PSR J0125−5854 is a neutron star spinning roughly forty times per second, locked in an 833-day orbit with the burned-out core of a star whose outer envelope it long ago pulled away. As of June 2026, it is the first millisecond pulsar that the Murchison Widefield Array — a low-frequency radio telescope spread across a stretch of red dirt 800 kilometres north of Perth — has ever found.
The discovery paper, led by Chia Min Tan of Curtin University, was posted to the arXiv preprint server on 17 June 2026 and has been accepted for publication in The Astrophysical Journal Letters. Reporting on the find noted that the pulsar emerged from an ongoing all-sky survey that has been quietly running on Curtin University’s supercomputers for years.
The MWA was not built with pulsar searches in mind. It was designed to image the low-frequency southern sky for cosmology and solar science. The fact that it is now turning up the most rapidly spinning class of neutron star, drawn out of a frequency band that most surveys have written off as too messy for the job, is a quiet validation of a decade of software work by a small team in Western Australia.
A first for the MWA
The Murchison Widefield Array sits at Inyarrimanha Ilgari Bundara, the CSIRO Murchison Radio-astronomy Observatory, on Wajarri Yamaji country in mid-west Western Australia. It is one of the precursor instruments to the low-frequency Square Kilometre Array being built on the same site, and it operates between roughly 80 and 300 MHz.
The Southern-sky MWA Rapid Two-metre survey, known as SMART, is the array’s flagship pulsar programme. It targets the sky south of +30° declination in a 140–170 MHz band, exploiting the MWA’s wide field of view and its voltage capture system to record raw antenna data that can be re-searched again and again as processing pipelines improve.
SMART has produced earlier discoveries. The first new pulsar found with the MWA, PSR J0036−1033, was reported in 2021 by a Curtin PhD student named Nick Swainston after he processed a small slice of the data; it was a long-period, low-luminosity object with a 0.9-second rotation. Every SMART discovery since had likewise been a slow, ordinary rotator.
PSR J0125−5854 breaks that pattern. It is the first millisecond pulsar identified with the MWA, and the first pulsar pulled out of what the SMART team calls a deep-pass search — a full-length, 80-minute integration of voltage data instead of the 10-minute chunks used in earlier shallow processing rounds.
Deep-pass searches are computationally heavier by an order of magnitude, but they reach much fainter signals. They are also the only realistic way to find a millisecond pulsar at MWA frequencies, where every additional minute on-source helps drag a faint, rapidly modulated pulse train out of the background.
What the pulsar looks like
The discovery paper reports a spin period of 24 milliseconds and a dispersion measure of 11.66 parsecs per cubic centimetre, which translates to an estimated distance of roughly 1,600 to 3,200 light-years based on standard models of the electron density of the Milky Way.
The pulsar sits at an unusually high galactic latitude of about −57°, well off the plane of the Milky Way. That puts it in the relatively clean galactic halo rather than buried in the noisy gas of the disk, which is one reason a low-frequency search could pick it up at all.
Follow-up observations with the MWA and with South Africa’s MeerKAT telescope show that PSR J0125−5854 is in a binary system with an orbital period of about 833 days, a projected semi-major axis of roughly 241 light-seconds, a minimum companion mass of about 0.41 solar masses, and an almost circular orbit. The discovery team’s analysis points to a helium white dwarf as the most likely companion — the inert, hydrogen-stripped remnant of a star that has finished donating mass to its neighbour.
That orbital architecture matters. Wide, nearly circular pulsar–helium white dwarf binaries are the textbook outcome of a particular evolutionary pathway, and finding one in the SMART data gives theorists a clean test case for the standard model of how millisecond pulsars are made.
How millisecond pulsars get spun up
Ordinary pulsars slow down over millions of years as they bleed rotational energy into radio emission and a magnetised wind. Millisecond pulsars run the process in reverse: they have been accelerated back up to extreme spin rates after being born. The accepted explanation is called the recycling scenario, and it requires a binary companion.
According to the Center for Astrophysics | Harvard & Smithsonian, millisecond pulsars increase their rotation rates by accreting material from a nearby companion star. The infalling matter carries angular momentum, which transfers to the neutron star and spins it up over hundreds of millions of years.
When the companion eventually exhausts its hydrogen envelope, what is left behind is often a helium white dwarf in a wide, nearly circular orbit — almost exactly the configuration Tan and his colleagues see around PSR J0125−5854. The system reads like a worked example pulled from a textbook chapter on binary stellar evolution.
A few hundred millisecond pulsars are now known across the Milky Way. A significant fraction sit inside globular clusters, where stellar densities are high enough to make binary capture and exchange relatively common. The newly discovered pulsar lies in the galactic field instead, part of the more dispersed but more numerous population that any all-sky low-frequency survey is statistically likely to dominate.
The pulsar map etched onto the cover of the Voyager Golden Record uses fourteen pulsars as cosmic landmarks, and millisecond pulsars are now the steadiest clocks in that landscape. PSR J0125−5854 will not be added to that map, but it joins the same broad family of objects.
Why low-frequency detections are hard
Most known millisecond pulsars were found at frequencies above 300 MHz, where their rapid pulses suffer relatively little from the smearing effects of the interstellar medium. At the MWA’s 140–170 MHz band, dispersion stretches the same pulse out over a much wider window, and any millisecond-scale signal has to survive that distortion to remain detectable.
For decades, that was the standard argument against using low-frequency arrays to hunt for millisecond pulsars: the physics works against you. Doing it anyway requires very long dwell times, careful coherent dedispersion, and a willingness to burn supercomputer hours on processing.
The SMART pipeline successfully recovered PSR J0125−5854’s pulse profile cleanly, and the subsequent MeerKAT follow-up confirmed it at higher frequencies. That is a proof of concept the SMART team has been waiting on.
Once complete, the SMART survey is projected to discover hundreds of new pulsars, including a meaningful share of new millisecond pulsars. The first deep-pass MSP is, in that sense, the leading edge of a much larger expected haul.
The broader stakes
Each new millisecond pulsar adds to a precision timing network that astronomers now use as a galaxy-scale instrument. Pulsar timing arrays exploit the metronome-like regularity of these objects to search for low-frequency gravitational waves, probe the interstellar medium, and constrain the equation of state of matter at neutron-star densities.
Population statistics also feed directly into one of the most contested problems in galactic astrophysics. The Galactic Centre Excess — a diffuse glow of GeV gamma rays around the centre of the Milky Way, first reported in Fermi-LAT data more than a decade ago — has two leading explanations. One is the annihilation of dark matter particles. The other is an unresolved population of faint millisecond pulsars in the galactic bulge.
Knowing how millisecond pulsars are distributed across the rest of the galaxy — their luminosities, their binary fractions, their formation efficiencies — directly affects how plausible the pulsar interpretation is. Surveys like SMART are the way that census gets built, one detection at a time.
What comes next
The SMART team plans to continue processing deep-pass data over the next several years. Less than a tenth of the survey volume has been pushed through deep-pass searches so far, which means the discovery rate of millisecond pulsars at low frequencies is essentially a tap that has just been opened.
The MWA itself has recently completed its Phase III upgrade, which brings a new correlator and a new receiver suite that allow all 256 of its tiles to be used simultaneously. The next round of surveys will be more sensitive than anything SMART has done to date, and they will feed into future searches with the low-frequency SKA being built on the same site.
For the SMART team, this discovery is the kind of result that justifies the long computational slog of a deep-pass search. One pulsar, by itself, does not rewrite any models. But it confirms that the pipeline works, that the MWA can do this kind of science, and that the southern sky still contains rapidly spinning neutron stars waiting to be heard.
The pulsar is still there, still spinning, still flashing its radio beam past Earth every twenty-four milliseconds. The white dwarf companion is still tracing its slow 833-day waltz around it. The antennas in the Western Australian outback are still recording, and somewhere in the next batch of voltage-capture data, more objects like PSR J0125−5854 are waiting to be untangled from the noise.
