We present the first determination of the x-dependent pion gluon distribution from lattice QCD using the pseudo-PDF approach. We use lattice ensembles with 2+1+1 flavors of highly improved staggered quarks (HISQ), generated by MILC Collaboration, at two lattice spacings a ≈ 0.12 and 0.15 fm and three pion masses Mπ ≈ 220, 310 and 690 MeV. We use clover fermions for the valence action and momentum smearing to achieve pion boost momentum up to 2.29 GeV. We find that the dependence of the pion gluon parton distribution on lattice spacing and pion mass is mild. We compare our results from the lightest pion mass ensemble with the determination by JAM and xFitter global fits.
I. INTRODUCTIONThe lightest bound state in quantum chromodynamics (QCD), the pion, plays a fundamental role, since it is the Nambu-Goldstone boson of dynamical chiral symmetry breaking (DCSB). Studies of pion and kaon structure reveal the physics of DCSB, help to reveal the relative impact of DCSB versus the chiral symmetry breaking by the quark masses, and are important to understand nonperturbative QCD. Studying the pion parton distribution functions (PDFs) is important to characterize the structure of the pion and further understand DCSB and nonperturbative QCD. Currently, our knowledge of the pion PDFs is less than the nucleon PDFs, because there are fewer experimental data sets, especially for the seaquark and gluon distributions. The future U.S.-based Electron-Ion Collider (EIC) [1], planned to be built at Brookhaven National Lab, will further our knowledge of pion structure [2,3]. In China, a similar machine, the Electron-Ion Collider in China (EicC) [4], is also planned to make impacts on the pion gluon and sea-quark distributions. In Europe, the Drell-Yan and J/ψ-production experiments from COMPASS++/AMBER [5] will aim at improving our knowledge of both the pion gluon and quark PDFs.Global analyses of pion PDFs mostly rely on Drell-Yan data. The early studies of pion PDFs were based mostly on pion-induced Drell-Yan data and use J/ψ-production data or direct photon production to constrain the pion gluon PDF [6-10]. There are more recent studies, such as the work by Bourrely and Soffer [11], that extract the pion PDF based on Drell-Yan π + W data. JAM Collaboration [12,13] uses a Monte-Carlo approach to analyze the Drell-Yan πA and leading-neutron electroproduction data from HERA to reach the lower-x region, and revealed that gluons carry a significantly higher momentum fraction (about 30%) in the pion than had been inferred from Drell-Yan data alone. The xFitter group [14] analyzed Drell-Yan πA and photoproduction data using their open-source QCD fit framework for PDF extraction and found that these data can constrain the valence distribution well but are not sensitive enough for the sea and gluon distributions to be precisely determined. The analysis done Ref. [15] suggests that the pion-induced J/ψ-production data has additional constraint on pion PDFs, particularly in the pion gluon PDF in the large-x region. All in all, the pion valenc...