First Measurement of the Σ Beam Asymmetry in η' Photoproduction off the Proton near Threshold - July 2014 di P. Levi Sandri, G. Mandaglio, O. Bartalini, V. Bellini, J. P. Bocquet, M. Capogni, F. Curciarello, A. D’Angelo, V. De Leo, J. P. Didelez, R. Di Salvo, A. Fantini, D. Franco, C. Gaulard, G. Gervino, F. Ghio, G. Giardina, B. Girolami, A. Giusa, A. Lapik, A. Lleres, F. Mammoliti, M. Manganaro, D. Moricciani, A. Mushkarenkov, V. Nedorezov, C. Randieri, D. Rebreyend, N. Rudnev, G. Russo, C. Schaerf, M. L. Sperduto, M. C. Sutera, A. Turinge, V. Vegna and I. Zonta (2014) Abstract The Σ beam asymmetry in η' photoproduction off the proton was measured at the GRAAL polarized photon beam with incoming photon energies of 1.461 and 1.480 GeV. For both energies the asymmetry as a function of the meson emission angle shows a clear structure, more pronounced at the lowest one, with a change of sign around 90°. The results are compared to the existing theories that fail to account for the data.

1. First Measurement of the Σ Beam Asymmetry in η Photoproduction off the Proton
near Threshold
P. Levi Sandri,1, ∗
G. Mandaglio,2, 3, 4, †
O. Bartalini,5, 6
V. Bellini,2, 7
J.-P. Bocquet,8
M. Capogni,5, 6, ‡
F. Curciarello,2, 3
A. D’Angelo,5, 6
V. De Leo,2, 3
J.-P. Didelez,9
R. Di Salvo,5
A. Fantini,5, 6
D.
Franco,5, 6, §
C. Gaulard,1, ¶
G. Gervino,10, 11
F. Ghio,12, 13
G. Giardina,2, 3
B. Girolami,12, 13
A. Giusa,2, 7
A. Lapik,14
A. Lleres,8
F. Mammoliti,2, 7
M. Manganaro,2, 3, ∗∗
D. Moricciani,5
A.
Mushkarenkov,14
V. Nedorezov,14
C. Randieri,2, 7
D. Rebreyend,8
N. Rudnev,14
G. Russo,2, 7
C.
Schaerf,5, 6
M.-L. Sperduto,2, 7
M.-C. Sutera,2, 7
A. Turinge,14
V. Vegna,5, 6, ††
and I. Zonta5, 6
(GrAAL Collaboration)
1
INFN - Laboratori Nazionali di Frascati, via E. Fermi 40, 00044 Frascati, Italy
2
INFN - Sezione di Catania, via S. Sofia 64, 95123 Catania, Italy
3
Dipartimento di Fisica e di Scienze della Terra - Universit`a di Messina, salita Sperone 31, 98166 Messina, Italy
4
Centro Siciliano di Fisica Nucleare e Struttura della Materia, Viale A. Doria 6, 95125 Catania, Italy
5
INFN - Sezione di Roma Tor Vergata, via della Ricerca Scientifica 1, 00133 Roma, Italy
6
Dipartimento di Fisica - Universit`a degli Studi di Roma Tor Vergata,
via della Ricerca Scientifica 1, 00133 Roma, Italy
7
Dipartimento di Fisica e Astronomia - Universit`a degli Studi di Catania, via S. Sofia 64, 95123 Catania, Italy
8
LPSC, Universit´e Grenoble-Alpes, CNRS/IN2P3, F-38026 Grenoble, France
9
IN2P3, Institut de Physique Nucl´eaire, Rue Georges Clemenceau, 91406 Orsay, France
10
INFN- Sezione di Torino via Pietro Giuria 1, 10125 Torino, Italy
11
Dipartimento di Fisica Sperimentale - Universit`a degli Studi di Torino, via Pietro Giuria 1, I-10125 Torino, Italy
12
INFN - Sezione di Roma, piazzale Aldo Moro 2, 00185 Roma, Italy
13
Istituto Superiore di Sanit`a, viale Regina Elena 299, I-00161 Roma, Italy
14
Institute for Nuclear Research, 60-letiya Oktyabrya prospekt 7a, 117312 Moscow, Russia
(Dated: July 21, 2014)
The Σ beam asymmetry in η photoproduction off the proton was measured at the GrAAL po-
larized photon beam with incoming photon energies of 1.461 and 1.480 GeV. For both energies the
asymmetry as a function of the meson emission angle shows a clear structure, more pronounced at
the lowest one, with a change of sign around 90o
. The results are compared to the existing theories
that fail to account for the data.
PACS numbers: 13.60.Le, 13.88.+e, 14.40.Be
Keywords:
I INTRODUCTION
The experimental study of nucleon excited states is
fundamental for the understanding of its internal struc-
ture. Important differences are still observed today be-
tween the experimental nucleon spectrum and the results
of the constituent quark models used to predict it[1–4].
Several predicted states have not been observed (missing
resonances). The nucleon excited states decay strongly
with meson emission; therefore meson photoproduction
experiments off the nucleon are an ideal way of searching
for missing resonances and complement the information
obtained with pion scattering experiments.
In pseudo-scalar meson photoproduction off the pro-
ton (γ + p → meson + p) we have eight possible combi-
nations of spin states. The scattering amplitude is thus
described by eight matrix elements, only four of which
are independent due to rotational invariance and parity
transformations. With these four complex amplitudes,
16 bilinear products can be constructed, corresponding
to 16 observables: the differential cross section, three sin-
gle polarization observables and twelve double polariza-
tion observables. To determine the scattering amplitude
thoroughly, the cross section, the three single polariza-
tion and four appropriately chosen double polarization
observables must be measured [5]. These observables can
be expressed in terms of helicity amplitudes and the fol-
lowing relations hold[6–9]:
dσ/dΩ ∼ |H1|2
+ |H2|2
+ |H3|2
+ |H4|2
Σ ∼ Re(H1H∗
4 − H2H∗
3 )
T ∼ Im(H1H∗
2 − H3H∗
4 )
P ∼ Re(H1H∗
3 − H2H∗
4 )
where dσ/dΩ is the differential cross section and Σ, T
and P are the beam, target and recoil asymmetries re-
spectively. From the above relations one can see that
the general structure of the scattering amplitude is al-
ready contained within the differential cross section, but
the details of the amplitude can be better highlighted by
studying the single polarization observables, where the
interference among the helicity amplitudes can play a
crucial role in revealing subtle effects[10].
The first data on η photoproduction cross section were
produced in 1968 in a bubble chamber experiment[11]
using an untagged photon beam, and confirmed in

2. 2
1976 with a streamer chamber setup and tagged pho-
tons at DESY[12]. Over 20 years later, the SAPHIR
collaboration[13] reported a more extended measure-
ment, based on 250 events, from which the masses and
widths of the dominating S11 and P11 resonances were
extracted. In more recent years, the CLAS experiment
at Jlab and the CB-ELSA-TAPS in Bonn have pro-
duced a rich amount of precise cross section data on the
proton[14–16]. The energy region from threshold (1.447
GeV) up to 2.84 GeV was measured, and total and dif-
ferential cross section data were produced.
As a consequence of this huge experimental effort, the
following facts were established:
i) the η N channel couples mainly to S11(1535) and
P11(1710). A marginal role is played by J=3/2 reso-
nances, namely P13(1720) and D13(1520)[14].
ii) gη NN = 1.3 − 1.5, a value consistent with existing
theoretical estimates[14].
iii) above 2 GeV, where the process is dominated by the
ρ and ω exchange, the dynamics of η photoproduction
are similar to those of η photoproduction[16].
From the theoretical point of view, four approaches are
available in the literature:
i) A relativistic meson-exchange model of hadronic
interactions[17], with t-channel mesonic currents (ρ and
ω) and s- and u-channel resonances contributions. The
resonances considered were S11, P11, D13 and P13, the
two latter being required to reproduce some of the de-
tails of the angular distribution. This approach was later
revisited[18] performing a combined analysis of η pro-
duction reactions and no D13 resonance was found nec-
essary in describing the data.
ii) In a reggeized model for η and η photoproduction[19],
the authors use the same ingredients, and the vector me-
son exchanges are treated in terms of Regge trajectories
to comply with the correct high-energy behavior.
iii) In a chiral quark-model[20] the process is governed
by S11(1535) and u-channel background. The inclusion of
S11(1920) improves the calculation close to threshold and
there is no evidence of contributions from P−, D13, F−
and G− resonances with masses around 2 GeV.
iv) In an isobar model[21] a good description of the exist-
ing cross section data is obtained by taking into account
the contributions of 6 high-angular-momenta heavy res-
onances alone.
All the abovementioned theoretical calculations give
a reasonable description of the data. In all cases the
authors stress that the cross section data alone are unable
to pin down the resonance parameters, while polarization
observables could be very helpful to better determine the
partial wave contributions in this reaction and impose
more stringent constraints on the parameter values of
the different models.
In this letter, we present the first measurement of the
single polarization observable Σ for η photoproduction
off the proton, at the incoming photon energies of 1.461
and 1.480 GeV, obtained with the Compton backscat-
tered photon beam of the GrAAL experiment. In section
II the apparatus is briefly described; the data analysis is
presented in section III; in section IV the results obtained
are discussed and compared with the available theoretical
approaches. Section V contains our conclusions.
II EXPERIMENTAL APPARATUS
The GrAAL experiment was located at the Euro-
pean Synchrotron Radiation Facility (ESRF) in Grenoble
(France), where it took data from 1995 to 2008. A lin-
early polarized photon beam impinged on a liquid H2
or D2 target, and the final products were detected by
the large solid angle detector LAGRANγE (Large Ac-
ceptance GRaal-beam Apparatus for Nuclear γ Experi-
ments).
The photon beam was produced by the Compton
backscattering of low-energy polarized photons from an
Argon laser, against the 6.03 GeV electrons circulating
inside the ESRF storage ring[22]. The UV laser line (3.53
eV) was used to produce a backscattered photon beam,
covering the energy range up to 1.5 GeV. By using of the
far-UV laser line, the investigated energy range was ex-
tended up to 1.55 GeV. A tagging system, located inside
the electron ring, provided an event-by-event measure-
ment of the photon beam energy, with a resolution of 16
MeV (FWHM). Since the electron involved in the Comp-
ton scattering is ultra-relativistic, its helicity is conserved
in the process at backward angles, and the outgoing pho-
ton retains the polarization of the incoming laser beam
(up to 96% for the UV laser line). The correlation be-
tween photon energy and polarization is calculated with
QED [23]. During data taking, the laser beam polariza-
tion was rotated by 90o
every 20 minutes approximately,
and unpolarized data from the Bremsstrahlung of the
electrons off the ESRF residual vacuum were collected as
well.
The LAGRANγE detector can be divided into two an-
gular regions;
1. the central region, for polar angles between 25o
and
155o
consists of: two cylindrical multi-wire proportional
chambers, for charged particle tracking, having an an-
gular resolution of 3.5o
and 4o
for polar and azimuthal
angles, respectively [24]; a thin (5mm) plastic scintillator
barrel divided into 32 modules, used for particle identifi-
cation; a BGO electromagnetic calorimeter (Rugby Ball),
with excellent energy resolution [25–27] consisting of 480
crystals arranged in a 15 x 32 matrix.
2. a forward region, for polar angles smaller than 25o
,
consisting of: two planar multi-wire proportional cham-
bers, for charged particle tracking with a resolution of
1.5o
and 2o
(FWHM) for polar and azimuthal angles, re-
spectively; a double wall of plastic scintillator bars, with
a time resolution of 300 ps, for time of flight (TOF)

3. 3
and impact coordinates measurement of charged parti-
cles, which may be used for proton/pion discrimination
and for the precise proton energy calculation from TOF
measurement; a shower wall, with a time resolution of
600 ps [28], for TOF and impact coordinates measure-
ment for both charged and neutral particles.
At the end of the beamline, two flux detectors were
used for beam monitoring. The first one, consisting of
two plastic scintillators preceded by an aluminum foil
to convert photons into electron-positron pairs, while a
third plastic scintillator before the aluminum foil was
used as a veto for the upstream background. Its detection
efficiency was low ( 3%) to avoid pile-up effects during
data taking. This monitor was used for relative intensity
measurements. The second flux monitor consisted of a
uniform array of plastic scintillating fibers and lead [29].
Its photon detection efficiency is close to 1, and it was
used to calibrate the efficiency of the former monitor,
with the low intensity Bremsstrahlung beam produced
in the residual vacuum of the synchrotron. A detailed
description of the LAGRANγE apparatus can be found
in [30].
III DATA ANALYSIS
Data were collected during many different stretches of
the GrAAL experiment. As the threshold for η pho-
toproduction off the proton is Eth = 1.447 GeV, only
the periods of measurement performed by using the UV
laser line (351 nm wavelength) and with the multi-lines
UV (364, 351 and 333 nm wavelengths) allow to reach
Eth and to explore the behavior of the asymmetry as
a function of the photon energy up to 1.5 GeV. The η
mesons were identified via γγ, π0
π0
η and π+
π−
η decay
modes and by requiring the fulfilment of the two-body
kinematics for the recoil proton.
The preliminary event selection required:
i) at least two photons measured in the Rugby Ball for
the invariant mass reconstruction;
ii) a tagging energy above Eth;
iii) a proton detected in the forward TOF wall with polar
angle θp lying in the acceptance region showed in Fig. 1
left.
The distribution of Fig. 1 (left) was produced with
an upgraded version of the event generator described in
[31]. As we can see, for the photon energies available
at GrAAL, the recoil proton is always detected in the
forward direction (θp ≤ 16o
). Moreover, the momen-
tum/energy ratio determined by the two-body kinemat-
ics is always below 0.4. We therefore detected non rel-
ativistic protons in the forward direction. In these con-
ditions, the resolution on the proton momentum for the
η photoproduction was estimated with a GEANT3[32]
simulation to be about 2.5%.
The η missing mass calculated from the recoil proton
0
100
200
300
400
500
(degree)pθ
0 5 10 15 20 25
(GeV)γE
1.42
1.44
1.46
1.48
1.5
1.52
1.54
1.56
1.58
1.6
)2
Missing mass (GeV/c
0.7 0.8 0.9 1 1.1
Counts
0
1000
2000
3000
4000
5000
6000 cuts i) and ii)
cuts i)-iii)
FIG. 1: (Color online) Left: energy of photon beam vs. the
proton polar angle θp for a simulated γp → η p. Right: miss-
ing mass spectrum from the recoil proton detection. The
black dashed curve shows the effects of selection cuts i) and
ii); while the solid blue curve is the result of all preliminary
selection cuts i), ii) and iii).
is shown in Fig. 1 right. The effects of the cuts i) and
ii) are shown as a black dashed line. The inclusion of
cut iii) gave as a result the blue solid line. The η peak
is clearly visible over a smooth background. This resid-
ual background was eventually suppressed by additional
constraints on the decay products of the η meson.
20
40
60
80
100
)2
Missing mass (GeV/c
0.92 0.94 0.96 0.98
)2
Invariantmass(GeV/c
0.8
0.9
1
1.1
γ2→’η
20
40
60
80
100
)2
Missing mass (GeV/c
0.92 0.94 0.96 0.98
)2
Invariantmass(GeV/c
0.8
0.9
1
1.1
γ6→0π0
πη→’η
(a) (b)
10
20
30
40
50
60
)2
Missing mass (GeV/c
0.94 0.96 0.98
)2
Invariantmass(GeV/c
0.45
0.5
0.55
0.6
0.65 -
π+
πη→’η
Missing mass (GeV)
0.94 0.96 0.98
Counts
0
100
200
300
400
500
600
700
η-
π+
π→’η
η0
π0
π→’η
γγ→’η
(c) (d)
FIG. 2: (Color online) Panels a), b) and c): Invariant mass
spectrum from photons (two photons in panels a and c, six
photons in panel b) in the BGO calorimeter vs. the missing
mass spectrum obtained from the measurement of the recoil
proton. There are no events in the white area. Panel (d):
Missing mass spectra from the recoil proton measurement af-
ter the selection of the events in panels a), b) and c).
The cleanest decay channel for LAGRANγE is the de-
cay η → γγ. The two final-state photons were detected
in the Rugby Ball and give rise, together with the recoil
proton, to the missing mass vs. invariant mass distribu-
tion of Fig. 2 (a). This decay mode has a rather small
branching ratio ( 2.18%) and the number of events col-
lected (3400) did not allow for the extraction of the beam
asymmetry with sufficiently good statistics. For this rea-
son, the decay channels involving two pions and one η me-
son were also included in the analysis. The η → π0
π0
η

4. 4
decay channel was included by requiring the detection of
six photons in the Rugby Ball reconstructing the η me-
son invariant mass (Fig. 2 (b)). For the inclusion of the
charged decay channel (η → π+
π−
η) we required the
invariant mass reconstruction from η meson decay into
two photons (Fig. 2 (c)) and two charged tracks in the
whole detector, identified as charged pions. All events
with extra spurious signals in the detector, charged or
neutral, were rejected.
The influence on the missing mass calculated from the
recoil proton of the selection on the decay products of the
η is shown in Fig. 2 (d). In all cases the missing mass
distribution has a Breit-Wigner behavior and the value
of the resulting η mass is in excellent agreement with the
literature [33]. At the end of the data reduction, 12121 η
events are available for asymmetry determination with a
residual background, estimated through simulation and
mainly due to non-resonant multi-meson photoproduc-
tion, of less than 4%. As the recoil proton angles are the
best measured ones, the emission angle of the meson in
the center-of-mass frame θη
c.m. was calculated from the
relevant proton angle θp
c.m..
IV RESULTS AND DISCUSSION
The selected η events were grouped into two energy
bins (the first bin is [1.447, 1.475] GeV with centroid
1.461 GeV; the second, with centroid 1.480 GeV, is
[1.475, 1.490] GeV), seven angular bins for θη
c.m., and
eight for the azimut angle φ. The beam asymmetry
Σ(Eγ, θη
c.m.) can be calculated by fitting the distribution
defined by the following ratio:
NV /FV
NV /FV + NH/FH
=
1
2
[1 + P(Eγ) · Σ · cos(2φ)]
where NV (NH) and FV (FH) are the number of events
and the total γ flux for vertical (horizontal) polarization
states and P(Eγ) is the calculated degree of polarization.
This procedure significantly decreases the systematic er-
rors of the extracted asymmetries by removing all the
detection and reconstruction efficiencies. In Fig. 3 we
give an example of this azimuthal distribution with the
performed fit.
The remaining systematic error, estimated as not more
than 0.03, is due to the uncertainties originating from
the possible deterioration of the laser light polarization
on the laser focusing system, from slightly different beam
profiles on the target for each polarization state, and from
the residual background. The stability of the results was
verified by changing the angular binning and by analysing
separately the subsets of events resulting from neutral or
charged decay modes. In all cases, the results were found
to be pleasantly stable[34].
The results are summarized in Fig. 4 together with the
calculations of 18–21. As one can see, the asymmetry is
(degree)c.m.
’η
φ
0 50 100 150 200 250 300 350
H
/FH+NV
/FVN
V
/FVN
0.4
0.45
0.5
0.55
0.6
0.65
>=1.461 GeVγ
, <E
°
>=65.79c.m.
’η
θ<
/ ndf = 1.51 / 72
χ
FIG. 3: (Color online) Azimuthal distribution at Eγ = 1.461
GeV and θη
c.m. = 65.79 o
-0.2
-0.1
0
0.1
0.2
Σ =1.461 GeVγE
W=1.903 GeV
(degree)
’η
c.m.θ
0 50 100 150
-0.2
-0.1
0
0.1
0.2
=1.480 GeVγE
W=1.912 GeV
FIG. 4: (Color online) Σ beam asymmetry at the incoming
photon energies of 1.461 and 1.480 GeV (corresponding to
a total center-of-mass energy W of 1.903 and 1.912 GeV re-
spectively) as a function of the meson emission angle in the
center-of-mass system compared to theoretical calculations:
red dotted line[19], blue dashed line[18] green dot-dashed[21],
orange long-dashed[20]. The solid black line is the result of a
fit performed with a function f(θ) = a·sin2
(θ)cos(θ). The fit
results for the free parameter are: a = 0.321 ± 0.063 at 1.461
GeV and a = 0.096 ± 0.051 at 1.480 GeV.
positive at forward angles and negative at backward an-
gles. Moreover, the data indicate a quite strong energy
dependence, the effect being more evident at 1.461 GeV,
closer to threshold. This behavior is compatible with a
∼ sin2
(θη
c.m.)cos(θη
c.m.) function, typical of a P-wave D-
wave (S-wave F-wave) interference. The existing calcu-
lations, whilst providing a reasonable description of the
measured cross section, cannot however reproduce these
data, especially in the first energy bin (Eγ = 1.461 GeV
corresponding to a total center-of-mass energy of 1.903

5. 5
GeV) where a change of sign in the asymmetry values
around 90o
for the meson center-of-mass emission angle
is clearly visible. A slightly better, but still not satisfac-
tory, agreement between data and calculation is obtained
at forward angles and at the highest energy bin (Eγ =
1.480 GeV corresponding to a total center-of-mass en-
ergy of 1.912 GeV) in [18, 20]. We must notice that the
theoretical curves presented here are the result of inter-
polations of the existing models at low energies, and that
none of these models contains D-wave or F-wave contri-
butions. It is also important to underline that, in con-
trast with the conclusions of [16] for higher energies, at
threshold the dynamics of η[35] and of η photoproduc-
tion processes are clearly different.
These results prove once again that the polarization
degrees of freedom play an essential role in accessing the
details of the interaction, and can lead to a better de-
termination of the partial wave contributions and to a
better comprehension of the reaction mechanism.
V CONCLUSION
The Σ beam asymmetry in the η photoproduction was
measured at the incoming photon energies of 1.461 and
1.480 GeV by using the highly linearly polarized GrAAL
photon beam and the large solid angle LAGRANγE de-
tector. This is the first measurement of this observ-
able for this reaction. The values obtained indicate a
P-wave D-wave (S-wave F-wave) interference, the closer
to threshold the stronger. Available calculations fail to
reproduce the observed behavior, regardless of the inter-
mediate resonance states involved in the models.
From the experimental point of view, new measure-
ments with a finer energy binning as well as an extended
energy range, would be highly desirable.
The authors would like to thank L. Tiator, F. Huang,
H. Haberzettl, K. Nakayama, X-H Zhong, Q. Zhao and V.
Tryasuchev for kindly providing the results of their mod-
els at the energies of this paper, and for helpful discus-
sions. It is a pleasure to thank the ESRF for the reliable
and stable operation of the storage ring and the technical
staff of the contributing institutions for essential help in
the realization and maintenance of the apparatus.
∗
Electronic address: paolo.levisandri@lnf.infn.it
†
Electronic address: gmandaglio@unime.it
‡
Present address: ENEA - C.R. Casaccia, Istituto
Nazionale di Metrologia delle Radiazioni Ionizzanti, Via
Anguillarese, 301 I-00123 Roma, Italy
§
Present address: IPNL - 43, Bd du 11 Novembre 1918,
Fr69622 Villeurbanne Cedex, France
¶
Present address: CSNSM, Universit´e Paris-Sud 11,
CNRS/IN2P3, 91405 Orsay, France
∗∗
Present address: Universidad de La Laguna, Instituto de
Astrof´ısica de Canarias, E-38205 La Laguna, Tenerife,
Spain
††
Present address: Universit¨at Bonn, Physikalisches Insti-
tut - Nußallee 12, Bonn, D-53115, Germany
[1] S. Capstick and N. Isgur, Phys.Rev. D34, 2809 (1986).
[2] S. Capstick, Phys.Rev. D46, 2864 (1992).
[3] S. Capstick and W. Roberts, Phys.Rev. D49, 4570
(1994).
[4] D. Riska and G. Brown, Nucl.Phys. A679, 577 (2001).
[5] W.-T. Chiang and F. Tabakin, Phys.Rev. C55, 2054
(1997).
[6] D. Drechsel, O. Hanstein, S. Kamalov, and L. Tiator,
Nucl.Phys. A645, 145 (1999).
[7] T. Feuster and U. Mosel, Phys.Rev. C59, 460 (1999).
[8] B. Saghai and F. Tabakin, Phys.Rev. C55, 917 (1997).
[9] C. Fasano, F. Tabakin, and B. Saghai, Phys.Rev. C46,
2430 (1992).
[10] R. A. Arndt, R. L. Workman, Z. Li, and L. D. Roper,
Phys.Rev. C42, 1853 (1990).
[11] R. Erbe et al. (ABBHHM Collaboration), Phys.Rev. 175,
1669 (1968).
[12] W. Struczinski et al. (AHHM Collaboration), Nucl.Phys.
B108, 45 (1976).
[13] R. Pl¨otzke et al. (SAPHIR Collaboration), Phys.Lett.
B444, 555 (1998).
[14] M. Dugger et al., Phys.Rev.Lett. 96, 062001 (2006).
[15] M. Williams et al., Phys.Rev. C80, 045213 (2009).
[16] V. Crede et al., Phys.Rev. C80, 055202 (2009).
[17] K. Nakayama and H. Haberzettl, Phys.Rev. C73, 045211
(2006).
[18] F. Huang, H. Haberzettl, and K. Nakayama, Phys.Rev.
C87, 054004 (2013).
[19] W.-T. Chiang, S. N. Yang, L. Tiator, M. Vanderhaeghen,
and D. Drechsel, Phys.Rev. C68, 045202 (2003).
[20] X.-H. Zhong and Q. Zhao, Phys.Rev. C84, 065204
(2011).
[21] V. Tryasuchev, Phys.Atom.Nucl. 76, 748 (2013).
[22] A. D’Angelo et al., Nucl.Instrum.Meth. A455, 1 (2000).
[23] D. Babusci, G. Giordano, and G. Matone, Phys.Lett.
B355, 1 (1995).
[24] A. Lleres et al., Eur.Phys.J. A31, 79 (2007).
[25] P. Levi Sandri et al., Nucl.Instrum.Meth. A370, 396
(1996).
[26] M. Castoldi et al., Nucl.Instrum.Meth. A403, 22 (1998).
[27] F. Ghio, B. Girolami, M. Capogni, L. Casano, L. Ciciani,
et al., Nucl.Instrum.Meth. A404, 71 (1998).
[28] V. Kuznetsov et al., Nucl.Instrum.Meth. A487, 396
(2002).
[29] V. Bellini et al., Nucl.Instrum.Meth. A386, 254 (1997).
[30] O. Bartalini et al. (GRAAL Collaboration), Eur.Phys.J.
A26, 399 (2005).
[31] P. Corvisiero et al., Nucl.Instrum.Meth. A346, 433
(1994).
[32] R. Brun, F. Carminati, and S. Giani, Tech. Rep. CERN-
W5013, CERN-W-5013 (1994).
[33] J. Beringer et al. (Particle Data Group), Phys.Rev. D86,
010001 (2012).
[34] G. Mandaglio et al., EPJ Web of Science 72, 00016
(2014).
[35] O. Bartalini et al. (GRAAL collaboration), Eur.Phys.J.
A33, 169 (2007), 0707.1385.