Probing Oxidation Mechanisms in Plasmonic Catalysis: Unraveling the Role of Reactive Oxygen Species.

Plasmon-induced oxidation has conventionally been attributed to the transfer of plasmonic hot holes. However, this theoretical framework encounters challenges in elucidating the latest experimental findings, such as enhanced catalytic efficiency under uncoupled irradiation conditions and superior oxidizability of silver nanoparticles. Herein, we employ liquid surface-enhanced Raman spectroscopy (SERS) as a real-time and in situ tool to explore the oxidation mechanisms in plasmonic catalysis, taking the decarboxylation of p-mercaptobenzoic acid (PMBA) as a case study. Our findings suggest that the plasmon-induced oxidation is driven by reactive oxygen species (ROS) rather than hot holes, holding true for both the Au and Ag nanoparticles. Subsequent investigations suggest that plasmon-induced ROS may arise from hot carriers or energy transfer mechanisms, exhibiting selectivity under different experimental conditions. The observations were substantiated by investigating the cleavage of the carbon-boron bonds. Furthermore, the underlying mechanisms were clarified by energy level theories, advancing our understanding of plasmonic catalysis.

Detection of Helicobacter pylori Infection in Human Gastric Fluid Through Surface-Enhanced Raman Spectroscopy Coupled With Machine Learning Algorithms.

Diagnostic methods for Helicobacter pylori infection include, but are not limited to, urea breath test, serum antibody test, fecal antigen test, and rapid urease test. However, these methods suffer drawbacks such as low accuracy, high false-positive rate, complex operations, invasiveness, etc. Therefore, there is a need to develop simple, rapid, and noninvasive detection methods for H. pylori diagnosis. In this study, we propose a novel technique for accurately detecting H. pylori infection through machine learning analysis of surface-enhanced Raman scattering (SERS) spectra of gastric fluid samples that were noninvasively collected from human stomachs via the string test. One hundred participants were recruited to collect gastric fluid samples noninvasively. Therefore, 12,000 SERS spectra (n = 120 spectra/participant) were generated for building machine learning models evaluated by standard metrics in model performance assessment. According to the results, the Light Gradient Boosting Machine algorithm exhibited the best prediction capacity and time efficiency (accuracy = 99.54% and time = 2.61 seconds). Moreover, the Light Gradient Boosting Machine model was blindly tested on 2,000 SERS spectra collected from 100 participants with unknown H. pylori infection status, achieving a prediction accuracy of 82.15% compared with qPCR results. This novel technique is simple and rapid in diagnosing H. pylori infection, potentially complementing current H. pylori diagnostic methods.

Charged Amino Acid Substitutions Affect Conformation of Neuroglobin and Cytochrome c Heme Groups.

Neuroglobin (Ngb) is a cytosolic heme protein that plays an important role in protecting cells from apoptosis through interaction with oxidized cytochrome c (Cyt c) released from mitochondria. The interaction of reduced Ngb and oxidized Cyt c is accompanied by electron transfer between them and the reduction in Cyt c. Despite the growing number of studies on Ngb, the mechanism of interaction between Ngb and Cyt c is still unclear. Using Raman spectroscopy, we studied the effect of charged amino acid substitutions in Ngb and Cyt c on the conformation of their hemes. It has been shown that Ngb mutants E60K, K67E, K95E and E60K/E87K demonstrate changed heme conformations with the lower probability of the heme planar conformation compared to wild-type Ngb. Moreover, oxidized Cyt c mutants K25E, K72E and K25E/K72E demonstrate the decrease in the probability of methyl-radicals vibrations, indicating the higher rigidity of the protein microenvironment. It is possible that these changes can affect electron transfer between Ngb and Cyt c.

Unusual Raman Enhancement Effect of Ultrathin Copper Sulfide.

In surface-enhanced Raman spectroscopy (SERS), 2D materials are explored as substrates owing to their chemical stability and reproducibility. However, they exhibit lower enhancement factors (EFs) compared to noble metal-based SERS substrates. This study demonstrates the application of ultrathin covellite copper sulfide (CuS) as a cost-effective SERS substrate with a high EF value of 7.2 × 104 . The CuS substrate is readily synthesized by sulfurizing a Cu thin film at room temperature, exhibiting a Raman signal enhancement comparable to that of an Au noble metal substrate of similar thickness. Furthermore, computational simulations using the density functional theory are employed and time-resolved photoluminescence measurements are performed to investigate the enhancement mechanisms. The results indicate that polar covalent bonds (Cu─S) and strong interlayer interactions in the ultrathin CuS substrate increase the probability of charge transfer between the analyte molecules and the CuS surface, thereby producing enhanced SERS signals. The CuS SERS substrate demonstrates the selective detection of various dye molecules, including rhodamine 6G, methylene blue, and safranine O. Furthermore, the simplicity of CuS synthesis facilitates large-scale production of SERS substrates with high spatial uniformity, exhibiting a signal variation of less than 5% on a 4-inch wafer.

Plasmonic Dodecahedral-Walled Elongated Nanoframes for Surface-Enhanced Raman Spectroscopy.

Here, elongated pseudohollow nanoframes composed of four rectangular plates enclosing the sides and two open-frame ends with four ridges pointing at the tips for near-field focusing are reported. The side facets act as light-collecting domains and transfer the collected light to the sharp tips for near-field focusing. The nanoframes are hollow inside, allowing the gaseous analyte to penetrate through the entire architecture and enabling efficient detection of gaseous analytes when combined with Raman spectroscopy. The resulting nanostructures are named Au dodecahedral-walled nanoframes. Synthesis of the nanoframes involves shape transformation of Au nanorods with round tips to produce Au-elongated dodecahedra, followed by facet-selective Pt growth, etching of the inner Au, and regrowth steps. The close-packed assembly of Au dodecahedral-walled nanoframes exhibits an attomolar limit of detection toward benzenethiol. This significant enhancement in SERS is attributed to the presence of a flat solid terrace for a large surface area, sharp edges and vertices for strong electromagnetic near-field collection, and open frames for effective analyte transport and capture. Moreover, nanoframes are applied to detect chemical warfare agents, specifically mustard gas simulants, and 20 times higher sensitivity is achieved compared to their solid counterparts.

2D TiS2-Nanosheet-Coated Concave Gold Arrays with Triple-Coupled Resonances as Sensitive SERS Substrates.

Herein, a hybrid substrate for surface-enhanced Raman scattering (SERS) is fabricated, which couples localized surface plasmon resonance (LSPR), charge transfer (CT) resonance, and molecular resonance. Exfoliated 2D TiS2 nanosheets with semimetallic properties accelerate the CT with the tested analytes, inducing a remarkable chemical mechanism enhancement. In addition, the LSPR effect is coupled with a concave gold array located underneath the thin TiS2 nanosheet, providing a strong electromagnetic enhancement. The concave gold array is prepared by etching silicone nanospheres assembled on larger polystyrene nanospheres, followed by depositing a gold layer. The LSPR intensity near the gold layer can be adjusted by changing the layer thickness to couple the molecular and CT resonances, in order to maximize the SERS enhancement. The best SERS performance is recorded on TiS2-nanosheet-coated plasmonic substrates, with a detectable methylene blue concentration down to 10-13 m and an enhancement factor of 2.1 × 109 and this concentration is several orders of magnitude lower than that of the TiS2 nanosheet (10-11 m) and plasmonic substrates (10-9 m). The present hybrid substrate with triple-coupled resonance further shows significant advantages in the label-free monitoring of curcumin (a widely applied drug for treating multiple cancers and inflammations) in serum and urine.

Multi-Functional Applications of H-Glass Embedded with Stable Plasmonic Gold Nanoislands.

Metal nanoparticles (MNPs) are synthesized using various techniques on diverse substrates that significantly impact their properties. However, among the substrate materials investigated, the major challenge is the stability of MNPs due to their poor adhesion to the substrate. Herein, it is demonstrated how a newly developed H-glass can concurrently stabilize plasmonic gold nanoislands (GNIs) and offer multifunctional applications. The GNIs on the H-glass are synthesized using a simple yet, robust thermal dewetting process. The H-glass embedded with GNIs demonstrates versatility in its applications, such as i) acting as a room temperature chemiresistive gas sensor (70% response for NO2 gas); ii) serving as substrates for surface-enhanced Raman spectroscopy for the identifications of Nile blue (dye) and picric acid (explosive) analytes down to nanomolar concentrations with enhancement factors of 4.8 × 106 and 6.1 × 105 , respectively; and iii) functioning as a nonlinear optical saturable absorber with a saturation intensity of 18.36 × 1015 W m-2 at 600 nm, and the performance characteristics are on par with those of materials reported in the existing literature. This work establishes a facile strategy to develop advanced materials by depositing metal nanoislands on glass for various functional applications.

Evolutionary Optimized, Monocrystalline Gold Double Wire Gratings as a Novel SERS Sensing Platform.

Achieving reliable and quantifiable performance in large-area surface-enhanced Raman spectroscopy (SERS) substrates poses a formidable challenge, demanding signal enhancement while ensuring response uniformity and reproducibility. Conventional SERS substrates often made of inhomogeneous materials with random resonator geometries, resulting in multiple or broadened plasmonic resonances, undesired absorptive losses, and uneven field enhancement. These limitations hamper reproducibility, making it difficult to conduct comparative studies with high sensitivity. This study introduces an innovative approach that addresses these challenges by utilizing monocrystalline gold flakes to fabricate well-defined plasmonic double-wire resonators through focused ion-beam lithography. Inspired by biological strategy, the double-wire grating substrate (DWGS) geometry is evolutionarily optimized to maximize the SERS signal by enhancing both excitation and emission processes. The use of monocrystalline material minimizes absorption losses and ensures shape fidelity during nanofabrication. DWGS demonstrates notable reproducibility (RSD = 6.6%), repeatability (RSD = 5.6%), and large-area homogeneity > 104 µm2. It provides a SERS enhancement for sub-monolayer coverage detection of 4-Aminothiophenol analyte. Furthermore, DWGS demonstrates reusability, long-term stability on the shelf, and sustained analyte signal stability over time. Validation with diverse analytes, across different states of matter, including biological macromolecules, confirms the sensitive and reproducible nature of DWGSs, thereby establishing them as a promising platform for future sensing applications.

Controlling Nanoparticle Distance by On-Surface DNA-Origami Folding.

DNA origami is a flexible platform for the precise organization of nano-objects, enabling numerous applications from biomedicine to nano-photonics. Its huge potential stems from its high flexibility that allows customized structures to meet specific requirements. The ability to generate diverse final structures from a common base by folding significantly enhances design variety and is regularly occurring in liquid. This study describes a novel approach that combines top-down lithography with bottom-up DNA origami techniques to control folding of the DNA origami with the adsorption on pre-patterned surfaces. Using this approach, tunable plasmonic dimer nano-arrays are fabricated on a silicon surface. This involves employing electron beam lithography to create adsorption sites on the surface and utilizing self-organized adsorption of DNA origami functionalized with two gold nanoparticles (AuNPs). The desired folding of the DNA origami helices can be controlled by the size and shape of the adsorption sites. This approach can for example be used to tune the center-to-center distance of the AuNPs dimers on the origami template. To demonstrate this technique's efficiency, the Raman signal of dye molecules (carboxy tetramethylrhodamine, TAMRA) coated on the AuNPs surface are investigated. These findings highlight the potential of tunable DNA origami-based plasmonic nanostructures for many applications.

Unmasking the Electrochemical Stability of N-Heterocyclic Carbene Monolayers on Gold.

N-heterocyclic carbene (NHC) monolayers are transforming electrocatalysis and biosensor design via their increased performance and stability. Despite their increasing use in electrochemical systems, the integrity of the NHC monolayer during voltage perturbations remains largely unknown. Herein, we deploy surface-enhanced Raman spectroscopy (SERS) to measure the stability of two model NHCs on gold in ambient conditions as a function of applied potential and under continuous voltammetric interrogation. Our results illustrate that NHC monolayers exhibit electrochemical stability over a wide voltage window (-1 V to 0.5 V vs Ag|AgCl), but they are found to degrade at strongly reducing (< -1 V) or oxidizing (>0.5 V) potentials. We also address NHC monolayer stability under continuous voltammetric interrogation between 0.2 V and -0.5 V, a commonly used voltage window for sensing, showing they are stable for up to 43 hours. However, we additionally find that modifications of the backbone NHC structure can lead to significantly shorter operational lifetimes. While these results highlight the potential of NHC architectures for electrode functionalization, they also reveal potential pitfalls that have not been fully appreciated in electrochemical applications of NHCs.

Interfacial Bonding Induced Charge Transfer in Two-Dimensional Amorphous MoO3-x/Graphdiyne Oxide Non-Van der Waals Heterostructures for Dominant SERS Enhancement.

Two-dimensional semiconductor-based nanomaterials have shown to be an effective substrate for Surface-enhanced Raman Scattering (SERS) spectroscopy. However, the enhancement factor (EF) tends to be relatively weak compared to that of noble metals and does not allow for trace detection of molecules. In this work, we report the successful preparation of two-dimensional (2D) amorphous non-van der Waals heterostructures MoO3-x/GDYO nanomaterials using supercritical CO2. Due to the synergistic effect of the localized surface plasmon resonance (LSPR) effect and the charge transfer effect, it exhibits excellent SERS performance in the detection of methylene blue (MB) molecules, with a detection limit as low as 10-14 M while the enhancement factor (EF) can reach an impressive 2.55×1011. More importantly, the chemical bond bridging at the MoO3-x/GDYO heterostructures interface can accelerate the electron transfer between the interfaces, and the large number of defective surface structures on the heterostructures surface facilitates the chemisorption of MB molecules. And the charge recombination lifetime can be proved by a ~1.7-fold increase during their interfacial electron-transfer process for MoO3-x/GDYO@MB mixture, achieving highly sensitive SERS detection.

An Exploration of Cysteamine as a Subphase Additive for the Fabrication of Uniform Gold Nanorod Arrays using Langmuir-Blodgett Deposition.

Gold nanorods (AuNRs) have attracted significant attention over the past several decades for a variety of applications and there has been steady progress with regards to their synthesis and modification. Despite these advances, the assembly of AuNRs into well-organized hierarchical assemblies remains a formidable challenge. Specifically, there is a need for tools that can fabricate assemblies of nanorods over large length scales at low cost with the potential for high-throughput manufacturing. Langmuir-Blodgettry is a monolayer deposition technique which has been primarily applied to amphiphilic molecules, but which has recently shown promise for the ordering of functionalized nanoparticles residing at the air-water interface. In this work, Langmuir-Blodgett deposition is explored for the formation of AuNR arrays for enhanced surface-enhanced Raman spectroscopy (SERS) sensing. In particular, both surface modification of the AuNRs as well as subphase modification with cysteamine were evaluated for AuNR array fabrication.

Structure and biological activity in vitro of Flagellin and its mutants from Escherichia coli Nissle 1917.

Bacterial flagellin is a potent immunomodulatory agent. Previously, we successfully obtained flagellin from Escherichia coli Nissle 1917 (FliCEcN) and constructed two mutants with varying degrees of deletion in its highly variable regions (HVRs). We found that there was a difference in immune stimulation levels between the two mutants, with the mutant lacking the D2-D3 domain pair of FliCEcN having a better adjuvant effect. Therefore, this study further analyzed the structural characteristics of the aforementioned FliCEcN and its two mutants and measured their levels of Caco-2 cell stimulation to explore the impact of different domains in the HVRs of FliCEcN on its structure and immune efficacy. This study utilized AlphaFold2, SERS (Surface-enhanced Raman spectroscopy), and CD (circular dichroism) techniques to analyze the structural characteristics of FliCEcN and its mutants, FliCΔ174-506 and FliCΔ274-406, and tested their immune effects by stimulating Caco-2 cells in vitro. The results indicate that the D2 and D3 domains of FliCEcN have more complex interactions compared to the D1-D2 domain pair., and these domains also play a role in molecular docking with TLR5 (Toll-like receptor 5). Furthermore, FliCΔ274-406 has more missing side chain and characteristic amino acid peaks than FliCΔ174-506. The FliCEcN group was found to stimulate higher levels of IL-10 (interleukin 10) secretion, while the FliCΔ174-506 and FliCΔ274-406 groups had higher levels of IL-6 (interleukin 6) and TNF-α (tumor necrosis factor-α) secretion. In summary, the deletion of different domains in the HVRs of FliCEcN affects its structural characteristics, its interaction with TLR5, and the secretion of immune factors by Caco-2 cells.

Determination of Trace Organic Contaminant Concentration via Machine Classification of Surface-Enhanced Raman Spectra.

Surface-enhanced Raman spectroscopy (SERS) has been well explored as a highly effective characterization technique that is capable of chemical pollutant detection and identification at very low concentrations. Machine learning has been previously used to identify compounds based on SERS spectral data. However, utilization of SERS to quantify concentrations, with or without machine learning, has been difficult due to the spectral intensity being sensitive to confounding factors such as the substrate parameters, orientation of the analyte, and sample preparation technique. Here, we demonstrate an approach for predicting the concentration of sample pollutants from SERS spectra using machine learning. Frequency domain transform methods, including the Fourier and Walsh-Hadamard transforms, are applied to spectral data sets of three analytes (rhodamine 6G, chlorpyrifos, and triclosan), which are then used to train machine learning algorithms. Using standard machine learning models, the concentration of the sample pollutants is predicted with >80% cross-validation accuracy from raw SERS data. A cross-validation accuracy of 85% was achieved using deep learning for a moderately sized data set (∼100 spectra), and 70-80% was achieved for small data sets (∼50 spectra). Performance can be maintained within this range even when combining various sample preparation techniques and environmental media interference. Additionally, as a spectral pretreatment, the Fourier and Hadamard transforms are shown to consistently improve prediction accuracy across multiple data sets. Finally, standard models were shown to accurately identify characteristic peaks of compounds via analysis of their importance scores, further verifying their predictive value.

Direct Observation of Particle-To-Particle Variability in Ambient Aerosol pH Using a Novel Analytical Approach Based on Surface-Enhanced Raman Spectroscopy.

The pH of atmospheric aerosols is a key characteristic that profoundly influences their impacts on climate change, human health, and ecosystems. Despite widely performed aerosol pH research, determining the pH levels of individual atmospheric aerosol particles has been a challenge. This study presents a novel analytical technique that utilizes surface-enhanced Raman spectroscopy to assess the pH of individual ambient PM2.5-10 aerosol particles in conjunction with examining their hygroscopic behavior, morphology, and elemental compositions. The results revealed a substantial pH variation among simultaneously collected aerosol particles, ranging from 3.3 to 5.7. This variability is likely related to each particle's unique reaction and aging states. The extensive particle-to-particle pH variability suggests that atmospheric aerosols present at the same time and location can exhibit diverse reactivities, reaction pathways, phase equilibria, and phase separation properties. This pioneering study paves the way for in-depth investigations into particle-to-particle variability, size dependency, and detailed spatial and temporal variations of aerosol pH, thus deepening our understanding of atmospheric chemistry and its environmental implications.

Digital Surface-Enhanced Raman Spectroscopy-Lateral Flow Test Dipstick: Ultrasensitive, Rapid Virus Quantification in Environmental Dust.

This study introduces a novel surface-enhanced Raman spectroscopy (SERS)-based lateral flow test (LFT) dipstick that integrates digital analysis for highly sensitive and rapid viral quantification. The SERS-LFT dipsticks, incorporating gold-silver core-shell nanoparticle probes, enable pixel-based digital analysis of large-area SERS scans. Such an approach enables ultralow-level detection of viruses that readily distinguishes positive signals from background noise at the pixel level. The developed digital SERS-LFTs demonstrate limits of detection (LODs) of 180 fg for SARS-CoV-2 spike protein, 120 fg for nucleocapsid protein, and 7 plaque forming units for intact virus, all within <30 min. Importantly, digital SERS-LFT methods maintain their robustness and their LODs in the presence of indoor dust, thus underscoring their potential for accurate and reliable virus diagnosis and quantification in real-world environmental settings.

Analysis and experimental assessment of an optimized SERS substrate used to detect thiabendazole in apples with high sensitivity.

Pesticides that linger in the environment and ecosystems for an extended period can cause severe and dangerous health problems in humans. To detect pesticides in foods, the development of high-sensitivity and quick screening technologies was required. This research investigated the performance of Au@Ag NPs with varying thicknesses of the silver shell for detecting trace quantities of thiabendazole (TBZ) in apples using surface-enhanced Raman spectroscopy (SERS). The Au@Ag NPs were synthesized by coating 32 nm gold seeds with different thicknesses of silver shell ranging from 2.4 to 8.7 nm, achieved by adjusting the incorporation of AgNO3 and ascorbic acid. The optimized Au@Ag NPs with a 7.3 nm silver shell demonstrated outstanding SERS activity, high sensitivity, and a detection limit of 0.05 µg/mL for TBZ. The R2 values, representing the goodness of fit, were found to be 0.990 and 0.986 for standard and real TBZ samples, respectively, indicating a strong correlation between the measured signal and the TBZ concentration. The recovery analysis showed a reliable and accurate detection capability (96 to 105%), suggesting good reliability and accuracy of the SERS-based detection using the optimal Au@Ag NPs. Overall, this research highlights the potential of SERS with optimal Au@Ag NPs for rapid and effective monitoring of pesticides in the food industry.

Simple and sensitive galactose monitoring based on capillary SERS sensor.

Galactosemia, a severe genetic metabolic disorder, results from the absence of galactose-degrading enzymes, leading to harmful galactose accumulation. In this study, we introduce a novel capillary-based surface-enhanced Raman spectroscopy (SERS) sensor for convenient and sensitive galactose detection. The developed sensor enhances SERS signals by introducing gold nanoparticles (Au NPs) onto the surface of silver nanoshells (Ag NSs) within a capillary, creating Ag NSs with Au NPs as satellites. Utilizing 4-mercaptophenylboronic acid (4-MPBA) as a Raman reporter molecule, the detection method relies on the conversion of 4-MPBA to 4-mercaptophenol (4-MPhOH) driven by hydrogen peroxide (H2O2) generated during galactose oxidation by galactose oxidase (GOx). A new SERS signal was observed, which was generated by H2O2 produced when galactose and GOx reacted. Our strategy yielded a quantitative change in the SERS signal, specifically in the band intensity ratio of 998 to 1076 cm-1 (I998/I1076) as the galactose concentration increased. Our capillary-based SERS biosensor provides a promising platform for early galactosemia diagnosis.

Detection of organic dyes using Ag NPAs/SMP SERS substrate produced via sandpaper template-assisted lithography and liquid-liquid interface self-assembly.

Surface-enhanced Raman spectroscopy (SERS) is a highly sensitive and reliable fingerprinting technique. However, its analytical capability is closely related to the quality of a SERS substrate used for the analysis. In particular, conventional colloidal substrates possess disadvantages in terms of controllability, stability, and reproducibility, which limit their application. In order to address these issues, a simple, cost-effective, and efficient SERS substrate based on silver nanoparticle arrays (Ag NPAs) and sandpaper-molded polydimethylsiloxane (SMP) was proposed in this work. Successfully prepared via template lithography and liquid-liquid interface self-assembly (LLISA), the substrate can be applied to the specific detection of organic dyes in the environment. The substrate exhibited good SERS performance, and the limit of detection (LOD) of rhodamine 6G (R6G) was shown to be 10-7 M under the optimal conditions (1000 grit sandpaper) with a relative standard deviation (RSD) of 7.76%. Moreover, the SERS signal intensity was maintained at 60% of the initial intensity after the substrate was stored for 30 days. In addition, the Ag NPAs/SMP SERS substrate was also employed to detect crystal violet (CV) and methylene blue (MB) with the LODs of 10-6 M and 10-7 M, respectively. In summary, the Ag NPAs/SMP SERS substrate prepared in this study has great potential for the detection of organic dyes in ecological environments.

Raman spectrum combined with deep learning for precise recognition of Carbapenem-resistant Enterobacteriaceae.

Carbapenem-resistant Enterobacteriaceae (CRE) is a major pathogen that poses a serious threat to human health. Unfortunately, currently, there are no effective measures to curb its rapid development. To address this, an in-depth study on the surface-enhanced Raman spectroscopy (SERS) of 22 strains of 7 categories of CRE using a gold silver composite SERS substrate was conducted. The residual networks with an attention mechanism to classify the SERS spectrum from three perspectives (pathogenic bacteria type, enzyme-producing subtype, and sensitive antibiotic type) were performed. The results show that the SERS spectrum measured by the composite SERS substrate was repeatable and consistent. The SERS spectrum of CRE showed varying degrees of species differences, and the strain difference in the SERS spectrum of CRE was closely related to the type of enzyme-producing subtype. The introduced attention mechanism improved the classification accuracy of the residual network (ResNet) model. The accuracy of CRE classification for different strains and enzyme-producing subtypes reached 94.0% and 96.13%, respectively. The accuracy of CRE classification by pathogen sensitive antibiotic combination reached 93.9%. This study is significant for guiding antibiotic use in CRE infection, as the sensitive antibiotic used in treatment can be predicted directly by measuring CRE spectra. Our study demonstrates the potential of combining SERS with deep learning algorithms to identify CRE without culture labels and classify its sensitive antibiotics. This approach provides a new idea for rapid and accurate clinical detection of CRE and has important significance for alleviating the rapid development of resistance to CRE.