The Bullet Cluster of galaxies might only have half as much dark matter than scientists thought, or perhaps even none at all, according to a new look at observations of this immense collision of galaxies from the James Webb Space Telescope (JWST).

“Our results indicate that the required amount of additional unseen mass – dark matter – is reduced,” astrophysicist Dong Zhang of the University of Bonn, who led the study, tells Physics World.

The Bullet Cluster is a central battleground for proponents of alternative models of gravity seeking to usurp dark-matter theory. The cluster itself is actually two clusters of hundreds of galaxies that have collided head-on. In 2006 the Hubble Space Telescope mapped how the Bullet Cluster’s mass was warping space and creating gravitational lensing, and from this it was possible to infer where the dark matter in the cluster was relative to the galaxies themselves and the hot, X-ray emitting gas. While the galaxies and hot gas remained in the middle, the dark matter from both colliding clusters had passed through unhindered and is now on opposite sides – exactly as dark-matter theory predicted.

In 2025 the JWST observed the Bullet Cluster and the initial analysis, led by Sangjun Cha of Yonsei University, found that while the distribution of mass in the cluster is more complex than expected, it still supports the dark-matter model.

“Those JWST observations of the Bullet Cluster have sparked a lot of interest,” says astrophysicist Richard Massey of Durham University, who was not involved in the new study.

Counting stellar remnants

Astrophysicists led by Zhang have now made their own analysis of the JWST data and arrived at a very different conclusion, as they report in Physical Review D.

Zhang’s team aimed to precisely measure the total mass of not just the stars, but also of dark stellar remnants – white dwarfs, neutron stars and black holes formed when stars die – to see if these objects could account for the extra gravity attributed to dark matter. They found that there would be enough stellar remnants to reduce the amount of dark matter required by half, at least. Furthermore, applying MOND (modified Newtonian dynamics) to the cluster potentially removes the need for dark matter entirely.

They utilized a model called the integrated galaxy-wide initial mass function (IGIMF), which is a variation on the standard IMF that describes how a collapsing molecular cloud fragments into clumps that produce stars. The standard IMF favours low-mass over high-mass stars.

However, variations in chemical compositions, stellar radiation fields and the rate at which gas cools in a given environment mean that the IMF varies from galaxy to galaxy. The IGIMF is designed to get around this by integrating over the full stellar population of a galaxy, or in this case, an entire cluster of galaxies.

“The IGIMF framework was developed independently of MOND and is motivated by observational constraints related to star formation and stellar populations,” says Zhang.

One of the quirks of the IGIMF is that it predicts that in the early universe, more massive stars formed in giant elliptical galaxies than form in galaxies today, consequently producing more stellar remnants than what the standard IMF predicts. Such giant galaxies exist in the core of clusters such as the Bullet. Since massive stars leave behind massive remnants when they go supernova, the gravity of these remnants could account for a large amount of the dark matter. This is supported, says Zhang, by the abundance of heavy elements in some galaxies that’s hard to explain without invoking an early population of very massive stars.

When the quantity of stellar remnants predicted by the IGIMF is applied to the standard model of cosmology “it does not appear sufficient to account for the strong lensing mass in the central regions of the Bullet Cluster,” says Zhang. It is only when applying MOND that things begin to align, he adds.

Questioning the mass function

However, not everyone is convinced by the choice to use the IGIMF over the standard IMF.

“It’s kind of a neat theoretical framework, in which aspects that could produce a total IMF are broken down into separate mathematical terms that can be modelled individually,” says Massey. “But it’s also kind of irrelevant for this sort of empirical analysis, since mainstream analyses already assume that the IMF varies between galaxies, and even in different regions of a galaxy.”

Traditionally, estimating the total stellar mass of a galaxy is accomplished by taking spectra or multi-band photometry and employing a technique called “stellar population synthesis” that best-fits a galaxy’s light to a particular distribution of stellar masses, Massey explains.

“However, the emitted light can be dominated by the brightest, most massive stars,” he says, meaning that the far greater number of less massive stars has to be extrapolated from a galaxy’s stellar mass function, which is derived from the IMF while factoring in the galaxy’s age.

Stellar remnants are even more difficult to detect directly, hence the dependency on the IMF or IGIMF.

“Stellar remnants cannot generally be detected in external galaxies beyond the Milky Way and its satellites,” says Kathy Romer, an astrophysicist at the University of Sussex who studies galaxy clusters. “Exceptions would be via gravitational-wave events or ultra-luminous X-ray binaries, but those are only the tip of the iceberg, so estimates of the total contribution of stellar remnants will always be an extrapolation of what we measure in the Milky Way and Magellanic Clouds.”

Like Massey, who describes IGIMF as being “less mainstream”, Romer is also sceptical about the new results, but acknowledges that “there are some strong aspects to their paper” and that “bringing fresh perspectives and challenging the orthodoxy is essential to keep science moving forward”.

Massey also highlights recent work from a team led by Gregor Rihtarŝiĉ and including Douglas Clowe of Ohio University who led the original Hubble observations of the Bullet Cluster, and who are running simulations to try and refine their maps of dark matter in the cluster.

The post Does the Bullet Cluster contain a lot less dark matter than we thought? appeared first on Physics World.

Through fresh analysis of supernovae data, a team of astronomers has discredited a recent study that suggested that dark energy has weakened since the early universe.

Led by Phil Wiseman at the UK’s University of Southampton, the team showed that the apparent weakening almost disappears after accounting for a known correlation between supernova brightness and galaxy mass. This latest work confirms that the universe’s expansion is accelerating – a discovery that bagged three astronomers the 2011 Nobel Prize for Physics.

Type 1a supernovae occur when a white dwarf accretes mass from a binary companion star. Eventually, it reaches a critical mass where the crushing weight of its gravity can no longer be supported by electron degeneracy pressure, triggering an explosion that can briefly outshine entire galaxies.

These events are especially valuable to cosmologists because white dwarfs always explode at the same critical mass – meaning type 1a supernovae always reach around the same peak brightness. As a result they are “standard candles”, whereby their distances from Earth is inferred by their apparently brightness. The rate at which a type 1A is moving away from us is given by the redshift of its light. By measuring the distance and motion of these objects, the rate of the universe’s expansion can be determined as a function of time.

Driving acceleration

Since the 1990s, researchers have observed thousands of type 1a supernovae. Contrary to expectations for a matter-dominated universe, these measurements revealed that the universe’s expansion is speeding up. This lead to the concept of dark energy – a hypothetical entity that is driving this acceleration.

But in 2025, a team of astronomers in South Korea, led by Junhyuk Son at Yonsei University, Seoul, made an observation that appeared to turn this long-standing picture on its head. “They claimed that a subtle correlation between the brightness of the supernova and the age of the galaxy it exploded in could, if left uncorrected, means that the overall measurements actually now indicate that the expansion is now slowing,” Wiseman explains. “In other words, that dark energy was strong billions of years in the past, but has subsequently weakened.”

In their study, Wiseman and his colleagues revisited the data used by Son’s team – accounting for a correction that the previous astronomers hadn’t applied. Regardless of redshift, the peak brightness of type 1a supernovae in lower-mass galaxies is known to be slightly fainter than those in higher-mass galaxies. Although the reasons for this correlation remain unclear, astronomers have routinely been correcting for it over the past 15 years.

Wiseman’s team applied the same correction – ensuring their calculated distances were unaffected by differences in galaxy mass. “We also used simulations, where we traced how the age of the exploding stars and the galaxies they explode in evolve with cosmic time,” Wiseman says.

Similar ages

With this correction applied, the claimed correlation between supernova brightness and host galaxy age became almost undetectable. “Even if we account for the prediction that the relationship between the supernova and the galaxy should change as we look further back in time, we end up with almost identical measurements of dark energy,” Wiseman continues. Their simulations also showed that supernova progenitor stars have exploded at consistently similar ages throughout cosmic history.

The results will likely come as a relief to many cosmologists, averting the dark energy crisis some had feared. All the same, the team says Son’s study provides a valuable test of existing approaches to measuring the universe’s expansion rate, given the limitations of current observational capabilities.

“In an ideal world, we think a correction based on the age of the galaxy would be slightly more accurate than using the galaxy mass, but these measurements are currently not available for more than a handful of supernova host galaxies in the distant universe,” Wiseman says. “Until then, a mass correction is adequate and the implication is that the universe is still accelerating due to dark energy.”

The research is described in Monthly Notices of the Royal Astronomical Society.

The post Cosmic crisis averted as supernovae put dark energy back on track appeared first on Physics World.

Physicists have developed a model that predicts how gravitational-wave emission changes when a binary black hole system moves through dark matter. Gravitational waves are ripples in spacetime generated by compact objects spiralling together in a cosmic dance that ends in a violent collision. In the presence of dark matter, these waves may carry an imprint of the surrounding environment. Despite making up over 80% of all matter, dark matter does not interact with the electromagnetic force, and its existence has been surmised only by its gravitational interactions, leaving its precise nature a mystery. Gravitational waves may provide a new way to probe the characteristics of dark matter.

One proposed dark-matter candidate is the light scalar boson, an extremely small particle orders of magnitude lighter than an electron, that can exhibit coherent wave-like behaviour. In the presence of spinning black holes, a process called superradiance can transfer rotational energy into the surrounding field of ultralight boson particles, amplifying the dark matter into a dense cloud around the black hole. These clouds can modify the dynamics of black hole binaries, encoding observable signatures into the gravitational-wave signals that they emit.

To understand what these signatures may look like, an international team of researchers developed a model for gravitational-wave emission from a binary black hole system moving through dark matter. Developing their model required meticulous numerical relativity simulations, but ensuring their results were accurate and reliable proved to be a challenge.

“The hardest part was making sure we truly understood what the simulations were telling us. The danger in numerics is fooling yourself,” explains Josu Aurrekoetxea of Massachusetts Institute of Technology. “You run a simulation, you see an effect, and you have to ask – is that dark matter, or is that my code?”

Working with Katy Clough of Queen Mary University of London and Pedro Ferreira of the University of Oxford, Aurrekoetxea spent several years developing these simulations, before joining forces with Soumen Roy of Université Catholique de Louvain and Rodrigo Vicente of the University of Amsterdam, whose expertise in analytical modelling and data analysis helped the team construct and validate the novel waveform model.

Most work in the field of gravitational waves and dark matter focuses on observations with future space-based detectors such as the Laser Interferometer Space Antenna. “Our motivation was to ask what could already be learned about specific dark-matter models using the LVK [LIGO–Virgo–KAGRA] gravitational-wave data available today,” says Vicente. “We were surprised to find that some of the more extreme models can already begin to be tested with signals observed by current experiments.”

Searching for these dark-matter imprints, they applied their model to 28 publicly available signals observed by the LVK collaboration, a global gravitational-wave detector network that has observed signals from hundreds of mergers. For each event, the team compared the signal to their predictive dark matter-modulated waveform and the standard waveform for black holes merging in a vacuum.

After rigorous testing, the researchers found that 27 of the 28 events were consistent with gravitational-wave signals expected from binary black holes evolving in empty space. One event, GW190728, showed a slight preference for the dark-matter model, indicating that its signal contains properties consistent with modulations from the presence of dark matter. However, the statistical evidence is not strong enough to claim a confident detection of a dark-matter imprint.

Although only one event demonstrated a preference for the predictive dark-matter model and no definitive claim of a dark-matter signature could be made, the researchers have demonstrated a novel method to probe and characterize dark matter through signatures encoded in gravitational-wave signals.

“Our findings from GW190728 suggest that, if this interpretation is correct and we are seeing a genuine environmental signature, similar effects may appear in future LVK events,” says Roy. Therefore, the next step for the researchers is to extend their analysis to LVK’s fourth observing run. Roy also says that if they can estimate the number and types of events that should carry dark-matter imprints, they can combine multiple signals to test their hypothesis more robustly.

“It has become clear that gravitational waves have tremendous potential to help us understand how matter behaves in the extreme gravitational fields surrounding black holes. This is particularly relevant for dark matter, which may interact only through gravity,” says Vicente. And as more gravitational-wave signals are observed, they provide a promising channel to uncover the elusive nature of dark matter.

The research is described in Physical Review Letters.

The post Novel gravitational-wave model sheds light on dark matter appeared first on Physics World.