Friday, December 6, 2013

Confirmation of SM Character Of Higgs Continues

Peter Woit summarizes the latest Higgs boson results from the Large Hadron Collider:
Both ATLAS and CMS have announced new data on tau-tau decays of the Higgs, providing stronger evidence for this signal than was available earlier. ATLAS sees a signal with significance 4.1 sigma, CMS at 3.4 sigma. These results are consistent with the SM, and rule out some SUSY alternatives in which the Higgs would behave differently. The Register headlines this Exotic physics takes an arrow to the knee.
Relative to the Standard Model expectation, at ATLAS the signal seen is 140% of the Standard Model value with margins of error to fit values between 100% and 190% of it.  At CMS, the signal seen is 87% of the Standard Model value with margins of error to fit between 58% and 116% of it.

The opposite direction of deviation from the Standard Model expectation value at two experiments simultaneously measuring the same thing discourages speculation that the true value differs materially in one direction or the other from the Standard Model Higgs boson tau-tau decay rate.  Most importantly, this result strongly disfavors speculative theories like a "leptophobic" or "leptophilic" Higgs boson that couples with different strengths relative to mass to different kinds of particles.

There are five kinds of decays that should be observable from the Higgs bosons (there are other kinds as well but they are expected to be vanishingly rare).  The discovery of the Higgs boson was based upon three of these decay channels, but the existence of Higgs boson decays in the tau-tau decay channel (a particle-antiparticle pair of third generation heavy electrons) and bottom-bottom (third generation down type quarks) channel remained less clear.  There was a three sigma signal of bottom-bottom decays from Higgs bosons at Tevatron before it was shut down (this is actually the main decay channel of Higgs bosons, but is harder to identify due to large backgrounds from other processes), but there was less data on the tau-tau channel.

In absolute terms, particular channels of Higgs boson decays have mostly been within about 50% of the expected values, which is within two sigma due to large margins of error.  Now that we have some reasonably significant observations in every expected decay channel and so far have not seen any unexpected decay products, the room for a non-Standard Model-like Higgs boson result narrows greatly.

Now, there is five plus sigma evidence of a Standard Model Higgs in three channels, better than four sigma evidence in the tau-tau channel, and three sigma evidence in the b-b channel.  But, the relative uncertainty in the b-b channel makes it hard to pin down the total decay spectrum of the Higgs boson experimentally.  If the b-b decay channel, for example, made up 55% rather than the expected 60% or so of expected Higgs decays, that would provide "room" for all sorts of other unexpected decays, so long as they don't involve W bosons or Z bosons or photons or charged Standard Model leptons.

The Standard Model predicts the probabilities of each of the Standard Model Higgs boson decay channels to a precision of something on the order of 1% of less for a Higgs boson with a mass known to the precision that it has been measured to date, so the prediction that the Standard Model Higgs will decay in a particular way is eminently testable.

Impact on Supersymmetry

Fortunately, many of the leading alternatives to the Standard Model, like supersymmetry (SUSY), also make quite specific predictions about how a spin-0 Higgs boson with zero electromagnetic charge and the observed mass (one of three or more neutral Higgs bosons present in SUSY theories), will behave in particular kinds of decay channels like tau-tau.

These models have moving parts (adjustable parameters) that can be used to tweak the predictions to the observed result, but the class of SUSY theories in which there is a Higgs boson that looks exactly like the Standard Model one and there are no other light Higgs bosons that can be detected with the searches that LHC has done so far is quite narrow (see also here).

The CMSSM version of SUSY, for example, isn't quite ruled out yet (and has a Higgs boson almost indistinguishable from the Standard Model Higgs boson), but is left to a steadily shrinking parameter space as SUSY particle exclusions from LHC grow larger and the anomalous magnetic moment of the muon limits how heavy its particles can get to evade LHC lower mass sparticle exclusions.

In a nutshell, it is becoming harder and harder for SUSY proponents to explain why there is still no meaningful experimental evidence of any of the myriad new particles that the theory implies.  While it is easy to devise a SUSY theory in which most of the particles are too heavy to ever be detected, it is much harder to devise one where none of them are so light that they are observable or will soon be observable.

UPDATE December 10, 2013:

Another paper rules out a cascade of SUSY Higgs bosons into the observed phenomena with experimental evidence and also places the limits on the cross-sections o$f any SUSY Higgs boson phenomena at the LHC.


Mitchell said...

Concerning the future of SUSY models, I have two issues, one standard, one just a gap in my own knowledge.

The standard issue is just naturalness. The point of weak-scale supersymmetry was to have a theory in which the electroweak scale is stable against quantum corrections due to heavy virtual particles, without finetuning of parameters. One can move the supersymmetry-breaking scale higher, but then finetuning is apparently necessary, removing one of the rationales for supersymmetry.

I know you know this, I'm just stating it as context for the second issue.

So here's the second issue. In general, the MSSM particle spectrum consists of the SM particle spectrum, plus superpartners; so if the superpartners all become sufficiently heavy, physics can look like pure SM up to quite high energies.

The exception to this is the Higgs boson. In the SM, it's the same Higgs doublet which couples to all the fermions. In the MSSM, there are *two* Higgs superfields, one for an up-type Higgs that couples to up-type quarks, the other for a down-type Higgs that couples to down-type quarks and to charged leptons. This is required both by constraints on the allowed terms and by anomaly cancellation; the details can be found e.g. in Stephen Martin's MSSM primer.

It might seem, therefore, that even if we assumed high-scale supersymmetry for the MSSM, we wouldn't get a single Higgs boson with SM-like couplings, because different Higgs fields are coupling to different fermions. Indeed, you get the 8-3=5 Higgs bosons left over, and it seems like their pattern of interactions and decays might be quite different from what we're seeing at LHC.

Now I think it surely *could* be different, but I'm also pretty sure (because I've seen it in the literature) that there's a region of MSSM Higgs parameter space which really *does* look like a single SM Higgs boson with SM couplings. But what I don't know, is what the MSSM Higgs parameters have to be like, to produce such an SM imitation.

Nor do I know how contrived that outcome is. Having all the superpartners heavy actually *is* fairly natural, it just requires that the main parameter controlling supersymmetry breaking falls in the right range. But I'm not aware of whether an SM-imitating Higgs sector in MSSM requires special tuning, or whether it can be achieved by a single condition (e.g. by having the mu parameter large?).

This is relevant for understanding the remaining prospects for SUSY - exactly what are the conditions required to have it look like pure SM? The answer seems to be, high-scale SUSY breaking, and, some further condition in the Higgs sector.

Maju said...

Andrew: have you read this?:

Superstrings Model strikes back by the hand of Maldacena and others: entanglement, at the level of black holes and fundamental particles alike, creates wormholes (always)... and also gravity, dumping Einstein but salvaging the Standard Model.

What does Mr. Quark think of this? ;-)

Maju said...

Erratum: I don't mean "salvaging" but "saving" or "keeping".

Mitchell said...

I can comment... That relationship is not yet being demonstrated for an empirically relevant theory like the standard model. There are numerous mathematical field theories, which are believed to be equivalent to string theories on a particular geometric background; this is the AdS/CFT duality. The cited papers are exhibiting detailed constructions that realize the Maldacena-Susskind idea, in the sense that pairs of entangled particles in the CFT, correspond to wormholes in the AdS geometry.

I think it's fundamental stuff, but the real world is not AdS geometry (it's more like dS, de Sitter space), and we don't know how holography works for dS. So in a sense this is all still math, not physics.

Maju said...

I see, Mitchell: you feel it's merely theoretical and may well be overridden soon, right?

However I feel that the quark entanglement (with gravitational wormhole) prediction can be tested empirically in Geneva, right? That's what particle accelerators are for. Beyond the theoretical debate about dS vs AdS, there is real stuff being predicted here that can be tested.

It may not be confirmed but what if it is?

Mitchell said...

"Beyond the theoretical debate about dS vs AdS, there is real stuff being predicted here that can be tested."

No, there is not. You are being misled by the reference to "quarks". Those are not the quarks of the real world; they are the fermions appearing on one side of the mathematical mapping. The quarks of the real world come in six "flavors", they have particular charges and mappings, they interact according to SU(3) gluons. The "quarks" of this CFT, like the quarks of the real world, would be spin 1/2 particles interacting via a gauge field, but that's about the extent of the commonality. The CFT here is a highly supersymmetric theory with particles and properties that are not part of reality (all the quarks are massless, for example).

For now this is just mathematics, some generalization of which may eventually be relevant to reality. Even then, the wormholes might be untestable in some sense. Remember that they provide an equivalent description of the physics. There would be no new particle behavior, just a mathematical proof that the same old behavior also corresponded to a description in term of wormholes. This aspect would lead to new predictions, only if the wormhole description allowed you to calculate something more easily than the entanglement description; this is how AdS/CFT *did* lead to some predictions (with mixed results), when it was applied as an approximation scheme to quark-gluon plasma. And even then, it wasn't being used as an exact model of the world, AdS/CFT was instead providing a calculable model system whose thermodynamics were believed to be similar to those of the real quark-gluon plasma.

Maju said...

But particle entanglement does exist, and has been demonstrated with all kind of particles.

"This aspect would lead to new predictions, only if the wormhole description allowed you to calculate something more easily than the entanglement description"...

They are saying that the wormholes create gravity. I know that gravity is hard to measure at this scale but there must be something to measure, right? An non-entangled quark (or whatever other particle) should be different than an entangled one IF this theory holds true.

I wonder whether it'd be easier to observe if we could find quantum entanglement at the level of black holes. Maybe the patient astronomy can do what particle accelerators cannot?

andrew said...

Sick today. Will comment someday.

andrew said...

Quantum entanglement is a mystery that requires one of three things to be violated, only two of which I can recall at the moment: locality, causality, or something else.

A wormhole is a resolution of this effect that chooses non-locality to resolve the paradox using a curved space-time singularity approach to do so. It is not the only solution and there is really no way to show that this rather than some other mechanism is at work.

There is really no experimental signature for this method of creating quantum entanglement versus any other, so it won't be validated or invalidated at the LHC.

andrew said...

"exactly what are the conditions required to have it look like pure SM? The answer seems to be, high-scale SUSY breaking, and, some further condition in the Higgs sector."

You need a bunch of parameters to have particular values that leave a pretty small corner of parameter space. Trickiest is that it is hard in more minimal or constrained versions of SUSY to get just one light Higgs and four heavy non-SM Higgs. You really need to have at least one be quite light. And, we haven't seen it. But, since we don't know precisely how it would act, it is hard to know what we are missing.

One solution is that both light Higgs bosons in SUSY have degenerate (i.e. almost identical) masses. For example, this would make sense if the current data favoring a scalar SM Higgs boson over a pseudo-scalar Higgs boson never resolved its current 2-1 bias in favor of the SM result as precision in the measurement improved which would suggest the presence of a SM-like Higgs boson and a pseudoscalar-SUSY extra neutral Higgs boson A of identical mass and couplings in 2-1 proportions to each other.

The exclusions of non-SM Higgs bosons at other masses so far seem entirely too timid to me. There just aren't any other bumps up to 600 GeV in the data and there should be at least one if there is a non-SM Higgs boson in that mass range and any reasonably constrained SUSY theory demands one somewhere under 1 TeV.

andrew said...

The literature of SUSY Higgs boson constraints is summarized here and here.

andrew said...

More discussion of Higgs property impacts on SUSY is found here.