betaFeedback

WorksPhysics and Astronomy

Inclusive weak decays of the J/ψJ/ψ: single-charm decay and weak annihilation

Abstract

Conventional estimates of J/ψJ/ψ weak decays, based on independent single-charm decays, suggest a branching fraction of order 10−810^{-8}. Searches with samples of approximately 101010^{10} J/ψJ/ψ events have so far yielded upper limits on exclusive modes. We calculate the inclusive weak branching fraction using the optical theorem, including single-charm decay, charged- and neutral-current annihilation, and their interference. The heavy-quark expansion identifies the leading pair-annihilation structure and motivates vacuum saturation for the leading color-singlet component. With tree-level hard kernels and leading-logarithmic Wilson coefficients at μ0=MJ/ψμ_0=M_{J/ψ}, we obtain Bweak=(7.96−0.43+3.43±1.24)×10−8B_{\rm weak}=(7.96^{+3.43}_{-0.43}\pm1.24)\times10^{-8}. The contribution mediated by the WW boson alone, i.e. single-charm decays and WW-exchange annihilation without the WW-ZZ interference, has a branching fraction of (2.37−0.42+1.64±0.13)×10−8(2.37^{+1.64}_{-0.42}\pm0.13)\times10^{-8}. For the flavor sectors (ΔC,∣ΔS∣)=(0,1)(ΔC,|ΔS|)=(0,1) and ∣ΔC∣=1|ΔC|=1, the branching fractions are (4.97−4.97+16.44±0.99)×10−10(4.97^{+16.44}_{-4.97}\pm0.99)\times10^{-10} and (1.86−0.06+0.20±0.02)×10−8(1.86^{+0.20}_{-0.06}\pm0.02)\times10^{-8}, respectively, including charge-conjugate channels. The first uncertainty estimates missing perturbative corrections through scale variation over 1 GeV≤μ≤3MJ/ψ1~\mathrm{GeV}\leqμ\leq3M_{J/ψ}. The second is a hadronic uncertainty, obtained by jointly varying each vacuum-saturation and kinetic/chromomagnetic input within relative ranges of ΛQCD/mc≃20%Λ_{\rm QCD}/m_c\simeq20\%. Pair annihilation dominates, with a substantial neutral-current contribution and scale-sensitive WW-ZZ interference.

Introduction

The J/ψJ/\psi allows weak interactions to be studied in a heavy quark–antiquark bound state dominated by strong and electromagnetic decays. Weak decay proceeds through the decay of either charm constituent or ccˉc\bar{c} annihilation, and the very small expected rates have motivated both theoretical estimates and dedicated searches [1, 2, 3, 4]. A widely used estimate treats the charm and anticharm as independently decaying constituents, with a charmed-meson lifetime as a proxy for the charm-quark decay time, so that the charm-changing width is [5]

Γ∣ΔC∣=1≃2ΓD.(1)\Gamma_{\lvert\Delta C\rvert=1} \simeq2\Gamma_D. \tag*{(1)}

The factor of two accounts for the two constituents; with ΓD=ΓD+\Gamma_D = \Gamma_{D^+}, the branching fraction is 2ΓD/ΓJ/ψ≃1.4×10−82\Gamma_D/\Gamma_{J/\psi} \simeq1.4 \times10^{-8}.

The applicability of the heavy-quark expansion (HQE) at the charm scale has been examined in detail, including radiative and subleading power corrections [6, 7, 8]. For the D+D^+ meson, destructive Pauli interference and cancellations between dimension-six and dimension-seven contributions make the absolute width so sensitive to higher-order corrections and four-quark matrix elements that, for several inputs, the truncated width has an unphysical negative central value [8, 9]. Agreement for lifetime ratios does not by itself establish the accuracy of absolute charm widths; a reorganization retaining phase-space-enhanced four-quark contributions in the leading term has also been proposed [10].

Lifetime differences among hadrons with the same heavy quark arise mainly from four-quark spectator contributions, suppressed by (ΛQCD/mQ)3(\Lambda_{\mathrm{QCD}}/m_Q)^3 but enhanced by a phase-space factor of order 16π216\pi^2 [6, 11]. For bottom hadrons, τ(B+)/τ(B0)=1.083±0.003\tau(B^+)/\tau(B^0) = 1.083 \pm0.003 and τ(Λb0)/τ(B0)=0.970±0.006\tau(\Lambda_b^0)/\tau(B^0) = 0.970 \pm0.006 [12, 13], in agreement with HQE predictions [14, 15]; for charmed hadrons, τ(D+)/τ(D0)≃2.52\tau(D^+)/\tau(D^0) \simeq2.52 and τ(Λc+)/τ(D0)≃0.49\tau(\Lambda_c^+)/\tau(D^0) \simeq0.49 [12]. Figure 1 shows that charm lifetimes span a factor of about seven, whereas bottom ones vary by only about 10%.

Singly heavy hadron lifetimes normalized to $\tau(D^0)$ and $\tau(B^0)$

Figure 1. Singly heavy hadron lifetimes normalized to τ(D0)\tau(D^0) (circles) and τ(B0)\tau(B^0) (squares) [12]; paired baryons share their light-quark content.

The Ωc0\Omega_c^0 is a striking example: long regarded as the shortest-lived singly charmed baryon, it was found to live nearly four times longer than the previous world average [16, 17, 18]. It has been argued that the HQE for the Ωc0\Omega_c^0 breaks down at order 1/mc41/m_c^4 and that a consistent expansion requires τ(Ωc0)>τ(Λc+)\tau(\Omega_c^0)>\tau(\Lambda_c^+) [19], a hierarchy accommodated by later analyses [20, 21]. Doubly charmed baryons show a similarly large splitting, driven by Pauli interference and WW exchange [22, 23, 24, 25, 26, 27, 28].

Exclusive weak channels of the J/ψJ/\psi have been studied using heavy-hadron spin symmetry [1], quark-model form factors combined with factorization [2], QCD sum rules [5, 29], covariant light-front form factors [30], QCD factorization with radiative corrections [31], and a perturbative-QCD approach [32]. These approaches relate individual decay rates to J/ψ→D(s)(∗)J/\psi\to D_{(s)}^{(*)} transition form factors and, for nonleptonic modes, rely on assumptions about hadronic factorization. The summed branching fraction of the nonleptonic two-body spectator channels can be as large as 1.3×10−81.3\times10^{-8} [5]. Such calculations provide channel-specific targets for experiment, but they address only charm-changing final states. Pair annihilation, which produces final states without open charm, requires an inclusive treatment in which single-charm decay and annihilation are organized together.

Substantial experimental effort has been devoted to these rare decays. Searches progressed from 5.77×1075.77\times10^7 J/ψJ/\psi events at BESII [3] to 2.25×1082.25\times10^8 events in early BESIII studies [4], and subsequently to about 101010^{10} events [33, 34, 35]. With these samples, BESIII obtained the 90% confidence-level limits B(J/ψ→D−e+νe+c.c.)<7.1×10−8\mathcal{B}(J/\psi\to D^-e^+\nu_e+\mathrm{c.c.})<7.1\times10^{-8} and B(J/ψ→Ds−e+νe+c.c.)<9.9×10−8\mathcal{B}(J/\psi\to D_s^-e^+\nu_e+\mathrm{c.c.})<9.9\times10^{-8} [33, 35], and limits of 8.0×10−78.0\times10^{-7} and 4.1×10−74.1\times10^{-7} on Ds−ρ+D_s^-\rho^+ and Ds−π+D_s^-\pi^+, including charge-conjugate modes [34]. The program also covers further hadronic modes containing a DD meson [36], the semimuonic channel J/ψ→Ds−μ+νμ+c.c.J/\psi\to D_s^-\mu^+\nu_\mu+\mathrm{c.c.} [37], and the flavor-changing neutral-current mode J/ψ→D0e+e−+c.c.J/\psi\to D^0e^+e^-+\mathrm{c.c.} [38]. No significant signal has been observed. A proposed super tau-charm facility would collect data samples about two orders of magnitude larger [39], bringing branching fractions of order 10−810^{-8} within reach.

§ 2–§ 5 construct the transition operator, develop the heavy-quark expansion, evaluate the matrix elements, and present the numerical results; § 6 concludes.

Theoretical framework

The weak Hamiltonian retained here is

Heff=[HNLc+HSLc+h.c.]+HNL+HZ,q+HZ,ℓ.(2)\mathcal{H}_{\mathrm{eff}} = \left[\mathcal{H}_{\mathrm{NL}}^{c} + \mathcal{H}_{\mathrm{SL}}^{c} + \mathrm{h.c.}\right] + \mathcal{H}_{\mathrm{NL}} + \mathcal{H}_{Z,q} + \mathcal{H}_{Z,\ell}. \tag*{(2)}

Here NL and SL denote nonleptonic and semileptonic decays, and h.c. denotes the Hermitian-conjugate Hamiltonians describing the corresponding single-cˉ\bar{c} decays. We use ΓLμ=γμ(1−γ5)\Gamma_{L}^{\mu} = \gamma^{\mu}(1-\gamma_{5}) and ΓRμ=γμ(1+γ5)\Gamma_{R}^{\mu} = \gamma^{\mu}(1+\gamma_{5}). The charged-current terms, with sums over q1,q2=d,sq_{1},q_{2}=d,s and ℓ=e,μ\ell=e,\mu, take the form [40]:

HNLc=GF2∑q1,q2Vcq1∗Vuq2[C1(qˉ1iΓLμci)(uˉjΓLμq2j)+C2(qˉ1iΓLμcj)(uˉjΓLμq2i)].(3)\mathcal{H}_{\mathrm{NL}}^{c} = \frac{G_{F}}{\sqrt{2}} \sum_{q_{1},q_{2}} V_{cq_{1}}^{*}V_{uq_{2}}\left[C_{1}(\bar{q}_{1i}\Gamma_{L}^{\mu}c^{i})(\bar{u}_{j}\Gamma_{L\mu}q_{2}^{j})+C_{2}(\bar{q}_{1i}\Gamma_{L}^{\mu}c^{j})(\bar{u}_{j}\Gamma_{L\mu}q_{2}^{i})\right]. \tag*{(3)}
HSLc=GF2∑q1,ℓVcq1∗(qˉ1iΓLμci)(νˉℓΓLμℓ).(4)\mathcal{H}_{\mathrm{SL}}^{c} = \frac{G_{F}}{\sqrt{2}} \sum_{q_{1},\ell} V_{cq_{1}}^{*}(\bar{q}_{1i}\Gamma_{L}^{\mu}c^{i})(\bar{\nu}_{\ell}\Gamma_{L\mu}\ell). \tag*{(4)}
HNL=GF2∑q1,q2Vcq1∗Vcq2[C1(qˉ1iΓLμci)(cˉjΓLμq2j)+C2(qˉ1iΓLμcj)(cˉjΓLμq2i)].(5)\mathcal{H}_{\mathrm{NL}} = \frac{G_{F}}{\sqrt{2}} \sum_{q_{1},q_{2}} V_{cq_{1}}^{*}V_{cq_{2}}\left[C_{1}(\bar{q}_{1i}\Gamma_{L}^{\mu}c^{i})(\bar{c}_{j}\Gamma_{L\mu}q_{2}^{j})+C_{2}(\bar{q}_{1i}\Gamma_{L}^{\mu}c^{j})(\bar{c}_{j}\Gamma_{L\mu}q_{2}^{i})\right]. \tag*{(5)}

Here GFG_{F} is the Fermi constant, VV is the CKM matrix, and C1,2C_{1,2} are Wilson coefficients. The superscript cc marks the charm-number-violating terms, while HNL\mathcal{H}_{\mathrm{NL}} conserves charm and induces WW exchange. The neutral-current terms take the form [41]:

HZ,q=2GF∑q,χ,χ′gZ,χcgZ,χ′q[CZχχ′(cˉiΓχμci)(qˉjΓχ′μqj)+C~Zχχ′(cˉiΓχμcj)(qˉjΓχ′μqi)].(6)\mathcal{H}_{Z,q} = \sqrt{2}G_{F}\sum_{q,\chi,\chi'}g_{Z,\chi}^{c}g_{Z,\chi'}^{q}\left[C_{Z}^{\chi\chi'}(\bar{c}_{i}\Gamma_{\chi}^{\mu}c^{i})(\bar{q}_{j}\Gamma_{\chi'\mu}q^{j})+\widetilde{C}_{Z}^{\chi\chi'}(\bar{c}_{i}\Gamma_{\chi}^{\mu}c^{j})(\bar{q}_{j}\Gamma_{\chi'\mu}q^{i})\right]. \tag*{(6)}
HZ,ℓ=2GF∑f,χ,χ′gZ,χcgZ,χ′f(cˉiΓχμci)(fˉΓχ′μf).(7)\mathcal{H}_{Z,\ell} = \sqrt{2}G_{F}\sum_{f,\chi,\chi'}g_{Z,\chi}^{c}g_{Z,\chi'}^{f}(\bar{c}_{i}\Gamma_{\chi}^{\mu}c^{i})(\bar{f}\Gamma_{\chi'\mu}f). \tag*{(7)}

The sums run over q=u,d,sq=u,d,s, f=e,μ,νe,νμ,ντf=e,\mu,\nu_{e},\nu_{\mu},\nu_{\tau}, and χ,χ′=L,R\chi,\chi'=L,R. Here i,ji,j are color indices, with Nc=3N_{c}=3 colors. The neutral-current Wilson-coefficient combinations appearing in eq. (6) are expressed in the CiZC_{i}^{Z} basis of [41] as

CZLL=CZRR=C1Z+C2Z2,C~ZLL=C~ZRR=C1Z−C2Z2,CZLR=CZRL=C4Z,C~ZLR=C~ZRL=C3Z−C4ZNc.(8)\begin{aligned} C_{Z}^{LL}=C_{Z}^{RR}&=\frac{C_{1}^{Z}+C_{2}^{Z}}{2}, & \widetilde{C}_{Z}^{LL}=\widetilde{C}_{Z}^{RR}&=\frac{C_{1}^{Z}-C_{2}^{Z}}{2},\\ C_{Z}^{LR}=C_{Z}^{RL}&=C_{4}^{Z}, & \widetilde{C}_{Z}^{LR}=\widetilde{C}_{Z}^{RL}&=\frac{C_{3}^{Z}-C_{4}^{Z}}{N_{c}}. \tag*{(8)} \end{aligned}

Here CiZC_{i}^{Z}, i=1,…,4i=1,\ldots,4, are the neutral-current Wilson coefficients at the scale μ\mu, normalized to C1Z(MZ)=1C_{1}^{Z}(M_{Z})=1. The chiral neutral-current couplings are

gZ,L=T3−QsW2,gZ,R=−QsW2.(9)g_{Z,L}=T_{3}-Q s_{W}^{2}, \qquad g_{Z,R}=-Q s_{W}^{2}. \tag*{(9)}

Here sW=sin⁡θWs_{W}=\sin\theta_{W}, QQ is the electric charge of the fermion, and T3T_{3} is its weak-isospin eigenvalue. For real Wilson coefficients, the ordered flavor sum makes HNL\mathcal{H}_{\mathrm{NL}} Hermitian, while HZ≡HZ,q+HZ,ℓ\mathcal{H}_{Z}\equiv\mathcal{H}_{Z,q}+\mathcal{H}_{Z,\ell} is Hermitian term by term. The ZZ field has been integrated out, so HZ\mathcal{H}_{Z} is represented by a local four-fermion vertex and no separate propagator factor is retained [41].

Throughout this work, Γweak\Gamma_{\mathrm{weak}} denotes the weak-only contribution at order GF2G_F^2. With the effective Hamiltonian specified above, the optical theorem expresses this inclusive weak decay width through the absorptive part of the forward-scattering amplitude with two Hamiltonian insertions [42, 43]:

Γweak(J/ψ)=12MJ/ψIm⁡i∫d4x ⟨J/ψ∣T{Heff(x)Heff(0)}∣J/ψ⟩=12MJ/ψ⟨J/ψ∣T∣J/ψ⟩.(10)\begin{aligned} \Gamma_{\mathrm{weak}}(J/\psi)=\frac{1}{2M_{J/\psi}}\operatorname{Im} i\int d^4x\,\langle J/\psi|\mathcal{T}\{\mathcal{H}_{\mathrm{eff}}(x)\mathcal{H}_{\mathrm{eff}}(0)\}|J/\psi\rangle\\ &=\frac{1}{2M_{J/\psi}}\langle J/\psi|\mathcal{T}|J/\psi\rangle. \tag*{(10)} \end{aligned}

Here MJ/ψM_{J/\psi} is the J/ψJ/\psi mass, and T\mathcal{T} denotes the inclusive transition operator generated by the double Hamiltonian insertion.

Substituting the effective Hamiltonian into eq. (10) and organizing the double insertions by the topologies in Figure 2 gives

Forward-scattering topologies entering the inclusive weak transition operator

Figure 2. Forward-scattering topologies entering the inclusive weak transition operator: (a) single-charm decay; (b) WW-exchange annihilation; (c) neutral-current annihilation; and (d) WW–ZZ interference. Dashed lines mark unitarity cuts; dots represent local weak-Hamiltonian insertions.

T=Tc+cˉ+Tccˉ,Tccˉ=TWE+TNC+Tint.(11)\mathcal{T}=\mathcal{T}_{c+\bar{c}}+\mathcal{T}_{c\bar{c}},\qquad\mathcal{T}_{c\bar{c}}=\mathcal{T}_{\mathrm{WE}}+\mathcal{T}_{\mathrm{NC}}+\mathcal{T}_{\mathrm{int}}. \tag*{(11)}

The first term contains the decay contributions of the charm and anticharm constituents. Each term in Tccˉ\mathcal{T}_{c\bar{c}} contains the complete heavy pair: TWE\mathcal{T}_{\mathrm{WE}}, TNC\mathcal{T}_{\mathrm{NC}}, and Tint\mathcal{T}_{\mathrm{int}} describe WW-exchange annihilation, neutral-current annihilation, and their interference.

After local matching, the transition operator is organized schematically in the full-QCD basis as

T=GF2mc5192π3[(c3O3+c5mc2O5)2+16π2∑n(cn(8)mc5On)4+⋯ ].(12)\mathcal{T}=\frac{G_F^2m_c^5}{192\pi^3}\left[\left(c_3\mathcal{O}_3+\frac{c_5}{m_c^2}\mathcal{O}_5\right)_2+16\pi^2\sum_n\left(\frac{c_n^{(8)}}{m_c^5}\mathcal{O}_n\right)_4+\cdots\right]. \tag*{(12)}

The first term in eq. (12) describes the single-charm decay sector through the order retained here, while the second describes the leading heavy-pair annihilation kernels. In the latter, QnQ_n denotes a full-QCD four-charm operator with two current derivatives and canonical dimension eight. The index nn labels the topology, chirality, and color contraction, and Cn(8)C_n^{(8)} is a dimensionless short-distance coefficient. The subscripts 2 and 4 mark two- and four-quark operators.

The leading two-quark term arises from the three-body cut of the partonic 1→31 \to3 single-charm decay, whereas the four-quark terms arise from the two-body cuts of the 2→22 \to2 heavy-pair annihilation processes. Relative to the common three-body normalization in eq. (12), the latter carry the conventional phase-space enhancement 16π216\pi^2 [7, 11]. The remaining numerical and mass-dependent factors are contained in the short-distance coefficients.

For Figure 2(a) and its charge-conjugate counterpart, contracting the hard final-state fields gives the standard local OPE [42, 43, 44]. Using the original relativistic charm field, the two-quark sector is

Tc+cˉ=GF2mc5192π3[c3O3+c5mc2O5+∑d,ncd,nmcd−3Od,n].(13)\mathcal{T}_{c+\bar c}=\frac{G_F^2m_c^5}{192\pi^3}\left[c_3\mathcal{O}_3+\frac{c_5}{m_c^2}\mathcal{O}_5+\sum_{d,n}\frac{c_{d,n}}{m_c^{d-3}}\mathcal{O}_{d,n}\right]. \tag*{(13)}

The first two operators in this expansion are

O3=cˉc,O5=cˉgs2σμνGμνc.(14)\mathcal{O}_3=\bar{c}c,\qquad\mathcal{O}_5=\bar{c}\frac{g_s}{2}\sigma_{\mu\nu}G^{\mu\nu}c. \tag*{(14)}

Here mcm_c is the charm-quark mass in the pole convention, and gsg_s is the QCD gauge coupling. Writing Aμ=AμaTaA_\mu=A_\mu^aT^a, we use Dμ=∂μ−igsAμD_\mu=\partial_\mu-ig_sA_\mu and [Dμ,Dν]=−igsGμν[D_\mu,D_\nu]=-ig_sG_{\mu\nu}, where Gμν=GμνaTaG_{\mu\nu}=G_{\mu\nu}^aT^a. The index aa is an adjoint color index, and the fundamental generators obey Tr⁡(TaTb)=δab/2\operatorname{Tr}(T^aT^b)=\delta^{ab}/2. Finally, σμν=i[γμ,γν]/2\sigma_{\mu\nu}=i[\gamma_\mu,\gamma_\nu]/2. In the last term, dd runs over the canonical dimensions d≥6d\ge6, and nn labels the independent operators at fixed dd.

At leading order, neglecting light-fermion masses and summing over the charged-current quark and charged-lepton flavors specified above, the two coefficients are

c3(μ)=(∣Vcs∣2+∣Vcd∣2){2+(∣Vud∣2+∣Vus∣2)[Nc(C12+C22)+2C1C2]},c_3(\mu)=\left(\lvert V_{cs}\rvert^2+\lvert V_{cd}\rvert^2\right)\left\{2+\left(\lvert V_{ud}\rvert^2+\lvert V_{us}\rvert^2\right)\left[N_c(C_1^2+C_2^2)+2C_1C_2\right]\right\},
c5(μ)=−2(∣Vcs∣2+∣Vcd∣2){2+(∣Vud∣2+∣Vus∣2)[Nc(C12+C22)+10C1C2]}.(15)c_5(\mu)=-2\left(\lvert V_{cs}\rvert^2+\lvert V_{cd}\rvert^2\right)\left\{2+\left(\lvert V_{ud}\rvert^2+\lvert V_{us}\rvert^2\right)\left[N_c(C_1^2+C_2^2)+10C_1C_2\right]\right\}. \tag*{(15)}

For the heavy-pair topologies in Figure 2(b)–(Figure 2d), the light fields are contracted between the two Hamiltonian insertions while all four charm fields remain uncontracted. The corresponding forward-scattering contributions, with TNC=TNC,q+TNC,ℓ\mathcal{T}_{\mathrm{NC}}=\mathcal{T}_{\mathrm{NC},q}+\mathcal{T}_{\mathrm{NC},\ell}, are

TWE=−GF212π∑q1,q2∣Vcq1∣2∣Vcq2∣2[C12Q~[L,L]+(NcC22+2C1C2)Q[L,L]].(16)\mathcal{T}_{\mathrm{WE}}=-\frac{G_F^2}{12\pi}\sum_{q_1,q_2}\lvert V_{cq_1}\rvert^2\lvert V_{cq_2}\rvert^2\left[C_1^2\widetilde{Q}[L,L]+(N_cC_2^2+2C_1C_2)Q[L,L]\right]. \tag*{(16)}
TNC,q=−GF23π∑q,χ,χ′,χ′′(gZ,χ′′q)2gZ,χcgZ,χ′c[KZZχχ′;χ′′Q[χ,χ′]+K~ZZχχ′;χ′′Q~[χ,χ′]].(17)\mathcal{T}_{\mathrm{NC},q}=-\frac{G_F^2}{3\pi}\sum_{q,\chi,\chi',\chi''}(g_{Z,\chi''}^{q})^2g_{Z,\chi}^{c}g_{Z,\chi'}^{c}\left[K_{ZZ}^{\chi\chi';\chi''}Q[\chi,\chi']+\widetilde{K}_{ZZ}^{\chi\chi';\chi''}\widetilde{Q}[\chi,\chi']\right]. \tag*{(17)}
TNC,ℓ=−GF23π∑f,χ,χ′,χ′′(gZχ′′f)2gZ,χcgZ,χ′cQ[χ,χ′](18)\mathcal{T}_{\mathrm{NC},\ell}=-\frac{G_F^2}{3\pi}\sum_{f,\chi,\chi',\chi''}(g_{Z\chi''}^{f})^2g_{Z,\chi}^{c}g_{Z,\chi'}^{c}\mathcal{Q}[\chi,\chi'] \tag*{(18)}
Tint=−GF23π∑q1,χ∣Vcq1∣2gZ,Lq1gZ,χc[KWZχQ[L,χ]+K~WZχQ~[L,χ]](19)\mathcal{T}_{\mathrm{int}}=-\frac{G_F^2}{3\pi}\sum_{q_1,\chi}|V_{cq_1}|^2g_{Z,L}^{q_1}g_{Z,\chi}^{c}\left[\mathcal{K}_{WZ}^{\chi}\mathcal{Q}[L,\chi]+\widetilde{\mathcal{K}}_{WZ}^{\chi}\widetilde{\mathcal{Q}}[L,\chi]\right] \tag*{(19)}

Gauge covariance requires the derivative of a color-open heavy-quark bilinear to be

Dμ(cˉjΓχici)=(cˉD←μ)jΓχici+cˉjΓχi(Dμc)i.(20)D_\mu(\bar{c}_j\Gamma_\chi^i c^i)=(\bar{c}\overleftarrow{D}_\mu)_j\Gamma_\chi^i c^i+\bar{c}_j\Gamma_\chi^i(D_\mu c)^i. \tag*{(20)}

We write both color contractions using covariant derivatives. For the color-singlet bilinears in the untilded contraction, these derivatives reduce to ordinary derivatives. The two color contractions are

Q[χ,χ′]=Dα(cˉiΓχρci)Dα(cˉjΓχρcj)−Dρ(cˉiΓχρci)Dσ(cˉjΓχσcj),(21)\mathcal{Q}[\chi,\chi']=D_\alpha(\bar{c}_i\Gamma_{\chi\rho}c^i)D^\alpha(\bar{c}_j\Gamma_\chi^\rho c^j)-D_\rho(\bar{c}_i\Gamma_\chi^\rho c^i)D_\sigma(\bar{c}_j\Gamma_\chi^\sigma c^j), \tag*{(21)}
Q~[χ,χ′]=Dα(cˉjΓχρci)Dα(cˉiΓχρcj)−Dρ(cˉjΓχρci)Dσ(cˉiΓχσcj).(22)\widetilde{\mathcal{Q}}[\chi,\chi']=D_\alpha(\bar{c}_j\Gamma_{\chi\rho}c^i)D^\alpha(\bar{c}_i\Gamma_\chi^\rho c^j)-D_\rho(\bar{c}_j\Gamma_\chi^\rho c^i)D_\sigma(\bar{c}_i\Gamma_\chi^\sigma c^j). \tag*{(22)}

Both Q[χ,χ′]\mathcal{Q}[\chi,\chi'] and Q~[χ,χ′]\widetilde{\mathcal{Q}}[\chi,\chi'] are symmetric under χ↔χ′\chi\leftrightarrow\chi'. The three unordered chirality pairs LLLL, LRLR, RRRR, together with the untilded and tilded color contractions, therefore give six independent full-QCD operators.

The combinations KZZ\mathcal{K}_{ZZ}, K~ZZ\widetilde{\mathcal{K}}_{ZZ}, KWZ\mathcal{K}_{WZ}, and K~WZ\widetilde{\mathcal{K}}_{WZ} entering the neutral-current and interference terms in eqs. (17)–(19) are defined by

KZZχχ′;χ′′=NcCZχχ′′CZχ′χ′′+CZχχ′′C~Zχ′χ′′+C~Zχχ′′CZχ′χ′′,K~ZZχχ′;χ′′=C~Zχχ′′C~Zχ′χ′′.(23)\mathcal{K}_{ZZ}^{\chi\chi';\chi''}=N_cC_Z^{\chi\chi''}C_Z^{\chi'\chi''}+C_Z^{\chi\chi''}\widetilde{C}_Z^{\chi'\chi''}+\widetilde{C}_Z^{\chi\chi''}C_Z^{\chi'\chi''},\qquad\widetilde{\mathcal{K}}_{ZZ}^{\chi\chi';\chi''}=\widetilde{C}_Z^{\chi\chi''}\widetilde{C}_Z^{\chi'\chi''}. \tag*{(23)}
KWZχ=(C1+NcC2)CZχL+C2C~ZχL,K~WZχ=C1C~ZχL.(24)\mathcal{K}_{WZ}^{\chi}=(C_1+N_cC_2)C_Z^{\chi L}+C_2\widetilde{C}_Z^{\chi L},\qquad\widetilde{\mathcal{K}}_{WZ}^{\chi}=C_1\widetilde{C}_Z^{\chi L}. \tag*{(24)}

All Wilson coefficients entering eqs. (23) and (24) use the same scale μ\mu.

Heavy-pair expansion through canonical dimension seven

The heavy-quark effective-field expansion identifies the leading annihilation structure of the four-charm operators and motivates the vacuum-saturation approximation used in section 4. For the velocity-projector construction we use the standard heavy-field expansion [45]. The heavy-pair power counting and color-singlet treatment follow [46, 47]. Let vμv^\mu be the heavy-pair four-velocity, normalized by v2=1v^2=1. The charm field is decomposed as

c(x)=e−imcv⋅xh+(x)+e+imcv⋅xh−(x)+O(1mc).(25)c(x)=e^{-im_cv\cdot x}h_+(x)+e^{+im_cv\cdot x}h_-(x)+\mathcal{O}\left(\frac{1}{m_c}\right). \tag*{(25)}

The corresponding projectors satisfy

P±h±=h±,P±=1±v̸2.(26)P_\pm h_\pm=h_\pm,\qquad P_\pm=\frac{1\pm\not{v}}{2}. \tag*{(26)}

The two annihilation currents and their transverse projections are defined by

(aχμ[0])ji=hˉ−,jΓχμh+i,(aχμ[1])ji=hˉ−,j(−iD̸←⊥)Γχμh+i+hˉ−,jΓχμi̸D⊥h+i,D⊥μ=Dμ−vμv⋅D,D←⊥μ=D←μ−vμv⋅D←,aχ⊥[r]μ=(gμν−vμvν)aχν[r],r=0,1.(27)\begin{aligned} \left(a_{\chi\mu}^{[0]}\right)^i_j&=\bar{h}_{-,j}\Gamma_{\chi\mu}h_+^i,\\ \left(a_{\chi\mu}^{[1]}\right)^i_j&=\bar{h}_{-,j}(-i\overleftarrow{\not{D}}_\perp)\Gamma_{\chi\mu}h_+^i+\bar{h}_{-,j}\Gamma_{\chi\mu}i\not{D}_\perp h_+^i,\\ D_\perp^\mu&=D^\mu-v^\mu v\cdot D,\qquad\overleftarrow{D}_\perp^\mu=\overleftarrow{D}^\mu-v^\mu v\cdot\overleftarrow{D},\\ a_{\chi\perp}^{[r]\mu}&=(g^{\mu\nu}-v^\mu v^\nu)a_{\chi\nu}^{[r]},\qquad r=0,1. \tag*{(27)} \end{aligned}

Here aχ[0]μa_{\chi}^{[0]\mu} is the leading local pair-annihilation current, while aχ[1]μa_{\chi}^{[1]\mu} contains one residual derivative.

The full current also contains number-conserving heavy-field bilinears. At leading order its frequency decomposition is

cˉjΓχμci=(nχ[0]μ)ji+e−2imcv⋅x(aχ[0]μ)ji+e+2imcv⋅x[(aχ[0]μ)†]ji+O(1mc),(nχ[0]μ)ji=hˉ+,jΓχμh+i+hˉ−,jΓχμh−i.(28)\bar{c}_{j}\Gamma^{\mu}_{\chi}c^{i}=(n_{\chi}^{[0]\mu})^{i}_{j}+e^{-2im_{c}v\cdot x}(a_{\chi}^{[0]\mu})^{i}_{j}+e^{+2im_{c}v\cdot x}[(a_{\chi}^{[0]\mu})^{\dagger}]^{i}_{j}+\mathcal{O}\left(\frac{1}{m_{c}}\right), \qquad (n_{\chi}^{[0]\mu})^{i}_{j}=\bar{h}_{+,j}\Gamma^{\mu}_{\chi}h_{+}^{i}+\bar{h}_{-,j}\Gamma^{\mu}_{\chi}h_{-}^{i}. \tag*{(28)}

The number-conserving current has no fast mass phase, so its total covariant derivative carries only residual momentum. The pair currents instead acquire a factor ∓2imcvα\mp2im_{c}v_{\alpha} when differentiated. Products with a net pair-creation or pair-annihilation number do not contribute to the forward matrix element in the sector with one charm and one anticharm. The leading connected pair contribution is therefore the product of a pair-creation current and a pair-annihilation current.

After eliminating the small heavy-field components at tree level, the pair component of the current is expanded through first order in the residual momentum as

cˉjΓχμci∣cˉc=e−2imcv⋅x[(aχ[0]μ)ji+12mc(aχ[1]μ)ji]+h.c.(29)\left.\bar{c}_{j}\Gamma^{\mu}_{\chi}c^{i}\right|_{\bar{c}c}=e^{-2im_{c}v\cdot x}\left[(a_{\chi}^{[0]\mu})^{i}_{j}+\frac{1}{2m_{c}}(a_{\chi}^{[1]\mu})^{i}_{j}\right]+\mathrm{h.c.} \tag*{(29)}

Applying the covariant derivative directly to the pair component in eq. (29) and retaining terms through one residual derivative, the derivative can be written as

Dμ(cˉjΓχμci)∣cˉc=e−2imcv⋅x[−2imcvμ(aχ[0]μ)ji+(Dμaχ[0]μ)ji−ivμ(aχ[1]μ)ji]+h.c.(30)\left.D_{\mu}(\bar{c}_{j}\Gamma^{\mu}_{\chi}c^{i})\right|_{\bar{c}c}=e^{-2im_{c}v\cdot x}\left[-2im_{c}v_{\mu}(a_{\chi}^{[0]\mu})^{i}_{j}+(D_{\mu}a_{\chi}^{[0]\mu})^{i}_{j}-iv_{\mu}(a_{\chi}^{[1]\mu})^{i}_{j}\right]+\mathrm{h.c.} \tag*{(30)}

For the heavy-pair contribution, separating the fast-phase and residual-derivative terms gives

Q[χ,χ′]=(2mc)2O(6)[χ,χ′]+2mcO(7)[χ,χ′]+⋯ ,Q[\chi,\chi']=(2m_{c})^{2}\mathcal{O}^{(6)}[\chi,\chi']+2m_{c}\mathcal{O}^{(7)}[\chi,\chi']+\cdots,
Q~[χ,χ′]=(2mc)2O~(6)[χ,χ′]+2mcO~(7)[χ,χ′]+⋯ .(31)\widetilde{Q}[\chi,\chi']=(2m_{c})^{2}\widetilde{\mathcal{O}}^{(6)}[\chi,\chi']+2m_{c}\widetilde{\mathcal{O}}^{(7)}[\chi,\chi']+\cdots. \tag*{(31)}

Here the ellipses denote terms with two or more residual derivatives and the associated gluonic structures; the displayed reduction is truncated through canonical dimension seven.

With Tr⁡c\operatorname{Tr}_{c} a trace over color indices, the dimension-six operators and the dimension-seven structures are

O(6)[χ,χ′]=Tr⁡c[(aχ⊥[0]ρ)†]Tr⁡c(aχ′⊥[0]ρ)+Tr⁡c[(aχ′⊥[0]ρ)†]Tr⁡c(aχ⊥[0]ρ),(32)\mathcal{O}^{(6)}[\chi,\chi']=\operatorname{Tr}_{c}[(a_{\chi\perp}^{[0]\rho})^{\dagger}]\operatorname{Tr}_{c}(a_{\chi'\perp}^{[0]\rho})+\operatorname{Tr}_{c}[(a_{\chi'\perp}^{[0]\rho})^{\dagger}]\operatorname{Tr}_{c}(a_{\chi\perp}^{[0]\rho}), \tag*{(32)}
O~(6)[χ,χ′]=Tr⁡c[(aχ⊥[0]ρ)†aχ′⊥[0]ρ]+Tr⁡c[(aχ′⊥[0]ρ)†aχ⊥[0]ρ],(33)\widetilde{\mathcal{O}}^{(6)}[\chi,\chi']=\operatorname{Tr}_{c}[(a_{\chi\perp}^{[0]\rho})^{\dagger}a_{\chi'\perp}^{[0]\rho}]+\operatorname{Tr}_{c}[(a_{\chi'\perp}^{[0]\rho})^{\dagger}a_{\chi\perp}^{[0]\rho}], \tag*{(33)}
O(7)[χ,χ′]=Tr⁡c[(aχ⊥[1]ρ)†]Tr⁡c(aχ′⊥[0]ρ)+Tr⁡c[(aχ⊥[0]ρ)†]Tr⁡c(aχ′⊥[1]ρ)+iTr⁡c[(aχ⊥[0]ρ)†]Tr⁡c(v⋅Daχ′⊥[0]ρ)−iTr⁡c[v⋅D(aχ⊥[0]ρ)†]Tr⁡c(aχ′⊥[0]ρ)+iTr⁡c[D⊥ρ(aχ⊥[0]ρ)†]Tr⁡c(v⋅aχ′)−iTr⁡c[v⋅(aχ′[0])†]Tr⁡c(D⊥ρaχ⊥[0]ρ)+h.c.,(34)\begin{aligned} \mathcal{O}^{(7)}[\chi,\chi']={}&\operatorname{Tr}_{c}[(a_{\chi\perp}^{[1]\rho})^{\dagger}]\operatorname{Tr}_{c}(a_{\chi'\perp}^{[0]\rho})+\operatorname{Tr}_{c}[(a_{\chi\perp}^{[0]\rho})^{\dagger}]\operatorname{Tr}_{c}(a_{\chi'\perp}^{[1]\rho})\\ &+i\operatorname{Tr}_{c}[(a_{\chi\perp}^{[0]\rho})^{\dagger}]\operatorname{Tr}_{c}(v\cdot D a_{\chi'\perp}^{[0]\rho})-i\operatorname{Tr}_{c}[v\cdot D(a_{\chi\perp}^{[0]\rho})^{\dagger}]\operatorname{Tr}_{c}(a_{\chi'\perp}^{[0]\rho})\\ &+i\operatorname{Tr}_{c}[D_{\perp}^{\rho}(a_{\chi\perp}^{[0]\rho})^{\dagger}]\operatorname{Tr}_{c}(v\cdot a_{\chi'})-i\operatorname{Tr}_{c}[v\cdot(a_{\chi'}^{[0]})^{\dagger}]\operatorname{Tr}_{c}(D_{\perp\rho}a_{\chi\perp}^{[0]\rho})+\mathrm{h.c.}, \tag*{(34)} \end{aligned}
O~(7)[χ,χ′]=Tr⁡c[(aχ⊥[1]ρ)†aχ′⊥[0]ρ+(aχ⊥[0]ρ)†aχ′⊥[1]ρ]+iTr⁡c[(aχ⊥[0]ρ)†(v⋅Daχ′⊥[0]ρ)]−iTr⁡c[(v⋅D(aχ⊥[0]ρ)†)aχ′⊥[0]ρ]+iTr⁡c[(D⊥ρ(aχ⊥[0]ρ)†)(v⋅aχ′)]−iTr⁡c[(v⋅(aχ′[0])†)(D⊥ρaχ⊥[0]ρ)]+h.c..(35)\begin{aligned} \widetilde{\mathcal{O}}^{(7)}[\chi,\chi']={}&\operatorname{Tr}_{c}[(a_{\chi\perp}^{[1]\rho})^{\dagger}a_{\chi'\perp}^{[0]\rho}+(a_{\chi\perp}^{[0]\rho})^{\dagger}a_{\chi'\perp}^{[1]\rho}]\\ &+i\operatorname{Tr}_{c}[(a_{\chi\perp}^{[0]\rho})^{\dagger}(v\cdot D a_{\chi'\perp}^{[0]\rho})]-i\operatorname{Tr}_{c}[(v\cdot D(a_{\chi\perp}^{[0]\rho})^{\dagger})a_{\chi'\perp}^{[0]\rho}]\\ &+i\operatorname{Tr}_{c}[(D_{\perp}^{\rho}(a_{\chi\perp}^{[0]\rho})^{\dagger})(v\cdot a_{\chi'})]-i\operatorname{Tr}_{c}[(v\cdot(a_{\chi'}^{[0]})^{\dagger})(D_{\perp\rho}a_{\chi\perp}^{[0]\rho})]+\mathrm{h.c.}. \tag*{(35)} \end{aligned}

In eqs. (34) and (35), h.c. applies to the entire preceding right-hand side. It supplies the reversed chirality ordering and makes each operator Hermitian and symmetric under interchange of χ\chi and χ′\chi'.

With γ⊥μ=(gμν−vμvν)γν\gamma_\perp^\mu= (g^{\mu\nu} - v^\mu v^\nu)\gamma_\nu, the heavy-field projectors give

P−γ⊥μγ5P+=0,aL[0]μ=aR[0]μ.(36)P_-\gamma_\perp^\mu\gamma_5P_+ = 0,\qquad a_L^{[0]\mu} = a_R^{[0]\mu}. \tag*{(36)}

The leading transverse pair current therefore has the vector spin structure appropriate to the 3S1{}^{3}S_1 J/ψJ/\psi. Equations (30) and (31) identify the leading contribution to the connected heavy-pair matrix elements of QQ and Q~\widetilde{Q} as pair annihilation followed by pair creation. It carries two powers of the hard mass from the fast phases, whereas terms involving only number-conserving currents carry residual derivatives. The operators O(7)\mathcal{O}^{(7)} describe the first residual-momentum correction to this leading annihilation structure.

The effective-field expansion also identifies the intermediate state to which vacuum saturation applies. Let Aχρ=Tr⁡c(aχ[0]ρ)A_\chi^\rho= \operatorname{Tr}_c(a_\chi^{[0]\rho}). The untilded dimension-six operator is O(6)[χ,χ′]=Aχρ†Aχ′ρ+Aχ′ρ†Aχρ\mathcal{O}^{(6)}[\chi,\chi'] = A_{\chi\rho}^\dagger A_{\chi'}^\rho+ A_{\chi'\rho}^\dagger A_\chi^\rho. Inserting a complete set of light states between the pair currents gives

⟨J/ψ∣Aχρ†Aχ′ρ∣J/ψ⟩=∑X⟨J/ψ∣Aχρ†∣X⟩⟨X∣Aχ′ρ∣J/ψ⟩≃⟨J/ψ∣Aχρ†∣0⟩⟨0∣Aχ′ρ∣J/ψ⟩.(37)\begin{aligned} \langle J/\psi|A_{\chi\rho}^\dagger A_{\chi'}^\rho|J/\psi\rangle = \sum_X \langle J/\psi|A_{\chi\rho}^\dagger|X\rangle\langle X|A_{\chi'}^\rho|J/\psi\rangle\\ &\simeq\langle J/\psi|A_{\chi\rho}^\dagger|0\rangle\langle0|A_{\chi'}^\rho|J/\psi\rangle. \tag*{(37)} \end{aligned}

For the leading color-singlet ccˉc\bar c component, AχρA_\chi^\rho annihilates the heavy pair and leaves the vacuum. Nonvacuum states in this sum arise from higher Fock components containing additional gluons or light partons. Retaining the dominant color-singlet component therefore motivates the vacuum-saturation approximation [47]. The exchanged-color operator also contains an independent color-octet contraction; its vacuum-to-meson current matrix element vanishes by color conservation, and its physical contribution is omitted in this leading color-singlet approximation.

At leading order in the hard decay kernels, no gluon is exchanged between the heavy annihilation current and the light final-state current, so the binding effects are contained in the physical decay constant. Independent color-octet inputs, nonvacuum intermediate states and hard radiative corrections lie beyond this approximation. Hence we will evaluate QQ and Q~\widetilde{Q} directly by vacuum saturation, retaining the physical pair momentum PμP^\mu and fixing the normalization by the full-QCD decay constant.

Hadronic matrix elements

We define the reduced forward matrix elements by

MQ[χ,χ′]≡12MJ/ψ⟨J/ψ(P,λ)∣Q[χ,χ′]∣J/ψ(P,λ)⟩,\mathcal{M}_Q[\chi,\chi'] \equiv\frac{1}{2M_{J/\psi}}\langle J/\psi(P,\lambda)|Q[\chi,\chi']|J/\psi(P,\lambda)\rangle,
M~Q[χ,χ′]≡12MJ/ψ⟨J/ψ(P,λ)∣Q~[χ,χ′]∣J/ψ(P,λ)⟩.(38)\widetilde{\mathcal{M}}_Q[\chi,\chi'] \equiv\frac{1}{2M_{J/\psi}}\langle J/\psi(P,\lambda)|\widetilde{Q}[\chi,\chi']|J/\psi(P,\lambda)\rangle. \tag*{(38)}

For any other local operator O\mathcal{O}, we use the same normalization:

MJ/ψ[O]=⟨J/ψ∣O∣J/ψ⟩2MJ/ψ.(39)\mathcal{M}_{J/\psi}[\mathcal{O}] = \frac{\langle J/\psi|\mathcal{O}|J/\psi\rangle}{2M_{J/\psi}}. \tag*{(39)}

Let Jχμ=cˉiΓχμciJ^\mu_\chi=\bar{c}_i\Gamma^\mu_\chi c^i and Jχ jμi=cˉjΓχμciJ^{\mu i}_{{\chi}\,j}=\bar{c}_j\Gamma^\mu_\chi c^i. Parity forbids the axial-vector current from creating the JPC=1−−J^{PC}=1^{--} state from the vacuum. Consequently, the left- and right-handed currents have the same vacuum-to-meson matrix element, fixed by the physical vector decay constant:

⟨0∣JLμ(0)∣J/ψ(P,λ)⟩=⟨0∣JRμ(0)∣J/ψ(P,λ)⟩=MJ/ψfψϵλμ,P⋅ϵλ=0,ϵλ∗⋅ϵλ=−1.(40)\langle0|J^\mu_L(0)|J/\psi(P,\lambda)\rangle=\langle0|J^\mu_R(0)|J/\psi(P,\lambda)\rangle=M_{J/\psi}f_\psi\epsilon^\mu_\lambda, \qquad P\cdot\epsilon_\lambda=0,\qquad\epsilon^*_\lambda\cdot\epsilon_\lambda=-1. \tag*{(40)}

In the color-singlet approximation, the open-color matrix element is δji/Nc\delta^i_j/N_c times that in eq. (40).

Translational covariance supplies the derivatives in the full-QCD operators without a heavy-field expansion:

⟨0∣Jχμ(x)∣J/ψ(P,λ)⟩=MJ/ψfψϵλμe−iP⋅x,\langle0|J^\mu_\chi(x)|J/\psi(P,\lambda)\rangle=M_{J/\psi}f_\psi\epsilon^\mu_\lambda e^{-iP\cdot x},
⟨0∣∂αJχμ(0)∣J/ψ(P,λ)⟩=−iPαMJ/ψfψϵλμ,\langle0|\partial_\alpha J^\mu_\chi(0)|J/\psi(P,\lambda)\rangle=-iP_\alpha M_{J/\psi}f_\psi\epsilon^\mu_\lambda,
⟨J/ψ(P,λ)∣∂αJχμ(0)∣0⟩=+iPαMJ/ψfψϵλ∗μ.(41)\langle J/\psi(P,\lambda)|\partial_\alpha J^\mu_\chi(0)|0\rangle=+iP_\alpha M_{J/\psi}f_\psi\epsilon^{*\mu}_\lambda. \tag*{(41)}

For the open-color current, the covariant derivative has the same matrix element with the factor δji/Nc\delta^i_j/N_c at the order retained. Equations (40) and (41) give

⟨0∣∂μJχμ∣J/ψ⟩=0,⟨0∣DμJχ jμi∣J/ψ⟩=0.(42)\langle0|\partial_\mu J^\mu_\chi|J/\psi\rangle=0,\qquad\langle0|D_\mu J^{\mu i}_{\chi\,j}|J/\psi\rangle=0. \tag*{(42)}

Thus the divergence products in both Q\mathcal{Q} and Q~\widetilde{\mathcal{Q}} vanish after vacuum saturation.

For the connected annihilation part of the normal-ordered four-charm products, vacuum saturation retains both assignments of the pair-creation and pair-annihilation currents. Suppressing the external-state labels,

⟨J/ψ∣Q[χ,χ′]∣J/ψ⟩VS≃⟨J/ψ∣∂αJχρ∣0⟩⟨0∣∂αJχ′∣J/ψ⟩+(χ↔χ′)=−2MJ/ψ4∣fψ∣2.(43)\langle J/\psi|\mathcal{Q}[\chi,\chi']|J/\psi\rangle_{\mathrm{VS}}\simeq\langle J/\psi|\partial_\alpha J^\rho_\chi|0\rangle\langle0|\partial^\alpha J^{}_{\chi'}|J/\psi\rangle+(\chi\leftrightarrow\chi') =-2M^4_{J/\psi}|f_\psi|^2. \tag*{(43)}

Each assignment contributes −MJ/ψ4∣fψ∣2-M^4_{J/\psi}|f_\psi|^2, using P2=MJ/ψ2P^2=M^2_{J/\psi} and ϵλ∗⋅ϵλ=−1\epsilon^*_\lambda\cdot\epsilon_\lambda=-1. For the exchanged color contraction, ∑i,jδjiδij/Nc2=1/Nc\sum_{i,j}\delta^i_j\delta^j_i/N_c^2=1/N_c, giving, for every chirality pair,

MQ[χ,χ′]≃−MJ/ψ3∣fψ∣2,M~Q[χ,χ′]≃−MJ/ψ3∣fψ∣2Nc.(44)\mathcal{M}_{\mathcal{Q}}[\chi,\chi']\simeq-M^3_{J/\psi}|f_\psi|^2,\qquad\widetilde{\mathcal{M}}_{\mathcal{Q}}[\chi,\chi']\simeq-\frac{M^3_{J/\psi}|f_\psi|^2}{N_c}. \tag*{(44)}

No separate residual-energy matrix element is needed here. Inserting eq. (44) into eqs. (16)–(19) gives

ΓWE=GF2MJ/ψ3∣fψ∣212πNc∑q1,q2∣Vcq1∣2∣Vcq2∣2(C1+NcC2)2,(45)\Gamma_{\mathrm{WE}}=\frac{G_F^2M^3_{J/\psi}|f_\psi|^2}{12\pi N_c}\sum_{q_1,q_2}|V_{cq_1}|^2|V_{cq_2}|^2(C_1+N_cC_2)^2, \tag*{(45)}
ΓNC,q=GF2MJ/ψ3∣fψ∣23π∑q,χ,χ′,χ′′(gZ,χ′′q)2gZ,χcgZ,χ′c[KZZχχ′;χ′′+K~ZZχχ′;χ′′Nc],(46)\Gamma_{\mathrm{NC},q}=\frac{G_F^2M^3_{J/\psi}|f_\psi|^2}{3\pi}\sum_{q,\chi,\chi',\chi''}(g^q_{Z,\chi''})^2g^c_{Z,\chi}g^c_{Z,\chi'}\left[\mathcal{K}^{\chi\chi';\chi''}_{ZZ}+\frac{\widetilde{\mathcal{K}}^{\chi\chi';\chi''}_{ZZ}}{N_c}\right], \tag*{(46)}
ΓNC,ℓ=GF2MJ/ψ3∣fψ∣23π∑ℓ,χ,χ′,χ′′(gZ,χ′′ℓ)2gZ,χcgZ,χ′c,(47)\Gamma_{\mathrm{NC},\ell}=\frac{G_F^2M^3_{J/\psi}|f_\psi|^2}{3\pi}\sum_{\ell,\chi,\chi',\chi''}(g^\ell_{Z,\chi''})^2g^c_{Z,\chi}g^c_{Z,\chi'}, \tag*{(47)}
Γint=GF2MJ/ψ3∣fψ∣23π∑q1,χ∣Vcq1∣2gZ,LqgZ,χc[KWZχ+K~WZχNc].(48)\Gamma_{\mathrm{int}}=\frac{G_F^2M^3_{J/\psi}|f_\psi|^2}{3\pi}\sum_{q_1,\chi}|V_{cq_1}|^2g^q_{Z,L}g^c_{Z,\chi}\left[\mathcal{K}^{\chi}_{WZ}+\frac{\widetilde{\mathcal{K}}^{\chi}_{WZ}}{N_c}\right]. \tag*{(48)}

The subscripts denote the same shorthand as in the transition operators.

Numerical analysis

We combine the short-distance coefficients with the matrix elements of § 4 to evaluate the inclusive weak width at the central renormalization scale μ0=MJ/ψ\mu_0 = M_{J/\psi}. We assign a scale uncertainty from the envelope obtained by varying a common scale over 1 GeV≤μ≤3MJ/ψ1\ \mathrm{GeV} \le\mu\le3M_{J/\psi}, with all masses and hadronic matrix elements held fixed. The upper and lower errors are the maximal upward and downward shifts from the central value over the full interval. For each total width or branching fraction, all contributions are evaluated at the same scale before taking the envelope. We quote two errors: the perturbative scale uncertainty and an assigned hadronic HQE uncertainty. The first estimates missing higher-order perturbative corrections; the hard kernels remain at tree level. The second combines vacuum saturation and the kinetic/chromomagnetic inputs through their joint variation at μ0\mu_0.

Table 1 summarizes the common inputs. The single-charm kernels and matrix-element estimates use mc=1.48 GeVm_c = 1.48\ \mathrm{GeV} [8]. The directly saturated annihilation matrix elements instead use the physical pair mass MJ/ψM_{J/\psi}, as in eq. (44). The value mb=4.186 GeVm_b = 4.186\ \mathrm{GeV} in Table 1 serves only as the flavor threshold in the leading-logarithmic evolution of the Wilson coefficients, which uses one-loop running. The masses, widths, CKM elements, and electroweak inputs are taken from the Review of Particle Physics [12], with the CKM elements from its global fit and sW2=sin⁡2θ^W(MZ)s_W^2 = \sin^2\hat{\theta}_W(M_Z) in the MS scheme. The decay constant follows from the leptonic width, Γ(J/ψ→e+e−)=4πα2Qc2fψ2/(3MJ/ψ)\Gamma(J/\psi\to e^+e^-) = 4\pi\alpha^2 Q_c^2 f_\psi^2/(3M_{J/\psi}), with Γ(J/ψ→e+e−)=5.53 keV\Gamma(J/\psi\to e^+e^-) = 5.53\ \mathrm{keV} [12], Qc=2/3Q_c = 2/3, and the effective coupling α(MJ/ψ2)=1/134.02\alpha(M_{J/\psi}^2) = 1/134.02 [48], which gives fψ=0.4065 GeVf_\psi= 0.4065\ \mathrm{GeV}.

ParameterValueParameterValue
GFG_F1.1663785×10−5 GeV−21.1663785 \times10^{-5}\ \mathrm{GeV}^{-2}NcN_c33
mcm_c1.48 GeV1.48\ \mathrm{GeV}MJ/ψM_{J/\psi}3.096900 GeV3.096900\ \mathrm{GeV}
MηcM_{\eta_c}2.98409 GeV2.98409\ \mathrm{GeV}ΓJ/ψtot\Gamma_{J/\psi}^{\mathrm{tot}}92.6 keV92.6\ \mathrm{keV}
∣Vcd∣\lvert V_{cd}\rvert0.225030.22503∣Vcs∣\lvert V_{cs}\rvert0.973450.97345
∣Vud∣\lvert V_{ud}\rvert0.974310.97431∣Vus∣\lvert V_{us}\rvert0.225170.22517
sW2s_W^20.231220.23122αs(MZ)\alpha_s(M_Z)0.11800.1180
MZM_Z91.1879 GeV91.1879\ \mathrm{GeV}MWM_W80.3625 GeV80.3625\ \mathrm{GeV}
mbm_b4.186 GeV4.186\ \mathrm{GeV}fψf_\psi0.4065 GeV0.4065\ \mathrm{GeV}

Table 1. Numerical inputs held fixed throughout the scale variation.

Across the bottom threshold, the leading-logarithmic evolution is

Uγ0(μH,μ)={[αs(5)(μH)αs(5)(μ)]γ0/(2β0(5)),μ≥mb,[αs(5)(μH)αs(5)(mb)]γ0/(2β0(5))[αs(4)(mb)αs(4)(μ)]γ0/(2β0(4)),μ<mb.(49)U_{\gamma_0}(\mu_H,\mu)= \begin{cases} \left[\dfrac{\alpha_s^{(5)}(\mu_H)}{\alpha_s^{(5)}(\mu)}\right]^{\gamma_0/(2\beta_0^{(5)})}, & \mu\ge m_b,\\[6pt] \left[\dfrac{\alpha_s^{(5)}(\mu_H)}{\alpha_s^{(5)}(m_b)}\right]^{\gamma_0/(2\beta_0^{(5)})} \left[\dfrac{\alpha_s^{(4)}(m_b)}{\alpha_s^{(4)}(\mu)}\right]^{\gamma_0/(2\beta_0^{(4)})}, & \mu< m_b. \end{cases} \tag*{(49)}
β0(nf)=11−2nf3.\beta_0^{(n_f)}=11-\frac{2n_f}{3}.

The coupling is continuous at mbm_b at this order. The charged-current coefficients are C1±C2=Uγ0(MW,μ)C_1 \pm C_2 = U_{\gamma_0}(M_W,\mu) with γ0=4,−8\gamma_0=4,-8, and the neutral-current coefficients are CiZ=Uγ0(MZ,μ)C_i^Z=U_{\gamma_0}(M_Z,\mu) with γ0=4,−8,−16,2\gamma_0=4,-8,-16,2 for i=1,…,4i=1,\ldots,4 [41]. Table 2 lists the running coupling, the charged-current and neutral-current Wilson coefficients, and the derived single-charm coefficients c3,c5c_3,c_5, with their central values at μ0=MJ/ψ\mu_0=M_{J/\psi} and their values at the two endpoints, μ=1 GeV\mu=1\ \mathrm{GeV} and μ=3MJ/ψ\mu=3M_{J/\psi}.

Quantityμ=1 GeV\mu=1\ \mathrm{GeV}μ=MJ/ψ\mu=M_{J/\psi}μ=3MJ/ψ\mu=3M_{J/\psi}
αs(μ)\alpha_s(\mu)0.35510.23170.1758
C1C_11.2421.1241.063
C2C_2−0.4802-0.4802−0.2796-0.2796−0.1570-0.1570
C1ZC_1^Z0.75830.84010.9012
C2ZC_2^Z1.7391.4171.231
C3ZC_3^Z3.0242.0071.516
C4ZC_4^Z0.87080.91660.9493
c3c_36.1175.3855.118
c5c_5−2.707-2.707−5.752-5.752−7.572-7.572

Table 2. Running coupling and short-distance coefficients at three renormalization scales. The middle column corresponds to the central scale μ0=MJ/ψ\mu_0=M_{J/\psi}.

For the single-charm sector, we estimate the kinetic matrix element from the spin-averaged 1S1S mass, in which the leading spin–spin interaction cancels, and the chromomagnetic matrix element from the hyperfine splitting [46]. We adopt estimated relative hadronic uncertainties of order εhad=ΛQCD/mc≃0.20\varepsilon_{\mathrm{had}}=\Lambda_{\mathrm{QCD}}/m_c\simeq0.20 for vacuum saturation and for each of these two inputs:

μπ2=(0.161±0.032) GeV2,\mu_\pi^2=(0.161\pm0.032)\ \mathrm{GeV}^2,
μG2=(−0.0835±0.0167) GeV2.(50)\mu_G^2=(-0.0835\pm0.0167)\ \mathrm{GeV}^2. \tag*{(50)}

The central kinetic estimate follows from 14(3MJ/ψ+Mηc)=2mc+μπ2/mc+⟨V⟩\frac{1}{4}(3M_{J/\psi}+M_{\eta_c})=2m_c+\mu_\pi^2/m_c+\langle V\rangle with the spin-independent potential energy ⟨V⟩\langle V\rangle neglected, and corresponds to ⟨v2⟩=μπ2/mc2≃0.073\langle v^2\rangle=\mu_\pi^2/m_c^2\simeq0.073. The central hyperfine estimate, μG2=−mc(MJ/ψ−Mηc)/2\mu_G^2=-m_c(M_{J/\psi}-M_{\eta_c})/2, assumes a tree-level spin–spin interaction and a common spatial wave function for J/ψJ/\psi and ηc\eta_c. The scalar and chromomagnetic matrix elements are [8, 46]

MJ/ψ[O3]=2(1−μπ22mc2+μG22mc2),MJ/ψ[O5]=2μG2.(51)\mathcal{M}_{J/\psi}[\mathcal{O}_3] = 2\left(1-\frac{\mu_\pi^2}{2m_c^2}+\frac{\mu_G^2}{2m_c^2}\right), \qquad\mathcal{M}_{J/\psi}[\mathcal{O}_5] = 2\mu_G^2. \tag*{(51)}

They enter the sum of the charm and anticharm decay widths as

Γc+cˉ=GF2mc5192π3[c3MJ/ψ[O3]+c5mc2MJ/ψ[O5]].(52)\Gamma_{c+\bar{c}} = \frac{G_F^2m_c^5}{192\pi^3}\left[c_3\mathcal{M}_{J/\psi}[\mathcal{O}_3]+\frac{c_5}{m_c^2}\mathcal{M}_{J/\psi}[\mathcal{O}_5]\right]. \tag*{(52)}

For the hadronic HQE uncertainty, we multiply all saturated MQ\mathcal{M}_Q and M~Q\widetilde{\mathcal{M}}_Q by one common factor 1+δVS1+\delta_{\mathrm{VS}}, with ∣δVS∣≤ϵhad|\delta_{\mathrm{VS}}|\leq\epsilon_{\mathrm{had}}. Together with δVS\delta_{\mathrm{VS}}, we vary μπ2\mu_\pi^2 and μG2\mu_G^2 independently by ϵhad\epsilon_{\mathrm{had}} of their magnitudes, keeping the same μG2\mu_G^2 in both matrix elements. At μ0\mu_0, the envelope over this three-input box gives the single HQE error quoted below. These assigned theory ranges are not statistical errors and are not combined in quadrature. For a common normalization, Table 3 lists dimensionless combinations obtained by dividing each reduced matrix element by the appropriate power of mcm_c. The corresponding unnormalized matrix elements enter the width formulas. The heavy-pair entries apply to every chirality pair χ,χ′\chi,\chi' at the order considered.

Dimensionless combinationCentral value with hadronic uncertainty
MJ/ψ[O3]\mathcal{M}_{J/\psi}[\mathcal{O}_3](1.89±0.02)(1.89\pm0.02)
MJ/ψ[O5]/mc2\mathcal{M}_{J/\psi}[\mathcal{O}_5]/m_c^2(−0.0762±0.0152)(-0.0762\pm0.0152)
MQ[χ,χ′]/mc5\mathcal{M}_Q[\chi,\chi']/m_c^5(−0.691±0.138)(-0.691\pm0.138)
M~Q[χ,χ′]/mc5\widetilde{\mathcal{M}}_Q[\chi,\chi']/m_c^5(−0.230±0.046)(-0.230\pm0.046)

Table 3. Dimensionless reduced matrix elements with assigned hadronic HQE uncertainties. The four-charm entries share a common vacuum-saturation normalization factor, and the single-charm entries share the same kinetic and chromomagnetic inputs.

At each scale, the weak width is organized into the single-charm sector and two heavy-pair annihilation sectors as

Γweak=Γ∣ΔC∣=1+ΓΔC=0, ∣ΔS∣=1+ΓΔC=0, ΔS=0.(53)\Gamma_{\mathrm{weak}}=\Gamma_{|\Delta C|=1}+\Gamma_{\Delta C=0,\,|\Delta S|=1}+\Gamma_{\Delta C=0,\,\Delta S=0}. \tag*{(53)}

The first term equals Γc+cˉ\Gamma_{c+\bar{c}} of eq. (52) and includes both cc and cˉ\bar{c} decays. In the WW-exchange contribution, diagonal flavor terms belong to ΔS=0\Delta S=0, while off-diagonal terms have ∣ΔS∣=1|\Delta S|=1 at the order considered. The neutral-current and WW–ZZ interference terms contribute only to ΔS=0\Delta S=0. Table 4 lists ΓWE\Gamma_{\mathrm{WE}}, ΓNC\Gamma_{\mathrm{NC}}, and Γint\Gamma_{\mathrm{int}} from Eqs. (45)–(48) by flavor sector.

Contribution∣ΔC∣=1|\Delta C|=1ΔC=0\Delta C=0, ∣ΔS∣=1|\Delta S|=1ΔC=0\Delta C=0, ΔS=0\Delta S=0
Γc+cˉ\Gamma_{c+\bar{c}}1.72−0.06+0.19±0.021.72^{+0.19}_{-0.06}\pm0.02––
ΓWE\Gamma_{\mathrm{WE}}–0.0460−0.0460+0.1522±0.00920.0460^{+0.1522}_{-0.0460}\pm0.00920.432−0.432+1.429±0.0860.432^{+1.429}_{-0.432}\pm0.086
ΓNC\Gamma_{\mathrm{NC}}––6.85−0.15+0.34±1.376.85^{+0.34}_{-0.15}\pm1.37
Γint\Gamma_{\mathrm{int}}––−1.68−1.75+2.89±0.34-1.68^{+2.89}_{-1.75}\pm0.34
Total1.72−0.06+0.19±0.021.72^{+0.19}_{-0.06}\pm0.020.0460−0.0460+0.1522±0.00920.0460^{+0.1522}_{-0.0460}\pm0.00925.60−0.47+3.01±1.125.60^{+3.01}_{-0.47}\pm1.12

Table 4. Signed contributions in units of 10−1210^{-12} GeV, classified by flavor change. Central values use μ0=MJ/ψ\mu_0=M_{J/\psi}; the first errors are scale envelopes over 1 GeV≤μ≤3MJ/ψ1\ \mathrm{GeV}\leq\mu\leq3M_{J/\psi}, followed by the combined hadronic HQE errors. Contributions share the underlying input variations, and each total is varied as a whole.

The flavor decomposition in Table 4 shows that the ΔC=0,ΔS=0\Delta C=0,\Delta S=0 sector dominates throughout the scale interval, with the neutral-current term providing the largest contribution within that sector. In the order of the three sectors in eq. (53), the fractions of the weak width at the central scale are (23.4%,0.62%,76.0%)(23.4\%, 0.62\%, 76.0\%), so pair annihilation provides 76.6%76.6\%. The single-charm width increases moderately as the scale is lowered, through the changes in c3,c5c_3,c_5. For WW exchange, color-singlet factorization makes the width proportional to (C1+NcC2)2(C_1+N_cC_2)^2; the smaller ∣C1+NcC2∣|C_1+N_cC_2| at 1 GeV1\ \mathrm{GeV} therefore suppresses both flavor sectors relative to the central values. This combination vanishes near μ=1.48\mu=1.48 GeV, so the lower scale envelopes of both WW-exchange contributions reach zero; their upper envelopes occur at μ=3MJ/ψ\mu=3M_{J/\psi}. The light-quark component drives the scale variation of the neutral-current contribution, while its leptonic component remains unchanged. The WW–ZZ interference depends linearly on C1+NcC2C_1+N_cC_2. It is destructive at the central scale μ0\mu_0 and becomes constructive below the zero of this combination. This sign change within the scale interval drives the large upward scale uncertainty.

We denote by ΓΔC=0\Gamma_{\Delta C=0} the total pair-annihilation width, i.e. the sum of the two ΔC=0\Delta C=0 terms in eq. (53), including the WW-exchange, neutral-current and WW–ZZ interference contributions. Branching fractions are defined as BX=ΓX/ΓJ/ψtot\mathcal{B}_X=\Gamma_X/\Gamma_{J/\psi}^{\mathrm{tot}}, using the total width in Table 1. We also quote the pure WW-exchange annihilation contribution BWE=ΓWE/ΓJ/ψtot\mathcal{B}_{\mathrm{WE}}=\Gamma_{\mathrm{WE}}/\Gamma_{J/\psi}^{\mathrm{tot}}, with ΓWE\Gamma_{\mathrm{WE}} summed over both strangeness sectors in Table 4. Adding the ∣ΔC∣=1|\Delta C|=1 single-charm decays gives the branching fraction of the contribution mediated by the WW boson alone, (ΓWE+Γ∣ΔC∣=1)/ΓJ/ψtot=(2.37−0.42+1.64±0.13)×10−8(\Gamma_{\mathrm{WE}}+\Gamma_{|\Delta C|=1})/\Gamma_{J/\psi}^{\mathrm{tot}}=(2.37^{+1.64}_{-0.42}\pm0.13)\times10^{-8}. Table 5 summarizes the results. For comparison, the measured DD-meson lifetimes give the central widths [12]

QuantityCentral value with separate uncertainties
ΓΔC=0\Gamma_{\Delta C=0} [GeV](5.65−0.35+2.99±1.13)×10−12(5.65^{+2.99}_{-0.35}\pm1.13)\times10^{-12}
Γweak\Gamma_{\mathrm{weak}} [GeV](7.37−0.40+3.18±1.15)×10−12(7.37^{+3.18}_{-0.40}\pm1.15)\times10^{-12}
BΔC=0\mathcal{B}_{\Delta C=0}(6.10−0.38+3.23±1.22)×10−8(6.10^{+3.23}_{-0.38}\pm1.22)\times10^{-8}
Bweak\mathcal{B}_{\mathrm{weak}}(7.96−0.43+3.43±1.24)×10−8(7.96^{+3.43}_{-0.43}\pm1.24)\times10^{-8}
BWE\mathcal{B}_{\mathrm{WE}}(5.16−5.16+17.07±1.03)×10−9(5.16^{+17.07}_{-5.16}\pm1.03)\times10^{-9}

Table 5. Pair-annihilation and inclusive weak widths and branching fractions, and the pure WW-exchange branching fraction, at μ0=MJ/ψ\mu_0=M_{J/\psi}. Errors are quoted as (scale, hadronic HQE).

(ΓD0,ΓD+,ΓDs+)=(1.60,0.637,1.31)×10−12 GeV.(54)(\Gamma_{D^0},\Gamma_{D^+},\Gamma_{D_s^+})=(1.60,0.637,1.31)\times10^{-12}\ \mathrm{GeV}. \tag*{(54)}

At the central scale, ΓWE=4.78×10−13\Gamma_{\mathrm{WE}}=4.78\times10^{-13} GeV and Γ∣ΔC∣=1=1.72×10−12\Gamma_{|\Delta C|=1}=1.72\times10^{-12} GeV, giving ΓWE+Γ∣ΔC∣=1=2.20×10−12\Gamma_{\mathrm{WE}}+\Gamma_{|\Delta C|=1}=2.20\times10^{-12} GeV. Using unrounded central inputs, the ratios of this sum to the widths in eq. (54) are 1.37, 3.45, and 1.67, respectively.

The present results omit radiative corrections to the hard kernels, hadronic effects beyond vacuum saturation, single-charm terms of dimension six and higher such as the Darwin contribution [8, 49], heavy-pair operators with independent relative-motion or gluonic structure, penguin and flavor-singlet mixing in the Wilson-coefficient evolution [50], and final-state masses. Uncertainties from the decay constant and the charm-mass prescription are not propagated.

Conclusions

We have calculated the inclusive weak width of the J/ψJ/\psi through the optical theorem, including single-charm decays, charged- and neutral-current pair annihilation, and their interference. The heavy-quark effective-field expansion identifies the leading pair-annihilation structure and motivates vacuum saturation, and the full-QCD matrix elements are fixed by the physical J/ψJ/\psi mass and decay constant. With tree-level hard kernels and leading-logarithmic Wilson coefficients at μ0=MJ/ψ\mu_0 = M_{J/\psi}, we obtain Bweak=(7.96−0.43+3.43±1.24)×10−8\mathcal{B}_{\mathrm{weak}} = (7.96^{+3.43}_{-0.43} \pm1.24) \times10^{-8}, where the errors are the scale envelope over 1 GeV≤μ≤3MJ/ψ1\ \mathrm{GeV} \le\mu\le3M_{J/\psi} and the hadronic HQE uncertainty. The contribution mediated by the WW boson alone is (2.37−0.42+1.64±0.13)×10−8(2.37^{+1.64}_{-0.42} \pm0.13) \times10^{-8}, and the (ΔC,∣ΔS∣)=(0,1)(\Delta C, |\Delta S|) = (0, 1) and ∣ΔC∣=1|\Delta C| = 1 sectors give (4.97−4.97+16.44±0.99)×10−10(4.97^{+16.44}_{-4.97} \pm0.99) \times10^{-10} and (1.86−0.06+0.20±0.02)×10−8(1.86^{+0.20}_{-0.06} \pm0.02) \times10^{-8}, respectively.

Pair annihilation dominates, providing about 77% of the central width, with the neutral-current term as the largest contribution. The WW–ZZ interference changes sign within the scale interval and drives the large upward scale uncertainty. Estimates based on single-charm decay alone therefore miss most of the J/ψJ/\psi weak width. Quantitative refinement requires radiative corrections to the hard kernels, consistent evolution of the hadronic matrix elements, and hadronic inputs beyond vacuum saturation.

Acknowledgments

This work was supported in part by the National Natural Science Foundation of China under Grant No. 12575096.

References

  1. [1]M.A. Sanchis-Lozano, On the search for weak decays of heavy quarkonium in dedicated heavy-quark factories, Z. Phys. C 62 (1994) 271–279.
  2. [2]K.K. Sharma and R.C. Verma, Rare decays of ψ and Υ, Int. J. Mod. Phys. A 14 (1999) 937–945 [hep-ph/9801202].DOI
  3. [3]BES collaboration, Search for the rare decays J/ψ → Dₛ⁻π⁺, J/ψ → D⁻π⁺, and J/ψ → D̄⁰K̄⁰, Phys. Lett. B 663 (2008) 297–301 [0707.3005].
  4. [4]BESIII collaboration, Search for the weak decays J/ψ → Dₛ⁽*⁾νₑν̄ₑ, Phys. Rev. D 90 (2014) 112014 [1410.8426].
  5. [5]Y.-M. Wang, H. Zou, Z.-T. Wei, X.-Q. Li and C.-D. Lü, Weak decays of J/ψ: the non-leptonic case, Eur. Phys. J. C 55 (2008) 607–613 [0802.2762].DOI
  6. [6]A. Lenz, Lifetimes and heavy quark expansion, Int. J. Mod. Phys. A 30 (2015) 1543005 [1405.3601].DOI
  7. [7]A. Lenz and T. Rauh, D-meson lifetimes within the heavy quark expansion, Phys. Rev. D 88 (2013) 034004 [1305.3858].DOI
  8. [8]D. King, A. Lenz, M.L. Piscopo, T. Rauh, A.V. Rusov and C. Vlahos, Revisiting inclusive decay widths of charmed mesons, JHEP 08 (2022) 241 [2109.13219].
  9. [9]D. Friday, E. Gersabeck, A. Lenz and M.L. Piscopo, Charm physics, arXiv:2506.15584.arxiv.org/abs/2506.15584
  10. [10]Th. Mannel, D. Moreno and A.A. Pivovarov, The heavy quark expansion for the charm quark, arXiv:2103.02058.arxiv.org/abs/2103.02058
  11. [11]M. Neubert and C.T. Sachrajda, Spectator effects in inclusive decays of beauty hadrons, Nucl. Phys. B 483 (1997) 339 [hep-ph/9603202].DOI
  12. [12]F. Takahashi et al. (Particle Data Group), Review of Particle Physics, Int. J. Mod. Phys. A 41 (2026) 2630011.
  13. [13]LHCb collaboration, Precision measurement of the ratio of the Λ⁰_b to B̄⁰ lifetimes, Phys. Lett. B 734 (2014) 122–130 [1402.6242].
  14. [14]J. Gratrex, A. Lenz, B. Melić, I. Nišandžić, M.L. Piscopo and A.V. Rusov, Quark–hadron duality at work: lifetimes of bottom baryons, JHEP 04 (2023) 034 [2301.07698].
  15. [15]J. Albrecht, F. Bernlochner, A. Lenz and A. Rusov, Lifetimes of b-hadrons and mixing of neutral B-mesons: theoretical and experimental status, Eur. Phys. J. Spec. Top. 233 (2024) 359–390 [2402.04224].arxiv.org/abs/2402.04224
  16. [16]LHCb collaboration, Measurement of the Ω⁰_c baryon lifetime, Phys. Rev. Lett. 121 (2018) 092003 [1807.02024].DOI
  17. [17]LHCb collaboration, Measurement of the lifetimes of promptly produced Ω⁰_c and Ξ⁰_c baryons, Sci. Bull. 67 (2022) 479–487 [2109.01334].
  18. [18]Belle II collaboration, Measurement of the Ω⁰_c lifetime at Belle II, Phys. Rev. D 107 (2023) L031103 [2208.08573].
  19. [19]H.-Y. Cheng, The strongest lifetime: a bizarre story of τ(Ω⁰_c), Sci. Bull. 67 (2022) 445–447 [2111.09566].DOI
  20. [20]J. Gratrex, B. Melić and I. Nišandžić, Lifetimes of singly charmed hadrons, JHEP 07 (2022) 058 [2204.11935].DOI
  21. [21]H.-Y. Cheng and C.-W. Liu, Study of singly heavy baryon lifetimes, JHEP 07 (2023) 114 [2305.00665].DOI
  22. [22]V.V. Kiselev, A.K. Likhoded and A.I. Onishchenko, Lifetimes of doubly charmed baryons: Ξ⁺⁺_cc and Ξ⁺_cc, Phys. Rev. D 60 (1999) 014007 [hep-ph/9807354].
  23. [23]B. Guberina, B. Melić and H. Štefančić, Inclusive decays and lifetimes of doubly charmed baryons, Eur. Phys. J. C 9 (1999) 213–219 [hep-ph/9901323].
  24. [24]H.-Y. Cheng and Y.-L. Shi, Lifetimes of doubly charmed baryons, Phys. Rev. D 98 (2018) 113005 [1809.08102].DOI
  25. [25]L. Dulibić, J. Gratrex, B. Melić and I. Nišandžić, Revisiting lifetimes of doubly charmed baryons, JHEP 07 (2023) 061 [2305.02243].
  26. [26]H.-Y. Cheng and C.-W. Liu, Study of doubly heavy baryon lifetimes, JHEP 08 (2026) 107 [2604.10939].
  27. [27]L. Dulibić, B. Melić and I. Nišandžić, New predictions for the lifetimes of doubly heavy baryons and the B_c meson, JHEP 09 (2026) 022 [2605.04967].
  28. [28]LHCb collaboration, Measurement of the lifetime of the doubly charmed baryon Ξ⁺⁺cc, Phys. Rev. Lett. 121 (2018) 052002 [1806.02744].DOI
  29. [29]Y.-M. Wang, H. Zou, Z.-T. Wei, X.-Q. Li and C.-D. Lü, The transition form factors for semi-leptonic weak decays of J/ψ in QCD sum rules, Eur. Phys. J. C 54 (2008) 107–121 [0707.1138].DOI
  30. [30]Y.-L. Shen and Y.-M. Wang, J/ψ weak decays in the covariant light-front quark model, Phys. Rev. D 78 (2008) 074012.DOI
  31. [31]J. Sun, L. Chen, Q. Chang, J. Huang and Y. Yang, J/ψ → DP, DV decays in the QCD factorization approach, Int. J. Mod. Phys. A 30 (2015) 1550094 [1603.00130].DOI
  32. [32]J. Sun, Y. Yang, J. Gao, Q. Chang, J. Huang and G. Lu, J/ψ → Ds,dπ, Ds,dK decays with perturbative QCD approach, Phys. Rev. D 94 (2016) 034029 [1709.05080].DOI
  33. [33]BESIII collaboration, Search for the rare semi-leptonic decay J/ψ → D−e+νe + c.c., JHEP 06 (2021) 157 [2104.06628].
  34. [34]BESIII collaboration, Search for the charmonium weak decays J/ψ → Ds−ρ+ + c.c. and J/ψ → Ds−π+ + c.c., JHEP 12 (2025) 077 [2506.09386].
  35. [35]BESIII collaboration, Search for the charmonium semi-leptonic weak decay J/ψ → Ds−e+νe + c.c., JHEP 07 (2026) 194 [2510.25100].
  36. [36]BESIII collaboration, Search for J/ψ weak decays containing a D meson, Phys. Rev. D 110 (2024) 032020 [2310.07277].arxiv.org/abs/2310.07277
  37. [37]BESIII collaboration, Search for the semi-muonic charmonium decay J/ψ → D−μ+νμ + c.c., JHEP 01 (2024) 126 [2307.02165].DOI
  38. [38]BESIII collaboration, Search for the rare decays J/ψ → D0e+e− + c.c. and ψ(3686) → D0e+e− + c.c., Phys. Rev. D 96 (2017) 111101 [1710.02278].
  39. [39]M. Achasov et al., STCF conceptual design report (Volume 1): Physics & detector, Front. Phys. 19 (2024) 14701 [2303.15790].
  40. [40]A.J. Buras and P.H. Weisz, QCD nonleading corrections to weak decays in dimensional regularization and ’t Hooft–Veltman schemes, Nucl. Phys. B 333 (1990) 66–99.
  41. [41]Y. Du, X.-G. He, J.-P. Ma and X.-Y. Du, Fundamental tests of P and CP symmetries using octet baryons at the J/ψ threshold, Phys. Rev. D 110 (2024) 076019 [2405.09625].arxiv.org/abs/2405.09625
  42. [42]I.I. Bigi, N.G. Uraltsev and A.I. Vainshtein, Nonperturbative corrections to inclusive beauty and charm decays: QCD versus phenomenological models, Phys. Lett. B 293 (1992) 430–436, Erratum Phys. Lett. B 297 (1992) 477 [hep-ph/9207214].
  43. [43]A.V. Manohar and M.B. Wise, Inclusive semileptonic B and polarized Λb decays from QCD, Phys. Rev. D 49 (1994) 1310–1329 [hep-ph/9308246].pubmed.ncbi.nlm.nih.gov/10017102
  44. [44]B. Blok and M.A. Shifman, The rule of discarding 1/Nc in inclusive weak decays (I), Nucl. Phys. B 399 (1993) 441–458 [hep-ph/9207236].DOI
  45. [45]M. Neubert, Heavy quark symmetry, Phys. Rept. 245 (1994) 259–396 [hep-ph/9306320].DOI
  46. [46]M. Beneke and G. Buchalla, The B_c meson lifetime, Phys. Rev. D 53 (1996) 4991 [hep-ph/9601249].DOI
  47. [47]G.T. Bodwin, E. Braaten and G.P. Lepage, Rigorous QCD analysis of inclusive annihilation and production of heavy quarkonium, Phys. Rev. D 51 (1995) 1125 [Erratum ibid. 55 (1997) 5853] [hep-ph/9407339].pubmed.ncbi.nlm.nih.gov/10018572
  48. [48]HPQCD collaboration, Charmonium properties from lattice QCD+QED: hyperfine splitting, J/ψ leptonic width, charm quark mass, and α_s, Phys. Rev. D 102 (2020) 054511 [2005.01845].
  49. [49]A. Lenz, M.L. Piscopo and A.V. Rusov, Contribution of the Darwin operator to non-leptonic decays of heavy quarks, JHEP 12 (2020) 199 [2004.09527].DOI
  50. [50]E.E. Jenkins, A.V. Manohar and P. Stoffer, Low-energy effective field theory below the electroweak scale: anomalous dimensions, JHEP 01 (2018) 084 [1711.05270].DOI

Paper details

vlnc:W_b9a8pxr6
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 1

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents