optics: general-purpose scintillator light response simulation code. E. Frlez, B.K. Wright, D. Pocanic.

PROGRAM SUMMARY
Title of program: optics
Catalogue identifier: ADNC
Ref. in CPC: 134(2001)110
Distribution format: tar gzip file
Operating system: DEC VMS V5.5, DEC OSF/1 V1.3A UNIX
High speed store required: 1MK words
Number of bits in a word: 32
Peripherals Required: disc
Number of lines in distributed program, including test data, etc: 8343
Keywords: Elementary particle physics, Detector design, Computer modeling and simulation, Monte Carlo simulation of scintillator response, Scintillation detectors, Computed tomography.
Programming language used: Fortran, Tk/Tcl
Computer: MicroVAX 3100 , DECstation 5000/200 .

Nature of problem:
Simulation of the volume and temporal light collection probability distributions given the geometrical shape plus bulk and surface optical properties of a scintillation detector.

Method of solution:
The code recognizes cylindrical, spherical, and parabolical as well as arbitrary polygonal scintillator shapes (and optional wrapping reflectors) that could couple via lightguides or wave-length shifters to photosensitive surfaces. The light-generating volume can be subdivided into the elementary cells. The photons generated within each cell are tracked through the scintillating volume taking into account specular, diffuse and rough surface reflections from lateral detector surfaces and wrapping reflectors, and the bulk attenuation and scattering effects from detector defects [1,2].
The code consists of 65 individual files containing the subroutines, data files and command files. A user can modify or expand the photon transport code as well as database files specifying default optical properties of the detector surfaces and the bulk media.
The program uses the CERNLIB programs in packlib library and kernlib Fortran callable libraries (optional).

Restrictions:
The statistical uncertainties of the simulated light collection probability distribution are limited by the practically tolerable running time (see below).

Typical running time:
The running time depends on the number of elementary volume cells chosen and the number of scintillating photons generated per cell and is therefore problem-dependent. For example, assuming a small-step volume subdivision into a 15x15x30 matrix with 6750 elements and aiming for better than 2 per cent average uncertainty in the three-dimensional light nonuniformity function typically requires 10**7 photon statistics per cell and running time of ~24 hours on a 200 MHz computer.

References:

 [1] B.K. Wright, Program optics (University of Virginia,                
     Charlottesville, 1992).                                             
 [2] B.K. Wright, Program tkoptics (University of Virginia,              
     Charlottesville, 1994).