Simultaneous Surface-Enhanced Raman Scattering with a Kerr Gate for Fluorescence Suppression

The combination of surface-enhanced and Kerr-gated Raman spectroscopy for the enhancement of the Raman signal and suppression of fluorescence is reported. Surface-enhanced Raman scattering (SERS)-active gold substrates were demonstrated for the expansion of the surface generality of optical Kerr-gated Raman spectroscopy, broadening its applicability to the study of analytes that show a weak Raman signal in highly fluorescent media under (pre)resonant conditions. This approach is highlighted by the well-defined spectra of rhodamine 6G, Nile red, and Nile blue. The Raman spectra of fluorescent dyes were obtained only when SERS-active substrates were used in combination with the Kerr gate. To achieve enhancement of the weaker Raman scattering, Au films with different roughnesses or Au-core-shell-isolated nanoparticles (SHINs) were used. The use of SHINs enabled measurement of fluorescent dyes on non-SERS-active, optically flat Au, Cu, and Al substrates.


Instrumentation
Continuous wave Raman spectra were recorded with an inVia Raman (Renishaw) spectrometer equipped with thermoelectrically cooled (−70 °C) CCD camera and a confocal Leica microscope.A He-Ne laser beam at 633 nm, 50×/0.75NA objective lens and 1800 lines/mm gratings were used.Laser power at the sample surface was restricted below 0.01 mW.All spectra were recorded using 10 s integration time, 1 scan accumulation and were corrected by polynomial function background subtraction.No smoothing function was applied.
Kerr-gated Raman measurements were carried out at ULTRA laser facilities at Central Laser Facility (Science and Technology Facilities Council [STFC], Rutherford Appleton Laboratories, UK) and the experimental set-up has been described previously. 1 Herein, Kerrgated Raman scattering was achieved under 633 nm laser pulse (2.6 mW, 10 kHz, 2 ps) with a 75 x 75 μm 2 spot size.The Kerr medium was activated by a gating pulse (800 nm, 240 μJ, 2 ps, 10 kHz).All spectra were collected and averaged over 4 repeats, each with an acquisition time of 60 s.The slit size was 300 µm and the spectra were collected under rastering conditions.The Raman shift was calibrated against the spectrum obtained for toluene.
UV/vis extinction/absorption spectra were collected in the range from 350 to 700 nm with an Evolution 201 UV/Visible spectrometer (Thermo Fisher Scientific).Transmission electron microscopy (TEM) was used for the characterization of the morphology of the NP and thickness of the coating using a JEM-2100 Plus (JEOL) operated at 200kV accelerating voltage.Scanning electron microscopy (SEM) images were acquired with a JSM 7001F FEG-

SEM (JEOL).
A conventional three-electrode electrochemical cell was used for the electrochemical studies.Charge measurements were carried out from the first scan CV using a VSP3 potentiostat (Biologic Science Instruments).The counter electrode was a gold wire of sufficiently large area, and all the potentials were measured against the Ag/AgCl (saturated KCl) reference electrode.The electrolyte was 0.5 M H2SO4 prepared in Milli-Q water (Millipore).All the experiments were conducted in an atmosphere of oxygen-free N2 gas using a sweep rate of 50 mV s -1 and scan range from -0.2 V to 1.3 V.

Materials & methods
All reagents were acquired from Sigma Aldrich in the maximum purity available (ACS reagent grade) and used without further purification.

Substrate identification and preparation
Five materials were used as substrates for the spectroscopic measurements in this work, in addition to microscope cover and their respective labels, and methods for preparation, are described below.
"Au film" -Au films (100 nm thick) were deposited onto microscope cover glass by thermal evaporation of Au wire (99.999%,Advent) with an Oerlikon Leybold Vacuum Univex 300.
"Au on wafer" -Au-coated (50nm layer thickness) Si wafers (Platypus Tech) were cut into 0.5 x 0.5 cm 2 pieces, cleaned with piranha solution, rinsed with Milli-Q water, and dried with N2 before use."Au foil" -Au foil (0.02 mm thickness on 5 μm polyester support, 99.9%, Goodfellow Cambridge Ltd.), was manually polished with alumina suspension down to 0.05 µm-particle and further sonicated in acetone and Milli-Q water before use."Al foil" and "Cu foil" -Al and Cu foils (0.035 mm thickness) were acquired from RS Components and used without processing.

SHIN synthesis and preparation procedure
Silica-coated Au NPs (SHell-Isolated Nanoparticles, SHINs) were synthesized following the method described elsewhere. 2Briefly, 55 nm Au NPs were synthesized by chemical reduction of HAuCl4.3H2O with sodium citrate. 3,4 he citrate stabilizing layer was then replaced by (3-aminopropyl)triethoxysilane and subsequently by a silica shell, under strict control of pH and temperature. 5The effective coating of the Au core was demonstrated by the conventional pinhole test using 10 mM pyridine (99.8%,Sigma-Aldrich) as indicator. 2r the spectroscopic measurements involving SHINs, a layer of the SHINs was deposited onto the substrate under investigation by drop-casting 20 µL of concentrated SHINs dispersion.The substrate was then dried under gentle vacuum conditions to remove water.

Chemisorbed Rhodamine 6G (Rh6G) preparation
For the spectra presented in Figure 1 [a(i), a(ii) and b], the Au film substrate was immersed in a solution of Rh6G in ethanol for 5 h, then rinsed with H2O (high purity) and dried under N2 before measurements.For spectra in Figure 1

Characterization of NPs
In SHINER spectroscopy, the structural parameters of the plasmonic core will determine the efficiency of the surface enhancement and, therefore, the enhancement of the signal in the vibrational spectrum of target analytes.Specifically, the wavelength corresponding to the localized surface plasmon resonance (LSPR) is conditioned by the nature of the metal, shape, size and degree of aggregation of the nanoparticles. 6The plasmonic resonance extinction peaks of 50 nm (average diameter) Au NPs (black) and SiO2-coated Au NPs (SHINs, grey) are shown in Figure S1a.Au showed resonant behavior when interacting with ultraviolet and visible (UV-vis) photons and the silica shell did not affect significantly the optical properties of Au NPs.The SiO2 shell, required to chemically isolate the metallic core and inhibit surface chemical or electrical interactions with the analyte needs to be robust as well as sufficiently thin as to preserve the relative short-range effect of the plasmonic electromagnetic field.
Herein, the shell thickness for the as-prepared SHINs was found to be uniform with approximately 3 nm thickness (Figure S1b).To achieve measurable absorption contributions from SHINs, these were deposited using multiple drops from a concentrated dispersion of SHINs.Additionally, to enable transmission mode measurement of UV-visible absorption spectra of Au films analogous to those used in the Kerr-gated Raman measurements, thinner (20 nm) Au layers were deposited to permit a reasonable optical transparency.The spectra B and C in panel (b) were collected at slightly different regions of the microscope slides (accounting for the non-uniform distribution of SHINs across the Au surface).The data presented here is a redrawing of select data presented originally in Figure 2 and Figure 3 where broad features in the 900-1200 cm -1 regions were observed.The dashed lines, and associated labels (in wavenumbers), highlight the apparent primary band maximum positions and their overlap with minor bands associated with the pure dye molecules (black traces in each panel).It is noted that the band shapes, relative intensities, and inconsistent appearance make reliable assignments to the dye molecules (or solvents) a challenge and their precise assignments are not fully understood.

Figure S1 .
Figure S1.Extinction spectra of 50 nm Au NPs (black) and SHINs (grey) showing their respective plasmonic resonance extinction peaks (a).High-resolution TEM image of a single 55nm Au nanoparticle coated with a 3 nm SiO2 layer, before and after exposure to Kerr-gated Raman laser (b & c, respectively).

Figure S2 .
Figure S2.UV-vis absorbance spectra of (a) SHINs deposited on a glass microscope slide and of (b) a glass microscope slide with a Au film (20 nm) without (black trace) and with (blue and orange traces) layers of SHINs.The difference spectra (c) for the Au layer with and without the addition of SHINs was calculated by subtraction of the Au film only spectrum (A) from the spectra of SHINs plus Au film (B and C).

Figure S4 .
Figure S4.Cyclic voltammograms of different Au substrates in 0.5 M H2SO4 (E = -0.2-1.35 V) at 50 mV s -1 .The geometrical area was 1 x 1 cm 2 in all cases.The areal charge associated with the cathodic peak at ca. -0.9 V, and the derived roughness factor, are provided in the table.

Figure S6 .
Figure S6.Kerr-gated Raman spectra of acetone and solutions of Nile red (NR) in acetone at different concentration (panels left) and ethanol and solutions of Nile blue (NB) in ethanol at different concentration (panels right).All performed using a cuvette, containing 500 L solution (0.2 cm optical path).

Figure S7 .
Figure S7.Different cell configurations used to acquire the Kerr-gated Raman spectra of a solution containing 30 nM Nile Red in acetone on Au foil substrate: SHINs on Au foil (a) and SHINs on cover-CaF2 window (b).

Figure S9 .
Figure S9.Comparison of additional broad features found in Kerr-gated Raman spectra with enhancement effects for Nile red (a) and Nile blue (b) solutions at different substrates compared with the Kerr-gated Raman spectra of the solid samples (black traces).