$D^0$-$D_s^+$ Elliptic-Flow Splitting under Event-Shape Engineering: A Probe of Sequential Charm Hadronization

The Gist

Imagine a giant, super-hot soup made of tiny particles, created when two heavy lead atoms smash into each other at nearly the speed of light. This "soup" is called a Quark-Gluon Plasma (QGP). Inside this soup, heavy particles called "charm quarks" swim around. As the soup cools down, these quarks grab onto lighter particles to form new, stable particles called "hadrons" (specifically, two types of D-mesons: the D⁰ and the D⁺ₛ).

For a long time, scientists thought all these new particles formed at the exact same moment, like a group of people stepping out of a building at the same time. But this paper suggests a different story: sequential hadronization.

Here is the simple breakdown of what the authors found, using everyday analogies:

1. The Two Stories: A Group Exit vs. A Staggered Exit

• The Old Story (Simultaneous): Imagine a crowd of people leaving a concert. Everyone walks out the door at the exact same time. If you look at two different groups of people (say, those wearing red hats vs. blue hats), they all get pushed by the crowd in the same way.
• The New Story (Sequential): Imagine the concert is ending, and the exit is crowded.
  • The D⁺ₛ particles are like people with "VIP passes" (they are tightly bound). They manage to squeeze out of the crowd earlier, when the room is still very hot and chaotic (around 1.2 times the critical temperature).
  • The D⁰ particles are like regular attendees. They stay inside a bit longer, swimming in the soup until the very end (at the critical temperature, $T_c$).
  • The Result: Because the D⁰s stayed in the soup longer, they got pushed around more by the swirling currents of the crowd. They picked up more "spin" or "flow" than the D⁺ₛs, who left early.

2. The Problem: How Do We See the Difference?

Scientists can measure how much these particles "spin" (this is called elliptic flow). However, there's a catch. The amount of spin depends on two things:

1. How the collision started: Was the collision a perfect head-on smash, or a glancing blow? (This is the "shape" of the event).
2. When they left: Did they leave early or late?

If you just look at all collisions mixed together, it's hard to tell if a particle has more spin because it left late, or just because the collision was shaped in a way that created more spin. It's like trying to guess if a runner is fast because they are a natural athlete or just because they had a tailwind.

3. The Solution: "Event-Shape Engineering" (The Wind Tunnel)

The authors used a clever trick called Event-Shape Engineering (ESE). Think of it like a wind tunnel.

• They took thousands of collisions and sorted them into two piles:
  • Big $q^2$ (Strong Wind): Collisions that started with a very strong, lopsided shape.
  • Small $q^2$ (Weak Wind): Collisions that started with a more round, gentle shape.
• By comparing these two piles, they could see how the particles reacted to the "wind" of the collision geometry.

4. The Discovery: The "Slope" Tells the Tale

When they looked at the data, they found a smoking gun that proved the "Staggered Exit" (Sequential) story is likely true:

• The "Slope" ($\chi$): Imagine plotting how much spin a particle gets as the "wind" gets stronger.
  • In the Sequential story (where D⁰s stay longer), the D⁰ particles are very sensitive to the wind. When the wind gets stronger, their spin goes up a lot. The D⁺ₛ particles, having left early, don't react as much.
  • The Rule: The "sensitivity slope" for D⁰ is steeper than for D⁺ₛ.
  • In the Simultaneous story (where they leave together), both particles react the same way. Their slopes would be identical.

The paper shows that in the semi-central collisions (the "sweet spot" where the soup lasts long enough but is still lopsided), the D⁰ particles indeed have a much steeper slope than the D⁺ₛ. This proves that D⁰s are staying in the soup longer to catch more flow.

5. Why It's Not Just About Numbers

The authors also checked if this was just a trick of numbers (like having more D⁰s than D⁺ₛs in certain collisions). They looked at the ratio of D⁺ₛ to D⁰.

• The Finding: The ratio stayed the same regardless of whether the "wind" was strong or weak.
• The Meaning: This confirms that the difference in spin isn't because there are more of one type of particle; it's purely a dynamical effect caused by when they left the soup.

Summary

This paper proposes that heavy particles don't all leave the hot soup at once. The "VIP" particles (D⁺ₛ) leave early, while the "regular" particles (D⁰) stay longer and get pushed around more.

By using a technique that sorts collisions by their shape (Event-Shape Engineering), the authors found a unique fingerprint: the "regular" particles react much more strongly to the shape of the collision than the "VIP" particles. This difference in reaction is the proof that they left the soup at different times, revealing the hidden timeline of how matter forms in the early universe.

Technical Summary

Technical Summary: $D^0$–$D_s^+$ Elliptic-Flow Splitting under Event-Shape Engineering

Problem Statement
The hadronization mechanism of open-charm hadrons in the quark-gluon plasma (QGP) remains a critical open question in relativistic heavy-ion physics. Conventional models typically assume simultaneous hadronization, where all charm-hadron species form on a common hypersurface at the critical temperature ($T_c$). However, recent theoretical proposals suggest sequential hadronization, wherein more tightly bound species (e.g., $D_s^+$) form earlier at higher temperatures (e.g., $\sim 1.2T_c$), while less tightly bound species (e.g., $D^0$) form later at $T_c$.

While the difference in elliptic flow, $\Delta v_2 \equiv v_2(D^0) - v_2(D_s^+)$, was previously proposed as a probe to distinguish these scenarios, inclusive measurements face a fundamental limitation: the elliptic flow of heavy-flavor hadrons depends jointly on the initial collision geometry and the duration of partonic evolution. This degeneracy makes it difficult to isolate the specific signal of hadronization-time ordering from global geometric effects.

Methodology
The authors utilize the SHELL framework, an event-by-event simulation combining:

1. (3+1)D Viscous Hydrodynamics: Using the CLVisc code to evolve the QGP medium, providing local temperature $T(x)$ and flow velocity $u^\mu(x)$.
2. Langevin Transport: Modeling charm-quark diffusion with a constant spatial diffusion coefficient $2\pi T D_s = 2.5$, including drag, diffusion, and medium-induced gluon radiation (Higher-Twist approach).
3. Hybrid Hadronization: A coalescence-plus-fragmentation model.
   • Sequential Scenario: $D_s^+$ formation is evaluated on an earlier hypersurface at $T \approx 1.2T_c$. If a charm quark fails to coalesce into $D_s^+$, it continues to propagate until $T_c$ to form $D^0$ (or other species), enforcing charm-number conservation.
   • Simultaneous Baseline: All species form on the same $T_c$ hypersurface.
4. Event-Shape Engineering (ESE): The study selects events within fixed centrality classes (0–10% and 30–50%) based on the reduced second-order flow vector $q_2$. Events are categorized into "large-$q_2$" (strong initial eccentricity) and "small-$q_2$" (weak initial eccentricity) classes (top and bottom 20% of the distribution). This allows for a differential measurement of the hadronization-time response independent of the overall centrality.

Key Contributions and Results

• Decoupling Geometry from Time: The study demonstrates that ESE provides a sharper discrimination between sequential and simultaneous hadronization than inclusive measurements. By varying $q_2$ within a fixed centrality, the authors separate the bulk geometry response from the species-specific hadronization-time response.
• Flow Splitting ($\Delta v_2$):
  • In the sequential scenario, a positive splitting $\Delta v_2(D^0 - D_s^+) > 0$ emerges at intermediate transverse momentum ($p_T \approx 3\text{--}4$ GeV/c). This arises because $D_s^+$ freezes out early, while $D^0$ parent quarks accumulate additional elliptic flow during the interval $1.2T_c \to T_c$.
  • In the simultaneous baseline, the splitting is near zero or negative, lacking the positive enhancement.
  • Crucially, the positive splitting in the sequential scenario grows systematically with $q_2$.
• The Response Slope ($\chi$): The authors define a response slope $\chi$ by fitting $v_2$ as a linear function of $q_2$ ($v_2 = \chi q_2 + v_2^{(0)}$).
  • Sequential Prediction: A distinct hierarchy $\chi(D^0) > \chi(D_s^+)$ is observed. Because $D^0$ forms later, it is more sensitive to the late-stage geometric evolution.
  • Simultaneous Prediction: The difference $\chi(D^0) - \chi(D_s^+)$ is approximately zero or negative.
  • This slope serves as a robust, model-independent discriminator that is insensitive to overall flow normalization.
• Optimal Centrality Window: The 30–50% semi-central class is identified as the optimal window for observation. While central collisions (0–10%) have longer QGP lifetimes, their smaller initial eccentricity limits the conversion of late-stage interactions into elliptic flow. Peripheral collisions have large eccentricity but short lifetimes. The 30–50% class offers the non-monotonic interplay necessary to maximize the late-stage flow conversion.
• Chemical Yield Stability: The $D_s^+/D^0$ yield ratio and its $q_2$ double ratio ($R_{q_2}$) remain close to unity across all selections. This confirms that the observed flow splitting is a dynamical flow effect driven by formation time, not a modification of chemical yields or coalescence fractions due to event shape.

Significance and Claims
The paper claims that $\Delta v_2(D^0 - D_s^+)$ under Event-Shape Engineering and the response slope $\chi$ constitute complementary differential probes of the space-time structure of charm hadronization near the QCD transition temperature.

The primary significance lies in establishing that:

1. Sequential hadronization naturally reverses the sign of the $D^0$–$D_s^+$ flow splitting relative to the conventional simultaneous baseline.
2. ESE isolates the hadronization-time response from the geometry-driven enhancement, providing a "two-knob" framework (geometry via $q_2$, time via formation temperature) that is not limited to open charm but can be generalized to other species.
3. The hierarchy $\chi(D^0) > \chi(D_s^+)$ provides a falsifiable, quantitative prediction. If future ALICE data analysis finds $\chi(D^0) \approx \chi(D_s^+)$, the sequential picture would be disfavored.
4. The results validate the semi-central regime as the cleanest environment to distinguish these mechanisms, as the interplay between QGP lifetime and initial eccentricity maximizes the observable signal.

The authors emphasize that these findings rely on the specific theoretical framework of sequential hadronization introduced in their previous work [19] and do not claim to have experimentally confirmed the effect, but rather provide the theoretical tools and specific observables required for such a test.