Abstract: Background: Gamma reaction history (GRH) is becoming important for inertial confinement fusion (ICF) diagnostic. The traditional design for GRH is by optical-ray-tracing method which is based on geometrical optics. However, the detector performances, such as detection efficiency and time response, are hard to improve as a result of lacking precise considerations of energy and angular distributions of secondary electrons. The optimization method genetic algorithm (GA) combining with GEANT4 can be used to enhance the detector performances during the design process. However, the computational time consumption of GEANT4 is great. Purpose: In order to accelerate the GEANT4 program, two continuation calculation methods, function fitting method and file read method, have been established. Methods: For the function fitting method, distributions of positions, directions, time and wavelengths of Cherenkov photons are fitted as functions, and then the fitted functions are sampled in GEANT4 program. For the file read method, positions, directions, time, and energy of Cherenkov photons are stored as a matrix, and then the matrix is read in GEANT4 program as the continuation source. Results: The difference between the acceleration efficiencies of the two methods is small. The file read method is more accurate than the function fitting method. Time spectrum of Cherenkov photons with the file read continuation source agrees well with time spectra calculated with 16.7-MeV gamma source. Deviations of the efficiency and full width half maximum (FWHM) are 1.4% and 2.1%. Conclusion: The computational time of GEANT4 can be reduced by more than 50%. Moreover, the continuation calculation method can be applied to other GEANT4 simulation problems with convergence difficulty by direct simulation.

Key words: GEANT4 simulation, Continuation calculation, DT fusion, Gas Cherenkov detection system