żěè¶ĚĘÓƵ

Out with a bang: The Tevatron’s deathbed revelation

With just months before the iconic particle accelerator is dismantled, physicists are racing to confirm a tantalising glimpse of unexpected new particles
The Tevatron's CDF detector was upgraded in 2001
The Tevatron’s CDF detector was upgraded in 2001
(Image: Fermilab)

With just months before the iconic particle accelerator is dismantled, physicists are racing to confirm a tantalising glimpse of unexpected new particles

TOM SCHWARZ describes his observations as a Lothario might describe his lovers. “They kind of come and go,” he says. Such insouciance is a necessary part of being a particle physicist: Schwarz’s workhorse, the at Fermilab in Batavia, Illinois, delivers information about millions of particle collisions every second. This deluge of data can often take on intriguing and seductive forms, only to be revealed as plain and uninteresting when more numbers have been crunched or the scrutiny is intensified.

Recently, though, Schwarz has been smitten with a result he first spotted as a PhD student in 2006. “It is the longest relationship I have ever had with a measurement,” he says. At first glance it seems a mundane affair. He and colleagues at the Tevatron’s and experiments have found that the heaviest of the basic building blocks of matter, the top quark, does not emerge evenly in all directions from particle collisions. It shows a distinct preference for one direction over another.

To particle physicists, such an asymmetry is a teasing display. If it doesn’t stop soon, it could well put their long-term relationship with their beloved “standard model” under strain, pointing the way to some deeper, more meaningful insight into the workings of nature. Experience says such flings usually don’t last – but this one is proving hard to end.

In its illustrious 25-year history, the Tevatron has deepened our understanding of nature’s fundamental particles and forces. Its high-water mark came in 1995, when both CDF and DZero spotted in the ejecta of head-on collisions between protons and antiprotons. This much more massive cousin of up and down quarks, which make up the protons and neutrons in the atomic nucleus, was the sixth and final quark to be discovered. Confirmation of its existence helped seal the deal on the standard model, our most successful theory of the world at its smallest. Of the particles predicted by the standard model, today only the Higgs boson remains at large.

Later this year, though, the Tevatron is due to be closed down and the contents of its 6.3-kilometre ring dismantled. Under such circumstances, it is not unusual for tantalising signs of new particles or effects to appear. Researchers know their kit inside out, and there is an understandable rush to squeeze every last morsel from the data and vie for the discovery that will cement an experiment’s legacy. It happened in late 2000 at the CERN laboratory near Geneva, Switzerland, when the Large Electron-Positron Collider was due to make way for the Large Hadron Collider (LHC). And just last month, CDF itself published details of another anomalous effect – a bump that suggests the existence of a new particle not predicted by the standard model (żěè¶ĚĘÓƵ, 16 April, p 8).

Sober judgement and a healthy scepticism are required, then, to separate what is real from what is wishful thinking. But there is good reason to believe that the asymmetry effect that has captivated Schwarz and others is genuine. For a start, there’s the fact that both CDF and DZero, two independent experiments that sit on opposite sides of the collider ring, have seen it. There’s also the fact that the effect was first spotted well before the smasher’s final collision date was known – and with more data, it is just getting stronger.

When protons and antiprotons collide at the scorching energies produced by the Tevatron, they are split open to expose their quark innards. These quarks can interact to produce other particles, including top quarks and their antiquark partners. The results from the Tevatron show that when such a pair is created, the top quark is between two and three times as likely to be emitted “forwards” – continuing in the same direction as the incoming proton – as “backwards”, in the opposite direction along the trajectory of the antiproton. Similarly, the top antiquark prefers the direction of the antiproton.

In 2008, CDF reported this effect with a statistical significance approaching 2 sigma, representing a roughly 1 in 20 chance that the apparently meaningful pattern could have arisen through a random fluctuation. DZero also saw the same effect, albeit with a slightly lower significance. Effects of this size come and go all the time in particle physics. At the end of last year, however, having worked through more data, CDF saw a result with a significance of 3.4 sigma for a select bunch of the most energetic top quark-antitop quark pairs – corresponding to odds of about 1 in 1500 that such a pattern could have emerged by chance alone (). Meanwhile, last year DZero in a larger data sample.

All this is highly problematic for the standard model. Quarks interact via the strong nuclear force, and the part of the standard model describing this force, a theory called quantum chromodynamics (QCD), embodies fundamental symmetries that mean quarks and antiquarks are treated equally. According to the simplest QCD calculation of how top quarks are produced, then, no forward-backward asymmetry should arise at all.

“The simplest calculations suggest there should be no asymmetry at all in how top quarks are produced”

In reality, the picture is more complex. When quarks and antiquarks interact through the strong force, they exchange particles known as gluons. Gluons can interact with themselves, or produce other quarks which can themselves beget more gluons, and so on. This leaves anyone using QCD to calculate the outcome of Tevatron collisions with a huge series of interfering quark and gluon configurations to tot up.

The best attempts to do so show that a forward-backward asymmetry is to be expected. A gluon radiated by an incoming quark, for instance, can interfere with a gluon radiated by an outgoing top quark, altering the outgoing top-antitop pair’s travel plans. Such activity means that between about 10 and 20 per cent more top quarks will be emitted forwards than backwards. But that is nowhere near enough to explain the measured asymmetry.

Compelling stuff

“For better or for worse, this effect is different to many others,” says Amnon Harel of the University of Rochester, New York, who works on the DZero experiment. , a theorist at Rutgers University in Piscataway, New Jersey, agrees. “It’s not obvious where there could be a big mistake,” he says. “As far as discrepancies from the standard model go, this is the most compelling that I have seen in many years.”

So assuming the result is not a cruel taunt of nature, one that will eventually up sticks and disappear, what is going on? The theorists have certainly been busy: since details of the asymmetry were first made public, some 25 papers have appeared road-testing explanations for the effect. Contenders range from particles straddling extra dimensions () to invisible and weird “unparticles” (), to the first indications of a 45-dimensional “grand unified theory” ().

A more conservative take is that the QCD calculations of the asymmetry have not yet been fully tamed. An unfortunate cocktail of incomplete calculations, statistical fluctuations of the data and some other experimental subtlety could just about conspire to explain the observed discrepancy, says Strassler.

That is the kind of caution advocated by of the Massachusetts Institute of Technology, who co-invented QCD in the early 1970s and shared a Nobel prize for his efforts. “I expect, and hope, that there is a mundane explanation for the effect” – if indeed there is an effect, he says. Given the detailed and precise tests that QCD has already passed, Wilczek thinks it unlikely that the theory of the strong force needs major rewriting.

Others, though, say it might be too late for such minimalist explanations. To theorist of Stanford University in California, the fact that the effect crops up specifically with the top quark – the heaviest known elementary particle – and especially at high energies, is suggestive. “It is exactly what you would expect to see if a new heavy particle that decays into top quarks were lurking just beyond the Tevatron’s reach,” she says.

“The signal is exactly what you would expect if a new heavy particle were lurking just beyond the Tevatron’s reach”

If there is, it’s something unexpected. The vaunted Higgs boson is not thought to produce such an effect, and nor is any of the particles predicted by simple versions of supersymmetry, the theory hotly tipped to tie up some of the loose ends left by the standard model.

, a theorist at the University of North Carolina in Chapel Hill, thinks he knows what this mystery particle might be. When he heard the news from the Tevatron, he says, he started jumping up and down. By way of explanation, he points to the licence plate of his car, which reads, simply, “AXIGLUON”.

“It’s one of my long-shot dreams of winning a Nobel prize,” he says. He thinks the signal seen by CDF and DZero has all the hallmarks of an exotic particle whose existence he predicted in 1987 together with of Boston University in Massachusetts. Whereas the familiar strong force carried by gluons is symmetric with respect to left and right (a so-called parity symmetry), axigluons would carry a skewed variant of the strong force capable of generating the asymmetry observed at the Tevatron – even without fancy calculations.

If this picture is correct, it would rock the symmetries underpinning the standard model. The strong force would be exposed as just one facet of a primordial superforce that split into two as the universe cooled and became less energetic – just as today’s weak nuclear and electromagnetic forces, we now know, were once combined in one “electroweak” force (see diagram).

Breaking symmetry

Eyes on the LHC

Not everyone is as keen on the idea as Frampton. Glashow, who shared the 1979 Nobel prize in physics for his work on the theory of the electroweak force, is not convinced that the axigluon can explain the Tevatron discrepancy, or even that the discrepancy is real. And of Michigan State University in East Lansing has concluded that the simplest axigluon models, such as Frampton’s, have already been ruled out. That is because axigluons should increase the rate at which particles made of a bottom quark, the top’s slightly lighter sister, and a down antiquark oscillate into their antimatter counterparts – yet no such effect has been observed ().

But Frampton reckons it is more credible to dig out a 20-year-old paper than to cook up a model with a new particle that has been customised to fit the data. That said, his original model predicted an asymmetry of the wrong sign, with more tops going backwards than forwards. He has fixed that, but only by postulating the existence of an unseen “fourth generation” of quarks even more massive than the top and bottom ().

All this leaves things a little in limbo. “We’re currently really living this to-and-fro of theory and experiment,” says Strassler. With just months left to run, the Tevatron seems unlikely to settle the issue conclusively, and all eyes are now on the LHC to do the job. After a year of running at reduced performance, the LHC is now flexing its muscles and has already produced thousands of top quarks, the first to be made outside of Illinois.

There is just one problem in measuring a forward-backward asymmetry at the LHC: it does not have a “forwards” or “backwards”. It does not smash antiprotons into protons, but protons into protons in a totally symmetric way that leaves no chance of reproducing the Tevatron effect. But the machine has so much juice that it should be able to produce the particles relevant to the anomaly directly. First indications are not good for proponents of the axigluon theory. In March, the LHC’s ATLAS experiment ruled out the existence of axigluons with a mass between 0.6 and 2.1 teraelectronvolts, right where Frampton’s putative particle should be ().

There are other ways to coax out an asymmetry – up and down quarks could turn directly into top quarks, for example, by exchanging a new particle. One candidate, snappily labelled the W’ (spoken “W prime”), is an exotic variant of the W bosons that carry the weak nuclear force (). Another claims to explain not only the top asymmetry but also CDF’s “mass bump”, announced last month ().

For Hewett, everything is still to play for. “No matter what piece of physics explains the top quark asymmetry, it has got to show up in this year’s LHC run,” she says. Working out what model any new particle belongs to, though, could take longer.

At the Tevatron, there are no signs anyone is tiring of the chase. In March, CDF reported in an independent set of top-quark data, and DZero is finalising its latest analysis. As the Tevatron tunnel is emptied this autumn, the two experiments will join forces to see how high a significance they can squeeze from their results.

Will this be a dalliance that they will come to regret? of the National Institute of Nuclear Physics in Pisa, Italy, doesn’t think so. One of the leaders of the CDF experiment, he is bullish about the prospect of using the full Tevatron dataset to establish an asymmetry with 5-sigma significance – the 1-in-a-million gold standard for announcing a discovery. Then it really will be love. “I don’t think we will go backward,” says Punzi. “The question is, how far we will go forward?”