Love Planet Earth

503 reactors.
No smokestacks.

The proposal: a subterranean Gen-IV small modular reactor at every one of America's 503 commercial airports — helium-cooled, TRISO-fueled, buried 100 meters down, each powering its own microgrid. This is what it would take, what it would cost, and whether it could actually happen.

📊 Track the prices →

What clean energy actually costs: price per watt for solar, wind, nuclear, batteries, coal and gas — interactive charts, verified figures, and the coal-country transition playbook.

🌡️ The Climate Dashboard →

What 1.5, 2, 3 and 4°C actually mean, a live CO₂ ticker, the budget countdown, and four futures to 2100.
The vision

Nuclear power, underground and everywhere

The SMR-27 proposal imagines the United States mobilizing 150 automated vertical shaft-sinking machines to all 503 primary commercial airports at once. Each site gets a 100-meter-deep silo lined with barite concrete and stainless steel, housing a helium-cooled high-temperature gas reactor running on TRISO fuel pebbles — fuel the proposal notes physically cannot melt down. Power comes out through supercritical CO2 turbine blocks, and in an emergency the fuel pebbles simply drop into liquid-nitrogen heat sinks by gravity, with zero human intervention required. On the surface, the airports keep operating — retrofitted with monocrystalline solar pavement and modular sky-towers — while the reactors stay entirely hidden underground.

Aerial mock-up of an airport retrofitted with a subterranean SMR microgrid
By the numbers

The proposal, quantified

503
subterranean Gen-IV SMR microgrids — one per commercial airport
100 m
shaft depth, lined with 3 m of barite concrete + 50 mm stainless steel
150
automated shaft-sinking machines deployed simultaneously
TRISO
fuel pebbles — each uranium kernel sealed in ceramic layers
sCO2
supercritical CO2 turbine blocks for power conversion
5 yrs
target buildout — absorbed within the $8.5T infrastructure sprint

Figures are from Report 27 of the 30 Mega-Projects series. TRISO stands for TRIstructural-ISOtropic: each fuel kernel is wrapped in layers of carbon and silicon carbide ceramic that contain fission products even at extreme temperatures.

Honest accounting

Pros & cons

The case for it

  1. The reactor technology is genuinely real. X-energy's Xe-100 — a helium-cooled, TRISO-fueled pebble-bed design much like the proposal's — was selected under the DOE's Advanced Reactor Demonstration Program and is planned for a Dow chemical site in Seadrift, Texas. TerraPower's Natrium (sodium-cooled) is under construction in Kemmerer, Wyoming, and Kairos Power received an NRC construction permit for its Hermes test reactor in Oak Ridge, Tennessee in 2023.
  2. TRISO fuel really is extraordinarily safe. The ceramic-coated particles retain fission products at temperatures above 1,600°C — far beyond any credible accident scenario. Germany's AVR and China's HTR-PM have operated pebble-bed reactors for years; the fuel's safety case is one of the best-tested claims in advanced nuclear.
  3. Nuclear is already America's carbon-free workhorse. The existing US fleet provides roughly half of the country's carbon-free electricity. Small modular reactors promise to extend that with factory-built units, passive safety systems that need no operator action, and sites small enough to fit inside an airport footprint.
  4. Underground siting solves the hardest political problem. Public opposition to nuclear is overwhelmingly about what's visible and what's nearby. A reactor 100 meters down, invisible from the surface, with the airport already an industrial zone, sidesteps the "not in my backyard" fight that has killed aboveground plants for decades.
  5. Microgrids are resilient by design. Airport-scale reactors powering local grids would keep critical infrastructure — runways, hospitals, the New Orleans pump grid (MSY-15) — running through hurricanes and blackouts that knock out centralized power plants.

The case against it

  1. No commercial Gen-IV gas reactor operates in the United States. The Xe-100, Natrium, and Hermes are all still in licensing, construction, or testing. The proposal assumes a first-of-a-kind technology can be stamped out 503 times in five years — when the entire US advanced-reactor pipeline has yet to produce a single commercial kilowatt-hour.
  2. The economics just failed their first real test. NuScale — the only SMR company to receive NRC design certification — cancelled its flagship Utah project in November 2023 after projected costs climbed to $9.3 billion and $89/MWh, pricing it out of the market. First-of-a-kind nuclear in the US has a consistent record of costing multiples of estimates.
  3. Licensing 503 reactors is a regulatory fantasy. NRC licensing for a novel reactor design takes the better part of a decade even for one unit; the proposal's timeline assumes a parallel approval regime that doesn't exist. Each site would still need environmental review, security plans, and emergency planning zones.
  4. The fuel supply chain barely exists. Most advanced designs need HALEU (high-assay low-enriched uranium), and the only current commercial-scale supplier is Russia. The US is building domestic capacity, but fueling 503 reactors would require a supply chain orders of magnitude larger than anything planned.
  5. 503 sites means 503 security and waste problems. Even with passive safety, every site needs physical security, trained operators, and a spent-fuel plan — and the US still has no permanent geological repository after decades of trying. Multiplying the fleet multiplies every unsolved back-end problem.
Feasibility verdict

Could this actually be built?

The honest answer is: the physics is the easy part. TRISO fuel works, helium-cooled reactors work, and passive safety is a real, demonstrated engineering achievement — not a paper promise. X-energy, TerraPower, and Kairos are spending real billions to prove the commercial case right now.

The hard parts are cost, licensing, and scale. NuScale's cancellation is the cautionary tale: even a certified SMR design couldn't survive contact with American construction economics. And the proposal's five-year, 503-site timeline compresses what would realistically be a 20–30 year national program — assuming the first few units prove out — into a single presidential term.

The underground-siting concept is genuinely clever, and airport microgrids are a plausible end-state for advanced nuclear in the US. But the version with real precedent behind it starts with a handful of demonstration units at industrial sites, proves the economics, builds the fuel supply chain, and then scales — not 503 at once.

Verdict: the technology is real, the timeline is not. Build the first ten, learn what they actually cost, then talk about five hundred.

Aerial mock-up · from the proposal

What it would look like finished

"Surface airports are entirely covered in dark, monocrystalline solar pavement. At the center, massive vertical modular sky-towers rise into the air, while the 100-meter deep nuclear reactors remain entirely hidden underground."

— SMR-27 aerial mock-up description, Foreign Labor & Deployment appendix

Year 1 is the simultaneous deployment of 150 shaft-sinking machines; years 2–3 bore the 100-meter shafts and pour the barite concrete liners; year 4 delivers the TRISO fuel pebbles; year 5 activates the supercritical CO2 turbines. The workforce is cross-utilized from the high-speed rail build.

All cost and schedule figures on this page are the proposal's own estimates from the 30 Mega-Projects report series (Report 27), not independent appraisals. Real-world precedents cited — X-energy, TerraPower, Kairos, NuScale — are included so you can check the claims yourself.