SERS Hotspot Engineering by Aerosol Self‐Assembly of Plasmonic Ag Nanoaggregates with Tunable Interparticle Distance

Abstract Surface‐enhanced Raman scattering (SERS) is a powerful sensing technique. However, the employment of SERS sensors in practical applications is hindered by high fabrication costs from processes with limited scalability, poor batch‐to‐batch reproducibility, substrate stability, and uniformity. Here, highly scalable and reproducible flame aerosol technology is employed to rapidly self‐assemble uniform SERS sensing films. Plasmonic Ag nanoparticles are deposited on substrates as nanoaggregates with fine control of their interparticle distance. The interparticle distance is tuned by adding a dielectric spacer during nanoparticle synthesis that separates the individual Ag nanoparticles within each nanoaggregate. The dielectric spacer thickness dictates the plasmonic coupling extinction of the deposited nanoaggregates and finely tunes the Raman hotspots. By systematically studying the optical and morphological properties of the developed SERS surfaces, structure–performance relationships are established and the optimal hot‐spots occur for interparticle distance of 1 to 1.5 nm among the individual Ag nanoparticles, as also validated by computational modeling, are identified for the highest signal enhancement of a molecular Raman reporter. Finally, the superior stability and batch‐to‐batch reproducibility of the developed SERS sensors are demonstrated and their potential with a proof‐of‐concept practical application in food‐safety diagnostics for pesticide detection on fruit surfaces is explored.


Table of Contents
Section S1: SERS sensing films fabrication by flame aerosol particle synthesis and deposition. .         The nanoparticle film porosity was calculated using the equation: where , , , , and ℎ are the film porosity, mass of the film, density of the particles, area of the film deposited on the cover glass, and film thickness, respectively. The mass of the film is obtained through weighing the cover glass before and after deposition. The density of particles was used as the mixed density of Ag and SiO2, considering Ag-SiO2 NPs has the SiO2 content of 6 wt%. The area of the film was taken as the cover glass area. The film thickness was measured using SEM instruments. Film porosity, (-) 96.5% 96.8% Figure S6. The Raman spectra from the cover glass and the SERS substrate without R6G deposition. The laser power was enhanced to 25 mW to obtain clear signals.
Section S2: The calculation of enhancement factor of the fabricated SERS substrate.
Enhancement factor (EF) is used to quantify the promoted Raman signal intensity per molecule by SERS compared to normal Raman. Here, EF is calculated using a hybrid approach, which is based on the measurements of normalized intensities and on the approximate estimation of the surface density of excited molecules. [1] = where , , , and are the Raman intensity, laser power, accumulation time, and the number of effective molecules, respectively. The subscripts, and , represent the measurements in the SERS test and the normal Raman test, respectively. The inputs and outputs of the calculation are listed in Table S2.
The SERS measurement was conducted on R6G molecules deposited on the Ag-SiO2 nanostructured SERS substrate ( Figure S7 a). 150 µL 10 -4 M R6G in ethanol solution was first dropped on the surface of the SERS substrate, and then spin coating was done to form uniform R6G layer. The size of Ag-SiO2 NPs is assumed as 8 nm, which is the same as the crystal size from the XRD measurement. Occupied Surface area of R6G is assumed as 1 nm 2 and R6G molecules are assumed to cover 2.3% of the totally surface of a nanoparticle. [2] The normal Raman measurement was done on R6G powders ( figure S7 b), which have the molecular weight of 479.01 g/mol and the density of 0.79 g/cm 3 . The effective molecules are considered as the solid powders in the laser spot aera.     Section S3: The SERS measurements using a portable Raman spectrometer system. Table S4. Comparison between the sophisticated Raman microscope system and the portable Raman spectrometer system used in this work. The excitation laser with a wavelength of 532 nm was used in both systems.

Raman microscope Raman Spectrometer
Laser power, P (mW) caused by the small detection depth of the used laser beam, which could not detect all the effective molecules exposed at the substrate surface. All data are represented mean ± SD (n =

4-12).
Section S4. SERS performance optimization by increasing the nanoparticle size.       Figure S16. Simulation of extinction spectra of fractal like aggregates with (a) varying primary particle size and (b) with the primary particles sizes and geometric standard deviations corresponding to those measured from TEM images of aggregates synthesized with a given silica weight percentage (shown in Figure S2). PP in (a) represents primary particle.