Diffractive Dijet and D* Production at HERA

Transcription

1 Diffractive Dijet and D* Production at HERA XVIIth Recontre de Blois XIth International Conference on Elastic and Diffractive Scattering towards High Energy Frontiers Château de Blois, France 2/22 June 24 Yuji Yamazaki (KEK, ZEUS) On behalf of the H and ZEUS collaborations 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA

2 Introduction: why jet and HQ production? Diffractive parton densities: Extracted from F 2 D(3) (DDIS) sensitive to uarks Gluons from scaling violation Poorer constraint e'(k') e(k) ( = k - k') Q 2 β X (M X ) p p' t = (p - p') 2 β: long. momentum fraction of the parton in the exchange x P : long. momentum fraction of the exchange in the proton In the following figures, the NLO calculations used the fit from the H data (red). 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA 2

3 dσ Why jet and HQ production () sensitive to gluons de Jet and HQ productions in pqcd: cross section using factorisation * p T x P Example: dσ/de T at given x P = i x x P d ˆ σ [ dz i * ( z, µ, x de ) f ( z, µ, x Assuming the factorisation holds, the jet and HQ cross sections give better constraint to the gluon density T 2 P D i 2 P, t)] Hard scale is given by E Tjet or HQ mass e e' g(x P, z P, Q 2 ) Dijet events can reconstruct z P longitudinal momentum of the parton to the hard scattering p p' 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA 3

4 Jets in DIS Agree with NLO using H 22 fit Factorisation works if the PDFs are correct, or Data constrain PDFs if factorisation holds [pb] dσ / dz IP (jets) H Diffractive Dijets (prel.) H fit 22, µ 2 r =p 2 T, µ 2 f =4 GeV 2 H Data p T,(2) >5(4) GeV DDISENT NLO *(+δ had. ) DDISENT NLO DDISENT LO longitudinal parton momentum reconstructed by jets (jets) z IP dσ / dq 2 [pb/gev 2 ] dσ / d log [pb] 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA Cone algorithm R =., *p frame E T > 5 GeV, E T2 > 4 GeV 4 < Q 2 < 8 GeV 2,. < y <.7 x P <.5 H Diffractive Dijets (prel.) H fit 22, µ 2 r =p 2 T, µ 2 f =4 GeV H Data p T,(2) >5(4) GeV DDISENT NLO *(+δ had. ) DDISENT NLO DDISENT LO Q 2 [GeV 2 ] log dσ / dw [pb/gev] lab dσ / d<η> [pb] jets W [GeV] lab <η> jets

5 D* cross sections (open charm) H ZEUS 2[] < Q 2 < [2] GeV 2 η D* <, p TD* > 2.[] GeV ( p frame) x P <.4[.35] etc. Giving constraints to PDFs H and ZEUS consistent dσ/dp T,D* [pb/gev] H Diffractive D * H 99- (prel.) 2 NLO QCD ZEUS (rescaled) F 2 D(3),cc (β,q 2,xIP ) Q 2 = 4 GeV 2 ZEUS Q 2 = 25 GeV 2 ZEUS 98- NLO QCD ACTW fit B.3<m c <.6 GeV ACTW, fit D ACTW, fit SG =.4 = p T,D* [GeV] log(β) 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA 5

6 Why jet and HQ production (2) factorisation test Dijet cross section at TeVatron: factor 5- lower than the QCD calculation using the HERA diffractive PDFs Multi-parton scattering (re-scattering)? F D jj 4 ( β ) = β g( β ) + ( β ) 9 A.D. Martin, talk in DIS5 p H fit ZEUS fit x z P p x P colour octet filling rapidity gap p' Important for the LHC 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA 6

7 Jets in photoproduction: controlling the size of the hadron = photon Jets in photoproduction (PHP): thought to be an ideal testing ground for rescattering e large (resolved) : hadron-like small (direct) : point-like x < p e' x P x z P remnant Suppressed x is reconstructed by jets: colour octet filling rapidity gap p' x jets = OBS 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA 7 x e p e' x ~ x P z P ( E Pz) = ( E Pz) jets hadrons No photon remnant p' Unsuppressed

8 New measurement in DIS: as close to the kinematical range of PHP Common phase space: DIS measurement was restricted to.3 < y <.65 PHP: Q 2 <. GeV < η lab jet < 2 DIS: 4 < Q2 < 8 GeV 3 < η jet * < Good agreement with NLO Comparison with PHP through NLO calculation dσ/dq 2 (pb GeV -2 ) dσ/dm X (pb GeV - ) H Preliminary correl. uncert. (a) H Diffractive DIS Dijets H 22 fit (prel.) DISENT NLO*(+δ had ) DISENT NLO DISENT LO Q 2 (GeV 2 ) M X (GeV) (c) dσ/dy (pb) dσ/dx jets (pb) (b) jets x (d) y 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA 8

9 The ratio data/nlo using the same PDF Cross sections are compared through the ratio to the NLO using the same PDFs Data/NLO H Diffractive Dijets (prel.) data bin correlated uncertainty H 22 fit (prel.) DIS PHP cross section is lower (w.r.t. the NLO calculation) Resolved suppressed? look more in detail p DIS NLO.5<µ r /E*,jet <2 T -.2 W=65 GeV W=242 GeV y 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA 9

10 PHP dijet: shape comparison with LO+PS k T algorithm in lab, E T >7.5, E T2 >6.5 GeV dσ/dy (pb) dσ/de jet T (pb/gev) ZEUS 99- Energy scale uncertainty RAPGAP H Fit 2.53 resolved (GRV-G-HO) y E jet T (GeV) ZEUS dσ/d (pb) dσ/dη jet (pb) η jet dσ/dm X (pb/gev) dσ/dx obs (pb) ZEUS 99- Energy scale uncertainty RAPGAP H Fit 2.53 resolved (GRV-G-HO) M X (GeV) 5 x x obs ZEUS dσ/dz obs IP (pb) ) data / (dσ/dx obs ) MC (dσ/dx obs z obs IP data/mc x obs possible sensitivity to DPDF x flat Shape of the cross section is well described by RAPGAP 3. MC normalised to the data Data/MC flat in x : no indication of resolved suppression Some excess at highest z P : sensitivity to the diffractive PDFs 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA

11 Comparison with NLO NLO suppose to give stable prediction in normalisation absolute cross section comparison scale uncertainty in band dσ/dx obs (pb) NLO, resolved suppressed Factor calculated from the CDF-H and fit 22 NLO, no resolved comparison: R =.34 by suppression (R = ) Kaidalov et al. ZEUS NLO (R=) NLO (R=) had. H 22 fit (prel.) ZEUS (prel.) 99- Energy scale uncertainty NLO (R=.34) NLO (R=.34) had. H 22 fit (prel.) Result: flat in x Consistent with LO+PS However, the data is lower than NLO by ~.6 Both direct & resolved are suppressed PDF uncertainty? unlikely DIS described by NLO with H fit 22 ) data / (dσ/dx obs ) NLO (dσ/dx obs NLO from Klasen, Kramer x obs 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA

12 PHP comparison with NLO other variables dσ/dx jets (pb) Global suppression (both dir+res) with factor ~.5 works also for other kinematical variables R =.34 resolved-only suppression fails clearly H Preliminary correl. uncert. H Diffractive p Dijets NLO(x jets <.9) jets x (a) H 22 fit (prel.) FR NLO*(+δ had ), (x jets <.9).34 dσ/dy (pb) NLO: Frixione, Ridolfi NLO(x jets <.9) (b) y H Preliminary correl. uncert. H Diffractive p Dijets H 22 fit (prel.) FR NLO*(+δ had ).5 FR NLO.5 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA 2 dσ/dp T jet (pb GeV - ) dσ/d η jet (pb) NLO.5 lab (pb) p jet T (GeV) (a) NLO η jet (c) dσ/d η jet dσ/dm 2 (pb GeV - ) NLO η lab jet (b) NLO M 2 (GeV) Both dir+res suppressed by ~.5 Both in H and ZEUS! (d)

13 Detailed comparison in PHP: x >.75 and x <.75 with NLO ZEUS ZEUS (dσ) data / (dσ) NLO (dσ) data / (dσ) NLO y y z obs IP η jet E jet T (GeV) ZEUS (prel.) 99- Energy scale uncertainty NLO had. NLO / (NLO had.) R=, H 22 fit (prel.) x obs.75 Direct enriched Data / NLO is flat, suppressed by ~ z obs IP E jet T (GeV) Exception in high-e T for x >.75 : resolved enhanced? Could also be photon PDFs 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA η jet ZEUS (prel.) 99- Energy scale uncertainty NLO had. NLO / (NLO had.) R=, H 22 fit (prel.) x obs <.75 Resolved enriched

14 Conclusion Dijets in DIS and D* cross sections: Agree with the NLO prediction with the H 22 diffractive PDFs Factorisation holds (assuming the PDF is correct) PHP dijet cross sections are measured to investigate the puzzle of Tevatron/HERA ~ /5 / Expectation: resolved PHP is suppressed while direct is not Data agree with LO and NLO in shape, also in x But data ~half of the NLO (with the H 22 PDFs) both resolved and direct suppressed in conflict with theoretical expectation Need more ideas to understand this! Jet calculation from saturation, SCI etc... 9 May 25 Y. Yamazaki (KEK) Diffractive Dijet and D* production at HERA 4