Université de Lyon Faculté des Sciences et Technologies École Doctorale PHAST THÈSE DE DOCTORAT - PDF

Description
Université de Lyon Faculté des Sciences et Technologies École Doctorale PHAST THÈSE DE DOCTORAT Présentée par Raphaël DUPRÉ pour obtenir le titre de Docteur ès Sciences de l Université de Lyon Specialité

Please download to get full document.

View again

of 150
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Literature

Publish on:

Views: 47 | Pages: 150

Extension: PDF | Download: 0

Share
Transcript
Université de Lyon Faculté des Sciences et Technologies École Doctorale PHAST THÈSE DE DOCTORAT Présentée par Raphaël DUPRÉ pour obtenir le titre de Docteur ès Sciences de l Université de Lyon Specialité : PHYSIQUE NUCLÉAIRE Sujet : Quark Fragmentation and Hadron Formation in Nuclear Matter Directeur de thèse: Guy CHANFRAY Institut de Physique Nucléaire de Lyon, Université de Lyon Encadrant local: Kawtar HAFIDI Physics Division, Argonne National Laboratory soutenue publiquement le 9 novembre 2011 à Villeurbanne Jury : Président : Jean-Yves GROSSIORD - IPNL (Villeurbanne, France) Directeur : Guy CHANFRAY - IPNL (Villeurbanne, France) Encadrant local: Kawtar HAFIDI - ANL (Argonne, IL, USA) Rapporteur : William BROOKS - UTFSM (Valparaíso, Chile) Rapporteur : Michel GUIDAL - IPNO (Orsay, France) Examinateur : François ARLEO - LAPTH (Annecy-le-Vieux, France) This thesis was prepared principally in the Medium Energy Physics Group, Physics Division Argonne National Laboratory 9700 South Cass Avenue, Argonne, IL, USA and occasionally in Hall B, Jefferson Lab Jefferson Avenue, Newport News, VA, USA and Institut de Physique Nucléaire de Lyon Université de Lyon, Université Claude Bernard Lyon 1, IN2P3, CNRS Domaine scientifique de la Doua 4, rue Enrico Fermi, Villeurbanne, France [L]e savant ne choisit pas au hasard les faits qu il doit observer. Il ne compte pas des coccinelles, comme le dit Tolstoï, parce que le nombre de ces animaux, si intéressants qu ils soient, est sujet à de capricieuses variations. Il cherche à condenser beaucoup d expérience et beaucoup de pensées sous un faible volume, et c est pourquoi un petit livre de physique contient tant d expériences passées et mille fois plus d expériences possibles dont on sait d avance le résultat. [...] Le savant n étudie pas la nature parce qu elle est utile ; il l étudie parce qu il y prend plaisir et il y prend plaisir parce qu elle est belle. Si la nature n était pas belle, elle ne vaudrait pas la peine d être connue, la vie ne vaudrait pas la peine d être vécue. Je ne parle pas ici, bien entendu, de cette beauté qui frappe les sens, de la beauté des qualités et des apparences ; non que j en fasse fi, loin de là, mais elle n a rien à faire avec la science ; je veux parler de cette beauté plus intime qui vient de l ordre harmonieux des parties, et qu une intelligence pure peut saisir. C est elle qui donne un corps, un squelette pour ainsi dire aux chatoyantes apparences qui flattent nos sens, et sans ce support, la beauté de ces rêves fugitifs ne serait qu imparfaite parce qu elle serait indécise et toujours fuyante. Au contraire, la beauté intellectuelle se suffit à elle-même, et c est pour elle, plus peut-être que pour le bien futur de l humanité, que le savant se condamne à de longs et pénibles travaux. Science et méthode Henri Poincaré 4 Fragmentation des Quarks et Formation des Hadrons dans la Matière Nucléaire Résumé: La formation des hadrons est, dans le cadre de la théorie quantique de couleur (QCD), un processus non-perturbatif ; cette caractéristique entraîne d importantes difficultés théoriques. C est pourquoi, les mesures expérimentales de fragmentation dans différents noyaux sont une nécessité afin d obtenir des progrès tangibles dans la compréhension des mécanismes de formation des hadrons. La thèse commence par les bases théoriques nécessaires à une telle approche, suivies des principaux modèles qui lui sont associés. La thèse se poursuit par l analyse de données de Jefferson Lab obtenues à l aide d un faisceau d électrons de 5 GeV incident sur différentes cibles ( 2 H, C, Al, Fe, Sn et Pb). Les produits de la réaction sont mesurés avec le spectromètre CLAS. Les principaux résultats de cette expérience sont : (a) l analyse multi-dimensionnelle des observables mesurées, qui permet une meilleure confrontation avec les modèles théoriques et l extraction d informations temporelles sur la fragmentation, et (b) l observation d une atténuation hadronique non-linéaire en fonction du rayon du noyau cible. Dans une partie plus théorique, le générateur d événements PyQM, développé dans le but de reproduire les données de la collaboration HERMES, est présenté. Les résultats sont mitigés, en effet la base théorique utilisée ne semble pas s appliquer au cas étudié, néanmoins certaines caractéristiques des données sont reproduites permettant de comprendre leurs origines parfois inattendues. Enfin, les possibilités d expériences futures, à Jefferson Lab et dans un collisionneur ion-électron (EIC), sont explorées. Mots-clef: Fragmentation, hadronisation, QCD, Jefferson Lab, CLAS, noyau, Monte-Carlo, perte d énergie des quarks, collisionneur électron-ion, EIC. 5 Quark Fragmentation and Hadron Formation in Nuclear Matter Summary: The hadron formation is, in the framework of the quantum chromodynamics theory (QCD), a non-perturbative process; this characteristic leads to important theoretical challenges. This is why experimental measurements of fragmentation in nuclei are a necessity in order to obtain substantial progress in our understanding of the mechanisms of hadron formation. The thesis begins with the introduction of theoretical background, followed by an overview of theoretical models. The thesis continues with the analysis of Jefferson Lab data obtained with a 5 GeV electron beam incident on various targets ( 2 H, C, Al, Fe, Sn and Pb). The reaction products are measured with the CLAS spectrometer of Hall B. The main results are: (a) a multi-dimensional analysis of the measured observables, which permits a better confrontation with theoretical models and the extraction of temporal information on fragmentation, and (b) the observation of a non linear hadronic attenuation as a function of the target s nuclear radius. The PyQM event generator, developed to reproduce the data from the HERMES collaboration, is also presented. The results are ambivalent, the theoretical basis used does not seem to apply to the studied case, however, some characteristics of the data are reproduced allowing to understand their origin, which is sometimes unexpected. Finally, the possibilities for future experiments, at Jefferson Lab and at an Electron-Ion Collider (EIC), are explored. Keywords: Fragmentation, hadronization, QCD, Jefferson Lab, CLAS, nuclei, Monte-Carlo, quark energy loss, Electron-Ion Collider, EIC. Acknowledgments A thesis work is not only a scientific work, it is also a human experience and I would like to thank here all those who made it possible and such a great experience! I would like to thank first the many people who supervised me one way or another. Especially, Kawtar Hafidi who taught me day after day during these three years. But also Guy Chanfray for accepting to be my supervisor abroad, Alberto Accardi for his guidance on the PyQM and EIC projects and Stepan Stepanyan for sharing his knowledge on detectors during the preparation of the eg6 run. I also would like to thank the members of my jury, especially W. Brooks and M. Guidal who accepted to be referee of my thesis. A great part of this thesis is due to the great environment that is Argonne National Laboratory. Therefore, I would like to thank Roy Holt, Don Geesaman and all the MEP group for welcoming me and for their numerous contributions during my various presentations along the years. I would like to credit the important participation of many other students and post-docs (in particular Aji, Hayk, Hyupwoo, Lamiaa and Taisiya) who worked on the eg2 data and spent countless hours discussing analysis details. Special thanks to William Brooks who organized the meetings and who was endlessly patient with us. I want to acknowledge the many members of the université Claude Bernard de Lyon who made my student years so enjoyable. Of course my many professors, of mathematics (T. Fack and M. Kibler) and quantum physics (S. Fleck and G. Chanfray) in particular. But also the professors, BIATOSS and students with which I worked in the various committees of the university, especially L. Collet who encouraged me to pursue my studies in the USA. I want also to credit all the people who make Turbulence (spécialement Ahmed, Fabien, Kevin, Mirsal et Samia) and the BVE (spécialement Maryline et Patrice) such great organizations, which made me appreciate fully the university. I also need to express my gratitude to all the people who visited me in America and helped me keep the connection. Samia, (Mirsal) 2, Thomas, Alexia et Marion merci d être venu! et merci aussi à mes nombreux correspondants : Samia (encore!), Aurélie, Anne-Ségo et tout particulièrement Kathalyne! Also to the Moroccan team: Ahmed, Brahim, Lamiaa and Kawtar; who made the stay both at Argonne and JLab so enjoyable. Finally, I would like to thank all my family members who supported me during these years. Remerciement spécial à Alain, Anne, Christophe, Clara, Elisabeth, Enzo, Magalie et Nicolas qui se sont sacrifiés en traversant l Atlantique! Contents Résumé 4 Summary 5 Acknowledgments 7 Introduction 13 1 Processes and Observables Deep Inelastic Scattering The Hadronization Process Hadronization in Vacuum Hadronization in Nuclei Observables Motivations Conclusion Theoretical Models Introduction Hadron Absorption Parton Energy Loss Medium Modified Fragmentation Functions Conclusion Overview and Interpretation of Existing Data Introduction Early Results SLAC Results EMC Results Fermi Lab Results Conclusion Recent Results HERMES Multiplicity Ratios HERMES Transverse Momentum Broadening HERMES Two Hadrons Multiplicity Ratio JLab Hall C Results Conclusion 10 Contents 4 PyQM Monte-Carlo Generator Presentation Technical Description of PyQM The Hard Scattering Quenching Weights Calculation Quenching Weights Implementation The Fragmentation Results Fermi-Momentum HERMES Description Conclusions The Hall B of Jefferson Laboratory The Accelerator The CEBAF Large Acceptance Spectrometer Generalities Drift Chambers Scintillator Counters Cherenkov Counters Electromagnetic Calorimeter Data Analysis Introduction Particle Identification Electron Identification π Identification π + Identification Target Determination Data Quality Extraction of Multiplicity Ratio and ΔP Method Preliminary Results Corrections Acceptance Correction Radiative Correction Isospin Correction Systematic Uncertainties Quality of the Detection Target Reconstruction Acceptance Normalization Error Contents 11 7 Results and Discussions Multiplicity Ratio A Dependence Cronin Effect ν Dependence z Dependence Q 2 Dependence ϕ h dependence Transverse Momentum Broadening A Dependence ν Dependence z Dependence Q 2 Dependence Future Experiments Introduction The CLAS12 Experiment The Electron Ion Collider Hadronization at EIC Parton Energy Loss Heavy Quarks Conclusion Conclusion 137 Bibliography 139 Introduction Quarks and gluons, namely partons, are confined inside hadrons and cannot be found isolated. This experimental fact known as confinement is tightly related to hadronization, the process by which partons transform into hadrons. Indeed, the strong force provokes the production of new hadrons when one tries to isolate a parton. This force is described by quantum chromo-dynamics (QCD) and had great successes in the perturbative regime since the beginning of its development in the 1960 s. However, at low energy, QCD cannot be treated perturbatively making reliable calculations very challenging, in particular for dynamics processes such as hadronization. In this regime, experimental measurements are, therefore, an important input to guide and constrain models. Hadrons are formed on distances of few femtometers, making the nucleus the best tool to study the space-time properties of hadronization. Indeed, using electron deep inelastic scattering (DIS) on nuclei allows to produce quarks with known kinematics, in a static medium with well understood properties. Comparing hadron production in light and heavy nuclei is equivalent to compare hadronization in vacuum with hadronization in nuclear medium. As the struck parton evolves through different stages, its interaction with the medium changes, therefore, the hadronization dynamics can be deduced from the variations of the physical observables, as function of the kinematics and the nucleus size. However, the intermediate stages of hadronization are not known a priori and the interpretation of the data is, most of the time, a challenge. Experimental investigations of hadronization, using DIS, started in the 1970 s, but most of the early results lacked statistical precision and allowed only qualitative interpretations. During the last decade, the HERMES collaboration published results with significant improvements, both in term of statistics and hadron separation. Consequently, a clearer picture of the hadronization process emerged and many models were excluded, but this was not enough and some very different models remain. The CLAS collaboration data, presented in this thesis, offer great statistics and a large variety of nuclear targets. With this new experiment, the goal is to provide stringent tests of the models and, therefore, advance our understanding of hadronization dynamics. Motivated by the recent improvements of the measurements quality, theorists improved their models by including new nuclear effects. This increasing sophistication favors the use of full Monte-Carlo simulation, to simplify the comparison with results presented in multi-dimensional bins. We developed such a simulation, called PyQM, the principal objective being to connect 14 Introduction traditional nuclear physics with relativistic heavy-ion collision physics. Such a comparison was made possible by recent theoretical developments, which link hadronization effects to the properties of the medium. Thus, we can make direct comparison between hot nuclear matter, such as Quark-Gluon Plasma (QGP), and normal nuclear matter, if we can provide a model fitting both kinds of data. Also, the effects observed in the PyQM simulation can help to interpret experimental data and understand the origin of certain observed features. The thesis is organized as follow. In chapter 1, the necessary background and the physics motivations are introduced, in chapter 2, the theoretical models are reviewed, and, in chapter 3, they are confronted to published results. The Monte-Carlo simulation of hadronization in nuclei, PyQM, is presented in chapter 4. Chapter 5 is an overview of the apparatus used in the CLAS experiment, it is followed, in chapter 6, by the analysis of the data and, in chapter 7, by the results, which are presented and discussed. Perspectives for future experiments are discussed in chapter 8 and, finally, a conclusion will summarize the results disclosed in this thesis. Chapter 1 Processes and Observables 1.1 Deep Inelastic Scattering To trigger the hadronization of a quark, a hard QCD process needs to be involved, in this thesis we concentrate on deeply inelastic scattering (DIS). Its general form is (k) + n(p) (k ) + X with a lepton and n a nucleon. Here, we will treat only the charged leptons and because the energy level is always much smaller than M Z or M W, photons are mediating the interaction 1. Therefore, this is a pure electromagnetic interaction and only charged constituent of the hadron target i.e. quarks are probed. As a consequence, the DIS process treated in this thesis is of the form shown in figure 1.1. Figure 1.1: Leading-order Feynman diagram describing DIS on a nucleon with photon exchange. The relevant variables for inclusive 2 measurements of DIS are: ˆ the 4-momentum transfer between the lepton and the nucleon squared Q 2 = q 2, ˆ the energy transfer ν = p q/m n (= E k E k in the target rest frame), ˆ the Bjorken scaling variable Bj = q2 2p q = Q2 2M n ν (dimensionless), 1 Very small effects remains from Z and W exchange, these are neglected here. 2 Inclusive means that only the scattered lepton is detected. If the scattered lepton and one hadron are measured, the reaction is semi-inclusive. If all products are identified, the reaction is called exclusive. 16 Chapter 1. Processes and Observables ˆ the ratio of the energy transferred to the total energy available y = p.q p.k (= ν E k in the target rest frame) (dimensionless), ˆ the mass of the total hadronic final state W = M 2 n Q2 + 2M n ν. Physically, Q 2 is the scale probed by the photon and W indicates the inelasticity of the reaction (W = M n is elastic). In the target rest frame, ν is the virtual photon energy and, in the infinite momentum frame, the Bjorken variable is the fraction of the nucleon momentum carried by the struck quark. By convention, in the case of nuclear targets, we calculate the kinematic variables considering that the target hadron is a nucleon (taken as the mean between proton and neutron). We do so because it facilitates comparisons and because, as DIS probes the nucleon constituents, it is more meaningful to keep it as reference. However, this choice is not perfect and several nuclear effects interfer, for example the EMC effect [Geesaman 1995], which introduces variations in the nucleon structure. The way around, in our hadronization studies, is to choose observables that are not sensitive to the modification of the initial state nucleon. With semi-inclusive measurements, it is possible to get information on the struck quark. The flavor content of the produced hadron gives information on the flavor of the struck quark and the hadron 4-momentum p h gives information on the quark kinematics and its hadronization. We list here the semiinclusive kinematic variables used in the thesis (see figure 1.2 for a graphic representation): ˆ the fraction z of the virtual photon energy transferred to the hadron: z = p h.p q.p (= E h in the target rest frame); ν ˆ the angle ϕ h between the leptonic plane, defined by the virtual photon and the outgoing lepton, and the hadronic plane, defined by the virtual photon and the detected hadron; ˆ the transverse momentum P of the hadron, defined in the target rest frame relative to the direction of the virtual photon; ˆ the Feynman scaling variable, F is the fraction of the maximum longitudinal momentum carried by the hadron F = P L P m L ˆ the Mandelstam variable t is the square of the 4-momentum transferred to the hadron: t = (q p h ) 2 ; ˆ the rapidity is defined as y = 1 2 ln E h+p L E h P L. Experimentally, the DIS events need to be separated from other processes like resonances and coherent production. ; In the case of resonances 1.1. Deep Inelastic Scattering 17 Figure 1.2: Semi-inclusive deep inelastic scattering on a nucleon. the photon is absorbed by the nucleon as a whole, exciting it to a state which will eventually decay. This leads, as in DIS, to the emission of one or more hadrons in the final state, but these are produced by internal, collective effects of the hadronic target. The spectroscopy of these resonances is extensively studied [Aznauryan 2011] but is not relevant to the study of hadronization and could contaminate our DIS sample. The scattering is expected to occur from quarks when the momentum transfer exceeds the QCD scale (Q 2 Λ QCD (300MeV) 2 ) and the energy of the final state exceeds all strong hadronic resonances (W 2 GeV). The quantum fluctuations of the virtual photon into hadronic states can also contaminate DIS samples. This is called diffractive production and comes mainly from vector mesons, it is concentrated at low Bj and high z ( 0.8). Theoretically, one can describe the DIS process easily using the factorization theorem (see [Brock 1995] for a complete review of DIS theory). Factorization permits to separate the DIS cross section in three independent parts: (a) the hard scattering cross section between the lepton and the parton, which, as it is an electromagnetic process,
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks