Generalized Fiber-Optic Surface-Plasmon-Resonance Modeling Through Ray-Tracing

Fiber-optic surface plasmon resonance (FO-SPR) sensors look at the absorbance of reflected light to measure changes in refractive index (RI). FO-SPR sensor modeling is essential in understanding the underlying processes that induce such RI changes. Despite two types of rays, i.e., skew and meridional rays, an FO-SPR model has been developed in the literature that only considers meridional rays. This meridional model has proven its applicability in several publications using pass-through FO-SPR sensors. However, this simplified FO-SPR model fails to simulate back-reflecting FO-SPR sensors properly, where diffuse light is delivered and collected at the same optical fiber end. Here, it is shown that a generalized FO-SPR model that includes skew rays more accurately simulates the spectra obtained in back-reflecting FO-SPR sensors. With the changing incidence plane of a skew ray in mind, the generalized FO-SPR model was built with three-dimensional polarization ray-tracing calculus. The necessary angular ray distribution of the back-reflecting FO-SPR sensor was acquired by a Monte Carlo three-dimensional ray-tracing simulation. Next, the effect of including optical components and deviations and model optimization by adjusting the gold relative permittivity and thickness was evaluated. The generalized model simulated FO-SPR absorbances with smaller widths than the experimental FO-SPR absorbances. The cause of this difference in absorbance is unclear and demands more research. Nevertheless, the skew ray incorporation in the generalized FO-SPR model enabled its application to a greater diversity of FO-SPR sensors compared to the simplified FO-SPR model both as a predictive and an analytic tool in the development of FO-SPR sensors.