Band-gap-tunable carbon quantum dots for surface-enhanced Raman scattering

Band-gap-tunable carbon quantum dots for surface-enhanced Raman scattering | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Band-gap-tunable carbon quantum dots for surface-enhanced Raman scattering Zhiming Liu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5962648/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Carbon quantum dots (CQDs) with photoluminescence properties have been widely recognized, however, their inherent surface-enhanced Raman scattering (SERS) activity is rare reported. Herein, we propose a type of band-gap-tunable CQDs derived from metal-phenolic network (MPN) that exhibits excellent SERS performance. MPN-CQDs may be the ideal nonmetallic SERS substrates to accurately elucidate the chemical mechanism (CM) due to their simply controllable bandgap structure. By screening the doped metal elements of MPN, MPN-CQDs can realize the optimal SERS effect with the maximum Raman enhancement factor of 5.5×10 4 , also exhibit outstanding SERS reproducibility and stability. We then systematically disclose the interfacial photo-induced charge transfer process and corresponding migration pathways between band-gap-tunable MPN-CQDs and analyte. This class of nonmetallic SERS substrates is finally applied for detection of hemoglobin with high sensitivity; further combined with machine learning algorithm, we have successfully achieved precise identification of the heterogeneity of hemoglobin. This is the first evidence for the tunable SERS performance in CQDs, which also offers the facile avenue for in-depth understanding of CM in nonmetallic materials. Physical sciences/Optics and photonics/Optical techniques/Optical spectroscopy/Raman spectroscopy Physical sciences/Materials science/Nanoscale materials/Quantum dots Physical sciences/Materials science/Materials for optics/Nanophotonics and plasmonics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Carbon quantum dots (CQDs) have attracted widespread attention as promising low-cost optical nanomaterials for diverse applications 1 – 3 . The photoluminescence (PL) features of CQDs have been exploited well that their PL emission spans from ultraviolet (UV) to near-infrared (NIR) region 1 . Several mechanisms involving quantum confinement effect, quantum size effect, molecular state, surface/edge state, and crosslinking-enhanced emission effect have been proposed to explain the diverse PL forms 4 , 5 , such as multicolor fluorescence, aggregation-induced emission, NIR-II emitting, photochromism, afterglow, and so on 6 – 10 . The PL process is also accompanied by a certain degree of non-radiative relaxation and oxidative effect, which expands the potential of CQDs for photothermal and photodynamic therapy 11 – 13 . Therefore, a more detailed understanding of the photophysical properties will substantially pave the way for versatile CQDs in practical applications. Recently, nonmetallic materials have developed as emerging candidates for surface-enhanced Raman scattering (SERS) analysis 14 . Compared to commonly noble metal-based SERS substrates, nonmetallic materials are more economic, biocompatible, stable and SERS-repeatable 15 . Nonmetallic SERS substrates can also reduce the Raman shift fluctuations and metal-catalyzed side reactions caused by strong metal-adsorbate interaction 16 , 17 . Diverse novel materials have been proven to possess SERS capabilities, e.g., black phosphorus, transition metal dichalcogenides/oxides/tellurides/nitrides, metal-organic frameworks, nanostructured organic semiconductor, amorphous TiO 2 nanosheets, some of which even demonstrate the Raman enhancement activities comparable to noble metal nanostructures 18 – 25 . The SERS origin can be largely ascribed to the interfacial charge transfer (CT) process between nonmetallic substrate and analyte, an explanation of chemical mechanism (CM) theory 21 . The SERS potential of carbon-based materials has also been explored, offering more economical choices for SERS sensing 26 . The SERS activities of carbon materials are closely associated with their intrinsic structural characteristics. Graphene gives the first evidence whose sp 2 -hybridized structure can provide a Raman enhancement factor (EF) of about 10 1 27 . After that, Zhang et al. developed a surfactant-free growth method for fabrication of graphdiyne hollow microspheres with both sp 2 and sp hybridized carbon atoms, achieving the EF value of about 3.7×10 7 28 . Chen et al. also push out a porous carbon nanowire array with high EF value (~ 10 6 ), reproducibility, durability and biocompatibility 29 . The SERS effects of carbon nanostructures with quantum-scale sizes have currently only been observed in graphene quantum dots (GQDs), while their EFs are no more than 5×10 2 30, 31 . Heteroatom doping may be a valid strategy to improve the SERS capability GQDs, like nitrogen-doping, achieving a higher EF value of 3.2×10 3 32 . CQDs are undoubtedly more cost-effective choices compared to GQDs, however, their SERS activities are barely reported. Herein, we offer a facile strategy for band-gap tuning of CQDs using low-cost metal-phenolic network (MPN) as the precursor, a classic metal coordination structure enabling the band-gap control of MPN-CQDs by implantation of various metal atoms (Fe, Cu, Mn, Zn, Cr, etc.). Metal-doping significantly reduces the Femi level of CQDs about 1 eV, which affords more efficient CT process between MPN-CQDs and analyte, reaching the maximum Raman enhancement of 3.2×10 4 . For detailed CM explanation, different metal atoms implantation in MPN-CQDs causes subtle fluctuations in Femi levels from − 5.01 to -5.30 eV, leading to diverse CT pathways accompanied by aeolotropic SERS generation. As a kind of nonmetallic SERS substrates, MPN-CQDs also exhibit excellent reproducibility and stability, indicating the promising potential for SERS sensing. As a proof-of-concept, we finally utilize MPN-CQDs as the sensitive SERS substrate for hemoglobin detection, reaching the limited of detection of 0.1 mg/mL. We further accomplish the molecular structural discrimination of seven derivatives of hemoglobin assisted by machine learning algorithm with the accuracy up to 100%, unveiling a new route for the diagnosis of blood-related diseases. Results Synthesis and characterization of MPN-CQDs MPN offers a facile and low-cost metal-doping strategy for MPN-CQDs, the latter is then synthesized through one-pot hydrothermal method, where the energy level control of MPN-CQDs can be effortlessly realized by adjusting the doped metal atoms within MPN (Fig. 1 ). Figure 2 illustrates the characterization of the typical MPN-CQDs with Fe-doping (FeP-CQDs). The transmission electron microscopy (TEM) image reveals the quasi-spherical nanostructures of FeP-CQDs with an average diameter of 3.2 nm (Fig. 2 A). ‌FeP-CQDs are partially aggregated, possibly due to the phenolic residues on their surface. In the high-resolution TEM (HRTEM) image, we can also notice the lattice fringe of 0.21 nm corresponding to the (100) crystal plane of graphitic carbon (Fig. 2 B) 3 3 . The UV-vis absorbance spectrum of Fe-MPN precursor displays a typical broad ligand-to-metal charge transfer (LMCT) band at approximately 575 nm assigned to the metal-phenolic coordination 3 4 , which then disappears after carbonization, and a characteristic CQDs peak at ~ 350 nm emerges (Fig. 2 C). Raman analysis also perceives the chemical composition alteration, where the Raman bands (1350 and 1485 cm −1 ) of Fe-MPN representing the skeletal vibrations of the benzene ring are replaced as the D (1360 cm −1 ) and G (1565 cm −1 ) bands in the Raman spectrum of FeP-CQDs (Fig. 2 D) 3 5 . Fourier transformed infrared (FTIR) spectrum further reveals the chemical structure of FeP-CQDs, where the absorption peaks at 3440, 2925, 1642, 1390, 1080 and 598 cm⁻¹ correspond to O-H, C-H, C = C/C = O, CH 2 /CH 3 /O-H, C-C/C-O-C, and Fe-O vibrations, respectively (Fig. 2 E) 3 5 , 3 6 . X-ray photoelectron spectroscopy (XPS) further investigated the precise elemental composition of the FeP-CQDs, like C, O, and Fe elements (Fig. 2 F). The high-resolution C 1s spectrum can be deconvoluted into three peaks at 284.8, 286.5, and 288.3 eV, corresponding to C = C, C-O, and C = O groups, respectively (Fig. 2 G) 3 7 . The peaks at 531.5, 533.9, and 535.5 eV in O 1s spectrum are assigned to C = O, C-O, and -COOH groups, respectively (Fig. 2 H) 3 8 . In the high-resolution Fe 2p spectrum, three fit peaks at 709.6, 713.0, and 724.0 eV corresponding to Fe 2p (II) 3/2, Fe 2p (III) 3/2, and Fe 2p 1/2, respectively can also be noticed (Fig. 2 I) 3 9 . In addition, the zeta potential of FeP-CQDs is measured to be -30.3 mV (Supplementary Fig. S1 ), indicating a good water dispersibility. SERS performance of FeP-CQDs To investigate the SERS properties of FeP-CQDs, methylene blue (MB) was used as the probe molecule for Raman analysis under 633 nm excitation (Fig. 3 A). The concentration-dependent SERS spectra of MB dye deposited onto FeP-CQDs are illustrated in Fig. 3 B, where the SERS signals gradually decrease with the declining dye concentration; and the major SERS peaks are still observed at the concentration of 5 × 10 − 8 M. Theoretical limited of detection (LOD) value was then calculated from fitting curves of typical SERS peaks (1624, 1398 and 770 cm − 1 ), obtaining a minimum LOD value to be 3.8 × 10 − 8 M (Fig. 3 C). We also calculated the Raman EFs of these bands at various dye concentrations, as displayed in Fig. 3 D. The maximum EF value is evaluated to be 5.5 × 10 4 at 770 cm − 1 , which is about 20–100 times higher than the data in previous reports on GQDs 30 – 32 . Figure 3 E shows the SERS spectral map of MB dye on FeP-CQDs acquired from 30 random spectral lines, where the relative standard deviation (RSD) values of the three bands at 1624, 1398 and 770 cm − 1 are calculated to be 1.21%, 1.65%, and 1.62%, respectively (Fig. 3 F), indicating the outstanding SERS reproducibility of FeP-CQDs. Moreover, SERS maps were plotted using these Raman bands, which depicted the boundary of the dye sample very well (Fig. 3 G), indicating a good SERS uniformity. For SERS stability study, a long-term Raman monitoring was performed. As demonstrated in Fig. 3 H, the SERS intensity of MB at 1624 cm − 1 reduces by only 6% after FeP-CQDs are stored at 4°C for 40 days, and after a 300-day observation, the SERS signal at 1624 cm − 1 remains at 77%. The superior SERS stability of FeP-CQDs highlights their considerable potential for real-world applications. To confirm the versality of FeP-CQDs, SERS experiments were finally carried out on diverse dye molecules, such as malachite green (MG), crystal violet (CV), rhodamine 6G (R6G), rhodamine B (RhB), IR-820, and IR-780. Figure 3 I shows the mean SERS spectra of these molecules at 5 × 10 − 5 M after deposited on FeP-CQDs, where intense and plentiful molecular fingerprint information can be noticed. SERS mechanism of FeP-CQDs In order to elucidate the underlying SERS mechanism, a SERS comparison experiment was carried out between FeP-CQDs and polyphenolic CQDs without metal doping (P-CQDs). As shown in Fig. 4 A, plentiful Raman fingerprint information can be obviously noticed after MB molecules (5 × 10 − 5 M) deposited on FeP-CQDs under 633 nm excitation. In marked contrast, P-CQDs only exhibits negligible Raman enhancement effect on MB. To explain the strong SERS effect of FeP-CQDs, we first evaluated the interaction of MB molecules with FeP-CQDs using UV-vis spectroscopy. Figure 4 B illustrates the characteristic absorption peak of MB at 662 nm with a shoulder at 614 nm, corresponding to the n-π* transition of MB molecule. Upon interaction with FeP-CQDs, an evident red-shift of these peaks occurs, indicating a strong interfacial interaction between MB and FeP-CQDs. Moreover, the ultraviolet photoelectron spectroscopy (UPS) was performed to measure the Fermi levels (E f ) of CQDs, which were calculated to be -4.07 and − 5.25 eV for P-CQDs and FeP-CQDs, respectively (Fig. 4 C). Then the energy level diagrams and charge transfer processes between MB and CQD-SERS sensors are plotted in Fig. 4 D. The highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) levels of MB are at -6.26 and − 4.55 eV, respectively. Charge transfer occurs when the contact between the molecule and substrate reaches equilibrium, resulting in rearrangement of the band structure 31 , 40 . According to Feynman's single-photon Raman scattering principle, the Raman scattering process involves three steps, each of which can be enhanced to amplify the Raman signals 41 : (1) excitation of ground state electron by incident light, (2) coupling of excited electron with phonon, and (3) emission of scattered light as electron relax back to the ground state. In MB/P-CQDs system, the Fermi level of P-CQDs (-4.07 eV) mismatches with the HOMO level of MB (-6.26 eV), indicating an inefficient ground-state charge transfer. However, the Fermi level of FeP-CQDs (-5.25 eV) is located much close to the HOMO level of MB (-6.26 eV). Due to the ground-state charge transfer process, the electrons near HOMO in MB/FeP-CQDs system show more possibility to contribute the Raman scattering process of MB, leading to the enhancement of electron-phonon coupling (second step of Feynman process) 31 . The magnified Raman signals originate from photo-induced charge transfer (PICT) process involves the charge transfer from MB’ HOMO level to FeP-CQDs’ Fermi level (µ i−CT ) and from E f(FeP−CQDs) to LUMO (MB) (µ k−CT ). PICT between MB and FeP-CQDs may also amplify the molecular polarizability tensor to enhance Raman scattering of molecules according to Herzberg-Teller theory 15 , 42 . The charge transfer across the interface between FeP-CQDs and MB was confirmed by density functional theory (DFT) calculation. As difficulty in explaining the nanostructures of CQDs prepared from hydrothermal method but retaining the precursor residues on their surface, we adopted Fe-MPN as the structural model of FeP-CQDs for DFT calculation. Figure 4 E exhibits the projected density of states (PDOS) of MB molecules before and after adsorption on SERS substrate, where the electronic structure of MB significantly changes after adsorption, implying the charge of MB is redistributed after interacting with the SERS substrate. The charge density difference map also confirms the redistribution of the electron cloud around MB after interfacial charge transfer (Fig. 4 F). Bader charge analysis shows that electrons are transferred from the SERS substrate to probe molecule with a transfer amount of 0.150 eV. Molecular resonance also contributes to the outstanding SERS effect of FeP-CQDs on MB (Fig. 4 D), as the excitation energy (633 nm, 1.96 eV) matches well with the HOMO-LUMO gap of MB (1.71 eV). We further compared the SERS performance of FeP-CQDs on MB under three traditional excitation lasers (514.5, 633, and 785 nm). As demonstrated in Fig. 4 G, the SERS signals of MB triggered by 785 nm is obviously weaker than that by 633 nm. It can be deduced that the energy of the 785 nm laser (1.58 eV) is insufficient to induce molecular resonance of MB but PICT occurs in MB/FeP-CQDs system. In marked contrast, almost no Raman enhancement effect on MB is noted under 514.5 nm laser irradiation, probably owing to its energy (2.41 eV) mismatches the energy for molecular resonance of MB and PICT pathways. Tunable SERS activities of MPN-CQDs and Raman enhancement mechanism We have proven that metal coordination in MPN precursor could remarkably reduce the Fermi level of CQDs (Fig. 4 D), which provides a facile strategy for band-gap control of MPN-CQDs. Besides FeP-CQDs, we also prepared other four kinds of MPN-CQDs doped with Cu, Cr, Zn, and Mn elements, namely: CuP-CQDs, CrP-CQDs, ZnP-CQDs, and MnP-CQDs, respectively. Above all display excellent SERS activities and repeatabilities to MB dye with the experimental LOD value of 5 × 10 − 8 M (Supplementary Figs. S2-3), further verifying the applicability of metal-doping strategy for development of CQDs-based SERS nanosensors. Then, a comparison among these MPN-CQDs was conducted. Figure 5 A shows the SERS spectra of MB molecules at 5 × 10 − 5 M deposited on various MPN-CQDs irradiated by 633 nm laser. The intensity of the SERS band at 1624 cm − 1 was further selected for quantitative analysis, which revealed a fluctuation in SERS intensity associated with metal-doping. The SERS intensities are ranked from highest to lowest as follows: CuP-CQDs > FeP-CQDs > CrP-CQDs > ZnP-CQDs > MnP-CQDs. The SERS effects of MPN-CQDs doped with various metals under 785 laser irradiation illustrate similar trend (Fig. 5 B, Supplementary Fig. S4). For explanation of the SERS mechanism of MPN-CQDs regarding to metal-doping, the interaction of MPN-CQDs with MB was first evaluated by the UV-vis absorbance spectroscopy. As displayed in Fig. 5 C, obvious red-shift of the characteristic absorbance bands of MB emerges in these spectral lines, which provides the preliminary evidence of the interfacial interaction between MPN-CQDs and MB. The Fermi levels of CuP-CQDs, CrP-CQDs, ZnP-CQDs and MnP-CQDs were further measured by UPS analysis, which were calculated to be -5.30, -5.08, -5.04 and − 5.01 eV, respectively (Fig. 5 D). Therefore, we can reasonably infer that the discrepant SERS activity of MPN-CQDs is closely related to their tunable band-gaps. The charge transfer processes between MB and MPN-CQDs with different Fermi levels are finally plotted in Fig. 5 E. PICT in MB/MPN-CQDs system includes charge transfer from MB’ HOMO level to MPN-CQDs’ Fermi level and from the Fermi level of MPN-CQDs to MB’ LUMO level. In addition, the molecular resonance at 633 nm laser irradiation also amplify the cross-section of MB, leading to better SERS signals compared to that under 785 nm laser. The Fermi levels of five MPN-CQDs are ranked as: CuP-CQDs < FeP-CQDs < CrP-CQDs < ZnP-CQDs < MnP-CQDs, which are opposite to their SERS performances (Fig. 5 A, B). Among them, the Fermi levels of CuP-CQDs (-5.30 eV) is closest to the HOMO level of MB (-6.26 eV), which may provide more electrons to participate in the ground-state charge transfer between the MB and CuP-CQDs, thereby achieving superior SERS performance. Moreover, the nearly symmetrical match of CuP-CQDs’ Fermi level to MB’ HOMO and LUMO levels further facilitates PICT in MB/CuP-CQDs system. These effects gradually subside as the Fermi level increases, reflecting decreasing Raman enhancement activity. SERS detection and discrimination of hemoglobin based on MPN-CQDs To investigate the potential of MPN-CQDs for bio-SERS analysis, hemoglobin (Hb), a functional protein for transportation of oxygen in blood circulation, was adopted as the biomolecule model. The SERS effect of FeP-CQDs on bovine Hb is first shown in Supplementary Fig. S5. It can be observed that the Raman signals of bovine Hb (1 mg/mL) are obviously enhanced after deposited on FeP-CQDs, which can be ascribed to the efficient PICT process between biomolecule and substrate, as well as the molecular resonance of Hb under 514.5 nm laser. The typical fingerprint features of Hb can be clearly identified, such as 1639 and 1586 cm − 1 (C α C m asymmetrical stretching), 1539 cm − 1 (C β C β stretching), 1372 and 1359 cm − 1 (pyrrole half-ring symmetrical stretching), and 674 cm − 1 (pyrrole deformation stretching), etc. (table S1 ) 43 . We also notice a concentration-dependent SERS effect of FeP-CQDs on bovine Hb with favorable SERS repeatability, where the SERS signals of Hb at 0.1 mg/mL are still detected (Supplementary Fig. S6). We then tested the oxygenated hemoglobin (OxyHb) extracted from fresh human blood using FeP-CQDs-based SERS technique, which was demonstrated in Fig. 6 A. Similar to bovine Hb, the SERS signals of OxyHb induced by FeP-CQDs under 532 nm irradiation linearly decrease as the molecule concentration declines with the minimum detectable concentration of 0.1 mg/mL (Fig. 6 B). The SERS spectral map of OxyHb on FeP-CQDs visually displays a good SERS reproducibility (Fig. 6 C) with the RSD value of 3.69% for the typical peak at 1586 cm − 1 (Fig. 6 D). SERS that can supply molecular fingerprint information has newly been exploited as an spectromics tool to decode the heterogeneity in chemical composition of biosamples 44 . The molecular structure of Hb determines its physiological and pathological functions. The structural abnormality of Hb is closely bound up with various hematologic diseases, like methemoglobinemia, sulfhemoglobinemia, sickle cell anemia, thalassemia, and hemoglobin C disease. To explore the potential of MPN-CQDs in revealing the heterogeneity of Hb-like proteins, we prepared four Hb variants—OxyHb, deoxyhemoglobin (DeoxyHb), methemoglobin (MetHb), and fluoromethemoglobin (MetHbF)—along with their corresponding prosthetic groups, protoporphyrin IX (PPIX) and ferroheme (Fig. 6 E). Then the SERS spectra of these molecules were acquired utilizing FeP-CQDs as the SERS sensor. As depicted in Fig. 6 F, although the SERS spectra of ferroheme and PPIX display comparable spectral patterns, distinct differences are observed in their characteristic SERS peaks. For example, the SERS bands of PPIX at 738, 1332 and 1603 cm − 1 evidently red-shift to 755 cm − 1 (pyrrole breathing stretching), 1369 cm − 1 (pyrrole half-ring symmetrical stretching) and 1627 cm − 1 (C α C m asymmetrical stretching) in the SERS spectrum of ferroheme, respectively, indicating substantial structural modification of PPIX following Fe-complexation. The SERS spectral patterns of OxyHb and DeoxyHb show a closer alignment with that of ferroheme. We can also observe that the Raman bands sensitive to oxygenation state at 1585 and 1637 cm − 1 are significantly stronger in the spectral line of OxyHb than that of DeoxyHb 45 . Furthermore, the peak at 1337 cm − 1 , corresponding to Fe 2+ , significantly diminishes, whereas the band at 1372 cm − 1 , associated with Fe 3+ , intensifies following the conversion of OxyHb to MetHb 46 . In addition, a blue-shift is noticed in the SERS band of MetHb at 1637 cm − 1 after treatment with NaF, which may be a result of the Fe-F − interaction 47 . Ultimately, a machine learning algorithm based on principal component analysis and linear discriminant analysis (PCA-LDA) was employed to assist SERS in the discernment of the subtle variations among these Hb derivatives. The SERS data were first reduced in dimensionality using the PCA algorithm and then classified through a supervised LDA model. 80% of the SERS data was utilized for model training, and it was observed that the recognition accuracy of the LDA model was directly dependent on the number of principal component (PC) loadings, reaching 100% when the number of PCs exceeded 10 (Supplementary Fig. S7). The remaining 20% of the data was allocated as the test set. As shown in the receiver operating characteristic (ROC) curves (Fig. 6 G), the areas under the ROC curve (AUCs) are calculated to be 1.00, 1.00, 1.00, 1.00, 1.00 and 0.95 for PPIX, ferroheme, OxyHb, DeoxyHb, MetHb and MetHbF, respectively, indicating excellent recognition performance of PCA-LDA algorithm. The confusion matrix was then drawn based on the structured and trained LDA model, revealing fantastic classification accuracy of 100% for all Hb derivatives (Fig. 6 H). In addition, a scatter plot following t-distributed stochastic neighbor embedding (TSNE) analysis is presented in Fig. 6 I to visualize the classification quality, demonstrating clear separation among all Hb molecules. Therefore, MPN-CQDs-based SERS technique combined with machine learning algorithm supplies an available strategy for analyzing the heterogeneity of biomolecules, further contributing to the understanding of molecule-related diseases. Discussion In summary, we have explored a novel optical characteristic of CQDs—the Raman enhancement effect—which fundamentally distinguishes them from their conventional photoluminescence properties. The inherent SERS activity of CQDs is originated from the charge transfer between the material and analyte. For systematical interpretation of the SERS mechanism of CQDs, we propose a type of MPN-derived CQDs with tunable band-gap which can be simply controlled by adjustment of coordinated metal in their polyphenol precursor. Metal-doping can dramatically reduce the Fermi level of polyphenol CQDs, and the precise Fermi level of MPN-CQDs can be further finely tuned by varying the type of doped metal. Five MPN-CQDs doped with Cu, Fe, Cr, Zn, and Mn, respectively, are prepared, exhibiting Fermi levels ranging from − 5.30 to -5.01 eV. The fluctuation in Fermi level may directly influence the interfacial photo-induced charge transfer between MPN-CQDs and analyte, leading to the variations in the SERS performance of MPN-CQDs. By optimizing the band-gap structure of MPN-CQDs, the maximum Raman enhancement factor of 5.5 × 10 4 is achieved, along with excellent SERS reproducibility and stability. Therefore, band-gap-tunable MPN-CQDs may be an ideal nonmetallic SERS substrate for elucidation of chemical mechanism in SERS. For bio-SERS analysis, MPN-CQDs can also serve as a low-cost SERS platform for the detection and discrimination of hemoglobin derivatives with the assistance of machine learning algorithm. The discovery of tunable SERS performance in CQDs uncovers a novel route for insight into the Raman enhancement mechanisms of nonmetallic materials. Methods Synthesis of MPN-CQDs MPN was fabricated via the simple self-assembly of polyphenols in the presence of metal ions, following a previously established protocol 48 . In brief, various metal ion solutions (Fe 3+ , Cu 2+ , Cr 2+ , Zn 2+ , and Mn 2+ ) at a concentration of 0.60 mM were added into the 0.24 mM tannic acid (TA) solution, and the pH was adjusted to greater than 7.4 using NaOH. The complexation products were formed within 1 h. Then, the mixtures were transferred to a poly(tetrafluoroethylene) (Teflon)-lined autoclave container and heated at 180°C for 6 h. The products were filtered using a 0.22 µm filter to remove large particles and purified using a dialysis bag (1000 Da) for 10 h. Characterization TEM analysis was carried out using a JEM-2100HR (JEOL, Tokyo, Japan) electron microscope operating at 200 kV. UPS spectra were measured using a photoelectron spectrometer (AXIS SUPRA, Shimadzu, Tokyo, Japan) equipped with angle-resolved capability. XPS spectra were recorded using an Escalab 250XI X-ray photoelectron spectrometer (Thermo Fisher Scientific, USA), with internal referencing to the C1s peak (sp² hybridized carbon binding energy) at 284.1 eV. UV-Vis absorption spectra were obtained using a UV-Vis spectrophotometer (UV-6100, MAPADA). FTIR spectra were measured using a Nicolet iS50 spectrometer (Thermo Fisher Scientific, USA). SERS experiments 2 µL MPN-CQDs was mixed with equal volume of dye solutions (MB, MG, CV, R6G, RhB, IR-820, and IR-780) at varying concentrations from 10 − 4 M to 10 − 8 M. The mixture was then deposited onto Si wafer for SERS measurement. The Raman spectra were recorded using a confocal Raman spectrophotometer (InVia, Renishaw, UK) equipped with 514.5, 532, 633, and 785 nm lasers. A 20 × objective was used to focus the laser beam and to collect the Raman signal. For bio-SERS analysis, bovine Hb molecules with different concentrations (0.1 ~ 8 mg/mL) were mixed with equal volume of FeP-CQDs and then detected under the Raman spectrophotometer with 514.5 nm laser. For classification of Hb derivatives, two prosthetic groups, PPIX and ferroheme, were purchased from Shanghai Macklin Biochemical Co., Ltd. OxyHb was extracted from fresh human blood and its derivatives (DeoxyHb, MetHb, and MetHbF) were prepared according to the established methods by Venkatesh et al 49 . PPIX, ferroheme, OxyHb, DeoxyHb, MetHb, and MetHbF after mixture with FeP-CQDs were analyzed using the Raman spectrophotometer with 532 nm laser. The PCA-LDA analysis of the SERS data of Hb was carried out in the Python software with the Scikit-Learn library (version 1.3.2). The raw Raman data were first baseline-corrected and randomly divide all spectral data into training and testing sets in an 8:2 ratio using train_test_split function. Subsequently, PCA was employed to reduce the dimensionality of the data, followed by LDA to further improve the model's accuracy, leading to the development of a PCA-LDA classification model. To visually assess the classification performance of the PCA-LDA model, the final model data was projected and visualized using t-SNE analysis, and scatter plots were generated to provide an intuitive representation of the classification outcomes among different hemoglobin derivatives. First principle calculation Theoretical calculation was performed using DFT as implemented in the Vienna Ab initio Simulation Package (VASP). The exchange-correlation potential was described using the generalized gradient approximation of Perdew-Burke-Ernzerhof (GGA-PBE). The projector augmented-wave (PAW) method was employed to treat interactions between ion cores and valence electrons. The plane-wave cutoff energy was fixed to 450 eV. Given structural models were relaxed until the Hellmann–Feynman forces smaller than − 0.02 eV/Å and the change in energy smaller than 10 − 5 eV was attained. Grimme’s DFT-D3 methodology was used to describe the dispersion interactions among all the atoms in adsorption models. The Г-centered Monkhorst-Packgrid k-points were set as 2 × 2 × 2 for structure optimization and static calculations, and 3 × 3 × 3 for DOS calculation, respectively. Declarations Data availability The data that support the findings of this study are available from the corresponding author upon request. Source data are provided with this paper. Competing interests The authors declare no competing interests. Additional information Supplementary information The online version contains supplementary material available at xxxxxx. Author contributions Z.L. conceived the study. Z. L., B.X., and Z.G. supervised the project. M.N., Z.J. and H.C. completed the experiments. Z.L. and M.N. designed the figures and drafted the manuscript. Z.L., M.N., Z.J., J.L and J.C. managed statistical analyses. K.X., Y.S., D.S. and B.Y. achieved the material characterization. Z.L., H.C. and Z.G. provided funding support. All authors discussed the results, provided critical feedback, and contributed to the final manuscript. Acknowledgements This work was supported by the National Natural Science Foundation of China (62175071 and 32071399), the Guangdong Basic and Applied Basic Research Foundation (2024A1515011675 and 2021A1515110265), the Science and Technology Program of Guangzhou (2024A04J4558). References Dordevic L, Arcudi F, Cacioppo M, Prato M (2022) A multifunctional chemical toolbox to engineer carbon dots for biomedical and energy applications. Nat Nanotechnol 17:112–130 Krasley AT, Li E, Galeana JM, Bulumulla C, Beyene AG, Demirer GS (2024) Carbon nanomaterial fluorescent probes and their biological applications. Chem Rev 124:3085–3185 Zhang Y, Lu S (2024) Lasing of carbon dots: Chemical design, mechanisms, and bright future. Chem 10:134–171 Wang B, Lu S (2022) The light of carbon dots: From mechanism to applications. Matter 5:110–149 Yang Z et al (2023) Zero-dimensional carbon nanomaterials for fluorescent sensing and imaging. Chem Rev 123:11047–11136 Jiang L-Y, Zhou Y-C, Zhang S-F, Shao H-C, Liang Y-C (2024) Time division colorful multiplexing based on carbon nanodots with modifiable colors and lifetimes. Nano Lett 24:8418–8426 Yang H et al (2019) Hydrophobic carbon dots with blue dispersed emission and red aggregation-induced emission. Nat Commun 10:1789 Yang M, Han Y, Bianco A, Ji D-K (2024) Recent progress on second near-infrared emitting carbon dots in biomedicine. ACS Nano 18:11560–11572 Zheng G-S et al (2024) Photooxidation triggered ultralong afterglow in carbon nanodots. Nat Commun 15:2365 Guo Z et al (2024) Multi-stimuli-responsive carbon dots with intrinsic photochromism and in situ radical afterglow. Adv Mater 36:2409361 Wang H et al (2024) Carbon dots with integrated photothermal antibacterial and heat-enhanced antioxidant properties for diabetic wound healing. Small 20:2403160 Zhang X et al (2024) Donor-acceptor type supra-carbon-dots with long lifetime photogenerated radicals boosting tumor photodynamic therapy. Angew Chem Int Ed 63:e202410522 Liu H et al (2023) Afterglow photodynamic therapy based on carbon dots embedded silica nanoparticles for nondestructive teeth whitening. ACS Nano 17:21195–21205 Gu Y et al (2023) Recent progress on noble-free substrates for surface-enhanced Raman spectroscopy analysis. Coord Chem Rev 497:215425 Zhang Y et al (2024) Phase-engineered transition metal dichalcogenides for highly efficient surface-enhanced Raman scattering. Nano Lett 24:14293–14301 Wang S et al (2024) Beyond the charge transfer mechanism for 2D materials-assisted surface enhanced Raman scattering. Anal Chem 96:9917–9926 Liang C et al (2022) Band structure engineering within two-dimensional borocarbonitride nanosheets for surface-enhanced Raman scattering. Nano Lett 22:6590–6598 Lin J et al (2015) Enhanced raman scattering on in-plane anisotropic layered materials. J Am Chem Soc 137:15511–15517 Sun H, Cong S, Zheng Z, Wang Z, Chen Z, Zhao Z (2019) Metal-organic frameworks as surface enhanced Raman scattering substrates with high tailorability. J Am Chem Soc 141:870–878 Song X et al (2024) Two-orders-of-magnitude enhancement of SERS activity via a simple surface engineering of quasi-metal single-crystal frameworks. Nano Lett 24:11683–11689 Tang X et al (2024) Exploring and engineering 2D transition metal dichalcogenides toward ultimate SERS performance. Adv Mater 36:e2312348 Li Y et al (2021) Lamellar hafnium ditelluride as an ultrasensitive surface-enhanced Raman scattering platform for label-free detection of uric acid. Photon Res 9:1039–1047 Guan H et al (2021) General molten-salt route to three-dimensional porous transition metal nitrides as sensitive and stable Raman substrates. Nat Commun 12:1376 Yilmaz M et al (2017) Nanostructured organic semiconductor films for molecular detection with surface-enhanced Raman spectroscopy. Nat Mater 16:918–924 Lyu X et al (2024) Molecularly confined topochemical transformation of MXene enables ultrathin amorphous metal-oxide nanosheets. ACS Nano 18:2219–2230 Liang X et al (2021) Carbon-based SERS biosensor: From substrate design to sensing and bioapplication. NPG Asia Mater 13:8 Ling X et al (2010) Can graphene be used as a substrate for Raman enhancement? Nano Lett 10:553–561 Zhang L, Yi W, Li J, Wei G, Xi G, Mao L (2023) Surfactant-free interfacial growth of graphdiyne hollow microspheres and the mechanistic origin of their SERS activity. Nat Commun 14:6318 Chen N et al (2020) Porous carbon nanowire array for surface-enhanced Raman spectroscopy. Nat Commun 11:4772 Cheng H, Zhao Y, Fan Y, Xie X, Qu L, Shi G (2012) Graphene-quantum-dot assembled nanotubes: A new platform for efficient Raman enhancement. ACS Nano 6:2237–2244 Liu D et al (2018) Raman enhancement on ultra-clean graphene quantum dots produced by quasi-equilibrium plasma-enhanced chemical vapor deposition. Nat Commun 9:193 Das R, Parveen S, Bora A, Giri PK (2020) Origin of high photoluminescence yield and high SERS sensitivity of nitrogen-doped graphene quantum dots. Carbon 160:273–286 Kumar VB, Mirsky SK, Shaked NT, Gazit E (2024) High High quantum yield amino acid carbon quantum dots with unparalleled refractive index. ACS Nano 18:2421–2433 Chen J et al (2022) Assembly of bioactive nanoparticles via metal-phenolic complexation. Adv Mater 34:2108624 Rahim MA et al (2017) Rust-mediated continuous assembly of metal-phenolic networks. Adv Mater 29:1606717 Liu H et al (2022) Carbon dots with intrinsic bioactivities for photothermal optical coherence tomography, tumor-specific therapy and postoperative wound management. Adv Healthc Mater 11:2101448 Li M et al (2024) Dynamic visualization monitoring of cell membrane damage using polarity-responsive amphiphilic carbon dots. Chem Eng J 482:149038 Guo D et al (2022) Photocatalytic Pt(IV)-coordinated carbon dots for precision tumor therapy. Adv Sci 9:e2205106 Liu Y et al (2022) Ultrasmall Fe-doped carbon dots nanozymes for photoenhanced antibacterial therapy and wound healing. Bioact Mater 12:246–256 Miao X et al (2019) Graphene quantum dots wrapped gold nanoparticles with integrated enhancement mechanisms as sensitive and homogeneous substrates for surface-enhanced Raman spectroscopy. Anal Chem 91:7295–7303 Ling X, Moura LG, Pimenta MA, Zhang J (2012) Charge-transfer mechanism in graphene-enhanced Raman scattering. J Phys Chem C 116:25112–25118 Ma H et al (2024) Surface-enhanced Raman spectroscopy: Current understanding, challenges, and opportunities. ACS Nano 18:14000–14019 Hu S, Smith KM, Spiro TG (1996) Assignment of protoheme resonance Raman spectrum by heme labeling in myoglobin. J Am Chem Soc 118:12638–12646 Cutshaw G, Uthaman S, Hassan N, Kothadiya S, Wen X, Bardhan R (2023) The emerging role of Raman spectroscopy as an omics approach for metabolic profiling and biomarker detection toward precision medicine. Chem Rev 123:8297–8346 Rusciano G (2010) Experimental analysis of Hb oxy–deoxy transition in single optically stretched red blood cells. Phys Med 26:233–239 Marzec KM, Rygula A, Wood BR, Chlopicki S, Baranska M (2014) High-resolution Raman imaging reveals spatial location of heme oxidation sites in single red blood cells of dried smears. J Raman Spectrosc 46:76–83 Asher SA, Vickery LE, Schuster TM, Sauer K (1977) Resonance Raman spectra of methemoglobin derivatives. Selective enhancement of axial ligand vibrations and lack of an effect of inositol hexaphosphate. Biochemistry 16:5849–5856 Ejima H et al (2013) One-step assembly of coordination complexes for versatile film and particle engineering. Science 341:154–157 Venkatesh B, Ramasamy S, Mylrajan M, Asokan R, Manoharan PT, Rifkind M (1999) Fourier transform Raman approach to structural correlation in hemoglobin derivatives. Spectrochim Acta Mol Biomol Spectrosc 55:1691–1697 Additional Declarations There is NO Competing Interest. Supplementary Files Supplementaryinformation.docx Supplementary information Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5962648","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":414771227,"identity":"23d7f5c5-cc43-47a9-9ef2-3fa98414b202","order_by":0,"name":"Zhiming Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuElEQVRIiWNgGAWjYBACPmYQaWDDw8/MfPABUVrYwFoq0mQk29mSDYjTAibPHLYxOM9jJkCcFnYe408325h5jA8zmDEw1NhEE+EwHgPj3DY2HrPDDGkPGI6l5TYQoyU5t40HpOW4AWPDYeK0HM5tk+AxbmZskyBWi2FzzhkDHgNmZjZitbAVM+dUJPBIHGZjNkggxi/8/Ic3f84x+G/P33/+44MPNTaEtaCCBNKUj4JRMApGwSjABQA/fzKPlKAB5AAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-0879-9438","institution":"College of Biophotonics, School of Optoelectronic Science and Engineering, South China Normal University","correspondingAuthor":true,"prefix":"","firstName":"Zhiming","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2025-02-05 06:05:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5962648/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5962648/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":76190380,"identity":"2306b4e4-87a7-45df-b907-0f2eaee2735b","added_by":"auto","created_at":"2025-02-13 09:30:24","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":268702,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSchematic illustration ofthe MPN-CQDs-based SERS system. \u003c/strong\u003e(A) Synthesis of band-gap-tunable MPN-CQDs. (B) The SERS performance and mechanism of MPN-CQDs.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-5962648/v1/6a441c501d869352f70fa829.png"},{"id":76190772,"identity":"f61bd237-1b44-449a-a281-33f069feb915","added_by":"auto","created_at":"2025-02-13 09:38:29","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":982694,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCharacterization of FeP-CQDs. \u003c/strong\u003e(A) TEM image and corresponding size distribution of FeP-CQDs. Insert shows the size distribution of FeP-CQDs. (B) HRTEM image of FeP-CQDs. UV-vis absorbance(C) and Raman (D) spectra of Fe-MPN and FeP-CQDs, respectively. (E) FTIR spectrum of FeP-CQDs. XPS full spectrum (F) and high-resolution C 1s (G), O 1s (H) and Fe 2p (I) spectra of FeP-CQDs, respectively.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-5962648/v1/7bd7a6eae7dccd1df65203c1.png"},{"id":76190383,"identity":"f128bb55-76c5-416a-91c5-afd5a024ca73","added_by":"auto","created_at":"2025-02-13 09:30:28","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":794889,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSERS performance of FeP-CQDs. \u003c/strong\u003e(A) Schematic diagram of the SERS analysis based on FeP-CQDs. (B) SERS spectra of MB molecules on FeP-CQDs at different concentrations. SERS intensities (C) and corresponding EFs (D) of three typical bands (1624, 1398, and 770 cm\u003csup\u003e-1\u003c/sup\u003e) as a function of concentration. (E) SERS spectral map of MB dye plotted by 30 random spectral lines. (F) RSD values of the bands at 1624, 1398, and 770 cm\u003csup\u003e-1\u003c/sup\u003e, respectively. (G) SERS maps of MB on FeP-CQDs. (H) Time-dependent SERS spectra of MB on FeP-CQDs during a 300-day observation. (I) SERS spectra of other dye molecules (MG, CV, R6G, RhB, IR-820, and IR-780) triggered by FeP-CQDs.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-5962648/v1/1e0eb6153e11b41e2bc6d708.png"},{"id":76190386,"identity":"86811ef7-7178-4f8b-a45f-770cc8cb31f3","added_by":"auto","created_at":"2025-02-13 09:30:28","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":541629,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRaman enhancement mechanism of FeP-CQDs. \u003c/strong\u003e(A) A comparison of the SERS activity between FeP-CQDs and P-CQDs. (B) UV-vis absorbance spectra of FeP-CQDs before and after interacting with MB. (C) UPS spectra of FeP-CQDs and P-CQDs. (D) Schematic illustration of theenergy level diagrams and charge transfer processes between MB and FeP-CQDs or P-CQDs. (E) Projected density of states of MB molecules before and after adsorption on SERS substrate. (F) Charge density difference of MB adsorbing on SERS substrate. (G) SERS spectra of MB adsorbed on FeP-CQDs under different lasers with the same condition (0.5 mW, 2 s). (H) Schematic illustration of the charge transfer processes between MB and FeP-CQDs under different lasers.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-5962648/v1/0652964293f035fbfc4d7761.png"},{"id":76190403,"identity":"e7ad0bbe-1e05-46ec-8e7d-2c44634857ed","added_by":"auto","created_at":"2025-02-13 09:30:30","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":383287,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTunable SERS performance of MPN-CQDs. \u003c/strong\u003eSERS spectra and corresponding quantitative data of MB molecules deposited on five different MPN-CQDs (CuP-CQDs, FeP-CQDs, CrP-CQDs, ZnP-CQDs, and MnP-CQDs) under 633 nm (A) and 785 nm (B) laser irradiation, respectively. (C) UV-vis absorbance spectra of different MPN-CQDs after interacting with MB. (D) UPS spectra of CuP-CQDs, CrP-CQDs, ZnP-CQDs, and MnP-CQDs. (E) Schematic illustration of the energy level diagrams and charge transfer processes between MB and MPN-CQDs with different Fermi levels.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-5962648/v1/0e70e11d1aafe7a5c67bffa6.png"},{"id":76190393,"identity":"86b16b7d-d9b1-4e99-89ce-1177c0dd9ad1","added_by":"auto","created_at":"2025-02-13 09:30:28","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":482253,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDiscrimination of hemoglobin using MPN-CQDs-based SERS platform. \u003c/strong\u003e(A) Concentration-dependent SERS spectra of OxyHb deposited on FeP-CQDs and (B) corresponding quantitative analysis. (C) SERS spectral map of OxyHb plotted by 20 random spectral lines. SERS spectra of MB molecules on FeP-CQDs at different concentrations. (D) RSD value of the band at 1585 cm\u003csup\u003e-1\u003c/sup\u003e. UV-vis absorbance spectra (E) and SERS spectra (F) of different Hb-related molecules. (G) ROC curves and corresponding AUC values analyzed by PCA-LDA algorithm. (H) Confusion matrix. (I) Scatter plot of the Hb derivatives after TSNE analysis\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-5962648/v1/991ffcc1b538e31887121647.png"},{"id":79608186,"identity":"1b772000-9d8e-4c9a-ba56-735a07590d60","added_by":"auto","created_at":"2025-03-31 16:40:50","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3785405,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5962648/v1/bd79b31f-10a7-43b8-b643-cbe207d03760.pdf"},{"id":76190404,"identity":"eb8d1ee0-abb9-40fa-9b76-f0d5e4d1e265","added_by":"auto","created_at":"2025-02-13 09:30:31","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2103098,"visible":true,"origin":"","legend":"Supplementary information","description":"","filename":"Supplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-5962648/v1/62375b7e8d0cf3d945ff0b6d.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Band-gap-tunable carbon quantum dots for surface-enhanced Raman scattering","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCarbon quantum dots (CQDs) have attracted widespread attention as promising low-cost optical nanomaterials for diverse applications\u003csup\u003e\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. The photoluminescence (PL) features of CQDs have been exploited well that their PL emission spans from ultraviolet (UV) to near-infrared (NIR) region\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Several mechanisms involving quantum confinement effect, quantum size effect, molecular state, surface/edge state, and crosslinking-enhanced emission effect have been proposed to explain the diverse PL forms\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e, such as multicolor fluorescence, aggregation-induced emission, NIR-II emitting, photochromism, afterglow, and so on\u003csup\u003e\u003cspan additionalcitationids=\"CR7 CR8 CR9\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. The PL process is also accompanied by a certain degree of non-radiative relaxation and oxidative effect, which expands the potential of CQDs for photothermal and photodynamic therapy\u003csup\u003e\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Therefore, a more detailed understanding of the photophysical properties will substantially pave the way for versatile CQDs in practical applications.\u003c/p\u003e \u003cp\u003eRecently, nonmetallic materials have developed as emerging candidates for surface-enhanced Raman scattering (SERS) analysis\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Compared to commonly noble metal-based SERS substrates, nonmetallic materials are more economic, biocompatible, stable and SERS-repeatable\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. Nonmetallic SERS substrates can also reduce the Raman shift fluctuations and metal-catalyzed side reactions caused by strong metal-adsorbate interaction\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Diverse novel materials have been proven to possess SERS capabilities, e.g., black phosphorus, transition metal dichalcogenides/oxides/tellurides/nitrides, metal-organic frameworks, nanostructured organic semiconductor, amorphous TiO\u003csub\u003e2\u003c/sub\u003e nanosheets, some of which even demonstrate the Raman enhancement activities comparable to noble metal nanostructures\u003csup\u003e\u003cspan additionalcitationids=\"CR19 CR20 CR21 CR22 CR23 CR24\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. The SERS origin can be largely ascribed to the interfacial charge transfer (CT) process between nonmetallic substrate and analyte, an explanation of chemical mechanism (CM) theory\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. The SERS potential of carbon-based materials has also been explored, offering more economical choices for SERS sensing\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. The SERS activities of carbon materials are closely associated with their intrinsic structural characteristics. Graphene gives the first evidence whose sp\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e-hybridized structure can provide a Raman enhancement factor (EF) of about 10\u003csup\u003e1 27\u003c/sup\u003e. After that, Zhang \u003cem\u003eet al.\u003c/em\u003e developed a surfactant-free growth method for fabrication of graphdiyne hollow microspheres with both sp\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e and sp hybridized carbon atoms, achieving the EF value of about 3.7\u0026times;10\u003csup\u003e7 28\u003c/sup\u003e. Chen \u003cem\u003eet al.\u003c/em\u003e also push out a porous carbon nanowire array with high EF value (~\u0026thinsp;10\u003csup\u003e6\u003c/sup\u003e), reproducibility, durability and biocompatibility\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. The SERS effects of carbon nanostructures with quantum-scale sizes have currently only been observed in graphene quantum dots (GQDs), while their EFs are no more than 5\u0026times;10\u003csup\u003e2 30, 31\u003c/sup\u003e. Heteroatom doping may be a valid strategy to improve the SERS capability GQDs, like nitrogen-doping, achieving a higher EF value of 3.2\u0026times;10\u003csup\u003e3 32\u003c/sup\u003e. CQDs are undoubtedly more cost-effective choices compared to GQDs, however, their SERS activities are barely reported.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eHerein, we offer a facile strategy for band-gap tuning of CQDs using low-cost metal-phenolic network (MPN) as the precursor, a classic metal coordination structure enabling the band-gap control of MPN-CQDs by implantation of various metal atoms (Fe, Cu, Mn, Zn, Cr, etc.). Metal-doping significantly reduces the Femi level of CQDs about 1 eV, which affords more efficient CT process between MPN-CQDs and analyte, reaching the maximum Raman enhancement of 3.2\u0026times;10\u003csup\u003e4\u003c/sup\u003e. For detailed CM explanation, different metal atoms implantation in MPN-CQDs causes subtle fluctuations in Femi levels from \u0026minus;\u0026thinsp;5.01 to -5.30 eV, leading to diverse CT pathways accompanied by aeolotropic SERS generation. As a kind of nonmetallic SERS substrates, MPN-CQDs also exhibit excellent reproducibility and stability, indicating the promising potential for SERS sensing. As a proof-of-concept, we finally utilize MPN-CQDs as the sensitive SERS substrate for hemoglobin detection, reaching the limited of detection of 0.1 mg/mL. We further accomplish the molecular structural discrimination of seven derivatives of hemoglobin assisted by machine learning algorithm with the accuracy up to 100%, unveiling a new route for the diagnosis of blood-related diseases.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eSynthesis and characterization of MPN-CQDs\u003c/h2\u003e \u003cp\u003eMPN offers a facile and low-cost metal-doping strategy for MPN-CQDs, the latter is then synthesized through one-pot hydrothermal method, where the energy level control of MPN-CQDs can be effortlessly realized by adjusting the doped metal atoms within MPN (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e illustrates the characterization of the typical MPN-CQDs with Fe-doping (FeP-CQDs). The transmission electron microscopy (TEM) image reveals the quasi-spherical nanostructures of FeP-CQDs with an average diameter of 3.2 nm (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). \u0026zwnj;FeP-CQDs are partially aggregated, possibly due to the phenolic residues on their surface. In the high-resolution TEM (HRTEM) image, we can also notice the lattice fringe of 0.21 nm corresponding to the (100) crystal plane of graphitic carbon (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB)\u003csup\u003e3\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. The UV-vis absorbance spectrum of Fe-MPN precursor displays a typical broad ligand-to-metal charge transfer (LMCT) band at approximately 575 nm assigned to the metal-phenolic coordination\u003csup\u003e3\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e, which then disappears after carbonization, and a characteristic CQDs peak at ~\u0026thinsp;350 nm emerges (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Raman analysis also perceives the chemical composition alteration, where the Raman bands (1350 and 1485 cm\u003csup\u003e\u0026minus;1\u003c/sup\u003e) of Fe-MPN representing the skeletal vibrations of the benzene ring are replaced as the D (1360 cm\u003csup\u003e\u0026minus;1\u003c/sup\u003e) and G (1565 cm\u003csup\u003e\u0026minus;1\u003c/sup\u003e) bands in the Raman spectrum of FeP-CQDs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD)\u003csup\u003e3\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Fourier transformed infrared (FTIR) spectrum further reveals the chemical structure of FeP-CQDs, where the absorption peaks at 3440, 2925, 1642, 1390, 1080 and 598 cm⁻\u0026sup1; correspond to O-H, C-H, C\u0026thinsp;=\u0026thinsp;C/C\u0026thinsp;=\u0026thinsp;O, CH\u003csub\u003e2\u003c/sub\u003e/CH\u003csub\u003e3\u003c/sub\u003e/O-H, C-C/C-O-C, and Fe-O vibrations, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE)\u003csup\u003e3\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, 3\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. X-ray photoelectron spectroscopy (XPS) further investigated the precise elemental composition of the FeP-CQDs, like C, O, and Fe elements (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF). The high-resolution C 1s spectrum can be deconvoluted into three peaks at 284.8, 286.5, and 288.3 eV, corresponding to C\u0026thinsp;=\u0026thinsp;C, C-O, and C\u0026thinsp;=\u0026thinsp;O groups, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG)\u003csup\u003e3\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. The peaks at 531.5, 533.9, and 535.5 eV in O 1s spectrum are assigned to C\u0026thinsp;=\u0026thinsp;O, C-O, and -COOH groups, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH)\u003csup\u003e3\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. In the high-resolution Fe 2p spectrum, three fit peaks at 709.6, 713.0, and 724.0 eV corresponding to Fe 2p (II) 3/2, Fe 2p (III) 3/2, and Fe 2p 1/2, respectively can also be noticed (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eI)\u003csup\u003e3\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. In addition, the zeta potential of FeP-CQDs is measured to be -30.3 mV (Supplementary Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e), indicating a good water dispersibility.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSERS performance of FeP-CQDs\u003c/h3\u003e\n\u003cp\u003eTo investigate the SERS properties of FeP-CQDs, methylene blue (MB) was used as the probe molecule for Raman analysis under 633 nm excitation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). The concentration-dependent SERS spectra of MB dye deposited onto FeP-CQDs are illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB, where the SERS signals gradually decrease with the declining dye concentration; and the major SERS peaks are still observed at the concentration of 5 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e M. Theoretical limited of detection (LOD) value was then calculated from fitting curves of typical SERS peaks (1624, 1398 and 770 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), obtaining a minimum LOD value to be 3.8 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e M (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). We also calculated the Raman EFs of these bands at various dye concentrations, as displayed in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD. The maximum EF value is evaluated to be 5.5 \u0026times; 10\u003csup\u003e4\u003c/sup\u003e at 770 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, which is about 20\u0026ndash;100 times higher than the data in previous reports on GQDs\u003csup\u003e\u003cspan additionalcitationids=\"CR31\" citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE shows the SERS spectral map of MB dye on FeP-CQDs acquired from 30 random spectral lines, where the relative standard deviation (RSD) values of the three bands at 1624, 1398 and 770 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e are calculated to be 1.21%, 1.65%, and 1.62%, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF), indicating the outstanding SERS reproducibility of FeP-CQDs. Moreover, SERS maps were plotted using these Raman bands, which depicted the boundary of the dye sample very well (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG), indicating a good SERS uniformity. For SERS stability study, a long-term Raman monitoring was performed. As demonstrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH, the SERS intensity of MB at 1624 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e reduces by only 6% after FeP-CQDs are stored at 4\u0026deg;C for 40 days, and after a 300-day observation, the SERS signal at 1624 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e remains at 77%. The superior SERS stability of FeP-CQDs highlights their considerable potential for real-world applications. To confirm the versality of FeP-CQDs, SERS experiments were finally carried out on diverse dye molecules, such as malachite green (MG), crystal violet (CV), rhodamine 6G (R6G), rhodamine B (RhB), IR-820, and IR-780. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI shows the mean SERS spectra of these molecules at 5 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;5\u003c/sup\u003e M after deposited on FeP-CQDs, where intense and plentiful molecular fingerprint information can be noticed.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eSERS mechanism of FeP-CQDs\u003c/h3\u003e\n\u003cp\u003eIn order to elucidate the underlying SERS mechanism, a SERS comparison experiment was carried out between FeP-CQDs and polyphenolic CQDs without metal doping (P-CQDs). As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, plentiful Raman fingerprint information can be obviously noticed after MB molecules (5 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;5\u003c/sup\u003e M) deposited on FeP-CQDs under 633 nm excitation. In marked contrast, P-CQDs only exhibits negligible Raman enhancement effect on MB. To explain the strong SERS effect of FeP-CQDs, we first evaluated the interaction of MB molecules with FeP-CQDs using UV-vis spectroscopy. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB illustrates the characteristic absorption peak of MB at 662 nm with a shoulder at 614 nm, corresponding to the n-π* transition of MB molecule. Upon interaction with FeP-CQDs, an evident red-shift of these peaks occurs, indicating a strong interfacial interaction between MB and FeP-CQDs. Moreover, the ultraviolet photoelectron spectroscopy (UPS) was performed to measure the Fermi levels (E\u003csub\u003ef\u003c/sub\u003e) of CQDs, which were calculated to be -4.07 and \u0026minus;\u0026thinsp;5.25 eV for P-CQDs and FeP-CQDs, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003eThen the energy level diagrams and charge transfer processes between MB and CQD-SERS sensors are plotted in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD. The highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) levels of MB are at -6.26 and \u0026minus;\u0026thinsp;4.55 eV, respectively. Charge transfer occurs when the contact between the molecule and substrate reaches equilibrium, resulting in rearrangement of the band structure\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. According to Feynman's single-photon Raman scattering principle, the Raman scattering process involves three steps, each of which can be enhanced to amplify the Raman signals\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e: (1) excitation of ground state electron by incident light, (2) coupling of excited electron with phonon, and (3) emission of scattered light as electron relax back to the ground state. In MB/P-CQDs system, the Fermi level of P-CQDs (-4.07 eV) mismatches with the HOMO level of MB (-6.26 eV), indicating an inefficient ground-state charge transfer. However, the Fermi level of FeP-CQDs (-5.25 eV) is located much close to the HOMO level of MB (-6.26 eV). Due to the ground-state charge transfer process, the electrons near HOMO in MB/FeP-CQDs system show more possibility to contribute the Raman scattering process of MB, leading to the enhancement of electron-phonon coupling (second step of Feynman process)\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. The magnified Raman signals originate from photo-induced charge transfer (PICT) process involves the charge transfer from MB\u0026rsquo; HOMO level to FeP-CQDs\u0026rsquo; Fermi level (\u0026micro;\u003csub\u003ei\u0026minus;CT\u003c/sub\u003e) and from E\u003csub\u003ef(FeP\u0026minus;CQDs)\u003c/sub\u003e to LUMO\u003csub\u003e(MB)\u003c/sub\u003e (\u0026micro;\u003csub\u003ek\u0026minus;CT\u003c/sub\u003e). PICT between MB and FeP-CQDs may also amplify the molecular polarizability tensor to enhance Raman scattering of molecules according to Herzberg-Teller theory\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. The charge transfer across the interface between FeP-CQDs and MB was confirmed by density functional theory (DFT) calculation. As difficulty in explaining the nanostructures of CQDs prepared from hydrothermal method but retaining the precursor residues on their surface, we adopted Fe-MPN as the structural model of FeP-CQDs for DFT calculation. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE exhibits the projected density of states (PDOS) of MB molecules before and after adsorption on SERS substrate, where the electronic structure of MB significantly changes after adsorption, implying the charge of MB is redistributed after interacting with the SERS substrate. The charge density difference map also confirms the redistribution of the electron cloud around MB after interfacial charge transfer (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF). Bader charge analysis shows that electrons are transferred from the SERS substrate to probe molecule with a transfer amount of 0.150 eV.\u003c/p\u003e \u003cp\u003eMolecular resonance also contributes to the outstanding SERS effect of FeP-CQDs on MB (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), as the excitation energy (633 nm, 1.96 eV) matches well with the HOMO-LUMO gap of MB (1.71 eV). We further compared the SERS performance of FeP-CQDs on MB under three traditional excitation lasers (514.5, 633, and 785 nm). As demonstrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG, the SERS signals of MB triggered by 785 nm is obviously weaker than that by 633 nm. It can be deduced that the energy of the 785 nm laser (1.58 eV) is insufficient to induce molecular resonance of MB but PICT occurs in MB/FeP-CQDs system. In marked contrast, almost no Raman enhancement effect on MB is noted under 514.5 nm laser irradiation, probably owing to its energy (2.41 eV) mismatches the energy for molecular resonance of MB and PICT pathways.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eTunable SERS activities of MPN-CQDs and Raman enhancement mechanism\u003c/h3\u003e\n\u003cp\u003eWe have proven that metal coordination in MPN precursor could remarkably reduce the Fermi level of CQDs (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), which provides a facile strategy for band-gap control of MPN-CQDs. Besides FeP-CQDs, we also prepared other four kinds of MPN-CQDs doped with Cu, Cr, Zn, and Mn elements, namely: CuP-CQDs, CrP-CQDs, ZnP-CQDs, and MnP-CQDs, respectively. Above all display excellent SERS activities and repeatabilities to MB dye with the experimental LOD value of 5 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e M (Supplementary Figs. S2-3), further verifying the applicability of metal-doping strategy for development of CQDs-based SERS nanosensors. Then, a comparison among these MPN-CQDs was conducted. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA shows the SERS spectra of MB molecules at 5 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;5\u003c/sup\u003e M deposited on various MPN-CQDs irradiated by 633 nm laser. The intensity of the SERS band at 1624 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e was further selected for quantitative analysis, which revealed a fluctuation in SERS intensity associated with metal-doping. The SERS intensities are ranked from highest to lowest as follows: CuP-CQDs\u0026thinsp;\u0026gt;\u0026thinsp;FeP-CQDs\u0026thinsp;\u0026gt;\u0026thinsp;CrP-CQDs\u0026thinsp;\u0026gt;\u0026thinsp;ZnP-CQDs\u0026thinsp;\u0026gt;\u0026thinsp;MnP-CQDs. The SERS effects of MPN-CQDs doped with various metals under 785 laser irradiation illustrate similar trend (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB, Supplementary Fig. S4).\u003c/p\u003e \u003cp\u003eFor explanation of the SERS mechanism of MPN-CQDs regarding to metal-doping, the interaction of MPN-CQDs with MB was first evaluated by the UV-vis absorbance spectroscopy. As displayed in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC, obvious red-shift of the characteristic absorbance bands of MB emerges in these spectral lines, which provides the preliminary evidence of the interfacial interaction between MPN-CQDs and MB. The Fermi levels of CuP-CQDs, CrP-CQDs, ZnP-CQDs and MnP-CQDs were further measured by UPS analysis, which were calculated to be -5.30, -5.08, -5.04 and \u0026minus;\u0026thinsp;5.01 eV, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). Therefore, we can reasonably infer that the discrepant SERS activity of MPN-CQDs is closely related to their tunable band-gaps. The charge transfer processes between MB and MPN-CQDs with different Fermi levels are finally plotted in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE. PICT in MB/MPN-CQDs system includes charge transfer from MB\u0026rsquo; HOMO level to MPN-CQDs\u0026rsquo; Fermi level and from the Fermi level of MPN-CQDs to MB\u0026rsquo; LUMO level. In addition, the molecular resonance at 633 nm laser irradiation also amplify the cross-section of MB, leading to better SERS signals compared to that under 785 nm laser. The Fermi levels of five MPN-CQDs are ranked as: CuP-CQDs\u0026thinsp;\u0026lt;\u0026thinsp;FeP-CQDs\u0026thinsp;\u0026lt;\u0026thinsp;CrP-CQDs\u0026thinsp;\u0026lt;\u0026thinsp;ZnP-CQDs\u0026thinsp;\u0026lt;\u0026thinsp;MnP-CQDs, which are opposite to their SERS performances (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, B). Among them, the Fermi levels of CuP-CQDs (-5.30 eV) is closest to the HOMO level of MB (-6.26 eV), which may provide more electrons to participate in the ground-state charge transfer between the MB and CuP-CQDs, thereby achieving superior SERS performance. Moreover, the nearly symmetrical match of CuP-CQDs\u0026rsquo; Fermi level to MB\u0026rsquo; HOMO and LUMO levels further facilitates PICT in MB/CuP-CQDs system. These effects gradually subside as the Fermi level increases, reflecting decreasing Raman enhancement activity.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eSERS detection and discrimination of hemoglobin based on MPN-CQDs\u003c/h3\u003e\n\u003cp\u003eTo investigate the potential of MPN-CQDs for bio-SERS analysis, hemoglobin (Hb), a functional protein for transportation of oxygen in blood circulation, was adopted as the biomolecule model. The SERS effect of FeP-CQDs on bovine Hb is first shown in Supplementary Fig. S5. It can be observed that the Raman signals of bovine Hb (1 mg/mL) are obviously enhanced after deposited on FeP-CQDs, which can be ascribed to the efficient PICT process between biomolecule and substrate, as well as the molecular resonance of Hb under 514.5 nm laser. The typical fingerprint features of Hb can be clearly identified, such as 1639 and 1586 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (C\u003csub\u003eα\u003c/sub\u003eC\u003csub\u003em\u003c/sub\u003e asymmetrical stretching), 1539 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (C\u003csub\u003e\u003cem\u003eβ\u003c/em\u003e\u003c/sub\u003eC\u003csub\u003e\u003cem\u003eβ\u003c/em\u003e\u003c/sub\u003e stretching), 1372 and 1359 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (pyrrole half-ring symmetrical stretching), and 674 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (pyrrole deformation stretching), etc. (table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e)\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. We also notice a concentration-dependent SERS effect of FeP-CQDs on bovine Hb with favorable SERS repeatability, where the SERS signals of Hb at 0.1 mg/mL are still detected (Supplementary Fig. S6). We then tested the oxygenated hemoglobin (OxyHb) extracted from fresh human blood using FeP-CQDs-based SERS technique, which was demonstrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA. Similar to bovine Hb, the SERS signals of OxyHb induced by FeP-CQDs under 532 nm irradiation linearly decrease as the molecule concentration declines with the minimum detectable concentration of 0.1 mg/mL (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). The SERS spectral map of OxyHb on FeP-CQDs visually displays a good SERS reproducibility (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC) with the RSD value of 3.69% for the typical peak at 1586 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSERS that can supply molecular fingerprint information has newly been exploited as an spectromics tool to decode the heterogeneity in chemical composition of biosamples\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. The molecular structure of Hb determines its physiological and pathological functions. The structural abnormality of Hb is closely bound up with various hematologic diseases, like methemoglobinemia, sulfhemoglobinemia, sickle cell anemia, thalassemia, and hemoglobin C disease. To explore the potential of MPN-CQDs in revealing the heterogeneity of Hb-like proteins, we prepared four Hb variants\u0026mdash;OxyHb, deoxyhemoglobin (DeoxyHb), methemoglobin (MetHb), and fluoromethemoglobin (MetHbF)\u0026mdash;along with their corresponding prosthetic groups, protoporphyrin IX (PPIX) and ferroheme (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE). Then the SERS spectra of these molecules were acquired utilizing FeP-CQDs as the SERS sensor. As depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF, although the SERS spectra of ferroheme and PPIX display comparable spectral patterns, distinct differences are observed in their characteristic SERS peaks. For example, the SERS bands of PPIX at 738, 1332 and 1603 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e evidently red-shift to 755 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (pyrrole breathing stretching), 1369 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (pyrrole half-ring symmetrical stretching) and 1627 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (C\u003csub\u003eα\u003c/sub\u003eC\u003csub\u003em\u003c/sub\u003e asymmetrical stretching) in the SERS spectrum of ferroheme, respectively, indicating substantial structural modification of PPIX following Fe-complexation. The SERS spectral patterns of OxyHb and DeoxyHb show a closer alignment with that of ferroheme. We can also observe that the Raman bands sensitive to oxygenation state at 1585 and 1637 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e are significantly stronger in the spectral line of OxyHb than that of DeoxyHb\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. Furthermore, the peak at 1337 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, corresponding to Fe\u003csup\u003e2+\u003c/sup\u003e, significantly diminishes, whereas the band at 1372 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, associated with Fe\u003csup\u003e3+\u003c/sup\u003e, intensifies following the conversion of OxyHb to MetHb\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. In addition, a blue-shift is noticed in the SERS band of MetHb at 1637 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e after treatment with NaF, which may be a result of the Fe-F\u003csup\u003e\u0026minus;\u003c/sup\u003e interaction\u003csup\u003e47\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eUltimately, a machine learning algorithm based on principal component analysis and linear discriminant analysis (PCA-LDA) was employed to assist SERS in the discernment of the subtle variations among these Hb derivatives. The SERS data were first reduced in dimensionality using the PCA algorithm and then classified through a supervised LDA model. 80% of the SERS data was utilized for model training, and it was observed that the recognition accuracy of the LDA model was directly dependent on the number of principal component (PC) loadings, reaching 100% when the number of PCs exceeded 10 (Supplementary Fig. S7). The remaining 20% of the data was allocated as the test set. As shown in the receiver operating characteristic (ROC) curves (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eG), the areas under the ROC curve (AUCs) are calculated to be 1.00, 1.00, 1.00, 1.00, 1.00 and 0.95 for PPIX, ferroheme, OxyHb, DeoxyHb, MetHb and MetHbF, respectively, indicating excellent recognition performance of PCA-LDA algorithm. The confusion matrix was then drawn based on the structured and trained LDA model, revealing fantastic classification accuracy of 100% for all Hb derivatives (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eH). In addition, a scatter plot following t-distributed stochastic neighbor embedding (TSNE) analysis is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eI to visualize the classification quality, demonstrating clear separation among all Hb molecules. Therefore, MPN-CQDs-based SERS technique combined with machine learning algorithm supplies an available strategy for analyzing the heterogeneity of biomolecules, further contributing to the understanding of molecule-related diseases.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn summary, we have explored a novel optical characteristic of CQDs\u0026mdash;the Raman enhancement effect\u0026mdash;which fundamentally distinguishes them from their conventional photoluminescence properties. The inherent SERS activity of CQDs is originated from the charge transfer between the material and analyte. For systematical interpretation of the SERS mechanism of CQDs, we propose a type of MPN-derived CQDs with tunable band-gap which can be simply controlled by adjustment of coordinated metal in their polyphenol precursor. Metal-doping can dramatically reduce the Fermi level of polyphenol CQDs, and the precise Fermi level of MPN-CQDs can be further finely tuned by varying the type of doped metal. Five MPN-CQDs doped with Cu, Fe, Cr, Zn, and Mn, respectively, are prepared, exhibiting Fermi levels ranging from \u0026minus;\u0026thinsp;5.30 to -5.01 eV. The fluctuation in Fermi level may directly influence the interfacial photo-induced charge transfer between MPN-CQDs and analyte, leading to the variations in the SERS performance of MPN-CQDs. By optimizing the band-gap structure of MPN-CQDs, the maximum Raman enhancement factor of 5.5 \u0026times; 10\u003csup\u003e4\u003c/sup\u003e is achieved, along with excellent SERS reproducibility and stability. Therefore, band-gap-tunable MPN-CQDs may be an ideal nonmetallic SERS substrate for elucidation of chemical mechanism in SERS. For bio-SERS analysis, MPN-CQDs can also serve as a low-cost SERS platform for the detection and discrimination of hemoglobin derivatives with the assistance of machine learning algorithm. The discovery of tunable SERS performance in CQDs uncovers a novel route for insight into the Raman enhancement mechanisms of nonmetallic materials.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eSynthesis of MPN-CQDs\u003c/h2\u003e \u003cp\u003eMPN was fabricated via the simple self-assembly of polyphenols in the presence of metal ions, following a previously established protocol\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. In brief, various metal ion solutions (Fe\u003csup\u003e3+\u003c/sup\u003e, Cu\u003csup\u003e2+\u003c/sup\u003e, Cr\u003csup\u003e2+\u003c/sup\u003e, Zn\u003csup\u003e2+\u003c/sup\u003e, and Mn\u003csup\u003e2+\u003c/sup\u003e) at a concentration of 0.60 mM were added into the 0.24 mM tannic acid (TA) solution, and the pH was adjusted to greater than 7.4 using NaOH. The complexation products were formed within 1 h. Then, the mixtures were transferred to a poly(tetrafluoroethylene) (Teflon)-lined autoclave container and heated at 180\u0026deg;C for 6 h. The products were filtered using a 0.22 \u0026micro;m filter to remove large particles and purified using a dialysis bag (1000 Da) for 10 h.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCharacterization\u003c/h2\u003e \u003cp\u003eTEM analysis was carried out using a JEM-2100HR (JEOL, Tokyo, Japan) electron microscope operating at 200 kV. UPS spectra were measured using a photoelectron spectrometer (AXIS SUPRA, Shimadzu, Tokyo, Japan) equipped with angle-resolved capability. XPS spectra were recorded using an Escalab 250XI X-ray photoelectron spectrometer (Thermo Fisher Scientific, USA), with internal referencing to the C1s peak (sp\u0026sup2; hybridized carbon binding energy) at 284.1 eV. UV-Vis absorption spectra were obtained using a UV-Vis spectrophotometer (UV-6100, MAPADA). FTIR spectra were measured using a Nicolet iS50 spectrometer (Thermo Fisher Scientific, USA).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eSERS experiments\u003c/h2\u003e \u003cp\u003e2 \u0026micro;L MPN-CQDs was mixed with equal volume of dye solutions (MB, MG, CV, R6G, RhB, IR-820, and IR-780) at varying concentrations from 10\u003csup\u003e\u0026minus;\u0026thinsp;4\u003c/sup\u003e M to 10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e M. The mixture was then deposited onto Si wafer for SERS measurement. The Raman spectra were recorded using a confocal Raman spectrophotometer (InVia, Renishaw, UK) equipped with 514.5, 532, 633, and 785 nm lasers. A 20 \u0026times; objective was used to focus the laser beam and to collect the Raman signal.\u003c/p\u003e \u003cp\u003eFor bio-SERS analysis, bovine Hb molecules with different concentrations (0.1\u0026thinsp;~\u0026thinsp;8 mg/mL) were mixed with equal volume of FeP-CQDs and then detected under the Raman spectrophotometer with 514.5 nm laser. For classification of Hb derivatives, two prosthetic groups, PPIX and ferroheme, were purchased from Shanghai Macklin Biochemical Co., Ltd. OxyHb was extracted from fresh human blood and its derivatives (DeoxyHb, MetHb, and MetHbF) were prepared according to the established methods by Venkatesh \u003cem\u003eet al\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. PPIX, ferroheme, OxyHb, DeoxyHb, MetHb, and MetHbF after mixture with FeP-CQDs were analyzed using the Raman spectrophotometer with 532 nm laser. The PCA-LDA analysis of the SERS data of Hb was carried out in the Python software with the Scikit-Learn library (version 1.3.2). The raw Raman data were first baseline-corrected and randomly divide all spectral data into training and testing sets in an 8:2 ratio using train_test_split function. Subsequently, PCA was employed to reduce the dimensionality of the data, followed by LDA to further improve the model's accuracy, leading to the development of a PCA-LDA classification model. To visually assess the classification performance of the PCA-LDA model, the final model data was projected and visualized using t-SNE analysis, and scatter plots were generated to provide an intuitive representation of the classification outcomes among different hemoglobin derivatives.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eFirst principle calculation\u003c/h2\u003e \u003cp\u003eTheoretical calculation was performed using DFT as implemented in the Vienna Ab initio Simulation Package (VASP). The exchange-correlation potential was described using the generalized gradient approximation of Perdew-Burke-Ernzerhof (GGA-PBE). The projector augmented-wave (PAW) method was employed to treat interactions between ion cores and valence electrons. The plane-wave cutoff energy was fixed to 450 eV. Given structural models were relaxed until the Hellmann\u0026ndash;Feynman forces smaller than \u0026minus;\u0026thinsp;0.02 eV/\u0026Aring; and the change in energy smaller than 10\u003csup\u003e\u0026minus;\u0026thinsp;5\u003c/sup\u003e eV was attained. Grimme\u0026rsquo;s DFT-D3 methodology was used to describe the dispersion interactions among all the atoms in adsorption models. The Г-centered Monkhorst-Packgrid k-points were set as 2 \u0026times; 2 \u0026times; 2 for structure optimization and static calculations, and 3 \u0026times; 3 \u0026times; 3 for DOS calculation, respectively.\u003c/p\u003e \u003c/div\u003e "},{"header":"Declarations","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eData availability\u003c/h2\u003e \u003cp\u003eThe data that support the findings of this study are available from the corresponding author upon request. Source data are provided with this paper.\u003c/p\u003e \u003c/div\u003e\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003cp\u003e \u003ch2\u003eAdditional information\u003c/h2\u003e \u003cp\u003eSupplementary information The online version contains supplementary material available at xxxxxx.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor contributions\u003c/h2\u003e \u003cp\u003eZ.L. conceived the study. Z. L., B.X., and Z.G. supervised the project. M.N., Z.J. and H.C. completed the experiments. Z.L. and M.N. designed the figures and drafted the manuscript. Z.L., M.N., Z.J., J.L and J.C. managed statistical analyses. K.X., Y.S., D.S. and B.Y. achieved the material characterization. Z.L., H.C. and Z.G. provided funding support. All authors discussed the results, provided critical feedback, and contributed to the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThis work was supported by the National Natural Science Foundation of China (62175071 and 32071399), the Guangdong Basic and Applied Basic Research Foundation (2024A1515011675 and 2021A1515110265), the Science and Technology Program of Guangzhou (2024A04J4558).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eDordevic L, Arcudi F, Cacioppo M, Prato M (2022) A multifunctional chemical toolbox to engineer carbon dots for biomedical and energy applications. Nat Nanotechnol 17:112\u0026ndash;130\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrasley AT, Li E, Galeana JM, Bulumulla C, Beyene AG, Demirer GS (2024) Carbon nanomaterial fluorescent probes and their biological applications. Chem Rev 124:3085\u0026ndash;3185\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Lu S (2024) Lasing of carbon dots: Chemical design, mechanisms, and bright future. Chem 10:134\u0026ndash;171\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang B, Lu S (2022) The light of carbon dots: From mechanism to applications. Matter 5:110\u0026ndash;149\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang Z et al (2023) Zero-dimensional carbon nanomaterials for fluorescent sensing and imaging. Chem Rev 123:11047\u0026ndash;11136\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang L-Y, Zhou Y-C, Zhang S-F, Shao H-C, Liang Y-C (2024) Time division colorful multiplexing based on carbon nanodots with modifiable colors and lifetimes. Nano Lett 24:8418\u0026ndash;8426\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang H et al (2019) Hydrophobic carbon dots with blue dispersed emission and red aggregation-induced emission. Nat Commun 10:1789\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang M, Han Y, Bianco A, Ji D-K (2024) Recent progress on second near-infrared emitting carbon dots in biomedicine. ACS Nano 18:11560\u0026ndash;11572\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng G-S et al (2024) Photooxidation triggered ultralong afterglow in carbon nanodots. Nat Commun 15:2365\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo Z et al (2024) Multi-stimuli-responsive carbon dots with intrinsic photochromism and in situ radical afterglow. Adv Mater 36:2409361\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang H et al (2024) Carbon dots with integrated photothermal antibacterial and heat-enhanced antioxidant properties for diabetic wound healing. Small 20:2403160\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X et al (2024) Donor-acceptor type supra-carbon-dots with long lifetime photogenerated radicals boosting tumor photodynamic therapy. Angew Chem Int Ed 63:e202410522\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu H et al (2023) Afterglow photodynamic therapy based on carbon dots embedded silica nanoparticles for nondestructive teeth whitening. ACS Nano 17:21195\u0026ndash;21205\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGu Y et al (2023) Recent progress on noble-free substrates for surface-enhanced Raman spectroscopy analysis. Coord Chem Rev 497:215425\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y et al (2024) Phase-engineered transition metal dichalcogenides for highly efficient surface-enhanced Raman scattering. Nano Lett 24:14293\u0026ndash;14301\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang S et al (2024) Beyond the charge transfer mechanism for 2D materials-assisted surface enhanced Raman scattering. Anal Chem 96:9917\u0026ndash;9926\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiang C et al (2022) Band structure engineering within two-dimensional borocarbonitride nanosheets for surface-enhanced Raman scattering. Nano Lett 22:6590\u0026ndash;6598\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin J et al (2015) Enhanced raman scattering on in-plane anisotropic layered materials. J Am Chem Soc 137:15511\u0026ndash;15517\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun H, Cong S, Zheng Z, Wang Z, Chen Z, Zhao Z (2019) Metal-organic frameworks as surface enhanced Raman scattering substrates with high tailorability. J Am Chem Soc 141:870\u0026ndash;878\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong X et al (2024) Two-orders-of-magnitude enhancement of SERS activity via a simple surface engineering of quasi-metal single-crystal frameworks. Nano Lett 24:11683\u0026ndash;11689\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang X et al (2024) Exploring and engineering 2D transition metal dichalcogenides toward ultimate SERS performance. Adv Mater 36:e2312348\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y et al (2021) Lamellar hafnium ditelluride as an ultrasensitive surface-enhanced Raman scattering platform for label-free detection of uric acid. Photon Res 9:1039\u0026ndash;1047\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuan H et al (2021) General molten-salt route to three-dimensional porous transition metal nitrides as sensitive and stable Raman substrates. Nat Commun 12:1376\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYilmaz M et al (2017) Nanostructured organic semiconductor films for molecular detection with surface-enhanced Raman spectroscopy. Nat Mater 16:918\u0026ndash;924\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLyu X et al (2024) Molecularly confined topochemical transformation of MXene enables ultrathin amorphous metal-oxide nanosheets. ACS Nano 18:2219\u0026ndash;2230\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiang X et al (2021) Carbon-based SERS biosensor: From substrate design to sensing and bioapplication. NPG Asia Mater 13:8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLing X et al (2010) Can graphene be used as a substrate for Raman enhancement? Nano Lett 10:553\u0026ndash;561\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang L, Yi W, Li J, Wei G, Xi G, Mao L (2023) Surfactant-free interfacial growth of graphdiyne hollow microspheres and the mechanistic origin of their SERS activity. Nat Commun 14:6318\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen N et al (2020) Porous carbon nanowire array for surface-enhanced Raman spectroscopy. Nat Commun 11:4772\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCheng H, Zhao Y, Fan Y, Xie X, Qu L, Shi G (2012) Graphene-quantum-dot assembled nanotubes: A new platform for efficient Raman enhancement. ACS Nano 6:2237\u0026ndash;2244\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu D et al (2018) Raman enhancement on ultra-clean graphene quantum dots produced by quasi-equilibrium plasma-enhanced chemical vapor deposition. Nat Commun 9:193\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDas R, Parveen S, Bora A, Giri PK (2020) Origin of high photoluminescence yield and high SERS sensitivity of nitrogen-doped graphene quantum dots. Carbon 160:273\u0026ndash;286\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumar VB, Mirsky SK, Shaked NT, Gazit E (2024) High High quantum yield amino acid carbon quantum dots with unparalleled refractive index. ACS Nano 18:2421\u0026ndash;2433\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen J et al (2022) Assembly of bioactive nanoparticles via metal-phenolic complexation. Adv Mater 34:2108624\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRahim MA et al (2017) Rust-mediated continuous assembly of metal-phenolic networks. Adv Mater 29:1606717\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu H et al (2022) Carbon dots with intrinsic bioactivities for photothermal optical coherence tomography, tumor-specific therapy and postoperative wound management. Adv Healthc Mater 11:2101448\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi M et al (2024) Dynamic visualization monitoring of cell membrane damage using polarity-responsive amphiphilic carbon dots. Chem Eng J 482:149038\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo D et al (2022) Photocatalytic Pt(IV)-coordinated carbon dots for precision tumor therapy. Adv Sci 9:e2205106\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Y et al (2022) Ultrasmall Fe-doped carbon dots nanozymes for photoenhanced antibacterial therapy and wound healing. Bioact Mater 12:246\u0026ndash;256\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiao X et al (2019) Graphene quantum dots wrapped gold nanoparticles with integrated enhancement mechanisms as sensitive and homogeneous substrates for surface-enhanced Raman spectroscopy. Anal Chem 91:7295\u0026ndash;7303\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLing X, Moura LG, Pimenta MA, Zhang J (2012) Charge-transfer mechanism in graphene-enhanced Raman scattering. J Phys Chem C 116:25112\u0026ndash;25118\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa H et al (2024) Surface-enhanced Raman spectroscopy: Current understanding, challenges, and opportunities. ACS Nano 18:14000\u0026ndash;14019\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu S, Smith KM, Spiro TG (1996) Assignment of protoheme resonance Raman spectrum by heme labeling in myoglobin. J Am Chem Soc 118:12638\u0026ndash;12646\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCutshaw G, Uthaman S, Hassan N, Kothadiya S, Wen X, Bardhan R (2023) The emerging role of Raman spectroscopy as an omics approach for metabolic profiling and biomarker detection toward precision medicine. Chem Rev 123:8297\u0026ndash;8346\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRusciano G (2010) Experimental analysis of Hb oxy\u0026ndash;deoxy transition in single optically stretched red blood cells. Phys Med 26:233\u0026ndash;239\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarzec KM, Rygula A, Wood BR, Chlopicki S, Baranska M (2014) High-resolution Raman imaging reveals spatial location of heme oxidation sites in single red blood cells of dried smears. J Raman Spectrosc 46:76\u0026ndash;83\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAsher SA, Vickery LE, Schuster TM, Sauer K (1977) Resonance Raman spectra of methemoglobin derivatives. Selective enhancement of axial ligand vibrations and lack of an effect of inositol hexaphosphate. Biochemistry 16:5849\u0026ndash;5856\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEjima H et al (2013) One-step assembly of coordination complexes for versatile film and particle engineering. Science 341:154\u0026ndash;157\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVenkatesh B, Ramasamy S, Mylrajan M, Asokan R, Manoharan PT, Rifkind M (1999) Fourier transform Raman approach to structural correlation in hemoglobin derivatives. Spectrochim Acta Mol Biomol Spectrosc 55:1691\u0026ndash;1697\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5962648/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5962648/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCarbon quantum dots (CQDs) with photoluminescence properties have been widely recognized, however, their inherent surface-enhanced Raman scattering (SERS) activity is rare reported. Herein, we propose a type of band-gap-tunable CQDs derived from metal-phenolic network (MPN) that exhibits excellent SERS performance. MPN-CQDs may be the ideal nonmetallic SERS substrates to accurately elucidate the chemical mechanism (CM) due to their simply controllable bandgap structure. By screening the doped metal elements of MPN, MPN-CQDs can realize the optimal SERS effect with the maximum Raman enhancement factor of 5.5\u0026times;10\u003csup\u003e4\u003c/sup\u003e, also exhibit outstanding SERS reproducibility and stability. We then systematically disclose the interfacial photo-induced charge transfer process and corresponding migration pathways between band-gap-tunable MPN-CQDs and analyte. This class of nonmetallic SERS substrates is finally applied for detection of hemoglobin with high sensitivity; further combined with machine learning algorithm, we have successfully achieved precise identification of the heterogeneity of hemoglobin. This is the first evidence for the tunable SERS performance in CQDs, which also offers the facile avenue for in-depth understanding of CM in nonmetallic materials.\u003c/p\u003e","manuscriptTitle":"Band-gap-tunable carbon quantum dots for surface-enhanced Raman scattering","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-02-13 09:30:19","doi":"10.21203/rs.3.rs-5962648/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"04fefc10-23d4-42fd-84f4-20c780c6513a","owner":[],"postedDate":"February 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":44229199,"name":"Physical sciences/Optics and photonics/Optical techniques/Optical spectroscopy/Raman spectroscopy"},{"id":44229200,"name":"Physical sciences/Materials science/Nanoscale materials/Quantum dots"},{"id":44229201,"name":"Physical sciences/Materials science/Materials for optics/Nanophotonics and plasmonics"}],"tags":[],"updatedAt":"2025-03-31T16:40:43+00:00","versionOfRecord":[],"versionCreatedAt":"2025-02-13 09:30:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5962648","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5962648","identity":"rs-5962648","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}