Heavy Flavor Production in Hadron Collisions (with a few leptons and photons thrown in)

a r X i v :h e p -e x /0408043v 1 12 A u g 2004

XXIV Physics in Collision -Boston,June 27-29,2004

HEAVY FLAVOR PRODUCTION IN HADRON COLLISIONS (WITH A FEW LEPTONS AND PHOTONS THROWN IN)

Harold G.Evans

Columbia University,New York,NY,USA

ABSTRACT

Substantial advances in our understanding of several aspects of QCD have been achieved in the recent past using heavy quarks as a tool.However,many open questions still remain.These successes and puzzles are highlighted by the latest measurements of heavy quark production at the Tevatron,HERA and ?xed tar-get experiments,which will be reviewed here.Results in both open heavy ?avor and heavy quarkonium production as well as evidence for new particles containing heavy quarks will be presented.The impact of these measurements on gaps in our understanding of QCD and how we hope to close these gaps in the future will be outlined.

1Why Study Heavy Quark Production?

Quantum Chromodynamics is universally acknowledged to be the theory of the strong force.However,its study continues to be a compelling area of research because of the di?culty of performing calculations in regions where the theory be-comes non-perturbative.This means that,although we understand the structure of the theory,we still cannot make accurate predictions for a wide range of impor-tant observables.Intellectually,this is frustrating(or an opportunity for the more optimistic).But it also has a more practical consequence.Our understanding of QCD processes is intimately entwined with our understanding of other aspects of the Standard Model because QCD is a part of all SM predictions,from estimates of backgrounds to corrections to electro-weak observables.

It turns out that the production of heavy quarks by the strong force is an excellent area to study some of the technical details of QCD that are so important in our tests of the Standard Model.To understand this,consider the production of a heavy quark-antiquark pair in the collision of two particles.Broadly speaking, this process consists of three components,which are all connected in real collisions: the structure of the incoming particles,the hard interaction producing the QˉQ pair and the subsequent parton shower and fragmentation of the?nal state partons to produce observable hadrons.

It is in the second of these entwined processes,the hard scattering,where heavy quarks make their contribution to QCD.Particle structure and hadronization are clearly governed by non-perturbative physics.However,they are also largely universal functions,appearing in a variety of processes.The hard-scatter is pro-cess dependent.But since the masses of heavy quarks are much larger than the QCD scale,this hard-scatter should be calculable using perturbative QCD.Heavy quark production measurements can therefore be used to probe our ability to do perturbative calculations or can be used as a tool to understand parton densities and fragmentation.

Before embarking on a discussion of speci?c heavy quark production re-sults,we should be clear as to exactly what a heavy quark is.In this paper,heavy quarks are taken to be b-and c-quarks.The obese t-quark is discussed in a separate contribution to these proceedings[1].Using this de?nition,heavy quarks are pro-duced at a variety of facilities.A comparison of those for which results are presented is given in Table1.

Table1:Comparison of experimental facilities with results presented here.

√Runs

Exp.or

Particles

γBeO96–97

Σ?,π?C,Cu96–97

p Cu96–97

p C,Al,Ti,W00,02–03

HERA Run I300,318130pb?1

HERA Run II318～70pb?1

pˉp92–96

pˉp02–04

LEP(I and II)90–210 3.6M bˉb

2b-and c-Quark Production

2.1History Lessons

The production of open b-and c-quarks has been one of the most troubling prob-lems in QCD for more than a decade.For a recent review of this problem see[2]. Particularly in the b-quark sector,calculations,which are done at next-to-leading order(NLO)inαs,were expected to provide a quite good description of the data.

A quick look at the data taken prior to2000[2],however,indicates that while the shape of the b-quark production cross-section is reasonably well modeled by NLO theory for pˉp→bˉb,e p→bˉb andγγ→bˉb data,the predictions underestimate the mag-nitude of the cross-sections by factors approaching three.Surprisingly,data and predictions for c-quark production showed much better agreement,although with larger uncertainties.

Over the past few years,the picture of b-quark production at the Tevatron, where the discrepancy was originally uncovered,has become much clearer.One important aspect of this understanding was the realization that experimentalists should report what they observe:B-hadron production cross-sections,rather than cross-sections corrected to the b-quark level.When the D?collaboration published a measurement of the p T distribution of jets containing b-quarks[3],signi?cantly better agreement with NLO predictions was found.Another piece of the puzzle was the correct incorporation of next-to-leading-log resummation of log(p T/m)terms with the NLO hard scatter calculation including massive quarks(FONLL)[4].Finally,the heavy quark fragmentation function was revisited by several groups[5,6]yielding a

calculation in the FONLL framework consistent with the hard scattering calculation and a reevaluation of parameters of the fragmentation function.2.2Open Beauty and Charm Production at the Tevatron

These new calculations [7]have

been compared to a preliminary measurement of the B -hadron cross-section by CDF,using H B →J/ψX decays.CDF selects H B →J/ψdecays in the J/ψ→μ+μ?mode from 37pb ?1of their Run II data using the position of the J/ψvertex with respect to the primary p ˉp interaction point to distinguish long-lived H B decays from prompt J/ψproduction.The resulting H B cross-section times branching ratio is shown on the left side of Figure 1while a comparison of this new result to older CDF measurements and to the FONLL prediction [7]is shown on the right side.As can be seen,the agreement between data and prediction is pared to FONLL predictions [9].The measurement uses 5.8pb ?1of hadronic charm decay triggers collected with the CDF Silicon Vertex Trigger.Prompt contributions

to the sample of reconstructed D 0,D ?+,D +and D +

s mesons are obtained using the

impact parameter of the charm meson candidate.The measured di?erential cross-sections in the rapidity region |y |≤1agree fairly well with FONLL predictions,as shown in Figure

2,although the data lie systematically on the high side of the theory.

Figure 2:The CDF di?erential charm meson cross-section measurements compared to the FONLL prediction.

2.3Open Beauty and Charm at Fixed Target

The Hera-B experiment has made new measurements of open b -and c -quark pro-duction in a ?xed target environment.Their preliminary measurement of the b ˉb cross-section uses J/ψ→e +e ?,μ+μ?decays with displaced vertices from a total of 320K J/ψcandidates in ～35%of their 02-03data sample.The new Hera-B measure-ment,σ(b ˉb )=12.3+3.5?3.2(stat)nb/nucleon,is lower than their previous result,32+14+6

?12?7nb/nucleon [10],which used the 40-times smaller 2000data sample.It agrees well with the prediction of Kidonakis,et al.[11],30±13nb/nucleon,although the errors on the prediction are still rather large.

Hera-B has also made a preliminary measurement of the open charm cross-section using 98±12D 0,189±20D +and 43±8D ?+fully reconstructed mesons.

The resulting D ˉD

cross-sections,in μb/nucleon,extrapolated to the full x F range,are σ(D +)=30.2±4.5±5.8and σ(D 0)=56.3±8.5±9.5,which are consistent with previous measurements but signi?cantly more accurate.

Measurements of charm production have also been made by the FOCUS

collaboration,which has produced new results on charm baryon/anti-baryon pro-duction asymmetries [12].The asymmetry is predicted to be vanishingly small by perturbative QCD.However “leading particle e?ects”,can enhance the production

of baryons sharing valence quarks with the target or projectile particles.The mea-sured integrated and di?erential asymmetries for Λ+c ,Λ+c (2625),Σ++(?)c and Σ0(?)

c agree poorly with predictions from the PYTHIA Monte Carlo.For example,FO-CUS measures the asymmetry in the production of Λ+c baryons an

d anti-baryons to b

e 0.111±0.018±0.012,1.8σaway from the prediction o

f 0.073.A better de-scription of older asymmetry measurements has been achieved usin

g heavy quark recombination models

[13].But this has yet to be compared to the FOCUS data.2.4Open Beauty and Charm in ep Collisions

Experimentally,both H1and ZEUS search for b -quark production in ep collisions using muon plus jet(s)events.ZEUS separates b -quark events from backgrounds using the component of the muon’s momentum transverse the the closest jet axis,P rel t ,while H1takes advantage of their silicon strip detectors to include the impact parameter of the muon track with respect to the primary interaction vertex,along with P rel t ,to their list of discriminating variables.

The experiments make measurements in two kinematic regions –the deep

inelastic scattering (DIS)regime,where photon virtuality is high (Q 2>1GeV 2)and the photo-production (PhP)regime,where there the photon is nearly real (Q 2<1GeV 2).Each of these regimes is sensitive to di?erent e?ects in heavy quark pro-duction and provide complementary input to the measurements from the Tevatron,where log(p T /m )e?ects,for example,are expected to be much more important.The results of preliminary H1measurements from 2003and 2004and of published ZEUS data [14]are shown in Figure 3.

p μ

t [GeV]

D a t a /T h e o r y

10

Q 2

[GeV 2

]d σ/d Q 2[p b /G e V 2]

Q 2 (GeV 2

)

d V /d Q 2 (p b /G

e V 2)

)

(x ) (p b )

10

10

10

1

1010

10

Figure 3:Measurements of b -production by H1and ZEUS in photo-production (left

plot)and DIS (middle and right plots).

Charm quark production is also measured by both experiments using D ?±

mesons.Preliminary results from H1(2003)and from ZEUS(2002)in PhP events as well as recently published ZEUS data

[15]are presented in Figure4.

η(D*)

d

σ

/

d

η

(

D

*

)

(

n

b

)

Figure4:Measurements of D?production by H1and ZEUS in DIS(left)and by ZEUS in photo-production(right)

Agreement between both beauty and charm data and NLO predictions is generally good within the relatively large experimental and theoretical errors. Some problems may arise in b-quark production at low p T and low Q2(see Figure 3).However,H1and ZEUS do not see the same discrepancies.In the charm PhP data,mild deviations between data and theory are observed in the medium p T and highηregions.ZEUS has studied these further in a preliminary measurement of the cross-section of jets containing D?±mesons,designed to reduce sensitivity to hadronization e?ects.As can be seen in Figure4some disagreement between data and predictions remains indicating that hadronization is unlikely to be the main cause of the problem.

2.5b-and c-Quark Summary

The general picture emerging from new measurements of beauty and charm produc-tion and from recent theoretical advances is of remarkably better agreement between data and https://www.sodocs.net/doc/3b2115957.html,parisons between measurement and prediction for the results discussed above show agreement to within about two sigma(taking into account both experimental and theoretical errors)for all recent measurements.This is ob-viously a big improvement over the situation a few years ago.However,optimism

should not be allowed to run rampant over caution.Uncertainties on nearly all measurements are dominated by systematic errors indicating that higher statistics alone will be unlikely to produce major improvements in accuracy.On the theoreti-cal side,uncertainties on the predictions are nearly always substantially larger than those on the measurements further adding to the di?culty of making quantitative comparisons.

3Heavy Quarkonium Production

3.1More History

As was the case with open beauty and charm,our understanding of the production of bound heavy quark-antiquark states has had a checkered past.(For a discussion of the decays of quarkonia see[16]).Until the late90’s the direct production of J/ψandΥstates was expected to proceed via a color singlet mechanism(CSM)where the QˉQ meson retains the quantum numbers of the QˉQ pair produced in the hard scatter.CDF measurements of prompt J/ψandψ(2S)production in Run I[17] were higher than CSM predictions by a factor of50though.This discrepancy was largely resolved by the introduction of the color octet model(COM)of quarkonium production[18].This model allows contributions from the production of QˉQ pairs in a color octet state,which evolve into color singlet states by the emission of a soft gluon.The COM also improved the agreement between the rate of J/ψproduction observed in ep collisions and predictions[19,20,21].

However,the introduction of the COM has a price:unlike the CSM,the COM predicts large values for the polarization of quarkonia states at high p T.These large polarizations have not been observed experimentally in J/ψorψ(2S)produc-tion at the Tevatron[22]or?xed target experiments.Measurements ofΥpolariza-tion tend to su?er from limited statistics and are generally not yet able to discrimi-nate signi?cantly between CSM and COM predictions for polarization.However,the NuSea collaboration?nds large polarization forΥ(2S,3S)states[23],in agreement with COM predictions.

3.2J/ψPolarization at Fixed Target

The NuSea collaboration has recently turned to the J/ψ,with a new polarization measurement of those mesons in proton–copper collisions[24].Approximately nine million J/ψ→μ+μ?candidates are selected allowing measurements of the polariza-tion to be made in several bins of x F.An average polarization of0.069±0.004±0.080

is found,which agrees with previous?xed target?ndings of very small polarization, but with substantially better accuracy.The measurement is lower than predictions based on the COM,which range from0.35to0.65.But J/ψmesons produced in decays of other particles(predicted to have small polarizations)have not been ex-cluded from this analysis,or from most of the other?xed target results.So direct comparisons with COM predictions are di?cult.

3.3J/ψProduction in ep Collisions

The ZEUS collaboration has released recent,preliminary results on the production of J/ψmesons in DIS events and their polarization in a PhP sample.While the polarization measurement has too low statistics to allow a distinction to be made between CSM and COM predictions,the DIS production measurement does have sensitivity to di?erences between the models.This measurement selects203±19 J/ψ→μ+μ?decays out of73pb?1of data and can be compared to a previously published H1result

[20]where458±30J/ψ→μ+μ?,e+e?decays were observed in 77pb?1of data.Measurements of the di?erential cross-section are shown in Figure 5for both the ZEUS and H1data.These data imply that the shape of the cross-section is better modeled by the CSM than by the COM,although errors on the predictions are still quite large.

1

2

3

4

0.40.60.8

z

1

/

V

d

V

/

d

z

500

1000

1500

2000

2500

3000

3500

4000

z

d

σ

/

d

z

(

p

b

)

Figure5:A comparison of H1and ZEUS J/ψdi?erential cross-sections in DIS with CSM and COM predictions.

3.4Heavy Quarkonium Summary

Despite recent measurements,our view of the production of heavy quarkonia states remains obscured.A COM description of the data is strongly preferred by measure-ments of J/ψproduction at the Tevatron and,to a lesser extent byΥpolarization measurements at?xed target.On the other hand,the color singlet model provides a better description of J/ψpolarization at the Tevatron and?xed target experiments as well as matching the shape of the J/ψdi?erential production cross-section in DIS events at HERA.Finally,the absolute normalization of the J/ψcross-section in DIS and PhP events is described well be neither model.As is the case with open beauty and charm production though,quarkonium measurements tend to be systematics limited and uncertainties on theoretical predictions are quite large.

4New Particles

4.1The X(3872)at the Tevatron

In the summer of2003,the Belle collaboration announced the observation of a new particle with a mass of around3872MeV in B+→K+X(3872)decays[25]. This particle,which like theψ(2S),decays to J/ψπ+π?has now been observed by several other experiments[16]including CDF[26]and D?[27].Both Tevatron experiments observe large signals,with CDF?nding730±90(11.6σ)events with a ?tted mass of3871.3±0.7±0.3MeV and D?seeing522±100(5.2σ)with a?tted mass of3871.8±3.1±3.0MeV(referenced to theψ(2S)mass).The large signal samples available to the Tevatron experiments(the original observation by Belle consisted of～35signal events)will allow detailed studies of the X(3872)to be made.D?has started this process by studying several kinematic properties,in production and decay,of their X(3872)sample,?nding that the X(3872)behaves very much like theψ(2S)within the statistics of their test.

4.2Charmed Pentaquarks?

Controversy continues to boil over the evidence for a pentaquark particle,Θ+,with a valence quark content of(uuddˉs)[28].Undeterred by this uncertainty,several groups have looked for a charmed pentaquark,Θ0c,with quark content(uuddˉc).The H1collaboration sees evidence for this particle in the decayΘ0c→D??p[29].As shown in Figure6,signi?cant signals are seen by H1in both DIS and PhP.They ?nd51±11(5.4σ)Θ0c candidates at a mass of3099±3±5MeV from a sample of ～8500D?mesons in75pb?1of data.

The primary experimental di?culty in the H1analysis is to avoid re?ec-tions from D ??→D ?πdecays,which peak in the 3100MeV region if the pion is misidenti?ed as a proton.H1avoids these re?ections by separating pions from pro-tons using dE/dx.They have performed many cross-checks to verify the reliability of this selection.

Motivated by H1’s result,ZEUS,CDF and FOCUS have conducted prelim-inary searches for the Θ0c .Despite having similar sensitivity to H1and larger samples of D ?mesons –43K,200K and 36K for ZEUS,CDF and FOCUS,respectively –none of these experiments observe any evidence for the H1signal.

10

20

30

40

M(D*p) [ GeV ]

E n t r i e s p e r 10 M e V

M (D*p) [ GeV ]

Figure 6:Evidence for the Θ0c by H1in DIS (left)and photo-production (right).4.3New Particles at SELEX

Two new particles have recently been sighted by the SELEX collaboration.Signi?-cant mass peaks for a doubly charmed baryon,Ξ+cc (3520),decaying to Λ+c K ?π+and

pD +K ?[30]and a charm-strange meson,D +sJ (2632),decaying to D +

s ηand D 0K +[31]are shown in Figure 7.The new measurement of the Ξ+cc (3520)supports a previ-ous SELEX observation of this particle,but has not been con?rmed by the FOCUS or E791collaborations.

Should the evidence for these particles hold up to further scrutiny,they

promise to provide some interesting physics.Both have rather strange properties.The decay length distribution of the Ξ+cc (3520)candidates indicates a lifetime signif-icantly shorter than expected and the relative branching ratios of the two observed

decay modes are inconsistent with phase space expectations.The two D +

sJ (2632)de-cay modes observed also show a large di?erence from phase space predictions and,even more mysteriously,the width of the particle is much narrower than expected,<17MeV at 90%C.L.

M(ccd)

E v e n t s /2.5 M e V /c

2

?M = M(K +K -π+ η) - M(K +K -π+)

0510

15

05MeV/c 2e v e n t s / 10 (M e V /c 2)

?M = M(K -π+K +) - M(K -π+)MeV/c 2

e v e n t s / 5 (M e V /c 2)

Figure 7:SELEX mass plots for the Ξ+cc (3520)(left),the D +sJ (2632)→D +

s η(middle)

and the D +

sJ (2632)→D 0K +(left).

5Where to Now?

After several years of particularly intense activity in the area of heavy quark pro-duction,our understanding of the topic has increased substantially.Problems that have plagued the comparison of b -quark data and predictions seem to have been largely resolved thanks to the e?orts of both experimentalists and theorists.

The ?eld should,by no means,slide into complacency though.Both ex-perimental systematic errors and theoretical uncertainties in beauty and charm pro-duction must be reduced before modeling of these processes can approach the level needed for understanding the next round of results from the LHC.Confusion also continues to reign in the area of heavy quarkonium production.Seemingly inconsis-tent experimental results across production and polarization measurements need to be resolved.And,as we have seen,surprising new particles,possibly pointing the way to interesting new phenomena,are waiting in the wings.

Fortunately,the future of heavy ?avor production physics looks bright.

Both the Tevatron and HERA accelerators have started new runs,which promise orders of magnitude more data than currently available,with upgraded detectors.Further down the road the LHC experiments,Atlas,CMS and LHCb,as well as BTeV at Fermilab should be able to collect heavy ?avor data sets that dwarf those foreseen from Run II at the Tevatron and HERA,allowing production studies using exclusive ?nal states.The optimism engendered by this possibility must be tempered by the knowledge that the physics goals of the upcoming experiments are not aimed primarily at heavy ?avor production.In particular,the ability to do this type of physics will be limited by the performance of trigger systems,which rely primarily on muon triggers to collect heavy ?avor data.BTeV,with its displaced vertex trigger,

is an exception,which deserves special attention here.Despite the challenges a heavy?avor production program presents,though,active e?orts have been started to study its possibilities[32].I believe that we can look forward to exciting reviews of heavy?avor production for many Physics in Collision conferences to come.

6Acknowledgments

The material presented here is the result of the sweat and toil of an army of physi-cists.Shamefully,I cannot acknowledge all who participated by name.But I bene-?ted particularly from the advice of H.Cheung,M.Corradi,R.Galik,V.Jain,R. Jesik,https://www.sodocs.net/doc/3b2115957.html,ndsberg,M.Leitch,M.zur Nedden,P.Newman,C.Paus,J.Russ,L. Silvestris,M.Smizanska,K.Stenson,A.Zieminski,D.Zieminska and A.Zoccoli.I would also like to thank the conference organizers for giving me an excuse to come to Boston and for putting together such a stimulating three days.

References

1.A.Hocker,these proceedings.

2.M.Cacciari,hep-ph/0407187.

3.B.Abbott,et al.(D?),Phys.Rev.Lett.85,5068(2000).

4.M.Cacciari,M.Greco,P.Nason,JHEP9805,007(1998).

5.J.Binnewies,B.Kniehl,G.Kramer,Phys.Rev.D58,034016(1998).

6.M.Cacciari,P.Nason Phys.Rev.Lett.89,122003(2002).

7.M.Cacciari,et al.,JHEP0407,033(2004).

8.D.Acosta,et al.(CDF),Phys.Rev.Lett.91,241804(2003).

9.M.Cacciari,P.Nason JHEP0309,006(2003).

10.I.Abt,et al.(Hera-B),Eur.Phys.J.C26,345(2003).

11.N.Kidonakis,et al.,Phys.Rev.D64,114001(2001).

12.J.M.Link,et al.(FOCUS),Phys.Lett.B581,39(2004).

13.E.Braaten,Y.Jia,M.Kusunoki and T.Mehen,hep-ph/0304280.

14.S.Chekanov,et al.(ZEUS),hep-ex/0312057.

S.Chekanov,et al.(ZEUS),hep-ex/0405069.

15.S.Chekanov,et al.(ZEUS),Phys.Rev.D69,012004(2004).

16.R.S.Galik,these proceedings.

17.F.Abe,et al.(CDF),Phys.Rev.Lett.79,572(1997).

18.E.Braaten and S.Fleming,Phys.Rev.Lett.74,3327(1995).

19.C.Adlo?,et al.(H1),Eur.Phys.J.C25,25(2002).

20.C.Adlo?,et al.(H1),Eur.Phys.J.C25,41(2002).

21.S.Chekanov,et al.(ZEUS),Eur.Phys.J.C27,173(2003).

22.T.A?older,et al.(CDF),Phys.Rev.Lett.85,2886(2000).

23.C.N.Brown,et al.(NuSea),Phys.Rev.Lett.86,2529(2001).

24.T.H.Chang,et al.(NuSea),Phys.Rev.Lett.91,211801(2003).

25.S.K.Choi,et al.(Belle),Phys.Rev.Lett.91,262001(2003).

26.D.Acosta,et al.(CDF),hep-ex/0312021.

27.V.M.Abazov,et al.(D?),hep-ex/0405004.

28.T.Nakano,these proceedings.

29.C.Aktas,et al.(H1),Phys.Lett.B588,17(2004).

30.A.Ocherashvili,et al.(SELEX),hep-ex/0406033.

31.A.V.Evdokimov,et al.(SELEX),hep-ex/0406045.

P.Cooper,these proceedings.

32.see,for example,the Hera-LHC Workshops,http://www.desy.de/～heralhc/.