A University of Oslo model says a mirror removed mid reflection leaves a superposition of photon counts behind, not a smaller photon, with the exact distribution set by how fast the mirror vanishes.
Imagine a single photon, a packet of light, traveling through empty space toward a mirror. The front half of its wavefunction hits the mirror and bounces back. The mirror is then abruptly removed before the back half arrives. The back half has nothing to reflect off. It passes through, unmoored, into the space where the mirror used to be. The electromagnetic field has been disturbed in a way that did not exist a moment before.
What emerges is not half a photon. It is a superposition of possibilities, containing one photon, two photons, or many, with the specific count weighted by how fast the mirror was removed, according to the Skaar group's model accepted to Physical Review Letters. In the limit of infinitely fast removal, the math produces an infinite number of new photons. At finite, physical speeds, the result is a distribution of photon counts skewed toward smaller numbers.
This is the thought experiment at the center of a new theoretical model from physicist Johannes Skaar and colleagues at the University of Oslo. Their paper, accepted to Physical Review Letters, treats photons as fundamental particles that cannot literally be cleaved in half. What can be interrupted is the photon's wavefunction, the mathematical description of where the photon could be detected. The team asks what happens when that interruption is timed to catch the wave in the middle of an interaction with a mirror.
The mechanism borrows from a known corner of quantum optics. Moving a mirror in empty space, it turns out, is not a free action. The electromagnetic field in even a perfect vacuum carries a small, persistent energy. Accelerate a mirror fast enough, and that energy can be tapped, materializing as real photons, a phenomenon called the dynamical Casimir effect. Skaar's model is a sharper version of the same idea. Rather than oscillating a mirror, the team considered what happens when a mirror is removed entirely, mid-reflection, while part of a photon's wave is still on its way to the surface.
The result is a perspective-dependent prediction that Skaar himself flags as the strangest part of the work. An observer who can see both sides of where the mirror used to be records a messy, multi-photon outcome. An observer with access to only one side records a single photon, or nothing at all. The mathematics admits both stories at once. "It is a bit strange by everyday standards, but not actually that weird by quantum standards," Skaar told Science News.
Independent physicist Daniele Faccio of the University of Glasgow had a more confrontational first reaction. "My first impression is that this is nonsense," he told Science News. Reading the paper changed his read. The technique, he concluded, is legitimate, and the result is the kind of foundational quantum-optics calculation that could matter for quantum sensing, the same broad class of work that powers the quantum squeezing used in gravitational-wave detectors. His caveat: the result is a theoretical model, not a measurement, and treating it as a near-term application for any specific detector would be premature.
A second-order caveat matters here. The "infinite photons" framing is the limit case, the answer the math gives when mirror removal is treated as instantaneous, an unphysical idealization. Real mirrors cannot be removed infinitely fast. Finite speeds produce superpositions of several photons, sometimes "a bunch," Skaar said, with the precise distribution set by the speed. Promoting the infinite case as the realistic prediction, or describing the model as if it had been observed in a lab, would misstate the work.
What the model does establish is a clean theoretical bridge between two familiar quantum effects. The dynamical Casimir effect shows that motion in vacuum can create light. The Skaar model shows that the sudden absence of a mirror, at exactly the wrong moment, can also create light, and that the count of new photons depends on the geometry of the disturbance and on where the observer stands. Skaar's group plans to extend the analysis to other wave-like particles, starting with electrons.
By the standards of a kitchen, the prediction is strange. By the standards of quantum mechanics, it is the same kind of thing the theory has always said: that what an observer measures depends on what the observer can see.