Event reconstruction
5.5 J/ψ reconstruction
track reconstruction (nhitsTRT) plus that of TRT hits associated with the track but not used (noutliersTRT ), i.e., n =nhitsTRT+noutliersTRT .
If |η| ≥1.9 and n ≤5, this requirement is not applied. If |η| ≥ 1.9 and n >5, noutliersTRT <0.9n is required.
If two opposite sign muons in an event satisfy the above requirements, the pair is used for the J/ψ reconstruction.
5.5 J/ψ reconstruction
Before starting the J/ψ vertex fit, the invariant mass of the selected muon pairs is required to be within the range of |M0(µ+µ−)−MPDG(J/ψ)| < 200 MeV. This requirement reduces a lot of irrelevant combinations of muons. Here, the index 0 ofM0 denotes the value calculated at the perigee. This requirement is loose enough to retain almost all J/ψ signals. The invariant mass distribution is shown in Figure 5.5 and the PDG average value is indicated as the red line. A partial sample corresponding to the integrated luminosity of 11.6pb−1 is used for the plot. A clear peak corresponding to the J/ψsignal can be seen, but the sample is still contaminated with background.
) [MeV]
µ+
µ -0( M
2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500
Entries / 2 MeV
0 200 400 600 800
1000 Data 2011 L dt = 11.6 pb-1
∫
TeV = 7 s
Mass (3096.916 MeV) ψ
J/
All CB muon pairs
Figure 5.5: Invariant mass distributions of the combined muon pairs.
Track parameters such as pT, η, and ϕ of the preselected muons are fitted with a constraint of the known J/ψ mass as described in Section 4.2. The fit is carried out by assuming that the muon pair originates from a common vertex. Then, the following requirements are imposed to select a pure J/ψ sample. The requirements are
5.5 J/ψ reconstruction
• χ2m <4,
• |ηm|<2.5,
where the index m denotes a value after the ‘m’ass constraint. Figure 5.6 shows the invariant mass distributions evaluated by using the original track parameters of the muon pairs. The distributions are separately shown for the muon pair samples with χ2m > 4 or |ηm| > 2.5, and χ2m < 4 and |ηm| < 2.5. The χ2m > 4 or |ηm| > 2.5 pairs are rejected as the background. The obtained signal (χ2m < 4 and |ηm| < 2.5) distribution is compared with simulation results in Figure 5.7. The left side plot shows the comparison with the Ω−b simulation and the right side one shows that with the Ξ−b simulation. The simulations are normalized to the number of entries. The distributions are basically in good agreement with a small shift of the peak position.
The peak position is 3095.1 ± 0.2 MeV for the data, and 3101.1 ± 0.1 MeV and 3100.8±0.1 MeV for the Ω−b and Ξ−b simulations, respectively. These mass differences are used for estimating the systematic uncertainties in Subsection 6.1.4.
) [MeV]
µ+
µ -0( M
2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500
Entries / 2 MeV
0 200 400 600 800 1000
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
> 4
m
χ2
Vertex
| > 2.5 ηm
or |
) [MeV]
µ+
µ -0( M
2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500
Entries / 2 MeV
0 200 400 600 800 1000
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
< 4
m
χ2
Vertex
| < 2.5 ηm
|
Figure 5.6: Invariant mass distributions of the combined muon pairs satisfying the quality requirements.
Figure 5.8 shows the pTm, ηm, and ϕm distributions of the muons composing the selected J/ψ candidates. There is no significant bias depending on the sign of the charge. The effective threshold of the selected muons is approximately 4 GeV due to the trigger. The dips at |ηm| ∼ 0 and 1 are due to the boundaries in the muon detector system and those at −2.5.ϕm .−1.5 are due to the supporting columns of the ATLAS detector. In both angular distributions, the shapes of µ+ and µ− are slightly different. This difference reflects the effect of the magnetic field. Figure 5.9 shows the χ2m, pTm, ηm, and ϕm distributions of the J/ψ candidates. The shaded
5.5 J/ψ reconstruction
) [MeV]
µ+
µ -0( M
2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500
Normalized Entries / 10 MeV
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
< 4 χ2
Vertex
| < 2.5 η0
| Data
sample
-Ωb
MC
) [MeV]
µ+
µ -0( M
2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500
Normalized Entries / 10 MeV
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
< 4 χ2
Vertex
| < 2.5 η0
| Data
sample
-Ξb
MC
Figure 5.7: Invariant mass distributions of the combined muon pairs in the data and the MC samples.
[MeV]
pTm
µ±
0 5000 10000 15000 20000
Entries / 100 MeV
0 200 400 600 800 1000 1200 1400
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
µ
-µ+
[MeV]
pTm
µ±
0 5000 10000 15000 20000
Entries / 100 MeV
0 500 1000 1500 2000 2500
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
high pTm
µ± Tm low
± p µ
ηm
µ±
-3 -2 -1 0 1 2 3
Entries / 0.1
0 500 1000 1500 2000 2500
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s µ
-µ+
φm
µ±
-3 -2 -1 0 1 2 3
Entries / 0.1
0 200 400 600 800 1000
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
µ
-µ+
Figure 5.8: pTm, ηm, and ϕm distributions of the muons composing the J/ψ candidates.
5.5 J/ψ reconstruction
m
χ2
candidates vertex ψ
J/
0 1 2 3 4 5 6 7 8 9 10
Entries / 0.1
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
[MeV]
candidates pTm
ψ J/
0 5000 10000 15000 20000 25000 30000 35000 40000
Entries / 100 MeV
0 100 200 300 400 500 600 700
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
ηm
candidates ψ
J/
-3 -2 -1 0 1 2 3
Entries / 0.1
200 400 600 800 1000 1200 1400 1600 1800 2000 2200
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
φm
candidates ψ
J/
-3 -2 -1 0 1 2 3
Entries / 0.1
0 100 200 300 400 500 600 700 800 900
2011 Data
L dt = 11.6 pb-1
∫
TeV = 7 s
Figure 5.9: χ2m,pTm,ηm, andϕmdistributions of theJ/ψcandidates.
Those pairs havingχ2m>4 are included in the χ2m distri-bution.
5.5 J/ψ reconstruction Figure 5.10 shows the distributions of the angular distance (dR) between the truth and the reconstructed J/ψ candidates in the MC simulations. Here, the truth values correspond to the production point in the event generation and dR is defined as dR ≡ √
(ηtruth−ηreco.)2+ (ϕtruth−ϕreco.)2. In the distributions, almost all of the reconstructed J/ψ candidates have dR < 0.01 and this value corresponds to the difference of 0.57◦ (dR ∼ dϕ = 0.01[rad]). When dR < 0.05 is required as a condition of matching between the truth J/ψand the reconstructedJ/ψ candidates, the efficiencies and the distributions in pT, η, and ϕ are shown in Figure 5.11. The left side plots are the Ω−b simulation and the right side ones from the Ξ−b simulation.
The efficiencies which depend on η and ϕ are roughly flat except for regions of the detector boundaries and columns, and the efficiency in pT depends on the pT value;
the efficiency increases from 0 to ∼0.3 in the pT range of 8 < pT <14 GeV, and the efficiency of approximately 0.3 is kept in 14< pT <30 GeV.
dR
0 0.01 0.02 0.03 0.04 0.05 0.06
Entries / 0.01
0 2000 4000 6000 8000 10000 12000 14000 16000
MC Simulation Sample
-Ωb
# Overflow 286
dR
0 0.01 0.02 0.03 0.04 0.05 0.06
Entries / 0.01
0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000
22000 MC Simulation
Sample
-Ξb
# Overflow 213
Figure 5.10: dR distributions between the truth J/ψ and the recon-structedJ/ψ candidates.
In the MC simulations, 61,377 and 85,827 J/ψ candidates are matched in the Ω−b and Ξ−b MC samples, respectively. Also, the corresponding J/ψ reconstruction efficiencies are 16.7% and 16.6%. In total, 1.5×107 events are selected at this stage from the data corresponding to the full luminosity.
5.5 J/ψ reconstruction
[MeV]
truth pT
ψ J/
0 10000 20000 30000 40000
Efficiency
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Entries / 1000 MeV
0 10000 20000 30000 40000 50000 60000 70000 MC Simulation
Sample
-Ωb Efficiency
Truth
Matched truth (dR < 0.05)
[MeV]
truth pT
ψ J/
0 10000 20000 30000 40000
Efficiency
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Entries / 1000 MeV
0 20 40 60 80 100
103
× MC Simulation
Sample
-Ξb Efficiency
Truth
Matched truth (dR < 0.05)
η truth ψ J/
-3 -2 -1 0 1 2 3
Efficiency
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Entries / 0.1
0 2000 4000 6000 8000 10000 12000 14000 MC Simulation
Sample
-Ωb Efficiency
Truth
Matched truth (dR < 0.05)
η truth ψ J/
-3 -2 -1 0 1 2 3
Efficiency
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Entries / 0.1
0 2000 4000 6000 8000 10000 12000 14000 16000 18000 MC Simulation
Sample
-Ξb Efficiency
Truth
Matched truth (dR < 0.05)
φ truth ψ J/
-3 -2 -1 0 1 2 3
Efficiency
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Entries / 0.1
0 1000 2000 3000 4000 5000 6000 7000 8000 MC Simulation 9000
Sample
-Ωb Efficiency
Truth
Matched truth (dR < 0.05)
φ truth ψ J/
-3 -2 -1 0 1 2 3
Efficiency
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Entries / 0.1
0 2000 4000 6000 8000 10000 12000 MC Simulation
Sample
-Ξb Efficiency
Truth
Matched truth (dR < 0.05)
Figure 5.11: J/ψ reconstruction efficiencies.