(Signal creation, energy loss, PID) Particle tracking and - - PowerPoint PPT Presentation

▶
signal creation energy loss pid
SMART_READER_LITE
LIVE PREVIEW

(Signal creation, energy loss, PID) Particle tracking and - - PowerPoint PPT Presentation

Feb 06, 2024 166 likes •901 views

Time Projection Chamber (Signal creation, energy loss, PID) Particle tracking and identification at high rates WS 16/17 Bogdan Blidaru Page 1 Motivation Standard detector courses describe to some extent how the signal is created and how

slide-1
SLIDE 1

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 1

Time Projection Chamber (Signal creation, energy loss, PID)

slide-2
SLIDE 2

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 2

The Flammarion Engraving

Motivation

  • Standard detector courses describe to some extent

how the signal is created and how the dE/dx variables are calculated.

  • They do not include however interdependencies that
  • ccur inside the gas chamber and that might influence

the signal.

  • We would like to go deeper and describe all the

processes that influence our measurements down to 0.2 % precision, which is our goal for dE/dx measurements in the ALICE TPC

slide-3
SLIDE 3

Time Projection Chamber

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 3

The Flammarion Engraving

Energy loss

slide-4
SLIDE 4

Time Projection Chamber

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 4

The Flammarion Engraving

Energy loss Transport

slide-5
SLIDE 5

Time Projection Chamber

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 5

The Flammarion Engraving

Energy loss Gas gain Transport

slide-6
SLIDE 6

Time Projection Chamber

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 6

The Flammarion Engraving

Energy loss Gas gain Transport Signal formation

slide-7
SLIDE 7

Time Projection Chamber

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 7

The Flammarion Engraving

Energy loss Gas gain Transport PID Signal formation

slide-8
SLIDE 8
  • Detection interaction with the medium
  • Charged particles lose energy

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 8

Detection mechanisms

slide-9
SLIDE 9

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 9

Energy loss

slide-10
SLIDE 10

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 10

Energy loss

Focus: the processes that influence the PID measurements & performance.

slide-11
SLIDE 11

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 11

Bethe Bloch formula

Bethe Bloch formula for heavy particles

  • don’t undergo: Bremsshtrahlung, emission of Cherenkov radiation;
  • nuclear reactions are extremely rare
  • elastic scattering off nuclei is less common compared to:

INELASTIC COLLISIONS WITH ATOMIC ELECTRONS

Density dependent [S/ρ] [MeV/g/cm2]

slide-12
SLIDE 12

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 12

Bethe Bloch formula

slide-13
SLIDE 13

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 13

Bethe Bloch formula

More or less the same for different materials (except for the Hydrogen)

slide-14
SLIDE 14

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 14

Bethe Bloch formula

More or less the same for different materials (except for the Hydrogen) Leads to a slow rise of ionization losses with the particle momentum (accounting for relativistic flattening of the electric field of the incoming particle)

slide-15
SLIDE 15

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 15

Bethe Bloch formula

More or less the same for different materials (except for the Hydrogen) Leads to a slow rise of ionization losses with the particle momentum (accounting for relativistic flattening of the electric field of the incoming particle) Maximum energy transfer to an electron (limited by E-p conservation laws)

slide-16
SLIDE 16

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 16

Bethe Bloch formula

More or less the same for different materials (except for the Hydrogen) Leads to a slow rise of ionization losses with the particle momentum (accounting for relativistic flattening of the electric field of the incoming particle) Maximum energy transfer to an electron (limited by E-p conservation laws) Density effect factor (polarization of the medium)

slide-17
SLIDE 17

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 17

Bethe Bloch formula

Differences between heavy charge particles and electrons in matter:

slide-18
SLIDE 18

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 18

Bethe Bloch formula

Differences between heavy charge particles and electrons in matter:

  • electrons are light and collide with other atomic electrons
  • large angle multiple scattering
  • large energy losses possible
  • indistinguishability (in a quantum sense)
  • electrons are relativistic at nuclear energies
slide-19
SLIDE 19

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 19

Bethe Bloch formula

Differences between heavy charge particles and electrons in matter:

  • electrons are light and collide with other atomic electrons
  • large angle multiple scattering
  • large energy losses possible
  • indistinguishability (in a quantum sense)
  • electrons are relativistic ar nuclear energies
  • electrons emit radiation as they lose energy
  • Bremsshtrahlung
  • Cherenkov radiation
slide-20
SLIDE 20

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 20

Bethe Bloch formula

Bethe Bloch formula for electrons

slide-21
SLIDE 21

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 21

Cherenkov radiation

Already included in the Bethe Bloch formula!

slide-22
SLIDE 22

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 22

Cherenkov radiation

Already included in the Bethe Bloch formula!

  • Density effect – result of polarization and screening of distant atoms

from the charged particle; more important at higher energies and more important for dense media

  • As the charged particle crosses through the material, polarization is

induced and then relaxes to zero after the particle has passed through  EM oscillation

  • the coherent sum at the shock wavefront, when the conditions are

possible for Cherenkov emission, is part of the density effect calculation

  • the dE/dx for the Cherenkov portion of the energy loss is only about

1% of the typical dE/dx value (in condensed materials)

slide-23
SLIDE 23

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 23

Bremsstrahlung

(photon emission by an electron accelerated in Coulomb field of nucleus)

Bremsstrahlung – dominant process for E > 10-30 MeV Within ALICE ITS & TPC: low material budget (~10% of radiation length for normal incident particles)

slide-24
SLIDE 24

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 24

c≈0.144 (ITS)

Low momenta

  • Bremsshtrahlung leads to deterioration of the momentum resolution.
  • So, the objective is to decrease the radiation length

 reduce the material budget as much as possible (RUN3 goal)

  • Typical momentum resolution at low momenta is about 1%.
  • For electrons, this effect is much bigger than the multiple scattering.

Bremsstrahlung

slide-25
SLIDE 25

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 25

Energy loss

Typical range of E for the operation of HEP TPC’s

slide-26
SLIDE 26

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 26

Energy loss

Region of minimum ionization approximately independent

  • f the material

Relativistic rise Fermi Plateau Energy loss basically depends

  • n the ratio Z/A

Ex: gases (TPC, TRD) dense media (Silicon detector, ITS)

slide-27
SLIDE 27

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 27

Energy loss

Pion/Kaon separation requires a dE/dx resolution of < 5% Simultaneous measurement of p and dE/dx defines m, the particle identity

  • Energy loss affected

by Landau distribution

  • A few measurements

necessary

  • Truncated mean

(samples with 40% highest values are ignored)

True in the first approximation…

slide-28
SLIDE 28

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 28

Ionization

Particle trajectory in gaseous media

slide-29
SLIDE 29

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 29

Ionization

Particle trajectory in gaseous media Primary ionization

slide-30
SLIDE 30

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 30

Ionization

Particle trajectory in gaseous media Primary ionization Secondary ionization

slide-31
SLIDE 31

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 31

Ionization

Particle trajectory in gaseous media Primary ionization Secondary ionization

ALICE TPC (Ne:CO2)

slide-32
SLIDE 32

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 32

Ionization

Electrons above threshold do not contribute to the observed energy loss Cutoff ~ 30 keV (1cm) Energy loss  how much the particle loses Energy deposit  restricted energy loss (within cutoff ~1cm) So, you can measure it if the energy loss is localized. What we really reconstruct  primary electrons in a small neighbourhood along the track.

slide-33
SLIDE 33

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 33

Electron attachment

  • Probability for an electron drifting to be captured by O2 molecule is 1% per 1m drift

per 1 ppm of O2. Signal attenuation ~ 2.5% for the full drift length

drift length mean free path

Leads to a decrease of signal

slide-34
SLIDE 34

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 34

Electron attachment

  • Probability for an electron drifting to be captured by O2 molecule is 1% per 1m drift

per 1 ppm of O2. Signal attenuation ~ 2.5% for the full drift length

drift length mean free path

Leads to a decrease of signal Ex.: in ALICE TPC:

  • After RUN1

something contaminated the gas mixture

  • Result: huge

electron attachment

  • Result: 3 times

smaller signal

  • This was the only

time the problem was reported

slide-35
SLIDE 35

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 35

Cluster size distribution

Distribution of E transfer for Ar at γ=1000

Integral cluster size distribution in Ar

Q(n) – probability that the cluster has more than n electrons

dN ~ 1/n2

Random process of primary e- creation  smeared out function The continuous lines are hand-drawn interpolations whereas the broken lines are extrapolations corresponding to the 1/n2 law for large n

In ALICE TPC some parametrisation is used for Ne: dN ~ 1/n2.2 Simulations Measurements

slide-36
SLIDE 36

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 36

Drift of electrons and ions

slide-37
SLIDE 37

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 37

Drift of electrons and ions

G~N~p/T

G and vD affected by p & T fluctuations

due to temperature modification ~400μm (compared to 200 μm of the resolution)

slide-38
SLIDE 38

Electrons

  • light mass

scatter, random direction

  • pick up extra

speed due to E

  • vD = μE

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 38

Drift of electrons and ions

Ions

  • ion drift

velocity « vD

  • cloud of ions

can distort field, affect gain

vD of order of cm/ms vD of order of cm/μs

*see presentation 1, 2

slide-39
SLIDE 39

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 39

Electron diffusion

ALICE TPC: Diffusion  important because it affects the part of the signal that we can measure

slide-40
SLIDE 40

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 40

Amplification of ionization

geometry, temperature, pressure dependent

Envelope of ions around the wire influences the shape of the induced signal (next slides)

slide-41
SLIDE 41

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 41

Amplification of ionization

geometry, temperature, pressure dependent

Typical distance between two interactions is λ~2.7 μm Ex: for G=10000, we have 13 generations of

  • multiplications. Thus, we have a distance of

~30 μm above the wire where the avalanche starts to form (comparable to the wire diameter)

Envelope of ions around the wire influences the shape of the induced signal (next slides)

slide-42
SLIDE 42

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 42

Gas gain influences

Diethorn formula Derived from cylindrical geometry Applicable in MWPC geometry of TPC

In practical cases varies between 5 and 8

slide-43
SLIDE 43

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 43

Working point of gas detectors

Ionization mode: full charge collection no multiplication; gain ≈ 1 Proportional mode: multiplication of ionization signal proportional to ionization measurement of dE/dx secondary avalanches need quenching gain ≈ 104 – 105 ~ linear with voltage Limited proportional mode [streamer]: strong photoemission requires strong quenchers or pulsed HV gain ≈ 1010 Geiger mode: massive photoemission full length of the anode wire affected discharge stopped by HV cut

slide-44
SLIDE 44

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 44

Working point of gas detectors

Ionization mode: full charge collection no multiplication; gain ≈ 1 Proportional mode: multiplication of ionization signal proportional to ionization measurement of dE/dx secondary avalanches need quenching gain ≈ 104 – 105 ~ linear with voltage Limited proportional mode [streamer]: strong photoemission requires strong quenchers or pulsed HV gain ≈ 1010 Geiger mode: massive photoemission full length of the anode wire affected discharge stopped by HV cut

slide-45
SLIDE 45

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 45

Working point of gas detectors

Ionization mode: full charge collection no multiplication; gain ≈ 1 Proportional mode: multiplication of ionization signal proportional to ionization measurement of dE/dx secondary avalanches need quenching gain ≈ 104 – 105 ~ linear with voltage Limited proportional mode [streamer]: strong photoemission requires strong quenchers or pulsed HV gain ≈ 1010 Geiger mode: massive photoemission full length of the anode wire affected discharge stopped by HV cut

slide-46
SLIDE 46

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 46

Working point of gas detectors

Ionization mode: full charge collection no multiplication; gain ≈ 1 Proportional mode: multiplication of ionization signal proportional to ionization measurement of dE/dx secondary avalanches need quenching gain ≈ 104 – 105 ~ linear with voltage Limited proportional mode [streamer]: strong photoemission requires strong quenchers or pulsed HV gain ≈ 1010 Geiger mode: massive photoemission full length of the anode wire affected discharge stopped by HV cut

slide-47
SLIDE 47

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 47

Gas gain influences

Diethorn formula  ideal in cylindrical chamber ALICE TPC has wire geometry

In the regions where the TPC is operating the dependence of the amplitude to voltage is assumed to be exponential.

These variations need to be calibrated!

slide-48
SLIDE 48

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 48

Gas gain influences

Variations of the gas gain (0.8 - 1.2) within

  • ne chamber due to mechanical imperfections

and chamber geometry (deformations, wire sag) can reach up to about 20%.

Diethorn formula  ideal in cylindrical chamber ALICE TPC has wire geometry

In the regions where the TPC is operating the dependence of the amplitude to voltage is assumed to be exponential.

These variations need to be calibrated!

slide-49
SLIDE 49

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 49

Gas gain calibration (Kr)

Characteristic decay spectrum for each pad + fit of main peak (local gain higher  spectrum wider)

slide-50
SLIDE 50

Signal formed by induction due to movement of charges towards cathode and anode

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 50

Signal formation and shape

IROC: wire to pad distance is 2mm, width of the induction signal is 2mm. OROC: wire to pad distance is 3mm, width of the induction signal is 3mm.

In ALICE TPC:

slide-51
SLIDE 51

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 51

Signal formation and shape

Signal almost entirely due to ions. Electrons reach the wire in ~0.2ns. Integration time of the signal ~200ns. The electrons cannot be detected. In drift tubes the signal can be calculated using:

Ions moving from the wire surface to the tube wall induce a current signal on the grounded wire

slide-52
SLIDE 52

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 52

Signal formation and shape

Signal almost entirely due to ions. Electrons reach the wire in ~0.2ns. Integration time of the signal ~200ns. The electrons cannot be detected. In drift tubes the signal can be calculated using:

Ions moving from the wire surface to the tube wall induce a current signal on the grounded wire Time development of a current pulse described by Qind The extremely sharp onset can be observed as well as the extremely long tail characteristic of such pulses

In TPC, with proper E field geometry, the induced signal is calculated numerically.

slide-53
SLIDE 53

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 53

Pad response function

pad width pad height surface charge density

Gatti formula (for one dimensional charge distribution)

anode-cathode separation

The width σ of the PRF(x) is given by the distance of the anode wire to the pad plane and the pad width

IROC, OROC: wire to pad distance is 2, 3mm

slide-54
SLIDE 54

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 54

Pad response function

  • 1. Induced charge distribution
  • 2. Pad response function
  • 2. Integrated charge over

the pad. Ions are created exclusively on top of the wire  radial position of the ions is fixed.

  • 1. It describes

what the signal induced on the pad plane will look like if there is a charge at a certain height.

pad plane wire pad

slide-55
SLIDE 55

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 55

Signal formation and shape

Fast electron collection -> peak Slow ion drift -> ion tail Fluctuations because history of events is fluctuating In high multiplicity environments it causes a degradation in the following signals due to signal pile-up. In turn, part of the pad signal remains under the zero suppression threshold and is lost. This loss of signal amplitude deteriorates dE/dx resolution and PID performance. It has to be corrected! (we need measurements of the ion tail)

slide-56
SLIDE 56

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 56

Signal formation and shape

Drift lines of the ions

Corresponding ion signals induced on the pads depends on the applied voltage Depending on gain, different fractions of the ions end up on pads, cathode or gate or are drifted away; we can get positive or negative signal depending

  • n where the ions are

drifting

Ion tails Induced signal in the pad plane has 2 parts: fast part (used to determine the track properties, dE/dx, …) and a slow part (that leads to the deterioration of the performance). The main signal is on the background of the following signal.

slide-57
SLIDE 57

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 57

Signal shape

Comparison of the measured and simulated signal shapes in five pads of a cluster where pad number 3 is the center of the cluster. Upper panel: pulse shape Lower panel: ion-tail Left: data Right: Garfield simulations

slide-58
SLIDE 58

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 58

Time response function

Ion tail effect offline correction using TRF If a particle loses less E than the tails from the previous particle, this particle is not detected. This effect has to be corrected -> hardware algorithms (e.g. ALTRO) -> unfortunately, not numerically stable. Offline emulation of the event and corrections.

3ADC threshold – whatever is below is not detected in the following reconstruction

slide-59
SLIDE 59

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 59

Particle IDentification

Subject to contamination from other particles

Has to be known very precise (0.2%) Different ways dE/dx is used for PID:

  • Reject background ( ) – right image
  • Assign each particle the weight to belong to a particle species
  • Spectra unfolding (left image)  Bayesian approach
slide-60
SLIDE 60

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 60

Probability of a particle to belong to a certain species (i=e,K,p,μ,π) (priors) – describe the relative concentrations

  • f particle species

Particle IDentification

slide-61
SLIDE 61

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 61

dE/dx calibration

  • From the measured charge in the reconstructed clusters,

an evaluation of the dE/dx for each track is done, following the above workflow.

  • The actual measurement is the charge in the cluster. From this

value an estimator for the initial emitted charge is evaluated, taking into account several corrections.

  • The dE/dx corresponding to the cluster is then evaluated, and the

dE/dx for one track is estimated from the measurements of the different track clusters.

  • The dE/dx can then be used at the analysis level, where further

corrections are done.

slide-62
SLIDE 62

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 62

dE/dx calibration

  • From the measured charge in the reconstructed clusters,

an evaluation of the dE/dx for each track is done, following the above workflow.

  • The actual measurement is the charge in the cluster. From this

value an estimator for the initial emitted charge is evaluated, taking into account several corrections.

  • The dE/dx corresponding to the cluster is then evaluated, and the

dE/dx for one track is estimated from the measurements of the different track clusters.

  • The dE/dx can then be used at the analysis level, where further

corrections are done.

slide-63
SLIDE 63

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 63

At the level of one cluster we have two estimators: Qmax & Qtot (depending on precision) From these one needs to extract the actual charge

  • f the particle (the number
  • f primary ionizations

corresponding to a cluster position)

Estimators

slide-64
SLIDE 64

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 64

At the level of one cluster we have two estimators: Qmax & Qtot (depending on precision) Take into account:

  • PRF, TRF
  • Diffusion
  • Inclination of the track

at the pad row

  • Distance w.r.t. pad center
  • (effective) Pad length
  • Threshold (only for Qtot)
  • missing charge inside the cluster

From these one needs to extract the actual charge

  • f the particle (the number
  • f primary ionizations

corresponding to a cluster position) Qtot – low int, low occ Qmax – high int, high occ

Estimators

slide-65
SLIDE 65

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 65

To transfer the charge into the dE/dx estimator, the cluster dependence

  • n the track topology and on the read out gain has to be removed.

+ gain map applied + gain correction (HV, p/T)

(pad by pad) (time dependent corrections)

dE/dx calibration

slide-66
SLIDE 66

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 66

  • Before combining the clusters, they are
  • rdered with increasing dE/dx values.
  • A truncated mean is evaluated, taking into

account only the first 60%.

  • Since 2010 Pb-Pb data, clusters below

threshold are also used in the algorithm. The missing clusters are clusters not observed, but with observed cluster in two adjacent

  • rows. They are replaced by the smallest
  • bserved charge in the track sample.
  • (*) Similar approach used in ALEPH, with a

two sided truncated mean. In ALICE no lower threshold is used, since this provokes a decrease in separation power.

60%

To disentangle between signal & noise (@2ADC) Signal below threshold (must be considered in order not to bias dE/dx)

dE/dx calibration

slide-67
SLIDE 67

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 67

The transfer function relates the reconstructed dE/dx to the input energy loss Required calibration: 0.2%

Transfer function

Transfer function parametrization is one approach that can be found in literature. However, this multidimensional function cannot be factorized. The effects are interdependent.

slide-68
SLIDE 68

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 68

Influence of threshold on TF

  • Bigger separation

for lower truncation

  • Relative RMS changing

as a function of truncation

  • Optimal separation

power is around threshold 55-70% for ALICE TPC Transfer function (no effects, just truncation)

slide-69
SLIDE 69

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 69

Influence of threshold on TF

At the current gain in TPC (8000), Qmax estimator has better separation power than Qtot

Transfer function ~5% spread Goal: 0.2%

slide-70
SLIDE 70

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 70

Conclusions

In order to use the TPC’s dE/dx measurements for the PID at relativistic rise, dE/dx should be calibrated with precision of 0.2 %. The knowledge of the dE/dx (βγ) curve is insufficient in order to characterize PID performance at ALICE TPC. To achieve this precision, many effects should be considered and calibrated.

slide-71
SLIDE 71
slide-72
SLIDE 72

Bogdan Blidaru Particle tracking and identification at high rates WS 16/17 Page 71

Backup Diethorn formula

Diethorn formula ∆V – potential difference through which an electron moves between successive ionization events a – anode radius b – cathode radius V – applied voltage Emin – minimal field needed for ionization ρ0 – normal gas density ρ – gas density