Germany-China Team Runs Time-Bin QKD Over 120 km of Fiber for Six Hours, Using a Telecom Quantum Dot as the Single-Photon Source
Researchers led by Leibniz Hannover, Stuttgart, and Nanjing have published the first time-bin QKD link driven by an on-demand telecom quantum dot, holding stable across 120 km of standard fiber for six hours straight.
Overview
An international team led by Leibniz Universität Hannover, the University of Stuttgart, and Nanjing University has published the first quantum key distribution (QKD) experiment that combines an on-demand telecom-wavelength quantum dot single-photon source with time-bin encoding, running over 120 km of standard optical fiber for more than six hours of continuous operation. The work appeared in Light: Science & Applications on February 25, 2026, and was promoted internationally through a press release issued by the journal’s publisher on 15 March 2026, with broader tech-press pickup arriving in April and May.
The paper is the first to put a deterministic semiconductor single-photon source — rather than the attenuated laser pulses that dominate commercial QKD — into the long-distance time-bin protocol that has become the workhorse of fiber-based quantum networks.
What we know
The demonstration was run by a 14-author team. Jipeng Wang, Jingzhong Yang, Michael Zopf, and group leader Fei Ding are based at the Institut für Festkörperphysik at Leibniz Universität Hannover; Joscha Hanel, Zenghui Jiang, and Eddy Patrick Rugeramigabo are also at Hannover. Raphael Joos, Michael Jetter, Simone Luca Portalupi, and Peter Michler contributed from the University of Stuttgart. Xiao-Yu Cao and Hua-Lei Yin are at Nanjing University (Yin is also affiliated with Renmin University of China), and Lei Shan is at Anhui University, according to the PMC open-access copy of the paper.
The authors state that the work “provides the first experimental validation of integrating a QD single-photon source with time-bin encoding in a telecom-band QKD system,” as recorded in the published paper. The matching arXiv preprint, revised in late February before publication, adds that the system reaches “the highest secure key rate among the time-bin QKDs based on single-photon sources,” according to the arXiv abstract.
The numbers
Reported in the paper:
- Distance: 120 km of standard single-mode fiber, with fiber loss measured at 0.1956 dB/km.
- Source: An InGaAs/GaAs quantum dot emitting in the telecom C-band at a central wavelength of 1560.6 nm, with a measured emitter lifetime of 1018 ps and a second-order autocorrelation g²(0) of 0.85% — close to a single-photon ideal.
- Operating rate: 75.947 MHz pulse repetition.
- Channel quality: Quantum bit error rate of 6.85% in the Z basis and 9.60% in the X basis at the 120 km output, with an average below 11%, per the press release.
- Key rate: Approximately 15 secure bits per second under finite-key analysis, with a per-pulse secure key contribution of 2 × 10⁻⁷.
- Runtime: More than six hours of continuous operation without active stabilization, the property the authors lean on hardest.
- Detector chain: Superconducting nanowire detectors with 74% efficiency and a dark-count floor around 100 counts per second.
SciTechDaily notes a maximum achievable transmission distance of 127 km from the same set-up, though the headline figures are quoted at the 120 km operating point.
Why time-bin and why a quantum dot
Time-bin encoding stores each qubit value in the arrival time of a single photon relative to a reference pulse, rather than in polarization or phase. Because temporal separations survive long fiber runs better than polarization rotations, the protocol “offers intrinsic stability against such channel fluctuations even without any complex compensation protocols,” the authors write, in language reproduced by ScienceDaily from the published paper. The six-hour stable run is the practical payoff: most fiber-QKD experiments require frequent polarization recalibration.
The novelty is the photon source. Commercial fiber-based QKD systems almost universally use attenuated laser pulses with decoy-state protocols, which work but cap secure-key rates and introduce side-channel assumptions about multi-photon emission. Solid-state quantum dots emit at most one photon per excitation, making them attractive for QKD — but until now, telecom-band quantum dots with the brightness needed for intercity distance had not been paired with time-bin encoding. The press release frames the contribution succinctly: “Telecom-band QDs with Purcell enhancement can provide high-brightness photons suitable for intercity fiber communication, making them promising candidates for integration into practical QKD systems,” the authors say via EurekAlert.
What we don’t know
The demonstration is a controlled bench experiment over a 120 km fiber spool, not a deployed link. Whether the long-term stability the authors report transfers to outdoor dark fiber — with its temperature drifts and splice losses — is the standard next question for any QKD platform; the paper does not address field deployment beyond noting the system’s robustness as a step toward it.
The 15 bits per second secure-key rate is also modest in absolute terms — sufficient for the “encrypted text messaging” use case cited in coverage, but well below the kilobit-per-second rates that decoy-state laser QKD systems already hit over similar distances. The authors do not claim a rate advantage; they claim a platform advantage. Whether that advantage compounds — through Purcell-enhanced sources, higher repetition rates, or wavelength multiplexing — is left to future work.
Finally, the paper does not benchmark against device-independent or measurement-device-independent QKD protocols, which target stronger security guarantees at the cost of throughput. The 120 km result sits squarely in the prepare-and-measure family.
Context
The European push to put deterministic single-photon sources into QKD has been visible for several years, with the Hannover-Stuttgart partnership a recurring presence in the literature. The 2026 paper does not match the throughput of the best decoy-state systems based on attenuated laser pulses, but by closing the time-bin gap for solid-state emitters, it positions semiconductor quantum dots as a credible building block for future networks that may eventually have to coexist with quantum repeaters and entanglement-based protocols — both of which need single-photon, not laser-pulse, sources.
In parallel, India’s National Quantum Mission has been building out a longer but lower-rate deployed network with weak-coherent-pulse QKD hardware, as previously reported. The Hannover-Stuttgart-Nanjing result complements that line of work rather than competing with it: it advances the source-side physics rather than the system integration.