Pion and kaon valence-quark quasiparton distributions
Algebraic Ansätze for the Poincaré-covariant Bethe-Salpeter wave functions of the pion and kaon are used to calculate their light-front wave functions (LFWFs), parton distribution amplitudes (PDAs), quasi-PDAs (qPDAs), valence parton distribution functions (PDFs), and quasi-PDFs (qPDFs). The LFWFs are broad, concave functions; and the scale of flavour-symmetry violation in the kaon is roughly 15%, being set by the ratio of emergent masses in the -and -quark sectors. qPDAs computed with longitudinal momentum GeV provide a semiquantitatively accurate representation of the objective PDA; but even with GeV, they cannot provide information about this amplitude’s endpoint behaviour. On the valence-quark domain, similar outcomes characterise qPDFs. In this connection, however, the ratio of kaon-to-pion -quark qPDFs is found to provide a good approximation to the true PDF ratio on , suggesting that with existing resources computations of ratios of quasi-parton-distributions can yield results that support empirical comparison.
Since the discovery of quarks in deep inelastic scattering (DIS) experiments at the Stanford Linear Accelerator Center Taylor:1991ew (); Kendall:1991np (); Friedman:1991nq (), parton distributions have occupied a central role in high-energy nuclear and particle physics; and today there is a vast international programme aimed at their measurement. Such measurements are possible owing to the existence of factorisation theorems Collins:1989gx (), which entail that the cross-sections for various processes can be written as the convolution of a piece calculable using QCD perturbation theory and a parton distribution function (PDF), which is independent of the process used. PDFs are therefore a characterising property of the chosen hadronic target. This means they are also essentially nonperturbative, i.e. their calculation is a problem in strong QCD (sQCD), with sound results demanding the use of a nonperturbative method with a traceable connection to QCD. The extent and importance of this computational challenge is canvassed elsewhere Holt:2010vj (); Brodsky:2015aia (); Horn:2016rip ().
An ab initio approach to sQCD is provided by the numerical simulation of lattice-regularised QCD (lQCD). However, a given PDF is mathematically defined as an expectation value of some bilocal operator evaluated along a light-like line, an object which cannot be evaluated using the methods of lQCD. This approach only provides access to the expectation value of local operators, i.e., in this context, to the Mellin moments of the PDF. That would not be an issue if every moment were accessible because a probability distribution is completely determined once all its moments are known. However, discretised spacetime does not possess the full rotational symmetries of the Euclidean continuum. Hence, only the lowest three non-trivial moments can readily be calculated; and they are insufficient to support a model-independent reconstruction of the PDF. A number of paths are being pursued to circumvent this problem Liu:1993cv (); Ji:2013dva (); Radyushkin:2016hsy (); Radyushkin:2017gjd (); Chambers:2017dov ().
Herein, using continuum methods in quantum field theory, we explore some aspects of the large (longitudinal-momentum) approach to the lQCD computation of parton distributions Ji:2013dva (), viz. since the maximum value of is bounded in any lQCD simulation, what is the lowest value of for which the quasi-distribution provides a realistic sketch of the true distribution; and given a quasi-distribution, is it possible to extract reliable information about the true distribution? These issues are considered, e.g. in Refs. Gamberg:2015opc (); Bacchetta:2016zjm () using spectator models of the proton. We, on the other hand, choose to focus on the pion and kaon because there has been significant progress in the continuum computation of the distribution amplitudes and functions of these systems in recent years Brodsky:2006uqa (); Chang:2013pq (); Cloet:2013tta (); Chang:2013nia (); Segovia:2013eca (); Shi:2014uwa (); Chang:2014lva (); Mezrag:2014jka (); Shi:2015esa (); Raya:2015gva (); Chen:2016sno (); Mezrag:2016hnp (); Li:2016dzv (); Raya:2016yuj (); Gao:2016jka (); Qin:2017lcd (); Gao:2017mmp (); deTeramond:2018ecg () and first lQCD results on some of their quasi-distributions are now available Zhang:2017bzy (); Chen:2017gck (). Our discussion complements the analysis in Refs. Nam:2017gzm (); Broniowski:2017wbr ().
Sec. II describes the framework used to represent - and -mesons as bound states in quantum field theory, specifying the elements at an hadronic scale, GeV. Herein, we do not consider perturbative QCD evolution Dokshitzer:1977 (); Gribov:1972 (); Lipatov:1974qm (); Altarelli:1977 (); Lepage:1979zb (); Efremov:1979qk (); Lepage:1980fj (): such evolution doesn’t affect our comparisons between quasi-distributions and their associated light-front distribution functions, which are all made at the same scale. Section III focuses on and leading-twist light-front wave functions, their derived parton distribution amplitudes and attendant quasi-distributions, introducing general formulae and providing numerical illustrations. The kindred analysis of valence-dressed-quark quasi-parton-distribution-functions (qPDFs) is presented in Sec. IV. We conclude in Sec. V.
Ii Pion and Kaon Bound States
Many insights into the character of the pion and kaon have been drawn using the following simple expressions for the relevant dressed-quark propagators and Bethe-Salpeter amplitudes Chang:2013pq (); Chang:2014lva (); Mezrag:2014jka (); Chen:2016sno ():
where is the dressed-quark mass evaluated in the neighbourhood , (we work in the isospin symmetric limit, so ); , ; , with , ; and is a spectral weight whose form determines the pointwise behaviour of the associated meson’s Bethe-Salpeter amplitude, with the related canonical normalisation constant. One of the strengths of these Ansätze is that they can be chosen to ensure that a primarily algebraic computation yields results which are pointwise similar to the most sophisticated predictions currently available for parton distribution amplitudes and functions, PDAs and PDFs. Notably, our approach to the continuum bound-state problem is Poincaré-covariant and hence, with complete generality, we may write
One branch of our analysis will focus on the leading-twist two-dressed-parton distribution amplitudes of the - and -mesons, the computation of which requires a projection onto the light-front of the given meson’s (unamputated) Bethe-Salpeter wave function. Working with the meson as an illustration, this wave function is ()
and the part which contributes to the leading-twist (twist-two) PDA is readily computed:
One may now introduce two Feynman parameters, combine the denominators into a single quadratic form, and thereby arrive at:
Formulae for the -meson are readily obtained by setting , , and using isospin symmetry.
As has long been known Pagels:1974se () and is demonstrated for parton distributions in, e.g. Refs. Segovia:2013eca (); Shi:2014uwa (); Shi:2015esa (); Li:2016dzv (); Li:2016mah (), distinctions between the - and -mesons are driven by dynamical chiral symmetry breaking (DCSB), expressed in Eq. (6) by the difference between the dressed - and quark masses: .
Iii Light-Front Wave Functions and Quasi-PDAs
iii.1 Algebraic Analysis
The pseudoscalar meson’s leading-twist two-dressed-parton light-front wave function (LFWF) can be written:
where is the kaon’s leptonic decay constant; the trace is over colour and spinor indices; ; ; and is a light-like four-vector, , . The twist-two PDA follows immediately:
Consider now the following Mellin moments:
where ; and is a Poincaré-invariant definition of the four-dimensional integral. Following a series of steps similar to those used above, one arrives directly at the following result:
The expression for is obvious by analogy and matches Eq. (20) in Ref. Radyushkin:2017gjd ().
iii.2 Numerical Illustrations
iii.2.1 Wave Functions
It is now possible to study the -evolution of the pointwise-form of meson qPDAs and chart their connection with the objective PDA. To proceed, it is necessary to specify the parameters and spectral densities in Eq. (1). For the latter, we use
where , , , are parameters, and is a derived constant that ensures unit normalisation of the density. This form is compact and yet has sufficient flexibility to produce pion and kaon valence-quark PDAs and PDFs whose features are consistent with contemporary predictions.
Regarding the parameters, we choose GeV, matching the infrared scale of the -quark mass function obtained using modern gap-equation kernels Binosi:2016wcx (); set , which is typical of the size obtained in phenomenologically efficacious continuum analyses Maris:1997tm (); Shi:2015esa (); float to fit the leptonic decay constants:
where GeV, GeV; and choose , such that the meson PDAs are broad, concave functions whose lowest nontrivial Mellin moments match those obtained in modern analyses Horn:2016rip ():
we obtain GeV, GeV, in agreement with experiment Olive:2016xmw (), and satisfy Eqs. (16). Recall that in connection with quantities that undergo QCD evolution, our models should be understood as producing results valid at an hadronic scale, GeV.
The pion and kaon leading-twist dressed-parton LFWFs, obtained using Eqs. (1), (11), (14), (17), are depicted in Fig. 1. Considered as a function of , with fixed, these wave functions are broad and concave. Conversely, at fixed , they fall as on . In QCD, the behaviour is (up to -corrections). Our model’s decay rate is amplified because Eq. (1b) retains only the piece of the pseudoscalar meson Bethe-Salpeter amplitude. Two additional “pseudovector” Dirac structures are prominent in symmetry-preserving solutions of the Bethe-Salpeter equations for light pseudoscalars Maris:1997tm (); and the omission of these components produces the decay at ultraviolet momenta Maris:1998hc (). This has a benefit: all integrals appearing herein are convergent. Restoring the pseudovector components, the LFWFs recover the decay characteristic of meson wave functions in QCD. Consideration of regularisation and renormalisation is then necessary; but that is straightforward and has no material effect on our discussion, which is why we exploit the simplicity of Eq. (1b).
The lower panel of Fig. 1 reveals that the LFWF is distorted, with its maximum located at , viz. displaced relative to that of and thereby indicating that the dressed -quark carries a larger fraction of the kaon’s momentum than the -quark. As noted elsewhere Segovia:2013eca (); Shi:2014uwa (); Shi:2015esa (); Li:2016dzv (); Li:2016mah (), the magnitude of this -flavour-symmetry breaking shift (%) is set by DCSB mass-scales, as expressed, e.g. in .
Having obtained the leading-twist LFWFs, one may compute the two-dressed-parton distribution amplitudes using Eq. (9a), with the results depicted in Fig. 2. Consistent with the LFWFs, the PDAs are broad, concave functions. Notably, although our models for the and are simple, they yield PDAs that agree qualitatively and semiquantitatively with results computed using more sophisticated approaches Chang:2013pq (); Segovia:2013eca (); Shi:2015esa (); Horn:2016rip (); Gao:2016jka (); Gao:2017mmp (). The peak in the kaon PDA lies at .
Notwithstanding the fact that a light-front Hamiltonian cannot in principle produce an eigenfunction of the form for any system because, e.g. it violates momentum conservation, such a product Ansatz is often used to produce numerical estimates of various quantities. Given a hadron , it is typically introduced in the form:
with , the profile function, often chosen to provide exponential decay. In order to judge the accuracy of estimates obtained therewith, Fig. 3 depicts the ratio:
evaluated at a number of -values and normalised at each such that . If a product Ansatz were a good approximation, then .
The upper panel of Fig. 3 depicts the ratio in Eq. (19) obtained for the pion, . It is not unity; and the discrepancy grows with increasing momentum until GeV, whereafter the ratio has a fairly static profile. On the other hand, the departure from unity is not great: the -deviation saturates at approximately %. One might therefore argue that Eq. (18), with appropriate power-law behaviour for , could be quantitatively useful for integrated properties of the pion and serve as a fair guide to the pointwise behaviour of . One should nevertheless bear in mind that any product Ansatz will be poorest on the domain of greatest correlation between the independent variables; and owing to momentum conservation, that domain is the neighbourhood of the endpoints, , as evident in Fig. 3.
The situation is somewhat different for the kaon. Depicted in the lower panel of Fig. 3, the ratio departs from unity by as much as 70%. The -deviation is 15% at , initially drops with increasing , but increases on GeV to reach a limiting value of %. In such circumstances, with a well-chosen power-law form for , Eq. (18) might provide a fair indication of integrated kaon properties, but it can only at best be a sketchy guide to pointwise features of .
iii.2.2 Quasi Parton Distribution Amplitudes
Eq. (13) can now be used to compute pion and kaon qPDAs, with the results depicted in Fig. 4. Focusing first on the pion (upper panel), it is evident that the result obtained with GeV does not closely resemble : the -difference between the two curves is 42% and the -moment obtained by integrating on is just 33% of the objective value.111The objective value for this moment is , Eq. (16). On physical grounds Segovia:2013eca (), the pion’s -moment should lie between the conformal limit value, , and the result obtained using constant, viz. . Using , the moment defined here takes the value : .
The step to GeV brings material improvement, so that the provides a qualitatively sound approximation to : the -difference between the two curves is 18%, the -moment is % of the objective value, and one can reasonably conclude that the target PDA is a broad, concave function.
Further increments in , however, do not bring much improvement. For example, with GeV, the -difference between the qPDA and the PDA is 10% and the -moment is % of the objective value. This outcome is a reflection of the fact that once the perturbative domain is entered, evolution in QCD is logarithmic.
It is noteworthy, too, that the pointwise forms of leak significantly from the domain . This prevents a determination of the target PDAs endpoint behaviour even with GeV. That behaviour is crucial because it fixes the magnitude of the leading-order, leading-twist perturbative QCD results for numerous observables Lepage:1980fj () and hence sets the benchmark against which existing and foreseen experiments aimed at testing solid QCD predictions must be compared Brodsky:2011yv (); Chang:2013nia (); Raya:2015gva (); Raya:2016yuj (); Gao:2017mmp (). Notably, in order to reach GeV in a lQCD simulation, one would need a lattice with roughly spatial sites and a spacing of fm.
Turning attention now to the kaon qPDAs, there are similarities with the pion case. Using GeV, provides some reliable qualitative information about : the -difference between the two curves is 28%, the qPDA peaks at , and its moment is 75% of the objective value. On the other hand, the -moment is just % of the goal. Once again, incrementing does not greatly improve the situation. Using GeV, the -difference between and is 20%, the qPDA peaks at , the moment is 82% of the objective value, but the -moment is only % of the goal. In this case, reducing the -difference between and to 10% would require GeV.
In closing this subsection we return to -flavour-symmetry violation in the kaon qPDAs, plotting in Fig. 5. Evidently, using modest values of , the qPDAs provide a fair pointwise description of the true difference on . Again, however, the behaviour on large domains near the endpoints is poorly represented. The figure also displays
The comparison of with highlights that the scale of flavour-symmetry breaking in the kaon distribution amplitudes measures differences between the emergent masses of - and -quarks in the Standard Model. The analogue of Eq. (20a) produced using Higgs-generated current-masses is an order-of-magnitude too large at the qPDAs’ extrema.
Iv Quasi Parton Distribution Functions
iv.1 Algebraic Analysis: qPDFs
In describing valence-dressed-quark parton distribution functions at an hadronic scale, , the impulse-approximation (handbag diagram) is inadequate because it omits contributions from the gluons which bind valence-quarks into a hadron. A remedy for this flaw is described and used to compute pion and kaon valence-quark distribution functions in Refs. Chang:2014lva (); Chen:2016sno (). Using the kaon as an illustration:
with , where is the charge conjugation matrix and denotes a transposed matrix. Expressions for analogous distributions in the are obtained by changing .
Canonical normalisation of the kaon’s Bethe-Salpeter amplitudes ensures
Consequently, using Eq. (21b), one finds immediately:
In obtaining these results, one must use mathematical features of the matrix trace, properties of propagators and Bethe-Salpeter amplitudes under charge conjugation, and the following identity: for ,
Analogous to the procedure in Sec. (III.1), the primary step is simply in the PDF formulae. However, the correction term, , is also needed. Its presence is suggested by the role of Eq. (25) in ensuring momentum conservation; and it guarantees, inter alia, Eq. (26b). Once again, analogous distributions in the are obtained by replacing .
iv.2 Numerical Illustrations: qPDFs
We now use Eqs. (1), (14), (17), (21) to compute the pion and kaon qPDFs. The calculation is straightforward, following the pattern in Sec. 1: one uses Feynman parametrisation to combine denominator products into a single quadratic form, Cauchy’s theorem to evaluate the integral, direct evaluation for , and finally numerical integration over the Feynman parameters. The results are depicted in Fig. 6. (The objective PDFs were obtained using the approach described in Ref. Chen:2016sno () and checked using the overlap representation Brodsky:1989pv ().)
A cursory comparison between Figs. 4 and 6 reveals that a valence-quark qPDF is typically a better approximation to the objective result than a qPDA at any given value of . Looking closer at the pion (Fig. 6, upper panel), the -differences are 19% (GeV), 9% (GeV), 5% (GeV), 4% (GeV). This series indicates that even with GeV, the pion’s valence-quark qPDF delivers a qualitatively sound approximation to the true result; and the step to GeV brings noticeable improvement; but, as with the qPDAs, improvement is slow on GeV.
Similar, too, is the pointwise behaviour of the valence-quark qPDFs in the neighbourhood of the endpoints: as with the qPDAs, the qPDFs leak significantly from the domain . This is important because one of the earliest predictions of the QCD parton model, augmented by features of perturbative QCD (pQCD), is that the valence-quark distribution function in a pseudoscalar meson behaves as follows Brodsky:1973kr (); Brodsky:1974vy (); Ezawa:1974wm (); Farrar:1975yb (); Berger:1979du (); Ball:2016spl ():
where is an anomalous dimension. Verification of Eq. (28) is an important milestone on the path toward confirmation of QCD as the theory of strong interactions Holt:2010vj (). In this connection we recall that Ref. Conway:1989fs () (the E615 experiment) reported a pion valence-quark PDF obtained via a leading-order pQCD analysis of their data, viz. , seemingly a marked contradiction of Eq. (28). Subsequent computations using continuum methods appropriate to QCD bound-states Hecht:2000xa () confirmed Eq. (28) and eventually prompted reconsideration of the E615 analysis, with the result that at next-to-leading order and including soft-gluon resummation Wijesooriya:2005ir (); Aicher:2010cb (), the E615 data can be viewed as being consistent with Eq. (28). New data are essential in order to check this reappraisal of the E615 data and settle the controversy. This goal is a focus of an approved tagged DIS experiment at the Thomas Jefferson National Accelerator Facility (JLab) Keppel:2015 (); McKenney:2015xis (); R.A.Montgomery:2017hab (). Such data could also be obtained with the common muon proton apparatus for structure and spectroscopy (COMPASS) detector at CERN Peng:2017ddf (); Bourrely:2018yck () and at a future electron ion collider (EIC) Holt:2000cv (); Horn:2018ghc ().
These observations emphasise that quantitatively reliable lQCD results bearing upon Eq.(28) would be very valuable. However, the challenge to delivering such outcomes using qPDFs is highlighted by Fig. 7. On a domain of valence-quark , this figure compares the pion qPDFs in the upper panel of Fig. 6 with the objective-PDF:
and another curve, whose -difference from the objective valence-quark PDF is just 2%, but which is at large-. Evidently, even the GeV qPDF is unable to distinguish between these two distinctively different results. (That is required before behaviour is visible in the pion’s valence-quark distribution was remarked upon earlier Holt:2010vj ().)
We now redirect our attention to kaon valence-quark qPDFs. As with kaon qPDAs, there are similarities with the pion. For instance, -differences are 25% (GeV), 12% (GeV), 8% (GeV), 6% (GeV), indicating, again, that even with GeV, the kaon’s qPDF delivers a qualitatively sound approximation to ; the step to GeV brings noticeable improvement, but changes are slow thereafter. The remarks made in connection with the pion qPDFs’ large- behaviour hold with equal force for the kaon.
It has been argued that the ratio serves as a sensitive probe of the difference between the gluon distributions in the pion and kaon Chen:2016sno (), and that this difference can reveal much about the emergence of mass in the Standard Model Roberts:2016vyn (). Experimental data on the ratio is available Badier:1980jq (), but one measurement is insufficient for complete confidence. Newer data would be welcome, in which connection tagged DIS at JLab might also be useful Horn:2016rip (); C12-15-006A (), as could the COMPASS detector at the CERN Peng:2017ddf (); Bourrely:2018yck () and a future EIC Holt:2000cv (); Horn:2018ghc (). With these things in mind, in Fig. 8 we depict the ratio . Evidently, for GeV, much as was the case with the qPDA asymmetry depicted in Fig. 5, the ratio of qPDFs is quantitatively a good approximation to the objective ratio on a material domain, viz. . This domain almost covers that upon which empirical data is available. We therefore anticipate that contemporary lQCD simulations could provide a sound prediction for this ratio before next generation experiments are completed.
V Summary and Perspective
Employing a continuum approach to bound-states in quantum field theory and practical algebraic Ansätze for the Poincaré-covariant Bethe-Salpeter wave functions of the pion and kaon, we computed the leading-twist two-dressed-parton light-front wave functions (LFWFs), ; parton distribution amplitudes (PDAs), ; quasi-PDAs (qPDAs), ; valence parton distribution functions (PDFs), ; and quasi-PDFs (qPDFs), , for these systems.
The LFWFs are broad, concave functions, with power-law -decay. Whilst the pion’s LFWF is symmetric about , peaks at , expressing -flavour-symmetry violation with a magnitude determined by differences between dynamical (not explicit) mass generation in the - and -quark sectors. Looking closely at the LFWFs, we found that a carefully constructed product Ansatz, viz. , although flawed in principle, can provide fair estimates of integrated , properties.
The LFWFs provide direct access to and PDAs and qPDAs; and for each system the qPDF provides a semiquantitatively reliable representation of the associated PDA when computed using a longitudinal momentum GeV. However, improvements thereafter are slow; and, notably, even with GeV, the qPDA cannot provide information about the true PDAs endpoint behaviour.
Regarding pion and kaon valence-quark PDFs and qPDFs, we found that at any given , a qPDF delivers a better representation of the associated PDF than does a qPDA of the objective PDA. In fact, even with GeV the qPDF provides a qualitatively clear picture of the PDF. However, as with qPDAs, differences between qPDFs and PDFs diminish slowly on GeV; and, similarly, even with GeV, qPDFs cannot be used to determine the objective PDF’s large- behaviour. On the other hand, the ratio does provide a good approximation to on , in consequence of which we expect that contemporary simulations of lattice-regularised QCD can deliver a reasonable prediction for this ratio before next generation experiments are completed.
It is natural to extend this analysis to the neutron and proton, for which analogous algebraic Ansätze for the bound-state Faddeev wave functions exist or can readily be developed Mezrag:2017znp (). We anticipate that the outcome will be qualitatively similar: in particular, that even using modest values of , a material valence-quark -domain will exist upon which ratios of qPDFs may provide sound representations of the PDF ratios measured empirically Hawker:1998ty (); Baillie:2011za (); Tkachenko:2014byy (); Reimer:2016dcd ().
Acknowledgements.We are grateful for constructive suggestions from J. Chen, T. Horn, L. Liu, C. Mezrag, J. Rodríguez-Quintero and S. Riordan. Work supported by: China Postdoctoral Science Foundation (under Grant No. 2016M591809); the Chinese Government’s Thousand Talents Plan for Young Professionals; the Chinese Ministry of Education, under the International Distinguished Professor programme; U.S. Department of Energy, Office of Science, Office of Nuclear Physics, under contract no. DE-AC02-06CH11357; and National Natural Science Foundation of China (under Grant Nos. 11475085, 11535005 and 11690030).
- (1) R. E. Taylor, Rev. Mod. Phys. 63, 573 (1991).
- (2) H. W. Kendall, Rev. Mod. Phys. 63, 597 (1991).
- (3) J. I. Friedman, Rev. Mod. Phys. 63, 615 (1991).
- (4) J. C. Collins, D. E. Soper and G. F. Sterman, Adv. Ser. Direct. High Energy Phys. 5, 1 (1989).
- (5) R. J. Holt and C. D. Roberts, Rev. Mod. Phys. 82, 2991 (2010).
- (6) S. J. Brodsky et al., (aXiv:1502.05728 [hep-ph]), QCD and Hadron Physics.
- (7) T. Horn and C. D. Roberts, J. Phys. G. 43, 073001 (2016).
- (8) K.-F. Liu and S.-J. Dong, Phys. Rev. Lett. 72, 1790 (1994).
- (9) X. Ji, Phys. Rev. Lett. 110, 262002 (2013).
- (10) A. Radyushkin, Phys. Lett. B 767, 314 (2017).
- (11) A. V. Radyushkin, Phys. Rev. D 95, 056020 (2017).
- (12) A. J. Chambers et al., Phys. Rev. Lett. 118, 242001 (2017).
- (13) I. Vitev, L. Gamberg, Z. Kang and H. Xing, PoS QCDEV2015, 045 (2015).
- (14) A. Bacchetta, M. Radici, B. Pasquini and X. Xiong, Phys. Rev. D 95, 014036 (2017).
- (15) S. J. Brodsky and G. F. de Teramond, Phys. Rev. Lett. 96, 201601 (2006).
- (16) L. Chang et al., Phys. Rev. Lett. 110, 132001 (2013).
- (17) I. C. Cloët, L. Chang, C. D. Roberts, S. M. Schmidt and P. C. Tandy, Phys. Rev. Lett. 111, 092001 (2013).
- (18) L. Chang, I. C. Cloët, C. D. Roberts, S. M. Schmidt and P. C. Tandy, Phys. Rev. Lett. 111, 141802 (2013).
- (19) J. Segovia et al., Phys. Lett. B 731, 13 (2014).
- (20) C. Shi et al., Phys. Lett. B 738, 512 (2014).
- (21) L. Chang et al., Phys. Lett. B 737, 23â29 (2014).
- (22) C. Mezrag et al., Phys. Lett. B 741, 190 (2015).
- (23) C. Shi et al., Phys. Rev. D 92, 014035 (2015).
- (24) K. Raya et al., Phys. Rev. D 93, 074017 (2016).
- (25) C. Chen, L. Chang, C. D. Roberts, S. Wan and H.-S. Zong, Phys. Rev. D 93, 074021 (2016).
- (26) C. Mezrag, H. Moutarde and J. Rodríguez-Quintero, Few Body Syst. 57, 729 (2016).
- (27) B. L. Li et al., Phys. Rev. D 93, 114033 (2016).
- (28) K. Raya, M. Ding, A. Bashir, L. Chang and C. D. Roberts, Phys. Rev. D 95, 074014 (2017).
- (29) F. Gao, L. Chang and Y.-X. Liu, Phys. Lett. B 770, 551 (2017).
- (30) S.-X. Qin, C. Chen, C. Mezrag and C. D. Roberts, Phys. Rev. C 97, 015203 (2018).
- (31) F. Gao, L. Chang, Y.-X. Liu, C. D. Roberts and P. C. Tandy, Phys. Rev. D 96, 034024 (2017).
- (32) G. F. de Teramond et al., (2018), Universality of Generalized Parton Distributions in Light-Front Holographic QCD, arXiv:1801.09154 [hep-ph].
- (33) J.-H. Zhang, J.-W. Chen, X. Ji, L. Jin and H.-W. Lin, Phys. Rev. D 95, 094514 (2017).
- (34) J.-W. Chen et al., (arXiv:1712.10025 [hep-ph]), Kaon Distribution Amplitude from Lattice QCD and the Flavor SU(3) Symmetry.
- (35) S.-i. Nam, Mod. Phys. Lett. A 32, 1750218 (2017).
- (36) W. Broniowski and E. Ruiz Arriola, Phys. Lett. B 773, 385 (2017).
- (37) Y. L. Dokshitzer, Sov. Phys. JETP 46, 641 (1977).
- (38) V. N. Gribov and L. N. Lipatov, Sov. J. Nucl. Phys. 15, 438 (1972).
- (39) L. N. Lipatov, Sov. J. Nucl. Phys. 20, 94 (1975).
- (40) G. Altarelli and G. Parisi, Nucl. Phys. B 126, 298 (1977).
- (41) G. P. Lepage and S. J. Brodsky, Phys. Lett. B 87, 359 (1979).
- (42) A. V. Efremov and A. V. Radyushkin, Phys. Lett. B 94, 245 (1980).
- (43) G. P. Lepage and S. J. Brodsky, Phys. Rev. D 22, 2157 (1980).
- (44) H. Pagels, Phys. Rept. 16, 219 (1975).
- (45) B.-L. Li, L. Chang, M. Ding, C. D. Roberts and H.-S. Zong, Phys. Rev. D 94, 094014 (2016).
- (46) D. Binosi, L. Chang, J. Papavassiliou, S.-X. Qin and C. D. Roberts, Phys. Rev. D 95, 031501(R) (2017).
- (47) P. Maris and C. D. Roberts, Phys. Rev. C 56, 3369 (1997).
- (48) C. Patrignani et al., Chin. Phys. C 40, 100001 (2016).
- (49) P. Maris and C. D. Roberts, Phys. Rev. C 58, 3659 (1998).
- (50) S. J. Brodsky, F.-G. Cao and G. F. de Teramond, Phys. Rev. D 84, 033001 (2011).
- (51) S. J. Brodsky and G. P. Lepage, Adv. Ser. Direct. High Energy Phys. 5, 93 (1989).
- (52) S. J. Brodsky and G. R. Farrar, Phys. Rev. Lett. 31, 1153 (1973).
- (53) S. J. Brodsky and G. R. Farrar, Phys. Rev. D 11, 1309 (1975).
- (54) Z. F. Ezawa, Nuovo Cim. A 23, 271 (1974).
- (55) G. R. Farrar and D. R. Jackson, Phys. Rev. Lett. 35, 1416 (1975).
- (56) E. L. Berger and S. J. Brodsky, Phys. Rev. Lett. 42, 940 (1979).
- (57) R. D. Ball, E. R. Nocera and J. Rojo, Eur. Phys. J. C 76, 383 (2016).
- (58) J. S. Conway et al., Phys. Rev. D 39, 92 (1989).
- (59) M. B. Hecht, C. D. Roberts and S. M. Schmidt, Phys. Rev. C 63, 025213 (2001).
- (60) K. Wijesooriya, P. E. Reimer and R. J. Holt, Phys. Rev. C 72, 065203 (2005).
- (61) M. Aicher, A. Schäfer and W. Vogelsang, Phys. Rev. Lett. 105, 252003 (2010).
- (62) D. Adikaram et al., (2015), Measurement of Tagged Deep Inelastic Scattering (TDIS), approved Jefferson Lab experiment E12-15-006.
- (63) J. R. McKenney, N. Sato, W. Melnitchouk and C.-R. Ji, Phys. Rev. D 93, 054011 (2016).
- (64) R. A. Montgomery et al., AIP Conf. Proc. 1819, 030004 (2017).
- (65) J.-C. Peng, W.-C. Chang, S. Platchkov and T. Sawada, (2017), Valence Quark and Gluon Distributions of Kaon from J/Psi Production, arXiv:1711.00839 [hep-ph].
- (66) C. Bourrely and J. Soffer, (arXiv:1802.03153 [hep-ph]), Statistical approach of pion parton distributions from Drell-Yan process.
- (67) R. J. Holt and P. E. Reimer, AIP Conf. Proc. 588, 234 (2001).
- (68) T. Horn, PoS DIS2017, 115 (2018).
- (69) C. D. Roberts, Few Body Syst. 58, 5 (2017).
- (70) J. Badier et al., Phys. Lett. B 93, 354 (1980).
- (71) J. Annand et al., (2017), Measurement of Kaon Structure Function through Tagged Deep Inelastic Scattering (TDIS), approved Jefferson Lab experiment C12-15-006A.
- (72) C. Mezrag, J. Segovia, L. Chang and C. D. Roberts, (arXiv:1711.09101 [nucl-th]), Parton distribution amplitudes: revealing diquarks in the proton and Roper resonance.
- (73) E. A. Hawker et al., Phys. Rev. Lett. 80, 3715 (1998).
- (74) N. Baillie et al., Phys. Rev. Lett. 108, 142001 (2012), [Erratum: Phys. Rev. Lett. 108, 199902 (2012)].
- (75) S. Tkachenko et al., Phys. Rev. C89, 045206 (2014), [Addendum: Phys. Rev. C 90, 059901 (2014)].
- (76) P. E. Reimer, EPJ Web Conf. 113, 05012 (2016).