Open AccessGeneralized method for seeking q-preserving configurations of multi-pass cells
Author(s) -
Zheng Yang,
Ming-Li Zou,
Liqun Sun
Publication year - 2019
Publication title -
optics express
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.394
H-Index - 271
ISSN - 1094-4087
DOI - 10.1364/oe.27.014054
Subject(s) - matrix (chemical analysis) , transfer matrix , gaussian , optics , mathematics , norm (philosophy) , identity matrix , physics , algorithm , computer science , eigenvalues and eigenvectors , materials science , quantum mechanics , political science , law , composite material , computer vision
A method is presented for seeking specified multi-pass cell (MPC) configurations with q-parameters of the input and output Gaussian beams that remain the same (q-preserving configurations). The Frobenius norm (F-norm) is chosen as the quantitative criterion of the deviation between the transfer matrix of the MPC configurations and identity matrix. When the deviation is down near to zero, the corresponding configuration is close to an ideal q-preserving configuration. In contrast to common 2 × 2 transfer matrixes that are only applicable to MPCs with negligible astigmatism, we adopt 4 × 4 transfer matrixes that are also workable for multi-pass systems with astigmatism. To demonstrate the validation of the method, several q-preserving structures of four-objective multi-pass matrix system (FO-MMS) and double-row circular MPC (DR-CMPC), which are two fundamentally different MPC types, are illustrated. The variation tendencies of the quantitative deviation between the transfer matrix and the identity matrix are discussed in detail for each structure as functions of structural parameters. The optimal q-preserving structures are found for FO-MMS and DR-CMPC, with the matrix deviation of 0.0047 and 0.0051, while the corresponding total optical path length (OPL) are 33.6 m and 67.8 m, respectively. The OPL of the optimal DR-CPMC is longer than the traditional CMPC in the state of the art, and the deviation is three orders of magnitude smaller if the similar spherical reflecting surfaces are adopted.