Measurements of hadron form factors at BESIII

Cristina Morales Morales, and

Citation: AIP Conference Proceedings 1735, 050006 (2016);

View online: https://doi.org/10.1063/1.4949425
View Table of Contents: http://aip.scitation.org/toc/apc/1735/1
Published by the American Institute of Physics
 Measurements of Hadron Form Factors at BESIII
Cristina Morales Morales on Behalf of BESIII Collaboration1,a)
1
Helmholtz-Institut Mainz, 55099 Mainz, Germany.
a)
Corresponding author: c.morales-morales@gsi.de

Abstract. BEPCII is a symmetric e+ e− -collider located in Beijing running at center-of-mass energies between 2.0 and 4.6 GeV.
This energy range allows the BESIII-experiment to measure hadron form factors both from direct e+ e− -annihilation and from
initial state radiation processes. In this paper, results on e+ e− → p p̄ based on data collected by BESIII in 2011 and 2012 are
presented. We also present preliminary results on e+ e− → ΛΛ̄ based on the same data samples at 4 center-of-mass energies. BESIII
results obtained from e+ e− → π+ π− using the initial state radiation technique at the center-of-mass energy of 3.773 GeV are also
summarized. Finally, expectations on the measurement of baryon electromagnetic form factors from the BESIII high luminosity
energy scan in 2015 and from initial state radiation processes at different center-of-mass energies are also explained.

INTRODUCTION
Form factors (FFs) account for the non point-like structure of hadrons. Depending on its spin, s, a hadron has 2s+1
form factors. The FFs are analytic functions of the momentum transferred by the virtual photon, q. They are real in the
space-like region (q2 < 0) and complex in the time-like region (q2 > 0) for q2 > 4m2π . FFs at q2 < 0 are determined by
elastic scattering of electrons from hadrons available as targets. FFs at q2 > 0 are measured in annihilation processes
e+ e− ↔ hh.
Spin 1/2 baryons have two electro-magnetic FFs. Here the Sachs FFs in the form of G E and G M are used. The
Born differential cross section of the e+ e− -annihilation into a baryon-antibaryon pair, e+ e− → BB, in center-of-mass
(c.m.) reads [1]
 
dσBorn (q2 , θ∗B ) α2 βC 2 ∗ 1 2 ∗
= (1 + cos θ )|G M (q 2 2
)| + sin θ |G E (q 2 2
)| , (1)
dΩ 4q2 B
τ B

√
where θ∗B is the polar angle of the baryon, m is its mass, τ = 4m2 /q2 and β = 1 − 1/τ. The Coulomb factor,
C = y/(1 − exp(−y)) with y = πα/β, accounts for the electromagnetic BB interactions of point-like baryons [2] and is
equal to 1 for neutral baryon pairs. Angular integration of the previous equation gives the total cross section:
 
4πα2 βC 1
σ (q ) =
Born 2
|G M (q )| + |G E (q )| .
2 2 2 2
(2)
3q2 2τ

Experiments usually quote measurements of |G M (q2 )| under the working hypothesis |G E (q2 )| = |G M (q2 )|. By defining
the effective form factor (EFF) as

2τ|G M (q2 )|2 + |G E (q2 )|2 σBorn (q2 )

|G(q2 )|2 = = , (3)
2τ + 1 (1 + 2τ
1
)( 4πα3q2βC )
2

it is possible to compare with older measurements. The simultaneous extraction of |G E | and |G M | without assumptions
is only possible by measuring the angular distributions of the outgoing particles (Eq. 1).
The process of e+ e− -annihilation can be accompanied by the emission of one or several high energy photons
from the initial state (ISR). The following equation relates the differential cross section of the ISR process and the
cross section of the direct annihilation:

XVIth International Conference on Hadron Spectroscopy

AIP Conf. Proc. 1735, 050006-1–050006-6; doi: 10.1063/1.4949425
Published by AIP Publishing. 978-0-7354-1389-4/$30.00

050006-1
 dσISR 1
= · W(s, x, θγ∗ ) · σBorn (q2 ), (4)
dq2 s
√ √
where x = 2Eγ∗ / s = 1 − q2 /s, s is the c.m. energy of the collider, and Eγ∗ and θγ∗ are the energy and polar angle
of the ISR photon in c.m. The radiator function W(s, x, θγ∗ ) (Equation 5), describes the probability of the ISR photon
emission ⎛ ⎞
∗ α ⎜⎜⎜⎜ 2 − 2x + x2 x2 ⎟⎟⎟⎟
W(s, x, θγ ) = ·⎜ − ⎟⎠ , (5)
πx ⎝ sin2 θγ∗ 2
with α the electromagnetic fine structure constant [3]. The emission of the ISR-photon at very small or very large
polar angle in c.m. and with low energies is favored. Depending on the energy of the ISR photon, the hadronic mass
in the final state is reduced and the hadronic cross section can be extracted for all masses below the actual c.m. energy
of the collider up to the production threshold of the hadronic state.

THE BESIII EXPERIMENT AND DATA SETS

√
BEPCII is a double ring e+ e− symmetric collider running at s from 2.0 to 4.6 GeV. The design luminosity is
−2 −1
1 × 10 cm s at a beam energy of 1.89 GeV. BESIII is a cylindrical detector which covers 93% of the full
33

solid angle [4]. It consists of the following detectors: a Multilayer Drift Chamber (MDC); a Time-of-Flight plastic
scintillator (TOF); a CsI(Tl) Electro-Magnetic Calorimeter (EMC); a superconducting magnet of 1T and finally a
Muon Chamber (MUC). BESIII has accumulated the world s largest samples of e+ e-collisions in the tau-charm region.
For the measurement of hadron form factors, the scan data between 2.0 and 4.6 GeV can be used for the study
of e+ e− → π+ π− , N N̄, ΛΛ̄, etc., while the data collected at different charmonium and XYZ-states above 3.77 GeV
(7.4 fb−1 ) can be used for the study of the corresponding ISR channels.

NUCLEON FORM FACTORS

Recently, BESIII has published the measurement of the channel e+ e− → p p̄ at 12 c.m. energies ranging from 2.2324
to 3.6710 GeV [5]. These data were collected in 2011 and 2012 and correspond to a luminosity of 157 pb−1 . The Born
cross section was extracted according to
Nobs − Nbkg
σBorn = , (6)
L · (1 + δ)
where the number of background events is subtracted from the observed signal event candidates, normalized with the
luminosity at each scan point, L, and corrected with the selection efficiency of the process, , times the ISR radiative
correction factor up to next-to-leading order (NLO) (1 + δ). The ConExc generator [6] was used both for the efficiency
and the√ radiative factor evaluation. The accuracy in the cross section measurements was between 6.0% and 18.9%
up to s < 3.08 GeV. The EFF was extracted according to Eq. 3 and is shown in Fig. 1 (a) together with previous
experimental results [11, 12, 13, 14, 15, 16, 17]. A fit to the polar angular distribution of the proton in c.m. was
performed according to Eq. 1 and the ratio of the FFs and |G M | was extracted for three energy points. The results from
this measurement are shown in Fig. 1 (b). An expansion in moments of the angular distribution of the proton [18] was
also used to evaluate the ratio of the FFs and |G M |. A comparison of the results obtained by the two methods is shown
in Table 1. While the measurements by the different experiments concerning the EFF show very good agreement, this
is not the case of the ratio of the FFs, where the measurements by BaBar [13] and PS170 [14] disagree for low q.
At present, the precision in the measurement of the ratio of the proton FFs is dominated by statistics. Using
the data collected by BESIII in 2015, accuracies below 10% are expected to be achieved. Considering the statistics
collected at different charmonium and XYZ-states above 3.77 GeV (7.4 fb−1 ), BESIII has higher visible luminosity
than BaBar (500 fb−1 ) for the study of the ISR channel e+ e− → p p̄γ. This is due to the fact that BESIII can also
perform untagged analysis of these events (ISR photon emitted at very low or very large polar angles) starting at q =
2.0 GeV due to the lower c.m. energies compared to BaBar. Comparable precision to BaBar is therefore expected in
the measurement of the hadronic section and EFF from the study of e+ e− → p p̄γ in BESIII.
Experimental results on the annihilation cross section into nn̄ are very scarce [7, 8, 9, 10]. Like in the case of
the proton, two different measurements are possible in BESIII: e+ e− → nn̄ and e+ e− → nn̄γ. In both cases, while the
n̄-annihilation releases a large amount of energy (more than twice the neutron mass) in a broad hadronic shower in

050006-2
 2.0
BESIII
BESIII

Effective Form Factor

BaBar 1.8
BaBar

-1
BESII 1.6 PS170
10 FENICE
1.4
CLEO

|GE/GM|
E760 1.2
E835 1.0
0.8
0.6
0.4
0.2
10-2 0.0
2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 2.0 2.2 2.4 2.6 2.8 3.0
q (GeV/c ) q (GeV/c)

FIGURE 1. Proton effective form factor (a) and ratio of proton electro-magnetic form factors (b) in the time-like region.

TABLE 1. Results on |G E /G M | and |G M | by fitting the proton angular dis-

tribution as well as by the method of moments at different c.m. energies.
√
s (GeV) |G E /G M | |G M | (×10−2 )
Fit on cos θ p
2.2324 0.87 ± 0.24 ± 0.05 18.42 ± 5.09 ± 0.98
2.4000 0.91 ± 0.38 ± 0.12 11.30 ± 4.73 ± 1.53
(3.0500-3.0800) 0.95 ± 0.45 ± 0.21 3.61 ± 1.71 ± 0.82
Method of moments
2.2324 0.83 ± 0.24 18.60 ± 5.38
2.4000 0.85 ± 0.37 11.52 ± 5.01
(3.0500-3.0800) 0.88 ± 0.46 3.34 ± 1.72

the EMC, neutrons do not produce such a broad shower and deposit much less energy. The sources of background
in detecting n and n̄ are other neutral particles, like photons, neutral kaons and mostly beam background and neutral
cosmic rays. In BESIII, the depth of the EMC amounts to 15X0 or 0.7λI , which means that the probability that a
nucleon interacts with the EMC material is of about 50%. Since the energy resolution in the EMC is 2.5% (5%) for
a 1 GeV photon in the barrel (endcaps), it is possible to at least reconstruct the n̄-annihilation. The selection strategy
could be based on the identification of the n̄ and the ISR photon through shower shape analysis in the EMC and then
detection of n in EMC with the help of the kinematics of the process. The detection efficiency of n̄ vs photons can be
extremely increased with the use of Multivariate Analysis Tools (Neural Networks). Also the use of other subdetectors
like the MUC and the TOF can be exploited such that the detection of the n is not needed to identify these events. The
probability of interaction of a nucleon inside the MUC is 96%. The TOF counters could be fired by the n̄-annihilation,
making the detection of n unnecessary. Using the available scan data from 2015 and the data collected at different
resonances above 3.77 GeV, BESIII can provide a measurement of the cross section of the Born annihilation process
in a wide continuous q range from nn̄-threshold up to 3.08 GeV. Furthermore, not only the EFF can be extracted but
also the ratio of the neutron FFs could be extracted for the first time using BESIII 2015 scan data.

HYPERON FORM FACTORS

BESIII has preliminary results on the measurement of the channel e+ e− → ΛΛ̄. The analysis is based on 40.5 pb−1
collected in 4 different scan points during 2011 and 2012. The lowest energy point is at 2.2324 GeV, only 0.4 MeV
above the ΛΛ̄-threshold. This makes it possible to measure the cross section almost at threshold. To use as much
statistics as possible, both events where Λ and Λ̄ decay to the charged mode (BR(Λ → pπ− ) = 64%) and events where
the Λ̄ decays to the neutral mode (BR(Λ̄ → n̄π0 ) = 36%) have been selected. In the first case, the identification relied
on finding two monoenergetic charged pions and a possible p̄-annihilation. In the second case, the n̄-annihilation
was identified through the use of Multivariate Analysis of EMC variables and additonally a monoenergetic π0 was

050006-3
 400 0.8

Cross section (pb)

Effective Form Factor

350 0.7
BESIII BESIII
300 BaBar 0.6 BaBar
250 DM2 0.5
200 BESIII preliminary 0.4 BESIII Preliminary
150 0.3
100 0.2
50 0.1
0 0.0
2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6
q (GeV/c) q (GeV/c)

FIGURE 2. Measurements of e+ e− → ΛΛ̄ cross section (a) and Λ effective form factor (b).

reconstructed to fully identify the channel. For the other 3 energy points, only the charged decay modes of Λ and Λ̄
were reconstructed by identifying the charged tracks and using the event kinematics. The preliminary results on the
measurement of the Born cross section are shown in Figure 2 (a) together with previous measurements [19, 20]. The
cross section at threshold is found to be 318 ± 47 ± 37 pb, clearly different from zero. Given that the Coulomb factor in
Eq. 1 is equal to 1 for neutral baryon pairs, the cross section is expected to go to zero at threshold. This result confirms
BaBar s measurement close to threshold [20] with the difference that BESIII measurement is not integrated within a
finite bin, like in the case of BaBar s measurement through ISR. The BESIII measurement improves at least by 10%
previous results at low q and even more above 2.4 GeV. From the measurement of the cross section, the extraction of
the EFF was also possible (Fig. 2 (b)). The cross section and EFF results are summarized in Table 2.

TABLE 2. Results on e+ e− → ΛΛ̄ cross section and Λ effective form factor

measurements.
√
s (GeV) σBorn (pb) |G| (×10−2 )
2.2324
Λ → pπ− , Λ̄ → p̄π+ 325 ± 53 ± 46
Λ̄ → n̄π0 (3.0 ± 1.0 ± 0.4) × 102
combined 318 ± 47 ± 37 63.2 ± 4.7 ± 3.7
2.4000 133 ± 20 ± 19 12.9 ± 1.0 ± 0.92
2.8000 15.3 ± 5.4 ± 2.0 4.2 ± 0.7 ± 0.3
3.0800 3.9 ± 1.1 ± 0.5 2.21 ± 0.31 ± 0.14

Using BESIII scan data from 2015, a full determination of the lambda FFs is possible. The imaginary part of
the FFs leads to polarization observables. Since the Λ decays through parity violation to proton and pion, the relative
phase between G E and G M , φ, can be extracted from

dN sin2θΛ sin2φ/τ 3
∝ 1 + αΛ Pn cosθ p and Pn = − = < cosθ p >, (7)
dcosθ p Rsin2 θΛ /τ + (1 + cos2 θΛ )/R αΛ

where N is the number of Λ → pπ− decays, θ p is the angle between the p and the polarization axis of the Λ in the Λ
c.m., αΛ is the Λ asymmetry, Pn is the Λ polarization, θΛ is the polar angle of the Λ in the c.m., and R is the ratio of the
Λ FFs. The same is also valid for the Λ̄ → p̄π+ , therefore both decays can be added and used in the polarizatioin mea-
surement. The expected statistical accuracies from the analysis of e+ e− → ΛΛ̄ from BESIII scan data in 2015 range
between 6 and 17%. The corresponding accuracies for the ratio of the Λ FFs range between 14 and 29%. Similar mea-
surements might also be possible in other hyperon channels like e+ e− → ΛΣ̄0 , Σ0 Σ̄0 , Σ+ Σ̄− , Ξ0 Ξ̄0 , Σ− Σ̄+ , Ω− Ω̄+ , Λ+c Λ̄−c .

050006-4
 45
BESIII fit KLOE 08 368.1 ± 0.4 ± 2.3 ± 2.2
40
BESIII
35 BaBar 09 376.7 ± 2.0 ± 1.9

2 30
|F
| KLOE 10 365.3 ± 0.9 ± 2.3 ± 2.2
25

20
KLOE 12 366.7 ± 1.2 ± 2.4 ± 0.8
15

10 BESIII 368.2 ± 2.5 ± 3.3

0.6 0.65 0.7 0.75 0.8 0.85 0.9 360 365 370 375 380 385 390 395
q (GeV/c) aπμπ,LO(600 - 900 MeV) [10-10]

FIGURE 3. Measured squared pion form factor and Gounaris-Sakurai fit parametrisation (a). Leading order hadronic vacuum
polarization 2π contributions to (g − 2)μ (b).

PION FORM FACTOR

Presently, the experimental and Standard Model (SM) values of the anomalous magnetic moment of the muon aμ =
(g − 2)μ /2 differ by more than 3 standard deviations [21, 22]. The SM calculation does not only receive contributions
from QED, but also from weak and strong interactions and probably also from Beyond SM interactions. In order to
improve the SM predictions of aμ , the most important uncertainty is the contribution of the strong interaction. The
hadronic contribution is split in two parts: the Hadronic Vacuum Polarization (HVP) and the Light-by-Light Scattering.
The measurements of the hadronic cross section in e+ e− -annihilation at low energies can be used to compute HVP by
means of dispersion integrals. In particular, e+ e− → π+ π− contributes to more than 70% to the total HVP contribution.
The ISR method was used by BESIII at 3.773 GeV (2.93 fb−1 ) to measure the channel e+ e− → π+ π− γ and access
the q region between 600 and 900 MeV, where the ρ resonance is located [23]. This energy range contributes more
than 70% to the two-pion contribution to aππ μ and to about 50% of the total hadronic vacuum polarization correction
of aμ . The analysis was performed by tagging the ISR photon and identifying the pions in the final state. The selection
criteria included the use of Multivariate Analyis to precisely separate pions from muons. For the extraction of the
cross section Eq. 6 was used, with the signal efficiency (including ISR), L the ISR luminosity, and additionally a
vacuum polarization factor. Phokhara NLO generator [24] was used to measure and the vacuum polarization factor.
The Gounaris-Sakurai parametrization [25] for the ρ − ω interference was applied to fit the π+ π− cross section and
obtain the pion form factor (Fig. 3 (a)). The obtained fit parameters agree well with existing results from PDG [26]
(Table 3).

TABLE 3.
Parameter BESIII Value PDG2014
mρ (MeV/c ) 2
776.0 ± 0.4 775.26 ± 0.25
Γρ (MeV) 151.7 ± 0.7 147.8 ± 0.9
mω (MeV/c2 ) 782.2 ± 0.6 782.65 ± 0.12
Γω (MeV) fixed to PDG 8.49 ± 0.08
|cρ |(10−3 ) 1.7 ± 0.2 -
|φω | (rad) 0.04 ± 0.13 -

As input for the calculation of aππ

μ the bare cross section is needed. It is obtained from the previous cross section
corrected for final state radiation (FSR). The FSR correction factor is determined with Phokhara NLO and theory [27].
The bare cross section and the pion form factor were measured with an accuracy of 0.9%. The contribution of the cross
section measurement to the hadronic contribution to aμ was also calculated for the energy range between 600 and 900
MeV. As summarized in Fig.3 (b), the BESIII result, aππ,LO
μ (600 − 900MeV) = (370.0 ± 2.5stat ± 3.3sys ), was found to
be in between the corresponding values of previous measurements [28, 29].

050006-5
 SUMMARY
BESIII is an excellent laboratory for the measurement of hadron FFs, since both ISR and scan measurements can be
performed. Results on the measurement of proton FFs from scan data from 2011 and 2012 and pion FFs from ISR
measurements at 3.773 GeV have been recently published. There are also preliminary results on lambda EFF based on
scan data from 2011 and 2012. In 2015 BESIII performed a high luminosity scan (555 pb−1 ) in the region between 2.0
to 3.08 GeV with the aim to measure baryon FFs and the inclusive hadronic cross section, Rincl , with unprecedented
accuracy. Currently there are 7.4fb−1 data above 3.77 GeV which could also be used for ISR measurements. The
visible luminosity of this data is comparable to BaBar s visible luminosity.

ACKNOWLEDGMENTS
This work was supported by German Research Foundation DFG under Collaborative Research Center CRC-1044.

REFERENCES
[1] A. Zichichi, S. M. Berman, N. Cabibbo and R. Gatto, Nuovo Cimento 24 170 (1962).
[2] C. Tzara, Nucl. Phys. B 18 216-252 (1970).
[3] G. Bonneau and F. Martin, Nucl. Phys. B 27 381 (1971).
[4] M. Ablikim et al. [BESIII Collaboration], Nucl. Instrum. Meth. A 614, 345 (2010).
[5] M. Ablikim et al. [BESIII Collaboration], Phys. Rev. D 91, 112004 (2015).
[6] R. G. Ping, Chin. Phys. C 38, 083001 (2014).
[7] B. Delcourt et al. [DM1 collaboration], Phys. Lett. B 86, 395 (1982).
[8] D. Bisello et al. [DM2 Collaboration], Nucl. Phys. B 224, 379 (1983); Z. Phys. C 48, 23 (1990).
[9] A. Antonelli et al. [FENICE Collaboration], Nucl. Phys. B 517, 3 (1998).
[10] N. Achasov et al. [SND Collaboration], Phys. Rev. D 90, 112007 (2014).
[11] M. Ablikim et al. [BES Collaboration], Phys. Lett. B 630, 14 (2005).
[12] T. K. Pedlar et al. [CLEO Collaboration], Phys. Rev. Lett. 95, 261803 (2005).
[13] J. P. Lees et al. [BABAR Collaboration], Phys. Rev. D 73, 012005 (2006); J. P. Lees et al. [BABAR Collabo-
ration], Phys. Rev. D 87, 092005 (2013); J. P. Lees et al. [BABAR Collaboration], Phys. Rev. D 88, 072009
(2013).
[14] G. Bardin et al. [PS170 Collaboration], Nucl. Phys. B 411 (1994) 3.
[15] T. A. Armstrong et al. [E760 Collaboration], Phys. Rev. Lett. 70 1212 (1993).
[16] M. Ambrogiani et al. [E835 Collaboration], Phys. Rev. D 60, 032002 (1999); M. Andreotti et al., Phys. Lett.
B 559, 20 (2003).
[17] R.R. Akhmetshin [CMD-3 Collaboration], arXiv:1507.08013 (2015).
[18] A. G. Frodesen, O. Skjeggestad and H. Tofte, Universitetsforlaget, Bergen, Norway (1979).
[19] B. Aubert et al. (BaBar Collaboration), Phys. Rev. D 76 092006 (2007).
[20] D. Bisello et al. (DM2 Collaboration), Z. Phys. C 48 23 (1990).
[21] M. Davier, A. Hoecker, B. Malaescu, and Z. Zhang, Eur. Phys. J. C71 1515 (2011), [Erratum-ibid. C72, 1874
(2012)].
[22] K. Hagiwara, R. Liao, A.D. Martin, D. Nomura and T. Teubner, J. Phys. G38 085003 (2011).
[23] M. Ablikim et al. [BESIII Collaboration], Phys. Lett. B753 629-638 (2016).
[24] G. Rodrigo, H. Czyz, J. H. Kuhn, M. Szopa, Eur. Phys. J C24, 71 (2002); H. Czyz, J. H. Kuhn and A.
Wapiehik, Phys. Rev. D 77, 114005 (2008).
[25] G. J. Gounaris and J.J. Sakurai, Phys. Rev. Lett. 21, 244 (1968).
[26] K. A. Olive et al. [Particle Data Group], Chin. Phys. C 38, 090001 (2014) and 2015 update.
[27] J. S. Schwinger, Particles, Sources and Fields, Vol. 3, Redwood City, USA Addison-Wesley, 99 (1989).
[28] A. Aloisio et al. [KLOE Collaboration], Phys. Lett. B 606, 12 (2005); F. Ambrosio et al. [KLOE Collabo-
ration], Phys. Lett. B 670, 285 (2009); F. Ambrosino et al. [KLOE Collaboration], Phys. Lett. B 700, 102
(2011); D. Babusci et al. [KLOE Collaboration], Phys. Lett. B 720, 336 (2013).
[29] B. Aubert et al. [BABAR Collaboration], Phys. Rev. Lett. 103, 231801 (2009); J. P. Lees et al. [BABAR
Collaboration], Phys. Rev. D 86, 032013 (2012).

050006-6