Self-Assembled Gold Nanoparticles as Reusable SERS Substrates for Polyphenolic Compound Detection

Abstract

Natural polyphenolic compounds play a pivotal role in biological processes and exhibit notable antioxidant activity. Among these compounds, chlorogenic acid stands out as one of the most widespread and important polyphenols. The accurate detection of chlorogenic acid is crucial for ensuring the quality and classification of the raw materials used in its extraction, as well as the final products in the food, pharmaceutical, and cosmetics industries that contain this bioactive compound. Raman spectroscopy emerges as a powerful analytical tool, particularly in field applications, due to its versatility and sensitivity, offering both qualitative and quantitative analyses. By using the self-assembly of gold nanoparticles at liquid-liquid interfaces and the developed "aqua-print" process, we propose a facile and inexpensive route to fabricate enhanced substrates for surface-enhanced Raman spectroscopy with high reproducibility. To ensure substrate reliability and accurate molecule detection in SERS experiments, a benchmarking procedure was developed. This process involved the use of non-resonant rhodamine 6G dye in the absence of charge transfer and was applied to all synthesized nanoparticles and fabricated substrates. The latter revealed the highest enhancement factor of 4 × 10⁴ for 72 nm gold nanoparticles among nanoparticle diameters ranging from 14 to 99 nm. Furthermore, the enhanced substrate was implemented in the detection of chlorogenic acid with a concentration range from 10 μM to 350 μM, demonstrating high accuracy (R² > 99%). Raman mapping was employed to validate the good uniformity of the signal (the standard deviation was below 15%). The findings of this study were also supported by DFT calculations of the theoretical Raman spectra, demonstrating the formation of the chlorogenic acid dimer. The proposed method is strategically important for the development of the class of in-field methods to detect polyphenolic compounds in raw materials such as plants, extracted plant proteins, and polyphenolic compounds.

Keywords: SERS; chlorogenic acid; density functional theory; gold nanoparticles; liquid–liquid interface; self-assembly.

Figure 1

Characterization of synthesized gold nanoparticles. (A) UV-Vis absorbance spectra, (B) particle size distributions and ζ-potentials, and (C) TEM images of synthesized AuNPs of maximal and minimal diameters. Insets: Bar diagrams of particle size distributions based on TEM image data processing (the red curve is the Gaussian fit). The labeling of the samples should be read as the following: Frens = nanoparticles synthesized by the Frens–Turkevich method using one reductant agent: Na₃Citr = sodium citrate, KAsc = potassium ascorbate, or HAsc = ascorbic acid; SMG = nanoparticles synthesized by Park’s method.

Figure 2

Morphological characterization of the fabricated enhanced substrates by SEM (A,D), TEM (B,E), and AFM (C,F). Silicon substrates coated with 14 nm AuNPs (A–C) and 72 nm nanoparticles (D–F).

Figure 3

Benchmarking the enhanced substrate made of 14 nm AuNPs with the non-resonant reporter molecule R6G at varying concentrations (the grey area marks out the peak used for plotting panel B). (A) Recorded surface-enhanced Raman spectra of R6G. (B) Calibration curve demonstrating the linear dependence of the Raman scattering of R6G at a selected wavenumber of 1360 cm⁻¹. (C,D) Mapping of the enhanced substrate at an R6G concentration of 150 μM: (C) peak position map and (D) peak intensity map at 1360 cm⁻¹.

Figure 4

Benchmarking the enhanced substrate made of 72 nm AuNPs with the non-resonant reporter molecule R6G at varying concentrations. (A) Recorded surface-enhanced Raman spectra of R6G (the grey area marks out the peak used for plotting panel B). (B) Calibration curve demonstrating the linear dependence of Raman scattering of R6G at a selected wavenumber of 1360 cm⁻¹. (C,D) Mapping of the enhanced substrate at an R6G concentration of 150 μM: (C) peak position map and (D) peak intensity map at 1360 cm⁻¹.

Figure 5

Dependence of the analytical enhancement factor on the diameter of AuNPs synthesized by different methods. A Gaussian curve was used as the trend line.

Figure 6

Detection of CGA dimer on the enhanced substrate made of 72 nm AuNPs by Raman spectroscopy. (A) Experimental spectra on SERS substrates for different concentrations of CGA dimer (the grey area marks out the peak used for plotting panel C). (B) Comparison of the DFT-calculated Raman spectrum with the experimental one. (C) Calibration curve demonstrating the linear dependence of Raman scattering of the CGA dimer at 1615 cm⁻¹. (D,E) Mapping of the enhanced substrate at a CGA concentration of 50 μM: (C) peak position map and (D) peak intensity map at 1615 cm⁻¹.

Scheme 1

Schematic of AuNP synthesis and SERS substrate preparation procedure. (A) Synthesis of AuNPs with different methods: Frens–Turkevich with trisodium citrate, ascorbic acid, or potassium ascorbate; seed-mediated growth from smallest seed particle obtained by Ferns–Turkevich method. (B) Enhanced substrate fabrication: (i) self-assembly of AuNPs into nanofilm using liquid–liquid interface (MELLDs), (ii) transfer of nanofilm to solid substrate (silicon) with aqua-print technology. Photos demonstrate visual appearance of MELLD and nanofilm on substrate.