1. Feasibility of the detection of D0 mesons in the NA61/SHINE experiment
Vertex Detector (VD) conception
D0 → K- + π+ (BR=3.87%).

2. Outline 1. Introduction → Input 2. Reconstruction → acceptance
1) physical input
2) detector
2. Reconstruction
→ acceptance
→ cut strategy
→ results on D0
3. Charged particle fluxes
4. Outer dimensions
5. Fluence estimates
6. Pixel occupancy
7. Cooling
8. Cost estimates
9. To do in perspective of application for founding (for discussion part)

3. Physical Input
AMPT predicts 0.01 of <D0> + <D0> per central Pb+Pb event.
→ this seems to be under-predicted value, e.g. PYTHIA run for N-N and scaled to central Pb+Pb gives 0.21
HDS predictions are consistent with PHYTIA, so we decided to stick with HDS predictions
We compensate low multiplicity by setting BR for D0 → K- + π+ to 1 (3.87%).
Probability of having more than 1 K-π pair from D0 decay in one central collision < 10-4
P. Braun-Munzinger, J. Stachel, PLB 490 (2000) 196

4. NA61/SHINE detector – top view
VTPC2
Vertex Detector (VD)

5. Design of the Zoom in Vertex Detector Pb target Helium Vessel
Helium Pipe
VDS1
VDS2
VDS3
VDS4
Pb target
beam pipe
VDS Stations are located at the distance of 5, 10, 15 and 20 cm respectively from the Target

6. AMPT Event: Pb+Pb at 158GeV
VTPC2
VTPC1

7. Reconstruction

8. Reconstruction 1. Track distance in VTPC1 + VTPC2 > 1m
2. Require hit at least in 3 stations
3. NA61 experimental momentum resolutions is assumed:
dp/p2 = 7.0 x 10-4(GeV/c)-1 (NIMA 430 (1999) 210)
4. Position resolution in VD is given as a parameter. Hits are spread in y and x around geant hit according to the Gaussian distribution ( σ=10 μm and 15 μm) Track line is taken from the fit to the spread points

9. Initial signal vs background
Combinatorial background is very large →need to to apply background suppression cuts.
The cuts should remove background and be friendly for signal.

10. Background suppression strategy
List of cuts in the order they are applied
Single particle cuts:
1. track pT cut
2. track d cut (track impact parameter)
Two particle cuts:
3. cuts in Armenteros-Podolanski space to remove background
from Ks and Λ
4. two track vertex cut Vz
5. reconstructed parent impact parameter cut D

11. Spectrum after selection Cuts
Reduction of background ≈ 106
Reduction of signal ≈ 3
D0 + D0

12. Reconstructed yield for D0→ K+ π-, 200k 0-10% cent. Pb+Pb at 158 AGeV

13. Reconstructed yield for D0→ K+ π-, 200k 0-10% cent. Pb+Pb at 40 AGeV
x 26

14. Charged Particle Fluxes

15. Charged Particle Fluxes
Sources of particles hitting VD:
VTPC2
1. Charged particles produced in Pb+Pb interactions.
- during spill the anticipated beam intensity is 105 Pb ions per second.
- for 200 μm Pb target interaction probability is 0.5% which leads to 500 Hz
interaction rate
- used AMPT to generate 100k min. bias Pb+Pb at 158 AGeV
2. Delta electrons produced mostly in target
- study 10k Pb ions passing through the lead target
- soft particles – surrounding material might be important
- production threshold cut in geant4: minimum distance that produced particle will travel in a given material → translates to cut on energy
If the distance is (too) small – a lot of soft particles is produced (CPU
consumption)
If the distance is (too) large – important component might not be
described
→ the influence of the production threshold cut has to be studied

16. Pb electromagnetic interaction – zoom in
VTPC2
VTPC1
Pb ion moving in helium

17. Charged particles produced in Pb+Pb interactions
Particle occupancies: δ-electrons and particles produced in Pb+Pb interactions
Delta electrons
(averaged over 10k Pb events)
Charged particles produced in Pb+Pb interactions

18. Particle Flux Hadronic interactions:
During spill the anticipated beam intensity is 105 Pb ions per second.
For 200 μm Pb target interaction probability is 0.5% which leads to 500 Hz interaction rate
Hadronic interactions:
flux = (105 * 0.005) event/s * 1.6 particles/mm2/event =
= 800 particles/mm2/s = 800 Hz/mm2
Electromagnetic interactions (δ-electrons):
flux = 105 event/s * particles/mm2/event =
= 4000 Hz/mm2
Rate of Flux is not critical → both silicon and gas micro-pattern detectors
can handle such fluxes (hundreds kHz/mm2)

19. Charged particles produced in Pb+Pb 0-10% central interactions
VTPC2
We can expect very high hit occupancy on the level of 5 hit/mm2/event in the most inner part of the vertex detector.
→ it suggests that silicon pixel sensors would provide a good solution for us.
VTPC1 1.

20. Outer dimensions

21. Signal track distribution at 158 AGeV in VDS1
VTPC2
2x4 cm2
VTPC1
The figure shows hits (x,y) distribution generated by signal tracks is Vds1. The dashed boxes represent the cuts. We found that ~99.5% of signal tracks is localized within the box 2x4cm2.
As you can see, to cover the remaining 0.5% we would need to extend the cut in the x direction for almost factor of 2.

22. Signal track distribution at 158 AGeV in VDS2
VTPC2
4x8 cm2
VTPC1
For stations Vds2-Vds4 we just extend size of the boxes in proportion to their distance from the target. So we got dimensions: 4x8 cm2, 6x12 cm2 and 8x16 cm2 for Vds2, Vds3 and Vds4, respectively. The signal lost is kept below 1 % for each station.
For Pb+Pb at 40 AGeV the signal lost is on the level of 4% for the same cuts.

23. Signal track distribution at 158 AGeV in VDS3 and VDS4
VTPC2
6x12 cm2
VTPC1
8x16 cm2

24. MIMOSA-26
The following conceptual drawings are based on MIMOSA-26 chip hosting sensitive area of about 1.06 x 2.12 cm2 with the pixel pitch equal 18.4 μm (~663.5k pixels/chip):
VTPC2
These pads are for testing purpose and can be removed
VTPC1
The readout speed of the whole fame in ~100 μs (10 kHz), zero suppression circuit.
The chips are available. We can just buy them from IPHC, Strasbourg (we know also the estimated price per chip)

25. Preliminary drawing of the 1-st station
VTPC2
VTPC1
Drawn blue boxes have dimensions of the sensitive area of MOMOSA-26 sensor (~1x2cm2)
Size of the dashed box is ~ 2x4 cm2. We have to cover this area to loose less than 0.5% / 3% of signal particles for 158 / 40 GeV

26. Preliminary drawing of the 2-nd station
VTPC2
Vds2
VTPC1
Size of the dashed box is ~ 4x8 cm2 –
full coverage of Vds2 area with MOMOSA-26 requires 20 sensors
Including Vds3 (6x12 cm2) and Vds4 (8x16 cm2) we will need about 126 sensors for the whole detector.

27. Fluence estimates

28. Performance of MIMOSA-26 → test on beam
Temperature: C
Readout Time: µs
Pitch size : µm
Irradiated with to
fluence = 3 X neq/cm2
For disc. Threshold= 5 mV:
detection efficiency ~ 99.8%,
fake hits < 10-4
resolution ~ 3.5 µm
(M.Winter, CBM Progress Report 2010)

29. Displacement Damage Function
Bulk damage exclusively depends upon non ionizing energy lose (NIEL). This is described by the displacement damage functions D(E)
Hadronic interactions:
flux = (105 * 0.005) event/s * 1.6 particles/mm2/event =
800 Hz/mm2
Electromagnetic interactions (δ-electrons):
flux = 105 event/s * particles/mm2/event =
4000 Hz/mm2
0.62
0.62/5
(A. Vasilescu, ROSE Internal Note ROSE/TN/97-2 (1997))

30. Fluence in NA61/SHINE Factor of 40 below the tested range
Φeq 1MeV = ϰΦ ϰ - radiation hardness parameter
ϰ = 0.62/5 for electrons
ϰ = for particles from hadronic interactions
Fluence for electrons in 1 month (upper limit):
Φeq 1MeV = 4 x 105 /cm2/sec X 0.62/5 X sec = 1.28 X 1011 neq/ cm2
For Spill of the beam (20%) = 2.6 X 1010 neq/cm2
Fluence for charged particles for 1 month (upper limit):
= 8 x 104 /cm2/sec X 0.62 X sec = 1.28 X 1011 neq/ cm2
For Spill of the beam (20%) = 2.6 X 1010 neq/cm2
Factor of 40 below the tested range

31. Pixel Occupancy

32. Pixel occupancy
As usually looking at the most critical area of Vds1 where the track occupancies are:
1. 5 tracks/mm2/event for central Pb+Pb collisions
tracks/mm2/event from averaging over minimum bias Pb+Pb collision
δ-electrons/mm2/event for Pb ion on 200 μm target
Assumed 100 μs reading time of the whole MIMOSA-26 sensor → 10 kHz reading frequency.
Interaction rate is 0.5 kHz → average interaction multiplicity in one frame is 0.05/10 = For Poisson distribution with average=0.05 we get:
P(0) = 95% empty frame
P(1) = 4.7% single event
P(2) = 0.12% pile-up P(2)/P(1) = 2.5%
Beam intensity of 100kHz will lead to 10 ions in 100 μs
Let us assume very challenging case, namely that 1 frame contains: 1 central event, 1 semi-peripheral, 10 δ-electron events → part/mm2 = 7 part/mm2
MIMOSA-26 sensor hosts 2953 pixels on one mm2 →
Single Pixel Occupancy = 0.25% (+0.01% contribution from fake hits)
→ Not very dense environment → probability of overlap low, however we need full simulation to prove the reconstruction feasibility

33. Cooling of sensors

34. Cooling
Energy dissipation:
Static dissipation: 300mW / sensor
Dynamic dissipation: 200mW / sensor per 1% occupancy level.
→ 500 mW / sensor
→ 63 W for the whole system
He gas:
ρ = g/l, Cp = 5.19 J/(gK)
Dla ΔT = 1K → 68 l/s
Dla ΔT = 10K → 6.7 l/s
H2O:
ρ = 103 g/l, Cp = 4.18 J/(gK)
Dla ΔT = 1K → 15 ml/s
Cooling with He gas is not sufficient. We need solution as this developed for ALICE.

36. Cost estimates 1. MIMOSA-26 sensors: 2. Read-out system:
126 sensors, lets take 150 (1.6k Euro/sensor) → 240 kEuro (~1 MPln)
2. Read-out system:
64 customized boards (CB) – each connected into two sensors (500 Euro/board)
8 TRB_v3 (2k Euro/TRB_v3)
64 cables (25 Euro/cable)
126 dedicated micro-cables (20 Euro/cable (?))
LV suppliers for TRB+CB: 10 kPln
LV suppliers for M-26: kPln
Firmware for TRB_v3 (?)
TOTAL: kPLN
3. Carbon fiber mechanical support + cooling
4. Helium vessel + movable support with remote control + dedicated tracks for cables
5. men power: tests, integration

37. To Do 1. Full simulation: → realistic track reconstruction in VD
→ matching with VTPC
2. Tests on beam with prototype (4-6 sensors):
→ stability, sensor temperature (to keep fake hits low),
→ read-out concept (synchronization with NA61 DAQ)
→ He vessel, connection between sensors and CB.
3. Based on 1. and 2. final design → final proposal

38. BACKUP SLIDES

39. D0→ K+ π- , 200k 0-10% cent. Pb+Pb at 158 AGeV
VTPC2
S/B=13
For 50M events:
SNR = 193
40k D0+D0bar
S/B=15
For 50M events:
SNR = 178
34k D0+D0bar
VTPC1
S/B=6.6
For 50M events:
SNR = 163
31k D0+D0bar
S/B=9.6
For 50M events:
SNR = 155
26k D0+D0bar