향상된 지열 프로젝트가 데이터 센터 건설 붐을 주도할 청정에너지 기가와트 규모 발전을 목표로 본격화되고 있다.
Deep in the Utah desert, down miles of gravel roads and hours from a major city, Fervo Energy bets it can economically harness heat from the Earth's core for power, potentially unlocking dozens of gigawatts of baseload, emissions-free electricity at a time when U.S. power demand is meaningfully growing for the first time in decades.
Geothermal power is not new; traditionally, it relied on specific geological formations that severely limited its potential. Fervo now aims to demonstrate that enhanced geothermal systems—leveraging drilling and hydraulic fracturing techniques pioneered by the oil and gas industry—can emerge as a foundational power source for the United States. Fervo’s Cape Station project in Milford is scheduled to begin delivering power to the grid next month, positioning it as the first enhanced geothermal system in the country to achieve commercial operation. Following its initial public offering in May and a newly announced offtake agreement with Google, the company must now demonstrate that it can deliver cost-competitive enhanced geothermal power at scale.
Fervo’s origins predate the current data center power surge. CEO Tim Latimer, who began his career as a drilling engineer in South Texas, co-founded the company in 2017 alongside Chief Technology Officer Jack Norbeck. According to Latimer, early attempts to secure venture capital were met with skepticism: investors frequently asked, "Why would we invest in a power generation company? The electric grid's not growing." Oil and gas firms similarly dismissed the venture, operating under the assumption that drilling hotter wells through harder granite was unfeasible. Less than a decade later, that narrative has completely reversed. "It's a total transformation from 10 years ago, when people were questioning even the need for more new electricity," Latimer told CNBC. "This is a once-in-a-generation boom moment where hyperscalers, utilities and other customers need [power] now, and it's our power that will be an important part of the electric grid."
Recognizing the strategic value of prime acreage, Latimer began securing geothermal rights across the Western U.S. during the company’s early days, ultimately assembling a portfolio of nearly 600,000 acres. Acquired at an average of $4 per acre, the land now commands prices exceeding $400 per acre—a move Latimer views as highly prescient. The acquisition spanned Beaver County, Utah, a region also targeted by the Department of Energy’s Frontier Observatory for Research in Geothermal Energy (FORGE). FORGE has been compiling data to help private developers commercialize enhanced geothermal technology. Fervo’s flagship Cape Station project is situated directly adjacent to the FORGE site.
During a tour of Cape Station, Latimer noted familiar sights reminiscent of the Permian Basin, including drill rigs supplied by Helmerich & Payne. The critical distinction, however, is that Utah’s operations are dedicated entirely to generating emissions-free power. Unlike traditional geothermal projects that depend on naturally occurring reservoirs, enhanced geothermal systems involve engineers artificially creating reservoirs within the rock. Fervo drills two wells exceeding two miles in depth, one of which features a lateral section extending over a mile horizontally. The company fractures the surrounding rock, then pumps water down an injection well where it absorbs heat from temperatures surpassing 400 degrees Fahrenheit. The heated fluid returns to the surface through a separate production well, passes through a heat exchanger to generate electricity, and is subsequently cooled by rows of industrial fans atop the power plant. Finally, the geothermal brine is reinjected into the ground via a closed-loop system. The entire cycle—from the moment non-potable water reaches the surface to its reinjection—takes only minutes.
Latimer described Fervo’s drilling operations as advancing beyond conventional shale techniques, driven by three primary challenges: harder rock formations, higher temperatures, and extended well lifespans. To address these hurdles, the company engineered custom solutions for drill motors, drilling fluids, and bit designs, notably deploying polycrystalline diamond compact drill bits. Latimer credited these innovations with driving a "step change in performance for geothermal drilling."
From its inception, Fervo’s mission has been to advance both the science and technology of geothermal energy—a sector Latimer believes has lagged technologically compared to other industries. The company continuously gathers operational data to minimize drilling time, which directly correlates with cost efficiency. Inside the rig’s driller cabin, monitors display real-time metrics transmitted from downhole sensors, with identical data streams monitored concurrently at Fervo’s Houston headquarters. An on-site geologist simultaneously analyzes shifting granite properties, ensuring operators maintain optimal drilling conditions.
As drilling volume increases, Fervo has steadily reduced its costs. Nevertheless, geothermal power remains pricier than alternative energy sources, despite retaining federal tax credits under President Donald Trump’s One Big Beautiful Bill Act. At Fervo’s Nevada pilot facility, the initial wells required 70 days to reach a measured depth of 11,220 feet. During Cape Station Phase I, average drilling time fell to 21 days for depths of 14,483 feet, bringing costs to approximately $7,000 per kilowatt. For Cape Station Phase II, the company targets $5,500 per kilowatt and plans to drill its deepest wells yet at 19,448 feet. Concurrently, rising reservoir temperatures and larger-diameter piping have boosted the electricity output per well.
Further cost reductions remain essential for geothermal to compete with established power generation methods. Looking ahead, Latimer projected that Fervo could eventually bring costs down to $3,000 per kilowatt. "[B]ecause we can drill consistently… we've removed a lot of the uncertainty and the exploration risk that has plagued geothermal for years," he said. "[We] put geothermal on the kind of learning curve that has yielded dramatic cost reductions in other industries … We're expecting dramatic cost reductions in both the drilling side of the business and the power plant side of the business through that modularity and standardization."
Fervo’s expenditures are divided between subsurface drilling and the adjacent power plants, both of which are seeing cost efficiencies. Components for the initial 33-megawatt power plant were delivered individually. For Phase II, which scales capacity to 50 megawatts, parts and designs have been standardized, positioning the project for a construction timeline reduction of more than 30 percent. Fervo’s long-term objective is to deploy these facilities anywhere, including behind-the-meter installations for data centers—a configuration that places energy production and storage systems directly on the customer’s property. "We are developing a model where we are both going to be pursuing at the right locations transmission and grid-connected projects, as well as customers who want faster delivery," Latimer said. "Behind-the-meter is a great opportunity there."
The Google agreement raises Fervo’s total contracted capacity to approximately 1 gigawatt. Latimer confirmed that the company is engaged in advanced commercial negotiations with multiple additional buyers, including Southern California Edison, Shell Energy, and NV Energy, all of whom have committed to purchasing Fervo’s power. While Fervo has not disclosed the specific terms of its power purchase agreements, hyperscalers have demonstrated a willingness to pay premium rates, driven largely by corporate climate objectives. The company also emphasizes that its accelerated development timelines differentiate it from other generation technologies. By leveraging established oil and gas supply chains, Fervo avoids the severe bottlenecks affecting the gas turbine market—where manufacturers like GE Vernova report multi-year backlogs. Additionally, the Organic Rankine Cycle turbines utilized in Fervo’s plants are not subject to those same delays.
Fervo completed its highly anticipated initial public offering in May, raising nearly $2 billion through an oversubscribed issuance. Although the stock reached an all-time high on May 15, shares have since declined by approximately 57 percent. The company stated it intends to deploy the IPO proceeds toward geographic and operational expansion. "We have project sites throughout the entire Western United States," Latimer said. "We now have the capital to pursue it, and we want to repeat these GeoBlocks all over the West and then all over the world until we realize a more reliable, affordable, and sustainable electric grid."