and attribution
December 21, 2025 at 9:46 PM
even if that means writing blatant loopholes and carveouts to legislation that are designed to explicitly disadvantage renewables to the benefit of fossil energy.

you see this e.g. in recent directives from DOE or DOI that apply arbitrary thresholds for capacity factor or areal power density
December 20, 2025 at 10:16 PM
permitting reform could help clean energy alongside fossil fuels. but it's not enough to beat fossil fuels in a "free market" — the market has never been free, fossil fuels always implicitly subsidized by a lack of emissions tax, and fossil oligarchs always have incentive to stack the deck further
December 20, 2025 at 10:01 PM
i think this is half true. as we're seeing especially for solar, it's possible for renewables to compete with fossil fuels on a pure cost basis, without subsidies.

but this still leaves carbon emissions as an unpriced externality. solar is fighting with one arm tied behind its back — and winning!
December 20, 2025 at 10:01 PM
184/ the more I learn about geochemistry the less I understand
December 18, 2025 at 4:04 AM