b bbar jet cross section

Abstract

We measure the inclusive bb jet production cross section by requiring two secondary vertex tagged jets within |h|< 1.2. One of these tagged jets has to have a corrected transverse energy greater than 30GeV, the other has to have a corrected transverse energy greater than 20GeV. We use the 2D SECVTX algorithm to tag heavy flavour jets and use a fitting technique to determine the fraction of these that are true b-jets. We compare our results to Leading Order (Pythia and Herwig) and Next to Leading Order (MC@NLO) predictions.

b Flavour Jet Correction

A Pythia Monte Carlo sample, simulating 2->2 processes with a ptmin of 18GeV, is used to find the ratio of hadronic jet Et to calorimetric jet Et for b flavour jets. The ratio is found to be:

1.157+/-0.015 (Systematic Error)
This is used as the b flavour correction. The study was repeated with Herwig and the difference was used as the systematic error.

Acceptance

The acceptance is calculated using the same sample as that used for the b jet correction, and has the trigger efficiency folded into it.

A_trig(|h|<1.2,Et₁ >30GeV, Et₂ >20GeV) = 1.03 +/- 0.02 (stat. error only)
Systematic errors are calculated by modifying each jet corrections by +/- 1 sigma and calculating the percentage change in the number of reconstructed b jets. The sytematic contribution from each correction is listed in the table below.

Correction	+ Sytematic	- Sytematic
Level 1	9.3%	10.8%
Level 2	0	0
Level 3	12.8%	13.7%
Level 4	0	0
Level 5	8.1%	8.5%
b correction	3.9%	4.0%
Total	18.2	19.8

A 4.5% systematic is assigned for the uncertainty in PDF's and combined in quadrature with the jet corrections systematic. The total systematic uncertainty is +18.7% -20.3%.

The acceptance is found as a function of:

leading jet Et (gif) (eps)
the azimuthal angle between the leading jet and the second jet (gif) (eps)
the invariant mass of the leading jet and second jet (gif) (eps)

SECVTX Tagging Efficiency

The tagging efficiency is found using a electron triggered sample, where Monte Carlo templates are used to find the b Fraction of the event.

Tagging efficiency in data as a function of Et (gif) (eps)

The errors from the fit are used as a systematic error, a 3.5% systematic error is assigned for the uncertainty in the b fraction and a 7.6% systematic error is assigned for the difference in efficiency between semileptonic b decays and all b decays.

b Fraction of Sample

The b fraction is found by fitting the combined secondary vertex mass spectrum of the lead and other tagged jet with templates made from Monte Carlo.

F_b = 0.83 +/- 0.04 The error quoted is that returned for the fit, which takes into account data and Monte Carlo statistic. It is used for the statisitical error on the cross section. A systematic error of 3.0% is assigned for variations in F_b with jet Et, this is found by fitting the secondary vertex mass spectrum for the lead jet and other jet seperately and taking the difference from the combined value.

Monte Carlo Templates (gif)(eps)
Fit to Secondary Vertex Mass in Data (gif)(eps)

Cross Section

N^events	716
F_b	0.83 +/- 0.04
e_b^lead	0.31
e_b^other	0.26
A_trig	1.03
Luminosity	64.5 pb^-1
s_bb(\|h\|<1.2, E_t¹>30GeV, E_t²>20GeV)	34.5 +/- 1.8nb
s (Pythia CTEQ 5l)	38.7 +/- 0.6nb
s (Herwig CTEQ 5l)	21.5 +/- 0.7nb
s (MC@NLO)	28.5 +/- 0.6nb