ZU-TH 15/09, IPPP/09/87, ETH-IPP-2009-11

Precise determination of the strong coupling constant at NNLO in QCD from the three-jet rate in electronpositron annihilation at LEP
G. Dissertoria, A. Gehrmann-De Ridderb , T. Gehrmannc , E.W.N. Gloverd, G. Heinrichd , H. Stenzele
Institute for Particle Physics, ETH Zurich, CH-8093 Zurich, Switzerland Institute for Theoretical Physics, ETH Zurich, CH-8093 Zurich, Switzerland c Institut fr Theoretische Physik, Universitt Zrich, CH-8057 Zrich, Switzerland u a u u Institute for Particle Physics Phenomenology, Department of Physics, University of Durham, Durham, DH1 3LE, UK e II. Physikalisches Institut, Justus-Liebig Universitt Giessen, D-35392 Giessen, Germany a (Dated: February 10, 2010)
b a

arXiv:0910.4283v2 [hep-ph] 10 Feb 2010

We present the rst determination of the strong coupling constant from the three-jet rate in e+ e annihilation at LEP, based on a next-to-next-to-leading order (NNLO) perturbative QCD prediction. More precisely, we extract s (MZ ) by tting perturbative QCD predictions at O(3 ) s to data from the ALEPH experiment at LEP. Over a large range of the jet-resolution parameter ycut this observable is characterised by small non-perturbative corrections and an excellent stability under renormalisation scale variation. We nd s (MZ ) = 0.1175 0.0020 (exp) 0.0015 (theo), which is more accurate than the values of s (MZ ) from e+ e event shape data currently used in the world average.
PACS numbers: 12.38.Bx, 13.66.Bc, 13.66.Jn, 13.87.-a

Jet observables in electronpositron annihilation play an outstanding role in studying the dynamics of the strong interactions [1], described by the theory of Quantum Chromodynamics (QCD, [2]). In particular, jet rates and related event-shape observables have been extensively used for the determination of the QCD coupling constant s (see [3, 4] for a review), mostly based on data obtained at the e+ e colliders PETRA, LEP and SLC at centre-of-mass energies from 14 to 209 GeV. Jets are dened using a jet algorithm, which describes how to recombine the particles in an event to form the jets. A jet algorithm consists of two ingredients: a distance measure and a recombination procedure. The distance measure is computed for each pair of particles to select the pair with the smallest separation in momentum space. If the separation is below a pre-dened resolution parameter ycut , the pair are combined according to the recombination procedure. The JADE algorithm [6] uses the pair invariant mass as distance measure. Several improved jet algorithms have been proposed for e+ e collisions: Durham [7], Geneva [8] and Cambridge [9]. The Durham algorithm has been the most widely used by experiments at LEP [1013] and SLD [14], as well as in the reanalysis of earlier data at lower energies from JADE [15]. The Durham jet algorithm clusters utilises the distance measure yij,D =
2 2 2 min(Ei , Ej )(1 cos ij ) 2 Evis

(1)

maining (pseudo-)particles are the jets. In experimental jet measurements, one studies the jet rates, i.e. jet cross sections normalised to the total hadronic cross section, as function of the jet-resolution parameter ycut . The theoretical prediction of jet cross sections is made within perturbative QCD, where the same jet algorithm is applied to the nal state partons. The QCD description of jet production is either based on a xed-order calculation or a parton shower. The xed order approach uses exact parton-level matrix elements including higher order corrections where available and/or analytical resummation of large logarithmic corrections for a given jet multiplicity. On the other hand, the parton shower starts with the leading-order matrix element for two-jet production and generates higher multiplicities in an iterative manner, thereby accounting only for the leading logarithmic terms from parton-level processes with higher multiplicity. In multi-purpose event generator programs [1618], such parton showers are complemented by phenomenological models which describe the transition from partons to hadrons. These programs provide a satisfactory description of multi-jet production rates but, since they generally contain many tunable phenomenological parameters, their predictive power is limited. Nevertheless, in order to compare parton level predictions with experimental hadronic data, these event generators are vital to estimate the eects due to hadronisation and resonance decays. Until recently, xed-order calculations were available up to next-to-next-to-leading order (NNLO) for two jets [1921] and up to next-to-leading order (NLO) for three [2224] and four jets [2528]. For ve and more jets, only leading order calculations are available [2931]. For jets involving massive quarks, NLO results are available for three-jet nal states [32]. The recent calculations of the 3 corrections (NNLO) for three-jet production [33 s 36] have already led to precise s determinations [3741],

for each pair (i, j) of particles, Evis denotes the energy sum of all particles in the nal state. The pair with the lowest yij,D is replaced by a pseudo-particle whose fourmomentum is given by the sum of the four-momenta of particles i and j (E recombination scheme). This procedure is repeated as long as pairs with invariant mass below the predened resolution parameter yij,D < ycut are found. Once the clustering is terminated, the re-

2 using event-shape observables measured by ALEPH and JADE. However, some of the event-shape variables still suer from a poor convergence of the perturbative expansion even at NNLO. Furthermore, the usage of event generators, which have been tuned to LEP data, for the determination of the hadronisation corrections may lead to a bias in the s measurements for some of the event shapes [41]. A comparison of dierent variables showed that jet broadening variables are most aected by missing higher orders and a potential hadronisation bias, while the dierential two-jet rate Y3 is most robust against these eects, and strongly motivates the present study of the three-jet rate. In this letter we describe a determination of the strong coupling constant from the three-jet rate measured by ALEPH [42] at LEP. We use the NNLO predictions as presented in [34]. There it was shown that: (i) For large values of ycut , ycut > 102 , the NNLO corrections turn out to be very small, while they become substantial for medium and low values of ycut ; (ii) The maximum of the jet rate is shifted towards higher values of ycut compared to NLO and is in better agreement with the experimental observations; (iii) The theoretical uncertainty is lowered considerably compared to NLO, especially in the region 101 > ycut > 102 relevant for precision phenomenology where the theory error is below two per-cent relative uncertainty; (iv) Finally, in this ycut region the parton level predictions at NNLO are already very close to the experimental measurements, indicating the need for only small hadronisation corrections. These ndings motivate a dedicated analysis of the three-jet rate, leading to a precise measurement of s . Our analysis closely follows the procedure described in [37, 41]. The ALEPH data [42] at LEP are based on the reconstructed momenta and energies of charged and neutral particles. The measurements have been corrected for detector eects, i.e. the nal distributions correspond to the so-called particle (or hadron) level, and for initial state photonic radiation. In the simulation of the detector response to particles, a bias is introduced by the choice of the physics event generator. This leads to a systematic uncertainty on the three-jet rate of about 1.5% for the relevant ycut range. Further experimental systematic eects are estimated by a variation of the trackand event-selection cuts as advocated in [42], giving an additional small systematic uncertainty of about 1%. We construct the perturbative expansion up to O(3 ) s as described in [41], with the coecients obtained from [34]. These are valid for massless quarks. We take into account bottom mass eects up to NLO [32], for a pole b-quark mass of Mb = 4.5 GeV. The latter is varied by 0.5 GeV in order to estimate the impact of the b-quark mass uncertainty on the value of the strong coupling. For the normalisation to the total hadronic cross section had we follow the procedure adopted in [41], which is based on a N3 LO calculation (O(3 ) in QCD) for had [43], s including mass corrections for the b-quark up to O(s ) and the leading mass terms to O(2 ). Weak corrections s to the three-jet rate were computed very recently [44]. They are at the one per-mille level for Q = MZ and are neglected here. The nominal value for the renormalisation scale x = /Q is unity. It is varied between 0.5 < x < 2 in order to assess the systematic uncertainty related to yet unknown higher order corrections. No attempt is made to combine the NNLO predictions with resummation calculations. At present, the resummation of the three-jet rate [7] is only fully consistent at leading logarithmic level [45], and resummation eects only become numerically relevant over xed-order NNLO for ln ycut < 4.5 (as can be seen from the Y3 transition parameter distribution [46]), which is below our region of interest. In order to compare the perturbative parton level thoretical prediction with the hadronic data, it is necessary to apply a correction for hadronisation and resonance decays. This bin-by-bin correction is computed with the PYTHIA [16], HERWIG [17] and ARIADNE [18] Monte Carlo generators, all tuned to global hadronic observables at MZ [47]. The parton level is dened by the quarks and gluons present at the end of the parton shower in PYTHIA and HERWIG and the partons resulting from the colour dipole radiation in ARIADNE. Our central values for the strong coupling constant are obtained with hadronisation corrections from PYTHIA, which are at the level of 5%. We dene the systematic uncertainty on s (MZ ) due to these hadronisation corrections as the biggest deviation observed when using any of the other generators. Motivated by the observations in [41], we veried that the shapes of the Monte Carlo parton level predictions are in fair agreement with those at NNLO, for reasonable choices of the strong coupling. Furthermore, the ratios of these predictions are relatively at over the relevant ycut range, giving further condence in the reliability of the hadronisation corrections. The corrected ALEPH measurements for the three-jet rate are compared to the theoretical calculation at particle level. Values for s (MZ ) are obtained by a leastsquares t, performed separately for each ycut value in the range listed in Table I (for the data at the Z peak), together with the uncertainties as described above. These results are also displayed in Fig. 1. We observe a nice stability of the results, within their total uncertainties, down to resolution parameters of ln ycut 4.5. Beyond that value we nd a fall-o of s (MZ ), most likely related to the onset of large logarithmic corrections from higher perturbative orders, which are not accounted for in our perturbative prediction. As nal result we quote our measurement for ycut = 0.02, which represents an optimal compromise between minimal systematic uncertainty and stability. We nd s (MZ ) = 0.1175 0.0020 (exp) 0.0015 (theo) where the rst uncertainty includes (in quadrature) the contributions from statistics, detector corrections and experimental selection cuts, and the second error is the quadratic sum of b-quark mass and renormalisa-

3 in e+ e annihilation at LEP, based on a NNLO perturbative QCD prediction. We nd a precise value of s (MZ ) with an uncertainty of 2%, consistent with the world average. This veries the expectations that the three-jet rate is an excellent observable for this kind of analysis, thanks to the good behaviour of its perturbative and non-perturbative contributions over a sizable range of jet-resolution parameters. Acknowledgements: This research was supported in part by the Swiss National Science Foundation (SNF) under contracts PP0022-118864 and 200020-126691, by the UK Science and Technology Facilities Council, by the European Commissions Marie-Curie Research Training Network under contract MRTN-CT-2006-035505 Tools and Precision Calculations for Physics Discoveries at Colliders and by the German Helmholtz Alliance Physics at the Terascale. EWNG gratefully acknowledges the support of the Wolfson Foundation and the Royal Society.

0.125 0.1225 0.12 0.1175 0.115 0.1125 0.11 0.1075 0.105 0.1025 0.1

s(MZ)

Q=MZ

central result with stat. uncertainty total uncertainty -6 -5 -4 -3 -2 ln(ycut) -1

0.005
total error perturbative statistical hadronisation

0.004 0.003 s(MZ) 0.002 0.001 0 -6

experimental

-5

-4

-3

-2 ln(ycut)

-1

FIG. 1: Determinations of s (MZ ) from the three-jet rate, measured by ALEPH at the Z peak, for several values of the jet-resolution parameter ycut . The error bars show the statistical uncertainty, whereas the shaded band indicates the total error, including the systematic uncertainty. The various contributions to the latter are displayed in the lower plot.

tion scale uncertainties (cf. Table I). We also performed similar measurements for the LEP2 energies between 133 and 206 GeV, where we nd consistent values for s (MZ ), but with considerably larger statistical uncertainties. Combining the errors in quadrature, yields s (MZ ) = 0.1175 0.0025 which is in excellent agreement with the latest world average value [4] of s (MZ ) = 0.1184 0.0007 that is based on a number of measurements from -decay, lattice gauge theory, Upsilon decay, DIS and e+ e data. As expected, our theoretical uncertainty is smaller than that obtained from ts of event-shape distributions, and even smaller than the experimental error, which is dominated by the modeldependence of the detector corrections. Our result is also more precise than the two extractions of s from e+ e event-shape data [40, 41] currently used in the world average [4]. In this letter we reported on the rst determination of the strong coupling constant from the three-jet rate

ln(ycut ) s (MZ ) -5.1 0.1110 0.1124 -4.9 -4.7 0.1147 -4.5 0.1153 0.1159 -4.3 -4.1 0.1170 -3.9 0.1175 -3.7 0.1179 0.1183 -3.5 -3.3 0.1184 -3.1 0.1179 0.1177 -2.9 -2.7 0.1180 -2.5 0.1169 -2.3 0.1166 0.1166 -2.1 -1.9 0.1191 -1.7 0.1173 0.1175 -1.5 -1.3 0.1159

stat. 0.0004 0.0004 0.0004 0.0004 0.0004 0.0004 0.0004 0.0004 0.0004 0.0004 0.0004 0.0004 0.0004 0.0005 0.0005 0.0006 0.0008 0.0010 0.0016 0.0037

det. 0.0013 0.0015 0.0015 0.0015 0.0016 0.0016 0.0016 0.0016 0.0015 0.0015 0.0016 0.0019 0.0020 0.0021 0.0019 0.0020 0.0021 0.0015 0.0005 0.0014

exp. 0.0008 0.0007 0.0008 0.0008 0.0009 0.0009 0.0011 0.0011 0.0009 0.0011 0.0013 0.0013 0.0013 0.0015 0.0018 0.0020 0.0019 0.0023 0.0029 0.0029

had. 0.0003 0.0003 0.0004 0.0005 0.0005 0.0005 0.0006 0.0006 0.0006 0.0008 0.0010 0.0010 0.0013 0.0013 0.0014 0.0015 0.0014 0.0016 0.0014 0.0018

mass 0.0004 0.0003 0.0003 0.0003 0.0003 0.0003 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.0001 0.0001 0.0001 0.0001 0.0002 0.0001 0.0001 0.0004

pert. 0.0020 0.0013 0.0012 0.0006 0.0010 0.0012 0.0014 0.0016 0.0018 0.0019 0.0021 0.0021 0.0020 0.0021 0.0021 0.0020 0.0016 0.0019 0.0017 0.0011

total 0.0025 0.0022 0.0022 0.0019 0.0022 0.0023 0.0025 0.0026 0.0026 0.0029 0.0031 0.0033 0.0034 0.0036 0.0037 0.0038 0.0036 0.0038 0.0040 0.0054

TABLE I: Results of s (MZ ) extracted from the three-jet rate measured by ALEPH at LEP1. The uncertainty contributions are given for the statistical error (stat.), the uncertainty related to the choice of the generator for the simulation of the detector response (det.), the quadratic sum of all other experimental systematic uncertainties arising from track and event selection cut variations (exp.), the hadronisation uncertainty obtained by the maximum dierence between either PYTHIA, HERWIG or ARIADNE (had.), the uncertainty on the b-quark mass correction procedure (mass) and the uncertainty for missing higher orders (pert.) estimated by a variation of the renormalisation scale.

[1] J. R. Ellis, M. K. Gaillard and G. G. Ross, Nucl. Phys. B 111 (1976) 253 [Erratum-ibid. B 130 (1977) 516]. [2] H. Fritzsch, M. Gell-Mann and H. Leutwyler, Phys. Lett. B 47 (1973) 365; D.J. Gross and F. Wilczek, Phys. Rev. D 8 (1973) 3633; H.D. Politzer, Phys. Rept. 14 (1974) 129.

[3] O. Biebel, Phys. Rept. 340 (2001) 165; S. Kluth, Rept. Prog. Phys. 69 (2006) 1771. [4] S. Bethke, Prog. Part. Nucl. Phys. 58 (2007) 351; arXiv:0908.1135. [5] R. Barate et al. [ALEPH Collaboration], Z. Phys. C 76 (1997) 1; R. Akers et al. [OPAL Collaboration], Z. Phys.

4
C 65 (1995) 367; P. Abreu et al. [DELPHI Collaboration], Z. Phys. C 59 (1993) 357. W. Bartel et al. [JADE Collaboration], Z. Phys. C 33 (1986) 23. S. Catani, Y.L. Dokshitzer, M. Olsson, G. Turnock and B.R. Webber, Phys. Lett. B 269 (1991) 432; N. Brown and W.J. Stirling, Phys. Lett. B 252 (1990) 657; Z. Phys. C 53 (1992) 629; W.J. Stirling et al., Proceedings of the Durham Workshop, J. Phys. G17 (1991) 1567. S. Bethke, Z. Kunszt, D.E. Soper and W.J. Stirling, Nucl. Phys. B 370 (1992) 310; 523 (1998) 681 (E). Y.L. Dokshitzer, G.D. Leder, S. Moretti and B.R. Webber, JHEP 9708 (1997) 001. A. Heister et al. [ALEPH Collaboration], Eur. Phys. J. C 35 (2004) 457. G. Abbiendi et al. [OPAL Collaboration], Eur. Phys. J. C 40 (2005) 287. P. Achard et al. [L3 Collaboration], Phys. Rept. 399 (2004) 71. J. Abdallah et al. [DELPHI Collaboration], Eur. Phys. J. C 29 (2003) 285. K. Abe et al. [SLD Collaboration], Phys. Rev. D 51 (1995) 962. P. Pfeifenschneider et al. [JADE collaboration], Eur. Phys. J. C 17 (2000) 19. T. Sjostrand, P. Eden, C. Friberg, L. Lnnblad, G. Miu, o S. Mrenna and E. Norrbin, Comput. Phys. Commun. 135 (2001) 238. G. Corcella et al., JHEP 0101 (2001) 010. L. Lnnblad, Comput. Phys. Commun. 71 (1992) 15. o C. Anastasiou, K. Melnikov and F. Petriello, Phys. Rev. Lett. 93 (2004) 032002. A. Gehrmann-De Ridder, T. Gehrmann and E.W.N. Glover, Nucl. Phys. B 691 (2004) 195. S. Weinzierl, Phys. Rev. D 74 (2006) 014020. R.K. Ellis, D.A. Ross and A.E. Terrano, Nucl. Phys. B 178 (1981) 421. Z. Kunszt, Phys. Lett. B 99 (1981) 429; J.A.M. Vermaseren, K.J.F. Gaemers and S.J. Oldham, Nucl. Phys. B 187 (1981) 301; K. Fabricius, I. Schmitt, G. Kramer and G. Schierholz, Z. Phys. C 11 (1981) 315. Z. Kunszt and P. Nason, in Z Physics at LEP 1, CERN Yellow Report 89-08, Vol. 1, p. 373; W. T. Giele and E.W.N. Glover, Phys. Rev. D 46 (1992) 1980; S. Catani and M. H. Seymour, Phys. Lett. B 378 (1996) 287. L.J. Dixon and A. Signer, Phys. Rev. Lett. 78 (1997) 811; Phys. Rev. D 56 (1997) 4031. Z. Nagy and Z. Trocsanyi, Phys. Rev. Lett. 79 (1997) 3604. [27] J. Campbell, M.A. Cullen and E.W.N. Glover, Eur. Phys. J. C 9 (1999) 245. [28] S. Weinzierl and D.A. Kosower, Phys. Rev. D 60 (1999) 054028. [29] K. Hagiwara and D. Zeppenfeld, Nucl. Phys. B 313 (1989) 560; F.A. Berends, W.T. Giele and H. Kuijf, Nucl. Phys. B 321 (1989) 39; N.K. Falck, D. Graudenz and G. Kramer, Nucl. Phys. B 328 (1989) 317. [30] S. Moretti, Nucl. Phys. B 544 (1999) 289. [31] F. Krauss, R. Kuhn and G. So, JHEP 0202 (2002) 044. [32] W. Bernreuther, A. Brandenburg and P. Uwer, Phys. Rev. Lett. 79 (1997) 189; G. Rodrigo, A. Santamaria and M. S. Bilenky, Phys. Rev. Lett. 79 (1997) 193; P. Nason and C. Oleari, Nucl. Phys. B 521 (1998) 237. [33] A. Gehrmann-De Ridder, T. Gehrmann, E.W.N. Glover and G. Heinrich, Phys. Rev. Lett. 99 (2007) 132002; JHEP 0711 (2007) 058; JHEP 0712 (2007) 094; JHEP 0905 (2009) 106. [34] A. Gehrmann-De Ridder, T. Gehrmann, E.W.N. Glover and G. Heinrich, Phys. Rev. Lett. 100 (2008) 172001. [35] S. Weinzierl, Phys. Rev. Lett. 101 (2008) 162001. [36] S. Weinzierl, JHEP 0906 (2009) 041; JHEP 0907 (2009) 009; arXiv:0909.5056. [37] G. Dissertori et al., JHEP 0802 (2008) 040. [38] T. Becher and M.D. Schwartz, JHEP 0807 (2008) 034. [39] R.A. Davison and B.R. Webber, Eur. Phys. J. C 59 (2009) 13. [40] S. Bethke, S. Kluth, C. Pahl and J. Schieck [JADE Collaboration], arXiv:0810.1389. [41] G. Dissertori et al., JHEP 0908 (2009) 036. [42] A. Heister et al. [ALEPH Collaboration], Eur. Phys. J. C 35 (2004) 457. [43] K. G. Chetyrkin, A. L. Kataev and F. V. Tkachov, Phys. Lett. B 85 (1979) 277; W. Celmaster and R. J. Gonsalves, Phys. Rev. Lett. 44 (1980) 560; M. Dine and J. R. Sapirstein, Phys. Rev. Lett. 43, 668 (1979); K.G. Chetyrkin, J.H. Khn and A. Kwiatkowski, Phys. u Rept. 277 (1996) 189. [44] A. Denner, S. Dittmaier, T. Gehrmann and C. Kurz, Phys. Lett. B 679 (2009) 219. [45] A. Ban, G.P. Salam and G. Zanderighi, JHEP 0201 (2002) 018; JHEP 0503 (2005) 073. [46] T. Gehrmann, G. Luisoni and H. Stenzel, Phys. Lett. B 664 (2008) 265. [47] R. Barate et al. [ALEPH Collaboration], Phys. Rep. 294 (1998) 1.

[6] [7]

[8] [9] [10] [11] [12] [13] [14] [15] [16]

[17] [18] [19] [20] [21] [22] [23]

[24]

[25] [26]