Trends in Application of SERS Substrates beyond Ag and Au, and Their Role in Bioanalysis

Abstract

This article compares the applications of traditional gold and silver-based SERS substrates and less conventional (Pd/Pt, Cu, Al, Si-based) SERS substrates, focusing on sensing, biosensing, and clinical analysis. In recent decades plethora of new biosensing and clinical SERS applications have fueled the search for more cost-effective, scalable, and stable substrates since traditional gold and silver-based substrates are quite expensive, prone to corrosion, contamination and non-specific binding, particularly by S-containing compounds. Following that, we briefly described our experimental experience with Si and Al-based SERS substrates and systematically analyzed the literature on SERS on substrate materials such as Pd/Pt, Cu, Al, and Si. We tabulated and discussed figures of merit such as enhancement factor (EF) and limit of detection (LOD) from analytical applications of these substrates. The results of the comparison showed that Pd/Pt substrates are not practical due to their high cost; Cu-based substrates are less stable and produce lower signal enhancement. Si and Al-based substrates showed promising results, particularly in combination with gold and silver nanostructures since they could produce comparable EFs and LODs as conventional substrates. In addition, their stability and relatively low cost make them viable alternatives for gold and silver-based substrates. Finally, this review highlighted and compared the clinical performance of non-traditional SERS substrates and traditional gold and silver SERS substrates. We discovered that if we take the average sensitivity, specificity, and accuracy of clinical SERS assays reported in the literature, those parameters, particularly accuracy (93-94%), are similar for SERS bioassays on AgNP@Al, Si-based, Au-based, and Ag-based substrates. We hope that this review will encourage research into SERS biosensing on aluminum, silicon, and some other substrates. These Al and Si based substrates may respond efficiently to the major challenges to the SERS practical application. For instance, they may be not only less expensive, e.g., Al foil, but also in some cases more selective and sometimes more reproducible, when compared to gold-only or silver-only based SERS substrates. Overall, it may result in a greater diversity of applicable SERS substrates, allowing for better optimization and selection of the SERS substrate for a specific sensing/biosensing or clinical application.

Keywords: LOD; SERS; aluminum; clinical applications; clinical sensitivity; detection of biomarkers; immunoassays; silicon.

Figure 1

(1) SERS spectrum (A) and below calibration plot with the pictures of the substrates (B) for the sandwich immunoassay of human immunoglobulin (hIgG) on Si substrates obtained by using a 633 nm laser. NBARI—normalized blank adjusted Raman intensity from 3 different sets of measurements carried out on different weeks. (2) Calibration curves obtained by 4 parameters logistic non-linear regression analysis for Au and Si substrates on the first week. (3) Representative AFM maps of increasing hIgG concentration: 0 (blank), 0.3 nM, 1 nM for A, B, C—silicon; D, E, F—gold, respectively. Reproduced from Kunushpayeva et al. under the Creative Commons CC BY license.

Figure 2

Scheme of specific (1. ERL or extrinsic Raman label) and non-specific (2. ERL) binding of Extrinsic Raman Labels (ERLs) to the capture antibody covered gold film substrate for SERS sandwich immunoassay. The detected cancer marker is an antigen. Non-specific binding to Si and even Al must be much weaker that non-specific binding to gold.

Figure 3

(A) LSPR spectra of nanoparticle arrays with identical geometries but varying metal (Al, Ag, Cu, or Au). Reprinted with permission from Chan et al. Copyright 2008 American Chemical Society. (B,C) Illustration of different surface binding mechanism of ssDNA molecules onto Au and Al substrates (D) SERS spectra of APhS and (E) PABA on Al (red) and Au (blue) substrates with normal Raman spectra (black) as reference. Reprinted with permission from Tian et al. Copyright 2017 American Chemical Society.

Figure 4

(A) Raman spectra of melamine (0.078–80 ppm concentration range) on commercial gold nanoparticles on Al foil (AuNPs@AlF) measured at 785 nm excitation wavelength (B) Calibration plot of 820 cm−¹ characteristic vibration peak area vs. logarithm of melamine concentration in ppb on gold film and Al foil. Reprinted from Gudun et al. under Creative Commons Attribution License. (C) AFM map of 10 × 10 micron of 80 nm commercial AuNPs@AlF substrate (D) The plot of blank adjusted Raman intensity vs. logarithm of urea concentration (log[urea]) on three different substrates: AuNPs@Au film, AuNPs@AlF, AuNPs@glass. Reprinted from Mukanova et al. Copyright © 2018 The Japan Society for Analytical Chemistry.

Figure 5

(A) HRTEM image of the prepared SiO₂ Nanoparticles, (B) The preparation of the dual-layers DTNB-modified SiO₂@Ag NPs. (C) The modification of the SiO₂@Ag SERS tags with SARS-CoV-2 S protein. (D) Operating principle of the SERS-LFIA strip for the detection of the SARS-CoV-2 IgM/IgG. (E) Enlarged HRTEM image of SiO₂@Au-seed NPs. (F) Raman spectra measured in the corresponding test lines and the enlarged viewport at the 1328 cm−¹ characteristic peak. (G) Element mapping results of the SiO₂@Ag NPs. (H) The images of the SERS-LFIA strips with a single Test line after application of the different S protein antibody concentrations (10–0.001 ng/mL). (I) The calibration curve of SiO₂@Ag SERS-based LFIA for the S protein antibody. Reproduced with permission Liu et al. Copyright ©2021 Elsevier.

Figure 6

(A) Raman spectra of Ag NRs wrapped with Al₂O₃ layer, serum, and serum with the substrate. (B) Principal component score plots of the SERS spectrum based on OPLS-DA for LAC and normal samples. (C) ROC curve of LAC samples. Reproduced with permission from Liu et al. Copyright ©2019 Elsevier.