Everything, everywhere, all at once in particle physics

Cosmic rays –high-energy protons and atomic nuclei – are launched into space by supernovae, active galactic nuclei, and other cosmic events. This makes cosmic particles incredibly useful for studying the universe, which scientists do by capturing them with space-based detectors. But there is a challenge: detectors only record particles that interact with their material. To understand all incoming cosmic rays, we need to know how likely they are to interact. This question extends beyond astrophysics – proton therapy in medicine, for example, relies on precise knowledge of how protons interact with human tissue to target cancer cells without unnecessary damage. Despite its importance, measuring and predicting these interaction probabilities is difficult. The DAMPE collaboration has now found a new way to measure how protons and other hadronic particles – that are particles that contain quarks – interact with heavy atomic nuclei. This effort is led by Prof. Tykhonov and Dr. Coppin.

The DAMPE detector was launched in 2015 to search for imprints of dark matter in cosmic rays. A big selling point of DAMPE among other space detectors is that it measures the energy and trajectory of incoming particles with high resolution. Originally, this was meant to identify particles at specific energies that might signal dark matter. But as often happens in science, DAMPE’s capabilities proved useful for something unexpected. Prof. Tykhonov and Dr. Coppin realized that DAMPE could also track where a cosmic ray interacts with the detector’s material. By analyzing thousands of such interactions, they could determine how likely a cosmic ray is to collide with atomic nuclei inside the detector – a quantity known as the cross section. Since DAMPE measures the energy of each incoming particle, they were able to map how this probability changes with energy.

Such wide-range measurements are not possible on Earth. While particle accelerators like the Large Hadron Collider at CERN can measure cross sections with high precision, they have limitations – they focus on one energy at a time and typically collide only a few types of particles, such as protons or lead ions. This is where space-based experiments like DAMPE have a major advantage. Cosmic rays arrive naturally at a wide range of energies and include not just protons, but also helium and heavier atomic nuclei. Just collect the data and get the cross section for each particle –sounds simple, right?

But as always, it’s easier said than done. The diversity of cosmic rays also brings new challenges. Particles arrive from all directions, at different energies, and all at once. To extract meaningful data, researchers must first distinguish protons from helium and other ions and then select only those that travel cleanly through the detector from top to bottom. Dr. Coppin recalls enjoying “exploring the data and testing different techniques” to tackle these problems. He was the first to apply machine learning to this type of measurement. Convolutional neural networks (CNN) analyze the energy patterns inside the detector, helping to identify particles and track their paths with high precision.

Prof. Tykhonov highlights another fundamental result of this work: measuring how protons interact with heavy nuclei advances our understanding of the strong force, one of nature’s fundamental forces alongside the electromagnetic, weak and gravitational forces. The strong force binds quarks together inside protons and neutrons, but its interactions are too complex to calculate directly. That’s why experimental data, like the cross sections measured by Prof. Tykhonov and Dr. Coppin, is so valuable. It helps us to learn more about the strong force in regimes that accelerators cannot reach.

Prof. Tykhonov and Dr. Coppin have demonstrated the power of DAMPE’s data – and this is just the beginning. The same dataset can be used to go beyond protons and helium, extending cross-section measurements to heavier particles. Ultimately, these measurements will refine simulations of how hadronic particles interact with matter, benefiting our exploration of the universe and societal applications such as medical physics, and helping to better understand the health risks that which astronauts face in space.

This work was supported by the ERC PeVSPACE grant and by the Swiss National Science Foundation (SNSF).