Sunlight Can Generate Quantum Entanglement, No Laser Required

In an outdoor enclosure at the Max Planck Institute for the Science of Light in Erlangen, Germany, behind fencing put up to blunt the wind, a large Fresnel lens rides on a motor that tracks the sun. The lens focuses sunlight onto the flat base of a cone-shaped piece of glass, and inside the taper the light bounces off the walls again and again, squeezing tighter as it slides toward the tip. At the tip it enters an optical fiber no thicker than a human hair, which carries the light into a darkened tent and onto a crystal with a clear aperture of a few millimeters. There, some of the sun’s photons split into pairs that are quantum mechanically entangled, and a new study says the process holds up under one of the strictest tests in physics.

The result, published August 6 in the journal Optica, is the first measurement of entanglement in light produced this way, from a collaboration between the Max Planck Institute for the Science of Light, the Max Planck Center for Extreme and Quantum Photonics, and the University of Ottawa. It undercuts a pair of assumptions that have kept quantum optics married to the laser.

They were not the first to feed sunlight to a nonlinear crystal. Earlier in 2026, a team at Xiamen University pumped down-conversion with sunlight and used the resulting position-correlated photon pairs for ghost imaging, reconstructing pictures with 90.7% visibility, not far off the 95.5% a laser managed at the same pump power, as SPIE reported in May. But that work never measured entanglement, and correlation is a weaker claim: ordinary classical light can be correlated. Demonstrating entanglement, and certifying it, is what the Erlangen and Ottawa researchers set out to do.

Scrambled in Space and Time, Orderly in Polarization

The process at work, spontaneous parametric down-conversion, or SPDC, sounds delicate: a photon from a pump beam enters the crystal, is absorbed, and reappears as two lower-energy photons whose properties stay linked however far apart they fly. Physicists had a principled reason to think the pump had to be a laser. A pump’s coherence in any degree of freedom, meaning how tightly its waves keep step in space, in time, or in polarization, sets a ceiling on the entanglement it can create in that same degree of freedom, and sunlight arrives thoroughly scrambled in space and time. But the ceiling applies property by property. Polarization is a separate account, and a polarizer can impose order there without disturbing the others, so nothing in principle blocks a scrambled but polarized pump from producing polarization-entangled pairs.

A Glass Cone Solved the Power Problem

The second objection was never about principle: sunlight is simply too diffuse to drive a nonlinear process inside a crystal a few millimeters wide. So the MPL group built a concentrator whose logic is entirely optical, gather wide, then squeeze. The system collects light over 1.4 square meters, compresses it in the glass cone through total internal reflection, and delivers it through the fiber to a crystal of periodically poled potassium titanyl phosphate sitting inside a polarization Sagnac interferometer, a loop of optics in which down-conversion running clockwise and counterclockwise becomes impossible to tell apart. That indistinguishability is what welds the daughter photons into a single entangled state.

Cheng Li, a PhD graduate from Robert Boyd’s group at the University of Ottawa, put the principle plainly. “The trick to harnessing sunlight is to keep different degrees of freedom of light from influencing each other during the process,” he says. “This means that sunlight is perfectly capable of generating entangled photons, as long as one can concentrate enough sunlight into a nonlinear crystal to induce SPDC.”

The Bell Test Passed by a Slim Margin

Photon pairs alone prove little, since ordinary correlated light can imitate some of their behavior, so the team characterized the two-photon state directly. Over three separate days they collected coincidences in two-minute blocks, noting how often the detectors fired within a nanosecond of each other: the counts clustered tightly at one time offset and fell to near zero elsewhere, meaning almost none were accidental pileups of unrelated photons. Reconstructing the state from measurements in sixteen polarization settings, a procedure called quantum state tomography, gave a fidelity of nearly 94% to the target Bell state, close to laser-pumped performance. The harder test measured correlations across polarization settings that no theory of local hidden variables can reproduce, the class of experiment honored by the 2022 Nobel Prize in Physics. The sunlight-pumped photons cleared the classical ceiling of 2 by 2.49 standard deviations, and their pair rate, normalized against the effective bandwidth of the nonlinear interaction, sat in the same range as laser-pumped sources.

That margin is thin. A 2.49 standard deviation violation clears the classical bound but sits far below five-sigma certainty, and the authors blame a low detection rate, nonmaximal entanglement from imperfect optics, and conditions nobody controls: seasonally low sun and drifting clouds. A laser experiment does not pause for weather.

Nor is the system unplugged. The crystal’s temperature control runs on electricity, the tracker’s motors run on electricity, and the authors state plainly that what they built is not a deployable architecture.

The Energy Arithmetic, and Its Asterisks

Even a proof of principle makes the energy case concrete. Lasers pay for their coherence with poor electrical-to-optical efficiency, stabilization electronics, and waste heat, and energy is starting to be treated as a figure of merit for quantum hardware rather than an afterthought, a cost one 2022 perspective called a blind spot; a RAND analysis projected on the order of 125 megawatts, admittedly a rough figure, to run a future code-breaking quantum computer. A pump that arrives free in the sky sidesteps that ledger. Even photovoltaics, the usual green workaround, convert less than half of sunlight at best, with the top confirmed research cell at 47.6%, so using sunlight directly as light keeps the energy in the form the experiment needs.

The strongest case is where sunlight is abundant and infrastructure is not. Satellites already traffic in entangled photons for cryptography, as China’s Micius mission showed by distributing entangled pairs to ground stations more than 1,200 kilometers apart, and a spacecraft in a Sun-synchronous orbit would find its pump overhead almost constantly. “Sunlight is an abundant and reliable resource in many environments, especially in space,” says Hanieh Fattahi, research group leader at MPL. “Being able to generate quantum-entangled photons directly from sunlight could enable simpler and more resilient quantum systems for satellites and future deep-space missions.”

The team treats the result as a beginning rather than a finish line: four-wave mixing and other nonlinear processes might also run on incoherent light, and better collectors, wider bandwidth, and cleaner optics could close the remaining gap with lasers. “The best part of this research is that it is only a beginning,” Boyd says. Whether the approach survives radiation tolerance, thermal management, and the mass budget of a real spacecraft is untested; for now the claim rests on a cone of glass, a fiber the width of a hair, and three days of whatever sun Erlangen offered.

FAQ