AMPT (model)

AMPT (model)
Name	AMPT
Full name	A Multi-Phase Transport
Developed by	[Zhang, Lin, Ko et al.]
Initial release	2004
Language	Fortran, C++
Operating system	Linux, macOS
License	research-use

AMPT (model) is a computational transport model designed for the simulation of relativistic heavy-ion collisions and the space-time evolution of partonic and hadronic matter. It integrates multiple theoretical modules to describe early parton production, partonic scatterings, hadronization, and hadronic rescatterings, and has been used by researchers studying results from Relativistic Heavy Ion Collider, Large Hadron Collider, Brookhaven National Laboratory, CERN, and related experimental collaborations. The code has influenced analyses in studies associated with Quark–Gluon Plasma, collective flow, jet quenching, and phenomenology compared across experiments such as STAR (detector), PHENIX, and ALICE (experiment).

Overview

AMPT was developed to bridge approaches used in transport simulations and hydrodynamic modeling for collisions involving nuclei such as Gold, Lead, Copper, and Uranium at beam energies spanning from the Super Proton Synchrotron regime through the LHC era. The model couples initial conditions inspired by perturbative Quantum Chromodynamics calculations and event generators like HIJING with partonic interaction modules and hadronic cascade codes like ART (model). Its architecture allows exploration of signals associated with deconfinement, chiral symmetry restoration, and the emergence of collective phenomena observed by collaborations including CMS, ATLAS, and PHENIX.

Theoretical Framework

AMPT’s theoretical basis synthesizes elements from perturbative Quantum Chromodynamics for minijet production, transport theory for parton cascades based on Boltzmann-type collision integrals, and phenomenological descriptions of hadronization inspired by the Lund string model. The model employs parton scattering cross sections derived from leading-order QCD matrix elements with screening masses that reflect medium effects studied theoretically by groups at institutions such as Institute of High Energy Physics (Chinese Academy of Sciences), Lawrence Berkeley National Laboratory, and RIKEN. Hadron transport follows principles used in cascade approaches developed alongside UrQMD, JAM (program), and RQMD, enabling comparisons with hadronic observables measured by detectors like PHOBOS and BRAHMS.

Model Components and Implementation

AMPT is organized into modules representing distinct physical stages: initial condition generation (via a HIJING-inspired routine), parton cascade (using a parton transport algorithm), hadronization (string melting or default string fragmentation), and hadronic rescattering (implemented via an ART (model)-like cascade). The string-melting option converts color strings into constituent partons to model deconfined matter and uses quark coalescence for hadron formation, a mechanism related to recombination ideas discussed in literature by authors affiliated with Columbia University, Duke University, and Tsinghua University. Practical implementations are coded in Fortran and optional C++ wrappers for parallel execution on clusters maintained by facilities such as National Energy Research Scientific Computing Center and Argonne National Laboratory.

Applications in Heavy-Ion Collisions

AMPT has been applied to interpret observables including anisotropic flow coefficients (v2, v3, v4) measured by STAR (detector), ALICE (experiment), and CMS, two-particle correlations investigated by PHENIX and ATLAS, and identified particle spectra compared to results from NA61/SHINE and SPS (accelerator) experiments. Studies using AMPT have addressed the ridge phenomenon observed in proton–lead collisions and proton–proton collisions at high multiplicity reported by CMS and ATLAS, as well as heavy-flavor transport by comparing with measurements from LHCb and heavy-ion groups at CERN. Model calculations have been used alongside hydrodynamic simulations developed by teams at McGill University, Brookhaven National Laboratory, and Lawrence Berkeley National Laboratory to disentangle initial-state fluctuations from final-state interactions.

Validation and Comparison with Data

Validation of AMPT entails systematic comparisons to data from experiments at RHIC and LHC across centrality, transverse momentum, rapidity, and particle species. Benchmarking exercises involve contrasts with event generators and transport models like HIJING, UrQMD, EPOS, and hydrodynamic hybrid frameworks produced by collaborations at Princeton University and University of Heidelberg. Parameters such as parton scattering cross section and hadronization schemes are tuned to reproduce elliptic flow from STAR (detector) and multiplicity distributions measured by ALICE (experiment). Sensitivity studies often reference theoretical developments from groups at MIT, Yale University, and University of Tokyo.

Limitations and Extensions

Known limitations include simplified treatments of in-medium parton energy loss relative to dedicated jet quenching formalisms developed at CERN and limited treatment of electromagnetic probes compared to approaches by researchers at Oak Ridge National Laboratory and GSI Helmholtz Centre for Heavy Ion Research. Extensions of AMPT have incorporated refined initial conditions inspired by Color Glass Condensate frameworks, event-by-event fluctuations influenced by work at Institut de Physique Théorique, and coupling to modern hadronization prescriptions developed at Rutgers University and University of California, Berkeley. Ongoing development efforts involve contributions from groups at Shanghai Jiao Tong University, Central China Normal University, and University of Science and Technology of China to address heavy-quark dynamics, viscosity matching, and GPU acceleration.

Category:Computational physics models Category:Heavy-ion physics