Entangled photons held across 38 miles of Maryland fiber. The test a real quantum network had to pass.

In early 2025, NIST researchers threaded pairs of entangled photons through 62 kilometers of ordinary commercial aerial fiber strung above Maryland's suburban streets, from the agency's Gaithersburg campus to the University of Maryland in College Park, and the fragile quantum link held. The result, now in the Journal of Optical Communications and Networking with the matching paper at arXiv:2601.11753, suggests the hard part of keeping a quantum state alive through real, deployed telecom fiber may no longer be the wall between lab demos and a working quantum network.

Entanglement, the "spooky action at a distance" Einstein spent decades trying to refute, is a pairing of two light particles such that measuring one instantly determines the result of measuring the other, no matter how far apart they are. That property is what makes entanglement useful. A quantum network built on it would let distant instruments share measurements rather than just data, in ways a classical internet cannot. The hitch has always been that real fiber, swaying in the wind, sitting in temperature swings, and getting stepped on by technicians, scrambles the polarization of light, the orientation of its electric field, which is the carrier of quantum information in this scheme. Lab-spooled fiber is short, calm, and temperature-controlled. A suburban aerial cable is none of those.

The Maryland run solved the polarization-drift problem with active feedback. The team tracked the polarization state at the receiver and sent correction signals back to the sender, which rotated the polarization of subsequent photons to compensate. The result is the technical core of the paper: a polarization-stabilized aerial fiber that holds entanglement over the kind of distance where a telecom company would actually deploy a customer link. Sixty-two kilometers is not a record for the lab. It is a record for the kind of fiber a real network would use.

What the achievement buys, in practice, is a working map of what entanglement links can do when they stop being fragile. The first application is long-baseline telescope interferometry: linking the light-collecting mirrors of two or more observatories so they act as one much larger instrument, sharp enough to image the surfaces of distant stars and the disks around black holes. The second is distributed quantum sensing, a dense web of entangled sensors that could pick up the faint strain signals of earthquakes, volcanic activity, and glacial movement, the way a phased array of radio dishes works today, but at sensitivities a classical sensor grid cannot reach. The third is networked quantum computing: small quantum processors spread across a campus or a region, linked by entanglement, behaving like one larger machine for drug and materials simulation. The fourth, and most often cited, is communications in which any eavesdropping attempt disturbs the quantum state and announces itself, rather than hiding in the noise.

The researchers themselves flag the limits. A single 62-kilometer run is not a network, and the polarization-stabilization hardware has to be made robust enough to survive the longer, less predictable fiber spans between cities. Repeaters that extend entanglement over hundreds of kilometers without destroying it remain an open engineering problem. So does the question of timing: synchronizing distant measurements well enough to use entanglement as a real sensor, not just a one-shot demonstration.

That is the practical gate. If polarization stabilization can be hardened for variable temperature, mechanical stress, and the patchwork ownership of long-haul fiber, the test in Maryland is the template for what comes next. If it cannot, the result is a beautiful milestone and a reminder that deployed fiber is a hostile environment for anything quantum. The team has a paper, a journal venue, and a route on the map. The next round of tests on real fiber, not lab spools, will determine which.