Bled Workshops in Physics Vol. 14, No. 1 p. 45
A
Proceedings of the Mini-Workshop Looking into Hadrons Bled, Slovenia, July 7 - 14, 2013

New results on quarkonium spectroscopy and exotic quarkonium-like resonances at B-factories*
Marko Petric**
Jozef Stefan Institute, Ljubljana, Slovenia
Abstract. These proceedings covers recent results from spectroscopy measurements with data collected at the Belle experiment, which has been operating at the KEKB asymmetric e+ e- collider at KEK in Tsukuba, Japan. The data sample can be used for various novel searches in spectroscopy. The paper discusses the discovery of exotic Zb states, the measurement of radiative hb (1,2P) —> r|b (1,2S) transitions including the first evidence for the hb (2S) state, and the discovery of the charmed state Z+ (3900).
1	Introduction
The Belle experiment [1] was taking data between 1999 and 2010 at the asymmetric e+e- collider KEKB [2] in Tsukuba, Japan. During this time more than 1 ab-1 of data was collected, mostly at the Y(4S) resonance, but also on resonances Y(1S), Y(2S) and Y(5S), as well as in the nearby continuum. The Belle detector was a large-solid-angle magnetic spectrometers that consisted of Drift Chambers, a silicon vertex detectors [3], an electromagnetic calorimeter and a superconducting solenoid that provided a 1.5 T magnetic field. The amount of collected experimental data and superb detector performance enables many interesting analyses, including searches for new hadronic states and studies of their properties. In this paper we will cover interesting spectroscopic measurements performed of charmonium(-like) and bottomonium(-like) states.
2	Discovery of charged and neutral Zb states
In the past years a myriad of new exotic states has been discovered at different experiments and different energies. A broad selection of interpretations for these states is being proposed, e.g. molecules of mesons, tetraquarks or hadrocharmo-nia [4]. The observed high rate of Y(5S) —» hb (mP)n+n- (m = 1,2) decays, which is expected to be suppressed compared to Y(5S) —» Y(nS)n+n- (n = 1,2,3) decays, since it requires a spin flip of one bottom quark [5] is a clear sign of an exotic decay mechanism in Y(5S) decays. Consequent studies of Y(5S) decays have shown the existence of two new charged states, Z+(10610) = Z+1
* Talk delivered by Marko Petric
** on behalf of the Belle Collaboration
and Z+(10650) = Z+2. These studies are based on the known initial energy at B-factories to observe unreconstructed particles in the recoil mass distribution of the reconstructed particles:
Mmiss = V ((Ebeam Erecon) Precon )
where Ebeam is half of the centre of mass energy, and all quantities are boosted to the centre of mass system of the colliding beams. This technique is not applicable in measurements at hadron colliders like the LHC. A direct observation of the Z+
Mb» - Mb = 45 MeV
Fig. 1. Discovery of the Zb states; (Left) M(Y(2S)rt)max spectrum in reconstructed Y(5S) —> Y(2S)n+ n- decays [6], (Middle) Mmis spectrum of reconstructed Y(5S) —> BnX decays [7], (Right) M(Y(2S)n)max distribution in reconstructed Y(5S) -> Y(2S)n0n0 decays [8].
signals is possible in the M(Y(nS)n)max distributions of exclusively reconstructed Y(5S) —> Y(nS)[^+^-]n+n- decays (Figure 1). The subscript 'max' denotes the choice of the Yn combination with the higher invariant mass. A Dalitz analysis with non-resonant Z+1 , Z+2 , f0(980) and f2 (1270) components in the variables M2(Y(nS)n+ )max x M2(n+ n-) favours the quantum numbers IG(JP) = 1 + (1+). This hypothesis is supported by a more complex six dimensional Dalitz analysis. During the analysis of the decays Y(5S) —> Zb1>2[hb(mP)nTthe recoil mass technique is applied twice: firstly the hb (mP) yield is determined from fits to the Mmis spectrum, which are performed in bins of Mmis, where the Z+ mass peaks can be observed. The measured masses and the widths of the Z+1 and Z+2 are in agreement for all channels. The averages are [6]:
Z+1 : M =(10607.2 ± 2.0)MeV/c2 r =(18.4 ± 2.5)MeV ,
Z+2 : M =(10652.2 ± 1.5)MeV/c2 r =(11.5 ± 2.2)MeV .
The observed new states have a mass close to the mass of B*B and B*B* pairs suggesting a molecular structure. In the analyses of decays Y(5S) —» Bn±X transitions of Zb1,2 to B(*'B are observed, where the B is a fully reconstructed B+ or B0 meson. The second B is measured in the recoil mass M^n distribution (Figure 1). The determined branching fractions are B(Y(5S) —» BBn) < 0.60% at 90% C. L., B(Y(5S) -> BB*n) = (4.25 ± 0.44 ± 0.69)% and B(Y(5S) -> B*B*n) = (2.12 ± 0.29 ± 0.36)%. To address the question if these decays proceed via the intermediate Z+1 2 states an amplitude analysis of the Mmis distributions in the B
and B* signal regions of Mmis was made. It is found that the BB*n sample can be described by the sum of a Z+1 and a Z+2 component or Z+1 and a non-resonant component. Whereas the B*B*n sample is described well by a Z+2 alone or a Z+2 with a non-resonant admixture. A significant Zb1,2 signal is found in all cases. Assuming that the only decay modes are to Y(nS)n+, hb(mP)n+ and B*B the Z+1 and Z+2 states decay predominantly to BB* and B*B* pairs with branching fractions of (86.0 ± 3.6)% and (73.4 ± 7.0)%, respectively [7].
The discovery of Z states prompted the search for their neutral counterparts in fully reconstructed decays Y(5S) —» Y(nS)[£+£-]n0n0, where Y(2S) was additionally reconstructed in the Y(1S)[£+£-]n+n- channel. From fits to the Mmis distributions the Y(nS) yields are extracted and the resulting branching fractions are B(Y(5S) -> Y(1S)n0n0) = (2.25±0.11 ±0.20) x 103, B(Y(5S) -> Y(2S)n0n0) = (3.79±0.24±0.49) x 103 and B(Y(5S) ^ Y(3S)n0n0) = (2.09±0.24±0.34) x 103. As in the charged channel a Dalitz analysis is performed. No significant Z^1 2 signal is found in the Y(1S)n0n0 sample, nor can it be excluded. However, a Z^1 signal is observed with in the Y(2S)n0n0 sample (4.9ct) and the Y(3S)n0n0 sample (4.3ct); with the mass being (10609 ± 8 ± 6)MeV/c2 which is consistent with the Z+1 mass. The Z^2 signal was not significant (2.9ct) [8].
3 Observation of hb(mP) —> nb(mS) transitions and first evidence for nb(2S)
Belle has recently discovered the hb (mP) states [5] which are expected to decay significantly via hb (mPH nb (m'P)y. This prompted the collaboration to search for these radiative transitions [9] and the measurement of the nb(m'P) masses and widths in the production chain Y(5S) —» Z+1 2n- —> hb(mP)n+n-. In this analysis only two pions and the photon from the hb (mP) decay are reconstructed. Events with a Z+1 2 are selected in the Z+1 2 mass window of the Mmis distribution and the hb (mP) yield is determined from fits to the M^is distribution (Figure 2). These fits are performed in bins of Mmrs(m) = Mm*T - Mms - M (hb (mP)). With the help of this transformation correlation of Mm£ and M^nY in the signal region are minimised. Figure 2 shows the n(2S) signal peak in the M^T (2) distribution, which is the first evidence for this state. The measured branching fractions are B(hb(1P) -> nb(IS)y) = (49.2 ± 5.7+|-f)%, B(hb(2P) ^ nb (1S)y) = (22.3 ± 3.8+3-3)%, B(hb(2P^ nb(2S)y) = (47.5 ± 10.5-f-|)%. The extracted masses and widths of the nb (m'S) states are:
nb(1S) : M =(9402.4 ± 1.5 ± 1.8)MeV/c2 r =(10.8-4 - 7-2 - 5)MeV ,
nb(2S) : M =(9999.0 ± 3.5-2 -9)MeV/c2 r<24MeV .
These results are needed for the calculation of the hyperfine splitting AMHF (mS) = MT(ms) — Mnb(mS), through which the spin dependence of bound state energy levels can be probed, and at the same time puts a constraint on theoretical descriptions of spin-spin interactions. The results are in agreement with lattice calculations [4,10].
lu
Fig. 2. Study of Y(5S) —> n+n X; (Left) Distribution of the recoil mass of the two pions, after subtraction of the combinatorial background - the peaks from left to right correspond to Y(1D), hb(2P), Y(2S) -> Y(1S) and Y(2S), (Right) Yield of hb(2P) mass fits to the Mmis distribution in bins of Mm™! - the significance of the r|b (2S) peak is 4.2a [9].
4 Observation of a Charged Charmonium-like State Z+ (3900)
The Belle collaboration measured the cross section for e+e- —»	be-
tween 3.8 GeV and 5.5 GeV on a data sample of 967 fb-1 [11]. In this analysis the Y(4260) state is observed, and its resonance parameters are determined. Additionly, an excess of	production around 4 GeV is observed, which
Fig.3. Unbinned maximum likelihood fit to the distribution of the Mmax(nJ/^). Points with error bars are data, the curves are the best fit, the dashed histogram is the phase space distribution and the shaded histogram is the non-rt+rt~J/^ background estimated from the normalized J/^ sidebands [11].
was parametrized with a Breit-Wigner distribution and the results are consistent with the state Y(4008) which was previously reported by Belle. In the subsequent Dalitz analysis of Y(4260) —»	decays, the collaboration observes
70
+ data
Mmax(nJ/¥) (GeV/c2)
a structure is in the M(n±J/^) mass spectrum with 5.2ct significance, with mass M = (3894.5 ± 6.6 ± 4.5) MeV/c2 and width r = (63 ± 24 ± 26) MeV/c2 [11] (Figure 3). This structure can be interpreted as a new charged charmonium-like state. This state is close to the DD* mass threshold; however, no enhancement is observed near the D*D* mass threshold. Since this Z state has a strong coupling to charmonium and is charged, it can be concluded that it cannot be a conventional cc state.
References
1.	A. Abashian et al., Nucl. Instrum. Methods A 479,117 (2002).
2.	S. Kurokawa and E. Kikutani, Nucl. Instrum. Methods A 499 (2003).
3.	Z. Natkaniec et al., Nucl. Instrum. Methods A 560,1 (2006).
4.	N. Brambilla et al., Eur.Phys. J. C 71,1534 (2011).
5.	I. Adachi et al. (Belle Collaboration), Phys. Rev. Lett. 108, 032001 (2012).
6.	A. Bondar et al. (Belle Collaboration), Phys. Rev. Lett. 108,122001 (2012).
7.	I. Adachi et al. (Belle Collaboration), arxiv:120 9.6450[hep-ph].
8.	I. Adachi et al. (Belle Collaboration), arxiv:120 7.4345[hep-ph].
9.	R. Mizuk et al. (Belle Collaboration), Phys. Rev. Lett. 109, 232002 (2012), arxiv:1205.6351[hep-ph] .
10.	S. Meinel, Phys. Rev. D 82,114502 (2010).
11.	Z. Liu et al. (Belle Collaboration), Phys. Rev. Lett. 110, 252002 (2013).