Probing early parton emissions in heavy ion collisions using the Lund jet plane

By analyzing the Lund jet plane of high-energy jets in PbPb collisions compared to pp collisions, this study finds no significant difference in the angular distribution of high-$k_\mathrm{T}$ emissions, suggesting that these early parton emissions occur before the jet interacts substantially with the quark-gluon plasma.

The Gist

The Cosmic Firework Show: A Story of Tiny Sparks and Thick Smoke

Imagine you are standing in a dark field, watching a massive, high-speed firework explode.

When that firework goes off, it doesn't just release one single light; it releases a "shower" of sparks. Some sparks fly out immediately, very fast and very straight. Others take a moment to develop, drifting slightly or changing direction as they travel through the air.

In the world of particle physics, scientists do something similar. They smash heavy atoms (like Lead) together at nearly the speed of light. This creates a "firework" of tiny particles called partons (quarks and gluons). These partons create a spray of particles we call a jet.

But there is a catch: in these heavy-ion collisions, the "air" isn't empty. The explosion creates a super-hot, super-dense "soup" called the Quark-Gluon Plasma (QGP). This soup is so thick and chaotic that it acts like a heavy, swirling fog or thick smoke that the sparks have to fight through.

---

The Goal: When did the sparks fly?

The big question this paper asks is: "Did the first, most important sparks fly before the thick smoke (the QGP) even formed?"

If the first sparks happen instantly, they should look exactly like they do in a "clean" environment (like a simple proton-proton collision where there is no thick smoke). If the smoke is already there when the sparks fly, the smoke will push them around, change their angles, and mess up their pattern.

The Tool: The "Lund Jet Plane" (The Spark Map)

To figure this out, scientists use a mathematical tool called the Lund Jet Plane.

Think of the Lund Jet Plane as a high-speed camera map. Instead of just taking a blurry photo of the explosion, this map records two specific things about every single spark:

1. How much energy the spark has (its "oomph").
2. What angle it flew off at compared to its parent particle.

By plotting these on a map, scientists can see a "fingerprint" of the jet. The "top-left" corner of this map represents the very first, most powerful, and earliest sparks.

The Discovery: The "Vacuum" Signature

The researchers compared the "spark maps" from the messy, smoky Lead-Lead collisions to the clean, clear Proton-Proton collisions.

Here is what they found:
When they looked at the most powerful, earliest sparks (the ones with high $k_T$), the maps looked almost identical in both the smoky collisions and the clean ones.

What does this mean?
It means the earliest, most intense part of the "firework" happens so incredibly fast that it finishes its business before the thick, hot soup of the Quark-Gluon Plasma has even had a chance to settle in.

It’s like throwing a firecracker into a thick fog: the initial "bang" and the first few bright flashes happen so quickly that the fog doesn't have time to dampen the sound or dim the light. The "quenching" (the slowing down of the particles) only happens later, as the remaining, smaller sparks try to struggle through the soup.

Why does this matter?

This is a huge win for our understanding of the universe. It confirms a major theory in physics: that we can separate the "initial explosion" from the "medium it travels through."

By proving that the earliest emissions are "vacuum-like" (meaning they behave as if they are in empty space), scientists can now more accurately use these jets as "probes." They can use the later, modified sparks to study the properties of the Quark-Gluon Plasma itself—the same kind of matter that filled our entire universe just microseconds after the Big Bang.

Technical Summary

Technical Summary: Probing Early Parton Emissions in Heavy Ion Collisions using the Lund Jet Plane

Reference: CMS Collaboration, CERN-EP-2025-260 / CMS-HIN-24-016, submitted to Physical Review Letters (2026).

---

1. The Problem: Timing the Jet-Medium Interaction

In ultrarelativistic heavy-ion collisions (such as Lead-Lead, PbPb), a Quark-Gluon Plasma (QGP) is formed. High-energy partons (quarks and gluons) traversing this medium undergo "jet quenching"—energy loss and radiation pattern modification due to interactions with the QGP.

A central theoretical question in high-energy nuclear physics is the timescale of jet evolution: Do the earliest, highest-virtuality parton emissions occur before the QGP is fully formed (vacuum-like), or are they immediately modified by the medium? Current theoretical models often assume a factorization where the initial shower is vacuum-like, followed by a medium-modification stage. This paper seeks to experimentally test this assumption by isolating the earliest emissions in the jet shower.

2. Methodology: The Lund Jet Plane (LJP)

To resolve the timing and scale of emissions, the study utilizes the Primary Lund Jet Plane (PLJP).

• The Observable: The PLJP is a two-dimensional phase-space representation of intrajet emissions. It maps the relative transverse momentum ($k_T$) and the angle ($\Delta$) of an emission with respect to its emitter.
• The Kinematic Link to Time: The formation time of an emission is approximated by $t_{\text{form}} \approx 2 / (k_T \Delta)$. Consequently, the upper-left corner of the Lund plane (high $k_T$, small $\Delta$) corresponds to the earliest-formed emissions.
• Iterative Declustering: The CMS experiment uses the Cambridge–Aachen (CA) algorithm to iteratively undo the jet clustering process, recording the kinematics of the "softer" subjet at each branching step.
• Data Selection & Cleaning:
  • Data Sets: PbPb collisions (2018) and proton-proton (pp) collisions (2017) at $\sqrt{s_{NN}} = 5.02$ TeV.
  • Hardness Selection: To specifically target the earliest vacuum-like stages, the analysis focuses on the hardest $k_T$ emission in each jet, sliced into two ranges: $k_T \in [10, 20]$ GeV and $k_T \in [20, 40]$ GeV.
  • Background Mitigation: In PbPb, the large Underlying Event (UE) is mitigated via constituent subtraction and a $k_T > 3$ GeV cutoff.
  • Unfolding: Detector-level distributions are unfolded to the particle level using a six-dimensional response matrix (D’Agostini iterative unfolding) to correct for detector effects and UE fluctuations.

3. Key Contributions

• First $k_T$-sliced LJP in Heavy Ions: This is the first measurement of the PLJP in heavy-ion collisions specifically partitioned into $k_T$ slices to separate soft and hard energy scales.
• Direct Test of Factorization: By selecting very hard splittings, the study provides a direct experimental probe into whether the initial stages of jet fragmentation are decoupled from the QGP medium.
• Model Comparison: The study provides a rigorous comparison between experimental data and several state-of-the-art jet quenching Monte Carlo generators (HYBRID, JETSCAPE, and JEWEL).

4. Results

• Angular Distributions: The angular distributions ($\ln(1/\Delta)$) of the hardest emissions in PbPb collisions are found to be similar in shape to those in pp collisions.
• The PbPb/pp Ratio: When comparing the two systems, the ratio of the distributions is flat (angle-independent) within experimental uncertainties.
  • In the $k_T \in [20, 40]$ GeV bin, a $\chi^2$ test yielded a p-value of 0.74, confirming the flatness.
  • The ratio is less than unity, indicating an overall suppression of emissions in PbPb, which is expected due to energy loss.
• Model Performance:
  • The HYBRID model (specifically the $L=0$ incoherent limit) describes the data best.
  • The JEWEL model also produces a nearly flat ratio, consistent with the data.
  • The data suggests that the medium resolution length ($L$) is small, meaning the medium resolves color charges effectively, but the shape of the earliest radiation remains vacuum-like.

5. Significance

The finding that the angular distribution of high-$k_T$ emissions is unchanged in PbPb compared to pp is highly significant. It provides strong experimental evidence that the earliest, most energetic parton emissions occur before the jet has significantly interacted with the QGP.

This supports the theoretical framework of factorized jet evolution, where the initial vacuum-like shower is established prior to the onset of medium-induced quenching. This result constrains the space-time evolution models of the QGP and provides a benchmark for future studies of medium-induced radiation and color coherence.