ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 1 Correlations in double parton distributions at small x MTA KFKI RMKI, Budapest, Hungary Dept. of Astronomy and Theoretical Physics, Lund University, Swede András Ster
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 1
Correlations in double parton distributions at small x
MTA KFKI RMKI, Budapest, Hungary Dept. of Astronomy and Theoretical Physics, Lund University, Sweden
András Ster
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 2
Content
• Introduction
• Double Parton Scattering & Double Parton Distributions
• The Lund Dipole Cascade Model
• Application to Double Parton Distributions
• Results
• Summary
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 3
Introduction - Motivation
Multiple hard parton (q,g) collisions are common in high energy pp collisions
For example:
• 4 jets in 63 AGeV events @ CERN ISR (AFS)
• 4 jets and 3 jets + in 1.8 TeV events @ Tevatron (CDF, D0)
? @ LHC
Larger probability found than expected for uncorrelated hard subcollisions
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 4
Literature
Connection between correlations in momentum and impact parameter (b) space have not yet been studied
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 5
Correlation studies
Our solution: detailed dynamical model for parton evolution (Lund Dipole Cascade Model)
Earlier Sjöstrand and van Zilj assumed that the dependence of double-parton density on kinematic variables (x, Q2) and on the separation in impact parameter space (b) factorizes.
Implemented in PYTHYA and HERWIG event generators
Problem: how to extrapolate to higher energies (LHC)
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 6
Correlation studies
Lund Dipole Cascade Model
Based on BFKL and Müller’s dipole cascade model
Correlation sources in the model: 1) from fluctuations (event by event) 2) “hot spots” makes the profile narrower
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 7
Double Parton Scattering & Double Parton Distributions - Experimental results
Measure of correlation is defined by eff according to
D(A,B) is the cross-section of the two hard processes
SA and S
B are the single inclusive cross-sections
If hard interactions were uncorrelated, eff would be equal to the total non-diffractive corss-section
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 8
Experimental results
CERN ISR (AFS) results:
eff 15 mb
Tevatron (CDF, D0) results:
eff 5 mb
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 9
The formalism:
Stirling and Gaunt:
Essential formulae
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 10
The Lund Dipole Cascade Model
We define:
With the constraint:
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 11
The Lund Dipole Cascade Model
Dipole cascades:
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 12
The Lund Dipole Cascade Model
We define:
With the constraint:
ZIMANYI-SCHOOL11, Budapest, 01/12/2011 A. Ster, RMKI, Hungary 20
Summary
In our analysis we have found that the two-parton correlation function F(b) in a non-trivial way on all kinematic variables x1,x2,Q2
1,Q22
Lund Dipole Cascade Model offers unique possibility to study gluon evolution inside hadrons at small x
Our results show a non-flat energy dependence within errors compatible with experiments in their energy range.