An optically pumped black-phosphorus device produced narrowband mid-infrared emission at room temperature, with a full width at half maximum—the width of a spectral peak at half its maximum intensity—below 3 nanometres under 1064-nanometre excitation, according to an arXiv preprint. The reported signal is consistent with distributed-feedback, or DFB, lasing, in which a grating supplies optical feedback, from black-phosphorus flakes integrated on patterned silicon-dioxide gratings.
Across the thickness series, reported lasing wavelengths ranged from 3.79 to 4.05 micrometres. The comparison used additional devices with 130-nanometre and 165-nanometre flakes, while a patterned 170-nanometre device was used for the reported minimum room-temperature threshold.
A narrow line, with room to tune
The paper reports the wavelength range across the tested thickness series; it does not establish that thickness alone caused the change. The same comparison tracked threshold fluence, meaning the optical pump energy delivered per unit area at the threshold used to identify the output regime. Within that comparison, extracted thresholds declined monotonically, with the 165-nanometre device reaching (0.40 ± 0.06) mJ/cm². The thresholds were determined by extrapolating fits of the linear regime of the output-power curves.
The 110 K test reported a lower threshold
For the patterned 170-nanometre device, the minimum reported room-temperature threshold fluence was (0.25 ± 0.07) mJ/cm². The paper obtained threshold fluences by extrapolating fits of the linear regime of the output-power curves, so the reported values are estimates from the measured output trend.
At 110 K, the temperature-testing device reached a reported threshold of (0.015 ± 0.005) mJ/cm²—approximately tenfold below the reported room-temperature value. That result indicates an association between the tested temperature and threshold in this setup; because the temperature-dependent device was distinct from the devices used in the thickness comparison, the study does not isolate temperature as the sole explanation for the difference.
The reported output performance included a differential slope efficiency of (14.4 ± 0.2)%, a peak output fluence of about 0.16 mJ/cm² and a peak output power density of 160 kW/cm². Slope efficiency expresses how quickly output rises as the pump increases; here it describes the measured optical output, not electrical efficiency.
The study also used a quasi-steady-state generation–recombination calculation to estimate the carrier concentration associated with the optical threshold. Assuming bimolecular recombination dominated at threshold, it placed the room-temperature threshold carrier concentration at about 1.1 × 10¹⁹ cm⁻³.
Using the same generation–recombination conversion, the authors estimated equivalent threshold current densities of about 100 kA/cm² at room temperature and about 6 kA/cm² at 110 K. These are model-based estimates that depend on recombination and injection assumptions; they are not measurements from electrically injected devices.
Behind the measurement
The devices used mechanically exfoliated black-phosphorus flakes transferred with a PDMS stamp onto patterned SiO₂/Si DFB gratings. The DFB structure was a second-order SiO₂/air grating with a 2-micrometre period, etched 600 nanometres into a 5-micrometre thermal SiO₂-on-Si substrate and made with a duty cycle of 0.7.
The flakes were encapsulated with a 10-nanometre Al₂O₃ layer deposited at 150°C. Fabrication took place under nitrogen, with oxygen and water vapour below 0.1 parts per million.
For photoluminescence measurements, the study used 808-nanometre continuous-wave excitation at 550 W/cm². Lasing spectra were collected with a Nd:YAG laser operating at 4.5 kHz and FTIR/InSb detection; the main narrowband result was measured under 1064-nanometre optical excitation.
Two-dimensional finite-element simulations modeled fundamental TE₀ transmission through a finite 80-micrometre black-phosphorus waveguide/grating as a function of flake thickness. The simulations accompanied the optical measurements as a model of transmission through the device structure.
What the preprint leaves open
The evidence is limited to the tested device geometries, optical excitation conditions and temperatures. It is an optical-pumping demonstration, not electrically injected lasing, and it does not show continuous-wave lasing or long-term operational stability.
The number of independent devices, flakes and replicate measurements is not reported, and no inferential statistical testing or prespecified analysis plan is reported. The threshold pattern and wavelength range should therefore be read as measurements from the tested structures, not as a statistically characterized estimate of device-to-device reproducibility.
Another caution is that thresholds were extracted by extrapolating the linear output regime. Device-to-device variability and edge-scattering differences may also affect threshold comparisons, so the reported values apply to the tested configurations rather than establishing a universal threshold for black-phosphorus devices.
The main unanswered step is whether stable continuous-wave or electrically injected mid-infrared lasing can be achieved. The study also leaves open how reproducible the thresholds and wavelength tuning would be across larger batches, and whether optical losses, grating absorption and carrier confinement can be reduced enough for practical electrical operation.
The document is identified as arXiv version 1, dated 20 Aug 2026. It lists funding from the Natural Sciences and Engineering Research Council of Canada and the Canada Research Chairs Program, with COMSOL licensing and fabrication support through the FABrIC project funded by the Government of Canada.
Paper data and sources
Original title: Mid-infrared distributed-feedback lasing from black phosphorus under nanosecond excitation
Authors: Julien Brodeur, Laure Sène, Stéphane Kéna-Cohen
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-20
DOI: Not available
Original paper · Full text