Fig. 1. (a) Calculated electric field in resonance condition at the BIC mode at the Γ-point of the Brillouin zone; top view of four unit cells with lattice constant a. (b) BIC amplitude over the PhCM with superimposed arrow maps of the electric field; as clearly visible, the electric field forms a lattice of vortices and antivortices that cannot couple to radiating waves, revealing the bound-in-the-continuum character of the calculated mode. (c) Intensity profile of the electric field (side view of one unit cell) showing that the BIC wave is mostly confined at the interface Si3N4/SiO2, but the electromagnetic field enhancement at Si3N4/air interface is expected to be large enough to provide strong light–matter interaction. Intensity enhancement of the simulation diverges in agreement with an ideal diverging Q-factor of an infinite structure.

Fig. 2. (a) Scanning electron microscopy image of the PhCM sample. The design consists of air cylindrical holes arranged in a square lattice (a=521  nm, r=130  nm, h=78  nm). (b) Sketch of the device: a PDMS microfluidic chamber was bonded to the PhCM. The inlet and outlet allow the controlled injection of the fluid.

Fig. 3. (a) Sketch of the experimental setup. SC source, supercontinuum source; P1, Glan–Thompson polarizer; R, automatic rotational stage; PhC, photonic crystal sample; P2, Glan–Thompson polarizer; S, spectrometer. (b) Experimental reconstructed band collected for p-polarized incident beam.

Fig. 4. (a) BIC resonance excited in the PhC metasurfaces without the micro-chamber by a normally incident incoming beam. By fitting the measured spectrum (blue dots) with a Lorentzian line shape (purple curve), a linewidth and a quality factor as large as c/γ=0.4  nm and Q≃2×103, respectively, were determined. (b) Two measured transmitted spectra collected from the sensor device corresponding to different RI. (c) Reconstructed sensitivity curve; the linear fit (red curve) to the experimental data (blue dots) revealed a bulk sensitivity S=178  nm/RIU (R-square 0.99). Error bars refer to spectrometer resolution, whereas the statistical error on spectral peak position is within the size of the symbols. (d) Sensitivity curves corresponding to θ=5° (red dots), θ=0° (black squares), and θ=−5° (blue triangles). The sensitivities determined by linearly fitting the data were found close to each other.

Fig. 5. Reconstructed sensitivity curve in the visible range; the measurements demonstrated the scalability of the device, which reveals a bulk sensitivity S=185  nm/RIU (R-square 0.96).

Fig. 6. Cross-polarized transmission spectrum of the PhCM-based sensor prior (black curve) and post-functionalization with the self-assembling monolayer of BPT (red curve). When the BPT monolayer is assembled, a redshift at resonance wavelengths 1 and 2 occurred. A sensitivity of 6 nm of shift after the BPT monolayer formation was measured.