First measurement and future prospects for measure-ments of single spin asymmetries in W production with the PHENIX experiment at RHIC

2010-07-21 1

Young Jin Kimfor the

PHENIX Collaboration

SPIN-Praha-2010

Outline

2010-07-21 2SPIN-Praha-2010

•Physics motivation Proton spin Single spin asymmetry of W bosons

•RHIC Accelerator PHENIX experiment

•Single spin asymmetry of W bosons from RHIC Run-9

•PHENIX forward moun trigger upgrade MuTRG RPC Projection

•Summary

Physics Motivation

2010-07-21 3SPIN-Praha-2010

Proton Structure

2010-07-21 4SPIN-Praha-2010

u

ud

zLG ΔΔΣ21

21

quark spin gluon spin orbital angular momen-tum

Spin ½ quarksSpin ½ anti-quarksSpin 1 gluons

With three valance quarks:• We understand

ChargeProton = 3 valance quarks + sea quarks + glu-ons

Momentum: carried by both quarks and gluons

• What about proton spin?Quark spin = ½

Spin Crisis:Quark Spin = 0.24 0.12 0.01EMC: Phys. Lett. B 206, 364 (1988); Nucl. Phys. B 328, 1 (1989)

Quark and Gluon Distribution Functions

2010-07-21 5SPIN-Praha-2010

)(xu

)(xd)(xG

)(xq

M. Hirai, S. Kumano, N. Saito, Phys. Rev. D 74, 014015 (2006)

Spin dependent quark distribution functions by the QCD analysis of DIS data

: well known

: only known with large uncertainties

Spin dependent gluon distribution function by the QCD analysis of DIS data

G=0.471.08 (largely unknown)

The latest extraction of polarized parton distributions (DSSV):D. de Florian, R. Sassot, M. Stratmann, W. Vogelsang, Phys. Rev. D 80, 034030 (2009)

)(

)(

xq

xq

proton

quark

pp

x

Semi-Inclusive DIS: e+p e+ h +X Quark & Anti-Quark Helicity Distributions

2010-07-21 6SPIN-Praha-2010

How well do we know hadron fragmentation functions ?

new analysis of e+e- data, Hirai, Kumano, Nagai, Sudohep-ph/0612009, INPC 2007 Extraction based on LO-QCD inspired

QPM at low Q2

Unspecified uncertainties from possible QCD NLO contributions and/or sub-leading twistLarge uncertainties from fragmentation functions

Semi-Inclusive DIS: e+p e+ h +X Quark & Anti-Quark Helicity Distributions

2010-07-21 7SPIN-Praha-2010

How well do we know hadron fragmentation functions ?

new analysis of e+e- data, Hirai, Kumano, Nagai, Sudohep-ph/0612009, INPC 2007 Extraction based on LO-QCD inspired

QPM at low Q2

Unspecified uncertainties from possible QCD NLO contributions and/or sub-leading twistLarge uncertainties from fragmentation functions

Recent progress on fragmentation functions (DSS):D. de Florian, M. Stratmann, R. Sassot, Phys. Rev. D76, 074033 (2007)

For the first time: Inclusion of e+e-, SIDIS and pp data in global analysis

Still significant uncertainties in s, heavy flavor and gluon fragmen-tation function

Several data sets at low Q2: validity of analysis at leading twist OPE and NLO QCD

Parity Violating W Production

2010-07-21 8SPIN-Praha-2010

• W boson production in p-p collision─ Large scale (~mW)─ flavor combination is almost fixed

• flavor separated measurement─ helicities of interacting partons are fixed

• spin measurement─ W → lepton decay includes no fragmen-

tation process• direct and clean measurement• Important to control hadron decays

and beam backgrounds

• Asymmetry measurement of W bo-son production is ideal method

─ High luminosity and longitudinally polar-ized p+p collisions at 500 GeV (integrated L = 300 pb-1, polarity = 70%)

Single Spin Asymmetry of W

2010-07-21 9SPIN-Praha-2010

WduAL

WudAL

dunpolarize

polarizedL

polarizeddunpolarizepolarizeddunpolarize

A

asymmetryspin Single

,

u

uAd

dAxx

ddA

uuAxx

xdxuxuxd

xdxuxuxdA

xuxdxdxu

xuxdxdxuA

WW

LLab

LLba

baba

babawL

baba

babawL

,:

,:

)()()()(

)()()()(

)()()()(

)()()()(

and for asymmetry spin Single

•Forward/backward region selects valance/sea quark helicity contribution•Flavor decomposed proton quark polarization can be probed using the single spin asymmetry of W+ and W- in longitudinally polarized p+p collisions

Relativistic Heavy Ion Collider (RHIC) and PHENIX

2010-07-21 10SPIN-Praha-2010

STAR

PHENIX

PHOBOSBRAHMS

RHIC, lenght: 3.83 km

Relativistic Heavy Ion Collider (RHIC)

2010-07-21 11SPIN-Praha-2010

Relativistic Heavy Ion Collider (RHIC)

2010-07-21 12SPIN-Praha-2010

RHIC is world’s unique polarized proton col-lider

PHENIX Detector

2010-07-21 13SPIN-Praha-2010

Central Arms:hadrons, photons, electrons |η| < 0.35 Δφ = π (2 arms x π/2)

Muon Arms: muons 1.2 < |η| < 2.2 Δφ = 2π

Global Detectors:Beam-Beam Counter (BBC)Zero Degree Calorimeter (ZDC)

Philosophy (initial design): High rate capability & granularity Good mass resolution & particle ID Sacrifice acceptance

e+/-

μ+/-

Expected Single Spin Asymmetry of W

in the PHENIX Acceptance

2010-07-21 14SPIN-Praha-2010

Central Arms: W measurement at RHIC Run-9Muon Arms: Forward muon trigger upgrade is ongoing and ready for RHIC Run-11

- - + +e- e+

W measurement at PHENIX Central Arms

2010-07-21 15SPIN-Praha-2010

W e

2010-07-21 16SPIN-Praha-2010

• RHIC Run-9: First 500 GeV Run, March 17 - April 13, 2009• Machine development in parallel with physics running to increase luminosity, po-

larization , reduce backgrounds• Integrated luminosity (with vertex cut) = 8.57 pb-1

• Polarization is <P> = 0.39±0.04 (scale uncertainty)

e

•EMCal (x~0.01x0.01) 4x4 tower sum trig-ger

•±30 cm vertex cut•High energy EMCal clusters matched to charged track•Loose timing cut to reduce cosmic ray back-ground•Loose E/p cut

Fully efficient at 12 GeV/c

Effect of Cuts

2010-07-21 17SPIN-Praha-2010

•Smooth spectrum of EMCal clusters after removing bad towers•Have a good track pointing to an EMCal cluster•Timing within start time of collision

• Reduces out of time backgrounds (cosmics) by ~80%•E/p cut

ClustersGood track matchTOF cutE/p

•QCD provides the most obvious background (W. Vogelsang)•Not shown here but very important

• Cosmics and photons (from meson decays and direct), which can have accidental matches to tracks or conversions

•c/b relatively small above 30 GeV, calculated at FONLL (Matteo Carciari)•Z/* background is estimated from PYTHIA (~1 count is expected in Run09).•We is also small

Expected W e Decay Signal

2010-07-21 18SPIN-Praha-2010

positive negative

RHICBOS: Nadolsky and Yuan, Nucl. Phys. B 666:31-55,2003

Comparison To Measured Spectra

2010-07-21 19SPIN-Praha-2010

Data and MC driven BG estimation:EMCal cluster distribution after subtracting cosmic background (Conversion + Accidental) Tracking Acceptance

+ (NLO Hadrons thru Geant + FONLL c/b)´ Normalization from fit to 10-20 GeV

+ -

•The same scale factor for PYTHIA was used for W/Z shape•W- e- signal has fewer counts than W+ e+ signal as expected

Isolation Cut

2010-07-21 20SPIN-Praha-2010

E < 2GeV

•Signature of a W event is that it is isolated•Sum up energy in a cone around electron and in cone on opposite hemisphere

+ -

•90+% of signal is kept (red histograms)•Factor ~5 reduction in jet dominated region

Raw Asymmetries (e+)

2010-07-21 21SPIN-Praha-2010

Background SignalpT Range (GeV/c) 12-20 30-50Raw Asymmetry 0.0350.047 -0.290.11

PHENIX AL(W+ e+)

2010-07-21 22SPIN-Praha-2010

• Using average polarization 0.39±0.04, we get

• Asymmetry is corrected for dilution by Z and QCD backgrounds

ALW+ = -0.83 ± 0.31

PHENIX Forward Muon Trigger Upgrade

2010-07-21 23SPIN-Praha-2010

Current PHENIX Muon Arms

2010-07-21 24SPIN-Praha-2010

Muon ID(MuID)

Muon Magnet

Muon Tracker (MuTr)

Absorber

Muon

Hadron

•Acceptance 1.2 < |η| < 2.2 Δφ = 2π

•MuTr 80 cm of steel & Copper absorber 3 stations of cathode strip chamber

3gap + 3 gap + 2 gap Each gap has non-stereo and stereo

strip plane Total 16 cathode strip planes

•MuID 5 gaps of larocci tube in x and y directions Total 80 cm of steel plates

•Muon Tracking and triggering

Tracking uses hit positions from each station Most hadrons are absorbed before reaching

MuID last gap Rejection power of the current MuID based

first level trigger ~ 100

PHENIX Muon Arms

2010-07-21 25SPIN-Praha-2010

Current Muon Trigger at PHENIX

Design Luminosity√s = 500 GeV σ=60mbL = 1.5x1032cm-2s-1

(L = 3.0x1032cm-2s-1, after luminosity upgrade)Total interaction rate = 9MHz (18 MHz)DAQ LIMIT = 2kHz (for Muon Arm)Required RF = 4500 (9000)MuID LL1 (trigger) = RF < 100

W

simulated muons into Muon Arms(2000 pb-1, with PYTHIA 5.7)

•In order to measure W at 500 GeV, we need a first level rejection factor 9000

•The current muon trigger accepts muons above 2 GeV/c and current DAQ cannot take full rate at 500 GeV

We need a momentum sensitive muon trigger and fast readout electronics for muon tracker

NSF funded trigger RPC stations JSPS funded MuTr front end electronics

Additional RPC and MuTr information in the trigger makes it possible to measure the sagitta of charged particles online

The upgraded first level trigger will select events with muons of high transverse momentum pt 20 GeV/c from W boson decay

PHENIX Muon Arms

2010-07-21 26SPIN-Praha-2010

Current Muon Trigger at PHENIX

W

simulated muons into Muon Arms(2000 pb-1, with PYTHIA 5.7)

PHENIX Muon Trigger Upgrade

2010-07-21 27SPIN-Praha-2010

RPC 1

RPC 3 RPC 3

MuTRG MuTRG

Adding 35 cm Fe absorber:reduce the lower momentum hadron punch through(S/B=3:1 instead of 1:3 without ab-sorber)

MuID Trigger (existing):Selecting momentum above 2 Gev/c

MuTRG (fully installed):Fast selection of high momentum tracks

RPC (being installed):Provide timing information and rough position infor-mation

PHENIX Muon Trigger Upgrade

2010-07-21 28SPIN-Praha-2010

RPC3

RPC1

• Momentum selectivity through online sagitta measurement Uses MuTr station 1, 2, 3 and RPC station 1, 3 Implement trigger using fast, parallel logic on FPGA’s Beam related background rejected by RPC timing information

PHENIX Muon Trigger Board Upgrade (MuTRG)

2010-07-21 29SPIN-Praha-2010

MuTRGADTX

MuTRGMRG

Level 1TriggerBoard

MuTrFEE

Resistive Plate Counter(RPC) (Φ segmented)

B

2 planes

5%

95%

Trigger

Trigger

Trigger

Interaction Region Rack Room

Optical

1.2Gbps

Amp/Discri.Transmit

DataMerge

MuTRG

RPCFEE

Trigger events with straight track(e.g. Dstrip <= 1)

RPC / MuTRG data arealso recorded on disk

PHENIX MuTRG Installation

2010-07-21 30SPIN-Praha-2010

MuTRG RackMuTRG ADTX

•Completed MuTRG installation and commissioned MuTRG•Evaluation of commissioning data is in progress

MuTRG Efficiency for Track

2010-07-21 31SPIN-Praha-2010

0.939

0.874

8.5GeV/c

12.2GeV/c

Red : Δs=1Black : Δs=0

= (# of MuTRG)/(# of MuTr hits)

•Sufficient efficiency can be achieved in MuTRG•Turn on point is enough low with Δs=0(12.2 GeV/c in p, ~6 GeV/c in pT)

PHENIX Muon Trigger RPC

2010-07-21 32SPIN-Praha-2010

Time resolution 3 nsAverage cluster size 2 strips

Efficiency 95%Rate capability 0.5 kHz/cm2

Number of streamers 10 %

PHENIX RPC Detector Require-ment

Characteristics of double gap RPC in avalanche mode Fast response: Suitable for the trigger device Good time resolution: 1 - 2 ns for M.I.P Good spatial resolution: typically ~ cm→ Determined by the read-out strip width and cluster size Efficiency: > 95 % for M.I.P Rate capability: 1 - 2 kHz/cm2

Low cost: bakelite (resistivity 1010 - 1011 cm) Noise rate: 0.5 - 5 Hz/cm2 for 1010 cm Typical strip multiplicity: 1.5 - 3 Typical gas mixture

95% C2H2F4 (Tetrafluoroethane, R134A, base gas for avalanche mode RPC) + 4.5% i-C4H10 (isobutane, photon quencher) + 0.5% SF6 (electron quencher)

20 – 40% relative humidity

PHENIX Muon Trigger RPC

2010-07-21 33SPIN-Praha-2010

RPC3station

RPC3station

RPC1(a,b)

stationhalf octant

RPC3

Use already established CMS Bakelite RPC technology and expertise

RPC 3ARPC 1A

PHENIX RPC Module QA

2010-07-21 34SPIN-Praha-2010

RPC test stand event display

cosmic ray trigger scintillators

RPC readout strip planes

cosmic ray trajectory

Raw TDC :1 unit = 100 ns/44 = 2.41 ns

PHENIX RPC Module Noise Rate

2010-07-21 35SPIN-Praha-2010

# of strips vs. noise rate• all 48 modules• H.V = 9.5 kV• FEE Threshold = 160 mV <noise rate> = 0.52 Hz/cm2

7 strips are over 10 Hz/cm2

# of strips vs. noise rate• Individual module• H.V = 9.5 kV• FEE Threshold = 160 mV

Black = module ABlue = module BRed = module C

All results are compliant with PHENIX RPC require-ments

PHENIX RPC Half Octant Assembly

2010-07-21 36SPIN-Praha-2010

RPC half octant placed on half octant assembly table at the RPC factory in BNL• All half octant parts are made of aluminum• Half octant contains three RPC modules• Final cable routing during half octant QA

PHENIX RPC Half Octant Noise Rate

2010-07-21 37SPIN-Praha-2010

•<noise rate> = 0.37 Hz/cm2

•Modules are in Faraday cage

PHENIX RPC-3 North Installation

2010-07-21 38SPIN-Praha-2010

PHENIX RPC-3 North Installation Completed

2010-07-21 39SPIN-Praha-2010

PHENIX RPC-3 North Integration Completed

2010-07-21 40SPIN-Praha-2010

RPC TDC RackRPC Integration completed

•RPC-3 north was commissioned in Run-10•Evaluation is in progress

PHENIX RPC Prototypes during RHIC Run-9

2010-07-21 41SPIN-Praha-2010

MuTRGMuTRG

PrototypeRPCs

RPC Prototypes

Absorber Prototype

Partially installed MuTRG

Timing Information from PHENIX RPC Prototype

2010-07-21 42SPIN-Praha-2010

•The peak below zero: tracks coming from the collision or the outgoing beam•The other peak above zero: background generated by incoming beam•Separation of the peaks: ~ 12 ns•Width for each peak: 3 ns•Individual RPC timing results required a timing correction from reference detec-tor

Muon cuts: tracks close to RPCs, at least 3 GeV momentum

~ 4.1ns ~ 5.3ns

No correction for collision vertex 30 cm ~ 2nsec.

PHENIX Muon Trigger Upgrade Schedule

2010-07-21 43SPIN-Praha-2010

•PHENIX RPC-3 south production completed•PHENIX RPC-3 south half octant QA is underway•PHENIX will be ready for W measurement for RHIC Run-11•Completion of RPC Project expected in 2011

RPC Installation for Station 3 Southby October+Absorber Installation

RPC Installation for Station 3 North by Oc-tober

RPC Production for Station 1 North by December

RPC Installation for Station 1 dur-ing shutdown

RHIC Run-11 AL of W Projection

2010-07-21 44SPIN-Praha-2010

•For RHIC Run-11, asking 10 weeks 500 GeV p+p collisions with 50% of polarizationExpecting integrated luminosity L = 50 pb-1

Exceptive Conservative

RHIC Run-11+12 AL of W Projection

2010-07-21 45SPIN-Praha-2010

•For RHIC Run-12, asking 8 weeks 500 GeV p+p collisions with 50% of polarizationExpecting integrated luminosity L = 150 pb-1 (together RHIC-11 and 12)

Exceptive Conservative

Summary

2010-07-21 46SPIN-Praha-2010

•PHENIX observed W e decay in the mid-rapidity region•First attempt to measure single spin asymmetry has detected a parity violating asymmetry leading to a preliminary value of AL•AL(W+ e+) at |ye| < 0.35 has been measured to be

Within errors, it is consistent with the predictions. •While the data sample is small, we got a first significant result• Gained experience on the backgrounds to the W in the PHENIX central arm• Learned how to handle multiple collisions from high luminosity • Learned how to handle very high momentum particles• Working on finalizing cross-sections, AL(W+), and AL(W-)

•Goal of the RHIC W program is 300 pb-1 and 70% polarization• Order of magnitude improvement in the error bars• Forward muon measurements in PHENIX will contribute significantly starting

next year with MuTRG, RPC, and absorber

ALW+ = -0.83 ± 0.31