Generation of primary photons through inverse Compton scattering using a Monte Carlo simulation code

Photon sources based on inverse Compton scattering, namely, the interaction between relativistic electrons and laser photons, are emerging as quasimonochromatic energy-tunable sources either as compact alternatives to synchrotron facilities for the production of low-energy (10–100 keV) x rays or to reach the 1–100 MeV photon energy range, which is inaccessible at synchrotrons. Different interaction layouts are possible for electron and laser beams, and several applications are being studied, ranging from fundamental research in nuclear physics to advanced x-ray imaging in the biomedical field, depending on the radiation energy range, intensity, and bandwidth. Regardless of the specific application, a reliable tool for the simulation of the radiation produced is essential for the design, the commissioning, and, subsequently, the study and optimization of this kind of source. Different computational tools have been developed for this task, based on both a purely analytical treatment and Monte Carlo simulation codes. Each of these tools has strengths and weaknesses. Here, we present a novel Monte Carlo code based on geant4 for the simulation of inverse Compton scattering in the linear regime. The code produces results in agreement with cain, one of the most used Monte Carlo tools, for a wide range of interaction conditions at a computational time reduced by 2 orders of magnitude. Furthermore, the developed tool can be easily embedded in a geant4 user application for the tracking of photons generated through inverse Compton scattering in a given experimental setup.

Figure 1

Example of a macro file for the ICS application.

Figure 2

Full and collimated spectrum of the BriXS photon beam simulated with cain and geant4.

Figure 3

Full and collimated spectrum of the NewSUBARU photon beam simulated with cain and geant4.

Figure 4

Full and collimated spectrum of the EGammaS-GBS photon beam simulated with cain and geant4.

Figure 5

Energy of emitted photons $E$ versus polar emission angle $\theta$ for the NewSUBARU photon beam, simulated with cain and geant4.

Figure 6

Spatial distribution of energy at a distance from the IP of 10 m for the BriXS photon beam simulated with cain and geant4 within a collimation angle of 3.5 mrad.

Figure 7

Intensity spatial distribution at a distance from the IP of 10 m for the NewSUBARU photon beam simulated with cain and geant4. The axis scale was set to highlight the most significant part of the distribution.

Figure 8

Intensity spatial distribution at a distance from the IP of 10 m for the EGammaS-GBS photon beam simulated with cain and geant4. The axis scale was set to highlight the most significant part of the distribution.

Figure 9

Profiles of the intensity spatial distribution for the EGammaS-GBS photon beam simulated with cain and geant4. The axis scale is set so as to highlight the most significant part of the distribution.

Figure 10

Time ($t_e$) required to simulate ${10}^6$ collimated photons with geant4 (blue curve) and cain (red curve), in the case of the BriXS source. The simulations were carried out on a server equipped with an Intel(R) Xeon(R) Gold 5220R CPU and 96 GB of RAM, and the reported execution times refer to single-thread mode. The number of rejected photons using the algorithm implemented in geant4 is also plotted (black curve).