NLO Quarkonium Impact Factor Analysis

• NLO Quarkonium Impact Factor is a key component in high-energy factorization that quantifies heavy quark-antiquark state production with both virtual and real emission contributions.
• Analytic methods incorporate dimensional regularization and subtraction schemes to cancel divergences, ensuring compatibility with both BFKL resummation and collinear factorization.
• Complete NLO impact factors for various NRQCD channels enable NLL-accurate predictions for forward quarkonium production at LHC and future high-energy colliders.

Heavy quarkonium production at high energies is a theoretically rich laboratory for the interplay of QCD factorisation, evolution, and resummation. The next-to-leading order (NLO) quarkonium impact factor, central to precision calculations in high-energy factorisation (HEF) and BFKL frameworks, quantifies the coupling of an external particle (gluon, photon, or a Reggeised gluon) to a heavy quark-antiquark ($Q\bar Q$) state in a specific nonrelativistic QCD (NRQCD) channel, including virtual and real emission effects. Its analytic structure is essential for the resummation of large logarithms in the high-energy (Regge) limit, and for matching to collinear factorisation (CF) at moderate kinematics. Rapid progress since 2023 has led to the first complete NLO BFKL impact factors for $S$-wave quarkonium states, opening the path to NLL-accurate phenomenology at the LHC and future colliders.

1. Factorisation Structure and Definition

In the high-energy ($\hat s \gg M^2$) or multi-Regge limit, hadronic and photon-induced production cross sections of quarkonium factorise as the convolution of two impact factors and a universal BFKL Green's function,

$$d\hat\sigma_{ab} = \int d^2q_1\,d^2q_2\, V^{(\cal Q)}_a(q_1,p_1,z_1)\, G(Y; q_1, q_2)\, V^{(\cal Q)}_b(q_2,p_2,z_2),$$

where $V^{(\cal Q)}_a$ is the process- and channel-dependent impact factor for parton $a$ producing a $Q\bar Q$ pair projected onto a definite NRQCD $[^1S_0^{[1,8]}]$ or $[^3S_1^{[8]}]$ state, $G$ is the BFKL Green’s function, and $Y$ is the rapidity interval (Fucilla et al., 7 Jan 2026). The impact factor contains all hard (short-distance) and finite-mass dependence of the transition $a+R\to Q\bar Q[m]$ with $R$ the Reggeised gluon, and $m$ the NRQCD state.

The HEF approach can be generalised to include real and virtual corrections at NLO, and is compatible with NRQCD and dipole frameworks. In this regime, large $\ln(\hat s/M^2)$ logarithms spoil fixed-order expansions and necessitate resummation. The NLO impact factor is central to this resummation, providing the process-dependent, non-universal building block for observables at NLL accuracy (Nefedov, 2023, Nefedov, 2024).

2. Leading and Next-to-Leading Order Impact Factors: Structure and Analytic Results

The LO impact factor for $S$-wave quarkonium production is determined by the squared amplitude for $j(q)+R(k)\rightarrow {\cal Q}[m](p)$, and encodes the necessary quark and gluon spin-color projectors per NRQCD. For color-singlet (CS) and color-octet (CO) channels, explicit analytic forms depend on $p_T$ and quark mass.

At NLO, the impact factor is a sum of one-loop virtual corrections (vertex and wave-function renormalisation) and real-emission processes ($j + R \to Q\bar Q[m] + g/q$) with local subtraction of divergences. Dimensional regularisation is employed; rapidity divergences are regulated via tilted Wilson lines (Nefedov, 2024), and the final results display the expected double and single poles ($1/\epsilon^2,\,1/\epsilon$), which are cancelled by proper subtraction and renormalisation procedures (Nefedov, 2023, Fucilla et al., 7 Jan 2026).

The NLO analytic results for the interference between one-loop and Born amplitudes can be written as

$$2{\rm Re}\, \frac{H_{1L\times LO}^m}{(\alpha_s/2\pi) H_{LO}^m} = \text{pole terms} + F^m(\tau) + O(r,\,\epsilon),$$

where $\tau = k_T^2/M^2$, $F^m(\tau)$ are finite remainder functions decomposed into $C_F$, $C_A$, and $n_f$ pieces, analytically given in terms of dilogarithms and logarithms of $\tau$ (Nefedov, 2023, Nefedov, 2024, Fucilla et al., 7 Jan 2026).

3. Regularisation, Subtraction, and Scheme Translations

Rapidity divergences arising from eikonal denominators in loop integrals are regulated by tilting the Wilson lines, introducing a small $r$ parameter that enters only as a single $\ln r$ in the final result, proportional to the one-loop gluon Regge trajectory,

$$\omega_g^{(1)}(-t) = \frac{\bar\alpha_s C_A}{2\pi} \frac{1}{\epsilon} \left(\frac{\mu^2}{-t}\right)^\epsilon,$$

up to $O(r,\,\epsilon)$ corrections (Nefedov, 2024). After subtracting corresponding Regge propagator corrections, all power and double logarithms in $r$ cancel, leaving schemes compatible with BFKL (energy-scale $s_0$), HEF (target- or projectile-side cutoffs), or shockwave regularisation (Nefedov, 2024).

Subtraction terms for real emission, especially in color-octet channels, ensure complete cancellation of soft and collinear divergences after combining with DGLAP counterterms and UV renormalisation. The full NLO IF is thus infrared safe and compatible with mass factorisation (Fucilla et al., 7 Jan 2026).

4. Logarithmic Structure, Resummation, and Matching

The NLO quarkonium impact factor encodes all single and double logarithms in $k_T^2/M^2$ and $\ln(\hat s/M^2)$, with the explicit pole and logarithm structure dictated by Regge factorisation and gluon Reggeisation. The presence of large single logs indicates that fixed-order calculations break down at high energy unless resummed (Nefedov, 2023, Flett et al., 2024).

Resummation to all orders of terms $[\alpha_s\ln(1/X)\ln(k_T^2/\mu_F^2)]^n$ (HEF, DLA) is achieved by exponentiating these contributions into a resummation factor $\mathcal{C}_{gi}(X, k_T^2,\mu_F)$. The NLO impact factor provides the first correction beyond DLA, necessary for NLL-accurate BFKL treatments (Nefedov, 2023, Fucilla et al., 7 Jan 2026). For exclusive processes and observables at moderate $X$, a matched scheme combines the fixed-order and resummed results with a smooth transition, ensuring stability of predictions (Flett et al., 2024).

5. Channels, Analytic Forms, and Phenomenological Implications

Comprehensive NLO impact factors have been constructed for the central NRQCD $S$-wave channels:

• ${}^1S_0^{[1]}$ (color-singlet)
• ${}^1S_0^{[8]}$ (color-octet, spin-0)
• ${}^3S_1^{[8]}$ (color-octet, spin-1)

Each channel displays a different analytic remainder structure in $F^m(\tau)$, with expressions in closed form involving dilogarithms $\mathrm{Li}_2(-\tau),\,\mathrm{Li}_2(-2\tau-1),$ logarithms, and constants, as detailed in (Nefedov, 2023, Nefedov, 2024, Fucilla et al., 7 Jan 2026). Table 1 summarises the available NLO impact factors:

Channel	NLO IF analytic form	Primary Reference
${}^1S_0^{[1]}$	$\Phi^{(1)}$ closed analytic	(Nefedov, 2023, Fucilla et al., 7 Jan 2026)
${}^1S_0^{[8]}$	$\Phi^{(1)}$ closed analytic	(Nefedov, 2023, Fucilla et al., 7 Jan 2026)
${}^3S_1^{[8]}$	$\Phi^{(1)}$ w/ extra LS	(Fucilla et al., 7 Jan 2026, Nefedov, 2024)

These explicit results allow, for the first time, complete NLL-resummed phenomenology: e.g., forward-backward quarkonium and quarkonium+jet production at large rapidity gaps, with the heavy-quark mass fully included (Fucilla et al., 7 Jan 2026).

6. Extensions: Exclusive Production and Dipole Picture

In exclusive quarkonium photoproduction and deep inelastic scattering, the NLO impact factor also enters the dipole picture, where amplitudes are factorized as convolutions of light-cone wave functions and dipole cross sections (Escobedo et al., 2019). The NLO light-cone wave function of quarkonium (including relativistic and radiative corrections) has been established, and, once combined with the NLO photon wave function, will enable precise predictions for exclusive vector meson and quarkonium production observables at NLO accuracy, consistent with small-$x$ evolution (B-JIMWLK) (Escobedo et al., 2019).

In collinear factorisation, NLO quarkonium impact factors appear in the coefficient functions for exclusive production and in the evolution of generalised parton distributions (GPDs), where a resummation-matched formula restores stability at small $x$ (Flett et al., 2024).

7. Outlook and Phenomenological Applications

The availability of complete analytic NLO quarkonium impact factors in the BFKL and HEF frameworks now enables robust, NLL-precise predictions at forward rapidities, and opens a path toward fully differential cross section calculations (in $p_T$, rapidity, etc.) for heavy-quarkonia at the LHC and future facilities (Fucilla et al., 7 Jan 2026). Their inclusion is mandatory for controlling factorisation-scale dependence, matching to collinear factorisation at intermediate $X$, and achieving theoretical uncertainties commensurate with upcoming experimental data.

These developments also enable systematic studies of process dependence, rapidity-cutoff scheme translations, and connections between factorisation schemes (HEF, BFKL, shockwave), as well as providing building blocks for approaches incorporating gluon saturation and small-$x$ non-linear effects (Nefedov, 2023, Nefedov, 2024).