DOMINANT :
Dom Optical-photon to Material INteraction ANd Tracking simulator
(based on GEANT4)



Update record:  
20070614   DOMINANT calibration using GDOM calibration page uploaded

What is DOMINANT?

  • How to construct PMT solid
  • Absorption Length vs wave length of photons in glass and gel
    (penetration ratios are given by Elisa Resconi)

Generated table

  • Acceptance curve as a function of wavelength
  • Angle dependence of acceptance

DOMINANT calibration using GDOM calibration page


What is DOMINANT?

In order to improve our knowledge about photon absorption and reflection inside DOM module, we developed full Monte-Carlo simulator based on Geant4 (DOMINANT). The DOMINANT traces photon path with refraction and reflection at the boundaries between different materials (ice and glass, etc.) (*1) and absorption inside the glass and gel. It also simulates photons from charged particles, for example, the cherenkov photons which produced by direct muon hits inside the DOM sphere.
Current DOMINANT equips borosilicate glass sphere, silicon gel, and a quartz block shaped into 10 inch PMT. The glass sphere is filled with the air, and no electric bord is installed.

(*1: 2005/02/10 Now we artificially exclude these reflection and refraction, because of a bug inside the program. )

(fig.1 schematic side view of DOMINANT [pdf])
(fig.2 actual shape of the PMT [tiff])

[top]

How to construct PMT solid

In order to obtain actual hit point of photons at PMT photo cathode, we constructed geometrical shape of PMT as one of the G4VSolid. Because our PMT shape is represented by some combination of Constructive Solid Geometry (CSG) Solids, we used G4UnionSolid to construct PMT.

(fig.3 parameter : [large picture] [pdf])
(fig.4 side view of the PMT [tiff])

[top]

Absorption Length vs wave length of photons in glass and gel

The DOMINANT requires absorption length and refractive index for all materials. Elisa Resconi measured the penetration ratios of glass and gel. We calculated the absorption length from her data.

[penet_glass.pdf] [penet_gel.pdf] [absorp_glass.pdf] [absorp_gel.pdf]

[top]


Generated table

Using the DOMINANT, we generated acceptance table for our 13inch DOM. Because the Photonics calculates an amplitude / 1m^2, the photon beam is injected to the DOM from 1m^2 beam spot with an off-axis angle theta. The acceptances as a function of wavelength and off-axis angle are obtained as follows.

[pdf]

[pdf]

The colors of spots indicates off-axis angle, angle between beam direction and PMT axis. 0 degree means beam injected from top of the photo-cathode of PMT. Taking a careful look of the figure, each shape of acceptance curve looks similar. It may be good approximation to separate these two variables, wavelength and off-axis angle, as follows:

final_acceptance(theta, lambda) =
angle_acceptance(theta) * wavelength_acceptance(lambda)

within all range of our assuming wavelength.
This expression indicates that we can regard the angle_acceptance(theta) as a normalization factor of wavelength_acceptance(lambda) curve and the highest normalization factor should be 1 at the off-axis angle theta = 0 deg.
We decided to generate two 1-dimensional tables about wavelength and acceptance for DOM, instead of one 2-dimensional convoluted acceptance table. The ROMEO, fast ROOT based OM simulator, is also using these two tables.

[top]


The acceptance of photons as a function of wavelength

The wavelength_acceptance is defined as follows

Number of photons arrived at PMT surface(lambda)/
Number of photons injected to DOM from 1m^2 photon source(lambda)

at off-axis angle theta = 0.

[pdf]

[top]


QE convolution (acceptance of photo-electrons)

Convoluting the acceptance table and QE curve, we can get acceptance of photo-electrons as a function of wavelength. This table would be used by AMANDA simulator.


[eps]

 

The acceptance of photo-electrons is calculated by

Number of photons arrived at PMT surface(lambda) * QE(lambda) /
Number of photons injected to DOM from 1m^2 photon source


[pdf]

[top]


The acceptance of photons as a function of injection angle

The angle acceptance is calculated by

acceptance_at_theta / acceptance_at_0degree

as a function of injection angle theta (at DOM local coordinate) and cos(theta).
Both functions are fitted by polynomial functions, the former is represented by pol(5) and the latter is pol(2). The ROMEO is using these fitting functions.

[top]



Photon Propagator
(Toy Monte-Carlo)

Ray trace&BG sim
(DOMINANT)

PMT Handy c++ sim
(ROMEO)

Back to the
Chiba Icecube Home