Catalog · Top / Higgs / EW · T005

T005 $\mathcal{B}(t\to c g)$

Flavor-changing neutral-current top decay t $ \to $ c g

Status SUBTLETY-ADDED VERIFIED High Code: NO Priority Low

PDG / equivalent values

Observable	Value	Year	Experiment / source	Provenance
ATLAS dataset for cg $ \to $ t single-top FCNC search	139 fb^-1 at sqrt(s) = 13 TeV integrated luminosity and collision energy	2022	PDG2025:T005:ATLAS2022:dataset	source ↑
$\sigma(c$ g $ \to $ t) * B(t $ \to $ b W) * B(W $ \to $ l $\nu$)	< 4.7 pb, 95% CL upper limit	2022	PDG2025:T005:ATLAS2022:cg_to_t_cross_section	source ↑
B(W $ \to $ l $\nu$) sum over e, $\mu$, $\tau$ modes used by ATLAS	0.325 branching fraction	2022	PDG2025:T005:ATLAS2022:w_leptonic_sum	source ↑
$\|C_{uG}^{ct}\|$ / $\Lambda^2$	< 0.14 TeV^-2, 95% CL upper limit	2022	PDG2025:T005:ATLAS2022:CuGct	source ↑
B(t $ \to $ c g), ATLAS AAD 2022T EFT interpretation	< 3.7e-4 branching fraction, 95% CL upper limit	2022	PDG2025:T005:ATLAS2022:t_cg	source ↑
CMS dataset for t-channel single top FCNC search	5.0 fb^-1 at sqrt(s) = 7 TeV and 19.7 fb^-1 at sqrt(s) = 8 TeV integrated luminosity and collision energy	2017	CMS2017:T005:dataset	source ↑
$\|\kappa_{tcg}\|$ / $\Lambda$	< 1.8e-2 TeV^-1, 95% CL upper limit	2017	CMS2017:T005:kappa_tcg	source ↑
B(t $ \to $ c g), CMS t-channel single-top interpretation	< 4.1e-4 branching fraction, 95% CL upper limit	2017	CMS2017:T005:t_cg	source ↑
Standard Model prediction for B(t $ \to $ c g), theory-context benchmark	4.6e-12 branching fraction	2004	AguilarSaavedra2004:T005:SM	source ↑
Generic ordinary-RS KK-gluon mass scale from flavor constraints, theory context	about 21 TeV TeV, theory-context mass scale	2008	CFW2008:T005:ordinary_RS_KK_gluon_context	source ↑
Composite pseudo-Goldstone Higgs setup KK-gluon mass scale from flavor constraints, theory context	about 33 TeV TeV, theory-context mass scale	2008	CFW2008:T005:pNGB_Higgs_KK_gluon_context	source ↑

Why this constrains the RS scan

This channel probes color-carrying top-charm flavor violation rather than the neutral-meson $\Delta F=2$ four-quark operators already audited in the quark scan. In anarchic warped flavor, the heavy color sector and the IR-localized top can generate non-universal mass-basis couplings and loop-level chromomagnetic dipoles. A $t\to c g$ limit is therefore a direct up-sector top-flavor diagnostic, complementary to $D^0$ mixing and to electroweak channels such as $t\to cZ$ or $t\to cH$.

What's changed since the original paper

The 2008 CFW reference point emphasized KK-gluon exchange and generic anarchic-flavor bounds near $21~\mathrm{TeV}$ for ordinary RS and $33~\mathrm{TeV}$ in the composite pseudo-Goldstone Higgs setup, but it did not provide a direct $t\to c g$ catalog limit (CsakiFalkowskiWeiler2008WarpedFlavor). Since then the important change is experimental: LHC searches now constrain the top-gluon FCNC operator directly. CMS used 7 and 8 TeV t-channel single-top data (CMS2017TopFCNCTUG), while ATLAS later used the full Run-2 $139~\mathrm{fb}^{-1}$ dataset at 13 TeV with a neural-network separation of $ug\to t$, $cg\to t$, and backgrounds (ATLAS2022TopFCNCTUG). The resulting limits remain many orders of magnitude above the SM prediction, so they are new-physics limits rather than SM precision tests.

Validity and model dependence

The experimental upper bound is clean in the sense that the SM rate is negligible. Its use as an RS scan cut is model dependent: ATLAS quotes a left-handed SMEFT $C_{uG}^{ct}$ interpretation, while CMS uses an older $\kappa_{tcg}$ convention. The conversion from single-top production to ${\cal B}(t\to c g)$ assumes one FCNC operator at a time, standard top decays, and a specific normalization of the chromomagnetic interaction. A model producing both chiralities, contact terms, or correlated heavy-gluon resonance effects needs a dedicated reinterpretation. In down-aligned or kaon-protected RS variants, top FCNCs become leading rather than secondary diagnostics.

Code coverage in this repo

NO. The required plan greps were rerun on 2026-05-16, followed by a targeted search for $t\to c g$, $tcg$, $tqg$, $C_{uG}$, and $cg\to t$ across quarkConstraints/, qcd/, flavorConstraints/, neutrinos/, yukawa/, warpConfig/, solvers/, scanParams/, and tests/. The only implementation-directory hits were generic FCNC prose/comments at quarkConstraints/PAPER\_0710\_1869.md:35 and tests/test\_paper\_couplings.py:258. The modern phenomenology surface enumerates only $\epsilon_K$, $K$, $B_d$, $B_s$, and $D^0$ at quarkConstraints/modern/phenomenology.py:23, and states that it is policy-only at quarkConstraints/modern/phenomenology.py:165.

Linked evidence (opens GitHub blob at flavor-catalog-website/2026q2):

Implementation difficulty

HIGH. Cataloging the external limit is straightforward, but live integration would need a top chromomagnetic FCNC operator convention, RS-to-SMEFT matching, decay-width and $cg\to t$ production normalization, PDF/collider acceptance assumptions or a validated recast, and QCD-running choices not present in the existing $\Delta F=2$ code.

Reason: Limit cataloging is MEDIUM, but production-grade integration is HIGH because the repo lacks a top chromomagnetic FCNC operator convention, RS-to-SMEFT matching, $t \to c$ g width and c $g \to t$ production normalization, collider/PDF reinterpretation machinery, and QCD-running choices for this operator.

Key references

PDG2025TopFCNCTUG; ATLAS2022TopFCNCTUG; CMS2017TopFCNCTUG; AguilarSaavedra2004TopFCNC; CsakiFalkowskiWeiler2008WarpedFlavor.

value 139 fb^-1 at sqrt(s) = 13 TeV

RESOLVED

Match snapshot line 19

L16: and kappa^{ctg}/Lambda" and lists values in TeV^{-1} with 95% CL.
L17: 
L18: PDG footnote 1, AAD 2022T / ATLAS:
L19: - Based on 139 fb^{-1} of pp data at sqrt(s) = 13 TeV.
L20: - Single top-quark production limits:
L21:   sigma(u g -> t) * B(t -> b W) * B(W -> l nu) < 3.0 pb.
L22:   sigma(c g -> t) * B(t -> b W) * B(W -> l nu) < 4.7 pb.

value < 4.7

RESOLVED

Match snapshot line 22

L19: - Based on 139 fb^{-1} of pp data at sqrt(s) = 13 TeV.
L20: - Single top-quark production limits:
L21:   sigma(u g -> t) * B(t -> b W) * B(W -> l nu) < 3.0 pb.
L22:   sigma(c g -> t) * B(t -> b W) * B(W -> l nu) < 4.7 pb.
L23: - EFT coupling limits:
L24:   |C_uG^{ut}| / Lambda^2 < 0.057 TeV^{-2}.
L25:   |C_uG^{ct}| / Lambda^2 < 0.14 TeV^{-2}.

value 0.325

RESOLVED

Match snapshot line 32

L29: - sigma(u+g -> t) * B(t -> Wb) * B(W -> l nu) < 3.0 pb.
L30: - sigma(c+g -> t) * B(t -> Wb) * B(W -> l nu) < 4.7 pb.
L31: 
L32: The arXiv abstract states that B(W -> l nu)=0.325 is the sum over the three
L33: leptonic W modes used in the interpretation.
L34: 
L35: EFT interpretation:

value < 0.14

RESOLVED

Match snapshot line 25

L22:   sigma(c g -> t) * B(t -> b W) * B(W -> l nu) < 4.7 pb.
L23: - EFT coupling limits:
L24:   |C_uG^{ut}| / Lambda^2 < 0.057 TeV^{-2}.
L25:   |C_uG^{ct}| / Lambda^2 < 0.14 TeV^{-2}.
L26: - Equivalent FCNC top-decay branching-fraction limits:
L27:   B(t -> u g) < 6.1e-5.
L28:   B(t -> c g) < 3.7e-4.

value < 3.7e-4

RESOLVED

Match snapshot line 28

L25:   |C_uG^{ct}| / Lambda^2 < 0.14 TeV^{-2}.
L26: - Equivalent FCNC top-decay branching-fraction limits:
L27:   B(t -> u g) < 6.1e-5.
L28:   B(t -> c g) < 3.7e-4.
L29: 
L30: PDG footnote 2, KHACHATRYAN 2017G / CMS:
L31: - Based on 5.0 and 19.7 fb^{-1} of pp data at sqrt(s) = 7 and 8 TeV.

value 5.0 fb^-1 at sqrt(s) = 7 TeV and 19.7 fb^-1 at sqrt(s) = 8 TeV

RESOLVED

Match snapshot line 22

L19: 
L20: The CMS analysis uses t-channel single top-quark events in pp collisions at
L21: sqrt(s) = 7 and 8 TeV, corresponding to integrated luminosities of
L22: 5.0 fb^{-1} and 19.7 fb^{-1}, respectively. Events with one muon and two or
L23: three jets are analyzed with a Bayesian neural-network discriminant.
L24: 
L25: For the FCNC couplings kappa_tug and kappa_tcg, the observed 95% CL upper

value < 1.8e-2

RESOLVED

Match snapshot line 28

L25: For the FCNC couplings kappa_tug and kappa_tcg, the observed 95% CL upper
L26: limits are:
L27: - |kappa_tug| / Lambda < 4.1e-3 TeV^{-1}.
L28: - |kappa_tcg| / Lambda < 1.8e-2 TeV^{-1}.
L29: 
L30: The corresponding FCNC top-decay branching-fraction limits are:
L31: - B(t -> u g) < 2.0e-5.

value 4.6e-12

RESOLVED

Match snapshot line 23

L20: ^{+2.1}_{-0.7}) x 10^{-12}.
L21: 
L22: Table 1 collects branching ratios for top FCN decays. The Standard Model
L23: entry for t -> c g is 4.6 x 10^{-12}. The table also lists representative
L24: extended-model scales, including t -> c g values of about 1.5e-7 for
L25: up-type quark singlets, about 1e-4 for a general 2HDM, about 1e-8 for a
L26: flavour-conserving 2HDM, about 8e-5 in the MSSM, and about 2e-4 with

value about 21 TeV

RESOLVED

Match snapshot line 20

L17: 
L18: The abstract says the authors re-examine flavor constraints in standard
L19: Randall-Sundrum-type scenarios and find that the KK gluon mass should
L20: generically be heavier than about 21 TeV. In their composite
L21: pseudo-Goldstone Higgs setup they find flavor-violating effects that make
L22: the corresponding KK-gluon scale about 33 TeV.
L23:

value about 33 TeV

RESOLVED

Match snapshot line 22

L19: Randall-Sundrum-type scenarios and find that the KK gluon mass should
L20: generically be heavier than about 21 TeV. In their composite
L21: pseudo-Goldstone Higgs setup they find flavor-violating effects that make
L22: the corresponding KK-gluon scale about 33 TeV.
L23: 
L24: The paper explains that the largest FCNC contribution in the standard RS
L25: setup is generated by exchange of heavy gauge bosons, especially the lightest