In-medium QCD splittings beyond the soft, large- N c N_{c} and harmonic-oscillator approximations all at once
Marco Leitão, José Guilherme Milhano, Alba Soto-Ontoso · 2026
Read In-medium QCD splittings beyond the soft, large- N c N_{c} and harmonic-oscillator approximations all at once by Marco Leitão, José Guilherme Milhano, Alba Soto-Ontoso (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-26039.txt.
Introduction
Marco Leitão, José Guilherme Milhano, and Alba Soto-Ontoso compute how quarks and gluons split inside quark-gluon plasma. Jets from heavy-ion collisions probe that hot, dense matter. A parton radiating inside the plasma loses energy in patterned ways. The BDMPS-Z formalism, now nearly thirty years old, describes the radiation. Standard uses of it lean on three approximations at once: soft emission, large colour count, and a harmonic-oscillator medium model.
Numerical solution replaces all three shortcuts. The team reports solving the BDMPS-Z equations numerically, completely, for the first time. Accuracy holds across the whole phase space. Finite-energy effects enter the results. Subleading colour contributions enter too, along with a realistic model of parton-medium interaction. Comparing against the standard approximations reveals substantial deviations across phase space. Sharper splitting functions should tighten jet-based measurements of plasma properties at the high-luminosity LHC.
Catalog record
| CREATOR | Marco Leitão, José Guilherme Milhano, Alba Soto-Ontoso |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.26039
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
In-medium QCD splittings beyond the soft, large- N c N_{c} and harmonic-oscillator approximations all at once Marco Leitão, José Guilherme Milhano, Alba Soto-Ontoso Abstract Nearly thirty years ago, Baier, Dokshitzer, Mueller, Peigné, Schiff, and Zakharov (BDMPS-Z) introduced a formalism to calculate the fully differential probability for a high-energy quark or gluon to radiate inside a finite-volume QCD plasma. We report on the first complete numerical solution of the BDMPS-Z equations for in-medium QCD splittings. Our numerical routines are precise across phase-space, enabling a determination of the in-medium splitting functions that is significantly beyond the state-of-the-art, including finite-energy effects, subleading-color contributions, and a realistic model for parton-medium interactions. We quantify the uncertainties associated with standard approximations in the literature, revealing substantial deviations across phase-space. This work opens a path toward more precise calculations of jet observables and for powerful new constraints of medium parameters from high-energy heavy-ion collider data. Extracting the material properties of quark-gluon matter at high temperatures is a long-standing objective of the heavy-ion collider programme. Among the most powerful probes of these properties are highly energetic particle jets, which are born from hard scatterings and evolve over space-time scales spanning several orders of magnitude before reaching the detector 11 ; 74 . This multi-scale sensitivity has been instrumental in discovering and characterizing the deconfined, chirally symmetric phase of matter known as the quark-gluon plasma (QGP) 43 ; 44 ; 1 ; 54 . The precision frontier enabled by the high-luminosity LHC era offers unprecedented opportunities to leverage jet observables for decoding the microscopic structure of hot, quark-gluon matter 42 . The theoretical description of jet evolution in the QGP is a multidisciplinary endeavor that integrates insights from both perturbative and non-perturbative techniques in the context of Quantum Chromodynamics (QCD) 65 . At the heart of this framework is the evaluation of in-medium QCD splitting kernels, which encode the probability for a high-energy parton, a , to branch into two daughter partons, b and c , within the medium, as illustrated in Fig. 1 . These splitting kernels serve as building blocks for calculating jet observables, whether through Monte-Carlo event generators or via analytical calculations. This work focuses on developing a method to compute these splitting kernels that is simultaneously precise and numerically efficient across the entire phase space. The foundational BDMPS-Z framework 19 ; 18 ; 17 ; 78 ; 79 established a path-integral formulation for in-medium QCD splitting kernels under several simplifying assumptions. At the kinematic level, longitudinal momenta are assumed to be much larger than any characteristic transverse scale ( close-to-eikonal approximation ). Concurrently, the QGP is modeled as a finite-volume, stochastic color field generated by static quasiparticle sources. Despite these simplifications, the sheer complexity of the resulting equations has so far restricted their solution to specific limiting cases. One class of approximations constrains the splitting kinematics, such as the soft ( hard ) limit, where the energy fraction z of the emitted parton is much smaller than (very close to) that of the parent parton, i.e. z\to 0 ( z\to 1 ) 31 ; 8 . Another class of approximations simplifies the color structure of the splitting by taking the number of colors in QCD to be large ( large- N_{c} approximation ) so as to reduce the complexity of the color algebra. A third class of approximations involves the modeling of parton-medium interactions; earlier calculations typically assumed either multiple soft momentum exchanges ( harmonic-oscillator approximation ) 53 ; 52 ; 14 ; 12 or a single hard momentum exchange ( N=1 opacity expansion ) 75 ; 51 ; 46 ; 67 , or an infinite QGP 13 ; 58 . More recently, substantial progress has been made towards relaxing these assumptions by incorporating finite- z corrections 7 ; 47 ; 56 ; 69 ; 5 , subleading- N_{c} contributions 56 ; 24 , and more realistic parton-medium interaction models 49 ; 48 ; 64 ; 3 ; 26 ; 55 ; 25 ; 61 ; 62 . However, these efforts have been strictly limited to lifting one (or at most two) approximation(s) at a time, leaving the problem of solving the BDMPS-Z equations in their full generality completely open. Figure 1: Diagrammatic representation of the time evolution of the splitting kernel computed in this Letter, as described by Eq. ( In-medium QCD splittings beyond the soft, large- N_{c} and harmonic-oscillator approximations all at once ). The dashed gray line separates amplitude from its complex conjugate. The inset illustrates how transverse momentum and color evolve at each time step due to interactions with the medium quasi-particles, while the light-cone longitudinal momentum p^{+} is conserved. This Letter introduces a novel numerical approach to solve the BDMPS-Z equations for arbitrary z at finite N_{c} , fully resumming jet-medium interactions to all orders in opacity with a realistic scattering potential. The starting point of our formulation is the established expression for the differential emission spectrum of the process depicted in Fig. 1 \displaystyle\frac{dI}{dzd^{2}\bm{p}}=\frac{\alpha_{s}P_{ba}(z)}{4\pi^{2}\omega^{2}}\text{Re}\Big[\int_{\bm{p}_{1},\bm{p}_{2},\bm{\bar{p}}_{2},\bar{\bm{p}}}\int_{0}^{\infty}dt_{1}\int_{t_{1}}^{\infty}dt_{2}\,\delta^{(2)}_{\bar{\bm{p}}+\bm{p}} \displaystyle\times(\bm{p}_{1}\cdot\bar{\bm{p}}_{2})\,{\cal K}(\bm{p_{2}},\bm{p_{1}}|t_{2},t_{1}){\cal Q}(\bm{p},\bar{\bm{p}},\bm{p_{2}},-\bar{\bm{p}}_{2}|t_{\infty},t_{2})\Big]\,, (1) where \bm{p} is the relative transverse momentum, \int_{\bm{p}}\equiv\frac{d^{2}\bm{p}}{(2\pi)^{2}} , p^{+} is the light-cone longitudinal momentum of the parent parton, \omega=z(1-z)p^{+} , and P_{ba}(z) is the vacuum Altarelli-Parisi splitting function. The strong coupling constant is denoted by \alpha_{s} , which we approximate as constant, and the intermediate momentum variables are defined as shown in Fig. 1 . The objects {\cal K}(t_{2},t_{1}) and {\cal Q}(t_{\infty},t_{2}) can be obtained by solving either: (i) a path integral over medium-averaged traces of Wilson lines 78 , (ii) Dyson-like integral equations 35 , or equivalently, (iii) a system of non-Hermitian Schrödinger-like equations for the corresponding Green’s functions 59 . We opt for the third strategy, which requires a series of manipulations to Eq. ( In-medium QCD splittings beyond the soft, large- N_{c} and harmonic-oscillator approximations all at once ). First, considering a medium of finite length, L , we can decompose the cross section into a sum over three distinct time intervals, depending on the ordering between the medium length and the light-cone times for the splitting in amplitude ( t_{1} ) and complex-conjugate amplitude ( t_{2} ). Naturally, when t_{1},t_{2}>L we recover the vacuum splitting kernel, meaning that Eq. ( In-medium QCD splittings beyond the soft, large- N_{c} and harmonic-oscillator approximations all at once ) reduces to \displaystyle\frac{dI^{\rm vac}}{dzd^{2}\bm{p}}\equiv\frac{dI}{dzd^{2}\bm{p}}\Theta(t_{1}>L)\Theta(t_{2}>L)=\frac{\alpha_{s}P_{ba}(z)}{2\pi^{2}\bm{p}^{2}}\,. (2) For the remaining contributions, we first compute their ratio with respect to Eq. ( 2 ) to isolate the medium modifications, denoting these quantities by {\cal R}(z,\bm{p}) such that the spectrum reads \displaystyle\frac{dI}{dzd^{2}\bm{p}}=\, \displaystyle\frac{dI^{\rm vac}}{dzd^{2}\bm{p}}\,\Big[1+{\cal R}^{\text{in-in}}(z,\bm{p})+{\cal R}^{\text{in-out}}(z,\bm{p})\Big]\, \displaystyle\equiv\, \displaystyle\frac{dI^{\rm vac}}{dzd^{2}\bm{p}}\,{\cal R}_{\mathrm{med}}(z,\bm{p}). (3) The two contributions to {\cal R}_{\mathrm{med}} are calculated as follows. In the t_{1}<L<t_{2} region Eq. ( In-medium QCD splittings beyond the soft, large- N_{c} and harmonic-oscillator approximations all at once ) reduces to \displaystyle{\cal R}^{\text{in-out}}(z,\bm{p}) \displaystyle=-\frac{1}{\omega}\mathrm{Re}\Bigg[\int_{0}^{L}dt_{1}\int_{\bm{p}_{1}}i\,(\bm{p}\cdot\bm{p}_{1})\,\mathcal{K}(\bm{p},\bm{p}_{1}|L,t_{1})\Bigg] \displaystyle\equiv\frac{1}{\omega}\text{Re}[\bm{p}\cdot{\cal A}(t=L,\bm{p})]\,, (4) where {\cal A} is obtained by solving the following evolution equation in 2+1 dimensions i\partial_{t}{\cal A}(t,\,\bm{{p^{\prime}}})=\Bigg[\dfrac{{\bm{p^{\prime}}}^{2}}{2\omega}-i(\tilde{v}_{ba}\ast\cdot)\Bigg]{\cal A}(t,\bm{p^{\prime}})-\bm{p^{\prime}}\,. (5) The operator (\tilde{v}_{ba}\ast\cdot) denotes a convolution in momentum space between the potential and the function \cal A , where the potential is process-dependent and encodes the elastic scattering cross section within the medium. Our algorithm is agnostic to the specific scattering potential model. The results in the main text adopt the Gyulassy-Wang (GW) description of the medium