Hadron Spin Polarization Produced by a Dynamical Spin-Orbit Interaction

Abstract

The spin observables in high energy hadron reactions, such as spin polarization, analyzing power and spin depolarization, are of increasing interest for extracting precise information about the structure functions of the quark parton in hadrons 1) and even in leptons and photon. 2) Almost two decades ago, DeGrand and Miettinen 3) proposed a kinematic model to predict the sign and to estimate the order of magnitude of the spin observables produced in inclusive high energy hadron collisions. In their model, they assumed an empirical rule which can be briefly expressed as, “fast spin up and slow spin down.” This rule refers to the preferential direction of spin polarization of the produced hadrons in relation to the velocity of participating partons. From the momentum distribution functions of the quark parton in hadrons, it is known that the valence parton has a larger velocity while the sea parton has a smaller velocity. DeGrand and Miettinen found systematic occurrence in the empirical data of spin polarization such that the fast (slow) parton(s) mostly carries up (down) spin; that is the fast (slow) parton(s) preferentially produces hadrons with spin-up (spin-down). They also showed that a spin-orbit interaction results from this rule through Thomas precession of the spin carrier parton(s), and they concluded that this is the cause of the observed spin asymmetries. Recently, we proposed a general formulation 4) to describe the spin observables of hadrons inclusively produced in high-energy hadron collisions, and we showed that our model, called the microscopic quark recombinaation (QRC) model, is successful in reproducing the experimentally obtained spin observables. In our formulation, we do not assume the DeGrand-Miettinen (DM) empirical rule, but, rather, we start from a relativistic expression for the transition amplitude of the participating quark parton’s distribution functions in hadrons. The QRC formulation provides spin observables as functions of the transverse momentum PT and Feynman’s variable xF. This is an advantage of the model, and it is extremely useful for extracting