In the inertial confinement fusion (ICF) implosion experiment, the 16.7 MeV deuterium-tritium (DT) fusion gammas provide a high-accuracy alternative to 14.1 MeV fusion neutrons for fusion reaction width and bangtime measurements. Gas Cherenkov detector (GCD) can be used to measure DT fusion gammas, which has the advantage of energy threshold to eliminate the interference of low-energy gamma photons. Previous studies mainly focus on optimizing system efficiency or time response of GCD. However, the system time delay and shield size of GCD are lacking in optimal design by simulation method. In the present study, we build a GCD simulation model using the Geant4 software, so as to optimize its structure considering the environment boundary of installation on the 100 kJ level laser facility. The influences of precursor signal and background interference on the fusion gamma measurement are analyzed. The GCD structure is optimized to increase the system sensitivity, and the system time delay and shield size are optimized to reduce the interference background. The measurement signal and performance changes of GCD are calculated by using the simulation model, which is helpful for configuring measurement parameters and estimating signal amplitude in implosion experiments conducted on the 100 kJ level laser facility.

A whole three-dimensional model of GCD is built by using the Geant4 software, including the conversion processes of "gamma photon-Compton electron-Cherenkov photon" and the collection process of Cherenkov photons. First, the electron conversion efficiency changing with converter material and thickness is studied to obtain more high-energy electrons within a small emission angle. The Cherenkov photons arriving at the end of the gas cell are calculated according to the gas length and gas diameter, so as to optimize the structure of the CO₂ gas cell. Meanwhile, the photon collection efficiency and the time waveform of collection photons are studied by changing the curvatures of the primary reflector R1 and the secondary reflector R2. Then, the influences of precursor signal and background interference on the main Cherenkov signal are analyzed, and the relationship between system time-delay tdelay (the peak time interval between the precursor signal and the main signal) and the distance from the secondary reflector to the first reflector L1 is calculated. Meanwhile, the tungsten shield size is determined by comparing the time waveforms of the collection Cherenkov photons before and after adding the tungsten shield. After that, the measurement signal of GCD installation on the 100 kJ level laser facility is calculated using the forward calculation method convoluting the collection Cherenkov photons, the impulse time response tIRF of photo multiplier tube (PMT), and the time spectrum tBW of fusion gamma emission. In addition, the detector sensitivity Sic (defined as collection Cherenkov photons per incident gamma photon on the convertor) and the system efficiency Sef (defined as collection Cherenkov photons per source gamma) are studied by changing the CO₂ pressure and the installation distance.

As the atomic number of material increases, the outcoming electrons within a small emission angle decrease (Fig. 3). A 15 mm thick carbon is selected as the gamma convertor according to the calculated electron conversion efficiency changing with the carbon thickness (Fig. 4). The CO₂, as the radiation medium, is optimized as that with a length of 100 cm and a diameter of 15 cm according to calculated curve of collected Cherenkov photons (Fig. 5). The optimal curvatures of the primary reflector and the secondary reflector are chosen as 34 cm and 600 cm, respectively, according to the calculated collection photons and the signal frontier proportion χ (the ratio of collection photons at the 30 ps ahead of peak time to photons at the peak time) changing with R1 and R2 (Fig. 8). The intrinsic time response trp [full width at half maximum (FWHM) of temporal discretization of collection photons] is evaluated as about 16 ps, and χ is about 5.5% (Fig. 9). In order to minimize the influence of the precursor signal on the main Cherenkov signal, tdelay is optimized as 0.71 ns with L1 of 10.4 cm (Fig. 11). The diameter and length of the tungsten shield are chosen as 68 mm and 80 mm, respectively. The time waveform of the main Cherenkov signal has no change, while the precursor signal is significantly suppressed (Fig. 12). The amplitude of the simulated signal is about 0.7 V, while the neutron yield Yn is 10¹³ with the PMT gain M of 5×10³ and threshold energy Eth of 6 MeV (Fig. 14). The FWHM of the measured signal is about 164 ps after convoluting tIRF of 105 ps and tBW of 100 ps. In addition, Sic will increase by three orders of magnitude by increasing the CO₂ pressure (Fig. 15), and it will decrease about 20% by changing the installation distance. Since the solid angle is inversely proportional to the square of the distance, Sef will decrease greatly (Fig. 16). In the implosion experiments with a higher yield, GCD can be installed farther to prevent PMT from outputting nonlinearly.

A whole three-dimensional model of GCD is built by using the Geant4 software, including the processes of "gamma photon-Compton electron-Cherenkov photon" and the boundary processes of photon reflection and transmission. The gamma converter and the CO₂ gas cell, as the radiation medium and the tungsten shield size, are optimized. A detector sensitivity of 0.21 photons per incident gamma photon and an intrinsic time response of 16 ps are achieved. The measurement signal and performance changes of GCD are calculated by using this simulation model, which is helpful for configuring measurement parameters and estimating signal amplitude in implosion experiments. The time response of GCD-coupled PMT can reach about 108 ps. The amplitude of the simulated signal is about 0.7 V, while the neutron yield is 10¹³ with a PMT gain of 5×10³ and a threshold energy of 6 MeV. The FWHM of the measured signal is about 164 ps after convoluting the fusion reaction width of 100 ps. The numerical calculation results indicate that the optimized GCD can meet the requirements of fusion gamma diagnostic in current implosion experiments on the 100 kJ level laser facility. In implosion experiments with high areal density, the instantaneous gammas activated by neutrons on the diagnostic devices will be strong. The influences of background interferences on the main Cherenkov signal are worth further study.