Wednesday, December 14, 2011

LHC releases data about new particle but is it Higgs?

The newest results about Higgs search using 4.9/fb of data were published yesterday and there are many articles in arXiv. The overall view is that there is evidence for something around 125 GeV. Whether this something is Higgs or some other particle decaying to Higgs remains to my opinion an open question. Lubos of course is strong in this faith on Higgs. Somewhat surprisingly Tommaso Dorigo seems the result as a firm evidence for Higgs. Matt Strassler is skeptic. The evidence comes basically from Higgs to γγ decays. There are some ZZ and WW events. CMS represented also data for more rare events. There are also indications about something at higher masses and the interpretation of them depends on the belief system of the blogger.

Bloggers were very active and Phil Gibbs certainly the most active one.

  • Phil Gibbs gave online comments and combinations of various data in his blog. In particular, Phil produced a combination of data from ATLAS, CMS, LEP, and Tevatron clearly supporting the existence of bump around 124 GeV. Also the other plots by Phil are very illustrative of the situation: for instance this.

  • Tommaso Dorigo commented the results. Tommaso gives many illustrations and sees the results as firm evidence for standard model Higgs and is skeptic about SUSY Higgs.

  • Lubos Motl has a nice -albeit over-optimistic from SUSY point of view - summary about the results and useful links to the articles by ATLAS and CMS.

  • I liked very much about Matt Strassler's critical comments making clear what is known and what is not known.

  • Also Resonaances had added comments on Higgs. And many other bloggers.

Overall view

It is good to try to summarize what has been found.

  1. According to ATLAS the bump is at 126 GeV. Altogether gamma-gamma and ZZ events give 3.6 sigma deviation reducing to 2.3-2.5 sigma by look-elsewhere effect.

  2. According to CMS the bump resides at 124 GeV. CMS has 2.6 sigma deviation reducing to 1.9 sigma when look- elsewhere effect is taken into account.

  3. The positions of bumps reported by ATLAS and CMS are not quite same so that there is still room for the possibility that the bumps are artifacts.

  4. Both collaborations publish also the results about signal for Higgs to ZZ decays. Fig 7 in ATLAS eprint reports three candidates Higss to ZZ events at 123.6 GeV, 124.3 GeV , and 124.8 GeV. CMS reports results about decays to bbbar, ττbar, WW, ZZ, γγ, combines the data from various channels, and compares signal cross sections to those predicted by standard model Higgs. There is structure around 145 GeV in some channels. There is also data about higher energies bump like structures both below and above 300 GeV. Maybe the standard model Higgs is not enough. SUSY indeed predicts several Higgs like states and M89 hadron physics entire meson spectroscopy.

γγ channel

Both ATLAS and CMS have published an reprint in arXiv about Higgs to γγ signal.

The ratio of the γγ signal cross section to the cross section predicted by standard model is given together with 1 and 2 sigma bands describing the background signal without Higgs contributions.

  1. Fig 8 ATLAS paper gives the observed and expected 95 per cent confidence level limits as a function of hypothesized Higgs boson mass.
  2. Fig 3 of CMS paper gives bump around 124 GeV.

I wish that I could understand the strange oscillating behavior for the ratio of signal cross section to cross section for signals mimicking Higgs predicted in absence of Higgs. There are bumps around 126 GeV, 139 GeV and 146 GeV. Is this an artifact produced - say - by a discretization in the data processing. There is also a small bump around 113 GeV.

I am a statistical dilettante so that I can make an innocent and possibly stupid question: Could these bumps be something real? For standard model Higgs this is certainly not the case but what about TGD inspired view about new physics at LHC. Going to another extreme: could all bumps with 126 GeV bump included be only data processing artifacts? We do not know yet.

Also the probability p0 that standard model without Higgs could explain the signal cross section is plotted.

  1. Fig 7 of ATLAS paper gives observed and expected p0 values as a function of the mass of hypothesized Higgs. Small p0 tells that standard model without Higgs contribution requires upwards fluctuation whose probability is p0 to explain the observed signal. There are strong downwards bumps at about 126 GeV and around 145 GeV. They are deeper than the prediction of standard model Higgs which might give rise to worries. There is also something very small at 139 GeV.

  2. Fig 4 of CMS paper gives similar plot. Now the bumps of p0 are around 123.5 GeV, 137 GeV, and 147.5 GeV.

If taken at face value, also these figures suggest three-bump structure. This might well be a statistical artifact but one can make questions and one fool like me can make more or them than the wise guys are able to answer. Here are two of them.

  1. Could M89 pion have higher excitations with a mass scale of 10-20 GeV? Could the pion-like state generating the signal besides ground state also excitations with excitation energy scale of order 20 GeV? Could these excitation assigned with the magnetic flux tube structures associated associated with scaled up u and d quarks.

    A rough guess for the p-adic prime of scaled up u and d quark in M89 hadron is k=113-18= 95 (k=113 corresponds to Gaussian Mersenne and nuclear p-adic length scale). This corresponds to the p-adic mass scale the estimate 16 GeV from electron's p-adic mass scale about .25 MeV. It however turns out that the actual mass must be by a factor two higher so that one would have 32 GeV mass scale.

    Could stringy excitations with string tension determined by 32 GeV scale be in question? If so then also ordinary pion should have similar fine structure in mass spectrum with energy scale of 31 MeV assignable with u and d quarks with k=113. I have a vague memory that Tommaso Dorigo had reported something about low energy excitations of pion but I failed to find anything about this in web and concluded that I must have been hallucinating.

  2. Shnoll effect is something which main stream colleagues certainly refuse to take seriously. In TGD framework one can develop a p-adic model for Shnoll effect which can be justified in terms of quantum arithmetics giving a first principle justification for the canonical identification playing a key role in p-adic mass calculations. The model predicts a number theoretic deformation of probability distributions characterized p-adic prime p. The modification replaces the rational valued parameters of distribution by quantum rationals. Typically a probabiity distribution with single bump decomposes to several ones and the phenomenon occurs also in nuclear physics.

    Could this deformation be at work even in particle physics? If so, it could cause the splitting of single very wide resonance bump around 125 GeV to several sharper bumps. Even the bump like structure at 113 GeV could correspond to this wide resonance bump. The original resonance bump could be rather wide: something like 30-40 GeV. Very naive guess would be that the width of leptopion obeys able to decay to ordinary quarks Γ ∼ αs(89) m(π89). Already for αs=.1 one could have a bump with width of about 15 GeV. For ordinary pion the impossibility of strong decays would not allow Shnoll effect. The splitting into sub-bumps by Shnoll effect would make this wide bump visible.

After a painful web search I managed to find an article titled Search for low-mass exotic mesonic structures: II. Attempts to understand the experimental results reporting that there is experimental support for narrow excited states of pion at masses 62, 80, 100, 140, 181, 198, 215, 227.7, and 235 MeV (authors mention that the last might be uncertain). The states at 100, 140, and 198 MeV are half octaves of the lightest state. The article fits the states to Regge trajectories but it is not possible to use single slope for all states. The mass differences vary between 10 and 40 GeV so that the scale is what one would expect from the above string argument. Also Shnoll effect might explain the existence of the bumps and if the explanations are consistent the spectrum of the pion states is dictated by number theoretical arguments to a high degree.

Combination of signals from all channels

CMS has also a preprint about the combination of signals from all decay channels of Higgs.

  1. CMS gives also a figure combination of all CMS searches (γγ, bbbar,ττbar, WW, ZZ).

  2. Figure 1 of CMS article shows a clear structure around 124 GeV. There is another structure around about 145 GeV. In standard model Higgs scenario the structure at 145 GeV would not be taken seriously since the cross section need to produce the bump would be much below the predicted one but if one accepts super-symmetric M89 hadron physics, the situation changes. There is also structure around 325 GeV and in the range 260-285 GeV. M89 hadron physics would assign these structures to vector mesons ρ and ω89 and corresponding smesons consisting of squark and anti-squark.

  3. CMS gives a plot comparing the ratio of best fit for signal cross section to the predicted cross section for Higgs to bbar, ττbar, γγ, ZZ, WW. The fit is rather satisfactory: for Higgs to γγ the signal cross section is about 1.7 times higher than predicted. One cannot deny that this can be seen as a strong support for standard model Higgs.

    The original idea behind M89 hadron physics was that it effectively replaces Higgs. If one takes the CMS result seriously this idea must be realized rather concretely: the predicted signal cross sections must be rather near to those predicted by standard model Higgs. The crucial tests are decay rates to fermion pairs and the possibly existing other resonances.

Higgsy character of 126 GeV bump is not proven!

Lubos has written a new post were he makes the strange assumption that if there is a signal it must be Higgs. Lubos also uses as a "proof" of Higgsyness the fit of Phil for which the gamma-gamma signal cross section at maximum equals to the prediction. This holds true because the fit forces it to hold true! For some reason Lubos "forgets" this!

By inspecting the figure more closely one finds that the observed cross section has a long tail unlike the predicted cross section. This long tail could correspond to the large width of resonance splitting into sub-bumps if Shnoll effect is present. If Higgs option is correct, this tail should disappear as statistics improves. Also the other structures which are present, should disappear.

What is of course remarkable that CMS paper shows that H→ γγ cross section is of the same order of magnitude and only about 1.7 times higher than the predicted cross section. This gives a constraint on M89 hadron physics, which it of course might fail to satisfy unless the idea about replacement of Higgs with M89 hadron physics is true at a rather quantitative level.

One should also keep in mind that the value of Higgs mass is at the lower bound for the range with stable Higgs vacuum. This is not a good sign. An interesting question is whether the mass for pion-like state of M89 hadron physics is in some sense also minimal and what this minimality could mean physically: some kind of criticality - maybe on instance of quantum criticality of TGD Universe- but not criticality against the decay of Higgs vacuum?

To sum up, one can agree with the official statement: the situation remains open. What is nice that there very probably is a signal and from TGD point view the nice thing is that this signal is still consistent with M89 hadron physics.

11 comments:

Satama said...

Matti, do you have a overview presentation of current state TGD-based particle physics, with tables etc., listing main differences with standard model and other models?

It would be nice to have around... :)

matpitka@luukku.com said...

Unfortunately not. The attempt to write anything like this unavoidably leads to a series of papers. I have indeed done this in Prespace-time Journal.

Also the chapters of p-Adic length scale hypothesis and dark matter hierarchy provide detailed comparisons with more standard views about physics expected at LHC.

Luboš Motl said...

Dear Matti, if you propose a particular theory what the bump could be different than a Higgs - a scalar with the interactions we expect from a Higgs (all of which seem necessary for the peaks to appear in the right channel and at the same strength) - I will be happy to discuss whether the LHC signals (combined with theoretical consistency requirements) leave some wiggle room for such alternatives. Without such a proposal, there's nothing to talk about.

I assure you that a thing decaying to 2 photons has to be an electrically neutral boson, so the spin has to be integer. Spin 1 would be a new gauge boson like Z' but it can't really be there because it would damage too many other things. So we're left with the scalar. OK. Now when it's a spin-zero field, we may discuss its couplings. You need a coupling to photons (via the W and top loops) because photons are seen, coupling to quarks because the particle is produced, and so on.

When it smells, looks, and talks like a Higgs, it's because it's a Higgs, especially if it's theoretically needed and high-precision experiments show that it shouldn't be too heavy, e.g. above 500 GeV, and everything else below 500 GeV is excluded.

Andrew Oh-Willeke said...

I have a discussion of why, if there is anything at 125 GeV that is producing diphoton channel signals, there is an exceedingly good reason to think that it is indistinguishable from the Standard Model Higgs boson.

In a nutshell, a 125 GeV diphoton decay can only be produced by the decay of something that has a zero electromagnetic charge, a spin of zero or two, and a mass of 125 GeV. It can't be a spin 1/2 or spin 1 particle and produce a diphoton decay. This is a fit to the theoretically predicted (35 years ago) profile and inconsistent with every other possibility that was proposed a priori.

I could be mistaken, but I think that the decays observed are also even consistent with distinguishing the result as CP-even rather than CP-odd as predicted.

Likewise, the result is inconsistent with any particle that has a net color charge since the end products lack it and net color charge is conserved.

There are tests one could perform (angular distribution, for example), to test the hypothesis further, but one really has to bend over backwards to find an alternative and given that we already known that a Higgs field with a vev of 246 GeV (on the same order of magnitude) exists, and that there a good reasons a priori to think that there should be this exact thing in that exact place (give or take a few GeV), the argument that this a Higgs rather than some other form of bump is pretty compelling.

matpitka@luukku.com said...

Dear Lubos,

It smells, looks, and talks like pseudoscalar or scalar state. This does not mean it is Higgs.

I of course make a detailed proposal for what these states are. I have been talking about this proposal in my blog this year!

a) The original -now 15 year vision- was that M_89 hadron physics -or more precisely (pseudoscalar) mesons of M_89 hadron physics- replaces Higgs and it seems that this is realized even at the level of orders of magnitudes. Technicolor could be seen analogous proposal but extending the gauge group.

b) The treatment M_89 mesons in effective QFT description in M^4 (assuming that this makes sense!) gives same couplings to weak bosons as for Higgs: this means that the rates H--> gammagamma and ZZ are same as for Higgs! So that in this respect these states smell like Higgs by mere gauge symmetry!

c) In the case of fermions standard dimensionless gradient coupling to meson gives coupling proportional to mass so that these mesons could be able to mimic Higgs like states to high degree. Here an independent dimensionless coupling enters the game. About fermions we have very little data yet.

One can test this picture against Higgs hypothesis.

a) The presence of charged companions of Higgs like states is of course something new and a test for the scenario. Gauge bosons are automatically massive and possess 3 polarization states so that they need not eat Higgses. This follows from the vision about generalized Feynman graphs which I have been also developing also in blog postings.

b) You can neglect the structures at higher masses only if you believe Higgs hypothesis. Otherwise you have structures for which signal cross section is above the standard model background. The structures around 140 GeV and at both sides of 300 GeV. There are very clearly visible as peaks also in p_0.

matpitka@luukku.com said...

Continuation of the reply to Lubos:


Supersymmetry in TGD sense is essential for the proposed interpretation of Higgs like signals in terms of mixtures of pion and spion of M_89 hadron physics.

The proposal inspired by the mysterious X and Y charmonium states is that SUSY is exact in the sense that p-adic mass scales of particle and spartner are same - even the masses could be same. 125 GeV bump would be mixture of pion and spion. The earlier bump at around 139 MeV would be neutral pion and bump around 144 GeV would correspond to charged pion. Rho, omega and the corresponding smeson states would be at both sides of 300 GeV.


What about ordinary hadrons? Is SUSY really possible for them?

a) Mesons and smesons consisting of squark pair mix and for large alpha_s the mixing is large and can make second eigenvalue of mass squared matrix negative. If so, these states disappears from spectrum. In any case one has two meson like states instead of single meson.

b) Transformation of second pionlike state to tachyon and disappearance from spectrum is not the only possibility. Just yesterday I however found experimental work claiming the existence of states analogous to ordinary pion with masses 60,80, 100,.... MeV. 100 GeV is first downwards half-octave of pion with mass about 140 MeV and also second half octave is there. Could it be that one of these states is second states predicted by TGD SUSY for ordinary hadrons? But what about other states?

Objection: SUSY signatures involving missing energy have not been observed.

a) Original argument. Strong interactions are faster than weak decays of squarks to quark and weak boson producing the usual signatures of SUSY so that shadronization would takes place instead of production of the SUSY signatures. This plus different realization of SUSY in TGD might explain the failure to find standard SUSY.

b) Objection: this is not enough. The weak decays of squarks producing right handed neutrinos as missing energy are still there! I am considering also the option that covariantly constant right handed neutrino which generates SUSY is replaced with color octet. Color leptons of leptohadron hypothesis would be sleptons which are color octets so that SUSY for leptons would have been seen already at seventies!

Covariantly constant right-handed neutrino as such would be gauge symmetry like super-generator annihilating the states: something very similar can occur in the reduction of ordinary SUSY algebra to sub-algebra familiar in string model context. By color confinement missing energy realized as color octet right handed neutrino would not be produced.

This is just one possibility. The situation is not completely settled and one must keep mind open.

To be continued....

matpitka@luukku.com said...

Continuation of reply to Lubos.


A further objection. The claimed pion like states at 60, 80, 100, 140,... MeV contain also other than those identifiable as spion llike state. What about their origin?

One can imagine two explanations which could be equivalent.

a) The states could be infrared Regge trajectories assignable to magnetic flux tubes of order Compton length of u and d quark (very long and with small string tension) could be the explanation. Hadron mass spectrum would have microstructure. Something very natural in fractal many-sheeted space-time. Similar fine structure for nuclei is predicted for nuclei in nuclear string model.

b) Shnoll effect would be a completely universal modification of probability concept based on the replacement of ordinary arithmetics with quantum arithmetics . Quite generally distributions with single peak would be replaced with many peaked ones with sub-peak structure having number theoretic origin. it would be really fascinating to see this deformation at work even in elementary particle physics.

For Higgs like signals IR-Regge trajectories / Shnoll effect would be visible as splitting of single wide bump for spion and pion of M_89 physics to narrower sub-bumps. This oscillatory bumpy structure is there but is regarded as a statistical artifact.

matpitka@luukku.com said...

To Andrew Oh-Willeke:

I agree with most what you say but not with the necessity to interpret the state as Higgs. Pseudo-scalar is to my view also possible interpretation. CP oddness is certainly an important difference.

Some hasty, amateurish, and certainly imprecise notes about couplings.

a) Gauge symmetry alone seems to imply identical couplings of charged pion like states to weak gauge bosons as in the case of Higgs.

b) For Higgss to gamma pair fermion and charged gauge boson loops involve HFFbar coupling and coupling of Higgs to W pair.

c) Also fore neutral pseudo-scalars the coupling to photon pair would be through fermion/sfermion loops and W boson loops. W bosons have same gauge coupling. Rate is also now proportional to alpha^2.

In M^4 QFT approximation one has coupling of pseudoscalar to fermion which as gradient coupling would be proportional to fermion mass. In fermion sector the only difference would be the value of dimensionless fermion coupling, which should not be far from that in standard model. Very similar results would be obtained except that the coupling proportional to fermion mass would not follow from Higgs vacuum expectation value but from gradient coupling.

Therefore theneutral pseudoscalar would behave very much like Higgs in QFT limit and if the dimensionless couplign to fermions is same, there might be no difference in gamma-gamma rate.

To be continued....

matpitka@luukku.com said...

Continuation to the reply to Andrew Oh-Willeke:

From my Iztykson-Zuber I learn that at IR limit at least the decay to gamma pair would be describable uniquely in terms of axial vector anomaly giving action which is product of the pseudo-scalar with E.B, the instanton density. The decay rate is proportional to 1/f_pi^2: the divergence of axial current is f_pi times the pion field.


From my Iztykson-Zuber I also read that sigma model for nucleon, pion and sigma meson involves the analog of Higgs mechanism so that it might be possible to develop rather close analogies. Questions.

a) Could sigma model description make sense as approximation in M_89 hadron physics at low energies?

b) Could sigma model approximation bring in non-perturbative aspects such as non-analytical dependence of the decay rate to gamma pair? Non-perturbative aspects for weak interactions would reduce to non-perturbative aspect of strong interactions understandable in terms of parton-string duality of generalized Feynman diagrams.

c) For instance, the vacuum expectation v of sigma field equals to the negative of f_pi and the decay width of pion to two-gamma is given by

Gamma = alpha^2m_pi^2/64pi^2f_pi^2, f_pi=-v

This should be compared to the decay rate of Higgs in standard model to get perspective. what is interesting is the non-perturbative dependence 1/v^2, v the sigma field vacuum expectation.

d) Could it be that sigma model description of M_89 hadron physics replaces the description in terms of standard model Higgs mechanism which fails because Higgs mass has the critical value at which vacuum ceases to be stable?

Ulla said...

This is maybe out of topic and too simple. Usually hbar is left out (G=c=hbar?) so maybe we forget this fact? The well is not symmetric, but the symmetry is emegent in the sphere or through pi?

Spin (quantum number S) is a vector quantity that represents the "intrinsic" angular momentum of a particle. It comes in increments of 1⁄2 ħ (pronounced "h-bar"). The ħ is often dropped because it is the "fundamental" unit of spin, and it is implied that "spin 1" means "spin 1 ħ". (In some systems of natural units, ħ is chosen to be 1, and therefore does not appear in equations). http://en.wikipedia.org/wiki/Meson

The spin of microscopic particles is so small it is measured in special units called "h-bar," related to Planck's constant, h, which is defined as 4.1 × 10-21 MeV seconds. h-bar is defined to be h divided by two and by pi (3.14159...).
spin can only have certain values. Another way of saying this is spin must be "quantized." Particles with spin values of one-half h-bar, three-halves h-bar, five halves h-bar, and so on are called fermions and described by a mathematical framework called Fermi-Dirac statistics

Particles with spin values of zero h-bar, one h-bar, two h-bar, and so on are called bosons, and are described by a mathematical framework called Bose-Einstein statistics. The quantization of spin means we have to add spins together carefully using special rules for addition of angular momentum in quantum mechanics.
http://science.jrank.org/pages/6383/Spin-Subatomic-Particles-Spin-in-quantum-mechanics.html#ixzz1goQqeYtY

(In actuality elementary particles do not rotate like spheres; it is only that the particle property termed spin obeys rules that mathematically are similar to those used to describe the rotation of macroscopic bodies.) The spin of elementary particles is measured in special units called "h-bar" (h-bar is Planck's constant divided by 2π), and = 1.1 × 10-34 Joule-seconds. A closer approximation in fractions is 355/113. Lambert found it was not rational = a transcendental number. That is, it is not the root of any polynomial equation with rational coefficients.

There are many infinite series that can be used to calculate approximations to π. One of these is where the denominators are the consecutive odd numbers. pages/5221/Pi.html
http://science.jrank.org/pages/6560/Subatomic-Particles.html#ixzz1gocfwzOG

the ratio of the circumference of a circle to the diameter: π = C/d. (h-bar is Planck's constant divided by 2π)
http://science.jrank.org/pages/5221/Pi.html#ixzz1godtcuZn

fermions and described by a mathematical framework called Fermi-Dirac statistics

bosons, and are described by a mathematical framework called Bose-Einstein statistics.

particles can also be represented mathematically using spinors. A spinor is like a vector, but instead of describing something's size and orientation in space, it describes the particle in a theoretical space called spin space

Ulla said...

http://arxiv.org/PS_cache/arxiv/pdf/1112/1112.3175v1.pdf 14.12.

Did Poincaré discover Special Relativity? December 18, 2011. http://physicsforme.wordpress.com/2011/12/18/did-poincare-discover-special-relativity/

The crucial difference between these formulations is that Poincar´e finds it necessary to refer to an observer, while Einstein does not.