The CMS experiment sets limits for BSM particles down to tens-of-MeV aided by machine learning techniques.
Axion-like particles (ALPs) are hypothetical light-weight neutral bosons that could solve deep mysteries in physics, from the strong CP problem (why is CP-symmetry conserved in strong interactions?) to the nature of dark matter. One compelling scenario would be if the Higgs boson acted as a portal to such particles: this would imply that the Higgs boson can decay into two ALPs, each further decaying into an electron and a positron, as illustrated in Figure 1.
The CMS Collaboration has launched a search for such light-weight hidden particles produced from Higgs boson decays—probing masses down to the tens-of-MeV range—in which the decay products are reconstructed as highly collimated electron-positron pairs.
Figure 1: An illustration of the physics signature: A Higgs boson produced via gluon fusion decays into two ALPs, each of which decays further into an electron-positron pair with individually distinguishable trajectories (represented by dashed curves), yet with energy deposits that merge in the CMS detector (represented by the colored squares).
Yet, finding these light-weight particles at the LHC is exceptionally challenging. “With a significantly higher momentum than the particle's mass, decay products of the particle can be so close together that their experimental footprints become indistinguishable and make it difficult to figure out whether there are actually one or two particles”, SangHyun Ko explains, who obtained his Ph.D. recently from Seoul National University.
To overcome this, CMS physicists developed a novel multivariate identification technique. By combining precise tracking information and advanced machine learning algorithms, the team was able to reconstruct these merged electron pairs.
No significant excess above the Standard Model prediction was observed in the full Run 2 dataset (Figure 2, left). However, the analysis sets new stringent limits on such exotic Higgs boson decays, excluding branching fractions down to the order of 10⁻⁵ (Figure 2, right). “This is the most sensitive search in the electron channel so far, and the first to set limits on such light-weight neutral particles from Higgs boson decays with masses down to 10 MeV”, says Yipin Wang, a Ph.D. student at Peking University who has focused on this search since his undergraduate thesis.
Figure 2: The measured mass spectrum from the two merged electron pairs fitted with a hypothetical signal of an ALP with a mass of 20 MeV and a decay length of 10 μm (left), and the full results re-interpreted using an effective theory description of the ALP model (right).
“The results also provide new constraints on the theoretical couplings governing ALPs, underlining the LHC's unique role as a potential axion factory.” Sitian Qian, a Ph.D. candidate at Peking University, explains.
Future LHC data, upgrades to CMS’s tracking and calorimetry systems, and cutting-edge analysis techniques could further sharpen the search for these elusive particles. In addition, this study demonstrates CMS’s capability to explore rare Higgs boson decays and highlights the experiment’s expanding reach into the light- and weakly-coupled sectors, opening windows to physics beyond the Standard Model.
Written by: Sitian Qian, for the CMS Collaboration
Edited by: Andrés G. Delannoy
Read more about these results:
-
CMS Physics Analysis Summary (EXO-24-031): "Search for light pseudoscalar a bosons in H→aa→4e decays in pp collisions at √s = 13 TeV"
-
@CMSExperiment on social media: Bluesky - Facebook - Twitter - Instagram - LinkedIn - TikTok - YouTube