Plasmonic magneto-optical 1D nanostructure: Wood's anomaly and the Faraday rotation for biosensing

Plasmonic nanostructures are actively investigated in order to make sensors portable and more sensitive [1]. In particular, the way to improve sensitivity is utilization of the magneto-optical multilayer in structure to... [ view full abstract ]

Plasmonic nanostructures are actively investigated in order to make sensors portable and more sensitive [1]. In particular, the way to improve sensitivity is utilization of the magneto-optical multilayer in structure to measure a magneto-optical (MO) response instead of the amplitude of transmitted or reflected wave [2].

In the present work, we demonstrate 500-mm-sized slabs of plasmonic 1D nanostructures for detecting low density lipoproteins (LDL): sample A–dielectric nanoridges coated with 40-nm-thick Au and sample B is a similar one based on bismuth-substituted yttrium iron garnet (Bi:YIG). We experimentally and theoretically studied their optical spectra, gave an explanation on resonant light coupling and optimized designs of the structures. For these designs, we measured and numerically analyzed optical and magneto-optical (MO) responses in the transmission geometries when modeling biomolecular binding for LDL detection with a low concentration. Numerical models of sample A and B were set up for one unit cell of the 1D grating; experiment and calculations were carried out for the E_x-polarized light (see Fig.1).

We have found that, by introducing Bi:YIG layer, the sensitivity can be significantly enhanced when measuring the MO response in the vicinity of the Fano peak. Indeed, a change in the angle of Faraday rotation Δθ = 1º was demonstrated for sample B with optimized parameters. The sensitivity of sample B was 1.56*10^-7 RIU when analyzing the direction of polarization plane. Note that the sensitivity of sample A was 2*10^-6 RIU when measuring the amplitude of transmitted wave.

Fig. 1. A sketch of the unit cells of experimental sample A (D = 650 nm, h_HSQ = 100 nm, d_Au = 40 nm, d_Ti = 5 nm, d_HSQ = 65 nm, h_HSQ = 80 nm) and the model of sample B (D = 500 nm, d_BiYIG= 140 nm, d_Au = 30 nm, h_R = 110 nm, d_R = 60 nm). Structural parameters for optimized samples are given in parentheses.

[1] B. Špačková , P. Wrobel , M. Bocková , J. Homola, Proceedings of the IEEE. 104 (2016) 2380 - 2408.

[2] A. V. Baryshev , K. Kawasaki , M. Inoue, Proc. BioPhotonics 2011, 12123238.