The elusive 'God particle'

PROF. ROLF HEUER (second right) with other international physicists at the Lepton Photon 2011 conference in Mumbai on August 25. - SHIRISH SHETE/PTI

The Lepton Photon conference in Mumbai ended with a feeling that the Higgs particle perhaps does not exist, but physicists are not giving up.

IT was widely expected that the recently concluded biennial XXV International Lepton Photon Symposium 2011 (LP11) at the Tata Institute of Fundamental Research (TIFR), Mumbai, would witness the historic announcement of the discovery of the Higgs particle that the high-energy physics community was desperately seeking. Unfortunately, it was not to be, and the particle continues to be elusive even after a nearly three-decade search. But new results from experiments engaged in the Higgs search provided more stringent limits on its existence than those presented as recently as July at the Europhysical Society's 2011 High-Energy Physics Conference (EPS-HEP) in Grenoble, France. The Mumbai conference ended with a distinct feeling that the particle perhaps does not exist, but physicists are not giving up as yet even as the window still available for Higgs to hide has really shrunk as it would otherwise throw the entire theoretical framework for fundamental particles and forces of nature completely out of gear.

The Higgs particle, or the God particle as it is commonly referred to in popular media, is the only crucial missing piece in the otherwise enormously successful theory of fundamental particles and the forces of interaction among them. The particles included in this framework are leptons (the light ones), which include electrons and neutrinos, and hadrons (the heavy ones), which are particles made up of quarks like protons and neutrons. The forces that the theory describes include the electromagnetic force mediated by the mass-less photon, the weak nuclear force mediated by the massive particles called W and Z vector bosons, and the strong nuclear force mediated by the eight mass-less particles called gluons.

Known as the Standard Model (SM), the theory has superbly held up to very high precision tests of its predictions assuming the existence of Higgs. It is through their interaction with the hypothetical Higgs (or the lack of it) that all other particles get their mass (or correspondingly remain mass-less). The Higgs particle, too, gets its mass in this way through self-interaction. Higgs can be imagined as an all-pervasive ether-like field, which endows particles with mass (or inertia) because of the drag that the field exerts on all particles by sticking to them as they move through space. More technically speaking, the mass arises through a mechanism known as spontaneous symmetry breaking by the Higgs field of a certain universal symmetry that prevailed at the Big Bang.

However, the SM does not predict a value for the mass of the Higgs particle though the model indirectly constrains the value to be under a certain value (186 gigaelectronvolt (GeV)) based on measurements of some processes, like particle decays, governed by the model. In high-energy physics, masses are measured in units of energy in accordance with Einstein's mass-energy relation. The mass of a proton, for example, is about 1 GeV, that of W and Z is about 100 GeV. Since the 1980s, physicists have been scanning the entire energy range available to the various high-energy particle accelerators in which high-energy particle collisions occur producing a multitude of particles. In the debris of these collisions, scientists look for signatures of Higgs.

As accelerator technologies have advanced a great deal during this period, and the energies accessible in these have correspondingly increased by orders of magnitude, techniques of detection and analysis, too, have become commensurately complex and sophisticated. But Higgs, without which it now seems impossible to explain the vast amount of particle data, has remained elusive. Higgs has thus become a sort of Holy Grail of current high-energy particle physics. Its existence, or an equivalent mechanism that not only gives rise to mass but is also consistent with all the precise measurements made on SM processes, is imperative for a consistent picture of the subatomic world.

At present, the most powerful particle accelerator is the Large Hadron Collider (LHC) at CERN (European Organisation for Nuclear Research) in Geneva, which after its initial mishap and early hiccups began running at a reduced total energy of 7 teraelectronvolt (3.5 TeV per proton beam) in November 2009. The LHC began taking data at 7 TeV in March 2010 and is expected to run through December 2012 at this energy when it will be shut down for about 18 months before it is brought up to its peak design value of 14 TeV (7 TeV per beam) in 2014. But at 7 TeV itself almost everything of interest with regard to Higgs and much of the other new physics that is expected to show up in the TeV energy scale, such as dark matter, dark energy, matter-antimatter asymmetry in the universe, super symmetry and extra dimensions, should show up. The combined results of the nearly identical multipurpose ATLAS and CMS experiments, which are key to the search for Higgs, were supposed to be the high point of the Mumbai meeting.

The front runner in the Higgs search was the Large Electron-Positron Collider (LEP), which operated between 1989 and 2000 and attained a maximum collision energy of 209 GeV. With the combined data from four of its experiments, the LEP ruled out the existence of Higgs with a mass less than 114.4 GeV at 95 per cent confidence level (CL). The terminology confidence level is a statistical measure and it refers to the number of times the result of the experiment repeated 100 times will meet the expectation within the specified range. That is, 95 per cent CL means one is likely to be proved wrong in this case Higgs-like signal will show up in that range five times out of 100. The LEP, however, produced tentative but inconclusive hints of Higgs with a mass of about 115 GeV before it shut down in 2000. The search was followed up by the next-generation accelerator at Fermilab, United States, called Tevatron proton-antiproton collider. The accelerator is so named because it brought proton and antiproton beams, each accelerated to about 1 TeV, to collide head on.

In order to put these ongoing searches in proper perspective, it should be appreciated that a discovery in particle physics is often a long, painstaking process, requiring sifting through huge amounts of data to isolate the signal of a rare process, such as the signal from Higgs. The lifetime of Higgs itself is very short, but it is its various decay channels into other subatomic particles, according to the SM, that will provide the signal. For example, the following channels are available for Higgs to decay into: H two photons; H two tau leptons; H WW; H ZZ and others (see picture). The real complication arises from the background events, which are those from other processes in the SM that mimic the characteristics of Higgs decay. A Higgs signal would appear as an excess number of events over the large background. As CERN's director-general Rolf Heuer said, It is like trying to capture the photograph of snowflakes against the backdrop of a snowfield. The effort is to pick up such excess, which are statistically significant. Statistical significance is measured in terms of what is called standard deviation (called sigma). For any discovery in particle physics, the signal should be at least at 5 sigma level over the background, which is equivalent to a CL of being wrong only one part in a billion.

To improve the statistics, we need more collisions so that events producing the signal we are looking for stack up and give a statistically significant peak of excess events over the background. At the Tevatron, for example, the number of collision events that the experimenters had recorded in their Higgs search from 2001 until now was 700,000 billion proton-antiproton collisions. According to the Italian physicist Marco Verzocchi of the National Institute of Nuclear Physics (INFN), Rome, who presented the latest Tevatron data at LP11 in Mumbai, We have already collected 95 per cent of the data [that Tevatron can potentially deliver]. Our data collection will end in the next one month [when the machine will shut down for ever].

To put the above in perspective, the total data collected by the LHC during the first few months of its running amounted to 3,000 billion proton-proton collisions. In April, the LHC overtook the highest beam intensity (called luminosity) achieved so far at Tevatron and recorded the highest intensity. A higher energy machine requires a much higher luminosity to yield far greater number of collisions because of the much larger number of channels available at the higher energy resulting in a much larger background. In June 2011, the LHC attained the key data milestone one inverse-femtobarn (fb -1) required for being able to see statistically significant physics results at 7 TeV, which is equal to 70 trillion proton-proton collision events. This means that the LHC has recorded this much of data since it began operations in March 2010. This was the target set for the LHC in 2010 for the entire 2011 runs, but this was achieved within a record time of three-four months. This data formed the basis for the results presented in Grenoble. The speed with which the LHC experiments have been analysing data has also been unprecedented. The Worldwide LHC Computing Grid, which links up computer centres around the world (including India), was routinely processing up to 200,000 physics analysis jobs concurrently.

Improved results

Between July and the date of the Mumbai conference, the data collected had nearly doubled, enabling much improved results to be presented at LP11. According to Vivek Sharma of the University of California at San Diego (UCSD), who presented the CMS results in Mumbai, the total amount collected by the LHC experiments up to August 4 was around 120-130 trillion collisions. This is only a tenth of the data that the LHC is capable of delivering at 7 TeV, which the LHC plans to reach by the year end to lay the basis for an even more improved data sample in 2012.

Figure 1 shows the excluded mass regions for Higgs as a result of these various experiments until March 2011, when the LHC results had not yet started coming in. The results on Higgs search from Tevatron's CDF and D-zero that were presented in Grenoble excluded Higgs in the mass ranges 100-109 GeV and 156-177 GeV at 95 per cent CL. From Tevatron's perspective, therefore, Higgs with mass in the range of 115-155 GeV is still not excluded. In January 2011, the U.S. Department of Energy (DoE) turned down scientists' request for a three-year extension up to 2014. The scientists had felt that a final assault on the still-allowed region of 115-155 GeV could have yielded definitive result about the existence of Higgs with mass in that range.

Since Tevatron had collected almost all the data that could be potentially delivered by the machine, and is due to be shut down in September, the results from its two experiments presented in Mumbai were virtually the same as those presented in Grenoble. We now have to really do the final analysis of the data. But we are a much smaller team as compared to the LHC and it takes a bit longer to analyse the data, said Verzocchi. We are at 98 per cent of our sensitivity and we are trying to get the last few per cent. That takes a little bit longer. We are trying to do the most difficult things. We are almost near saturation in what can be achieved at Tevatron.

Before the LHC results were presented in Grenoble, an internal note claiming significant excess events indicative of a Higgs signal somehow leaked out into a blog, Not Even Wrong, maintained by a mathematician, Peter Woit, which led to a rumour among people not associated with the experiments and the media of Higgs having probably been seen. The note described highly preliminary data from observations which were yet to be vetted, scrutinised for the quality of the data and statistically analysed to see whether there was a true signal or not before making the data public. The claim was that in the channel H 2 photons, the ATLAS experiment was seeing more events than expected (almost 30 times!) at a mass value of 115 GeV.

Since this was the value at which the LEP had seen hints of a Higgs signal, the rumour gained ground pretty fast. It had to be finally quashed by an official statement from ATLAS. Signals of the kind reported in the memo, said the ATLAS spokeswoman, show up quite frequently in the course of data analysis and are later falsified after more detailed scrutiny. Only official ATLAS results, which have undergone all the necessary scientific checks by the collaboration, should be taken seriously. The ATLAS collaboration involves 3,000 scientists from all over the world who analyse every detail of the data before it can be publicised.

In Grenoble in July, the LHC finally entered the Higgs game and presented its first results on the excluded mass regions for Higgs from the ATLAS and CMS experiments. ATLAS ruled out the regions 155-190 GeV and 295-450 GeV at 95 per cent CL and CMS excluded the mass regions 149-206 GeV and 300-440 GeV at 95 per cent CL. But more tantalising was the moderate excess events seen in the data presented for the Higgs decay into WW channel in the low mass region. ATLAS saw excess events in the mass range of 120-145 GeV. Interestingly, the entirely independent CMS experiment too saw excess events roughly in the same mass region. The intriguing aspect of this is that CMS was completely unaware of the ATLAS data before the conference.

However, the statistical significance of these excesses seen in both the experiments was about 2.7 sigma. This implies an 8 per cent chance of the excess being produced by statistical fluctuations in the data. This, as we mentioned earlier, is far below the 5 sigma gold standard for a discovery. But the fact that both the ATLAS and CMS experiments had seen the same kind of excess did cause some excitement and flutter among the 750 physicists gathered in Grenoble, leading to unwarranted hype in the news media as a possible hint for a Higgs boson. The ATLAS group did not attach much importance to this blip but merely noted in its release, In some regions, there are small excesses above expectations. Likewise the CMS release said, It should be noted that modest excess of events is observed for Higgs boson masses below 145 GeV, and added, With the data that we will collect in the next few months, we will be able to distinguish between the production of a Higgs boson or a statistical fluctuation of the backgrounds.

ATLAS' Higgs search results presented in Mumbai covered as many as nine decay channels. In particular, a new channel, H 2 tau leptons, was analysed (which is plagued by background at lower data rates) and the channel known as the golden mode, where H 4 leptons were analysed in much greater detail with the increased data. According to Sharma, the latter is so called because it is a clean decay channel for analysis as both the ATLAS and CMS detectors were built to detect exactly these events and reconstruct them fully. This means all the four particles can be reconstructed completely and the mass of Higgs that gave rise to these can be determined very accurately.

The latest data, however, excluded more mass ranges beyond those presented at the EPS in July. As the ATLAS release said, 85 per cent of all mass regions below 466 GeV are excluded at 95 per cent CL. The windows that remained open for possible Higgs discovery, it said, were 115-146 GeV, 232-256 GeV, 282-296 GeV and any mass above 466 GeV. Similarly, the expanded exclusion regions for CMS were 145-216 GeV, 226-288 GeV and 310-400 GeV. What this means is that there is no excess that we think is relevant for the first hint of any signal, said Aleandro Nisati of INFN, who presented the new ATLAS data. This is not the same as saying that there is none. What we are saying is that in the entire region of 114-200 GeV we have not reached that level of confidence to say that there is a signal, he added.

The CMS release similarly pointed out, For the quantity of data we have collected, on average we would expect to exclude the range 130-440 GeV in the absence of a signal. We believe that the differences between the expected and observed exclusion mass ranges are consistent with statistical fluctuations. And at 90 per cent CL, CMS excluded the mass region 144-440 GeV. What about the region 440-600 GeV? Sharma pointed out that in that region, the experiments did not have much sensitivity and in any case there were other constraints that indirectly ruled it out though one would like to exclude it by direct experimentation perhaps at the peak energy of 14 TeV.

The elusive Higgs particle, if it exists, is running out of places to hide, said the CERN press release of August 22. It announced the excluded region as roughly 145 to 466 GeV by combining the results of both ATLAS and CMS. Actually, combing the results requires a detailed statistical analysis, which is yet to be done. Indeed, new data (Figure 2) have excluded vast swathes of the real estate for Higgs to hide. CMS and ATLAS are equals, by construction and design. They should have the same sensitivity. If you take the ATLAS and CMS results together, Higgs cannot hide at least in the high mass region [above 155 GeV], Sharma pointed out. According to him, even the couple of narrow allowed windows that may seem to be there at higher masses will disappear if the two sets of data were combined with proper statistical analysis.

Indeed, this was what the high-energy physics community was expecting to see at the Mumbai conference. After several months of taking additional data, the options [for this conference] were to either combine the ATLAS and CMS data, or independently present new data from both detectors, said Heuer. It was decided to independently present the analysis from additional data, and we see a convergence in the exclusion zones from the two experiments. The data excludes the higher mass range, and the mass range window has become narrow, he added.

From the ATLAS and CMS data, one can now see that only the narrow low-energy window where Higgs, if it exists, could be lurking is between 115 GeV and 145 GeV. According to Heuer, Higgs can show itself in many ways in the lower mass range. The problem, he said, is that Higgs boson signatures in the low mass range are less frequent and they also look very similar to the anticipated background, thus proving hardest to find where it is now expected to be observed.

In fact, in the low-energy region, Tevatron may have a slight edge over the LHC in being able to pick up Higgs if it shows up in the next one month of Tevatron's analysis. If the mass is, say, below 125 GeV, we would be in a neck-and-neck competition, said Verzocchi. If it is below 120 GeV it could be a little bit easier for us than for the LHC because of the background and the process you are looking at. For example, below 125 GeV, the LHC has to look for Higgs 2 photons and this happens to be very luminosity-sensitive. We look mostly at Higgs b-quark + b-antiquark. At Tevatron we don't have so much background in this channel. We can do with less luminosity. So in that specific region there is complementarity between the two machines. The LHC will take a few more years, but for us that channel will be able to say something if it is between 115 and 125 GeV, Verzocchi said.

The excess seen in the low-mass region in Grenoble has persisted in the results presented in Mumbai, though there has been no enhancement of this excess in spite of higher luminosity and nearly double the data collected. The ATLAS release, however, said, but we are still in the early days in this search. The CMS release said the issue should be resolved with increased data collection in the coming months. Actually speaking, the excess may have marginally reduced; from a 2.7 sigma enhancement, in Mumbai it was down to about 2.1-2.2 sigma.

Two sigma excess is something that you notice but should not get excited about, Nisati pointed out. What we hope to get by combining more than one big experiment is 4-5 sigma effect. By the end of the year, we should be able to tell if it is excluded from here as well.

Similarly, CMS observed a slight excess in the Golden Channel at Higgs' mass values of 122 GeV, 142 GeV and 165 GeV with the new data. And only around 142 it smells like that it has come from Higgs, the rate at which it has come, said Sharma. The others are more likely to be from the background, from some other process.

The next step will be to focus on all regions, including the excluded regions, in order to increase the confidence level that there is indeed nothing there. The same final states that you used to search for Higgs can be used for searching something else, pointed out Nisati. If you see some signal even if at a much lower rate than what is predicted for SM Higgs, it could be through some other kind of new physics. For SM Higgs, the study has to focus on a region which is very narrow [115-145 GeV] where a lot of effort will now be focussed, Nisati added.

We will triple the data set by end of October, said Sharma. Tripling the data set means statistics fluctuations are going to be that much smaller. And if you combine both the CMS and ATLAS data, we should be in a position to close the gap and we will know if Higgs doesn't exist. That is, it was just the ether of the 21st century. We will know if it was indeed science fiction by the end of the year, he added.

What if Higgs is not found at all? Then we have chaos, said Sharma. It cannot get more explosive than this. For more than 20-25 years, we have done very precise measurements of all the quantities that the SM predicts and almost every one of them has been right on the money. So if Higgs isn't there, there is no other theory that I know which completely explains all the known measurements and says there is no Higgs. Super symmetry, a theory describing a higher level of symmetry in the universe beyond the symmetry in the SM, is such a theory. Super symmetry requires super-partners to known particles of the SM (called sparticles) to exist and, if these exist, they are expected to show up at the LHC. However, the simplest super symmetry model seems very unlikely as the data from the LHC experiments show.

The LHCb experiment made high-precision measurements of decays of particles called B-mesons. If super symmetric particles existed, B-mesons should have decayed more often, violating SM predictions. Earlier measurements at Tevatron had suggested that the decay of B-mesons was more than SM predictions and was indicative of evidence of sparticles. However, the new LHCb data have shown that these decays are in full agreement with the SM with barely any room to accommodate sparticles. It has thus reinforced its earlier data in Grenoble that had overturned Tevatron's claim. ATLAS and CMS too had ruled out the existence of sparticles below 900 GeV mass at 95 per cent CL at Grenoble. The new data should, therefore, come as a big blow for proponents of super symmetry.

But the current energetic and frenetic experimental activity at the LHC is bound to throw up new physics, if not Higgs, pretty soon. As Sharma said, like a true experimentalist, I don't want to miss Higgs if it exists and I don't want to find Higgs if it does not exist. But not finding Higgs or super symmetric particles are also significant discoveries in themselves. In addition to the already existing outstanding questions, these negations will only expand the unknown vistas in high-energy physics.