Room-Temperature Single Photons, From a Dot in a Fiber

On 28 October 2026, the Los Alamos quantum-dot manufacturer UbiQD announced a $200,000 grant through New Mexico’s Quantum Technologies Award to prototype a fiber-coupled, room-temperature single-photon source built from its copper indium sulfide (CuInS₂) colloidal quantum dots. The money, from the state’s Economic Development Department through its Technology and Innovation Office, will fund optical test benches and cleanroom packaging at the company’s Los Alamos facility. The work builds on feasibility trials run with Los Alamos National Laboratory in 2024–2025 and on a fiber-coupled broadband light-source patent, US 12,085,443.

Read the announcement carefully and it says something narrower than the headline suggests. This is a prototype grant, not a device result. No purity figure, no emission rate, and no indistinguishability measurement has been published for this source, because no source exists yet to measure. The team will first measure how reliably its dots produce single photons, then refine the materials until they meet industry performance targets. What is being funded is the measurement program and the packaging — the two steps between a material that can emit single photons and a source that anyone can use.

That distinction is worth holding on to, because the idea itself is a good one. Room-temperature, fiber-integrated, solution-processed single-photon sources attack exactly the two objections that have kept colloidal quantum dots out of quantum hardware: cost and packaging. The physics behind the bet is real, and recent peer-reviewed results in this material class make it more than a slogan.

The problem: three metrics, two objections

A single-photon source is judged on three metrics, and a source that excels at one may fail at another. Purity asks whether the source emits exactly one photon per trigger, measured by the second-order correlation function g(2)(0); values below 0.5 indicate single-photon emission, and values approaching zero indicate high purity. Indistinguishability asks whether photons from different emission events are identical enough to interfere, tested by Hong–Ou–Mandel interference. Brightness asks how often a trigger actually yields a usable photon in the desired mode — collected into a fiber, not merely emitted in a random direction.

Epitaxial quantum dots in III–V semiconductors satisfy all three at cryogenic temperatures, and they remain the benchmark — but they require cooling to a few kelvin, molecular-beam epitaxy, and sophisticated nanofabrication to extract the light. Colloidal quantum dots, synthesized in solution and deposited from a beaker, invert the economics: cheap, solution-processable, size-tunable. Their problem has always been the physics — blinking, spectral diffusion, and broad emission lines — and room temperature makes all of it worse, since thermal energy at 300 K (~26 meV) drives phonon scattering and dephasing that degrade interference between successive photons. Any room-temperature source claim has to be read with the three metrics held separately.

The UbiQD announcement is framed, correctly, as an attack on cost and packaging rather than on the physics record. The company’s pitch is that nothing available today addresses all the needs at once: safe materials, room-temperature operation, tunable emission, fiber coupling, and reliability at scale. The first three are material properties. The last two are engineering. The grant funds the engineering.

The material: copper indium sulfide

UbiQD’s dots are I–III–VI semiconductors: copper indium sulfide, CuInS₂, with a zinc sulfide shell in the thick-shell variants studied in the literature. The composition matters for two reasons. First, it is free of cadmium, lead, and mercury — the heavy metals that dominate high-performance colloidal dots (cadmium selenide, lead halide perovskites) and that carry regulatory and handling burdens a manufacturer would rather avoid. Second, it is a ternary compound, and its photophysics is messier than that of a binary dot.

The relevant history starts at Los Alamos. Before joining UbiQD, the grant’s principal investigator, Nikolay Makarov, now the company’s Senior VP of R&D, co-authored a 2017 study from Los Alamos National Laboratory on thick-shell CuInS₂/ZnS quantum dots, published in Nano Letters. The paper explained why CuInS₂ ensemble emission is so broad — typically more than 300 meV — by doing spectroscopy on individual dots. With a thick ZnS shell, single dots showed strongly suppressed blinking and single-particle room-temperature linewidths down to about 60 meV, far narrower than the ensemble. The authors attributed the emission to recombination of a delocalized band-edge electron with a hole localized on a copper-related defect, and showed that random positioning of the emitting center within the dot could shift the emission energy by more than 300 meV. The announcement characterizes this 2017 work as showing that CuInS₂ dots can emit single photons at room temperature; the paper itself is a single-dot photophysics study, and the antibunching claim rests on the announcement’s characterization rather than on the paper’s headline results. That is a small distinction, but it is the kind that matters when a company announcement reaches back a decade for its scientific foundation.

The more direct evidence is much newer. In September 2026, a paper in Nano Letters reported room-temperature single-photon emission from individual gallium-doped multinary CuInS₂/ZnS core–shell quantum dots, with Hanbury Brown–Twiss measurements showing a minimum g(2)(0) of 0.06 — a single-photon purity of 94 percent — and an emission linewidth of 35 meV, with blinking strongly suppressed. That result is independent of UbiQD and its grant; it is a laboratory spectroscopy result, not a device. But it establishes the key premise the grant depends on: dots in this material class can emit single photons at room temperature with high measured purity. What it does not establish is indistinguishability — the paper reports no Hong–Ou–Mandel measurement — nor brightness in a packaged device, nor reproducibility across many dots.

There is a real tension in the material to keep in mind. The same defect-localized hole that gives CuInS₂ its broad tunability also gives it broad ensemble linewidths and a radiative mechanism intrinsically less clean than a band-edge exciton in a binary dot. Gallium doping and thick shells narrow the single-dot line, as the 2026 result shows, but the spread from dot to dot remains a manufacturing problem: a source built from one dot at a time needs every dot to behave, or a way to select the good ones.

The grant: what $200,000 buys

The funding mechanics are straightforward. The grant is non-dilutive state money through the New Mexico Quantum Technologies Award, administered by the Economic Development New Mexico Department’s Technology and Innovation Office. It is one of six $200,000 allocations in a $1.2 million round to regional quantum firms — the other recipients named in trade coverage are Mesa Quantum, Photon Queue, Bandelier Technologies, Mesa Photonics, and Conductor Quantum — and recipients must maintain active facilities and resident personnel in New Mexico for at least two years after performance. The scale is modest: trade coverage notes the grant beside a $20 million Series B round UbiQD raised in 2025 and a subsequent $6 million Silicon Valley Bank loan. Two hundred thousand dollars is seed-scale money for hardware, not a product budget; it should be read as the state buying a measurement campaign and a prototype, not a production line.

What the campaign consists of is described in unusual detail for a grant announcement. The money funds optical equipment — test benches for measuring single-photon emission — and device prototyping, including cleanroom packaging, at UbiQD’s Los Alamos headquarters. The sequence is explicit: first measure how reliably the dots produce single photons, then refine the materials until they meet industry performance targets. That is the right order, and it is also a quiet admission that the current reliability is unknown. The announcement quotes no target g(2)(0), rate, or efficiency, and neither should any coverage of it.

The groundwork is a 2024–2025 partnership with Los Alamos National Laboratory under the state’s Technology Readiness Gross Receipts (TRGR) Initiative, which the announcement says produced some of the first detailed measurements of this kind for UbiQD’s dots — single-dot-level characterization of the I–III–VI material’s photon statistics. That is the continuity the grant is buying: the same material, the same laboratory ecosystem, now with its own test benches instead of borrowed beamtime. UbiQD was founded in 2014 and is headquartered in Los Alamos; its founder and CEO, Hunter McDaniel, worked in quantum computing as an undergraduate researcher in Nobel laureate John Martinis’ group at UC Santa Barbara.

The intellectual-property position is part of the story the company wants told. UbiQD owns or exclusively licenses 17 granted US patents and 23 granted international patents, with 27 applications pending. The patent the announcement singles out is US 12,085,443, “Fiber-coupled Broadband Light Source,” also granted in Korea, Canada, the Netherlands, Germany, and France. Its claimed relevance is dual-use: faster connections between servers inside AI data centers, where electrical wiring is becoming a bottleneck, and quantum hardware.

Why fiber-coupling is the point

It is easy to miss why the fiber matters as much as the dot. In most single-photon experiments, the emitter sits in free space or in a cryostat, and the photon is collected by a microscope objective, aligned by hand, and routed through bulk optics. That works in a laboratory and fails as a product: every alignment is a failure point, and nothing about it scales to dozens or hundreds of sources.

A fiber-coupled source inverts the problem. If the dot is integrated directly into a standard single-mode fiber assembly — the approach the patent covers — the photon is born inside the mode it will travel in. There is no alignment to drift, no objective to stabilize, and the output is a connector, not an optical table. For quantum key distribution, where the photon’s job is to travel down a fiber anyway, that is the natural form factor. For photonic quantum computing, where many sources must feed one processor, it is the only form factor that plausibly scales.

The trade is that fiber integration makes everything about the emitter harder to optimize after the fact. A dot in a cryostat can be repositioned, re-excited, and re-filtered. A dot packaged into a fiber assembly is fixed; its brightness is whatever the coupling efficiency is, and its purity is whatever the material gives. That is why the grant’s sequencing — measure first, package second — is the correct order. The history of quantum-dot sources is full of emitters that looked excellent under a microscope objective and mediocre once packaged. The fiber is where that history gets tested.

What would change the picture

The honest way to cover a prototype grant is to list the measurements that would turn it into a result, in the order they matter.

First, a measured g(2)(0) on the actual prototype, reported with its excitation conditions and without background subtraction. The 2026 literature result of 0.06 was measured on individual research-grade dots under laboratory conditions; the question is what UbiQD’s packaged dots give. A value below 0.1 on a fiber-coupled device would be a genuine milestone for this material class. A value below 0.5 would establish single-photon emission and not much more.

Second, indistinguishability — a Hong–Ou–Mandel measurement, reported with its conditions. This is the measurement most likely to disappoint, because room-temperature phonon dephasing broadens the emission line far beyond the Fourier-transform limit, and no cavity or Purcell enhancement is mentioned in the announcement. If the program never reports indistinguishability, the source’s natural application is quantum key distribution, which tolerates moderate indistinguishability — not linear-optical quantum computing, which demands it.

Third, brightness in the fiber: what fraction of excitation pulses yields a photon in the fiber mode, and at what rate. A pure source that delivers one photon in a hundred triggers is a demonstration, not a component. No rate or efficiency figure exists yet, and none should be assumed.

Fourth, reproducibility: how many dots in a batch meet the target. Colloidal synthesis produces distributions, not clones, and the defect-localized emission mechanism in CuInS₂ makes the distribution wide. A prototype built from a hand-selected dot proves the physics; a process that yields good dots routinely proves the product.

Fifth, the emission wavelength and its tunability. The announcement describes tunable emission as one of the requirements, but gives no wavelength range for the planned source. For quantum communication, the interesting wavelengths are the telecom bands around 1.3 and 1.55 micrometres, where fiber loss is lowest; CuInS₂ emission is typically in the visible to near-infrared, and whether it can be pushed to telecom wavelengths is an open materials question.

For now, the contribution is bounded and clear. A Los Alamos company with a decade of quantum-dot manufacturing, a real patent position, and a recent laboratory result showing that its material class can emit high-purity single photons at room temperature has won modest state funding to find out whether that material can become a packaged, fiber-coupled source. The physics says it is possible. The engineering says it is not yet done. The $200,000 is the price of finding out which of those two statements matters more.

Sources and further reading

• UbiQD, Inc., “UbiQD Secures New Mexico Quantum Technologies Award to Advance Novel Light Source for Quantum Computing, Expands Patent Portfolio Strategy,” PR Newswire, 28 October 2026. Primary source for the grant amount, funding agency, work plan, TRGR partnership, patent portfolio figures, and patent US 12,085,443.
• Quantum Computing Report, “UbiQD Receives $200,000 New Mexico State Grant to Prototype Room-Temperature Quantum Dot Single-Photon Sources,” 28 October 2026. Trade coverage; source for the six-award, $1.2 million round, the other recipients, and the two-year New Mexico presence requirement.
• HPCwire, “UbiQD Receives New Mexico Grant to Develop Quantum Dot Single-Photon Source,” 28 October 2026. Republication of the announcement.
• H. Zang, H. Li, N. S. Makarov, K. A. Velizhanin, K. Wu, Y.-S. Park, and V. I. Klimov, “Thick-Shell CuInS₂/ZnS Quantum Dots with Suppressed ‘Blinking’ and Narrow Single-Particle Emission Line Widths,” Nano Letters 17, 1787–1795 (2017). Single-dot spectroscopy of CuInS₂/ZnS; source for the suppressed blinking, ~60 meV single-dot linewidths, and the defect-localized-hole emission model.
• “Narrow-Line-Width Room-Temperature Single-Photon Emission from Multinary Copper-Based Quantum Dots,” Nano Letters 26 (39), 13348–13356 (2026), published online 18 September 2026. Gallium-doped CuInS₂/ZnS core–shell dots; source for the g(2)(0) of 0.06, 94 percent purity, and 35 meV linewidth figures. Independent of UbiQD.
• Dealroom, “UbiQD lands $200K state grant to build single-photon source for quantum computing,” 28 October 2026. Source for the Series B and loan scale context.
• Review literature on single-photon sources and quantum-dot photonics, including reviews in Nature Photonics, Reviews of Modern Physics, and Nature Reviews Physics, covering purity, indistinguishability, and brightness metrics.
• Standard quantum optics references on the second-order correlation function g(2), photon antibunching, and Hong–Ou–Mandel interference.
• Note on verification: all grant, funding, personnel, and patent figures in this article are taken from the UbiQD announcement of 28 October 2026 and the trade coverage cited above. No purity, rate, indistinguishability, brightness, or wavelength figure is claimed for the planned UbiQD source, because none has been published. The g(2)(0) of 0.06 and 35 meV linewidth figures belong to the independent 2026 Nano Letters study, not to UbiQD’s device. The 2017 Nano Letters paper is described from its published abstract; the characterization of it as showing room-temperature single-photon emission from CuInS₂ dots is the announcement’s, not the paper’s headline claim.