Toward a mechanistic understanding of plasmon-mediated photocatalysis

James L. Brooks; Christopher L. Warkentin; Dayeeta Saha; Emily L. Keller; Renee R. Frontiera

doi:10.1515/nanoph-2018-0073

Open Access Published by De Gruyter August 29, 2018

Toward a mechanistic understanding of plasmon-mediated photocatalysis

James L. Brooks , Christopher L. Warkentin , Dayeeta Saha , Emily L. Keller and Renee R. Frontiera

From the journal Nanophotonics

https://doi.org/10.1515/nanoph-2018-0073

Abstract

One of the most exciting new developments in the plasmonic nanomaterials field is the discovery of their ability to mediate a number of photocatalytic reactions. Since the initial prediction of driving chemical reactions with plasmons in the 1980s, the field has rapidly expanded in recent years, demonstrating the ability of plasmons to drive chemical reactions, such as water splitting, ammonia generation, and CO₂ reduction, among many other examples. Unfortunately, the efficiencies of these processes are currently suboptimal for practical widespread applications. The limitations in recorded outputs can be linked to the current lack of a knowledge pertaining to mechanisms of the partitioning of plasmonic energy after photoexcitation. Providing a descriptive and quantitative mechanism of the processes involved in driving plasmon-induced photochemical reactions, starting at the initial plasmon excitation, followed by hot carrier generation, energy transfer, and thermal effects, is critical for the advancement of the field as a whole. Here, we provide a mechanistic perspective on plasmonic photocatalysis by reviewing select experimental approaches. We focus on spectroscopic and electrochemical techniques that provide molecular-scale information on the processes that occur in the coupled molecular-plasmonic system after photoexcitation. To conclude, we evaluate several promising techniques for future applications in elucidating the mechanism of plasmon-mediated photocatalysis.

Keywords: nanomaterials; photocatalysis; plasmons; spectroscopy

1 Introduction

The growing necessity for clean and renewable forms of energy production has had a significant effect on developing new technologies capable of achieving environmentally conscious and energetically efficient methodologies for driving industrial catalytic reactions. Photocatalysis, which relies on harvesting an abundance of photons from an external source, has been repeatedly demonstrated as a plausible option for driving energetically demanding chemical reactions. The concept of directly converting solar energy to chemical energy has proven promising, as it negates the need of devoting a high amount of electrical or other forms of energy to carry out the process [1]. Initially, the groundbreaking research performed by Fujishima and Honda in 1972 ushered in a wave of studies focused on implementing semiconducting materials for solar-driven photocatalysts [2]. However, semiconducting materials generally absorb in the ultraviolet, a relatively inefficient region in the solar spectrum; despite extensive research, photoconversion efficiencies have remained below the limit needed for widespread application. Present-day catalyst design has been focused on implementing materials capable of harvesting the most abundant region of the solar spectrum, the visible region [3]. Still, the current options for photocatalysts leave much to be desired regarding their efficiencies and energetic demands that are required to power them.

In the development of industrially relevant photocatalytic processes, challenges impeding their progress must be considered. Many of the desirable catalytic processes have large energetic barriers, require the transfer of multiple electrons, and frequently require bimolecular collisions on the catalyst surface. Therefore, these reactions typically require high temperature, high pressure, and lots of time to produce desired products. For example, steam methane reforming, a key method in hydrogen production and the Fischer-Tropsch process, is highly endothermic and requires reaction conditions exceeding 1000 K at high pressure (15–30 atm) [4]. Other key industrial catalytic reactions require high specificity to produce products with high purity, as in the hydrogenation of acetylene to ethylene, a key feedstock for many industrial products. In addition to meeting strict reaction conditions, the scale of such processes must continually rise to meet increasing global demands. Table 1 provides a glance at the relative scale of several important industrial catalytic processes, which reflects the vast amount of capital invested in the development of the necessary infrastructure for their implementation. In view of these challenges, for photocatalytic processes to be competitive with standard industrial practices, highly efficient and robust catalysts that can maintain their catalytic activity under harsh reaction conditions are required. The application of photocatalytic processes will likely require integration into existing infrastructure and must prove to be financially viable at large scales to be competitive and comply with ever-expanding global consumer demands. Additionally, significant research effort has been directed toward implementing photocatalysts into energy-demanding situations that do not require mass-scale production. For example, photocatalysts have the potential to be used in the production of electricity on an individual household scale. Regardless, the development of plasmonic photocatalysts could eventually leave a dramatic impact on both large- and small-scale catalytic processes.

Table 1:

List of plasmon-mediated industrially-relevant catalytic processes.

Process	Endo/exo	Standard catalyst	Approx. production scale (megaton per year)	Plasmon-mediated (prominent catalyst)
Acetylene hydrogenation	Exothermic	Pd/Al₂O₃	N/A	Yes, AINC-Pd NP [5]
Haber-Bosch process (ammonia generation)	Exothermic	Fe/K₂O	200 [6]	Yes, Au NP coupled systems [7], [8]
Reverse water-gas shift reaction	Endothermic	Cu	N/A	Yes, Au NP/Ti0₂ [9]
Steam methane reforming	Endothermic	Ni/Al₂O₃	215 (H₂) [10], [11]	Yes, Au-Pt NP [12]
Water electrolysis	Endothermic	Ni & Pt	20 (H₂) [11]	Yes, Au NR/Ti0₂ [13]

Significant progress has been made in recent years toward the application of plasmonic materials as catalysts for driving chemical reactions, as highlighted by examples in Table 1 [5], [13], [14], [15], [16], [17], [18], [19]. These materials are considered promising candidates for driving highly selective chemical processes due in part to their ability to host surface plasmons. The extinction spectrum of a plasmonic material can be finely tuned to match the output of the solar spectrum by changing the size or shape of the nanomaterial. Plasmonic materials harvest energy from resonant photons and partition it into multiple different pathways. Once generated, the surface plasmon produces highly enhanced localized electromagnetic fields, creates elevated thermal environments, ejects highly energetic hot carriers, and/or modifies the potential energy landscape of a nearby molecular species [20], [21], [22]. Each of these possible pathways of energy partitioning may contribute to mediating a catalytic process, with the preferential pathway heavily dependent on the targeted chemical reaction and the plasmonic substrate’s design. Due to the complexity of these potential pathways, designing a plasmonic system capable of achieving optimal turnover numbers and yields is a nontrivial task. Ultimately, to make a significant leap in device fabrication, in the absence of an unpredicted technological development, it is imperative to study the individual mechanistic contributions and dynamics during the energy transformation after plasmon excitation. With this highly pertinent information in hand, plasmonic substrates may be designed to preferentially dictate the flow of energy as the surface plasmon decays and efficiently channel it into mediating a chemical reaction. Herein, we have thoroughly discussed and reviewed the leading spectroscopic and electrochemical techniques used to explore the mechanism behind plasmon-mediated photocatalysis.

One of the most promising aspects of plasmon-mediated photocatalysis is the ability to achieve a high level of chemical selectivity. Chemical selectivity is used to quantify how successful a given catalytic system is at producing the desired product over undesired byproducts. In many commercial catalytic systems, the chemical selectivity of the catalyst can be quite poor due to the existence of a number of competing chemical processes, requiring expensive separation processes. However, plasmonic materials are suitable materials for achieving high selectivity compared to most thermally activated catalytic processes, as they may be specifically designed to selectively generate charge carriers containing the required energies [5]. It is crucial to employ a system capable of producing highly selective products to reach a level of optimization for achieving practical applications of plasmonic photocatalysts. Once fully optimized, one can envision the prospect of a tunable solar-driven photocatalytic device capable of selectively driving a wide range of chemical reactions.

Throughout this review, we have provided a discussion and critique of the current literature that is focused on studying plasmon-mediated photocatalysis from a molecular viewpoint. Probing the interactions between strongly coupled molecule-plasmon systems may help provide a beneficial insight into better understanding the nontrivial mechanism driving fundamental plasmonic processes. Whereas a singular technique may not entirely provide a definitive description of plasmon-molecule interactions on its own, the collective knowledge garnered from various experimental approaches may produce a cohesive picture of how surface plasmons and molecules behave as the plasmon decays, leading to the rational design of highly optimized and selective plasmonic photocatalysts.

The goal of this review is to provide a narrative centralized on highlighting the capabilities and future promise of a number of advanced experimental methods employed to study the transient energetic dynamics of plasmonic materials and their effect on mediating catalytic processes. Ultimately, we aim to recognize the current challenges found in the field of plasmon-mediated photocatalysis, to discuss the notable scientific efforts to better understand the interactions between plasmons and molecules, and to provide an evaluation on the future landscape for applied plasmonic photocatalysts. We begin with a review of the relevant plasmon dynamics and possible energetic decay pathways, highlight a number of exceptional experimental studies that were successful in further explaining the intricate details found in plasmonic systems, and conclude with an outlook on the future of plasmon-mediated photocatalysis.

2 Fundamentals of plasmon generation and decay

2.1 Surface plasmons

Plasmons are the collective oscillation of the free charge density within a material with negative real and small positive imaginary dielectric components, which include materials such as gold (Au), silver (Ag), aluminum (Al), titanium nitride [23], metal oxides [24], and copper (Cu) chalcogenides [25], among a growing list of many others. Surface plasmons are confined to the surface of nanostructured materials [26]. After photoexcitation, the free electron charge density begins to oscillate at the surface of the plasmonic material, leading to the focusing of the far-field radiation to highly localized and dramatically enhanced nanoscale electromagnetic fields. Materials that support surface plasmons have the ability to amplify an electromagnetic wave from free space within an effective volume well below the diffraction limit [27], allowing for a wide range of applications, including fueling photocatalytic reactions [5], [13], [15], [16], [17], [18], [28], [29], [30], increasing the efficiency of photovoltaics [31], [32], [33], [34], [35], [36], [37], and serving as vehicles for photothermal therapy [38], [39], [40], [41], [42], [43], [44].

2.2 Plasmonic energy partitioning

A coherent surface plasmon can be generated by introducing an incident photon source that is on-resonant with the nanoparticle’s extinction spectrum. The wavelength of light used to interact with the particles is much larger than the size of the material itself and produces a coherent oscillation of the electron density locally confined to the nanosized particle. After the surface plasmon has been photoexcited, its energy may rapidly decay through either radiative or nonradiative pathways. The surface plasmon only maintains its coherence for 1–10 fs and begins to convert its energy into multiple pathways as it decays [45]. However, depending on the substrate’s composition, shape, and size, multiple distinct chemical and physical processes may be triggered as the plasmon relaxes. The following sections will detail the various decay pathways the plasmon may undergo as it loses coherence and relaxes. Once the plasmonic energy is dispersed through the preferential pathways, the energy can be used to power a plethora of applications (Figure 1).

Figure 1:

Present-day applications for plasmonic materials.

(Top left) Plasmonic nanoparticles have been employed to target cancerous cells within live organisms. Once photoexcited, the plasmonic energy is intentionally converted into a thermal energy to annihilate the targeted cells. (Top right) Sensing platforms constructed with plasmonic materials are used to detect specific analytes at an ultralow concentration. (Bottom left) Nanolevel architectural design can be achieved by inducing chemical reactions that link two or more individual nanoparticles to assemble a new plasmonic structure. (Bottom right) Plasmonic nanostructures are capable of mediating highly selective catalytic processes.

2.2.1 Enhanced electromagnetic fields

Surface plasmons give rise to dramatically enhanced electromagnetic fields near the surface of the nanostructure [27]. The plasmonic nanoparticles can be specifically engineered to host various nanoscale physical structures that allow for extreme light concentration to subwavelength dimensions, resulting in highly localized electric fields. Resonant interactions between the oscillating free electron density near the surface of the nanostructure and the incident electric field produce these exponentially decaying fields. The regions of amplified electromagnetic fields typically form between the crevice of two or more physical feature or at a sharp, distinct edge on the surface. Examples of a plasmonic field enhancement can be seen in constructs such as the nanoparticle oligomer and bow-tie antenna substrates shown in Figure 2 [46], [47]. When positioned within a region of dramatically enhanced electromagnetic fields, a molecule may experience dramatically enhanced scattering due to the subwavelength confinement of electromagnetic light. This phenomenon led to the development of multiple surface-dependent processes and applications, such as surface-enhanced Raman [48], [49], infrared (IR) [50], [51], and fluorescence [52], [53] spectroscopies and the invention of light-harvesting plasmonic photovoltaics [31], [32], [33], [34], [35], [36], [37]. Once photoexcited, the coherent electromagnetic field decays rapidly and effectively acts like an AC field being driven by the photoexcitation source, making it unlikely to have a direct impact on the photocatalytic reaction pathway.

Figure 2:

Calculated electric field enhancements for Au: (A) nanoparticle oligomers [46] and (B) bow-tie antennas [47].

The maximized field enhancement for both systems is produced at the interfaces between the nanoparticles and the regions containing sharp structural features. (A) Reprinted with permission from Ref. [46]. Copyright 2013 American Chemical Society. (B) Adapted with the permission from Ref. [47]. OSA Publishing.

2.2.2 Hot carrier generation

After the surface plasmon loses coherence, the nonradiative decay pathway produces a distribution of hot carriers that may be used in initiating photocatalytic reactions. This multistep decay process occurs primarily via Landau damping [54], where energy is transferred from a coherent plasmon to individual electron-hole pair excitations. Initially, the hot carrier distribution is nonthermal and contains charged species far from the Fermi level of the material [55], [56]. Then, hot electrons and holes rapidly thermalize, reaching a Fermi-Dirac distribution that corresponds to a high effective electron temperature. This initial thermalization is carried out through a redistribution of the energy via electron-electron scattering interactions during the next several hundred femtoseconds (1–100 fs) [57], [58]. During this time, the hot electrons and holes may contain energies ranging from the Fermi level to the work function. These charge carriers are sufficient in quantity and lifetime to initiate external chemical processes (Figure 3) [55]. The hot carriers further dephase through an additional relaxation mechanism consisting of electron-phonon interactions over a timescale of 1–10 ps [59], [60]. It is during these two time intervals that the charge carriers may contain sufficient energy to transfer to a nearby chemical species to initiate a single or multistep chemical reaction.

Figure 3:

Theoretical modeling of the hot carrier distribution as a function of their energy in AgNPs with diameters of (A) 15 nm and (B) 25 nm.

An abundance of hot electrons (red) and holes (blue) are generated via plasmon decay in the 15 nm particle, whereas the hot carrier generation is significantly diminished in the 25 nm particle. Reprinted with permission from Ref. [55]. Copyright 2015 American Chemical Society.

The ultrafast timescales described above can present challenges in using plasmonic substrates to drive photocatalytic processes. A large majority of industrial significant catalytic processes rely on bimolecular reactions that are induced through collisions in the gas phase. However, due to the extremely short lifespans of plasmonic hot carriers, current reaction activities and turnover frequencies are limited by the low probability of interactions between molecules in free space and the hot carriers generated near the surface. Plasmon-mediated catalysis would be more efficient if the reactants can be preloaded onto the surface, as is common in most heterogeneous catalytic processes, dramatically increasing the likelihood of inducing the interactions between the hot carriers and targeted reactants. Unfortunately, achieving controlled preloading is quite difficult for the most prevalent plasmonic materials (Au and Ag) [61]. Coupling these plasmonic materials with more reactive metals may be the most promising route for achieving high-efficiency catalysis [5].

2.2.3 Localized heating

After the electrons have undergone the fast electron-electron and electron-phonon scattering events (10 fs–10 ps) and have reached a thermalized distribution, the remaining energy is transferred to the localized environment (both solvent and adsorbed molecules) in the form of heat [20], [21]. Plasmonic photothermal heating has received a considerable amount of attention in the past due to its potential of using plasmonic nanoparticles as an excellent agent for delivering thermal energy in a highly localized manner [14], [62], [63]. In fact, an exciting field using photoinduced plasmonic heating is plasmonic photothermal therapy [38], [39], [40], [41], [42], [43], [44]. This form of cancer therapy employs nanoparticles with plasmons specifically tailored to the near-IR to minimize absorption by any tissue above the nanoparticles. The particles are targeted at cancerous cells and irradiated with an external near-IR light source to initiate a phase of cell death via localized thermal heating and induce tumor remission. In addition to photothermal therapy, localized plasmonic heating has also been applied in other applications such as drug delivery and release [64], steam generation [65], photothermal therapeutics [66], and growth manipulation of nanoscale structures [67], [68].

Localized plasmonic heating has also been demonstrated as a fundamental mechanism for carrying out or assisting certain catalytic reactions [14], [19], [69]. A surplus of localized heat is typically a vital condition for a variety of endothermic reactions. During the plasmon decay process, the formation of highly localized thermal environments, both on the nanoparticle’s surface and on the surrounding medium, can be an excellent source of thermal heat transferred to a nearby molecular species. In fact, localized heating is one of the few benefits of the lossy nature of plasmons [70]. While working in tandem with the plasmonic hot carriers, the released thermal energy could play an integral role in reducing energetic barriers preventing a reaction from taking place. A consideration of the interplay between local heating and hot carrier generation is necessary for successfully mediating a photocatalytic process.

2.2.4 Modified molecular potential energy surfaces

Another aspect of plasmonic photocatalysis is the formation of new energy levels as a molecule adsorbs to the surface of a plasmonic material. In this scenario, the strong coupling between the plasmon and molecule may lead to an alteration to the molecular resonances. An analogous effect was discussed by Boerigter et al. in 2016 [71]. Boerigter et al.’s work described new hybridized metal-adsorbate states that form as a result of molecular chemisorption to a metal surface, producing a new charge transfer pathway from the nanoparticle to the molecule. These chemical interactions on the surface resulted in a highly modified coupled system that allowed for resonant energy transfer between the plasmon and the adsorbate. A more detailed discussion of this study is provided in the surface-enhanced Raman spectroscopy (SERS) section (3.1.1) further on in this review. This proposed pathway was successful in providing empirical evidence of the effects of molecule-surface interactions on plasmonic photocatalysis. In 2018, Kazuma et al. were successful in identifying a plasmon-induced single-molecule dissociation of dimethyl disulfide [(CH₃S)₂] on Ag and Cu surfaces with scanning tunneling microscopy (STM) [72]. Here, the authors found that the dissociation was mediated by an energy transfer from the localized surface plasmon to the adsorbed molecular species. The direct adsorption of the (CH₃S)₂ onto the metal substrates resulted in the LUMOs of (CH₃S)₂ to be weakly hybridized with the metal. The weak hybridization significantly reduced the likelihood of excited-state relaxation to the ground state, which gave rise to an accessible dissociative potential energy surface. There is a substantial literature within the surface science community that has detailed the interactions between molecular adsorbates and surfaces, which may be quite helpful for further understanding the interplay between surface and plasmonic effects on the electronic states of adsorbed molecules [73], [74], [75], [76], [77]. The results from these studies ultimately laid a groundwork for understanding surface and molecular plasmonic effects. However, a discussion into these interactions is beyond the scope of this review. Due to this complexity of the coupled molecular-plasmonic system, a variety of technical approaches are needed to examine the transient mechanisms relevant to plasmonic photochemistry.

3 Select methods to elucidate the mechanism of plasmon-mediated photocatalysis

3.1 Spectroscopy

3.1.1 Surface-enhanced Raman spectroscopy (SERS)

3.1.1.1 Ensemble SERS

One of the more effective and frequently used techniques to explore plasmon-molecule interactions is SERS. SERS is a vibrationally sensitive technique that exploits the fundamental nature of plasmonic systems by enhancing the Raman-scattered signal generated by molecules positioned within the substrate’s regions of most dramatic field enhancement, which are known as hot spots in the SERS community [78], [79]. SERS is an excellent method for probing and monitoring plasmon-mediated photoreactions in situ and in real time due to its remarkably high signal enhancement, which can enable microsecond acquisition times [80]. These experiments are typically highly sensitive to the signal arising from the most enhanced hot spots on the plasmonic substrates [81]. Traditionally, the nanostructures are noble metals (e.g. Au and Ag) that have been specifically fabricated or synthesized to host sharp, nanoscale features that couple together to form hot spots.

The magnitude of a collected signal can be quantified by calculating an SERS enhancement factor (EF) for the plasmonic substrate used in the experiment. The SERS EF can be calculated by the following equation [82]:

EF=ISERS/NsurfINRS/Nvol

where I_SERS and I_NRS are the Raman intensities on the SERS substrate and under normal conditions, respectively. N_surf is the number of molecules probed on the substrate and N_vol is the number of molecules probed during the normal Raman collection. This equation is evaluated at a set excitation wavelength for a specific Raman-active vibrational mode. Multiple assumptions, such as the molecular probe’s surface coverage and estimated Raman cross-sections, can be made during these calculations and may lead to discrepancies in the values reported throughout literature. Understanding SERS EFs is essential for appropriately evaluating the work performed within this field, as it can be a useful metric of the localized electric fields experienced by the molecular probes [83]. However, it is important to note that the EF in nearly all SERS experiments are ensemble averaged. This ensemble averaging can include contributions from hot spots with EFs ranging across approximately six orders of magnitude, although the strongest enhancing sites produce a vast majority of the collected Raman signal. Fang et al. published an exceptional study that examined the distribution of individual hot spots and how they contribute to the overall SERS intensity [81]. They found that fewer than 0.01% of probed hot spots in a single SERS exposure account for nearly 25% of the collected SERS signal. Therefore, the reported SERS EFs in most publications are typically a lower bound to overall signal magnitude due to the assumption that every adsorbed molecule contributes to the SERS signal. There is currently a huge focus in the literature on developing new SERS substrates [84], [85], [86], [87], [88], [89], [90], but until atomically defined nanostructures are available in geometries large enough to support a surface plasmon, they will fundamentally be limited by this ensemble averaging, making absolute quantitative analysis difficult.

In fact, SERS has a myriad of advantages for examining the mechanism behind plasmon-mediated photocatalysis. Being a vibrationally sensitive spectroscopic technique, SERS offers much more detail and information about the molecular probe’s structure than electronic spectroscopic techniques. SERS is capable of discerning any minor alterations to the molecule’s orientation on the surface through discrete changes in the spectra. Also, the rapid decay of the electromagnetic fields near the plasmonic nanostructure’s surface ensures that the signals are generated from molecules on or near (r⁻¹⁰ distance dependence for spherical nanoparticles) the surface [78], [91], [92], making it a highly selective technique. The selectivity of SERS ensures that the probed molecules are the most susceptible to interactions with any plasmon-induced effects.

However, the inherent limitations of SERS must also be addressed. The selectivity of this technique comes with a cost: the molecular probe typically contains specific functional groups that have a strong affinity to the surface. Certain chemical moieties (thiols, amines, N-heterocyclic carbenes) undergo chemisorption onto the surface of traditional plasmonic metals and may be confined within the intense electromagnetic fields produced during excitation. Another constraint is the limited options currently available for surface plasmon-hosting substrates. The most commonly used substrates are Au, Ag, Cu, and, more recently, Al due to their enhancement capabilities that cover most of the visible and near-IR spectra. Nevertheless, the dramatic signal enhancement found in a wide range of Au and Ag substrates has resulted in these metals being the leading materials for exploring plasmon-mediated reactions with SERS.

One of the most interesting aspects of using SERS to study surface plasmons and their subsequent dynamic processes is how the technique is capable of simultaneously initiating a photocatalytic reaction while also probing the vibrational features of reacting adsorbates. The most meticulously studied chemical reaction using SERS is the dimerization of both 4-nitrobenzenethiol (4-NBT) and 4-aminothiolphenol to 4,4′-dimercaptoazobenzene (DMAB) [83], [93], [94], [95], [96], [97], [98], [99], which will be a common reaction discussed throughout this review. Most variants of SERS techniques collect signals that effectively describe how the molecules are behaving in the steady-state regime, which can contain data rich with information about reaction yields, efficiencies, and rates. However, this experimental constraint may be viewed as a limitation as it does not allow for an analysis pertaining to the transient dynamics that occur during a chemical reaction [55]. To address this concern, multiple research groups are aggressively attempting to develop ultrafast SERS techniques that can monitor the step-by-step mechanism of various plasmonic dynamics (Section 3.1.3).

An elegant study published by Boerigter et al. used SERS to examine the mechanism of charge transfer from a plasmonic system to adsorbates [71]. Their findings suggest that, by considering the effects of molecular adsorption on the flow of charge within the plasmonic nanostructure, the yield of extracted hot electrons potentially could be substantially higher than the theoretical yields derived from traditional metal-centric models. The conventional models describe the process of charge excitation in plasmonic nanoparticles and the subsequent transfer to an adsorbed species, which disregards any potential influence the adsorbate may have on the transient dynamics. Here, the authors proposed an alternative mechanism for charge transfer that introduces a new pathway capable of a direct and on-resonant charge transfer into the high-energy adsorbate states (Figure 4A and B). This alternative mechanism circumvents the “losses” that occur during the electron-electron and electron-phonon thermalization pathways that are present in the present-day mechanism [70]. The experimental work to support this proposal examined the anti-Stokes and Stokes scattered photons to simultaneously measure the vibrational temperatures within the chemical adsorbates and metal nanoparticle (Figure 4C and D). During these studies, they found that methylene blue experienced an elevated vibrational temperature when excited by 785 nm photons and not 532 nm photons. These results hint at the possibility of an observed charge transfer between the metal and the adsorbate during 785 nm excitation, which is resonant with the proposed hybrid metal-adsorbate states that are formed upon chemisorption.

Figure 4:

Formation of an alternative pathway of resonant energy transfer for plasmonic charge carriers into the high-energy states of the molecular adsorbates.

(A) This pathway is generated when an adsorbate or semiconductor interacts with the metal’s surface and can circumvent the thermalization of hot electrons before transfer occurs. (B) Density of state visualization of the proposed mechanism of direct, resonate transfer. (C and D) Data representing the anti-Stokes (blue) and Stokes (red) spectra collected with (C) 785 nm and (D) 532 nm lasers. There is a noticeable increase in the 785 nm anti-Stokes signal due to a resonant charge transfer into the adsorbates. Reprinted with permission from Ref. [71]. Copyright 2016 American Chemical Society.

The conclusions proposed in this work suggest that the community may need to direct more attention toward the effects of molecular adsorption and hybridized states on plasmonic charge transfer leading to photocatalysis. The hybridized metal-adsorbate states formed during chemisorption may give rise to the possibility of a direct, ultrafast pathway from the metal to the adsorbate at the interface between the two materials.

Initiating and monitoring chemical reactions across a wide range of localized field enhancements may provide helpful insights toward optimizing plasmon-mediated photochemistry. Recently, Brooks and Frontiera published a study that looked into the effect of field enhancement on the plasmon-mediated dimerization reaction of 4-NBT to DMAB [83]. The ensemble-averaged field enhancement was effectively tuned at multiple different spatial locations across a heterogeneous Au film-over-nanosphere substrate. Changes in the localized surface plasmon resonance (LSPR) for each probed region were present due to the random variation of nanosphere packing defects, producing a range of EFs from 5×10⁶ to 3×10⁷. Interestingly, no identifiable correlations were found between the reaction rate and the yield with an increased plasmonic field enhancement, suggesting that plasmon-driven processes are not rate limiting for this reaction (Figure 5). The photoreaction rate is likely limited by the local concentration of protons used in the initial reduction of 4-NBT. Additionally, this work found that the presence of alternative plasmon-induced processes, such as a molecular degradation and/or atomic-level substrate alteration, may dramatically impede the reaction’s rate or potentially hinder the overall molecular conversion.

Figure 5:

Field effects on plasmon-mediated dimerization of 4-NBT to DMAB tracked with SERS.

(A) Time-dependent SERS spectra of the dimerization process monitored in real-time. (B) Plot of the final reaction yield’s dependence on the ensemble SERS field enhancement. The final reaction yield is a ratio of the final amplitudes of the product and reactant peaks. (C) Comparison of the rate constants for the reactant loss (red) and product formation (green) against the ensemble SERS field enhancement. Reprinted with permission from Ref. [83]. Copyright 2016 American Chemical Society.

These findings argue that plasmon-mediated chemical processes are not always most effectively powered in the strongest regions of field enhancement. Having a firm grasp on the energy partitioning after plasmon excitation for a specific plasmonic system is a necessity for optimizing the targeted output. Redirecting the flow of energy away from the undesired pathways (local environmental heating, molecular cleavage/dissociation, alternative chemical reactions, etc.) and toward the preferred chemical reaction or catalytic process is the ideal, yet nontrivial, solution for producing plasmonic systems suitable for integration into energetically demanding industrial processes. SERS can also quantify the spectral relationship between the far-field LSPR and near-field Raman scattering events that occur during the typical SERS experiment. Work performed by Kleinman et al. provided a unique viewpoint on how to consider the interplay between far-field and near-field scattering in hot spot-dominated SERS collections [46]. They found that the presence of hot spots in single-particle SERS studies can clearly modify the optimal excitation wavelength for maximized SERS enhancement. To study these far-field and near-field interactions, they developed a correlated LSPR-transmission electron microscopy surface-enhanced Raman excitation spectroscopy technique. Their results clearly identify that the spectral dependence in individual nanoantennas, which are aggregated nanoparticles encapsulated within a silica coating, and their SERS intensity is dramatically dominated by near-field Raman scattering produced within the hot spots between the two or three aggregated Au nanospheres. In fact, they found that the maximum enhancement was completely independent of the far-field LSPR spectra of both individual Au nanoantennas and an ensemble-averaged collection in their experiments. The maximum ensemble-averaged EF of 5.0×10⁷ peaked at an excitation wavelength of 830 nm, although the far-field LSPR scattering intensity was most intense at ~600 nm.

This report is an excellent example of carefully evaluating hot spot-dominated systems and identifying how they can be optimally used for future experiments. As the authors appropriately described, the commonly accepted assumption that far-field spectral scattering determines the ideal excitation wavelength for SERS enhancement may not always hold true. This work clearly demonstrates the importance of using complimentary computational modeling methods and experimental techniques [55], [100] to develop a full description of the plasmonic modes, both bright and dark, for any given system.

3.1.1.2 Single-molecule SERS (SMSERS)

The preceding section evaluated several SERS experiments that were all performed with a monolayer or near monolayer of molecules residing on the plasmonic substrate’s surface. Results collected with this level of molecular concentration have clearly been used to construct fully developed and highly impactful conclusions pertaining to a wide range of photochemical and catalytic reactions. However, as the name implies, an ensemble-averaged measurement can only describe the physical nature of the adsorbed molecules as a whole, which can hinder the identification of clear structure-function relationships. For instance, ensemble-averaged studies are limited to gathering information that is generated from a remarkably wide range of localized field enhancements rather than exploring the sensitive behavior of individual molecules as they interact with their highly specific and unique local environment. SMSERS was first reported in two pioneering articles in 1997 [86], [101] and the field has vastly grown since its discovery [102], [103], [104], [105], [106], [107].

Several studies have been key in the development of SMSERS over the last 20 years. A bianalyte method was developed by the Etchegoin lab in 2006 by statistically comparing the unique spectral information collected from two distinct molecules [104]. A year later, the Van Duyne group developed a similar method that used two isotopologues of a single chemical species [105]. When working in ultralow concentrations of both molecular species, on average, only one variant of the isotopologues is adsorbed to the metal within the effective probing area. Each isotopologue produces a vibrational spectrum containing Raman bands that make it easily distinguishable from the other. Since then, similar techniques have been used to study single-molecule dynamics on plasmonic substrates. Recently, SMSERS has been used in the detection of plasmon-driven electron transfer. In 2017, Sprague-Klein et al. presented a study where 4,4′-bipyridine (BPY-h₈) and its deuterated isotope (BPY-d₈) were adsorbed between Au nanosphere oligomers and subjected to a single-particle pump-probe experimental probing technique [108]. The single particles were exposed to a 532 nm pump CW laser to excite the monomer plasmon while simultaneously being probed by a 785 nm CW laser for the signal collection. They observed a delayed (>2 min) plasmon-mediated charge transfer of surface electrons from the Au oligomers to produce long-lived BPY radical anion products (Figure 6A and B). The authors suggested that an accumulation of surface electrons was required before a neutral BPY molecule could successfully accept a nearby hot electron. In fact, a low 3% yield of electron transfer to an adjacent BPY molecule was reported. A similar study was published in 2017 and was also successful in identifying hot electron-driven redox chemistry at a single-molecule level using methyl viologen [110].

Figure 6:

Time-dependent single-molecule SERS.

(A and B) Anion formation event for BPY on a Au nanoantenna substrate with the associated isotopologue analysis to confirm single-molecule limit and (C) single-molecule step transitions after the formation of DMAB in real time. (A and B) Reprinted with permission from Ref. [108]. Copyright 2017 American Chemical Society. (C) Reprinted with permission from Ref. [109]. Copyright 2016 American Chemical Society.

Typical SMSERS signals are inherently intricate and difficult to analyze due to dramatic temporal and spectral fluctuations. These fluctuations are caused by several sources, including variations of molecular orientation in an isolated hot spot, atomic-scale surface modifications throughout prolonged photoexposure, and possible Brownian diffusion of the molecule [105], [111], [112]. Recently, an innovative SERS-based method to suppress these issues was published by Choi et al. [109]. Here, they explored a single-molecule catalytic reaction of 4-NBT to DMAB in real time by monitoring discrete step-like transitions of the reaction’s products. These experiments implemented a self-assembled Ag nanoparticle (AgNP)-4-NBT-Au thin-film substrate to induce and monitor the dimerization of 4-NBT to DMAB in the junctions between the nanoparticles and Au film, which were capable of reaching SERS EFs of 1.7×10⁸. Rather than attempting to isolate distinct SERS signals from their collected spectra, the authors developed an innovative method to identify discrete transitions tracking the formation and annihilation of individual DMAB molecules within a single, intense hot spot. Figure 6C depicts a time-resolved SERS trajectory of the DMAB species as they react within the nanoparticle-thin-film junctions. In conjunction with computational modeling, the transitions present in time-dependent data are clear spectral evidence of a single-molecule plasmon-mediated photocatalytic reaction.

Developing new methods to monitor individual molecules as they undergo a chemical transformation is nontrivial. Choi et al.’s unique approach of data analysis was capable of extracting data rich with information describing the behavior of individual molecules in a highly complex environment [109]. They addressed the common difficulties found in SMSERS experiments (spectral and temporal fluctuations) and designed a plasmonic system that systematically removed a majority of the randomness by ensuring that the adsorbed molecules were situated in the center of well-characterized hot spots. If anything, these studies identify the need for a new class of intricately designed plasmonic substrates that employ well-defined, easily accessible, and dramatically enhancing hot spots. Recent studies have done an excellent job of using cucurbit[n]uril structures to generate well-defined hot spots for ensemble [113] and single-molecule [114] SERS detection. Constructing a system that directs the flow of hot electrons to these regions of enhancement should provide a significant improvement to the present-day yields and efficiencies reported. This level of nanoscale design will allow the community to probe the globally relevant photochemical and catalytic reactions, which are currently limited by their low Raman cross-sections in the single-molecule regime.

The advancement of SERS and its ability to probe single molecules has left a clear impact on the field of plasmonics and, more specifically, plasmon-mediated photochemical processes. Having a technique that is inherently linked to surface plasmon generation has ignited a boom in publications heavily invested in better understanding the complex molecule-plasmon interactions. For example, SERS studies were crucial in the identification of the role that metal-adsorbate hybridized states may play in modifying the preferred pathway of charge transfer from a plasmonic surface to the molecular probe. Implementing an equal level of well-defined experimental design may help in identifying new phenomena and SERS can continue to be used as an excellent technique for further elucidating the complex mechanism of plasmon-mediated photocatalysis.

3.1.2 Tip-enhanced Raman spectroscopy (TERS)

Another valuable vibrationally sensitive method used to explore plasmon-mediated photocatalysis is TERS. TERS is a spectroscopic technique capable of imaging physiosorbed or chemisorbed molecules with nanoscale or even submolecular spatial resolution while simultaneously providing the ability to manipulate the highly localized environment surrounding the molecules. The concept of coupling scanning probe microscopy with SERS was first introduced in the 1980s [115]. This technique was successfully demonstrated in decades to be followed by multiple independent research groups, each reporting Raman signatures belonging to surface-bound molecules positioned within plasmonic tip-enhanced localized fields [115], [116], [117], [118].

The power of TERS lies in its inherent ability to provide spatially resolved and spectroscopic information pertaining to one [119], [120] or many molecules positioned on a surface. TERS combines the advantages found in SERS and scanning probe microscopy to produce dramatic electromagnetic field enchantments capable of amplifying the chemically sensitive Raman scattered photons while reaching subdiffraction spatial resolution. This is achieved by replacing the nanoparticle substrates used in traditional SERS experiments with a sharp tip made of or coated with a plasmonic metal. These tips typically have an extremely sharp radius of curvature (<50 nm) and can generate highly localized electromagnetic field enhancements in the effective volume between the tip apex and the substrate used to host the molecular probes [121]. The sharp, needle-like tip apex serves as a physical point-source for the surface plasmon generation and can generate a localized field that rapidly decays as it extends away from the tip [122]. When the tip is brought within close proximity to a flat metal substrate, the highly confined electromagnetic field is further amplified due to coupling between the tip and the surface. This phenomenon, which is known as the gap-mode effect, is achievable when the tip is positioned approximately 1–2 nm away from the metal surface. The enhancement produced in this confined gap-mode effect experiences a d⁻¹⁰ distance dependence, where d is the distance between the tip apex and the flat metal surface [123], [124].

One of the more helpful aspects of TERS is its ability to selectively probe individual or multiple molecules at well-defined regions on a planar substrate composed of both metallic and nonmetallic materials [125]. In fact, if the proper molecular analytes are selected, TERS is capable of initiating and tracking plasmon-mediated catalytic reactions in real time in a similar manner as SERS [126], [127], [128], [129]. However, if the goal of the experiment is to study the effects of the tip alone, the technique can be modified to replace the metallic substrate with a material that does not have the proper dielectric constants to host a surface plasmon. This removes the inherent requirement to have the molecules locally adhered to the plasmonic substrate and gives more flexibility in the experimental approaches the user may explore.

One key difference between SERS and TERS is that the SERS signal is typically produced from the regions of highest field enhancement and is generally ensemble averaged and has much higher throughput. In TERS, the dramatically high spatial resolution provides a more detailed analysis of the studied environment. In a similar manner, the hot electrons (or holes) can transfer to the excited state of the adsorbed molecules and drive a chemical reaction. In this configuration, the tip may be selectively positioned over specific molecules or clusters for the controlled transport of hot carriers to the desired molecules. When using STM-TERS, the tip can also be used as an additional source of electrons with a tunable energy range, allowing for a more detailed analysis of the studied material or chemical reaction.

Unfortunately, a handful of complications may be present when attempting to mediate and characterize plasmon-driven chemical reactions with TERS. One of the more difficult challenges in performing TERS experiments is reproducibly fabricating the nanoscale plasmonic tips [130]. Presently, the most common method for tip fabrication is electrochemically etching Ag and Au wires [131]. A solution-based etching method such as this will produce tips that are highly heterogeneous in shape, especially near the tip apex. This lack of reproducibility introduces a wide range of local environments for enhancing the collected Raman signal or mediating a chemical reaction, making it extremely difficult to produce consistent results across multiple tips. The tip stability can even be quite poor during the course of a typical TERS measurement, resulting in data that are potentially flawed or inconsistent in nature. Another challenge when approaching plasmon-mediated photocatalysis with TERS is quantifying any contributions that may arise from electrons tunneling from the metal tip to the adsorbed molecule. To work around this issue, users have also coated the plasmonic TERS tips with an ultrathin oxide film. The oxide layer protects and isolates the plasmonic tip away from the probed surface and minimizes undesirable interactions [129].

Plasmon-mediated photocatalytic reactions were first tracked with TERS in 2012. van Schrojenstein Lantman et al. published a report that monitored the well-studied dimerization reaction of 4-NBT to DMAB [127]. Here, the authors implemented a dual-wavelength approach to monitor the photocatalytic process. They employed a 633 nm excitation source to monitor the vibrational spectra corresponding to both 4-NBT and DMAB, whereas short, discrete periods of 532 nm laser light were directed toward a Ag-coated tip to initiate the dimerization. Using the 532 nm laser source alone to probe the surface-bound molecules resulted in an instantaneous conversion to DMAB and no relevant information could be extracted. However, by exciting the plasmon with short bursts of low-power 532 nm laser light, they could significantly slow down the rate of the ensemble-averaged reaction and monitor the molecular conversion during a slower timescale. By starting with a stable spectrum of 4-NBT, the authors were able to produce time-dependent TERS measurements that tracked the growth and subsequent decay of the DMAB and 4-NBT, respectively (Figure 7A). Ultimately, this study demonstrated the feasibility of using TERS to study the molecular dynamics that are occurring at the nanoscale level, clearly highlighting the potential to employ TERS to extract information that may help describing the interactions between plasmons and molecules.

Figure 7:

Plasmon-mediated photocatalytic reactions monitored with TERS.

(A, left) Time-dependent TERS measurements before and after irradiation with 532 nm laser light. Raman signatures begin to appear in the TERS spectra after the tip is illuminated with a 532 nm excitation source. (A, right) Raman spectra corresponding to 4-NBT (black) and DMAB (red). (B) Topography image of the Ag substrate (left) and TERS chemical reactivity map of DMAB after inducing dimerization of pMA (right). (A) Reprinted with permission from Ref. [127]. Copyright 2012. (B) Reproduced from Ref. [129] with permission of The Royal Society of Chemistry.

Kumar et al. also reported an impressive study that experimentally mapped out the catalytic activity in the photoinduced reaction with nanoscale spatial resolution using atomic force microscopy-TERS [129]. The authors studied a similar dimerization reaction, where they reacted two p-mercaptoaniline (pMA) moieties together to form DMAB. To prevent the molecules from physically interacting with the tip, they protected an Ag-coated tip with an ultrathin layer of alumina film. The alumina film prevented any interactions between the Ag tip and the thiol-containing molecules, which include chemisorption and unintentional spontaneous catalysis, and allowed for the acquisition of chemical reactivity maps with nanoscale resolution on nanostructured Ag substrates. With the alumina-coated Ag tip, they were able to probe a nanostructured Ag substrate and identify the regions of greatest photocatalytic activity by monitoring the production of DMAB (Figure 7B). The experimental design allowed for probing conditions capable of achieving a spatial resolution of 20 nm, allowing the authors to map the chemical reactivity on individual plasmonic nanoparticles. This was the first reported publication that successfully mapped the substrate-dependent catalytic activity of a reaction in a TERS configuration. In addition, the TERS configuration allowed the users to identify the catalytically active and inactive regions on a plasmonic substrate with a 20 nm spatial resolution, which was unprecedented in the field of plasmonics.

Both of these studies are excellent examples of using TERS to induce a chemical process and monitor as the molecules undergo a physical transformation. The plasmonics community stands to gain valuable insights once TERS is able to study industrially relevant catalytic processes. However, it is clear that using TERS to reproducibly study alternative chemical reactions or heterogeneous catalytic processes will be challenging, and improvements in tip fabrication methods would be extremely useful for quantitative measurements.

3.1.2 Ultrafast SERS

Ultrafast SERS is a burgeoning field that combines the plasmonic enhancements discussed in the previous sections with ultrafast time-resolved Raman spectroscopies to investigate the dynamics of plasmon-molecule coupled systems on picosecond and femtosecond timescales (Figure 8A). Here, in the first portion of this section, we used the term ultrafast SERS to refer to all pump-probe plasmon-enhanced Raman scattering spectroscopies with time resolution in the picosecond to femtosecond range.

Figure 8:

Ultrafast SERS probing of the photophysical responses of plasmon-molecule coupled systems.

(A) Representation of the transient dynamics that ultrafast SERS is capable of measuring and tracking on aggregated AgNPs. (B) Ultrafast SERS describing the behavior of 4-NBT molecules as the AuNP’s plasmon experiences a photoinduced energy shift. (C) Fano-like signatures indicate that the energy transfer between the plasmon and the adsorbates is an indirect mechanism. (A and C) Reprinted with permission from Ref. [132]. Copyright 2016 American Chemical Society. (B) Modified with permission from Ref. [133]. Copyright 2017 American Chemical Society.

In recent years, these techniques have been applied toward understanding plasmon-molecule interactions [132], [133], [134], [135] and single-molecule dynamics [136], [137]. The field of ultrafast SERS has been discussed at length in recent reviews [138], [139]. Here, to fit the scope of this review, we will be highlighting and discussing the recent applications, developments, and challenges in ultrafast SERS as applied to plasmonic photocatalysis.

With a time resolution on the order of molecular vibrational periods, and thus the timescale of the nuclear motions involved in chemical reactions, ultrafast SERS has the capability to monitor key photophysical processes and reaction dynamics of such phenomena in real time, elucidating underlying mechanisms and providing a pathway toward their optimization. Of interest are the complex interactions between adsorbed molecules and plasmon generated phenomena: hot electrons and localized hot spots, which can be observed via transient changes in intensity and line shape in ultrafast SERS [132], [133], [136], [140]. In conjunction with complimentary spectroscopic techniques and theoretical modeling, ultrafast SERS has great potential as a tool for observing these interactions across a complete reaction coordinate, monitoring how energy partitions with variations in environment and plasmonic substrate morphology. Ultimately, the transient data extracted from ultrafast SERS studies may help to elucidate fundamental principles in plasmon coupled systems, leading to improved photocatalytic technologies and plasmonic substrates as well as providing empirical data toward the development of a unified electromagnetic and chemical SERS theory [141].

A recent study by Keller and Frontiera has shown promise toward the time-resolved quantification of hot electron generation at plasmonic surfaces [133]. In this study, Keller et al. obtained ultrafast SERS of 4-NBT reporter molecules on 70±10 nm Au nanoparticles (AuNPs). Peak depletions from 4-NBT ring breathing (1079 cm⁻¹) and NO₂ symmetric stretch (1343 cm⁻¹) vibrational modes were observed, maximized as the pump and probe pulses overlap, and decayed over a period of 50 ps (Figure 8B). In accordance with transient absorption (TA), electrochemical, and theoretical models, it was determined that the transient peak depletions were caused by a red shift in the structure’s LSPR, resulting from the delocalization of the hot electrons. The calculations suggested that the charge density shift resulted in the displacement of approximately 10⁹ free electrons in the metal. Although they did not directly observe the product of the hot electron-induced LSPR shift, they proposed that the charge delocalization may be due to a charge transfer between the hot electrons and the adsorbed molecules, uneven distribution of the hot electrons across the probed regions, or potential electron-phonon interactions that result in the heating of the metal lattice. This study corresponds well with a similar study by Brandt et al., which used the same ultrafast SERS methodology to probe the plasmon-driven photochemical dimerization of 4-NBT to DMAB on AgNPs [132]. This study proposed a hot electron charge transfer mechanism for the dimerization, evidenced by transient Fano line shapes and transient changes in 4-NBT/DMAB signal amplitudes. Here, as seen in Figure 8C, the Fano features are attributed to the coupling of the broad AgNP emission and the more narrow SERS photons. The relevant amplitudes of product and reactant in the transient spectra compared to the ground state spectra proved that the nanoscale regions generating the broadband emission were most efficient at driving the photoreaction. Thus this work indirectly proved the role of localized hot electrons in driving the plasmon-induced process.

Together, these studies use the time resolution of ultrafast SERS to provide a description of charge delocalization and charge transfer from a plasmonic substrate to an adsorbed molecule. Interestingly, the Fano line shapes observed by Brandt et al. were not apparent in the study by Keller et al. suggesting a different mechanism of broadband light emission in AuNPs and AgNPs. These results not only display the capability of ultrafast SERS to observe and interpret changes to the coupled plasmon-molecule system in real time but also highlight the dependence of ultrafast SERS measurements on the substrate composition. This is critical to the development of application-specific ultrafast SERS substrates, where differentiation in the identity and morphology of the plasmonic substrate may allow the tuning of interactions between scattered photons and other light-induced processes. Unfortunately, due to the fact that collected SERS signal is spontaneous, the aforementioned ultrafast SERS techniques are limited to probing transient dynamics that occur on the picosecond timescale, as with spontaneous Raman the spectral resolution and temporal resolution are inversely related. This resolution is sufficient to resolve many of the nuclear structural changes but unable to track the early plasmonic dynamics that occur after photoexcitation.

Other recent work has been devoted to the optimization of ultrafast SERS techniques that implement stimulated Raman spectroscopies to achieve femtosecond time resolution for probing the effects of photoexcited plasmons on proximal adsorbates. This is enabled through the use of surface-enhanced coherent anti-Stokes Raman scattering (SE-CARS) and surface-enhanced femtosecond stimulated Raman scattering (SE-FSRS) experiments [134], [136]. Both techniques are similar in experimental design but involve different Raman scattering pathways to provide vibrational information describing the molecular analyte, optimally with time resolution in the 10–100 fs regime. Both spectroscopic techniques employ two to three laser pulses that interact at the sample of interest to produce the stimulated Raman signal. A CARS experiment is designed to generate a stimulated beam of blue-shifted signal (Figure 9A), whereas the four-wave mixing that is dominant in FSRS stimulates Stokes-shifted photons (Figure 9B). Although both techniques are excellent in producing stimulated signals useful for both molecular imaging and time-resolved measurements, they are both generally limited to a high concentration of molecules or materials with large Raman cross-sections. To bypass this issue, multiple attempts have been made to locally enhance the Raman signal by the means of plasmonic surface enhancement.

Figure 9:

Surface-enhanced coherent Raman spectroscopies.

Energy diagram depicting the photon interactions leading to the generation of surface-enhanced (A) CARS and (B) FSRS signal. (B) Comparison of ps-SE-CARS spectra (red and black traces) and cw-SERS spectra (gray, shaded peaks). (D) Raman spectra of BPE using cw-SERS (red) and SE-FSRS (black). (C) Reprinted with permission from Ref. [136]. Copyright 2016 American Chemical Society. (D) Reprinted with permission from Ref. [142]. Copyright 2017 American Chemical Society.

Crampton et al. demonstrated the novelty of SE-CARS to study plasmonic materials and their enhancement capabilities [136]. In this study, the authors probed core-shell Au-silica dumbbells (Au@SiO₂-NTs) with trans-1,2-bis(4-pyridyl)ethylene (BPE) adsorbed at the nanojunction of the participating Au nanospheres (Figure 9C). The results highlight several conclusions regarding the development of surface-enhanced nonlinear spectroscopies. The enhanced local electric field results in a limited accessible power range between observation of signal and damage to the sample. Also, the stimulation of the Raman field in the SE-CARS process is easily saturated using relatively low pulse energies (100 fJ in a 100 fs pulse). The authors also found that, when working in a few-molecule regime with SE-CARS, the time-domain and frequency-domain measurements are not direct Fourier transforms, with the time-domain measurements revealing more in-depth information on the molecular dynamics.

An alternative stimulated Raman technique is SE-FSRS, which was first introduced in 2011 by Frontiera et al. [142]. The SE-FSRS signals have Fano line shapes, which result from coupling between the broadband plasmonic response from colloidal nanoparticles and the narrowband vibrational coherences in the molecular adsorbates (Figure 9D) [140], [143], [144]. Interestingly, the vibrational coherence’s dephasing time was independent of the coupling between the plasmon and the molecule [140]. This initial study laid the groundwork for the future attempts toward applying and further understanding the mechanism behind SE-FSRS. Follow-up work by Gruenke et al. described the optimal concentration of particles and identified the path length dependence in SE-FSRS measurements [134], [139]. They emphasized that extinction and enhancement in SE-FSRS, and other surface-enhanced resonance Raman techniques, must be appropriately balanced to achieve optimal signal intensity and signal-to-noise ratios, particularly in the transmission geometry used here. These studies serve to elucidate fundamental factors in designing efficient ultrafast SERS experiments. Understanding the factors that play a role in signal intensity, sensitivity, degradation, and interpretation are key to the development of a technique that can be readily adopted and applied by the scientific community at large. However, currently both SE-CARS and SE-FSRS studies have been primarily limited to Au@SiO₂ nanoantennas, which has limited applications. The continuation of such diagnostic work is key toward the characterization and development of reliable ultrafast SERS experimentation.

Despite recent advances, some hurdles must still be overcome in the development of femtosecond SERS techniques. Present studies have been limited to a narrow range of model systems, which still require optimization for expansion into studying more complex plasmon-molecule interactions. Furthermore, numerous processes that occur within plasmonic-molecular systems, particularly in the interrogation in the few molecule limit, can greatly complicate data interpretation [145], [146]. Although the above techniques have come a long way to understanding these interactions, there is still much fundamental work that needs to be done to make them more efficient and effective as analytical tools for understanding plasmon-molecule dynamics. To date, there is no study that has been successful in obtaining structural snapshots of evolving chemical reactants on plasmonic surfaces, although, once the technical hurdles are surmounted, these techniques should be tremendously useful approaches for identifying the vibrational character of a molecular species as it undergoes a plasmon-mediated chemical reaction. Having the tools to spectroscopically track the intermediate states as a molecular transformation takes place would dramatically increase our understanding of these nontrivial mechanisms involved in plasmonic photocatalysis. However, complicated in experimental design, the advancement of ultrafast SERS has far-reaching implications in fully realizing applications of plasmon-related processes and deserves continued attention in coming years.

3.1.4 Transient absorption (TA)

The previous sections have contained numerous experimental examples where all the significant data were collected by probing a set of molecules or an individual species as they interact with a photoexcited surface plasmon. However, to effectively probe and understand the interactions between surface plasmons and the electronic states of a given material, a different flavor of ultrafast spectroscopy is required. Here, we discuss the efficacy of TA spectroscopy as a tool to elucidate the key parameters of plasmon-induced processes. The configuration of TA spectroscopy used to study plasmonic materials employs two femtosecond laser pulses: a pump pulse for photoexcitation and a probe pulse to track the subsequent plasmon-induced dynamics. For a more detailed description, Berera et al. [147] and Ruckebusch et al. [148] have provided excellent reviews on the subject. By evaluating the entire collection of transient spectra, one can deduce the effect of plasmonic excitation on adsorbed species. Specific parameters such as energy transfer rates and efficiencies can be extracted by monitoring the differential spectra. To date, most TA experiments on plasmonic systems have looked at the effect of plasmon excitation on proximal semiconductor materials. Studies on transient plasmon-molecule dynamics are challenging as the optical cross-sections of plasmonic materials far exceed the absorption cross-sections of molecules, meaning that the plasmonic response obscures the transient molecular signal. Therefore, directing a concerted effort toward the development of new plasmonic-semiconductor nanostructures for TA studies may help with the pursuit of better understanding the fundamental mechanism behind plasmon-mediated photocatalysis [149], [150].

TA spectroscopy of plasmon-semiconductor systems has been used to directly probe interfacial charge transfer reactions, as seen by Wu et al. in 2015 [151]. This alternative route for plasmon-induced hot-electron transfer (PHET) is plasmon-induced metal-to-semiconductor interfacial charge transfer transitions (PICTT), which combines the strong light-absorbing power of a plasmonic transition with the charge separation properties of a semiconductor-metal system (Figure 10A). Due to the strong coupling between the plasmonic material and the semiconductor’s electronic states, a new decay pathway is formed between the two materials. The first demonstration for the PICTT pathway was provided by TA studies involving colloidal quantum-confined CdSe-Au nanorod (NR) heterostructures, where the PHET occurs within 20 fs (Figure 10B) [151]. The plasmon decays directly by the transfer of an electron to CdSe; therefore, the quantum yield of the charge separation process is independent of the excess energy above the CB edge, as long as the excitation energy is above the absorption threshold. The recorded high yield (>24%) of the electron transfer provides a potential bridge to evade the various competitive channels into which the plasmonic energy may partition, providing an alternative design protocol for fabricating efficient plasmon-mediated devices.

Figure 10:

Transient absorption studies of plasmon-induced interfacial charge transfer.

(A) Energy schematic of PICTT in CdSe-Au NR systems, where the Au tip is strongly damped. Green dashed arrow corresponds to the interband absorption that occurs in the visible region and red arrow indicates the intraband transition in the IR region. (B) Kinetic trace highlighting the intraband absorption in the (probed at ~3000 nm, red circles) and the 1Σ-exiton-bleach (probed at ~580 nm, green line) after 800 nm excitation. From Ref. [151]. Reprinted with permission from AAAS.

TA spectroscopy has also been used to monitor energy transfer between a metal and a semiconductor through a plasmon-induced resonance energy transfer (PIRET) process [29], [152], [153], [154. Here, rather than transferring electrons, the surface plasmon decays by donating its energy to a nearby species, analogous to Förster resonance energy transfer (FRET). PIRET was first directly observed by Cushing et al. in 2012, where they monitored Au@SiO₂@Cu₂O heterojunctions using TA [29]. The spectral overlap between Au and Cu₂O allows for PIRET through dipole-dipole interactions even with the SiO₂ barrier. By monitoring PIRET with TA spectroscopy, they observed PIRET with SiO₂ barrier thicknesses over which direct electron transfer would not be possible. The high-energy transfer efficiency at and below the band edge has the potential to increase the Cu₂O photoconversion range, which is independent of the charge transfer process. The ability to transfer plasmonic energy to a blue-shifted region opens up a wide possibility for improving light-harvesting yields. Even for the smallest ~1.5 nm SiO₂ barrier, PIRET excited at 650 nm created ~1.4 times the number of charge carriers in Cu₂O as the above-bandgap excitation of the same incident flux, with the enhancement spanning the entire plasmon distribution. With proper modifications of the device aimed at minimizing the back transfer of energy by FRET, as well as maintaining a sufficient spectral overlap between the plasmon and the thin-film semiconductor, this method can be used in effective solar light-harvesting devices over a wide spectral range with high yield.

These examples highlight the importance of employing TA to study plasmon-mediated processes to help elucidate the underlying mechanism for charge separation in plasmonic systems. The use of semiconductors to separate plasmon-generated charges may lead to more efficient harvesting of these charges for plasmonic photocatalysis. However, with the scope of the review article in mind, it is likely that present-day TA techniques will not readily surpass ultrafast Raman approaches in determining chemical mechanisms of catalytic steps in plasmonic photocatalysis. Yet, the results compiled with both of these highly useful spectroscopic approaches could be further supplemented by employing or integrating nonspectroscopic methods, such as electrochemistry.

3.2 Electrochemistry

A number of electrochemical techniques have recently been employed to examine plasmon-induced processes, particularly with regard to understanding charge transfer and heating processes in the coupled plasmon-molecule system. By monitoring overpotentials, it is possible to determine the energy levels of reactive intermediates [13]. In addition, scanning electrochemical microscopy (SECM), which monitors the current at an ultramicroelectrode as it is scanned across an active surface, has played a crucial role in determining the contributions of heating during plasmon-mediated processes [155], [156]. Ultimately, electrochemical techniques can be employed to provide a wealth of information about the mechanism of plasmonic photocatalysis and to what extent electrons play a role in driving reactions [157].

A common approach is to pair electrochemical potential control with dark-field microscopy, which monitors the scattering from the plasmonic sample [158], [159], [160]. This has the advantage of providing precise control over the charge carrier density by altering the applied potential as well as allowing for the active monitoring of the LSPR and its response to charge density changes [159], [160], [161]. Most of these techniques measure the charging of plasmonic nanostructures by adsorbing them to a conductive substrate, which is typically indium tin oxide (ITO) [159], [160]. Upon manipulation of the charge density of a plasmonic nanostructure, various changes to the plasmonic nanoparticle and the local environment can occur, including ionic rearrangement of the electrolyte solution, modifications to the particle morphology, and the ability to mediate specific chemical reactions at the particle surface [162], [163], [164]. As an example of the sensitivity of this approach, Collins et al. correlated a 0.02 nm blue shift in the LSPR to the addition of 110 electrons to a single Au NR in an ion gel device [163]. Dark-field microscopy used in tandem with an applied electrochemical potential is clearly a powerful tool for monitoring electron transfer events by closely tracking the LSPR, which may allow for a deeper analysis of the rate-limiting steps and catalyst-specific reaction mechanisms.

Additional modifications to this technique, such as including dynamic potential control and achieving single nanoparticle analysis, has allowed for an increased understanding of the electrochemical tuning of nanoparticles and the role of particle heterogeneity in the various responses [159], [162]. Byers et al. used hyperspectral dark-field imaging to show that, upon electrochemical tuning, a population of nanoparticles can undergo several different processes ranging from nanoparticle charging, as described previously, to electrochemical reactions, such as chloride ion oxidation and hydrogen evolution reaction (HER) (Figure 11) [162]. These responses vary based on a combination of nanoparticle or nanoparticle/substrate properties that either favor nanoparticle charging or charge transfer. Depending on the local environment and nanoparticle properties, nanoparticles also undergo irreversible or reversible electrochemical reactions that may include HER under negative potentials or metal-halide formation under positive potentials [162], [164]. Although there is still room for further understanding of the fundamental principles driving these phenomena, it is clear that these techniques can be used to illustrate a picture of the chemistry occurring around plasmonic nanostructures.

Figure 11:

Heterogeneity observed during electrochemical tuning of individual AuNPs on ITO.

(A) Hyperspectral dark-field image of AuNPs at open circuit potential. (B) Organization of AuNPs into three categories based on the behavior during potential scan with corresponding spectral response of a nanoparticle for (C) reaction 1 (irreversible electrochemical reaction), (D) reaction 2 (reversible electrochemical reaction), and (E) charge density where electrons are added to the AuNP. (F) LSPR shift reported as energy (meV) as a function of potential scanned for seven individual nanoparticles averaged together. Reprinted with permission from Ref. [162]. Available at: https://pubs.acs.org/doi/abs/10.1021/jp504454y. Copyright 2014 American Chemical Society.

In addition to monitoring how adding or removing charge can manipulate plasmonic behavior, electrochemical techniques can also be used to monitor plasmonic photochemical reactions. When the plasmons are intentionally photoexcited with an external source, the electrons in the systems undergo a perturbation and the photocurrent of an integrated plasmonic device or overpotential changes for the half-reaction can successfully be recorded. For solar water splitting, the plasmonic electrochemical cell can be modified such that either the HER is plasmon driven or the device can be modified to form a plasmonic photoanode where the oxidation reaction occurs [13], [17]. For example, Lee et al. built a plasmonic photoanode capable of splitting water by modifying Au NRs. In this device, they successfully separated the hot carriers and used the hot holes to generate O₂ from water. When they compared the measured photocurrent of their device with the production of H₂, they related the various kinetics they observed to cell efficiency [13]. Since then, multiple groups have performed similar measurements where the enhanced photocurrent upon illumination due to integrated plasmonic nanostructures is used as a metric for device efficiency toward solar water splitting [17], [165], [166], [167]. These efforts are excellent demonstrations of the novelty and the potential impact of using the power of plasmonic photocatalysts to mediate chemical processes that are relevant to society today.

Many of these studies also use cyclic voltammetry (CV) to measure how illumination may affect the oxidation or reduction reactions of a redox couple. CV is an electrochemical measurement in which the current is monitored as the working electrode is ramped linearly and cycled through a potential range [157]. In plasmon-enhanced photochemical reactions, illumination of the cell leads to several deviations from traditional CV characteristic traces, such as discrete shifts in the onset potential for the half-reactions and an increase in the photocurrent, which provide useful information pertaining to the energetic barriers for the reaction [168]. Additionally, the redox energies for a studied molecule can be significantly altered by adsorption to the plasmonic material, which is observable in the distinct formation of new features in the CV [169]. Quantifying the changes observed in CV upon illumination of a plasmonic electrochemical cell may help identify how plasmons enhance and modify the observed reactions. However, these techniques focus on ensemble measurements of the system. Due to the potential heterogeneity of plasmonic samples, techniques that measure a smaller subset of the system are necessary to unravel the contributions of different mechanisms.

SECM uses an ultramicroelectrode or a nanoelectrode to sample a more localized environment at interfaces compared to other electrochemical techniques. SECM probes surface reactions and can provide information regarding electron transfer kinetics. For example, Yu et al. separated the individual contributions of hot carriers and thermal effects for the plasmon-driven oxidation of ferrocyanide to ferricyanide [155]. They controlled the reaction by holding the electrodes at a fixed potential at various regimes where the hot carriers and thermal effects worked synergistically or competitively (Figure 12). Using this technique, they were even able to quantify the degree of heating that occurred during the plasmon-driven oxidation of ferrocyanide [156]. For many plasmon-enhanced chemical reactions, the ability to probe surface reactions and measure the surface-dependent current changes allow for a better understanding of the underlying mechanisms that influence plasmon-mediated photochemistry.

Figure 12:

Proposed mechanisms for photoinduced oxidation based on the contributions of hot holes and thermal effects.

(A) Isolating hot hole contribution for photoinduced oxidation by comparing how the reaction potential changes at various laser intensities. Substrate 1 (black) contains contributions from hot holes and heating. Substrate 2 (blue) only has thermal contributions. Hot hole contribution for substrate 1 (red) is isolated by subtracting out the thermal contributions. (B) Pictorial depiction of proposed mechanisms: (i) hot hole generation and (ii) thermal contribution leading to altered equilibrium potential of redox couple. Reprinted with permission from Ref. [155]. Copyright 2018 American Chemical Society.

Although SECM probes a smaller population of a plasmonic system, it is still an ensemble measurement due to the heterogeneous nature of the plasmonic substrate. With the heterogeneity of plasmonic nanostructures and the utility of site-specific catalytic activity, techniques with higher spatial resolution may be necessary. One approach may be to couple electrochemical techniques, such as sample biasing, with techniques such as SERS or TERS, which may aid the tracking of electron transfer events at the molecular level [170]. However, it is necessary to characterize the changes in TERS and SERS signals under these conditions as shown by Kurouski et al. [171] and Di Martino et al. [172], where, depending on changes of the LSPR due to electrochemical tuning or changes between the redox couple, the Raman signal magnitudes can diminish or disappear. Ultimately, the development of an electrochemically integrated TERS technique may be an optimal route for explaining these phenomena.

Electrochemical techniques hold promise for elucidating plasmon-mediated reactions especially with their ability to manipulate and monitor the flow of electrons and the resulting chemical effects, as shown in the above examples. These techniques face some difficulties due to the ensemble nature of these measurements and the fact that majority of these studies are performed with well-understood electrochemical reactions. Although broader applications of these techniques to other plasmon-mediated reactions may be challenging, the information that may be gathered from such studies would greatly benefit the understanding of plasmon-mediated processes and their underlying mechanisms.

4 Outlook

The maturation of the above experimental techniques, both spectroscopic and electrochemical, has been instrumental for the advancement of new applications and fundamental insight in the plasmonics field. Enormous strides have been made in the past decade, as evidenced by the exceptional growth in the number of plasmonically powered processes. However, there is still crucial quantitative information that needs to be extracted to help propel plasmon-mediated catalysis toward becoming a competitive and rational option for industrial applications. To help confront these elusive critical parameters in our collective knowledge, we would like to briefly highlight a few technical approaches that clearly hold abundant promise or may be fruitful to explore.

The experimental approaches covered in this review have focused on spectroscopic and electrochemical characterizations of interactions between plasmonic nanostructures and molecular species. These techniques have a number of advantages that make them well suited for understanding plasmonic photocatalysis, particularly due to their ability to probe the molecular response of the evolving coupled system. However, there are a number of other experimental approaches that have great promise for understanding the mechanism of plasmonic photocatalysis. Magnetic resonance approaches have recently found great success as a characterization technique for the growth and evolution of plasmonic particles [173], and the molecular specificity could find great use in quantifying and tracking the progress of some plasmon-driven reactions. Other experimental approaches such as electron paramagnetic resonance, time-resolved terahertz spectroscopy, or pump-probe IR spectroscopy could also provide new insight into the mechanism of plasmonic photocatalysis.

Although this review has focused on experimental approaches to monitor plasmonic photocatalysis, theory and computational work has guided much of the current understanding of the decay processes and branching ratios in molecular-plasmonic systems, and many of the experimental interpretations presented here would not be possible without theoretical support. A major challenge in furthering theoretical approaches to the study of plasmon-driven photocatalysis lies in the large scale of the system coupled with the importance of quantum mechanical effects and difficulties in determining the landscape of highly reactive molecular potential energy surfaces. A 20 nm diameter Au sphere contains ~60,000 Au atoms, some of which may be strongly coupled and have significant wavefunction overlap with adsorbed molecular species. High-level quantum mechanical calculations on a system of this magnitude are generally intractable with current processing power. However, a number of reasonable approaches have led to significantly improved understanding of these large and complex systems. For example, work by the Nordlander group has provided quantitative values for the hot carrier energy distribution [55] and the relaxation pathways and dynamics of these photoexcited carriers [174]. On the other end of the spectrum, work by the Carter group has applied embedded correlative wavefunction approaches to calculate energy levels of adsorbed molecular species, which has been highly effective at understanding the mechanism of some plasmon-induced processes [18], [175], and in guiding the design of more efficient plasmonic photocatalysts. This approach has recently been applied to complex multistep multielectron reactions such as N₂ dissociation [176], [177]. Recent theoretical work aimed at understanding the complete and strongly coupled molecular-plasmonic system includes atomistic electrodynamics and quantum mechanical models [178], [179], many-body Green’s function formalism approaches to probe quantum emitters coupled to plasmonic systems [180], and combination of Drude model plasmonic systems with time-dependent density functional theory molecular calculations [181], among many others. There are many exciting possibilities for future directions for theoretical models of plasmonic photocatalysis, particularly with the incorporation of methods to track reactive molecular potential energy surfaces, such as the ab initio multiple spawning approach [182], which has been highly successful in looking at reaction dynamics of molecular species in vacuum.

Although the recent focus has been directed toward engineering useful plasmonic nanostructures to serve as catalysts, there is an immense amount of promise in integrating traditional catalysts to enhance the practical usage of plasmon-mediated catalysis. With this mindset, efforts are currently being made to develop plasmon-metal catalyst coupled systems to effectively improve the metal catalyst’s general efficacy. For instance, Swearer et al. synthesized a heterometallic antenna-reactor complex composed of an Al nanocrystal (AlNC) antenna and Pd catalytic nanoparticle to initiate highly selective photocatalysis [5]. This reactor complex is driven by the direct coupling of the plasmonic AlNC with the Pd nanoparticle, which dramatically enhanced the absorption character in the catalytic metal. A depiction of this nanostructured architecture can be seen in Figure 13A. The enhanced production of HD during H₂ dissociation was more strongly correlated to the optical wavelengths that were on resonant to the AlNC’s absorption cross-section, clearly demonstrating the plasmon’s ability to substantially enhance the Pd nanoparticle’s ability to facilitate a photocatalytic reaction. Improved selectivity was also achieved when acetylene was added during H₂ exposure. The authors noted a remarkable 40:1 increase in the relative selectivity of ethylene/ethane production (Figure 13B). Ultimately, their work launched a new wave of studies tailored toward engineering similar reactor complexes that are fueled by a near-field coupling between a traditional metal catalyst and plasmonic material. An original and inventive design such as this is essential for identifying new pathways for refining the present-day attempts of selective plasmon-mediated photocatalysis.

Figure 13:

Plasmon-mediated photocatalytic reactions using a heterometallic antenna-reactor complex.

(A) Cartoon depiction of the coupled AlNC-Pd nanoparticle system. (B) This fabricated substrate is capable of achieving a high level of chemical selectivity, reaching a maximum 40:1 relative selectivity in ethylene/ethane production. Modified with permission from Ref. [5]. Copyright 2016 National Academy of Sciences.

A practical and rational integration of plasmon-mediated photocatalysis into widespread applications may be on the horizon, as the recent findings and scientific advancements have been heavily influential for highly efficient plasmonic catalytic platforms. To achieve this goal, it is imperative to direct a united front toward understanding the mechanism of plasmon decay and the subsequent preferential partitioning of the desired energy. Throughout this review, we hope the readers have grown aware of the nontrivial balance between this partitioning, as the plasmonically generated hot carriers, thermal elevations, and enhanced local electric fields can all play a role in mediating a given chemical reaction. Once the necessary understanding of this dynamic mechanism is obtained, we believe the field will find great success in designing highly tunable and selective plasmonic catalytic systems to carry out significant light-driven chemical reactions.

Acknowledgments

We acknowledge the support from the Air Force Office of Scientific Research under AFOSR award no. FA9550-15-1-0022. E.L.K. acknowledges the support from the University of Minnesota doctoral dissertation fellowship.

References

[1] Schultz DM, Yoon TP. Solar synthesis: prospects in visible light photocatalysis. Science 2014;343:1239176.10.1126/science.1239176Search in Google Scholar PubMed PubMed Central

[2] Fujishima A, Honda K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972;238:37.10.1038/238037a0Search in Google Scholar PubMed

[3] Pelaez M, Nolan NT, Pillai SC, et al. A review on the visible light active titanium dioxide photocatalysts for environmental applications. Appl Catal B Environ 2012;125:331–49.10.1016/j.apcatb.2012.05.036Search in Google Scholar

[4] Dry ME. High quality diesel via the Fischer-Tropsch process – a review. J Chem Technol Biotechnol 2002;77:43–50.10.1002/jctb.527Search in Google Scholar

[5] Swearer DF, Zhao H, Zhou L, et al. Heterometallic antenna-reactor complexes for photocatalysis. Proc Natl Acad Sci U S A 2016;113:8916–20.10.1073/pnas.1609769113Search in Google Scholar PubMed PubMed Central

[6] Galloway JN, Townsend AR, Erisman JW, et al. Transformation of the nitrogen cycle: recent trends, questions, and potential solutions. Science 2008;320:889–92.10.1126/science.1136674Search in Google Scholar PubMed

[7] Oshikiri T, Ueno K, Misawa H. Plasmon-induced ammonia synthesis through nitrogen photofixation with visible light irradiation. Angew Chem Int Ed Engl 2014;53:9802–5.10.1002/anie.201404748Search in Google Scholar PubMed

[8] Ali M, Zhou FL, Chen K, et al. Nanostructured photoelectrochemical solar cell for nitrogen reduction using plasmon-enhanced black silicon. Nat Commun 2016;7:11335.10.1038/ncomms11335Search in Google Scholar PubMed PubMed Central

[9] Tahir M, Tahir B, Amin NAS. Gold-nanoparticle-modified TiO₂ nanowires for plasmon-enhanced photocatalytic CO₂ reduction with H₂ under visible light irradiation. Appl Surf Sci 2015;356:1289–99.10.1016/j.apsusc.2015.08.231Search in Google Scholar

[10] Ursua A, Gandia LM, Sanchis P. Hydrogen production from water electrolysis: current status and future trends. Proc IEEE 2012;100:410–26.10.1109/JPROC.2011.2156750Search in Google Scholar

[11] Balat M. Potential importance of hydrogen as a future solution to environmental and transportation problems. Int J Hydrogen Energy 2008;33:4013–29.10.1016/j.ijhydene.2008.05.047Search in Google Scholar

[12] Song H, Meng X, Dao TD, et al. Light-enhanced carbon dioxide activation and conversion by effective plasmonic coupling effect of Pt and Au nanoparticles. ACS Appl Mater Interfaces 2018;10:408–16.10.1021/acsami.7b13043Search in Google Scholar PubMed

[13] Lee J, Mubeen S, Ji X, Stucky GD, Moskovits M. Plasmonic photoanodes for solar water splitting with visible light. Nano Lett 2012;12:5014–9.10.1021/nl302796fSearch in Google Scholar PubMed

[14] Adleman JR, Boyd DA, Goodwin DG, Psaltis D. Heterogenous catalysis mediated by plasmon heating. Nano Lett 2009;9:4417–23.10.1021/nl902711nSearch in Google Scholar PubMed

[15] Christopher P, Xin H, Linic S. Visible-light-enhanced catalytic oxidation reactions on plasmonic silver nanostructures. Nat Chem 2011;3:467–72.10.1038/nchem.1032Search in Google Scholar PubMed

[16] Hung WH, Aykol M, Valley D, Hou W, Cronin SB. Plasmon resonant enhancement of carbon monoxide catalysis. Nano Lett 2010;10:1314–8.10.1021/nl9041214Search in Google Scholar PubMed

[17] Liu Z, Hou W, Pavaskar P, Aykol M, Cronin SB. Plasmon resonant enhancement of photocatalytic water splitting under visible illumination. Nano Lett 2011;11:1111–6.10.1021/nl104005nSearch in Google Scholar PubMed

[18] Mukherjee S, Libisch F, Large N, et al. Hot electrons do the impossible: plasmon-induced dissociation of H₂ on Au. Nano Lett 2013;13:240–7.10.1021/nl303940zSearch in Google Scholar PubMed

[19] Zheng Z, Tachikawa T, Majima T. Plasmon-enhanced formic acid dehydrogenation using anisotropic Pd-Au nanorods studied at the single-particle level. J Am Chem Soc 2015;137:948–57.10.1021/ja511719gSearch in Google Scholar PubMed

[20] Hartland GV. Measurements of the material properties of metal nanoparticles by time-resolved spectroscopy. Phys Chem Chem Phys 2004;6:5263–74.10.1039/b413368dSearch in Google Scholar

[21] Aruda KO, Tagliazucchi M, Sweeney CM, Hannah DC, Weiss EA. The role of interfacial charge transfer-type interactions in the decay of plasmon excitations in metal nanoparticles. Phys Chem Chem Phys 2013;15:7441–9.10.1039/c3cp51005kSearch in Google Scholar PubMed

[22] Zhang Y, He S, Guo W, et al. Surface-plasmon-driven hot electron photochemistry. Chem Rev 2017;118:2927–54.10.1021/acs.chemrev.7b00430Search in Google Scholar PubMed

[23] Naik GV, Schroeder JL, Ni XJ, Kildishev AV, Sands TD, Boltasseva A. Titanium nitride as a plasmonic material for visible and near-infrared wavelengths. Opt Mater Express 2012;2:478–89.10.1364/OME.2.000478Search in Google Scholar

[24] Garcia G, Buonsanti R, Runnerstrom EL, et al. Dynamically modulating the surface plasmon resonance of doped semiconductor nanocrystals. Nano Lett 2011;11:4415–20.10.1021/nl202597nSearch in Google Scholar PubMed

[25] Zhao YX, Burda C. Development of plasmonic semiconductor nanomaterials with copper chalcogenides for a future with sustainable energy materials. Energy Environ Sci 2012;5:5564–76.10.1039/C1EE02734DSearch in Google Scholar

[26] Kelly KL, Coronado E, Zhao LL, Schatz GC. The optical properties of metal nanoparticles: the influence of size, shape, and dielectric environment. J Phys Chem B 2003;107:668–77.10.1021/jp026731ySearch in Google Scholar

[27] Schuller JA, Barnard ES, Cai W, Jun YC, White JS, Brongersma ML. Plasmonics for extreme light concentration and manipulation. Nat Mater 2010;9:193–204.10.1038/nmat2630Search in Google Scholar PubMed

[28] Aslam U, Chavez S, Linic S. Controlling energy flow in multimetallic nanostructures for plasmonic catalysis. Nat Nanotechnol 2017;12:1000–5.10.1038/nnano.2017.131Search in Google Scholar PubMed

[29] Cushing SK, Li J, Meng F, et al. Photocatalytic activity enhanced by plasmonic resonant energy transfer from metal to semiconductor. J Am Chem Soc 2012;134:15033–41.10.1021/ja305603tSearch in Google Scholar PubMed

[30] Yu S, Wilson AJ, Heo J, Jain PK. Plasmonic control of multi-electron transfer and C-C coupling in visible-light-driven CO₂ reduction on Au nanoparticles. Nano Lett 2018;18:2189–94.10.1021/acs.nanolett.7b05410Search in Google Scholar PubMed

[31] Ferry VE, Sweatlock LA, Pacifici D, Atwater HA. Plasmonic nanostructure design for efficient light coupling into solar cells. Nano Lett 2008;8:4391–7.10.1021/nl8022548Search in Google Scholar PubMed

[32] Ferry VE, Verschuuren MA, Li HB, et al. Light trapping in ultrathin plasmonic solar cells. Opt Express 2010;18(Suppl 2):A237–45.10.1364/OE.18.00A237Search in Google Scholar PubMed

[33] Gan Q, Bartoli FJ, Kafafi ZH. Plasmonic-enhanced organic photovoltaics: breaking the 10% efficiency barrier. Adv Mater 2013;25:2385–96.10.1002/adma.201203323Search in Google Scholar PubMed

[34] Munday JN, Atwater HA. Large integrated absorption enhancement in plasmonic solar cells by combining metallic gratings and antireflection coatings. Nano Lett 2011;11:2195–201.10.1021/nl101875tSearch in Google Scholar PubMed

[35] Pala RA, White J, Barnard E, Liu J, Brongersma ML. Design of plasmonic thin-film solar cells with broadband absorption enhancements. Adv Mater 2009;21:3504.10.1002/adma.200900331Search in Google Scholar

[36] Reineck P, Lee GP, Brick D, Karg M, Mulvaney P, Bach U. A solid-state plasmonic solar cell via metal nanoparticle self-assembly. Adv Mater 2012;24:4750–5.10.1002/adma.201200994Search in Google Scholar PubMed

[37] Wang W, Wu S, Reinhardt K, Lu Y, Chen S. Broadband light absorption enhancement in thin-film silicon solar cells. Nano Lett 2010;10:2012–8.10.1021/nl904057pSearch in Google Scholar PubMed

[38] Ali MR, Rahman MA, Wu Y, et al. Efficacy, long-term toxicity, and mechanistic studies of gold nanorods photothermal therapy of cancer in xenograft mice. Proc Natl Acad Sci U S A 2017;114:E3110–8.10.1073/pnas.1619302114Search in Google Scholar PubMed PubMed Central

[39] Chen J, Wang D, Xi J, et al. Immuno gold nanocages with tailored optical properties for targeted photothermal destruction of cancer cells. Nano Lett 2007;7:1318–22.10.1021/nl070345gSearch in Google Scholar PubMed PubMed Central

[40] Huang P, Lin J, Li W, et al. Biodegradable gold nanovesicles with an ultrastrong plasmonic coupling effect for photoacoustic imaging and photothermal therapy. Angew Chem Int Ed Engl 2013;52:13958–64.10.1002/anie.201308986Search in Google Scholar PubMed PubMed Central

[41] Huang X, El-Sayed IH, Qian W, El-Sayed MA. Cancer cell imaging and photothermal therapy in the near-infrared region by using gold nanorods. J Am Chem Soc 2006;128:2115–20.10.1021/ja057254aSearch in Google Scholar PubMed

[42] Loo C, Lin A, Hirsch L, et al. Nanoshell-enabled photonics-based imaging and therapy of cancer. Technol Cancer Res Treat 2004;3:33–40.10.1177/153303460400300104Search in Google Scholar PubMed

[43] Mackey MA, Ali MR, Austin LA, Near RD, El-Sayed MA. The most effective gold nanorod size for plasmonic photothermal therapy: theory and in vitro experiments. J Phys Chem B 2014;118:1319–26.10.1021/jp409298fSearch in Google Scholar PubMed

[44] Wang S, Huang P, Nie L, et al. Single continuous wave laser induced photodynamic/plasmonic photothermal therapy using photosensitizer-functionalized gold nanostars. Adv Mater 2013;25:3055–61.10.1002/adma.201204623Search in Google Scholar PubMed

[45] Hartland GV. Optical studies of dynamics in noble metal nanostructures. Chem Rev 2011;111:3858–87.10.1021/cr1002547Search in Google Scholar PubMed

[46] Kleinman SL, Sharma B, Blaber MG, et al. Structure enhancement factor relationships in single gold nanoantennas by surface-enhanced Raman excitation spectroscopy. J Am Chem Soc 2013;135:301–8.10.1021/ja309300dSearch in Google Scholar PubMed

[47] Lesina AC, Vaccari A, Berini P, Ramunno L. On the convergence and accuracy of the FDTD method for nanoplasmonics. Opt Express 2015;23:10481–97.10.1364/OE.23.010481Search in Google Scholar PubMed

[48] Jeanmaire DL, Van Duyne RP. Surface Raman spectroelectrochemistry: part 1. Heterocyclic, aromatic, and aliphatic-amines adsorbed on anodized silver electrode. J Electroanal Chem 1977;84:1–20.10.1016/S0022-0728(77)80224-6Search in Google Scholar

[49] Le Ru EC, Etchegoin P. Principles of surface-enhanced Raman spectroscopy and Related Plasmonic Effects. Oxford, UK, Elsevier Science, 2008.10.1016/B978-0-444-52779-0.00005-2Search in Google Scholar

[50] Hartstein A, Kirtley JR, Tsang JC. Enhancement of the infrared-absorption from molecular monolayers with thin metal overlayers. Phys Rev Lett 1980;45:201–4.10.1103/PhysRevLett.45.201Search in Google Scholar

[51] Osawa M, Ataka K, Yoshii K, Nishikawa Y. Surface-enhanced infrared-spectroscopy – the origin of the absorption enhancement and band selection rule in the infrared-spectra of molecules adsorbed on fine metal particles. Appl Spectrosc 1993;47:1497–502.10.1366/0003702934067478Search in Google Scholar

[52] Liebermann T, Knoll W. Surface-plasmon field-enhanced fluorescence spectroscopy. Colloid Surf A 2000;171:115–30.10.1016/S0927-7757(99)00550-6Search in Google Scholar

[53] Gersten J, Nitzan A. Spectroscopic properties of molecules interacting with small dielectric particles. J Chem Phys 1981;75:1139–52.10.1063/1.442161Search in Google Scholar

[54] Li XG, Xiao D, Zhang ZY. Landau damping of quantum plasmons in metal nanostructures. N J Phys 2013;15:023011.10.1088/1367-2630/15/2/023011Search in Google Scholar

[55] Manjavacas A, Liu JG, Kulkarni V, Nordlander P. Plasmon-induced hot carriers in metallic nanoparticles. ACS Nano 2014;8:7630–8.10.1021/nn502445fSearch in Google Scholar PubMed

[56] Brongersma ML, Halas NJ, Nordlander P. Plasmon-induced hot carrier science and technology. Nat Nanotechnol 2015;10:25–34.10.1038/nnano.2014.311Search in Google Scholar PubMed

[57] Link S, El-Sayed MA. Size and temperature dependence of the plasmon absorption of colloidal gold nanoparticles. J Phys Chem B 1999;103:4212–7.10.1021/jp984796oSearch in Google Scholar

[58] Sonnichsen C, Franzl T, Wilk T, et al. Drastic reduction of plasmon damping in gold nanorods. Phys Rev Lett 2002;88:077402.10.1103/PhysRevLett.88.077402Search in Google Scholar PubMed

[59] Ahmadi TS, Logunov SL, ElSayed MA. Picosecond dynamics of colloidal gold nanoparticles. J Phys Chem 1996;100:8053–6.10.1021/jp960484eSearch in Google Scholar

[60] Aruda KO, Tagliazucchi M, Sweeney CM, Hannah DC, Schatz GC, Weiss EA. Identification of parameters through which surface chemistry determines the lifetimes of hot electrons in small Au nanoparticles. Proc Natl Acad Sci U S A 2013;110:4212–7.10.1073/pnas.1222327110Search in Google Scholar PubMed PubMed Central

[61] Li K, Hogan NJ, Kale MJ, Halas NJ, Nordlander P, Christopher P. Balancing near-field enhancement, absorption, and scattering for effective antenna-reactor plasmonic photocatalysis. Nano Lett 2017;17:3710–7.10.1021/acs.nanolett.7b00992Search in Google Scholar PubMed

[62] Harris N, Ford MJ, Cortie MB. Optimization of plasmonic heating by gold nanospheres and nanoshells. J Phys Chem B 2006;110:10701–7.10.1021/jp0606208Search in Google Scholar PubMed

[63] Herzog JB, Knight MW, Natelson D. Thermoplasmonics: quantifying plasmonic heating in single nanowires. Nano Lett 2014;14:499–503.10.1021/nl403510uSearch in Google Scholar PubMed

[64] Pissuwan D, Valenzuela SM, Cortie MB. Therapeutic possibilities of plasmonically heated gold nanoparticles. Trends Biotechnol 2006;24:62–7.10.1016/j.tibtech.2005.12.004Search in Google Scholar PubMed

[65] Neumann O, Feronti C, Neumann AD, et al. Compact solar autoclave based on steam generation using broadband light-harvesting nanoparticles. Proc Natl Acad Sci U S A 2013;110:11677–81.10.1073/pnas.1310131110Search in Google Scholar PubMed PubMed Central

[66] Zharov VP, Mercer KE, Galitovskaya EN, Smeltzer MS. Photothermal nanotherapeutics and nanodiagnostics for selective killing of bacteria targeted with gold nanoparticles. Biophys J 2006;90:619–27.10.1529/biophysj.105.061895Search in Google Scholar PubMed PubMed Central

[67] Cao L, Barsic DN, Guichard AR, Brongersma ML. Plasmon-assisted local temperature control to pattern individual semiconductor nanowires and carbon nanotubes. Nano Lett 2007;7:3523–7.10.1021/nl0722370Search in Google Scholar PubMed

[68] Rontzsch L, Heinig KH, Schuller JA, Brongersma ML. Thin film patterning by surface-plasmon-induced thermocapillarity. Appl Phys Lett 2007;90:044105.10.1063/1.2432282Search in Google Scholar

[69] Zhang X, Li X, Reish ME, et al. Plasmon-enhanced catalysis: distinguishing thermal and nonthermal effects. Nano Lett 2018;18:1714–23.10.1021/acs.nanolett.7b04776Search in Google Scholar PubMed

[70] Khurgin JB. How to deal with the loss in plasmonics and metamaterials. Nat Nanotechnol 2015;10:2–6.10.1038/nnano.2014.310Search in Google Scholar PubMed

[71] Boerigter C, Aslam U, Linic S. Mechanism of charge transfer from plasmonic nanostructures to chemically attached materials. ACS Nano 2016;10:6108–15.10.1021/acsnano.6b01846Search in Google Scholar PubMed

[72] Kazuma E, Jung J, Ueba H, Trenary M, Kim Y. Real-space and real-time observation of a plasmon-induced chemical reaction of a single molecule. Science 2018;360:521–6.10.1126/science.aao0872Search in Google Scholar PubMed

[73] Somorjai GA, Li Y. Introduction to surface chemistry and catalysis. Hoboken, New Jersey, John Wiley & Sons, 2010.Search in Google Scholar

[74] Kolasinski KW. Surface science: foundations of catalysis and nanoscience. Chichester, West Sussex, UK, John Wiley & Sons, 2012.10.1002/9781119941798Search in Google Scholar

[75] Ertl G, Knözinger H, Weitkamp J. Handbook of heterogeneous catalysis. Weinheim, Germany, Wiley-VCH, 1997.10.1002/9783527619474Search in Google Scholar

[76] Bonn M, Funk S, Hess C, et al. Phonon-versus electron-mediated desorption and oxidation of Co on Ru(0001). Science 1999;285:1042–5.10.1126/science.285.5430.1042Search in Google Scholar PubMed

[77] Diebold U. The surface science of titanium dioxide. Surf Sci Rep 2003;48:153–229.10.1016/S0167-5729(02)00100-0Search in Google Scholar

[78] Stiles PL, Dieringer JA, Shah NC, Van Duyne RP. Surface-enhanced Raman spectroscopy. Annu Rev Anal Chem 2008;1:601–26.10.1146/annurev.anchem.1.031207.112814Search in Google Scholar PubMed

[79] Moskovits M. Surface-enhanced spectroscopy. Rev Mod Phys 1985;57:783–826.10.1103/RevModPhys.57.783Search in Google Scholar

[80] Goddard G, Brown LO, Habbersett R, et al. High-resolution spectral analysis of individual SERS-active nanoparticles in flow. J Am Chem Soc 2010;132:6081–90.10.1021/ja909850sSearch in Google Scholar PubMed PubMed Central

[81] Fang Y, Seong NH, Dlott DD. Measurement of the distribution of site enhancements in surface-enhanced Raman scattering. Science 2008;321:388–92.10.1126/science.1159499Search in Google Scholar PubMed

[82] Kleinman SL, Frontiera RR, Henry AI, Dieringer JA, Van Duyne RP. Creating, characterizing, and controlling chemistry with SERS hot spots. Phys Chem Chem Phys 2013;15:21–36.10.1039/C2CP42598JSearch in Google Scholar PubMed

[83] Brooks JL, Frontiera RR. Competition between reaction and degradation pathways in plasmon-driven photochemistry. J Phys Chem C 2016;120:20869–76.10.1021/acs.jpcc.6b02314Search in Google Scholar

[84] Li P, Ma B, Yang L, Liu J. Hybrid single nanoreactor for in situ SERS monitoring of plasmon-driven and small Au nanoparticles catalyzed reactions. Chem Commun 2015;51:11394–7.10.1039/C5CC03792ASearch in Google Scholar PubMed

[85] Liu HW, Zhang L, Lang XY, et al. Single molecule detection from a large-scale SERS-active Au79Ag21 substrate. Sci Rep 2011;1:112.10.1038/srep00112Search in Google Scholar PubMed PubMed Central

[86] Nie S, Emory SR. Probing single molecules and single nanoparticles by surface-enhanced Raman scattering. Science 1997;275:1102–6.10.1126/science.275.5303.1102Search in Google Scholar PubMed

[87] Petti L, Capasso R, Rippa M, et al. A plasmonic nanostructure fabricated by electron beam lithography as a sensitive and highly homogeneous SERS substrate for bio-sensing applications. Vib Spectrosc 2016;82:22–30.10.1016/j.vibspec.2015.11.007Search in Google Scholar

[88] Xie W, Walkenfort B, Schlucker S. Label-free SERS monitoring of chemical reactions catalyzed by small gold nanoparticles using 3D plasmonic superstructures. J Am Chem Soc 2013;135:1657–60.10.1021/ja309074aSearch in Google Scholar PubMed

[89] Yin Z, Wang Y, Song C, et al. Hybrid Au-Ag nanostructures for enhanced plasmon-driven catalytic selective hydrogenation through visible light irradiation and surface-enhanced Raman scattering. J Am Chem Soc 2018;140:864–7.10.1021/jacs.7b11293Search in Google Scholar PubMed

[90] Zhang X, Dai Z, Si S, et al. Ultrasensitive SERS substrate integrated with uniform subnanometer scale “hot spots” created by a graphene spacer for the detection of mercury ions. Small 2017;13:1603347.10.1002/smll.201603347Search in Google Scholar

[91] Joshi GK, White SL, Johnson MA, Sardar R, Jain PK. Ultrashort, angstrom-scale decay of surface-enhanced Raman scattering at hot spots. J Phys Chem C 2016;120:24973–81.10.1021/acs.jpcc.6b08242Search in Google Scholar

[92] Masango SS, Hackler RA, Large N, et al. High-resolution distance dependence study of surface-enhanced Raman scattering enabled by atomic layer deposition. Nano Lett 2016;16:4251–9.10.1021/acs.nanolett.6b01276Search in Google Scholar PubMed

[93] Shin KS, Lee HS, Joo SW, Kim K. Surface-induced photoreduction of 4-nitrobenzenethiol on Cu revealed by surface-enhanced Raman scattering spectroscopy. J Phys Chem C 2007;111:15223–7.10.1021/jp073053cSearch in Google Scholar

[94] Liu X, Tang L, Niessner R, Ying Y, Haisch C. Nitrite-triggered surface plasmon-assisted catalytic conversion of p-aminothiophenol to p,p′-dimercaptoazobenzene on gold nanoparticle: surface-enhanced Raman scattering investigation and potential for nitrite detection. Anal Chem 2015;87:499–506.10.1021/ac5039576Search in Google Scholar

[95] Lee SJ, Kim K. Surface-induced photoreaction of 4-nitrobenzenethiol on silver: influence of SERS-active sites. Chem Phys Lett 2003;378:122–7.10.1016/S0009-2614(03)01269-7Search in Google Scholar

[96] Kim K, Choi JY, Shin KS. Photoreduction of 4-nitrobenzenethiol on Au by hot electrons plasmonically generated from Ag nanoparticles: gap-mode surface-enhanced Raman scattering observation. J Phys Chem C 2015;119:5187–94.10.1021/acs.jpcc.5b00033Search in Google Scholar

[97] Huang YF, Zhu HP, Liu GK, Wu DY, Ren B, Tian ZQ. When the signal is not from the original molecule to be detected: chemical transformation of para-aminothiophenol on Ag during the SERS measurement. J Am Chem Soc 2010;132:9244–6.10.1021/ja101107zSearch in Google Scholar PubMed

[98] Dong B, Fang Y, Xia L, Xu H, Sun M. Is 4-nitrobenzenethiol converted to p,p′-dimercaptoazobenzene or 4-aminothiophenol by surface photochemistry reaction? J Raman Spectrosc 2011;42:1205–6.10.1002/jrs.2937Search in Google Scholar

[99] Dong B, Fang Y, Chen X, Xu H, Sun M. Substrate-, wavelength-, and time-dependent plasmon-assisted surface catalysis reaction of 4-nitrobenzenethiol dimerizing to p,p′-dimercaptoazobenzene on Au, Ag, and Cu films. Langmuir 2011;27:10677–82.10.1021/la2018538Search in Google Scholar PubMed

[100] Sundararaman R, Narang P, Jermyn AS, Goddard WA, Atwater HA. Theoretical predictions for hot-carrier generation from surface plasmon decay. Nat Commun 2014;5:5788.10.1038/ncomms6788Search in Google Scholar PubMed PubMed Central

[101] Kneipp K, Wang Y, Kneipp H, et al. Single molecule detection using surface-enhanced Raman scattering (SERS). Phys Rev Lett 1997;78:1667–70.10.1103/PhysRevLett.78.1667Search in Google Scholar

[102] Kleinman SL, Ringe E, Valley N, et al. Single-molecule surface-enhanced Raman spectroscopy of crystal violet isotopologues: theory and experiment. J Am Chem Soc 2011;133:4115–22.10.1021/ja110964dSearch in Google Scholar PubMed

[103] Camden JP, Dieringer JA, Wang Y, et al. Probing the structure of single-molecule surface-enhanced Raman scattering hot spots. J Am Chem Soc 2008;130:12616–7.10.1021/ja8051427Search in Google Scholar PubMed

[104] Le Ru EC, Meyer M, Etchegoin PG. Proof of single-molecule sensitivity in surface enhanced Raman scattering (SERS) by means of a two-analyte technique. J Phys Chem B 2006;110:1944–8.10.1021/jp054732vSearch in Google Scholar PubMed

[105] Dieringer JA, Lettan RB, 2nd, Scheidt KA, Van Duyne RP. A frequency domain existence proof of single-molecule surface-enhanced Raman spectroscopy. J Am Chem Soc 2007;129:16249–56.10.1021/ja077243cSearch in Google Scholar PubMed

[106] Xu HX, Bjerneld EJ, Kall M, Borjesson L. Spectroscopy of single hemoglobin molecules by surface enhanced Raman scattering. Phys Rev Lett 1999;83:4357–60.10.1103/PhysRevLett.83.4357Search in Google Scholar

[107] Kneipp K, Kneipp H, Deinum G, Itzkan I, Dasari RR, Feld MS. Single-molecule detection of a cyanine dye in silver colloidal solution using near-infrared surface-enhanced Raman scattering. Appl Spectrosc 2016;52:175–8.10.1366/0003702981943275Search in Google Scholar

[108] Sprague-Klein EA, McAnally MO, Zhdanov DV, et al. Observation of single molecule plasmon-driven electron transfer in isotopically edited 4,4′-bipyridine gold nanosphere oligomers. J Am Chem Soc 2017;139:15212–21.10.1021/jacs.7b08868Search in Google Scholar PubMed

[109] Choi HK, Park WH, Park CG, Shin HH, Lee KS, Kim ZH. Metal-catalyzed chemical reaction of single molecules directly probed by vibrational spectroscopy. J Am Chem Soc 2016;138:4673–84.10.1021/jacs.6b01865Search in Google Scholar PubMed

[110] de Nijs B, Benz F, Barrow SJ, et al. Plasmonic tunnel junctions for single-molecule redox chemistry. Nat Commun 2017;8:994.10.1038/s41467-017-00819-7Search in Google Scholar PubMed PubMed Central

[111] Sonntag MD, Chulhai D, Seideman T, Jensen L, Van Duyne RP. The origin of relative intensity fluctuations in single-molecule tip-enhanced Raman spectroscopy. J Am Chem Soc 2013;135:17187–92.10.1021/ja408758jSearch in Google Scholar PubMed

[112] Zrimsek AB, Wong NL, Van Duyne RP. Single molecule surface-enhanced Raman spectroscopy: a critical analysis of the bianalyte versus isotopologue proof. J Phys Chem C 2016;120:5133–42.10.1021/acs.jpcc.6b00606Search in Google Scholar

[113] Taylor RW, Lee TC, Scherman OA, et al. Precise subnanometer plasmonic junctions for SERS within gold nanoparticle assemblies using cucurbit[n]uril “Glue”. ACS Nano 2011;5:3878–87.10.1021/nn200250vSearch in Google Scholar PubMed

[114] Kim NH, Hwang W, Baek K, et al. Smart SERS hot spots: single molecules can be positioned in a plasmonic nanojunction using host-guest chemistry. J Am Chem Soc 2018;140:4705–11.10.1021/jacs.8b01501Search in Google Scholar

[115] Wessel J. Surface-enhanced optical microscopy. J Opt Soc Am B 1985;2:1538–41.10.1364/JOSAB.2.001538Search in Google Scholar

[116] Anderson MS. Locally enhanced Raman spectroscopy with an atomic force microscope. Appl Phys Lett 2000;76:3130–2.10.1063/1.126546Search in Google Scholar

[117] Hayazawa N, Inouye Y, Sekkat Z, Kawata S. Metallized tip amplification of near-field Raman scattering. Opt Commun 2000;183:333–6.10.1016/S0030-4018(00)00894-4Search in Google Scholar

[118] Stockle RM, Suh YD, Deckert V, Zenobi R. Nanoscale chemical analysis by tip-enhanced Raman spectroscopy. Chem Phys Lett 2000;318:131–6.10.1016/S0009-2614(99)01451-7Search in Google Scholar

[119] Sonntag MD, Klingsporn JM, Garibay LK, et al. Single-molecule tip-enhanced Raman spectroscopy. J Phys Chem C 2011;116:478–83.10.1021/jp209982hSearch in Google Scholar

[120] Bailo E, Deckert V. Tip-enhanced Raman spectroscopy of single RNA strands: towards a novel direct-sequencing method. Angew Chem Int Ed Engl 2008;47:1658–61.10.1002/anie.200704054Search in Google Scholar PubMed

[121] Zhang Z, Sheng S, Wang R, Sun M. Tip-enhanced Raman spectroscopy. Anal Chem 2016;88:9328–46.10.1021/acs.analchem.6b02093Search in Google Scholar PubMed

[122] Hayazawa N, Saito Y, Kawata S, Picardi G, Schuster R. Detection and characterization of longitudinal field for tip-enhanced Raman spectroscopy. Appl Phys Lett 2004;85:6239–41.10.1063/1.1839646Search in Google Scholar

[123] Pettinger B, Domke KF, Zhang D, Picardi G, Schuster R. Tip-enhanced Raman scattering: influence of the tip-surface geometry on optical resonance and enhancement. Surf Sci 2009;603:1335–41.10.1016/j.susc.2008.08.033Search in Google Scholar

[124] Kazemi-Zanjani N, Vedraine S, Lagugne-Labarthet F. Localized enhancement of electric field in tip-enhanced Raman spectroscopy using radially and linearly polarized light. Opt Express 2013;21:25271–6.10.1364/OE.21.025271Search in Google Scholar PubMed

[125] Deckert-Gaudig T, Richter M, Knebel D, et al. A modified transmission tip-enhanced Raman scattering (TERS) setup provides access to opaque samples. Appl Spectrosc 2014;68:916–9.10.1366/13-07419Search in Google Scholar PubMed

[126] Sun M, Zhang Z, Zheng H, Xu H. In-situ plasmon-driven chemical reactions revealed by high vacuum tip-enhanced Raman spectroscopy. Sci Rep 2012;2:647.10.1038/srep00647Search in Google Scholar PubMed PubMed Central

[127] van Schrojenstein Lantman EM, Deckert-Gaudig T, Mank AJG, Deckert V, Weckhuysen BM. Catalytic processes monitored at the nanoscale with tip-enhanced Raman spectroscopy. Nat Nanotechnol 2012;7:583.10.1038/nnano.2012.131Search in Google Scholar PubMed

[128] Zhang Z, Chen L, Sun M, Ruan P, Zheng H, Xu H. Insights into the nature of plasmon-driven catalytic reactions revealed by HV-TERS. Nanoscale 2013;5:3249–52.10.1039/c3nr00352cSearch in Google Scholar PubMed

[129] Kumar N, Stephanidis B, Zenobi R, Wain AJ, Roy D. Nanoscale mapping of catalytic activity using tip-enhanced Raman spectroscopy. Nanoscale 2015;7:7133–7.10.1039/C4NR07441FSearch in Google Scholar PubMed

[130] Yeo B-S, Stadler J, Schmid T, Zenobi R, Zhang W. Tip-enhanced Raman spectroscopy – its status, challenges and future directions. Chem Phys Lett 2009;472:1–13.10.1016/j.cplett.2009.02.023Search in Google Scholar

[131] Ren B, Picardi G, Pettinger B. Preparation of gold tips suitable for tip-enhanced Raman spectroscopy and light emission by electrochemical etching. Rev Sci Instrum 2004;75:837–41.10.1063/1.1688442Search in Google Scholar

[132] Brandt NC, Keller EL, Frontiera RR. Ultrafast surface-enhanced Raman probing of the role of hot electrons in plasmon-driven chemistry. J Phys Chem Lett 2016;7:3179–85.10.1021/acs.jpclett.6b01453Search in Google Scholar PubMed

[133] Keller EL, Frontiera RR. Monitoring charge density delocalization upon plasmon excitation with ultrafast surface-enhanced Raman spectroscopy. ACS Photonics 2017;4:1033–9.10.1021/acsphotonics.7b00082Search in Google Scholar

[134] Gruenke NL, McAnally MO, Schatz GC, Van Duyne RP. Balancing the effects of extinction and enhancement for optimal signal in surface-enhanced femtosecond stimulated Raman spectroscopy. J Phys Chem C 2016;120:29449–54.10.1021/acs.jpcc.6b10727Search in Google Scholar

[135] Zhang Y, Zhen YR, Neumann O, Day JK, Nordlander P, Halas NJ. Coherent anti-Stokes Raman scattering with single-molecule sensitivity using a plasmonic Fano resonance. Nat Commun 2014;5:4424.10.1038/ncomms5424Search in Google Scholar PubMed

[136] Crampton KT, Zeytunyan A, Fast AS, et al. Ultrafast coherent Raman scattering at plasmonic nanojunctions. J Phys Chem C 2016;120:20943–53.10.1021/acs.jpcc.6b02760Search in Google Scholar

[137] Yampolsky S, Fishman DA, Dey S, et al. Seeing a single molecule vibrate through time-resolved coherent anti-Stokes Raman scattering. Nat Photonics 2014;8:650–6.10.1038/nphoton.2014.143Search in Google Scholar

[138] Keller EL, Brandt NC, Cassabaum AA, Frontiera RR. Ultrafast surface-enhanced Raman spectroscopy. Analyst 2015;140:4922–31.10.1039/C5AN00869GSearch in Google Scholar PubMed

[139] Gruenke NL, Cardinal MF, McAnally MO, Frontiera RR, Schatz GC, Van Duyne RP. Ultrafast and nonlinear surface-enhanced Raman spectroscopy. Chem Soc Rev 2016;45:2263–90.10.1039/C5CS00763ASearch in Google Scholar PubMed

[140] Frontiera RR, Gruenke NL, Van Duyne RP. Fano-like resonances arising from long-lived molecule-plasmon interactions in colloidal nanoantennas. Nano Lett 2012;12:5989–94.10.1021/nl303488mSearch in Google Scholar PubMed

[141] Ding SY, You EM, Tian ZQ, Moskovits M. Electromagnetic theories of surface-enhanced Raman spectroscopy. Chem Soc Rev 2017;46:4042–76.10.1039/C7CS00238FSearch in Google Scholar PubMed

[142] Frontiera RR, Henry AI, Gruenke NL, Van Duyne RP. Surface-enhanced femtosecond stimulated Raman spectroscopy. J Phys Chem Lett 2011;2:1199–203.10.1021/jz200498zSearch in Google Scholar PubMed

[143] Mandal A, Erramilli S, Ziegler LD. Origin of dispersive line shapes in plasmonically enhanced femtosecond stimulated Raman spectra. J Phys Chem C 2016;120:20998–1006.10.1021/acs.jpcc.6b03303Search in Google Scholar

[144] McAnally MO, McMahon JM, Van Duyne RP, Schatz GC. Coupled wave equations theory of surface-enhanced femtosecond stimulated Raman scattering. J Chem Phys 2016;145:094106.10.1063/1.4961749Search in Google Scholar PubMed

[145] Itoh T, Yamamoto YS. Recent topics on single-molecule fluctuation analysis using blinking in surface-enhanced resonance Raman scattering: clarification by the electromagnetic mechanism. Analyst 2016;141:5000–9.10.1039/C6AN00936KSearch in Google Scholar PubMed

[146] Lombardi JR, Birke RL, Haran G. Single molecule SERS spectral blinking and vibronic coupling. J Phys Chem C 2011;115:4540–5.10.1021/jp111345uSearch in Google Scholar

[147] Berera R, van Grondelle R, Kennis JT. Ultrafast transient absorption spectroscopy: principles and application to photosynthetic systems. Photosynth Res 2009;101:105–18.10.1007/s11120-009-9454-ySearch in Google Scholar PubMed PubMed Central

[148] Ruckebusch C, Sliwa M, Pernot P, de Juan A, Tauler R. Comprehensive data analysis of femtosecond transient absorption spectra: a review. J Photochem Photobiol C 2012;13:1–27.10.1016/j.jphotochemrev.2011.10.002Search in Google Scholar

[149] Furube A, Hashimoto S. Insight into plasmonic hot-electron transfer and plasmon molecular drive: new dimensions in energy conversion and nanofabrication. NPG Asia Mater 2017;9:e454.10.1038/am.2017.191Search in Google Scholar

[150] Jiang R, Li B, Fang C, Wang J. Metal/semiconductor hybrid nanostructures for plasmon-enhanced applications. Adv Mater 2014;26:5274–309.10.1002/adma.201400203Search in Google Scholar PubMed

[151] Wu K, Chen J, McBride JR, Lian T. Efficient hot-electron transfer by a plasmon-induced interfacial charge-transfer transition. Science 2015;349:632–5.10.1126/science.aac5443Search in Google Scholar PubMed

[152] Li JT, Cushing SK, Meng FK, Senty TR, Bristow AD, Wu NQ. Plasmon-induced resonance energy transfer for solar energy conversion. Nat Photonics 2015;9:601.10.1038/nphoton.2015.142Search in Google Scholar

[153] Li J, Cushing SK, Zheng P, Meng F, Chu D, Wu N. Plasmon-induced photonic and energy-transfer enhancement of solar water splitting by a hematite nanorod array. Nat Commun 2013;4:2651.10.1038/ncomms3651Search in Google Scholar PubMed

[154] Cushing SK, Li J, Bright J, et al. Controlling plasmon-induced resonance energy transfer and hot electron injection processes in metal@TiO₂ core-shell nanoparticles. J Phys Chem C 2015;119:16239–44.10.1021/acs.jpcc.5b03955Search in Google Scholar

[155] Yu Y, Sundaresan V, Willets KA. Hot carriers versus thermal effects: resolving the enhancement mechanisms for plasmon-mediated photoelectrochemical reactions. J Phys Chem C 2018;122:5040–8.10.1021/acs.jpcc.7b12080Search in Google Scholar

[156] Yu Y, Williams JD, Willets K. Quantifying photothermal heating at plasmonic nanoparticles by scanning electrochemical microscopy. Faraday Discuss 2018. DOI: 10.1039/C8FD0005.Search in Google Scholar

[157] Bard AJ. Electrochemical methods: fundamentals and applications. New York, Wiley, 1980.Search in Google Scholar

[158] Chirea M, Collins SS, Wei X, Mulvaney P. Spectroelectrochemistry of silver deposition on single gold nanocrystals. J Phys Chem Lett 2014;5:4331–5.10.1021/jz502349xSearch in Google Scholar PubMed

[159] Hoener BS, Zhang H, Heiderscheit TS, et al. Spectral response of plasmonic gold nanoparticles to capacitive charging: morphology effects. J Phys Chem Lett 2017;8:2681–8.10.1021/acs.jpclett.7b00945Search in Google Scholar PubMed

[160] Novo C, Funston AM, Gooding AK, Mulvaney P. Electrochemical charging of single gold nanorods. J Am Chem Soc 2009;131:14664–6.10.1021/ja905216hSearch in Google Scholar PubMed

[161] Brown AM, Sheldon MT, Atwater HA. Electrochemical tuning of the dielectric function of Au nanoparticles. ACS Photonics 2015;2:459–64.10.1021/ph500358qSearch in Google Scholar

[162] Byers CP, Hoener BS, Chang WS, Yorulmaz M, Link S, Landes CF. Single-particle spectroscopy reveals heterogeneity in electrochemical tuning of the localized surface plasmon. J Phys Chem B 2014;118:14047–55.10.1021/jp504454ySearch in Google Scholar PubMed PubMed Central

[163] Collins SS, Wei X, McKenzie TG, Funston AM, Mulvaney P. Single gold nanorod charge modulation in an ion gel device. Nano Lett 2016;16:6863–9.10.1021/acs.nanolett.6b02696Search in Google Scholar PubMed

[164] Hoener BS, Byers CP, Heiderscheit TS, et al. Spectroelectrochemistry of halide anion adsorption and dissolution of single gold nanorods. J Phys Chem C 2016;120:20604–12.10.1021/acs.jpcc.6b00650Search in Google Scholar

[165] Ingram DB, Linic S. Water splitting on composite plasmonic-metal/semiconductor photoelectrodes: evidence for selective plasmon-induced formation of charge carriers near the semiconductor surface. J Am Chem Soc 2011;133:5202–5.10.1021/ja200086gSearch in Google Scholar PubMed

[166] Robatjazi H, Bahauddin SM, Doiron C, Thomann I. Direct plasmon-driven photoelectrocatalysis. Nano Lett 2015;15:6155–61.10.1021/acs.nanolett.5b02453Search in Google Scholar PubMed

[167] Thomann I, Pinaud BA, Chen Z, Clemens BM, Jaramillo TF, Brongersma ML. Plasmon enhanced solar-to-fuel energy conversion. Nano Lett 2011;11:3440–6.10.1021/nl201908sSearch in Google Scholar PubMed

[168] Zheng JW, Lu TH, Cotton TM, Chumanov G. Photoinduced electrochemical reduction of nitrite at an electrochemically roughened silver surface. J Phys Chem B 1999;103:6567–72.10.1021/jp990928hSearch in Google Scholar

[169] Wilson AJ, Molina NY, Willets KA. Modification of the electrochemical properties of Nile Blue through covalent attachment to gold as revealed by electrochemistry and SERS. J Phys Chem C 2016;120:21091–8.10.1021/acs.jpcc.6b03962Search in Google Scholar

[170] Zaleski S, Wilson AJ, Mattei M, et al. Investigating nanoscale electrochemistry with surface- and tip-enhanced Raman spectroscopy. Acc Chem Res 2016;49:2023–30.10.1021/acs.accounts.6b00327Search in Google Scholar PubMed

[171] Kurouski D, Mattei M, Van Duyne RP. Probing redox reactions at the nanoscale with electrochemical tip-enhanced Raman spectroscopy. Nano Lett 2015;15:7956–62.10.1021/acs.nanolett.5b04177Search in Google Scholar PubMed

[172] Di Martino G, Turek VA, Lombardi A, et al. Tracking nanoelectrochemistry using individual plasmonic nanocavities. Nano Lett 2017;17:4840–5.10.1021/acs.nanolett.7b01676Search in Google Scholar PubMed

[173] Marbella LE, Millstone JE. NMR techniques for noble metal nanoparticles. Chem Mater 2015;27:2721–39.10.1021/cm504809cSearch in Google Scholar

[174] Liu JG, Zhang H, Link S, Nordlander P. Relaxation of plasmon-induced hot carriers. ACS Photonics 2017;5:2584–95.10.1021/acsphotonics.7b00881Search in Google Scholar

[175] Libisch F, Cheng J, Carter EA. Electron-transfer-induced dissociation of H₂ on gold nanoparticles: excited-state potential energy surfaces via embedded correlated wavefunction theory. Z Phys Chem 2013;227:1455.10.1524/zpch.2013.0406Search in Google Scholar

[176] Martirez JM, Carter EA. Excited-state N₂ dissociation pathway on Fe-functionalized Au. J Am Chem Soc 2017;139:4390–8.10.1021/jacs.6b12301Search in Google Scholar PubMed

[177] Martirez JMP, Carter EA. Prediction of a low-temperature N₂ dissociation catalyst exploiting near-IR-to-visible light nanoplasmonics. Sci Adv 2017;3:eaao4710.10.1126/sciadv.aao4710Search in Google Scholar PubMed PubMed Central

[178] Chulhai DV, Hu Z, Moore JE, Chen X, Jensen L. Theory of linear and nonlinear surface-Enhanced Vibrational Spectroscopies. Annu Rev Phys Chem 2016;67:541–64.10.1146/annurev-physchem-040215-112347Search in Google Scholar PubMed

[179] Hu Z, Chulhai DV, Jensen L. Simulating surface-enhanced hyper-Raman scattering using atomistic electrodynamics-quantum mechanical models. J Chem Theory Comput 2016;12:5968–78.10.1021/acs.jctc.6b00940Search in Google Scholar PubMed

[180] Litz JP, Brewster RP, Lee AB, Masiello DJ. Molecular-electronic structure in a plasmonic environment: elucidating the quantum image interaction. J Phys Chem C 2013;117:12249–57.10.1021/jp403047tSearch in Google Scholar

[181] Neuman T, Esteban R, Casanova D, Garcia-Vidal FJ, Aizpurua J. Coupling of molecular emitters and plasmonic cavities beyond the point-dipole approximation. Nano Lett 2018;18:2358–64.10.1021/acs.nanolett.7b05297Search in Google Scholar PubMed

[182] Ben-Nun M, Quenneville J, Martínez TJ. Ab initio multiple spawning: photochemistry from first principles quantum molecular dynamics. J Phys Chem A 2000;104:5161–75.10.1021/jp994174iSearch in Google Scholar

Received: 2018-06-16

Revised: 2018-07-26

Accepted: 2018-08-06

Published Online: 2018-08-29

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Toward a mechanistic understanding of plasmon-mediated photocatalysis

Abstract

1 Introduction

2 Fundamentals of plasmon generation and decay

2.1 Surface plasmons

2.2 Plasmonic energy partitioning

2.2.1 Enhanced electromagnetic fields

2.2.2 Hot carrier generation

2.2.3 Localized heating

2.2.4 Modified molecular potential energy surfaces

3 Select methods to elucidate the mechanism of plasmon-mediated photocatalysis

3.1 Spectroscopy

3.1.1 Surface-enhanced Raman spectroscopy (SERS)

3.1.1.1 Ensemble SERS

3.1.1.2 Single-molecule SERS (SMSERS)

3.1.2 Tip-enhanced Raman spectroscopy (TERS)

3.1.2 Ultrafast SERS

3.1.4 Transient absorption (TA)

3.2 Electrochemistry

4 Outlook

Acknowledgments

References

Journal and Issue

Articles in the same Issue