1 E895 - correlation analysis - Status Report Mike Lisa, The Ohio State University E895 Motivation and Measurement Status of HBT analysis Summary and.

1

E895 - correlation analysis - Status ReportMike Lisa, The Ohio State University

• E895 Motivation and Measurement

• Status of HBT analysis

• Summary and plans

2

Lawrence Berkeley Lab D. Best, T. Case, K. Crowe, D. Olson, G. Rai, H.-G. Ritter, L. Schroeder, J. Symons, T. Wienold

Brookhaven National Lab S. Gushue, N. Stone

Carnegie Mellon University M. Kaplan, Z. Milosevich, J. Whitfield

Columbia University I. Chemakin, B. Cole, H. Hiejima, X. Yang, Y. Zhang

U.C. Davis P. Brady, B. Caskey, D. Cebra, J. Chance, J. Draper, M. Heffner, J. Romero, L. Wood

St. Mary’s College J. Kintner

Harbin Institute (China) L. Huo, Y. Liu, W. Zhang

Kent State Univeristy M. Justice, D. Keane, H. Liu, S. Panitkin, S. Wang, R. Witt

Lawrence Livermore Lab V. Cianciolo, R. Sotlz

Ohio State University A. Das, M. Lisa, R. Wells

University of Auckland (NZ) D. Krofcheck

Purdue University M. Gilkes, A. Hirsch, E. Hjort, N. Porile, R. Scharenberg, B. Srivastava

S.U.N.Y. Stony Brook N.N. Ajitanand, J. Alexander, P. Chung, R. Lacey, J. Lauret, E. LeBras, B. McGrath, C. Pinkenburg

3

Systematics/meta-analysis suggest approach to maximum AGS energy interesting...

P. Braun-Munzinger and J. Stachel,NPA606, 320 (1996)

4

Perhaps some signals only apparent near threshold

D. Rischke, NPA 610, c88 (1996)

E895 flow statusdiscussed by R. Lacey

5

Ideally, HBT gives a measure of source size

C k kP k k

P k P k

d x d x x x e

d x d x x x

Q

i k k x x

2 1 21 2

1 2

31

32 1 23

13

2 1 2

2

1

1

2 1 2 1

( , )( , )

( ) ( )

( ) ( )

( ) ( )

~( )

( ) ( )

r rr r

r r

r r

r r

r

r r r r

=⋅

= ± ⋅ ⋅ ⋅∫ ⋅⋅ ⋅ ⋅∫

= ±

⋅ − ⋅ −ρ ρρ ρ

ρ

PionSource ρ( )x

C (Qinv)

Qinv (GeV/c)

1

2

0.05 0.10

Width ~ 1/R

6

HBT systematics at AGS interesting in themselves,

& can look for suprises

“”“c”

Rischke & GyulassyNPA 608, 479 (1996)

May miss signal at “too high” Ebeam

7

HBT - another handle onmean field effects at AGS

Generated by H. Liu and S. Panitkin

8

AGS BevalacParticle reconstruction upgrades have taken huge effort,but have born fruit...

9

PID via dE/dxfor primaries

(negative particles cleaner)

2 AGeV

4 AGeV

8 AGeV

10

Strange neutrals reconstructed(& provide sensitive diagnostic of data quality)

plots from P. Chung, SUNY-SB

p + K0

11

Non-uniform trigger in dataset analysedWill be possible toselect top ~5% forall energies offline

Current analysis:2 GeV: b 0-8 fm4 GeV: b 0-8 fm8 GeV: b 0-3 fm

Otherwise seems OKe.g. log increase ofmultiplicity with Ebeam:Ebeam Ebeam

Mmax Mmax + 50M

max Mmax + 15

12

Singles coverage for pions

13

Large acceptance many -

But...phase space meansmost are at large Q

Background (denominator)generated with standardevent-mixing (15 previous)

4 GeV central

14

Pairwise cuts to remove track splitting effects“Raw” correlation functionshows encouraging structureat low Qinv

Simulations: requirement that> 50% of track is seenkills truly found pairs.

(Au+Be event)

15

Pairwise cuts, cont’Track-splitting virtually eliminated by pairwise cut:

require that sum of % track seen > 100% (applied to “real” and “mixed” pairs)

Next low-Q problem:track-merging.

16

Merging effect reduced by cut on projected seperation at exit of TPC.

Real pairs

Mixed pairs

17

Require particles to exit TPC 10 cm apart.

0 cm cut

5 cm cut

10 cm cut

15 cm cut

18

Overview of E895 HBT AnalysisOverview of E895 HBT Analysis

raw data

(pass1)

TRKS

HBT_SWHBT_EVENT_CUTHBT_TRK_CUTHBT_PAIR_CUT

• Embed MC pairs into raw data• perform pass1• correlate embedded, extracted particles

fit of singles distribution

generation of MC pairs(kuip macro files)

EMBED_PARTSTRKS

AM_PID AM_HBT

histogramsacceptance correctionscoulomb correctiondiagnostic ntuplescorrelation functions

19

Corrections - I

Ideally... C k kP k k

P k P kP k k

P k ktrue

mixed2 1 2

1 2

1 2

1 2

1 2

( , )( , )

( ) ( )( , )( , )

r rr r

r rr rr r=

⋅=

“Background” pair distribution contains all physicsand detector effects except for the BE symmetrization

Well-known deviation from this is due to final-state Coulomb repulsion...

Approximate correction - Gamow factor:

G Qe

m e

Qinvinv

( ) ;=−

=1

η ηη

h

(Better to do full Coulomb integration)

20

Corrections - IIDetector acceptance effects are more subtle, especially with a tracking detector

Original pion pair

k1

k2

MC Scattering

Pattern recognition

Digitization, thresholds

Pixel noise

Measuredparticle(s)

k1’, k2

’, (k3’...)

Track merging and splitting and momentum resolution and distortion.

Hit the low-Q pairs hardest, and affect the correlation signal significantly.

Effects depend on k1, k2 (six-dimensional!), as well as track and pair cuts!!!

Correcting for or minimizing these 2-particle effects requires detailed simulation.

21

Generating the Acceptance /Resolution Correction - I

C2(ideal)

C2(reconstructed)=

R(k1,k2)B(k1,k2)

R(k1,k2)B(k1,k2)

Kacceptance =

B(k1,k2) =d6N

d3k1d3k2

R(k1,k2) =d6N

d3k1d3k2

• C2(k1,k2)

for someset of cuts:

only phase space (k) cut applied(no track quality or 2-track cuts)

B(k1,k2) =d6N

d3k1d3k2

R(k1,k2) =d6N

d3k1d3k2

• C2(k1,k2)

apply same track and pair cuts as applied to data

(weighting by C2(k1,k2) implies foreknowledge of correlation function iterative approach)

22

Understanding close pairs

•Close pairs are embedded into real data events at pixel levelwith measured momentum distribution, to get correct noisetrack density environment•Full event reconstruction run•Gives momentum distortions, pair loss...

23

Resolution from the 4 GeV simulations...

10 MeV/c resolution(includes MCS)

finite resolution + phase spacegive Q distortion at low Q

(Qin > 40 MeV/c)

24

Pair loss from 4 GeV simulations

Pair loss constant above 50 MeV/c(statistical loss of single track)

Single pair in...

Lost pair Split track

25

Finally: “Correction to the Coulomb Correction”

In the measured ratios, we apply the Coulomb correction (currently the Gamowcorrection) according to the measured Q, not the true Q.

With the simulated pairs, we have the true and reconstructed momenta, socan account for this.

Then, the full acceptance/resolution correction function is:

R(k1,k2)B(k1,k2)

R(k1,k2)B(k1,k2)

G(k1,k2)

G(k1,k2)

26

Corrections for 10 cm exit seperation

27

Corrections for 2 cm exit separation

28

Acceptance/resolution well understood & accounted for

2 cm exit separation cut10 cm exit separation cut

Different cuts give very different raw correlation functions.But corrected correlation function is robust.

2 GeV results2 GeV results

29

Data points consistent - fits are sensitive

30

4 GeV results stable (and reasonable) as well

10 cm exit separation cut 2 cm exit separation cut

31

8 GeV results not stable or reasonable(under study)

10 cm exit separation cut 2 cm exit separation cut

32

SummarySummary• E895 can measure low-Q correlations well

• Difficulties of close pairs (splitters/mergers) largely addressed through pairwise cuts

• Detailed simulations generate corrections that track with cuts– These corrections are significant and important

• Different quality cuts very different measured correlation functions very different measured corrections NOT different corrected correlation functions

• Must figure out what is going on at high energy

• Multi-dimensional HBT and phase space cuts come next

(present analysis on < 5% of data)