QCD studies in e+e(cid:0) annihilation from 30 GeV to 189 GeV M. Acciarri, P. Achard, O. Adriani, M. Aguilar-Benitez, J. Alcaraz, G. Alemanni, J. Allaby, A. Aloisio, G.M. Alviggi, G. Ambrosi, et al. To cite this version: M. Acciarri, P. Achard, O. Adriani, M. Aguilar-Benitez, J. Alcaraz, et al.. QCD studies in e+e(cid:0) annihilation from 30 GeV to 189 GeV. Physics Letters B, Elsevier, 2000, 489, pp.65-80. <in2p3- 00005412> HAL Id: in2p3-00005412 http://hal.in2p3.fr/in2p3-00005412 Submitted on 2 Oct 2000 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. EUROPEAN ORGANIZATION FOR NUCLEAR RESEARCH CERN-EP/2000-064 May 04, 2000 QCD Studies in e+e− Annihilation from 30 GeV to 189 GeV The L3 Collaboration Abstract We present results obtained from a study of the structure of hadronic events recorded by the L3 detector at various centre-of-mass energies. The distributions of event shape variables and the energy dependence of their mean values are measured from 30 GeV to 189 GeV and compared with various QCD models. The energy dependence of the moments of event shape variables is used to test a power law ansatz for the non-perturbative component. We obtain a universal value of the non- perturbative parameter (cid:11) = 0.537 (cid:6) 0.073. From a comparison with resummed 0 O((cid:11)2) QCD calculations, we determine the strong coupling constant at each of the s selected energies. The measurements demonstrate the running of (cid:11) as expected in s QCD with a value of (cid:11) (m ) = 0.1215 (cid:6) 0.0012 (exp) (cid:6) 0.0061 (th). s Z Submitted to Phys. Lett. B 1 Introduction LEP operated at centre-of-mass energies around 91.2 GeV from 1989 to 1995 and then moved up to six di(cid:11)erent centre-of-mass energies between 130 GeV and 189 GeV in the following three years. Thus a study of the process e+e− ! hadrons at LEP o(cid:11)ers a unique environment to test the predictions of the theory of the strong interaction (QCD) over a wide energy range. The energy range has been extended by using hadronic events from Z decays with isolated high energy photons in order to probe the structure of hadronic events at reduced centre-of-mass energies down to 30 GeV [1,2]. The high energy photons are radiated early in the process through initial state radiation (ISR) or through quark bremsstrahlung whereas the hadronic shower develops over a longer time scale. Wereporthere measurements ofevent shapedistributions andtheir moments using the data p collected with the L3 detector [3]. We update the published results at s = 161, 172 and 183 p GeV [4,5] with an improved selection method for hadronic events and present new results at s = 130, 136 and 189 GeV. The measured distributions are compared with predictions from event generators based on an improved leading log approximation (Parton Shower models including QCD coherence e(cid:11)ects). Three such Monte Carlo programs (Jetset PS [6], Herwig [7] and Ariadne [8]) have been used for these comparisons. We also compare our measurements with predictions from QCD models with no coherence e(cid:11)ects (Cojets [9]). These Monte Carlo programs use di(cid:11)erent approaches to describe both the perturbative parton shower evolution and non-perturbative hadronisation processes. They have been tuned to reproduce the global event shape distributions and the charged particle multiplicity distribution measured at 91.2 GeV [10]. The moments of event shape variables are measured between 30 GeV and 189 GeV. The perturbative and non-perturbative QCD contributions are obtained from a (cid:12)t using the power correction formula [11]. This approach was (cid:12)rst applied by the DELPHI collaboration [12]. The strong coupling constant (cid:11) is also determined at each of these centre-of-mass energies s by comparing the measured event shape distributions with predictions of second order QCD calculations [13] containing resummed leading and next-to-leading order terms [14]. Section 2 describes the selection of hadronic events. Measurements of event shape variables and estimation of systematic errors are described in section 3. Section 4 presents a comparison of the data with predictions from various QCD models, a study of the power correction ansatz and a determination of (cid:11) from event shape distributions. The results are summarised in s section 5. 2 Event Selection The selection of e+e− ! hadrons events is based on the energy measured in the electromagnetic and hadron calorimeters. We use energy clusters in the calorimeters with a minimum energy of 100 MeV. We measure the total visible energy (E ) and the energy imbalances parallel (E ) vis k andperpendicular (E ) tothebeamdirection. Backgrounds aredi(cid:11)erent forhadronicZdecays, ? hadronic events at reduced centre-of-mass energies and at high energies. This is reflected in the di(cid:11)erent selection cuts used for these three types of data sets. We use Monte Carlo events to estimate the e(cid:14)ciency of the selection criteria and purity of the data sample. Monte Carlo events for the process e+e− ! qq(cid:22)(γ) have been generated by the parton shower programs Jetset and Pythia [15] and passed through the L3 detec- tor simulation [16]. The background events are simulated with appropriate event generators: 2 Pythia and Phojet [17] for two-photon events, Koralz [18] for the (cid:28)+(cid:28)−(γ) (cid:12)nal state, Bhagene [19] and Bhwide [20] for Bhabha events, Koralw [21] for W-pair production and Pythia for Z-pair production. p Details of event selection at s (cid:25) m and at reduced centre-of-mass energies have been p Z described earlier [1,2]. At s (cid:25) m , we have used only a small subset of the complete data Z sample(8.3pb−1 outof142.4pb−1 ofintegratedluminosity)whichstillprovidesanexperimental error three times smaller than theoretical uncertainties. p Dataat s = 130and136GeV were collected intwo separate runs during 1995[4]and1997. The main background at these energies comes from ISR resulting in a mass of the hadronic system close to m . This background is reduced by applying a cut in the two dimensional plane Z p of j E j =E and E = s. In the current analysis, data sets from the two years have been k vis vis combined and the cuts are optimised to get the best e(cid:14)ciency times purity. p For the data at s (cid:21) 161 GeV, additional backgrounds arise from W-pair and Z-pair production. A substantial fraction ((cid:24) 80%) of these events can be removed by a speci(cid:12)c selection [5] based on: (cid:15) forcing the event to a 4-jet topology using the Durham algorithm [22], (cid:15) performing a kinematic (cid:12)t imposing the constraints of energy-momentum conservation, (cid:15) making cuts on energies of the most and the least energetic jets and on yD, where yD is 34 34 the jet resolution parameter for which the event is classi(cid:12)ed as a three-jet rather than a four-jet event. These cuts have also been optimised at each energy point. For centre-of-mass energies at or p above 130 GeV, hadronic events with ISR photon energy larger than 0.18 s are considered as background. The integrated luminosity, selection e(cid:14)ciency, purity and number of selected events for each of the energy points are summarised in Table 1. 3 Measurement of Event Shape Variables We measure (cid:12)ve global event shape variables for which improved analytical QCD calculations [14] are available. These are thrust (T), scaled heavy jet mass ((cid:26)), total (B ) and wide (B ) T W jet broadening variables and the C-parameter. For Monte Carlo events, the global event shape variables are calculated before (particle level) and after (detector level) detector simulation. The calculation before detector simulation takes into account all stable charged and neutral particles. The measured distributions at detector level di(cid:11)erfromtheones atparticlelevel because ofdetector e(cid:11)ects, limited acceptance and resolution. After subtracting the background obtained from simulations, the measured p distributions for all energies except s (cid:25) m are corrected for detector e(cid:11)ects, acceptance Z and resolution on a bin-by-bin basis by comparing the detector level results with the particle level results. The level of migration is kept at a negligible level with a bin size larger than the p experimental resolution. At s (cid:25) m , the detector e(cid:11)ects are unfolded for these event shape Z variables using a regularised unfolding method [23]. We also correct the data for initial and (cid:12)nal state photon radiation bin-by-bin using Monte Carlo distributions at particle level with and without radiation. 3 The systematic uncertainties in the distributions of event shape variables arise mainly due to uncertainties in the estimation of detector correction and background estimation. The un- certainty in the detector correction has been estimated by several independent checks: (cid:15) Thede(cid:12)nitionofreconstructedobjectsusedtocalculatetheobservableshasbeenchanged. Instead of using only calorimetric clusters, the analysis has been repeated with objects obtained from a non-linear combination of energies of charged tracks and calorimetric p clusters. At s (cid:25) m , we use a track based selection and the event shape variables are Z constructed from the tracks. (cid:15) The e(cid:11)ect of di(cid:11)erent particle densities in correcting the measured distribution has been estimated by changing the signal Monte Carlo program (Herwig instead of Jetset). (cid:15) The acceptance has been reduced by restricting the events to the central part of the detector (jcos((cid:18) )j < 0:7, where (cid:18) is the polar angle of the thrust axis relative to the T T beam direction) where the energy resolution is better. The uncertainty on the background composition of the selected event sample has been esti- p mated di(cid:11)erently for the three types of data sets. At s (cid:25) m , the background contamination Z is negligible and the uncertainty due to that has been neglected. For data samples at reduced centre-of-mass energies, the systematic errors arising from background subtraction have been estimated [2] by: (cid:15) varying, by one standard deviation, the background scale factor which takes into account the lack of isolated (cid:25)0 and (cid:17) production in the Monte Carlo sample, (cid:15) varying the cuts on neural network probability, jet and local isolation angles, and energy in the local isolation cone. At high energies, the uncertainty is determined by repeating the analysis with: (cid:15) an alternative criterion to reject the hard initial state photon events based on a cut on the kinematically reconstructed e(cid:11)ective centre-of-mass energy, (cid:15) avariationoftheestimatedtwo-photoninteractionbackgroundby(cid:6)30%andbychanging the background Monte Carlo program (Phojet instead of Pythia), and (cid:15) a variation of the W+W− background estimate by changing the W-pair rejection criteria. The systematic uncertainties obtained from di(cid:11)erent sources are combined in quadrature. At high energies, uncertainties due to ISR and W+W− backgrounds are the most important ones. They are roughlyequal andare2-3 times larger thanthe uncertainties due to thedetector correction. p Apart from the data set at s (cid:25) m , statistical fluctuations are not negligible in the Z estimation of systematic e(cid:11)ects. The statistical component of the systematic uncertainty is determined by splitting the overall Monte Carlo sample into luminosity weighted sub-samples and treating each of these sub-samples as data. The spread in the mean position gives an estimate of the statistical component and is taken out from the original estimate in quadrature. 4 4 Results 4.1 Comparison with QCD models Figure 1 shows the corrected distributions for thrust, scaled heavy jet mass, total and wide p jet broadening and the C-parameter obtained at s = 189 GeV. The data are compared with predictions from QCD models Jetset PS, Herwig and Ariadne at particle level. The agreement is satisfactory. Animportant test ofQCD models is acomparison of theenergy evolution ofthe event shape variables. The energy dependence of the mean event shape variables arises mainly from two sources: the logarithmic energy scale dependence of (cid:11) and the power law behaviour of non- s perturbative e(cid:11)ects. The (cid:12)rst moments of the (cid:12)ve event shape variables are shown in Figure 2 and Table2. Also shown aretheenergy dependences of these quantities aspredicted by Jetset PS, Herwig, Ariadne, Cojets and Jetset ME (O((cid:11)2) matrix element implementation). s All the models with the possible exception of Jetset ME give a good description of the data. 4.2 Power Law Correction Analysis The energy dependence of moments of the event shape variables has been described [11] as a sum of the perturbative contributions and a power law dependence due to non-perturbative contributions. The (cid:12)rst moment of an event shape variable f is written as hfi = hf i + hf i ; (1) pert pow where the perturbative contribution hf i has been determined to O((cid:11)2) [24]. The power pert s correction term [11], for 1−T, (cid:26), and C, is given by hf i = c P ; (2) pow f where the factor c depends on the shape variable f and P is supposed to have a universal f form: (cid:20) p (cid:18) p (cid:19)(cid:21) 4C (cid:22) p (cid:11)2( s) s K P = FMpI (cid:11) ((cid:22) )−(cid:11) ( s)−(cid:12) s ln + +1 (3) (cid:25)2 s 0 I s 0 2(cid:25) (cid:22) (cid:12) I 0 p for a renormalisation scale (cid:12)xed at s. The parameter (cid:11) is related to the value of (cid:11) in the 0 s non-perturbative region below an infrared matching scale (cid:22) (= 2 GeV); (cid:12) is (11N −2N )=3, I 0 c f where N is the number of colours and N is the number of active flavours. K = (67/18 − c f (cid:25)2/6)C − 5N =9 and C , C are the usual colour factors. The Milan factor M is 1.49 for N A f F A f = 3. For the jet broadening variables, the power correction term takes the form hf i = c FP ; (4) pow f where (cid:18) (cid:19) (cid:25) 3 (cid:12) p F = p + − 0 −0:6137+O( (cid:11) ) (5) 2 aCF(cid:11)CMW 4 6a CF s and a takes a value 1 for B and 2 for B and (cid:11) is related to (cid:11) [11]. T W CMW s Wehave carriedout(cid:12)tstothe(cid:12)rst momentsofthe(cid:12)ve event shapevariablesseparately with (cid:11) (m ) and (cid:11) as free parameters. The diagonal terms of the covariance matrix between the s Z 0 5 di(cid:11)erent energy points are constructed by summing in quadrature the systematic uncertainty and the statistical error. The o(cid:11)-diagonal terms are obtained from the common systematic errors. The results of the (cid:12)ts are summarised in Table 3 and shown in Figure 3. The (cid:12)ve values of(cid:11) obtained fromtheevent shape variables agreewithin errors, supporting 0 the predicted universality of the power law behaviour. The theoretical predictions for event shapevariables, beingincomplete, givedi(cid:11)erentestimatesof(cid:11) and(cid:11) . Sincethemeasurements 0 s arefullycorrelated,thebestestimatesoftheoverallvaluesareobtainedbytakinganunweighted average: (cid:11) = 0:537 (cid:6) 0:070 (cid:6) 0:021 ; (6) 0 (cid:11) (m ) = 0:1110 (cid:6) 0:0045 (cid:6) 0:0034 : (7) s Z The (cid:12)rst error on each measurement is experimental and is obtained from the average of the (cid:12)ve errors on (cid:11) and (cid:11) . To estimate theoretical uncertainties we vary the renormalisation 0 p s p scale between 0:5 s and 2:0 s and (cid:11) and (cid:11) (m ) vary on average by (cid:6)0:021 and (cid:6)0:0033 0 s Z respectively. A variation of (cid:22) in the range from 1 to 3 GeV gives an additional uncertainty I on (cid:11) (m ) of (cid:6) 0.0010. These two estimates of theoretical uncertainties are combined in s Z quadrature and quoted as the second error. We have also measured the second moments of these shape variables which are summarised in Table 4. The energy dependence of these moments has been analysed in terms of power law corrections. For variables 1−T, (cid:26) and C, the following result is expected to hold [25]: (cid:18) (cid:19) 1 hf2i = hf2 i + 2hf ic P + O : (8) pert pert f s This assumes that the non-perturbative correction to the distributions causes only a shift. For jet broadenings the power corrections are more complicated. The O(1) term has been s parametrised as A =s and is expected to be small for 1−T, (cid:26) and C. Fits have been performed 2 to the second moments where (cid:11) and (cid:11) have been (cid:12)xed to the values obtained from the 0 s corresponding (cid:12)ts to the (cid:12)rst moments. Figure 4 shows the second moments compared to these (cid:12)ts. The contributions of the O(1) term are non-negligible for 1−T and C, in contradiction s with the expectation. The (cid:12)ve values of A , as obtained from the (cid:12)ts, are summarised in 2 Table 3. 4.3 α from Event Shape Distributions s In order to derive (cid:11) from event shape variables at each energy point we (cid:12)t the measured s distributions to theoretical calculations based on O((cid:11)2) perturbative QCD with resummed s leading and next-to-leading order terms. These calculations are performed at parton level and do not include heavy quark mass e(cid:11)ects. To compare the analytical calculations with the experimental distributions, the e(cid:11)ects of hadronisation and decays have been corrected for using Monte Carlo programs. The (cid:12)t ranges used take into account the limited statistics at high energy as well as the reliability of the resummation calculation and are given in Table 5. In this analysis, we deter- p mine (cid:11) at s = 130, 136 and 189 GeV for the (cid:12)rst time. We also include the measurements s p done at s = 161, 172 and 183 GeV since the experimental systematic uncertainties are con- siderably reduced by using an improved selection method and by subtracting the statistical component of the systematic uncertainties. All the measurements are summarised in Table 5. These measurements supersede those published previously [5]. 6 Theexperimentalerrorsincludethestatisticalerrorsandtheexperimentalsystematic uncer- tainties. The theoretical error is obtained from estimates [5] of the hadronisation uncertainty and of the errors coming from the uncalculated higher orders in the QCD predictions. The estimate of the theoretical error does not always reflect the true size of uncalculated higher order terms. An independent estimate is obtained from a comparison of (cid:11) measurements s from many event shape variables which are a(cid:11)ected di(cid:11)erently by higher order corrections and hadronisation e(cid:11)ects. To obtain a combined value for the strong coupling constant we take the unweighted average of the (cid:12)ve (cid:11) values. We estimate the overall theoretical error from the s simple average of the (cid:12)ve theoretical errors or from half of the maximum spread in the (cid:12)ve (cid:11) s values. Both estimates yield similar results. The combined results are summarised in Table 6. p The earlier measurements at s = m and at reduced centre-of-mass energies determined (cid:11) Z s from four event shape variables only: T, (cid:26), B and B . For comparison we also provide in T W Table 6 the mean from these four measurements. We compare the energy dependence of the measured (cid:11) values with the prediction from s QCD in Figure 5a. The theoretical errors are strongly correlated between these measurements. The error appropriate to a measurement of the energy dependence of (cid:11) can then be considered s to be experimental. The experimental systematic errors on (cid:11) are dominated by the back- s ground uncertainties. These are similar for all the individual low energy or high energy data points but di(cid:11)er between the low energy, Z peak and high energy data sets. The experimental systematic errors are then di(cid:11)erent and uncorrelated between the three data sets, but are taken as fully correlated between individual low energy or high energy measurements. The thirteen measurements in Figure 5a are shown with experimental errors only, together with a (cid:12)t to the QCD evolution equation [26] with (cid:11) (m ) as a free parameter. The (cid:12)t gives a (cid:31)2 of 13.5 for 12 s Z degrees of freedom corresponding to a con(cid:12)dence level of 0.34 with a (cid:12)tted value of (cid:11) : s (cid:11) (m ) = 0:1215 (cid:6) 0:0012 (cid:6) 0:0061 : (9) s Z The (cid:12)rst error is experimental and the second error is theoretical. On the other hand, a (cid:12)t with constant (cid:11) gives a (cid:31)2 of 65.1 for 12 degrees of freedom. The value of (cid:11) (m ) thus s s Z obtained is in agreement with the value obtained in the power law ansatz analysis considering the experimental and the theoretical uncertainties. Figure 5b summarises the (cid:11) values determined by L3 from the (cid:28) lifetime measurement [27], s Z lineshape [28] and event shape distributions at various energies, together with the QCD prediction obtained from a (cid:12)t to the event shape measurements only. These measurements support the energy evolution of the strong coupling constant predicted by QCD. The slope in the energy evolution of (cid:11) depends on the number of active flavours. We have s performed a (cid:12)t with N as a free parameter along with (cid:11) and obtain the number of active f s flavours: N = 5:0 (cid:6) 1:3 (cid:6) 2:0 ; (10) f where the (cid:12)rst error is experimental and the second is due to theoretical uncertainties. The errors have been estimated by using the covariance matrix determined from experimental and overall errors on (cid:11) in the (cid:12)t. This result agrees with the expectation N = 5. s f 5 Summary We have measured distributions of event shape variables in hadronic events from e+e− annihi- lation at centre-of-mass energies from 30 GeV to 189 GeV. These distributions as well as the 7 energy dependence of their (cid:12)rst moments are well described by parton shower models. The energy dependence of the (cid:12)rst two moments has been compared with second order perturbative QCD with power law corrections for the non-perturbative e(cid:11)ects. The (cid:12)ts of the (cid:12)ve event shape variables agree with a universal power law behaviour giving (cid:11) = 0.537 (cid:6) 0.070 0 (exp) (cid:6) 0.021 (th). We (cid:12)nd a non-negligible contribution from an O(1) term in describing the s second moments of 1−T, B and C. T The event shape distributions are compared to second order QCD calculations together with resummed leading and next-to-leading log terms. The data are well described by these calculations at all energies. The measurements demonstrate the running of (cid:11) as expected s in QCD with a value of (cid:11) (m ) = 0.1215 (cid:6) 0.0012 (exp) (cid:6) 0.0061 (th). From the energy s Z dependence of (cid:11) , we determine the number of active flavours to be N = 5.0(cid:6) 1.3 (exp) s f (cid:6) 2.0 (th). 6 Acknowledgments We express our gratitudeto the CERNaccelerator divisions forthe excellent performance of the LEP machine. We acknowledge with appreciation the e(cid:11)ort of the engineers, technicians and support sta(cid:11) who have participated in the construction and maintenance of this experiment. 8 Author List The L3 Collaboration: M.Acciarri2,6 P.Achard1,9 O.Adriani,16 M.Aguilar-Benitez2,5 J.Alcaraz2,5 G.Alemanni2,2 J.Allaby,17 A.Aloisio2,8 M.G.Alviggi2,8 G.Ambrosi1,9 H.Anderhub4,8 V.P.Andreev6,,36 T.Angelescu,12 F.Anselmo9, A.Are(cid:12)ev2,7 T.Azemoon,3 T.Aziz1,0 P.Bagnaia3,5 A.Bajo2,5 L.Baksay,43 A.Balandras4, S.V.Baldew2, S.Banerjee1,0 Sw.Banerjee1,0 A.Barczyk4,8,46 R.Barill(cid:18)ere,17 L.Barone3,5 P.Bartalini,22 M.Basile9, R.Battiston,32 A.Bay2,2 F.Becattini1,6 U.Becker1,4 F.Behner4,8 L.Bellucci,16 R.Berbeco3, J.Berdugo2,5 P.Berges,14 B.Bertucci3,2 B.L.Betev4,8 S.Bhattacharya1,0 M.Biasini3,2 A.Biland4,8 J.J.Blaising,4 S.C.Blyth3,3 G.J.Bobbink2, A.Bo¨hm,1 L.Boldizsar1,3 B.Borgia3,5 D.Bourilkov,48 M.Bourquin1,9 S.Braccini1,9 J.G.Branson3,9 V.Brigljevic4,8 F.Brochu4, A.Bu(cid:14)ni1,6 A.Buijs4,4 J.D.Burger1,4 W.J.Burger3,2 X.D.Cai1,4 M.Campanelli,48 M.Capell,14 G.Cara Romeo9, G.Carlino2,8 A.M.Cartacci1,6 J.Casaus2,5 G.Castellini1,6 F.Cavallari,35 N.Cavallo3,7 C.Cecchi,32 M.Cerrada2,5 F.Cesaroni2,3 M.Chamizo,19 Y.H.Chang5,0 U.K.Chaturvedi1,8 M.Chemarin,24 A.Chen5,0 G.Chen,7 G.M.Chen7, H.F.Chen,20 H.S.Chen7, G.Chiefari,28 L.Cifarelli,38 F.Cindolo9, C.Civinini,16 I.Clare1,4 R.Clare,14 G.Coignet4, N.Colino2,5 S.Costantini5, F.Cotorobai1,2 B.de la Cruz2,5 A.Csilling1,3 S.Cucciarelli,32 T.S.Dai1,4 J.A.van Dalen3,0 R.D’Alessandro1,6 R.deAsmundis,28 P.D(cid:19)eglon1,9 A.Degr(cid:19)e,4 K.Deiters,46 D.della Volpe2,8 E.Delmeire,19 P.Denes,34 F.DeNotaristefani,35 A.DeSalvo4,8 M.Diemoz3,5 M.Dierckxsens,2 D.van Dierendonck,2 F.Di Lodovico,48 C.Dionisi,35 M.Dittmar,48 A.Dominguez3,9 A.Doria2,8 M.T.Dova,18,(cid:93) D.Duchesneau4, D.Dufournaud4, P.Duinker2, I.Duran4,0 H.El Mamouni,24 A.Engler3,3 F.J.Eppling1,4 F.C.Ern(cid:19)e,2 P.Extermann,19 M.Fabre4,6 R.Faccini,35 M.A.Falagan2,5 S.Falciano3,5,17 A.Favara1,7 J.Fay,24 O.Fedin3,6 M.Felcini,48 T.Ferguson,33 F.Ferroni3,5 H.Fesefeldt,1 E.Fiandrini,32 J.H.Field,19 F.Filthaut1,7 P.H.Fisher,14 I.Fisk,39 G.Forconi1,4 K.Freudenreich4,8 C.Furetta2,6 Yu.Galaktionov,27,14 S.N.Ganguli1,0 P.Garcia-Abia5, M.Gataullin3,1 S.S.Gau1,1 S.Gentile3,5,17 N.Gheordanescu1,2 S.Giagu3,5 Z.F.Gong2,0 G.Grenier2,4 O.Grimm,48 M.W.Gruenewald8, M.Guida3,8 R.van Gulik,2 V.K.Gupta3,4 A.Gurtu1,0 L.J.Gutay,45 D.Haas5, A.Hasan2,9 D.Hatzifotiadou9, T.Hebbeker8, A.Herv(cid:19)e1,7 P.Hidas1,3 J.Hirschfelder3,3 H.Hofer4,8 G. Holzner4,8 H.Hoorani3,3 S.R.Hou5,0 Y.Hu3,0 I.Iashvili,47 B.N.Jin7, L.W.Jones,3 P.deJong2, I.Josa-Mutuberr(cid:19)(cid:16)a,25 R.A.Khan1,8 M.Kaur1,8,♦ M.N.Kienzle-Focacci1,9 D.Kim,35 J.K.Kim4,2 J.Kirkby,17 D.Kiss,13 W.Kittel3,0 A.Klimentov1,4,27 A.C.K¨onig3,0 A.Kopp4,7 V.Koutsenko1,4,27 M.Kr¨aber,48 R.W.Kraemer,33 W.Krenz,1 A.Kru¨ger4,7 A.Kunin1,4,27 P.Ladron de Guevara2,5 I.Laktineh2,4 G.Landi,16 K.Lassila-Perini4,8 M.Lebeau,17 A.Lebedev1,4 P.Lebrun2,4 P.Lecomte,48 P.Lecoq,17 P.Le Coultre4,8 H.J.Lee,8 J.M.Le Go(cid:11),17 R.Leiste,47 E.Leonardi,35 P.Levtchenko,36 C.Li,20 S.Likhoded4,7 C.H.Lin5,0 W.T.Lin5,0 F.L.Linde,2 L.Lista,28 Z.A.Liu7, W.Lohmann,47 E.Longo,35 Y.S.Lu7, K.Lu¨belsmeyer,1 C.Luci1,7,35 D.Luckey,14 L.Lugnier,24 L.Luminari,35 W.Lustermann4,8 W.G.Ma2,0 M.Maity,10 L.Malgeri1,7 A.Malinin1,7 C.Man~a2,5 D.Mangeol3,0 J.Mans,34 P.Marchesini,48 G.Marian,15 J.P.Martin2,4 F.Marzano3,5 K.Mazumdar1,0 R.R.McNeil,6 S.Mele1,7 L.Merola2,8 M.Meschini1,6 W.J.Metzger3,0 M.von derMey1, A.Mihul1,2 H.Milcent1,7 G.Mirabelli,35 J.Mnich,17 G.B.Mohanty,10 P.Molnar8, T.Moulik,10 G.S.Muanza2,4 A.J.M.Muijs,2 B.Musicar,39 M.Musy,35 M.Napolitano2,8 F.Nessi-Tedaldi,48 H.Newman3,1 T.Niessen,1 A.Nisati,35 H.Nowak4,7 G.Organtini3,5 A.Oulianov2,7 C.Palomares,25 D.Pandoulas1, S.Paoletti3,5,17 P.Paolucci,28 R.Paramatti3,5 H.K.Park3,3 I.H.Park4,2 G.Passaleva,17 S.Patricelli2,8 T.Paul1,1 M.Pauluzzi3,2 C.Paus1,7 F.Pauss,48 M.Pedace3,5 S.Pensotti2,6 D.Perret-Gallix,4 B.Petersen3,0 D.Piccolo,28 F.Pierella9, M.Pieri1,6 P.A.Pirou(cid:19)e3,4 E.Pistolesi,26 V.Plyaskin,27 M.Pohl,19 V.Pojidaev2,7,16 H.Postema1,4 J.Pothier1,7 D.O.Proko(cid:12)ev4,5 D.Proko(cid:12)ev3,6 J.Quartieri3,8 G.Rahal-Callot4,8,17 M.A.Rahaman1,0 P.Raics,15 N.Raja1,0 R.Ramelli,48 P.G.Rancoita2,6 A.Raspereza,47 G.Raven,39 P.Razis,29D.Ren,48 M.Rescigno,35 S.Reucroft1,1 S.Riemann,47 K.Riles,3 A.Robohm4,8 J.Rodin,43 B.P.Roe3, L.Romero,25 A.Rosca8, S.Rosier-Lees,4 J.A.Rubio1,7 G.Ruggiero,16 D.Ruschmeier,8 H.Rykaczewski,48 S.Saremi,6 S.Sarkar3,5 J.Salicio1,7 E.Sanchez1,7 M.P.Sanders3,0 M.E.Sarakinos2,1 C.Scha¨fer1,7 V.Schegelsky3,6 S.Schmidt-Kaerst1, D.Schmitz1, H.Schopper4,9 D.J.Schotanus3,0 G.Schwering1, C.Sciacca2,8 D.Sciarrino1,9 A.Seganti9, L.Servoli,32 S.Shevchenko3,1 N.Shivarov4,1 V.Shoutko2,7 E.Shumilov2,7 A.Shvorob3,1 T.Siedenburg1, D.Son4,2 B.Smith3,3 P.Spillantini1,6 M.Steuer,14 D.P.Stickland3,4 A.Stone6, B.Stoyanov4,1 A.Straessner,1 K.Sudhakar1,0 G.Sultanov1,8 L.Z.Sun,20 H.Suter4,8 J.D.Swain1,8 Z.Szillasi,43,¶ T.Sztaricskai4,3,¶ X.W.Tang7, L.Tauscher5, L.Taylor1,1 B.Tellili2,4 C.Timmermans,30 Samuel C.C.Ting1,4 S.M.Ting1,4 S.C.Tonwar1,0 J.T(cid:19)oth1,3 C.Tully1,7 K.L.Tung,7Y.Uchida1,4 J.Ulbricht4,8 E.Valente3,5 G.Vesztergombi,13 I.Vetlitsky,27 D.Vicinanza3,8 G.Viertel4,8 S.Villa1,1 M.Vivargent4, S.Vlachos5, I.Vodopianov3,6 H.Vogel3,3 H.Vogt4,7 I.Vorobiev2,7 A.A.Vorobyov3,6 A.Vorvolakos2,9 M.Wadhwa5, W.Wallra(cid:11),1 M.Wang1,4 X.L.Wang2,0 Z.M.Wang2,0 A.Weber1, M.Weber1, P.Wienemann1, H.Wilkens,30 S.X.Wu1,4 S.Wynho(cid:11)1,7 L.Xia,31 Z.Z.Xu2,0 J.Yamamoto3, B.Z.Yang2,0 C.G.Yang7, H.J.Yang7, M.Yang7, J.B.Ye,20 S.C.Yeh5,1 An.Zalite3,6 Yu.Zalite3,6 Z.P.Zhang2,0 G.Y.Zhu7, R.Y.Zhu3,1 A.Zichichi9,,17,18 G.Zilizi,43,¶ M.Z¨oller1. 9