Illustrative image: a 64-pixel superconducting nanowire single-photon detector (SNSPD) array. The Paderborn system uses eight SNSPDs multiplexed across 128 time modes. Image: NASA/JPL-Caltech, public domain (via Wikimedia Commons).
TL;DR: Photon counting 9000 photons explained in one line: researchers at Paderborn University built a detector that counts individual photons in a laser pulse from zero to more than 9,000, beating the Poisson noise limit by at least 4.1 dB. It uses eight superconducting nanowire detectors split across 1,024 time bins, and keeps sub-single-photon precision up to 276 photons per pulse.
Key Highlights
- Paderborn University researchers counted photons from 0 to over 9,000 per pulse — an order of magnitude beyond previous multiplexed single-photon counting systems.
- The detector beats the Poisson noise limit by at least 4.1 dB across that whole range, and measures pulses of up to 276 photons with an uncertainty smaller than a single photon.
- The trick: one pulse is split into 128 time-delayed slices and spread over eight superconducting nanowire single-photon detectors (SNSPDs), making 1,024 detection bins.
- A conventional multiplexed system would need about 37,000 detector elements to match the same precision.
- Published in the journal Optica on 12 November 2026 — the next step is testing it on bright quantum states of light, not just laser pulses.
Photon counting 9000 photons explained: what Paderborn actually built
Counting light sounds simple until you try to do it exactly. A photon is the smallest packet of light, and most detectors can only tell you whether light is present or absent — not how many photons arrived. That is fine for a camera, but quantum computing and precision sensing need the exact number.
The team at Paderborn University in Germany — led by Timon Schapeler, with Fabian Schlue, Isabell Mischke, Michael Stefszky, Benjamin Brecht, Christine Silberhorn and Tim J. Bartley — built a detector system that counts photons one by one, from zero all the way past 9,000 in a single pulse. Their paper, "Macroscopic photon counting beating the Poisson noise limit", was published in the peer-reviewed journal Optica on 12 November 2026. The underlying preprint has been on arXiv since 30 April 2026; the journal publication is the new, peer-reviewed result.
To test it, the researchers fired laser pulses at the device with average photon numbers ranging from one to 15,000 per pulse, and measured how precisely the detector could report the count. At a repetition rate of 80 kHz, the system can precisely count photons corresponding to an optical power of about 71 picowatts — bridging the gap between single-photon measurements and high-sensitivity optical power meters.
How the 1,024-bin detector works
The system combines two ideas that had each been studied on their own, but never together at this scale. First, superconducting nanowire single-photon detectors (SNSPDs): wires cooled until they superconduct, which snap out of superconductivity for a split instant when a photon hits, producing a measurable electrical pulse. SNSPDs are known for high efficiency, very high timing resolution and low noise — and the eight used here are themselves intrinsically photon-number-resolving.
Second, multiplexing: instead of asking one detector to swallow a whole pulse, the pulse is split into 128 smaller pulses separated in time, then spread across the eight detectors. That gives 128 × 8 = 1,024 individual detection bins, each holding only a small share of the original photons — few enough for the detectors to count accurately.
The third ingredient is a model. The team characterised every one of the 1,024 bins with a detector model developed with Mariia Sidorova from PTB, Germany's national metrology institute, so each bin's quirks are corrected for. They also performed quantum detector tomography to reconstruct the full measurement description of the device — a matrix with 1.38 × 108 elements.
A superconducting nanowire single-photon detector built for NASA's Deep Space Optical Communications, coupled to the 200-inch Hale Telescope. SNSPDs are the detector technology at the heart of the Paderborn result. Image: NASA/JPL-Caltech, public domain (via Wikimedia Commons).
What "beating the Poisson limit" means
Light from an ordinary laser arrives with built-in randomness: even a perfectly steady laser produces pulses whose photon numbers fluctuate. That fluctuation is called Poisson noise, and it sets a floor on how precisely any ordinary measurement can know the photon number. Beating that limit means the detector is extracting more information than the noise of the light itself would normally allow.
| Metric | Previous multiplexed systems | Paderborn detector (Optica, 2026) |
|---|---|---|
| Photons counted per pulse | Far below 9,000 | 0 to over 9,000 |
| Precision vs Poisson limit | At or above the limit | Beats it by at least 4.1 dB across the range |
| Sub-single-photon precision | Only for small photon numbers | Up to 276 photons per pulse |
| Detector elements needed | ~37,000 for the same precision | 1,024 bins (8 SNSPDs × 128 time modes) |
| Optical power counted | Single-photon regime only | Down to ~71 pW at 80 kHz |
Two numbers in that table deserve care. The headline range — 0 to over 9,000 photons — is the range the detector can count across. The sub-single-photon precision figure — uncertainty below one photon — holds up to 276 photons per pulse, not all the way to 9,000. Both are real results; they describe different things, and the paper reports them separately.
Why it matters for quantum computing
"Large-scale measurement of individual photons is a cornerstone of photonic quantum computing and single-photon-level imaging," said Tim Bartley from Paderborn University. Photonic quantum computers encode information in light, so a machine that cannot measure its own light precisely cannot check its own answers. A detector that counts thousands of photons accurately is a piece of infrastructure those machines need as they scale up — the same logic applies to ultra-sensitive cameras that work photon by photon.
For India, the result lands in familiar territory: the National Quantum Mission is funding quantum computing and sensing work across Indian institutes, and photonic hardware is one of its tracks. Paderborn's detector is not an Indian project, and no Indian lab is involved — but the measurement problem it solves is the same one Indian quantum-photonics teams face, and the paper is openly available for them to build on.
The honest limit: so far the system has only been tested on laser pulses. Laser light is the easy case — its photons arrive independently. The team's stated next step is to apply the detector to bright quantum states of light, where photons are correlated and the quantum effects the technology is ultimately for actually show up. Until that test, this is a measurement advance, not a quantum computer.
Should you care?
Verdict: If you follow quantum computing, yes — this is a genuine, peer-reviewed measurement record that removes a real bottleneck for photonic machines. If you are waiting for a quantum computer you can use, no — this is lab infrastructure, and the quantum-state tests that matter most have not happened yet. Worth knowing about; not worth overhyping.
What happens next
Three things to watch. First, the bright-quantum-state tests: if the detector keeps its precision on entangled or squeezed light, its value jumps. Second, whether other labs reproduce the 4.1 dB figure — independent replication is the standard step after any record claim, and none has been reported yet. Third, integration: SNSPDs need cryogenic cooling, so the practical question is how this 1,024-bin design fits into a real photonic quantum computer rather than a bench experiment.
FAQs
What is photon counting?
Photon counting is measuring light photon by photon — reporting the exact number of photons in a pulse, rather than just whether light is present. It is a cornerstone of quantum optics, because photonic quantum computers and ultra-sensitive sensors need to know precisely how much light they are dealing with.
How do you count 9,000 photons?
The Paderborn team splits one light pulse into 128 time-delayed slices and spreads them over eight superconducting nanowire single-photon detectors, creating 1,024 detection bins. Each bin receives only a few photons, which the detectors can count accurately, and a calibrated model recombines the bins into a total.
What is an SNSPD?
A superconducting nanowire single-photon detector is a wire cooled until it superconducts; a hitting photon briefly breaks the superconductivity and produces an electrical pulse. SNSPDs are valued for high efficiency, very precise timing and low noise, which is why they are used in quantum communication and deep-space optical links.
What is the Poisson noise limit?
It is the statistical floor on measurement precision set by the random arrival of photons in ordinary laser light. Even a perfect laser's pulses vary in photon number, and that variation — Poisson noise — limits how precisely the count can be known. Beating it by 4.1 dB means the Paderborn detector extracts more information than that floor normally allows.
Takeaway
Counting light exactly is hard, and it matters — photonic quantum computers cannot check their work without it. Paderborn's 1,024-bin detector now counts from zero to over 9,000 photons and beats the Poisson limit, a real record with a clear next test: bright quantum light.
Curious how the rest of quantum hardware is scaling? Read our explainer on Qunova's 80-qubit chemistry result, which matched the best classical calculation, and the Berlin proof that shallow quantum circuits can out-learn classical ones. For AI-designed quantum hardware, see DeepMind's AI-designed quantum programs.
Sources
Schapeler et al., "Macroscopic photon counting beating the Poisson noise limit", Optica (2026) — DOI: 10.1364/optica.604389
arXiv:2604.27761 — "Macroscopic photon counting beating the Poisson noise limit" (abstract, submitted 30 April 2026)
arXiv:2604.27761 — full text (HTML)
phys.org — "Multiplexed detector system precisely counts more than 9,000 photons" (12 November 2026)
AZoQuantum — "Researchers Extend Photon Counting Beyond Single Photons" (12 November 2026)
Comments
Post a Comment