Two Monte Carlo Approaches for Generation of Scintillation DRFs

1
Two Monte Carlo Approaches for Generation of
Scintillation DRFs
Zhijian Wang Daniel Speaker Robin P. Gardner
North Carolina State University
2
Outline

• Introduction of DRFs
• G03 approach (Daniel Speaker)
• MCNP5 approach (Zhijian Wang)

3
Introduction to DRFs

• Definition the pulse height spectrum is caused
  by a monoenergetic energy, E. The Response
  function R(E,E) spans all energies pertinent.
  And each energy E has a corresponding pulse
  height energy that ranges from 0 to E?E. R(E,E)
  is defined as a PDF, and is NOT a function of any
  materials outside the detector
• Advantages and Use (1) Serves as a significant
  variance reduction approach in Monte Carlo codes
  that calculate photon spectra, (2) has a natural
  smoothing effect, and (3) Yields better accuracy
  since there are presently some unknown features
  in DRF

4
Heath Experiments Empirical Detector Response
Function
Compressed Neoprene Rubber
5
G03 Originally

• Monte Carlo simulation of the detector as a bare
  crystal
• Simulates Detector non-linearity and the variable
  flat continua
• 3 Detectors NaI, Ge, Si(Li)
• 2 Different sizes 3x3, 6x6 (minimal)
• Source at variable distances, on axis with the
  right circular cylinder
• Advantages vs. MCNP (1)Could put in nonlinearity
  very easily, (2) easy to put in flat continuum,
  (3) more easy to manipulate input to examine
  details of DRF, and (4) speed
• Disadvantages (1) Cannot model complex geometry,
  (2) Does not use rigorous electron transport, and
  (3) Does not simulate detector can

6
Details Of G03 Monte Carlo Simulation

• Particle tracking done completely in Monte Carlo
• Cross Section Data originally done in functional
  form
• Rayleigh scattering not considered
• Simple Electron Range Relationship
• The variable a in the relation above signifies
  the multiplier applied to the range
• Each component of the DRF is calculated
  separately and then added up for the total
  response

7
(No Transcript)
8
Examples of Original G03 With and Without Flat
Continuum vs. Heath
9
MCNP Simulations for Investigation of Can
Influence on Pulse Height

• Source Isotropic
• Positioned 10 cm from Detector
• 20 Different Energies from 0.344 to 10 MeV
• Surface tally (f2) for both electrons and photons
  between the can and crystal
• Pulse Height Tally for the detector
• Surface tally on the side of can
• 3 Different Simulations bare crystal, crystal
  with front face of can, and crystal with the
  whole can

10
Pulse Height for Various Detectors With 0.662 MeV
Incident Energy
BGO
NaI
LaBr
Note Backwards Continua Of Same Energy
11
Big Differences Between the Detector With Can and
a Bare Crystal

• Compton Continuum
• Auger Electrons
• Overall Spectrum
• Photons and Electrons of different energies
  Incident on Crystal
• As much as 5 difference in the number of total
  counts
• The Valley region between the Photopeak and the
  Compton Edge

12
Photons Incident on NaI Crystal After Going
Through the Front Face of the Can
13
Electrons Incident on NaI Crystal After Going
Through the Front Face of the Can
14
Photons Incident on NaI Crystal Coming from the
Side of the Can
15
Electrons Incident on NaI Crystal Coming from the
Side of the Can
16
Improvements ongoing and future

• More Detectors added LSO, BGO, LaBr3, LaCl2
• Cross Section data obtained more efficiently
• Can not physically added but patch to include the
  difference for incident photons and electrons
• Different Geometry 1-Cylindrical detector on the
  side
• Different Geometry 2-Cylindrical detector of
  varying sizes (LD1)
• Different Geometry 3-Box of any dimension

17
Different Types of Detectors Implemented

• Things to be changed (1) Densities (2) Yield From
  Bremsstrahlung (3) Range Relationships Added (4)
  Non-Linear Parameters
• Range Relationships from Turner
• Interesting Note La detectors are very near
  linear

18
Implantation of Photons And Electrons in the
Crystal

• Remember Components of the DRF are calculated
  separately and then added up to get the total
  response in that particular spectra
• Front Face and Sides addition to spectra for
  electrons directly by code elec_in
• The above is done by converting the spectra into
  direct spectra that is the response from the
  electrons and extrapolated between incident
  energies
• Bounce in Photons
• Front Face Photons

19
Radioactive Component to DRF

• If the DRF is defined as interactions that happen
  inside the detector only
• What happens when a detector is Radioactive and
  it contributes to the spectrum like LaBr and LaCl
• Simple implantation--- add the spectra directly
  like the electrons earlier

20
Detector Response Function Development for MCNP
Use
21
DRFs Development for MCNP USE

• DRFs development in MCNP5
• Application in 6X6 NaI detector
• Application in Raytheon Cargo simulation
• Gamres in MCNP4
• GADRAS

22
Heath Experiments Empirical Detector Response
Function
23
A modified MCNP5 For DRFs

• Code benchmarked by Heath Experiment
• The flat continua adjustment
• ---An empirical factor operator is added to the
  electron cross section in the electron transport
  in the detector cell
• The NaI scintillation nonlinearity
• ---The Nonlinear factor is added before the
  electron is tallied

24
Flat Continua Adjustment
25
A modified MCNP5 For DRFs

• Code benchmarked by Heath Experiment
• The flat continua adjustment
• ---An empirical factor operator is added to the
  electron cross section in the electron transport
  in the detector cell
• The NaI scintillation nonlinearity
• ---The Nonlinear factor is added before the
  electron is tallied

26
Nonlinearity for Na-24 (2.754 MeV)

• Both spectra are shifted
• Nonlinearity has a broadening effect on the peaks

27

• Experimental Results of 6x6 NaI Detector

A)0.04 Aluminum B)0.125 Silicone sponge
rubber C)0.02 White teflon D) NaI Crystal Side
between crystal to Aluminum magnesium Oxide
powder.
28
Experimental Results of 6x6 NaI Detector
29
Raytheon Cargo

• 2X4X16 NaI detectors
• Shielding with Pb, Iron and wood

30
Gamres

• (n, g) patch, tally the photons based on the
  creation mechanism.
• Cosine bins can be specified in either radians or
  degrees.(F2).
• F8 pulse-height tally is modified to put the
  source energy into energy bins specified on an
  fu8 card so that a gamres-like detector map can
  be generated.
• One of the beauties you get an unbroaden
  response map which can be smeared with Gaussian
  convolution to any desire resolution.

31
Pertinent References

• R.P. Gardner and A. Sood, 2004 A Monte Carlo
  Simulation for Generating NaI Detector Response
  Functions (DRFs) that Accounts for Nonlinearity
  and Variable Flat Continua, Nucl. Instr. Meth.,
  Vol. 213, pp 87-99 .
• R.L. Heath, 1964, Scintillation Spectrometry
  Gamma-Ray Spectrum Catalogue, IDO-16880-1,AEC
  and Development Report, Physics, TID-4500.
• G.F. Knoll, 2000, RADIATION DETECTION AND
  MEASUREMENT, 3rd Edition, John Wiley Sons, Inc.
• MCNP5 manual vols I, II and III, LA-UR-04-2506,
  Los Alamos National Lab.
• D.E. Peplow, R.P. Gardner, and K. Verghese, 1994,
  Sodium Iodide Detector Reponse Functions Using
  simplified Mote Carlo Simulation and principal
  Components, Nuclear Geophysics, Vol. 8, No. 3,
  pp.243-259.
• A. Sood, 2000, A new Monte Carlo Assisted
  Approach to Detector Response Functions, PhD
  Thesis, North Carolina State University, Raleigh
  NC.