MICOS a CHERNE collaboration in applied research involving industry

1
MICOSa CHERNE collaboration in applied research
involving industry

• François Tondeur, Lara Marwaha,
• Isabelle Gerardy, Jonathan Baré
• ISIB
• José Rodenas, Sergio Gallardo
• UPV

2
MICOSminiaturisation of the Compton spectrometer

• PRINCIPLE
• Measurement on the scattered X-ray beam to
  avoid detector saturation

3
The Compton spectrometer in the 1990s

• Matscheko et al RTI Electronics
• A big, heavy device adapted to a Ge detector

4
The Compton spectrometerin CHERNE

• UP Valencia Gallardo, Rodenas, Verdu Monte Carlo
  simulation of the Compton scattering technique
  applied to characterize diagnostic X-ray spectra
• U. Bologna Fernandez, Scot, Baré et al
• Reconstruction of the X-ray tube spectrum from a
  scattering measurement
• ISIB Bruxelles Marwaha, Tondeur et al
• present work

5
MICOS

• Development of a small portable device
• for routine spectrometric controls of X-ray
  generators
• 4-year FIRST project supported by the Walloon
  region 2008-2012
• Industrial cooperation with Balteau NDT and
  Canberra Benelux
• Academic collaboration with UP Valencia

6
MICOS partnership

• Project coordinated by ISIB, and executed at ISIB
  lab of nuclear and radiation physics
• research fellow Lara Marwaha (ex-FHA)
• UPV training of the research fellow to MC
  simulation and to the  old  spectrometer
• Balteau manufacturer of the shielding sales in
  the industrial sector
• Canberra provided detectors for the project
  sales in the medical sector

7
MICOS main steps

• Design of a simplified shielding
• Tests of detectors and choice of a detector
• MC simulation and validation
• Choice of the methodology for unfolding

8
Shielding design

• 3 options according to energy maximum 150, 300
  or 450 keV
• Design with MC simulation
• Steel/lead/copper for 150 kV and 300 kV
• 450 kV W cover above the 300 kV device

9
(No Transcript)
10
Choice of the detector

• Limited choice of  pen-type  detectors diam.lt20
  mm, reasonable cost
• Tested CZT, CdTe, NaI
• CZT- CdTe not easy to simulate with MC (problems
  with dead layers, variability from one detector
  to another, response depending on electronics
  settings)
• Choice NaI easy to simulate with MC

11
Monte Carlo simulation

• Geometrical model validated with radioactive
  sources 241Am, 109Cd, 57Co, 137Cs
• thickness of front
• window fitted to
• experimental data
• Response matrix of 150 keV prototype calculated
  with 1 keV interval

12
(No Transcript)
13
Deconvolution

• A specific algorithm similar to GRAVEL gives good
  results in few iterations
• sensitive to good energy calibration

14
MICOS going on

• 300/450 keV prototype soon manufactured
• simulation-validation-unfolding-testing
• Back to CZT/CdTe
• improve the geometrical model and the
  simulation of the response
• Students from CHERNE welcome for a master thesis

15
Thank you