Geant4 Simulation of Test-Mass Charging in the LISA Mission

1
Geant4 Simulationof Test-Mass Chargingin the
LISA Mission
Henrique Araújo Alex Howard Tim Sumner Imperial
College, London
Geant4 Workshop 2002
2
Test-Mass Charging
p
3
Geometry I
spacecraft
payload area
4
Geometry II
optical bench
5
Geometry III
Ti housing
proof mass
Mo caging
glass bench
6
LISA Environment
1. Protons from solar flares Localised in
time 2. Extragalactic cosmic rays Protons,
Alphas Isotropic flux MeV TeV
energies peak 500 MeV/nucleon
Jursa AS (ed) 1985
previous G3 simulations
Jafry et al. 1996, 1997
G4 simulation
7
Physics List
EM processes (LowE) Electrons, Gammas, etc Atomic
de-excitation Hadrons (no hFluorescence)
Hadronic processes Elastic (LEHE) Inelastic
(LEHE) Nuclear de-excitation Absorption/Annihilat
ion Photonuclear (g, e, m) Neutrons (HP, LE, HE)
Decays Decay in flight (no RDM)
Secondaries Cuts (250 eV), 1um - 5um Kill e-
outside caging
8
GEANT4 on the Grid

• extensive physics
• wide primary energy range
• low production cuts
• large statistics required
• 1 CPU-Year for 60 s
• of cosmic-ray exposure !

G4 Parallelisation !!!
First significant use of UK particle physics
Grid Sat 11 May 2002 1618 BST. Posted by Steve
Lloyd Alex Howard of Imperial College has made
the first significant use of the UK particle
physics Grid. Alex has submitted over a hundred
jobs to 16 Grid nodes at Imperial and 8 Grid
nodes at RAL running EDG middleware. GRIDPP News
9
Results I
45 s exposure 10,000,000 events 1 charging
event 2,000 events 40,000 events/day/CPU
Charging rate
G4 result 58 e/s
(G3 result 11 e/s)
52 e/s
6 e/s
10
Results II
Charge spectrum
The main contributions

• low energy electrons
• nuclear reaction

• proton/alpha stopping
• EM showers

G3
G3
11
Results III
Cuts
1 um 8.77 keV 41 e/s
(0.1 um) 250 eV 52 e/s
2 um 30.7 keV 40 e/s
5 um 53.2 keV 42 e/s
Energy of Primary, MeV
Decreasing the production threshold from 10 keV
(G3, G4EM) to 250 eV (G4LowE) leads to a 20
increase of the charging rate!
12
Conclusion

• Charging rate increased 5 times relative to G3
  simulation
• more physics, wider primary energy, geometry and
  lower cuts

• Geant4 there is room for improvement
• idiosyncrasies - hFluorescence, cuts by region,
  etc
• comprehensive physics testing hadronics at
  100 MeV
• yet more physics - (a,n) reactions, more
  materials, etc
• speed ?

• Geant4/GRID partnership is very promising
• huge computational power is becoming available
• GRID tools need development - job submission and
  retrieval
• Geant4 may need changes - data files in exec?
• G4Parallelisation is under way!