What can we learn from η production in proton-proton collisions? Joe Seele MIT and University of Colorado.

What can we learn from η production in proton-proton collisions?

Joe SeeleMIT and University of

Colorado

J. Seele - SPIN 2008 2

Outline

• Proton spin puzzle• η production in (polarized) proton-proton collisions

• Measurements of η production at PHENIX

• Extraction of η fragmentation functions

• Constraining ΔG with η production in polarized proton-proton collisions at PHENIX


The Proton Spin Puzzle

€

1

2=

1

2Δq∑ + Lq

z + ΔG + Lgz

Fairly well measuredonly ~30% of spin

A future challenge

The proton is viewed as being a “bag” of bound quarks and gluons interacting via QCD

Spins + orbital angular momentum needto give the observed spin 1/2 of proton

Beginning to be measuredat RHIC (and hopefully EIC)


Double Helicity Asymmetries

€

ALL =σ ++ −σ +−

σ ++ + σ +−=

Δfa ⊗Δfb

a,b,c

∑ ⊗ d ˆ σ fa fb → fc X ⋅ ˆ a LLfa fb → fc X ⊗D fc

h

fa ⊗ fb

a,b,c

∑ ⊗ d ˆ σ fa fb → fc X ⊗D fc

h

€

ALL ≈ aggΔG2 + aqgΔqΔG + aqqΔqΔq'and translating…

Take the asymmetry of proton helicity configurations


Cross Section and Fragmentation Functions

High pT particle production provides information about the

parton distribution functions and fragmentation functions

€

dσ = fa ⊗ fb

a,b,c

∑ ⊗ d ˆ σ ab →cX ⊗Dch


Why the η?

Want a final state that:1) Produced in large quantities2) Theoretical Motivation

PHENIX - PRC 75 024909


Why the η?

Want a final state that:1) Produced in large quantities2) Theoretical Motivation

€

π 0 = uu − dd

€

η =uu + dd − 2ss

Different flavor and gluon dependence of structure and fragmentation should give different sensitivities to hard

subprocesses


RHIC

Year s [GeV] Recorded L Pol [%] FOM (P4L)

2003 (Run 3) 200 .35 pb-1 32 3.7 nb-1

2004 (Run 4) 200 .12 pb-1 45 4.9 nb-1

2005 (Run 5) 200 3.4 pb-1 49 200 nb-1

2006 (Run 6) 200 7.5 pb-1 57 690 nb-1


PHENIX

Central Arms:• γ/π0/η detection

– Electromagnetic Calorimeter (-0.35 < η < 0.35)

– PC3 - Charge VetoGlobal Detectors:• Relative Luminosity

– Beam-Beam Counter (BBC) (+- 3.1 < η < 4.0)

– Zero-Degree Calorimeter (ZDC) (+- 6.9 < η < infinity)

• Local Polarimetry - ZDC


Measuring the η at PHENIX

Channel BR (%)

η->2γ 39.39

η->π0π0π0 32.52

η->π+π-π0 22.68

η->π+π-γ 4.69

Choose a decay channel that couples wellwith the capabilities of PHENIX


Measuring the η at PHENIX

The number of η’s in each pT bin are determined by fitting the peak+background in the di-photon

invariant mass spectrum

2-3 3-4 4-5 5-6

6-7 7-9 9-11


η Cross Section - Results

Run3 and Run5 cross section results


Measuring Helicity Asymmetries

€

ALL =1

PB PY

N ++ − RN +−

N ++ + RN +−

€

R =L++

L+−

€

ALL =σ ++ −σ +−

σ ++ + σ +−=

Δσ

σ

What to measure :1) N’s - Final state - the η2) P’s - Polarization at IR3) R - Relative Luminosity

PHENIX Central Arms

PHENIX ZDCs + RHIC Polarimeters

PHENIX ZDCs+BBCs


η Asymmetry - Results

Scaling uncertainty (Run5 = 9.4% and Run6 = 8.3%) from polarization not shown

*pT bins identical up to

pT=6. Above that they use different binning.


η Fragmentation Functions

None previously existed, so Marco Stratmann, Christine Aidala and myself

extracted the η FFs from data on e++e- and p+p.


η FFs - Comparison

Describes e+e- data very well over a large range in energies.

Plan to incorporate BABAR, and possibly

some UA1/UA2 data.


η Cross Section and Theory

NLO+η FF (μ=pT) calculation done by Marco Stratmann

*In a way this is more of a self-consistency check than a prediction.


η and π0 Subprocess Fractions


η Asymmetry and Theory

Calculations by Marco Stratmann


Constraining ΔG with the η

A series of constrained fits were done by

Vogelsang and Stratmann in the GRSV model.

They constrained the integral of ΔG at the

input scale.

€

C = dxΔg(x,Q02)

0

1

∫



Each set yields a different ALL

Does not give much power for the negative ΔGs



Allows a small positive or strongly negative ΔG.

It should be noted that this is only for the GRSV fits.

The data needs to be incorporated into a flexible global fit.


Constraining ΔG with the η and π0


The Polarization of the Proton’s Glue

• PHENIX has measured the double helicity asymmetry in η production at mid-rapidity

• With the new fragmentation functions it will provide a constraint on ΔG

Thanks!


Backup


η FFs - DataExperiment System Energy (GeV) # Points

ALEPH ’92 e+e- 91.2 8

ALEPH ’00 e+e- 91.2 18

ALEPH ‘02 e+e- 91.2 5

L3 ‘92 e+e- 91.2 3

L3 ’94 e+e- 91.2 8

OPAL e+e- 91.2 9

ARGUS e+e- 10 6

CELLO e+e- 35 4

HRS e+e- 29 13

JADE ’85 e+e- 34.4 1

JADE ‘’90 e+e- 34.9 3

MARK II e+e- 29 7

PHENIX 2 p+p 200 12

PHENIX 3π p+p 200 6

PHENIX ’05 prelim.

p+p 200 19


Relative Luminosity at PHENIX

€

R =NBBC

++

NBBC+−

=L++σ BBC

++

L+−σ BBC+−

=L++

L+−Measured using the Beam-Beam Counter (BBC) Coincidences

Systematic Studies Include :1. Spin Independent BBC cross

section2. Effects from multiple

collisions per bunch crossing

Currently δR ~ O(10-4)

€

δALL ∝δN ++

N ++

⎛

⎝ ⎜

⎞

⎠ ⎟

2

+δN +−

N +−

⎛

⎝ ⎜

⎞

⎠ ⎟

2

+δR

R

⎛

⎝ ⎜

⎞

⎠ ⎟2

Statistical uncertainty in δR is negligible, but systematics could become a problem in the near future


Local Polarimetry at PHENIX

The neutrons are measured in the PHENIX ZDCs

Spin rotators are not perfect

and leave a small transverse component

Exploit a forward neutron single spin asymmetry in transversely polarized

p+p collisions


Polarimetry at RHIC

RHIC uses 2 types of transversely polarized elastic scattering

€

p↑ + C → p + C

€

p↑ + H → p + H

•Provides fast, high statistics measurement•Gives polarization profile of beam•Analyzing power empirically determined

•Provides self-calibrating, low statistics measurement •Analyzing power unknown •Provides calibration for the p+C polarimeter

A tour de force measurement as this is the first time bunch-by-bunch measurements were performed at

any high energy collider

€

p + H↑ → p + H


η Asymmetry

Because the eta peak sits on a background we need to

subtract a possible background asymmetry

€

r =NBG

NBG + NS

€

ALLη +BG = (1− r)ALL

η + rALLBG

700-800 MeV

300-400 MeV€

ALLη =

ALLη +BG − rALL

BG

1− r

The background asymmetry is estimated

using the blue sideband regions and subtracted from the

peak asymmetry

What can we learn from η production in proton-proton collisions? Joe Seele MIT and University of Colorado.

Documents