286
J. Eur. Opt. Society-Rapid Publ. 22, 28( 2026)
Table 1. Process parameters for metal layer deposition and thermal annealing.
Material |
Thickness( nm) |
Deposition rate( Å / s) |
Pressure( mbar) |
Thermal annealing conditions |
Ag |
15 |
0.3 |
2.5 10-6 |
250 ° C, 30 min |
|
20, 25 |
|
|
250 ° C, 1 h |
Al |
2, 4, 6 |
0.1 |
> 10 �6 |
– |
Fig. 1. SEM micrographs showing the morphology after the Al layer deposition process:( a) 2 nm,( b) 4 nm, and( c) 6 nm.
aluminium undergoes a morphological transition from isolated islands to a gradually interconnected percolated network of nanostructures as thickness increases from 2 to 6 nm. The statistical analysis( inset to Fig. 1) reveals a systematic increase of the mean particle size with increasing aluminium thickness. Specifically, the mean size increases from 11.2 nm for the 2 nm Al layer to 24.1 nm for the 6 nm Al layer, while the corresponding standard deviation increases from 2.2 nm to 4.7 nm, indicating enhanced particle coalescence and growth at higher nominal thicknesses. The resulting size histograms also show the FWHM values for each layer( 9 – 14, 10 – 15, and 18 – 29 nm).
The enhancement factor varies depending on the shape and distance of the metallic nanoaggregates [ 32 ]. Taking this into account, we collected the fluorescence emission across the entire surface of the nanostructured platform. Due to the resulting surface filling factor and the consistent shapes and sizes of the nanoaggregates achieved under the same deposition or annealing conditions, the MEF remains relatively unchanged when analysing new samples, ensuring high reproducibility. The steady-state fluorescent emission spectra of CV, NR, and RB on nanostructured Al films were compared to a reference consisting of a silicon substrate without metasurfaces coated with fluorophore. Figure 2 shows the obtained fluorescence spectra, with an inset graph on a logarithmic scale illustrating the magnitude orders of the best enhancement result. The fluorescence enhancement depends on spectral overlap between excitation / emission bands and plasmonic scattering / nearfield response. For all three fluorophores, nanostructured Al induced changes in fluorescence intensity without significantly affecting the characteristic spectral band profiles. This indicates that the dominant interaction is photonic( near-field / plasmonic) and not a chemical modification of the fluorophore. Such behaviour is largely associated with MEF, in which local electromagnetic fields and modified radiative decay rates can increase emission intensity without necessarily changing spectral maxima.
The EF are calculated relative to the respective emission maxima of each fluorophore. The values vary depending on the type of fluorophore – specifically, their quantum efficiency – rather than the layer thickness. For NR, we obtain a maximum EF of approximately 3.2 when we use a thin film of 4 nm thickness. In the meantime, for CV and RB, we obtain maximum EF values of 2.8 and 7.7, respectively, when we use a thickness of 6 nm.
Silver thin films with thicknesses of 15 nm, 20 nm, and 25 nm were deposited on silicon and subsequently subjected to thermal annealing. This treatment induced the formation of metal aggregates, resulting in randomly distributed meta-atom structures. Figure 3 shows SEM images of the annealed Ag layers at all three thicknesses, revealing the morphology of the formed nanostructures. SEM analysis demonstrates a systematic thickness-dependent morphological transition in the annealed Ag films. The 15 nm film exhibits isolated, quasi-ellipsoidal nanoparticles with minimal inter-particle connectivity. At 20 nm, partial aggregation occurs, forming bigger anisotropic islands with increased contact between adjacent structures. The 25 nm film displays complete percolation, characterized by an interconnected spiderweb-like metallic network.
Similar to Al films, the particle size statistics for Ag nanoaggregates( see inset in Fig. 3) show an increase in mean particle size with layer thickness: 124 nm for the 15 nm layer, 189 nm for the 20 nm layer, and 234 nm for the 25 nm layer. The corresponding standard deviation increases as well, from 34 nm to 57 nm. Furthermore, the surface filling factor, which is defined as the fraction of the substrate surface covered by Ag nanoparticles, increases with layer thickness. Specifically, it increases from 32 % for the 15 nm layer, to 37 % for the 20 nm layer, and to 76 % for the 25 nm layer. This increase in surface coverage reflects