Template for Electronic Submission to ACS Journals · Web viewThe perturbation of vibrational spectra by electromagnetic fields has been known for many years, which is predicted to

Template for Electronic Submission to ACS Journals

Frequency Shifts in SERS-Based Immunoassays: Mechanistic Insights and Application in Protein Carbonylation Detection

Hao Ma,1 Songlin Liu,1 Naiqing Zheng,2 Yawen Liu,1 Xiao Xia Han,1 Chengyan He,3 Hui Lu4* and Bing Zhao1*

1. State Key Laboratory of Supramolecular Structure and Materials, Jilin University, Changchun 130012, P. R. China.

2. National Engineering Laboratory for AIDS Vaccine, School of Life Science, Jilin University, Changchun 130012, P. R. China

3. China-Japan Union Hospital of Jilin University, Changchun 130033, P. R. China.

4. School of Biological Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester M13 9PT, UK.

ABSTRACT: Frequency-shift based surface-enhanced Raman spectroscopy (SERS) has exhibited great potential applications in bioanalytical chemistry and biomedicine in recent years. The basis and the crucial factors determining frequency shifts are, however, still unclear. Herein, we have systematically investigated how solvents, antigens and antibodies affect the band shifts in SERS-based immunoassays. By applying the charge transfer theory together with Stark effect and time-dependent density functional theory (TDDFT) calculation, mechanistic insights into the frequency shifts in immunoreactions is proposed and discussed in details. Accordingly, the experimental condition is further optimized and is successfully applied for the first time to detect carbonylated proteins, promising diagnostic biomarkers for human diseases. This study provides theoretical guidance for designing SERS frequency shift-based immunoassays and paves a new avenue for the further applications of the strategy in clinical diagnosis.

PAGE 2

Surface-enhanced Raman spectroscopy (SERS) has been proved to be a powerful tool for chemical analysis, material science and biomedicine in recent years.1-6 As a novel readout method with excellent spectral reproducibility, SERS frequency shifts have been applied into immunoassays since 2012.7 Recent years, the fast growing research field allows widespread applications in the detection of cancer biomarkers,8, 9 DNA,10 miRNA11 and protein self-assembles.12 On the other hand, carbonylated proteins are promising biomarkers for human diseases (Alzhermer’s disease and diabetes et al.) 13, 14 and have been attracting the interest of numerous laboratories regarding detection and identification. A fast, highly sensitive and selective assay is of great importance for detection of these biomarkers in clinical diagnosis.

The key factors that control the frequency shifts have been explored, but the related mechanism is poorly understood. Olive et al. reported the vibrational frequency shifts in the antibody-reporter complexes.7 Their theory is based on harmonic oscillator model and it is demonstrated that this phenomenon is influenced by pH, hydrophobic interactions between proteins and pressure. Moreover, the changes in the polarizability of molecule-metal complexes were found to contribute to the frequency shifts.10, 15 Our previously studies suggested that hydrogen binding and enzymatic reaction may cause frequency shifts of the Raman reporters, in which the charge-transfer effect was considered as a dominant factor.16-18

We recently reported an approach for the detection of alpha-fetoprotein (AFP) by SERS frequency shifts of p-mercaptobenzoic acid (MBA).19 Here, the frequency shifts were for the first time found to be a comprehensive reflection of different vibrational modes of MBA, named as a ratio metric method. More recently, we found the frequency shifts are solvent dependent as demonstrated in the related study by Zhao et al.11 Moreover, the SERS frequencies suffer obvious shifts when the Raman reporter and antigen are fixed and a different antibody is introduced. Thus, all these factors may influence the frequency shifts, and a better understanding of the mechanism is of great importance for experimental design and improving the accuracy of SERS-based immunoassays.

In this study, we systematically investigated the factors (solvents, antigens and antibodies) that control the band shifts of a Raman reporter, MBA in SERS-based immunoassays. Together with the charge transfer theory, Stark effect and time-dependent density functional theory (TDDFT) calculation, the mechanism of the frequency shifts in immunoreactions is explored. Accordingly, the experimental condition is further optimized and is successfully applied for the first time to detect carbonylated proteins.

The analysis is based on aforementioned solvatochromic trends of Ag-MBA chip, a model system prepared via a self-assembled method.16 The experimental details are summarized in the supporting information. As seen in Figure 1, the SERS spectra of MBA are dominated by the peaks at 1075 cm-1, and 1584 cm-1, which can be assigned to in-plane ring breathing mode coupled with C-S vibration (1075 cm-1) and aromatic vibration (1584 cm-1) mode respectively.20 The band at 1075 cm-1 shifted to 1078.3 cm-1 in the most nonpolar solvent (CCl4), and to 1072.2 cm-1 in the most polar solvent (H2O). All the other bands were located between the above two wavenumbers when the Ag-MBA chip was exposed to the solvents of C2H5OH, MeCN, and DMSO. This tendency is attributed to the effective electric field created by the polarization of the solvent. Interestingly, the intensities of the bands at 998 and 1022 cm-1 ascribed to the in-plane breathing modes (b2 modes) reveal a similar solvatochromic effect, which increase with the polarization of the solvents. The consistent results from frequency shifts and band intensities suggested the mechanistic corrections between the two of them. Moreover, as can be seen in Figure 1C, D, we also observed same tendencies

Figure 1. Effect of solvents on the SERS spectra of the MBA. (A) SERS spectra of the MBA-Ag films in different solvents: a. CCl4, b. C2H5OH, c. MeCN, d. DMSO, e. H2O, f. No solvent (gas); (B) Zoomed-in image of the dotted line in (A); (C) Zoomed-in image of the intensity change of b2 mode (pink region) in (A); (D) SERS spectra of antiAFP-MBA-Ag exposure to different solvents (MeCN, DMSO, H2O); (E) SERS spectra of antiAFP conjugated with (HRP)- MBA-Ag exposure to different solvents (MeCN, DMSO, H2O).

for MBA-antiAFP with kinds of polar solvent: MeCN, DMSO, H2O as well, indicating the solvent effect is demonstrated in Ag-MBA-antibody system.

We have shown that the charge transfer (CT) process would selectively increase nontotally symmetric intensities (b1 mode), which is a resonance Raman-like process.21, 22 As well known, MBA can adsorb on silver surface through a Ag-S bond, and thus the symmetry of complex is C2v if hydrogen interactions were ignored. Noted that the CT dipole moment allowed excited states of the complex are A1 and B2 respectively. According to Herzberg-Teller selection rules proposed by Lombardi,23, 24 if a metal-to-molecule CT occurs, a normal mode of b2 (A1*B2) would be allowed. Hence, in the present system, the observed intensity changes of two bands (b2 modes) at 998 and 1022 cm-1 indicates that the metal-to-molecule CT process occurs in the Ag-MBA complexes. To further investigate this phenomenon, a TDDFT analysis was carried out to explore the solvent effect on the CT process. The calculation details are provided in the supporting information. Figure 2A shows the calculated electronic transition states of the Ag3-MBA complex with the hole-electron distributions, which is performed by using Multiwfn software.25 Obviously, there is a molecule-to-metal CT process in gas phase. When the solvents are introduced into the Ag3-MBA system, the electric fields created by the solvent polarization will disturb the electronic structure of the complex, inhibiting the CT of molecule-to-metal. In another word, the introduction of the solvents induced a metal-to-molecule CT process. Meanwhile, the energy gap between the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) is changed dramatically as illustrated in Figure S1. Because of the larger dipole moment of HOMO, the influence on the ground state of Ag3-MBA can be neglected, which is stabilized by the solvents. On the other hand, in the electron density point of view, a molecule-to-metal CT process should induce a decrease of electron density (ρ) in MBA, whereas an increase of electron density from metal-to-molecule CT process. Hence, compared with the MBA without solvents, the molecule with certain solvent contains higher ρ.

Moreover, in order to compare the CT contribution from different assemblies more directly and accurately, we applied the concept

Figure 2. The solvent effect on different assembles. (A) Hole-electron distribution for Ag3-MBA complex with consideration of the solvent effect. The blue and red isosurfaces represent the hole and electron distributions, respectively. The isosurface value is taken as 0.002; (B) ρCT and (C) frequency shifts of the four assemblies in different solvents of H2O, DMSO, MeCN, C2H5OH and CCl4.

of the ‘degree of CT (ρCT)’ proposed by Lombardi et al.23 Based on our previous work, (k) of a k-band can be expressed as following equation:

(1)

where Ik(CT) is the intensity of a band which is ascribed to SERS intensity by CT resonance. Also, it is crucial to choose a reference band without CT contribution to SERS intensity, the line of Ik(CT), which originated from SPR. Mostly, Ik(SPR) is very small or zero when the CT vibrations are nontotally symmtric. In this case, we selected the ring breathing A1 mode at 1075 cm-1 as the peak of I0(SPR), which is enhanced in the whole spectrum. As for Ik(CT), we selected the in-plane vibration band at 992 cm-1 (B2 mode) for investigating CT of metal-to-molecule. Thus, Eq. 1 can be expressed as follows:

(2)

Clearly, compared with the frequency shifts of the four assemblies shown in Figure 2C, a similar trend is observed. Therefore, it is deduced that there is an intrinsic correction between these two SERS transformations. The perturbation of vibrational spectra by electromagnetic fields has been known for many years, which is predicted to be promising to explain various results observed in SERS.26 The effect of an electric field on the frequency splitting of a molecule vibration (Stark effect) has been studied in different systems.27-30 As it has been well established by Boxer and coworkers,26 the expansions of field-induced Stark shift can be expressed as the following equation:

(3)

Here represents the difference in the dipole moment, and represents the polarizability difference. F is the electric field. Here, we use this theory to interpret the SERS transformation together with the general expression for the polarizability tensor.23, 28, 31, 32 In the present system Eq. 3 can be expressed as:

+

(4)

Briefly, two CT processes are involved in the present system: B tensor (molecule-to-metal CT process) and C tensor (metal-to-molecule CT process). Where is the electronic transition from HOMO to LUMO in MBA. is the electronic transition from HOMO to the Fermi level of silver. represents the electronic transition from Fermi level to LUMO. As discussed above, the molecule-to-metal CT is largely inhibited. The polarizability tensor is mainly affected by the metal-to-molecule CT process. Assuming the electric field is the same, this CT contribution is directly proportional to the frequency shift. Thus, it is reasonable for the observed similar trend of CT and the frequency shift. Furthermore, we observed a similar tendency in a Au-MBA system as discussed in Figure S2, which suggested that the metals affect more remarkably the ρCT than the electric field.

However, it is obvious that the solvent effect is not the only factor that influences the frequency shifts of Raman probe. The diversity of the antibodies and antigens will also affect to some extent the SERS band shifts. As shown in Figure 2, the antibodies (antiAFP and conjugated antiAFP) display different ρCT and the frequency shifts. Moreover, the similar results were observed when introducing different antigens (data not shown). To further explore these controlling factors, we designed a SERS-based imunoassay for carbonylated proteins. In the present study, there are two antigens (DNPH and DNP-BSA) and two antibodies (IgG and IgE). As illustrated in Figure 3A and Figure S3, when introducing DNPH into the proposed Ag-MBA-antibody chip, the frequency shifts of MBA with IgG and IgE are almost the same, indicating that such small molecule cannot induce deformation or change of the polarizability of MBA. In contrast, when the DNP-BSA bound into the Ag-MBA-IgG chip, an upshifted (2 cm-1) band was observed as referred to the case of DNPH. Therefore, we deduced that the BSA with high molecule weight may induce a deformation of MBA (Figure 3B). A similar frequency shift was observed for the characteristic peak located at 1111 cm-1.

Figure 3. Effect of antigen and antibody on the SERS spectra of the MBA. (A) SERS spectra for proposed chips with IgG and different antigens; (B) schematic diagram of the MBA deformation with different antigens; (C) the ρCT versus different antibodies; (D) DFT calculation of Ag3-MBA with consideration of the isotype effect.

To verify our hypothesis, the density functional theory was performed to identify the expected vibrations. All the calculations were under M062X/6-311+G*/LanL2DZ level. With the consideration of the isotope effect, we tried to figure out the effect of antigen weight to the vibrations. As shown in Figure 3D, we found that the attached antigen indeed can induce frequency shift up to 2 cm-1, which is consistent with the observed experimental result in Figure 3A. In addition, we found that different antibodies induced different frequency shifts when introducing DNP-BSA as well as the ρCT (Figure 3C). As for the IgG, the upshifts were attributed to the attractive hydrophobic interactions between IgG molecules, which thereby causing the relaxing of the deformed MBA and thus in turn change the polarizability and the ρCT. The reason why the SERS frequency in the IgE-DNP-BSA chip appeared a downshift is probably the lack of intermolecular interactions, which further induced deformation of the MBA. As a result, we find that the theory of Olive definitely makes sense. However, to a large extent, it depends on the type of antibodies and antigens.

Carbonylation of proteins is an irreversible oxidative damage, which is associated with a large number of diseases. Currently, most of the assays for detecting protein carbonylation involve deriving carbonyl groups with 2,4-dinitrophenylhydrazine(DNPH), which can subsequently be detected by the enzyme-linked immunosorbent assay (ELISA) or Western blotting. However, these methods usually lead to false-positive or false-negative results, because it is challenging to remove unreacted DNPH fully and DNPH can also act as an antigen. Moreover, the current and traditional assays for carbonylated protein detection always require large amount of protein samples with many washing steps or suffer bad quantitative ability.

As illustrated in Figure 4A-C, given the fact that the SERS method is simple and fast without the interference from DNP, it is expected to establish a novel analytical method for carbonylated proteins based on SERS frequency shifts. To this end, we optimized the experimental condition and introduced IgG as the capture antibody. As shown in Figure 4E, it was feasible to detect carbonylated proteins at the concentrations ranging from 0.18 to 9 nmol/mL. We found the detection curve became saturated at the concentration of 9 nmol/mL, a threshold value. According to the results, the proposed SERS-based method is more sensitive than other related assays with wider dynamic range (Table S2), thus it may be applied to analyze clinic samples.

Figure 4. Assays for protein carbonylation: (A) a colormetric approach, (B) Western blotting, and (C) the proposed SERS frequency-shift method. (D) DNP-BSA-dependent SERS spectra of the Ag-MBA-IgG chip: a. blank; b, 0.18 nmol/mL; c, 0.9 nmol/mL, d, 2.25 nmol/mL, e, 9 nmol/mL, f, 18 nmol/mL DNP-BSA; (E) frequency shifts of the peak at 1075 cm-1 versus the carbonyl content. Each data point represents an average of five measurements and the error bars indicate standard derivation.

In summary, we performed a systematic analysis of transformation of the Raman probe MBA in SERS-based immunoassays. By combination of the CT and Stark theories, we inferred that both of the metal-to-molecule and molecule-to-metal CT processes can induce frequency shifts by changing the polarizability of MBA. Moreover, with the aid of DFT theory, we demonstrated that the factors controlling frequency shifts are not only related to the solvent effect and CT but also involved in the isotype effect, intermolecular interaction and types of antigens and antibodies. Our study has clarified some fundamental questions of the immunoreaction induced SERS transformation, and paves a new avenue for the further applications of frequency shift-based immunoassays in clinical diagnosis.

ASSOCIATED CONTENT

Supporting Information

Detailed information of preparation of self-assemble chips; preparation and determination of the carbonyl proteins; computational details; energy gaps in different solvant; comparison of relevant assays; SERS spectra of the Au-MBA with different solvents and the Ag-MBA-IgE with different antigens; UV-Vis spectra of the DNP-BSA.

The Supporting Information is available free of charge on the ACS Publications website.

AUTHOR INFORMATION

Corresponding Author

* [email protected] (H.L.); [email protected] (B.Z.)

Author Contributions

The manuscript was written through contributions of all authors. / All authors have given approval to the final version of the manuscript.

NotesThe authors declare no competing financial interest.

ACKNOWLEDGMENT

This work was supported by the National Natural Science Foundation (Grant Nos. 81572082, 21773080, 21711540292 and 21773079) of P. R. China; Project for Science & Technology Development of Jilin Province (20180101201JC, 20190201215JC and 20190701003GH). It was also supported by the Royal Society (IE150835 to HL).

REFERENCES

1. Lane, L. A.; Qian, X.; Nie, S., SERS Nanoparticles in Medicine: From Label-Free Detection to Spectroscopic Tagging. Chem. Rev. 2015, 115 (19), 10489-10529.

2. Hu, G.; Feng, Z.; Han, D.; Li, J.; Jia, G.; Shi, J.; Li, C., Charge Transfer between Triphenyl Phosphine and Colloidal Silver: A SERS Study Combined with DFT Calculations. J. Phys. Chem. C 2007, 111, 6.

3. Hackler, R. A.; McAnally, M. O.; Schatz, G. C.; Stair, P. C.; Van Duyne, R. P., Identification of Dimeric Methylalumina Surface Species during Atomic Layer Deposition Using Operando Surface-Enhanced Raman Spectroscopy. J. Am. Chem. Soc. 2017, 139 (6), 2456-2463.

4. Han, X. X.; Jia, H. Y.; Wang, Y. F.; Lu, Z. C.; Wang, C. X.; Xu, W. Q.; Zhao, B.; Ozaki, Y., Analytical Technique for Label-Free Multi-Protein Detection Based on Western Blot and Surface-Enhanced Raman Scattering. Anal. Chem. 2008, 80, 6.

5. Zong, C.; Xu, M.; Xu, L. J.; Wei, T.; Ma, X.; Zheng, X. S.; Hu, R.; Ren, B., Surface-Enhanced Raman Spectroscopy for Bioanalysis: Reliability and Challenges. Chem. Rev. 2018, 118 (10), 4946-4980.

6. Cialla-May, D.; Zheng, X. S.; Weber, K.; Popp, J., Recent progress in surface-enhanced Raman spectroscopy for biological and biomedical applications: from cells to clinics. Chem. Soc. Rev. 2017, 46 (13), 3945-3961.

7. Kho, K. W.; Dinish, U. S.; Kumar, A.; Olivo, M., Frequency Shifts in SERS for Biosensing. ACS Nano 2012, 6, 11.

8. Tang, B.; Wang, J.; Hutchison, J. A.; Ma, L.; Zhang, N.; Guo, H.; Hu, Z.; Li, M.; Zhao, Y., Ultrasensitive, Multiplex Raman Frequency Shift Immunoassay of Liver Cancer Biomarkers in Physiological Media. ACS Nano 2016, 10 (1), 871-879.

9. Ma, H.; Sun, X.; Chen, L.; Han, X. X.; Zhao, B.; Lu, H.; He, C., Antibody-Free Discrimination of Protein Biomarkers in Human Serum Based on Surface-Enhanced Raman Spectroscopy. Anal Chem 2018, 90 (21), 12342-12346.

10. Guerrini, L.; Pazos, E.; Penas, C.; Vazquez, M. E.; Mascarenas, J. L.; Alvarez-Puebla, R. A., Highly sensitive SERS quantification of the oncogenic protein c-Jun in cellular extracts. J. Am. Chem. Soc. 2013, 135 (28), 10314-7.

11. Zhu, W. F.; Cheng, L. X.; Li, M.; Zuo, D.; Zhang, N.; Zhuang, H. J.; Xie, D.; Zeng, Q. D.; Hutchison, J. A.; Zhao, Y. L., Frequency Shift Raman-Based Sensing of Serum MicroRNAs for Early Diagnosis and Discrimination of Primary Liver Cancers. Anal. Chem. 2018, 90 (17), 10144-10151.

12. Zhu, W.; Wang, Y.; Xie, D.; Cheng, L.; Wang, P.; Zeng, Q.; Li, M.; Zhao, Y., In Situ Monitoring the Aggregation Dynamics of Amyloid-β Protein Aβ42 in Physiological Media via a Raman-Based Frequency Shift Method. ACS Appl. Bio Mater. 2018, 1 (3), 814-824.

13. Levine, R. L.; Garland, D.; Oliver, C. N.; Amici, A.; Climent, I.; Lenz, A.-G.; Ahn, B.-W.; Shaltiel, S.; Stadtman, E. R., Determination of Carbonyl Content in Oxidatively Modified Proteins. Methods Enzymol 1990, 186 (186), 15.

14. Dalle-Donne, I.; Rossi, R.; Giustarini, D.; Milzani, A.; Colombo, R., Protein carbonyl groups as biomarkers of oxidative stress. Clinica Chimica Acta 2003, 329 (1-2), 23-38.

15. Saikin, S. K.; Olivares-Amaya, R.; Rappoport, D.; Stopa, M.; Aspuru-Guzik, A., On the chemical bonding effects in the Raman response: Benzenethiol adsorbed on silver clusters. Phys. Chem. Chem. Phys. 2009, 11 (41).

16. Wang, Y.; Ji, W.; Sui, H.; Kitahama, Y.; Ruan, W.; Ozaki, Y.; Zhao, B., Exploring the Effect of Intermolecular H-Bonding: A Study on Charge-Transfer Contribution to Surface-Enhanced Raman Scattering of p-Mercaptobenzoic Acid. J. Phys. Chem. C 2014, 118 (19), 10191-10197.

17. Yu, Z.; Chen, L.; Park, Y.; Cong, Q.; Han, X.; Zhao, B.; Jung, Y. M., The mechanism of an enzymatic reaction-induced SERS transformation for the study of enzyme–molecule interfacial interactions. Phys. Chem. Chem. Phys. 2016, 18 (46), 31787-31795.

18. Wang, Y.; Yu, Z.; Ji, W.; Tanaka, Y.; Sui, H.; Zhao, B.; Ozaki, Y., Enantioselective Discrimination of Alcohols by Hydrogen Bonding: A SERS Study. Angew. Chem. Int. Edit. 2014, 53 (50), 13866-13870.

19. Ma, H.; Sun, X.; Chen, L.; Cheng, W.; Han, X. X.; Zhao, B.; He, C., Multiplex Immunochips for High-Accuracy Detection of AFP-L3% Based on Surface-Enhanced Raman Scattering: Implications for Early Liver Cancer Diagnosis. Anal. Chem. 2017, 89 (17), 8877-8883.

20. Smith, G.; Girardon, J.-S.; Paul, J.-F.; Berrier, E., Dynamics of a plasmon-activated p-mercaptobenzoic acid layer deposited over Au nanoparticles using time-resolved SERS. Physical Chemistry Chemical Physics 2016, 18 (29), 19567-19573.

21. Han, X. X.; Ji, W.; Zhao, B.; Ozaki, Y., Semiconductor-enhanced Raman scattering: active nanomaterials and applications. Nanoscale 2017, 9 (15), 4847-4861.

22. Ji, W.; Kitahama, Y.; Xue, X.; Zhao, B.; Ozaki, Y., Generation of Pronounced Resonance Profile of Charge-Transfer Contributions to Surface-Enhanced Raman Scattering. The Journal of Physical Chemistry C 2012, 116 (3), 2515-2520.

23. Lombardi, J. R.; Birke, R. L., A Unified Approach to Surface-Enhanced Raman Spectroscopy. The Journal of Physical Chemistry C 2008, 112, 12.

24. Lombardi, J. R.; Birke, R. L., Theory of Surface-Enhanced Raman Scattering in Semiconductors. The Journal of Physical Chemistry C 2014, 118 (20), 11120-11130.

25. Lu, T.; Chen, F., Multiwfn: A multifunctional wavefunction analyzer. Journal of Computational Chemistry 2012, 33 (5), 580-592.

26. Korzeniewski, C.; Randall B, S.; Pons, S., Field-Induced Infrared Absorption in Metal Surface Spectroscopy: The Electrochemical Stark Effect. The Journal of Physical Chemistry 1985, 89, 2.

27. Chattopadhyay, A.; Boxer, S. G., Vibrational Stark Effect Spectroscopy. J Am Chem Soc 1995, 117, 2.

28. Andrews, S. S.; Boxer, S. G., Vibrational Stark Effects of Nitriles II. Physical Origins of Stark Effects from Experiment and Perturbation Models. The Journal of Physical Chemistry A 2002, 106, 8.

29. Suydam, I. T.; Snow, C. D.; Pande, V. S.; Boxer, S. G., Electric Fields at the Active Site of an Enzyme: Direct Comparison of Experiment with Theory. Science 2006, 313, 5.

30. Fafaman, A. T.; Sigala, P. A.; Herschlag, D.; Boxer, S. G., Decomposition of Vibrational Shifts of Nitriles into Electrostatic and Hydrogen-Bonding Effects. J Am Chem Soc 2010, 2010 (132), 3.

31. Ma, H.; Wang, H.; Li, P.; Wang, X.; Han, X.; He, C.; Zhao, B., Interfacial Charge Transfer in TiO2/PTCA/Ag Revealed by Surface-Enhanced Raman Spectroscopy. The Journal of Physical Chemistry C 2018, 122 (27), 15208-15213.

32. Gieseking, R. L. M.; Lee, J.; Tallarida, N.; Apkarian, V. A.; Schatz, G. C., Bias-Dependent Chemical Enhancement and Nonclassical Stark Effect in Tip-Enhanced Raman Spectromicroscopy of CO-Terminated Ag Tips. The Journal of Physical Chemistry Letters 2018, 9 (11), 3074-3080.

Table of Contents

1

6