Pith. sign in

REVIEW 3 major objections 5 minor 9 references

Zb tetraquark channel and $B\bar B^*$ interaction from lattice QCD

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A lattice QCD computation shows that B and B* attract each other in the Zb tetraquark channel, and with a fitted single-channel potential the two mesons form a near-threshold virtual state about 32 MeV below the BB* threshold.

desk verdict The lattice attraction between B and B* is the robust, citable result; the Zb virtual-state and deep-bound-state poles are fragile outputs of a hand-picked potential extrapolation and should not be quoted as lattice predictions. read the letter →

arxiv 1909.02356 v2 pith:THI3GUHR submitted 2019-09-05 hep-lat hep-ph

S. Prelovsek , H. Bahtiyar , J. Petkovic This is my paper · ORCID
classification hep-lathep-ph
keywords latticeQCDtetraquarkZb(10610)Zb(10650)Born-OppenheimerapproximationstaticheavyquarksBB*interactionexotichadrons
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper asks whether the exotic Zb tetraquarks, quark content $\bar b\bar b\bar u d$, can be understood as $B\bar B^*$ systems held together by a strong attraction. It reports a preliminary lattice QCD calculation in which the heavy $b$ quarks are held fixed at a separation $r$ and the energies of the light $u/d$ quarks and gluons are computed. The key result is that the $B\bar B^*$-dominated eigenstate lies substantially below $m_B + m_{B^*}$ for $r$ between 0.1 and 0.4 fm, a signal of sizable attraction. Treating that eigenstate as the pure $B\bar B^*$ channel yields a potential, and solving the non-relativistic Schrödinger equation with a fitted potential produces a virtual bound state about 32 MeV below threshold, a plausible relative of the experimentally observed $Z_b(10610)$, plus a deep bound state about 403 MeV below threshold. The near-threshold result matters because it offers a first-principles route to the $Z_b$ states, while the deep state is a sharp prediction that can be searched for in the $\Upsilon(1S)\pi$ invariant-mass distribution.

What carries the argument

The load-bearing machinery is the Born-Oppenheimer factorization of the $\bar b\bar b\bar u d$ system: the heavy $b$ and $\bar b$ are static at separation $r$, and the light $u/d$ quarks and gluons form eigenstates $E_n(r)$. The relevant eigenstates are identified from a $6\times6$ correlation matrix of operators resembling $B\bar B^*$, $\Upsilon\pi(\vec p=0)$, $\Upsilon\pi(|\vec p|=2\pi/L)$, $\Upsilon\pi(|\vec p|=4\pi/L)$, a derivative-type $B\bar B^*$, and $\Upsilon b_1$, extracted with the full-distillation method and a variational (GEVP) analysis. The energy $E_{B\bar B^*}(r)$ supplies $V(r)=E(r)-m_B-m_{B^*}$; the second step promotes the heavy quarks to finite mass and solves the non-relativistic Schrödinger equation to find scattering-matrix poles. The two assumptions carrying the argument are that the $B\bar B^*$-dominated eigenstate has no other Fock components, and that the potential below the lattice spacing is described by the exponential form with exponent $p=3/2$.

What would settle it

Compute the same correlation matrix on a finer lattice ($a\approx0.05$ fm) so that $r=0.1$ fm is resolved by several lattice points, and check whether the $B\bar B^*$-dominated eigenstate remains tens of MeV below $m_B+m_{B^*}$ at $r\approx0.2$ fm; also measure its overlaps with the $\Upsilon\pi$ operators and verify they stay below about 2%. If the energy shift shrinks, or the overlap to $\Upsilon\pi$ grows, the single-channel potential and the two derived poles are artifacts of the assumed form.

Watch

Extended reading notes

Core claim

The paper's central result is the $r$-dependent energy spectrum of the $\bar b\bar b\bar u d$ system with $I=1$, $S_{\rm heavy}=1$, and conserved product $C\cdot P=-1$, computed on a 2 fm lattice with two dynamical quark flavors at $m_\pi\simeq 266$ MeV. In this spectrum the state whose overlap is dominated by the $B\bar B^*$ operator (the red-cross eigenstate in the figures) has energy significantly below the non-interacting $B\bar B^*$ threshold for $r\in[0.1,0.4]$ fm, and approaches the threshold at larger $r$. Under the stated assumption that this eigenstate consists exclusively of $B\bar B^*$, the paper extracts the potential $V(r)=E(r)-m_B-m_{B^*}$ and fits it to $V(r)=-A e^{-(r/d)^{3/2}}$ with $A=0.99(5)$ and $d=1.84(10)$. Solving the non-relativistic Schrödinger equation with the measured $B$ and $B^*$ masses then gives an s-wave virtual bound-state pole at $-32^{+29}_{-5}$ MeV below threshold and a deep s-wave bound state at $-403\pm70$ MeV below threshold. The paper presents the first as a plausible lattice description of $Z_b(10610)$, noting the similarity between the predicted $B\bar B^*$ rate peak and the observed rate, and the second as a surprising prediction that has not yet been seen experimentally.

Load-bearing premise

Everything after the spectrum—the potential, the virtual pole, and the deep bound state—depends on the belief that the red-cross eigenstate is purely $B\bar B^*$, so its energy can be read directly as the $B\bar B^*$ potential, and on the guessed form of that potential for distances smaller than the lattice spacing (about 0.12 fm).

Editorial extensions

If this is right

  • A near-threshold enhancement in the $B\bar B^*$ scattering rate is predicted, with a shape resembling the observed $Z_b(10610)$ peak, making the virtual pole directly testable in measured invariant-mass distributions.
  • The deep bound state at roughly 400 MeV below the $B\bar B^*$ threshold is a specific prediction that can be searched for in $\Upsilon(1S)\pi^+$ invariant-mass spectra; current data are not flat but do not resolve it.
  • Because the $B\bar B^*$ eigenstate approaches the non-interacting energy for $r\gtrsim0.5$ fm, the attraction is short-range; accurate large-$r$ data would be needed to determine whether one-pion exchange plays any role.
  • The pole positions depend on the unknown short-distance behavior of $V(r)$; the fitted form $V(r)=-A e^{-(r/d)^{3/2}}$ with $p=3/2$ is an assumption, and a better-motivated short-range potential is the stated next step.
  • The lower $\Upsilon\pi$ and $\Upsilon b_1$ eigenstates show no statistically significant energy shifts, so within current precision the $\Upsilon\pi$ interaction is unresolved and does not visibly mix into the $B\bar B^*$ state.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A finer lattice and larger volume would allow the potential to be measured at $r\approx0.1$ fm rather than extrapolated; I would expect the near-threshold virtual pole to shift less than the deep pole, since the deep pole is controlled mainly by the unmeasured $r\to0$ region.
  • The present result fixes $S_{\rm heavy}=1$; the physical $Z_b$ states could mix with $S_{\rm heavy}=0$, so a complete account would likely require a potential matrix spanning both heavy-spin sectors and the $\Upsilon\pi$ channel.
  • A high-statistics search in $\Upsilon(1S)\pi^+$ invariant mass is a clean discriminator: a broad structure roughly 400 MeV below the $B\bar B^*$ threshold would be evidence for the deep state, while a smooth background would disfavor it.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This proceedings paper reports a preliminary N_f=2 lattice QCD study of the \bar b\bar b d\bar u system with static b quarks and heavy-quark spin S_heavy=1. The authors compute eigen-energies E_n(r) as a function of the b-\bar b separation using six interpolating operators and GEVP; they identify an eigenstate dominated by B\bar B^*. Its energy lies significantly below m_B+m_{B*} for r between 0.1 and 0.4 fm, which is interpreted as a sizable attractive potential. Assuming this state is exclusively B\bar B^*, they extract V(r)=E(r)-m_B-m_{B*}, fit it to V(r)=-A exp(-(r/d)^{3/2}) using four lattice separations, and solve the non-relativistic Schrödinger equation with physical B meson masses. They find an s-wave virtual bound state 32^{+29}_{-5} MeV below threshold and a deep bound state 403±70 MeV below threshold. The paper emphasizes that the attraction result is the robust conclusion and that the pole positions rest on additional simplifying assumptions.

Significance. The direct observation of strong attraction in the static limit is the paper's main strength and is genuinely valuable: it is visible in the raw eigen-energies and does not depend on the potential fit. The paper is also explicitly worded about its approximations and labels the study as preliminary. If the short-range potential were constrained reliably, the Schrödinger-equation step would be a promising route to connect lattice potentials to Z_b phenomenology. However, the quantitative pole predictions are not yet robust lattice observables: they depend on an unmeasured extrapolation of V(r) below the lattice cutoff and on the exclusive B\bar B^* assumption. In its current form the paper therefore supports the qualitative conclusion, while the specific pole masses require additional robustness checks before they can be considered established.

major comments (3)
  1. [Section 5, Eq. (5.1)] The fit of the potential uses only lattice separations r/a = 1..4 and fixes the exponent p = 3/2 by hand. Since the lattice data constrain V(r) only for r >= a ~ 0.124 fm, the bound-state poles obtained from the Schrödinger equation are controlled by an unmeasured extrapolation below a. The quoted uncertainties on W_B reflect only the statistical fit errors of A and d, not the systematic choice of the functional form. A sensitivity test with alternative forms (e.g., p=1, p=2, or a potential capped at V(a) for r<a) is necessary; without it, the claims of a virtual bound state at -32^{+29}_{-5} MeV and a deep bound state at -403±70 MeV are not robust. Please add such tests or, if they are not available, explicitly demote these numbers to illustrative values.
  2. [Section 5, Fig. 2 and footnote 2] The extraction V(r)=E(r)-m_B-m_{B*} assumes that the red-cross eigenstate is exclusively B\bar B^*. The normalized overlaps quoted in footnote 2 support dominance, but they do not exclude small admixtures of \Upsilon\pi or \Upsilon b1 that could shift the eigen-energy and hence V(r). Since the pole positions in the single-channel Schrödinger equation inherit this assumption, the paper should either quantify the effect of possible mixing or state clearly that the pole prediction assumes exactly zero mixing; the current text mentions the assumption but does not assess its impact.
  3. [Section 5, paragraph containing W_B = -403±70 MeV] The deep bound state at -403±70 MeV is presented as a 'surprising' finding and a potentially falsifiable prediction. Given that it is driven entirely by the extrapolated short-range part of the ansatz (5.1), and that the authors state more physical potentials will appear in a forthcoming publication, presenting this number with a statistical-only uncertainty of ±70 MeV is misleading. At minimum the systematic uncertainty from the choice of V(r) below a should be included; otherwise the deep state should be described as a model-dependent consequence of the ansatz rather than a lattice prediction.
minor comments (5)
  1. [Section 4, second paragraph] The sentence beginning 'The eigenstate dominated by B\bar B^*...' contains the typo 'but has it has'; it should read 'but it has'.
  2. [Section 5, paragraph on scattering matrix] 'cmf momenta' should be written as 'center-of-mass momenta' (c.m.).
  3. [Eq. (5.1)] The units of A and d are not specified; they should be stated explicitly (e.g., A in GeV and d in fm) to make the fit parameters unambiguous.
  4. [Introduction and Section 5] The relation to the earlier studies [4,5] could be stated more clearly; the text says the only preliminary study was reported in [4,5], but it is not obvious to the reader which elements are new in the present work beyond the inclusion of \Upsilon\pi(\vec p \neq 0) operators.
  5. [Footnote 2] The definition of the normalized overlap is terse; one explanatory sentence defining \tilde Z^i_n and its role in identifying the dominant Fock component would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the lattice potential is extracted from simulated eigen-energies and the bound-state poles are genuine outputs of a subsequent Schrödinger solve, not refitted inputs.

full rationale

The paper's derivation chain is self-contained against its own inputs. The lattice QCD calculation determines eigen-energies E(r) for static b and bbar; the potential is then defined as V(r)=E(r)-m_B-m_B* under an explicitly stated assumption that the red-cross eigenstate is exclusively B Bbar* (Section 5). This is an extraction, not a construction designed to reproduce Zb masses. The parameters A and d in V(r)=-A exp(-(r/d)^(3/2)) are fitted only to the lattice-determined V(r) at r/a=1..4, and the Schrödinger equation is solved afterward; the virtual bound state and deep bound state are outputs of that solve, not fitted targets. The comparison with Belle data is qualitative and interpretive, appearing only after the poles are obtained. The cited prior lattice studies [4,5] are by different authors and serve as methodological inspiration/context, not as load-bearing evidence for the central result. The assumptions about the small-r form of V and the exclusive B Bbar* Fock content are acknowledged limitations, not circular reductions. Therefore no circular step meets the standard of quoting a specific equation that is equivalent to its inputs by construction.

Assumptions & free parameters 3 free parameters · 5 assumptions · 1 invented entities

The central derivation leans on three categories of input: lattice parameters that define the simulated theory, the single-channel reduction that turns one eigen-energy into a potential, and the assumed functional form for short distances. Free parameters A, d and p are fitted or fixed by hand to the lattice potential, not derived from data; the pole positions therefore inherit their model dependence.

free parameters (3)
  • A = 0.99(5)
    Amplitude in the assumed potential (5.1), fitted to lattice V(r) for r/a = 1..4; it controls the depth of attraction and therefore the bound-state pole positions.
  • d = 1.84(10)
    Range parameter in the assumed potential (5.1), fitted to lattice V(r); it sets where the attraction turns off and influences the pole energies.
  • p = 3/2 (chosen by hand)
    Exponent in the assumed potential (5.1), fixed by hand rather than derived; it controls the short-range shape and strongly affects the deep bound-state energy.
assumptions (5)
  • domain assumption Born-Oppenheimer separation of heavy and light degrees of freedom, with b quarks fixed as static sources.
    Used in Section 1 and Section 5; valid for large m_b, but finite-mass corrections are not estimated.
  • ad hoc to paper The B B* eigenstate is exclusively B B*, with negligible coupling to Upsilon-pi and Upsilon-b1 states.
    Stated as a serious simplifying approximation in Section 5; overlap ratios are small (< 0.02) but not zero, and coupled-channel effects are deferred to future work.
  • ad hoc to paper The potential has the analytic form V(r) = -A exp(-(r/d)^p) with p = 3/2 for all r, including r below the lattice cutoff where it is not measured.
    Eq. (5.1); the authors state more physical forms will be considered in a forthcoming publication. The deep bound-state result depends on this extrapolation.
  • domain assumption The N_f = 2 gauge ensemble with m_pi = 266 MeV, a = 0.124 fm and L = 2 fm approximates the physical heavy-light system.
    Section 3; unphysical pion mass and small volume shift thresholds and may affect the energy levels.
  • domain assumption The non-relativistic Schrodinger equation with measured B and B* masses describes the two-heavy-meson dynamics.
    Section 5; standard Born-Oppenheimer input, but spin-dependent and relativistic corrections are not included.
invented entities (1)
  • Deep B B* bound state near -400 MeV independent evidence
    purpose: Emerges from the assumed potential as a bound state 403 +/- 70 MeV below the B B* threshold; proposed as a possible explanation of a non-flat Upsilon(1S) pi+ invariant-mass distribution.
    The paper specifies a concrete search channel, Zb to Upsilon(1S) pi+, where existing Belle data could reveal or exclude such a state; this gives a falsifiable handle, though the state is model-dependent.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Zb tetraquark channel and $B\bar B^*$ interaction from lattice QCD." pith.science (2026). https://pith.science/paper/THI3GUHR

@misc{pith2026190902356,
  author       = {Pith},
  title        = {Pith review of: Zb tetraquark channel and $B\bar B^*$ interaction from lattice QCD},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/THI3GUHR}},
  note         = {Machine review of arXiv:1909.02356}
}
abstract

Two tetraquark candidates $Z_b(10610)$ and $Z_b(10650)$ with flavor structure $\bar bb\bar du$ were discovered by Belle experiment in 2011. We present a preliminary $N_f=2$ lattice study of the $\bar bb\bar du$ system in the approximation of static $b$ quarks, where the total spin of heavy quarks is fixed to one. The ground and the excited eigen-energies are determined as a function of separation $r$ between $b$ and $\bar b$. The lower eigenstates are related to a bottomonium and a pion. One of the higher eigenstates is dominated by $B\bar B^*$: its energy is significantly below $m_B+m_{B*}$ for r=[0.1,0.4] fm, which suggests sizable attraction. The attractive potential $V(r)$ between $B$ and $\bar B^*$ is extracted assuming that this eigenstate is related exclusively to $B\bar B^*$. Assuming a certain form of the potential and solving non-relativistic Schrodinger equation, we find a bound state pole below $B\bar B^*$ threshold. For certain parametrizations, the bound state is very close to the $B\bar B^*$ threshold - this feature could be related to $Z_b(10610)$ in the experiment.

Figures

Figures reproduced from arXiv: 1909.02356 by the authors.

Figure 1
Figure 1. (a) System considered. (b-d) Two-hadron Fock components relevant in the system with [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Eigen-energies of bb¯ du¯ system (Fig. 1a) for various separations r between static quarks b and b¯ are shown by points. The label on the right indicates which two-hadron Fock component dominates each eigenstate. The dot-dashed lines indicate related two-hadron energies E n.i. (4.1) in the limit when two hadrons (1.2) do not interact. The most important conclusion based on this spectra is that BB¯∗ eigenstate (red c… view at source ↗
Figure 3
Figure 3. Left: The potential V(r) between B and B¯∗ as function of separation r extracted from our lattice simulation. It is based on a simplifying approximation discussed in Section 5 and can be prone to lattice discretization errors for r/a ' 1. Right: Fit assuming the form (5.1) for central values of parameters. Lattice spacing is a ' 0.124 fm. V(r)]u(r) =Wu(r) for finite (measured) B (∗) meson masses and 1/µ = 1/m exp B … view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: (a) The expected rate NBB¯∗ ∝ kσBB¯∗ ∝ sin2 δ/k based on our lattice results for the potential (5.1) with A = 1.02, d = 1.9 that are within the uncertainty range. (b) Rate related to Zb → BB¯∗ and NBB¯∗ by Belle [3] (c) Rate related to Zb → ϒ(1s)π by Belle ( [PITH_FUL…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

9 extracted references · 1 canonical work pages

  1. [1]

    Bondar et al., Phys

    Belle, A. Bondar et al., Phys. Rev. Lett. 108, 122001 (2012), [arXiv:1110.2251]

    arXiv 2012

  2. [2]

    Garmash et al., Phys

    Belle, A. Garmash et al., Phys. Rev. D91, 072003 (2015), [arXiv:1403.0992]

    arXiv 2015

  3. [3]

    Garmash et al., Phys

    Belle, A. Garmash et al., Phys. Rev. Lett. 116, 212001 (2016), [arXiv:1512.07419]

    arXiv 2016

  4. [4]

    Investigation of $B\bar B$ four-quark systems using lattice QCD

    A. Peters, P. Bicudo, K. Cichy and M. Wagner, J. Phys. Conf. Ser. 742, 012006 (2016), [arXiv:1602.07621]

    work page Pith review arXiv 2016

  5. [5]

    Peters, P

    A. Peters, P. Bicudo and M. Wagner, EPJ Web Conf. 175, 14018 (2018), [arXiv:1709.03306]

    arXiv 2018

  6. [6]

    Braaten, C

    E. Braaten, C. Langmack and D. H. Smith, Phys. Rev. D90, 014044 (2014), [arXiv:1402.0438]

    arXiv 2014

  7. [7]

    Brambilla, G

    N. Brambilla, G. Krein, J. Tarrus Castella and A. Vairo, Phys. Rev. D97, 016016 (2018), [arXiv:1707.09647]

    arXiv 2018

  8. [8]

    Peardon et al., Phys

    Hadron Spectrum, M. Peardon et al., Phys. Rev. D80, 054506 (2009), [arXiv:0905.2160]

    arXiv 2009

Show all 9 references
  1. [9]

    Wang et al., Phys

    Q. Wang et al., Phys. Rev. D98, 074023 (2018), [arXiv:1805.07453]. 6

    2018 arXiv

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.