Despite knowing with a high degree of certainty that dark matter exists, we do not (yet) know what it is. Credit: Ralf Kaehler / SLAC National Accelerator Laboratory
It is now understood that all known matter, i.e., studied by science and harnessed by technology, constitutes only 5% of the content of the universe. The rest is composed of two unknown components: dark matter (about 27%) and dark energy (about 68%). This calculation, confirmed decades ago, continues to surprise both lay people and scientists alike.
In the case of dark matter (DM), there is abundant evidence that it really exists, all resulting from its gravitational interaction with ordinary matter. This evidence comes from sources such as the rotation curves of stars in galaxies, discrepancies in the movement of galaxies in clusters, the formation of large-scale structures in the universe, and cosmic background radiation, which is distributed uniformly throughout space.
Despite knowing with a high degree of certainty that DM exists, we do not know what it is. Several models proposed thus far have failed.
A new study by researchers at the University of S茫o Paulo (USP) in Brazil proposes an inelastic DM model that interacts with ordinary matter through a vector mediator similar to a photon, but with mass. The aim is to open a new window of observation. An article on the subject is in the Journal of High Energy 麻豆淫院ics.
"In this work, we consider a DM model composed of a dark sector with light particles that interact weakly with the known particles of the Standard Model [SM]," says Ana Luisa Foguel, a Ph.D. student at the 麻豆淫院ics Institute (IF-USP) and the first author of the article.
Initially, the search for DM focused on heavy candidates with masses much greater than that of an electron or even the heaviest particles in the SM. The idea was that, because they were so massive, these particles could not be produced by particle colliders, which did not yet have sufficient energy. However, even with the experiments carried out at the Large Hadron Collider (LHC) at the European Organization for Nuclear Research (CERN, in its official French acronym), no new particles beyond those of the SM have been observed.
Consequently, some in the scientific community shifted their focus to searching for light particles with extremely weak interactions. The idea was that such particles had not yet been observed because they interact very weakly with ordinary matter. To investigate signs of these particles, experiments needed to move toward the so-called "intensity frontier," meaning they would have to measure couplings and interactions with increasing precision to detect any discrepancies that might signal the existence of something new.
Thermal freeze-out
The study is moving in this direction.
"When considering a new DM model, the first thing is to know how it was possible to produce the right amount of such a component. This amount is now measured very precisely, with data from cosmic background radiation, for example. And several mechanisms are known that could have led to the generation of DM in the early universe. One of the most theoretically motivated is the so-called 'thermal freeze-out,'" says Foguel.
In particle physics and cosmology, thermal freeze-out is the moment when certain particles decouple from the thermal bath, meaning the interactions that transform these particles into other SM particles are no longer sufficient. After this point, since there are no processes that can alter the number of these particles, their abundance "freezes," remaining virtually unchanged.
"This mechanism is interesting and well known, as we have several examples of SM particles whose abundance was generated in this way. Therefore, it's natural to consider that the components of DM were generated by a similar mechanism," comments the researcher.
In this mechanism, DM candidate particles are in a "thermal bath" with ordinary matter particles shortly after the beginning of the universe. In other words, all particles interact very quickly to share the same temperature. As the universe expands and cools, the particles lose this thermal contact. This process is called "freeze-out."
"The exact moment of decoupling depends on the probability of interactions between DM particles and SM particles. This probability is parameterized by a variable we call the sigma shock section. If sigma is very small, DM particles decouple very early and their abundance is very high. Conversely, if it's very large, the DM remains in thermal contact longer, annihilating itself into SM particles, so that when it decouples later, it doesn't have sufficient abundance," points out Foguel.
In the case of light DM, the interaction with ordinary matter occurs through a portal. In other words, DM particles do not couple directly with all SM particles, but rather with a mediator particle that facilitates interaction between DM and the SM. The sigma shock section of this interaction is proportional to the mass of the DM and inversely proportional to the mass of the portal particle.
Thus, for a light candidate to exist at an energy level below gigaelectronvolts, the portal cannot be too heavy. Therefore, the SM bosons that mediate weak interactions (W+, W-, and Z0) would not function as portals. A new dark particle must be introduced to mediate between the DM and the SM.
"In our model, this particle that mediates the relationship between the two sectors is a vector boson (ZQ). It behaves like a photon, the particle that mediates electromagnetic interactions, but it has mass. In addition, the difference in this model is that this mediator also interacts directly with other SM particles," says the researcher.
This mediator would connect the SM particles to the DM particles. According to the proposed model, there are two types of these particles: a stable particle (蠂鈧), which would make up DM itself, and a slightly heavier unstable particle (蠂鈧). These particles would always interact with the ZQ mediator together. In other words, the mediator would interact with both at the same time. This would constitute a specific type of DM called "inelastic dark matter."
In addition, 蠂鈧 could decay into 蠂鈧 and SM particles. This work demonstrates that these arrangements can explain the abundance of DM in the universe while circumventing the experimental limits that prevent its detection.
"It's worth noting that models such as ours, with inelastic DM, are interesting because in addition to explaining the efficient generation of DM through the freeze-out mechanism, they also make it possible to circumvent the current limits of direct and indirect detection, as well as the limits of cosmology. The reason comes from the fact that as 蠂鈧 isn't stable and interactions depend on 蠂鈧, there isn't enough 蠂鈧 population during the recombination epoch to inject energy into the plasma, which could have modified the cosmic background radiation.
"And there's also no 蠂鈧 in the current universe to decay or annihilate with 蠂鈧, producing signals that enable indirect detection. Furthermore, for 蠂鈧 to interact in direct detection experiments, it'd have to transform into 蠂鈧, which is very difficult because 蠂鈧 is more massive," Foguel explains.
Overcoming the 'vanilla' model
According to the researcher, the proposed new model would serve as an alternative to the "vanilla" model of inelastic DM, which considers a mediator that does not couple directly with DM particles. In particle physics, "vanilla" is used to designate the most basic and minimalist version of a model with the fewest theoretical ingredients possible.
"The vanilla model has already been practically ruled out, because almost all of the parameters that reproduce the correct abundance of DM have been discarded by experimental searches. Thus, the main objective of our work was to show that by considering a simple modification of this model鈥攁llowing mediators with direct rather than indirect couplings鈥攚e can potentially 'save' inelastic DM," Foguel explains.
"Considering the proposed models, we first calculated the abundance of DM using the freeze-out process and made a code available online that allows these calculations to be reproduced, showing the regions of the parameter space that produce inelastic DM for different choices of Q load, with correct abundance. After that, we focused on the limits of different experiments.
"We concluded that for certain models, new regions of the parameter space are 'unlocked,' that is, there are parameters that reproduce the correct abundance of DM and haven't yet been excluded. Some of these parameter regions could be investigated in future experiments."
Renata Zukanovich Funchal, a full professor at IF-USP, Foguel's advisor, study coordinator, and co-author of the article, summarizes, "The use of more general vector mediators opens a new window for viable models of inelastic DM, with direct consequences for decay rates, experimental signatures, and cosmological limits."
More information: Ana Luisa Foguel et al, Unlocking the inelastic Dark Matter window with vector mediators, Journal of High Energy 麻豆淫院ics (2025).
Provided by FAPESP