Unveiling the Unprecedented Optical Properties of Citrate-Stabilized Hollow AgAu Nanoshells Under Photothermal Irradiation

Abstract

Metallic nanoshells heat efficiently on excitation of the localized surface plasmon resonance (LSPR). Whilst investigating the photothermal properties of citrate-stabilized hollow gold nanoshells (HGNs) synthesized using a sacrificial silver nanoparticle (AgNPs), the LSPR undergoes a distinct blueshift (70 ± 20 nm (0.20 ± 0.06 eV)) when photothermally irradiated. Notably, when functionalized with a Raman reporter, the surface-enhanced Raman scattering (SERS) signal unexpectedly and dramatically increases 8 ± 2-fold upon plasmonic heating, despite the LSPR shifting away from the excitation wavelength. This unprecedented enhancement of the SERS signal is absent in samples lacking citrate or prepared using a cobalt nanoparticle template, underscoring the importance of citrate, heat, and AgNPs in eliciting these effects. It is hypothesized that aqueous silver ions near the surface of the HGNs react with the citrate and form a complex that is both light and temperature sensitive. The formation of silver deposits, observed by electron microscopy, alters the core-to-shell thickness ratio, resulting in a blueshift in the LSPR, and change the scattering to absorption properties, enabling an improved SERS performance. This new optical phenomenon has now been understood and will be of significant interest to future studies in harnessing the properties of HGNs.

Keywords: citrate; extinction spectroscopy; hollow gold nanoshells; photothermal irradiation; plasmonics; silver; surface‐enhanced Raman scattering.

Figure 1

A) and B) TEM images of HGNs prepared with a 3:6:1 ratio of AgNPs: ddH₂O: K/Au, equivalent to adding 1.00 mL of K/Au per 3.00 mL of AgNPs. C) Extinction spectra of HGNs prepared with different volumes of K/Au. D) Change in temperature profiles for those samples under 785 nm (280 mW) irradiation for 60 min.

Figure 2

A) and B) TEM images of irradiated HGNs prepared with a 3:6:1 ratio of AgNPs: ddH₂O: K/Au, equivalent to adding 1.00 mL of K/Au per 3.00 mL of AgNPs. C) Extinction spectra of irradiated HGNs prepared with different volumes of K/Au. D) Comparison of the LSPR shift before and after irradiation for 60 min (785 nm, 280 mW). SERS spectra before and after 60 min of irradiation for HGNs functionalized with E) malachite green isothiocyanate, F) 4‐mercaptopyridine, and G) 4‐bromothiophenol. H) Change in the normalized SERS intensity for the peaks indicated in E–G over 60 min of photothermal irradiation. SERS spectra were acquired under 785 nm (50 mW) excitation, 1 s acquisition, are an average of 10 accumulations, and baseline corrected.

Figure 3

Comparison of HGNs that were prepared without adding citrate after the synthesis with those that had citrate added. A) Bulk temperature profiles for photothermal heating for 60 min under 785 nm (280 mW) irradiation, with B) the corresponding extinction from before heating as the dashed spectrum and after heating the solid lines. C) SERS spectra from different batches of HGNs as (A) and (B) functionalized with MGITC after having been photothermally irradiated for 45 min. D) Temporal analysis of the SERS intensity of the 1170 cm⁻¹ vibrational mode of MGITC normalized to the intensity prior to irradiation. SERS spectra were acquired under 785 nm (50 mW) excitation, 1 s acquisition, are an average of 10 accumulations, and baseline corrected.

Figure 4

Comparison of normalized extinction spectra of A) citrate‐stabilized and B) citrate‐free HGNs heated with an oil bath at defined temperatures for 20 min. Each spectrum is the average of 3. Images of vials containing citrate‐free (left), and citrate‐stabilized (right) HGNs taken after C) 0 min, D) 5 min, and E) 10 min of heating in an oil bath at 70 °C. Comparison of the normalized extinction spectra of citrate‐stabilized HGNs before and after photothermal heating and heating at 70 °C in the oil bath for 20 min for HGNs prepared with volume ratios of F) 3:6:1, G) 3:5.5:1.5, and H) 3:5:2 for AgNPs: ddH₂O: K/Au. I) Temperature heating profiles for citrate‐stabilized HGNs that underwent 45 min of continuous 785 nm (280 mW) irradiation or “blinking” whereby the shutter was opened and then closed repeatedly for 30 and 90 s durations respectively. J) Changes in the SERS intensities for MGITC functionalized HGNs after continuous or “blinking” exposure. K) Variation in the intensity of the 1174 cm⁻¹ vibrational mode of MGITC over the duration of photothermal irradiation normalized to the intensity prior to irradiation. SERS spectra were acquired under 785 nm (50 mW) excitation, 1 s acquisition, are an average of 10 accumulations, and baseline corrected. The spectra in (J) were offset for clarity.

Figure 5

Photothermal, optical, and SERS properties of HGN_Cos prepared with different concentrations of HAuCl₄. A) Temperature curves for samples at an optical density of ≈1 and photothermally irradiated for 1 h. Normalized extinction spectra before and after one hour of photothermal irradiation (785 nm, 280 mW) for samples prepared with 50 mL of B) 992, C) 940, and D) 868 μM HAuCl₄. SERS spectra of HGN_Cos functionalized with MGITC before and after one hour of photothermal irradiation for samples prepared with 50 mL of E) 992, F) 940, and G) 868 μM HAuCl₄·3H₂O. SERS spectra were acquired under 785 nm (50 mW) excitation, 1 s acquisition, are an average of 10 accumulations, and baseline corrected.

Figure 6

HAADF images of an HGN taken under constant exposure A) initial and B) 50 s later. C) EDX image taken at the end of the measurement. The purple arrows indicate a location where a silver deposit formed during the measurement. Ratios of the calculated scattering cross‐section to absorption cross‐section for the different HGNs using a Mie theory calculator for the formation of a D) interior and E) exterior silver shell.