Inclusion of specific effects associated with constituent binding in hadronic wave functions is shown to lead to important nonscaling, nonfactorizing 1/Q2 contributions to cross sections for semi-inclusive deep-inelastic scattering, the Drell-Yan process, and other hard scattering reactions. These 1/Q2 higher-twist terms are predicted to be dominant in well defined kinematic regions such as large x and/or large z. They provide angular distributions typical of longitudinally polarized virtual photons and W's, including sin2theta terms in meson induced Drell-Yan processes and in e/sup +/e− .-->. .pi. X, as well as unusual (1-y) terms in deep-inelastic scattering. Calculations are also presented of the quark structure functions of the pion q/sub .pi./(x, Q2) and for the quark to pion fragmentation function D/sub .pi.(z, Q2). Predictions are made for the azimuthal angle dependence of the cross sections for .pi. N .-->. .mu. anti .mu. X and lN .-->. l'.pi. X.

In this paper, we study the transverse momentum dependent (TMD) evolution of the Collins azimuthal asymmetries in e+e- annihilations and semi-inclusive hadron production in deep inelastic scattering (SIDIS) processes. All the relevant coefficients are calculated up to the next-to-leading logarithmic (NLL) order accuracy. By applying the TMD evolution at the approximate NLL order in the Collins- Soper-Sterman (CSS) formalism, we extract transversity distributions for u and d quarks and Collins fragmentation functions from current experimental data by a global analysis of the Collins asymmetries in back-to-back di-hadron productions in e+e- annihilations measured by BELLE and BABAR Collaborations and SIDIS data from HERMES, COMPASS, and JLab HALL A experiments. The impact of the evolution effects and the relevant theoretical uncertainties are discussed. We further discuss the TMD interpretation for our results, and illustrate the unpolarized quark distribution, transversity distribution, unpolarized quark fragmentation and Collins fragmentation functions depending on the transverse momentum and the hard momentum scale. Finally, we give predictions and discuss impact of future experiments.

Although perturbative QCD has progressed much in the last few years, the notion of diquarks is still useful and vital and many short cuts induced by diquarks can still be very efficient for simulating nonperturbative effects in the intermediate energy region (from a few GeV to some 10-20 GeV), where a whole new generation of machines is being planned and will, hopefully, become operative in the near future.

We study the QCD evolution for the twist-three quark-gluon correlation functions associated with the transverse momentum odd quark distributions. Different from that for the leading twist quark distributions, these evolution equations involve more general twist-three functions beyond the correlation functions themselves. They provide important information on nucleon structure, and can be studied in the semi-inclusive hadron production in deep inelastic scattering and Drell-Yan lepton pair production in pp scattering process.