Friday 31 October 2008

On CDF Anomaly

Mhm, it seems that I chose the wrong side of the Atlantic. First, the LHC produces a big firework display instead of small black holes, and then the CDF collaboration at the Tevatron discovers new physics. About the CDF anomaly in multi-muon events, see Tommaso's post or the original paper. Together with a few CERN fellows we had an impromptu journal club today, and we have reached the conclusion that, well, we don't know :-). The anomaly occurs in a theoretically difficult region, the B-baryon spectrum is poorly known, the local Monte Carlo magicians are very sceptical about modeling the b-quarks, etc, etc. It does not mean, of course, that one should shrug it off. Whether we want it or not, the CDF anomaly will dominate particle model-building for the next few months.

Meanwhile, there is already one model on the market that, incidentally:-), looks relevant for the anomaly: SuperUnified Theory of Dark Matter. One can immediately cook up $e^N$ variations of that model, but there seem to be 3 basic building blocks:
1) The "visible" sector that consists of the usual MSSM with the supersymmetry breaking scale $M_{MSSM} \sim$ few hundred GeV.
2) The dark sector with a smaller supersymmetry breaking scale $M_{dark} \sim $ GeV. It includes a dark gauge group with dark gauge bosons and dark gauginos, a dark Higgs that breaks the dark gauge group and gives the dark mass to the dark gauge bosons of order 1 dark GeV. In fact it's all dark.
3) The dark matter particle that is charged under the dark group and has a large mass, $M_{DM} \sim $ TeV. Unlike in a typical MSSM-like scenario, dark matter is not the lightest supersymmetric particle, but rather some new vector-like fermion whose mass is generated in the similar fashion as the MSSM mu-term.

The dark group talks to the MSSM thanks to a kinetic mixing of the dark gauge bosons with the Standard Model photon, that is via lagrangian terms of the type $f_{\mu\nu} F_{\mu \nu}$. Such mixing terms are easily written down when the dark group is U(1), although for non-abelian gauge groups there is a way to achieve that too (via higher-dimensional operators). Once the dark gauge boson mixes with the photon, it effectively couples to the electromagnetic current in the visible sector. Thanks to this mixing, the dark gauge boson can decay into the Standard Model particles.

The SuperUnified model is tailored to fit the cosmic-ray positron excess PÀMELA and ATIC/PPB-BETS. The dark matter particle with a TeV scale mass is needed to explain the positron signal above 10 GeV (as seen by PAMELA) all the way up to 800 GeV (as suggested by ATIC/PPB-BETS), see here. The dark gauge bosons with a GeV mass scale play a two-fold role. Firstly, they provide for a long range force that leads to the Sommerfeld enhancement of the dark matter annihilation rate today. Secondly, the 1 GeV mass scale ensures that the dark matter particle does not annihilate into protons/antiprotons or heavy flavors, but dominantly into electrons, muons, pions and kaons. The second point is crucial to explain why PAMELA does not see any excess in the cosmic-ray antiprotons. Supersymmetry does not play an important role in the dynamics of dark matter, but it ensures "naturalness" of the 1 GeV scale in the dark sector, as well as of the electroweak scale in the visible sector. I guess that analogous non-supersymmetric constructions based, for example, on global symmetries and axions will soon appear on ArXiv.

What connects of this model to the CDF anomaly is the prediction of "lepton jets". In the first step, much as in the MSSM, the hadron collider produces squarks and gluinos that cascade down to the lightest MSSM neutralino. The latter mixes into the dark gauginos, by the same token as the dark gauge boson mixes with the visible photon. The dark gaugino decays to the dark LSP and a dark gauge boson. Finally, the dark gauge boson mixes back into the visible sector and decays into two leptons. At the end of this chain we obtain two leptons with the invariant mass of order 1 GeV and a small angular separation, the latter being due the Lorentz boost factor $\gamma \sim M_{MSSM}/M_{dark} \sim 100$.

The perfect timing of the "lepton jets" prediction is unlikely to be accidental. A new spying affair is most welcome, now that the paparazzi affair seems to by dying out. While waiting for CDF to find the traitor and hang him on the top pole, I keep wondering if the SuperUnified model does indeed explain the CDF excess. If you take a look at the invariant mass distribution of the anomalous muon-pair events (right panel) it does not resemble a 2-body decay of a narrow-width particle (for comparison, admire the J/Psi peak in the left panel), which it should if the muons come from a decay of the dark gauge boson. Or am I missing something? Furthermore, it has been experimentally proved that bosons are discovered in Europe, while only fermions can be discovered in the US. This is obviously inconsistent with the Tevatron finding the dark gauge boson ;-)

Thanks to Bob, Jure and Tomas for the input.
See also Lubos' post on the SuperUnified model.
For more details and explanations on the CDF anomaly, see the posts of Peter and Tommaso and John.

Thursday 30 October 2008

PAMELA's coming-out

Yesterday, PAMELA finally posted on ArXiv her results on the cosmic-ray positron fraction. In the last months there was a lot of discussion whether it is right or wrong to take photographs of PAMELA while she was posing. Here at CERN, people were focused on less philosophical aspects: a few weeks ago Marco Cirelli talked about the implications for dark matter searches, and Richard Taillet talked about estimating the positron background from astrophysical processes in our galaxy. Finally, PAMELA had her coming-out seminar two days ago. PAMELA is a satellite experiment that studies cosmic-ray positrons and anti-protons. She has a better energy reach (by design up to 300GeV, although the results presented so far extend only up to 100 GeV) and much better accuracy than the previous experiments hunting for cosmic anti-matter. Thanks to that, she was able to firmly establish that there is an anomaly in the positron flux above 10 GeV, confirming the previous hints from HEAT and AMS.

Here are the PAMELA positron data compared with the theoretical predictions. The latter assume that the flux is dominated by the secondary production of positrons due to collisions of high-energy cosmic rays with the interstellar medium. The two lines are almost perpendicular to each other :-). In fact, the discrepancy below 10 GeV is not surprising, and is interpreted as being due to solar modulation. It turns out that the solar wind modifies the spectrum of low-energy cosmic rays, and in consequence the flux depends on solar activity which changes in the course of the 22-years solar cycle. Above 10 GeV the situation is different, as solar modulation is believed to produce negligible effects. Even though the secondary production of positrons has large theoretical uncertainties, one expects that it decreases with energy. Such a power-law decrease has been observed in the flux of anti-protons who also may originate from secondary production. The positron fraction, instead, significantly increases above 10 GeV.

Thus, PAMELA shows that the secondary production is not the dominant source of high-energy positrons. The excess can be due to astrophysical sources, for example young near-by pulsars have been proposed as an explanation. But what makes particle physicists so aroused is that dark matter annihilation is a plausible explanation too. It might be that PAMELA is a breakthrough in indirect dark matter searches. It is less known that there are other experiments that see some excess in the cosmic ray flux. Most interestingly, two balloon-borne experiments called ATIC and PPB-BETS see an excess in the total electron+positron flux (they cannot distinguish the two) with a peak around 700 GeV. This adds to the EGRET gamma-ray excess at a few GeV, and to the WMAP haze - an excess of diffuse microwave background from the core of our galaxy.

A dark matter candidate that fits the PAMELA excess, must have rather unexpected properties. If the observed dark matter abundance has a thermal origin, the dark matter annihilation cross section naively seems too small to explain the observed signal. As usual, theorists have magic tricks to boost the annihilation rate today. One is using the so-called boost factor: if dark matter clumps, its density is locally higher than average, and then the average annihilation rate also increases with respect to the case of a uniform distribution. However, this does not save the day for the most popular dark matter candidates. For example, the MSSM neutralino typically requires a boost factor of order a few hundred, which is probably stretching the point. The latest trick is called the Sommerfeld enhancement: if the dark matter particle feels some attractive long range forces (other than electromagnetism, of course), a pair of particles may form a bound state, which enhances the annihilation rate.

Another challenge for particle models is the fact that PAMELA sees no excess in the antiproton flux. This means that the dark matter particle must be hadrophobic, that is to say, it should decay preferentially into leptons. Again, the most popular dark matter candidates, like the MSSM neutralino, do not satisfy this criterion. However, particle models compatible with the PAMELA data do exist, for example Minimal Dark Matter (though this one is not compatible with the ATIC/PPB-BETS peak), or recent Exciting Dark Matter.

So, it seems, we have to wait and see till the smoke clears up. Certainly, a single indirect detection signal has to be taken with all due scepticism (so many have died before). Only combined efforts of several experiments can lead to a convincing conclusion. As for the moment, if somebody pointed a gun to my face and made me choose one answer, I would probably go for an astrophysical explanation. On the other hand, if the PAMELA excess is really a manifestation of dark matter, the LHC could concentrate on more interesting issues than discovering and undiscovering the MSSM. It seems that astrophysicists have at least one more year to sort this thing out by themselves.

Tuesday 21 October 2008

A Long-Expected Party

Today, soon after publishing the damage report, CERN is organizing the LHC Inauguration Ceremony. Given that the restart date is unclear (in private conversations, the estimate September 2009 appears most often), some lesser souls may feel dissonance. However, CERN is here to push the frontiers of science, and organizing an opening of a damaged accelerator is truly innovatory. The current DG himself must have had some doubts, as he cautiously writes "Dear Council Delegates, I would like to thank you for your reactions to my suggestion to maintain the LHC Inauguration Ceremony on October 21 2008, as initially foreseen...". Fortunately, the diplomats vehemently supported the idea, since they were already promised molecular cuisine.

Thus, CERN is overrun today by men in suits normally unseen at this latitude. This must be the first time foreign diplomats ever visit Geneva, so that unprecedented security measures were taken. All CERN parkings had to be emptied from cars, and those that remained are likely to be exploded. The roads connecting CERN to Geneva and France are now closed. The public buses that normally take this route run with a police escort, and they are not allowed to stop near CERN. It is not clear if the trespassers will be shot, or only impaled.

If you're curious what's on the menu, there will be a webcast here. This time there will be no live commentary, unless something funny happens. There is a rumor that during the ceremony the current DG will give a speech and vanish.