On September 4, 2026, a four-page preprint titled *A photonic source with half-a-GHz single-photon flux* appeared on arXiv (arXiv:2609.05387). Three days later, Sparrow Quantum's press release followed, and the quantum community took notice: for the first time, a deterministic single-photon source pushed more than half a billion photons per second into a single-mode fiber. The previous best publicly reported figure was 76 million per second.
What This Device Is For
Photonic quantum computing, quantum key distribution (QKD), and multi-photon interference experiments all share one common building block: a machine that emits photons one at a time.
Ordinary light sources emit "bundled" photons — a laser pointer sprays trillions per second, with unknown numbers packed into each pulse. Quantum protocols demand singletons: exactly one photon per pulse, because the mathematics of quantum interference only holds for single photons; two-photon contamination directly corrupts results.
The field has spent decades chasing one number: how many "clean" single photons per second can be produced. The stated ambition is interference experiments with ten to twenty photons — every added channel raises difficulty, and every channel needs feeding.
Speed vs. Efficiency: A Thirty-Year Trade-off
The record-setting route is the quantum dot: a nanoscale "artificial atom" grown in gallium arsenide that emits one photon per excitation. Two metrics have never improved together:
- Speed: how densely you can pulse. This dot's 220-ps emission lifetime permits up to ~4 GHz excitation in theory.
- Efficiency: how many photons can be captured into the fiber. Catching more than half from a nanoscale point source is elite.
- At 80 MHz: multi-photon contamination (g2) under 4%
- At 500 MHz: 3.63%
- At 1 GHz: 11.13%, with photon indistinguishability dropping from 82% to 55%
Historically these sat on a seesaw. A 2024 demonstration hit 2.5 GHz excitation but with fiber coupling of only a fraction of a percent (a few hundred thousand usable photons per second); efficiency-focused designs stalled around 76–80 MHz clocks. The 76-million-per-second mark — set by a 2024 intercity QKD experiment — was the highest reported fiber-coupled rate for a quantum-dot source. Rate times efficiency equals usable photons, and for thirty years the two factors never stood up simultaneously.
What the Danish Team Did
Sparrow Quantum, spun out of the Niels Bohr Institute in 2016, is led by founder Peter Lodahl (professor of quantum optics, Chief Quantum Officer). The paper has 13 authors — 11 from the company, 2 from Ruhr University Bochum, which supplied the epitaxial materials. First author Patrik Zahálka is an industrial PhD at the company. This is a chip company showing its work.
The technology has three engineering layers:
1. Excitation: Two cascaded electro-optic modulators carve a continuous laser into 40-ps pulses with >40 dB extinction ratio, FPGA-programmable up to 1 GHz, precisely kicking the quantum dot once per photon. 2. Collection: The dot sits in a photonic-crystal waveguide whose periodic holes act as a funnel, capturing the emission into the waveguide mode and onward into single-mode fiber. At the full 1 GHz rate, measured fiber efficiency is 51.5% — read directly with an ordinary commercial power meter. 3. No filtering: The paper's cleverest move is deliberately omitting spectral filters (which normally purify at the cost of lost photons). The team collected everything, reported a lower bound, and even counted phonon sideband light in the denominator. The measured fiber power of 110 pW, divided by the known photon energy, yields efficiency directly — skipping the superconducting-detector calibration chain entirely.
For scale, 110 pW is two billion times smaller than a button battery's output — invisible to any camera. But in the single-photon world, this brightness moves from "countable" (with expensive single-photon counters) to "measurable" (with a power meter). Lodahl's press-release quote: every improvement came from solid engineering, not redefining the technology; the source was driven to the emitter-physics limit while collecting every photon it produces, with quality preserved.
The Fine Print at the Record Point
"Quality preserved" holds at 500 MHz, but discounts at the record-setting 1 GHz full speed. The paper's own data:
A fair statement: quantity and quality coexist today at the 500 MHz setting — roughly 269 million photons per second into fiber with >96% purity. Even that tier alone is 3.5 times the old record.
Flux evolution (fiber-coupled, millions of photons per second): Sparrow 2023 product ~20; 2024 global record 76; this paper at 500 MHz 269; this paper at 1 GHz 515.
What Half a Billion Photons Per Second Can Do
The paper claims two uses: serving as an optical power standard for metrology, and time-multiplexing into multiple channels to feed 10–20-photon quantum simulation experiments. The arithmetic is vivid: a 515-million-per-second conveyor split ten ways gives ~50 million per channel — about two-thirds of the entire 2024 global best single source. A day's data collection shrinks to under an hour.
The press release sketches broader prospects — linear optical quantum computing, detector calibration, quantum networks — but these extend beyond the paper. And one caveat must be stated plainly: this is a September 4 preprint, not peer-reviewed; the company's own release labels it "the preprint."
A Small Company's Small Mountain
Sparrow was spun out of the Bohr Institute in 2016, raised €4.1M in seed funding in 2023, and €21.5M in a Series A in 2025 (extended to €27.5M by year-end, led by a Danish pension fund). Its 55+ person team sells the Sparrow Core single-photon source chip and Sparrow Nest integration platform. The new record is 25 times the company's own 2023 product spec — a chip company welding paper numbers into its product catalog year after year.
Lodahl named the next mountain in the press release: entangling multiple such sources to operate together "with the same level of control and reliability." The single-photon source is only the first building block of photonic quantum computing; once a half-billion-per-second supply line exists, the bottleneck shifts to entanglement distribution and multi-channel integration — a slower battle.
Records matter not for the record itself, but for pushing back the boundary of "impossible" one notch — and letting everyone see the new terrain.
---
Sources: arXiv:2609.05387, *A photonic source with half-a-GHz single-photon flux* (2026-09-04, v1 preprint, 13 authors: Sparrow Quantum ×11, Ruhr University Bochum ×2); Sparrow Quantum press release *Enough light, at last* (2026-09-07); The Quantum Insider coverage (2026-09-07); Sparrow Quantum website team and product pages.