AINEWS
Eden GeoPower, a Massachusetts startup, is firing high voltage pulses between deep electrodes to fracture rock. The field test is a first attempt to crack hydrogen's 50 year energy balance problem.
Hydrogen only emits water when burned or run through a fuel cell. Making it currently costs more energy than it yields, and the cheapest route, steam-methane reforming, is itself a potent greenhouse gas. That energy-balance problem has stalled the hydrogen economy since the 1970s, and it is the constraint a Massachusetts field test by startup Eden GeoPower is trying to crack.
Eden's technique is "electrical reservoir stimulation." The company lowers paired high-voltage electrodes into separate deep boreholes at the same depth, then fires a series of pulses between them. Each pulse is a brief, intense arc through the rock, a "subterranean lightning strike" in the company's own words. The arc heats the surrounding stone and shatters it into a spiderweb fracture network hundreds of meters below ground.
Eden is running the test on a horse farm outside Boston. The setup is small by industry standards: an electrode roughly half a meter long, lowered into a borehole, with a paired electrode in a second hole nearby. The site is a first field deployment, not a pilot plant, and the company tagline "We break rocks with electricity" is closer to a service pitch than a finished product. The Massachusetts test is the first time the technique is being run end to end with the explicit goal of making hydrogen in place.
The pitch: a fracture network in hard, low-permeability rock gives water a path it would not otherwise have. Run water through that network, pressurize it, and the rock itself becomes the reactor. Hot, fractured stone may be able to drive chemical reactions that split water or release hydrogen from iron-bearing minerals. The appeal is that the energy goes into the rock, not into a steel tank and a catalyst bed. If the geology cooperates, the same pulse train that breaks the rock also heats it, and a single piece of equipment does two jobs that a conventional plant does separately.
The visible IEEE Spectrum excerpt does not include yield numbers, energy-in versus hydrogen-out ratios, or cost projections. Those are the metrics that decide whether "subterranean hydrogen" is a category or a curiosity. The article itself flags the open question: hydrogen today costs more energy to make than it yields, and the cheapest path uses methane. Eden's positioning of the method as a way to flip that equation is the company's own claim, not independent validation.
Eden's application list runs from geothermal heat to carbon storage to mineral mining. The same fracture network is the asset, and the hydrogen application shares infrastructure with the others. A field test that proves the rock-fracturing part of the equation still leaves the chemistry, the durability, and the cost structure open.
The horse-farm test will show whether the electrodes, the pulse train, and the borehole geometry can reliably fracture the targeted rock, and whether water circulated through the resulting network behaves the way the chemistry predicts. It will not, on its own, settle whether underground hydrogen can be made cheaply enough to compete with steam-methane reforming, and it will not retire the 50-year energy-balance question. The test's value is that it converts a simulation into a real fracture pattern, and a real fracture pattern into the next measurement.
The first reported hydrogen yield per pulse, and the energy that had to be put in to get it, is the ratio that will decide the rest. The Massachusetts site is built to produce exactly that number.