betaFeedback

WorksPhysics and Astronomy

Overview of the latest developments in understanding the initial state and thermalization

Abstract

A proper description of the non-equilibrium matter preceding the quark-gluon plasma (QGP) in heavy-ion collisions and its observable consequences remain a major theoretical challenge, while at the same time offering new opportunities for experimental exploration. In these proceedings, I provide an overview of studies presented in talks and posters at Quark Matter 2025 on this topic. We will focus on the latest developments regarding the features and the numerical description of the non-equilibrium pre-QGP matter, as well as the potential to use hard probes as a means to study the hydrodynamization dynamics of the QCD plasma.

Introduction

In high-energy heavy-ion collisions, strongly interacting matter formed by quarks and gluons is created under extreme conditions between the collided nuclei, and it is described by Quantum Chromodynamics (QCD). As the medium cools during its evolution, it undergoes different phases, starting from an early pre-equilibrium regime, followed by a fluid-like quark-gluon plasma (QGP) phase, and eventually hadronization. The pre-equilibrium or pre-QGP phase is of particular interest since it probes the dynamical and quantum aspects of QCD.

From this perspective, the overarching goal is to learn about the real-time properties of QCD in such extreme environments. Two central research questions arise. First, what are the details of the initial stages of the pre-QGP in heavy-ion collisions? Considerable progress has been made in recent years through the interplay of initial-state modeling, classical-statistical simulations, and QCD kinetic theory, which together form the standard weak-coupling picture of hydrodynamization (see Figure 1 and, e.g., [1]). Current efforts aim to refine this picture further by extending studies to fully three-dimensional dynamics and embedding them in phenomenological frameworks. Second, how can these stages be probed experimentally, and what are their signatures? In this context, it is essential to study how the pre-QGP dynamics affects experimental observables. Particularly hard probes may provide new opportunities for small and large systems.

Figure 1. An overview of different stages of the pre-QGP medium from the weak-coupling perspective including their physical interpretation and theoretical description. (figures from [2, 3])

Acknowledgements: I would like to thank all the speakers and colleagues on the topic of initial stages at Quark Matter 2025, especially with whom I had the pleasure to discuss their work. This work is funded in part by the Austrian Science Fund (FWF) under Grant DOI 10.55776/P34455.

In these proceedings, we will summarize recent developments in the theoretical understanding of initial stages that were presented at Quark Matter 2025, discuss possible experimental signatures, and highlight new directions for future studies.

Pre-equilibrium features of the initial stages

From a weak-coupling perspective λ≪1\lambda\ll1, the pre-QGP medium probes the very nature of quantum physics. As illustrated in Figure 1, immediately after the collision, the system is dominated by strong gluon fields with high occupancy ∼1/λ\sim1/\lambda that behave as classical waves and form the so-called Glasma [4]. It is characterized by large gauge potentials and longitudinal flux tubes with transverse domains of size ∼1/Qs\sim1/Q_s, with the saturation scale QsQ_s, that persist up to the time τ≲1/Qs∼0.1−0.2\tau\lesssim1/Q_s \sim0.1 - 0.2 fm/c. Later, the system evolves towards a QCD kinetic theory description in terms of scatterings of quasiparticles [5] that describe the hydrodynamization process of the bottom-up scenario [3, 6, 7]. The medium passes through dense and dilute stages and exhibits strong momentum anisotropies, with typical local thermalization timescales τ≲1\tau\lesssim1 fm/c. Although the Glasma and kinetic theory regimes are often matched, in reality, the transition is smooth passing by a phase of plasma instabilities [8, 9] before arriving at a shared regime where both descriptions are applicable [7, 10]. Importantly, the aforementioned distinct features of the Glasma and kinetic stages may, in principle, be experimentally testable. It should be emphasized that the picture presented here relies on weak-coupling assumptions and can be contrasted with the strong-coupling holographic approach [1], since the more realistic coupling λ∼10\lambda\sim10 is neither weak nor strong.

Initial stages go spatially 3D

Recent progress has been made toward describing and understanding the spatial structure of the initial stages beyond boost-invariant approximations. Extensions of Glasma simulations to three dimensions reveal efficient descriptions of longitudinal structures in the dilute regime [11, 12], building on earlier results of the more general case in [2, 13]. Developments in kinetic theory include new approaches to solving the QCD evolution in full 3+3D (spatial and momentum) using machine learning techniques [14]. Hybrid frameworks are also advancing: the KØMPØST approach that adds energy-momentum tensor fluctuations to a homogeneous QCD kinetic theory background [15] has been generalized to two and three spatial dimensions, in a multi-stage description of the evolution [16] complemented by the CGC-based McDipper for the initial state [17] and CLVisc for 3+1D viscous hydrodynamics. Moreover, Boltzmann-type transport approaches focusing on 2↔22 \leftrightarrow2 processes have been extended to 3+1D, showing the emergence of dynamical attractors in systems beyond the boost-invariant limit [18].

Better understanding of the pre-QGP era

At the level of kinetic theory, new insights have been obtained in simplified descriptions of the early stages. A reduced Boltzmann equation in the diffusion approximation for both elastic and inelastic scatterings reproduces the full QCD effective kinetic theory evolution during bottom-up hydrodynamization to reasonable accuracy, providing a more efficient and simple framework for longitudinally expanding systems [19]. A deeper understanding of the hydrodynamization process can be gained by studying the eigenvalues and eigenmodes of an effective Hamiltonian of the kinetic description in the adiabatic hydrodynamization framework [20], which has now been extended to include inelastic scatterings [21]. Complementary studies [22] highlight the role of quasinormal modes that are related to those of the adiabatic hydrodynamization framework during the nonthermal fixed point [23] of the bottom-up stage. Finally, highly occupied plasmas have recently been shown to exhibit thermalization of magnetic modes triggered by chaotic dynamics [24].

Hard probes: window into the initial stages

Hard probes offer valuable opportunities to study the properties of the pre-equilibrium QGP. Electromagnetic probes, such as direct photons and dileptons, are produced throughout the evolution and escape the medium largely unmodified, making them sensitive to the earliest stages [25]. In contrast, QCD probes including jets, heavy flavors, and quarkonia interact strongly with the medium at all times and therefore accumulate information from different stages of the evolution [26].

Electromagnetic probes have received renewed attention as signatures of pre-equilibrium dynamics. Recent studies have shown that dileptons can provide information on the early stages [27] and particularly on plasma anisotropies, with polarization emerging as a signal of early non-equilibrium distributions [28]. Similarly, the polarization of photons has been suggested as a signature of the early stages [29].

Jet and heavy-quark transport coefficients

Once created, jets and heavy quarks interact with the medium and thus provide an important window into the properties of the non-equilibrium pre-QGP matter. Their evolution can be characterized by transport coefficients that quantify momentum broadening. For heavy quarks with masses M≫pM \gg p, the relevant quantity is the heavy-quark diffusion coefficient κi=ddτ⟨pi2⟩\kappa_i = \frac{\mathrm{d}}{\mathrm{d}\tau}\langle p_i^2\rangle, which describes stochastic momentum kicks and can enter Lindblad evolution equations for quarkonium dynamics. In analogy, the transverse momentum broadening of jets with p≫Qs,Tp \gg Q_s, T is described by the jet quenching parameter q^i=ddt⟨p⊥,i2⟩\hat{q}_i = \frac{\mathrm{d}}{\mathrm{d}t}\langle p_{\perp,i}^2\rangle in the direction ii. The time dependence of these coefficients during the pre-equilibrium evolution can leave observable imprints on hard probes: q^i(τ)\hat{q}_i(\tau) affects jet energy loss and substructure, while κi(τ)\kappa_i(\tau) influences heavy-quark diffusion, nuclear modification factors, and the in-medium dynamics of quarkonia. Recent studies emphasize the importance of incorporating pre-equilibrium dynamics into phenomenological models, as transport coefficients from the early stages can impact the interpretation of jet quenching and heavy-flavor suppression in experimental data.

These transport coefficients have been computed in detail during the pre-equilibrium stages. In the Glasma phase, classical-statistical simulations allow the extraction of κi\kappa_i and q^i\hat{q}_i [36, 37, 38, 39, 40, 41, 42, 43, 44, 45], leading to large anisotropic values κz>κT\kappa_z > \kappa_T and q^z>q^y\hat{q}_z > \hat{q}_y with beam direction zz. A notable development is the identification of a mass dependence of κ\kappa, which extends spectral function studies from light [46] to heavy quarks [44]. Spectral reconstruction methods, where gauge-invariant transport coefficients are compared with integrals over gauge-fixed spectral functions [38, 45], have successfully explained the observed properties of their evolution. For example, an initial increase κi∼τ\kappa_i \sim\tau is due to a coherence effect, while in two-dimensional Glasma-like systems the observed negative κz<0\kappa_z < 0 originates from narrow scalar excitations and the observed large values of κT\kappa_T are additionally enhanced due to broad gluonic excitations and a novel transport peak [45, 47]. During the kinetic phase, q^(τ)\hat{q}(\tau) seems to interpolate quite smoothly from Glasma to hydrodynamic values, whereas κ\kappa exhibits much larger deviations [48, 49, 50, 51]. Both coefficients retain the same anisotropic ordering as in the Glasma, with κz>κT\kappa_z > \kappa_T and q^z>q^y\hat{q}_z > \hat{q}_y, however, each component connecting to the Glasma phase in a less smooth manner. Importantly, the contribution from the pre-equilibrium regime is not negligible: the total amount of heavy-quark momentum broadening is found to be of the same parametric order as in the hydrodynamic phase, ⟨Δp2⟩neq∼⟨Δp2⟩eq\langle\Delta p^2\rangle_{\mathrm{neq}} \sim\langle\Delta p^2\rangle_{\mathrm{eq}} [40, 48, 52].

Impact of QCD probes on observables

The anisotropies and dynamics of the pre-QGP medium leave characteristic traces on observables. For instance, correlations of heavy-quark pairs in the Glasma, the spectrum of a jet parton splitting into a ccˉc\bar{c} pair, and the jet-medium interaction potential during the kinetic regime all reveal nontrivial azimuthal dependences, as illustrated in Figure 2 [30, 31, 32, 33, 34, 35]. It was also suggested that medium anisotropies could manifest themselves in the polarization of jets [53] or in the features of the differential gluon radiation rates [54]. Related studies have investigated medium-induced gluon radiation in Glasma models [55] and in evolving plasmas with time-dependent q^(τ)\hat{q}(\tau) [56].

Figure 2. Left: A sizable azimuthal correlation of ccˉc\bar{c} pairs created in the Glasma remains even at relatively late times [30, 31]. Center: One observes large deviations for the (differential) spectrum of a jet parton splitting into a ccˉc\bar{c} pair in an anisotropic plasma as compared to the isotropic case [32]. Right: The jet collision kernel (that enters radiative energy loss and splitting rate calculations) of a pre-QGP plasma in the kinetic regime (using the isoHTL screening prescription [33]) shows strong azimuthal (ϕ\phi) and transverse momentum (q⊥q_\perp) variations at early times [34, 35].

Further developments have been made for heavy flavors and quarkonia. Predictions for DD-meson observables have been obtained from IP-Glasma hybrid frameworks [52]. Modified Wong equations have been used to describe quarkonium melting in the Glasma, including memory effects [57, 58]. Pre-equilibrium effects have also been studied in smaller systems, such as charm and beauty diffusion in the Glasma in pA collisions, where considerable quarkonium melting has been seen already at the pre-equilibrium stage [59].

Conclusion

The initial stages of heavy-ion collisions provide unique insights into the real-time dynamics of QCD. From a weak-coupling perspective, the pre-equilibrium medium evolves from coherent gluon fields to quasiparticles and kinetic theory. Recent work has extended this picture into full spatial dynamics and revealed novel features suitable for embedding into multi-stage frameworks. Hard probes—both electromagnetic and QCD—offer a promising way to test these developments, but require further studies on both theoretical and experimental frontiers. A recent theory workshop (“High energy probes of the initial stages”) has additionally stimulated progress in this direction, which may provide compelling opportunities for the high-luminosity LHC Runs 4 and 5, small collision systems, and CBM-FAIR experiments.

References

  1. [1]J. Berges, M. P. Heller, A. Mazeliauskas and R. Venugopalan Rev. Mod. Phys. 93 (2021) no. 3 035003.
  2. [2]A. Ipp and D. Müller Phys. Lett. B 771 (2017) 74.
  3. [3]A. Kurkela and Y. Zhu Phys. Rev. Lett. 115 (2015) no. 18 182301.
  4. [4]F. Gelis Int. J. Mod. Phys. A 28 (2013) 1330001.
  5. [5]P. B. Arnold, G. D. Moore and L. G. Yaffe JHEP 01 (2003) 030.
  6. [6]R. Baier, A. H. Mueller, D. Schiff and D. T. Son Phys. Lett. B 502 (2001) 51.
  7. [7]J. Berges, K. Boguslavski, S. Schlichting and R. Venugopalan Phys. Rev. D 89 (2014) no. 7 074011.
  8. [8]S. Mrowczynski Phys. Lett. B 314 (1993) 118.
  9. [9]P. Romatschke and R. Venugopalan Phys. Rev. Lett. 96 (2006) 062302.
  10. [10]J. Berges, K. Boguslavski, S. Schlichting and R. Venugopalan Phys. Rev. D 89 (2014) no. 11 114007.
  11. [11]A. Ipp, D. I. Müller, S. Schlichting and P. Singh Phys. Rev. D 104 (2021) no. 11 114040.
  12. [12]A. Ipp et. al. Phys. Rev. D 109 (2024) no. 9 094040.
  13. [13]D. Gelfand, A. Ipp and D. Müller Phys. Rev. D 94 (2016) no. 1 014020.
  14. [14]S. Barrera Cabodevila, A. Kurkela and F. Lindenbauer arXiv:2506.19632 [hep-ph].DOI
  15. [15]A. Kurkela et. al. Phys. Rev. Lett. 122 (2019) no. 12 122302.
  16. [16]O. Garcia-Montero, S. Schlichting and J. Zhu Phys. Rev. D 111 (2025) no. 7 076029.
  17. [17]O. Garcia-Montero, H. Elfner and S. Schlichting Phys. Rev. C 109 (2024) no. 4 044916.
  18. [18]V. Nugara, V. Greco and S. Plumari Eur. Phys. J. C 85 (2025) no. 3 311.
  19. [19]S. Barrera Cabodevila, X. Du, C. A. Salgado and B. Wu arXiv:2503.24291 [hep-ph].arxiv.org/abs/2503.24291
  20. [20]J. Brewer, L. Yan and Y. Yin Phys. Lett. B 816 (2021) 136189.
  21. [21]K. Rajagopal, B. Scheihing-Hitschfeld and R. Steinhorst arXiv:2507.21232 [hep-ph].arxiv.org/abs/2507.21232
  22. [22]M. De Lescluze and M. P. Heller Phys. Rev. Lett. 135 (2025) no. 9 091601.
  23. [23]J. Berges, K. Boguslavski, S. Schlichting and R. Venugopalan Phys. Rev. Lett. 114 (2015) no. 6 061601.
  24. [24]H. Pandey, R. Shanker and S. Sharma arXiv:2407.09253 [hep-lat].DOI
  25. [25]F. Geurts and R. A. Tripolt Prog. Part. Nucl. Phys. 128 (2023) 104004.
  26. [26]L. Apolinário, Y.-J. Lee and M. Winn Prog. Part. Nucl. Phys. 127 (2022) 103990.
  27. [27]M. Coquet et al. Phys. Lett. B 821 (2021) 136626.
  28. [28]M. Coquet et al. Phys. Rev. Lett. 132 (2024) no. 23 232301.
  29. [29]S. Hauksson and C. Gale arXiv:2306.10307 [nucl-th].arxiv.org/abs/2306.10307
  30. [30]D. Avramescu et al. Phys. Rev. Lett. 134 (2025) no. 17 172301.
  31. [31]D. Avramescu et al. Phys. Rev. D 111 (2025) no. 7 074036.
  32. [32]J. Barata, C. A. Salgado and J. M. Silva JHEP 12 (2024) 023.
  33. [33]K. Boguslavski and F. Lindenbauer Phys. Rev. D 110 (2024) no. 7 074017.
  34. [34]A. Altenburger, K. Boguslavski and F. Lindenbauer arXiv:2509.03868 [hep-ph].arxiv.org/abs/2509.03868
  35. [35]F. Lindenbauer arXiv:2509.09897 [hep-ph].arxiv.org/abs/2509.09897
  36. [36]A. Ipp, D. I. Müller and D. Schuh Phys. Lett. B 810 (2020) 135810.
  37. [37]A. Ipp, D. I. Müller and D. Schuh Phys. Rev. D 102 (2020) no. 7 074001.
  38. [38]K. Boguslavski, A. Kurkela, T. Lappi and J. Peuron JHEP 09 (2020) 077.
  39. [39]M. E. Carrington, S. Mrówczyński and B. Schenke Phys. Rev. C 95 (2017) no. 2 024906.
  40. [40]M. E. Carrington, A. Czajka and S. Mrowczynski Phys. Lett. B 834 (2022) 137464.
  41. [41]M. E. Carrington, A. Czajka and S. Mrowczynski Phys. Rev. C 105 (2022) no. 6 064910.
  42. [42]P. Khowal, S. K. Das, L. Oliva and M. Ruggieri Eur. Phys. J. Plus 137 (2022) no. 3 307.
  43. [43]D. Avramescu et al. Phys. Rev. D 107 (2023) no. 11 114021.
  44. [44]H. Pandey, S. Schlichting and S. Sharma Phys. Rev. Lett. 132 (2024) no. 22 222301.
  45. [45]L. Backfried, K. Boguslavski and P. Hotzy Phys. Rev. D 110 (2024) no. 11 114013.
  46. [46]K. Boguslavski, T. Lappi, M. Mace and S. Schlichting Phys. Lett. B 827 (2022) 136963.
  47. [47]K. Boguslavski, A. Kurkela, T. Lappi and J. Peuron JHEP 05 (2021) 225.
  48. [48]K. Boguslavski et al. Phys. Rev. D 109 (2024) no. 1 014025.
  49. [49]K. Boguslavski et al. Phys. Lett. B 850 (2024) 138525.
  50. [50]K. Boguslavski et al. Phys. Rev. D 110 (2024) no. 3 034019.
  51. [51]K. Boguslavski et al. Phys. Lett. B 852 (2024) 138623.
  52. [52]M. Singh et. al. arXiv:2509.18647 [nucl-th].DOI
  53. [53]S. Hauksson and E. Iancu JHEP 08 (2023) 027.
  54. [54]J. Barata, K. Boguslavski, F. Lindenbauer and A. V. Sadofyev, in preparation .
  55. [55]J. Barata, S. Hauksson, X. Mayo López and A. V. Sadofyev Phys. Rev. D 110 (2024) no. 9 094055.
  56. [56]S. P. Adhya and K. Tywoniuk arXiv:2409.04295 [hep-ph].arxiv.org/abs/2409.04295
  57. [57]M. Ruggieri, Pooja, J. Prakash and S. K. Das Phys. Rev. D 106 (2022) no. 3 034032.
  58. [58]Pooja et. al. Phys. Rev. D 110 (2024) no. 9 094018.
  59. [59]L. Oliva, G. Parisi, V. Greco and M. Ruggieri Phys. Rev. D 112 (2025) no. 1 014008.

Paper details

vlnc:W_xf3tw8qh
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

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