Yashee Mathur, PhD ’26: The Search for Clean, Affordable Hydrogen Energy

Yashee Mathur, PhD ’26, is hunting for an energy source that most people don’t know exists naturally underground: hydrogen gas.

While energy companies race to manufacture “green” hydrogen using renewable electricity, she is betting on a novel path to deliver the same clean fuel at a fraction of the cost — if she can find it.

Her mission, if successful, could rewrite the global energy story. The risk of failure is high. However, the potential rewards are considerable: unearthing a primary energy source that could be more abundant and affordable than natural gas.

Raised in an energy‑scarce region near Dhanbad, India’s coal capital, Mathur is driven to provide affordable, clean energy for traditionally overlooked communities. “I come from a place where we don’t have access to electricity 24/7,” she says. “The possibility of expanding energy access really resonated with me.”

From Conventional Energy to Geologic Hydrogen

Mathur’s startup, Hydrify, applies the physics and machine learning skills she developed in conventional subsurface exploration to the task of finding a new target: geologic hydrogen, or hydrogen that forms naturally underground. She first learned about geologic hydrogen while completing a PhD in energy sciences engineering at Stanford, becoming one of the first researchers to specialize in it.

After researching geochemistry in Alberta, Canada, and working as a petrophysicist in India, Mathur was recruited to Stanford initially to use machine learning to enhance oil and gas exploration. When Mathur came to Stanford, her focus expanded to exploring green applications such as carbon capture and storage or geothermal energy. Yet something clicked when she learned about the untouched frontier of geologic hydrogen.

Also known as “white” hydrogen, geologic hydrogen could solve a critical problem: existing hydrogen production is either dirty or expensive. “Grey” hydrogen — the dominant production method — uses steam methane reformation and emits a significant amount of carbon. “Green” hydrogen is produced via electrolysis from water and emits near-zero emissions when powered by renewables, but it costs too much for widespread use. Geologic hydrogen offers the best of both worlds: near-zero emissions at an estimated cost below $1 per kilogram — cheaper than grey hydrogen with a fraction of the carbon.

Even a small fraction of available underground hydrogen could provide more energy than all of the proven natural gas reserves on Earth, according to the U.S. Geological Survey. That could be enough to enable net-zero, carbon-free energy for two centuries.

However, subsurface hydrogen is underexplored and has attracted only a tiny slice of total capital for hydrogen exploration. Mathur aims to change that by addressing the main challenge of identifying hidden deposits.

Hydrogen trapped in underground rock formations is typically located hundreds of meters or more below the surface, like natural gas. These reservoirs require precise conditions, including a tight geological seal. Underground hydrogen deposits formed when water reacted with iron-rich rock deep underground. Yet the same pathways the water took also allow tiny hydrogen gas molecules to escape.

Rather than deterring Mathur, these technical challenges drew her in. Her skill set, blending subsurface knowledge with experience in machine learning and artificial intelligence, is unusually well‑matched to this core challenge. “There were just a handful of researchers around the world who were looking at natural hydrogen,” she says. “It was still very niche.”

Aided by advisor Tapan Mukerji, codirector of Stanford’s Center for Earth Resources Forecasting, Mathur chose the risk of laying foundations in a wide-open field with a potentially outsize impact. She completed her doctoral thesis on the exploration and commercialization of natural hydrogen and has a patent pending for a novel hydrogen detection methodology.

Building Hydrify at Stanford

Launching a venture to translate her research into real-world impact required a different kind of support. The Stanford Climate Ventures course helped Mathur envision how to spin out a company. That, and serving as copresident of the Stanford Hydrogen Club, introduced Mathur to her early team members. A grant followed from the Stanford TomKat Center for Sustainable Energy in September, helping her team create a hypothesis and validate assumptions to develop the business.

Mathur spent a summer exploring the commercial potential through Stanford Ecopreneurship’s EcoSprint program. Then, a Stanford Impact Founder (SIF) fellowship enabled her to pursue the startup full-time. The SIF program has provided Mathur with tactical advice, reliable income, and room to experiment. “The most important thing for me is the freedom to operate — to work full time on the business and still experiment,” she says.

Senior members of the Stanford community, including deep-tech founders and energy investors, have been generous with their stories and advice. If she has questions about a contract, Mathur can meet with an experienced founder over coffee for advice. “The community as a whole really helps,” she says. “You can reach out to these people, talk to them, and brainstorm ideas.”

Energy Access, Equity, and the Long Game

Mathur’s hypothesis follows the first principles of how hydrogen forms, migrates, and accumulates. Hydrify flips the industry’s dated norm of pursuing rare signs of the gas seeping from the surface. Instead, Mathur applies AI to large geological and geophysical datasets to predict where hydrogen is likely to occur underground, rather than where it’s already been found. Mathur’s algorithm ingests maps, soil surface samples, satellite images, geologic models, and other data to pinpoint sites, and does it faster.

Hydrify has already put these models to the test in the field. Once hydrogen is found, the extraction and transport infrastructure can largely borrow from existing oil, gas, or geothermal industries.

The implications for Hydrify stretch far beyond the startup itself. If geologic hydrogen scales, it could become one of the cheapest forms of power on today’s grid, while dramatically reducing emissions from high-carbon industries.

“This could be an affordable, reliable source of energy for regions the traditional grid has left behind,” Mathur says. She personifies that promise: Raised amid energy scarcity, Mathur is now building a company around a resource that could rewrite whom the energy system is built for.

“It could be huge,” she says. “It could also be a geological curiosity. But the potential was too big to ignore.”