Phi Radiative decays at KLOE Camilla Di Donato

Phi Radiative decays at KLOE Camilla Di Donato* for the KLOE Collaboration *Sezione I. N. F. N. Napoli Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006

Outline Scalar mesons at f factory: § f f 0(980) [PLB 634 (2006) 148] § f f 0(980) Pseudoscalar mesons at f factory: § mass measurement § BR(f ) and mixing angle Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 2

Scalar mesons at a f factory How a f factory can contribute to the understanding of the scalar mesons Mass (Me. V/c 2) Scalar Mesons Spectroscopy: f 0(980), s(600) and a 0(980) are accessible (k not accessible) through S ; Questions: 1. Is s(600) needed to describe the mass spectra ? 2. “couplings”of f 0(980) and a 0(980) to f |ss> and to KK, and . 4 -quark vs. 2 -quark states Camilla Di Donato f(1020) 1000 a 0(980) f 0(980) K*0(800) “k” 500 Novosibirsk, Feb 27 Mar 2, 2006 f 0(600) “s” 0 I=1/2 I=1 3

The scalar amplitude description We use three different approaches in the description of the scalar amplitude: 1. 2. 3. Kaon-loop KL (Achasov-Ivanchenko, NPB 315 1989): for each scalar meson S there are three free parameters of the fit: g. S , g. SKK, MS No-Structure NS (by G. Isidori and L. Maiani): a modified BW + a polynomial continuum: g S , g. SKK, MS + pol. cont. parameters e+ Scattering Amplitudes SA e(M. E. Boglione, M. R. Pennington, Eur. Phys. , J. C 30 (2003)) A (a 1+b 1 m 2+c 1 m 4) T( ) + (a 2+b 2 m 2+c 2 m 4) T( KK) pole residual g Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 K+ p S p K- g KK g. SPP p V S g. VS g. Spp p 4

@ • large angle: looking for f e+e → events with the photon at large angle (45 < <135 ) Events/1. 2 Me. V • Main contributions: – ISR (radiative return to , ) – FSR M( ) (Me. V) • Search for the f 0 signal as a deviation of M ( ) spectrum from the expected ISR + FSR shape • 676, 000 events selected (2001+2002) f 0(980) region Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 M( ) 5(Me. V)

Fit to the m( ) spectrum: (491 bins, 1. 2 Me. V wide, m( ) = 420 to 1009 Me. V) KL fit NS fit Fit : ISR + FSR + r + scalar ± interf(SCAL+ FSR) Kaon Loop and No Structure fits: Good description in both cases of signal and background (Kuhn Santamaria); “negative” interference; The introduction of a s(600) doesn’t improve the fit. Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 6

Scattering Amplitude Fit gf = 6. 6 × 10 4 BR ( f f 0 ) × BR ( f 0 ) ~ 3 × 10 5 [similar conclusion from Boglione Pennington analysis of data (KLOE + SND)] Summarizing: The peak at ~980 Me. V is well interpreted in both KL and NS approaches as due to the decay f 0(980) with a negative interference with FSR. The couplings suggest the f 0(980) to be strongly coupled to kaons and to the . Scattering gives a. Feb marginal agreement. Camilla Di Donato Amplitude. Novosibirsk, 27 Mar 2, 2006 7

Forward Backward asymmetry A = (N(q+>90 o) – N(q+<90 o)) / (N(q+>90 o) + N(q+<90 o)) + system: A(ISR) C odd A(FSR) & A(scalar) C even Cross section: |A(tot)|2 = |A(ISR) + A(FSR) + A(scalar)|2 zoom FB asymmetry vs. m( ): Clear signal ~ 980 Me. V Interesting comparison with Phokara MC simulation Data Simulation FSR+ISR+ scalar(KL) Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 8

final state: VMD vs scalar 450 pb-1 from 2001 – 2002 data taking Two main contributions to final state @ Mf: →S → 0 0 ~ 400 k events ary elimin r p E O KL √s = 1019. 6 Me. V e+e-→w 0→ 0 0 New analysis scheme: Allow for interference between e+ e- wp 0 and f Sg Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 Bi-dimensional analysis of Dalitz-plot : m(p 0 p 0) vs m(p 0 g) 9

final state: VMD vs scalar • Fit with No Structure approach: in progress • Fit Dalitz plot with improved Kaon loop parametrization: (Achasov Kiselev hep ph/0512047) Insertion of a KK scattering phase New parametrization of scattering phase Scalar contributions: f 0(980) and s(600) Combined fit to scattering data and to already published KLOE data on f S gives six sets of parameters able to describe both distributions Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 10

KLOE preliminary K Loop fit: f 0+s (dalitz slices) @ 1019. 6 Me. V § Free parameters: Mf 0, g_f 0 + , g_f 0 K+K , VMD description § /KK phases and s(600) fixed to Achasov’s results Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 11

KL fit results y r a n i m i l OE pre KL • parameter values are from best fit • The first error is statistical error from fit, the second one reflects the changes related to the other fit variants with acceptable c 2 Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 12

Pseudoscalar mesons at f factory ● ● ● mass measurement; Br(f )/Br(f ) with + 7 final state; Pseudoscalar mixing angle. Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 13

mass measurement Large discrepancy between the two most precise measurements GEM measurement M = (547. 311 ± 0. 028 ± 0. 032) Me. V/c 2 Reaction used: p + d → 3 He + [M. Abdel Bary et al. , Phys. Lett. B 619 (2005) 281 287] NA 48 measurement Using → 3 from p → n : M = (547. 843 ± 0. 030 ± 0. 041) Me. V/c 2 [A, Lai et al. , Phys. Lett. B 533 (2002) 196] Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 14

Measurement method Using the decays: f f E 3 cross checking purpose A kinematic fit is performed imposing: Px, Py, Pz, Etot t-r/c of clusters E 1 conservation compatible with light velocity E 2 As a consequence of the kinematic fit the mass measurement is almost independent from the energy of the clusters, it is dominated by the cluster positions. The f momentum and the vertex position are precisely determined run by run from the study of the Bhabha scattering at large angle e+e – (90000 events for each run). Absolute scale determined buy f line shape and f mass Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 15

M measurement @ KLOE • Kinematic fit applied on f → events § and 0 selected by looking at different Dalitz plot regions c 2/ndf = 304/257 0 0 m = ( 134. 956 0. 018 ) Me. V M (Me. V) s = ( 1. 66 0. 005 ) Me. V E 1<E 2<E 3 Camilla Di Donato c 2/ndf = 146/161 m = ( 547. 708 0. 014 ) Me. V M (Me. V) Novosibirsk, Feb 27 Mar 2, 2006 16 s = ( 2. 143 0. 012 ) Me. V M (Me. V)

M (Me. V) 2001 2002 data sample divided into 8 periods, each of 50 pb 1 M (Me. V) M measurement @ KLOE 547. 95 y r a n i m i l OE pre KL 547. 90 547. 85 547. 80 M( 0) = ( 134990 6 stat 30 syst ) ke. V M( 0)PDG = ( 134976. 6 0. 6 ) ke. V 547. 75 547. 70 M( ) = ( 547822 5 stat 69 syst ) ke. V § Systematics mainly from √s and vertex position EMC linearity in progress § NA 48 compatibility: 0. 24 s Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 17

BR( )/ BR( ) ' , ' + - , 3 0 ' 0 0 , + - 0 § , 3 0 § Signal selection: charged vertex in a cylinder with a 4 cm radius and a 16 cm height around the interaction point 7 prompt photons |t – r/c| < 5 st and 21 < q < 159 anti KSKL tag No way to separate radiative and from Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 E (Me. V) 18

Signal extraction: data 2001 2002 427 pb 1 @ √s = Mf ' , ' + -6 wrong combination Nobs = 3750 observed Nbkg = 345 estim. bkg Nsig=(Nobs - Nbkg)=3405 ± 65 stat ± 28 syst ' mass distribution after wrong mass combination subtraction Ks + Kl 0 0 0 ary n i m i l e r LOE p Ks 0 0 Kl + - 0 K Ks + - Kl 0 0 0 Signal Camilla Di Donato Novosibirsk, 27 Mar 2, 2006 M Feb(Me. V) 19 M (Me. V)

Results ry a n i m i l e OE pr KL Source Systematics dominated by knowledge of , ’ branching ratios Previous KLOE results: Phys. Lett. B 541 (2002) 45 51 Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 Syst. Err. Filfo Evcl 1% TRK 1% VTX 1% Back. Sub. 0. 1% e /e ´ 0. 4% c 2 1. 5% BR” 3% K 1% total 4% 20

Mixing angle and ’ gluonic content Using the Bramon approach (A. Bramon et. al. Eur. Phys. J. C 7, 271(1999)) we extract mixing angle: ry a n i m i l e OE pr KL Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 21

Conclusions Extensive study of f 0(980) properties with 2001 2002 KLOE data: • Clear evidence of f 0 + - both in invariant mass and in the forward backward asymmetry. M fitted with three different theoretical approaches: [PLB 634 (2006) 148] • Preliminary results on S 0 0 : fit to Dalitz plot with improved kaon loop model Perspectives with our new data sets: • 2 fb 1 @ M : search for f 0 / a 0 KK • Energy scan (4 points 10 pb 1 each in the 1010 1030 Me. V range): improved study of the √s dependence of the cross section • Off peak data (~200 pb 1 1000 Me. V): search for 0 0 Measurement of mass: KLOE preliminary result is in agreement with NA 48 measurement in + - + 7 final state, in agreement with the KLOE previous Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 result in + - + 3 final state; Pseudoscalar mixing angle 22

KL fit results: f 0+s (the six variants), S term Fixed M, G s Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 23

KL fit results: VDM/S compositions Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 24

Fit function: the Achasov parametrization Camilla Di Donato Novosibirsk, Feb 27 Mar 2, 2006 25