Deep under your feet, Earth is hot. Really hot. Some of that heat is left over from the planet’s violent formation. More is continually created by the splitting of radioactive atoms in the rocks below. In a few places, Earth’s hidden furnace dramatically makes itself known, bursting through the crust as volcanoes, geysers and rivers of magma.
Spectacular stuff, but also extremely useful – if we can get at it. Conventional geothermal power works best wherever geology has done the hard work for us: where fault lines crack the crust and hot water or steam rises close to the surface. That’s why countries like Iceland, New Zealand, Kenya and Indonesia can turn Earth’s heat into warmth for homes and electricity for grids, while much of the rest of the world has barely touched it.
Its appeal is obvious. Unlike solar power, geothermal doesn’t fade when the sun sets, and unlike wind power, it doesn’t stall when the air is still.
Advertisement
Over the past decade, advances in drilling have begun to loosen geothermal power from its geological shackles. Some projects aim to improve on nature, creating cracks and pathways where none existed. Others are more ambitious still: they want to drill so deep that geothermal power could work almost anywhere. “Ultimately, the whole world is the goal,” says , co-founder of geothermal energy company Quaise Energy. That would turn it from a niche resource into something far bigger – a way to tap the planet’s hidden heat wherever we happen to need it.
How we harness Earth’s heat
To convert heat from Earth’s interior into usable energy, engineers pump cold water into naturally occurring reservoirs in hot rock. There, it warms up and returns to the surface as hot water or steam, which can be used to drive turbines and generate electricity. The water is then sent back underground to begin the cycle again.
It’s a simple concept, and unlike fossil fuels, geothermal heat sources are practically inexhaustible on human timescales. “If you measure in terms of the energy stored within the Earth’s crust, geothermal heat exceeds by orders of magnitude all hydrocarbons,” says Houde. The trouble is that traditional geothermal requires three ingredients to work: heat, water and rock that is permeable or fractured enough for water to move through it. “Traditional geothermal systems are very constrained to countries and regions within countries where you have this unique combination of conditions,” he says.
That is why geothermal has flourished in places like Iceland. Sitting astride the mid-Atlantic ridge, the country has those core ingredients in unusual abundance. The global oil crisis in the 1970s pushed Iceland to exploit that advantage, and geothermal now makes up over . That may not sound like much, but Iceland’s geothermal plants can keep generating whatever the weather, making them a steady backbone for the grid. “Geothermal is really well placed to be the glue that will make low-carbon electricity systems work,” says , a sustainable energy expert at Imperial College London.
However, most countries don’t share Iceland’s good fortune. They may have heat or water or fractured rock, but not all three at depths that are easy to reach. Traditional geothermal plants are limited to mining heat from less than 2 kilometres underground. Go much deeper, and the engineering becomes brutal. Drills must chew through hard, hot rock. The vertical shafts that connect the surface to heat reservoirs, known as wells, must survive punishing temperatures and pressures. Even after all that, the rock may still be too tight and unbroken for water to flow through it.
Still, the prize is enormous. The International Energy Agency estimates that geothermal’s technical potential is around , assuming we can exploit heat down to about 8 kilometres. Conventional geothermal typically taps pre-existing reservoirs of hot water or steam in permeable, fractured rock less than 2 kilometres underground, leaving much of that potential out of reach. The implication is clear: to expand the resource, engineers must drill deeper – and create the underground plumbing that nature hasn’t provided.
Enhanced geothermal
The first of the new geothermal techniques does exactly that. Enhanced geothermal systems target hot but impermeable rock, typically 2 to 10 kilometres underground. Engineers pump in hot water at high pressures to open or widen fractures and create pathways for fluid to circulate. This is known as an enhanced geothermal system (EGS), and it is the most mature technology in geothermal’s new era.
But this approach has a downside. Forcing water into deep rock doesn’t just create useful cracks. , it can also disturb older, pre-existing faults that are already under strain. Many faults are held still by friction, like a heavy book resting on a tilted table. Pump water into them, and the pressure in their pores and fractures rises. That partly props the two sides of the fault apart, reducing the force clamping them together. If the fault was already close to slipping, that can be enough to induce an earthquake.
An investigation concluded that a nearby EGS project triggered a magnitude-5.5 earthquake in the city of Pohang in 2017. More than 80 people were injured and the earthquake caused an estimated (£148 million) of damage, making it the most destructive in the country’s history.
at Stanford University in California was part of the team that helped establish what role the geothermal plant played in the disaster. “In Pohang, they drilled a couple of wells that were a long way apart,” he says. The hope was that pumping water into one would fracture the rock enough to connect with the other, creating a route for water to pass between them and gather heat. But trying to force that kind of connection through the subsurface is risky, particularly if the drilling intersects faults that are already primed to slip.
That is why newer projects are trying to be more deliberate, says Ellsworth. In Iceland, for instance, engineers try to create EGS reservoirs more gently without blasting them open all at once by pumping cold water slowly and at relatively low pressure. Over months, the cooling rock contracts and cracks, says , executive vice president of resources at Icelandic energy company HS Orka.
Iceland also has an advantage born of necessity: it is a seismically restless place, so the country has already invested heavily in monitoring. At HS Orka, that now includes using geothermal wells themselves as early-warning sensors. When magma begins to move, it compresses the surrounding rock, creating a telltale signal in the well. “It’s a really cool design,” says Magnúsdóttir.
But even where EGS projects have learned to manage seismic risk, hard rock still makes an unreliable basis for a plumbing system. Fluid can leak away and contaminate groundwater, for instance. And the same fractures that make the system useful can also make it dangerous.
Advanced geothermal
One way around this is to stop asking the rock to carry water at all. Where natural fractures and reservoirs don’t exist, an alternative to EGS is to drill a sealed loop through hot rock. Engineers drill down, then sideways, until the boreholes connect underground, before lining and sealing them with steel and cement. That creates something like a buried radiator where fluid circulates, but never comes into direct contact with the rock itself.
This closed-loop system is known as an advanced geothermal system (AGS), and it has been tested successfully in Germany, where Canadian energy company Eavor has developed , which began construction in 2022 and started producing electricity last year.
Because closed-loop systems don’t force fractures open in the rock, they should reduce the risk of induced earthquakes. And because they don’t depend on continually pumping water into and out of those fractures, they should reduce it further still.
There is another possible advantage. If the working fluid stays inside a sealed pipe, it doesn’t have to be water. Companies can, in principle, use fluids that pick up and release heat more efficiently. Eavor’s working fluid is a company secret. The firm China Huaneng Group, meanwhile, has built the first geothermal demonstration plant that uses supercritical carbon dioxide – CO2 that is compressed and heated until it becomes not quite a gas and not quite a liquid. In that state, it is dense and flows easily, and is very good at moving heat. The firm’s test facility began operating in May this year in Zhengzhou, China.
But these systems have their own difficulties. Eavor has had to scale back the number of wells it planned to drill after rock debris clogged parts of the pipe network. Its plant works, but not yet as efficiently as hoped. That points to a broader question: can closed-loop systems circulate enough fluid, and draw heat from the rock fast enough, to be commercially useful? “The jury is still out,” says Ellsworth. “Time will tell whether these ideas of having just a single closed loop will be actually successful.”
And while AGS could be safer and cleaner than inducing fractures deep underground, it is still limited to places where hot rock lies within reach, usually a few kilometres down. To make geothermal truly global, companies will have to drill much, much deeper.
Supercritical geothermal
The deeper you go, the hotter Earth gets. Go deep enough and water crosses a strange threshold. At about 374°C (705°F), under sufficient pressure, it transforms into a supercritical fluid – akin to the CO2 in the China Huaneng demonstration plant – and can carry far more heat than ordinary hot water or steam.
“There are a few countries where geothermal does provide a sizable mix of the energy supply. But strictly the heat itself, which is fundamentally what geothermal energy is, is actually available anywhere if you go deep enough,” says Houde.
Supercritical temperatures are usually found at depths of at least 5 to 20 km, and often deeper. The rocks become harder the further you go: basalt or granite rather than the softer sedimentary layers often found closer to the surface. This rock chews through even the hardest tungsten carbide or diamond drill bits. Then there’s the time spent not drilling at all but instead replacing worn-down drill bits or equipment that begins to fail at higher temperatures and greater depths. These delays quickly push up costs, says Houde.
Quaise, which is based in Houston, Texas, wants to get around this by drilling in a different way – without using a drill bit at all. Its plan is to use conventional drilling for the upper part of the well, then switch to high-powered microwave energy once the rock becomes too hard. The hole would be lined with corrugated steel, helping the microwaves travel downwards. When that energy hit the rock, it would heat it until the rock cracked, broke apart or even vaporised. “There’s no drill bit mechanically destroying the rock here, it’s just the interaction of the energy with the rock,” says Houde.
The rock would be reduced into a fine ash that is purged from the well with a nitrogen-rich gas and collected at the surface. Quaise isn’t alone in pursuing non-contact drilling: Slovakia-based GA Drilling is also trialling laser- and plasma-based systems, and carried out its first field tests in 2023.
Once the hole is open, Quaise wants to create a deep, EGS-style reservoir. Water would be sent through fractures in superhot rock, picking up heat before returning to the surface to generate electricity. In principle, it is geothermal freed from geography.
Quaise’s test holes are still just a couple of kilometres deep, but the company sees them as a proof of concept for something much larger. Its first planned superhot geothermal plant, located in Oregon, is intended to come online by 2030, accessing temperatures up to 400°C (752°F) at a depth of 3 to 5 km. A potential future site would push deeper, potentially to 10 km. “By 2035, we want the first superhot well producing energy at a depth greater than 10 kilometres,” says Houde, who thinks his company’s technology could transform any region into geothermal country.
But there are still serious uncertainties. One is whether such deep wells can be kept open under the immense pressures found kilometres below ground. “It’s unclear whether that is possible,” warns Ellsworth. Houde accepts that the hardest engineering lies ahead. “I think it’s very viable for getting to these initial depths to prove superhot geothermal,” he says. “I think it’ll become increasingly more challenging to push the frontier on how deep we can go.”
Iceland offers a glimpse of how these different strands of geothermal might converge. Instead of simply resting on its laurels as one of the world’s great geothermal success stories, the country is fast becoming a test bed for what comes next, says Magnúsdóttir.
Its deepest geothermal well – the second well by the Iceland Deep Drilling Project, a consortium of the country’s leading energy companies and its energy authority – has already brushed against the promise of superhot geothermal. It was completed in 2017 and goes down 4.7 kilometres. “It is the first well in the world where supercritical conditions were confirmed in the geothermal well,” says Magnúsdóttir. Icelandic operators are also using EGS techniques to get more from existing fields, stimulating deeper faults to reach another kilometre or so of heat.
Next year, the Iceland Deep Drilling Project plans to drill its third well. “There, they just have to go down to a little over 2 kilometres to reach supercritical conditions,” says Magnúsdóttir. But even at that relatively shallow depth, the engineering challenges remain fierce. The main problem is well integrity. At such high temperatures, the steel casings that line the well can expand, deform and fail. In Iceland, engineers have been testing fixes, including flexible connectors between sections of casing.
This sort of engineering will be crucial if geothermal is to win over investors. Projects are expensive up front and slow to prove, and often need permits before developers know whether a site will work commercially. But the reward for patience can be unusually durable. “It’s very, very costly,” says at the European Geothermal Energy Council. “But once you’ve drilled that hole in the ground, that lasts for over 100 years.”
That is the promise and the problem in miniature. “We know that very deep geothermal energy would be fabulous if we could reach it,” says Ellsworth. “We just don’t yet know if it’s possible.”
