Proof That Nothing Has Power: The Casimir Effect

The experiment that proved empty space carries energy — and the ones that showed the vacuum can push, pull, and even create light.

A Problem with Colloids

The story begins with glue.

In the 1940s, Hendrik Casimir was a theoretical physicist at the Philips Research Laboratory in Eindhoven, the Netherlands. He'd studied under Paul Ehrenfest at Leiden and spent eighteen months in Copenhagen with Niels Bohr — world-class training for a career that landed him in an industrial lab solving practical problems. One of those problems involved colloids: tiny particles suspended in liquid, the kind you find in paint, milk, and certain adhesives.

The particles in these colloids were supposed to be electrically neutral, yet they attracted each other more strongly than existing theories could explain. In 1947, Casimir and his colleague Dirk Polder set out to calculate the van der Waals force between two neutral atoms more carefully, accounting for the finite speed of light — what physicists call retardation effects.

The calculation worked. But then Casimir had a conversation that changed everything.

He discussed the result with Niels Bohr, who suggested it might have something to do with zero-point energy. Casimir went back and reformulated the problem from scratch — not as a force between atoms, but as a consequence of the quantum vacuum itself.

In 1948, he published a short paper predicting that two perfectly conducting, uncharged, parallel plates placed in a vacuum would experience an attractive force. Not from any charge or magnetism. Not from any classical effect. From the vacuum.

How It Works

The explanation is elegant.

Empty space is filled with electromagnetic vacuum fluctuations — zero-point oscillations of the electromagnetic field at every possible wavelength. In open space, all wavelengths are present. But between two conducting plates, there's a constraint: only wavelengths that fit between the plates as standing waves are allowed. A wavelength larger than twice the plate separation simply can't exist there.

This means the vacuum between the plates has fewer available modes than the vacuum outside. Fewer modes means lower energy density between the plates. The energy difference between inside and outside creates a net pressure — an inward push that drives the plates together.

Casimir calculated the force per unit area:

F/A = -(pi^2 * hbar * c) / (240 * d^4)

Where d is the plate separation. The force scales as the inverse fourth power of distance — it's vanishingly small at everyday separations, but at around 100 nanometers, it's roughly equivalent to one atmosphere of pressure. At the nanoscale, the vacuum pushes hard.

There's an important nuance here. In 2005, MIT physicist Robert Jaffe showed that the Casimir force can be calculated entirely without invoking zero-point energy — as a relativistic van der Waals force between the charges in the plates. Both approaches give identical results. The debate isn't about whether the force is real, but about the best way to think about why it's real. Most physicists consider the two descriptions complementary: different languages for the same physics.

The Long Wait for Measurement

Casimir's prediction sat for almost half a century before anyone measured it convincingly.

The first attempt came in 1958, when Marcus Sparnaay — working at the same Philips lab where Casimir had made his prediction — tried to measure the force between parallel plates. The result was, in Sparnaay's careful wording, "not inconsistent with Casimir's theory." In practice, his error bars were around 100%. He could see something was there, but he couldn't pin down the magnitude.

The problem was brutally practical. To measure the Casimir force with parallel plates, you need the plates perfectly flat, perfectly parallel, perfectly clean, and incredibly close together — all while avoiding every other force (electrostatic, gravitational, vibrational) that would swamp the signal. At the nanoscale separations required, even a single speck of dust is a mountain.

For nearly four decades, the Casimir effect remained a theoretical prediction everyone believed but nobody could measure precisely.

Lamoreaux's Breakthrough

In 1997, Steve Lamoreaux at the University of Washington found an elegant way around the parallel-plate problem: don't use parallel plates.

Lamoreaux used a torsion pendulum — a horizontal rod balanced on a thin wire — with a gold-coated spherical lens on one end and a flat gold-coated quartz plate nearby. The sphere-plate geometry eliminated the need for perfect parallelism. Only one point on the sphere is closest to the plate at any moment, and the theory for the Casimir force in this geometry was already worked out.

His apparatus measured the force at separations between 0.6 and 6 micrometers. The setup was, by the standards of modern physics, relatively simple. A torsion pendulum, some careful gold coatings, a lot of patience with vibration isolation.

The result: agreement with Casimir's prediction to within 5%.

The paper landed in Physical Review Letters in January 1997, made The New York Times, and immediately settled any remaining doubt. The Casimir effect was real, measurable, and quantitatively correct.

Lamoreaux went on to become the Eugene Higgins Professor of Physics at Yale, where he now leads the HAYSTAC dark matter experiment. His Casimir measurement remains one of the most beautiful experiments in modern physics — one measurement that turned a fifty-year-old theoretical curiosity into established fact.

Around the same time, Umar Mohideen and Anushree Roy at UC Riverside performed independent measurements with similar precision, confirming Lamoreaux's result.

Getting More Precise

The field accelerated after Lamoreaux.

In 2001, a team at the University of Padua became the first to successfully measure the Casimir force between true parallel plates using microresonators — finally achieving what Sparnaay had attempted in 1958.

In 2013, a collaboration between Hong Kong, Harvard, MIT, and Oak Ridge National Laboratory demonstrated something remarkable: a Casimir force measurement on a single silicon chip. Their compact integrated platform required no external alignment and enabled measurements with complex geometries that would have been impossible with traditional setups.

By 2021, the same group had pushed into truly exotic territory. They measured the Casimir force between interpenetrating nanoscale rectangular gratings — two comb-like structures sliding into each other — and found that the geometry dependence was 500 times stronger than the standard approximation predicted. The shape of the boundaries matters enormously, in ways we're still learning to calculate.

Today, the Casimir effect has been confirmed to better than 1% precision across multiple independent labs using different techniques. It's as well-established as any force measurement in physics.

When Nothing Pushes Back: Repulsive Casimir Forces

In 2009, Federico Capasso's group at Harvard published the cover story of Nature with a result that sounds impossible: they'd reversed the Casimir force. Instead of attraction, they measured repulsion. The vacuum was pushing objects apart.

The trick was choosing the right materials. The standard Casimir effect involves two identical metallic surfaces in vacuum. But the theoretical physicist Evgeny Lifshitz had predicted decades earlier that the force could change sign if you introduced a liquid between the plates with the right dielectric properties.

Capasso's team immersed a gold sphere and a silica plate in bromobenzene. In this configuration, the liquid has a dielectric response intermediate between gold and silica. The result: the vacuum fluctuations in the liquid push the surfaces apart rather than pulling them together. They measured repulsive forces of tens of piconewtons at separations up to 40 nanometers.

This wasn't just a curiosity. In nanotechnology, the attractive Casimir force causes a serious problem called stiction — tiny machine parts snap together and can't be separated. A repulsive Casimir effect could provide the "quantum lubrication" that keeps nanomachines running.

More fundamentally, it showed that the sign of the vacuum force depends on material properties and geometry. The vacuum isn't a fixed background — it's a responsive medium whose behavior can be engineered.

Creating Light from Nothing

If the Casimir effect proved that vacuum fluctuations produce forces, the dynamic Casimir effect proved something even more startling: vacuum fluctuations can produce real particles.

The idea is simple in principle. Virtual photon pairs constantly pop in and out of existence in the vacuum. Normally they annihilate each other before they can become real. But if a mirror moves fast enough, it can catch a virtual photon pair at just the right moment, tearing them apart before they can recombine. The virtual photons become real.

There's a catch. "Fast enough" means a significant fraction of the speed of light. No physical mirror can move that fast.

In 2011, a team led by Christopher Wilson at Chalmers University in Sweden found a way around the problem. They used a SQUID — a superconducting quantum interference device — to create an "effective mirror" in a microwave transmission line. By modulating the electrical boundary condition at about 11 GHz, they simulated a mirror oscillating at roughly 5% of the speed of light.

Real microwave photons appeared. They came in correlated pairs, with frequencies adding up to the driving frequency — exactly as theory predicted. The quantum vacuum had been shaken hard enough to produce real, detectable light.

The result, published in Nature in November 2011, was the first observation of the dynamic Casimir effect. Subsequent confirmations came in 2013 using a Josephson metamaterial and in 2019 using an optical fiber.

Think about what this means. Empty space, agitated in the right way, produces real energy. Not metaphorical energy. Not theoretical energy. Photons you can count.

What This Means

The Casimir effect, in all its forms, establishes three things beyond any reasonable doubt.

First, vacuum energy is physically real. It produces forces between objects that have no classical explanation. This has been measured to better than 1% precision by multiple independent groups worldwide.

Second, the vacuum is not fixed. Its properties depend on boundaries, materials, and geometry. The same vacuum that attracts two gold plates will repel gold from silica in the right liquid. The vacuum state can be engineered.

Third, the vacuum can be driven to release real energy. The dynamic Casimir effect converts vacuum fluctuations into detectable photons under the right conditions. The energy of nothing can become the energy of something.

These are not speculative claims. They're established physics, backed by decades of increasingly precise experiments. They don't prove that vacuum energy can be harvested at useful scales — that's a much harder question. But they prove that the vacuum is an active, physical system with energy, forces, and responses to manipulation.

The question isn't whether vacuum energy exists. The question is whether we can do anything useful with it.

Next in the series: Can We Actually Use It? — Forward's vacuum battery, Puthoff's controversial claims, DARPA's investments, and the honest scorecard on extracting energy from nothing.