X-ray polarization detection is an important means to study the astrophysical properties of intense X-ray sources such as black holes, pulsars, and related gamma-ray bursts. The development of X-ray polarization detectors with excellent performance is the technical basis for related research. Early X-ray polarization detectors were mainly Thomson scattering polarimeters and Bragg polarimeters. However, due to the low modulation factor and narrow detection energy range, the ideal polarization measurement results were not obtained. In 2001, Costa et al. proposed a new way of X-ray polarization detection using the photoelectric effect, in which the X-ray polarization information was obtained by imaging the photoelectron track produced by X-ray photons through a gas detector. The polarimetric photoelectric process is the key physical process for the detector to realize polarization detection. It is of great significance to clarify the photon-gas interaction process and the distribution law of emitted photoelectrons for further understanding the working mechanism of the detector. The polarimetric photoelectric process is an important research content in the development of this type of X-ray polarization detector. Different types of gases have various properties, which will affect the particle transport in the polarimetric photoelectric process and further leads to different detection efficiencies. Therefore, it is necessary to simulate the polarimetric photoelectric process under different conditions. This can provide a theoretical basis and data support for the structure design of X-ray polarization detectors.

We simulate the polarimetric photoelectric process of 2-10 keV linearly polarized X-ray photons in several commonly used working gases by the Monte Carlo code Geant4. The selected working gas combinations include He+C₃H₈, Ne+CF₄, Ne+DME, Ar+CH₄, Ar+CO₂, Xe+CO₂, CF₄+C₄H₁₀, and DME+CO₂. The response relationship of the emission position and azimuthal angle distribution of photoelectron with the polarization direction and energy of the incident photon is discussed. Moreover, the effects of gas thickness, gas component, gas ratio, and photon energy on the detection efficiency are analyzed.

First, the response relationship of the emission position and azimuthal angle distribution of the photoelectron with the polarization direction and energy of the incident photon is clarified. The emission direction distribution probability of the photoelectron is the largest in the polarization direction of the incident photon, and the azimuthal angle distribution can be approximated as a cosine squared function. With the increase in photon energy, the counts of photoelectrons at each angle decrease in different degrees, but all of them show a statistical law that the maximum values occur when the azimuthal angle is 0 or π (-π) (Fig. 6). Moreover, the effects of gas thickness, gas component, gas ratio, and photon energy on the detection efficiency are revealed and quantified. For 2 keV photons entering into 90%Ne+10%DME gas mixture, when the gas thickness is small, the detection efficiency increases rapidly with the increase in gas thickness, from less than 0.1 at 0.1 cm to 0.64 at 1 cm (Fig. 7). When the gas thickness increases to 3 cm, the detection efficiency is greater than 0.9. Then, with the increase in gas thickness, the detection efficiency gradually approaches 1. For the CF₄+C₄H₁₀, Ne+CF₄, Ne+DME, DME+CO₂, and He+C₃H₈, the detection efficiency decreases with the increase in photon energy, and the large average atomic number of gas can lead to a high detection efficiency (Fig. 8). While for the Xe+CO₂, Ar+CO₂, and Ar+CH₄, when the photon energy is greater than the binding energy of certain shell electrons of Xe or Ar atoms, the detection efficiency will be improved to a certain extent because the corresponding shell electrons begin to be ejected. In addition to the Ar+CO₂ which is affected by the electron emission in K-shell, the detection efficiency in each energy range can be effectively improved by increasing the proportion of gas with high atomic number (Fig. 9).

We simulate the polarimetric photoelectric process of 2-10 keV linearly polarized X-ray photons in several commonly used working gases by the Monte Carlo code Geant4. The response relationship of the emission position and azimuthal angle distribution of the photoelectron with the polarization direction and energy of the incident photon is clarified. The emission direction distribution probability of the photoelectron is the largest on the polarization direction of the incident photon, and the azimuthal angle distribution can be approximated as a cosine squared function. With the increase in photon energy, the counts of photoelectrons at each angle decrease in different degrees, but all of them show a statistical law that the maximum values occur when the azimuthal angle is 0 or π (-π). Moreover, the effects of gas thickness, gas component, gas ratio, and photon energy on the detection efficiency are revealed and quantified. The larger gas thickness and larger average atomic number can lead to higher detection efficiency. In addition, the increase in photon energy can result in a decrease in detection efficiency. However, for the working gases composed of Xe or Ar, when the photon energy is greater than the binding energy of a certain shell electron, the detection efficiency will be improved to a certain extent because the corresponding shell electrons begin to be ejected. The results in this paper can provide some theoretical basis and data support for the structure design of X-ray polarization detectors. In the actual selection of working gases, the drift properties of electrons in gases, the effect of photoelectron drift and diffusion on track thickness and length, and the reconstruction efficiency of the track reconstruction algorithm should also be considered.