The protocol describes how to monitor electrochemical events on single nanoparticles using surface-enhanced Raman scattering spectroscopy and imaging.
Studying electrochemical reactions on single nanoparticles is important to understand the heterogeneous performance of individual nanoparticles. This nanoscale heterogeneity remains hidden during the ensemble-averaged characterization of nanoparticles. Electrochemical techniques have been developed to measure currents from single nanoparticles but do not provide information about the structure and identity of the molecules that undergo reactions at the electrode surface. Optical techniques such as surface-enhanced Raman scattering (SERS) microscopy and spectroscopy can detect electrochemical events on individual nanoparticles while simultaneously providing information on the vibrational modes of electrode surface species. In this paper, a protocol to track the electrochemical oxidation-reduction of Nile Blue (NB) on single Ag nanoparticles using SERS microscopy and spectroscopy is demonstrated. First, a detailed protocol for fabricating Ag nanoparticles on a smooth and semi-transparent Ag film is described. A dipolar plasmon mode aligned along the optical axis is formed between a single Ag nanoparticle and Ag film. The SERS emission from NB fixed between the nanoparticle and the film is coupled into the plasmon mode, and the high-angle emission is collected by a microscope objective to form a donut-shaped emission pattern. These donut-shaped SERS emission patterns allow for the unambiguous identification of single nanoparticles on the substrate, from which the SERS spectra can be collected. In this work, a method for employing the SERS substrate as a working electrode in an electrochemical cell compatible with an inverted optical microscope is provided. Finally, tracking the electrochemical oxidation-reduction of NB molecules on an individual Ag nanoparticle is shown. The setup and the protocol described here can be modified to study various electrochemical reactions on individual nanoparticles.
Electrochemistry is an important measurement science for studying charge transfer, charge storage, mass transport, etc., with applications in diverse disciplines, including biology, chemistry, physics, and engineering1,2,3,4,5,6,7. Conventionally, electrochemistry involves measurements over an ensemble — a large collection of single entities such as molecules, crystalline domains, nanoparticles, and surface sites. However, understanding how such single entities contribute to ensemble-averaged responses is key for bringing forth new fundamental and mechanistic understandings in chemistry and related fields because of the heterogeneity of electrode surfaces in complex electrochemical environments8,9. For example, ensemble reduction has revealed site-specific reduction/oxidation potentials10, the formation of intermediates and minor catalysis products11, site-specific reaction kinetics12,13, and charge carrier dynamics14,15. Reducing ensemble averaging is particularly important in improving our understanding beyond model systems to applied systems, such as biological cells, electrocatalysis, and batteries, in which extensive heterogeneity is often found16,17,18,19,20,21,22.
In the past decade or so, there has been an emergence of techniques to study single-entity electrochemistry1,2,9,10,11,12. These electrochemical measurements have provided the capabilities to measure small electrical and ionic currents in several systems and revealed new fundamental chemical and physical characteristics23,24,25,26,27,28. However, electrochemical measurements do not provide information about the identity or structure of molecules or intermediates at the electrode surface29,30,31,32. Chemical information at the electrode-electrolyte interface is central to understanding electrochemical reactions. Interfacial chemical knowledge is typically obtained by coupling electrochemistry with spectroscopy31,32. Vibrational spectroscopy, such as Raman scattering, is well-suited to provide complementary chemical information on charge transfer and related events in electrochemical systems that predominately utilize, but are not limited to, aqueous solvents30. Coupled with microscopy, Raman scattering spectroscopy provides spatial resolution down to the diffraction limit of light33,34. Diffraction presents a limitation, however, because nanoparticles and active surface sites are smaller in length than optical diffraction limits, which, thus, precludes the study of individual entities35.
Surface-enhanced Raman scattering (SERS) has been demonstrated to be a powerful tool in studying interfacial chemistry in electrochemical reactions20,30,36,37,38. In addition to providing the vibrational modes of reactant molecules, solvent molecules, additives and the surface chemistries of electrodes, SERS provides a signal that is localized to the surface of materials that support collective surface electron oscillations, known as localized surface plasmon resonances. The excitation of plasmon resonances leads to the concentration of electromagnetic radiation at the surface of the metal, thus increasing both the flux of light to and the Raman scattering from surface adsorbates. Nanostructured noble metals such as Ag and Au are commonly used plasmonic materials because they support visible light plasmon resonances, which are desirable for detecting emission with highly sensitive and efficient charge-coupled devices. Although the largest enhancements in SERS come from aggregates of nanoparticles39,40, a new SERS substrate has been developed that allows SERS measurements from individual nanoparticles: gap-mode SERS substrate (Figure 1)41,42. In gap-mode SERS substrates, a metallic mirror is fabricated and coated with an analyte. Next, nanoparticles are dispersed over the substrate. When irradiated with circularly polarized laser light, a dipolar plasmon resonance formed by the coupling of the nanoparticle and substrate is excited, which enables SERS measurements on single nanoparticles. SERS emission is coupled to the dipolar plasmon resonance43,44,45, which is oriented along the optical axis. With the parallel alignment of the radiating electric dipole and collection optics, only high-angle emission is collected, thus forming distinct donut-shaped emission patterns46,47,48,49 and allowing the identification of single nanoparticles. Aggregates of nanoparticles on the substrate contain radiating dipoles that are not parallel to the optical axis50. In this latter case, low-angle and high-angle emissions are collected and form solid emission patterns46.
Here, we describe a protocol for fabricating gap-mode SERS substrates and a procedure to employ them as working electrodes to monitor electrochemical redox events on single Ag nanoparticles using SERS. Importantly, the protocol using gap-mode SERS substrates allows for the unambiguous identification of single nanoparticles by SERS imaging, which is a key challenge for current methodologies in single nanoparticle electrochemistry. As a model system, we demonstrate the use of SERS to provide a readout of the electrochemical reduction and oxidation of Nile Blue A (NB) on a single Ag nanoparticle driven by a scanning or stepped potential (i.e., cyclic voltammetry, chronoamperometry). NB undergoes a multi-proton, multi-electron reduction/oxidation reaction in which its electronic structure is modulated out of/in resonance with the excitation source, which provides a contrast in the corresponding SERS spectra10,51,52. The protocol described here is also applicable to non-resonant redox-active molecules and electrochemical techniques, which may be pertinent to applications such as electrocatalysis.
1. Gap-mode SERS substrate preparation
2. Gap-mode SERS substrate characterization
3. Preparation of the electrochemical cell
4. Bulk cyclic voltammetry measurements
5. Single-nanoparticle electrochemical SERS microscopy and spectroscopy measurements
6. Imaging analysis
7. Nanoparticle size analysis
8. Spectroelectrochemical data analysis
Figure 2A shows Ag thin film substrates prepared using an electron beam metal deposition system. The "good" substrate shown in Figure 2A has a homogenous coverage of Ag metal over the glass coverslip, while the "bad" substrate has a non-uniform coverage of Ag. The ultraviolet-visible spectrum of the "good" Ag thin film is shown in Figure 2B, which demonstrates that the film is partially transparent for the visible portion of the electromagnetic spectrum. The "good" Ag thin film substrate has an optical transparency of 34% for the 642 nm laser light that is used for the spectroelectrochemistry experiments in the current protocol. Figure 2C shows a representative AFM image of a 10.8 µm x 10.8 µm area of the "good" substrate. The root mean square roughness value of the representative area is 0.7 nm, which indicates that the Ag thin film is atomically smooth. The variation in the height of the Ag thin film substrate is represented by the line profile shown in Figure 2D, further demonstrating the uniformity and smoothness of the film.
Figure 3A shows a representative SEM image of Ag nanoparticles drop-cast and air-dried on a Si wafer. From an analysis of 243 nanoparticles, the average diameter of the Ag nanoparticles used in this protocol was 79.2 nm ± 8.4 nm. It should be noted that different sizes of Au or Ag nanoparticles could also be used55. Additionally, this protocol uses highly monodisperse nanoparticles, but there is no dispersity requirement, as this protocol enables the measurement of single nanoparticles. To construct a gap-mode SERS substrate, in this work, the Ag nanoparticles were deposited onto the surface of an Ag thin film substrate that had been previously incubated with NB (Figure 3B).
A gap-mode SERS substrate was used as the working electrode to construct an electrochemical cell, as shown in Figure 4A. The electrochemical cell was immobilized on a microscope stage and connected to a potentiostat, as shown in Figure 4B. With the electrochemical cell mounted on an inverted optical microscope, a 642 nm laser was focused onto the gap-mode SERS substrate working electrode in an epi-illumination geometry. Individual Ag nanoparticles on the Ag thin film in air can be unambiguously identified by a donut-shaped emission pattern, as shown in Figure 5A. These donut-shaped emission patterns can be reliably used as a signature to identify individual Ag nanoparticles49. If more than a single nanoparticle (dimer, trimer, or multimer) is present in the illumination volume, a solid emission pattern is observed, as shown in Figure 5B. Upon the introduction of the electrolyte solution, the donut-shaped emission pattern typically is converted to a solid emission pattern. The reason for this is that the dipolar plasmon modes within the single nanoparticle (not aligned with the optical axis) radiate emission from the solvent and electrolyte molecules in all directions. Therefore, the emission pattern is a superposition of high-angle NB SERS emission from the nanoparticle-substrate gap and low-angle SERS emission from the electrolyte and solvent molecules. The removal of the electrolytic solution recovers the donut-shaped emission patterns. In this protocol, following the identification of a single nanoparticle by SERS imaging, SERS spectroscopy is used to identify the redox probe molecule. The SERS spectrum in Figure 5C corresponds to the donut-shaped emission pattern shown in Figure 5A. The vibrational modes represent a fingerprint for the NB molecules.
Figure 6A displays representative cyclic voltammograms of NB in phosphate buffer (pH = 5) obtained using an Ag disk working electrode and a Pt wire counter electrode. A cyclic voltammogram is obtained prior to spectroelectrochemistry measurements of single nanoparticles to understand the ensemble redox behavior of the probe molecules — NB in this case. In this work, as the applied potential was swept from 0 to −0.6 V, a cathodic peak was observed at −0.27 V versus Ag/AgCl (3 M KCl). As the potential was swept back to 0 V, an anodic peak was observed at −0.21 V. The same applied potential range was used for the spectroelectrochemical measurements, as shown in Figure 6B. After the identification of a single Ag nanoparticle exhibiting a donut-shaped emission pattern, the electrolyte solution was pipetted into the electrochemical cell. Under laser illumination, the SERS spectra were then continuously collected as the applied potential was swept between 0 to −0.6 V at a scan rate of 50 mV/s (Figure 6B). The NB molecules in and around the gap between the Ag nanoparticle and the Ag film were electrochemically reduced (off state), and the SERS intensity decreased, as shown in the waterfall plot of the SERS spectra (also an inset in Figure 6A). As the applied potential was swept from −0.6 to 0 V, the SERS intensity increased, as the NB molecules were electrochemically oxidized (on state). The modulation in SERS signals represents a method to determine the reduction and oxidation potentials of NB on a single nanoparticle. Other electrochemical techniques can be substituted for voltammetry to further characterize redox reactions. Figure 7A shows the SERS response from NB when the potential of the working electrode was stepped to −0.4 V (i.e., chronoamperometry). When the electrode potential was stepped to −0.4 V, the SERS signal decayed due to the reduction of NB. This spectroelectrochemical technique enables one to investigate the transient behavior of redox reactions at the single-nanoparticle level. Figure 7B demonstrates how the reduction kinetics were altered by the magnitude of the electric bias applied, as evidenced by the decay of the area under the 592/cm peak. Interestingly, the sharp variations in the normalized area demonstrate how stochastic events play a larger role at this scale. As demonstrated with conventional voltammetry and chronoamperometry, the protocol described in this article allows researchers to track the vibrational modes of molecules as they are electrochemically reduced or oxidized on a single nanoparticle. Further, vibrational analyses of molecules on the surface of single nanoparticles allow for the differentiation between chemical and electrochemical steps, which is useful in studying reaction mechanisms.
Figure 1: Gap-mode SERS substrate. Schematic of a gap-mode substrate prepared by placing individual metal nanoparticles on a metal mirror. Please click here to view a larger version of this figure.
Figure 2: Ag thin film substrate characterization. (A) Digital photographs of a good and a bad Ag thin film substrate prepared by an electron beam metal evaporation system. (B) An ultraviolet-visible transmittance spectrum of a good substrate. (C) An AFM image of a representative 10.8 µm x 10.8 µm area of a good substrate. (D) A line profile of the AFM image indicated by the black dashed line shown in (C). Please click here to view a larger version of this figure.
Figure 3: Ag nanoparticle characterization. (A) SEM image of an aqueous Ag nanoparticle colloid drop cast and air dried on a Si wafer. The average diameter of the nanoparticles is 79.2 nm, with a standard deviation of 8.4 nm. (B) Schematic diagram of the gap-mode SERS substrate. The blue stars represent NB molecules. Please click here to view a larger version of this figure.
Figure 4: Preparation of the spectroelectrochemical cell. (A) A representative spectroelectrochemical cell prepared using a gap-mode SERS substrate as the working electrode. (B) A spectroelectrochemical cell immobilized on an inverted optical microscope stage for single-nanoparticle electrochemical spectroscopy and microscopy experiments. Please click here to view a larger version of this figure.
Figure 5: Identification of a single Ag nanoparticle on the Ag thin film substrate. (A) A donut-shaped NB SERS emission pattern, indicating that the signal originates from an individual Ag nanoparticle. (B) A solid NB SERS emission pattern, indicating that the signal originates from more than a single nanoparticle. (C) The SERS spectrum of the donut-shaped emission shown in (A), showing the characteristic peak at 592/cm from the ring deformation vibrational mode of NB52. Please click here to view a larger version of this figure.
Figure 6: Electrochemistry and spectroelectrochemistry of NB. (A) Cyclic voltammograms of 0.5 mM NB in 0.1 M phosphate buffer (pH = 5) using an Ag disk working electrode. The insets show electrochemical SERS images of NB on an individual Ag nanoparticle on the gap-mode SERS substrate at the NB oxidation (bottom image) and reduction (top image) potentials. The scale bars represent 300 nm. (B) Electrochemical modulation of the NB SERS spectrum by cyclic voltammetry on a single Ag nanoparticle on the gap-mode SERS substrate. A Pt wire and an Ag/AgCl (3 M KCl) electrode were used as the counter and reference electrodes, respectively. Please click here to view a larger version of this figure.
Figure 7: Potential step spectroelectrochemistry of NB. (A) Electrochemical modulation of the NB SERS spectrum by a potential step from 0 to −0.4 V (vs. Ag/AgCl) applied at t = 0 (dashed line). The intensity of the peak at 592/cm decreases with time due to the reduction of the NB molecules near the Ag nanoparticle. (B) Transient profile of the normalized area under the 592/cm peak as a function of the applied potential: −0.2 V (blue curve), −0.4 V (green curve), and −0.6 V (red curve). Please click here to view a larger version of this figure.
Depositing Cu and Ag thin metal films on clean coverslips is vital to ensure that the final film has a roughness no greater than two to four atomic layers (or a root mean square roughness less than or equal to around 0.7 nm). Dust, scratches, and debris present on the coverslip prior to metal deposition are common issues that prevent the fabrication of the smooth film required to produce donut-shaped emission patterns. Hence, it is recommended to sonicate the coverslips in different solvents before the metal deposition and, if possible, to perform this process in a cleanroom. Further, careful attention should be paid to the deposition procedure. It may be necessary to clean all the surfaces inside the vacuum chamber (including the crucible holder drawer) and the crucibles, since these parts tend to accumulate dust and debris.
The high deposition rates used during the metal deposition process allow for the deposited film to be atomically smooth but may also be more difficult to control. The incorrect reading of the film thickness sensors may lead to inhomogeneous, excessively thick, or excessively thin films. If the metal films are too thin, islands of material may be deposited instead of a continuous surface. Films that are too thick will result in substrates that are opaque, which will prevent the excitation light from efficiently exciting the NB molecules and impede the collection of emission light; this, in turn, will decrease the overall sensitivity of the method and yield poor-quality SERS images and spectra with low signal-to-noise ratios. The deposition of Cu prior to Ag is crucial for the adhesion of the latter metal, but depositing excess Cu will reduce the optical transparency of the substrate, while an insufficient amount of Cu will lead to the delamination of the Ag from the glass coverslips. Further, if the sample platen's dimensions are greater than those of the shutter, evaporated metal may deposit on the coverslips while the shutter is closed, resulting in inhomogeneity of the substrate, as shown in Figure 2A.
The concentrations and incubation times for the NB solutions and Ag nanoparticle suspensions play a key role in producing a good-quality gap-mode SERS substrate. The use of NB solutions with concentrations higher than recommended in the protocol or the use of longer incubation times can lead to high background signals and, thus, pose challenges to locating individual Ag nanoparticles. On the other hand, a low NB solution concentration and a short incubation time will lead to low coverage of the NB molecules on the Ag thin film, which will make identifying single Ag nanoparticles a time-consuming process. Similarly, the use of an Ag nanoparticle suspension with a concentration higher than recommended in the protocol or the use of longer incubation times will lead to agglomeration of the Ag nanoparticles on the Ag thin film; this agglomeration will, thus, lead to SERS substrates that produce a high percentage of solid emission patterns and a decrease in the number of substrate sites that can be identified as single nanoparticles. In contrast, the use of a lower-concentration Ag nanoparticle suspension or a shorter incubation time will lead to a low coverage of Ag nanoparticles. In this case, a greater fraction of the SERS emission patterns will originate from single Ag nanoparticles, but the throughput of the experiment will be reduced.
For the successful implementation of the single-nanoparticle electrochemical SERS imaging and spectroscopy technique described in this paper, special attention needs to be paid to the spectroelectrochemical experimental setup. First, the identification of single Ag nanoparticles on the gap-mode SERS substrate using donut-shaped emission patterns is central for the successful use of the described method. Greater than 100x magnification of the optical image and a microscope objective with a high numerical aperture (e.g., 1.45) are typically required to observe the donut-shaped emission patterns. The high numerical aperture is particularly important for collecting high-angle emissions. Second, it is important to synchronize the collection of the SERS spectra with the electrochemical program. In this protocol, a transistor-transistor logic pulse is sent from the spectrometer detector to the potentiostat to trigger the simultaneous collection of the SERS spectra and electrochemical data. In addition, the readout time of the detector must be considered to accurately correlate the applied potentials with the SERS spectra in voltammetry.
The interpretation of vibrational modes is an important component of SERS spectroscopy. Ag is prone to forming oxides, which may affect the electrochemical process under study56. No oxide layers were detected by SERS in this protocol, but prolonged air exposure or oxidizing potentials cause the formation of oxides on the Ag mirror and/or nanoparticles. The oxide layers may alter the adsorption of redox-active molecules, thus inducing shifts in the vibrational modes. In the current protocol, we did not observe any shifts in vibrational modes between the NB molecules adsorbed onto the Si, the Ag film, or the Ag nanoparticles. We also note that the excitation of plasmon resonances, such as those of the gap-mode substrate, results in the generation of non-equilibrium hot electrons and hot holes that can participate in redox reactions57,58,59,60. To minimize interference from light-induced hot charge carriers, low light fluences are encouraged.
The technique described here can address the limitations of other single-nanoparticle techniques such as collision-based electrochemistry61,62,63, scanning electrochemical microscopy64,65,66, and scanning electrochemical cell microscopy67,68,69. It is possible to measure the electrochemical response of single nanoparticles using these electrochemical techniques; however, it is not possible to directly obtain the identity of and structural information about the reactants, intermediates, and products. The technique described in this protocol allows for tracking electrochemical reactions on single nanoparticles and obtaining chemical information through vibrational spectroscopy. However, this gap-mode SERS electrochemical method yields the best results when the SERS substrates are prepared using the most SERS-active metals under visible light excitation: Ag and Au. This may limit the choice of metals that can be employed in the technique. Further, while gap-mode SERS provides chemical information on electrochemical processes occurring on single nanoparticles, it only yields ensemble-averaged electrochemical information, as the current response is measured over the entire substrate. Nevertheless, the technique demonstrated in this paper is a powerful tool that can be used to gain fundamental mechanistic knowledge in diverse areas of electrochemistry, including in the fields of electrocatalytic reactions, which are important for energy storage70,71, chemical feedstock synthesis72,73, and sensors74,75.
The authors have nothing to disclose.
This work was supported by start-up funds from the University of Louisville and funding from Oak Ridge Associated Universities through a Ralph E. Powe Junior Faculty Enhancement Award. The authors thank Dr. Ki-Hyun Cho for creating the image in Figure 1. The metal deposition and SEM were performed at the Micro/Nano Technology Center at the University of Louisville.
Acetone, microelectronic grade | J. T. Baker | 9005-05 | |
Adjustable pipette, Eppendorf Reference 2 5000 mL | Eppendorf | 4924000100 | |
Analytical Balance, AB54-S/FACT | Metter Toledo | N.A. | |
Atomic Force Microscope, Easy scan 2 | Nanosurf | N.A. | |
AXXIS Electron Beam Thin Film Deposition System | Kurt J. Lesker | N.A. | |
Cary 60 UV-Vis Spectrophotometer | Agilent | N.A. | |
Conductive epoxy, two part | Electron Microscopy Sciences | 12642-14 | |
Copper pellets, 99.99% pure | Kurt J. Lesker | EVMCU40EXE | |
Copper wire, bare, 18 AWG | VWR | 66248-040 | |
Crucible, Graphite E-Beam | Kurt J. Lesker | EVCEB-23 | |
Diamond Scriber | Ted Pella | 54484 | |
EMCCD Camera, ProEM HS: 1024BX3 | Teledyne Princeton Instruments | N.A. | |
Epoxy, Clear | Gorilla Glue | N.A. | |
Glass Tube Cutter | Wheeler-Rex | 69012 | |
Glass Tube, Borossilicate (OD 0.75", ID 0.62", L 12") | McMaster-Carr | 8729K45 | |
Immersion oil, Type-F | Olympus | IMMOIL-F30CC | |
Inverted Microscope, IX73 | Olympus | N.A. | |
Laser, Excelsior One 642 nm Free space | Spectra-Physics | N.A. | |
LightField | Teledyne Princeton Instruments | N.A. | |
MATLAB 2022b | MathWorks | N.A. | |
Micro cover glass (coverslips), 24×60 mm No. 1 | VWR | 48404-455 | |
Microscope Smartphone Camera Adapter | qhma | QHMC017A-S01 | |
Nile Blue A, pure | Acros Organics | 415690100 | |
Nitrogen, Ultra Pure, Compressed | Specialty Gases | N.A. | |
Objective, UPLanXApo 100× Oil Immersion | Olympus | 14-910 | |
Polyimide Film, Kapton | 3M | 16089-4 | |
Potassium Phosphate Monobasic | VWR | P285 | |
Potentiostat, 660E | CH Instruments | N.A. | |
Pt wire | Alfa Aesar | 10956-BS | |
Scanning Electron Microscope, Apreo C SEM | Thermo Fischer Scientific | N.A. | |
Si wafer | Ted Pella | 16006 | |
Silver nanoparticles (nanospheres), NanoXact 0.02 mg/mL in 2 mM citrate | nanoComposix | AGCN60 | |
Silver pellets, 99.99% pure | Kurt J. Lesker | EVMAG40EXE-A | |
Slide Rack, Wash-N-Dry | Diversified Biotech | WSDR-2000 | |
Smartphone, iPhone 13 mini | Apple | N.A. | |
Sodium Phosphate Dibasic Heptahydrate | VWR | 0348 | |
Spectrometer, IsoPlane SCT320 | Teledyne Princeton Instruments | N.A. | |
Tissue Wipers, Light-duty | VWR | 82003-820 | |
Tweezers, KS-04 | Kaisi Hardware | N.A. | |
Utrasonic Generator, sweepSONIK | Blackstone-NEY Ultrasonics | 809379 | |
Water Ultrapurifier, Sartorius Arium mini | Sartorius | N.A. |