Gammaray tracking part III

1
Gamma-ray tracking - part III part IV

• Tracking algorithms
• Forward tracking
• Backtracking
• Simulation tool and ingredients
• Tracking performance
• Agata Data Acquisition
• Demonstrator performance

2
Aim of tracking

• Read for each event the list of deposited
  energies and positions of all the
• interactions points in AGATA
• e1, x1, y1,z1
• e2, x2, y2,z2
• .
• en, xn, yn,zn
• Disentangle the interaction points i.e
  reconstruct individual photon trajectories
• and write out photon energies, incident and
  scattering directions
• E1, (????inc-1? (????sc-1
• E2, (????inc-2? (????sc-2
• ..
• Ei, (????inc-i? (????sc-i

3
Processes in Germanium
100 keV 1
MeV 10 MeV
?-ray energy
Photoelectric Compton Scattering Pair
Production
4
Compton scattering
assuming that the e- is at rest, from
conservation of energy momentum
incident energy at i
cos(?i) 1 mec2 (1/E?i 1/E?(i-1))
scattered energy at i E?(i-1)-ei
5
Compton scattering Rules
Is the event complete ? Is track order 0-1-2-3
?
Original photon energy E?0 e1 e2 e 3 (if
totally absorbed !!)
from source interaction positions
1)
E?1,pos
from energy deposition incident energy
E?1 E?? e1
E?2,pos
2)
E?2 E?1 - e2
2
2
E?n - E?n,pos
?2 ?
Track order Permutation with best
?E
n1
6
Identification is not 100 sure
gt spectra will always contain background gt
Acceptance value determines the quality of the
spectrum gt Use R Efficiency x P/T to qualify
the reconstructed spectrum
7
What limits tracking performance ?

• Interaction position ? position of energy
  deposition

?inc
?sc
?ion/?Brem Ee- (MeV) /21.8
e-
?

• Rayleigh scattering (relevant at low gamma
  energies and end of track)
• gt change in incident direction

8
Electron Momentum Profile
E?/E?
Biggs et al., At. Data and Nucl. Data Tab. 3
(1975) 16
9
Some more complications
From PSA uncertainty in position of
interaction
x
x
x
position resolution
x
From preamplifier energy threshold (5 keV)
x
x
x
From preprocessing energy resolution
10
Forward tracking

• G. Schmidt et al., Nucl. Instr. Methods 430
  (1999) 69
• Forward peaking of Compton scattering
  (Klein-Nishina)
• Clusterisation of interaction points in
• (???) space
• gt Interaction points with an angular distance
  ? between each other (LINK ALGORITHM) or with
  respect to a given point (LEADER ALGORITHM)
  constitute a CLUSTER

11
Forward tracking

• Create cluster pool gt for each cluster, E??
  sum of energy depositions in the cluster
• Find most probable sequence of interaction points
  for each cluster
• Which sequence satisfies best the Compton
  scattering rules ?
• 3. Accept or reject clusters on the basis of Nth
  root of likelihood

2
E?n - E?n,pos
-
N
?E
L ?????Pn exp
n1
Probability for Compton or photoelectric
interaction and for travelling a given distance
in Germanium
12
Forward tracking

• Ideal 4? shell

27 ??s detected 23 in photopeak16
reconstructed 14 in photopeak
13
Backtracking

• J. Van der Marel and B. Cederwall,Nucl. Instr.
  Meth. 437 (1999) 538
• Photoelectric energy deposition
• is independent of incident energy
• Peaks around 100-250 keV
• gt interaction points within a given deposited
  energy interval
• (emin lt ei ltemax) will be considered as the LAST
  INTERACTION
• of a fully absorbed photon track

14
Bactracking

• 1. Create photoelectric interaction pool emin lt
  ei ltemax
• 2. Find closest interaction j to photoelectric
  interaction i
• distance between interaction points lt 2 ?
• Einc eiej, Escei
• 3. Find incident direction
• cos(?) 1 - mec2(1/Esc - 1/Einc)
• 4. Find previous interaction k or source along
  direction
• cos(?energy)-cos(?geometry) lt limit
• Prob(Compton interaction) gt PComp,min
• distance between interaction points lt 3 ?
• Einceiejek, Esceiej

15
Common treatment

• Single interaction points
• Does the interaction point satisfy the
  photoelectric conditions ?
• (e1, depth, distance to other interaction
  points,)
• Pair production interaction points
• Do the interaction points correspond to a pair
  production event ?
• e1 E?? - 2 mec2
• E? 4 MeV, M? 1
• 74 reconstruction efficiency (ph. eff 6.4)
• 99 P/T

16
Development of tracking algorithms

• Simulate detection of gammas in ideal
  detectors (the standard shell Ri15 cm, Ro24
  cm)
• Make the simulated data realistic considering
  experimental limitations
• Energy resolution and energy threshold
• Position resolution ? Packing Smearing
• Use simple events containing only one gamma to
  develop the basic methods
• Consider the case of high gamma multiplicity
  events
• Consider realistic detectors with dead spaces and
  materials.

17
Simulation tool
Simulation of PSA packing, smearing, energy
threshold, energy resolution
(ei,xi,yi,zi)
(ek,xk,yk,zk,?
Simulation of interactions of M? photons in a 4?
shell or Agata (with the Agata Geant4 code)
Tracking codes
(En, incident scattered directions)
E. Farnea, LNL Annual Report (2003)
18
Relative perfomances
A. Lopez-Martens et al., Nucl. Instr. Meth. A 533
(2004) 454

• Geant4 gt Ph. Eff. 76
• P/T 81

Photon Energy 1 MeV Multiplicity 1 and
30 Smearing - packing - energy threshold
Backtracking the last points of the sequence are
of low energy and close to each other ? bad
position resolution and easily packed together
19
Performances vs position resolution gamma
multiplicity
E? 1.33 MeV Packing Smearing
The biggest losses are due to multiplicity
(mixing points)
5 mm
5mm is a realistic packing smearing assumption
20
Effect of energy threshold
21
Doppler correction capabilities
22
Background rejection
1m
(same data used)
23
Effect of neutrons
J. Ljungvall and J. Nyberg, Nucl. Instr. Meth. A
550 (2005) 379

• Neutrons Gammas
• similar signals
• similar distribution of interaction points

• gt Large effect on P/T for
• low gamma multiplicities
• gt Ph. Eff -1 /n
• gt Can only discriminate ns and ??s with good
  timing

24
Effect of encapsulation and other dead materials
gt Careful design of ancillary devices !
25
Effect of simplified PSA
?
26
Effect of simplified PSA
30 photon band in Agata
27
Online implementation of PSA and tracking codes
Front-end electronic and pre-processing
Pulse Shape Analysis
Event Builder
Ancillary Detector
Post-Event Builder
Tracking
Storage
28
Data rates in AGATA _at_ 50 kHz singles
20 ?s/event
300 kHz of M? 30 -gt 50 kHZ singles
29
Data rates in AD _at_ 10 kHz singles
1 ms/event
15 detectors, GL-trigger, Ancillary gt 10 kHz
singles
30
Current computing time
PSA complex experimental event in 1
crystal Adaptive Grid Search 2ms/evt gt lt 5mm
pos. res. Genetic Algorithm 1s/evt gt 5mm
pos. res other algorithms not yet
tested/optimized Tracking complex simulated
event in 4? (30 1-MeV photons)
3ms/evt

• Specifications met for demonstrator

31
Expected performances of the Agata demonstrator

• Nominal target-detector distance 23.5 cm

target
E. Farnea
32
Expected performances of the Agata demonstrator
target
E. Farnea
33
Expected performances of the Agata demonstrator

• Exisiting arrays at Legnaro, Ganil and GSI

34
Agata demonstrator vs Rising
54Cr Coulex _at_ 135 MeV
FWHM6.9 keV
FWHM 4.9 keV
835
FWHM 3.1 keV
A. Buerger
35
Agata demonstrator vs Exogam
Secondary beam fragmentation 37Ca -gt 33Cl _at_ 65
AMeV
2/3 peaks resolved
A. Buerger
36
Agata demonstrator vs Exogam

• Fusion-evaporation reaction 58Ni(76Kr,4p)130Nd

Total projection
Gated spectra
Exogam
Demo
O. Stezowski
37
Agata demonstrator vs Clara

• Deep inelastic collision (massive transfer)
  82Se238U -gt 78Ge

E. Farnea and N. Marginean
38
Conclusion

• Next generation gamma-ray spectrometers based on
  the gamma-ray tracking principle
• 4? germanium arrays, no suppression shields
• Very high efficiency and good spectrum quality
• Increase of sensitivity by orders of magnitude
• Explore unknown territory in Nuclear Structure
  and Astrophysics
• Possible applications to gamma-ray imaging

39
Application of Gamma-ray tracking

• Position sensitive Germanium detectors and
  tracking
• techniques allow to image the origin of gamma-ray
• sources
• Medical imaging (i.e of positron sources)
• Portable gamma-ray imager (i.e detection of 235U)
• Gamma-ray astronomy

40
Imaging of gamma-ray sources from overlaping
Compton scattering cones
events
41
Imaging with Germanium detectors
Existing technology relies on BGO scintillator
technology. Limited position resolution High
patient dose requirement Poor energy resolution
Will not function in large magnetic field gt
significant room for improvement (SmartPET
project in Liverpool)