Full Jet-Reconstruction in Heavy-Ion Collisions at RHIC and LHC

Size: px

Start display at page:

Download "Full Jet-Reconstruction in Heavy-Ion Collisions at RHIC and LHC"

Joanna Copeland
5 years ago
Views:

1 Full Jet-Reconstruction in Heavy-Ion Collisions at RHIC and LHC And is there a consistent jet-quenching picture emerging at RHIC? Jörn Putschke (Yale University)

2 single particle time di-hadron 0-% control-exp. no trigger 3-particle corrl. full jet reco. 0.<pt,assoc<.0 GeV SAR preliminary SAR preliminary pt per grid cell Open symbols p+p jet-hadron η ϕ

3 single particle di-hadron 0-% control-exp. no trigger 3-particle corrl. Understanding/consistency? full jet reco. 0.<pt,assoc<.0 GeV SAR preliminary SAR preliminary pt per grid cell Phenix preliminary Open symbols p+p jet-hadron η ϕ

4 Outline Short intro: Jets and Jet-Finding Algorithms Jets as a calibrated probe: p+p and d+au reference Our new tool: Jets in heavy-ion collisions (RHIC) Background in heavy-ion collisions: fake-jets and fluctuations Inclusive jet spectrum and jet RAA Jet energy profile (R=0./0.4) Di-Jet coincidence measurements More this week: Y. Lai and M. Ploskon Consistency? Connection to single/di-hadron results!? Outlook: Jets at the LHC Summary More this week: M. McCumber More this week: A. Morsch 3

Goal: re-associate (measurable) hadrons to accurately reconstruct partonic kinematics pqcd calculates partons

5 Jets connect theory and experiment Jets are the experimental signatures of quarks and gluons. hey reflect the kinematics and topology of partons. Goal: re-associate (measurable) hadrons to accurately reconstruct partonic kinematics pqcd calculates partons experiments measure fragments of partons: hadrons ool: Jet-finding algorithms: Apply same algorithm to data and theoretical calculations pqcd factorization/jet spectrum: PDF Partonic x-section 4

6 Jet definition Jet algorithm he construction of a jet is unavoidably ambiguous. On at least two fronts: which particles get put together into a common jet? How do you combine their momenta? 5

7 Jet definition Jet algorithm he construction of a jet is unavoidably ambiguous. On at least two fronts: Jet Definition which particles get put together into a common jet? How do you combine their momenta? {pi} {jk} particles 4-momenta, calorimeter towers jet algorithm + parameters (at least the cone radius/ resolution parameter R) + recombination scheme jets 5

8 Jet definition Jet algorithm he construction of a jet is unavoidably ambiguous. On at least two fronts: which particles get put together into a common jet? How do you combine their momenta? Sequential Recombination bottom-up successively undoes QCD branching Modern Jet Finder Algorithms Cone top-down centred around idea of an ʻinvariantʼ, directed energy flow k algorithm anti-k algorithm Cambidge-Aachen algorithm CDF JetClu CDF MidPoint D0 (run II) Cone Gaussian Filter ALAS Cone PyCell/CellJet GetJet SISCone CMS Iterative Cone 5

9 Further conditions a jet algorithm has to fulfill... Snowmass Accord (990): Recent Additions (> 000): Physical results independent of your choice of jet definition Jets should be invariant with respect to certain modifications of the event: collinear splitting infrared emission Jet-finding algorithms discussed (fulfilling the requirements): - k/anti-k (recombination algorithms) (- SISCone (infrared, seedless cone algorithm)) - Gaussian Filter (PHENIX) } Known as infrared and collinear safety Y. S. Lai, B. Cole, arxiv: } M. Cacciari, G.P. Salam, G. Soyez, JHEP 04, 005 (008), Jörn Putschke 6

/π dσ/(dηdp) [pb/gev] data / theory 8 7 6 5 4 3.8.4.0 0.6 0. 0 c 3 ) ) (mb GeV 3!

10 /π dσ/(dηdp) [pb/gev] data / theory c 3 ) ) (mb GeV 3!/(dp 3 d E p+p reference: inclusive jet cross-section SAR, PRL 97 (006), 500 Jets are unavoidable at hadron (a) colliders, e.g. fromsar parton scattering s=00 p+p g jet + X GeV p Combined MB Combined H NLO QCD (Vogelsang) midpoint-cone r cone=0.4 0.<η<0.8 Systematic Uncertainty heory Scale Uncertainty jet (b) p [GeV/c] PHENIX Preliminary 3 Run 5 p + p s = 00 GeV/c 4 Gaussian filter,! = width not a p uncertainty 6 jet RHIC PHENIX,!=0.3 PYHIA, K =.5,!=0.3 NLO SCA, R = 0.3 (Vogelsang) SAR H, R=0.4 (PRL 97, 500) (GeV/c) p p Jets from scattering of partons [nb/(gev/c)] JE dp JE! / dy d D= JE p K JE -6.6< y <. (" ) - CDF data ( L =.0 fb ) Systematic uncertainties NLO: JERAD CEQ6.M corrected to hadron level JE µ R = µ F = max p / = µ 0 PDF uncertainties JE 0.7< y <. JE -3.< y <.6 (" ) JE 6 y <0. (" ) JE 3 0.< y <0.7 (" ) [GeV/c] Jet cross section: data and theory agree over many orders of magnitude probe of underlying interaction RHIC evatron ALAS performance thus LHC!" between two highest p jets in a event shows expected back-to-back structure p+p inclusive jet cross-section described over orders of magnitude and at energies ranging from 00 GeV to 7 ev by NLO/ Pythia pqcd calculations! Aaron Angerami, Heavy-ion physics with the ALAS detector (5/5/) R g 7

11 p+p reference: fragmentation functions (focus on RHIC) H. Caines (SAR), QM09 /N jet dn/d! /N jet dn/d!.8 SISCone K.6 Anti-K ! (=ln(p jet SISCone K Anti-K ! (=ln(p jet A. immins (SAR), SQM09 Prelim inary Pythia 6.4 p-p data Pythia 6.4 p-p data /p Part /p Part ) ) R=0.4 /N jet dn/d! /N jet dn/d! SISCone K Anti-K Pythia 6.4 p-p data ! (=ln(p jet 00. GeV/c Data not corrected to particle level. PYHIA = PYHIA+ GEAN R=0.7.5 SISCone K Anti-K Prelim inary Prelim inary Pythia 6.4 p-p data /p Part ! (=ln(p jet /p Part ) ) /dz c(pjet ) N jet dn charged PHENIX Preliminary Run-5 p + p s = 00GeV Gaussian filter σ = 0.3 charged ratio < 0.9 > 3 particle 8.0 GeV/c 5.4 GeV/c 3. GeV/c. GeV/c PYHIA p jet = 5 GeV/c data overall syst. uncertainty D(z) =Nz α ( z) β γ (+ z) fit to data z Figure 5. PHENIX Run-5 p + p at s = 00 GeV charged (with electron rejection) jet fragmentation function with respect to z = w filter p /p jet and vertically scaled Fragmentation function c(p measurements jet by ) = k,k = 0,..., 3. he shaded box to the left indicates the overall normalization systematic uncertainty, shaded boxes associated with data points indicate point-to-point systematic uncertainties, and error bars indicate statistical uncertainties. Biases from the jet level cuts are fully at RHIC in agreement with Pythia! 8

12 Cold nuclear matter effects: k broadening in d+au J. Kapitan (SAR), ISMD 09 P/$^$./$5QR: "#"$"#('GF( $#%&$"#('GF( k=p,*sin(δϕ) '(%)*+,-./0/,3.//$5Q./$^$S/$5QR: Measure σk via di-jets in d+au (p+p): p+p : σk,raw=.8 ± 0.GeV/c d+au (0-0%): σk,raw=3.0 ± 0.GeV/c No significant cold nuclear effect on k broadening observed Remark: Quantitative consistency with estimates based on di-hadron measurements from PHENIX (and SAR) 9

13 Full-Jet reconstruction in HI collisions p+p JP trigger ~ GeV SAR preliminary pt per grid cell [GeV] η ϕ

14 Full-Jet reconstruction in HI collisions p+p JP trigger ~ GeV ~ GeV SAR preliminary SAR preliminary pt per grid cell [GeV] pt per grid cell [GeV] η ϕ η ϕ

15 Full-Jet reconstruction in HI collisions SAR preliminary ~ GeV Phenix preliminary pt per grid cell [GeV] η ϕ Full jet reconstruction in HI collisions is a challenge due to the underlying background!

16 A word of caution (especially in HI): Jet Definition Jet Algorithm Y. Lai QM009 R AA PHENIX Preliminary 0 0% # 0 0 %, "z! = 0.7 (PRL, 630) Run 5 Cu + Cu s NN = 00 GeV Gaussian filter, $ = (GeV/c) rec pp p A jet is what you ask for! Jet-finder based on (unmodified) jet-shapes/jet core veto against modified/quenched jets!? p cut to minimize background bias towards non-interacting jets!(?) Anti-quenching biases are everywhere! S. Salur HP08 (")*#!+,-./.0+# '"# *M4HN# )%3)%#45647########!#895"+.:# #!###; <.0 #(=-,>#?3?# (")*#!+,-./.0+# '"# )%3)%#45647########!#895"+.:# #!###; <.0 #(=-,>#?3?# here are no shortcuts! We have to deal with the full complexity of the heavy-ion background! Good news: he tools are available and our understanding is improving!

he challenge: Heavy-ion Background ρ (GeV/area) A jet in HI collisions schematically: p (Jet Measured) = p (Jet) + HI Bkg. ± F(A) SAR Preliminary hree main components:.

17 he challenge: Heavy-ion Background ρ (GeV/area) A jet in HI collisions schematically: p (Jet Measured) = p (Jet) + HI Bkg. ± F(A) SAR Preliminary hree main components:. HI background: for example determine energy density per unit area ρ (event-by-event) with A the jet area (determined by FastJet algorithm) ρ A ~ 45 GeV for RC=0.4 (S/B ~0.5 for 0 GeV jet). Fake jets = signal in excess (due to jet clustering) of background model from random association of uncorrelated soft particles (i.e. not due to hard scattering) 3. Background fluctuations: A priori unknown background fluctuation distribution F(A). In a gaussian (random area) approximation: ~ 6-7 GeV for RC=0.4 Background fluctuations [Gev] Multiplicity SAR Preliminary Rc

18 Fake-Jet contribution Fake jets: signal in excess of background model from random association of uncorrelated soft particles (i.e. not due to hard scattering) Inclusive jet spectrum (SAR): Spectrum of jets after randomizing HI event in ϕ and removing leading jet particle More details: M. Ploskon Di-Jet / Fragmentation function (SAR): Background di-jet rate = Fake + Additional Hard Scattering Estimated using jet spectrum at 90 deg. SAR Preliminary rigger jet p > GeV p cut,particle = 0. GeV Au+Au H 0-0% SAR Preliminary PHENIX (gaussian filter): Gaussian fake-jet rejection; use overall shape of jets for discrimination Caveat: If quenching distorts jet-shape substantially, danger of vetoing quenched jets! More details: Y. Lai dn/d"! N jet 3.5 PHENIX Preliminary g > 4.9 (GeV/c).5 Run 5 Cu + Cu 0 0% s NN = 00 GeV/c Gaussian filter, # = 7.5 (GeV/c) >.5 (GeV/c) > 7.8 (GeV/c) g > 7.4 (GeV/c) "! 3

Fake-Jet contribution Fake jets: signal in excess of background model from random association of uncorrelated soft particles (i.e. not due to hard scattering) Inclusive jet spectrum (SAR): Spectrum of jets after randomizing HI event in ϕ and removing leading jet particle More details: M.

19 Fake-Jet contribution Fake jets: signal in excess of background model from random association of uncorrelated soft particles (i.e. not due to hard scattering) Inclusive jet spectrum (SAR): Spectrum of jets after randomizing HI event in ϕ and removing leading jet particle More details: M. Ploskon Di-Jet / Fragmentation function (SAR): Background di-jet rate = Fake + Additional Hard Scattering Estimated using jet spectrum at 90 deg. SAR Preliminary rigger jet p > GeV p cut,particle = 0. GeV Au+Au H 0-0% SAR Preliminary PHENIX (gaussian filter): Gaussian fake-jet rejection; use overall shape of jets for discrimination Caveat: If quenching distorts jet-shape substantially, danger of vetoing quenched jets! More details: Y. Lai Conceptual difference between SAR and PHENIX! (Quantitative difference!?) dn/d"! N jet 3.5 PHENIX Preliminary g > 4.9 (GeV/c) Run 5 Cu + Cu 0 0% s NN = 00 GeV/c Gaussian filter, # = 7.5 (GeV/c) >.5 (GeV/c) > 7.8 (GeV/c) g > 7.4 (GeV/c) 0. 3

20 Background fluctuations!"#$%&'!"#$%&'()*&+*,' Jet spectrum in Au+Au (schematically):!"#$%&'-./0,*,' dσ AA dp t = dσ pp dp t F (A, p t ) Effect of background fluctuations F(A,pt) substantial feed-up in the jet x-section Generalized probe embedding (conceptually the same in SAR and PHENIX) Systematic extension of random region-to-region fluctuation estimate. Embed probes (single particles, pythia jets, p+p jets...) into Au+Au/Cu+Cu events and measure the fluctuations spectrum (used for unfolding). akes the effect of clustering/jet-finding algorithms into account; conceptually higher bound for fluctuations (diluted due to random embedding; has to be estimated; and what about v!?) Statistical description (strictly lower bound, in context of estimating systematics) Conceptually F(A,pt) for stat. independent thermal (exp.) particle emission : F (A, p t )=P oisson((m(a)) Γ(M(A), p t ) More details/first data comparison: E. Bruna (SAR), AGS Users Meeting 0 4

21 Background fluctuations!"#$%&'!"#$%&'()*&+*,' Jet spectrum in Au+Au (schematically):!"#$%&'-./0,*,' dσ AA = dσ pp Fakeʼs and background F (A, p t ) dp t fluctuation dp t corrections are work in progress and will be discussed in more detail during this meeting! F (A, p t )=P oisson((m(a)) Γ(M(A), p t ) Effect of background fluctuations F(A,pt) substantial feed-up in the jet x-section (M. Ploskon, Y. Lai and A. Morsch) We have to discuss the conceptual differences, biases (concerning fakes ) and their quantitative effects! Generalized probe embedding (conceptually the same in SAR and PHENIX) Systematic extension of random region-to-region fluctuation estimate. Embed probes (single particles, pythia jets, p+p jets...) into Au+Au/Cu+Cu events and measure the fluctuations spectrum (used for unfolding). akes the effect of clustering/jet-finding algorithms systematic into account; errors conceptually associated higher bound to for these fluctuations corrections! (diluted due to random embedding; has to be estimated; and what about v!?) But,we also have to discuss how to estimate conservative Statistical description (strictly lower bound, in context of estimating systematics) Conceptually F(A,pt) for stat. independent thermal (exp.) particle emission : More details/first data comparison: E. Bruna (SAR), AGS Users Meeting 0 4

22 Example SAR: Spectrum Unfolding Corrections for smearing of jet p t due to HI bkg. nonuniformities: M.Ploskon QM009 ) raw spectrum ) removal of fake - correlations 3) unfolding (bayesian) of HI bkg. fluctuations 4) correction for p resolution SAR Preliminary 5

23 What do we learn from the Au+Au/Cu+Cu jet spectrum? Energy shi7? p+p Absorp)on? Au+Au Cross sec)on ra)o AuAu/pp Momentum and energy is conserved even for quenched jets If full jet reconstruction in heavy-ion collisions are unbiased Inclusive jet spectrum scales with Nbinary relative to p+p Caveat: Initial state nuclear effects at large x; EMC effect can be measured in d+au 6

24 Inclusive jet x-section in heavy-ion collisions ) dy) ((GeV/c) dp dn/(p N evt (!) Cu + Cu 0 0% 0 40% 40 60% 60 80% Y. Lai QM009 p + p" AB 0 0% 0 40% 40 60% 60 80% PHENIX Preliminary Run 5 Cu + Cu s NN = 00 GeV/c Gaussian filter, # = 0.3 uncorrected p + p compared to background unfolded Cu + Cu (GeV/c) rec pp p Au+Au collisions 0-% lines=unfolding uncertainties SAR Preliminary M.Ploskon QM009 (Yue Shi Lai, for the PHENIX Collaboration) APS DPF Meeting 009, Heavy Ions III 9 / 35 Inclusive Jet spectrum measured in central Au+Au and Cu+Cu collisions at RHIC Extended the kinematical reach to study jet quenching phenomena to jet energies > 40 GeV Remark: New high statistics Au+Au runs on tape (Phenix and SAR) will increase significantly the kinematic reach! 7

25 Jet RAA in central Au+Au and Cu+Cu M.Ploskon QM009 Y. Lai QM009 Jet R AA - Au+Au and p+p at Au+Au: % most central s NN =00 GeV/c kt R=0.4 anti-kt R=0.4 Inclusive RAA SAR Preliminary Jet p (GeV/c) R AA PHENIX Preliminary.8 0 0% # 0 0 %, "z! Run 5 Cu + Cu s NN = 00 GeV Gaussian filter, $ = 0.3 = 0.7 (PRL, 630) (GeV/c) rec pp p SAR sees a substantial fraction of jets in Au+Au - in contrast to x5 suppression for light hadron RAA Strong suppression (similar to single particle) in Cu+Cu measured by PHENIX 8

26 First look at the jet energy profile M.Ploskon QM009 Yield Ratio: R=0./R= Au+Au and p+p at Au+Au: % most central SAR Preliminary s NN =00 GeV/c Au+Au kt Au+Au anti-kt p+p kt p+p anti-kt R=0.4 R=0. p+p: Narrowing of the jet structure with increasing jet energy Au+Au: Deficit of jet energy of jets reconstructed with R= Jet p (GeV/c) Strong evidence of broadening in the jet energy profile 9

27 Recoil jet spectrum RAA E. Bruna QM009 rigger jet Recoil jet SAR Preliminary Selecting biased trigger jet maximizes pathlength for the back-to-back jets: extreme selection of jet population Significant suppression in di-jet coincidence measurements! 0

28 Recoil Fragmentation Function in Au+Au collisions large uncertainties due to background (further systematic evaluation needed) E. Bruna QM009 AntiKt R=0.4 pt,rec(auau)>5 pt,rec(auau)>5 GeV GeV < < pt,rec(pp)> pt,rec(pp)> ~ ~ 5 5 GeV GeV SAR SAR Preliminary Jet Fragmentation process Hard scatter Small/ no modification in the fragmentation function for jet pt < pt,rec(pp)> ~ 5 GeV at high z!

29 Di-jet azimuthal correlation in Cu+Cu Jet Δϕ Recoil jet Y. S. Lai, arxiv: v Figure 5: Comparison between the PHENIX Run-5 Cu + Cu at 3.5 s NN = 00 GeV R AA derived from unfoldi symbols) and embedding (open symbols). he shaded box to the left indicates the p + p Cu + Cu systematic un PHENIX Preliminary in the jet energy scale, shaded boxes to the right shows centrality 60 80% dependent systematic uncertainty between em and unfolding, Run 5 Cu+Cu s = 00 GeV/c 3 shaded boxes associated NN with data points indicate 40 60% point-to-point systematic uncertainties, and e indicate statistical Gaussian uncertainties. filter, # = 0.3 Note that the flatness of R AA makes a comparison across different energy scales 0 40% symmetric jet jet.5 rec 0 0% 7.5 < p <.5 GeV/c Centrality Width dn/d"! area normalized to "! Small k broadening of surviving parton in Cu+Cu Figure 6: Comparison between the central PHENIX Run-5 Cu + Cu at s NN = 00 GeV jet R AA derived from unfolding and the π 0 R AA. he shaded box to the left indicates the p + p Cu + Cu systematic uncertainty in the jet energy scale, shaded boxes to the right shows centrality dependent systematic uncertainty between embedding and unfolding, shaded boxes associated with data points indicate point-to-point systematic uncertainties, and error bars indicate statistical uncertainties. Note that while the flatness of R AA makes a comparison across different energy scales possible, π 0 with z =0.7 has a different energy scale. 0 0% 0.3 ± % 0.3 ± % 0.60 ± % 0.53 ± able I Widths of Gaussian fit to the PHENI Cu + Cu at s NN = 00 GeV azimuthal angular tion for jets with 7.5 GeV/c < p CuCu <.5 GeV therefore allows a comparison. We observe a R AA that becomes gradua pressed with increasing centrality. he level pression in the most central 0% centraliti R AA and comparable to that of π Cu + Cu jet-jet azimuthal correl he Cu + Cu jet-jet azimuthal correlatio tracted by correcting for the acceptance effe the area-normalized mixed event yield (e.g. [ dn( φ) d φ = dn raw ( φ) A( φ) d φ

30 Di-jet azimuthal correlation in Cu+Cu Jet Δϕ Recoil jet Y. S. Lai, arxiv: v Figure 5: Comparison between the PHENIX Run-5 Cu + Cu at 3.5 s NN = 00 GeV R AA derived from unfoldi symbols) and embedding (open symbols). he shaded box to the left indicates the p + p Cu + Cu systematic un PHENIX Preliminary in the jet energy scale, shaded boxes to the right shows centrality 60 80% dependent systematic uncertainty between em and unfolding, Run 5 Cu+Cu s = 00 GeV/c 3 shaded boxes associated NN with data points indicate 40 60% point-to-point systematic uncertainties, and e indicate statistical Gaussian uncertainties. filter, # = 0.3 Note that the flatness of R AA makes a comparison across different energy scales 0 40% symmetric jet jet.5 rec 0 0% 7.5 < p <.5 GeV/c Centrality Width dn/d"! area normalized to "! Small k broadening of surviving parton in Cu+Cu Figure 6: Comparison between the central PHENIX Run-5 Cu + Cu at s NN = 00 GeV jet R AA derived from unfolding and the π 0 R AA. he shaded box to the left indicates the p + p Cu + Cu systematic uncertainty in the jet energy scale, shaded boxes to the right shows centrality dependent systematic uncertainty between embedding and unfolding, shaded boxes associated with data points indicate point-to-point systematic uncertainties, and error bars indicate statistical uncertainties. Note that while the flatness of R AA makes a comparison across different energy scales possible, π 0 with z =0.7 has a different energy scale. Are we biasing our (di-)jet measurements towards non-interacting jets? Or is our HI jet energy underestimated due to jet broadening!? 0 0% 0.3 ± % 0.3 ± % 0.60 ± % 0.53 ± able I Widths of Gaussian fit to the PHENI Cu + Cu at s NN = 00 GeV azimuthal angular tion for jets with 7.5 GeV/c < p CuCu <.5 GeV therefore allows a comparison. We observe a R AA that becomes gradua pressed with increasing centrality. he level pression in the most central 0% centraliti R AA and comparable to that of π Cu + Cu jet-jet azimuthal correl he Cu + Cu jet-jet azimuthal correlatio tracted by correcting for the acceptance effe the area-normalized mixed event yield (e.g. [ Can we test this with an independent measurement!? dn( φ) d φ = A( φ) dn raw ( φ) d φ

31 Jet-Hadron correlations (JH) 0-0% Au+Au High ower rigger (H): tower 0.05x0.05 (ηxϕ) with Et> 5.4 GeV 0.<pt,assoc<.0 GeV.0<pt,assoc<.5 GeV Jet axis /NJet dn/dδϕ SAR Preliminary 0-0% Au+Au /NJet dn/dδϕ SAR Preliminary 0-0% Au+Au As an example: gaus+ constant bkg. fit rigger jet Δϕ Δϕ Assoc. pt,assoc>.5 GeV Δϕ=ϕJet ϕassoc. /NJet dn/dδϕ SAR Preliminary 0-0% Au+Au ϕjet = H trigger jet-axis found by Anti-kt with R=0.4, pt,cut> GeV and pt,rec(jet)>0 () GeV No Δη triangle acceptance applied (in progress...) Corrected for single particle tracking efficiencies rigger jet energy: correction for tracking eff. (p+p vs. Au+Au), background fluctuations ~ GeV; uncertainties: p+p=au+au and + GeV in jet energy (p+p relative to Au+Au) Δϕ 3

32 Jet-Hadron correlations (JH) 0-0% Au+Au High ower rigger (H): tower 0.05x0.05 (ηxϕ) with Et> 5.4 GeV 0.<pt,assoc<.0 GeV GeV.0<pt,assoc<.5 GeV GeV Jet axis /NJet dn/dδϕ SAR Preliminary SAR Preliminary 0-0% 0-0% Au+Au Au+Au /NJet dn/dδϕ SAR Preliminary SAR Preliminary 0-0% 0-0% Au+Au Au+Au As an example: gaus+ constant bkg. fit rigger jet Open symbols p+p Δϕ Open symbols p+p Δϕ Assoc. /NJet dn/dδϕ pt,assoc>.5 GeV pt,assoc>.5 GeV SAR SAR Preliminary Preliminary 0-0% 0-0% Au+Au Au+Au Open symbols p+p Δϕ=ϕJet ϕassoc. ϕjet = H trigger jet-axis found by Anti-kt with R=0.4, pt,cut> GeV and pt,rec(jet)>0 () GeV No Δη triangle acceptance applied (in progress...) Corrected for single particle tracking efficiencies rigger jet energy: correction for tracking eff. (p+p vs. Au+Au), background fluctuations ~ GeV; uncertainties: p+p=au+au and + GeV in jet energy (p+p relative to Au+Au) Δϕ Increased kinematics in JH due to jet requirement! Different systematics in bkg. correction compared to full-jet measurements! Can be used to study jet-finding biases in di-jets! 3

33 Short remark about ZYAM and broad correlation structures Narrow peaks: well-separated /d"! AB dn A /N Illustration bkg N.S. + bkg A.S. + bkg N.S. + A.S. + bkg Broad peaks overlap: ZYAM bkg. too high /d"! AB dn A /N bkg N.S. + bkg A.S. + bkg N.S. + A.S. + bkg ZYAM bkg (v ± %) "! (rad) "! (rad) Relatively small bias where peaks are separated (peripheral, p+p, high p). N.B.: bkg. modulation also typically small Background overestimated where broad peaks merge, subtracted shape highly sensitive to v uncertainty for weak correlations (central, low p) 4

34 First look at Jet v SAR mid-rapidity (PC+EMCal): Anti-k R = 0.4, p track,tower > GeV/c, p jet > GeV/c SAR Preliminary Au+Au H Is it possible to remove the jet particles from the EP calculation? First attempts: A. Ohlson (SAR), JE Collaboration Summer School All Par)cles Used to Calculate EP Jet Cone Removed Random Jet Cone Removed Specific Jet Cone Removed Jet Cone at Different η Removed he presence of a jet influences the EP calculation! For the jet definition used: jet v ~ v{} (used in jet-hadron correlations; max v uncertainties: no v and +50% of v Jet *v Assoc {}) Next steps: Using forward detectors to calculate the EP (FPC, BBC, ZDC-SMD), to suppress non-flow! 5

35 JH: Near-side IAA and energy balance DAA... rigger jet energy selection SAR Preliminary 0-0% Au+Au <p,rec Jet <5 GeV (effect smaller for larger jet energies) SAR Preliminary 0-0% Au+Au D AA (p assoc )=Y AA (p assoc ) p assoc,aa Y pp (p assoc ) p assoc,pp B = dp assoc D AA (p assoc ) ΔB~0.4 ± 0. (stat.) GeV for jet energies -5 GeV (ΔB~.6 GeV w/o p+p energy shift) Is the near-side suffering energy loss even with these stringent jet-finding criteria!? Are we just biasing the p+p like fragmentation after energy loss and do not see the energy loss in this analysis and di-hadrons, because it happens below GeV? Caveat: Δη study on near-side required; in progress... 6

36 JH: Away-side width and IAA Remember: his is not z!!! Significant (gaussian) jet broadening for recoil jets Softening of jet fragmentation : suppression at high p and enhancement at low p (p< GeV) Measurements/conclusions robust wrt to background subtraction Further studies: jet energy scale/uncertainties on near-side (Δη study), included in systematics 7

37 JH: Away-side DAA vs jet energy D AA (p assoc )=Y AA (p assoc ) p assoc,aa Y pp (p assoc ) p assoc,pp B = dp assoc D AA (p assoc ) Jet energy [GeV] ΔB [GeV] (stat. only) Away-side yields enhancement/suppression not fully balanced, more energy at low p in Au+Au But significant amount of energy ~3-4 GeV at low p compensated by high-p suppression! Jet-quenching at work! 8

38 But what about di-hadron correlations at lower pʼs? SAR, arxiv Increasing pt,trigger Same observation in: PHENIX, arxiv:0.77 (PRL in publication) $ "#%&%3!"#%&%3 -./0 4556, %&%'"!%()*+, %&%$"!%()*+, '"!%&%3 -./0 %&%7"!%()*+, 7"!%&%3 -./0 %&%8"!%()*+, 8"!%&%3 -./0 %&%$!"!%()*+,!"# Increasing pt,assoc!!"8!"7!"!!"8!"7!"!!"'!" $"!%&%3 $"#%&%3 "#%&%3 4556, 4556, 4556, %&%$"#%()*+, %&%"#%()*+, %&%7"!%()*+, 9:;9: 9:;9:%<"!<%&%!"= >;9: 0-% Au+Au Away-side structure dep. on p rig!!"$!!$! $ ' 7 #!$! $ ' 7 #!$! $ ' 7 #!$! $ ' 7 # trig assoc 9

39 A simple model: Mono-energetic Pythia jet in thermal bkg. model In general: wo-component (ZYAM) approach Jet axis ΔR In simple model: riggers Assoc. wo cases: (i) hjet-h: rigger associated to jet (ΔR<0.4) (ii) h-h: All trigger particles in event 30

40 Simple MC model results A. Adare (SAR), RHIC AGS Users Meeting 0 p A -3 GeV/c p B - GeV/c /N A dn AB /dδϕ "! (rad) 3

41 Simple MC model results A. Adare (SAR), RHIC AGS Users Meeting 0 p A -3 GeV/c p B - GeV/c o start: produce h-h correlations in pythia. /N A dn AB /dδϕ "! (rad) 3

42 Simple MC model results A. Adare (SAR), RHIC AGS Users Meeting 0 p A -3 GeV/c p B - GeV/c o start: produce h-h correlations in pythia. /N A dn AB /dδϕ Add isotropic thermal background; calculate hjet-h. rigger particles are inside ΔR = RC = "! (rad) Bkg. subtracted 3

43 Simple MC model results A. Adare (SAR), RHIC AGS Users Meeting 0 p A -3 GeV/c p B - GeV/c o start: produce h-h correlations in pythia. /N A dn AB /dδϕ Add isotropic thermal background; calculate hjet-h. rigger particles are inside ΔR = RC = 0.4. Background pedestal calculated "! (rad) 4 Bkg. subtracted "! (rad) 3

44 Simple MC model results A. Adare (SAR), RHIC AGS Users Meeting 0 p A -3 GeV/c p B - GeV/c o start: produce h-h correlations in pythia. /N A dn AB /dδϕ Add isotropic thermal background; calculate hjet-h. rigger particles are inside ΔR = RC = 0.4. Background pedestal calculated hjet-h: Pedestal subtraction recovers PYHIA yield (dark points) "! (rad) Bkg. subtracted "! (rad) 4 3

45 Simple MC model results A. Adare (SAR), RHIC AGS Users Meeting 0 p A -3 GeV/c p B - GeV/c o start: produce h-h correlations in pythia. /N A dn AB /dδϕ N trig ("R<0.4) / N trig = 0.4 Add isotropic thermal background; calculate hjet-h. rigger particles are inside ΔR = RC = 0.4. Background pedestal calculated hjet-h: Pedestal subtraction recovers PYHIA yield (dark points) "! (rad) 4 Bkg. subtracted 3 Inclusive h-h: many fake/uncorrelated background trigger particles (at low p ) - peak yield is suppressed by /f = 0.4 (/f=nrig Jet /Nrig Bkg ) - pedestal raised by /pi *(-/f) NJet Assoc = 0.67 riggers 0 Fake trigger Assoc "! (rad) 3

46 hjet-h vs. h-h in H Au+Au, p+p A. Adare (SAR), RHIC AGS Users Meeting 0 Anti-k R = 0.4, p track,tower > GeV/c, p jet > GeV/c /d"! AB dn A /N SAR preliminary p A B p GeV/c GeV/c h-h: Event contains a + GeV jet, but no ΔR cut hjet-h: Same events, with ΔR< Same v currently used for both as initial estimation subtracted) /d"! (v AB dn v = SAR event-plane, v{4} average 0-% H Au+Au: "R < % H Au+Au: no "R cut H p+p: "R < 0.4 H p+p: no "R cut ZYAM applied for consistency with SAR and PHENIX h-h analyses A /N "! (rad) 3

47 hjet-h vs. h-h in H Au+Au, p+p A. Adare (SAR), RHIC AGS Users Meeting 0 Anti-k R = 0.4, p track,tower > GeV/c, p jet > GeV/c /d"! AB dn A /N SAR preliminary p p A B GeV/c GeV/c h-h: Event contains a + GeV jet, but no ΔR cut hjet-h: Same events, with ΔR<0.4 Same v currently used for both as initial estimation subtracted) /d"! (v AB dn A /N v = SAR event-plane, v{4} average % H Au+Au: "R < 0.4"! 0-% H Au+Au: no "R cut H p+p: "R < 0.4 H p+p: no "R cut "! (rad) ZYAM applied for consistency with SAR and PHENIX h-h analyses 3

48 hjet-h vs. h-h in H Au+Au, p+p A. Adare (SAR), RHIC AGS Users Meeting 0 Anti-k R = 0.4, p track,tower > GeV/c, p jet > GeV/c /d"! AB dn A /N subtracted) /d"! (v AB dn A /N 4.5 SAR preliminary p A GeV/c p B GeV/c v = SAR event-plane, v{4} average % H Au+Au: "R < 0.4 "! (rad) 0-% H Au+Au: no "R cut H p+p: "R < 0.4 H p+p: no "R cut h-h: Event contains a + GeV jet, but no ΔR cut hjet-h: Same events, with ΔR<0.4 Same v currently used for both as initial estimation ZYAM applied for consistency with SAR and PHENIX h-h analyses "! (rad) 3

49 hjet-h vs. h-h in H Au+Au, p+p A. Adare (SAR), RHIC AGS Users Meeting 0 Anti-k R = 0.4, p track,tower > GeV/c, p jet > GeV/c /d"! AB dn A /N subtracted) SAR preliminary A p GeV/c /d"! (v AB dn A /N p B GeV/c v = SAR event-plane, v{4} average % H Au+Au: "R < 0.4 "! (rad) 0-% H Au+Au: no "R cut H p+p: "R < 0.4 H p+p: no "R cut "! (rad) h-h: Event contains a + GeV jet, but no ΔR cut hjet-h: Same events, with ΔR<0.4 Same v currently used for both as initial estimation ZYAM applied for consistency with SAR and PHENIX h-h analyses hjet-h and h-h correlations similar at highest trigger p! HI trigger and associated background complex... What do we measure with di-hadrons at lower trigger pʼs? 3

50 So, what do we learn? A. Adare (SAR), RHIC AGS Users Meeting 0 SAR Phys. Rev. Lett. 97 (006) 530 /d"! AB dn A /N p A B p GeV/c GeV/c SAR preliminary (a)p/! + 0-% Au+Au 60-80% Au+Au d+au - (b) p /! e + e - :(p+p)/(! + +! - ) Hwa:Recombination Fries:Coalescence+Jet 4 4 Ratios 40 - subtracted) /d"! (v AB dn A /N % H Au+Au: "R < % H Au+Au: no "R cut H p+p: "R < 0.4 H p+p: no "R cut "! (rad) ransverse Momentum p (GeV/c) Secondary (n-th) hard-scattering reduces h-h due to different jet energy scales sampled wrt hjet-h! If h-h is the true Au+Au jet correlation dominated by semi-hard scatterings! But there is the B/M enhancement! So some dilution due to fake triggers expected in h-h! We can not have both! 33

51 So, what do we learn? A. Adare (SAR), RHIC AGS Users Meeting 0 SAR Phys. Rev. Lett. 97 (006) 530 /d"! AB dn A /N p A B p GeV/c GeV/c SAR preliminary (a)p/! + 0-% Au+Au 60-80% Au+Au d+au - (b) p /! e + e - :(p+p)/(! + +! - ) Hwa:Recombination Fries:Coalescence+Jet 4 4 Ratios 40 - V) subtracted) /d"! (v AB dn A /N AMP Au+Au 00GeV % < 3 (a) ))! 0-% H Au+Au: "R < % H Au+Au: no "R cut H p+p: "R < 0.4 H p+p: no "R cut "! (rad) 3 = $cos(3(#-"! 3 $v <N part < AMP 60<N part <00 Au+Au 00GeV 40<N part <80 % < <N part < B. Alver, G. Roland arxiv: p (GeV) (b) ransverse Momentum p (GeV/c) Secondary (n-th) hard-scattering reduces h-h due to different jet energy scales sampled wrt hjet-h! If h-h is the true Au+Au jet correlation dominated by semi-hard scatterings! But there is the B/M enhancement! So some dilution due to fake triggers expected in h-h! We can not have both! My take on that: At lower trigger p we are dominated by bulk correlations! angular flow, v 3, as a function of transverse momentum, p, in bins of number of s at mid-rapidity ( η < ) in s NN = 00 GeV Au+Au collisions from the AMP Remark: More quantitative studies needed, but I think we have all the tools/measurements to answer this unambiguously! rs. 33

52 Figure 4: Run-5 Cu+Cu φ distribution for symmetric dijets with 7.5 GeV/c Discussion Jet R AA - Au+Au and p+p at Au+Au: % most central SAR Preliminary s NN =00 GeV/c kt R=0.4 anti-kt R=0.4 kt R=0. anti-kt R= p Jet (GeV/c) (?) Algorithmic differences R AA PHENIX Preliminary 0 0% # 0 0 %, "z! = 0.7 (PRL, 630) Run 5 Cu + Cu s NN = 00 GeV Gaussian filter, $ = (GeV/c) rec pp p (Yue Shi Lai, for the PHENIX Collaboration) RHIC/AGS Users Meeting, Workshop 6 0 / 30 SAR Preliminary Methodical issues or bulk effects (v3)!? Initial effect!? ests possible with 6 vs. 00 GeV! Ridge in JH!? Jet shapes! (?) 0.<pt,assoc<.0 GeV SAR Preliminary Open symbols p+p Yield Ratio: R=0./R= Au+Au and p+p at Au+Au: % most central SAR Preliminary s NN =00 GeV/c Au+Au kt Au+Au anti-kt p+p kt p+p anti-kt (?) (?) At low p, we are maybe not looking at jet physics! Methodical issues or bulk effects (v3)!? (?) SAR Preliminary 0-% dn/d"! area normalized to PHENIX Preliminary Run 5 Cu+Cu s NN = 00 GeV/c Gaussian filter, # = 0.3 symmetric jet jet 7.5 < p <.5 GeV/c rec Jet p (GeV/c) Bias or broadening? (can be tested with JH) 60 80% 40 60% 0 40% 0 0% SAR Preliminary SAR Preliminary "!

53 Figure 4: Run-5 Cu+Cu φ distribution for symmetric dijets with 7.5 GeV/c Discussion Jet R AA - Au+Au and p+p at Au+Au: % most central SAR Preliminary s NN =00 GeV/c kt R=0.4 anti-kt R=0.4 kt R=0. anti-kt R= p Jet (GeV/c) (?) Algorithmic differences R AA PHENIX Preliminary 0 0% # 0 0 %, "z! = 0.7 (PRL, 630) Run 5 Cu + Cu s NN = 00 GeV Gaussian filter, $ = (GeV/c) rec pp p (Yue Shi Lai, for the PHENIX Collaboration) RHIC/AGS Users Meeting, Workshop 6 0 / 30 SAR Preliminary (?) Methodical issues or bulk effects (v3)!? 0.<pt,assoc<.0 GeV Au+Au and p+p at s 0.8 NN =00 GeV/c Picture SAR Preliminary emerging: Au+Au: % most central SAR Preliminary Jet Initial quenching effect!? measurements 0.5 at RHIC can ests possible with vs. 00 GeV! be (qualitatively) (?) explained in a consistent 0.3 Ridge in JH!? 0. Jet shapes! Open symbols p+p 0. picture by significant broadening and Jet p (GeV/c) softening of the jet structure caused by (?) SAR Preliminary 0-% dn/d"! area normalized to Yield Ratio: R=0./R= PHENIX Preliminary Run 5 Cu+Cu s NN = 00 GeV/c Gaussian filter, # = 0.3 symmetric jet jet 7.5 < p <.5 GeV/c rec Au+Au kt Au+Au anti-kt p+p kt p+p anti-kt At low p, we are maybe not looking Bias or broadening? partonic energy loss in at jet physics! (can the be tested medium! with JH) Methodical issues or bulk effects (v3)!? (?) 60 80% 40 60% 0 40% 0 0% "! SAR Preliminary SAR Preliminary

54 From RHIC to LHC: he hard probes factory! Heavy ions at the LHC: LO p+p y=0 (h + +h - )/ s = 5500 GeV π 0 00 GeV fireball hotter and denser, lifetime longer than at RHIC 7 GeV Huge increase in yield of hard probes! RHIC LHC SPS 35

55 For example: Jet x-section measurement in ALICE ] - d! [mb GeV/c /dp jet d " /# ALICE EMCal PPR (009) qpyhia spectrum systematic uncertainty (uncorrelated) systematic uncertainty (0% correlated errors from p+p, uncorrelated bkg fluctuations) Pb+Pb s =5.5 ev NN % Central nb EMCal acceptance: η <0.7, Δφ= o EMCal needed for triggering and neutral jet energy component: Large kinematical reach in year ALICE running % stat. and syst. uncertainty Anti-k R=0.4 q=7 GeV /fm.8 Pb+Pb cross section uncertainty jet p [GeV/c] Precise measurement: Effect of background fluctuations in jet spectrum suppressed due to harder underlying partonic spectrum! Remark: DCal upgrade funded; would allow di-jet studies in ALICE! 36

56 Inclusive jet cross-section jet-quenching measurements Jet AA R 0. ALICE EMCal PPR (009) ALICE EMCal PPR (009) JE R AA systematic error Pb+Pb s NN = 5.5 ev systematic error % Central Pb+Pb cross section uncertainty p+p cross section uncertainty Anti-k, R=0.4 qpyhia [q] = GeV /fm q = q = 6 q = 7 q = 6 ALICE+EMCal year of data Jet p LHC (GeV/c) d!(r=0.)/d!(r=0.4). JE BROADENING qpyhia [q] = GeV /fm Pb+Pb s NN = 5.5 ev unfolding systematic uncertainty only Vacuum q = q = 6 q = 7 q = 6 ALICE+EMCal year of data Jet p % Central Anti-k LHC (GeV/c) Large kinematic reach: p Jet =50-00 GeV/c study QCD evolution of jet quenching Reduced systematic uncertainties due to heavy-ion background wrt to RHIC measurements Entering quantitative area concerning jet quenching measurements in heavy-ion collisions! 37

57 γ-jet and full-jet fragmentation functions Simulated result APQ G. Roland, HP08 ALICE CD/3 (008) E γ >0GeV Precise modified jet fragmentation function measurements over a large jet energy range by combining γ-jet and full-jet will place quantitative constraints on partonic energy loss models at the LHC! 38

A word of caution: Initial state effects at LHC... peripheral central y~3 at RHIC probes similar x as at mid-rapidity at LHC Suppression/de-correlation at y~3 in central d+au collisions at RHIC!

58 A word of caution: Initial state effects at LHC... peripheral central y~3 at RHIC probes similar x as at mid-rapidity at LHC Suppression/de-correlation at y~3 in central d+au collisions at RHIC! Onset of CGC!? Control experiment p+pb at LHC necessary to measure with high precision initial state effects to allow an unambiguous interpretation of jet-quenching measurements in Pb+Pb collisions! Remark: Can we learn more about the initial effects from other measurements before the p+pb run? 39

59 Summary Qualitative picture (so far) from RHIC: Jet-quenching measurements can be consistently explained by jet-broadening/softening due to radiative energy loss in the medium! Landscape of hard probes: Large kinematical reach and precise (full) jet measurements at the LHC: Quantitative constraints on underlying partonic energy loss mechanisms (for light quarks)! RHIC and LHC jet measurements will be complementary! But this is just the start! he landscape of hard probes is rich at the LHC (and RHIC II)! Measure heavy quark energy loss (b-tagged jets), still open theoretical issue to describe heavy and light flavor energy loss in a consistent framework! 40

Exploring Jet Properties in p+p Collisions at 200 GeV with STAR

Exploring Jet Properties in p+p Collisions at 200 GeV with SAR Helen Caines - Yale University - for the SAR Collaboration Outline Quark Matter 2009 Knoxville,N, USA March 30 th -April 4 th What we know