Skip to content

Generate optical-photon primaries with a transverse polarization - #277

Open
cesarjesusvalls wants to merge 1 commit into
next-exp:masterfrom
cesarjesusvalls:optical-photon-polarization
Open

cesarjesusvalls wants to merge 1 commit into
next-exp:masterfrom
cesarjesusvalls:optical-photon-polarization

Conversation

@cesarjesusvalls

Copy link
Copy Markdown

ScintillationGenerator, LambertianGenerator and SingleParticleGenerator set the polarization of optical-photon primaries to G4RandomDirection(). That is an isotropic 3D unit vector, in general not perpendicular to the photon's momentum. Light is transverse, and Geant4 uses the vector as given: G4OpBoundaryProcess::DielectricDielectric (Geant4 11.3.0, G4OpBoundaryProcess.cc:1186-1192) splits it into its component along the s axis and the magnitude of the rest, so any component along the momentum is treated as p-polarized. At the first boundary each primary photon meets, the light is then on average one third s and two thirds p, instead of half and half for unpolarized light, which changes the Fresnel reflection and transmission there. Later interactions are not affected: Geant4 recomputes the polarization at each one.

This PR adds a function RandomPolarization(direction) to RandomUtils. It returns a linear polarization perpendicular to the given direction and uniform in azimuth, as G4Scintillation (Geant4 11.3.0) builds it. The three generators now use it, and a Catch2 test checks that the polarization has unit length, is perpendicular to the direction and is uniform in azimuth.

In NEXT-100 S1 simulations (Next100OpticalGeometry at 15 bar, ScintillationGenerator at fixed vertices in ACTIVE), the light at the PMTs changes by up to +0.8 % for vertices near the gate and −0.5 % near the cathode. Leaving the polarization unset gives the same light as this change: Geant4's G4PrimaryTransformer then assigns a random transverse polarization itself, but warns (ZeroPolarization) at every run.

Random-number sequences change, so outputs are not bit-identical to before. Built against Geant4 11.3.0; the new test passes, and the other tests give the same results as on master.

ScintillationGenerator, LambertianGenerator and SingleParticleGenerator
set the polarization of optical-photon primaries to G4RandomDirection(),
an isotropic 3D vector that is in general not perpendicular to the
momentum.

The new RandomUtils function RandomPolarization(direction) returns a
linear polarization perpendicular to the given direction, uniform in
azimuth around it. The three generators now use it, and a new test in
RandomUtilsTest checks it.

@gonzaponte gonzaponte left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This looks perfect to me. Thank you!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants