Atomic-photonic chips for quantum technologies
Quantum sensing, built in glass
We write a 3D chamber inside solid glass with light — and fill it with atoms that sense.
What QSENSATO makes
Atomic sensors measure magnetic fields, time, rotation and radio waves by watching how atoms respond to light. Every one of them needs a sealed, transparent chamber of atomic vapor. That chamber is what we make.
Custom all-glass atomic vapor cells. The internal geometry is written in 3D inside a transparent substrate, then etched hollow and filled — shape, size, cell material and gas species tailored to the sensor they go into.
Vapor cells integrated with photonic structures — laser-written waveguides and fibre interfaces — in the same glass chip. Photonic integration is claimed in the patent family; microfluidic channels alongside the cell are a potential application.
Full sensing systems built around QSENSATO cells and chips are on the roadmap from 2027. They are not sold today.
QSENSATO is also the exclusive Italian distributor for Twinleaf instruments (third-party products).
Laser-written vapor cells · LWVC
Our patented approach uses femtosecond laser irradiation followed by chemical etching (FLICE) to build a vapor cell inside a single block of glass — maskless, three-dimensional, and based on the same femtosecond laser writing that inscribes optical waveguides.
01
Femtosecond pulses are focused inside fused silica. Non-linear absorption modifies the glass only at the focal spot. The laser traces the surfaces of the chamber, layer by layer, bottom to top.
The glass is modified, not removed — it is still solid after writing.
02
The block goes into hydrofluoric acid. The laser-modified glass etches far faster than pristine glass, so the acid preferentially hollows out the written pattern — entering through the reservoir’s open face.
≈18 h (2022) and ≈10 h (2024) at 35 °C, 20 % HF.
03
Optical facets are polished so probe light can pass cleanly through the glass into the sensing channel.
Residual roughness of the inner input/output facets: 50–100 nm RMS (2022).
04
A rubidium dispenser pill is placed in the reservoir; a glass plate is bonded over the open face in a nitrogen atmosphere.
Compatible with other bonding methods for harsher requirements.
05
Heating the getter with a laser releases rubidium and absorbs excess nitrogen — no vacuum apparatus needed. Vapor diffuses into the channel.
Buffer-gas pressure can in principle be set by the activation step.
Process as published in Lucivero et al., Optics Express 30, 27149 (2022) and Zanoni et al., J. Appl. Phys. 136, 144401 (2024) .
1 · Pump
Circularly polarized light tuned to rubidium spin-polarizes the atoms along the beam. Polarized atoms absorb less, so more light gets through.
2 · Precess
A magnetic field across the beam makes the spins rotate (precess) away from the beam axis. The bigger the field, the less polarization survives.
3 · Read
Less polarization means more absorption, so the transmitted light dips. Modulating the field and demodulating with a lock-in turns the dip into a signed, linear signal near zero field. This is a single-beam zero-field-resonance magnetometer, operated inside a magnetic shield.
3D versatility
Conventional microfabricated (MEMS) cells are built by bonding planar silicon and glass wafers, which limits 3D shape and optical access. Laser writing places chambers, reservoirs and conduits where the sensor needs them — at chosen depths in the glass.
500 × 500 µm channel, 9 mm long, 750 µm below the surface; five 250 µm conduits keep the reservoir and its residue away from the optical path. ≈1 pT/√Hz at 10 Hz (lab set-up, 96 °C, four-layer magnetic shield).
Zanoni et al., J. Appl. Phys. 136, 144401 (2024)Chambers that light can enter along more than one axis (patent claims 2, 7). Planar MEMS cells are limited here.
Buried channels, reservoirs and connecting conduits at different depths in one block (2022, 2024).
Laser-written waveguides share the same fabrication technique, giving a route to monolithic atom + photonics chips.
The same process is a mature microfluidics technology; channels for liquid samples can sit sub-mm from the atoms.
Atomic-photonic chips
The femtosecond laser that writes the vapor cell can also write optical waveguides in the same glass. Light can then be guided to the atoms, interact, and be guided out — one monolithic chip instead of a cell glued onto a photonic circuit.
Today’s state of the art
Hybrid assemblies: a separately microfabricated cell placed on top of a photonic chip, with gratings or mode converters coupling light in and out — often at low efficiency.
Laser-written route
Cell and waveguides written in one substrate; fibres attached directly. Claimed in the patent as an “integrated atomic/photonic device”.
In developmentOne fabrication platform
The same fabrication process — not the same cell — serves different sensors. Each architecture needs its own geometry, gas fill and surrounding optics.
Circularly polarized light spin-polarizes the atoms; a magnetic field makes the spins precess; the transmitted light (2024) or its polarization rotation (2022) changes.
Demonstrated in LWVCs (2022, 2024).
Applications
Each application starts with a physical quantity. The vapor cell is the part of the sensor that turns it into a change in light. All four families below are potential applications of the platform: they describe what laser-written cells can enable, not products QSENSATO sells today.
Femto- to picotesla magnetic fields from the brain, heart and muscles
Technology enablement, not a medical device. QSENSATO does not sell approved clinical products.
Time, rotation, RF electric fields and magnetic anomalies
ESA OSIP and ESA BIC are funded development projects (2026), not deployments.
Atomic transitions as absolute frequency references
Fields from microlitre samples, nanoparticles and bacteria
LWVC + microfluidics on one chip is a potential application; not yet published.
Custom development
QSENSATO designs vapor cells around the sensor: shape, size, cell material, gas species and integration. Sketch your requirements below and send them to the team.
Science
Laser-written vapor cells were born from a collaboration between ICFO (Barcelona) and CNR-IFN (Milan). QSENSATO was founded to take them out of the lab.
Optics Express 30, 27149 · 2022
V. G. Lucivero, A. Zanoni, G. Corrielli, R. Osellame, M. W. Mitchell
J. Appl. Phys. 136, 144401 · 2024
A. Zanoni, K. Mouloudakis, M. C. D. Tayler, G. Corrielli, R. Osellame, M. W. Mitchell, V. G. Lucivero
ICFO (Barcelona) and CNR-IFN (Milan) launch the research behind laser-written vapor cells.
Optics Express: femtosecond-laser-written cell; sub-Doppler spectroscopy and single-beam magnetometry.
EP application filed 16 Feb 2022 (ICREA, ICFO, CNR, Politecnico di Milano).
J. Appl. Phys.: 500 µm channel, 1 pT/√Hz at 10 Hz magnetometer.
Spin-off of the University of Bari Aldo Moro; exclusive licence to the patent family. US patent granted (Dec).
€500k pre-seed (LIFTT, Quantum Italia); rapid prototyping; EP patent granted (May); first public projects.
Prototype vapor cells for sensors, MVP, business validation. €1M SAFE; in-house fab at TecnoPolis under construction.
Develop complete sensors; grow the team.
Enter the global sensing market.
Intellectual property
“An atomic vapor cell, an integrated atomic/photonic device and apparatus comprising the atomic vapor cell, and a method for fabricating an atomic vapor cell”
| Jurisdiction | Number | Status |
|---|---|---|
| Europe | EP4231032B1 ↗ | Granted · 14 May 2025 |
| United States | US12181541B2 ↗ | Granted · 31 Dec 2024 |
| China | CN116605827A ↗ | Application · pending |
Status from Google Patents (October 2026); not a legal opinion.
View patent ↗Laser writing the chamber inside a host substrate — buried or open, planar or 3D — plus reservoir, connecting channels, etching through an open end, filling and sealing.
Laser-written chambers with optical access along one or more axes, with buffer gas or wall treatments against depolarising collisions.
The cell combined with waveguide-based photonic components written into, or slotted into, the same substrate; fibre-coupled variants.
Spectroscopy systems, frequency references, atomic clocks, OPMs, magnetic microscopes, spin-based sensors, gyroscopes.
Commercial status · October 2026
We are in our go-to-market year: prototype vapor cells for sensors, a minimum viable product and business validation. An in-house fabrication facility at TecnoPolis (Valenzano, Bari) is expected to be operational by the end of 2026.
Laser-written cells designed for your sensor, built with you during our 2026 go-to-market phase.
Atomic-photonic integration and sensor architectures developed together with research and industrial partners.
Joint projects with universities, metrology institutes and space programmes.
Future roadmap Complete, off-the-shelf QSENSATO sensors are not available yet; they are planned from 2027.
Momentum
Equity, SAFE, public funding and grants are different instruments and are not added together.
Team
Founder-led deep tech from Bari: one of the cell’s inventors, specialists who machine glass with light, and the people turning both into components.
CEO & Founder
University of Bari Aldo Moro
First author (2022) and senior, corresponding author (2024) of the LWVC papers; co-inventor.
Co-Founder
Associate Professor, Politecnico di Bari
Co-Founder & Tech Advisor
Board: Vito Giovanni Lucivero · Annalisa Volpe · Vincenzo Mazzilli · Marco Cravetto (LIFTT) · Simone Valorani (Quantum Italia). Scientific roots: Prof. Morgan W. Mitchell (ICFO), Dr Roberto Osellame and Dr Giacomo Corrielli (CNR-IFN) and Andrea Zanoni — co-inventors of the patent with Prof. Lucivero.
Ecosystem
Contact
Researchers, sensor builders and industrial teams: describe your application and we’ll tell you honestly whether a laser-written cell can help — and what is available now versus in development.
QSENSATO S.R.L.
Via Giovanni Amendola 173
70126 Bari, Italy
VAT 08949670726 · Spin-off of the University of Bari Aldo Moro