Estimation of initial state structures in high energy heavy-ion collisions using Principal Component Analysis (PCA)

TL;DR

This paper applies Principal Component Analysis to study initial state structures in high-energy heavy-ion collisions, demonstrating sensitivity of PCA modes to spatial clustering in the initial collision zone.

Contribution

It introduces a formalism for implementing spatial clusters at the partonic level in the AMPT model and analyzes their impact using PCA on various particle distributions.

Findings

PCA modes are sensitive to initial spatial clustering.

Eigenvalues vary with collision centrality.

Clustering affects the distributions of charged particles.

Abstract

In high-energy heavy-ion collisions, structures in the initial collision zone are a matter of intense investigation, both from theory and experimental points of view. A large number of models have been developed to represent the initial state of the collision including Glauber model, Colour Glass Condensate (CGC) among others. Another important aspect of the study is to investigate proper observables that will be sensitive to the initial collision zone. In this work, we have discussed a formalism to implement the spatial clusters at the partonic level in the string melting version of the AMPT model for PbPb collisions at $\sqrt{s_{NN}}$ = 200 GeV. These clusters are then propagated through the AMPT hadronization scheme. The Principal Component Analysis (PCA) has been used on the $\eta$, $\phi$ and $p_T$ distributions of the produced charged particles and the eigenvalues have been compared before and after the implementation of the clustering. It is found that for all these three different distributions, all the prominent PCA modes have shown sensitivity to the clustering. A centrality dependent study has also been performed on those eigenvalues.