Catching fire

Past efforts to coax geothermal energy from hot, dry rock deep underground have faltered. But new techniques could crack the problem

Go to content
Eric Larson/Flash Point SLC
issue cover image
Table of contents
A version of this story appeared in Science, Vol 377, Issue 6603.Download PDF

Milford, Utah—The day started inauspiciously for John McLennan, as he tried to break the curse haunting a 45-year quest to coax abundant energy from deep within Earth.

First came news of an overnight accident that left one researcher recuperating in a hotel with a sore back. Then reports trickled in that seismic sensors dangling inside holes bored deep into the Escalante Desert here were malfunctioning. Repairs were delayed by gale-force winds that whipped the sagebrush-covered hills and buffeted a drilling rig that rose 50 meters from the desert like a misplaced lighthouse. Workers were already a day behind schedule, and each day burned an additional $350,000.

Finally, shortly before sunset, McLennan, a geomechanics engineer at the University of Utah, was ready to take a critical step in advancing a $218 million project, 4 years in the making, known as FORGE (Frontier Observatory for Research in Geothermal Energy). If successful, FORGE will help show how to transform dry, intensely hot rock found belowground all over the world into a major renewable source of electricity—and achieve a technical triumph where many others, over many years, have failed.

Gusts no longer rocked the trailer where McLennan, eyes baggy with fatigue and wearing the same brown sweater as the day before, faced five computer screens. The trailer’s door opened and a co-worker—a giant of a man wearing a white hard hat—looked in. “You ready for me to go?”

“Yeah,” McLennan replied. “We’re ready.”

With that, powerful pumps nearby sprang to life and began pushing thousands of liters of water down a hole drilled 3 kilometers into the hard granite below.

The concept of using Earth’s internal heat to generate electricity is attractively simple. Temperatures in the planet’s core approach those found at the surface of the Sun, and the heat leaks outward. In places this geothermal energy emerges at Earth’s surface as molten lava, steaming vents, and hot springs. More often, however, it remains trapped in deep sediments and rock.

There’s plenty of it. By one recent estimate, more than 5000 gigawatts of electricity could be extracted from heat in rock beneath the United States alone. That’s nearly five times the total currently generated by all U.S. power plants. Geothermal energy is also attractive because it doesn’t burn fossil fuels, isn’t imported, and can run around the clock, unlike solar panels and wind turbines.

Aerial photo of the geothermal plant.
In Utah, researchers are testing new ways to use hot, dry basement rock to produce energy.Eric Larson/Flash Point SLC

Tapping that heat, however, has proved difficult. Some nations—notably volcanically active Iceland—siphon hot groundwater to heat buildings and generate electricity. In most places, however, the rock lacks the water or the cracks needed to easily move heat to the surface. For decades engineers have sought to coax heat from this hard, dry basement rock, which can reach temperatures of more than 250°C. But those efforts have largely failed, often at huge expense—and sometimes after causing damaging earthquakes. As a result, geothermal energy provides just 0.33% of the world’s electricity, little changed from 1990, according to the International Energy Agency.

In recent years, new hope for this renewable energy source has come from an unlikely source: new technologies developed by the oil and gas industry. The same methods that have boosted fossil fuel production in the United States, such as targeted drilling and fracking—artificially fracturing deep rock with high pressure fluids—can, it’s hoped, be put to work to efficiently and safely extract energy from hot, dry rock. Government agencies and private companies are pouring hundreds of millions of dollars into the approach, called enhanced geothermal systems (EGS), though it, too, has had setbacks. Now, FORGE, situated on a remote patch of land in southwestern Utah, has become a closely watched effort to demonstrate and fine-tune EGS technologies—and finally break the losing streak.

“Geothermal isn’t going to work if we can’t make this [EGS] work,” says geologist Joseph Moore of the University of Utah, who leads FORGE. “That’s really the bottom line.”

For McLennan, FORGE brings a sense of déjà vu. In 1983, when he was an engineer at an oilfield company, he worked with scientists from the Department of Energy’s (DOE’s) Los Alamos National Laboratory on a pioneering attempt to exploit hot, dry rock in New Mexico’s Jemez Mountains. The scientists had hoped to create what amounted to artificial hot springs, by injecting water into deep fractures and then channeling the heated water back to the surface via a nearby exit well. But much of the injected water never resurfaced; researchers later concluded that they had misread the underlying geology, and the water disappeared into undetected cracks.

“It was a disappointment,” McLennan recalls. And it was one of many.

Portrait of John McLennan at geothermal energy plant. He has a broad smile and gray stubble, and wears a hard hat.
Geomechanics engineer John McLennan is part of the FORGE team.Eric Larson/Flash Point SLC

Much the same thing happened decades later to a $144 million geothermal plant in Australia’s arid Cooper Basin. Water pumped into the wells flowed into a previously unknown fault, and the project shut down in 2016, after just 5 years.

In some places, EGS projects had more dramatic failures, as high-pressure water injected for fracking caused existing faults to slip, setting off earthquakes. In 2006, engineers shuttered a project beneath Basel, Switzerland, after earthquakes caused minor damage. Eleven years later, a magnitude 5.5 quake struck Pohang, South Korea, killing one person, injuring dozens, and causing more than $75 million in damage. It was traced to a new, nearby EGS project where, despite a series of tremors, operators had injected fluid at high pressures near a previously unknown natural fault.

The high cost of drilling into hot, dry rock is also a challenge. Equipment designed for the softer, cooler sedimentary rock often found in oil fields falters in the extremes of hot, hard rocks such as granite.

Today, just three EGS power plants—all near the border of France and Germany—produce electricity. In total, they generate less than 11 megawatts, enough to power about 9000 homes.

EGS “always has been fraught with technological challenges,” says Jamie Beard, an attorney and executive director of the new nonprofit Project InnerSpace, which is seeking donations to help geothermal startups. And “EGS in its pure form like FORGE is hard,” she adds.

Even as EGS projects have struggled, however, new techniques have emerged from the oil and gas industry. Engineers learned to drill horizontally instead of just vertically. Today they can create wells that can resemble rollercoaster routes, curving and doubling back on themselves. Sophisticated steering systems allow drillers to target their fracking to release oil and gas from veins of rock as narrow as 5 meters. The advances have prompted investors and governments to take a fresh look at EGS.

U.S. map showing the location of Milford in southwest Utah.

Plumbing the earth

Scientists are trying to extract the abundant heat trapped belowground in places where hot water doesn’t flow through the bedrock. Their goal: to create artificial systems that act like hot springs, commonly known as enhanced geothermal systems (EGS).

Diagram of an EGS plant.
1 In EGS, a cold fluid such as water is injected into wells deep enough to reach hot, dry rock, often kilometers down.
2 Water is heated as it flows into cracks in the rock. In EGS, high-pressure fluids are used to create new fractures or enlarge existing ones, a technique much like the fracking used in the oil and gas industry.
3 Hot water is pumped to the surface via exit wells drilled not far from the injection well. The water turns to steam as the pressure drops closer to the surface.
4 The steam is directed to turbines that generate electricity. Most of the water is reused and continues the cycle.
5 EGS-like techniques can also be used to create water that’s not hot enough to efficiently generate electricity. But the artificial hot springs can be used to warm homes and commercial buildings, and drive a variety of industrial processes.
C. Bickel/Science

In the United States, more than two dozen geothermal companies have emerged since 2020, Beard says; that’s more startups than she counts in the previous decade. In Germany, the Helmholtz Association of German Research Centers announced in June it is putting €35 million into a new underground laboratory dedicated to geothermal research in deep crystalline rock, including EGS. And DOE in April announced plans to spend $84 million on four EGS pilot projects. They’ll be placed in different geological settings in the United States to study the best ways to extract heat from different types of rock.

Those plants will build on the results of FORGE, which DOE launched in 2014 with a competition to create a laboratory for honing EGS tools. In 2018, DOE announced the University of Utah and partners had won the funding to build the facility near the small railroad town of Milford, Utah, where Earth’s feverish interior creeps close to the surface.

The Milford Valley’s veneer of vegetation is so thin that much of its geologic history is exposed like an open book. Moore has spent more than 40 years reading that tome. Earlier this year, he stood next to a low cliff of silica that runs north to south along the top of a small hill. “This is a fundamental boundary,” he declared.

The wall divides what are, for the purposes of geothermal energy, two different worlds. To the east, the ground quickly rises to the flanks of the Mineral Mountains, rounded peaks speckled with granite outcroppings and juniper trees. A flat-topped mountain devoid of granite marks the top of a long-dead volcano, one of nine that testify to the heat still trapped beneath the ground in this region.

Between the cliff and the mountains, abundant hot groundwater flows close to the surface. A squat, brown building tucked into the foothills holds a 38-megawatt conventional geothermal power plant whose wells tap into that water. Just beyond, the steaming vent of a hot spring emerges next to the rock-walled ruins of a crudely built “resort” from the late 1800s. It catered to miners from the nearby Horn mine, once declared the world’s richest silver deposit. At Moore’s feet sat opal gemstones formed about 1600 years ago when hot springs saturated with silica spilled to the surface, leaving behind rocks candy-striped in yellow, red, and white.

West of what is known as the Opal Mound fault, the groundwater is blocked by an underground wall of solid granite that reaches temperatures of 235°C—truly hot, dry rock. FORGE’s drilling rig perches above that granite, its metal skeleton dwarfing the trucks and one-story buildings clustered around it.

Starting in 2020, crews used a similar rig to drill an injection well. The completed shaft, 22 centimeters in diameter, extends for 3.3 kilometers. The well includes features that are standard in fracking operations but still cutting edge in EGS. For example, FORGE’s shaft dives into the target granite at an angle that is close to horizontal—chosen to intersect with natural stresses in the rock in a way that would enable engineers to amplify tiny existing fractures.

At its deepest point, the shaft pierces rock that is 1.7 billion years old. Conventional metal drill bits struggled to cut through this stone, and the younger granite above, advancing just 3 meters per hour and frequently disintegrating. That prompted a switch to tougher drills tipped with synthetic diamonds—a first for geothermal drilling in granite. The bits sliced through the rock 10 times faster, and are “definitely a breakthrough,” says Peter Meier, an engineer and CEO of the Swiss geothermal company Geo Energie Suisse, who visited FORGE earlier this year to help with seismic monitoring. “This is already a very big result of the project.”

50-meter-tall drilling rig in Utah's Milford Valley
In Utah’s Milford Valley, a 50-meter-tall drilling rig marks where scientists are working to extract geothermal heat from deep in the rock below. Water is pumped through the rig into a well drilled 3.3 kilometers down into hot, dry granite.Eric Larson/Flash Point SLC

The FORGE well is lined with a steel casing that’s standard in oil fracking. Such linings make it easier to use specially designed gaskets to seal off sections of pipe in which operators detonate small explosives, shattering the pipe and exposing the surrounding rock. That helps FORGE fracture rock bit by bit—another novelty in EGS.

That piecemeal approach could help EGS projects avoid fracking in seismically sensitive areas that could trigger nearby faults, says William Ellsworth, a Stanford University geophysicist who has studied drilling-induced tremors, including the Pohang quakes. But he cautions that spotting problem faults in hard basement rock is “an exceedingly difficult imaging problem.”

At FORGE, moreover, the granite’s high heat has crumbled the gaskets typically used in cooler oil and gas wells. It’s also fried seismic sensors essential to tracking the fracking operation. So, the team has been testing special high-temperature gaskets and new monitoring tools, including fiber optic cables able to withstand the heat while detecting tiny vibrations in the earth.

Universities, government labs, and companies are currently developing other technologies they hope to test at FORGE. One is a small device, resembling a motorized skateboard, that would drive deep into the well to open and shut “windows” in the steel casing that expose nearby rock for fracking, another technique for targeting specific regions of rock. Such “tool development … is absolutely critical” to moving EGS into the mainstream, Moore says.

On 16 April, McLennan and the FORGE team were ready for one of their first big tests: seeing whether they could pump enough water into the deepest part of the well, under enough pressure, to enlarge tiny cracks or create new ones in pockets of granite.

The equipment malfunctions and high winds had initially thrown the schedule into disarray. But by late in the day, McLennan was perched in front of his computer screens, like an air traffic controller ready to guide a plane in for a landing.

Voices crackled over radios. Beside him, Kevin England, a veteran petroleum engineer, issued short bursts of commands.

Outside, a full Moon rose over a row of water tanks, each nearly as big as a school bus. The roar of motors filled the air as powerful truck-mounted pumps moved water to the well through a spaghetti of pipes.

Water pours from three rusted pipes.
At FORGE, a vast network of pipes helps move water around the drilling site.Eric Larson/Flash Point SLC

On McLennan’s screen, numbers began to climb, tracking the water flooding into the hole, where the pressure would hit the last 60 meters of rock, left exposed without a steel shell. The engineers hoped that would allow them to create a focused, dense cloud of fractures, like an acupuncturist inserting needles into a specific nerve.

A red line crawled across the screen, marking the gradual rise in water pressure. Eventually, the line wavered, bouncing around a pressure of about 28 megapascals, more than 250 times the atmospheric pressure. The flutter was good news: It probably meant the rock was giving way. “We’re getting some action,” McLennan announced. “This is nice.”

Over the next half hour, the signals continued to be encouraging. For the first time in days, McLennan appeared at ease. “This is beautiful,” he said. “It fractures, it stops, then it propagates again.”

Portrait of Dr. Joseph Moore at a geothermal plant wearing a hard hat.
Joseph Moore, a University of Utah geologist, manages research at the Department of Energy’s Frontier Observatory for Research in Geothermal Energy.Eric Larson/Flash Point SLC

It was hard to know exactly what was happening nearly 3 kilometers below. But Jim Rutledge, a seismologist at a nearby screen, was gathering clues. Clusters of black dots appeared on a grid, marking tiny earthquakes detected by sensors in a monitoring well half a kilometer away. The tiny tremors were no cause for alarm—just a sign that the fracking was going as planned. “We have a big cloud,” Rutledge said.

At 77 minutes in, the volume of water pouring down the well had grown to 50 barrels per minute—an aspirational target some had predicted the team wouldn’t reach because of the unyielding rock.

“Let’s go to 60,” Moore pushed.

But McLennan urged caution: “Let’s stick to the plan.” An hour later, the pumps went quiet.

Moore and McLennan were buoyant the day after the April test: They had pulled off their first successful frack. Now, the FORGE team is sifting through the mountains of data collected during that test as well as two subsequent fracks at locations higher up in the same well. What they learn will shape their next steps.

A 3D map of the tiny seismic events triggered during the tests, for example, will help them decide where to drill a second well. If all goes as planned, in 2023 they will pump water down the first hole and then see what, if anything, flows back up the second.

Others are watching FORGE closely for lessons. Meier is leading plans for an EGS project in Switzerland. He hopes FORGE’s technique of executing smaller, segmented fracks will point the way to reducing the risk that EGS will cause damaging earthquakes, like those that shut down his company’s previous work in Basel.

Others are eager to see whether FORGE can identify ways to make EGS more commercially attractive by solving problems that today scare off would-be investors, including time-consuming drilling, broken equipment, and prolonged uncertainty over whether a well can produce hot water.

“That’s where we’re focusing all of our time and energy—taking away the risk from the [geothermal energy] community,” says Lauren Boyd, acting director of DOE’s Geothermal Technologies Office.

Others, however, see more immediate commercial promise in other strategies. Call them EGS 2.0. Beard, for example, argues for targeting softer, slightly cooler rocks at shallower depths, familiar territory for oil drilling. The approach would still rely on engineered hot springs and possibly fracking. But the formations are easier to work in, Beard says, and drillers have gained expertise from boring tens of thousands of wells into such geology.

Sage Geosystems is one company pursuing that strategy. Founded in 2020 by scientists and executives from the oil giant Shell, the Houston-based company aims to drill and frack a single well in sedimentary rock and use a set of concentric pipes to pump cool fluids into the rock and draw out hot ones. Instead of water, the firm might use liquid carbon dioxide, because it has a lower boiling point. The resulting steam would drive turbines specially designed to operate with carbon dioxide.

One longtime EGS proponent is trying a different, less technically challenging approach. Chemical engineer Jeff Tester of Cornell University helped run the Los Alamos work in the 1970s and was the lead author of a 2006 DOE report touting EGS. Now, he’s overseeing a program that this summer started to drill a test well in sedimentary rock on the Cornell University campus in New York state. Although rock temperatures could top out at just 100°C, that would be enough to produce hot water to heat all the university’s buildings, Tester says. And these lower temperatures are found in rock in many more places. “It doesn’t have to be high temperature,” he says. “That’s the beautiful feature of using [cooler rock] directly for heating.”

Around the FORGE site, researchers believe the hot, dry rock just a few kilometers below holds enough heat to power a city the size of Salt Lake City. Given its mission as a testbed, however, the facility might never produce enough power to light a single bulb.

That doesn’t trouble Moore. “The purpose of things like this is not to solve all the problems,” Moore said the day after the first frack, as he stood on a dirt road a short distance from the drill rig. Instead, he said, FORGE’s goal is to see whether it can “take [EGS] to the point where the private sector can see its viability.”

Support nonprofit science journalism

Help News from Science publish trustworthy, high-impact stories about research and the people who shape it. Please make a tax-deductible gift today.


Not Now