The $37 Million Lesson: Methane-Eating Microbes Worked, the Fertilizer Business Didn't

Windfall Bio raised $37 million on a simple promise: feed methane to bacteria, harvest the fertilizer, and let farms turn a greenhouse gas into a margin. In April 2026 the company stopped operating. Its bioreactors in Texas are now listed for auction.

Methane drives a large share of the warming locked in this decade, and agriculture is one of its biggest sources. Methanotrophs, the microbes that eat the gas and excrete nitrogen, sit at the center of a bet that the cheapest way to destroy methane is to sell what it becomes.

The chemistry was never the bottleneck

While the company was scaling steel, the science kept getting cheaper. Three 2026 results show the direction of travel.

In Trends in Biotechnology, researchers described engineering a methane-producing archaeon, Methanosarcina acetivorans, to run its metabolism in reverse. The reversal let the organism consume methane instead of emitting it, converting the gas into value-added products. The work sidesteps a long-standing constraint: most methane-eating archaea are hard to culture and hard to engineer, so the field has leaned on a narrow set of aerobic bacteria.

A second study in Frontiers in Microbiology enriched a heat-loving methanotroph consortium from ordinary compost. Inside a biofilter, the consortium oxidized methane at roughly 140 milligrams per kilogram of compost per hour without added nutrients, a result that points toward low-cost passive treatment of landfill and composting emissions.

A third project, SynYeast at Aston University, is funded through the UK's Horizon Europe guarantee and aims to build methane-based biomanufacturing into yeast. If it holds, methane conversion stops depending on exotic microbes and starts running on the same fermentation equipment used for beer and insulin.

A 2026 review in Energy & Environment Nexus catalogued the wider menu: methanotrophic communities converting greenhouse gas into green plastics, animal feed, and fuels. The framing matters. Methane stops being a liability to neutralize and becomes a feedstock with several possible exits.

Other work points at passive systems that need no bioreactor at all. A framework published in Trends in Microbiology proposed treating aerated caves, and possibly decommissioned mines, as natural methane biofilters, where surface bacteria oxidize the gas as air circulates through the rock. It is speculative, but it sketches a version of methane removal with almost no steel in it.

A third thread works on the supply side of the problem. A 2026 preprint reported that adding gypsum to anaerobic digesters redirects electron donors away from methane production and toward sulfate reduction, cutting emissions before they form. Interventions like that compete with Windfall's model on cost rather than on product.

What $37 million actually bought

The company's arc is unusually well documented for a startup that failed, which makes it a case study rather than a eulogy.

Josh Silverman co-founded the company in 2022 in San Mateo, California, with Carla Risso and Louis Stenmark. Silverman had already built Calysta, a gas-fermentation company that turns methane into high-protein animal feed. It took the same feedstock and aimed at fertilizer instead.

The funding arrived quickly. A $9 million seed in 2023 was followed by a $28 million Series A in April 2024, led by Prelude Ventures, with Amazon's Climate Pledge Fund, Breakthrough Energy Ventures, Bessemer Venture Partners, Mayfield, SOSV, and Global Brain taking part. Seventeen institutional backers in total.

The milestones came too. By January 2025 the company had scaled fermentation of its methane-eating microbes to 17,000 liters, enough to produce the microbes by the metric ton. In August 2025 its FOUNDATION fertilizer earned certification from the Organic Materials Review Institute (OMRI), the gatekeeper for organic farming inputs. In October 2025 it broke ground on a demonstration plant in Houston; the roughly 5,000-square-foot facility, a $5 million investment, opened in December with 10 full-time roles. In November, field trials run by Ag Metrics Group showed FOUNDATION lifting Romaine lettuce yields 23–34% over feather meal.

Silverman framed the logic plainly that November. "Many people who were turning methane into electricity have turned off their generators because it doesn't make the money anymore," he told AgFunderNews. "We said, can we do something better?"

Many people who were turning methane into electricity have turned off their generators because it doesn't make the money anymore. We said, can we do something better?— Josh Silverman, co-founder and CEO, Windfall Bio

The answer turned out to be: maybe not yet. By spring, the company was raising a Series B. The round never closed. The company ceased operations in April 2026, and by late May auctioneer Silicon Valley Disposition was selling off LC/MS/MS systems, HPLC units, freeze dryers, and bench-top bioreactors in San Mateo, plus large-scale bioreactors at the Texas site.

No acquirer stepped in for the platform. That is the part that should worry the rest of the sector. The collapse was a cost problem. The fertilizer worked. Making it was expensive, and no patient buyer wanted the machine that made it.

As we wrote in August, the moat in gas fermentation is hardware, not the microbe.

The structural problems are easy to name. Point-source methane, the kind found at manure lagoons and landfills, is dilute, so capturing it costs more per tonne than the gas is worth as a commodity. On-site bioreactors push capital spending onto the customer, and farms and landfills are not venture-funded. Organic fertilizer competes on price with commodity nitrogen, which is cheap when natural gas is cheap. Carbon credits could bridge the gap, but they depend on registries and policy that shift with elections. Against all that, a startup needs a decade of patient capital. The 2025–2026 funding market offered quarters.

The next cohort is betting lighter

The companies still working in methane-to-value are mostly avoiding Windfall's vertical integration. Instead of owning a farm's bioreactors, they sell a strain, an additive, or a service.

C1iN, an Arlington, Virginia company, designs methanotroph strains and fermentation processes and pitches them as inputs for agriculture, ingredients, and materials; it entered an accelerator in April 2026. SynBioBlox, based in Edmonton, is building methane-based synthetic biology for fuels, agricultural additives, and industrial materials. Both are asset-light next to a demo plant.

On the feed side, Volta Greentech raised about €1.7 million in early 2026 to launch Lome, a feed additive that suppresses methane-producing microbes in a cow's rumen. Provectus Algae, an Australian synthetic-biology company, is scaling Surf'N'Turf, an algae-based methane-reducing supplement, after a $10.1 million Series A. These products slot into existing farm operations rather than requiring a new industrial facility on site.

Calysta, Silverman's earlier company, remains the counter-example that worked. It sells high-protein animal feed made by gas fermentation, treating the feedstock as a route to a premium product rather than a climate obligation.

Sources: AgFunderNews; company disclosures, 2026.

Signals to track

The collapse will be read as a verdict on methane-eating microbes. It is a verdict on a business model. The microbe ate the methane. The machine that fed it, and the market that was supposed to pay for the fertilizer, are what failed. Investors watching the category should keep the two apart.