Intense electromagnetic (EM) hot-spots arising at the junctions or gaps in plasmonic nanoparticle assemblies can drive ultrahigh sensitivity in molecular detection by surface-enhanced spectroscopies. Harnessing this potential however requires access to the confined physical space at the EM hot-spots, which is a challenge for larger analytes such as biomolecules. Here, we demonstrate self-assembly derived gold nanoparticle cluster arrays (NCAs) on gold substrates exhibiting controlled interparticle (<1 nm wide) and intercluster (<10 nm wide) hot-spots as highly promising in this direction. Sensitivity of the NCAs toward detection of small (<1 nm) or large (protein–receptor interactions) analytes in surface-enhanced Raman and metal-enhanced fluorescence assays is found to be strongly impacted by the size of the cluster and the presence of reflective substrates. Experiments supported by numerical simulations attribute the higher sensitivity to higher EM field enhancements at the hot-spots, as well as greater analyte leverage over EM hot-spots. The best-performing arrays could push the sensitivity down to picomolar detection limits for sub-nanometric organic analytes as well as large protein analytes. The investigation paves the way for rational design of plasmonic biosensors and highlights the unique capabilities of a molecular self-assembly approach toward catering to this objective.

All-semiconductor plasmonic gratings are investigated by spectroscopic nanoimaging in the vicinity of the plasma frequency, where the material behaves as an epsilon near-zero (ENZ) material. Both phase-sensitive scattering type nanoscopy (s-SNOM) and photoinduced force microscopy (PiFM) are carried out on this structure. The obtained data and models reveal that PiFM, as for s-SNOM, can have a mostly dispersive line shape, in contrast with recent near-field spectra obtained with photothermal AFM nanoscopic imaging on ENZ material where absorption maxima are observed. On the obtained result, PiFM signal exhibited better sensitivity to the dielectric function variation while interferometric s-SNOM can provide additional phase information. Localized surface plasmon resonances (LSPR), highly confined on the structure edges were also observed with both techniques. A higher sensitivity was observed with PiFM for both dielectric contrast imaging and LSPR observation. In addition, for both microscopies, the near-field response is phenomenologically described using a similar formalism based on dipole-image dipole approach. In this model, the sensitivity difference between both techniques is mostly accounted for by probes having different polarizabilities.

Novel plasmonic structures are on the rise, with applications varying from sensing and spectroscopy to solar cells and biological therapies. In this work, we introduce a plasmonic metasurface with a very rich dispersion spectrum, measured both experimentally and numerically. It shows a tunable absorption that depends on the folding angle and periodicity. A detailed numerical analysis identifies the presence of quasi-omnidirectional absorption. This broad directional absorption mode matches a Fabry–Perot resonance of a surface plasmon polariton along an elementary segment of the periodic structure. This geometry induced wide directional absorption is highly promising for a variety of photonic, light harvesting, and sensing applications.

High sensitivity in plasmon-enhanced spectroscopies results not only from high electromagnetic (EM) field enhancements at the vicinity of nanostructured metal surfaces but also the ability to leverage on these enhancements via analyte colocalization with the EM hot spots. However, promising configurations for EM hot spots such as metal nanogaps are spatially restrictive for adsorption of larger analytes such as proteins. This results in an adverse spatial trade-off in the design of EM hot spots, viz., increasingly confined geometries sought to drive high EM enhancements and space necessary to accommodate binding of large biomolecular analytes. In this direction, we demonstrate gold nanoparticle cluster arrays (NCAs) exhibiting interparticle (<1 nm wide) and intercluster (<10 nm wide) EM hot spots, with cluster size and densities engineered to enhance the leverage of biomolecular analytes over EM hot spots. Surface-enhanced Raman and fluorescence assays of streptavidin−dye conjugates on biotinylated NCAs show higher sensitivity to be achieved using NCA geometry exhibiting greater density of intercluster hot spots. Quartz crystal microbalance measurements reveal the higher sensitivity to be achieved despite lower nanoparticle as well as analyte surface densities as compared to less sensitive NCA geometries. The results are aligned with the expectation that the large dimensions of protein analytes allow better leverage of intercluster rather than interparticle hot spots and consequently be detected with better sensitivity on NCAs presenting higher intercluster hot-spot densities. The investigation supports the need to factor in the size of the analyte into the design of the EM hot spots to achieve high sensitivity in plasmon-enhanced spectroscopic sensors.

Nanoplasmonic sensors based on surface-enhanced spectroscopies carry profound promise toward ultrasensitive detection of biomolecular analytes within miniaturized measurement footprints. High sensitivity in these sensors is achieved by intense electromagnetic (EM) enhancements at hot spots and colocalization of analytes with such hot spots. However, EM hot spots exhibiting high EM enhancements often present confined areas that render them inaccessible to large analytes such as biomolecules. Addressing this requires rational engineering of a nanoplasmonic interface that factors in the analyte surface concentrations and how they are distributed with respect to the EM hot spots. Here we demonstrate combination of metal-enhanced fluorescence (MEF) with a quartz crystal microbalance (QCM) through fabrication of highly resolved plasmonic nanoarrays directly on the QCM sensor. QCM and MEF are simultaneously employed to monitor in situ, real-time binding of fluorescently labeled protein to receptor-functionalized plasmonic arrays. The correlation between the QCM and MEF responses is used to quantify surface density of protein that contributes to the MEF signal intensities and obtain analyte distribution with respect to the EM hot spots by geometric modeling. Results further reveal the MEF assays to be sensitive down to ∼2 zmol of protein within measurement footprints that are 8 orders of magnitude smaller than that of the QCM.