Evidence for Narrow N*(1685) Resonance in Quasifree Compton Scattering on the Neutron

Abstract

The first study of quasi-free Compton scattering on the neutron in the energy
range of $E_{\gamma}=0.75 - 1.5$ GeV is presented. The data reveals a narrow
peak at $W\sim 1.685$ GeV. This result, being considered in conjunction with
the recent evidence for a narrow structure at $W\sim 1.68$GeV in the $\eta$
photoproduction on the neutron, suggests the existence of a new nucleon
resonance with unusual properties: the mass $M\sim 1.685$GeV, the narrow width
$\Gamma \leq 30$MeV, and the much stronger photoexcitation on the neutron than
on the proton.

Full text

Page 1

arXiv:1003.4585v3 [hep-ex] 21 Feb 2011
Evidence for a Narrow N∗(1685) Resonance in Quasifree Compton Scattering on the
Neutron
V. Kuznetsov1,2, M. V. Polyakov3,4, V.Bellini5,6, T. Boiko7, S. Chebotaryov1, H.-S.Dho1,
G.Gervino8,9, F.Ghio10,11, A.Giusa5,6, A. Kim1,12, W. Kim1, F.Mammoliti5,6, E. Milman1, A. Ni1,
I.A. Perevalova13, C.Randieri5,6, G.Russo5,6, M.L.Sperduto5,6, C.M.Sutera5, and A.N. Vall13
1Kyungpook National University, 702-701, Daegu,Republic of Korea
2Institute for Nuclear Research, 117312, Moscow, Russia
3Institute f¨ ur Theoretische Physik II, Ruhr-Universit¨ at Bochum, D - 44780 Bochum, Germany,
4Petersburg Nuclear Physics Institute, Gatchina, 188300, St. Petersburg, Russia,
5INFN - Sezione di Catania, via Santa Sofia 64, I-95123 Catania, Italy
6Dipartimento di Fisica ed Astronomia, Universit´ a di Catania, via Santa Sofia 64, I-95123 Catania, Italy
7Belarussian State University, 220030, Minsk, Republic of Belarus
8Dipartimento di Fisica Sperimentale, Universit´ a di Torino, via P.Giuria, I-00125 Torino, Italy
9INFN - Sezione di Torino, I-10125 Torino, Italy
10INFN - Sezione di Roma, piazzale Aldo Moro 2, I-00185 Roma, Italy
11Instituto Superiore di Sanit´ a, viale Regina Elena 299, I-00161 Roma, Italy
12Thomas Jefferson National Accelerator Facility, Jefferson Av., 23606 VA, USA and
13Physics Department, Irkutsk State University, Karl Marx str. 1, 664003, Irkutsk, Russia
(Dated: February 22, 2011)
The first study of quasi-free Compton scattering on the neutron in the energy range of Eγ =
0.75 −1.5 GeV is presented. The data reveals a narrow peak at W ∼ 1.685 GeV. This result, being
considered in conjunction with the recent evidence for a narrow structure at W ∼ 1.68 GeV in the
η photoproduction on the neutron, suggests the existence of a new nucleon resonance with unusual
properties: the mass M ∼ 1.685 GeV, the narrow width Γ ≤ 30 MeV, and the much stronger
photoexcitation on the neutron than on the proton.
PACS numbers:
Many properties of known baryons were transparently
explained by the constituent quark model(CQM) [1] that
treats baryons as bound system of three valence quarks
in the ground or excited state. Some baryon properties
remain a mystery: almost half of the CQM-predicted nu-
cleon and ∆ resonances [2] still escape the reliable experi-
mental identification [3] (so-called “missing resonances”).
The chiral quark soliton model (χQSM) is an alterna-
tive view of baryons which are treated as space/flavor
rotational excitations of a classical object - a chiral
mean-field. χQSM predicts the lowest-mass multiplets
of baryons to be the 1/2+octet and 3/2+decuplet - ex-
actly as CQM does. The χQSM predictions for higher
multiplets are different from CQM [4].
Thus, the experimental study of baryon resonances
provides benchmark information for the development
of theoretical models and for finding relations between
them.
In this context the possible observation of a new nar-
row resonance N∗(1685) is of potential importance. Re-
cently, four groups - GRAAL [5], CBELSA/TAPS[6],
LNS [7], and Crystal Ball/TAPS [8] - reported evidence
for a narrow structure at W ∼ 1.68 GeV in the η photo-
production on the neutron. The structure was observed
as a bump in the quasi-free cross section and as a peak
in the invariant-mass spectrum of the final-state η and
the neutron M(ηn) [5, 6, 8]. The width of the bump
in the quasi-free cross section is close to the smearing
caused by Fermi motion of the target neutron bound in a
deuteron target [5]. The width of the peaks observed in
the M(η,n) spectra is close the instrumental resolution
of the corresponding experiments [5, 6, 8].
Furthermore, a sharp resonant structure at W ∼
1.685 GeV was found in the GRAAL beam asymme-
try data for the η photoproduction on the free proton
[9, 10](see also [11]). Such structure is not (or poorly)
seen in the γp → ηp cross section [13]. Any resonance
whose photoexcitation on the proton is suppressed by
any reason, may manifest itself in polarization observ-
ables due to interference effects.
In Refs. [5, 9, 10, 14–16], the combination of the ex-
perimental findings was interpreted as a possible signal
of a nucleon resonance with unusual properties: the mass
near M ∼ 1.68 GeV, the narrow width, and the strong
photoexcitation on the neutron. Alternatively, the au-
thors of Refs. [17, 18] explained the bump in the quasi-
free γn → ηn cross section in terms of the interference
of well-known resonances or as the virtual sub-threshold
KΛ and KΣ photoproduction [19].
If the narrow N∗(1685) does really exist, it can be seen
not only in the η photoproduction but also in other re-
actions on the neutron, e.g. Compton scattering or the
π0photoproduction. On the contrary, the narrow bump
cannot be generated by the interference of wide reso-
nances in these reactions, as they receive contributions
of different (from the η photoproduction) resonances.
In this paper, we present the first measurement of
Compton scattering on the neutron at the photon en-
ergies of Eγ = 0.75 − 1.5 GeV (W ∼ 1.5 − 1.9 GeV)
focusing on the search for the signal of N∗(1685). Simul-

Page 2

2
FIG. 1: Schematic view of the GRAAL detector.
taneously, we investigate the photoproduction of neutral
pions on the neutron and the isospin-mirrored reactions
γp → γp and γp → π0p.
The existing data base for Compton scattering and π0
production on the neutron is scarce. The available data
for Compton scattering is limited to lower energies Eγ≤
400 MeV [20–22]. There is no published data for γn →
π0n cross section in the domain of W ∼ 1.6 − 1.7 GeV.
New preliminary data on the γn → π0n cross section
were presented in [23, 24].
The data was collected at the GRAAL facility [25].
The GRAAL polarized and tagged photon beam is pro-
duced by backscattering of laser light on 6.04 GeV elec-
trons circulating in the storage ring of the ESRF (Greno-
ble, France). The 4π detector (Fig. 1) is designed for the
detection of neutral and charged particles. It is com-
posed of a cylindrically symmetrical central part, for the
detection of the particles emitted at θlab = 25 − 155◦
with respect to a beam axis, and of a forward part for
the detection of the particles emitted at θlab≤ 25◦.
The central part consists of two coaxial cylindrical wire
chambers, a 5 mm thick plastic scintillator barrel, which
provides ∆E information for particle identification, and
a BGO ball made of 480 crystals each of 21 radiation
length. The energy resolution for the detection of pho-
tons at 1 GeV is 3%(FWHM).
The forward part consists of two planar multiwire
chambers, which provide tracking angular resolution of
∼0.5◦for charged particles, and a double hodoscope wall
made of two layers of 3 cm thick plastic scintillator bars
covering an area of 3x3 m2and located 3 m away from
the target. The hodoscope wall is followed by the TOF
lead-scintillator wall which is an assembly of 16 mod-
ules covering the same area as the hodoscope wall. Each
module is a composition of four 300x19x4 cm3scintillator
bars separated by 3 layers of 3 mm thick lead converter.
The wall provides the detection of photons, neutrons and
charged particles with an angular resolution of 3◦and ∆E
information. The TOF resolution is 600 ps (FWHM) for
charged particles and 700-800 ps for neutrons. The esti-
mated efficiency of the detection of photons and neutrons
is about 95 and 22% respectively. The particle identifi-
cation (photons, neutrons, protons or charged pions) in
the forward assembly is achieved by means of coincidence
(anticoincidence) of the corresponding signals in the lead-
scintillator wall and the preceding planar chambers and
the hodoscope wall, and using ∆E-TOF relations. Mo-
menta of the charged particles and neutrons can be recon-
structed from the measured TOF and angular quantities.
Both d(γ,γ′n)p and d(γ,γ′p)n reactions were mea-
sured simultaneously in the kinematics that emphasize
the quasi-free reaction. Scattered photons were detected
in the BGO crystal ball [26]. Recoil neutrons and protons
emitted at Θlab= 3− 23◦were detected in the assembly
of the forward detectors (Fig. 1).
As the first step, the identification of γN final states
was achieved using the criterion of coplanarity, cuts on
the neutron(proton) and photon missing masses, and
comparing the measured TOF and the angle of the recoil
nucleon with the same quantities calculated assuming the
γN → γN reaction. The sample of the selected events
was still contaminated by the events from the π0photo-
production. The π0cross section is about two orders of
magnitude larger than that of Compton scattering.
At the second step, two types of the π0background
were taken into consideration:
i) Symmetric π0→ 2γ decays. The pion decays in two
photons of nearly equal energies. Being emitted in a nar-
row cone along the pion trajectory, such photons imitate
a single-photon hit in the BGO ball;
ii) Asymmetric π0→ 2γ decays. One of the photons
takes the main part of the pion energy. It is emitted
nearly along the pion trajectory. The second photon is
soft and is emitted into a backward hemisphere relative
to the pion track. Its energy depends on the pion energy
and may be as low as 6 − 10 MeV.
The symmetric events were efficiently rejected by an-
alyzing the distribution of energies deposited in crystals
attributed to the corresponding cluster in the BGO ball.
The efficiency of this rejection was verified in simulations
and found to be 99%.
The asymmetric π0→ 2γ decays present the ma-
jor problem. The GRAAL detector provides the low-
threshold (5 MeV) detection of photons in the nearly 4π
solid angle. If one (high-energy) photon is emitted at
backward angles Θlab = 130 − 150◦, the second (low-
energy) photon is detected in the BGO ball or in the for-
ward lead-scintillator wall (Fig. 1). This feature makes it
possible to suppress the π0photoproduction at backward
angles θcm= 150 − 165◦. At more forward angles one of
the photons may escape out from the detector through
the backward hole. Consequently, the background rejec-
tion deteriorates dramatically.
For the further selection of events the missing energy
Emiswas employed
Emis= Eγ− Eγ′ − TN(θN),(1)

Page 3

3
0
20
40
60
80
100
120
140
160
-0.100.1
Emis (GeV)
0.2
Counts
0
100
200
300
400
500
600
700
800
900
-0.10 0.1
Emis (GeV)
0.2
Counts
FIG. 2: On the left: Simulated spectrum of missing energy
for a free-proton target. The area colored in magenta shows
Compton events. The blue area corresponds to the photo-
production of π0s. On the right: Spectrum of missing energy
measured in experiment with a free-proton target.
where Eγdenotes the energy of the incoming photon, Eγ′
is the energy of the scattered photon, and TN(θN) is the
kinetic energy of the recoil neutron(proton).
The simulated spectrum of the missing energy for the
free proton is shown in the left panel of Fig. 2. π0events
form a wide distribution. Compton events generate a
narrow peak centered at Emis= 0. The events in the re-
gion of this peak mainly belong to Compton scattering.
On the contrary, the cut Emis≥ 0.05 GeV selects only
π0events. The right panel of Fig. 2 shows the same spec-
trum measured with the free-proton target. This spec-
trum is similar to the simulated one.
The right column of Fig. 3 shows the missing energy
spectra corresponding to reactions on the free proton,
(the first row), the quasi-free proton (the second row),
and the quasi-free neutron (the third row). The data
obtained on the quasi-free nucleons are smeared by Fermi
motion.
The left and central columns show the distributions
of events which correspond to the cuts −0.05 GeV ≤
Emis≤ 0.04 GeV and 0.07 GeV ≤ Emis≤ 0.15 GeV re-
spectively. The first cut selects events around the Comp-
ton peak. These events mostly correspond to Compton
scattering with some contamination of π0events. The
second cut selects mostly π0events.
The distributions of π0events obtained on the free and
quasi-free proton are similar and exhibit a wide bump
near W ∼ 1.65 GeV. This bump is well seen in the pub-
lished data for this reaction [25]. The Compton events
on the proton indicate a similar structure. This structure
was also seen in the previous measurements [28]. On the
contrary, the distribution of π0events on the neutron
is flat. This observation is in agreement with the pre-
liminary results from Crystal Ball/TAPS [23] and LNS
Collaborations [24].
The distribution of Compton events on the neutron
(lower row, left column of Fig. 3) reveals a narrow peak
at W ∼ 1.685 GeV. The peak is similar to that observed
in the η photoproduction on the neutron.
In left panel of Fig. 4 the 2nd-order-polynomial (the
0
100
200
300
400
500
1.5 1.6 1.7 1.8
W (GeV)
Counts
0
100
200
300
400
500
1.5 1.6 1.7 1.8
W (GeV)
0
200
400
600
800
-0.1 0 0.1 0.2
Emis (GeV)
0
100
200
300
1.5 1.6 1.7 1.8
W (GeV)
Counts
0
100
200
300
1.5 1.6 1.7 1.8
W (GeV)
0
200
400
-0.1 0 0.1 0.2
Emis (GeV)
0
20
40
60
1.5 1.6 1.7 1.8
W (GeV)
Counts
0
20
40
60
1.5 1.6 1.7 1.8
W (GeV)
0
20
40
60
-0.1 0 0.1 0.2
Emis (GeV)
FIG. 3: Experimental data obtained on the free proton (up-
per row), quasi-free proton(middle row), and quasi-free neu-
tron (lower row). Right column: spectra of missing energy.
Magenta and blue areas indicate cuts used for the selection
of Compton and π0events respectively. Middle column: W
distributions of events corresponding to blue areas in the
missing-energy spectra (π0events). Left column: distribu-
tion of events corresponding to magenta areas in the missing-
energy spectra (dominance of Compton events).
background hypothesis) fit for Compton events on the
neutron in the interval W = 1.585−1.888 GeV is shown
by the dashed line. The solid line in the same figure
shows the background-plus-Gaussian fit. The χ2of both
fits are 3.7/6 and 18.5/9 respectively. The log likelihood
ratio of these two hypotheses (?2ln(LB+S/LB)) corre-
sponds to the confidence level of ∼ 4.6σ The extracted
peak position is M = 1686 ± 7stat± 5systMeV. and the
r.m.s is σ ∼ 12 ± 5 MeV (Γ ≈ 28 ± 12 MeV). The sys-
tematic uncertainty in the mass position is due to the
uncertainties in the calibration of the GRAAL tagger.
The middle panel of the Fig. 4 shows the similar distri-
bution obtained with the wider cut on the missing energy
−0.1 ≤ Emis≤ 0.075GeV. The contamination of the π0
background is increased (espesially at the higher ener-
gies) while the peak at W ∼ 1.685 GeV remains almost
unaffected.
The right panel of the Fig. 4 presents the simulated
yield of events obtained with the same cuts as in the left
panel of the same figure. The event generator used in
MC included a flat Compton cross section. Neither of
any peak appeared in the W spectrum of events.

Page 4

4
0
10
20
30
40
50
60
1.51.61.71.8
W (GeV)
Counts
0
10
20
30
40
50
60
70
1.51.61.7 1.8
W (GeV)
Counts
0
10
20
30
40
50
60
70
1.51.61.71.8
W (GeV)
Counts
FIG. 4: Left panel: The W spectrum of events obtained with
the cut on the missing energy −0.05 ≤ Emis ≤ 0.04GeV .
Solid line indicates the Gaussian-plus-2-order-polynomial fit.
Dashed line corresponds to the 2-order-polynomial fit only.
Middle panel: The W spectrum of events obtained with the
cut −0.1 ≤ Emis ≤ 0.075GeV . Right panel: The simulated
W spectrum obtained with the same cuts as in the left panel.
The observation of the narrow peak, its position and
width, being considered togerther with the high-statistics
results on the η photoproduction on the neutron [5–8]
and the beam asymmetry data on the free proton [9,
10], supports the existence of a narrow nucleon N∗(1685)
resonance and challenges the explanations [17, 18] of the
bump structure in the quasi-free γn → ηn cross section
in terms of the interference of well-known resonances.
The assumption on the virtual sub-thershold KΛ and
KΣ photopoduction [19] cannot be exluded. However,
it requires the explanation why this effect occurs in the
γn → ηn and γn → γn reactions and is not seen in
γn → π0n.
The putative N∗(1685) resonance is dominantly pho-
toexcited on the neutron whereas its photoexcitation on
the proton is suppressed. Such feature was suggested
in Ref. [29] as the benchmark signature of a resonance
belonging to the flavour SU(3) antidecuplet of exotic
baryons predicted by χQSM [4]. Interestingly, the mass,
the narrow width, and the isospin of N∗(1685) are also
in agreement with the predictions for this member of the
antidecuplet [30–32].
The decisive identification of N∗(1685), in particular
its definite association with the second member of the
exotic antidecuplet, requires further efforts and more ex-
perimental data. A critical point is to determine the spin
and the parity of this state. It is worthwhile to note that
the fit of the beam asymmetry data for the η on the pro-
ton resulted in three possible quantum numbers, namely
P11, or P13, or D13[9, 10].
In summary, we report the evidence for a narrow res-
onance structure in the Compton scattering on the neu-
tron. This structure is quite similar to that observed
in η photoproduction on the neutron. The combination
of experimental observations suggests the existence of a
narrow nucleon resonance with unusual properties: the
mass M ≈ 1.685 GeV, the narrow width Γ ≤ 30 MeV,
the much stronger photoexcitation on the neutron than
on the proton, and the suppressed branching ratio to πN
final states.
It is our pleasure to thank the staff of the Euro-
pean Synchrotron Radiation Facility (Grenoble, France)
for the stable beam operation during the experimental
runs.This work was supported by Basic Science Re-
search Program throgh the National Research Founda-
tion of Korea (NRF) funded by the Ministry of Educa-
tion, Science and Technology (grant 2010-0013430), and
by SFB/Transregio 16 (Germany). The work of M.V.P.,
I.A.P. and A.N.V. is also supported by the grant 2010-
1.5-508-005 of Russian Ministry for Education and Re-
search. The authors are grateful to N. Sverdlova for her
comments on the manuscript, and to Jiyoung Ha for the
administrative support of this work.
[1] N. Isgur and G. Karl, Phys.Rev. D 18, 4187, (1978);
N. Isgur and G. Karl, Phys. Lett. B74, 353, (1978).
[2] S.Capstick and W.Roberts, Prog. Part. Nucl. Phys. 45,
S241 (2000) .
[3] C.Amsler et al., Phys. Lett. B667, 1 (2008).
[4] D. Diakonov, V. Petrov and M. V. Polyakov, Z. Phys. A
359, 305 (1997).
[5] V. Kuznetsov et al., Phys. Lett. B 647, 23 (2007).
[6] I. Jaegle et al. [CBELSA Collaboration and TAPS Col-
laboration], Phys. Rev. Lett. 100, 252002 (2008).
[7] F. Miyahara et al., Prog. Theor. Phys. Suppl. 168, 90
(2007).
[8] D. Werthmuller [for the Crystal Ball/TAPS collabora-
tions], Chin. Phys. C 33, 1345 (2009).
[9] V. Kuznetsov et al., Acta Phys. Polon. B 39, 1949 (2008).
[10] V. Kuznetsov and M. V. Polyakov, JETP Lett. 88, 347
(2008).
[11] It is worth to noting that the authors of Ref. [12] arrived
at a different conclusion. The detailed critique of the re-
sults from Ref. [12] is available in Ref. [9]. This critique
remains unreplied.
[12] O. Bartalini eta al., Eur. Phys. J. A 33, 169 (2007).
[13] V. Crede et al. [CBELSA/TAPS Collaboration], Phys.
Rev. C 80, 055202 (2009); F. Renard et al. [GRAAL
Collaboration], Phys. Lett. B 528 215 (2002); M. Dugger
et al. [CLAS Collaboration], Phys. Rev. Lett. 89, 222002
(2002).
[14] Y. I. Azimov, et al., Eur. Phys. J. A 25, 325 (2005).
[15] A. Fix, L. Tiator and M. V. Polyakov, Eur. Phys. J. A
32, 311 (2007).
[16] K. S. Choi, S. i. Nam, A. Hosaka and H. C. Kim, Phys.
Lett. B 636, 253 (2006).
[17] V. Shklyar, H. Lenske and U. Mosel, Phys. Lett. B 650,
172 (2007).
[18] A. V. Anisovich et al., Eur. Phys. J. A 41, 13 (2009).
[19] M. Doring and K. Nakayama, Phys. Lett. B 683, 145
(2010).
[20] K. W. Rose et al., Nucl. Phys. A 514, 621 (1990).
[21] N.R. Kolb et al., Phys. Rev. Lett. 85, 1388, 2000.
[22] K. Kossert et al., Eur. Phys. J. A 16 259 (2003).

Page 5

5
[23] B. Krusche, Talk at Int. Workshop “Narrow Nucleon res-
onances: Predictions, Evidences, Perspectivs”, June 8 -
10, 2009, Edinburg, Scotland.
[24] H. Shimizu, Talk at Int. Workshop “Narrow Nucleon res-
onances: Predictions, Evidences, Perspectivs”, June 8 -
10, 2009, Edinburgh, Scotland.
[25] General description of the GRAAL facility is available in:
O. Bartalini et al., Eur. Phys. J. A26, 399 (2005).
[26] F. Ghio et al., Nucl. Inst. a. Meth. A404, 71 (1998).
[27] V. Kouznetsov et al., Nucl. Inst. a. Meth. A487, 396
(2002).
[28] T. Ishii et al., Nucl. Phys. B 254, 458 (1985);
Y. Wada et al., Nucl. Phys. B 247, 313 (1984).
[29] M. V. Polyakov and A. Rathke, Eur. Phys. J. A 18, 691
(2003).
[30] D. Diakonov and V. Petrov, Phys. Rev. D 69 094011
(2004).
[31] R. A. Arndt et al., Phys. Rev. C 69, 035208 (2004).
[32] J. R. Ellis, M. Karliner and M. Praszalowicz, J. High.
Energy Phys. 04 (2004) 002; M. Praszalowicz, Acta Phys.
Polon. B 35, 1625 (2004); Ann. Phys. 13, 709 (2004).