The neutrino observatory in Antarctica sees evidence of Glashow resonance

The giant underground neutrino telescope IceCube in Antarctica sees a phenomenon called Glashow resonance that was predicted more than 60 years ago. The existence of this process marks yet another validation of the standard model of particle physics.

The IceCube laboratory at the South Pole. This building holds the computer servers that collect data from IceCube’s sensors under the ice.

PHYSICISTS working with the giant instrument called the IceCube Neutrino Observatory— which is buried a couple of kilometres deep under a huge mass of Antarctic ice and is designed to detect ultra-high-energy (UHE) neutrinos from outer space—may have found evidence of a rare neutrino interaction with matter, something that had been predicted to exist more than six decades ago.

The existence of this process, called Glashow resonance, marks yet another validation of the standard model (SM) of particle physics, which has been so remarkable in describing the subatomic world in terms of elementary particles and the fundamental forces of nature (minus gravity) since it was formulated and put on a firm footing in the1970s.

In 1959, the physicist Sheldon Lee Glashow predicted that the collision between electron-antineutrino (the antiparticle version of the electron-neutrino) and an electron at rest via the weak (nuclear) force can, at a particular incident antineutrino energy, give rise to the formation of the (then) hypothetical particle called W-boson through a resonance process. The W-boson, being extremely short-lived, would quickly decay into other particles characteristic of the process, which should be observable.

But why is it called a resonance? We all know what resonance means, for instance, in mechanical systems and musical instruments. It results in the enhancement of an effect at an instant when a certain variable of the mechanical system or the musical instrument, say its frequency, gets tuned to a particular value. In particle physics, it refers to a sudden enhancement in the probability of occurrence of some particle interaction at a particular total energy of the interacting particles. In the scattering between electron-antineutrino and electron that Glashow considered, the production of W-boson shows up as a peak in the cross section (a measure of probability) of that process at a particular energy. We will return to this later.

When Glashow made this prediction, the formulation of even the basic structure of what we today call the standard model was still about a decade away. Of course, Glashow would later become one of the chief architects of the SM itself. On the basis of the independent works of Glashow, Steven Weinberg and Abdus Salam, the SM got properly formulated around 1968, for which the trio was awarded the Nobel Prize in 1979. A bit of history of the theory of weak interactions will help us understand this apparent anachronism in Glashow’s prediction and its relevance to the SM today.

A bit of history

Weak (nuclear) interactions cause the radioactive decay of nuclei (the beta decay) and other elementary particles. In the 1950s, physicists were in the quest for a proper theory to describe these weak nuclear processes. In 1933, the great Enrico Fermi postulated what is known as the four-fermion interaction to describe nuclear beta decay, which was the only weak process known then, with the (four) particles involved in the process interacting at a single point.

Through the 1940s and the 1950s, as more weak processes were discovered and studied, the search for a universal weak interaction theory to describe the various weak processes was vigorously on. Also, elementary particle processes were now beginning to be understood in terms of the exchange of specific particles that acted as transmitters or carriers of particular forces, much like a basketball being lobbed between players, and not as single point interactions. Force between interacting particles was understood to arise from a continual exchange of force carrier particles which are not real but virtual in the sense that energy need not be conserved at the point where the force carrier is emitted and at the point where it is absorbed. Quantum theory allows for such virtual processes of “exchange force” because of the energy uncertainty arising from the Heisenberg uncertainty principle.

The basic approach towards building a universal weak interaction theory at that time was to mimic quantum electrodynamics (QED), the quantum theory of electromagnetism, which had been successfully formulated in the 1940s. In QED, the photon is the carrier of the electromagnetic force, and quantum electromagnetic processes result from the exchange of photons between charged particles.

In a brief paper that was published in April 1960 in Physical Review , Glashow considered the process in the context of what was called the intermediate vector boson (IVB) theory of weak interactions that many people were studying then. In the IVB theory, a charged massive vector particle, the W-boson, mediates weak interaction processes much like the photon (also a vector particle) mediates electromagnetic processes.

Assuming the (energy equivalent of the) mass of the W-boson, the carrier of the weak force, to be between about 500 million electronvolts and 1 billion, or giga, electronvolts (GeV), Glashow calculated that W-boson would be formed as a resonance effect when the incident electron-antineutrino has an energy in the range of 10 ¹¹ to 10 ¹² eV (0.1-1 trillion, or tera, electronvolt (TeV)). He also argued that such high-energy antineutrinos in cosmic rays would invariably be accompanied by muons whose count rate with suitable underground cosmic ray muon detectors would be sizeable enough (0.1-2 per day) to study collisions between electron-antineutrinos and electrons at these energies and look for W-boson formation events. Its detection, he said, would be proof of the IVB theory.

As mentioned earlier, in particle physics, a resonance is said to occur when there is a peak at a particular energy in the graph of number of process events against collision energy. These peaks are usually associated with certain very short-lived particles, with the width of the peak being inversely proportional to the particle lifetime. Since the neutrino and its antiparticle, the antineutrino, have zero charge, the collision of electron and electron-antineutrino that Glashow analysed can take place only through the weak force, which is supposed to occur normally by the exchange of a virtual W-boson. But, in Glashow’s words, “at some [incident] antineutrino energy there will be a resonance, occasioned by the real production of an intermediary boson”. In the context of the IVB theory, the antineutrino energy required for resonance, which goes as the square of the W-boson mass, was some fraction of a TeV.

But it was quickly realised that the IVB theory was only a precursor to a more complete and self-consistent theory. This turned out to be the SM, which evolved subsequently through the decade of the 1960s. The SM, in fact, goes beyond being just a proper weak interaction theory. It unifies both electromagnetism and weak interactions in a single mathematical framework. The mathematical structure of the SM also necessitates invoking an additional massive but chargeless vector particle called the Z ^o -boson as the third weak force carrier besides the charged W ^± -boson.

In weak processes involving the exchange of the neutral Z-boson, there is thus no exchange of charge between the initial and the final states unlike in processes such as beta decay or collision of electrons and electron-antineutrinos when charged W-bosons are exchanged and there is a transfer of charge. The former are called neutral-current (NC) reactions (whose existence was, in fact, a prediction of the SM) and the latter charged-current (CC) reactions. While NC interaction was confirmed in 1973, the actual discovery of the vector bosons W ⁺ , W ^- and Z ^o came in 1983.

So, the W-boson is no longer hypothetical; it is part of the ingredients of the SM. Only its properties differ from what Glashow had assumed, and correspondingly, the experimental set-up to discover the “resonant production” of the W-boson becomes drastically different. The mass of the W-boson in the SM is 80.4 GeV, which is about 80 times what Glashow had assumed for the IVB theory. If you do the calculations for the SM, the resonance production of W-boson should occur for an incident antineutrino energy of about 6.32 peta electronvolts (1 PeV = 10 ¹⁵ eV).

And that is huge, way beyond the range of present-day particle accelerators. The highest energy attained in a terrestrial particle accelerator is 6.5 TeV, in a proton beam at the Large Hadron Collider at CERN in Geneva. The energies reached in accelerator neutrino beams are even less, in hundreds of GeV, four orders of magnitude lower than what is needed to observe Glashow resonance in the laboratory.

However, antineutrinos produced in cosmic processes—such as active galactic nuclei, gamma-ray bursts and violent phenomena such as exploding stars and black hole mergers—can reach energies of up to exa electronvolts (1 EeV = billion billion, or 10 ¹⁸ , eV). But fluxes of cosmic neutrinos and antineutrinos in the desired UHE range are low, and these fluxes also decrease rapidly with increasing neutrino/antineutrino energy (Fig. 1). That is why large neutrino detector volumes are required to observe such UHE neutrinos, and the IceCube neutrino detector has been built with that objective.

A landmark achievement

Situated at the Amundsen-Scott South Pole Station in Antarctica, IceCube comprises a natural transparent detector medium in the form of a cubic kilometre of ice 1.45 km to 2.45 km beneath the surface. Since 2011, when it became fully operational, it has measured neutrino fluxes spanning six orders of magnitude of energy, between 10 GeV and 10 PeV, and is stated to be sensitive to neutrinos with energies beyond 1 EeV.

On March 10, the international IceCube collaboration reported in Nature (a) the observation of an UHE antineutrino with 6.3 PeV energy from an astrophysical source and (b) its interaction with an electron in the atoms within the ice mass of the detector, resulting in what may well be the first observation of Glashow resonance.

The above conclusions were based on the identification of a single event from 4.6 years (55 months) of neutrino/antineutrino interaction data gathered between May 2012 and May 2017. The event, which had the tell-tale signatures of a resonant W-boson production, occurred on December 8, 2016, at around 07:18 hrs IST (Fig. 2). After over three years of intense analysis to confirm that this was indeed a resonance event, the team announced the finding. It is only the third event that IceCube has detected with an incident neutrino energy greater than 5 PeV. The detection of Glashow resonance is clearly a landmark achievement in the decade of research with IceCube.

“IceCube sees about 4,000 events every second,” pointed out Mohamed Rameez of the Tata Institute of Fundamental Research, Mumbai, who was part of the team that carried out the data analysis. “The fact that we could select one [event] that turned out to be special is because of all the knowledge we have built up over the years to characterise the astrophysical neutrino flux,” he said.

“It is truly a major milestone in neutrino astrophysics,” said the particle physicist Vernon D. Barger, Van Vleck Professor of Physics at the University of Wisconsin, Madison, and an expert in neutrino physics, in an email to Frontline . “I compare its importance with the first Omega-minus event [in 1964] that confirmed the Eightfold Way classification of decuplet states in SU(3) by [Murray] Gell-Mann and [Yuval] Ne’eman that led to the quark model,” Barger added.

Since neutrinos (and antineutrinos) are uncharged, they can only be detected indirectly. IceCube detects them by the Cherenkov radiation emitted by the secondary charged particles produced by neutrino/antineutrino interactions in ice (see box). The emitted Cherenkov radiation is detected by 5,160 basketball-sized digital optical modules (DOMs) containing light sensors, which are hung using 86 vertical strings each carrying 60 DOMs. The strings are spaced 125 m apart, forming a hexagonal array in the cubic kilometre of ice that forms the detector. The light gathered by DOMs around the event and the timings of detection by them (which have a high resolution of two nanoseconds) are used to reconstruct the following: the wave front of the emitted Cherenkov radiation, the total visible (light) energy deposited in the instrument by the secondary particles of the event and the incoming direction of the primary neutrino/antineutrino.

“The events for the analysis were chosen based on the signatures they leave within the detector,” explained Rameez. “Roughly speaking, we chose to select only events that clearly started within the detector, though they can go outside afterwards,” he added. A threshold of 4 PeV was set for the deposited visible energy, and the search for such above threshold neutrino/antineutrino events from the entire 4.6 years of data resulted in only one event being picked out. This is consistent with the estimated 1.55 events for that period based on the SM prediction, known neutrino/antineutrino flux on earth and the detector characteristics.

The total light energy this selected event deposited in the instrument was calculated to be about 6.02 PeV. Accounting for an estimated 5 per cent of the energy that would be taken up by particles that do not produce detectable Cherenkov radiation, the authors conclude that this is consistent with a resonant W-boson of 6.3 PeV energy decaying into a shower of strongly interacting particles, such as nucleons and mesons (collectively called hadrons).

As a further consistency check, the team reconstructed the main hadronic shower on the basis of simulations of such events in the SM and the observed Cherenkov pulses. The reconstruction found that some DOMs closest to the event vertex had detected light pulses earlier than the light wave front due to the main shower. As the authors explain, these leading pulses are due to muons arising from decays of mesons in the hadron shower (see box). The authors also performed another consistency check by track tracing and directional reconstruction using the early pulses alone. This showed that the muons and the hadron shower travelled along the same general direction as would be expected from relativistic kinematics considerations. From the light energy amount deposited by the early pulses, the energy of the leading muon was estimated to be about 26.5 GeV, which is stated to be consistent with simulations of hadronic shower from the Glashow resonance.

The authors considered two possible event scenarios that could, in principle, mimic the above observations: One, an UHE cosmic ray–induced atmospheric muon deposits 6 PeV of energy on the detector surface and creates the observed cascade as well as the early pulses. Two, UHE charmed mesons directly decay in the atmosphere into UHE neutrinos accompanied by muons. But the expected rate of such events turns out to be exceedingly small: 1-2 × 10 ^-7 events in 4.6 years. Given the highly negligible background arising from these, the authors conclude that the event was the result of a single UHE astrophysical antineutrino interaction.

Similarly, the major confounding backgrounds to the hadronic shower from a Glashow resonance event are the hadronic final states from CC and NC interactions of electron-neutrinos (and antineutrinos) with nucleons in the detector medium via the exchange of (virtual) W ^± bosons and Z ^o bosons respectively. However, the authors reject these alternatives in favour of Glashow resonance on the following grounds. The leading muon energy measured in the event is an order of magnitude higher than the muon energy expected in a CC event. In the NC case, on the other hand, much higher incident neutrino/antineutrino energy is required to produce the observed energy of the hadronic final state. Also, according to simulations, the expected event rates from these channels are much lower in comparison.

According to the authors, the conclusion that the antineutrino is of astrophysical origin is, in statistics parlance, at a 5-sigma level, which is equivalent to saying that the odds against it are one in 3.5 million. On the other hand, the conclusion that the event is the decay of Glashow resonance at 6.3 PeV could be made only at a 2.3-sigma level, which is equivalent to the odds being one in hundred. That is, there is 1 per cent chance that the event could have been due to off-resonance CC and NC interactions of neutrino/antineutrino with nucleons, called deep inelastic scattering.

In particle physics, a 5-sigma statistical significance is considered a definitive discovery. So, while the observed antineutrino is undoubtedly of cosmic origin, how unambiguous is the claim of its being a Glashow resonance at 2.3-sigma statistical significance?

“The event is not unambiguously a GR event,” said Rameez. “We acknowledge that there is a 1 per cent chance that it is a charged current interaction. However, the 1 per cent is completely determined by the known astrophysical neutrino flux as well as the standard model prediction for the charged current interaction cross section and the Glashow resonance cross section. We hope to see more events in the future to be able say that the GR exists with more confidence.”

“The probability that it is on the resonance rather than off resonance is compelling,” said Francis Halzen, the principal investigator of the IceCube observatory, in an email to Frontline . “But to separate the resonance from normal deep inelastic scattering events will take more than one event, of course,” he added.

The IceCube press release has quoted Glashow himself as saying: “To be absolutely sure, we should see another event at the same energy as the one that was seen. So far there’s one, and someday there will be more.” “There are two ways to increase the number of events: better performance of [IceCube]-Gen1 and building Gen2,” said Halzen. “We are pursuing both. With Gen2 we will see more than one event per year,” he added. IceCube-Gen2 will be the upgraded version of the present detector and will be able to see 10 times more events of interest than Gen1 currently does every year. Costing $350 million, Gen2 is expected to be completed by 2033.

The detection of Glashow resonance is also important for the development of neutrino astronomy itself. Neutrino experiments so far have been unable to discriminate between neutrinos and antineutrinos. Since the event is unique for electron-antineutrinos, the finding constitutes the first measurement of the antineutrino component in the astrophysical flux. “A statistically significant measurement of the Glashow resonance event rate thus directly probes the antineutrino fraction and helps constrain the neutrino production mechanism(s),” says the Nature paper.

“The nature of sources producing the ultra-high energy neutrinos is as yet unknown,” said Raj Gandhi, a neutrino physics specialist at the Harish-Chandra Research Institute, Prayagraj, Uttar Pradesh. “The finding will help in understanding the nature and modelling [of] the physics of these sources, and help unravel the mystery of the origin of the highest energy neutrinos, especially as more such events accumulate in the IceCube detector,” he added.

The present finding of Glashow resonance corresponds almost exactly to the theoretically predicted energy of 6.3 PeV. It thus represents the highest energy test of a specific process in the SM. Notwithstanding this, and the SM’s enormous success otherwise in describing the world with great precision, it remains an incomplete theory. Its deficiencies include its inability to include gravity among the forces it unifies in a single mathematical framework and its failure to explain, for example, matter-antimatter asymmetry in the universe, dark matter/dark energy and the fact that all neutrinos have mass, however small. Its validity even at the PeV energy scale only tells us that even higher energies need to be probed for any evidence of “physics beyond the standard model”.