WorksPhysics and Astronomy
Photoproduction and detection of decays in ultra-peripheral collisions and at an electron-ion collider
Abstract
Vector meson photoproduction is an important probe of nuclear structure. Light vector mesons are most sensitive to low structure, as long as they are not too light for perturbative QCD calculations. The is of interest as an intermediate mass state (between the and ) that is easier to detect than the . Using HERA data on proton targets, we make projections for lead/gold targets in UPCs at the Large Hadron Collider and RHIC, and for and collisions at a future Electron-Ion Collider (EIC). We compare the UPC projections with ALICE data, and constrain the branching ratio divided by the square of the photon- coupling. The data prefer large couplings and small branching ratio, probably less than 25%. The photon-meson coupling predicted by generalized vector meson dominance does not fit the data. The HERA and ALICE UPC Pb data exhibit very similar mass spectra, indicating that, if the system is composed of two resonances, the products of their photon couplings with their four-pion branching ratios are similar. The predicted rates are high for both UPCs and the EIC. The decay can be observed at the EIC with high efficiency. In collisions at the highest energy, the forward B0 detector is needed to observe this channel down to the lowest achievable Bjorken values.
Introduction
Vector meson photoproduction has been studied extensively at fixed-target accelerators [1], the HERA collider [2], and with ultra-peripheral collisions (UPCs) at heavy-ion colliders [3]. It will also be an important probe of nucleons and nuclei at a future electron-ion collider (EIC) [4, 5, 6]. The Good-Walker paradigm relates coherent vector meson production to the average nuclear configuration, while incoherent vector meson photoproduction is related to fluctuations in the nuclear configuration, including gluonic hotspots [7, 8, 9].
The dependence of exclusive production is an important signature of saturation [10]. Definitive conclusions about saturation will require studies of different mesons, with different wave functions and masses.
The is straightforward to reconstruct [11], but from the theory perspective is rather light, limiting the applicability of perturbative QCD (pQCD) based calculations. The is heavy enough that saturation phenomena are greatly reduced [10]. The is attractive because it has an intermediate mass (between the and the ).
Early plans for exclusive vector meson production at a U. S. EIC focused on the and [4]. However, production at low is hard to reconstruct [5, 12] because the main channel, , suffers from a low value, with the daughter kaons having a momentum of only 127 MeV/ in the rest frame. Other final states have either low branching ratios or include a long-lived . For UPCs, the situation is similar, with coherent photoproduction difficult to observe [13, 14].
The states are attractive alternatives to the , as they also have intermediate masses, between the and . However, these states have a more complex wave function and a more complex phenomenology. There are likely two overlapping resonances, the and the [15]. These resonances can decay to many different final states, but both have a significant branching ratio to . Since the final state is easy to reconstruct, we will focus on it. Most photoproduction analyses have fit the mass spectrum to a single resonance, so we will perform do the same here.
The final state has been studied at fixed-target accelerators, using collisions at HERA and in ultra-peripheral collisions (UPCs) of gold ions at RHIC and lead ions at the LHC [16]. Using HERA and fixed-target data on proton targets as input, we will use a Glauber calculation to predict the RHIC and LHC cross sections for ion targets.
We also make predictions about the cross sections in and collisions at the EIC, and estimate the reconstruction efficiency using a simple model of the proposed ePIC detector. As with UPCs, the cross sections depend on the branching ratio.
Modeling of photoproduction
The states are radial excitations of the , with the same quantum numbers. The first studies of photoproduction were done in fixed-target experiments at photon energies from 3 GeV to 70 GeV[17, 18, 19, 20, 21, 22, 23]. Four pion photoproduction was first studied at collider energies by the Solenoidal Tracker at RHIC (STAR) collaboration [24], and later by the ALICE detector at the LHC [25]. The H1 collaboration has also studied 4-pion photoproduction, and observed a resonance with similar parameters to those measured by STAR. Both STAR and H1 found that the data was well fit by a single resonance [26, 27]. We characterize the resonance using the STAR results: mass MeV/, and width, MeV/, with a momentum-dependent width. We use the H1 cross section data as input to make predictions for ion targets in UPCs and at the EIC.
Proton targets
We fit H1 and fixed-target data on to a two-component Reggeon + Pomeron model [28, 29]:
Since the branching ratios are unknown, the data are for the cross sections times branching ratio.
The first term represents Pomeron exchange (with strength and power-law exponent ), while the second is for Reggeon exchange (with strength and power-law exponent ). Br is the branching ratio for the specified decay. Figure 1 shows the data, with both statistical and systematic errors. The H1 systematic uncertainties include point-to-point correlations. Also shown are three fits to the data, with the parameters given in Tab. I. These parameters assume a branching ratio of 100% for . For smaller branching ratios, and should be scaled up by dividing by the branching ratio.
In Model I, all four parameters in Eq. (1) were allowed to float. This led to a fairly large value of of 0.87, considerably above expectations for Pomeron exchange. In this fit, the Pomeron contribution was generally small, with the cross section dominated by the Reggeon component, even at H1 energies.
The second fit, Model II, added a nuisance term, based on the value for found by H1. A term was added to the in the fit, where and are taken from the fit in Fig. 4 of Ref. [27].

Figure 1. H1 [26, 27] and fixed-target data [17, 18, 23, 30, 31] on , along with our fit to the data. Model I is fit to Eq. (1) with all four parameters treated as free parameters. Model II is the reference fit used for the rest of the calculations in this paper. This fit has an additional term in the , pulling it toward the value from H1 [26, 27], as discussed in the text. Model III uses the set of parameters used in Ref. [32]. Fit results are given in Tab. I.
The rationale for this was that was dominated by systematic errors; the H1 fit correctly accounted for the correlated systematic errors in their analysis, so was more precise than if we just used their data points and errors. This fit pulled to 0.28, slightly larger than that found by H1, and many other studies of light mesons [33]. The downside is that this approach uses the H1 data twice - in the H1 and in our fit. This decreases the importance of the fixed-target data. Since that data is far away in energy from the region of interest, this should not cause problems.
The third fit, Model III, is from Ref. [32]. It differs from our fit in that it includes a slightly larger selection of fixed-target data, especially at very low energies. It found an intermediate between Model I and Model II.
Models I and III both have large values of , and so, at large energies, the cross section will grow overly rapidly, unlike other light vector mesons. They also have a rather large fractional Reggeon contribution ( is small), in tension with the H1 fit and expectations based on other vector mesons. For these reasons, the remainder of the paper uses the Model II fit.
The differences in predictions between the three fits for energies above 20 GeV are generally within 15%. The for Model 2 is 1.736, which is marginally higher than that of Model 1 (1.440) and significantly lower than that of Model 3 (3.049).
The branching ratio in Eq. (1) converts the cross section into the cross section. The branching fraction is important for ion targets, since a Glauber calculation, discussed below, does a non-linear mapping of into , the cross section on an
| Model I | 0.02 | 0.87 | 16.61 | 0.99 |
| Model II | 0.26 | 0.28 | 20.78 | 1.21 |
| Model III | 0.16 | 0.41 | 23.0 | 1.4 |
caption: TABLE I. Fit parameters, assuming a 100% branching ratio for . These fits results are for 100% branching ratio. For other branching ratios, and should be divided by the branching ratio.
ion target. Because of the non-linearity, it is necessary to remove the branching ratio before this mapping.
The branching ratio to is poorly known. Ref. [34] suggested , based on the inferred cross-section for decaying into in gold-gold collisions at RHIC, about one-third of that for the . Another study, estimated a branching ratio of (summing nonresonant and ) [35]. We consider branching ratios between 10% and 100%. 100% is the obvious maximum, while a branching ratio below 10% would require that the cross section is larger than that for the , which seems unlikely.
Production on ion targets
The cross section for photoproduction on nuclear targets is found using a quantum Glauber calculation [36]. The cross section for forward production, is equal to , where is the slope of at small . This forward cross section has two parts: the probability for a photon to fluctuate into the , and the -nucleon elastic scattering cross section:
We take , taken from the coherent (softer) exponent in Figure 6 of Ref. [27] to convert from the total cross section to .

Figure 6. Cross-sections for and collisions at EIC energies, presented as a function of rapidity for collisions at GeV (left) and collisions at /nucleon GeV (right). The inclusive (solid line) and various ranges (shaded areas) are indicated.
The vector-meson photon coupling is usually determined from the meson coupling to :
where is the vector meson mass, is the partial width for that meson to decay to and is the fine structure constant. The coupling is unmeasured for the .
As will be shown, the ratio of proton-target to ion-target cross sections are sensitive to . There are two prior estimates. Ref. [32] estimated and for the . Alternately, Generalized Vector Meson Dominance (GVMD) predicts that [37]
This leads to and , or about four times greater than Ref. [32]. This GVMD treatment lacks off-diagonal elements that couple different mesons, which may be important for [38]. The GVMD estimate of also leads to a surprisingly small cross section, as will be discussed below. The use of two resonances, each with the GVMD coupling, would not reduce these discrepancies. For these reasons, we use the coupling from Ref. [32] as our baseline, while considering other possibilities.
The optical theorem is then used to find the total cross section.
At , , assuming a 100% branching ratio, and 57 mb for a 10% branching ratio. In comparison, the cross sections for the and are 24 and 26 mb respectively [28]. It seems unlikely that the cross section is twice that of the . This disfavors scenarios with low branching ratios.
The cross section is then found using a quantum Glauber calculation [34]:
where is the nuclear thickness function.
The optical theorem can be used again, to convert into . Then, we use the nuclear form factor from Ref. [39] to get the total cross-section.
This is then combined with the photon flux in UPCs from Ref. [39] or for collisions in Ref. [40] to find the cross sections for UPC or collisions.
For PbPb collisions at LHC energies, using the two different estimates for results in cross sections that differ by about a factor of four. Varying the branching ratio from 10% to 100% changes the cross section by approximately a factor of five. In comparison, the 15% uncertainty from the choice of is negligible. As we will see, ALICE data significantly narrows the possible combinations.
' production in ultra-peripheral collisions
The cross section for photoproduction in UPCs is given by combining the (Eq. (1)) or cross sections (Eq. (7)) with the photon flux for protons [41] or ions [39]. This code is implemented in the STARlight Monte Carlo code [29].
Figure 3 shows the calculated for the for possible branching ratio between 10% and 100% and for the two different possible s. The calculations are compared with ALICE data on 5.02 TeV lead-lead collisions [25]. Unfortunately, the STAR data on the final state suffers from large experimental uncertainty [24] so it cannot contribute to the comparison.
For the Ref. [32] coupling, the data matches the cross section with about a 15 % branching ratio.
For the GVMD-predicted coupling, the best-fit branching ratio is unreasonably low, requiring a very large total cross section. The GVMD prediction in Eq. (4) must be too high. A GVMD calculation with off-diagonal elements might do better. Alternately, a Glauber-Gribov calculation with inelastic shadowing (cross-section fluctuations) might also fit the data better [42]. However, neither of these approaches is likely to lead to big enough changes to remove the tension.
Figure 4 shows the range of allowed values for vs. branching ratio. The red line shows the data, surrounded by and contours. If we assume that the branching ratio must be greater than 10%, then a large is preferred. However, the branching ratio cannot be too much more than 10%, unless the coupling is much larger than expected. As was previously noted, small branching ratios correspond to large meson-nucleon scattering cross-sections, with even a 10% branching ratio requiring a surprisingly large cross section. This leaves a fairly small preferred range, with branching ratios around 15% and coupling around the value from Ref. [32].
If the is composed of two resonances, then they are are likely to have different branching ratios and . However, as Figure 2 shows, the mass spectra for proton and lead targets are very similar. This indicates that the photon-meson couplings times the branching ratio to are similar for the two mesons. Otherwise, the Glauber mapping would distort the mass spectra for ion targets. More detailed comparisons of the resonance shape and substructure should clearly show if there is one resonance or two. High-statistics data from LHC Runs 3 and 4 [43] should allow for a definitive comparison.

Figure 2. Comparison of the invariant mass distribution of four pions from H1 collaboration (shown as gray squares) [26, 27] and the ALICE collaboration [25] in arbitrary units, normalized for easy shape comparison. Error bars are not shown for H1, but they are smaller than the error bars for ALICE.
We modeled the final state using phase space to determine the acceptance, using simple models for four LHC detectors. The model parameters are given in Table II, along with the resulting detector efficiencies. The efficiency for detecting would be slightly different than for this phase space distribution [24], but would not alter the conclusions here.
For all four detectors, the signal covers a much wider rapidity range than the detectors, reducing the overall acceptance. Figure 5 shows the signal acceptance in for the different detectors.

Figure 5. Differential cross-section for Pb-Pb UPCs at TeV. The solid blue line represents the results with no cuts applied, while the dotted red, dashed green, and dashed-dotted yellow lines correspond to the cross sections in the ALICE [25], LHCb [44], CMS [45] and expected ALICE 3 [46] acceptance, respectively.
| Experiment | Kinematic coverage | Acceptance |
| ALICE | , | 2.2% |
| LHCb | , | 1.8% |
| CMS | , | 12% |
| ALICE 3 | , | 43% |
Table II. Geometric acceptance criteria for four LHC experiments [25, 44, 45, 46] with their respective pseudorapidity () and transverse momentum () acceptance for charged particles, as well as the corresponding acceptances for .
photoproduction in collisions
Photoproduction in collisions adds another dimension: the photon . There is no data on cross sections for virtual photons, so we model the evolution of the cross-section following the [40, 47]:
where .
production is assumed to follow vector meson dominance, with the final state linearly polarized transverse to the beam direction at , but with an increasing longitudinal polarization as rises. The rate of this increase is not known for the , but we use the approach described in Ref. [40] here, assuming that the spin matrix for the is the same as for the .
Simulations were performed in the eSTARlight framework [40]. Table III shows the calculated cross sections for the top EIC energies: 18 GeV electrons colliding with 275 GeV protons or 110 GeV/nucleon lead ions. In addition to the total ( integrated) cross section, cross sections are given for photoproduction () and electroproduction (). The total cross section is dominated by photoproduction. Figure 6 shows the corresponding .

(a)

(b)
Figure 3. Differential cross-section of for Pb-Pb UPCs at TeV (top) and Pb collisions at GeV (bottom). The curves show different branching ratios (B.R.) of 100%, 50%, 30%, and 10%. The orange marker indicates the ALICE data for the Pb-Pb collisions at TeV. For the UPCs (top plot), the left hand curves () use the coupling in Ref. [32], while the right hand curves () use the GVMD predictions. For Pb collisions (bottom plot), we use the coupling in Ref. [32].

Figure 4. Two-dimensional scan of the differential cross section (in mb) as a function of the branching ratio (B.R.) and the coupling . The color map shows the interpolated cross section values on a logarithmic scale. Overlaid lines indicate the central value of the ALICE measurement (red solid) together with the (black), (blue), and (orange) contours, where the bands represent the ALICE measurement uncertainties (stat syst). Horizontal solid black lines mark the reference couplings and .
The cross section for production in Pb collisions is about 45 times that for collisions. This is a smaller ratio than for the or , but is similar to the ratio found for and collisions in Ref. [32]. For collisions, the cross-section in collisions is about 1/7 of the cross section, and about 1/3 of the cross section. For Pb collisions, the ratios are lower, with the cross section 1/30 of the cross section, and half that for the . In both cases, the rates are high enough (7 or 1.5 billion events per of integrated luminosity) to allow high-precision differential measurements.
The rates are high enough to measure final states with small branching ratios, including . With this measurement of , it will be possible to determine the branching ratio. This data will also make it possible to compare high and low mass and final states, to determine if there are one or two resonances.
The reconstruction efficiency for the EPIC detector is shown in Figure 7. ePIC is modeled with two tracking components: a central barrel tracker sensitive to and the B0 detector, which covers . For both components, we required track MeV/, but this cut had little effect on the efficiency. Figure 7 show the geometric detection efficiency for photoproduction events. We do not consider detector inefficiency within the geometric acceptance. This inefficiency should be small.

Figure 7. Accessible cross sections for collisions at GeV (left) and collisions at /nucleon GeV (right), with specific kinematic cuts based on the current EPIC detector tracker design. The blue shaded area with dotted lines represents the coverage including the B0 tracker.
The efficiency is high for , but falls off at larger , as expected. The detector plays an important role for collisions in the region ; otherwise the efficiency would be near zero for . Even though the efficiency is fairly low in this region, the rates are high enough that high statistics data should be achievable.
The rapidity is related to the target Bjorken-
where is the proton mass and is the Lorentz boost of the proton beam at maximum EIC energy. For UPCs, a similar relation holds, except that there is a sign in the exponential, since the photon direction is unknown.
The central detector cutoff, , corresponds roughly to (as can be seen in Figure 7), or .
For collisions, covering the full range of Bjorken- requires acceptance out to rapidity , corresponding to . As Figure 7 shows, the B0 tracker is needed to cover this low- region. Nuclear shadowing (beyond that present in the Glauber calculation) or saturation would manifest itself as a reduction in cross-section with decreasing , i.e. with increasing rapidity at low/moderate [10]. There would also be changes in [4]. The rates are high enough to permit these measurements, even at the lowest .
In short, the is copiously produced and relatively easy to reconstruct, showing promise for use in saturation studies. The rates given in Table III assume a branching ratio of 100%. For a more likely branching ratio around 10%, the rates are reduced by a factor of about 8, but are still high enough for high-statistics measurements.
| Collision System | Range (GeV) | Total Cross Section | Events for () | Acceptance | Acceptance with |
| at GeV | Inclusive | 696 nb | 0.39 | 0.41 | |
| 686 nb | 0.39 | 0.41 | |||
| 10.0 nb | 0.52 | 0.53 | |||
| at GeV | Inclusive | 31.3 b | 0.73 | 0.73 | |
| 30.6 b | 0.73 | 0.73 | |||
| 0.639 b | 0.76 | 0.76 |
Table III. Projected cross sections and event rates for and collisions at EIC energies, based on the current EPIC detector design, for different ranges. The integrated luminosities () used are for and for collisions. Acceptance values are based on simulations using the current detector design.
As with UPCs, the main uncertainties are from the unknown and branching ratios. Fortunately, the coupling to can be directly determined experimentally, by measuring that final state. Other uncertainties are much smaller, making the an excellent medium-mass candidate for mapping out shadowing as a function of
Discussion and conclusions
We have calculated the cross section for on ion targets, using proton-target data for photoproduction on proton targets as input. The resulting ion-target cross sections depend on the photon-meson couplings and the branching ratio for the to decay to . The couplings predicted using a GVMD model lead ion-target cross-sections that are high enough to require implausibly small branching ratios. Using the coupling from Ref. [32], ALICE data on , prefers branching ratios around 15%. This estimate uses a Glauber calculation, but small deviations will not have a significant effect on this conclusion. There is some tension between the cross-section ratio and the inferred meson-nucleon scattering cross section.
With LHC Run 3/4 and EIC data, it should also be possible to measure the coupling to , using higher pairs from incoherent photoproduction, to avoid backgrounds from . It should also be possible to investigate whether a GVMD model with off-diagonal intermediate states might better predict the photon-meson couplings [48, 49, 50].
We also made predictions for at the proposed U. S. EIC. The production rates are high, and the meson is an easy-to-reconstruct probe of nuclear structure. The is intermediate in mass between the and the , so it should show significant saturation, but also be amenable to pQCD calculations. These features should also make it an attractive target at the proposed Chinese electron-ion collider, EiCC [51] or the proposed LHeC collider [52]. For the LHeC, excellent forward instrumentation is critical to be able to observe photoproduction and electroproduction at the highest energies, corresponding to the lowest Bjorken-. The wide range at the LHeC data will better constrain the energy dependence of photoproduction.
The approach developed here, comparing and collisions may be applicable for determining absolute branching ratios for other mesons, as long as the photon-meson coupling is known.
This work is supported in part by the U.S. Department of Energy, Office of Science, Office of Nuclear Physics, under contract numbers DE-AC02-05CH11231 and DE-FG02-96ER40991.
References
- [1]T. H. Bauer, R. D. Spital, D. R. Yennie, and F. M. Pipkin, Rev. Mod. Phys. 50, 261 (1978), [Erratum: Rev.Mod.Phys. 51, 407 (1979)].
- [2]I. P. Ivanov, N. N. Nikolaev, and A. A. Savin, Phys. Part. Nucl. 37, 1 (2006), arXiv:hep-ph/0501034.
- [3]S. Klein and P. Steinberg, Ann. Rev. Nucl. Part. Sci. 70, 323 (2020), arXiv:2005.01872 [nucl-ex].DOI
- [4]A. Accardi et al., Eur. Phys. J. A 52, 268 (2016), arXiv:1212.1701 [nucl-ex].
- [5]R. Abdul Khalek et al., (2021), arXiv:2103.05419 [physics.ins-det].DOI
- [6]J. Adam et al. (ATHENA), JINST 17, P10019 (2022), arXiv:2210.09048 [physics.ins-det].arxiv.org/abs/2210.09048
- [7]H. I. Miettinen and J. Pumplin, Phys. Rev. D 18, 1696 (1978).
- [8]H. Mäntysaari and B. Schenke, Phys. Rev. Lett. 117, 052301 (2016), arXiv:1603.04349 [hep-ph].DOI
- [9]S. R. Klein and H. Mäntysaari, Nature Rev. Phys. 1, 662 (2019), arXiv:1910.10858 [hep-ex].DOI
- [10]H. Mäntysaari and R. Venugopalan, Phys. Lett. B 781, 664 (2018), arXiv:1712.02508 [nucl-th].DOI
- [11]C. Adler et al. (STAR), Phys. Rev. Lett. 89, 272302 (2002), arXiv:nucl-ex/0206004.DOI
- [12]J. Arrington et al., (2021), arXiv:2102.08337 [nuc-ex].arxiv.org/abs/2102.08337
- [13]S. Acharya et al. (ALICE), Phys. Rev. Lett. 132, 222303 (2024), arXiv:2311.11792 [nucl-ex].
- [14]V. Chekhovsky et al. (CMS), (2025), arXiv:2504.05193 [nucl-ex].
- [15]R. L. Workman et al. (Particle Data Group), PTEP 2022, 083C01 (2022).
- [16]S. Navas et al. (Particle Data Group), Phys. Rev. D 110, 030001 (2024).
- [17]H. H. Bingham et al., Phys. Lett. B 41, 635 (1972).
- [18]P. Schacht, I. Derado, D. C. Fries, J. Park, and D. Yount, Nucl. Phys. B 81, 205 (1974).
- [19]G. Alexander, O. Benary, J. Gandsman, D. Lissauer, A. Levy, Y. Oren, and L. M. Rosenstein, Phys. Lett. B 57, 487 (1975).
- [20]D. P. Barber et al. (LAMP2 Group), Z. Phys. C 4, 169 (1980).
- [21]D. Aston et al. (Bonn-CERN-Ecole Poly-Glasgow-Lancaster-Manchester-Orsay-Paris-Rutherford-Sheffield), Nucl. Phys. B 189, 15 (1981).
- [22]M. Atkinson et al. (Omega Photon), Z. Phys. C 26, 499 (1985).
- [23]M. S. Atiya et al., Phys. Rev. Lett. 43, 1691 (1979).
- [24]B. I. Abelev et al. (STAR), Phys. Rev. C 81, 044901 (2010), arXiv:0912.0604 [nucl-ex].
- [25]S. Acharya et al. (ALICE), (2024), arXiv:2404.07542 [nucl-ex].
- [26]S. Schmitt (Presented at Deep Inelastic Scattering, 2018).
- [27]H1 Collaboration, “Exclusive photoproduction of 2π+2π− final state at her”, (2018), H1 prelim 18-211.
- [28]S. Klein and J. Nystrand, Phys. Rev. C 60, 014903 (1999), arXiv:hep-ph/9902259.DOI
- [29]S. R. Klein, J. Nystrand, J. Seger, Y. Gorbunov, and J. Butterworth, Comput. Phys. Commun. 212, 258 (2017), arXiv:1607.03838 [hep-ph].DOI
- [30]M. Davier, I. Derado, D. Fries, F. Liu, and R. Mozley, Nuclear Physics B 58, 31 (1973).
- [31]D. Aston, M. Atkinson, R. Bailey, A. Ball, and B. Bouquet, Nuclear Physics B 189, 15 (1981).
- [32]M. Klusek-Gawenda and J. D. Tapia Takaki, Acta Phys. Polon. B 51, 1393 (2020), arXiv:2005.13624 [hep-ph].
- [33]D. Szuba (H1, ZEUS), PoS EPS-HEP2009, 333 (2009).
- [34]L. Frankfurt, M. Strikman, and M. Zhalov, Phys. Rev. C 67, 034901 (2003), arXiv:hep-ph/0210303.DOI
- [35]V. Andreev et al. (H1), Eur. Phys. J. C 80, 1189 (2020), arXiv:2005.14471 [hep-ex].
- [36]This differs from Ref. [32] which used a classical Glauber calculation.
- [37]L. Frankfurt, V. Guzey, and M. Strikman, Phys. Rev. D 58, 094039 (1998), arXiv:hep-ph/9712339.
- [38]J. M. Bronstein, High-Energy Photoproduction of the ρ and ρ′ Vector Mesons, Ph.D. thesis, Illinois U., Urbana (1977).
- [39]S. R. Klein and J. Nystrand, Phys. Rev. Lett. 84, 2330 (2000), arXiv:hep-ph/9909237.DOI
- [40]M. Lomnitz and S. Klein, Phys. Rev. C 99, 015203 (2019), arXiv:1803.06420 [nucl-ex].DOI
- [41]S. R. Klein and J. Nystrand, Phys. Rev. Lett. 92, 142003 (2004), arXiv:hep-ph/0311164.DOI
- [42]L. Frankfurt, V. Guzey, M. Strikman, and M. Zhalov, Phys. Lett. B 752, 51 (2016), arXiv:1506.07150 [hep-ph].DOI
- [43]Z. Citron et al., CERN Yellow Rep. 7, 1159 (2019), arXiv:1812.06772 [hep-ph].arxiv.org/abs/1812.06772
- [44]LHCb-PAPER-2024-042.
- [45]CMS-DP-2024-011.
- [46]Letter of intent for ALICE 3: A next generation heavy-ion experiment at the LHC, Tech. Rep. (CERN, Geneva, 2022) arXiv:2211.02491.arxiv.org/abs/2211.02491
- [47]F. D. Aaron et al. (H1), JHEP 05, 032 (2010), arXiv:0910.5831 [hep-ex].
- [48]T. Kobayashi, Prog. Theor. Phys. 49, 282 (1973).
- [49]H. Fraas and M. Kuroda, Phys. Rev. D 16, 2778 (1977).
- [50]L. Frankfurt, M. Strikman, and M. Zhalov, Acta Phys. Polon. B 34, 3215 (2003), arXiv:hep-ph/0304301.arxiv.org/abs/hep-ph/0304301
- [51]X. Chen, F.-K. Guo, C. D. Roberts, and R. Wang, Few Body Syst. 61, 43 (2020), arXiv:2008.00102 [hep-ph].DOI
- [52]P. Agostini et al. (LHeC, FCC-he Study Group), J. Phys. G 48, 110501 (2021), arXiv:2007.14491 [hep-ex].DOI