WorksPhysics and Astronomy
Polarized Nuclear DVCS at the EIC
Abstract
The Electron-Ion Collider (EIC) will enable a series of measurements at unprecedented energies and luminosities, providing new opportunities to investigate the microscopic structure of nucleons and nuclei at small . Exclusive processes such as Deeply Virtual Compton Scattering (DVCS) offer unique access to the three-dimensional structure of hadrons through Generalized Parton Distributions (GPDs), while polarized electron and ion beams further enable detailed studies of spin-dependent structure. A model for coherent DVCS on polarized Heis developed and applied to simulations of for -GeV He collisions at the EIC. Using this framework, the statistical precision achievable is estimated for measurements of beam-spin asymmetries and for the extraction of the Compton Form Factors (CFFs) and . Early EIC data are found to enable precise differential measurements of the unpolarized CFF and to provide significant constraints on its real and imaginary components. By contrast, meaningful constraints on the polarized CFF require substantially larger integrated luminosities. The kinematics of the recoil He nucleus are also examined, and the far-forward detector capabilities at the EIC required to tag the intact nucleus and perform fully exclusive measurements of coherent nuclear DVCS are discussed.
Introduction
An important aim of nuclear physics in the modern era is to advance our understanding of the three-dimensional partonic structure of hadrons, encoded within the framework of Generalized Parton Distributions (GPDs) [1, 2, 3, 4, 5, 6], which carry both momentum and spatial information about the quarks and gluons within a hadronic state. GPDs can be accessed via a number of exclusive scattering processes, the most commonly used of which is Deeply Virtual Compton Scattering (DVCS). DVCS is the hard exclusive electroproduction of a real photon from a hadronic target and represents a clean probe of GPDs. Many measurements of this process have been performed and are ongoing at HERA, HERMES, JLab, and COMPASS [7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29].
The majority of existing DVCS measurements have been carried out on the proton and are aimed at accessing its GPDs. Relatively few measurements exist to constrain the three-dimensional partonic structure of nuclei [30, 31], and even fewer can reliably distinguish the coherent signal from the incoherent background. The partonic structure of nuclei remains an outstanding topic in nuclear physics [32, 33, 34, 35, 36, 37, 38, 39], and DVCS has the potential to broaden this topic into the area of tomography.
Exclusive processes such as DVCS are an important component of the physics program of the future Electron-Ion Collider (EIC) [40, 41, 42, 43, 44, 45], which aims to constrain the three-dimensional structure of the proton and nuclei across a significantly broader kinematic range in and than that accessible with existing data sets, as shown in Figure 1. The EIC will utilize a series of polarized and unpolarized electron-proton () and electron-nucleus () collisions to access higher center-of-mass energies (20–141 GeV) [40] than those available in fixed-target experiments. The current design of the EIC project detector, ePIC, can precisely determine the kinematics of key observables, such as the incoming and scattering electrons
Figure 1. Kinematic coverage in and of coherent nuclear DVCS measurements. Blue circles and red squares show the coverage of existing measurements from CLAS [30] and HERMES [31] respectively. The shaded regions show the kinematic coverage possible with different electron-ion systems at the EIC, including at 9 GeV166 GeV/nucleon (pink), at 9 GeV130 GeV/nucleon (yellow), and at 9 GeV100 GeV/nucleon (green), each requiring .
and the produced real photon in the DVCS process. The ePIC far-forward region detectors, which cover small angles mrad in the forward hadron-going direction [40], are crucial for performing exclusive measurements by tagging particles experiencing very small momentum-transfer. This small-angle tagging is also a key capability enabling the EIC to perform exclusive measurements of nuclei. By measuring nucleons and nuclear fragments from the dissociation of beam nucleus, coherent and incoherent nuclear reactions may be distinguished.
Formalism and Model
The theoretical formalism relating the DVCS cross section to the target GPDs has been extensively developed [3, 46, 47, 48], including its extension to nuclear targets [49]. In this study, we follow the leading-twist formulations of Ref. [50] and Ref. [47] for spin-0 and spin-1/2 hadrons, respectively, while neglecting contributions from gluon transversity. The total cross section for the exclusive leptoproduction of a photon from nucleus , , is given by
Here we define the lepton momentum-transfer , the hadron momentum-transfer , the nuclear Bjorken variable , the lepton energy-fraction , and the target-mass kinematic factor , along with the azimuthal angle between the lepton and hadron scattering planes in the nuclear rest frame, where we follow the Trento convention [51] in the following. Also relevant is the “skewness” , which characterizes the longitudinal momentum-transfer to the nucleus. The lepton has initial and final four-momentum and , while the nucleus has initial and final four-momentum and , respectively. Figure 2 illustrates the kinematics of the reaction as well as the convention for used here.
Figure 2. Kinematics of the coherent DVCS process in the initial-state ion rest frame. The azimuthal angle is defined between the lepton and hadron planes using the Trento convention.
A key feature of the DVCS process is its irreducible interference with the purely QED Bethe Heitler (BH) process, in which the real photon is emitted by either the incoming or outgoing lepton. As a result, the squared amplitude for the process can be written as
where DVCS-BH interference term is defined by
In practice, the BH contribution dominates over most of the accessible kinematic phase space. Consequently, the DVCS amplitude is most effectively accessed through the DVCS-BH interference term, which can be isolated via polarization and/or lepton-charge asymmetries.
More generally, the BH, DVCS, and interference contributions to the cross section can be decomposed into Fourier harmonics of the hadronic azimuthal angle :
where the prefactors and the harmonic coefficients and differ for BH, DVCS, and the interference term. The harmonic coefficients have different expressions for spin-0 and spin-1/2, given by Ref. [50] and Ref. [47] respectively, neglecting higher-order kinematical effects [52] which are suppressed in the high-energy kinematics of the EIC. The BH coefficients depend bilinearly on the electromagnetic form factors of the hadron, while the DVCS coefficients depend bilinearly on the “Compton Form Factors” (CFFs), which are related to sums over the individual quark GPDs within a target weighted by the square of the quark charge. The interference coefficients are determined by products of electromagnetic and Compton form factors. In general, they also depend on the lepton helicity and the polarization state of the target nucleus.
In this study, we focus on the nuclei and , which have spin 0 and spin 1/2, respectively. For , we employ the charge form factor calculated in Ref. [53]. For , we use the electric and magnetic form factors obtained from the phenomenological fits of Ref. [54].
Nuclear CFFs are considerably less constrained than their nucleonic counterparts, with only limited experimental data and theoretical calculations currently available. In this work, we evaluated the nuclear CFFs with the impulse approximation, expressing them as convolutions of the corresponding nucleon CFFs. At leading twist, a spin-1/2 nucleus is characterized by four CFFs. The dominant contribution, which is also the only leading-twist CFF for a spin-0 nucleus, is the unpolarized helicity-conserving CFF . For both and , the nuclear CFF can be related to the nucleon CFFs through the convolution formalism of [55, 56]:
where is the off-diagonal light-cone momentum distribution for nucleon in nucleus . For small values of the momentum-transfer and skewness , the off-forward light-cone momentum distribution factorizes approximately as:
Here, is the light-cone momentum distribution of nucleon in nucleus , with momentum fraction , while characterizes the spatial distribution of nucleon through its contribution to the nuclear form factor.
For and other spin-1/2 nuclei, an additional leading-twist CFF, the polarized helicity-conserving CFF , contributes. Within the impulse approximation, this CFF can be expressed in terms of the corresponding nucleon CFFs:
where is the net polarized off-diagonal light-cone momentum distribution. In the absence of detailed spin-dependent nuclear structure calculations, we approximate the polarized off-forward distribution by multiplying the corresponding unpolarized distribution by the effective nucleon polarization . For , we adopt and [57]. In addition, spin-1/2 nuclei are characterized by unpolarized and polarized helicity-flip CFFs and . These quantities contribute to transverse-spin observables and are sensitive to more complex nuclear dynamics. Accordingly, they are not considered in the present analysis.
Combining the cross-section formalism outlined above with appropriate models of nucleon and nuclear structure, we construct complete descriptions of coherent photon leptoproduction from spin-0 and spin-1/2 nuclei. The resulting cross sections are implemented in a Monte Carlo event generator to produce pseudodata samples, which can be used both to benchmark the model against existing measurements and to evaluate projected rates and sensitivities for future experiments. In the following sections, we first compare our calculations for with measurements from HERMES and CLAS experiments to assess the validity of the convolution approach employed in this study. We then apply the validated framework to predict coherent DVCS observables for polarized and to evaluate the corresponding experimental sensitivities at the Electron-Ion Collider (EIC) using the ePIC detector acceptance.
As input to the model, we employ the KM15 parametrization of the nucleon CFFs [58], using the parameter values reported in Ref. [59]. The nuclear form factors are obtained by Fourier transforming the coordinate-space nucleon densities of Ref. [60]. For , the nonrelativistic spectral function of Ref. [61] is used to construct the associated light-cone momentum distribution. The light-cone momentum fraction is approximated by
which exhibits the correct asymptotic behavior and remains well behaved in the limit of large internal nuclear momentum . For , light-cone momentum density is derived from the momentum distribution of Ref. [60], assuming that the system remains bound. To account for nuclear shadowing effects, we further implement the -dependent rescaling of introduced in Ref. [62] when evaluating the nuclear CFFs.
We emphasize that the process considered in this work is coherent DVCS, in which the nucleus remains intact in the final state. This exclusivity condition is essential for isolating the static off-forward components of the nucleus encoded in its generalized parton distributions. By contrast, the incoherent DVCS (and BH) process, , involves nucleus excitation or breakup and is therefore sensitive to nuclear fluctuations and the properties of bound nucleons. Consequently, coherent and incoherent photon-production processes must be distinguished experimentally. Although incoherent scattering is not considered explicitly in the present study, it constitutes a potential background to coherent DVCS when complete final-state exclusivity cannot be enforced. This background becomes increasing important with increasing and eventually dominates the total cross section at sufficiently large momentum transfer.
Comparison with He in Fixed-Target Experiments
To validate both the model described above and its implementation in the event generator, we compare our calculations for with existing measurements from CLAS [30] and HERMES [31]. In both experiments, information on the nuclear CFF was accessed through the lepton single-spin asymmetry,
where denotes the differential cross section for positive (negative) lepton helicity, and is the degree of lepton polarization. Changing the helicity of the lepton beam does not affect the BH or DVCS contributions to the total cross section, but it reverses the sign of the helicity-dependent harmonic coefficients in the interference term, most notably . Since the total cross section is typically dominated by the BH contribution, the beam-spin asymmetry can be approximated, to leading order, by
These measurements therefore focused on the harmonic coefficient, , in order to constrain the imaginary component of as a function of the kinematic variables. Throughout this study, we extract this coefficient by fitting the -dependence of within a given kinematic bin according to
where higher-order harmonic contributions have been neglected, apart from the leading modulation in the denominator. The latter arises primarily from the azimuthal dependence of the unpolarized cross section and retains some sensitivity to the real part of ().
In Figure 3, we compare the predictions of our model with the results reported in Refs. [30, 31]. For the HERMES measurements, the event selection is not fully exclusive and therefore likely contains some contamination from incoherent scattering. Accordingly, we restrict our comparison to the region , where the incoherent contribution is expected to be small. In addition, HERMES used a positron beam rather than an electron beam. Since the DVCS-BH interference term changes sign with the lepton charge, the measured beam-spin asymmetry also reverses sign. To facilitate a direct comparison with our calculations, which assume an electron beam, we therefore compare against for the HERMES results. Overall, the calculations are broadly consistent with the measured data, although they tend to systematically overpredict the magnitude of the observed asymmetry. Such deviation may arise from higher-order kinematic effects beyond the present convolution approximation or from additional nuclear-structure effects that are not included in the current model.
Figure 3. measurements from He using 6-GeV electrons at CLAS [30] (blue circles) and 27.6-GeV positrons at HERMES [31] (red squares), compared with calculations using matching kinematics (blue solid and red dashed lines, respectively), as a function of momentum-transfer .
Physics Projection for Polarized He at the EIC
At the EIC, a wide variety of ion beams with be available for nuclear physics studies [40, 63], including He, which serves as an effective source of highly polarized neutrons. For a spin- nucleus such as He, the nuclear spin gives rise to additional asymmetries beyond the lepton asymmetry . These observables provide sensitivity to the polarized CFFs and , which are probed by longitudinal and transverse nuclear polarization, respectively. The simplest nuclear-spin observable, which we investigate in this section, is the nuclear longitudinal single-spin asymmetry,
where is the differential cross section for a target nucleus polarized parallel (antiparallel) to the lepton beam, and is the degree of nuclear polarization. At leading order, this asymmetry follows
where here denotes the interference -coefficient in the presence of longitudinal polarization of the target nucleus. In addition to the longitudinal beam-spin and target-spin asymmetries discussed above, one may also define the longitudinal double-spin asymmetry,
which is sensitive to the real part of (). Additional observables involving transverse nuclear asymmetries provide access to the helicity-flip CFFs and . These asymmetries are beyond the scope of the present work and are therefore deferred to future studies.
In order to assess the sensitivity of the EIC to this process, we generate events corresponding to the anticipated beam configuration during the early stages of EIC operation, consisting of a 9-GeV electron beam colliding with a 166-GeV/nucleon He beam. Consistent with the current EIC accelerator design, both the electron and ion beam polarizations are assumed to be . Events are generated within the kinematic range , , and . Final-state electrons and photons are required to fall within the central ePIC detector acceptance, defined by and . The scattered He ions are required to satisfy mrad with respect to the beamline, matching the acceptance of the Roman Pot detectors in the ePIC far-forward region [40]. Events passing these selection criteria are weighted according to the total cross section and the assumed integrated luminosity. The resulting event yields are then used to estimate rates in individual kinematic bins and to evaluate the projected sensitivity to asymmetries and CFFs.
Figure 4 shows the projected measurement of the electron-spin asymmetry as a function of in several bins of , assuming the early-science integrated luminosity of divided evenly between positive- and negative-helicity electron-beam configurations. Event yields are calculated in bins of , , and , and the -dependent fit of Eq. (11) is used to propagate the statistical uncertainties to the extracted moment . The predicted electron-spin asymmetry increases markedly with decreasing , reflecting the growth of . The coherent Bethe–Heitler and DVCS event rates are expected to be sufficiently large to enable a precise measurement of the dependence of up to approximately , beyond which the coherent cross section falls rapidly. We also observe that, at small , the growth of both and causes the DVCS contribution to become increasingly important relative to the Bethe–Heitler process. Consequently, the total cross section is no longer strongly dominated by Bethe–Heitler, and the leading-order approximations for the beam- and target-spin asymmetries cease to provide an accurate description.
Figure 4. Projected electron-spin asymmetry -moment for collisions at GeV, presented as a function of momentum-transfer in several bins of . Uncertainties are the projected statistical uncertainties assuming of luminosity and electron polarization of .
To determine the statistical precision with which and can be extracted in this nonlinear regime, we employ a more complete description of the -dependence of . Following the approach of [30], we parameterize the asymmetry as
with
For a spin-1/2 nucleus such as , additional contributions involving , , and are present. However, these terms are suppressed by powers of and are therefore neglected in the present analysis. This parameterization, together with the projected event yields in bins of , , and , enables the propagation of statistical uncertainties on the measured yields to the extracted values of and . While is primarily constrained by the odd harmonic components of , is constrained more weakly through the even harmonic terms appearing in the denominator. Figure 5 shows the projected measurements of as a function of for several values of the skewness , along with an inset displaying the corresponding projections for . The nonlinear dependence of on the CFFs in this kinematic regime, combined with the presence of as a nuisance parameter in the fit, leads to larger relative uncertainties on than on the extracted moment . We predict that an integrated luminosity of will enable precise differential measurements of up to at several values of skewness . The corresponding constraints on are expected to be less precise, although additional sensitivity may be obtained through measurements of the absolute cross section. Such measurements, however, are experimentally more challenging and are not considered further in this study.

Figure 5. Main figure: Projected measurements of using collisions at GeV, presented as a function of momentum-transfer at several values of skewness . Uncertainties are the projected statistical uncertainties assuming of luminosity and electron polarization of . Inset: Projected measurements of using the same settings and luminosity.
The ion-spin asymmetry and the resulting sensitivity to are also examined. The event yields as a function of , , and are used to determine the statistical uncertainties on the extracted value of in the same manner as for . Owing to the substantially smaller asymmetry signal, we assume an integrated luminosity of , divided equally between the two nuclear-polarization settings configurations. Although the -dependence of is sensitive to and in addition to , we assume that the components of have been constrained independently through measurements of . Under this assumption, the statistical uncertainties on dominate the extraction of . Figure 6 shows the projected measurements of as a function of for several values of the skewness . The projected precision on is significantly worse than that obtained for either or , despite the larger integrated luminosity assumed in this study. This reduced sensitivity arises from the relatively small magnitude of compared with , which leads to a substantially weaker target-spin asymmetry than the corresponding electron-spin asymmetry. With an integrated luminosity of , the projected measurements provide meaningful constraints on the and dependence of , particularly at larger values of and . More precise measurements of would require substantially larger data sets of to be collected at EIC.

Figure 6. Projected measurements of using collisions at GeV, presented as a function of momentum-transfer at several values of skewness . Uncertainties are the projected statistical uncertainties assuming of luminosity and ion polarization of . A dashed grey line is shown at for reference.
An important experimental requirement for DVCS measurements is the exclusive reconstruction of final-state particles, including detection of the scattered ion. Such exclusivity is crucial for suppressing the substantial background from production, and in the case of nuclear DVCS, for separating the coherent process from incoherent nuclear breakup.
The capability of the ePIC far-forward detectors to probe the scattered ion is examined. Figure 7 presents the angular distribution of the final-state ions in simulated events. The expected scattering angles are found to lie well within the nominal outer acceptance of the Roman Pot detectors. At the same time, the small momentum-transfer typical of coherent scattering results in ions are often nearly collinear with the beam. Consequently, the efficiency for detecting these ions will be driven by the minimum-angle acceptance of the Roman Pot detectors near the beamline, which has not yet been fully established. The ability to detect ions with such small deflections should therefore be regarded as an important design consideration for the far-forward detectors at the EIC. This capability is required not only for coherent nuclear DVCS, but also for a broader program of exclusive nuclear measurements, including Deeply Virtual Meson Production [64, 65, 66] and the photoproduction of heavy vector mesons [67, 68].
Figure 7. Angular distribution of the final-state ion in at GeV.
Conclusion
We have presented a model for coherent DVCS on polarized and applied it to simulations of the process at the EIC. Projections for measurements of the electron-spin asymmetry from , as well as for the extraction of CFF components , , and , have been performed using this framework. Our results indicate that early data will enable precise differential measurements of and provide significant constraints on . By contrast, meaningful constraints on will require substantially larger integrated luminosities, particularly for access to small . The kinematics of the scattered ion in this process has been examined as well. The exclusivity requirement of the coherent nuclear process places stringent demands on the minimal-angle acceptance of the EIC far-forward detectors in order to enable precision measurements of coherent nuclear DVCS. Looking ahead, the present framework can be incorporated into a full ePIC detector simulation to quantify detector acceptance, reconstruction efficiencies, and systematic effects under realistic experimental conditions. The model may also be extended to include transverse-spin observables, providing access to the helicity-flip CFFs and and enabling a more complete investigation of the spin structure of . An equally important direction is the development of a corresponding description of incoherent nuclear DVCS. In addition to constituting a background to coherent scattering, the incoherent process provides direct sensitivity to the structure of bound nucleons and offers unique opportunities for polarized-neutron tomography. Together, coherent and incoherent DVCS on polarized will provide complementary probes of nuclear and nucleonic structure, forming a unified program for three-dimensional imaging at the EIC.
Acknowledgments
This work was supported by the U.S. Department of Energy through the Los Alamos National Laboratory. Los Alamos National Laboratory is operated by Triad National Security, LLC, for the National Nuclear Security Administration of U.S. Department of Energy (Contract No. 89233218CNA000001). Research presented in this article was supported by the Laboratory Directed Research and Development program of Los Alamos National Laboratory under project number 20260377ER.
References
References
- [1]D. Müller, D. Robaschik, B. Geyer, F.-M. Dittes, and J. Hořejší, Fortschr. Phys. 42, 101 (1994).
- [2]X.-D. Ji, Phys. Rev. Lett 78, 610 (1997).
- [3]X.-D. Ji, Phys. Rev. D 55, 7114 (1997).
- [4]A. Radyushkin, Phys. Lett. B 385, 333 (1996).
- [5]A. Radyushkin, Phys. Rev. D 56, 5524 (1997).
- [6]M. Diehl, Physics Reports 388, 41–277 (2003).
- [7]A. Aktas et al. (H1), Eur. Phys. J. C 44, 1 (2005), arXiv:hep-ex/0505061.
- [8]F. D. Adloff et al. (H1), Phys. Lett. B 681, 391 (2009), arXiv:0907.5289 [hep-ex].
- [9]S. Chekanov et al. (ZEUS Collaboration), JHEP 05, 108 (2009).
- [10]A. Airapetian et al. (HERMES), Phys. Rev. Lett. 87, 182001 (2001), arXiv:hep-ex/0106068.DOI
- [11]A. Airapetian et al. (HERMES), Phys. Rev. D 75, 011103 (2007), arXiv:hep-ex/0605108.DOI
- [12]A. Airapetian et al. (HERMES), JHEP 06, 066, arXiv:0802.2499 [hep-ex].DOI
- [13]A. Airapetian et al. (HERMES), JHEP 11, 083, arXiv:0909.3587 [hep-ex].DOI
- [14]A. Airapetian et al. (HERMES), JHEP 06, 019, arXiv:1004.0177 [hep-ex].DOI
- [15]A. Airapetian et al. (HERMES), Phys. Lett. B 704, 15 (2011), arXiv:1106.2990 [hep-ex].DOI
- [16]A. Airapetian et al. (HERMES), JHEP 07, 032, arXiv:1203.6287 [hep-ex].DOI
- [17]S. Stepanyan et al. (CLAS Collaboration), Phys. Rev. Lett. 87, 18202 (2001).
- [18]S. Chen et al. (CLAS Collaboration), Phys. Rev. Lett. 97, 072002 (2006).
- [19]F. X. Girod et al. (CLAS Collaboration), Phys. Rev. Lett. 100, 162002 (2008).
- [20]G. Gavalian et al. (CLAS Collaboration), Phys. Rev. C 80, 035206 (2009).
- [21]S. Pisano et al. (CLAS Collaboration), Phys. Rev. D 91, 052014 (2015).
- [22]C. M. Camacho et al. (JLab Hall A Collaboration), Phys. Rev. Lett. 97, 262002 (2006).
- [23]H. S. Jo et al. (CLAS Collaboration), Phys. Rev. Lett. 115, 212003 (2015).
- [24]M. Defurne et al. (JLab Hall A Collaboration), Phys. Rev. C 92, 055202 (2015).
- [25]M. Defurne et al. (JLab Hall A Collaboration), Nat. Commun. 8, 1408 (2017).
- [26]F. Georges et al. (JLab Hall A Collaboration), Phys. Rev. Lett. 128, 252002 (2022).
- [27]M. Benali, C. Desnault, M. Mazouz, Z. Ahmed, H. Albataineh, et al., Nature Phys. 16, 191 (2020).
- [28]R. Akhunzhanov et al. (COMPASS), Phys. Lett. B 793, 188 (2019), [Erratum: Phys.Lett.B 800, 135129 (2020)], arXiv:1802.02739 [hep-ex].
- [29]F. Gauthieron et al. (COMPASS Collaboration), COMPASS-II Proposal, CERN/SPSC-2010-014, SPSCP340 (2010).
- [30]M. Hattawy et al., Physical Review Letters 119, 10.1103/physrevlett.119.202004 (2017).
- [31]A. Airapetian et al. (HERMES Collaboration), Phys. Rev. C 81, 035202 (2010).
- [32]J. Aubert et al., Phys. Lett. B 123, 275 (1983).
- [33]R. G. Arnold et al., Phys. Rev. Lett. 52, 727 (1984).
- [34]J. Ashman et al., Phys. Lett. B 202, 603 (1988).
- [35]M. Arneodo et al., Nucl. Phys. B 333, 1 (1990).
- [36]J. Gomez et al., Phys. Rev. D 49, 4348 (1994).
- [37]J. Seely et al., Phys. Rev. Lett. 103, 202301 (2009).
- [38]O. Hen et al., Rev. Mod. Phys. 89, 045002 (2017).
- [39]B. Schmookler et al. (CLAS Collaboration), Nature 566, 354 (2019).
- [40]R. Abdul Khalek et al., Nuclear Physics A 1026, 122447 (2022).
- [41]A. Bylinkin et al., Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 1052, 168238 (2023).
- [42]J. K. Adkins et al., Nucl. Instrum. Meth. A 1073, 170240 (2025), arXiv:2209.02580 [physics.ins-det].arxiv.org/abs/2209.02580
- [43]J. Adam et al. (ATHENA), JINST 17 (10), P10019, arXiv:2210.09048 [physics.ins-det].arxiv.org/abs/2210.09048
- [44]E. C. Aschenauer et al., Phys. Rev. D 112, 036010 (2025).
- [45]M. Boër et al., Three-dimensional imaging of hadrons with hard exclusive reactions: advances in experiment, theory, phenomenology, and lattice QCD (2025), arXiv:2512.15064 [hep-ph].
- [46]M. Vanderhaeghen, P. A. M. Guichon, and M. Guidal, Physical Review D 60, 10.1103/physrevd.60.094017 (1999).
- [47]A. Belitsky, D. Müller, and A. Kirchner, Nuclear Physics B 629, 323–392 (2002).
- [48]S. Boffi and B. Pasquini, La Rivista del Nuovo Cimento 30, 387–448 (2008).
- [49]A. Kirchner and D. Müller, The European Physical Journal C 32, 347–375 (2003).
- [50]A. V. Belitsky, D. Müller, A. Kirchner, and A. Schäfer, Physical Review D 64, 10.1103/physrevd.64.116002 (2001).
- [51]A. Bacchetta, U. D’Alesio, M. Diehl, and C. A. Miller, Phys. Rev. D 70, 117504 (2004).
- [52]A. V. Belitsky and D. Müller, Phys. Rev. C 82, 074010 (2010).
- [53]R. B. Wiringa, Phys. Rev. C 43, 1585 (1991).
- [54]S. K. Barcus, Extraction and Parametrization of Isobaric Trinucleon Elastic Cross Sections and Form Factors, Ph.D. thesis, William-Mary Coll. (2019).
- [55]S. Scopetta, Physical Review C 70, 10.1103/physrevc.70.015205 (2004).
- [56]S. Scopetta, Physical Review C 79, 10.1103/physrevc.79.025207 (2009).
- [57]F. Bissey, V. Guzey, M. Strikman, and A. Thomas, Physical Review C 65, 10.1103/physrevc.65.064317 (2002).
- [58]K. Kumerički and D. Müller, Nuclear Physics B 841, 1–58 (2010).
- [59]L. Calero Diaz and D. Keller, Phys. Rev. D 112, 096001 (2025).
- [60]R. B. Wiringa, R. Schiavilla, S. C. Pieper, and J. Carlson, Phys. Rev. C 89, 024305 (2014).
- [61]C. Ciofi degli Atti and L. P. Kaptari, Physical Review Letters 95, 10.1103/physrevlett.95.052502 (2005).
- [62]I. Vitev, T. Goldman, M. B. Johnson, and J. W. Qiu, Physical Review D 74, 10.1103/physrevd.74.054010 (2006).
- [63]G. Atoian et al. (EPIOS Scientific Consortium), Phys. Rev. C 113, 060501 (2026).
- [64]M. Lomnitz and S. Klein, Phys. Rev. C 99, 015203 (2019).
- [65]N. Devi, J. Seger, M. Kim, and S. R. Klein, Phys. Rev. C 113, 015207 (2026).
- [66]M. Kesler, A. I. Sheikh, R. Ma, Z. Tu, T. Ullrich, and Z. Xu, Physics Letters B 879, 140585 (2026).
- [67]W. Chang, E.-C. Aschenauer, M. D. Baker, A. Jentsch, J.-H. Lee, Z. Tu, Z. Yin, and L. Zheng, Phys. Rev. D 104, 114030 (2021), arXiv:2108.01694 [nucl-ex].DOI
- [68]O. Gryniuk, S. Joosten, Z.-E. Mezziani, and M. Vanderhaeghen, Phys. Rev. D 102, 014016 (2020), arXiv:2005.09293 [hep-ph].




