Eavor’s Geretsried Pivot Raises Hard Questions About Next Gen Closed-Loop Geothermal

The recent GeoExPro interview about Eavor’s next-generation geothermal Geretsried project lands less like an update and more like a stress test result. Eavor was one of the more serious next-generation geothermal companies I had assessed, but that was never the same thing as saying it had solved geothermal. It had real engineers, real investors, real drilling, a real German project, and a real claim that could be tested. Now that test is producing evidence. The issue was never whether geothermal heat exists, or whether horizontal drilling exists, or whether surface equipment such as ORC units and heat pumps can be bought. Eavor’s specific claim was that it could build a closed, sealed, multilateral subsurface radiator, circulate a working fluid through it, extract enough useful heat, generate electricity or provide district heating, and do so at a cost and reliability level that would support repeatable commercial projects. Geretsried was supposed to prove that claim. Instead, the interview suggests that the proof point has become the problem.

The source matters here. GeoExPro is not a general news outlet parachuting into geothermal. It is a subsurface-focused energy publication, and the article was written by Henk Kombrink, GeoExPro’s editor-in-chief and a geoscientist who writes regularly on geothermal, subsurface storage, oil and gas, and the practical realities of drilling underground. Kombrink’s technical knowledge and insights complement mine in energy generation, and I’ve referenced his points on Eavor in the past. They informed aspects of my lengthy white paper on global geothermal prospects published last year.

Geretsried was always the test. The attractive part of Eavor’s story was that it avoided some of the hardest problems in conventional geothermal. It did not require a naturally productive hydrothermal reservoir. It did not need to produce hot brine full of dissolved minerals. It did not require the same hydraulic stimulation logic as enhanced geothermal systems. It promised low seismicity, a sealed loop, predictable operation, and heat extraction from hot rock rather than hot water. In policy and investment circles, that is a powerful story. It makes geothermal look less like a risky mining-adjacent subsurface gamble and more like a manufactured energy asset. The problem is that the subsurface does not become manufacturing because a pitch deck says so.

My earlier risk assessment focused on the hard bits. Eavor had to drill deep. It had to drill long. It had to steer accurately. It had to connect laterals. It had to seal open-hole rock. It had to avoid clogging. It had to keep hydraulic resistance low enough that parasitic pumping did not eat the product. It had to maintain thermal output. It had to keep drilling cost under control. It had to operate for decades in a hot, wet, chemically active, mechanically stressed underground environment. That is a long list of gates. Passing one or two does not pass the course.

The new GeoExPro interview on Geretsried changes the emphasis from risk assessment to evidence. Eavor’s first large commercial demonstration in Bavaria was supposed to be the proof point. It was backed by a large project finance stack, including a €91.6 million EU Innovation Fund grant, an EIB loan of about €44 million to €45 million, roughly €87 million of loans involving JBIC, ING and Mizuho, and other equity or project funding to bring the headline package to about €350 million. The EIB project page has a total project cost somewhat higher, at €368 million. Either way, this was not a small pilot. It was not a benchtop experiment or a shallow demonstration in friendly geology. It was meant to be the commercial reference plant.

The physical plan matters as much as the financing. The original idea was to drill four injector-producer well pairs, each with associated closed loops branching off underground. So far, Eavor has completed one of those four well pairs. Within that one completed pair, GeoExPro reports six completed horizontal loops instead of the planned twelve, with only three to four materially contributing. Two loops were clogged by rock fragments and could not be cleared, while another appears to contribute only partly. That is not a minor construction delay. Eavor’s economics depend on enough long, clean, productive loops adding up to enough heat-transfer area. A closed-loop geothermal plant is not valuable because one borehole gets warm. It is valuable only if the installed underground heat exchanger works as a large, durable, low-resistance radiator.

The reported power performance reinforces the completion problem. Geretsried was framed around roughly 60 MW to 64 MW of thermal output and about 8 MW of electricity. GeoExPro reports current gross electrical output of only about 0.5 MW to 1 MW, while plant parasitic demand is around 0.5 MW. That means the plant can be barely net-positive or effectively not net-positive depending on operating conditions. Against an 8 MW electrical promise, that is not a shortfall. It is an order-of-magnitude miss.

The most telling part of the new Eavor story is not only underperformance. It is the pivot. The new CEO appears to be moving Eavor away from being the developer and operator and toward being a technology provider. That phrase sounds tidy. It suggests a maturing technology company realizing that its role is to license intellectual property while infrastructure specialists build and operate assets. That would be reasonable if Geretsried were a successful reference plant. But Geretsried is not a completed, high-performing asset looking for routine operations and maintenance. It is an underbuilt, underperforming first-of-kind project that still appears to need more drilling, more completion work, more remediation, more capital, and more proof that the core subsurface system can deliver.

This is not a normal O&M handoff. It does not appear to mean only that Eavor does not want to run a finished turbine hall or handle district heating customer service. It means Eavor is stepping away from the practical, physical, risk-bearing parts of its own technology proposition. Drilling, completion, sealing, flow assurance, remediation, loop reliability, thermal delivery and net energy output are not side issues. They are the technology in the only sense that matters. A closed-loop geothermal company that retreats from subsurface delivery is no longer selling a working energy system. It is selling a recipe.

The obvious next question is who would take the job. A competent drilling or subsurface operator would not view Geretsried as a normal operating contract. It would see an unfinished, distressed, first-of-kind completion problem with unclear performance upside and large downside risk. The remaining work is not merely running pumps and maintaining a turbine. It is taking responsibility for more drilling, loop completion, blockage risk, flow assurance, and future operating performance on a project whose original budget has already been spent. That is not a role many serious firms would accept except on time-and-materials terms, with narrow liability and no guarantee of thermal or electrical output.

That point cuts straight through the licensing pivot. If Eavor is looking for a third party to finish and operate the project, the third party will price the risk. It will want payment for equipment, crews, mobilization, drilling time, non-productive time, geological surprises, lost tools, borehole problems, cuttings management and remediation attempts. It will not want to guarantee Eavor’s promised heat output, thermosiphon performance, net generation or project economics. The organizations that understand drilling risk best are the least likely to underwrite someone else’s unproven energy model for a fixed price.

The hard question is what is left in Eavor’s technology package that anyone should pay much for? Much of the stack is not proprietary in any meaningful sense. Thermosiphon is physics. It is not an invention. Organic Rankine cycle power generation is standard equipment. Heat pumps are standard equipment. Surface heat exchangers, pumps, controls, district heating integration and grid interconnection are normal industrial engineering. Subsurface radiative and conductive heat modelling is standard in geothermal and oil and gas-adjacent engineering. Horizontal drilling is standard. Directional drilling, magnetic ranging and borehole steering are standard. Lengths of laterals and spacing between laterals are design variables, not a moat.

That does not mean Eavor has no technology. It means the real technology claim is narrow. The core proprietary element appears to be Rock-Pipe, or the broader wellbore synthesis approach. In simple terms, Eavor’s system needs to turn open-hole lateral wellbores into sealed underground pipes without casing all of them in the conventional way. That matters because casing tens of kilometres of laterals would add cost and likely reduce heat transfer. If Eavor can reliably seal the rock wall itself so that the wellbore behaves like a durable pipe, that is a meaningful engineering accomplishment. It is the closest thing in the stack to a crown jewel.

But a crown jewel is not a moat by itself. Subsurface sealing chemistry is not an empty field. Oil and gas, geothermal, underground storage and other subsurface industries have worked on sealing, grouting, mineralization, permeability control, wellbore strengthening and fluid-loss management for years. Variants exist. Some subsurface energy storage concepts have related approaches. A competent drilling and completions organization can look at the functional need and ask how to reproduce enough of it without copying Eavor’s exact patent claims. Seal the rock. Reduce permeability. Maintain flow. Manage solids. Avoid thermal degradation. These are hard tasks, but they are not beyond first principles.

The moat problem is that Eavor’s most distinctive claimed technology is also the thing now under question. If Rock-Pipe and the associated completion workflow reliably created sealed, clean, durable, high-flow laterals, Geretsried should be the evidence. Instead, the reported evidence is incomplete well pairs, incomplete loops, clogged loops, weak contribution from some completed loops, low current output and exhausted budget. That does not prove the method can never work. It does show that the method has not yet been demonstrated as a repeatable commercial delivery system. Rock-Pipe may be the crown jewel, but a crown jewel that cannot be delivered repeatedly, kept clean, and tied to bankable output is not a moat.

A business moat requires more than patents. It requires customers who cannot easily substitute away, competitors who cannot route around the claims, high switching costs, unique execution capability, unique data, regulatory lock-in, or proven superior economics. Eavor appears weak on most of these. Conventional hydrothermal geothermal still works where the resource is good. Enhanced geothermal systems are advancing through firms such as Fervo. Other closed-loop systems can use different completion architectures. District heating can use industrial heat pumps, waste heat, sewage heat, thermal storage, resistive backup and renewable electricity. Firm clean power has many competitors, including hydro, nuclear in some jurisdictions, batteries, demand response, grid interties and combustion of scarce sustainable fuels for rare backup.

The lack of obvious copycats does not prove Eavor has a strong moat. It more likely proves the opposite. If there were a large, obvious, high-margin market behind Eavor’s exact architecture, sophisticated subsurface firms would be trying harder to copy it, route around it, or acquire their way into it. The absence of a rush may be a market signal. Competitors may not be blocked by Eavor’s IP. They may simply have decided that deep, sealed, multilateral closed-loop radiators are too expensive, too risky, too geology-specific, or too hard to finance until someone else proves them. Geretsried was supposed to change that. So far, it has not.

That creates a buyer problem for the licensing pivot. What developer would pay Eavor a meaningful upfront licence fee now? A private geothermal developer would ask why it should pay for unproven IP while also carrying the drilling, completion, output, financing and operational risk. An oil and gas company would ask the same question with more confidence, because it already knows subsurface risk. A district heating utility would ask who guarantees delivered heat. A bank would ask where the reference plant is. An insurer would ask what exactly is being warranted. A public agency might pay for demonstration value, but public subsidy is not a commercial moat.

A rational buyer could hire much of the required capability directly. It could hire geothermal reservoir engineers, drilling engineers, completions experts, ORC suppliers, heat-pump integrators, project managers and subsurface modelling specialists. It could hire Eavor alumni if Eavor is reducing headcount. Public reporting has Eavor cutting staff from 147 to 80. That matters. If the company’s value lies in tacit knowledge, and the people who hold part of that knowledge are leaving, buyers do not have to pay full corporate licensing rates to access every lesson. They still have to respect patents and confidentiality, but labour-market leakage weakens Eavor’s negotiating power.

Chubu Electric is the one buyer that still makes strategic sense. Chubu invested in Eavor in 2022, reportedly between ¥1 billion and ¥5 billion, about $7 million to $34 million at the time. It also has exposure to the Geretsried project. Japan has real geothermal reasons to care. It has strong geothermal resources, but conventional development runs into national parks, hot spring interests, local opposition, land constraints and permitting complexity. A closed-loop system that avoids produced brine, reduces hot spring conflict and promises lower seismicity has political and strategic appeal. Chubu is not irrational for wanting the option.

But Chubu’s alternative to a negotiated agreement is strong. This is where Negotiation Genius, by Deepak Malhotra and Max Bazerman, is useful. It is the only negotiation book I routinely recommend because it focuses on preparation, incentives, value creation and deal structure instead of chest-thumping tactics. One of its central lessons is that bargaining power comes from the BATNA, the best alternative to a negotiated agreement. The party with the stronger alternative should not pay as though it has no choice.

Chubu’s BATNA is not “pay Eavor or abandon geothermal.” It can wait. It can observe Geretsried. It can learn from its existing investment and project exposure. It can work with Japanese industrial partners. Kajima, a major Japanese construction and engineering firm, has also invested in Eavor. Chubu can hire drilling contractors, engineering firms, geothermal consultants and former Eavor staff. It can explore ways to route around parts of Eavor’s IP. It can support other geothermal pathways, including conventional geothermal, enhanced geothermal, other closed-loop variants and supercritical geothermal research. It can let Eavor carry the cost of further proof before committing more capital.

That gives Chubu a strong negotiating hand. If Eavor wants a large upfront licence fee, Chubu can ask a simple question: for what? Not for thermosiphon. Not for ORC equipment. Not for horizontal drilling. Not for subsurface thermal modelling. Not for district heating integration. Not for a proven commercial reference plant. Maybe for Rock-Pipe, project data, failure lessons, patents and technical advice. Those have value. But they do not command platform-company pricing when the first full-scale platform has not performed.

This is where the contingent contract lesson from Negotiation Genius is directly applicable. When two parties disagree about the future, they can turn disagreement into deal structure. If Eavor believes its technology will perform and Chubu is skeptical, the contract should make payment contingent on performance. Eavor should receive documented engineering fees for actual work. It might receive a small access fee if the patents are unavoidable. It could earn success payments if agreed milestones are met: completed well pairs, completed loops, hydraulic performance, thermal output, low parasitic load, delivered heat, net MWh, availability and sustained performance over several seasons. If Eavor is confident, it should welcome upside tied to results. If it demands large upfront fees without performance obligations, that is itself a signal.

Eavor can still sell something. There is value in scar tissue. It can sell first-of-kind lessons. It can sell patents. It can sell design support. It can sell Rock-Pipe know-how. It can sell data from Geretsried. It can sell a permitting and public-acceptance narrative around no fracking, no produced brine and lower seismicity. It can sell the story that it has already made the mistakes others should avoid. But scar tissue is not the same as a bankable technology platform.

The evidence that would change my view is clear. Complete the Geretsried loop field. Show stable net-positive output near the original commercial claims. Deliver heat at scale into a real district heating system. Publish credible drilling cost reductions. Show that loops can be completed repeatedly without clogging. Show low parasitic load. Show sustained performance across seasons. Finance a second project with normal private-sector risk allocation. Find a customer that pays meaningful licence fees without requiring a heavy public subsidy scaffold or a rescue-style contingent deal. Those would be meaningful signals.

Until then, the most reasonable conclusion is that Eavor has moved from next generation infrastructure company to troubled IP and services company. That may improve its survival odds for a while because it reduces the amount of capital it must directly deploy. But it also weakens the core claim. The company that was supposed to prove closed-loop geothermal by building and operating the first commercial system is now stepping away from the very tasks that determine whether the technology works.

I do not dismiss the intelligence of the people involved or the importance of geothermal in general. I want more clean heat and power in the world. I want more serious subsurface innovation. But wanting a technology to work is not the same as seeing a commercial entity with a defensible moat, strong buyers and proven execution. Eavor’s current position looks weak. Its main IP appears narrow. Its first large-scale project has arguably failed badly. Eavor may struggle to find a credible party willing to take on completion and operations except on time-and-materials terms, with narrow liability and no output guarantee. Its likely best buyer has a strong BATNA. Its licence value should be contingent, discounted and negotiated hard. At this point, I do not rate Eavor’s chances of remaining a functioning commercial entity for much longer unless a strategic investor, public funder or distressed buyer decides that the optionality is still worth carrying.

CleanTechnica's Comment Policy

Share this story!

• Share on LinkedIn (Opens in new window) LinkedIn
• Share on WhatsApp (Opens in new window) WhatsApp
• Share on Facebook (Opens in new window) Facebook
• Share on Bluesky (Opens in new window) Bluesky
• Email a link to a friend (Opens in new window) Email
• Share on Reddit (Opens in new window) Reddit