Pion valence structure from Ioffe-time parton pseudodistribution functions

Abstract

We present a calculation of the pion valence quark distribution extracted using the formalism of reduced Ioffe time pseudo-distributions or more commonly known as pseudo-PDFs. Our calculation is carried out on two different 2+1 flavor QCD ensembles using the isotropic-clover fermion action, with lattice dimensions 24 3 × 64 and 32 3 × 96 at the lattice spacing of a = 0.127 fm, and with the quark mass equivalent to a pion mass of mπ 415 MeV. We incorporate several combinations of smeared-point and smeared-smeared pion source-sink interpolation fields in obtaining the lattice QCD matrix elements using the summation method. After one-loop perturbative matching and combining the pseudo-distributions from these two ensembles, we extract the pion valence quark distribution using a phenomenological functional form motivated by the global fits of parton distribution functions. We also calculate the lowest four moments of the pion quark distribution through the "OPE without OPE". We present a qualitative comparison between our lattice QCD extraction of the pion valence quark distribution with that obtained from global fits and previous lattice QCD calculations.arXiv:1909.08517v1 [hep-lat]

Pion valence quark distribution from current-current correlation in lattice QCD

“…5 the effects of other possible correction terms such as m 2 π , a 2 , Le −m π ðL−ξÞ , ffiffiffi ffi L p e −m π ðL−ξÞ are observed to be very mild. We choose ω cut ¼ 1.0 as used in [61]; other choices of ω cut were observed to have no effect on the final band in the physical limit and vanishing higher-twist Oðξ 2 Þ contributions. The blue band in Fig.…”

“…The parameters in the PDF parametrization equation 17are determined by fitting the convolution of the model PDF and the NLO perturbative kernel to σ VA ðωÞ in a manner similar to Ref. [61], where the Ioffe-time zero point is fixed in this calculation by the LO þ NLO perturbative kernel. The isolation of σ VA ðωÞ is a multistep process, and begins by performing a correlated fit of lattice data from all four ensembles according to Eq.…”

“…For the above fit, we use α s ¼ 0.303 at the initial scale μ 0 ¼ 2 GeV [66]. Systematic uncertainties in each PDF parameter set are estimated by a 10% variation in α s as in [61]. The 2-parameter fit, by fixing γ ¼ 0 in Eq.…”

Machine-learning prediction for quasiparton distribution function matrix elements

“…In this work, we are interested in finding out how well the ML approach would work with the difficult computational situation in LaMETtype observables and whether computational cost can be saved for future large lattices (say, 64 3 × 192 and above). Although this paper focuses on the discussion with quasi-PDF, what we learn here also applies to the pseudo-PDF 1 [27][28][29][30] correlators since the building blocks of matrix elements are the same.…”

Generalized parton distributions and pseudodistributions

“…The matching relations for the pseudo-PDFs have been already used in lattice calculations [10,21,[47][48][49], while these for the pion DA and GPDs will be used in the ongoing lattice calculations.…”