Transcript 20120301_BhamAtlasWeekly_Diff

Update on Diffractive Dijet
Production Search
Hardeep Bansil
University of Birmingham
Birmingham ATLAS Weekly Meeting
01/03/2012
Contents
•
•
•
•
•
Theory & Motivation
Analysis
Comparison with SM inclusive jets analysis
Forward gap distributions
Next steps
2
Diffractive dijets
• Look for single diffractive events (pppX)
– Involve a rapidity gap due to colourless
exchange with vacuum quantum numbers:
“pomeron”
• Then look for dijet system within X
– Hard diffraction
• Sensitive to the diffractive structure
function (dPDF) of the proton
• Studied at HERA and Tevatron
– At Tevatron, ratio of yields of SD to inclusive
dijets ≈ 1%
– Likely to be smaller than this at LHC
3
Motivation
• Understand the structure of the diffractive exchange by
comparison with predictions from electron-proton data and
be able to get a measure of FDjj
• Measure the ratio of the single diffractive to inclusive dijet
events
• Gap Survival Probability – the chance of the gap between the
intact proton and diffractive system being lost due to
scattering (affects measured structure function)
– Tevatron have Gap Survival Factor of 10 smaller than H1 predictions
– Khoze, Martin and Ryskin predict LHC to have GSF around 30
Rescatter
with p?
Comparison of
Tevatron results to
H1 predictions
(ξ)
Gap destruction
by secondary
scattering
4
Gap Finding Algorithm
• Gap finding based on Soft Diffraction analysis
– Divides calorimeter into 49 rings of 0.2 in η
– Identifies calorimeter cells where energy significance (= cell
energy/noise) large enough that probability of noise cell studied in
event is small
– Where no cells in ring found above ESig threshold  ring is ‘empty’
– Full details in blue box
• Determine the size of the biggest forward gap
Detector gap definition
•Calorimeter: no cell above threshold E/σ
> Sth - probability of noisy cell in ring
smaller than 10-4 (electronic noise only,
no pile-up environment), cluster that cell
came from has ET > 200 MeV
Example Single Diffractive Topology
ΔηF:3.4
|ηStart|:4.9
•Tracker: no good track above pT > 200
MeV, |η| < 2.5
Truth gap definition
•No stable particle above pT > 200 MeV
5
Analysis
• Using Athena version AtlasProduction-16.6.4.5
• Using L1Calo stream data10 period A and B ESDs
– Run 153030 (period B) excluded (due to noise bursts)
– Total ∫L dt = 7.22 nb-1 - calculated using online iLumiCalc tool with
L1_MBTS_2 live trigger and L1_J5 ref. trigger
• Average <μ> for selected runs < 0.15  currently ignore pile-up
• Anti-Kt jets with R=0.6 or R=0.4:
–
–
–
–
–
Require >= 2 jets in event passing loose jet quality cuts
ET Jet1,2 |η| < 4.4
ET Jet1 > 30 GeV, ET Jet2 > 20 GeV
Cut values suggested based on 2010 SM dijet analysis
Jet ET Jet2 limit and η cuts based on 2010 jet energy scale systematic
• Ask for a forward gap: |ηstart| = 4.9, ΔηF ≥ 2.0
6
Monte Carlo for Analysis
• Use POMWIG generator for Single Diffractive Dijets
– Modifies HERWIG ep photoproduction so ee+γ becomes pp+IP
– No rapidity gap destruction built in
– Generates QCD 22 process within diffractive system in different pT
ranges (8-17, 17-35, 35-70, 70+ GeV) for SD (system dissociating in ±z
direction)
– Using MC samples generated by myself (4000 events of each
POMWIG sample)
• Use PYTHIA 6 and PYTHIA 8 Dijet samples as Non Diffractive
(inclusive) Dijets
– Made in same pT ranges as Pomwig (8-17, 17-35, 35-70, 70-140 GeV)
– PYTHIA 8 J0 sample (8-17 GeV) generated by myself (5000 events) as
only available on Grid with pile-up present
• Prague and Birmingham groups now looking to make new
official samples
– Derived AOD format where clusters have most significant cell
information (no longer working directly with cells)
– Apply filters to boost statistics at higher gap sizes
7
Normalisation Issues
•
Previously finding big disagreement between data and Pythia 6/8 samples
(factor 10, expect it to be 2 at most)
Uncorrected Gap Size
Distribution
POMWIG SD, PYTHIA 6
& 8 Jets weighted
relative to luminosity
of data runs used and
then plotted against
MinBias Data
(triggered on MBTS_2)
•
•
•
•
Ldata – Luminosity of data
Ngen – Number of events generated
σgen – Csx of events generated
Nrec – Number of events reconstructed
to run over
•
Ngen , Nrec,σgen all taken from AMI, checked multiple times but still get big
factor difference between data and MC
Weight 
Ldata
 N rec N gen 



 N gen  
gen 

8
Comparison with SM10 Inclusive Jet & Dijet Analysis
•
Identify source of disagreement by comparing my results to full 2010 (37
pb-1) Standard Model inclusive jet & dijet analysis paper
•
Compare double differential cross sections to AntiKt6 dijets (plots of mjj
with separate y*=|∆y|/2 ranges)
Take lowest y* range ( 0.0 < y* < 0.5)
•
•
•
Can see good agreement between SM analysis and Pythia 6/8 samples
(statistics permitting) but data sample is not correct
Consistent results when using full y* ranges, also trying AntiKt4/6 jets and
with inclusive jets and dijets
9
Comparison with SM10 Inclusive Jet & Dijet Analysis
•
Change data streams and triggering strategy to use L1Calo stream and
L1_J5 as main trigger
–
–
–
L1_J5 unprescaled across Periods A and B
However may not be fully efficient for jets at low pT
Smaller number of events across L1Calo but larger fraction of the events containing
dijets passing the cuts
Trigger
Selection
L1Calo
stream Ak4
L1Calo
stream Ak6
MBTS_2
296858
563794
J5
293468
513628
Trigger
Selection
MinBias
stream Ak4
MinBias
stream Ak6
MBTS_2
44471
86848
J5
42255
73583
•
Agreement is a lot better now between L1Calo data and SM analysis
(but not quite what SM analysis did)
10
More like SM10 Inclusive Jet & Dijet Analysis?
•
•
•
•
•
One option is to use both L1Calo and MinBias stream to get full statistics
but avoid overlap of events within streams
(L1Calo stream & J5) + (MinBias stream & MBTS_2 & !J5)
Trigger
Strategy
L1Calo
stream Ak4
L1Calo
stream Ak6
MinBias
stream Ak4
MinBias
stream Ak6
Combined
streams Ak4
Combined
streams Ak6
MBTS_2
296858
563794
44471
86848
341329
650642
J5
293468
513628
42255
73583
335723
587211
No Overlap
293468
513628
2459
13689
295927
527317
In SM analysis, do per-jet triggering (pT & y dependent)
Took a long time for Prague to implement this
Will need to convince Soft QCD
group that this is more
complicated than needs to be
Could just use MBTS_1 or
MBTS_2 and on both
streams and do some clever
event overlap myself
11
Uncorrected Gap Size Distribution
POMWIG SD,
PYTHIA 6 & 8 Jets
weighted relative
to luminosity of
data runs used and
then plotted
against L1Calo
Data
L1Calo Data
Pomwig SD
Pythia 8 Jets
Pythia 6 Jets
Drop in number of
events with ΔηF ≥ 6,
cuts into phase space
Biggest ND
contribution at
small ΔηF
•
•
Data agrees much better with Pythia distribution
Ratio of MC to Data suggests a GSF of 3 in majority of bins (prev. ≈20)
•
By ΔηF of 6, ξ = 10-4.5  MX = 39.4 GeV – cut out phase space for
producing pair of 20 GeV jets so get drop in events after this point
•
Dijet samples lacking statistics at higher gap sizes with new gap algorithm
– Still observe forward gaps with sizes of 4, 5, 6 in PYTHIA 6/8 but errors large
12
First Bin Scaling of Gap Distribution
• POMWIG SD, PYTHIA 6 & 8 Jets weighted relative to
luminosity of data runs used and then plotted against L1Calo
Data (all scaled by first bin)
L1Calo Data
Pomwig SD
Pythia 8 Jets
Pythia 6 Jets
Before forward gap cuts, 20 GeV
• Before gap cuts, observe that PYTHIA 8 is best describing
L1Calo data (until larger forward gap sizes)
13
First Bin Scaling & Background Subtraction
• First bin scaling used and then PYTHIA 6 & 8 Jets subtracted
from L1Calo Data and plotted against POMWIG SD
• If (Data – PYTHIA) ≤ 0.0 then do not plot data point
• Best measure of determining minimum Gap Survival Factor
Data – Py6
Pomwig SD
20 GeV, L1Calo Data – Pythia 6
Data – Py8
Pomwig SD
20 GeV, L1Calo Data – Pythia 8
• Due to how PYTHIA samples describe L1Calo data, get
unusual shape in first few bins (MC/Data too large at ΔηF=2.0)
• Best opportunity for studying diffractive dijets may lie with
selected candidates having 4.0 < ΔηF < 6.0 (GSF ≥ 3 here)
14
Differential Cross Sections
• MC samples weighted to lumi of data runs - Differential cross
section as a function of forward gap size
• Some odd bins due to acceptance but otherwise time to look
at systematics
MinBias Data
Pomwig SD
Pythia 6
Pythia 8
Combined Acceptance
Weights applied to different
samples based on lumi Migrations can then cause the
acceptances to be larger or
smaller than expected
Needs more statistics
MC/Data ratio suggests GSF of approx. 3
15
Next steps
• Present work to Soft QCD Working Group
• Find out appropriate trigger strategy
• Make sense of new gap survival factors
• Get official production of MC samples in new format
• Start looking at systematic uncertainties
• Focus on publishing rapidity gap distribution cross
section measurements
16