Suppression Of High Transverse Momentum Hadrons At RHIC And LHC

The Quantum Chromodynamics (QCD) predicts the existence of Quark-gluon plasmas (QGP), which is a new phase of hadron matter at high temperature and high density where the quark degrees of freedom normally confined within hadrons are lib-erated. QGP is thought to exist in the very early universe after the Big Bang. Thus studying the properties of QGP is very important for our deeper understanding of the evolution of universe as well as the physics of QCD. Now it’s believed that QGP has been created in the experiments of heavy ion collisions in RHIC (Relativistic Heavy-Ion Collider) at BNL and LHC (Large Hadron Collider) at CERN. There are mainly two kinds of observables to study the matter created in heavy ion collisions, the col-lective flow of soft hadrons and the suppression of large transverse momentum spectra of hard hadrons (jet quenching). This thesis focuses on the later one.In heavy-ion collisions, high pt parton produced by hard scattering will pene-trate the formed QGP and lose a large amount of energy through multiple scatterings with thermal partons in QGP, which results in the suppression of large transverse mo-mentum spectra of hadrons. In perturbative QCD the transverse momentum spectra of hadrons can be calculated by the convolution of distribution functions of colliding partons, cross-section of two partons and the fragmentation function which represents the probability of the final parton fragmentates into hadrons. Owing to the fact that jet quenching mainly stems from final interactions, it is natural to introduce the modified fragmentation functions (mFF) to express the softening of the hadron spectra.In the process of deep inelastic scattering (DIS) of a large nucleus, the outgoing parton struck by the virtual photon will also lose its energy when passing though the remainder cold nuclear matter. In higher-twist approach, the interactions of the hard parton with the medium will induce additional gluon bremsstrahlung which will mod-ify the splitting functions in vacuum. Then mFF can be solved out by the evolution of the modified DGLAP (mDGLAP) equations which were derived in DIS. In Ref.[45] the mDGLAP equations were solved numerically for the first time and the application in DIS exhibits its validity in describing the experimental data at HERMES.In this thesis we extend the framework form DIS to heavy-ion collisions. We realize that the initial condition which is the input to solve mDGLAP equations is very important. If we use the naive initial condition in Ref.[45] we can’t explain the experimental data at RHIC.Other groups don’t obtain mFF by solving mDGLAP equations. They get mFF by the convolution of fragmentation functions in vacuum with the quenching weight which is calculated by the induced gluon spectra. We can also calculate induced gluon spectra and quenching weight in higher-twist approach. Using the initial condition by convolution we can describe the experimental data in DIS at HERMES and heavy-ion collisions at RHIC.After the data at LHC released, people find there are some difficulties to fit the data well if we simply extend the calculation at RHIC to LHC. Usually, we will over-rate the suppression at large pT. It’s realized latter that a reduced coupling constant at LHC can explain the problem. After we using a running coupling constant instead of a fixed one, we can indeed fit the data at RHIC and LHC simultaneously.Moreover, we find a new mechanism which can settle the problem as well. The higher-twist modification mainly comes from the contribution of small Q2because the higher-twist term is suppressed by1/Q2, where the Q2in mDGLAP equations can be served as the virtuality of the final parton. Parton with more energy needs more time to lose its virtuality to Q2than the parton with less energy. That means at a given Q2the parton with more energy locates at the later period of QGP. Because the QGP dilutes quickly, the parton with more energy will have less interactions with the medium. We develop a simple model of parton shower in medium base on the tactics of PYTHIA to get the information of the virtuality of the final parton on its path. This information can modify our calculations on medium induced splitting function and help us to explain the experimental data at LHC.