CR010 $pp \to (T,B) (\bar{T},\bar{B}) \,\text{doublet}\, \to \,\text{mixed final states}\,$
Pair production of VLQ doublet (T,B) Status REVIEWED VERIFIED High Code: NO Priority Low
PDG / equivalent values
| Observable | Value | Year | Experiment / source | Provenance |
|---|---|---|---|---|
| $m_T$ lower limit for weak-isospin (T,B) doublet | m_T > 1.37 TeV 95% CL (lower_limit) | 2018 | ATLAS | source ↑ |
| $m_B$ lower limit for weak-isospin (T,B) doublet | m_B > 1.37 TeV 95% CL (lower_limit) | 2018 | ATLAS | source ↑ |
| $m_T$ lower limit for B(T $ \to $ W b) = 1 | m_T > 1.70 TeV 95% CL (lower_limit) | 2024 | ATLAS | source ↑ |
| $m_B$ lower limit for B(B $ \to $ H b) = 1 | m_B > 1.57 TeV 95% CL (lower_limit) | 2024 | CMS | source ↑ |
| $m_B$ lower limit for B(B $ \to $ W t) = 1 | m_B > 1.56 TeV 95% CL (lower_limit) | 2023 | CMS | source ↑ |
| $m_B$ lower limit for B(B $ \to $ Z b) = 1 | m_B > 1.54 TeV 95% CL (lower_limit) | 2024 | CMS | source ↑ |
| $m_T$ lower limit valid for all T decays to third-generation quarks in the CMS leptonic VLQ scan | m_T > 1.48 TeV for all decays to third-generation quarks 95% CL (lower_limit) | 2023 | CMS | source ↑ |
Why this constrains the RS scan
Custodial Randall--Sundrum models contain heavy vector-like fermion partners
of the third generation, and the direct searches constrain those partner
masses whenever the spectrum, charges, and prompt decays resemble the LHC
simplified models. The most robust piece is the QCD pair-production rate;
the interpretation still depends on whether the first custodial partners are
identified with the searched \(T\) and \(B\), whether they are approximately
mass degenerate, and which of \(Wb,Zt,Ht,Wt,Zb,Hb\) saturate the branching
fractions.
For the anarchic-flavor scan in this repo, CR010 is a cross-check rather than
the leading mass setter. The quark-scan methodology note reports
\(\Mkk^{\min}(p50,\gs=3)=47.26~\mathrm{TeV}\) for the low-energy
\(\Delta F=2\) envelope, far above the \(1\)--\(2~\mathrm{TeV}\) direct VLQ
partner reach. Collider VLQ searches therefore matter most for non-anarchic
or custodial spectra in which the low-energy flavor pressure is softened, or
for diagnosing whether a nominally allowed point predicts third-generation
partners that should already have been seen.
What's changed since the original paper
Early reinterpretation work after the theoretical VLQ search taxonomy showed
how limits depend on branching fractions rather than only on a single decay
mode. Garberson--Golling used CMS multilepton data to establish the
branching-fraction-plane logic used by later searches. CMS then moved to
13 TeV boosted-object searches; the 2017 single-lepton analysis extended the
early doublet-scenario \(T\) reach.
The major Run-2 consolidation was the ATLAS 2018 combination
arXiv:1808.02343, which combined \(T\bar T\) and \(B\bar B\) channels
and set the current doublet-specific \(1.37~\mathrm{TeV}\) mass limit. Full
Run-2 analyses then improved simplified-mode reach: CMS
arXiv:2209.07327 used single-lepton, same-sign dilepton, and
multilepton final states with \(138~\mathrm{fb}^{-1}\), and ATLAS
arXiv:2401.17165 pushed the pure \(T\to Wb\) endpoint to
\(1.70~\mathrm{TeV}\). The CMS 2025 review arXiv:2405.17605
summarizes the Run-2 VLQ search program and records the move toward combined
and projection studies.Validity and model dependence
The published pair-production limits assume prompt decays, narrow or
analysis-compatible resonance widths, and decay topologies dominated by
third-generation Standard Model bosons plus quarks. The doublet limit also
assumes the weak-isospin \((T,B)\) pattern and \(|V_{Tb}|\ll |V_{tB}|\); it is
not a model-independent lower bound on every custodial-RS fermion. Cascades
through other KK states, compressed spectra, nonstandard decays, or large
single-production contributions require a dedicated reinterpretation.
For RS use, the collider observable is therefore best treated as a mass-limit
surface conditional on spectrum and branching-fraction inputs, not as a
single universal \(M_{KK}\) cut. It constrains partner masses directly; only
a model-specific spectrum calculation can translate that into a bound on the
gauge KK scale or the flavor scan's \(\Mkk\).
Code coverage in this repo
NO. A refined grep for VLQ, vector-like, top-partner, bottom-partner,
ATLAS/CMS collider-limit, and reinterpretation-tool keywords found no CR010
implementation in
quarkConstraints/, qcd/,
flavorConstraints/, neutrinos/, yukawa/,
warpConfig/, solvers/, scanParams/, or
tests/. The only ATLAS/CMS-adjacent grep hit was
tests/test\_alpha\_s.py:89, a CMS/RunDec \(\alpha_s\) example, not a
collider constraint.
Adjacent evidence confirms the gap. quarkConstraints/scan.py:359 through
quarkConstraints/scan.py:441
computes \(\Mkk\), evaluates \(\Delta F=2\) constraints, and records
\(\epsilon_K\), \(B_d\), \(B_s\), and \(D\)-mixing pass/fail ratios, with no
ATLAS/CMS mass-exclusion gate. quarkConstraints/validation.py:108
through quarkConstraints/validation.py:145
uses the same \(\Delta F=2\) summary for benchmark validation.Implementation difficulty
HIGH. A useful live CR010 constraint would need a spectrum-to-final-
state map, branching fractions, efficiencies or simplified-likelihood inputs,
and a recast layer such as CheckMATE, MadAnalysis5, or SModelS. A hard-coded
mass threshold would be misleading because the published exclusion changes
substantially across branching fractions, mass splittings, widths, and
nonstandard RS decay chains.
Reason: A live CR010 constraint requires spectrum and branching-fraction matching plus a recast or simplified-likelihood layer; direct hard-coded mass cuts would miss the strong branching, width, mass-splitting, and cascade-dependence of LHC VLQ searches.
Key references
PDG2025\_TprimeListing;
PDG2025\_BprimeListing;
AguilarSaavedra2009\_TopPartners;
GarbersonGolling2013\_ExoticQuarks;
CMS2017\_TBBoosted;
ATLAS2018\_TBCombination;
CMS2023\_VLQLeptonic;
ATLAS2024\_VLQLeptonJets;
CMSReview2025\_VLQ.