Modeling of the propagation of Bessel beams in a uniaxial crystal at different positions of the crystal axis

Abstract

Abstract. The numerical study of the propagation of the Bessel beams in anisotropic media with different orientations of the crystal axis and different polarizations of the input beams has been carried out using distributed computing on supercomputers. The analysis is based on two models: the geometric optics model based on ray tracing and implemented using the ZEMAX software, and the wave model in the approximation of thin optical elements based on plane waves expansion and implemented using the VectorAnisotropicPropagators package. High-performance computing resources have been necessary to implement the wave model. The results obtained can be used in various fields of optical design, for example to determine the position of the crystal axis and for the development of devices that perform polarization conversion. IntroductionOne of the major challenges for optical information science is the calculation of the propagation of electromagnetic waves in different media. Thus, optical devices that allow certain properties of electromagnetic radiation to be converted have acquired growing interest and practical use. Most often, mode and polarization conversions are needed. One tool for such transformations are anisotropic crystals [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20]. One method of modelling the propagation of electromagnetic waves in anisotropic media is the method of plane waves expansion [21][22][23][24][25][26]. As a rule, the use of this method causes large computational problems due to the need to calculate the direct and inverse Fourier transforms for all electromagnetic field components. One solution is the use of various fast algorithms, including FFT, which significantly reduces computation time. However, the use of the FFT algorithm has drawbacks associated with the fixed discreteness of input and output signals and the ability to calculate only the transverse distribution. Another option is the use of parallel algorithms using high performance computing resources (supercomputers).