Probing the charm-quark dead cone effect in heavy ion collisions

arXiv:2611.01285 · nucl-ex, hep-ex · Submitted 2026-10-31 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Nuclear Experiment. The day in nuclear experiment: beams, targets and what came out.

Kai: We're looking at the paper: "Probing the charm-quark dead cone effect in heavy ion collisions".

Mira: Here's the gist of the paper. The first experimental study of the dead-cone effect for the charm quark in heavy ion collisions is presented The angular distribution of late-kT emissions in PbPb collisions is found to be shifted towards larger angles in D0 jets relative to inclusive jets, indicating a suppression of collinear splittings for D0 jets This shift, which is expected from the dead-cone effect, is consistent with a similar observation made in proton-proton collisions The new measurement disfavors hypotheses that predict a filling of the charm-quark dead cone by medium-induced radiation in heavy ion collisions Experimental Setup and Observable The analysis is based on a sample of lead-lead (PbPb) collisions at a center-of-mass energy per nucleon pair of 5.02 TeV, recorded by the CMS experiment at the CERN LHC in 2018 and corresponding to an integrated luminosity of 1.7 nb−1 The studied observable is the angular distribution of hard, collinear intrajet emissions that are selected with the late-kT grooming algorithm The analysis is performed using inclusive jets and a sample of jets containing prompt D0 mesons Events of interest are selected using a two-tiered trigger system Events with high-pT jets are collected with an exclusive combination of the minimum bias trigger, requiring coincident signals in the HF calorimeters on both sides of the interaction point, and a trigger that requires a calorimeter object with pT > 100 GeV The collision centrality, expressed as a percentile of the total inelastic hadronic cross section

31: , is evaluated from the transverse energy deposited in both HF calorimeters To focus on medium modification effects, the analysis is restricted to the 30% most central collisions Probing Quark Mass Effects The dead-cone effect in quantum chromodynamics (QCD) is a suppression of gluon emissions off a quark in a cone of angle θd = mq/Eq around the emitting quark where mq and Eq are the mass and energy of the radiating quark, respectively The effect is more pronounced for heavy quarks, being negligible for light quarks It was measured in proton-proton (pp) collisions at the CERN LHC by ALICE

10: and CMS

11: for c quark initiated jets (c jets), and by LHCb

12: and CMS

13: for b quark initiated jets (b jets) The dead-cone effect implies a reduced number of emissions in heavy-flavor jets, leading to harder fragmentation and fewer intrajet constituents with respect to inclusive jets Jet substructure techniques allow the radiation pattern of a jet to be examined by reconstructing its successive splittings, providing an experimental proxy for the parton shower through the primary Lund jet plane By selecting emissions at smaller angles and lower energies within the jet, these techniques enhance the sensitivity to quark-mass effects, enabling the study of the charm-quark dead cone even in high-energy jets Analysis Methodology The D0 candidates are reconstructed by combining pairs of oppositely charged tracks with invariant mass within ±0.2 GeV of the D0 mass The D0 candidates must have pT > 4 GeV and rapidity y < 1.2 To distinguish the D0 → π+K− (and charge conjugate) signal from the background, we use a multivariate analysis technique based on a boosted decision tree (BDT) The remaining background is statistically subtracted through fits to the D0 invariant mass distributions After using the two charged daughter tracks to reconstruct the four-momentum of the D0 candidate, the daughters are removed from the particle list and replaced by the D0, which is treated as a single constituent in the subsequent jet finding and CA reclustering The late-kT groomer selects the last emission in the declustering process whose relative transverse momentum kT is larger than 1 GeV If the leading prong of the selected splitting contains the D0 candidate as a constituent, the “late-kT splitting angle”, θl, defined as the angular distance between the two prongs in the pseudorapidity–azimuthal angle plane, is registered; otherwise the jet is rejected Results and Conclusions The D0 signal yield is measured, in each p jetT and ln(1/θl) bin, by fitting the D0 invariant mass distributions in the 1.7–2.0 GeV range The analysis provides the angular distribution of the intrajet emissions, as selected by the latekT algorithm, for D0 jets and inclusive jets After using the two charged daughter tracks to reconstruct the fourmomentum of the D0 candidate, the daughters are removed from the particle list and replaced by the D0, which is treated as a single constituent in the subsequent jet finding and CA reclustering The D0 signal yield extracted from fits of the invariant mass distribution includes both a prompt term, from the fragmentation of a charm quark, and a nonprompt term, from decays of b hadrons The prompt D0 signal, which includes feed-down contributions from heavier charm hadrons, is obtained using template fits to the distance-of-closest-approach (DCA) significance distribution In heavy ion collisions, broad jets are known to lose more energy than narrow ones, leading to a selection bias effect However, the energy loss of charged particles and of D0 mesons has been seen to be compatible The PbPb-to-pp double ratio is compatible with unity This supports that, in PbPb collisions, the charm-quark dead-cone region remains approximately vacuum-like The results are consistent with this expectation and with theoretical scenarios in which the charm-quark dead-cone region remains approximately vacuum-like Model Comparison The lower panels of Fig.

Kai: 3 compare the measurements with HYBRID

57: and LIDO

58: models, in their “vacuum” and “medium” scenarios The vacuum parton-shower baselines are PYTHIA for both models, with next-to-leading order corrections

59: included in the LIDO model Both models predict a similar hierarchy between prompt-D0-tagged and inclusive jets and a PbPb-to-pp double ratio close to unity The similarity of the final predictions suggests that, within the present observable and kinematic range, the flavor dependence is relatively robust against the different implementations of jet quenching considered here In the kinematic regime considered here, the charm-quark dead-cone angle is expected to be smaller than the medium coherence angle, θd < θc, suggesting that the medium induced radiation may occur predominantly outside the charm dead-cone region The present results are consistent with this expectation and with theoretical scenarios in which the charm-quark dead-cone region remains approximately vacuum-like Unfolding and Uncertainties The unfolding recovers the particle-level distributions by correcting for detector resolution and underlying-event fluctuations that migrate jets between bins of p jetT and ln(1/θl) It also corrects for UE effects, quantified using a four-dimensional response matrix built from the particle- and detector-level simulated distributions The total systematic uncertainties range from 8.6–21.5% for D0 jets and 3.3–7.8% for inclusive jets, dominated by the prompt D0 fraction, signal yield extraction, and PF scale for D0 jets The statistical uncertainties are 14.3–16% for D0 jets and 0.7–2% for inclusive jets Double Ratio Compatibility The PbPb-to-pp double ratio is compatible with unity We have also tested the hypothesis that this double ratio is constant and equal to unity over the four angular bins, retaining the available PbPb statistical bin-to-bin covariance and treating the pp uncertainties as diagonal The result is χ2 = 1.55 for the four tested bins, corresponding to a p-value of 0.82 Since the bin-to-bin covariance information is not available for the published pp reference measurement, the test uses the total bin-by-bin uncertainties, treating the PbPb and pp measurements as independent The double ratio is compatible with unity Figure Interpretation The D0 jet angular distribution shows a trend towards larger splitting angles (smaller ln(1/θl) values) relative to the inclusive jet distribution, reflecting the convolution of the energy distribution of the selected splittings and their individual dead-cone angles A similar trend was seen in pp collisions

3: and attributed to the relatively high charm quark mass The compatibility of the D0-to-inclusive ratio with unity is quantified using a χ2 test over the four angular bins common to the PbPb and pp measurements The test gives a p-value of 0.

Title and authors: Kai: So we’re looking at this paper, "Probing the charm-quark dead cone effect in heavy ion collisions." It seems like a direct look at how that dead cone idea plays out when you have these massive lead-lead collisions happening.

Mira: It's really about testing that specific prediction—the suppression of gluon emissions off a quark based on its mass and energy—in the context of a hot, dense medium. The authors are checking if the charm quark behaves the way QCD says it should in heavy ion environments.

Lev: From my side, I’m thinking about what this means for real hardware. They're looking at how many emissions you get in jets with D zero mesons compared to regular jets, and that’s a direct probe of those mass-dependent suppression effects.

Kai: Exactly. The title hints at the core investigation: using heavy ion collisions to see if the charm quark dead cone effect is suppressed or modified by the medium. It’s about checking if what we see in a proton-proton collision translates over to something different in lead-lead collisions because of that plasma.

Mira: And it brings up a key question for us: are we seeing a vacuum-like region around the charm quark, or is the medium filling that dead cone? That distinction is vital for understanding how the plasma interacts with heavy quarks.

Lev: If you want to run this on real hardware, you need to model those jet quenching effects very precisely because they're messing with the radiation patterns we're measuring.

The paper's summary: Kai: So, what the authors are summarizing is that in lead-lead collisions at five point zero two teraelectronvolt per nucleon pair, they found that the angular distribution of late ka-t emissions in D zero jets is shifted towards larger angles compared to inclusive jets.

Mira: That shift suggests a suppression of those collinear splittings for those D zero jets, which is precisely what we expect from the dead cone effect happening. They are finding consistency with similar observations from proton-proton collisions as well.

Lev: That’s interesting because it means the medium isn't completely washing out the mass dependence in this specific way, at least not in this range of measurements. We have to be careful about how we interpret that suppression.

Kai: Right, and they explicitly state that this new measurement disfavors some hypotheses that predict the charm-quark dead cone region gets filled up by medium-induced radiation when you go into heavy ion collisions.

Mira: That’s a big deal for our theoretical models because it suggests the medium might not be doing what some of those early predictions suggested regarding filling that cone. It keeps the vacuum-like idea alive for now in this specific observation.

Lev: From an error correction standpoint, if you're trying to isolate that effect, you have to account for how broad jets lose energy compared to narrow ones, but they say the energy loss of charged particles and d zero mesons is compatible with what’s seen.

The paper's improvements: Kai: The authors suggest a few ways to improve this kind of study, mostly by looking closer at the jet substructure. They focus on using techniques like measuring the late ka-t splitting angle, denoted as theta l.

Mira: That’s an improvement because it lets them directly probe the radiation pattern of the jet by reconstructing its successive splittings, which is a better proxy for that parton shower than just looking at inclusive jets alone.

Lev: From an engineering view, measuring that angular distance between the two prongs in the pseudorapidity azimuth angle plane requires very high precision in tracking and calorimetry to get those small angles right. It’s a real measurement challenge.

Kai: And they also talk about using more sophisticated unfolding methods, like the D agostini iterative unfolding, to correct for detector resolution and underlying event fluctuations that can move jets between different energy bins.

Mira: That correction is important because if you don't fix those bin-to-bin migrations, your measured yield of prompt d zero mesons could be skewed by noise rather than actual physics. It’s about getting the particle-level distributions right.

Lev: Those unfolding methods are crucial for getting a clean measurement on the charm quark fragmentation, because if you don't correct for those fluctuations, you end up with an inaccurate picture of what's actually happening inside the medium.

Conclusion: Kai: So to wrap up this paper on "Probing the charm-quark dead cone effect in heavy ion collisions," they show that the D zero jet angular distribution does indeed shift towards larger angles relative to inclusive jets, which supports the expectation from the dead cone effect.

Mira: They confirm that this is consistent with proton-proton results and that it doesn't support models where medium radiation fills the charm quark dead cone region. It reinforces the idea that it stays vacuum-like in this kinematic regime.

Lev: From a real hardware standpoint, this means for error correction, you can still treat the charm fragmentation as relatively stable compared to more extreme quenching scenarios. The compatibility of the lead-lead to proton-proton double ratio being close to one is a good sign for your simulations.

Kai: It really means that when we look at the charm quark dead cone effect, we can take these results seriously and keep those vacuum models in our consideration while we push for more complex modeling.

Mira: And it sets up the next step: focusing on how the medium coherence angle compares to that charm dead-cone angle, theta d less than theta c. That comparison will tell us if medium radiation is mostly happening outside that cone or not.

Lev: For me, it’s about setting up the simulation framework where we can test those models against these experimental constraints on how the jet yields and patterns are modified by the plasma.

CMS Collaboration

European Organization for Nuclear Research (CERN)

nucl-ex, hep-ex

Submitted: 2026-10-31

Updated: 2026-10-31

Comments: Submitted to Physical Review Letters. All figures and tables can be found at http://cms-results.web.cern.ch/cms-results/public-results/publications/HIN-25-013 (CMS Public Pages)

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 73/100

The gist: The first experimental study of the dead-cone effect for the charm quark in heavy ion collisions is presented The angular distribution of late-kT emissions in PbPb collisions is found to be shifted…

Key concepts

Dead-cone effect
This is a phenomenon in QCD where a quark suppresses gluon radiation into a narrow cone around it, defined by an angle related to the quark's mass and energy. It means fewer gluons are emitted at small angles relative to the quark's direction of motion, which is more pronounced for heavier quarks like charm.
Late-kT grooming algorithm
This technique is used to select specific emissions within a jet by focusing on those with high transverse momentum (kT > 1 GeV). It helps isolate the last splitting in a jet's evolution, allowing physicists to probe the detailed structure of how the quark fragments into particles.
PbPb-to-pp double ratio
This ratio compares measurements taken in heavy ion collisions (PbPb) with those from proton-proton collisions (pp). When this ratio is close to one, it suggests that the modification caused by the dense medium in heavy ion collisions does not significantly alter the fundamental properties of charm quark fragmentation compared to vacuum conditions.
Jet substructure analysis
This involves examining how a jet splits into smaller components. By reconstructing successive splittings, researchers can map out the radiation pattern of a jet. This technique is used here to experimentally test the dead-cone effect by looking at the angular distribution of these emissions in D0 jets.

Terminology

Summary

The first experimental study of the dead-cone effect for the charm quark in heavy ion collisions is presented The angular distribution of late-kT emissions in PbPb collisions is found to be shifted towards larger angles in D0 jets relative to inclusive jets, indicating a suppression of collinear splittings for D0 jets This shift, which is expected from the dead-cone effect, is consistent with a similar observation made in proton-proton collisions The new measurement disfavors hypotheses that predict a filling of the charm-quark dead cone by medium-induced radiation in heavy ion collisions

Experimental Setup and Observable

The analysis is based on a sample of lead-lead (PbPb) collisions at a center-of-mass energy per nucleon pair of 5.02 TeV, recorded by the CMS experiment at the CERN LHC in 2018 and corresponding to an integrated luminosity of 1. p. 0

Where the paper says it

The analysis is based on a sample of lead-lead (PbPb) collisions at a center-of-mass energy per nucleon pair of 5. 02 TeV, recorded by the CMS experiment at the CERN LHC in 2018 and corresponding to an integrated luminosity of 1.

7 nb−1 The studied observable is the angular distribution of hard, collinear intrajet emissions that are selected with the late-kT grooming algorithm The analysis is performed using inclusive jets and a sample of jets containing prompt D0 mesons Events of interest are selected using a two-tiered trigger system Events with high-pT jets are collected with an exclusive combination of the minimum bias trigger, requiring coincident signals in the HF calorimeters on both sides of the interaction point, and a trigger that requires a calorimeter object with pT > 100 GeV The collision centrality, expressed as a percentile of the total inelastic hadronic cross section [31], is evaluated from the transverse energy deposited in both HF calorimeters To focus on medium modification effects, the analysis is restricted to the 30% most central collisions

Probing Quark Mass Effects

The dead-cone effect in quantum chromodynamics (QCD) is a suppression of gluon emissions off a quark in a cone of angle θd = mq/Eq around the emitting quark where mq and Eq are the mass and energy of the radiating quark, respectively The effect is more pronounced for heavy quarks, being negligible for light quarks It was measured in proton-proton (pp) collisions at the CERN LHC by ALICE [10] and CMS [11] for c quark initiated jets (c jets), and by LHCb [12] and CMS [13] for b quark initiated jets (b jets) The dead-cone effect implies a reduced number of emissions in heavy-flavor jets, leading to harder fragmentation and fewer intrajet constituents with respect to inclusive jets Jet substructure techniques allow the radiation pattern of a jet to be examined by reconstructing its successive splittings, providing an experimental proxy for the parton shower through the primary Lund jet plane By selecting emissions at smaller angles and lower energies within the jet, these techniques enhance the sensitivity to quark-mass effects, enabling the study of the charm-quark dead cone even in high-energy jets

Analysis Methodology

The D0 candidates are reconstructed by combining pairs of oppositely charged tracks with invariant mass within ±0.2 GeV of the D0 mass The D0 candidates must have pT > 4 GeV and rapidity y < 1. p. 3

Where the paper says it

The D0 invariant mass is evaluated for two hypotheses: assuming that one track is a pion and the other is a kaon or vice-versa. The D0 candidates must have pT > 4 GeV and rapidity |y| < 1.

2 To distinguish the D0 → π+K− (and charge conjugate) signal from the background, we use a multivariate analysis technique based on a boosted decision tree (BDT) The remaining background is statistically subtracted through fits to the D0 invariant mass distributions After using the two charged daughter tracks to reconstruct the four-momentum of the D0 candidate, the daughters are removed from the particle list and replaced by the D0, which is treated as a single constituent in the subsequent jet finding and CA reclustering The late-kT groomer selects the last emission in the declustering process whose relative transverse momentum kT is larger than 1 GeV If the leading prong of the selected splitting contains the D0 candidate as a constituent, the “late-kT splitting angle”, θl, defined as the angular distance between the two prongs in the pseudorapidity–azimuthal angle plane, is registered; otherwise the jet is rejected

Results and Conclusions

The D0 signal yield is measured, in each p jetT and ln(1/θl) bin, by fitting the D0 invariant mass distributions in the 1.7–2.0 GeV range The analysis provides the angular distribution of the intrajet emissions, as selected by the latekT algorithm, for D0 jets and inclusive jets After using the two charged daughter tracks to reconstruct the fourmomentum of the D0 candidate, the daughters are removed from the particle list and replaced by the D0, which is treated as a single constituent in the subsequent jet finding and CA reclustering The D0 signal yield extracted from fits of the invariant mass distribution includes both a prompt term, from the fragmentation of a charm quark, and a nonprompt term, from decays of b hadrons The prompt D0 signal, which includes feed-down contributions from heavier charm hadrons, is obtained using template fits to the distance-of-closest-approach (DCA) significance distribution In heavy ion collisions, broad jets are known to lose more energy than narrow ones, leading to a selection bias effect However, the energy loss of charged particles and of D0 mesons has been seen to be compatible The PbPb-to-pp double ratio is compatible with unity This supports that, in PbPb collisions, the charm-quark dead-cone region remains approximately vacuum-like The results are consistent with this expectation and with theoretical scenarios in which the charm-quark dead-cone region remains approximately vacuum-like

Model Comparison

The lower panels of Fig. 3 compare the measurements with HYBRID [57] and LIDO [58] models, in their “vacuum” and “medium” scenarios The vacuum parton-shower baselines are PYTHIA for both models, with next-to-leading order corrections [59] included in the LIDO model Both models predict a similar hierarchy between prompt-D0-tagged and inclusive jets and a PbPb-to-pp double ratio close to unity The similarity of the final predictions suggests that, within the present observable and kinematic range, the flavor dependence is relatively robust against the different implementations of jet quenching considered here In the kinematic regime considered here, the charm-quark dead-cone angle is expected to be smaller than the medium coherence angle, θd < θc, suggesting that the medium induced radiation may occur predominantly outside the charm dead-cone region The present results are consistent with this expectation and with theoretical scenarios in which the charm-quark dead-cone region remains approximately vacuum-like

Unfolding and Uncertainties

The unfolding recovers the particle-level distributions by correcting for detector resolution and underlying-event fluctuations that migrate jets between bins of p jetT and ln(1/θl) It also corrects for UE effects, quantified using a four-dimensional response matrix built from the particle- and detector-level simulated distributions The total systematic uncertainties range from 8.6–21.5% for D0 jets and 3. p. 0

Where the paper says it

The studied observable is the angular distribution of hard, collinear intrajet emissions that are selected with the late-kT grooming algorithm. The analysis is performed using inclusive jets and a sample of jets containing prompt D0 mesons.

3–7.8% for inclusive jets, dominated by the prompt D0 fraction, signal yield extraction, and PF scale for D0 jets The statistical uncertainties are 14. p. 6
Where the paper says it

8% for inclusive jets, dominated by the prompt D0 fraction, signal yield extraction, and PF scale for D0 jets, and the physics model for inclusive jets. The statistical covariance matrix is propagated through the unfolding and normalization procedures.

3–16% for D0 jets and 0. p. 0
Where the paper says it

The studied observable is the angular distribution of hard, collinear intrajet emissions that are selected with the late-kT grooming algorithm. The analysis is performed using inclusive jets and a sample of jets containing prompt D0 mesons.

7–2% for inclusive jets

Double Ratio Compatibility

The PbPb-to-pp double ratio is compatible with unity We have also tested the hypothesis that this double ratio is constant and equal to unity over the four angular bins, retaining the available PbPb statistical bin-to-bin covariance and treating the pp uncertainties as diagonal The result is χ2 = 1.55 for the four tested bins, corresponding to a p-value of 0. p. 8

Where the paper says it

7 uncertainties as diagonal; the result is χ 2 = 1. 55 for the four tested bins, corresponding to a p-value of 0.

82 Since the bin-to-bin covariance information is not available for the published pp reference measurement, the test uses the total bin-by-bin uncertainties, treating the PbPb and pp measurements as independent The double ratio is compatible with unity

Figure Interpretation

The D0 jet angular distribution shows a trend towards larger splitting angles (smaller ln(1/θl) values) relative to the inclusive jet distribution, reflecting the convolution of the energy distribution of the selected splittings and their individual dead-cone angles A similar trend was seen in pp collisions [3] and attributed to the relatively high charm quark mass The compatibility of the D0-to-inclusive ratio with unity is quantified using a χ2 test over the four angular bins common to the PbPb and pp measurements The test gives a p-value of 0.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved Jet Substructure Analysis for Heavy Ions: AI systems can now perform jet substructure observables like late-kT splitting angle measurement to distinguish between prompt D0 jets and inclusive jets, as the analysis shows that D0 jets exhibit a suppression of small-angle emissions, consistent with the expected dead-cone effect.

  2. Medium Interaction Modeling: AI systems can incorporate physics models like HYBRID or LIDO to predict jet quenching effects in heavy ion collisions, allowing them to determine whether the charm-quark dead-cone region remains approximately vacuum-like based on the comparison of prompt D0 over inclusive angular distribution ratios measured in pp and PbPb collisions.

  3. Unfolding and Background Correction: AI systems can utilize multidimensional unfolding methods, such as the D’Agostini iterative unfolding, to correct for detector resolution and underlying event fluctuations, leading to a more accurate extraction of prompt D0 yields by addressing bin-to-bin migrations between p jetT and ln(1/theta l).

Abstract

The first experimental study of the dead-cone effect for the charm quark in heavy ion collisions is presented. The analysis is based on a sample of lead-lead (PbPb) collisions at a center-of-mass energy per nucleon pair of 5.02 TeV, recorded by the CMS experiment at the CERN LHC in 2018 and corresponding to an integrated luminosity of 1.7 nb-1. The studied observable is the angular distribution of hard, collinear intrajet emissions that are selected with the late- k T grooming algorithm. The analysis is performed using inclusive jets and a sample of jets containing prompt D 0 mesons. The angular distribution of late- k T emissions in PbPb collisions is found to be shifted towards larger angles in D 0 jets relative to inclusive jets, indicating a suppression of collinear splittings for D 0 jets. This shift, which is expected from the dead-cone effect, is consistent with a similar observation made in proton-proton collisions. The new measurement disfavors hypotheses that predict a filling of the charm-quark dead cone by medium-induced radiation in heavy ion collisions.

Sources

Related papers