Methods For The Analysis Of Rapidity Correlations In High-energy Heavy-ion Collisions

Hot and dense matter is considered to be produced in high-energy heavy-ion collisions,and maybe quark-gluon plasma?QGP?is considered to be created in the reaction.In quantum chromodynamics?QCD?,there is no single and separate quark without a hadron.Therefore,QGP can not be investigated directly in experiments which we can get only some information such as types and velocities of hadrons after the interactions.The experimental results are used indirectly to investigate the physical process of early stages of the interactions.Rapidity is a variable used for describing the longitudinal velocity of a hadron,and correlation among nuclear matter may survive through interactions.Therefore,two-particle rapidity correlation was considered to be an important probe of physical process of early stages.In this thesis,emphases are put on the influence of centrality fluctuation on rapidity correlation,and some methods are raised to reduce or remove the bias from centrality fluctuation.At first,the correlations of charged particle multiplicities between forward-backward rapidity windows?”FB correlation” for short?are calculated with a method called FB_average,and the relationship between correlation strength and centrality window is discussed.Positive FB correlation strengths bfbwere observed obviously in Au+Au collisions at?sN N= 200 GeV.For the most central collisions,strong correlations are observed.We argue that the obviously positive correlations may be caused by the mixing of events of different centralities in centrality windows.A Monte-Carlo model with the particles emitting randomly and independently is built,and positive correlations are seen,even there are no correlations in the model.FB correlation strength branunder the random event assumption is deduced for a centrality window.bfband branof AMPT model are similar,and the similarity is more obvious for bigger partonic scatter sections.In sum,it is suggested that FB correlation strengths are dominated by centrality-window widths,and we argued that the correlations from early stage of collisions are overwhelmed by thermalization.Then,we discuss the influence on FB correlation measurement of centrality fluctuation caused by statistical error of reference multiplicity.For removing or reducing the influence of centrality fluctuation,a method called FBprofilewas raised.The core idea of this method is that forward multiplicity N_f,backward multiplicity N_b,N_fN_b,and N2 fare considered to be functions of reference multiplicity,and are fitted to remove or reduce the influence of centrality fluctuation.It is argue that the essence of this method is similar with that FB correlation strengths calculated in some narrow reference multiplicity windows,and an average is made with them in a centrality window.A Monte Carlo model without any correlation is used,and some positive correlations are seen with this method.It is seen that the FB correlation strengths decrease with the increasing reference-window width and depend on the fitting range.In sum,we argue that some positive correlations are added because of centrality fluctuation caused by statistical error of reference multiplicity,and the results are influenced by fitting range greatly.Then,for removing or reducing the influence of centrality fluctuation,a new method called relative multiplicity method(FB_relative)is raised.For a centrality window,relative multiplicities are introduced so that events with different centralities are unified.And,the biases of reference multiplicities are modified.On the other hand,a Monte Carlo model with adjustable correlation is made.The correlation strengths of the FB_average,FBprofile,and FB_relativemethods are compared with this model.It is seen that the FB_relativemethod extract the set correlations successfully,and the results of the other two methods are all more positive.With this method,the FB correlations of HIJING model at RHIC energy are calculated,and the correlations are very weak and nearly independent of centrality.At last,the method for calculating two-particle rapidity correlation function?2PC?is discussed.Beside the forward-backward multiplicity correlation strength,two-particle rapidity correlation function is also an important probe reflecting the relationship of particles.In addition,asymmetry of the two-ion collision was investigated by decomposing2 PC.But the centrality fluctuation is still a problem in the calculation.Two new methods denoted by Cdefinitionand Crelativeare raised to remove the influence.In Cdefinitionmethod,the correlation function is deduced from the definition of 2PC,which is applicable for a centrality window?not wider than 5%?.In Crelativemethod,relative multiplicity is introduced to remove the influence of centrality fluctuation just like the core idea of FB_relative method.A granular source model is built to simulate the event-by-event fluctuation of rapidity distribution,and the methods are tested with this model.With these two methods,variances of parameters are extracted successfully by decomposing the correlation functions.