Hunting for geologic hydrogen

Start with why the prize is so large, because it is a different claim from “cheap hydrogen.” Hydrogen today is an energy carrier, meaning you spend energy to make it. Electrolysis puts 50 to 55 megawatt-hours of electricity in to get a metric ton of hydrogen holding about 33 megawatt-hours back out, roughly 60% efficient at best. Making it from natural gas costs about 40 megawatt-hours of thermal energy for the same 33. Either route ends with less energy than it started with, which is why hydrogen functions as storage and transport rather than as a source. Koloma’s estimate for a natural hydrogen field is 1 to 3 megawatt-hours of parasitic energy, spent on compressors, pumps and purification, to bring out that same 33. That is the range it takes to extract gas, coal or oil, and it is what would make geologic hydrogen primary energy. Kann calls it the first significant new source of primary energy in a century and Johnson agrees, dating the previous one to nuclear in the 1930s and 1940s. One qualifier of Johnson’s is worth keeping attached: he expects production cost similar to natural gas on a volumetric basis, not on an energy basis, because natural gas holds about three times as much energy per unit volume as hydrogen.

The reason to doubt any of this is not the chemistry. Water reacting with iron-rich rock to release hydrogen, a process called serpentinization, is uncontested, it is exothermic so it needs no catalyst, and hydrogen seeps turn up all over the planet. The debate is entirely about whether that hydrogen accumulates anywhere you could tap. Johnson gives three reasons it should be harder than oil and gas. Oil and gas are a one-ingredient system, buried organic matter, whereas hydrogen is a two-reactant system, so the pathway water took into the rock is a pathway the hydrogen can take back out. Hydrogen is a small molecule and needs tighter seals than oil or gas. And hydrogen is readily metabolized by microbes, so a filled reservoir can be emptied by biology in a way that never threatens helium. His rebuttal to the sealing objection is the cleanest move in the episode: helium is smaller still, and we do find helium trapped underground, so nature demonstrably makes seals tight enough. The apparent advantage over oil and gas, that the reaction is ongoing rather than a one-time ancient burial, he immediately qualifies. It is fast like a snail and not like a tree, on the order of thousands to hundreds of thousands of years, so a producing reservoir should not be assumed to refill at any useful rate.

What is publicly proven is, in his own word, very little. The theoretical resource is enormous, since roughly 10% of the earth’s crust is the mafic rock that can produce hydrogen, and the US Geological Survey has put the scale on the order of trillions of tons and hundreds of years of energy, a figure Johnson calls probably a bit aggressive against Koloma’s own view. Proven reserves are close to nil. A water well drilled in Mali in 1987 struck hydrogen and has produced high-purity hydrogen for decades, but enough to power a village rather than enough to matter. The Australian company Gold Hydrogen twinned an old well, drilling a new one beside a decades-old hole that had shown hydrogen, and flowed gas from it, which demonstrates flowable hydrogen rather than a large reservoir. Johnson then defines the watershed moment precisely, and it is a two-stage test borrowed straight from oil and gas: a discovery well that flows through a multi-month production run before pressure declines, then three to five appraisal wells around it, then a third-party audit certifying proven reserves large enough to sustain production over a 15 to 30 year life. Large enough means roughly 10,000 tons a year to support liquid hydrogen, or around 50,000 tons a year to support an ammonia plant. As of this recording, nobody has publicly announced even the first stage.

Two further objections he treats as answerable rather than answered. On geography, the resource will not sit under existing demand, but hydrogen is the dominant cost in the things hydrogen makes, about 75% of the cost of ammonia, so a greenfield clean ammonia or synthetic fuel plant has a real reason to move to the resource instead. His analogy is Houston, which became a refining center because the oil was found there rather than the other way round. The hedge is that this holds for most locations inside the United States and not for the middle of Greenland, and that it requires confidence the hydrogen will keep flowing for 20 years. On leakage, hydrogen is an indirect greenhouse gas because it interferes with the breakdown of methane in the atmosphere; Johnson says plainly that he is not an atmospheric scientist and that the warming numbers are a moving target with low and high estimates in circulation. His case for it being manageable is empirical rather than theoretical: industrial gas companies have operated hydrogen wells and pipelines for four decades, and the best of them lose about a quarter of a percent between wellhead and customer, so the 20% leakage figure he has seen circulating does not match operating practice. Underneath all of it sits an asymmetry he keeps returning to. He thinks nothing being worked on anywhere could do more for decarbonization, and he also thinks his job as a chief executive is to talk people down, because good exploration programs run by the best oil and gas companies in the world take years and cannot be compressed by spending more money.