Mplanfl

Metal evaporation was performed with Plassys MEB 400 (Plassys, Bestek, France). A hot plate (Thermo Fisher Scientific, Waltham, MA, USA) was used for annealing under clean room conditions.
Nanostructured coverslips were characterized with a scanning electron microscope (SEM) (FEG-SU8030, Tokyo, Japan) and an atomic force microscope (AFM) (Bruker ICON, Billeric, MA, USA) with cantilever ScanAsyst-Air in silicon nitride with a tip height of 2.5–8.0 mm. A spring constant of 4 N/m and a reflective aluminum coating on the back side in standard ScanAsyst-Air mode were used to characterize the morphology of AuNPs (data not shown).
SERS spectra were recorded with backscattering geometry using a modified Jobin-Yvon LabRAM (Horiba scientific, Longjumeau, France) and an excitation wavelength of 632.8 nm (11 mW) from the He–Ne laser source, and all the spectra were recorded with a 10× objective Olympus MPlanFl with a 5.2 μm² laser spot area. The acquisition time varied from 10 to 120 s, and all the spectra were recorded 3 times with a D filter range between 0 and 0.3.
For sterilization, a Tuttnauer Autoclave Steam Sterilizer 2540ML (Tuttnauer, Villenoy France) was used. The samples were dried in an oven provided by VWR company (DRY-Line drying oven DL 53), and all operations were made under a biological hood provided by Thermo-scientific MSC 1,2 ADV (Illkirch Cedex, France).

Scattering properties of the fabricated nanostructures were studied using spatially resolved linear reflection spectroscopy. The spectroscopic reflection analysis was performed on a BX51 microscope (Olympus) equipped with a halogen light source, polarizers and a fiber-coupled grating spectrometer QE65000 (Ocean Optics) with a wavelength resolution of 1.6 nm. The reflected light was collected in backscattering configuration using MPlanFL (Olympus) objectives with magnifications × 50 (NA = 0.75). The microscope images (1600 × 1200 pixels) were captured with a LC20 digital color camera (Olympus). The experimental data in Fig. 3D represent the reflectivity determined as the ratio R_str/R_ref, where R_str is the reflection measured from the structure and R_ref is the reference spectrum recorded from the reference broadband laser mirror (Edmund Optics, 11 NT64-114) that exhibits an average reflection of 99% between 350 and 1100 nm of light wavelengths.

A hyperspectral imaging system (Cytoviva HISV3) was adapted to perform the reflectance measurements. Reflectance spectra were recorded by the spectrometer (Horiba iHR550) with a ×10 objective (Olympus MPlanFL, NA = 0.25). Relative reflectivity was utilized to represent the reflectance spectra, which was obtained by dividing the reflected intensity of the sample by that of the substrate. Nonpolarized white light passed through a linear polarizer (Thorlabs LPNIRE100-B) with the polarization direction along the x-axis. As the normal incident white light was confined by the objective, the in-plane wavevector k_x was not zero despite the value being small compared with the momentum provided by gratings. However, this issue could not be neglected as it resulted in a splitting of uncoupled SPP modes (SPP₁ and SPP₂).