Every home has dozens of lasers. They read the data on Blu-ray discs, carry phone calls through fiber optic cables, scan groceries at the checkout counter, and guide surgical instruments to sub-millimeter precision. Their cousin the maser, the microwave counterpart of the laser, has no such résumé. Seven decades after its invention, the maser remains locked in physics laboratories, requiring cryogenic temperatures or vacuum chambers to function. But that may finally be changing. For the first time, researchers have demonstrated a semiconductor maser that works at room temperature, built from silicon carbide, a material already mass-produced for power electronics and LED lighting.
The breakthrough, published in Nature Communications, represents more than just a temperature record. It suggests that masers could follow the same path lasers did in the 1960s and 1970s: from bulky laboratory apparatus to compact, mass-manufactured components embedded in everyday technology.
Masers and lasers share the same fundamental physics. Both rely on stimulated emission, the process by which an excited atom or molecule is triggered by a passing photon to release an identical photon, amplifying the signal. The difference is wavelength. Lasers produce visible or infrared light. Masers produce microwaves and radio waves. The first maser, built by Charles Townes in 1954, used ammonia gas and could fit in a breadbox. It was elegant physics but impractical engineering. To work, the ammonia molecules had to be separated by electric fields inside a vacuum chamber, yielding only a few nanowatts of power.
Solid-state masers improved matters but introduced a different obstacle: the need for liquid helium, which boils at 4 kelvin. The crystal lattice of a solid generates vibrations that destroy the population inversion necessary for maser action, so the entire device must be kept near absolute zero. This cryogenic requirement made masers exquisitely sensitive detectors for radio astronomy but completely unsuited to commercial use. Lasers, by contrast, found their escape hatch in semiconductors. The semiconductor laser diode, demonstrated in 1962 and commercialized over the following decades, could be made small, efficient, and cheap.
The new maser aims to do the same for microwaves. It uses a crystal of 4H-silicon carbide, a polytype of SiC that is already widely used in high-voltage power electronics and electric vehicle inverters. The maser action comes from a specific type of defect in the crystal lattice: a silicon vacancy, where a single silicon atom is missing from its site, leaving behind a spin-active electronic center. These Vsi centers mimic the role of the ruby crystal in the first laser, but with an important advantage. They sit inside a semiconductor that can, in principle, be electrically pumped.
The researchers demonstrated continuous-wave operation above room temperature at 315 kelvin (42 degrees Celsius). The trick was an active feedback loop that multiplies the quality factor of the microwave resonator by a factor of 5, effectively trapping the microwave field long enough for the weak gain from the silicon vacancies to build into sustained oscillation. At 110 kelvin, the device achieved gain exceeding 10 decibels. At room temperature with 617 milliwatts of optical pump power at 808 nanometers, a wavelength produced by inexpensive laser diodes, the output power reached minus 80 dBm per hertz, a level comparable to the signals encountered in practical microwave receivers.
To prove the maser was more than a laboratory curiosity, the team demonstrated three distinct applications using the same device. The first was a low-noise amplifier with a noise figure between 1 and 3.5 decibels. For comparison, conventional room-temperature microwave amplifiers struggle to get below 6 decibels. The maser amplifier achieves its low noise because stimulated emission, unlike resistive amplification, adds almost no thermal noise to the signal. This could make it useful for quantum readout circuits, radio astronomy receivers, and 5G or 6G base stations.
The second demonstration was even more unusual: a microwave refrigerator. The maser, when tuned to absorb microwaves rather than amplify them, acts as a heat pump that extracts energy from the cavity mode. The researchers measured the cavity mode temperature dropping by 40 kelvin below the ambient temperature of the device. This is a form of active cooling at microwave frequencies, analogous to what a laser cooling trap does for atoms, and it could help protect sensitive quantum circuits from thermal noise without requiring a dilution refrigerator.
The third demonstration was a magnetometer. The silicon vacancy centers are sensitive to magnetic fields due to their spin properties, but reading out that spin state via conventional optical methods is noisy. The maser provides an alternative: the spin state modulates the maser gain, which in turn modulates the output power. This readout achieved a sensitivity of 20 picotesla per square root hertz. The contrast-to-linewidth ratio improved by nine orders of magnitude over conventional optical readout. Nine orders of magnitude is the difference between a wristwatch and the age of the universe.
The choice of silicon carbide over diamond, which has been used in previous maser experiments, is deliberate. Diamond masers suffer from a zero-field splitting in the gigahertz range, meaning the spin energy levels are already widely separated even in zero magnetic field. This makes it difficult to tune the maser frequency or switch between amplification and absorption. Silicon carbide has a zero-field splitting of only 70 megahertz, roughly 100 times smaller. A change in magnetic field of just 2.5 millitesla in either direction is enough to toggle the device between amplifying and absorbing behavior. This tunability is essential for practical applications. A refrigerator that only cools one fixed frequency is far less useful than one that can be tuned to match the specific thermal noise peak of the circuit it is protecting.
The broader significance lies in material compatibility. Silicon carbide is not a specialized crystal grown only for quantum experiments. It is a standard semiconductor, fabricated in foundries that already produce millions of wafers per year for electric vehicles, solar inverters, and high-voltage power supplies. The infrastructure for manufacturing, doping, and contacting SiC is mature. This raises the prospect of an electrically driven maser diode, analogous to the laser diode. Instead of requiring an external optical pump at 808 nanometers, such a device would inject electrons and holes directly into the silicon carbide, creating the population inversion electrically. A semiconductor maser diode, small enough to fit on a chip and cheap enough to embed in consumer electronics, would bring the same revolution to microwaves that the laser diode brought to light.
What lasers did for light (shrinking, industrializing, and democratizing a fundamental physical effect), the semiconductor maser may now do for microwaves. The first laser was a ruby rod inside a helical flashlamp, a device that filled a lab bench. Today, a laser diode the size of a grain of salt powers the internet. The first semiconductor maser is still bulkier than that, but it works at room temperature, it is built from a material already in mass production, and it amplifies, cools, and senses. The history of the laser suggests that once a device can be built from a semiconductor, the path to ubiquity is open.
Reference: Nature Communications 17, 7267 (2026). DOI: 10.1038/s41467-026-75446-2