Figure 1. Schematics of the Raman scattering effect. The interaction of electron and incident photon (green color) with a vibrational mode; gaining or losing the energy of photon produces Stokes (red color) and anti-Stokes (blue color) scattering, respectively. Dotted lines show the vibrational sublevels of the virtual energy state.

Figure 2. Schematic of Raman scattering enhancement mechanisms: (a) RRS, (b) SRS, (c) CARS, and (d) SERS.

Figure 3. Resonance enhancement of the Raman signal. Dependence of the Raman signal of BBQ650-NHS (RRS probe) (A) on excitation wavelength and (B) on concentration. Reprinted from [23] under a CC-BY 4.0 license.

Figure 4. (A) Schematic diagram of the SPR effect of plasmonic nanostructures: (a) hotspot in plasmonic dimer, (b) plasmonic junction for single-molecule detection, (c) plasmonic tip, and (d) plasmonic nanopore. Reprinted from [53] under a CC-BY 4.0 license. (B) Use of SHINERS on a gold nanocrystal electrode surface [54]. (C) Quantitative bioassay on aptamer-functionalized nanopillars using SERS. Reprinted from [55] under a CC-BY 3.0 license. (D) Raman reporters can be encoded and used as fingerprints for multiplexed assays. Existence of these reporters is represented from a mode or set of modes from the SERS spectra of the analyte [56].

Figure 5. Different approaches and techniques for fabrication of SERS substrates. Top-down approaches. (A) Etching. Reprinted from [76] under a CC-BY 2.0 license. (B) Ball milling. Reprinted from [77] under a CC-BY 4.0 license. (C) Laser ablation. Adapted from [78]. (D) Electron beam lithography. Adapted from [79]. Bottom-up approaches. (E) Solgel process: recent applications in solgel synthesis. Reprinted from [80] under a CC-BY 3.0 license. (F) Chemical deposition. Reprinted from [81] under a CC-BY 3.0 license. (G) Solvothermal/hydrothermal method. Adapted from [82]. (H) Templating method. Reprinted from [83] under a CC-BY 3.0 license.

Figure 6. Different kinds of SERS substrates based on their properties and detection area. (A) Paper-based SERS substrate. Adapted from [100]. (B) TERS substrate. Adapted from [101]. (C) Self-cleaning-based substrate. Adapted from [102]. (D) Immunoassay-based substrate. Adapted from [103]. (E) Various shaped nanostructures. Reprinted from [104] under a CC-BY 4.0 license. (F) Dimers using DNA origami substrates. Reprinted from [105] under a CC-BY 3.0 license. (G) Core–shell-based substrates. Adapted from [106]. (H) Flexible tape-based substrate. Reprinted from [68] under a CC-BY 3.0 license. (I) Molecular imprinting-based approach. Adapted from [107].

Figure 7. TERS biosensing. (A) Schematic of TERS platform; the plasmonic field, generated by laser excitation in the TERS tip vicinity, produces SERS of a sample with nanoscale resolution. (B)–(D) Example of TERS application for DNA imaging and sensing. The approach curve in terms of Raman intensity as the tip approaches the DNA sample is shown in (B). The integrated Raman intensity versus the tip distance is shown in (C). The TERS spectrum is shown in red, and the spectrum from retracted tip is shown in black (D). (B)–(D) Reprinted from [124] under a CC-BY 4.0 license.

Figure 8. (A) Raman spectrometer from the end of the last century. (B) Current Raman microscope and (C) a handheld model of Raman spectrometer. Images copyright Thermo Fisher Scientific. Used with permission. (D) Raman probe for real-time surgical guidance [144].

Figure 9. (A) Schematic diagram of the optical setup of the counter-propagating dual-beam optical trap combined with Raman spectroscopy. Adapted from [152]. (B) High-speed CARS microscope for lipid research. Reprinted from [153] under a CC-BY 4.0 license. (C) Concept of SRS microscope optical train and intensity modulation transfer from the pump to the Stokes/probe beam. Adapted from [154]. (D) SORS concept to probe diffusely scattering samples at depth. Adapted from [155].

Figure 10. Overview of Raman-based biomolecular quantification at the organelle level. In initial steps, diverse subcellular structures are labeled by specific fluorescence probes and located by high-resolution microscopy. Then, Raman spectra are collected from the labeled organelles, and the BCA algorithm is applied to extract biomolecular profiles. In the last step, the biomolecular composition is analyzed; as an example, the chart shows levels of lipid unsaturation parameter (LSU) plotted against the absolute weight of lipids in apparatus endoplasmic reticulum of HeLa cells, as indicated. AG, apparatus Golgi; ER, endoplasmic reticulum; mito, mitochondrion; LSU, lipids unsaturation parameter. Reprinted with permission from Lita et al., Anal. Chem. 91, 11380-11387 (2019) [188]. Copyright 2019 American Chemical Society, https://doi.org/10.1021/acs.analchem.9b02663.

Figure 11. Quantification of major biomolecular classes in single organelles by Raman BCA. (A) Raman spectra were acquired in organelles located by fluorescent probes; the endoplasmic reticulum (ER) is shown as an example. The red dot inside a punctuated white outline indicates the spectral acquisition site. (B) A preprocessed Raman spectrum contains contributions from all biomolecules at the spectral acquisition site. (C) BCA algorithm selectively identifies the contributions of different biomolecules to the acquired spectrum in (B). (D) Concentrations of major biomolecule groups in ER were derived by BCA and arranged into an R dataset. The table lists concentrations of proteins (Prot), DNA, RNA, glycogen (Gly), lipids (Lip), saturated phospholipids (L-St), unsaturated phospholipids (L-USt), phosphatidylcholine sphingomyelin (PC), cholesterol and cholesteryl esters (CL), and triglycerides (TG) in mg/ml. In addition, the lipid unsaturation parameter (LSU), the average number of double bonds per unsaturated phospholipid (LSU-n), and trans/cis ratio (TCP) were determined. All resolvable molecular parameters are collectively referred to as the R dataset. (E) Single-cell R datasets are then compiled into a cross-omics R matrix characterizing a cellular population or a cell line. Adapted from [5].

Figure 12. Ramanomics with ML and AI prospects. Here, the Raman measurements are arranged in datasets and used for training of AI algorithms to generate a “black box” model that identifies, but does not explain, the measured patterns. These models are further used for designing new experiments and unbiased interpretation of findings.

Figure 13. (A) AFM images of amyloids of WT and L34T. (B) TERS spectra of WT and L34T and oG37C amyloid specimens. (C) Proportion of TERS spectra of spectral fingerprints of β-sheet secondary structures in the three amyloid forming peptides. WT, wild-type Aβ_1-42 peptide; L34T mutant forming fibrils that are relatively less toxic; oG37C mutant forming highly toxic oligomers. Adapted from [243].

Figure 14. The distribution of proteins, lipids, DNA and RNA in control and apoptotic HeLa cells visualized by multimodal CARS/two-photon excited fluorescence (TPEF) imaging. During the imaging live cells were maintained at the physiological conditions. Proteins and lipids were observed in the CARS mode at their characteristic vibrations of 2928 and 2890 cm^-1, respectively. Nucleic acids, stained by acridine orange, were acquired in the red (RNA) and green (DNA) fluorescence channels in the TPEF mode. In the right panels, schematics of the macromolecular organization of cells are represented. The CARS signal from proteins is represented in the left panels. The panels in the middle represent merged signals of the proteins (red), RNA (green), DNA (blue), and lipids (gray). The white-outlined areas in the protein channel are enlarged underneath. The upper row represents non-treated (control) cells. The signal from proteins is accumulated in the nucleolus (inset, arrowhead) and the nuclear lamina (arrows). In the rest of the nuclear volume, the intensity of the protein signal is nearly uniform. The lower row represents cells at the final stage of the apoptotic development. Proteins abandon the nucleolus and demonstrate a highly irregular distribution in the nucleoplasm; the genomic DNA is condensing to chromatin bodies and partially segregates from the proteins. Adapted from [187].

Figure 15. Raman spectroscopy for plant disease identification. Raman spectra of healthy maize kernels (green) and maize kernels infected by Aspergillus niger (brown), A. flavus (blue), Diplodia spp. (black), and Fusarium spp. (red). The Raman spectra of healthy maize kernels have characteristic modes from lignin, carbohydrates, proteins, and carotenoids. However, infection by Fusarium leads to the degradation of lignin and leads to the almost complete disappearance of the 1600 and 1633 cm^-1 modes of lignin (red). Infections by A. flavis and A. niger also results in the reduction in lignin peaks, but the extent is less compared with Fusarium infection. The presence of pathogens can directly be detected from the deposition of proteins. which can be observed by the presence of an amide I peak at 1658 cm^-1. The presence of pathogens is also associated by the degradation or fragmentation of carotenoids, which is characterized by changes in peak intensities in the region 1520–1550 cm^-1. Thus, Raman spectroscopy can be used to ascertain the maize kernel health and differentiate between healthy and diseased kernels. Adapted from [283].

Figure 16. (A) Schematic of the fluorescence−SERS dual-modal aptasensor for detection of hepatotoxin microcystin-LR and (B) (i),(ii) the near-field intensity and the electric field distance dependence profile from the gold nanostar SERS probes. Adapted from [288]. (C) Portable leaf-clip Raman sensor and its application in plants Adapted from Gupta et al., Sci. Rep. 10 (2020), https://doi.org/10.1038/s41598-020-76485-5.

Figure 17. (A) Experimental setup for super-resolution Raman imaging and schematic description of breaking the diffraction limit for Raman imaging where a picosecond pump and femtosecond probe pulse interact to create vibrational coherences. (B) Vibrational coherences in the ring region are destroyed by a spatially shaped decoherence beam to generate a resonant Raman signal. Adapted from [338]. (C) Energy diagram of a direct coupling of stimulated emission depletion (STED) and stimulated Raman excited fluorescence (SREF) and (D) the STED-SREF images of Rh800-stained E. coli cells. Reprinted from [339] under a CC-BY 4.0 license.

Figure 18. (A)–(E) SERS-based immunoassay using gold nanostars. The figures show the gold nanostar probe, the extinction spectra of the nanostar as compared with a gold nanosphere probe, the electron microscope image of the SERS probe, the SERS spectra of the probes, and the SERS immunoassay for multiplexed detection of circulating biomarkers, respectively. Adapted from [288]. (F) (a) Example of a portable Raman spectrometer for Raman immunoassay applications and (b) schematic of portable SERS-based lateral flow immunoassay reader. Reprinted from [341] under a CC-BY 4.0 license.