Thursday, October 10, 2013

Glueball Found?

The Possible Glueball Resonance.

One of the most striking predictions of the Standard Model of particle physics not yet confirmed by experiment is the existence of a class of particles called glueballs.

Earlier this year, a new paper (updated today) noted that a particle resonance dubbed X(3020) for its mass of 3.02 GeV in neutral B meson decays that has been observed by the BABAR collaboration is a strong candidate to be the first experimentally observed "glueball."

Nine glueball states predicted by the Standard Model have masses which are predicted to be consistent to within margin of error (due to incomplete calculations and uncertainties in fundamental constant measurements) of 3020 MeV.  Three are variations of the excited 2-+ tensor glueball state made of two gluons.  Three are variation of the excited 1-- vector glueball state made of three gluons.  Three a variation of the excited 1+- vector glueball state made of three gluons.

While confirmation that the particle resonance observed at 3.02 GeV is indeed a glueball would not be as profound as the discovery of the Higgs boson, it would be considerably more profound in terms of what it would mean for our fundamental understanding of physics than the synthesis of a new atomic element or the discovery of a new (QCD predicted) meson or baryon.


Quantum chromodynamics (QCD) is the part of the Standard Model of Particle Physics that described how the nuclear strong force binds six different kinds quarks (up, down, charm, strange, top, bottom), each of which have one of three strong force color charges (often called red, green and blue although the names are purely arbitrary and have nothing to do with the electromagnetic photon frequency based colors of those names).  Anti-quarks of each type and color are possible as well.  Hence, one might have a blue up quark, or an anti-red strange quark.

Color charge interactions between quarks are transmitted by force carrying particles called gluons, which like photons which transmit the electromagnetic force have zero rest mass.  Gluons do not interact via the electromagnetic force that is transmitted by photons, nor by the weak nuclear force that is transmitted by W bosons and Z bosons.  There are eight different strong nuclear force color charges for gluons that are possible, normally described as pairs of quark colors (e.g. red-antigreen).  Gluons interact with each other via the strong force as well as with color charged quarks.

While gluons have zero rest mass, they do transmit carry energy between quarks and that mass-energy contributes substantially to the mass-energy of mesons and baryons.  About 98% of the mass of a proton or neutron (about 1 GeV) comes from the gluon fields binding the quarks together rather than the masses of the three component quarks (about 13 MeV +/- 50%).  But, after controlling for the masses of constituent quarks, ground state mesons and baryons with the same spin and electromagnetic charges in the constituent quarks have about the same mass contribution attributable to gluons.

All composite particles observed to date are gluon bound systems quarks: two quarks (mesons), three quarks (baryons such as protons and neutrons), or possibly four quarks (tetraquarks), although the last observation appears to be better explained as a two meson "molecule" rather than a true composite four quark particle

The Prediction

One of the most distinctive predictions of quantum chromodynamics (i.e. nuclear strong force physics) that has yet to be confirmed experimentally is the existence of "glueballs" which would be composite particles made entirely of QCD-color charged gluons without any quarks.  QCD predicts a variety of possible glueballs with well defined properties (also, e.g., here and here and here), both in terms of their properties as particles and the kinds of decays that would follow from them.  As the paper notes:
The glueball (G) is a bound state that contains no valence quark but gluons only. This is because gluons, which are charged with colors in QCD and force carriers to bind quarks becoming mesons and baryons, can also glue themselves together to form a bound state. Since it is a unique feature purely for the non-Abelian gauge fields, whether the existence of the gluon condensates can be well established or not appears to be a real test for QCD.
So far, in the Standard Model, the aphorism that everything that is possible is mandatory, has held up well.  But, this is a possibility that has not been detected, although the experimental challenges to detecting glueballs make this lack of a direct observation of them less troubling.

Still, while the issue has attracted very little scholarly attention in the form of proposed theoretical physics models, a model that would rule out glueball states while allowing for the exquisite confirmation of QCD in terms of observed mesons and baryons that are predicted by QCD, would require a major tweak to the Standard Model of Particle physics (such as some new quantum number conservation law) with other consequences (e.g. an implied goldstone boson to correspond to the new symmetry) and implications possible grand unified theories and the like.

Of course, ruling out a prediction of QCD which is so strongly theoretically favored by the Standard Model, despite the fact that QCD is profoundly less precise in its predictions and parameters than the electroweak part of the Standard Model, would call for truly definitive experimental evidence and truly accurate theoretical predictions - so this would be very difficult to prove.


All glueballs have electromagnetic charges aka Q(e) of zero because all of their component gluons have zero electromagnetic charges.

The various possible glueballs predicted by QCD are described by three quantum number (J, P, and C) and a mass commonly stated in mega-electron volts.  All glueballs must be bosons rather than fermions (i.e. must have integer spin).  Two gluon glueballs can have J=0 (scalar or pseudo-scalar), or 2 (tensor).  Three gluon glueballs can have J=1 (vector) or 3.   P and C refer to parity and C-parity in this context (the concepts are rather technical).  These three numbers plus a mass (typically stated in MeV/c^2 units) fully describe a particular type of glueball.

QCD allows for various glueball states that have the same JPC quantum numbers but different masses in excited states (meson and baryons can also have excited states).

Glueball Discovery Efforts Have Been Inconclusive

The paper explains that efforts to match experimental data to glueball candidates have been inconclusive so far:
With the predicted mass around 1.7 GeV, the lightest scalar glueball with the quantum number of J P C = 0++ is allowed to mix with nearby qq¯ mesons in the spectrum.  Since there are two states, f0(1500) and f0(1710), proposed to be composed of the glueball in different mixing scenarios, the identification is obscure. 
The lightest tensor glueball with J P C = 2++ is believed to have a mass close to 1.3 GeV in the MIT bag model and 2.4 GeV in the lattice QCD calculation. For the former, both f2(1270) and f'2'(1525) as the ground states of the 2++ mesons are argued to have the 2++ glueball content, while for the later, fJ(2220) (J = 2 or 4) and f2(2340) are considered to be the candidates, in which the existence of fJ(2220) is still questionable. 
Unlike 0++ and 2++, the difficulty to establish the lightest 0−+ pseudoscalar glueball is that the predicted mass around 2.6 GeV in the lattice QCD calculation has no correspondence with the data.  Nonetheless, η(1405) seems to be a perfect candidate. Particularly, the unseen in γγ reactions reflects that its components are gluons. In addition, X(1835), measured first in the J/Ψ → γpp¯ decays [15], is another possible glueball state at a mass below 2 GeV.  Interestingly, instead of taking the candidates as the pure glueballs, the η − η′ − G and ηc − G mixing scenarios for η(1405) and X(1835) are able to allow their own glueball components to be at least 2 GeV, respectively. Due to the two mixing scenarios, it is not easy to draw a clear conclusion about the glueball state.
This conclusion confirms previous reviews of the literature (also here exhaustively reviewing experimental hints to date as of 2013) on the experimental evidence for glueballs.

While there have been hints that particle resonances might be glueballs since 2005, eight years later, the evidence is still inconclusive and this study, while identifying a promising glueball candidate, does not change that assessment.  These particles, predicted by the Standard Model, remain elusive.

Experimental Challenges

The trouble is that a particle zoo of mesons, baryons and hypothetical glueballs, many of which have similar properties.  There are roughly a hundred ground states of composite mesons and baryons that are possible in QCD and these states have also "excited" state equivalents that are identical in all respects except mass.

Moreover, while some properties of a particle, like its electromagnetic charge, are easy to measure experimentally, others like its spin, charge and parity, can be challenging to reconstruct experimentally.

Any particle with an electromagnetic charge and any particle that behaves like a fermion can't be a glueball.

Distinguishing a glueball from a meson is more challenging.  Mesons which are made of two quarks, like glueballs, are always bosons.  They can have J=0 (scalar and pseudo-scalar) or can have J=1 (vector).

The ease of measuring electromagnetic charge makes it easy to rule out many of these particles as potential glueballs and to make a first rough sort of potential quark components.  But, there are still dozens of potential neutral mesons and baryons that are harder to distinguish from glueball states at first glance, and there are scores of unclassified particle resonances that are clearly not fundamental particles and appear to be bosons, but whose exact composition is undetermined.

A large number of QCD predicted glueball states have masses that cluster around a quite narrow range of masses, mostly between or near the mass of a charm quark (about 1.3 GeV) and the mass of the next heavier bottom quark (about 4.2 GeV).

For comparison purposes, all mesons are at least 134 MeV and the lighest vector meson is 775 MeV and the heaviest definitively identified meson is 6227 MeV.  The proton at 938 MeV is the lightest baryon and the heaviest definitively identified baryon is about 6072 MeV.  Theoretically baryons made up of three bottom quarks could have masses of as much as about 15 GeV give or take.

All of these mass estimates flow from conjectures about how mass would be generated in glueballs that have not be experimentally confirmed in a glueball context where some unconsidered factor might play a part.
The current paper on X(3020), a resonance that has been detected with a four sigma significance, argues that it may be easier to match an "excited" heavier mass state with experimental data than the "ground state" of a glueball type, basically because the particle spectrum isn't so crowded in the case of particles with masses in excess of 3 GeV.  Even excited states of mesons containing only light quarks (u, d and s) are probably too light to fit the observed particle with the observed decay width although an excited Phi meson isn't entirely ruled out.

Better characterization of the particle could clarify the nature of X(3020).  If it is found to be a vector spin particle, it would be the first plausible experimental evidence for such a glueball, as previous hints of possible glueball resonances have focused on only possible spin-0 and spin-2 glueballs.

There are many mesons with zero electromagnetic charge and JPC numbers of 1--.  For example, neutral rho, the omega, the phi, the J/Psi, and the upsilon mesons all have these quantum numbers, although the ground states of each of these mesons are mostly far different from 3020 MeV.  Only the J/Psi at 3096.91 is even remotely close and the decay patterns observed from X(3020) are inconsistent with those of the J/Psi (which is a meson made of a charm quark and an anti-charm quark).

There are several particles that have been observed to have JPC numbers of 1+- which could be mesons, but not have been well characterized.

There are no massive particles predictive by the Standard Model that have quark components with a tensor spin (J=2) other than tetraquarks, so any particle experimentally observed to have tensor spin would be a very strong candidate for glueball status, since the only other candidates are new fundamental particles.
But, distinguishing a particle with tensor spin from other bosons is a subtle and difficult task experimentally.

1 comment:

andrew said...

FYI, I recently made a major addition to the wikipedia article on glueballs.