Constraining scenarios of the soft/hard transition for the pion electromagnetic form factor with expected data of 12-GeV Jefferson Lab experiments and of the Electron–Ion Collider S.V. Troitsky1 and V.E. Troitsky2 5 1 0 1Institute for Nuclear Research of the Russian Academy of Sciences, 2 60th October Anniversary Prospect 7a, Moscow 117312, Russia∗ r a M 2D.V. Skobeltsyn Institute of Nuclear Physics, 3 M.V. Lomonosov Moscow State University, Moscow 119991, Russia† ] (Dated: December 31, 2014) h p Abstract - p e It has been shown previously [1] that a non-perturbative relativistic constituent-quark model h [ for the π-meson electromagnetic form factor allows for a quantitative description of the soft/hard 2 v transition, resulting in the correct Quantum-Chromodynamical asymptotics, including normal- 2 1 ization, from the low-energy data without further parameter tuning. This happens universally 7 2 whenever the constituent-quark mass is switched off. The energy range where the transition hap- 0 . 1 pens is therefore determined by the quark-mass running at intermediate energies and is not tightly 0 5 constrained theoretically. Here we consider possible ways to pin down this energy range with com- 1 : v ing experimental data. We demonstrate that expected experimental uncertainties of the 12-GeV i X Jefferson-Lab data are larger than the span of predictions of the model, so these data might be r a used for testing the model but not for determination of the soft/hard transition scale. Contrary, the projected Electron-Ion Collider will be capable of pinning down the scale. PACS numbers: 13.40 Gp, 14.40 Be, 12.39 Ki, 11.10 Jj ∗ [email protected] † [email protected] 1 I. INTRODUCTION Making connections between high-energy (hard) and low-energy (soft) models of hadrons, that is between the Quantum Chromodynamics (QCD) and effective theories of strong in- teractions working in the infrared limit, is one of the major challenges of modern particle theory. The electromagnetic form factor of the charged π meson, F , represents a particu- π larly interesting observable in this context. On one hand, its high-energy asymptotics is well defined within the perturbative QCD [2–4]. On the other hand, precise experimental data in the soft region allow to directly trace the evolution of the observable with the momentum transfer, Q2, up to Q2 (cid:38) 2 GeV2, that is close to the range where one expects the hard behavior to start settling down. The F form factor is remarkable in one more aspect which we will exploit here. It is π probably the only observable for which a successful low-energy theory exists which gives the correct QCD asymptotics quantitatively without a dedicated parameter tuning [1]. The soft/hard transition is governed byswitching theconstituent-quarkmass M off inthe model, and wherever the mass is switched off, the QCD asymptotics settles down, provided the low-energy description works well. Therefore, no additional parameters are required to be tuned to reproduce quantitatively the QCD asymptotics. The model uses the M(Q2) dependence as an input, and various ways of switching the mass off are allowed, just because the asymptotics is universal and is determined by the infrared, and not intermediate-scale, parameters. Therefore, the model, in its present form, does not predict the energy scale at which the soft/hard transition, that is switching M(Q2) off, takes place; this scale is to be determined either by a detailed model for M(Q2) or experimentally. Unfortunately, detailed models for the M running available in the literature contain a number of free parameters and can hardly be used for this purpose. Here, we explore prospects for experimental determination of this scale from future F measurements. π The rest of the paper is organized as follows. In Sec. II, we briefly review the model for the pion form factor which gives the correct QCD asymptotics starting from the low-energy physics. In Sec. III, we describe how the unknown ingredient of the model, the dependence of the effective constituent-quark mass on Q2, is constrained. We proceed in Sec. IV with estimates of the impact future data may have on these constraints, working within two representative scenarios; then we briefly conclude in Sec. V. 2 II. THE MODEL FRAMEWORKS We adopt a well-elaborated model for the pion form factor [1, 5–9] originally developed as a low-energy theory based on the Poincar´e invariant constituent-quark model. It exploits the instant form of Relativistic Hamiltonian Dynamics (see e.g. Ref. [10]). The use of the Modified Impulse Approximation [6] provides for the full relativistic invariance and eliminates certain drawbacks of the original instant form. The pion form factor, F (Q2), is π given by rather cumbersome but explicit expressions [6] collected in Ref. [1], which we do not quote here. The low-energy model has two free parameters, the constituent-quark mass M and the wave-function confinement scale b (the actual choice of the wave function does not have any significant effect on the result, see Refs. [5, 8]). These two parameters are tuned to reproduce correct experimental values of the pion decay constant f and of the pion charge π radius. It is remarkable that these parameters were fixed from the low-energy data (actually in 1998, Ref. [5], by making use of the F (Q2) measurements at Q2 (cid:46) 0.26 GeV2 [11]), so π no room to tune them remained. The predictions of Ref. [5] have been subsequently verified by new measurements of F (Q2) up to Q2 (cid:39) 2.5 GeV2, an order of magnitude higher, and π are in excellent agreement with all present-day data. In parallel with this phenomenological success, the model has an interesting, if not mirac- ulous, theoretical advantage. It has been noted in Ref. [12] that the asymptotical behavior Q2F (Q2) ∼ const, predicted by QCD [13, 14], is obtained in this model at M → 0. This is π however not a full story: introducing explicit M(Q2) dependence in such a way that M(0) is taken from the original model but M(∞) = 0, we obtained in Ref. [1] the numerical coef- ficient of this asymptotics which appeared to reproduce the QCD predictions quantitatively, without any parameter tuning, for every possible way of switching the quark mass off. The QCD asymptotics [2–4] is Q2F (Q2) (cid:39) 8πα(1)(Q2)f2, (1) π s π where α(1)(Q2) is the one-loop QCD coupling constant (extension to higher loops is not s straightforward, see e.g. Refs. [15, 16]). The right-hand side of Eq. (1) is determined by two parameters, f (determined in the low-energy theory) and the QCD scale Λ (which is π QCD related, though not explicitly, to the low-energy confinement parameter b). The fact that, by fixing f and the charge radius in the low-energy theory, we immediately reproduce Eq. (1) π quantitatively, is an important advantage of the model, not seen in other approaches. 3 The idea of switching the constituent-quark mass M off in order to obtain the form- factor behaviour at high Q2 was put forward in Ref. [17] in the frameworks of the light-front approach[18]; however,Eq.(1)wasnotobtainedthere. InRef.[1],weusedaparametrization for M(Q2) inspired by Ref. [17] but corrected for effects of a one-gluon exchange, 1+e−µ2/λ2 M(Q2) = M(∞)+(M(0)−M(∞)) L(Q2), (2) 1+e(Q2−µ2)/λ2 1 L(Q2) = . (3) 1+log Q2+µ2 µ2 The soft/hard transition is therefore governed by two parameters, µ and λ, of the M(Q2) function. The µ and λ parameters determine the position and the steepness of the transition between M(0) and M(∞). The boundary value M(∞) → 0 while M(0) = 0.22 GeV is fixed, like in previous studies, to reproduce f and the pion charge radius correctly. We note that π the parametrization (2), (3) describes well the M(Q2) dependence obtained, within certain assumptions about free parameters, in complicated non-perturbative dynamical models [19, 20], see Ref. [1] for details and illustrations. To summarize this section, we have a predictive quantitative model for the pion form factor whose low-energy parameters are fixed and determine the correct high-energy asymp- totics automatically, but the soft/hard transition is parametrized by M(Q2), that is by two parameters (µ,λ). We turn now to constraining these parameters. III. CONSTRAINING M(Q2) We start with theoretical constraints which are determined by the limits of applicability of the model. While, technically, Eq. (2) implies that M is always decreasing for arbitrary µ and λ, it appears that it cannot decrease too slow. Indeed, for each fixed Q2, there exists a value M (Q2) beyond which the form factor, considered as a function of M, ceases to max be monotonic. This, in turn, implies that the form factor as a function of Q2 may cease to be monotonically decreasing. We calculate this M (Q2) numerically and determine, max from the requirement M(Q2) < M (Q2), the corresponding restriction on the parameters max (µ,λ). This bound (“the consistency limit”), applicable in any case, is presented in Fig. 1 as a long-dashed line: the allowed region is to the left of this line. Note that the requirement of perturbativity at large Q2 we used in Ref. [1] is qualitatively similar to this bound. 4 FIG.1. Allowedregionsoftheplaneofparametersµandλwhichgovernthequark-massevolution, see text. Long-dashed line: the consistency limit (region to the left of the line is allowed; relevant for all cases). Full black line: the 95% CL bound from the present data (region to the right of the line is allowed). Other lines: 95% CL example bounds from future data for the ”low-scale” (dashed; the allowed region is bound by the lines) and “high-scale” (dot-dashed; the allowed region is to the right from the lines) scenarios of the soft/hard transition; thin lines assume 12-GeV JLab data, thick gray lines assume EIC data. The values of µ and λ assumed for the “low-scale” and “high-scale” scenarios are shown by the triangle and the diamond, respectively. 5 The other kind of constraints come from experimental measurements of F at relatively π high Q2. As we have already pointed out, the original model with M = const predicted the form-factor values up to Q2 ∼ 2.6 GeV2 with high accuracy, hence an early departure from the constant-mass scenario might result in a disagreement with data. The corresponding bound on M(Q2) is obtained from the requirement of agreement, at the 95% confidence level (CL), of the corresponding F (Q2) function with the data points, tested by means of π the ususal chi-square method. For demonstration purposes, we also define the “soft/hard transition scale” Q2 as the value of Q2 at which the difference between the predicted trans Q2F (Q2) and Eq. (1) is one half of its maximal value, that is the form factor is half way π from its nonperturbative values to the QCD asymptotics. By making use of all present-day data described in Ref. [21], we obtain constraints on M(Q2) which are presented in Fig. 1 in terms of µ and λ (the full line; the region to the right of the line is allowed). The corresponding range of allowed F (Q2) is shown as a gray π band in Fig. 2 (see Ref. [1] for more plots). In terms of the soft/hard transition scale, this constraint is Q2 > 8.5 GeV2. trans Having determined the constraints on M(Q2) from the present data, we are ready to discuss prospective bounds from future experiments. IV. EXPECTED CONSTRAINTS FROM FUTURE DATA Experimental prospects of the measurements of the pion form factor are briefly summa- rized in Ref. [24]. They include the approved E12-06-101 experiment at the upgraded Jef- ferson Laboratory (12-GeV JLab) fascility and measurements at the projected Electron-Ion Collider (EIC). More details may be found in the experimental proposal [25] for the 12-GeV JLab and in the talk [26] for EIC. Hereafter, we will use the information about the Q2 reach and projected error bars of F measurements at these fascilities given in Refs. [25, 26] and π reproduced in Ref. [24] (for EIC, we assume the energy of the ion beam of 5 GeV, the lowest one considered there). With the 12 GeV energy, JLab will be able to measure F for the π momentum transfers up to ∼ 6 GeV2 with the accuracy of ∼ 4%. For EIC, there exist various proposals under consideration; for the 5 GeV proton energy, F might be measured π up to Q2 ∼ 15 GeV2 with the accuracy of ∼ 10%. To proceed further, we restrict ourselves to two particular representative scenarios cor- 6 FIG. 2. The range of predictions of our model for the charged pion form factor allowed by the present data and model consistency for various scenarios of quark-mass running (gray band) to- gether with predictions of Refs. [22] (area between two dotted lines) and [23] (dash-dotted line). Two dashed lines represent two particular scenarios for M(Q2) corresponding to the “low-scale” and “high-scale” soft/hard transitions discussed in the text. Existing experimental data points (see Ref. [1] for their description and list of references) are shown by black dots with thin error bars. A typical simulated example realization of expected 12-GeV JLab data, corresponding to the “high-scale” scenario, is shown by gray dots with thick gray error bars. It is evident that the JLab data would not help to choose between the allowed scenarios of quark-mass running within our model but would provide a good test of the model versus others. responding to the “low-scale” and “high-scale” soft/hard transition. The (µ,λ) parameters of these scenarios are shown in Fig. 1 by symbols. The low-scale scenario corresponds to the lowest transition scale which agrees, at the 95% CL, with the present data (µ = 2.79; λ = 0.715). The high-scale one is a typical representative point inside the allowed region (µ = 3.5; λ = 1.0). The corresponding F (Q2) functions are shown in Figs. 2, 3 by dashed π lines. Having assumed particular values for µ and λ, and therefore a particular F (Q2) model π curve, we simulate fake data points for a given experiment, scattered around the theoretical curve with the Gaussian distribution. The width of the distribution is determined by the error bars quoted in Refs. [25, 26]; the values of Q2 for these fake “measurements” are also 7 FIG.3. Therangeofpredictionsofourmodelforthechargedpionformfactorallowedbysimulated EIC data and model consistency for two particular scenarios for M(Q2) (dashed lines) correspond- ing to the “low-scale” and “high-scale” soft/hard transitions discussed in the text (gray bands bound by dotted lines). Existing experimental data points (see Ref. [1] for their description and list of references) are shown by black dots with thin error bars. Two typical simulated example realizationsof expected EICdata, correspondingtothe “high-scale”(black)and “low-scale”(gray) scenarios, are shown by dots with thick error bars. JLab 12-GeV simulated data are not shown for clarity. It is evident that the EIC data would make it possible to choose between the allowed scenarios of quark-mass running within our model. taken from there. Then, these fake data are processed in the joint chi-square fit with the existing data. The results are presented in Figs. 1, 2, 3. For12-GeVJLab, onemayseefromFig.2thattheexpectederrorbarsoftheF measure- π ments exceed the width of the region allowed by the present data for our model. Therefore, these data are not expected to contribute much into the determination of the soft/hard transition scale, as illustrated in Fig. 1 in terms of (µ,λ). However, we point out that the 12-GeV JLab data will be of crucial importance for testing the model itself. In Fig. 2, predictions of two alternative scenarios describing soft/hard transitions are also shown (see Ref. [1] for a more detailed discussion). Clearly, the precision of the expected 12-GeV JLab data would be sufficient to confirm or exclude the model we use here. Contrary, the 5-GeV EIC data, though having large expected error bars, will be able to 8 disentanglethelow-scaleandhigh-scaletransitionscenarioswithinourapproach, seeFigs.1, 3. The 95% CL constraints obtained in the two example scenarios are clearly separated. There still remains a formal degeneracy for Q2 (cid:38) 16 GeV2, but this range of momentum trans transfer, with the energy scale of order the b-quark mass, most probably corresponds to the perturbative QCD regime anyway. V. CONCLUSIONS We considered the effect of future experimental data on our understanding of the depen- dence of the pion electromagnetic form factor, F , on the momentum transfer squared, Q2, π paying a special attention to the “soft/hard” transition region where the QCD asymptotics should settle down. We took advantage of a particular low-energy model which describes excellently the existing data and predicts the QCD asymptotics automatically, without pa- rameter tuning. Given the estimated precision of 12-GeV JLab and of EIC, as well as the expected range of the momentum transfer accessible to the instruments, we conclude that the coming JLab data may confirm or exclude our model while the EIC measurements would be able to pin down the soft/hard transition scale. ACKNOWLEDGMENTS The work of ST was supported in part by the RFBR grant 12-02-01203. [1] S. V. Troitsky and V. E. Troitsky, “Transition from a relativistic constituent-quark model to the quantum-chromodynamical asymptotics: a quantitative description of the pion electro- magnetic form factor at intermediate values of the momentum transfer,” Phys. Rev. D 88 (2013) 093005 [arXiv:1310.1770 [hep-ph]]. [2] G. R. Farrar and D. R. Jackson, “The Pion Form-Factor,” Phys. Rev. Lett. 43 (1979) 246. [3] A. V. Efremov and A. V. Radyushkin, “Factorization and Asymptotical Behavior of Pion Form-Factor in QCD,” Phys. Lett. B 94 (1980) 245. [4] G. P. Lepage and S. J. Brodsky, “Exclusive Processes in Quantum Chromodynamics: Evolu- 9 tion Equations for Hadronic Wave Functions and the Form-Factors of Mesons,” Phys. Lett. B 87 (1979) 359. [5] A.F.KrutovandV.E.Troitsky,“Onapossibleestimationoftheconstituentquarkparameters from Jefferson Lab experiments on pion form-factor,” Eur. Phys. J. C 20 (2001) 71 [hep- ph/9811318]. [6] A. F. Krutov and V. E. Troitsky, “Relativistic instant form approach to the structure of two-body composite systems,” Phys. Rev. C 65 (2002) 045501 [hep-ph/0204053]. [7] A. F. Krutov and V. E. Troitsky, “Relativistic instant form approach to the structure of two- body composite systems. 2. Nonzero spin,” Phys. Rev. C 68 (2003) 018501 [hep-ph/0210046]. [8] A. F. Krutov, V. E. Troitsky and N. A. Tsirova, “Nonperturbative relativistic approach to pion form factor versus JLab experiments,” Phys. Rev. C 80 (2009) 055210 [arXiv:0910.3604 [nucl-th]]. [9] A. F. Krutov and V. E. Troitsky, “Instant form of Poincare-invariant quantum mechanics and description of the structure of composite systems,” Phys. Part. Nucl. 40 (2009) 136. [10] B. D. Keister and W. N. Polyzou, “Relativistic Hamiltonian dynamics in nuclear and particle physics,” Adv. Nucl. Phys. 20 (1991) 225. [11] S. R. Amendolia et al. [NA7 Collaboration], “A Measurement of the Space - Like Pion Elec- tromagnetic Form-Factor,” Nucl. Phys. B 277 (1986) 168. [12] A. F. Krutov and V. E. Troitsky, “Asymptotic estimates of the pion charge form-factor,” Theor. Math. Phys. 116 (1998) 907 [Teor. Mat. Fiz. 116 (1998) 215]. [13] V. A. Matveev, R. M. Muradyan and A. N. Tavkhelidze, “Automodelity in strong interac- tions,” Lett. Nuovo Cim. 5 (1972) 907. [14] S. J. Brodsky and G. R. Farrar, “Scaling Laws at Large Transverse Momentum,” Phys. Rev. Lett. 31 (1973) 1153. [15] A. P. Bakulev, A. V. Pimikov and N. G. Stefanis, “Pion Form Factor in QCD Sum Rules with Nonlocal Condensates and in the Local-Duality Approach,” Mod. Phys. Lett. A 24 (2009) 2848 [arXiv:0910.3077 [hep-ph]]. [16] G. P. Lepage and S. J. Brodsky, “Exclusive Processes in Perturbative Quantum Chromody- namics,” Phys. Rev. D 22 (1980) 2157. [17] L. S. Kisslinger, H. -M. Choi and C. -R. Ji, “Pion form-factor and quark mass evolution in a light front Bethe-Salpeter model,” Phys. Rev. D 63 (2001) 113005 [hep-ph/0101053]. 10