Measurement of isolated prompt photon production in pp and p-Pb collisions at the LHC

The ALICE Collaboration measured the inclusive production cross section of isolated prompt photons in pp and p-Pb collisions at various LHC energies, observing a low-transverse-momentum suppression in p-Pb collisions consistent with nuclear shadowing effects predicted by perturbative QCD.

The Gist

Imagine the universe as a giant, high-speed racetrack. At CERN's Large Hadron Collider (LHC), scientists smash particles together at speeds close to the speed of light to see what happens when the "cars" (protons and atomic nuclei) collide.

This paper is a report from the ALICE experiment, one of the teams watching these crashes. Their goal? To understand how the "traffic" (particles) behaves when a single car crashes into a massive truck (a lead nucleus) versus when two cars crash into each other.

Here is the story of their discovery, broken down into simple concepts.

1. The Star of the Show: The "Flashlight" Photon

In these collisions, thousands of particles are created. Most of them are like debris from a car crash—messy, interacting with everything around them. But sometimes, a photon (a particle of light) is born directly from the initial crash.

Think of this photon as a super-fast flashlight that flies out of the crash zone without hitting anything.

• Why is it special? Because light doesn't interact with the "traffic jam" of other particles, this flashlight gives us a clear, unobstructed view of the very first moment of the collision. It tells us what the "engine" (the partons inside the protons) was doing before the crash.
• The Problem: Sometimes, other particles (like neutral pions) decay into photons that look just like our special flashlight. It's like trying to spot a specific flashlight in a stadium full of people holding flashlights.
• The Solution (Isolation): To find the real flashlight, the scientists used a rule called "Isolation." They only counted a photon if it was flying alone in a small cone. If there was too much "debris" (other particles) nearby, they ignored it. This ensured they were only looking at the clean, direct flashes from the crash.

2. The Experiment: Cars vs. Trucks

The team compared two types of crashes:

1. Car vs. Car (pp collisions): Two protons smashing together. This is the "control group" or the baseline.
2. Car vs. Truck (p–Pb collisions): A proton smashing into a Lead nucleus (which is like a heavy truck made of 208 protons and neutrons stuck together).

They did this at two different speeds (energies): a "fast" speed (8.16 TeV) and a "medium" speed (5.02 TeV).

3. The Big Question: Does the Truck Change the Crash?

When a proton hits a lead nucleus, does the "stuff" inside the lead nucleus (the gluons and quarks) change how the crash happens?

Scientists have a theory called Gluon Shadowing. Imagine the lead nucleus is a dense forest. If you shine a light into the forest, the trees (gluons) might block or "shadow" some of the light, making it harder to see deep inside.

• The Prediction: At lower speeds (lower energy), the proton should probe the "deep forest" of the lead nucleus. The scientists expected to see a suppression—meaning fewer isolated photons coming out of the truck crash than the car crash, because the "forest" was blocking the view.
• The High-Speed Expectation: At very high speeds, the proton should zoom past the trees so fast that the shadowing effect disappears, and the results should look just like the car-vs-car crash.

4. The Results: What Did They Find?

The ALICE team measured the number of "isolated flashlights" coming out of these crashes.

• The High-Speed Zone (High Energy, High Momentum): When the photons had a lot of energy (moving fast), the number of flashlights in the "Car vs. Truck" crash was exactly the same as in the "Car vs. Car" crash. The lead nucleus didn't seem to block anything here. This matches the predictions of standard physics (Quantum Chromodynamics).
• The Low-Speed Zone (Lower Energy, Lower Momentum): When they looked at photons with lower energy, they saw something interesting. There were fewer flashlights coming out of the truck crash than expected—about 20% fewer.
  • The Significance: This isn't a huge difference, but it's statistically noticeable (about 2 to 2.3 times more likely to be real than a random fluke).
  • The Meaning: This confirms the "Gluon Shadowing" idea. The dense "forest" of the lead nucleus is indeed blocking or shadowing the view of the inner workings at lower energies.

5. Why Does This Matter?

This paper is like a new, high-resolution map of the inside of an atomic nucleus.

• New Territory: Previous experiments could only see the "edge" of the forest. This experiment, by using lower energy thresholds, allowed them to see deeper into the "forest" (down to very small distances inside the nucleus).
• Testing the Theory: The fact that the data matches the complex mathematical predictions (using something called "Parton Distribution Functions") gives scientists confidence that their understanding of how matter is built is correct.
• Future Impact: By understanding how the "shadow" works, scientists can better refine their models of the universe's fundamental building blocks. It helps them understand how the "glue" (gluons) holds the nucleus together.

Summary Analogy

Imagine you are trying to count how many people are inside a crowded room (the lead nucleus) by shining a laser pointer through it.

• If you shine the laser from far away (high energy), the beam is so focused and fast that it passes through the crowd without hitting anyone. You see exactly what you expect.
• But if you shine the laser from closer up (lower energy), the beam hits the people in the front row (the gluons). They block some of the light.
• The ALICE team found that at the "closer" distance, the light was indeed dimmer (suppressed) by about 20%, proving that the crowd is dense enough to cast a shadow. This confirms our theories about how the crowd is arranged.

In short: The universe behaves exactly as our best theories predict, even when we look at the densest matter we can create in a lab.

Technical Summary

1. Problem and Motivation

The primary goal of this research is to understand the dynamics of partons (quarks and gluons) within nuclear matter. While inclusive particle and jet production are standard probes, they are complicated by final-state interactions (energy loss in the hot medium). Photons, interacting only electromagnetically, serve as a "clean" probe because they escape the collision zone without significant final-state modification.

Key Scientific Challenges:

• Initial State Effects: In proton-nucleus (p–Pb) collisions, the production rates of prompt photons can be modified by nuclear effects in the initial state, such as gluon shadowing (modification of parton distribution functions inside the nucleus) and saturation.
• Low-$x$ Reach: Previous measurements at the LHC had limited reach to low Bjorken-$x$ (momentum fraction of the parton), where shadowing effects are expected to be most significant.
• Background Suppression: Distinguishing "prompt" photons (produced directly in hard scatterings) from "decay" photons (from $\pi^0 \to \gamma\gamma$) and "fragmentation" photons (from parton fragmentation) is difficult. The paper aims to minimize fragmentation contributions to isolate the direct photon signal.

2. Methodology

Experimental Setup:

• Detector: The ALICE detector at the Large Hadron Collider (LHC).
• Collision Systems and Energies:
  • pp collisions: $\sqrt{s} = 8$ TeV (2012 data) and $\sqrt{s} = 5.02$ TeV (2017 data, referencing previous work).
  • p–Pb collisions: $\sqrt{s_{NN}} = 5.02$ TeV (2013 data) and $\sqrt{s_{NN}} = 8.16$ TeV (2016 data).
• Kinematic Range: Measurements performed at midrapidity ($|y| < 0.7$) with transverse momentum ($p_T$) ranges of $12 < p_T < 60$ GeV/c (for 5.02 TeV) and $12 < p_T < 80$ GeV/c (for 8.16 TeV). This extends the low-$x$ reach in Pb nuclei to $x \approx 2.9 \times 10^{-3}$.

Reconstruction and Identification:

• Photon Candidate Selection: Photons are reconstructed using the Electromagnetic Calorimeter (EMCal) and the DiJet Calorimeter (DCal). Clusters are formed from energy depositions in towers.
• Shower Shape: A variable $\sigma^2_{long}$ (length of the long axis of the shower ellipse) is used to distinguish narrow single-photon showers from broader $\pi^0/\eta$ decay showers. Cuts: $0.1 < \sigma^2_{long} < 0.3$.
• Charged Particle Veto (CPV): Tracks from the Inner Tracking System (ITS) and Time Projection Chamber (TPC) are extrapolated to the calorimeter to reject clusters associated with charged particles. (Note: CPV was not used for the 5.02 TeV p–Pb data to maintain consistency with the reference pp dataset where TPC was not fully active).

Isolation Criteria:
To suppress fragmentation photons and decay backgrounds, an isolation momentum ($p^{iso, ch}_T$) is calculated within a cone of radius $R=0.4$ around the photon:
$$p^{iso, ch}_T = \sum_{track \in \Delta R < 0.4} p^{track}_T - \rho \times \pi \times 0.4^2$$

• Only charged particles are included to maximize acceptance and reduce autocorrelation with neutral decays.
• The underlying event (UE) density $\rho$ is subtracted using either the perpendicular cone method or the $k_T$ jet-finding algorithm.
• Selection: Candidates are selected if $p^{iso, ch}_T < 1.5$ GeV/c.

Purity and Efficiency Corrections:

• Efficiency: Determined via PYTHIA 8 simulations ($\gamma$-jet events) embedded into p–Pb background events (DPMJET).
• Purity (Signal Extraction): Two data-driven methods were employed to determine the fraction of prompt photons in the selected sample:
  1. ABCD Method: Used for 8 TeV pp and 8.16 TeV p–Pb. It uses a 2D distribution of shower shape vs. isolation momentum to extrapolate background from control regions.
  2. Template Fit Method: Used for 5.02 TeV p–Pb. Fits the shower shape distribution using signal (MC) and background (data sideband) templates.

Theoretical Comparison:
Results are compared to Next-to-Leading Order (NLO) perturbative QCD (pQCD) calculations using the JETPHOX program.

• PDFs: NNPDF4.0 for protons and nNNPDF3.0 (or nCTEQ15HQ) for nuclear PDFs.
• Fragmentation: BFG II fragmentation functions.

3. Key Contributions

1. Extended Low-$x$ Reach: The measurement probes gluon densities down to $x \approx 2.9 \times 10^{-3}$, extending the reach in Pb nuclei by nearly a factor of two compared to previous p–Pb measurements.
2. New Cross Sections: Provides precise inclusive production cross sections for isolated prompt photons in pp (8 TeV) and p–Pb (5.02 and 8.16 TeV) collisions.
3. Nuclear Modification Factor ($R_{pA}$): Presents the first high-precision $R_{pA}$ measurements for isolated prompt photons at these energies, specifically focusing on the low-$p_T$ region where nuclear effects are expected.
4. Methodological Consistency: Demonstrates the consistency of purity extraction methods (ABCD vs. Template Fit) and validates the use of charged-track isolation in heavy-ion environments.

4. Results

Cross Section Measurements:

• The measured cross sections for both pp and p–Pb collisions are in good agreement with NLO pQCD calculations using recent PDFs and nPDFs.
• The data validates the theoretical description of isolated prompt photon production across a wide range of energies.

Nuclear Modification Factor ($R_{pA}$):
$R_{pA}$ is defined as the ratio of the p–Pb cross section to the pp cross section, scaled by the number of binary collisions.

• High $p_T$ ($> 20$ GeV/c): $R_{pA}$ is consistent with unity, indicating no significant nuclear modification at high momentum.
• Low $p_T$ ($< 20$ GeV/c):
  • A suppression ($R_{pA} < 1$) of up to 20% is observed.
  • Significance:
    • At $\sqrt{s_{NN}} = 8.16$ TeV: Suppression has a significance of 1.8$\sigma$ for $R_{pA} < 1$ and 2.3$\sigma$ for a non-zero downward slope.
    • At $\sqrt{s_{NN}} = 5.02$ TeV: Suppression has a significance of 1.1$\sigma$ for $p_T < 14$ GeV/c.
• Comparison with Theory: The observed suppression trend (increasing as $p_T$ decreases) is consistent with NLO pQCD predictions incorporating gluon shadowing in the lead nucleus (nNNPDF3.0 and nCTEQ15HQ).

Comparison with ATLAS:
The results agree with ATLAS measurements in the overlapping $p_T$ range ($25 < p_T < 80$ GeV/c). However, ALICE extends the measurement down to 12 GeV/c, demonstrating superior capability for low-$p_T$ photon reconstruction due to its smaller material budget.

5. Significance

• Validation of Nuclear PDFs: The results provide strong constraints on nuclear Parton Distribution Functions (nPDFs), particularly in the shadowing regime (low $x$). The agreement with nPDF-based calculations supports the current understanding of how nuclear matter modifies parton densities.
• Initial State Probe: The observation of suppression at low $p_T$ (high $x$ in the nucleus, low $x$ in the projectile) confirms the presence of initial-state nuclear effects, specifically gluon shadowing, without the confounding factor of final-state energy loss.
• Future Constraints: The demonstrated sensitivity of isolated prompt photons to nuclear effects offers a promising independent probe for future global fits of nuclear PDFs, helping to reduce uncertainties in the shadowing regime which are critical for interpreting heavy-ion collision data (e.g., quark-gluon plasma studies).
• Precision Benchmark: The paper establishes a high-precision benchmark for isolated photon production in p–Pb collisions, resolving previous discrepancies and validating the use of NLO pQCD for these systems.