Drell-Yan PerspectivesDrell-Yan Perspectives

at FAIRat FAIR

Marco Destefanis Università degli Studi di Torino

Drell-Yan Scatteringand the Structure of Hadrons

Trento (Italy)May 21-25, 2012

for the PANDA Collaboration

OverviewOverview

• Motivation

• Drell-Yan cross section and azimuthal asymmetries

• Experimental scenarios

• Drell-Yan process and background @ PANDA

A. Bianconi Drell-Yan generator

Cut studies

• Investigation of Drell-Yan asymmetries

• Towards antiproton polarized beams

• Summary

MotivationMotivation

Complete description of the nucleonic structure

requires:

➠ Parton Distribution Functions (PDF)

➠ Fragmentation Functions (FF)

Including kT dependence

➠ Transverse Momentum Dependent (TMD) PDF and

FF

➠ Test of Universality

SIDISf

TDYf

T ff || 11

J. C. Collins and D. E. Soper, Nucl. Phys. B 194, 445 (1982)

Helicity Helicity basebase

Twist-2 PDFs: ),()(1 xuxf u ),()(1 xuxg u )()(1 xuxhu

1/2 1/2

L L

+1/2 1/2

R R

q(x) spin averaged

a

aaa xfxfxexxFxF )()()(2)( 212

1/2 1/2

L L-

1/2 1/2

R R

Δq(x) helicity difference

Helicity Helicity basebase

Twist-2 PDFs: ),()(1 xuxf u ),()(1 xuxg u )()(1 xuxhu

Helicity Helicity basebase

Twist-2 PDFs: ),()(1 xuxf u ),()(1 xuxg u )()(1 xuxhu

-

δq(x) helicity flip

off diagonal in the helicity base

LR

LR

uuu

uuu

2

12

1

TransversiTransversityty

δq(x): a chirally-odd, helicity flip distribution function

δg(x): no gluon transversity distribution; transversely polarised nucleon shows transverse gluon effects at twist-3 (g2) only

SOFFER INEQUALITYSOFFER INEQUALITY

An upper limit:

• can be violated by factorisation at NLO• inequality preserved under evolution to larger scales only

1 1 1

1( ) ( ) ( )

2h x f x g x

TMD: KTMD: KTT-dependent Parton Distributions-dependent Parton Distributions

Twist-2 PDFs )k,x(fkd)x(f T1T2

1

Distribution functions

Chirality

even odd

Twist-2

U

L

T

, h1,

Transversity

Boer-Mulders

Sivers

1hL1hT1h

Tf1 Tg1

1f

1g

TMD PDF InvestigationTMD PDF Investigation

➠ Process SIDIS → convolution with FF

Drell-Yan → PDF only

pp annihilations: each valence quark can

contribute to the diagram

➠ Energies

@ FAIR unique energy range up to s~30 GeV2 with PANDA up to s~200 GeV2 with PAX

@ RHIC much higher energies → big contribution from sea-quarks

Drell-Yan ProcessDrell-Yan Process

• Drell-Yan: pp -> +-X

Collins-Soper frameCollins-Soper frame

Kinematics

x1,2 = mom fraction

of parton1,2

= x1 • x2 =

M2/s

xF = x1 - x2

Collins-Soper frame: Phys. Rev. D16 (1977) 2219.

Drell-Yan ProcessDrell-Yan Process

Gives access tochirally odd functions

Transversity

No convolution with FF

Chirally odd functions are not suppressed (as in DIS)

DIS ProcessDIS Process

• RHIC energies: √s = 100 GeV -> τ ≤ 10-2 -> small x1 and/or x2

• Too small Soffer upper bound on ATT (percent level[1])

• ATT @ RHIC very small, smaller √s would help[2]

[1] Martin et al, Phys.Rev. D60 (1999) 117502.[2] Barone, Colarco and Drago, Phys.Rev. D56 (1997) 527.

a

aaaTT hhe

XppXpp

XppXppA 11

2

DIS ProcessDIS Process

• RHIC energies: √s = 100 GeV -> τ ≤ 10-2 -> small x1 and/or x2

• Too small Soffer upper bound on ATT (percent level[1]):

• ATT @ RHIC very small, smaller √s would help[2]

[1] Martin et al, Phys.Rev. D60 (1999) 117502.[2] Barone, Colarco and Drago, Phys.Rev. D56 (1997) 527.

a

aaaTT hhe

XppXpp

XppXppA 11

2

2 20Q 0.23 GeV

2 2Q 25 GeV

QCD higher order contributionsQCD higher order contributions

s=45 GeV2

s=210 GeV2

s=45 GeV2

s=210 GeV2

[1] Shimizu et al., hep-ph/0503270 .

• cross section affected• K-factors almost spin independent

QCD contributions to AQCD contributions to ATTTT

• Contributions drop increasing the energy[1]

[1] Shimizu et al., hep-ph/0503270 .

Perturbative CorrectionsPerturbative Corrections

• Smaller at higher energies[1]

2GeV 30s 2GeV 200s

[1]H. Shimizu et al., Phys. Rev. D71 (2005) 114007

Double Spin Asymmetries @ s = 30, 45 GeVDouble Spin Asymmetries @ s = 30, 45 GeV22

• ATT small at large √s and M2

due to slow evolution of h1a(x,Q2)

• Large ATT expected[1] for √s and M2 not too large and τ not too small

[2]M. Anselmino et al., Phys. Lett. B594 (2004) 97.

2GeV 30s

2GeV 45s

220

20

201q

[2] GeV/c 23.0Q @ )Qx,(q)Qx,(h Assuming

[1] Shimizu et al., Phys.Rev.D 71 (2005) 114007.

[3] Efremov et al, Eur. Phys. J. C 35 (2004) 207.

Q2 = 5 GeV2

Q2 = 9 GeV2Q2 = 16 GeV2

NLO pQCD: λ 1, 0, υ 0

Lam-Tung sum rule: 1- λ = 2ν

• reflects the spin-½ nature of the quarks

• insensitive top QCD-corrections

Experimental data [1]: υ 30 %

[1] J.S.Conway et al., Phys. Rev. D39 (1989) 92.

Di-Lepton Rest Frame

Drell-Yan Drell-Yan AsymmetriesAsymmetries

Expected polar Expected polar distributionsdistributions

E772 @ Fermilab

λ, μ, ν measured[1] in p N→μ+μ - X

Perfect agreement with pQCD exptectations!

[1] McGaughey, Moss, JCP, Annu. Rev. Nucl. Part. Sci. 49 (1999) 217.

)cos1( 20

dd

E537 @ Fermilab

Anassontzis et al., Phys. Rev. D38 (1988) 1377

vs coscos

d

d

p

pπ

π

vs coscos

d

d

Expected polar Expected polar distributionsdistributions

Conway et al, Phys. Rev. D39 (1989) 92

E615 @ Fermilab

-N +-X @ 252 GeV/c

-0.6 < cos < 0.6

4 < M < 8.5 GeV/c2

Angular distribution in CS Angular distribution in CS frameframe

Conway et al, Phys. Rev. D39 (1989) 92

30% asymmetry observed for -

E615 @ Fermilab -N +-X @ 252 GeV/c

Angular distribution in CS Angular distribution in CS frameframe

NA10 coll., Z. Phys. C37 (1988) 545

NA10 @ CERN -N +-X @ 286 GeV/c

Deuterium

Tungsten

Nuclear Nuclear effects?effects?

NA10 @ CERN[1]λ, μ, ν measured in π N→μ+μ – X

υ involves transverse spin effects at leading twist [2]

If unpolarised DY σ is kept differential on kT , cos2φ contribution to angular distribution provide:

2 21 2, 1 1h (x κ ) h (x ,κ )

[1] NA10 coll., Z. Phys. C37 (1988) 545

Violation of Lam-Tung sum Violation of Lam-Tung sum rulerule

[2] D. Boer et al., Phys. Rev. D60 (1999) 014012.

[1] L. Zhu et al, PRL 99 (2007) 082301; [1] D. Boer, Phys. Rew. D60 (1999) 014012.

Boer-Mulders

T-odd Chiral-odd TMD

• ν > 0 → valence h 1 has same sign in π

and N

• ν(π-W→μ+μ-X) ~ h 1 (π)valence x h 1

(p)valence

• ν(pd→μ+μ-X) ~ h 1 (p)valence x h 1

(p)sea

• ν > 0 → valence and sea h 1 has same

sign, but sea h 1

should be significantly smaller

[1]

Drell-Yan Drell-Yan AsymmetriesAsymmetries

Drell-Yan Drell-Yan AsymmetriesAsymmetries

λ 1, 0

Even unpolarised beam on polarised p, or polarised on unpolarised p

are powerful tools to investigate кT dependence of QDF

D. Boer et al., Phys. Rev. D60 (1999) 014012.

pp

Transverse Single Spin Transverse Single Spin AsymmetriesAsymmetries

# of partons in polarized proton depends on

Sivers effect

p pS p k \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

PDFX

p\\\\\\\\\\\\\\p

pS\\\\\\\\\\\\\\

k

qp

X

p\\\\\\\\\\\\\\p

qS\\\\\\\\\\\\\\

k

qppartons’ polarisation in unpolarized proton depends on

Boer Mulders effect

q qS p k

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

FF

X

p\\\\\\\\\\\\\\q

S

\\\\\\\\\\\\\\

p̂

q

Polarising Fragmentation Function

qS p p \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

Hadrons’ polarisation coming from

unpolarised quarks depends on

X

p\\\\\\\\\\\\\\q

qS\\\\\\\\\\\\\\

p\\\\\\\\\\\\\\

qfragmentation of polarised quark depends on

Collins effect

q qS p p \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

Transverse Single Spin Transverse Single Spin AsymmetriesAsymmetries

d d

d dNA

All those terms contribute to the Single Spin Asymmetry

SINGLE-POLARISED

UNPOLARISED

.

DOUBLE-POLARISED

Drell-Yan Cross SectionDrell-Yan Cross Section

R.D. Tangerman and P.J. Mulders, Phys. Rev. D51, 3357-3372 (1995)

Azimuthal Azimuthal AsymmetriesAsymmetries

)2cos(

)sin(),sin(

2cos

21

22

SS

SS

• Unpolarized

• Single polarized

• Double polarized

U = N(cos2φ>0)

D = N(cos2φ<0) DUDU

A

Asymmetry

Primary beams:• Proton• Heavy Ions• Factor 100-1000 over present in intensity

Future GSI andFuture GSI andFacility for Antiproton Facility for Antiproton

and Ion Researchand Ion Research

Secondary Beams:• Radioactive beams• Antiprotons 3 - 30 GeV 1-2 107 /s

Storage and Cooler Rings:

• Radioactive beams• e – A collider

• 1011 stored and cooled

0.8 - 14.5 GeV antiprotons

High Energy Storage High Energy Storage RingRing

ElectroncoolerE<8 GeV

Injection

HESR

High res. mode: L = 1031 cm-2 s-1 p/p < 10-5

High lum. mode: L = 2·1032 cm-2 s-1 p/p < 10-4

Cooling: electron/stochastic

Pmax = 15 GeV/cLmax = 2·1032 cm-2 s-1

Ø < 100 mp/p < 10-5

internal target

Characteristics70 m

185

m

1011 stored and cooled 0.8-15 GeV/c antiprotons

Detector

HESR: asymmetric collider layout

Asymmetric double-polarised collider mode proposed by PAX people:

• APR (Antiproton Polariser Ring): polarising antiprotons, p > 0.2 GeV/c• CSR (Cooled Synchrotron Ring): polarised antiprotons, p = 3.5 GeV/c• HESR: polarised protons, p = 15 GeV/c

The PANDA DetectorThe PANDA Detector

STT Detectors

Physics Performance Report for PANDA arXiv:0903.3905

Cherenkov

Polarized Antiproton eXperiments

Asymmetric collider (√s=15 GeV): polarized protons in HESR (p=15 GeV/c) polarized antiprotons in CSR (p=3.5 GeV/c)

The PAX DetectorThe PAX Detector

V. Barone et al., Technical Proposal for PAX, 2005

http://www2.fz-juelich.de/ikp/pax/portal/index.php?id=103

CERN NA51 450 GeV/c

Fermilab E866 800

GeV/c

Di-Lepton ProductionDi-Lepton Production

pppp -> -> ll++ll--XX

A. Baldit et al., Phys. Lett. 332-B, 244 (1994)

R.S. Towell et al., Phys. Rev. D 64, 052002 (2001)

Phase space for Drell-Yan processesPhase space for Drell-Yan processes

x1,2 = mom fraction of parton1,2

= x1 • x2

xF = x1 - x2

= const: hyperbolaexF = const: diagonal

PAX @ HESR

symmetric HESR collider

1

1.5 GeV/c2 ≤ M ≤ 2.5 GeV/c2

PANDA

Drell-Yan Process and BackgroundDrell-Yan Process and Background

• Background studies: needed rejection factor of 107

• Drell-Yan: pp -> +-X

cross section 1 nb @ s = 30 GeV2

• Background: pp -> +-X, 2+2-X,……

cross section 20-30 b

m = 105 MeV/c2; m 145 MeV/c2

average primary pion pairs: 1.5

A. Bianconi Drell-Yan Generator for ppA. Bianconi Drell-Yan Generator for pp• Antiproton beam

• Polarized/Unpolarized beam and target

• Drell-Yan cross section from experimental data

• Selects event depending on the variables:

x1, x2, PT, , , S

from a flat distribution

• Cross section: ST

T

,,AP'Sx,xSS

K

ddPdxdx

d

2121

A. Bianconi, Monte Carlo Event Generator DY_AB4 for Drell-Yan Events with Dimuon Production in Antiproton and Negative Pion Collisions with Molecular Targets,

internal note (PANDA collaboration)

A. Bianconi, M. Radici, Phys. Rev. D71, 074014 (2005) & D72, 074013 (2005)

A. Bianconi, Nucl.Instrum.Meth. A593: 562-571, 2008

DY @ 15 GeV/c — pp->DY @ 15 GeV/c — pp->++--XX

• layout studies for muon id with ABDYG (1.5 MEv)

A. Bianconi Drell-Yan Generator

[1]A. Bianconi and M. Radici, Phys. Rev. D71 (2005) 074014

[1]

5.5s GeVDilepton Mass Distribution

Mdilepton [GeV/c2]

Focus on one single M range

1.5 MEv in 1.5 < M< 2.5 (GeV/c2)

σ04≤M≤9 ~ 0.4 pb

σ01.5≤M≤2.5 ~ 0.8

nbxP

xPx2

x1

DY Asymmetries @ VertexDY Asymmetries @ VertexUNPOLARISED SINGLE-POLARISED

500KEv included in asymmetries

Acceptance

corrections crucial!

1 < qT < 2 GeV/c

2 < qT < 3 GeV/c

xP xP

xP

xP

xPxP

Physics Performance Report for PANDA arXiv:0903.3905

R = L·σ·ɛ

= 2·1032cm-2s-1 × x 0.8·10-33cm2× 0.33

= 0.05 s-1 ~ 130

Kev/month

Statistical errors for 500KEv generated

xP

)

)xP

xPPhysics Performance Report for PANDA arXiv:0903.3905

DY Asymmetries @ VertexDY Asymmetries @ Vertex

The Road to The Road to

Polarized AntiprotonsPolarized Antiprotons

How to polarize antiprotons?How to polarize antiprotons?

Intensity loss,

but it works

H. Ströher, PoS(STORI11) 030D. Oellers, PoS(STORI11) 008

Spin filtering

Λbar decay

Not feasible

Not enoughtintensity

Spin flip

Spin dependent reaction Under study

Experimental setupExperimental setup

H. Ströher, PoS(STORI11) 030D. Oellers, PoS(STORI11) 008

Beam PolarimenterBeam Polarimenter

• pd elastic scattering

detection in two (L-R) symmetric Silicon Tracking Telescopes

Deuteron separation

H. Ströher, PoS(STORI11) 030D. Oellers, PoS(STORI11) 008

Results – Beam PolarizationResults – Beam Polarization

Measurements after a storing period of 5000 s

H. Ströher, PoS(STORI11) 030D. Oellers, PoS(STORI11) 008

• Polarization lifetime: τp > 105 s

• Spin-flip efficiency: 0.9887 ± 0.0001

SummarySummary

New physics from unpolarized DY

Wide Single Spin Asymmetry program

Double Spin Asymmetries, if p can be polarized

• Interest on Drell-Yan studies• 1.5 < Mμμ < 2.5 GeV/c2

• Cuts for background rejection

Rejection factor achieved for secondary background: >

5 106

Kinematically constrained refit still to be investigated

• Few months of data taking are enough to: evaluate unpolarised and single-spin asymmetries with good accuracy investigate their dependence on qT,μμ

• Next step: Polarized Antiprotons