Eurofiber completes entanglement-ready dark fiber loop between Delft and Rotterdam

Eurofiber and Delft Networks have completed the entanglement-ready dark fiber loop between Delft and Rotterdam first reported by Data Center Dynamics, a project the two companies present as a testbed rather than a service.

Dark fiber is unlit glass, capacity laid before anyone decides what runs over it, and here no customer is named riding it. According to Eurofiber, the loop offers a setting to test long-range quantum connections over already-deployed fiber and to evaluate entanglement distribution across operational networks for performance, stability and scalability. The work draws on Eurofiber's fiber infrastructure and Delft Networks' quantum networking expertise, and Eurofiber calls it a significant step toward a future quantum internet.

Eurofiber lists the constraints it expects to meet: attenuation, synchronization between distant nodes, and integration into carrier-grade infrastructure. Entanglement, as the company describes it, generates secure correlations and fast coordination between far-apart nodes, the property that would let quantum computers and sensors be joined into a network. Distributed quantum computing, ultra-precise sensing and time synchronization using quantum clocks are the applications it points to. Whether any of them become demand that pays for fiber is what the loop is built to inform.

Marc Hulzebos, Eurofiber's innovation manager, said: "Our project demonstrates how existing fiber infrastructure can evolve to support the next generation of quantum technologies." He added that the company is preparing its infrastructure for a future in which quantum devices are integral to communication and computation. Ronald Hanson, chief executive of Delft Networks, said deploying entanglement over operational telecom networks lets the company understand and address practical constraints and accelerate the shift from research to scalable deployment.

The commercial question for a fiber operator is what the glass is for. Every additional payload a metro route can carry makes the same duct work harder, and quantum channels would be one more, with the difference that the traffic they carry has no price today. Eurofiber frames the combination of quantum transmission and deployed fiber as a way to explore how existing telecom networks can be augmented for what it calls fundamentally new levels of privacy, trust and coordination. It is a claim about capability rather than about billings.

The announcement does not say what the project cost, how much fiber the loop covers, when results will be published, or who funded it. The language around it is the language of a commercial build — deployment, completion, carrier-grade — while the items an underwriter prices are absent: length, capacity, contract, counterparty. For a research loop the omission is unremarkable. Read as a fiber build, there is nothing to underwrite.

The cost structure carries the logic. The deployment rides on Eurofiber's existing fiber, so the incremental spending is on equipment and engineering at each end rather than on the civil works of laying new glass. An operator that already owns the glass can hold a long-dated technical option cheaply, and the option is on demand that does not exist yet: quantum memories separated by metropolitan distances are, in Eurofiber's framing, the prerequisite for everything downstream.

Eurofiber's balance sheet is not the constraint. In August the company locked in a €2.2bn sustainability-linked facility that replaced a €1.5bn platform, a financing that tested whether open digital infrastructure can hold infrastructure pricing without a hyperscaler anchor. A quantum testbed neither answers nor aggravates that test; it adds a technical claim to the asset base rather than a contracted cash flow, and the announcement offers no framework for what entanglement readiness is worth to a lender.

The market has been moving the other way, toward assets anchored by a named tenant. Tenant quality, not megawatts, now sets the cost of capital. September supplied the corroboration: a £2.45bn package from 13 banks for Virtus Data Centres, including £1.2bn of green capex, funded a 78MW campus at Saunderton with no named tenant against the megawatts. Set against that, a two-city quantum loop running on owned fiber is likely a small fraction of the capital committed to a single campus, which is why an operator can carry the experiment for years while the market forms.

Whether any of this becomes a fiber product depends on whether entanglement distribution can run alongside paying traffic. If it can, the capability sits on top of glass that metro operators already own; if it cannot, it is a dedicated-network business with a smaller addressable market and a longer build. Eurofiber's own list of constraints is where that question gets settled, and Delft Networks says the point of the project is to compress the distance between research and scalable deployment. No results have been reported from the loop, and no date is attached to that transition; the next useful item out of Delft and Rotterdam would be measured performance over a stated distance.