Charged Bottomoniumlike States at Belle Alex Bondar Belle

Charged Bottomonium-like States at Belle Alex Bondar Belle Collaboration Novosibirsk, NSU/BINP (ICNFP 2017, August 22, 2017)

Motivation Observation of e+e- → + - hc by CLEO PRL 107 (2011) 041803 Ryan Mitchell @ CHARM 2010 Energy dependence of the cross section Enhancement of (hc + -) @ Y(4260) (hb + -) is enhanced @ Yb? Belle search for hb in (5 S) data 2

3

Energy dependence of [e+e– hadrons] CUSB PRL 54, 377(1985) (1 S) (e+e– hadrons) R= (4 S) (hadrons) ( + –) -1 30 fb -1 (2 S) 24 fb (3 S) (5 S) (4 S) (6 S) 2 M(B) _ + – e e (4 S) BB Belle 711 fb-1 e+ e– 2 M(Bs) Study rare B decays and CP violation _ _ _ ( ) (5 S) BB, BB*, B*B*, BB* , B*B* , Bs * Bs(*), … Belle 121 fb-1 4

hb reconstruction Missing mass to system Mhb(n. P) = ( P (5 S) – P + -)2 MM( + -) hb(1 P) Simple selection : + - : good quality, positively identified (1 S) (2 S) hb(2 P) (3 S) Suppression of continuum events FW R 2<0. 3 Search for hb(n. P) peaks in MM( + -) spectrum 121. 4 fb-1 5

Observation of hb(1 P, 2 P) e+e- (5 S) X + – reconstructed, use Mmiss( + -) (Pe+e- – P + -)2 11. 00 PRL 108, 032001(2012) (10860) + - 10. 75 raw distribution (4 S) hb(2 P) residuals (11020) hb(1 P) 10. 50 b(3 S) 10. 25 10. 00 b(2 S) MHF(1 P) 2 M(B) hb(2 P) hb(1 P) 9. 75 MHF(1 P) = +0. 8 1. 1 Me. V consistent with zero, MHF(2 P) = +0. 5 1. 2 Me. V as expected 9. 50 b(1 S) JPC=0 -+ (1 S) 1 -- 1 - + (0, 1, 2)++ Large hb(1, 2 P) production rates c. f. CLEO e+e- (4170) hc + 6

Observation of hb(1 P, 2 P) e+e- (5 S) hb(n. P) + – reconstructed, use Mmiss( + -) (Pe+e- – P + -)2 11. 00 PRL 108, 032001(2012) (10860) + - 10. 75 raw distribution (4 S) hb(2 P) residuals (11020) hb(1 P) 10. 50 10. 25 % 19 b(2 S) MHF(1 P) = +0. 8 1. 1 Me. Vconsistent with zero, MHF(2 P) = +0. 5 1. 2 Me. Vas expected Large hb(1, 2 P) production rates c. f. CLEO e+e- (4170) hc + - 9. 50 b(1 S) JPC=0 -+ hb(1 P) 41 9. 75 hb(2 P) % 13% 10. 00 b(3 S) 2 M(B) (1 S) 1 -- 1 - + (0, 1, 2)++ hb(n. P) decays are a source of b(m. S) 7

Observation of hb(1 P, 2 P) b(1 S) e+e- (5 S) hb(n. P) + – reconstruct M (1 S) HF b(1 S) Belle : 57. 9 2. 3 Me. V 3 PDG’ 12 : 69. 3 2. 8 Me. V hb(1 P) b(1 S) Ba. Bar (3 S) CLEO (3 S) b(1 S) p. NRQCD LQCD (n) + (4 S) b(3 S) 10. 25 10. 00 9. 75 Mmiss ( + - ) (10860) 10. 75 Ba. Bar (2 S) 10. 50 hb(2 P) (11020) 11. 00 b(2 S) Mizuk et al. Belle PRL 109 (2012) 232002 hb(2 P) hb(1 P) Kniehl et al, PRL 92, 242001(2004) Meinel, PRD 82, 114502(2010) 9. 50 2 M(B) b(1 S) JPC=0 -+ (1 S) 1 -- MHF(1 S) 1 - + (0, 1, 2)++ Belle result decreases tension with theory +4. 0 +4. 5 Me. V First measurement = 10. 8 – 3. 7 – 2. 0 as expected 8

Observation of hb(1 P, 2 P) b(1 S) PRL 101, 071801 (2008) e+e- (5 S) hb(n. P) + – reconstruct M (1 S) HF Ba. Bar (3 S) b(1 S) Belle : 57. 9 2. 3 Me. V 3 ISR (1 S) PDG’ 12 : 69. 3 2. 8 Me. V b hb(1 P) b(1 S) Ba. Bar (3 S) Ba. Bar (2 S) hb(2 P) CLEO (3 S) b(1 S) (n) PRL 103, 161801 (2009) Ba. Bar (2 S) b(1 S) ISR (1 S) b p. NRQCD Mmiss ( + - ) b(1 P) LQCD Kniehl et al, PRL 92, 242001(2004) Meinel, PRD 82, 114502(2010) PRD 81, 031104 (2010) Mizuk et al. Belle PRL 109 (2012) 232002 Belle result decreases tension with theory +4. 0 +4. 5 Me. V First measurement = 10. 8 – 3. 7 – 2. 0 as expected CLEO (3 S) 9

First evidence for b(2 S) e+e- (5 S) hb(2 P) + – b(2 S) Mizuk et al. Belle PRL 109 (2012) 232002 +4. 0 Me. V MHF(2 S) = 24. 3 – 4. 5 First measurement p. NRQCD LQCD Belle 4. 2 w/ syst In agreement with theory (2) Mmiss ( + - ) (2 S) = 4 8 Me. V, < 24 Me. V @ 90% C. L. expect 4 Me. V Branching fractions Expectations +5. 6 BF[hb(1 P) b(1 S) ] = 49. 2 5. 7 – 3. 3 % 41% Godfrey Rosner PRD 66, 014012(2002) BF[hb(2 P) b(1 S) ] = 22. 3 3. 8+3. 1 % 13% – 3. 3 BF[hb(2 P) b(2 S) ] = 47. 5 10. 5+6. 8 % 19% – 7. 7 c. f. BESIII BF[hc(1 P) c(1 S) ] = 54. 3 8. 5 % 39% 10

_ Anomalies in (5 S) (bb) + – transitions (11020) + – (4 S) 2 10. 50 330 (3 S) 10. 00 9. 75 430 b(2 S) (5 S) hb(1, 2 P) + – are not suppressed hb(2 P) 1 190 (2 S) hb(1 P) 290 Belle: PRL 108, 032001 (2012) -fl ip 260 2 M(B) 10. 25 [ (5 S) (1, 2, 3 S) + –] >> [ (4, 3, 2 S) (1 S) + –] (10860) in Mass, Ge. V/c 2 10. 75 Belle: PRL 100, 112001 (2008) 100 sp 11. 00 b(1 S) Heavy Quark Symmetry _ Rescattering of on-shell B(*) ? 6 /hb(2 P) partial (ke. V) 9. 50 expect suppression QCD/mb (1 S) JPC = 0 -+ 1 -- 1 -+ hb production mechanism? Study resonant structure in hb(m. P) + – 11

Resonant substructure of (5 S) hb(1 P) + - P(hb) = P (5 S) – P( + -) M(hb +) = MM( -) measure (5 S) hb yield in bins of MM( ) phase-space MC data PHSP combine Fit function Results M 1 = M 2 = Me. V/c 2 Me. V _ ~BB* threshold a= _ ~B*B* threshold = degree Significances 2 vs. 1 : 7. 4 (6. 6 w/ syst) 2 vs. 0 : 18 (16 w/ syst) non-res. amplitude ~0 12

Resonant substructure of (5 S) hb(2 P) + phase-space MC data PHSP combine hb(1 P) + M 1 = hb(2 P) + - a= = Me. V/c 2 Me. V co 2 = ns Me. V/c 2 M 2 = te Me. V is 1 = Me. V/c 2 nt Me. V/c 2 degree Significances 2 vs. 1 : 2. 7 (1. 9 w/ syst) 2 vs. 0 : 6. 3 (4. 7 w/ syst) 13

Exclusive (5 S) -> (n. S) + (5 S) (n. S) + (n = 1, 2, 3) (n. S) + (3 S) (2 S) (1 S) reflections 14

Results: (2 S)π+πsignals reflections M( (2 S)π+), Ge. V M( (2 S)π-), Ge. V 15

Results: (3 S)π+π- M( (3 S)π+), Ge. V M( (3 S)π-), Ge. V 16

_ Resonant structure of (5 S)→(bb) + – (5 S) hb(1 P) + no non-res. contribution (5 S) hb(2 P) + - Two peaks are observed in all modes! phsp Belle: PRL 108, 232001 (2012) phsp Zb(10610) and Zb(10650) should be multiquark states M[ hb(1 P) π ] (5 S) (1 S) + - M[ hb(2 P) π ] Dalitz plot analysis (5 S) (2 S) + - (5 S) (3 S) + - note different scales 17

Summary of Zb parameters Belle: PRL 108, 232001 (2012) Average over 5 channels M 1 = 10607. 2 2. 0 Me. V 1 = 18. 4 2. 4 Me. V M 2 = 10652. 2 1. 5 Me. V 2 = 11. 5 2. 2 Me. V M 1 – (MB+MB*) = + 2. 6 2. 1 Me. V M 2 – 2 MB* = + 1. 8 1. 7 Me. V Zb(10610) yield ~ Zb(10650) yield in every channel Relative phases: 0 o for and 180 o for hb 18

_ Anomalies in (5 S) (bb) + – transitions (11020) – (4 S) 2 10. 50 + (3 S) 10. 00 9. 75 430 b(2 S) Z+b hb(2 P) 1 (2 S) hb(1 P) 290 6 b(1 S) Belle: PRL 108, 032001 (2012) expect suppression QCD/mb Heavy Quark Symmetry _ Rescattering of on-shell B(*) ? /hb(2 P) partial (ke. V) 9. 50 (5 S) hb(1, 2 P) + – are not suppressed -fl ip 260 2 M(B) 10. 25 [ (5 S) (1, 2, 3 S) + –] >> [ (4, 3, 2 S) (1 S) + –] (10860) sp Mass, Ge. V/c 2 10. 75 Belle: PRL 100, 112001 (2008) 100 in 11. 00 (1 S) JPC = 0 -+ 1 -- 1 -+ hb production mechanism? (5 S) hb(1, 2 P) + – are not suppressed due to Zb intermediate states! But it’s not enough for (n. S) + –. 19

(5 S)→ (2 S)π+π–: JP Results (2 S)π+π Data Toy MC with various JP JP = 1 + JP = 1 - JP = 2 + JP = 2 - 20 20

JP Results 21

Heavy quark structure in Zb A. B. , A. Garmash, A. Milstein, R. Mizuk, M. Voloshin PRD 84 054010 (ar. Xiv: 1105. 4473) Wave func. at large distance – B(*)B* Explains • Why hb is unsuppressed relative to • Relative phase ~0 for and ~1800 for hb • Production rates of Zb(10610) and Zb(10650) are similar • Widths –”– • Dominant decays to B(*)B* Other Possible Explanations • Coupled channel resonances (I. V. Danilkin et al, ar. Xiv: 1106. 1552) • Cusp (D. Bugg Europhys. Lett. 96 (2011), ar. Xiv: 1105. 5492) • Tetraquark (M. Karliner, H. Lipkin, ar. Xiv: 0802. 0649) 22

Zb -> Open Beauty MC: B*Bπ r. M(B ) = (Ee+e-– EB )2 – (pe+e-– p. B )2 MC: B*B*π (shifted by 45 Me. V) 23

Zb -> Open Beauty Reconstruct one B meson, use Zb(10650) Zb(10610) Mmiss( ) = (Ee+e-– E )2 – (p e+e-– p )2 _ BB* _ B*B* r. M(B ) , Ge. V/c 2 24

Observation of _ Zb(10610) BB* _ Zb(10650) B*B* No signal of _ Zb(10650) BB* 25

Belle PRL 108, 122001(2012) PRL 116, 212001(2016) PRL 117, 142001(2016) JP = 1+ 6 D amplitude analysis PRD 91, 072003(2015) MZb(10610) – (MB+MB*) = +2. 6 2. 1 Me. V MZb(10650) – 2 MB* = +1. 8 1. 7 Me. V _ Zb(10610) = B B* _ Zb(10650) = B*B* _ B* B Zb(10610) = 18. 4 2. 4 Me. V Zb(10650) = 11. 5 2. 2 Me. V Decays into constituents dominate (smoking gun of molecular structure? ) JP = 1+ _ B(*)B* in S-wave 26

Heavy quark structure in Zb A. B. , A. Garmash, A. Milstein, R. Mizuk, M. Voloshin PRD 84 054010 (ar. Xiv: 1105. 4473) Wave func. at large distance – B(*)B* Explains • Why hb is unsuppressed relative to • Relative phase ~0 for and ~1800 for hb • Production rates of Zb(10610) and Zb(10650) are similar • Widths –”– Predicts • Existence of other similar states 27

Voloshin PRD D 84 (2011) 031502, ar. Xiv: 1105. 5829 12 Ge. V U(? S) 11. 5 Ge. V U(6 S) U(5 S) r w U b w U hb b r Uw b w Ur b 0 -(1+) 1+(1+) Wb 0 r r w Zb 0+(0+) 1 -(0+) Uw Ur Uw BB* Ur Xb Wb 1 0+(1+) 1 -(1+) B*B* Wb 2 0+(2+) 1 -(2+) 0 -(1 -) BB IG(JP) 28

Charm vs Beauty: I JP = 1+ for both, arxiv: 1706. 04100 29

Charm vs Beauty: II BESIII, arxiv: 1703. 08787 ZC(4020)? ? ZC(4020)? 30

Summary The first exotic bottomonium-like Zb+ states were discovered in decays to (1 S) +, (2 S) +, (3 S) +, hb(1 P) +, hb(2 P) + Spin parity of Zb states is 1+ Zb states mainly decay to BB* and B*B*final states Zb(10610) dominantly decays to BB*, but Zb(10650) to B*B* Decay fraction of Zb(10650) to BB* is currently not statistically significant, but at least less than to B*B* Phase space of Y(5 S)->B(*)B* is tiny, relative motion B(*)B*is small, which is favorable to the formation of the molecular type states Y(5 S) [and possible Y(6 S)] is ideal factory of molecular states In heavy quark limit we can expect more molecular states in vicinity of the BB, BB* and B*B* thresholds. To search the new states we need the energy up to 12 Ge. V Studies of Zb states properties may help us to understand exotic states in charm sector too 31

Back up slides

Super. KEKB Belle II e+ New IR New superconducting /permanent final focusing quads near the IP New beam pipe & bellows Replace short dipoles with longer ones (LER) e. Add / modify RF systems for higher beam current Redesign the lattices of HER & LER to squeeze the emittance Low emittance positrons to inject Positron source Damping ring Low emittance gun Low emittance electrons to inject Ti. N-coated beam pipe with antechambers New positron target / capture section To aim × 40 luminosity 33

Tetraquark? M ~ 10. 2 – 10. 3 Ge. V Ying Cui, Xiao-lin Chen, Wei-Zhen Deng, Shi-Lin Zhu, High Energy Phys. Nucl. Phys. 31: 7 -13, 2007 (hep-ph/0607226) M ~ 10. 5 – 10. 8 Ge. V Tao Guo, Lu Cao, Ming-Zhen Zhou, Hong Chen, (1106. 2284) M ~ 9. 4, 11 Ge. V M. Karliner, H. Lipkin, (0802. 0649) 34

Coupled channel resonance? I. V. Danilkin, V. D. Orlovsky, Yu. Simonov ar. Xiv: 1106. 1552 No interaction between B(*)B* or is needed to form resonance No other resonances predicted B(*)B* interaction switched on individual mass in every channel? 35

Cusp? D. Bugg Europhys. Lett. 96 (2011) (ar. Xiv: 1105. 5492) Amplitude Line-shape Not a resonance 36

37

(5 S) (n. S) 0 0 (1, 2, 3 S) + -, e+e-, (2 S) (1 S) + Y(1 S)[l+l-] + - 0 0 e+e- 0 0 + - 0 0 (2 S) (1 S) (2 S) (3 S) reflection (1 S) [e+e- (5 S) (1 S) 0 0] = (1. 16 0. 06 0. 10) pb [e+e- (5 S) (2 S) 0 0] = (1. 87 0. 11 0. 23) pb [e+e- (5 S) (3 S) 0 0] = (0. 98 0. 24 0. 19) pb Consistent with ½ of Y(n. S) + ar. Xiv: 1308. 2646, accepted for publication in Phys. Rev. D 38

39

Selection Decay chain + (5 S) Zb hb (n. P) + reconstruct b(m. S) R 2<0. 3 Hadronic event selection; continuum suppression using event shape; 0 veto. Require intermediate Zb : 10. 59 < MM( ) < 10. 67 Ge. V bg. suppression 5. 2 Zb 40

Mmiss( + -) spectrum requirement of intermediate Zb Update of M [hb(1 P)] : MHF [hb(1 P)] = (+0. 8 1. 1)Me. V/c 2 (2 S) 3 S 1 S hb(1 P) Previous Belle meas. : ar. Xiv: 1103. 3411 MHF [hb(1 P)] = (+1. 6 1. 5)Me. V/c 2 41

Results of fits to Mmiss( + -) spectra hb(1 P) yield b(1 S) Peaking background? MC simulation none. (2 S) yield no significant structures Reflection yield 42

Calibration Use decays B+ c 1 K+ (J/ ) K+ Photon energy spectrum MC simulation hb(1 P) b(1 S) c 1 J/ cos Hel > – 0. 2 cos Hel ( c 1)> – 0. 2 match energy of signal & callibration channels 43

Calibration (2) Resolution: double-sided Crystal. Ball function with asymmetric core data E s. b. subtracted M(J/ ), Ge. V/c 2 Correction of MC mass shift fudge-factor for resolution – 0. 7 0. 3 +0. 2 – 0. 4 Me. V 1. 15 0. 06 44

Integrated Luminosity at B-factories (fb-1) asymmetric e+e- collisions > 1 ab-1 On resonance: (5 S): 121 fb-1 Bs (4 S): 711 fb-1 (3 S): 3 fb-1 (2 S): 24 fb-1 (1 S): 6 fb-1 Off reson. /scan : ~100 fb-1 530 fb-1 On resonance: (4 S): 433 fb-1 (3 S): 30 fb-1 (2 S): 14 fb-1 Off reson. /scan : ~54 fb-1 45

Description of fit to MM( + -) Three fit regions Example of fit Residuals 2 3 BG: Chebyshev polynomial, 6 th or 7 th order Signal: shape is fixed from + - data “Residuals” – subtract polynomial from data points KS contribution: subtract bin-by-bin M( + -) (1 S) 1 kinematic boundary Ks generic (3 S) MM( + -) 46

Belle Detector 47

Heavy quarkonium – approximately a non-relativistic system. System Ground triplet state Name Mass, Me. V (v/c)2 POSITRONIUM e+ e- Ortho- 1 5 10 -15 ~0. 0001 QUARKONIUM uu, dd r 800 150 ~1. 0 ss f 1000 4 ~0. 8 cc 3100 0. 09 ~0. 25 bb 9500 0. 05 ~0. 08 Good approximation for heavy quarkonium – potential models. 48

Quarkonium Basics 49

_ _ Classification of cc and bb levels is the same as in positronium: L, S, n r : confinement chromoelectric tube asymptotic freedom one-gluon exchange string breaking Non-relativistic effects for low excitations are small. 50

Charmonium table Potential models States observed in 1974 -1980 2(1 D) S=0 S=1 S=0 _ DD States observed after 2002 S=1 JPC L=0 L=1 L=2 _ States below DD or other charmonia) _ threshold are narrow (annihilation _ _ States above DD threshold are broad ( DD, DD*, . . . ) 51

Charmonium table Y(4660) X(4700) X(4500) X(4160) X(3915) X(3940) X(4274) X(4140) Y(4360) Y(4260) 2(1 D) X(3872) S=0 S=1 Z(4430)+ Z(4250)+ Z(4200)+ Z(4050)+ Zc(4020)+ _Zc(3900)+ DD S=0 L=1 S=1 L=2 States observed in 1974 -1980 States observed after 2002 States with unexpected properties JPC L=0 Potential models Z+ _ _ = |ccud manifestly exotic _ States below DD or other charmonia) _ threshold are narrow (annihilation _ _ States above DD threshold are broad ( DD, DD*, . . . ) 52

Puzzles of (5 S) decays Anomalous production of (n. S) + - with 21. 7 fb-1 (Me. V) PRL 100, 112001(2008) PRD 82, 091106 R(2010) 102 Dedicated energy scan shapes of Rb and ( ) different Rb (5 S) is very interesting and not yet understood Finally Belle recorded 121. 4 fb-1 data set at (5 S) 53