Zap Rocks. Add Water. Get Clean Hydrogen.

On the edge of a horse farm outside Boston, a crane lowers a half-meter electrode into a borehole. High-voltage generators fire. The rock hundreds of meters below splits into a spiderweb of fractures. The goal isn't geothermal heat or mineral extraction - it's hydrogen^[1].

That's the pitch from Eden GeoPower, a Somerville, Massachusetts startup whose tagline is blunt: "We break rocks with electricity."^[1] The company calls the underlying process electrical reservoir stimulation, or ERS - a patented technique that uses high-voltage pulses to fracture tight subsurface rock with more precision than conventional hydraulic fracturing^[1]. The hydrogen application is the newest and most ambitious use case, and it sits at the intersection of two things the energy industry badly needs: a clean fuel that doesn't cost a fortune to make, and a production pathway that doesn't depend on electrolyzers, renewable power contracts, or government subsidies to stay solvent.

Why the Hydrogen Industry Needs a New Idea

The context here matters. Global hydrogen demand in 2024 reached approximately 100 million tonnes, containing energy equal to roughly 3 percent of annual world energy consumption^[1]. Almost none of it is clean. Unabated fossil fuels dominate the supply chain, with nearly two-thirds of production coming from natural gas reforming and most of the rest from coal gasification concentrated in China and India. Low-emissions hydrogen accounts for just 1 percent of total production.

The obvious fix - splitting water with renewable-powered electrolyzers - has been tried at scale and largely failed to deliver. A 2025 study published in Nature Energy, tracking 190 planned green hydrogen projects over three years, found that only 7% of globally announced capacity reached completion on schedule. That's not a rounding error; it's a structural problem. Every unit of clean electricity directed into an electrolyzer is a unit that can't power homes or factories directly, and the economics rarely pencil out without subsidy. The repeal of the U.S. 45V hydrogen production tax credit under the One Big Beautiful Bill Act in July 2025 removed the primary mechanism for achieving cost parity with grey hydrogen, leaving unsubsidized green hydrogen production costs at $4.50 to $7.00 per kilogram in many markets.

That's the gap geologic hydrogen is trying to fill.

What Serpentinization Actually Is

The underlying chemistry isn't new. Hydrogen is liberated when water reacts with reduced iron (iron 2+) in deep iron-rich continental and oceanic rocks, converting it to oxidized iron (iron 3+) and producing serpentine minerals in the process - a reaction known as serpentinization. It happens naturally, slowly, wherever groundwater contacts iron-bearing minerals like olivine. The problem is timescale: natural serpentinization operates over geological epochs, not project finance horizons.

Iron-rich minerals within 3 km of the surface could theoretically generate approximately 150,000 gigatonnes of hydrogen if all ferrous iron were fully oxidized - and 1 Gt of hydrogen contains enough energy to power the entire United States for a year. The resource isn't the constraint. The reaction rate is.

That's what stimulated geologic hydrogen, or engineered hydrogen, is trying to solve. The approach involves injecting water into iron-rich rock formations and accelerating the serpentinization reaction - producing hydrogen as a byproduct - at timescales that make commercial extraction plausible^[1]. Fracturing the rock first is critical: it creates the network of conduits that lets water reach iron-bearing minerals in the first place^[1].

How Eden's Approach Differs

Most stimulated hydrogen proposals involve injecting fluid under pressure - hydraulic fracturing - or adding chemical catalysts to speed the reaction. Eden's ERS technology is the only electricity-based approach in the ARPA-E geologic hydrogen program^[1].

The mechanism: two electrodes are lowered into separate boreholes at the same depth. High-voltage generators fire pulses between them. The rock between the electrodes heats, pressure builds, and it fractures - creating a precise, controllable network of cracks^[1]. The key advantage over hydraulic fracturing is directional control. Eden's ERS technology grants precise control over the directionality of the fractures created, addressing the problem of uncontrollable fractures that can dominate fluid flow and prevent re-stimulation in areas that still need fracturing - increasing fluid flow and ensuring more targeted reservoir stimulation. By using electricity rather than high-pressure fluid, the company can also reduce the water consumption, carbon emissions, and seismicity risks that come with traditional hydraulic fracturing.

In 2024, ARPA-E awarded $20 million to 16 teams aiming to advance stimulation technologies and research. Winning ideas included fracturing rocks with fluid pressure or mechanical stimuli, exposing them to catalysts to speed hydrogen-generating reactions, and manipulating native microbial communities to enhance production. Eden's rock-breaking project - the lone electricity-based approach - received $900,000.

Eden's two ARPA-E projects are titled "Electric-based Mechanical and Thermal Stimulation to Increase Geologic Hydrogen Reaction Rates in the Samail Ophiolite in Oman" and "Engineered Geologic Hydrogen Battery for Long-Term Energy Storage." The Oman work is significant: the Samail Ophiolite is one of the world's largest exposed sections of oceanic crust, rich in the peridotite rock types that serpentinize most readily. Eden's electrical stimulation method could produce significant surface area enhancement while also increasing the local temperature to promote reaction conditions suitable for hydrogen production.

Photo: Brandon Stoll / Unsplash

The Cost Case - and Its Caveats

The economics of stimulated geologic hydrogen are still largely theoretical, but the numbers being cited are striking. The U.S. Department of Energy estimates geologic hydrogen could be produced for less than $1 per kilogram, compared to $3.50-$6.00 for current green hydrogen, and cheaper than fossil fuel-derived hydrogen. ARPA-E's geologic hydrogen program has set aspirational targets of under $1/kg at the wellhead with less than 0.45 kg CO₂ emitted per kg of H₂ produced.

Existing studies suggest geologic hydrogen could have near-zero carbon intensity, particularly when not co-produced with methane - with hydrogen at 85% purity potentially carrying a carbon intensity as low as 0.4 kilograms of CO₂ equivalent per kilogram of hydrogen.

Those are the targets. The honest read of where the industry actually stands is more cautious. The industry has very limited real-world experience. Most extraction methods are still experimental, and the only operating project - in Mali - produces just a few tons of hydrogen per year for local electricity. That proof of concept is important but insufficient to demonstrate whether geologic hydrogen can scale to meet global demand.

For stimulated hydrogen specifically, the gap between lab and field is even wider. Accelerating geologic hydrogen production from geological to human timescales via enhanced serpentinization remains a formidable scientific and technical challenge. Serpentinization rate is hindered by low porosity and permeability of source rocks, suboptimal temperatures, unfavorable water chemistry, inefficient Fe²⁺-to-Fe³⁺ conversion, thermodynamic constraints, and low reactive surface area.

Eden's Boston field test is a proof-of-concept for the fracturing technique, not a demonstration of commercial hydrogen production. No electrolyzer size, no production rate, no commissioning date, and no offtake agreement have been announced. The company has raised $12 million in seed funding and $1.4 million in ARPA-E grants - meaningful for early-stage research, but a long way from the capital required to build a commercial-scale operation^[1].

What Needs to Be True for This to Work

Stimulated geologic hydrogen is genuinely novel - novel enough that the scientific community is still working out the basic mechanistic questions. There isn't yet enough understanding of the chemical reactions that happen during serpentinization: how iron loses electrons and reduces water or protons to hydrogen, whether it happens at mineral surfaces, and whether the process is catalyzed by other metals - these are material science, chemistry, and Earth science questions that still need answers.

For Eden's ERS approach to become a commercial hydrogen production technology, several things need to be demonstrated that haven't been yet:

• Fracturing at scale. The Boston field test shows the electrode can crack rock. It doesn't show that the resulting fracture network is large enough, or connected enough, to sustain meaningful water-rock contact over time.
• Reaction rate acceleration. Fracturing creates the conduits; it doesn't guarantee the serpentinization reaction will proceed fast enough to be economically useful. Temperature, water chemistry, and mineral composition all matter.
• Hydrogen capture. Getting hydrogen out of fractured subsurface rock without losing it to migration or microbial consumption is a separate engineering challenge.
• Reservoir replenishment. Unlike a conventional gas well, a stimulated hydrogen reservoir could theoretically regenerate as long as iron-bearing minerals and water are present. Whether that happens at useful rates in practice is unproven.

Several geological uncertainties continue to hamper the search for viable hydrogen reservoirs. Although the processes that generate native hydrogen are well established in theory, their real-world rates, spatial continuity, and environmental dependencies are still relatively poorly understood. The mechanisms of hydrogen accumulation, a key prerequisite for economically viable extraction, remain particularly under-researched.

The Bigger Picture

The reason this deserves attention despite all those caveats is the scale of what's theoretically on offer. Some estimates suggest that trillions of tons of geologic hydrogen may exist underground. Even accessing a small fraction could satisfy global hydrogen demand for centuries. Certain reservoirs may even replenish through ongoing geological processes, though this remains under study.

That's the kind of resource base that justifies serious research investment even at long odds. The ARPA-E program's $20 million across 16 teams is exactly the right scale for this stage: enough to run rigorous experiments, not enough to paper over fundamental scientific uncertainty with capital.

Eden's contribution - using electricity rather than fluid pressure to fracture rock - is a genuinely differentiated approach. Geologic hydrogen's risk is concentrated at the front end of the value chain. Unlike manufacturing an electrolyzer, finding and producing natural hydrogen is a probabilistic endeavor akin to early oil and gas exploration. The ERS technique, if it works at scale, could give operators more control over where and how fractures propagate - which matters enormously for a resource where the geometry of the rock-water interface determines whether you get hydrogen or nothing.

The Boston horse farm is a long way from a commercial hydrogen well. But it's the kind of experiment the industry needs to be running right now - methodical, field-tested, and honest about what it doesn't yet know.