Efficient numerical representation of the optical field for the propagation of partially coherent radiation with a specified spatial and temporal coherence function

Abstract

We propose a method to narrow the gap between the rigorous methods for the propagation of partially coherent light, which require excessive computational capacity, and the numerical methods used in practical engineering applications, where it is not clear how to handle spatial and temporal coherence in a statistically correct manner. As is the case for the latter methods, the numerical method described can deal with fields with a large spatial and temporal extent, which is necessary in practical applications such as laser fusion or optical lithography. However, the method also takes a few steps toward a more rigorous, yet efficient, representation of the optical field, which depends on detailed specified coherence properties of the radiation. The described method uses a set of independent monochromatic fields at different oscillation frequencies. The frequencies are chosen such that the statistical properties of the integrated intensity closely resemble those from a full-time trace treatment. Finally, we demonstrate the capabilities and limitations of the method with a few numerical examples of the propagation of a large field with a specified spatial and temporal coherence.

Numerical sampling rules for paraxial regime pulse diffraction calculations

“…We believe the analysis presented in this paper complements that in [15], providing an intuitive graphical picture that describes how undersampling the pulse results in numerical artifacts in the spatial domain. Related work on the propagation of partially coherent radiation is discussed in [40] and the references therein.…”

The Structure of Partially Coherent Fields

“…Another technique for generating spatio-temporal realizations was introduced by Gbur (2006), who constructed the realizations from a collection of pulses of appropriate spatial and temporal shape. Around the same time another method was introduced by Rydberg and Bengtsson (2006) in which realizations are generated from a finite superposition of independent monochromatic fields.…”

Full characterization of a high-power semiconductor disk laser beam with simultaneous capture of optimally sized focus and farfield

“…For a branched system, where at least one intermediate plane is involved, a method where the arrival plane sampling distance can be more freely chosen is most convenient. In this work one such method was used: the FFT-based two-step propagation method [22], which is a formalized version of the strategy described in [16] and mathematically identical with the propagation method based on fractional Fourier transforms described in [23]. Just like the conventional methods, it can easily be modified for inverse propagation, i.e.…”