key: cord-0118629-zk6y0utn
authors: Abrams, MARATHON Collaboration D.; Albataineh, H.; Aljawrneh, B. S.; Alsalmi, S.; Aniol, K.; Armstrong, W.; Arrington, J.; Atac, H.; Averett, T.; Gayoso, C. Ayerbe; Bai, X.; Bane, J.; Barcus, S.; Beck, A.; Bellini, V.; Bhatt, H.; Bhetuwal, D.; Biswas, D.; Blyth, D.; Boeglin, W.; Bulumulla, D.; Butler, J.; Camsonne, A.; Carmignotto, M.; Castellanos, J.; Chen, J.-P.; Cohen, E. O.; Covrig, S.; Craycraft, K.; Cruz-Torres, R.; Dongwi, B.; Duran, B.; Dutta, D.; Fuchey, E.; Gal, C.; Gautam, T. N.; Gilad, S.; Gnanvo, K.; Gogami, T.; Gomez, J.; Gu, C.; Habarakada, A.; Hague, T.; Hansen, J.-O.; Hattawy, M.; Hauenstein, F.; Higinbotham, D. W.; Holt, R. J.; Hughes, E. W.; Hyde, C.; Ibrahim, H.; Jian, S.; Joosten, S.; Karki, A.; Karki, B.; Katramatou, A. T.; Keith, C.; Keppel, C.; Khachatryan, M.; Khachatryan, V.; Khanal, A.; Kievsky, A.; King, D.; King, P. M.; Korover, I.; Kulagin, S. A.; Kumar, K. S.; Kutz, T.; Lashley-Colthirst, N.; Li, S.; Li, W.; Liu, H.; Liuti, S.; Liyanage, N.; Markowitz, P.; McClellan, R. E.; Meekins, D.; Beck, S. Mey-Tal; Meziani, Z.-E.; Michaels, R.; Mihovilovic, M.; Nelyubin, V.; Nguyen, D.; Nuruzzaman,; Nycz, M.; Obrecht, R.; Olson, M.; Owen, V. F.; Pace, E.; Pandey, B.; Pandey, V.; Paolone, M.; Papadopoulou, A.; Park, S.; Paul, S.; Petratos, G. G.; Petti, R.; Piasetzky, E.; Pomatsalyuk, R.; Premathilake, S.; Puckett, A. J. R.; Punjabi, V.; Ransome, R. D.; Rashad, M. N. H.; Reimer, P. E.; Riordan, S.; Roche, J.; Salme, G.; Santiesteban, N.; Sawatzky, B.; Scopetta, S.; Schmidt, A.; Schmookler, B.; Segal, J.; Segarra, E. P.; Shahinyan, A.; Sirca, S.; Sparveris, N.; Su, T.; Suleiman, R.; Szumila-Vance, H.; Tadepalli, A. S.; Tang, L.; Tireman, W.; Tortorici, F.; Urciuoli, G. M.; Wojtsekhowski, B.; Wood, S.; Ye, Z. H.; Ye, Z. Y.; Virginia, J. Zhang University of; University, Texas AM; University, North Carolina AT State; University, Kent State; University, King Saud; Zagreb, University of; University, California State; Laboratory, Argonne National; Laboratory, Lawrence Berkeley National; University, Temple; William,; Mary,; Tennessee, University of; Technology, Massachusetts Institute of; Catania, INFN Sezione di; University, Mississippi State; University, Hampton; University, Florida International; University, Old Dominion; Lab, Jefferson; University, Tel Aviv; Connecticut, University of; University, Tohoku; University, Columbia; University, Cairo; University, Ohio; University, Stony Brook; Pisa, INFN Sezione di; University, Syracuse; Center-Negev, Nuclear Research; Sciences, Institute for Nuclear Research of the Russian Academy of; Hampshire, University of New; Regina, University of; Ljubljana, University of; Institute, Jozef Stefan; Gutenberg-Universitat, Johannes; College, Saint Norbert; Vergata, University of Rome Tor; Tech, Virginia; Carolina, University of South; Physics, Institute of; Technology,; University, Norfolk State; University, Rutgers; Roma, INFN Sezione di; Perugia, University of; Institute, Yerevan Physics; Technology, Shandong Institute of Advanced; University, Northern Michigan; Illinois-Chicago, University of; Vergata, INFN Sezione di Roma Tor; Perugia, INFN Sezione di
title: Measurement of the Nucleon $F^n_2/F^p_2$ Structure Function Ratio by the Jefferson Lab MARATHON Tritium/Helium-3 Deep Inelastic Scattering Experiment
date: 2021-04-12
journal: nan
DOI: nan
sha: 31ea6d0e4f6501d646e5816c5f8e4b09aa0c42d8
doc_id: 118629
cord_uid: zk6y0utn

The ratio of the nucleon $F_2$ structure functions, $F_2^n/F_2^p$, is determined by the MARATHON experiment from measurements of deep inelastic scattering of electrons from $^3$H and $^3$He nuclei. The experiment was performed in the Hall A Facility of Jefferson Lab and used two high resolution spectrometers for electron detection, and a cryogenic target system which included a low-activity tritium cell. The data analysis used a novel technique exploiting the mirror symmetry of the two nuclei, which essentially eliminates many theoretical uncertainties in the extraction of the ratio. The results, which cover the Bjorken scaling variable range $0.19<x<0.83$, represent a significant improvement compared to previous SLAC and Jefferson Lab measurements for the ratio. They are compared to recent theoretical calculations and empirical determinations of the $F_2^n/F_2^p$ ratio.

(Dated: June 10, 2021)

The ratio of the nucleon F2 structure functions, F n 2 /F p 2 , is determined by the MARATHON experiment from measurements of deep inelastic scattering of electrons from 3 H and 3 He nuclei. The experiment was performed in the Hall A Facility of Jefferson Lab and used two high resolution spectrometers for electron detection, and a cryogenic target system which included a low-activity tritium cell. The data analysis used a novel technique exploiting the mirror symmetry of the two nuclei, which essentially eliminates many theoretical uncertainties in the extraction of the ratio. The results, which cover the Bjorken scaling variable range 0.19 < x < 0.83, represent a significant improvement compared to previous SLAC and Jefferson Lab measurements for the ratio. They are compared to recent theoretical calculations and empirical determinations of the F n 2 /F p 2 ratio. The nucleon structure functions, found from deep inelastic scattering (DIS) of electrons by protons and deuterons, have been of fundamental importance in establishing the internal quark structure of the nucleon [1]. First measurements occurred in a series of DIS experiments at the Stanford Linear Accelerator Center (SLAC) circa 1970 [2] , which showed the existence of pointlike entities within the nucleons. Further studies of muon-nucleon and neutrino-nucleon DIS experiments at CERN [3] [4] [5] [6] and Fermilab [7, 8] established the quarkparton model (QPM) for the nucleon [9] , and provided supporting evidence for the emerging theory of quantum chromodynamics (QCD).

The cross section for deep inelastic electron-nucleon scattering, where the nucleon breaks up, is given, in the one-photon-exchange approximation, in terms of the structure functions F 1 (ν, Q 2 ) and F 2 (ν, Q 2 ) of the nucleon. In the lab frame and in natural units it reads [9] :

where σ M = 4α 2 (E ) 2 Q 4 cos 2 θ 2 is the Mott cross section, α is the fine-structure constant, E is the incident electron energy, E and θ are the scattered electron energy and angle, ν = E − E is the energy trans- * Present address: Canon Medical Research USA Inc., Vernon Hills, IL 60061 fer, Q 2 = 4EE sin 2 (θ/2) is the negative of the fourmomentum transfer squared, and M is the nucleon mass. The invariant mass of the final hadronic state is W =

The scattering process is mediated through the exchange of virtual photons. The cross section can also be written in terms of those for the absorption by the nucleon of longitudinally, σ L , or transversely, σ T , polarized photons. The functions F 1 and F 2 are related to the ra- [2] . All of the above formalism can also be applied to the case of DIS by a nucleus, with F 1 and F 2 becoming the structure functions of the nucleus in question. It should be noted that the ratio of DIS cross sections of different nuclear targets is equivalent to the ratio of their F 2 structure functions if R is the same for all nuclei. The latter has been confirmed experimentally within inherent experimental uncertainties [10] .

The basic idea of the QPM [11, 12] is to represent DIS as quasi-free scattering of electrons from the nucleon's partons/quarks, in a frame where the nucleon possesses infinite momentum. The fractional momentum of the nucleon carried by the struck quark is then given by the Bjorken "scaling" variable, x = Q 2 /(2M ν). In the limit where ν → ∞, Q 2 → ∞ with x finite between 0 and 1, the nucleon structure functions become:

, where e i is the fractional charge of quark type i, f i (x)dx is the probability that a quark of type i carries momentum in the range between x and x + dx, and the sum runs over all quark types. For the Gell-Mann/Zweig quarks, the F 2 (x) structure function for the proton (p) becomes F p 2 (x) = x[(4/9)U + (1/9)D + (1/9)S], and due to isospin symmetry, that of the neutron (n) F n 2 (x) = x [(1/9)U + (4/9)D + (1/9)S], where U = u +ū, D = d +d, and S = s +s, with bars denoting antiquarks [9] .

The positivity of the structure functions dictates that the ratio of the neutron to proton F 2 functions is bounded for all values of x: (1/4) ≤ F n 2 /F p 2 ≤ 4, a relationship known as the Nachtmann inequality [13] . This relationship was verified in the pioneering SLAC experiments E49a and E49b circa 1970 [14] , which found that the ratio approaches unity at x = 0 and approximately 1/4 at x = 1. The SLAC findings showed that at low x the three quark-antiquark distributions are equal, dominated by sea quarks, and that at large x the u (d) quark distribution dominates in the proton (neutron). These findings were surprising as the expectation, at the time, from SU(6) symmetry was that F n 2 /F p 2 should be equal to 2/3 for all x. The behavior of the ratio at x = 1 was justified by the diquark model of Close [15] , and Regge phenomenology, initiated by Feynman [16] . In Close's diquark model, the diquark configuration with spin 1 is suppressed relative to that with spin 0. The phenomenological suppression of the d quark distribution, which results from the F n 2 /F p 2 value of 1/4 at x = 1, can be understood in the quark model of Isgur [17] in terms of the color-magnetic hyperfine interaction between quarks, which is also responsible for the N -∆ mass splitting. It should be noted that perturbative QCD arguments [18] and a treatment based on quark-counting rules [19] suggest that the nucleon F 2 ratio should have the larger value of 3/7 at x = 1.

The original considerations of the magnitude of the nucleon F 2 ratio were called into question in the 1990s when a re-examination of the subject by Whitlow et al. [20] , who, using the original SLAC data [14] and a plausible model of the EMC effect in which the deuteron, medium and heavy nuclei scale with nuclear density [21] , found a strong sensitivity in the determination of the ratio at large x. The EMC effect, discovered at CERN [22] and quantified precisely at SLAC [23] , characterizes the modification of the nucleon structure functions in nuclear matter. The above strong sensitivity was subsequently confirmed in a relativistic re-analysis of the SLAC data, which assumes the presence of minimal binding effects in the deuteron [24] . In Ref. [20] , it also became evident that the nucleon F 2 ratio was very sensitive to the choice of the nucleon-nucleon (N-N) potential model governing the structure of deuterium, later confirmed in Refs. [25, 26] . The large uncertainty in the extraction of the F n 2 /F p 2 ratio at large x calls into question the presumption that F n 2 /F p 2 and D/U tend to 1/4 and zero, respectively, as x approaches 1.

These difficulties in the F n 2 /F p 2 determination can be remedied using a method proposed by Afnan et al. [27, 28] , which determines the F n 2 /F p 2 ratio from DIS measurements on 3 H (triton) and 3 He (helion), exploit-ing the isospin symmetry and similarities of the two A = 3 mirror nuclei. In the absence of Coulomb interactions and for an isospin symmetric world, the properties of a proton (neutron) bound in the 3 He nucleus should be identical to that of a neutron (proton) bound in the 3 H nucleus. Defining the EMC-type ratios for the F 2 structure functions of helion (h) and triton (t) by:

, one can write the ratio of these ratios as R ht = R h /R t , which directly yields the F n 2 /F p 2 ratio as:

The F n 2 /F p 2 ratio found from this Equation depends on the ratio of the EMC effects in 3 He and 3 H. Since the neutron and proton distributions in the A = 3 nuclei are similar, the ratio can be calculated reliably with the expectation that R ht 1 [28, 29] , once F h 2 /F t 2 is measured experimentally. The seeming dependence of the process on the F n 2 /F p 2 input is actually artificial. In practice, one can employ an iterative procedure to eliminate this dependence altogether. Namely, after extracting F n 2 /F p 2 from the data using some calculated R ht , the extracted F n 2 /F p 2 can then be used to compute a new R ht , which is then used to extract a new and better value of F n 2 /F p 2 . This procedure is iterated until convergence is achieved and a self-consistent solution for the extracted F n 2 /F p 2 is obtained. The convergence of the procedure was confirmed in Refs. [29, 30] . The above technique was used in the JLab E12-10-103 MARATHON experiment [30] (initiated in 1999 [31] ), which took data in the winter/spring of 2018 using the Electron Accelerator and Hall A Facilities of the Lab. Electrons scattered from light nuclei were detected in the Left and Right High Resolution Spectrometers (LHRS and RHRS) of the Hall [32] . The beam energy was fixed at 10.59 GeV, and its current ranged from 14.6 to 22.5 µA. The experiment detected DIS events from the proton, deuteron, helion, and triton particles using a cryogenic gaseous target system [33] . The LHRS was operated at a momentum of 3.1 GeV/c with angles between 16.81 • and 33.55 • . The RHRS was operated at a single setting of 2.9 GeV/c and 36.12 • .

The target system consisted of four high-pressure cells, of length 25.0 cm and diameter 1. 27 35 .02 atm, resulting in densities (determined from data-supported virial models [34] ) of 2.129 ± 0.021, 3.400 ± 0.010, 5.686 ± 0.022, and 2.832 ± 0.011 kg/m 3 , respectively. The target assembly also contained an empty cell and a "dummy target" consisting of two Al foils separated by 25.0 cm, which were used to measure the contribution to the scattered electron yields from the Al end-caps of the cells. The cells were cycled many times in the beam for each kinematic setting in order to minimize effects of possible drifts of the beam diagnostic or other instrumentation (e.g. the beam current monitors).

Scattered particles were detected in the HRSs using two planes of scintillators for event triggering, two drift chambers for particle tracking, and a gas thresholď Cerenkov counter and a lead-glass calorimeter for particle identification. Electrons were identified as electrons on the basis of i) a minimal pulse height in theČerenkov counter, and ii) the energy deposited in the calorimeter, consistent with the momentum as determined from the drift chamber track using the spectrometers' optical properties. The detector efficiencies for both spectrometers were found to be stable and independent of the gas target used. A small fraction of events with two or more drift chamber tracks (1-2% of the total) were not included in the data analysis, as they were dominated by electrons passing through the edges of the exit of the Al vacuum pipe of the spectrometers. Details on the Hall A Facility, beam line, and detector instrumentation as used in MARATHON, including calibrations, are given in Refs. [35] [36] [37] [38] [39] [40] .

Because of the low density of the gas targets, the electron counting rate was dominated, for all kinematics, by events originating from the target cell Al end-caps, as the total thickness of the two end-caps of the 3 He, 3 H, 2 H and 1 H cells was 0.55, 0.60, 0.51, and 0.64 mm, respectively. In order to reject electrons originating from the end-caps, a software, scattering-angle-dependent target position reconstruction "cut" was imposed, which resulted in an effective, usable target length of 21 cm, on the average. All events properly identified as electrons originating from the gas inside each target cell were binned by Bjorken x, resulting in the formation of an electron yield, Y (x), defined as:

where N e is the number of scattered electrons, N e is the number of incident beam electrons, ρ t is the density of the gas target, L t is the selected target length, and C cor = C det C cdt C den C tec C psp C rad C cde C bin C dth . Here, C det is the correction for trigger and detector inefficiency, C cdt is the computer dead-time correction (1.001 to 1.065), C den is a correction to the target density due to beam heating effects (1.066 to 1.125), C tec is a correction for falsely-reconstructed events originating from the end-caps (0.973 to 0.998), C psp is a correction for events originating from pair symmetric processes (0.986 to 0.999), C rad is a correction for radiative effects (0.826 to 1.173), C cde is a correction for Coulomb distortion effects (0.997 to 1.000), C bin is a bin-centering correction (0.995 to 1.001), and C dth is a correction for the beta decay of tritons to helions, applicable only to the tritium yield [0.997 (0.989) at the beginning (end) of the experiment]. A cross section model by Kulagin and Petti (K-P), based on the works of Refs. [41] [42] [43] , was adopted [44] for the bin-centering correction, and the Coulomb correction (which used the Q 2 -effective approximation as outlined in Ref. [45] ). When forming ratios of electron yields from different targets, which are equivalent to cross section ratios, the effective gas target length L t (18.0-22.5 cm) and the correction C det cancel out. In general, the corrections to the ratios from each effect become minimal, and in some cases, so do the associated systematic uncertainties. For example, the radiative effect correction, ranges from 0.997 (at the highest x) to 1.015 (at the lowest x) for the h/t cross section ratio. The dominant point-to-point systematic uncertainties for the yield ratios are those from the beam-heating gas target density changes [±(0.1%-0.5%)], the radiative correction [±(0.25%-0.45%)], and the choice of spectrometer acceptance limits (±0.2%). The total point-to-point uncertainty ranged from ±0.4% to ±1.0% for the d/p cross section ratio, and ±0.3% to ±0.5% for the h/t ratio. Details on the determination of the yields, and all associated corrections and uncertainties, can be found in Refs. [36] [37] [38] [39] [40] .

The experiment also collected DIS data for the proton and deuteron (d) over the x range from 0.19 to 0.37 for the purpose of finding the F n 2 /F p 2 ratio in the vicinity of x = 0.3, where it is known that nuclear corrections are minimal [41, 43] , and comparing it with the F n 2 /F p 2 ratio found using DIS by the triton and helion. The measured values of the σ d /σ p ratio are given, together with associated uncertainties, in Table 1 of the Appendix. The σ d /σ p = F d 2 /F p 2 values, plotted in Figure 1 , are compared to reference measurements from the seminal SLAC E49b and E87 experiments [46] taken with similar beam ener- [56] , and a band based on the fit of the SLAC data as provided in Ref. [46] , for the MARATHON kinematics [Q 2 = 14 · x (GeV/c) 2 ] (see text). All three experimental data sets include statistical, point to point systematic, and normalization uncertainties.

gies. It is evident from Fig. 1 that the JLab and SLAC data are in excellent agreement, at the 10 −3 level. Given [29, 42] . The R d ratio used in the MARATHON F d 2 /F p 2 data analysis is from the model by Kulagin and Petti based on Refs. [41, 42] . The results of this model are, in the vicinity of x = 0.3, in excellent agreement with determinations using data from the JLab BoNuS [47] and SLAC E139 [23] experiments, and two distinct calculations based on studies of data from DIS off nuclei, described in Refs. [43] (using nuclei with A ≥ 4) and [48] (using nuclei with A ≥ 3).

The focus of MARATHON was to study DIS from helion and triton in order to extract the F n 2 /F p 2 ratio in the range 0.19 < x < 0.83 using the measured σ h /σ t = F h 2 /F t 2 ratio and model-calculated values of the super-ratio R ht . The values used for R ht come from the theoretical model by Kulagin and Petti [42, 43] , which provides a global description of the EMC effect for all known targets (for a review see Ref. [49] ). The K-P model includes a number of nuclear effects out of which the major correction for the relevant kinematics comes from the smearing effect with the nuclear energy-momentum distribution, described in terms of the nuclear spectral function, together with an off-shell correction to the bound nucleon F 2 [43] . The underlying nucleon structure functions come from the global QCD analysis of Ref. [50] , which was performed up to NNLO approximation in the strong coupling constant including target mass corrections [51] as well as those due to higher-twist effects. For the spectral functions of the 3 H and 3 He nuclei, the results of Ref. [29] have been used. In order to evaluate theoretical uncertainties, the 3 He spectral function of Ref. [52] was used. Reasonable variations of the high-momentum part of the nucleon momentum distribution in 3 H and 3 He were considered, and uncertainties in the off-shell correction of Ref. [43] , as well as in the nucleon structure functions, were accounted for. The maximum resulting uncertainty in R ht is estimated to be up to ±0.4% (at x = 0.8), contributing minimally to the total uncertainty in the final F n 2 /F p 2 values. The K-P calculations were performed prior to the analysis of the MARATHON data.

The comparison of F n 2 /F p 2 as extracted from σ h /σ t and σ d /σ p was done at x = 0.31, where nuclear corrections contribute negligibly to EMC-type ratios like R d and R ht , as σ A /A = σ d /2 [53] (determined by the A ≥ 3 data of Refs. [23, 54, 55] and taking into account the quoted normalization uncertainties therein). The K-P models used, predicted a value of 1.000 at x = 0.31 for both R ht and R d with uncertainties of ±0.1% and ±0.2%, respectively. The recent work of Ref. [48] , based on a global analysis of nuclear DIS data where the EMC effect is accounted for through nucleon short-range correlations, found R ht (x = 0.31) = 1.001, with a similar uncertainty. The values of σ d /σ p and σ h /σ t at x = 0.31 were determined by weighted fits to the MARATHON data, which included statistical and point-to-point uncertainties added in quadrature. In order to match the F n 2 /F p 2 values found using the two different sets of nuclei, the σ h /σ t ratio at x = 0.31 had to be normalized by a multiplicative factor of 1.025±0.007. Consequently, all values of σ h /σ t reported in this work have been normalized upwards by 2.5%.

The normalized σ h /σ t values are given in Table 2 of the Appendix, together with associated uncertainties. The F n 2 /F p 2 values are given in Table 3 of the Appendix, together with associated uncertainties. Shown also in Table 3 are the R ht super-ratio values used to find F n 2 /F p 2 . The R ht uncertainty was incorporated in quadrature with the point-to-point F n 2 /F p 2 uncertainty. Figure 2 shows the MARATHON results for the F n 2 /F p 2 ratio, along with data from the JLab Hall B BoNuS experiment [56] for W ≥ 1.84 GeV/c 2 , evolved to the Q 2 of MARATHON [20] , and results from early SLAC measurements with W ≥ 1.84 GeV/c 2 [14, 46] . The SLAC results are presented as a band, the width of which at high x is dominated primarily by uncertainties due the choice of the N-N potential used for the evaluation of the deuteron wave function [20, 25, 26] . The MARATHON data are in good agreement with the BoNuS data, and fall well within the SLAC results band. The highest-x points are consistent with the F n 2 /F p 2 ratio tending to a value between 0.4 and 0.5 at x = 1. This is consistent with the predictions of perturbative QCD and quark counting rules (for which this ratio is 3/7 at x = 1), and with a recent prediction [57] that treats strong interactions using the Dyson-Schwinger equations, where diquark correlations in the nucleons are consequences of dynamical chiral symmetry breaking (for which the nucleon F 2 ratio lies, at x = 1, between 0.4 and 0.5). It is also consistent with a covariant quark-diquark model which also predicts that this ratio should be 3/7 at x = 1 [58] .

The MARATHON F n 2 /F p 2 ratio values are in excellent agreement, as quantified by a χ 2 per degree of freedom (df) of 0.8, with those predicted by Kulagin and Petti, which were used in the determination of R ht . For this reason, an iterative procedure, as described earlier, was not necessary. A comparison between the MARATHON F n 2 /F p 2 results and the K-P prediction is shown in Figure  3 . Shown also in the Figure are the σ t /σ h MARATHON values compared with the K-P prediction. The predicted σ t /σ h values by K-P, which were also used in the determination of R ht , are in excellent agreement with the MARATHON data, as quantified by a χ 2 /df of 0.8. Also shown in Figure 3 is the nuclear DIS determination of F n 2 /F p 2 by Segarra et al. [48] , the latest calculation for F n 2 /F p 2 by the CTEQ-JLab (CJ) Collaboration [59] , and a recent prediction for σ t /σ h by Tropiano et al. [60] , which includes isovector components in the off-shell effects for the bound nucleons in the two A = 3 nuclei, resulting in different corrections for the proton and neutron.

In summary, the JLab MARATHON experiment has provided a precise determination of the nucleon F n 2 /F p 2 ratio, which is expected to constrain theoretical models of the few body nuclear structure functions, and to be used in algorithms which fit [41, 59, 61] hadronic data to determine the essentially unknown (u +ū)/(d +d) ratio at large Bjorken x. These new data will also improve our knowledge of the nucleon parton distributions, which is needed for the interpretation of high-energy collider data. We acknowledge the outstanding support of the staff of the Accelerator Division and Hall A Facility of JLab, and work of the staff of the Savannah River Tritium Enterprises and the JLab Target Group. We thank Dr. M. E. Christy for useful discussions on the optical properties of the HRS systems. We are grateful to Dr. W. Melnitchouk The helion to triton DIS cross section ratio (after normalization, see text) at the x, Q 2 , and W kinematics of MARATHON. Listed also are the statistical (stat), point-to-point (ptp) and overall/scale systematic (syst) components of the total (tot) error. The latter is the quadrature of the three components.

x R ht ∆R ht F n 2 /F p Listed also are the ratio's statistical (stat), point-to-point (ptp), and overall/scale (syst) components of the total (tot) error. The latter is the quadrature of the three components. Also listed are the values for the R ht super-ratio and their uncertainties used in the F 2 2 /F p 2 ratio extraction (see text).

An Introduction to Quarks and Partons

SLAC-E49a)

Proceedings of 3rd International Conference of High Energy Collisions

JLab PR12-10-103 MARATHON Proposal: MeAsurement of the F n 2 /F p 2 , d/u RAtios and A=3 EMC Effect in Deep Inelastic Electron Scattering Off the Tritium and Helium MirrOr Nuclei

Proceedings of Workshop on Experiments with Tritium at JLab

Conceptual Design of a Tritium Gas Target for JLab (JLab report)

Electron Scattering from Complex Nuclei