WW cross sections and |Vcs| measurement

WW cross sections and |Vcs| measurement

UCLEAR PHYSiC~ PROCEEDINGS SUPPLEMENTS Nuclear Physics B (Proc. Suppl.) 115 (2003) 249-254 ELSEVIER www.elsevier.com/locate/npe W W cross sections...

407KB Sizes 0 Downloads 1 Views

UCLEAR PHYSiC~

PROCEEDINGS SUPPLEMENTS Nuclear Physics B (Proc. Suppl.) 115 (2003) 249-254

ELSEVIER

www.elsevier.com/locate/npe

W W cross sections and Vcs m e a s u r e m e n t A.Ealet a aCNRS/IN2P3, Centre de Physique des Particules, Marseille, FRANCE Data collected by the four LEP experiments at collision energies up to 209 GeV have been analysed to extract the W pair production cross section. Combining all LEP2 centre-of-mass energies has allowed the determination of the W decay branching ratios into leptons and hadrons and to derive the value of the CKM matrix element IVy81. A review of existing direct and indirect IVc~I measurements is also presented.

1. I n t r o d u c t i o n In this paper, the determination of the W W production cross section is reviewed and the W branching ratio into leptons and hadrons is extracted. This is then used to set the most precise value of the CKM matrix element IVcsI. A review of other IVcsI measurements is also presented. These results are crucial for precision tests of the Standard Model (SM). Between 1996 and 2000, LEP increased its centre-of-mass energy from 161 up to 209 GeV. As summarised in Table 1, an integrated luminosity of approximatively 700 p b - l w a s recorded by each experiment in that period. The data from 2000 have been averaged in 2 sets with mean centre-of-mass energies of 204.9 and 206.5 GeV. Most of the results quoted in this paper are the combined values of the four L E P experiments and are preliminary unless otherwise specified. The results are compared with new theoretical predictions recently available through the work of the LEP2 Monte-Carlo workshop [1] .

2.

Event

selections

At LEP, W's are pair-produced and can decay leptonically or hadronically, giving three possible final states. The selections are based on sequential cuts or more elaborate, using likelihood or Neural Networks approach [2,4]. The fully leptonic channels W W -~ lvlv (10.6 % of the decays) are characterized by 2 acoplanar leptons and missing momentum. The selections

are in general cut based analyses and give typical efficiencies of 60-80 % for a low background level of ~ 150 fb. The semileptonic channels account for 43.8 % of final states and are chaxacterised by 2 hadronic jets, an isolated lepton and missing energy from the neutrino. Selections are based on the identification of an isolated lepton (e, # or T jet) and on missing momentum. To improve upon cut based selections, the experiments use multivariate techniques based on probability or likelihood. Very high purity (~ 90 %) and very low background (,~100 fb) are achieved in those analyses for electron and muon lepton samples. In the Tvqq channel, the efficiency is around 60 % with a background of 200 fb. The fully hadronic channel has four hadronic jets, no missing energy with a decay fraction of 45.6%. However, analyses have also to fight against a huge QCD background (e+e qqT(gg)). To enhance background rejection, all experiments use sophisticated analyses based on neural network techniques or likelihood analyses. The neural network from A L E P H is shown as an example on Fig. 1. The event selection efficiencies are around 85 % with a typical background of 1-2 pb. 3. W W c r o s s s e c t i o n s 3.1.

results

Method

The W W cross section in each channel is extracted from a maximum likelihood fit using the number of observed events, a cross-efficiency ma-

0920-5632/03/$ - see front matter © 2003 Published by Elsevier Science B.M doi:l 0.1016/S0920-5632(02)01995-3

250

A. Ealet/Nuclear Physics B (Proc. Suppl.) 115 (2003) 249-254

Table 1 LEP recorded energies and luminosity YEAR

1997 I 1998

1996

1999

2000

V~ (GeV)

161

172

183

189

192

196

200

202

205

207

lumi.(pb -1)

10

10

55

180

30

80

80

40

80

140

3.2. S y s t e m a t i c s Systematic uncertainties for all experiments are dominated by the modeling uncertainties of jet fragmentation, FSI decay and by the uncertainties of detector calibration and event reconstruction. These effects account for ~ 60 lb. In leptonic channels, the lepton identification and reconstruction uncertainties are at the level of 30 fb. Other effects such as statistics, luminosity and theory account for approximatively 40 fb. The total systematic error is at the level of 100 fb for each experiment which is roughly half the average statistical error.

e~

0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 4q NN output

Figure 1. An example of Neural Network ditribution for qqqq selection in ALEPH.

trix between channels, the luminosity and the expected number of background events. The signal WW events is defined via the so-called CC03 diagrams[3] in an unambiguous way. These diagrams belong to the more general 4 fermions final states and the interference with other terms is corrected in the cross section measurement. Combining all channels in one fit allows the extraction of the total WW production cross section and of the leptonic and hadronic W branching fractions.

3.3. R e s u l t s The 4 experiments have given preliminary results at all energies [4,5]. The combined LEP results for the total WW production cross section at the various centre-of-mass energies are shown in Figure 2. The eight points above 190 GeV are preliminary for all experiments, the lower points are final. The LEP combination is extracted for the eight points from a global fit of the 32 values of each experiment taking into account the inter energy and experiment correlations. The systematic uncertainty from the jet fragmentation is assumed to be fully correlated between years and experiments. The luminosity and theory uncertainties are treated as year-by-year correlated. Other systematic contributions are considered as uncorrelated. The experimental points are compared with the most recent calculations using the so-called Double Pole Approximation (DPA) [6], RACOONWW [7] and YFSWW 1.16 [8] which includes the O ( a ) electroweak radiative corrections. These new codes have been shown to agree at a level better than 0.5 %[1] at LEP2 energies . The shaded band

251

A. Ealet/Nuclear Physics B (Proc. Suppl.) 115 (2003) 249-254

of figure 2 represents the theoretical uncertainty, which decreases with the centre-of-mass energy from 0.7 % at 170 GeV to 0.4 % at 210 GeV. Since the DPA calculation can only be used away from the production threshold, a larger theoretical uncertainty is assigned below 170 GeV. All measured cross sections are in good agreement with the two actual predictions.

PRELIMINARY Measured c~ww/ K o r a l W iii]i ii 183 G e V

1.011 + 0.023

189 G e V

0 . 9 6 7 + 0.013 0.991 +- 0.028

192 G e V 08/0712001

LEP

]

196 G e V

1.o15 + o.o18

Preliminar 200 G e V

20

RacoonWW / YFSWW 1.16 0 989 ± 0.024

202 G e V _._

205 G e V

o.

L

0.955 +_o o~8 :

207 G e V

O. 989 +- 0.014

/¢)~'/ I .....

0.979 + 0.009

LEP combined

/

0.9

1.

1.1

LEP WW Working Group Summer 2001

160

170

180

190

200

210

Ecru[GeV]

Figure 3. Ratio of LEP combined pair cross section measurements to the expectations according to Koralw.

Figure 2. W W cross section results of the LEP combined data compared to the predictions of RacoonWW, YFSWW.

3.4. R a t i o o f m e a s u r e d a n d p r e d i c t e d c r o s s sections The ability to test the standard model from the W W measurement can be defined as the ratio meas

Rww-

°ww

o.theo WW

which test the agreement of the measured Wpair cross section with its expectation in a given model.

The standard generator used in all experiments is KORALW[9]which includes ISR LL O(c~3), FSR O ( a 2) via P H O T O S but not the non-leading O(~) electroweak radiative corrections which are included only in YFSWWand RacoonWW. The difference in the ratio when comparing these generators is then a very good test of these extra terms. The R w w ratio combination is performed using the 32 cross sections measured by each experiment at the eight energies. The full covariance matrix is used with the same assumption for systematic sources as that used for the cross section measurement. The results are shown on Figure 3 and 4: in one fit, eight ratio values are extracted at each energy and averaged over the four experiments. Another

252

A. E a l e t / N u c l e a r Physics B (Proc. Suppl.) 115 (2003) 2 4 9 - 2 5 4

seen that in this combination, the systematic and statistical errors are at the same level, mostly because of the size of the correlated errors.

PRELIMINARY Meas u red ~ ww / Racoon W W 183 GeV

[

1.028 + 0.023

189 GeV

..e_i

0.985 + 0.013

ji e

192 GeV

1.012 +_0.029

iiii

196 GeV

~

200 GeV

1.037 +_0.019

~

0.992 i 0.018

i

202 GeV 205 GeV

1.012 + 0.025

~i

o.978+_o.o18

2o7Gev

j,._

LEP combined

~

1.o1,+_o.o15

1

,

0.9

.

,

.

i

1.

Using the data over the full range of centreof-mass energies, a fit to extract the W branching ratios can be performed with and without the assumption of lepton universality. In the fit with lepton universality, the branching ratio to hadrons is determined from that to leptons by constraining the sum to unity. The results from each experiment are given in Table 2 together with the LEP combined value. The leptonic branching ratios without universality are consistent with each other within the errors. The quoted errors include the systematic uncertainty. Assuming lepton universality, the leptonic and hadronic branching ratios are:

1.000 + 0.009 .

,

.

,

1.1

L E P W W Working Grotq~ S u m m e r 2001

Figure 4. Ratio of LEP combined pair cross section measurements to the expectations according to RacoonWW.

fit is made to extract a single value of the ratio to measure the global agreement of the measured and predicted cross-sections over the whole energy range. The global result leads for KORALW and RacoonWW respectively: RKoralW -WW

4. W Branching ratios

0.9788 ± 0.0060(stat) =t=0.0067(syst)

RRa~oWW _ 1.000 ± 0.0061(stat) ± 0.0069(syst) WW YFSWW gives a similar results to that of RacoonWW. As already mentioned, the data agree very well with expectations that include (.9(c~) and disfavour KoralW by almost 2.5 ~. This result is fundamental as it is a good test of the SM at the one loop level. It can also be

B ( W --4 lu) = 10.69 ± 0.06(stat) ± 0.07(syst)% B ( W --4 qq) = 67.92 ± 0.17(stat) :t= 0.21(syst)% Results agree well with the SM expectations which are 10.83 % and 67.51% respectively. The correlated errors between the different channels of each experiment are taken into account in the averaging procedure together with the QCD component which is assumed to be fully correlated among experiments. The systematic errors receive equal contributions from correlated and uncorrelated errors.

5. IVcsl m e a s u r e m e n t s 5.1. IVc~l f r o m B ( W --4 qq) Within the Standard Model, the W hadronic branching ratio can be related to the CKM matrix element IVcs I using B(W -4 qq) 1 - B(W -4 qq)

=

(1 + as(m~v)/Tr )

E

I jI 2

i=u,c j = d , s , b

without the need of a CKM unitarity constraint. Using the world average value of a8 = 0.121 + 0.002 [10] and the sum of the experimental values of CKM matrix elements involving light quarks

A. Ealet/Nuclear Physics B (Proc. Suppl.) 115 (2003) 249-254

253

Table 2 Summary of W branching ratios results in % using all data from 161 GeV up to 208 GeV extracted from the W-pair production cross sections. Exp I B ( W --+ e~,) B ( W --+ #1,') B ( W ~ TV) B ( W ~ qq) 10.95 10.36 10.40 10.40 10.54

A D L 0 LEP

44444-

11.11 4- 0.29 10.62 4- 0.28 9.72 4- 0.31 10.61 4- 0.35 10.54 4- 0.16

0.31 0.34 0.30 0.35 0.17

10.57 10.99 11.78 11.18 11.09

44444-

0.38 0.47 0.43 0.48 0.22

67.33 68.10 68.34 67.91 67.92

Rw_

IVc~] = 0.993 4- 0.013

OPAL

Z

10:

O.l

0.2

0,3

0.4

0.5

0.6

0.7

0.8

0.9

1

_, ;

' ~' i ;

, ~,

~ , i~ , ~ i ,e , , , I . . . .

i ....

] ....

I ....

j ....

Using the direct measurement of the other CKM matrix elements [10], Rc leads to the direct estimation:

( b ) . Ef~.iency o Purity

0



o o "

~ o.6

o •

o

o

o

o •

I~ 0.4

o

o o

o

o

o

o

o

o

o

I

. . . .

o

o

o

°

o

*

°

IVcsl = 0.93 + 0.08(stat) + 0.06(syst)

I ....



0.2

This result is completely independent from the determination of IVcsl from the W branching ratio and provides a good cross check even if the precision is much worse.

*

5.3. . . . .

I

0.1

. . . .

0.2

I

. . . .

0.3

I

0.4

. . . .

I

0.5

. . . .

I

. . . .

0.6

~

0.7

. . . .

I

I

i

0.8

0.9

i D

Combined Likelihood Output

Figure 5. Output of the combined likelihood used to tag charm in OPAL.

IV~sl f r o m

Re A direct measurement of IV~sl can be made by measuring the charm content of hadronic W de5.2.

[V¢ol2+lvcsl 2+lvcb[ 2 ~ , .... IVijl2

OPAL has used data up to 189 GeV[11] to provide a measurement of Rc using jet properties, c-quark lifetime information and leptonic charm decays. Figure 5 shows the combined output likelihood of these variables. The combined result gives

Combined Likelihood Output 1

_

R W = 0.48 4- 0.042(stat) 4- 0.032(syst).

0

-r.

r(W-~cX) FW ~ qq)

j=d,s,b

which is the best existing measurement. The error is dominated by the experimental uncertainty of the W branching ratio measurement.

10:

0.47 0.52 0.52 0.61 0.27

cays as expressed in the ratio

= 1.0477 4- 0.0074 [10], the above result is then interpreted as

'~

44444-

IVcsl s u m m a r y

The OPAL result can be combined with the previous published values from ALEPH, D E L P H I and L3 at lower energies (Fig 6) and gives an average direct LEP mesurement of

Iv~sl

- 0.95 4- 0.08.

This value can be compared to the actual Particle Data Group (PDG) [10] value which is 1.044-0.16. 6. C o n c l u s i o n

Over a five year period, LEP has successfully collected a large amount of data at 10 new en-

A. Ealet/Nuclear Physics B (Proc. Suppl.) 115 (2003) 249-254

254

Preliminary- Vcs

ac ALEPH 11831GeV

~:

DELPHI [1721GeV

.ll~

L3 [183] GeV

~

1.00 ± 0.13 0.87_+ 0.26 0.98_+ 023

!iiiii!i!$1

OPAL[183-189]GeV

hadronic W decays has been compared with the indirect value extracted from the W branching ratio. The results are in excellent agreement and drastically improve the actual PDG value. The LEP collaborations are working to finalize their analyses and to improve the procedure to combine systematic errors for the final LEP result.

~

0.93 ± 0.10

LEP DIRECT

~

0.95_+ 0.08

PDG D-->KIv

~ _ ~

1.040 + 0.1600

i t •

0.996 + 0.0130 0.975 + 0.0005

Acknowledgments I wish to thank my ALEPH and LEP colleagues and the LEP 4-fermion working group for preparing the results presented here.

REFERENCES LEP W~qq Unitarity 0.8

Figure 6. (from Rc) branching the actual

1.0 Vcs

1.5

Summary of IVcsl values from direct and indirect measurements (from W ratio). Results are also compared to PDG value.

ergy points corresponding to more than 10000 W pairs. The study of these data by each experiment and the combination of the results has led to a precise estimation of the WW cross sections at each energy. The experimental values have been compared with theoretical predictions and agree well. The ratio R w w used to compare new models with the data is determined with a comparable precision from experiment and theory and allows to test the SM model at the loop level. Leptonic and hadronic W branching ratio are also extracted and agree well with SM predictions. Finally in a review of the IVcsl measurements, a direct measurement from the charm content of

1. M.W Grunewald, G. Passarino et al., 1999/2000; hep-ph/0005309. 2. ALEPH coll., Phys. Left. B 484, 205 (2000); DELPHI coll., Phys. Lett. B 479,89 (2000); L3 coll., CERN-EP/2000-104; OPAL coll.,Phys. Lett. B 493, 249 (2000). 3. D.Bardin et al., in Physics at LEP2, G.Altarelli et all eds,CERN 96-01(1996), Vol. 2, p. 3, hep-ph/9709270. 4. ALEPH coll., ALEPH 2000-005 CONF 2000002; DELPHI coll., DELPHI 2000-140 CONF 439; L3 coll., L3 Note 2514; OPAL coll., physics Note PN437 and PN420. 5. ALEPH coll., ALEPH 2001-013 CONF 2001010; DELPHI coll., DELPHI 2001-104 CONF 532; L3 coll., L3 Note 2599; OPAL coll., physics Note PN469. 6. W. Beenakker et al., Nucl. Phys. B548, 3 (1999); see also[l] for discussions. 7. A. Denner et al., Phys. Lett. B 475, 127 (2000); 131-TP 2000/06, hep-ph/0006307. 8. S. Jadach et al., Phy. Rev. D 61 113010 (2000); Comp. Phy. Comm. 140, 432 (2001). 9. M.Skrzypek, S.Jadach, W. placzek and Z. Was, Comp. Phy. Comm. 94, 216 (1996). 10. Particle Data Group, D.E. Groom et al. Eur. Phys. J. C 15, 1 (2000). 11. OPAL coll., CERN-EP 2000/100.