Astronomers close in on the source of the highest energy particles

Here on Earth, if you want to observe particles at the highest possible energies, you have two potential approaches you can choose from. You can isolate charged particles in a laboratory and accelerate them with a combination of electric and magnetic fields, either linearly or in a circular path, to higher and higher energies before either releasing them in a particular direction or smashing them into other particles. These particle physics experiments have created enormous numbers of high-energy particles and given us enormous amounts of data that’s useful for studying nature, allowing us to understand the building blocks of reality at a fundamental level. We’ve accelerated particles up to GeV and even TeV energies in the lab: to billions (10⁹) or even trillions (10¹²) of electron-volts.

But nature, even in the depths of space, has ways to far surpass anything humans can achieve on Earth. Natural particle accelerators — in the form of star-forming regions, black holes, supernovae, and even pulsing neutron stars — frequently reach energies much greater than even those found at the Large Hadron Collider. While the LHC caps out at around 7 TeV (tera-electron-volts) of energy per particle, cosmic rays are often found at PeV (peta-electron-volts, or 10¹⁵ electron-volts) scales, or even higher. For a long time, we could only speculate as to where these particles originated. But now, astronomers are closing in on the origins of these very high energy cosmic particles, and our own galaxy looks like the culprit.

The energy spectrum of the highest energy cosmic rays, by the collaborations that detected them. The results are all incredibly highly consistent from experiment to experiment, with the “knee” (or 1st knee) representing PeV-scale phenomena. That, alone, is already hundreds of times the maximum energies achieved by the LHC, and cosmic rays of that energy are thought to originate from within our own Milky Way.

Cosmic rays are one of those things that we observed long before we gained a theoretical understanding of what must be out there in the Universe to create them. Early particle physics experiments conducted via hot air balloons, followed by terrestrial experiments with cloud chambers, revealed a slew of particles possessing a wide variety of properties. When a magnetic field was applied, cosmic ray particles bent in different fashions from one another. When taken together, the data revealed:

• particles of different energies,
• particles with differing electric charges and charge-to-mass ratios,
• and particles with different origins: from our Sun, from the galaxy, and randomly across the sky.

Over time, astronomers and physicists who studied these cosmic rays began to measure them more and more accurately. We acquired direct measurements of cosmic rays by sending detectors up into space. We detected cosmic ray showers by observing the particles produced when a cosmic ray strikes Earth’s upper atmosphere. We’ve detected cosmic neutrinos directly with apparatuses such as IceCube. And we detect Cherenkov light — or light from fast-moving particles that enter the atmosphere and exceed the speed of light in that medium — that allows us to reconstruct both the energy and, sometimes, the direction of origin of the particles that spawned these observed phenomena.

This animation showcases what happens when a relativistic, charged particle moves faster than light in a medium. The interactions cause the particle to emit a cone of radiation known as Cherenkov radiation, which is dependent on the speed and energy of the incident particle. Detecting the properties of this radiation is an enormously useful and widespread technique in experimental particle physics, and also in astronomy for detecting atmospheric cosmic rays.

One of the more interesting results to come out of cosmic ray studies is a specifically-shaped diagram that describes the abundance, or flux, of cosmic rays as a function of the energy of those rays. Known as the cosmic ray energy spectrum, the spectrum follows a simple, straightforward curve the way you might expect:

• with greater numbers of cosmic particles at lower energies,
• and then with fewer numbers as you scale to higher energies,
• all following a simple relationship that makes a straight line when you plot it on a log-log scale.

But above a certain energy threshold — roughly a few PeV in energy — what has up until that point been a straight line relationship between how many particles there are in each specific energy range suddenly changes: a phenomenon that astronomers call the “knee” (or sometimes the “first knee”) in the cosmic ray energy spectrum. Because of various effects that matter and radiation are known to experience when they travel through intergalactic space at very high energies, it’s long been thought that these highest-energy cosmic rays, of PeV energies and upward, must have originated from somewhere within our own galaxy, rather than from any extragalactic source.

When a neutrino interacts in the clear Antarctic ice, it produces secondary particles that leave a trace of blue light as they travel through the IceCube detector. IceCube is a series of 86 strings with over 5000 detectors embedded in the ice, capable of detecting the Cherenkov photons produced by particle showers arising from characteristic neutrino interactions. If a supernova were to go off inside the Milky Way, IceCube alone would detect many millions of neutrinos.

This picture got a big boost a few years ago from what might seem, on the surface, to be an unrelated source of information: the IceCube neutrino detector. IceCube is an under-ice neutrino detector in Antarctica, with over one billion tons (a gigaton) of ice — more than a full kilometer on a side in all three dimensions — serving as the volume for neutrinos to interact with. When they do, they produce other particles that propagate through the ice, and in particular particles that move faster than the speed of light in the medium of that ice. When those particles go faster than light in this medium, they emit a cone of blue light: Cherenkov radiation, for as long as they travel through this medium at superluminal speeds.

Within the ice, IceCube has installed a network of more than 5000 digital optical modules, or “strings” that detect the presence of that Cherenkov light generated by the particle traveling through the ice. The data collected by those optical modules allow scientists to reconstruct the path, momentum, energy, and other properties of those charged particles generated by interacting neutrinos. Remarkably, IceCube recently determined that there was a very high (4.5-sigma, or only a 0.00034% chance of a fluke) probability that high-energy neutrinos, of TeV energies and above, were coming to us after originating from within the Milky Way’s galactic plane.

Based on many years of IceCube data, we’ve been able to map out the origin location of neutrinos in the sky, identifying a clear signature of the galactic plane in neutrinos. This points out a number of candidate origin sources for various galactic ultra-high-energy cosmic rays, with complementary observations in the electromagnetic spectrum strengthening the case.

Independently, back in April of 2019, the Large High Altitude Air Shower Observatory (LHAASO), located in China, began science operations. At very high altitudes, three enormous pools of water, containing a dozen telescopes between them, are designed to capture those same high-energy Cherenkov photons emitted by fast-moving cosmic rays that strike those pools. Upon beginning science operations, LHAASO began seeing extremely high energy cosmic particles, above even 1 PeV in energy, in significant abundance. Because of their design and capabilities, LHAASO is capable of determining the direction of origin of the highest energy cosmic rays that come in, even leading to the identification of new structures within our own galaxy.

The first LHAASO catalogue was publicly released in 2024, wherein it featured a whopping 43 ultra-high-energy cosmic ray sources. The types of cosmic rays that LHAASO is sensitive to are primarily focused on what were initially gamma-ray photons, which arise when more common particles (protons, ions, electrons, etc.) get accelerated by strong electric and magnetic fields in their initial astrophysical environments and smash into other particles present there, creating ultra-high-energy gamma-ray photons in the PeV energy range in the process. Those photons then travel in a straight line, and when they happen to strike Earth’s atmosphere, they produce the signals seen by LHAASO.

This map shows a 1-year view of the entire gamma-ray sky from NASA’s Fermi satellite. The growing-and-shrinking sources are active galaxies powered by supermassive black holes, but the transient “blips” that appear are gamma-ray bursts. The galactic plane remains an intriguing source of many signals, including gamma-rays that may be linked to the ultra-high-energy cosmic rays that we find.

The big question about these PeV-and-up energy cosmic rays used to be, “Do PeVatrons, which would be the astrophysical engines that produce these ultra-high-energy cosmic rays, exist within our galaxy?” We now have the ability to determine the answer to a few important questions.

• Are detectable neutrinos, including neutrinos at high energies, arriving from a particular location in the sky?
• When we see cosmic ray showers originating from gamma-rays striking the atmosphere, do they point back to an object of interest that exists within our galaxy, and a possible source of those gamma-rays?
• And when we combine the signals that we do and don’t see, do they support a scenario where both electrons and hadrons (i.e., particles made of quarks) were produced, or where only electrons were produced?

The ability to answer questions like these allows us to make a remarkable leap: from asking whether PeVatrons exist within our galaxy to asking the much more pointed question, “What are the PeVatrons and where are they?” It turns out there are a few different candidates for the types of astrophysical environments that can create these PeV-energy particles.

In this artistic rendering, a blazar is accelerating protons that produce pions, which produce neutrinos and gamma rays. Photons of all energies are also produced. Extreme events in energy are generated by processes occurring around the largest supermassive black holes known in the Universe when they’re actively feeding, as well as around highly magnetic neutron stars and actively feeding stellar mass black holes. Active black holes, even within our galaxy, can be incredibly powerful sources of energetic cosmic rays.

If you see neutrinos, that tells you that hadrons were part of the story. You only produce neutrinos from processes that involve decays from hadronic states, and so that means you have to have an environment that’s capable of accelerating protons and other heavier atomic nuclei. That means you need the extreme conditions found only in a few cosmic environments:

• in new star-forming regions,
• in recent supernovae and new supernova remnants,
• and around active, feeding black holes.

In these hot environments, atoms become ionized, and then the charged particles get accelerated by the strong electromagnetic fields present within them. They get accelerated to such great energies that when they slam into the material surrounding them, they produce all sorts of secondary particles — including neutrinos and gamma-ray photons — that fly off in all directions.

But it’s also possible to have an environment that doesn’t efficiently accelerate heavy particles like protons or ions, but instead only is effective at accelerating the lightest charged particles: particles with the highest charge-to-mass ratios. Those particles are known as electrons (and their antimatter counterpart, positrons), and there is a type of location that accelerates them and them alone: pulsar wind nebulae.

This full-scale view of the Crab Nebula, from upper-right to lower-left, spans about 11-12 light-years in extent at the nebula’s distance of ~6,500 light-years. The outer shells of gas are expanding at around ~1500 km/s, or about 0.5% the speed of light. This is perhaps the best studied supernova remnant of all-time.

The Crab Nebula, above, is shown in a rather unfamiliar view: in infrared light as seen by JWST. Although many of the features might seem familiar, there’s something new that appears to JWST’s eyes: the wispy, smoke-like strands filling the interior. Those are accelerated electrons, being sped up by the electromagnetic fields generated by the intensely spinning central neutron star. Sure, those electrons can slam into their surroundings and produce high-energy photons — gamma rays — that can then travel in straight lines just like any other gamma ray: right up until the moment they arrive here on Earth and make cosmic ray particle showers that our instruments can then detect.

That’s a key aspect to understand: when you have electrons only generated by your astronomical source, such as from a pulsar wind nebula, you won’t get neutrinos or any other hadronic signature. Although there are several possible astrophysical environments that could serve as PeVatrons, the only candidate that accelerates electrons alone, and not protons or ions, is a pulsar wind nebula.

This is why it was so exciting, just a few months ago, when a gamma ray-induced cosmic ray event, 1LHAASO J0343+5254u, was successfully traced back to a place in the sky, within the galactic plane, that just happened to house a candidate location for its emergence: a pulsar wind nebula.

By tracing a Cherenkov telescope event back to its point of origin in space, and then by observing that location in X-ray light, scientists were able to associate and identify a pulsar wind nebula with a recent LHAASO event. We are now mapping out the origins of galactic cosmic rays at the highest energies that they appear in.

The key was to conduct follow-up observations with an X-ray telescope: in this case, XMM-Newton. When you see cosmic rays that come in with energies of around ~100 TeV or more, that tells you that the original event that generated it was probably at least 10 times as energetic, making it a PeVatron event. This one event had energies that went up to about 200 TeV, and it appeared as an extended source in the sky: spanning nearly a third of a degree. There wasn’t a low-energy counterpart to be seen, which disfavors most of the hadronic options for the cosmic ray’s origin, but the discovery of a pulsar wind nebula as a likely X-ray counterpart is profound.

For one thing, it establishes this particular PeV-energy event as a pulsar wind driven cosmic ray source: one that’s purely leptonic with no hadronic counterpart. For another, it takes us into the era where we can actually trace back these arriving ultra-high-energy cosmic rays to a source within our own Milky Way, and in at least one instance, it’s uncovered the nature of the source. Note that there are other identifications that have been made: molecular clouds have been detected in the region of space associated with the PeVatron event LHAASO J0341+5258: a likely hadronic event. And with follow-up observations by Swift, the PeVatron event 1LHAASO J1928+1813u was discovered to have a low-energy counterpart, further evidence for a hadronic option.

This Fermi-LAT map of the region near where the LHAASO cosmic ray arrived from shows off very strongly a candidate location for the emergence of these cosmic rays: according to a Test Statistic map, this is the likely origin source.

It’s funny that when most of us hear the term “multi-messenger astronomy,” we still think about the merging neutron star pair we found back in 2017, where we detected:

• gravitational waves from the inspiral and merger,
• gamma-rays emitted just 1.7 seconds after the merger ended,
• and then a follow-up afterglow from all across the electromagnetic spectrum.

That was indeed an example of multi-messenger astronomy, but any combination of at least two of light, particles, or gravitational waves is also an example. If you can synthesize together observations of Cherenkov light arising from cosmic rays (particles), neutrinos (as seen by IceCube), and then add in direct data from X-ray and gamma-ray observatories, you’ll have the opportunity to uncover the source from where these signals originated.

Some of them will be star-forming regions. Others will be stellar cataclysms. Some will be supernovae; others will be black holes. Some of them — the ones with no neutrinos — will even have pulsar wind nebulae as the source of their origins. As Prof. Shou Zhang, whose group at Michigan State leads these identification and association efforts, put it, “Through identifying and classifying cosmic ray sources, our effort can hopefully provide a comprehensive catalogue of cosmic ray sources with classification. That could serve as a legacy for future neutrino observatory and traditional telescopes to perform more in-depth study in particle acceleration mechanisms.” Science is a truly global endeavor, and when we share our data and results openly, we can put the pieces together to solve puzzles whose solutions would have remained elusive if we hadn’t combined the evidence from all the different observatories out there.