NIST physicists have packed roughly 50 chips carrying 10,000 photonic circuits — each circuit outputting a unique wavelength — onto a single silicon wafer the size of a beer coaster. Each chip is fingernail-sized. The result, published in Nature on April 15, 2026, is a monolithic three-dimensional integration of two nonlinear optical materials on standard silicon, a fabrication route the researchers say points toward lasers at essentially any wavelength a designer specifies.

The problem this work attacks is narrow and well known to anyone specifying laser sources: high-quality, compact, efficient semiconductor lasers exist at only a few colors. Diode lasers perform well at 980 nm, just beyond human vision. But optical atomic clocks and quantum computers demand wavelengths matched to specific atomic transitions, and today those wavelengths come from lasers that are large, expensive and power-hungry — which is why these quantum instruments remain confined to a handful of specialized labs.

The atom-by-atom requirements are unforgiving. Rubidium, a workhorse atom in quantum computers and clocks, responds only to red light at 780 nm. Strontium, another popular choice, requires blue light at 461 nm. Shine any other color at these atoms and nothing happens. Every atomic species needs its own bespoke laser, which has been a major obstacle to moving optical clocks and quantum computers into the field.

Scott Papp's group at NIST, with Grant Brodnik as lead author and collaborators including Octave Photonics, a Louisville, Colorado startup founded by former NIST researchers, built the chip as a layered stack. The substrate is a standard silicon wafer coated with silicon dioxide (glass) and lithium niobate, a nonlinear material that converts light from one color to another. The researchers patterned metal electrodes onto the lithium niobate to electrically control that wavelength conversion, and built additional metal–lithium niobate interfaces that switch light on and off rapidly — the function needed for data processing and high-speed routing.

The key material advance is tantalum pentoxide, or tantala, deposited as the top layer. Tantala is strongly nonlinear: it can take in a single laser color and generate supercontinuum output spanning the full visible rainbow plus a wide range of infrared wavelengths. Papp's group spent years developing deposition techniques that pattern tantala circuits without heating, so the material can be laid down onto finished layers without damaging them. Because the researchers patterned the materials in a 3D stack rather than side by side, a single chip routes light efficiently between layers, combining tantala's wavelength conversion with lithium niobate's electrical controllability.

"The real power is that tantala can be added to existing circuitry," Brodnik said, describing the technique as allowing "seamless integration."

The economics follow from the wafer numbers. Ten thousand circuits per wafer, each producing a designer-chosen color through circuit design rather than through a different laser crystal or cavity, is the difference between a bespoke lab instrument and a photonic IC. "We can create all these different colors, just by designing circuits," Papp said.

The application set NIST names is broad. Portable, low-power optical clocks could aid earthquake and volcanic eruption prediction, provide a GPS-independent positioning and navigation alternative, and support searches for dark matter. Quantum computers could open new routes to simulating the physics and chemistry of drugs and materials. Beyond quantum, Papp sees the chips shuttling signals between the specialized processors tech firms deploy — potentially making AI tools more efficient — and improving virtual reality displays.

One caution for buyers and system integrators: these chips are not production-ready. The paper demonstrates a fabrication technique, not a product. Octave Photonics is now working to scale the process, and no yield, output power, linewidth or efficiency figures accompanied the announcement — the figures that will decide whether tantala-on-lithium-niobate circuits displace the bulky lasers now serving rubidium and strontium systems.

The open question is manufacturing adoption. If Octave Photonics can carry the low-temperature tantala process into volume foundry flows, wavelength-flexible laser sources become a design choice rather than a procurement constraint — and the bill of materials for field-deployed optical clocks shrinks accordingly.

Reference: G. M. Brodnik et al., "Monolithic 3D integration of tantalum pentoxide nonlinear photonics," Nature, published online April 15, 2026. DOI: 10.1038/s41586-026-10379-w.