New Photon Results from CDF Costas Vellidis Fermilab DIS 2012, Marseilles, April 22

Photon analyses at CDF Photon-related analyses have been hot topics at CDF ~30 papers published using CDF Run II data on a wide variety of photon-related topics. o Cross section measurements o Searches 4/24/13DIS 2013 – C. Vellidis2 H  γγ X  Inclusive- 

Diphoton cross sections 4/24/13DIS 2013 – C. Vellidis3 p   p _

Prompt  production in hadron colliders Born:  2 Compton+radiation  s  2 D  /q ~  /  s Fragmentation:  2 Suppressed by isolation cut “Box”: Dominant at the LHC Hard QCD (“direct”  production): colinear singularity 4/24/134DIS 2013 – C. Vellidis

Prompt  production in hadron colliders Born:  2 Compton+radiation  s  2 D  /q ~  /  s Fragmentation:  2 Suppressed by isolation cut “Box”: Dominant at the LHC Hard QCD (“direct”  production): Possible heavy resonance decays: Higgs boson colinear singularity Extra dimensions 4/24/135DIS 2013 – C. Vellidis

Identified the importance of resummation, q  fragmentation in the modeling of diphoton cross sections. 6 PRL 107 (2011) PRD 84 (2011) fb -1 4/24/13DIS 2013 – C. Vellidis Previously published results – CDF

Previously published results – D0 Sherpa describes data the best in the intermediate P T (  ) and low   regions. 7 P T1(2) >18(17) GeV/c, |η 1,2 | 0.4, E T iso <2.5 GeV 4/24/13DIS 2013 – C. Vellidis arXiv: Full Run II data set

Previously published results – ATLAS 8 JHEP 1301 (2013) 086 P T1(2) >25(22) GeV/c, |η 1,2 | 0.4 4/24/13DIS 2013 – C. Vellidis

Previously published results – CMS DIPHOX discrepancy for P T (  )>30 GeV and  ( ,  )<π/2 9 JHEP 1201 (2012) 133 4/24/13DIS 2013 – C. Vellidis

Collinear diphoton production Fragmentation – a higher-order effect o The pQCD cross section is divergent when q and  are collinear  logarithmic enhancement of the cross section o Handled with a fragmentation function – MCFM, DIPHOX o Affects low m(  ), moderate P T (  ) and low   regions Higher order subprocesses (2  3 at 1-loop and 2  4 at “tree” level) needed to describe the enhancement 104/24/13DIS 2013 – C. Vellidis

Resummation Remove singularities [P T (  )->0] by adding initial gluon radiation o RESBOS: Low-P T analytically resummed calculation (NNLL) matched to high-P T NLO o PYTHIA and SHERPA: Use parton showering to add gluon radiation in a Monte Carlo simulation framework which effectively resums the cross section (LL) o Affects low P T (  ) and   =  regions 114/24/13DIS 2013 – C. Vellidis PRD 76, (2007) or Fixed-order calculation contains singular terms at and M(  ) ≠ 0 of the form

Updated diphoton cross section measurements Use the full 9.5 fb -1 CDF run II dataset Select isolated diphoton events o Background subtraction using track isolation information Pythia evaluation of efficiency/acceptance/unfolding Compare results with new predictions 124/24/13DIS 2013 – C. Vellidis

The Tevatron and CDF Tevatron: Proton-antiproton accelerator √s = 1.96 TeV Delivered ~12 fb -1 Recorded ~10 fb -1 for each experiment CDF Collider Detector at Fermilab Tracking (large B field): ◦ Silicon tracking ◦ Wire Chamber Calorimetry: ◦ Electromagnetic (EM) ◦ Hadronic Muon system 134/24/13DIS 2013 – C. Vellidis A big thank you to Accelerator Division!

Photon identification and event selection 14  Used dedicated diphoton triggers with optimized efficiency  Photons were selected offline from EM clusters, reconstructed in a cone of radius R=0.4 in the  –  plane, and requiring: Fiducial to the central calorimeter: |  |<1.1 E T  17,15 GeV (  events) Isolated in the calorimeter: I cal = E tot (R=0.4) - E EM (R=0.4)  2 GeV Low HAD fraction: E HAD /E EM   E tot /GeV At most one track in cluster with p T trk  1 GeV/c  E T  /c Shower profile consistent with predefined patterns:  2 CES  20 Only one high energy CES cluster: E T of 2 nd CES cluster  2.4 GeV  E T γ CP2: pre-shower CES: shower maximum profile EM Cal HAD Cal Isolation cone: R=0.4 rad Imply that  R( ,  ) or  R( ,j)  0.4 4/24/13DIS 2013 – C. Vellidis

Theoretical predictions 15 PYTHIA LO parton-shower calculation – including  and  j with radiation [T. Sjöstrand et al., Comp. Phys. Comm. 135, 238 (2001)] SHERPA LO parton-shower calculation with improved matching between hard and soft physics [T. Gleisberg et al., JHEP 02, 007 (2009)] MCFM: Fixed-order NLO calculation including non-perturbative fragmentation at LO [J. M. Campbell et al., Phys. Rev. D 60, (1999)] DIPHOX: Fixed-order NLO calculation including non-perturbative fragmentation at NLO [T. Binoth et al., Phys. Rev. D 63, (2001)] RESBOS: Low-P T analytically resummed calculation matched to high-P T NLO  T. Balazs et al., Phys. Rev. D 76, (2007)  NNLO calculation with q T subtraction [L. Cieri et al., (2011)] 4/24/13DIS 2013 – C. Vellidis

Theoretical predictions 16 PYTHIA LO parton-shower calculation – including  and  j with radiation [T. Sjöstrand et al., Comp. Phys. Comm. 135, 238 (2001)] SHERPA LO parton-shower calculation with improved matching between hard and soft physics [T. Gleisberg et al., JHEP 02, 007 (2009)] MCFM: Fixed-order NLO calculation including non-perturbative fragmentation at LO [J. M. Campbell et al., Phys. Rev. D 60, (1999)] DIPHOX: Fixed-order NLO calculation including non-perturbative fragmentation at NLO [T. Binoth et al., Phys. Rev. D 63, (2001)] RESBOS: Low-P T analytically resummed calculation matched to high-P T NLO  T. Balazs et al., Phys. Rev. D 76, (2007)  NNLO calculation with q T subtraction [L. Cieri et al., (2011)] 4/24/13DIS 2013 – C. Vellidis Integrated cross section (pb) Data (CDF)12.3 ± 0.2 stat ± 3.5 syst RESBOS11.3 DIPHOX10.6 MCFM11.5 SHERPA12.4 PYTHIA  +  j 9.2 NNLO11.8

m(  ) Good agreement between data and theory for M  >30 GeV/c 2 except PYTHIA  174/24/13DIS 2013 – C. Vellidis

P T (  ) 184/24/13DIS 2013 – C. Vellidis

P T (  ) - ratios RESBOS agrees with low P T (  ) data the best SHERPA agrees with low P T (  ) data well NNLO and SHERPA describe the “shoulder” of the data at P T (  ) = 20 – 50 GeV/c (the “Guillet shoulder”) 19 NB: Vertical axis scales are not the same DIPHOX RESBOS PYTHIANNLO MCFM SHERPA 4/24/13DIS 2013 – C. Vellidis

 (  ) 204/24/13DIS 2013 – C. Vellidis

 (  )- ratios RESBOS and SHERPA describe  (  ) =  region Fixed order calculations do not describe  (  ) =  region NNLO describes  (  ) =  region 21 NB: Vertical axis scales are not the same DIPHOX RESBOS PYTHIANNLO MCFM SHERPA 4/24/13DIS 2013 – C. Vellidis

Summary of diphoton cross sections High precision  cross sections are measured using the full CDF Run II dataset The data are compared with all state-of-the-art calculations The SHERPA calculation, overall, provides good description of the data, but still low in regions sensitive to nearly collinear  emission (very low mass, very low Δ ϕ ) The RESBOS calculation provides the best description of the data at low P T and large Δ ϕ, where resummation is important, but fails in regions sensitive to nearly collinear  emission The NNLO calculation provides the best description of the data at low Δ ϕ, but still not very good at very low mass and at high P T More in PRL 110, (2013) (supplemental material online) 224/24/13DIS 2013 – C. Vellidis

Photon+heavy flavor (b/c) cross sections 4/24/13DIS 2013 – C. Vellidis23 p  p _ b-jet

 +b/c+X production Photon produced in association with heavy quarks provides valuable information about heavy flavor excitation in hadron collisions o LO contribution: Compton scattering (Qg  Q  ) dominates at low photon p T o NLO contribution: annihilation (qq  QQ  ) dominates at high photon p T 24 Compton scattering ~  S Annihilation ~  S 2 Q g q q - - 4/24/13DIS 2013 – C. Vellidis -

Previous results – D0 25 PRL 102, (2009) − 1 fb -1 Good agreement for  +b+X Discrepancy for  +c+X PLB 714, 32 (2012) – 8.7 fb -1  +b+X PLB 719, 354 (2013) – 8.7 fb -1 Discrepancies in both channels.  +c+X 4/24/13DIS 2013 – C. Vellidis

Previous results – CDF Measure low p T cross section using a special trigger  +b+X agrees with NLO up to 70 GeV 26 CDF: PRD 81, (2010) pb -1 4/24/13DIS 2013 – C. Vellidis

Analysis overview Measure  +b/c+X cross section using 9.1 fb -1 inclusive photon data collected with CDF II detector Use ANN (artificial neural network) to select photon candidates o Fit ANN distribution to signal/background templates to get photon fraction Use SecVtx b-tag to select heavy-flavor jets o Fit secondary vertex invariant mass to get light/c/b quark fractions Use Sherpa MC to get efficiency/unfolding factor o Photon ID efficiency, b-tagging efficiency, detector acceptance and smearing effects Cross section o 274/24/13DIS 2013 – C. Vellidis

4 theoretical predictions NLO – direct-photon subprocesses and fragmentation subprocesses at O(  s 2 ), CTEQ6.6M PDFs [T.P. Stavreva and J.F. Owens, PRD 79, (2009)] k T -factorization – off-shell amplitudes integrated over k T - dependent parton distributions, MSTW2008 PDFs [A.V. Lipatov et al., JHEP 05, 104 (2012)] Sherpa – tree-level matrix element (ME) diagrams with one photon and up to three jets, merged with parton shower, CT10 PDFs [T. Gleisberg et al., JHEP 02, 007 (2009)] Pythia – ME subprocesses: gQ   Q, qq   g followed by gluon splitting: g  QQ, CTEQ5L PDFs [T. Sjöstrand et al., JHEP 05, 026 (2006)] 28 _ _ 4/24/13DIS 2013 – C. Vellidis

 +b+X cross sections 29 NLO fails to describe data at large photon Et – perhaps gluon splitting is treated at LO k T -factorization and Sherpa agree with data reasonably well Pythia with doubled gluon splitting rate to heavy flavor describes the shape 4/24/13DIS 2013 – C. Vellidis NB: Vertical axis scales are not the same

 +c+X cross sections 304/24/13DIS 2013 – C. Vellidis NLO fails to describe data at large photon Et – perhaps gluon splitting is treated at LO k T -factorization and Sherpa agree with data reasonably well Pythia with doubled gluon splitting rate to heavy flavor describes the shape NB: Vertical axis scales are not the same

Summary of photon+b/c cross sections High precision  +b/c cross sections are measured using the full CDF Run II dataset The data are compared with parton shower, fixed-order and kt- factorization calculations NLO does not reproduce data most likely because of its limitation in modeling gluon splitting rates. k T -factorization and Sherpa agree with data reasonably well Pythia with doubled gluon splitting rates to heavy flavor describes the data shape 314/24/13DIS 2013 – C. Vellidis

Conclusions The CDF experiment has produced a wealth of QCD physics results and analysis techniques, which is a legacy for the current and future high energy physics experiments We have achieved an unprecedented level of precision for many photon-related observables Those results provide valuable information to the HEP community, e.g. the diphoton results can help the precision measurements of H boson in the  channel. … and we are not done yet!! 4/24/13DIS 2013 – C. Vellidis32

4/24/13DIS 2013 – C. Vellidis33

Interesting kinematic variables o Search for resonances. o Sensitive to activity in the event. o Sensitive to production mechanism. 4/24/13DIS 2013 – C. Vellidis34 PT1PT1 PT2PT2  =0   p p _ 

Interesting kinematic variables o Search for resonances. o Sensitive to activity in the event. o Sensitive to production mechanism. Fragmentation/higher order diagrams o Two  ’s go almost collinear o Low m(  ), intermediate P T (  ), low  (  ) Resummation o Low P T (  ), high  (  ) 4/24/13DIS 2013 – C. Vellidis35  =0   p p _  Special case  =0   p p _ PT1PT1 PT2PT2

Background subtraction using track isolation Sensitive only to underlying event and jet fragmentation (for fake  ) Immune to multiple interactions (due to z-cut) and calorimeter leakage Good resolution in low-E T region, where background is most important Uses charged particles only 36 Signal: direct diphotons Background: jets misidentified as photons – j  jj Signal Probability (I trk <1 GeV) 4/24/13DIS 2013 – C. Vellidis Background Probability (I trk <1 GeV)

Background subtraction For a single , a weight can be defined to characterize it as signal or background: o  = 1 (0) if I trk  (  ) 1 GeV/c o  s = signal probability for I trk  1 GeV/c o  b = background probability for I trk  1 GeV/c For , use the track isolation cut for each photon to compute a per-event weight under the different hypotheses ( ,  +jet and dijet): 37 Both photons fail Leading fail, trailing passes Leading passes, trailing fails Both photons pass e.g. leading passes/trailing fails 4/24/13DIS 2013 – C. Vellidis Transfer matrix Function of  s and  b

Average  40% Better at high mass: o 60-80% for m(  )  GeV/c 2 o  80% for m(  )>150 GeV/c 2 Better at high P T (  ): o  70% for P T (  ) >100 GeV/c 15-30% sys. errors 384/24/13DIS 2013 – C. Vellidis Signal fractions

Efficiency×Acceptance Estimated using detector- and trigger-simulated and reconstructed PYTHIA events Procedure iterated to match PYTHIA kinematics to the data 39 Uncertainties in the efficiency estimation: 3% from material uncertainty 1.5% from the EM energy scale 3% from trigger efficiency uncertainty 6% (3% per photon) from underlying event (UE) correction Total systematic uncertainty: ~7-15% 4/24/13DIS 2013 – C. Vellidis

Experimental systematic uncertainties Total systematic uncertainty  15-30%, smoothly varying with the kinematic variables considered Main source is background subtraction, followed by overall normalization (efficiencies: 7%; integrated luminosity: 6%; UE correction: 6%) 404/24/13DIS 2013 – C. Vellidis

4/24/13DIS 2013 – C. Vellidis41 Integrated cross section (pb) Data (CDF)12.3 ± 0.2 stat ± 3.5 syst RESBOS11.3 ± 2.4 DIPHOX10.6 ± 0.6 MCFM11.5 ± 0.3 SHERPA12.4 ± 4.4 PYTHIA gg+gj9.2 NNLO – 0.6

Comparison with D0 4/24/13DIS 2013 – C. Vellidis42

43 A closer look at fragmentation: DIPHOX isolation study iso < 2 GeV Fragmentation strength is missing from the DIPHOX calculation possibly because of the approximate application of the isolation requirement at the parton level 4/24/13DIS 2013 – C. Vellidis

44 Total Direct 1-frag 2-frag E T iso < 2 GeV E T iso < 10 GeV A closer look at fragmentation: DIPHOX isolation study 4/24/13DIS 2013 – C. Vellidis

Event selection Use inclusive photon trigger to select photon events o Trigger efficiency is approximately 100% for  E T >30 GeV Interaction vertex in the fiducial region Photon candidate must pass a neural-net based photon ID o ANN>0.75 o |  |<1.05, 30<E T <300 GeV, divided into 8 E T bins Jets are reconstructed with JetClu cone size 0.4 and must be positively tagged. o |  | 20 GeV  R( ,jet)> /24/13DIS 2013 – C. Vellidis

ANN photon ID Trained with TMVA (Toolkit for Multivariate Data Analysis) 7 input variables to take into account difference between  and  0 /  isolation (2), lateral shower shape (3), Had/Em, CES/CEM ANN ID improves signal efficiency by 9% at the same background rejection compared with the standard cut-based ID. Use MC with full detector simulation to get templates o Signal – prompt photons o Background – jets with prompt photons removed 46 prompt photons  0,  4/24/13DIS 2013 – C. Vellidis

True photon fraction Fit data ANN distribution using signal and background templates to get true photon fraction 474/24/13DIS 2013 – C. Vellidis

True photon fraction (continued) Systematics o Photon energy scale o Vary inputs to photon ID ANN according to their uncertainties o Vary Photon ID ANN template binning to test sensitivity to shapes o 6% at low E T, 2% at high E T. 484/24/13DIS 2013 – C. Vellidis

Standard b-jet identification B-hadrons are long-lived – search for displaced vertices Fit displaced tracks and cut on L xy significance (σ ~ 200  m) Charm hadrons have similar tag behavior but lower efficiency Use “tag mass” to deduce the flavor composition of a sample of tagged jets o Mass of the tracks forming the secondary vertex o B-hadrons are heavy: will have higher m tag spectrum than charm or light jet fakes 494/24/13DIS 2013 – C. Vellidis

Light/c/b-jet fractions Fit data secondary vertex mass using MC templates Shape of secondary vertex mass for event with fake photon is taken from di-jet data 504/24/13DIS 2013 – C. Vellidis

Light/c/b-jet fractions (continued) Results from fitter. 514/24/13DIS 2013 – C. Vellidis

Systematics on b/c-jet fractions Jet energy scale: affects acceptance Uncertainty in tracking efficiency: scale secondary vertex mass templates by ±3% o Dominant systematic effect Difference between single-quark and di-quark jets Total systematic error is ~20% 524/24/13DIS 2013 – C. Vellidis

Efficiency×Acceptance Use Sherpa MC to unfold photon ID efficiency, b-tagging efficiency, detector acceptance and smearing effects. Systematic effects evaluated o photon energy scale and ID o jet energy scale o b-tagging efficiency o Generator o PDF 534/24/13DIS 2013 – C. Vellidis