Can We Harness Zero-Point Energy? Inside Thorsten Ludwig’s Casimir Research

September 25, 2026

Can we turn the energy of empty space into electricity? For physicist Dr. Thorsten Ludwig, that question starts with a sphere just 200 micrometers across, a flexible cantilever, and a laser that reveals movements too small to see. In an interview with Tim Ventura, Ludwig described the instruments he’s building to investigate Casimir forces, the semiconductor devices attracting attention to vacuum-energy research, and the patterned surfaces he’s testing for possible propulsion effects. His work offers a view of a field where genuine laboratory capabilities sit alongside ambitious, unresolved possibilities—and where measuring a force is only the beginning of learning how to use it.

A Niche Within a Niche

Ludwig describes practical vacuum-energy research as a “niche within a niche.” The larger Casimir community investigates quantum electromagnetic interactions, material properties, temperature effects, and increasingly precise force measurements. A smaller group asks whether those interactions can become useful components in energy and propulsion systems. Ludwig identifies Fabrizio Pinto, Garret Moddel, and Harold “Sonny” White among the researchers pursuing those applications. They don’t all use the same mechanism, and their projects aren’t at the same stage, but they share an interest in moving beyond measurement toward functioning technology.

His own background combines university research with independent experimentation. Ludwig earned his physics diploma in 2001 and doctorate in 2005 at the Technical University of Berlin, where he also worked as a research scientist. Since 2006, he has consulted on energy and propulsion projects internationally. His experience includes developing energy-research projects associated with the Technical University of Clausthal and coauthoring a study of unconventional energy technologies for the German government. His current experiments draw on that practical experience: finding components, adapting instruments, preparing surfaces, and testing whether a proposed interaction survives measurement.

His connection to the subject also includes meeting Hendrik Casimir himself. Ludwig recalls attending a conference in the late 1990s, although he’s uncertain about the exact year, where hundreds of researchers had gathered to discuss the effect bearing Casimir’s name. The encounter left an impression less because of scientific celebrity than because of the older physicist’s demeanor. Ludwig remembers a “very humble man,” delighted that so many people were exploring his work. He also recalls Casimir crediting a conversation with Niels Bohr for encouraging him to consider zero-point energy.

The scientific history reaches back to Philips’ Natuurkundig Laboratorium, or NatLab, in Eindhoven. In 1948, Casimir and Dirk Polder published their analysis of how finite light-propagation time changes interactions between atoms and between an atom and a conducting surface. Later that year, Casimir published his sole-authored paper predicting attraction between perfectly conducting plates. Ludwig emphasizes the industrial setting behind those discoveries: Casimir was thinking about practical material problems at Philips, not merely pursuing an abstract curiosity. A problem involving matter and surfaces became a lasting connection between quantum theory and engineering.

What Empty Space Can Make a Surface Do

The familiar Casimir illustration shows two uncharged conducting plates facing one another across a small gap. Their presence changes the electromagnetic fluctuations permitted by the field’s boundary conditions, producing an interaction that depends on separation. In the ideal parallel-plate configuration, that interaction is attractive. Ludwig explains it through waves that don’t fit between closely spaced surfaces: as the gap narrows, the available electromagnetic spectrum changes. It’s an accessible picture of the effect, although the glowing waves used in illustrations aren’t literal light emitted by stationary plates.

This isn’t ordinary ambient radio-frequency energy. Zero-point fluctuations belong to the quantum description of a field in its lowest-energy state, rather than a broadcast signal waiting to be intercepted. The ideal parallel-plate pressure varies with the inverse fourth power of distance, so halving the gap increases its magnitude sixteenfold. Calculated for perfect conductors at zero temperature, the pressure is approximately 0.0013 pascal at a one-micrometer separation and 13 pascals at 100 nanometers. Those are theoretical benchmarks, not specifications for Ludwig’s apparatus, but they show why the smallest positioning errors matter.

Ludwig stresses that materials deserve as much attention as distance. He describes van der Waals interactions through “dancing charges”: fluctuating dipoles that influence neighboring matter. He contrasts that picture with ideal metallic boundaries, where charges redistribute rapidly, and emphasizes that real metals are never perfect conductors. Their response changes with frequency, and sufficiently high-frequency electromagnetic radiation can penetrate them. For an experimenter, that means a cavity isn’t defined solely by its width. Its surfaces, coatings, and frequency-dependent electromagnetic properties help determine the interaction.

His material-centered explanation is one way of introducing the subject. The broader Casimir–Lifshitz framework treats conductors and dielectrics within a unified description that includes material response, temperature, and retardation—the consequences of finite propagation speed. It doesn’t impose a universal boundary where van der Waals forces end and Casimir forces begin. For Ludwig’s work, the practical consequence is clear: an unexpected signal cannot be interpreted from geometry alone. Measuring a force requires understanding the entire arrangement, and it doesn’t directly reveal how much of the vacuum’s theoretical energy could be converted into useful power.

A Laboratory Built Around a Tiny Sphere

Ludwig’s force sensor borrows an idea from atomic force microscopy but changes its purpose. A conventional AFM often uses a sharp tip to scan a surface and build an image. His Casimir arrangement replaces that tip with a sphere approximately 200 micrometers in diameter—one-fifth of a millimeter—glued to the cantilever. The sphere approaches a flat substrate while the instrument measures the interaction. Ludwig credits Umar Mohideen’s work for pioneering this kind of AFM-based Casimir measurement, which became an important alternative to larger, mechanically demanding apparatus.

The sphere is an alignment solution. Bringing two extended flat plates extremely close together requires exceptional control of their relative angle. A smooth sphere, by contrast, has no preferred rotational orientation. Ludwig gives the example of a 200-micrometer sphere approaching within 100 nanometers of a surface. At that separation, the gap is one-thousandth of the sphere’s radius, and the closest region appears nearly flat. The geometry preserves a manageable interaction while avoiding the challenge of keeping two entire planes parallel across a nanoscale gap.

Obtaining the sphere is easier than obtaining every property needed to model it. Ludwig uses commercially available microspheres, but their suppliers don’t necessarily characterize them for Casimir experiments. They may guarantee diameter while leaving important material details unspecified. He typically coats plastic spheres with gold or aluminum to create the metallic surface he wants. He has also experimented with larger objects, including ballpoint-pen balls, because increasing the sphere’s radius can strengthen the interaction at a fixed small gap. Their weight, however, makes them unsuitable for some delicate cantilevers.

The readout turns tiny bending into a measurable optical displacement. A laser reflects from the cantilever toward a detector, and changes in the cantilever’s angle move the reflected spot. Ludwig describes an optical path of roughly five centimeters, with a photodiode arrangement tracking the beam’s position. The experiment doesn’t need to image atoms or map surface topography. Its central task is simpler: change the sphere–surface separation, observe the cantilever, and determine how the response varies as the two objects approach.

The Watch Component That Became a Force Sensor

A second sensing approach begins with something found in ordinary electronics: a quartz tuning fork. Ludwig buys commercially manufactured resonators, opens their small packages, and adapts them for force measurements. The exposed devices are only a few millimeters in size. Their original application is timekeeping, but their stable mechanical resonance makes them interesting for an entirely different measurement problem. Instead of detecting the bending of a soft cantilever, an experimenter can look for a force-gradient-induced change in a resonator’s frequency.

The attraction is partly mechanical. A flexible cantilever is sensitive because it bends easily, but bringing it sufficiently close to an attractive surface can trigger a sudden jump into contact. Ludwig describes the resulting difficulty of pulling the probe away without breaking it. A quartz tuning fork offers a stiffer structure that can resist this instability while retaining a sensitive resonant response. That doesn’t make it immune to every experimental problem, but it changes the balance between mechanical compliance, stability, and readout sensitivity.

The trade-off is the complexity of measuring frequency. Ludwig explains that scanning through resonance takes time, especially when a small shift must be resolved accurately. Holding the drive at one frequency and watching the amplitude provides another option, but amplitude can change for reasons unrelated to the desired interaction. Phase measurements can also help, although they add their own demands. Even component procurement becomes less straightforward than it looks: packages sold for equivalent electronic functions may contain resonators with different physical dimensions, something Ludwig discovers only after opening them.

Those considerations have pushed his latest work toward a custom, stripped-down laser-and-cantilever instrument. “I don’t need to make the picture,” he says. He wants a controlled scan of separation, not the full three-dimensional imaging capabilities of an AFM. Removing unnecessary functions improves access to the experimental region and reduces cost and complexity. His published sensor research, including the 2012 paper Quantum Field Energy Sensor based on the Casimir Effect, sets out that broader ambition: build practical instruments for investigating quantum-field-related effects before attempting more elaborate applications.

Robert Forward’s Battery and the Energy Accounting Problem

The question of converting Casimir interactions into electrical work has a history of its own. In 1984, Robert L. Forward proposed a vacuum-fluctuation battery built from a stack of charged conducting leaves. Charge of the same polarity would make the leaves repel one another electrically, opposing their Casimir attraction. By controlling the electrical repulsion so that it remained slightly weaker than the attraction, the stack could contract while adding energy to the electrical system. It was an engineering proposal for controlled conversion, not a report of a completed generator.

Forward’s design also included recharging. Increasing the electrostatic repulsion would force the leaves apart again, restoring their separation at an energy cost. That step is central to interpreting the device. A conservative, reversible cycle that returns the complete system to its original state doesn’t produce a net energy gain; real losses make the balance less favorable. The useful analogy is energy storage, not an endlessly replenished fuel supply. Casimir attraction can do work during an approach, but reopening the same arrangement must be included in the accounting.

Ludwig’s interest in the proposal led him to examine the quantities involved. In the interview, he describes idealized calculations in which an effective area of one square kilometer, brought to a 100-nanometer separation, corresponds to roughly 0.4 joule of interaction energy. Reducing the separation to one nanometer raises the figure to about 400,000 joules. With only one square meter of effective area at that one-nanometer gap, the figure falls back to approximately 0.4 joule. His point is the extraordinary sensitivity to geometry—and the difficulty of obtaining large usable energies from practical structures.

These are scaling calculations, not achievable-output guarantees. At nanometer separations, real-material and short-range effects make perfect-conductor extrapolations unreliable as engineering specifications. Maintaining uniform gaps over vast effective areas presents another obstacle. Ludwig also points to Pinto’s proposals for using lasers to change semiconductor properties instead of relying entirely on physical plate motion. Such approaches offer different control mechanisms, but their illumination, switching, and reset requirements still belong in the energy budget. The hard question isn’t simply whether a Casimir interaction exists; it’s whether a complete device can deliver useful net performance.

MicroSparc Moves the Discussion Onto a Wafer

White’s MicroSparc program approaches the problem through semiconductor fabrication. Casimir, Inc. has publicly presented wafers and images of repeated device structures, and Ludwig recognizes that distinction: this isn’t merely a sketch of a possible machine. The disclosed geometry contains opposing conductive structures with narrow pillars between them, features Ludwig compares to skyscraper-like needles. These structures are intended to create an electrical function within an engineered Casimir environment. The existence of fabricated hardware, however, is a separate question from demonstrating its claimed source of power.

The company’s May 2026 announcement described a five-by-five-millimeter chip designed to supply 1.5 volts at 25 microamps. Those figures correspond to 37.5 microwatts, placing the initial ambition in the world of ultra-low-power electronics rather than household electricity. The release identified tire-pressure monitors, embedded sensors, and wearables as early applications and announced a $12 million seed round led by Scout Ventures. These are company-reported specifications, funding details, and development objectives—not measurements Ludwig independently verified during the interview.

For him, the investment matters because sophisticated fabrication costs money. He contrasts his own purchase of approximately $20 cantilevers—a box of 30 becomes a $600 decision—with White’s ability to discuss custom designs directly with manufacturers. Purpose-built structures allow a researcher to explore geometries that aren’t available off the shelf. Ludwig welcomes that capability and the attention it brings to the field. His enthusiasm is directed toward the opportunity to perform better experiments, not a claim that he has inspected every result or established the commercial readiness of the device.

“What I’m a little bit missing,” he says, is experimental disclosure. He has seen the wafer images and close-up geometry, but not enough measurements to assess the amount of power being produced or the mechanism behind it. “I couldn’t say much because I just don’t know anything about it,” he explains when discussing the project’s quantitative status. That reservation sits alongside a clear desire to see White succeed. Ludwig’s position is neither dismissal nor endorsement of the power claim: fabricated prototypes deserve examination, and evaluating their significance requires accessible performance data.

Moddel’s Search for an Electrical Signature

Garret Moddel’s work at the University of Colorado Boulder offers a different approach, with results described in scientific publications. Ludwig says he has visited Moddel’s laboratory and had recent telephone conversations with him, but he also acknowledges that he hasn’t studied the displayed device or its data in detail. His comments therefore establish familiarity with the researcher and interest in the project, not an independent technical validation. Understanding the experiment requires going to Moddel’s papers.

The devices combine a metal–insulator–metal junction with an adjoining optical cavity. In the published architecture, nickel and palladium electrodes are separated by thin insulating material, with a transparent dielectric and aluminum mirror forming the cavity above. The research investigates whether changing that nearby electromagnetic environment changes electrical behavior. Unlike Forward’s contracting leaves, this is a solid-state structure whose proposed operation involves charge transport through extremely thin layers.

In the 2021 paper Optical-Cavity-Induced Current, Moddel and colleagues reported electrical output without an applied bias and consistent trends across more than 1,000 devices fabricated in 21 batches. One device with a 33-nanometer cavity yielded an inferred maximum output of 1.4 picowatts over a 0.02-square-micrometer active area. The researchers described eight categories of artifact tests and a four-hour short-circuit-current measurement without a decline. They interpreted the results through a proposed mechanism involving quantum fluctuations and changes in carrier injection.

The interpretation remains a distinct scientific question. In a 2024 analysis, G. Jordan Maclay argued that an idealized cavity calculation did not support the proposed mechanism, predicting different behavior for current direction and cavity-width dependence. He also called for a more complete model incorporating real material parameters and the entire device. That disagreement doesn’t establish an alternative explanation for every observation, but it identifies a substantive theoretical problem. The published record contains electrical measurements, a proposed account of their origin, and a published challenge to that account—not a single, settled demonstration of continuous vacuum-energy extraction.

Sawtooth Surfaces and the Search for Directional Forces

Ludwig’s own propulsion-related experiments center on work with Robert DeBiase, whom he explicitly credits for the ideas he’s testing. Tom Valone brought the two researchers together after DeBiase began looking for a way to fabricate a sawtooth structure with metal on one side of each groove and dielectric material on the other. The intended asymmetry changes the electromagnetic boundary conditions. Ludwig’s task includes turning that proposal into samples, arranging sensitive measurements, and determining whether the geometry produces the directional effects they’re seeking.

He has taken the idea down to his microspheres. Using heated optical gratings, he imprints grooves into the plastic before selectively metallizing the surface. The process demands a narrow temperature window: too cold and the grating won’t leave the desired pattern; too hot and it damages the sphere. Ludwig says he has built an entire setup around preparing these samples. Patterning also removes one advantage of a smooth sphere. Once the surface has grooves, it has an orientation, and rotational alignment becomes part of the experiment again.

The most encouraging observations he reports involve weighing assemblies in different orientations. Ludwig says reversing a structure produces a predictable, repeatable difference in the scale reading. He and DeBiase have also investigated rotation on water, although he is more cautious about that result. “I’m still struggling a little bit to replicate that,” he says, adding that he had begun another series of experiments that day. The interview doesn’t provide numerical effect sizes, a complete uncertainty budget, or enough environmental-control information to establish the cause of the reported weight differences.

Lateral Casimir forces themselves aren’t hypothetical: a 2002 experiment measured them between suitably corrugated surfaces. But a force between components is not automatically thrust on a complete isolated apparatus. Establishing propulsion requires accounting for the whole system and any momentum exchanged with its surroundings. Ludwig recognizes another important distinction when he discusses the appeal of a moving test device. “One thing is a force and the other is energy,” he says. A repeatable movement against a known load would reveal more than an orientation-dependent reading alone, though its energy source and mechanism would still need to be established.

Casimir Devices Don’t Have to Be Power Generators

The application landscape is already broader than the search for continuous electrical power. During the interview, Ventura refers to a catalogue of 47 Casimir-device research leads, including both constructed apparatus and proposed designs. That isn’t a list of 47 proven generators. Published examples show several other ways an interaction can become useful. In 2001, Ho Bun Chan, Federico Capasso, and colleagues demonstrated a micromechanical actuator in which attraction between a metallic spherical surface and a polysilicon plate produced torque, rotating the plate about thin torsional supports.

A more recent experiment at Purdue used three modified cantilevers to create a transistor-like mechanical system. Two outer cantilevers carried 70-micrometer spheres, while a thin central cantilever served as the intervening plate. Controlling its modulation allowed researchers to switch energy transfer between the outer elements, and active feedback enabled amplification. The resemblance to a transistor concerns control of an input-to-output interaction, not ordinary semiconductor conduction. The Casimir effect supplies coupling; the modulation and feedback provide the active control and energy needed for those functions.

Optical structures offer another route. In 2021, researchers at Chalmers demonstrated self-assembled cavities formed by charged metallic nanoflakes in liquid. Casimir attraction balanced electrostatic repulsion at separations of approximately 100–200 nanometers, creating pairs of mirrors that functioned as optical resonators. Introducing an appropriate material into the cavity produced hybrid light–matter states called polaritons. The useful result wasn’t a new fuel source. It was a structure assembled and stabilized by competing interactions, with an optical function that could be measured.

Even propulsion proposals need to be distinguished by mechanism. Maclay and Forward’s 2004 dynamical-Casimir spacecraft concept relies on radiation associated with deliberately accelerated boundaries and the recoil accompanying that radiation. It isn’t the same proposal as a passive arrangement of asymmetric surfaces, and the authors described extremely small thrust in the version they analyzed. Taken together, these examples reveal a field with several distinct engineering questions: how to measure an interaction, how to control it, how to use it for a device function, and whether any particular architecture can support an entirely new source of power.

The Experiment Ludwig Hasn’t Finished

The limitations of conventional Casimir-energy calculations helped lead Ludwig toward his most speculative proposal: testing whether human intention can influence Casimir-force measurements. He says he first proposed the idea in 2007 and draws inspiration from the Princeton Engineering Anomalies Research program. He interprets that work as evidence that intention can produce small statistical changes in random processes, discussing an effect on the order of 0.1 percent. Extending that interpretation to vacuum fluctuations is Ludwig’s hypothesis, not an experimentally established consequence of the earlier research.

He is unusually clear about what remains unfinished. After struggling to obtain funding, he decided to build the experiment himself. The Casimir apparatus is operating, he says, but participant interaction, computer problems, and changes to the experimental interface have delayed the intended study. “I don’t have the experimental data,” he states. The distinction is essential: he has constructed a force-measurement platform and intends to use it for a consciousness-related test, but he isn’t reporting a demonstrated intention-induced Casimir effect.

Testing the idea would require more than instrument sensitivity. Blinded conditions, randomized sessions, environmental monitoring, and an analysis specified before looking at the results would help distinguish a reproducible association from drift, expectation, or chance. Ludwig acknowledges that working with people is different from working with resonators: thoughts and intentions aren’t easy to stabilize or characterize. The enormous vacuum-energy estimates that motivate him also aren’t measurements of an accessible reserve. Even a reliable change in a force signal would leave separate questions about its cause, energetic significance, and practical use.

His immediate priorities are continuing the patterned-surface experiments and completing the intention study. Both return him to the hands-on work that defines the interview: preparing samples, opening component packages, controlling a gap, watching a reflected laser, and repeating a measurement. The possibility of harnessing zero-point energy gives that work its ambition, but it doesn’t settle its outcome. For Ludwig, the next step isn’t a sweeping declaration that empty space can power the world. It’s another experiment capable of showing what the vacuum-related interaction does—and what it doesn’t.