Inclusion of up-to-date parton distribution function and nuclear shadowing

Transcription

1 Inclusion of up-to-date parton distribution function and nuclear shadowing in the AMP model Chao Zhang 1 ( 张潮 ) ZiWei Lin 1,, ShuSu Shi 1 and Liang Zheng 1 Central China Normal University 1 East Carolina University SCU AMP Workshop 1

2 Outline 1. Motivation. Methods & Strategies. Results Øwo component model parameter tuning ØFragmentation parameter tuning 4. Summary & Outlook SCU AMP Workshop

3 Motivation A+B PDF Cteq6M Initial conditions Parton Interactions Hadronization Hadron cascade Final particle spectra SCU AMP Workshop

4 PDF: Parton Distribution Function Duke-Ovens: adequate fordescription at RHIC energies. Outdated AMP model: valid for wide energy range, especially LHC energies when minijet production reaches to a very small-x region, where gluon distribution is much higher than Duke-Owens parametrization. Update the PDF. HIJING.0 work:grv94l PDF. SCU AMP Workshop 4

5 Nuclear modification antishadowing EMC effect shadowing EPS, [hep-ph] Defined vs. CEQ6.1M Free p PDFs. R g Pb (x,q 0 ): nuclear modification for gluon, obvious depress shadowing effect at small x. SCU AMP Workshop 5

6 Nuclear modification factor vs. Impact parameter In the center of the nucleus the modifications are slightly stronger than the average. By construction, nuclear modifications vanish at the edge of the nucleus. HEHS, [hep-ph] SCU AMP Workshop 6

7 Preliminary comparison without tuning ) dp evt dn/(n dη) evt dn/(n cteq61m cteq6m duke PbPb 0 p η ) dp evt dn/(n dη) evt dn/(n 1-1 π + PP 0 p 4 6 η PbPb collision:.76 ev, Factor ~ larger for both p " and pseudorapidity, mainly within eta <5. PP collision: 1 ev, No significant change vs. PbPb, 0% level increase eta ~0. No significant differencebetween cteq61m and cteq6m SCU AMP Workshop 017/7/4 7

8 Methods and strategies Parameter tuning strategy 1. otal and inelastic cross section fitting: to get the key parameter input p # and σ &'() in the two component model.. Lund fragmentation parameter (a&b) tuning: use the chargedparticle pseudorapidity distribution and transverse momentum spectra. Step have no influence on the step 1, thus we can do the step 1 first and then step. SCU AMP Workshop 017/7/4 8

9 wo component model σ./ σ )') p * Hard interactions (PQCD) σ +,- p # soft interactions σ &'() Key input parameters in the model PHYSICAL REVIEW C 8, (011) SCU AMP Workshop 017/7/4 9

10 uning method HIJING input p # σ &'() σ.0_1) σ )')_1) σ.0_bc- σ )')_BC- compare C. Patrignani et al. (Particle Data Group), Chin. Phys. C, 40, 0001 (016). A relative residual sum of squared is defined as the target function to be minimized allowed p # and σ &'() parameters Minimize χ = (5 676_89:6 ;5 676_<=6 )? 5 676_<=6 + )? SCU AMP Workshop 017/7/4

11 χ plots E= GeV E=15.8GeV E=167GeV E=1965GeV SCU AMP Workshop 017/7/4 11

12 p # and σ &'() tuning his fit is done for the the PDF Cteq6m. When collision energy S HH > GeV, it is matched with both σ )') and σ.0, however when S HH < GeV, we only fit to the Inelastic cross section. Jet cross section is completely switched off below GeV in HIJING. SCU AMP Workshop 017/7/4 1

13 results he ampt results compared to pp data. SCU AMP Workshop 017/7/4 1

14 Experimental data Charged particle pseudorapidity distributions Invariant transversemomentumspectra ) (mb/gev σ/dp Ed 1 1 s = 00GeV pp s = 500GeV pp s = 900GeV pp s = 1800GeV pp s = 7000GeV pp p (GeV/c) he data wereused to tune the fragmentation parameters (a&b) SCU AMP Workshop 017/7/4 14

15 Parameter (a) contribution to η distribution Default version; pp collision We fix b=0.5 and vary a=0.15 ~ 0.6 range Larger (a) gives overall larger charge particle density. dη) evt dn/(n 4 1 NSD cut made with < η <5 for UA5 data and at least one track within η <1 (INEL>0) for LHC data 8 a=0.15 a=0. a=0.6 b= 0.5 ALICE data a=0.15 a=0. a=0.6 UA5 data η dη) evt dn/(n SCU AMP Workshop 017/7/4 15 η b= 0.5

16 Parameter (b) contribution to η distribution We fix a=0. and vary b=0.4 ~ 0.6 range Charge particle density becomes slightly larger with smaller b at η~0, but the change is not strong enough to explain the difference to data from 0.4 to a= 0. NSD cut made with both ends of < η <5 accepts charged tracks for UA5 ppbar data at 00 GeV dη) evt dn/(n 1 b=0.5 b=0.4 b=0.6 UA5 data η SCU AMP Workshop 017/7/4 16

17 a&b contribution to the P " spectra he p spectra is not so sensitive to the variation of a. It agrees better with b=0.5; smaller b (b=0.4) leads to wider p tail. ) (mb/gev σ/dp Ed b=0.5 a=0. a=0.4 a=0.5 UA1 NSD 0 1 p ) (mb/gev σ/dp Ed a=0. b=0.4 b=0.5 b=0.6 UA1 NSD p SCU AMP Workshop 017/7/4 17

18 a&b tuning 8 AMP,NSD,a=0.,b=0.5 UA5 s NN =00GeV a=0., b=0.5 agrees with data in general, he mid-rapidity η~0 region is always smaller than data at different energy scales. dη) evt dn/(n 6 4 UA5 UA5 CDF CMS ALICE s NN =546GeV s NN =900GeV s NN =1800GeV s NN =7000GeV s NN =00GeV η UA5 Collaboration, G.J. Alner et al. Z. Phys. C Particles and Fields, 1-6(1986). CDF Collaboration F.D. Snider et al. Phys. Rev. D 41, 0 (1990). CMS Collaboration V. Khachatryan et al. PRL 5, 000 (0). ALICE Collaboration Physics Letters B 751 (015) SCU AMP Workshop 017/7/4 18

19 a&b tuning he p spectra is consistent with data at all energies ranging from 00 GeV to 1 ev ) (mb/gev σ/dp Ed 4 1 (h + +h - )/ AMP,NSD,a=0.,b=0.5 UA1 s NN =00GeV (x 0.1) UA1 s NN =546GeV (x 1) UA1 s NN =900GeV (x ) CDF s NN =1800GeV (x 0) dη) N/(dp )d )1/(πp evt 1 AMP,NSD,a=0.,b=0.5 CMS s NN =7000GeV ALICE s NN =00GeV (x ) 5 1/(N p 7 UA1 Collaboration, C. Albajar, et al.nuclear Physics B5 (1990) CDF Collaboration, F. Abe, Phys. Rev. Lett. 61, 1819 (1988) SCU AMP Workshop 017/7/4 19 p

20 a&b tuning String melting version; pp collision a=0.5, b=0.5 for energy below 1800 GeV, b=0.5 for others. he overall feature agrees with data in all energies. dη) evt dn/(n AMP,string melting,nsd,a=0.5 UA5 s NN =00GeV, b=0.5 UA5 s NN =546GeV, b=0.5 UA5 s NN =900GeV, b=0.5 CDF s NN =1800GeV, b=0.5 CMS s NN =7000GeV b=0.5 ALICE s NN =00GeV, b= η ) (mb/gev σ/dp Ed UA1 s NN =00GeV (x 0.1) UA1 s NN =546GeV (x 1) UA1 s NN =900GeV (x ) CDF AMP,string melting,nsd,a=0.5,b=0.5 s NN =1800GeV (x 0) p dη) N/(dp )d )1/(πp evt 1/(N AMP,string melting,nsd,a=0.5,b=0.5 ALICE CMS s NN =7000GeV s NN =00GeV (x ) p SCU AMP Workshop 017/7/4 0

21 Summary & Outlook Summary: 1. he necessity for updated PDFand nuclear shadowing modification.. Built up the systematic strategy to determine the parameter.. Fit the energy dependence of p # and σ &'() as well as tuning the parameters with the latest dataset. Outlook: 1. Implement inelastic -> interactions in the current ZPC model.. Determination of the Lund fragmentation parameter a and b for AA collision on the string melting mode.. Study the Heavy Flavor results with the updated AMP model. SCU AMP Workshop 017/7/4 1

22 SCU AMP Workshop 017/7/4

23 SCU AMP Workshop 017/7/4

24 SCU AMP Workshop 017/7/4 4