Wednesday, July 2, 2014

Reanalysis of Tevatron Data Disfavors Exotic Particles Under 120 GeV In Higgs Decays

Once again, the prospects of beyond the Standard Model physics at experimentally accessible scales is looking ever more bleak.
Tevatron Measurements on Standard Model Higgs

Federico Sforza, on behalf of the CDF, D0 Collaborations(Submitted on 1 Jul 2014)

We present the study of the SM Higgs properties obtained from the combined analysis of the up-to 10 fb−1 dataset collected by the CDF and D0 experiments during the pp¯ collision at s√=1.96~TeV of Tevatron Run II. The observed local significance for the SM Higgs boson signal is of 3.0σ at mH=125 GeV/c2. 
After a brief review of analysis channels contributing the most, where the Higgs boson decays to a pair of W bosons or to a pair of b-quarks jets, the signal production cross section and its couplings to fermions and vector bosons are analyzed. Other presented results are the recent study of the spin and parity of the SM Higgs performed by the D0 collaboration, leading to 3σ level expected exclusion of the JP=0− and JP=2+ hypothesis, and the investigation of exotics final states with invisible decay products of the Higgs, excluded by the CDF collaboration for masses below 120 GeV.

From here.

The Background Context and Minor Conclusions That Reaffirm Prior Data

Reanalysis of Tevatron experiment data after it shut down and after the LHC discovered the Higgs boson has revealed three sigma evidence of the 125 GeV mass Higgs boson discovered by the LHC.

This data also shows that the Higgs boson has W boson and b quark pair decays consistent with a Standard Model Higgs boson's decays.  The b quark pair decays are actually the dominant form of Higgs boson decay but are hard to detect in high energy collisions because lots of other non-Higgs boson decays have very similar end states.  But, since b quark pair decays account for something like 57% of Higgs boson decays and are the only form of decay into quarks that we expect should be detectable with current experiments, evidence that these decays happen at about the right frequency is critical to confirming the Standard Model nature of the Higgs boson and to ruling out, for example, "lepto-phillic" models of non-Standard Model Higgs bosons.

The data rule out pseudo-scalar (at the 3.1 sigma level) and tensor boson (at the 3.2 sigma level) versions of the Higgs boson, leaving the true scalar Higgs boson total angular momentum and parity of the Standard Model Higgs boson as the sole remaining viable option at the 3 sigma level (a few holdout skeptics note that a mix of scalar and pseudo-scalar Higgs bosons is not ruled out quite as strongly, but few really think that this is what we are seeing when all other evidence points so strongly to a Higgs boson indistinguishable from the Standard Model prediction of many decades ago from electro-weak unification theory).

All of this is basically old news, however, that has been revealed in previous papers from Tevatron and the LHC previously discussed at this blog.  Just click on the Higgs boson tag if you would like to review the prior results on these points.

The Big News

The critical new development announced in this paper is that it discusses "the investigation of exotics final states with invisible decay products of the Higgs, excluded by the CDF collaboration for masses below 120 GeV." This exclusion is at the 95% confidence level (i.e. two sigma).

Ruling out the production of exotic particles with masses of less than 120 GeV in Higgs boson decays is huge!

This disproves a huge swath of beyond the Standard Model theories.

For example, many beyond the Standard Model theories attempting to provide a dark matter candidate assume the existence of sterile neutrinos that don't couple to the gauge bosons of any of the three Standard Model forces, but do couple to gravity and the Higgs boson.  This result largely rules out sterile neutrino candidates of this type with masses of less than 120 GeV.

This also extends the exclusion range of an ordinary "fertile" fourth generation neutrino from about 45 GeV to about 120 GeV.  There are very sound interpretations of the current experimental evidence to suggest that the heaviest of the three currently known neutrino mass states is lighter than 0.2 eV.  Thus, the minimum mass of a fourth generation ordinary neutrino is now about 600,000,000,000 times that of the third generation neutrino (i.e. 6*10^11 times as heavy).

More generally, this rules out pretty much all plausible forms of "Higgs portal" dark matter, which has been a leading way to integrate dark matter candidates into the Standard Model framework in a minimalist manner.

Impact On SUSY Exclusions

Increasingly overwhelming experimental evidence, including this latest data point, strongly disfavor the existence of non-Standard Model SUSY particles at the electroweak scale, which was the "natural" mass range at which many original proponents of SUSY theories expected to discover these new particles.

Many still popular theories in the parameter space of supersymmetry (SUSY) theories call for "invisible" SUSY particles with masses of under 120 GeV.  This new Tevatron data regarding Higgs boson decay products therefore, impacts many theories that might otherwise be considered viable.  For example, Lubos Motl recently discussed a SUSY scenario with at least two invisible sparticles of less than 120 GeV as an attractive model.

Also, one of the main reasons to devise SUSY theories in the first place was to address the "hierarchy problem" of the seeming unnaturalness of the many contributions to the Higgs boson mass almost completely cancelling out, through mathematically obvious additional couplings to the Higgs boson that transparently cancel each other out.  So, it doesn't work to devise a SUSY theory in which sparticles don't couple to a Higgs boson and can thus escape the impact of the decay product exclusions of the new Tevatron data analysis.

The ATLAS experiment at the LHC has already set higher mass exclusions on SUSY particles, but in more model dependent ways that may not capture as many kinds of Higgs boson decay paths as the Tevatron data:

In R-parity-violating simplified models with decays of the lightest supersymmetric particle to electrons and muons, limits of 1350 GeV and 750 GeV are placed on gluino and chargino masses, respectively. In R-parity-conserving simplified models with heavy neutralinos decaying to a massless lightest supersymmetric particle, heavy neutralino masses up to 620 GeV are excluded. Limits are also placed on other supersymmetric scenarios.

Meanwhile, CMS experiment at the LHC has ruled out a large range of possible masses for heavy, Standard Model-like Higgs bosons of the type predicted in almost all SUSY theories:

A SM-like Higgs boson is excluded in the mass range 248-930 GeV at 95% CL using the shape-based analysis. . . . For the less sensitive cut-and-count analysis we obtain an observed exclusion of 268-756 GeV[.]

Other constraints, for example from the anomalous magnetic moment of the muon and the total absence of strong SUSY signals in any of the great numbers of channels studies, are discussed here, and notes that exclusions are particularly strong for sparticles of less than 300 GeV. Similarly, there is no evidence of SUSY predicted deviations from the Standard Model running of any of the three gauge coupling constants.

Matt Strassler's recent paper also identified a broad and quite comprehensive exclusion range for SUSY particle masses in "natural SUSY" models.

Direct dark matter detection experiments have also ruled out WIMP dark matter in the mass range from about 5 GeV to 1000 GeV, the bulk of the low end of the SUSY parameter space for light sparticles.

The fact that searches from multiple methodologies all confirm the absence of SUSY particles at these mass scales provides further confidence that this result is a robust one.

Exclusions of beyond the Standard Model SUSY particles at single digit TeV scales and higher masses aren't nearly so strong.  But, constraints like the expected amount of neutrinoless beta decay in many of those models, and the absence of dark matter models that can fit such heavy lightest supersymmetric particles, as well as their overall minimal impact of experimentally accessible phenomenology from the Standard Model (some of which SUSY theories were motivated to address), all make these versions of SUSY far less interesting and far less well motivated theoretically.

UPDATE July 3, 2014:

ATLAS has released four pre-prints with new, more demanding SUSY exclusion papers last night.  Needless to say, no hint of any SUSY particle was detected in any of the scenarios investigated.  Gluinos with masses of less than a bit more than 1 TeV and stop squarks with masses in the low hundreds of GeVs are excluded.

SM and SUSY Particles Reviewed

There are 18 particles in the Standard Model (ignoring color charge variants, antimatter variants of fermions and parity variants of charged fermions):
* one scalar spin-0 fundamental particle in the Standard Model (the Higgs boson) that is electrically neutral, lacks color charge and is massive, with antimatter counterparts;
* six massive spin-1/2 quarks that are electrically and color charged in three color charged variations each and antimatter variants;
* three spin-1/2 charged leptons that lack color charge, with antimatter counterparts;
* three spin-1/2 neutrinos that lack electrical or color charge,
* one massless electrically neutral spin-1 photons without color charge, with anti-matter counterparts;
* one massless electrically neutral, color charged spin-1 gluon that has their eight color charge variants;
* two massive electrically charged spin-1 W bosons without color charge which are antiparticles of each other; and
* a massive electrically neutral spin-1 Z boson without color charge.

Most non-SUGRA quantum gravity theories add one massless spin-2 graviton to the eighteen Standard Model particles.  The Standard Model lacks any particle that is a good dark matter candidate.  It would be possible to extend the Standard Model with quantum gravity to include a gravitino or right handed sterile neutrinos (and some boson to all their transition to other particle types or between each other), however, without otherwise disrupting the Standard Model.

The many particles predicted to exist in a Minimal Supersymmetric Standard Model aka MSSM (and almost all other less minimal SUSY models) are reviewed at Wikipedia.

In addition to a full set of Standard Model particles, the MSSM has five classes of superpartners: squarks (spin 0 bosons that are quark partners), sleptons (spin 0 bosons that are lepton partners), gluinos (spin 1/2 gluon partners), charginos (charged spin 1/2 superpartners of the W bosons and charged supersymmetric Higgs bosons), and neutralinos (four electrically neutral spin 1/2 that are mixes of the superpartners of the W bosons, Z bosons, and Higgs boson). There are also four additional Higgs bosons beyond those in the Standard Model (a positive charged spin zero Higgs ("H+"), a negatively charged spin zero Higgs (H-), a pseudo-scalar Higgs with neutral electric charge (usually denoted "A"), and an extra scalar Higgs with neutral electric charge that is either heavier or lighter than the Standard Model Higgs (usually denoted "H" for the heavier one and "h" for the lighter one).

Thus, the MSSM expanded into SUGRA has:
* six new scalar bosons that are color and electrically charged (the squarks) in three color charge variation each,
* five new scalar bosons are electrically charged but not color charged (the charged leptons, H+ and H-)
* five new scalar bosons electrically neutral and not color charged (the sneutrinos, h and A);
* one new electrically neutral spin 1/2 fermions (the gluinos) in eight color charge variations;
* four new charged spin 1/2 fermions (the charginos) that lack color charge;
* four new electrically neutral spin 1/2 fermions (the neutralinos) that lack color charge;
* one spin 3/2 fermion (the gravitino) that lacks electrical charge or color charge; and
* one massless spin-2 boson (the graviton).

The MSSM fermions presumably have anti-matter counterparts.

The MSSM conserves R-parity, which has one value of superpartners and the opposite value for non-superpartners, while this is broken dynamically in other SUSY models.  There are also SUSY theories that do not perfectly conserve R-parity, although these theories generally provide that the LSP decays to Standard Model particles in R-parity violating decays only very slowly, or with an LSP that is created from Standard Model particles almost as rapidly as it decays into Standard Model particles.

In all there are 25 BSM particles in the MSSM (ignoring color charge variants, antimatter variants of fermions, and parity variants of some or all fermions) and an additional 2 in SUGRA (although some people would call the graviton an "honorary" Standard Model particle as it is present in more or less identical forms in almost all supergravity theories) (ignoring antimatter variants of gravitinos, if any).

The are experimental hints of the existence of a graviton, but there are no experimental hints of the existence of any of the other 26 BSM particles in the MSSM and SUGRA.

The masses of these 26 BSM particles are key parameters, and another key parameter which related to the relationship between the H and h mass, and to the relationship between sfermion and sboson masses in SUSY called tan beta, is another key parameter.

None of the MSSM particles are stable except for the lightest supersymmetric particle which cannot decay further because R-parity prevents it from doing so.  The LSP then serves as a dark matter candidate.  The lightest neutralino and the gravitino are the most attactive as LSP candidates that are also good dark matter candidates because they are massive fermions, lack color charge and electric charge, and are stable.

Minimal supergravity (SUGRA) theories have the MSSM particles and also the graviton (massless, spin-2), and a gravitino (the massive, spin-3/2 superpartner of the graviton).  Non-minimal SUSY and SUGRA theories often have additional superpartners and non-Standard Model Higgs bosons.  Many also have right handed neutrinos and their superpartners, sometimes with a see-saw boson to handle interactions between left and right handed neutrinos that impart mass to neutrinos via a see saw mechanism, and most SUSY and SUGRA theories have Majorana rather than Dirac neutrinos, and conserve B-L number rather than conserving Baryon number and Lepton number separately.

BSM particle masses in SUSY and SUGRA theories.

The only massless BSM particle in the MSSM or SUGRA is the graviton which is also found in Standard Model extensions with quantum gravity.

Typically, the mass hierarchy of squarks and sleptons is inverted in SUSY models.  Third generation superpartners are the lighest, followed by second generation superpartners, followed by first generation superpartners.  Similarly, superpartners of bosons, which are fermions, all have masses greater than their Standard Model (or supersymmetric Higgs boson) counterparts.  The possibility that there are any superpartners that are lower in mass than any Standard Model counterparters of the same type has pretty much been completely ruled out experimentally.

This does not necessarily mean that the rank order of all Standard Model fermion partner masses is exactly inverted from the mass ranking of the particles in the Standard Model.  SUSY theories do not a priori demand, for example that a stau slepton be heavier than a stop squark.  But, to simplify matters, many more minimal SUSY models make this assumption.  In this case, for example, the lightest sfermion in such a SUSY theory will always be the stop squark, which helps to explain why the search for it is emphasized so heavily relative to other sfermion searches.

Also, since most dark matter theories prefer fermions to bosons as dark matter candidates, there should be at least one BSM SUSY fermion which is a partner of a Standard Model (or a partner of a SUSY Higgs boson or graviton) which is lighter than the stop squark (or than the lighest sfermion if different).  A SUSY theory in which the stop squark is the LSP is problematic when it comes to providing a good dark matter candidate.  Generally speaking, this means that either the lighest neutralino or the gravitino, or both, should be lighter than the stop squark.  Since neutralinos are linear combinations of superpartners of Standard Model particles with masses of 80 to 126 GeV, and experimental exclusions for neutralinos appear to be for masses greater than these, this puts an effective floor on the stop squark mass, in a model where the neutralino is the LSP dark matter candidate.  If the gravitino is the dark matter candidate (even if the neutralino is actually the LSP), then R-parity conservation for the neutralino is not necessary.

One particularly hot issue of scientific debate at the moment is whether there are any flaws in the methodology of SUSY particle searches that would allow any BSM particles in the MSSM or SUGRA with a mass of less than the Higgs boson or top quark to be missed by particle accelerator searches that purport to exclude BSM SUSY or SUGRA particles in the mass ranges.  Long lived BSM particles or particles with very low cross-sections of interaction could escape detection, but SUSY and SUGRA theories quite strongly constrain the cross-sections of interaction of the BSM particles that they mandate and provide a framework within which it is possible to qualitatively evaluate particle stability, although SUSY and SUGRA's many parameters do not allow for quantitative evaluations of this stability without additional assumptions about this parameter space not strongly supported by experimental data in any particular case.

Thus, a failure to detect a stop, sbottom, stau or tau sneutrino below a certain mass threshold generally also implies that this exclusion also applies lower generation squarks.  For example, conclusion that there are no stop squarks below 200 GeV would also imply that there are no scharm or up squarks below 200 GeV.

In SUSY theories, all gluinos should have identical masses, the charged Higgs bosons should have masses identical to each other, and the charginos should consist of pairs of particles with opposite charges and identical masses.  An exclusion of neutralinos below a certain mass is implied to exclude both the lighest neutralino and all of the other, heavier neutralinos.  The differences in couplings between the two scalar Higgs bosons in SUSY theories varies.

The New Search Results

* Stops squarks ("No significant excess over the Standard Model prediction is observed. A stop with a mass between 210 and 640 GeV decaying directly to a top quark and a massless LSP is excluded at 95% confidence level, and in models where the mass of the lightest chargino is twice that of the LSP, stops are excluded at 95% confidence level up to a mass of 500 GeV for an LSP mass in the range of 100 to 150 GeV. Stringent exclusion limits are also derived for all other considered stop decay scenarios, and generic upper limits are set on the visible cross-section for processes beyond the Standard Model.");

third generation squarks ("A top squark of mass up to about 240 GeV is excluded at 95% confidence level for arbitrary neutralino masses, within the kinematic boundaries. Top squark masses up to 270 GeV are excluded for a neutralino mass of 200 GeV. In a scenario where the top squark and the lightest neutralino are nearly degenerate in mass, top squark masses up to 260 GeV are excluded.");

UPDATE July 8, 2014: Another underwhelming analysis rules out a certain kind of stop squark with masses of under 180 GeV ("[F]or R-parity conserving supersymmetry . . . . Stop masses below ~ 180 GeV can now be ruled out for a light neutralino.") END UPDATE.

* superpartner particles generally ("No excess is observed with respect to the Standard Model predictions . . . Gluino masses up to 1340 GeV are excluded, depending on the model, significantly extending the previous ATLAS limits."); and

* more superpartner particles generally ("No excess above the Standard Model background expectation is observed in the various signal regions and 95% confidence level upper limits on the visible cross section for new phenomena are set. The results of the analysis are interpreted in several SUSY scenarios, significantly extending previous limits obtained in the same final states. In the framework of minimal gauge-mediated SUSY breaking models, values of the SUSY breaking scale Lambda below 63 TeV are excluded, independently of tan beta. Exclusion limits are also derived for an mSUGRA/CMSSM model, in both the R-parity-conserving and R-parity-violating case. A further interpretation is presented in a framework of natural gauge mediation, in which the gluino is assumed to be the only light coloured sparticle and gluino masses below 1090 GeV are excluded.").

Note that the natural SUSY breaking scale Lambda for a SUSY theory that naturally and fully solve the "hierarchy problem" which was one of the important reasons that SUSY was proposed in the first place is 1 TeV.  This is impossible when this Lambda value is at least 63 TeV.

4 comments:

andrew said...

Today, there is a new CMS diphoton mass of 124.70 0 ± 0.34 GeV = 124.70 ± 0.31 (stat) ± 0.15 (syst) GeV. The ATLAS diphoton mass is 126.0 +/- 0.2 +/- 0.28 (combined MOE +/- 0.344). The discrepancy between these measurements is 2.7 sigma. The raw average of these two measurements (which is appropriate given nearly identical margins of error) is 125.35 GeV.

The CMS four lepton mass is 125.8 +/- 0.5 +/- 0.2 (combined MOE 0.54) The ATLAS four lepton mass is 124.5 +0.6/-0.5 +/- 0.05 (combined MOE +/- 0.55). The discrepancy between these measurements is 1.7 sigma. The raw average of these two measurements (which is appropriate given nearly identical margins of error) is 125.15 GeV.

The raw average of the four measurements is 125.25 GeV, but this isn't quite appropriate because the diphoton measurements are a bit more accurate than the four lepton measurements. So the proper weighted average is a bit more than 125.25 GeV but still substantially less than 125.35 GeV. Rounded off the global average value for the Higgs boson mass is 125.3 GeV, although I'm not sure how to calculate the MOE and suspect that the MOE is underestimated given the considerable spread of the four measurements from 124.5 to 126.0 GeV. Notably, there is not systemic bias between the experiments. ATLAS has one high value (the diphoton value) and CMS has one high value (the four lepton value), that almost counterbalance each other.


Anonymous said...

Andrew,
I think that you are mistaken in your statement that all particles of mass less than 120 GeV (and that couple to the Higgs Boson) are ruled out. The authors state that a Higgs mass of <120 GeV/c2 is ruled out...not that there aren't sterile neutrinos of mass less than 120 GeV/c2.

Remember, the lighter the mass of the dark particle particle, the lower the branching ratio in Higgs decays. A sterile neutrino of 2-10 keV would have a really small branching ratio, and hence would not be ruled out from experiments so far.
In fact, we are barely even able to measure experimentally the branching of the charm quark...which is much heavier than a 2-10 keV sterile neutrino.

andrew said...

As I read it, the authors are saying that "invisible decay products" of less than 120 GeV from a Higgs boson are ruled out, although perhaps with conditions that they are only ruled out in some subtype of decays involving some visible component. They only look at 115 GeV or less in the chart because the trend below that is clear, but that is the way it looks to me.

I am really quite certain that they aren't talking about the mass of the 125 GeV Standard Model Higgs boson.

I guess what you are saying is that if the 125 GeV decays to another invisible Higgs which is less than 120 GeV, and then in turn to something else (possibly invisible) that this is ruled out.

I don't see anything that seems to indicate that they are talking about a light scalar Higgs boson of a second Higgs doublet, but if you could point out language to that effect, I'd appreciate the pointer if that is the case and I have misinterpreted it.

Anonymous said...

The branching ratios for decay products from a Higgs Boson are proportional to the mass of the particle (if it is a Standard Model Higgs boson decaying to a Standard Model fermion.)
As such, it will be decades to centuries before we could rule out a a 2-10 keV sterile neutrinos based off of looking at its production from a decaying Higgs Boson. We will hopefully be able to confirm or rule out a 2-10 keV sterile neutrino (or some other 2-10 keV dark matter fermion) via some other method before we are able to confirm or rule out based off of Higgs decay.

Andrew, you should read the entire paper, not just the abstract. The authors are not arguing that all invisible decay products below 120 GeV are ruled out. That would be a Nobel Prize winning achievement...that they have not done here. The authors are simply stating that the mass of the Higgs Boson is definitely not less than 120 GeV.
Here's a quote: "Masses of the Higgs below 120 GeV/c2 are excluded at 95% confidence level if BR(H → invisible) = 100% is imposed on the signal."