betaFeedback

WorksPhysics and Astronomy

NRQCD analysis of charmonium production with pion and proton beams at fixed-target energies

Abstract

We present an analysis of hadroproduction of J/ψJ/ψ and ψ(2S)ψ(2S) at fixed-target energies in the framework of non-relativistic QCD (NRQCD). Using both pion- and proton-induced data, a new determination of the color-octet long-distance matrix elements (LDMEs) is obtained. Compared with previous results, the contributions from the qqˉq \bar{q} and color-octet processes are significantly enhanced, especially at lower energies. A good agreement between the pion-induced J/ψJ/ψ production data and NRQCD calculations using the newly obtained LDMEs is achieved. We find that the pion-induced charmonium production data are sensitive to the gluon density of pions, and favor pion PDFs with relatively large gluon contents at large xx.

Introduction

Since the discovery of the J/ψJ/\psi resonance in 1974, the study of ccˉc\bar{c} and bbˉb\bar{b} quarkonium states has significantly improved our understanding of the strong interaction [1, 2, 3, 4, 5]. Production of heavy-quark bound states offers important testing grounds for both perturbative and nonperturbative aspects of QCD. From the experimental perspective, quarkonia are readily detected through their dilepton decay, and the associated production cross sections are large compared to other hard processes. Since hadroproduction of quarkonium receives important contribution form gluon-gluon fusion it could serve as a tool for accessing the gluon distributions of interacting hadrons, and particularly for the unstable pion. Our current knowledge on the pion parton distribution functions (PDFs) mainly comes from pion-induced Drell-Yan data. As the Drell-Yan cross sections are dominated by qˉ\bar{q}-qq annihilation, they essentially probe the valence quark distribution in the pion, but leave the sea and the gluon distributions unconstrained. The mounting interest in the pion PDFs is reflected in many publications in the past few years. The pion PDFs were extracted using new and refined global analyses [6, 7, 8], or calculated from new theoretical developments, such as the continuum Dyson-Schwinger Equations [9, 10], light-front holographic QCD [11], and light-front Hamiltonian [12]. First lattice QCD results on the xx-dependence of pion valence structure are also becoming available [13, 14, 15, 16, 17, 18].

A significant amount of pion-induced charmonium data has been collected in the past. Recently, we investigated the sensitivity of the fixed-target J/ψJ/\psi production data to the pion PDFs [19]. The pion-induced J/ψJ/\psi data on hydrogen and several light-mass nuclear targets were compared to next-to-leading-order (NLO) Color Evaporation Model (CEM) [20] calculations using the available global-fit parametrizations of the pion PDFs. Over the energy range considered, the calculations with pion PDFs determined by SMRS [21], GRV [22], JAM [6] and xFitter [7], are all in reasonable agreement with the total cross sections data. In contrast, for the longitudinal momentum xFx_F distributions, we found that the agreement between data and calculations strongly depends on the magnitude and shape of the gluon distribution in the pion. The data favor pion PDFs with larger gluon strength at large xx.

To examine a possible model dependence of our previous observations [19], we extend our work to a more elaborate formalism for quarkonia production, namely, the non-relativistic QCD (NRQCD) [23] approach. This approach is based on a factorization of the heavy-quark pair QQˉQ\bar{Q} production and its subsequent hadronization. The production of the QQˉQ\bar{Q} pair proceeds through a short-distance partonic interaction, calculated using perturbative QCD. The probability of hadronization of a QQˉQ\bar{Q} pair into some quarkonium bound state depends on its spin, color and angular momentum. Being of a non-perturbative nature, the hadronization is described by the associated long-distance matrix elements (LDMEs). The LDMEs, assumed to be universal, are extracted from the experimental data. The values for the color-singlet (CS) LDMEs can be obtained from the decay widths of charmonia or model calculations, while the color-octet (CO) ones are determined from fits to data of transverse-momentum (pTp_T) differential cross sections, polarization, and pTp_T-integrated cross sections of charmonium production.

The NRQCD approach is based on the assumption that the factorization holds [23]. However, for low pTp_T additional effects might have large contributions. Analysis of fixed-target data could provide hints for such effects by assessing the universality of the LDMEs. Such study, focusing exclusively on charmonium hadroproduction from fixed-target experiments, was reported by Beneke and Rothstein [24]. The calculation was performed at partial next-to-leading order (up to O(αs3)\mathcal{O}(\alpha_s^3)). Some LDMEs were taken from fits to collider data of pTp_T-differential cross sections [25, 26] while others were fitted to the proton-induced cross section. The resulting LDMEs were used to calculate the pion-induced total cross sections. The calculation was found to lie systematically below the data, for both J/ψJ/\psi and ψ(2S)\psi(2S) production. The failure in describing the pion-induced data was speculated to be due to either inaccurate gluon distributions of pion PDFs or a difference in the higher twist effects between the proton and the pion data [24].

More recently, Maltoni et al. [27] re-analyzed the data of J/ψJ/\psi and ψ(2S)\psi(2S) fixed-target proton-induced cross sections. The pion-induced data were not included in the analysis. Their NRQCD calculation was performed at NLO using the formalism described in Ref. [28]. The initial CO parameters were taken from the Tevatron data [29, 30], but the S-wave CO LDMEs for J/ψJ/\psi and ψ(2S)\psi(2S) were allowed to vary. A good description of the data was achieved, at the expense of a strong reduction of the color-octet LDMEs by a factor of 5 to 10, depending on the proton PDFs used.

By using the NRQCD formalism in the fixed-target domain, the primary motivation of our present work is twofold. First, we aim at the determination of a new set of LDMEs that can reproduce both pion and proton-induced J/ψJ/\psi production cross sections. As the pion contains a valence antiquark, the J/ψJ/\psi cross section for a pion beam has a larger qqˉq\bar{q} contribution than that for a proton beam. Thereby, the pion-induced data provide a stronger constraint on the LDMEs that are responsible for the qqˉq\bar{q} part of the cross section. Second, we use this new set of LDMEs to explore the sensitivity of the pion-induced charmonium data to the parametrizations of the pion PDFs. The most recent global fits [6, 7] lead to pion gluon distributions which significantly differ from the older pion PDF determinations [21, 22], at large and medium parton momenta [19]. According to our previous results [19], these new sets of pion PDFs bring worse agreement between the CEM calculations and the data. It is important to validate these observations with a NRQCD calculation.

To perform this study, we adopt the NRQCD framework explicitly formulated in Ref. [24]. We start with a comparison of NRQCD calculations with the total cross section data for J/ψJ/\psi and ψ(2S)\psi(2S) productions with proton and pion beams. We then vary the CO LDMEs for J/ψJ/\psi and ψ(2S)\psi(2S) production to achieve a consistent description of both proton and pion data. In addition, we use the new LDMEs to explore the sensitivity of the pion-induced J/ψJ/\psi data to the gluon distribution in the pion. This paper is organized as follows. In Sec. 2, we introduce the NRQCD framework for the calculations of charmonia cross sections in hadroproduction. A new fit to the fixed-target J/ψJ/\psi and ψ(2S)\psi(2S) production data to determine a new set of LDMEs is presented in Sec. 3. We further discuss several findings from the present study in Sec. 4, followed by a summary in Sec. 5.

Charmonium Production and NRQCD

In NRQCD, the differential cross section as a function of Feynman xx (xFx_F), dσ/dxFd\sigma/dx_F for a charmonium state HH (H=J/ψH = J/\psi, ψ(2S)\psi(2S), or χcJ\chi_{cJ}) from the hNhN, (h=ph = p, pˉ\bar{p}, or π\pi) collisions, where hh is the beam hadron and NN the target nucleon, is [31]

dσHdxF=∑i,j=q,qˉ,G∫01dx1dx2δ(xF−x1+x2)×fih(x1,μF)fjN(x2,μF)σ^[ij→H](x1Ph,x2PN,μF,μR,mc),(1)\begin{aligned} \frac{d\sigma^{H}}{dx_{F}}=\sum_{i,j=q,\bar{q},G}\int_{0}^{1} dx_{1}dx_{2}\delta(x_{F}-x_{1}+x_{2}) \\ &\times f_{i}^{h}(x_{1},\mu_{F})f_{j}^{N}(x_{2},\mu_{F})\hat{\sigma}[ij\to H](x_{1}P_{h},x_{2}P_{N},\mu_{F},\mu_{R},m_{c}), \tag*{(1)} \end{aligned}
σ^[ij→H]=∑nCccˉ[n]ij(x1Ph,x2PN,μF,μR,mc)×⟨OnH[2S+1LJ]⟩(2)\hat{\sigma}[ij\to H]=\sum_{n}C_{c\bar{c}[n]}^{ij}(x_{1}P_{h},x_{2}P_{N},\mu_{F},\mu_{R},m_{c})\times\langle\mathcal{O}_{n}^{H}[{}^{2S+1}L_{J}]\rangle \tag*{(2)}
xF=2pL/s,x1,2=xF2+4Mccˉ2/s±xF2(3)x_{F}=2p_{L}/\sqrt{s},\quad x_{1,2}=\frac{\sqrt{x_{F}^{2}+4M_{c\bar{c}}^{2}/s}\pm x_{F}}{2} \tag*{(3)}

where ii and jj label the type of interacting partons (gluons, quarks and antiquarks), and the ccˉc\bar{c} pair is denoted by its color (n)(n), spin (S)(S), orbital angular momentum (L)(L) and total angular momentum (J)(J). Here mcm_{c} and MccˉM_{c\bar{c}} are the charm quark and ccˉc\bar{c} pair masses, fhf^{h} and fNf^{N} are the incoming hadron and the target nucleon parton distribution functions, evaluated at their respective Bjorken-xx values, x1x_{1} and x2x_{2}. The μF\mu_{F} and μR\mu_{R} are the factorization and renormalization scales. The total cross sections are obtained by integrating over xFx_{F}.

The production cross section σ^[ij→H]\hat{\sigma}[ij\to H] is the sum of the products of the short-distance coefficients Cccˉ[n]ijC_{c\bar{c}[n]}^{ij}, calculated perturbatively in powers of αs(μR)\alpha_{s}(\mu_{R}), and the nonperturbative parameters, LDMEs ⟨OnH[2S+1LJ]⟩\langle\mathcal{O}_{n}^{H}[{}^{2S+1}L_{J}]\rangle. The coefficients Cccˉ[n]ijC_{c\bar{c}[n]}^{ij} describe the production of a ccˉc\bar{c} pair in a specific spin-color state [n][n], while the parameters ⟨OnH[2S+1LJ]⟩\langle\mathcal{O}_{n}^{H}[{}^{2S+1}L_{J}]\rangle account for the hadronization of the ccˉ[n]c\bar{c}[n] pair into the charmonium state HH. The LDMEs are mainly determined by a fit to data and assumed to be universal, independent of beam or target hadrons and the energy scale.

In this work, we use the expressions given in Ref. [24] for computation of J/ψJ/\psi, ψ(2S)\psi(2S), and χcJ\chi_{cJ} production via GGGG, qqˉq\bar{q} and qGqG processes. The scattering processes qqˉ→QQˉq\bar{q}\to Q\bar{Q} and GG→QQˉGG\to Q\bar{Q} at O(αs2)\mathcal{O}(\alpha_{s}^{2}) produce QQˉQ\bar{Q} pairs in an S-wave CO state or P-wave CS state. Starting from O(αs3)\mathcal{O}(\alpha_{s}^{3}), the S-wave or P-wave CS QQˉQ\bar{Q} pairs can be produced through the scattering processes GG→QQˉGGG\to Q\bar{Q}G and qG→QQˉqqG\to Q\bar{Q}q [26]. We use this NRQCD framework for calculating the total cross sections of J/ψJ/\psi and ψ(2S)\psi(2S) production in hadronic collision.

The relationship of LDMEs to the scattering processes for each charmonium state is summarized in Table 1. For the qqˉq\bar{q} process, the ccˉc\bar{c} pairs are produced at O(αs2)\mathcal{O}(\alpha_{s}^{2}) in color octet states, which then hadronize into various charmonium states with the LDMEs ⟨O8H[3S1]⟩\langle\mathcal{O}_{8}^{H}[{}^{3}S_{1}]\rangle. For the GGGG process, both J/ψJ/\psi, and ψ(2S)\psi(2S) can be produced from either the CO ccˉc\bar{c} at O(αs2)\mathcal{O}(\alpha_{s}^{2}) or the CS ccˉc\bar{c} at O(αs3)\mathcal{O}(\alpha_{s}^{3}). The corresponding hadronization LDMEs for the CO and CS states are Δ8H=⟨O8H[1S0]⟩+3/mc2⟨O8H[3P0]⟩+4/(5mc2)⟨O8H[3P2]⟩\Delta_{8}^{H}=\langle\mathcal{O}_{8}^{H}[{}^{1}S_{0}]\rangle+3/m_{c}^{2}\langle\mathcal{O}_{8}^{H}[{}^{3}P_{0}]\rangle+4/(5m_{c}^{2})\langle\mathcal{O}_{8}^{H}[{}^{3}P_{2}]\rangle and ⟨O1H[3S1]⟩\langle\mathcal{O}_{1}^{H}[{}^{3}S_{1}]\rangle, respectively. The χc0\chi_{c0}, χc1\chi_{c1} and χc2\chi_{c2} can be produced from CS ccˉc\bar{c} states in the GGGG process at either O(αs2)\mathcal{O}(\alpha_{s}^{2}) (χc0\chi_{c0}, χc2\chi_{c2}) or O(αs3)\mathcal{O}(\alpha_{s}^{3}) (χc1\chi_{c1}) with the corresponding hadronization LDMEs ⟨O1H[3P0]⟩/mc2\langle\mathcal{O}_{1}^{H}[{}^{3}P_{0}]\rangle/m_{c}^{2}, ⟨O1H[3P1]⟩/mc2\langle\mathcal{O}_{1}^{H}[{}^{3}P_{1}]\rangle/m_{c}^{2} and ⟨O1H[3P2]⟩/mc2\langle\mathcal{O}_{1}^{H}[{}^{3}P_{2}]\rangle/m_{c}^{2}. At O(αs3)\mathcal{O}(\alpha_{s}^{3}), the qGqG process can contribute to the production of χc1\chi_{c1} via the CS ccˉc\bar{c} state which hadronizes with the LDME ⟨O1H[3P1]⟩/mc2\langle\mathcal{O}_{1}^{H}[{}^{3}P_{1}]\rangle/m_{c}^{2}.

HHqqˉq\bar{q}GGGGqGqG
J/ψJ/\psi, ψ(2S)\psi(2S)⟨O8H[3S1]⟩\langle\mathcal{O}^{H}_{8}[{}^{3}S_{1}]\rangle (O(αs2)\mathcal{O}(\alpha_{s}^{2}))Δ8H∗\Delta_{8}^{H*} (O(αs2)\mathcal{O}(\alpha_{s}^{2})) ⟨O1H[3S1]⟩\langle\mathcal{O}^{H}_{1}[{}^{3}S_{1}]\rangle (O(αs3)\mathcal{O}(\alpha_{s}^{3}))
χc0\chi_{c0}⟨O8H[3S1]⟩\langle\mathcal{O}^{H}_{8}[{}^{3}S_{1}]\rangle (O(αs2)\mathcal{O}(\alpha_{s}^{2}))⟨O1H[3P0]⟩\langle\mathcal{O}^{H}_{1}[{}^{3}P_{0}]\rangle (O(αs2)\mathcal{O}(\alpha_{s}^{2}))
χc1\chi_{c1}⟨O8H[3S1]⟩\langle\mathcal{O}^{H}_{8}[{}^{3}S_{1}]\rangle (O(αs2)\mathcal{O}(\alpha_{s}^{2}))⟨O1H[3P1]⟩\langle\mathcal{O}^{H}_{1}[{}^{3}P_{1}]\rangle (O(αs3)\mathcal{O}(\alpha_{s}^{3}))⟨O1H[3P1]⟩\langle\mathcal{O}^{H}_{1}[{}^{3}P_{1}]\rangle (O(αs3)\mathcal{O}(\alpha_{s}^{3}))
χc2\chi_{c2}⟨O8H[3S1]⟩\langle\mathcal{O}^{H}_{8}[{}^{3}S_{1}]\rangle (O(αs2)\mathcal{O}(\alpha_{s}^{2}))⟨O1H[3P2]⟩\langle\mathcal{O}^{H}_{1}[{}^{3}P_{2}]\rangle (O(αs2)\mathcal{O}(\alpha_{s}^{2}))

Table 1. Relationship of LDMEs and the associated orders of αs\alpha_{s} to the scattering processes for various charmonium states in the NRQCD framework of Ref. [24]. *: Δ8H=⟨O8H[1S0]⟩+3mc2⟨O8H[3P0]⟩+4mc2⟨O8H[3P2]⟩\Delta_{8}^{H} = \langle\mathcal{O}^{H}_{8}[{}^{1}S_{0}]\rangle+ \frac{3}{m_{c}^{2}}\langle\mathcal{O}^{H}_{8}[{}^{3}P_{0}]\rangle+ \frac{4}{m_{c}^{2}}\langle\mathcal{O}^{H}_{8}[{}^{3}P_{2}]\rangle

The number of independent LDMEs is further reduced by applying the spin symmetry relations [24, 27]:

⟨O8J/ψ,ψ(2S)[3PJ]⟩=(2J+1)⟨O8J/ψ,ψ(2S)[3P0]⟩for J=2⟨O8χcJ[3S1]⟩=(2J+1)⟨O8χc0[3S1]⟩for J=1,2⟨O1χcJ[3PJ]⟩=(2J+1)⟨O1χc0[3P0]⟩for J=1,2.(4)\begin{aligned} \langle\mathcal{O}^{J/\psi,\psi(2S)}_{8}[{}^{3}P_{J}]\rangle&= (2J+1)\langle\mathcal{O}^{J/\psi,\psi(2S)}_{8}[{}^{3}P_{0}]\rangle\quad\text{for } J=2 \\ \langle\mathcal{O}^{\chi_{cJ}}_{8}[{}^{3}S_{1}]\rangle&= (2J+1)\langle\mathcal{O}^{\chi_{c0}}_{8}[{}^{3}S_{1}]\rangle\quad\text{for } J=1,2 \\ \langle\mathcal{O}^{\chi_{cJ}}_{1}[{}^{3}P_{J}]\rangle&= (2J+1)\langle\mathcal{O}^{\chi_{c0}}_{1}[{}^{3}P_{0}]\rangle\quad\text{for } J=1,2. \tag*{(4)} \end{aligned}

The best-fit LDMEs for the description of proton-induced J/ψJ/\psi and ψ(2S)\psi(2S) production from Ref. [24] are summarized in Table 2. In Ref. [24], the CS LDMEs ⟨O1H[3S1]⟩\langle\mathcal{O}^{H}_{1}[{}^{3}S_{1}]\rangle for J/ψJ/\psi and ψ(2S)\psi(2S), and ⟨O1H[3P0]⟩\langle\mathcal{O}^{H}_{1}[{}^{3}P_{0}]\rangle for χc0\chi_{c0}, are given by the potential models [25], and the CO LDMEs ⟨O8H[3S1]⟩\langle\mathcal{O}^{H}_{8}[{}^{3}S_{1}]\rangle for J/ψJ/\psi, ψ(2S)\psi(2S) and χc\chi_{c}, are taken from the fits to Tevatron collider data of pTp_{T} spectra [26]. The Δ8H\Delta_{8}^{H} parameters for the individual J/ψJ/\psi and ψ(2S)\psi(2S) production were determined by a fit of NRQCD calculation to the proton-induced data.

HH⟨O1H[3S1]⟩\langle\mathcal{O}^{H}_{1}[{}^{3}S_{1}]\rangle⟨O1H[3P0]⟩/mc2\langle\mathcal{O}^{H}_{1}[{}^{3}P_{0}]\rangle/m_{c}^{2}⟨O8H[3S1]⟩\langle\mathcal{O}^{H}_{8}[{}^{3}S_{1}]\rangleΔ8H\Delta_{8}^{H}
J/ψJ/\psi1.166.6×10−36.6 \times10^{-3}3×10−23 \times10^{-2}
ψ(2S)\psi(2S)0.764.6×10−34.6 \times10^{-3}5.2×10−35.2 \times10^{-3}
χc0\chi_{c0}0.0443.2×10−33.2 \times10^{-3}

Table 2. NRQCD LDMEs for the charmonium production in Ref. [24], all in units of GeV3^{3}.

A later study by Maltoni et al. [27] was performed with a full NLO calculation [28]. Starting with the LDMEs determined from the collider data of pTp_{T} differential cross sections [29, 30], an additional reduction factor of about 0.1 for ⟨O8H[3S1]⟩\langle\mathcal{O}^{H}_{8}[{}^{3}S_{1}]\rangle and ⟨O8H[1S0]⟩\langle\mathcal{O}^{H}_{8}[{}^{1}S_{0}]\rangle of J/ψJ/\psi and ψ(2S)\psi(2S), was required in the calculations to achieve a good description of fixed-target proton-induced data.

The CS LDMEs used in these two studies are either identical or very similar, as they were obtained from a potential model [25]. The CO ⟨O8H[3S1]⟩\langle\mathcal{O}^{H}_{8}[{}^{3}S_{1}]\rangle for χc0\chi_{c0} are nearly identical. In contrast, the remaining CO LDMEs obtained in Ref. [24] are about a factor of 1.5-5 larger than those determined in Ref. [27].

With the information of LDMEs, the direct production cross sections of J/ψJ/\psi, ψ(2S)\psi(2S) and three χcJ\chi_{cJ} states can be calculated as shown in Eq. (1). Furthermore, taking into account the direct production of J/ψJ/\psi and the feed-down from hadronic decays of ψ(2S)\psi(2S) and radiative decays of three χcJ\chi_{cJ} states, the total J/ψJ/\psi cross section is estimated as follows,

σJ/ψ=σJ/ψdirect+Br(ψ(2S)→J/ψX)σψ(2S)+∑J=02Br(χcJ→J/ψγ)σχcJ(5)\sigma_{J/\psi} = \sigma^{\mathrm{direct}}_{J/\psi} + Br(\psi(2S) \to J/\psi X)\sigma_{\psi(2S)} + \sum_{J=0}^{2} Br(\chi_{cJ} \to J/\psi\gamma)\sigma_{\chi_{cJ}} \tag*{(5)}

For the branching ratios we take the PDG 2020 values [32]

Br(ψ(2S)→J/ψX)=61.4%,Br(χc0→J/ψγ)=1.4%,Br(\psi(2S) \to J/\psi X) = 61.4\%, \qquad Br(\chi_{c0} \to J/\psi\gamma) = 1.4\%,
Br(χc1→J/ψγ)=34.3%,Br(χc2→J/ψγ)=19.0%.(6)Br(\chi_{c1} \to J/\psi\gamma) = 34.3\%, \qquad Br(\chi_{c2} \to J/\psi\gamma) = 19.0\%. \tag*{(6)}

Using this NRQCD framework, we calculate below the J/ψJ/\psi and ψ(2S)\psi(2S) production cross sections and compare them to the available fixed-target data. We explore if a set of LDMEs can be identified to achieve a good description of both proton-induced and pion-induced data of charmonia production.

Results of NRQCD calculations

In this section, we compare our NRQCD calculations with the fixed-target data of J/ψJ/\psi and ψ(2S)\psi(2S) production by proton and pion beams. The cross sections for proton-induced J/ψJ/\psi and ψ(2S)\psi(2S) and their ratios Rψ=σ(ψ(2S))/σ(J/ψ)R_{\psi} = \sigma(\psi(2S))/\sigma(J/\psi) are taken from Ref. [27]. The pion-induced data for J/ψJ/\psi and ψ(2S)\psi(2S) are taken from Refs. [1, 33]. In addition, three measurements of RψR_{\psi} from HERA [34] and NA38 [35], and the pion-induced data of σ(J/ψ)\sigma(J/\psi) and σ(ψ(2S))\sigma(\psi(2S)) from NA38 [36] and WA92 [37] are added.

In spite of a caveat of double counting by including the data of RψR_{\psi} in the analysis [27], the cancellation of systematic uncertainties in this quantity brings additional constraining power in the fit. The NRQCD calculations presented in this section are performed using the nucleon CT14nlo PDFs [38] and the pion GRV NLO PDFs [22] under the LHAPDF framework [39, 40]. The cross sections are evaluated with a charm quark mass mc=1.5 GeV/c2m_c = 1.5\ \mathrm{GeV}/c^2 and renormalization and factorization scale μR=μF=2mc\mu_R = \mu_F = 2m_c [24].

Reference fit

We compare these data with the NRQCD calculations using the LDMEs determined in Ref. [24], named as "Fit-R" in the following. The difference between the proton PDFs selected for the study of Ref. [24] (CTEQ3L) and here (CT14nlo) has a negligible effect on the final results. The results are plotted as black dashed lines in Figure 1. The total reduced χ2/ndf\chi^2/\mathrm{ndf}, together with the deviations between data and calculations per data point (χ2/ndp\chi^2/\mathrm{ndp}) for each data set, are displayed in Figure 1 and listed in Table 3. The reduced χ2/ndf\chi^2/\mathrm{ndf} of the entire data sets is 16.8. Clearly, the pion-induced data below s=20 GeV\sqrt{s} = 20\ \mathrm{GeV} are significantly underestimated by the calculations as already observed in Ref. [24].

Image not extracted

Figure 1. J/ψJ/\psi and ψ(2S)\psi(2S) production cross sections and the ψ(2S)/(J/ψ)\psi(2S)/(J/\psi) ratios in the p+Np + N reactions, and J/ψJ/\psi and ψ(2S)\psi(2S) production cross sections in the π−+N\pi^- + N reactions, labeled as (a)-(e) in the plot. The dashed (black), dot-dashed (blue) and solid (red) curves represent the NRQCD results using the LDMEs obtained in "Fit-R", "Fit-1" and "Fit-2", respectively. The reduced χ2/ndf\chi^2/\mathrm{ndf} for all data are displayed in the bottom-right. The values of χ2\chi^2 divided by the number of data point (ndp) for each data set are also shown.

Fit-RFit-1Fit-2
χtotal2/ndf\chi^2_{\mathrm{total}}/\mathrm{ndf}16.86.03.3
χ2/ndp∣σ(J/ψ)p\chi^2/\mathrm{ndp}|_{\sigma(J/\psi)}^{p}9.24.15.4
χ2/ndp∣σ(ψ(2S))p\chi^2/\mathrm{ndp}|_{\sigma(\psi(2S))}^{p}2.21.41.7
χ2/ndp∣R(ψ(2S))p\chi^2/\mathrm{ndp}|_{R(\psi(2S))}^{p}1.10.71.0
χ2/ndp∣σ(J/ψ)π\chi^2/\mathrm{ndp}|_{\sigma(J/\psi)}^{\pi}46.815.33.7
χ2/ndp∣σ(ψ(2S))π\chi^2/\mathrm{ndp}|_{\sigma(\psi(2S))}^{\pi}2.80.90.7
⟨O8J/ψ[3S1]⟩\langle\mathcal{O}_{8}^{J/\psi}[{}^{3}S_{1}]\rangle6.6×10−36.6 \times10^{-3}(1.47±0.07)×10−1(1.47 \pm0.07) \times10^{-1}(9.5±0.4)×10−2(9.5 \pm0.4) \times10^{-2}
Δ8J/ψ\Delta_{8}^{J/\psi}3×10−23 \times10^{-2}(0±8)×10−4(0 \pm8) \times10^{-4}(1.8±0.2)×10−2(1.8 \pm0.2) \times10^{-2}
⟨O8ψ(2S)[3S1]⟩\langle\mathcal{O}_{8}^{\psi(2S)}[{}^{3}S_{1}]\rangle4.6×10−34.6 \times10^{-3}(2.5±0.2)×10−2(2.5 \pm0.2) \times10^{-2}(2.6±0.2)×10−2(2.6 \pm0.2) \times10^{-2}
Δ8ψ(2S)\Delta_{8}^{\psi(2S)}5.2×10−35.2 \times10^{-3}(0±8)×10−4(0 \pm8) \times10^{-4}(4±6)×10−4(4 \pm6) \times10^{-4}

Table 3. The reduced χ2/ndf\chi^2/\mathrm{ndf} of the whole data sets, the χ2\chi^2 divided by the number of data point (ndp) for each data set in "Fit-R", "Fit-1" and "Fit-2" NRQCD calculations and the corresponding input or best-fit LDMEs. All LDMEs are in units of GeV3\mathrm{GeV}^3.

Fit of the proton-induced data

It is known that the qqˉq\bar{q} contribution plays an important role in the pion-induced J/ψJ/\psi production near threshold because the parton density at large xx is dominated by the valence antiquarks in pions [19]. In our NRQCD framework, the ⟨O8H[3S1]⟩\langle\mathcal{O}_{8}^{H}[{}^{3}S_{1}]\rangle LDMEs are responsible for the qqˉq\bar{q} contribution, as shown in Table 1, and their values were taken from the fits to the Tevatron data [26] at high energies. The observed underestimation of low-energy pion-induced data in "Fit-R" could arise from too small a value for the input ⟨O8H[3S1]⟩\langle\mathcal{O}_{8}^{H}[{}^{3}S_{1}]\rangle.

To seek an improved description of the pion-induced data, we take a different approach, called "Fit-1", of leaving both ⟨O8H[3S1]⟩\langle\mathcal{O}_{8}^{H}[{}^{3}S_{1}]\rangle and Δ8H\Delta_{8}^{H} LDMEs for J/ψJ/\psi and ψ(2S)\psi(2S) as free parameters in the global fit of the proton-induced data. The CO LDME of χc0\chi_{c0} is fixed at the value given in Ref. [24] and the best-fit LDMEs are required to be positive-definite. The resulting fit is shown as blue dot-dashed lines in Figure 1. The cross sections for both proton- and pion-induced data are significantly enhanced, compared to the results of "Fit-R". The agreement between the data and calculation is greatly improved. As shown in Table 3, the overall χ2/ndf\chi^2/\mathrm{ndf} is reduced from 16.8 to 6.0, compared to "Fit-R". While the pion-induced data are not used in the global fit for the LDMEs determination, these data are included in the evaluation of χ2/ndf\chi^2/\mathrm{ndf} for comparison purposes.

More specifically, the agreement between the pion-induced J/ψJ/\psi data alone and the calculations with best-fit LDMEs of "Fit-1" is improved by a factor of 3, compared to "Fit-R". The values of newly determined ⟨O8H[3S1]⟩\langle\mathcal{O}_{8}^{H}[{}^{3}S_{1}]\rangle LDMEs are 1.5×10−11.5 \times10^{-1} and 2.5×10−22.5 \times10^{-2} for J/ψJ/\psi and ψ(2S)\psi(2S), respectively. Both are significantly larger than the "Fit-R" values of 6.6×10−36.6 \times10^{-3} and 4.6×10−34.6 \times10^{-3} determined from collider data. The increase of the values of ⟨O8H[3S1]⟩\langle\mathcal{O}_{8}^{H}[{}^{3}S_{1}]\rangle clearly accounts for better agreement between the NRQCD calculation and the pion-induced data, even though the pion data were not included in the fit.

The values of the CO Δ8H\Delta_8^H LDMEs for J/ψJ/\psi and ψ(2S)\psi(2S) resulting from "Fit-1" are compatible with zero, as shown in Table 3. Despite an improved description of data in this approach, "Fit-1" finds vanishing values of Δ8H\Delta_8^H for both J/ψJ/\psi and ψ(2S)\psi(2S). It appears that these LDMEs cannot be determined from the proton-induced J/ψJ/\psi and ψ(2S)\psi(2S) production data alone. This suggests the need to include also the pion data in the global fit, as discussed next.

Fit of both the pion- and proton-induced data

Because of the different nature of valence quarks in the protons and pions, the energy dependence of the relative contributions of qqˉq\bar{q} and GGGG processes is different for the J/ψJ/\psi and ψ(2S)\psi(2S) production, especially at low energies. Under the assumption that higher-twist effects are negligible, a combined fit of these two data sets should further constrain the CO LDMEs. The results of this new fit, referred to as "Fit-2", are shown in Table 3 and displayed as the solid red lines in Figure 1.

Comparing the results of "Fit-2" and "Fit-1", it is found that the description of the pion-induced data is improved, while maintaining a good agreement between the proton data and the calculation. The CO matrix elements are also better constrained. The total reduced χ2/ndf\chi^2/\mathrm{ndf} is further decreased to about 3.3. The agreement between the pion-induced J/ψJ/\psi data and the NRQCD calculation is improved by a factor of 4, from a χ2/ndp\chi^2/\mathrm{ndp} of 15.3 in "Fit-1" to 3.7 in "Fit-2". The values of the newly determined CO ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle LDMEs are 9.5×10−29.5 \times10^{-2} and 2.6×10−22.6 \times10^{-2} for J/ψJ/\psi and ψ(2S)\psi(2S), respectively, either smaller than or consistent with those obtained in "Fit-1".

With the inclusion of the pion-induced data, non-zero values of the CO Δ8H\Delta_8^H LDMEs can now be obtained. As shown in Table 3, the values of CO Δ8H\Delta_8^H LDMEs are found to be 1.8×10−21.8 \times10^{-2} and 4.0×10−44.0 \times10^{-4} for J/ψJ/\psi and ψ(2S)\psi(2S), respectively. The best-fit ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle LDMEs responsible for the contribution of the qqˉq\bar{q} process are larger by about a factor of 10, while the Δ8H\Delta_8^H related to the contribution of the GGGG process are reduced by a factor of 2-10, in comparison with the LDMEs determined from collider data [26, 29]. The new CO LDMEs indicate that the qqˉq\bar{q} contribution determined by the fixed-target data is significantly larger than the corresponding contribution at collider energies.

The systematic uncertainties of these results are studied by setting mcm_c to 1.2 GeV/c2c^2 or 1.5 GeV/c2c^2 and the normalization scale μ=μR=μR\mu= \mu_R = \mu_R to mcm_c, 2mc2m_c, and 3mc3m_c. Figure 2 shows the comparison of the pion-induced J/ψJ/\psi data and NRQCD calculation with the corresponding settings of mcm_c and μ\mu. The total cross sections and the qqˉq\bar{q}, GGGG, and qGqG contributions are denoted as black, blue, red and green lines, respectively.

Image not extracted

Figure 2. The NRQCD results from an analysis of both proton-induced and pion-induced data sets under variation of charm quark mass mcm_c, renormalization scale μR\mu_R and factorization scale μF\mu_F, compared with the pion-induced data of J/ψJ/\psi production as a function of s\sqrt{s}. The total cross sections and qqˉq\bar{q}, GGGG, and qGqG contributions are denoted as black, blue, red and green lines, respectively. The values of mcm_c, μ=μR=μF\mu= \mu_R = \mu_F in the NRQCD calculation as well as the best-fit χ2/ndp\chi^2/\mathrm{ndp} are displayed in each plot.

The parameter of the charm quark mass mcm_c plays a significant role in the systematic effect. With mcm_c set to 1.2 GeV/c2c^2, the CO LDMEs as free parameters are not well constrained, and the quality of fit significantly deteriorates as seen from the increased χ2/ndp\chi^2/\mathrm{ndp}. Judging from the contributions of various subprocesses, the GGGG process is enhanced too much to provide a good description of data in the calculations with this reduction.

of mcm_c. When mcm_c is set as 1.5 GeV/c21.5\ \mathrm{GeV}/c^2, the quality of the fit is equally good with μ\mu varying among mcm_c, 2mc2m_c, and 3mc3m_c. Even though one of CO LDMEs cannot be obtained with good accuracy when μ\mu is set at mcm_c, the best-fit LDMEs for three different scales are consistent with having large values of ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle LDMEs, reflecting a non-negligible qqˉq\bar{q} contribution.

Discussion

Fractions of individual contributions in hadroproduction

A new set of LDMEs for J/ψJ/\psi and ψ(2S)\psi(2S) production at fixed-target energies has been obtained in an analysis of data with proton and pion beams. Our analysis differs from that of Ref. [24] in two aspects. First, the LDMEs ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle are now allowed to vary in the global fit. This leads to a much improved description of fixed-target data with proton beam. Second, the pion data are included in the global fit. This allows for the determination of the LDMEs Δ8H\Delta_8^H.

To better understand the reasons for the significantly improved description of the J/ψJ/\psi and ψ(2S)\psi(2S) production data, it is instructive to compare the NRQCD calculations using the "Fit-R" LDMEs of Ref. [24] and the "Fit-2" LDMEs of the present analysis. In particular, we examine the decomposition of the J/ψJ/\psi production cross section into individual contributions in three fashions: (i) qqˉq\bar{q}, GGGG, and qGqG subprocesses; (ii) color singlet versus color octet ccˉc\bar{c} states; (iii) direct production of J/ψJ/\psi versus feed-down from ψ(2S)\psi(2S) and χc\chi_c.

The decomposition of J/ψJ/\psi production cross sections for proton and pion beams into the qqˉq\bar{q}, GGGG and qGqG processes is shown in Figure 3. We note that the qGqG contributions remain unchanged in the new analysis since the qGqG process only contributes to the formation of the χc1\chi_{c1} states and the LDMEs for χcJ\chi_{cJ} are identical for "Fit-R" and "Fit-2". Figure 3 also shows that the GGGG contribution is dominant in the J/ψJ/\psi production with proton beam at all energies, except near the threshold. In contrast, the qqˉq\bar{q} contribution for pion-induced data is enhanced due to the increased antiquark content in pion’s valence region. Therefore, the inclusion of the pion data in the global fit provides additional constraints on those LDMEs which are sensitive to the qqˉq\bar{q} process. The low-energy fixed-target pion data are particularly important for the determination of the ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle LDMEs.

Image not extracted

Figure 3. Total cross sections (black) and contributions from qqˉq\bar{q} (blue), GGGG (red) and qGqG (green) processes for J/ψJ/\psi production as a function of s\sqrt{s} in (a) pNpN and (b) π−N\pi^{-}N interactions. The dashed and solid curves represent the "Fit-R" and "Fit-2" results. The fractions of each sub-process cross section are displayed at the bottom of each plot.

In comparison with "Fit-R", the ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle LDME is increased, whereas Δ8H\Delta_8^H is decreased. These changes lead to an enhancement of the CO qqˉq\bar{q} contribution and a reduction of the CO GGGG contribution. The increase of the fraction of qqˉq\bar{q} contribution, especially at low-energies, accounts for the improvement in describing the pion data. The opposite trend in the variations of the two CO LDMEs ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle and Δ8H\Delta_8^H leads to significant changes in the energy dependence of J/ψJ/\psi production cross sections, as shown in Figure 3.

Figure 4 shows the decomposition of the J/ψJ/\psi cross sections into contributions from the color octet and color singlet states. As the CS contribution (blue lines) in our study is fixed, the enlarged ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle LDMEs

Image not extracted

Figure 4. Same as Figure 3 for the decomposition of contributions from CS (blue) and CO (red) processes.

significantly enhance the CO contribution (red lines) at low energies, while in "Fit-2" the reduced Δ8H\Delta_8^H results in a reduction of the CO contribution at high energies. Through the interplay of these two CO LDMEs and the parton luminosities, the CO contribution remains similar at high energies but is enhanced near threshold for the proton-induced production. In the case of pion-induced production, the CO contribution is slightly suppressed at high energies.

Figure 5 shows the decomposition of the J/ψJ/\psi cross sections into the contributions from direct production (red lines) and the feed-down from heavier charmonium states of ψ(2S)\psi(2S) (blue lines) and χc\chi_c (green lines). We note that the LDMEs for the three χc\chi_c states are kept unchanged for "Fit-R" and "Fit-2". The most notable change between the calculations with "Fit-2" and the "Fit-R" is the enhancement of the direct J/ψJ/\psi production at low energies, as a consequence of an enlarged ⟨O8H(3S1)⟩\langle\mathcal{O}_8^H({}^3S_1)\rangle. Taking into account the decay branching ratios, the contributions to the J/ψJ/\psi production in descending order of importance are direct production, χc\chi_c, and ψ(2S)\psi(2S).

Image not extracted

Figure 5. Same as Figure 3 for the decomposition of contributions of J/ψJ/\psi production from direct production of J/ψJ/\psi (red) and feed-down from ψ(2S)\psi(2S) (green) and all χc\chi_c states (blue).

To recapitulate the main findings at this point, we note that the inclusion of the low-energy pion-induced total cross section data of J/ψJ/\psi and ψ(2S)\psi(2S) production to the analysis of NRQCD provides an important constraint of the ⟨O8H(3S1)⟩\langle\mathcal{O}_8^H({}^3S_1)\rangle LDMEs via the qqˉq\bar{q} contributions. A good description of both proton- and pioninduced data by NRQCD can be achieved. The qqˉq\bar{q} and CO contributions from the NRQCD calculations with the new LDMEs are greatly enhanced at low energies with proton and pion beams, compared with results found in the earlier studies [24, 27].

Sensitivity to the pion PDFs

The new set of LDMEs is now used to study the sensitivity of the J/ψJ/\psi data to the various pion PDFs. We have considered four pion PDFs, namely, SMRS [21] and GRV [22], representative of the most widely used pion PDFs thus far, as well as JAM [6] and xFitter [7], obtained from very recent global analyses. For SMRS, we select the default one in which the sea quarks carry 15% of the pion momentum at Q2=4Q^2 = 4 GeV2^2. As illustrated in Ref. [19], SMRS, JAM, and xFitter have similar valence-quark distributions, while the magnitude of the GRV distribution is smaller by about 20%–30%. For the gluon distributions, SMRS and GRV have similar shapes and magnitudes, while xFitter and JAM have significantly smaller magnitudes by a factor of 2–4.

The NRQCD calculation with each of the four pion PDFs is compared with the data in Figure 6. Overall, the total cross sections (black lines) for the four pion PDFs exhibit similar s\sqrt{s} dependencies. However, the individual terms differ strongly. The qqˉq\bar{q} contribution dominates near thresholds and the GGGG contribution increases rapidly at higher energies, while the qGqG component is relatively negligible over the whole energy range. The relative fractions of qqˉq\bar{q} and GGGG contributions as a function of s\sqrt{s} vary for each pion PDFs, reflecting the differences among their parton distributions. For SMRS and GRV the GGGG contribution starts to dominate the cross section around s=20\sqrt{s} = 20 GeV. For xFitter and JAM the corresponding values are larger at ∼s=35\sim\sqrt{s} = 35 GeV because of their relatively reduced gluon strength in the valence region.

Image not extracted

Figure 6. The NRQCD J/ψJ/\psi production cross sections at xF>0x_F > 0 for the π−N\pi^-N reactions, calculated for four pion PDFs (SMRS, GRV, JAM and xFitter) using LDMEs of "Fit-2". The black, blue, red, and green curves represent the calculated total cross sections, and the qqˉq\bar{q}, GGGG, and qGqG contributions, respectively. The shaded bands on the xFitter and JAM calculations come from the uncertainties of the corresponding PDF sets. The SMRS and GRV PDFs contain no information on uncertainties.

Table 4 lists the χ2\chi^2 values of NRQCD calculations for various data sets and best-fit CO LDMEs for each pion PDF. We find that the χ2\chi^2 of the pion-induced J/ψJ/\psi data strongly correlates with the gluon density of pions over the valence quark regions [19]. The GGGG contributions are similar for GRV and SMRS, while those

SMRSGRVJAMxFitter
χtotal2/ndf\chi^2_{\mathrm{total}}/\mathrm{ndf}3.13.44.84.3
χ2/ndp∣σ(J/ψ)\chi^2/\mathrm{ndp}|_{\sigma(J/\psi)}2.73.78.46.4
χ2/ndp∣σ(ψ(2S))\chi^2/\mathrm{ndp}|_{\sigma(\psi(2S))}0.40.70.40.2
⟨O8J/ψ(3S1)⟩\langle\mathcal{O}_8^{J/\psi}({}^3S_1)\rangle(6.9±0.3)×10−2(6.9 \pm0.3) \times10^{-2}(9.5±0.4)×10−2(9.5 \pm0.4) \times10^{-2}(8.3±0.3)×10−2(8.3 \pm0.3) \times10^{-2}(7.4±0.9)×10−2(7.4 \pm0.9) \times10^{-2}
Δ8J/ψ\Delta_8^{J/\psi}(2.5±0.2)×10−2(2.5 \pm0.2) \times10^{-2}(1.8±0.2)×10−2(1.8 \pm0.2) \times10^{-2}(2.0±0.2)×10−2(2.0 \pm0.2) \times10^{-2}(2.2±0.2)×10−2(2.2 \pm0.2) \times10^{-2}
⟨O8ψ(2S)(3S1)⟩\langle\mathcal{O}_8^{\psi(2S)}({}^3S_1)\rangle(2.1±0.4)×10−2(2.1 \pm0.4) \times10^{-2}(2.6±0.2)×10−2(2.6 \pm0.2) \times10^{-2}(2.6±0.1)×10−2(2.6 \pm0.1) \times10^{-2}(2.3±0.4)×10−2(2.3 \pm0.4) \times10^{-2}
Δ8ψ(2S)\Delta_8^{\psi(2S)}(1.7±1.0)×10−3(1.7 \pm1.0) \times10^{-3}(4.0±6.2)×10−4(4.0 \pm6.2) \times10^{-4}(3.7±3.4)×10−4(3.7 \pm3.4) \times10^{-4}(0.9±6.0)×10−3(0.9 \pm6.0) \times10^{-3}

Table 4. The χ2\chi^2 values for the entire data sets and the individual J/ψJ/\psi and ψ(2S)\psi(2S) data set from the NRQCD calculations. The best-fit LDMEs for each pion PDF are in units of GeV3^3.

for xFitter and JAM are 50–80% smaller due to their weaker gluon strength at x=0.1−0.6x = 0.1 - 0.6, relative to GRV and SMRS [19]. The deficiency of xFitter and JAM in the GGGG contribution leads to an underestimation of the NRQCD calculations against the data over s=15−25 GeV\sqrt{s} = 15 - 25\ \mathrm{GeV}, as shown in Figure 6.

Table 4 shows that the dependence of the best-fit LDMEs for J/ψJ/\psi and ψ(2S)\psi(2S) to the pion PDFs is rather mild. We also checked that the overall reduced χ2/ndf\chi^2/\mathrm{ndf} of data for each pion PDF has a very small variation when the calculations are done with the best-fit LDMEs obtained using a different set of pion PDFs.

Summary

We have analyzed J/ψJ/\psi and ψ(2S)\psi(2S) hadroproduction data in fixed-target experiments within the framework of NRQCD. The previously reported difficulty [24] of obtaining a consistent description of proton and pion data by NRQCD, is resolved by allowing the CO LDMEs for J/ψJ/\psi and ψ(2S)\psi(2S) to be determined by a simultaneous fit to both proton- and pion-induced data. The pion-induced data, especially at low energies, have a sizable qqˉq\bar{q} contribution and thus provide a strong constraint on the ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle LDMEs. This sensitivity is much reduced if the analysis is restricted to the proton-induced data alone. Consequently, the best-fit ⟨O8H[3S1]⟩\langle\mathcal{O}_8^H[{}^3S_1]\rangle values for both J/ψJ/\psi and ψ(2S)\psi(2S) are found to be about ∼10\sim10 times larger than that same values obtained using the collider data. In contrast, the best-fit values for Δ8H\Delta_8^H, partly responsible for the GGGG contribution, are compatible for J/ψJ/\psi but more than 10 times smaller for ψ(2S)\psi(2S), compared to the results of Ref. [24].

The resulting LDMEs combined with four different pion PDFs were used to compare the NRQCD calculation with the pion-induced J/ψJ/\psi production cross-sections. The pronounced differences between the predicted individual quark-antiquark annihilation and gluon-gluon fusion terms result from the different shapes and magnitudes of the corresponding PDF parametrizations. When compared to the total cross section, the SMRS or GRV PDFs still provide a slightly better description of the data than JAM or xFitter, suggesting that the data favor those PDFs with larger gluon contents at medium and large xx. All these results are in line with our earlier findings [19] obtained with the CEM.

We note that there are the recent state-of-the-art NLO NRQCD fits [41, 42, 43, 44] to collider data. The ⟨O8J/ψ[3S1]⟩\langle\mathcal{O}_8^{J/\psi}[{}^3S_1]\rangle LDMEs found in Refs. [41, 43] are substantially smaller than our results. In the fixed-target domain considered here the mean pTp_T values are lower than the charmonium mass. In this low-pTp_T region the factorization assumption of NRQCD may no longer be valid. Higher-order corrections or other additional contributions could therefore spoil the universality of the LDMEs. It will be interesting to perform similar studies, including the pion-induced data of charmonium production with the full NLO NRQCD formalism and make a comparison with the current results in the future.

In the near future, new measurements of Drell-Yan as well as J/ψJ/\psi data in π−N\pi^-N reactions will be available from the CERN COMPASS [45] and AMBER [46] experiments. While future theoretical advances are needed to further reduce the model dependence in describing the J/ψJ/\psi production, we emphasize the importance of including the pion-induced J/ψJ/\psi data in future pion PDF global analysis.

Acknowledgments

We thank Nobuo Sato and Ivan Novikov for providing us with LHAPDF6-compatible grid files of JAM and xFitter PDFs. This work was supported in part by the U.S. National Science Foundation (Grant PHY18-22502) and the Ministry of Science and Technology of Taiwan.

References

  1. [1]G. A. Schuler, Quarkonium production and decays, arXiv:hep-ph/9403387.DOI
  2. [2]G. A. Schuler, Production of heavy quarks and heavy quarkonia, Z. Phys. C 71 (1996) 317.
  3. [3]N. Brambilla et al., Heavy Quarkonium: Progress, Puzzles, and Opportunities, Eur. Phys. J. C 71 (2011) 1534.DOI
  4. [4]A. Andronic et al., Heavy-flavour and quarkonium production in the LHC era: from proton-proton to heavy-ion collisions, Eur. Phys. J. C 76 (2016) 107.
  5. [5]J. P. Lansberg, New Observables in Inclusive Production of Quarkonia, Phys. Rept. 889 (2020) 1.
  6. [6]P. C. Barry, N. Sato, W. Melnitchouk, and C. R. Ji, First Monte Carlo Global QCD Analysis of Pion Parton Distributions, Phys. Rev. Lett. 121 (2018) 152001.DOI
  7. [7]I. Novikov et al., Parton Distribution Functions of the Charged Pion Within The xFitter Framework, Phys. Rev. D 102 (2020) 014040.
  8. [8]C. Bourrely, F. Buccella and J. C. Peng, A new extraction of pion parton distributions in the statistical model, Phys. Lett. B 813 (2021) 136021.DOI
  9. [9]K. D. Bednar, I. C. Cloët, and P. C. Tandy, Distinguishing Quarks and Gluons in Pion and Kaon PDFs, Phys. Rev. Lett. 124 (2020) 042002.DOI
  10. [10]Z. F. Cui et al., Kaon and pion parton distributions, Eur. Phys. J. C 80 (2020) 1064.
  11. [11]G. F. de Teramond et al. [HLFHS Collaboration], Universality of Generalized Parton Distributions in Light-Front Holographic QCD, Phys. Rev. Lett. 120 (2018) 182001 (2018).DOI
  12. [12]J. Lan, C. Mondal, S. Jia, X. Zhao, and J. P. Vary, Parton Distribution Functions from a Light Front Hamiltonian and QCD Evolution for Light Mesons, Phys. Rev. Lett. 122 (2019) 172001.DOI
  13. [13]J. H. Zhang, J. W. Chen, C. Jin, H. W. Lin, A. Schäfer, and Y. Zhao, First direct lattice-QCD calculation of the x-dependence of the pion parton distribution function, Phys. Rev. D 100 (2019) 034505.DOI
  14. [14]R. S. Sufian, J. Karpie, C. Egerer, K. Orginos, J. W. Qiu, and D. G. Richards, Pion Valence Quark Distribution from Matrix Element Calculated in Lattice QCD, Phys. Rev. D 99 (2019) 074507.DOI
  15. [15]T. Izubuchi, L. Jin, C. Kallidonis, N. Karthik, S. Mukherjee, P. Petreczky, C. Shugert, and S. Syritsyn, Valence parton distribution function of pion from fine lattice, Phys. Rev. D 100 (2019) 034516.
  16. [16]B. Joó, J. Karpie, K. Orginos, A. V. Radyushkin, D. G. Richards, R. S. Sufian, and S. Zafeiropoulos, Pion valence structure from Ioffe-time parton pseudodistribution functions, Phys. Rev. D 100 (2019) 114512.DOI
  17. [17]R. S. Sufian et al., Pion Valence Quark Distribution from Current-Current Correlation in Lattice QCD, Phys. Rev. D 102 (2020) 054508.DOI
  18. [18]X. Gao et al., Valence parton distribution of the pion from lattice QCD: Approaching the continuum limit, Phys. Rev. D 102 (2020) 094513.DOI
  19. [19]W. C. Chang, J. C. Peng, S. Platchkov and T. Sawada, Constraining gluon density of pions at large x by pion-induced J/ψ production, Phys. Rev. D 102 (2020) 054024.DOI
  20. [20]M. B. Einhorn and S. D. Ellis, Hadronic Production of the New Resonances: Probing Gluon Distributions, Phys. Rev. D 12 (1975) 2007; H. Fritzsch, Producing Heavy Quark Flavors in Hadronic Collisions: A Test of Quantum Chromodynamics, Phys. Lett. 67B (1977) 217; F. Halzen, CVC for Gluons and Hadroproduction of Quark Flavors, Phys. Lett. 69B (1977) 105.DOI
  21. [21]P. J. Sutton, A. D. Martin, R. G. Roberts, and W. J. Stirling, Parton distributions for the pion extracted from Drell-Yan and prompt photon experiments, Phys. Rev. D 45 (1992) 2349.
  22. [22]M. Gluck, E. Reya, and A. Vogt, Pionic parton distributions, Z. Phys. C 53 (1992) 651.
  23. [23]G. T. Bodwin, E. Braaten, and G. P. Lepage, Rigorous QCD analysis of inclusive annihilation and production of heavy quarkonium, Phys. Rev. D 51 (1995) 1125; 55 (1997) 5853(E).
  24. [24]M. Beneke and I. Z. Rothstein, Hadroproduction of quarkonia in fixed target experiments, Phys. Rev. D 54 (1996) 2005; 54 (1996) 7082(E).
  25. [25]E. J. Eichten and C. Quigg, Quarkonium wave functions at the origin, Phys. Rev. D 52 (1995) 1726.
  26. [26]P. L. Cho and A. K. Leibovich, Color octet quarkonia production. II, Phys. Rev. D 53 (1996) 6203.
  27. [27]F. Maltoni et al., Analysis of charmonium production at fixed-target experiments in the NRQCD approach, Phys. Lett. B 638 (2006) 202.DOI
  28. [28]A. Petrelli, M. Cacciari, M. Greco, F. Maltoni and M. L. Mangano, NLO production and decay of quarkonium, Nucl. Phys. B 514 (1998) 245.
  29. [29]M. Beneke and M. Krämer, Direct J/ψ and ψ′ polarization and cross-sections at the Tevatron, Phys. Rev. D 55 (1997) 5269.DOI
  30. [30]P. Nason et al., Bottom production, 1999 Workshop on Standard model physics (and more) at the LHC, 231-304, arXiv:hep-ph/0003142.
  31. [31]R. Vogt, The x(F) dependence of ψ and Drell-Yan production, Phys. Rev. C 61 (2000) 035203.DOI
  32. [32]P.A. Zyla et al. (Particle Data Group), Prog. Theor. Exp. Phys. 2020 (2020) 083C01.
  33. [33]R. Vogt, J/ψ production and suppression, Phys. Rept. 310 (1999) 197.
  34. [34]I. Abt et al. [HERA-B Collaboration], A Measurement of the ψ′ to J/ψ production ratio in 920-GeV proton-nleus interactions, Eur. Phys. J. C 49 (2007) 545.DOI
  35. [35]C. Baglin et al. [NA38 Collaboration], psi-Prime and J/ψ production in p W, p U and S U interactions at 200-GeV/nucleon, Phys. Lett. B 345 (1995) 617.
  36. [36]M. J. Corden et al., Experimental Results on J/ψ Production by π±, K±, p and p̄ Beams at 39.5-GeV, Phys. Lett. 96B (1980) 411.DOI
  37. [37]Y. Alexandrov et al. [BEATRICE Collaboration], Inclusive J/ψ and ψ′ production in π− - nucleus interactions at S(1/2) approximately equals 26-GeV, Nucl. Phys. B 557 (1999) 3.
  38. [38]S. Dulat et al., New parton distribution functions from a global analysis of quantum chromodynamics, Phys. Rev. D 93 (2016) 033006.DOI
  39. [39]M. R. Whalley, D. Bourilkov and R. C. Group, The Les Houches accord PDFs (LHAPDF) and LHAGLUE, arXiv:hep-ph/0508110.
  40. [40]A. Buckley, J. Ferrando, S. Lloyd, K. Nordström, B. Page, M. Rüfenacht, M. Schönherr and G. Watt, LHAPDF6: parton density access in the LHC precision era, Eur. Phys. J. C 75 (2015) 132.DOI
  41. [41]M. Butenschoen and B. A. Kniehl, Reconciling J/ψ production at HERA, RHIC, Tevatron, and LHC with NRQCD factorization at next-to-leading order, Phys. Rev. Lett. 106 022003 (2011).
  42. [42]Y. Q. Ma, K. Wang and K. T. Chao, J/ψ(ψ′) production at the Tevatron and LHC at O(αₛ⁴v⁴) in nonrelativistic QCD, Phys. Rev. Lett. 106 (2011) 042002.
  43. [43]M. Butenschoen and B. A. Kniehl, World data of J/ψ production consolidate NRQCD factorization at NLO, Phys. Rev. D 84 (2011) 051501.DOI
  44. [44]Y. Q. Ma, K. Wang and K. T. Chao, A complete NLO calculation of the J/ψ and ψ′ production at hadron colliders, Phys. Rev. D 84 (2011) 114001.
  45. [45]M. Aghasyan et al. [COMPASS Collaboration], First measurement of transverse-spin-dependent azimuthal asymmetries in the Drell-Yan process, Phys. Rev. Lett. 119 (2017) 112002; C. Y. Hsieh, JPS Conf. Proc. 26 (2019) 031003.
  46. [46]B. Adams et al., Letter of Intent: A New QCD facility at the M2 beam line of the CERN SPS (COMPASS++/AMBER), arXiv:1808.00848.arxiv.org/abs/1808.00848

Paper details

vlnc:W_gxtbdgyu
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents