Highlights from Faraday Discussion 300: hot electron science and microscopic processes in plasmonics and catalysis, London, UK, February 2019

Abstract

The 300th Faraday discussion on hot electron processes in plasmonics and catalysis brought together a wide spectrum of delegates from across the globe. Scientists from across multiple disciplines, with a broad spectrum of experience and expertise, joined together in hope of identifying and answering the latest problems within the ever growing hot electron science community. In this paper we present a brief overview of the themes and topics presented at this meeting, highlighting the major findings of each presentation.

---

Discussions day 1

The ability to effectively confine strong electromagnetic (EM) fields within nanoscale cavities through plasmonic systems, beyond optical diffraction limits, has led to many recent advances and discoveries in science and technology. Although plasmonic systems have been relatively well understood and used for decades, this discovery brought about a rapid new wave of multi-disciplinary development of plasmonic devices over the last quarter of a decade.

More recently, the realization that the electronic parts of plasmonic decays can be exploited for physical and chemical processes has captured the focus of publications. Namely, a decay from a plasmonic excitation can allow an energy transfer to an electron–hole pair within the material. For a short period of time, generally understood to be below a picosecond, the carriers are able to stay “hot” in the sense of a non-equilibrium distribution of energy. The use of plasmonic nanostructures has allowed the extraction of these hot carriers prior to lattice thermalization. The past three years specifically has shown a range of fascinating applications of hot carriers experimentally. In addition, the theoretical understanding of generation, transport and extraction of hot carriers has jointly attracted recent attention to prove the existence of hot carriers. Yet, the proof of the existence of hot carriers has not been accepted across the whole scientific community. There is still a strong opinion that the effect assigned to hot carriers can simply be explained by thermal lattice effects, giving rise to its own “heated” discussions. To build upon this, in an attempt to re-connect physicists, chemists and ab initio theoreticians from a wide range of various sub-disciplines in the field, to discuss, challenge and define the latest problems within plasmonics and hot electron processes a Faraday Discussion meeting was planned. The 18th of February 2019, within the London residence of the Royal Society of Chemistry (RSC), brought the beginning of a landmark event, the 300th Faraday Discussion meeting, dedicated to the topic of hot electron science.

Faraday Discussions have been bringing together the best scientists from around the globe for the past 112 years in a unique format. Research papers are distributed to all the delegates prior to the conference, allowing the majority of the time of the meeting to be devoted to discussing the presented papers. Within the meeting speakers were given five minutes to briefly summarize the main findings of their work submitted to the discussions. Since the first meeting in 1907, the gatherings have only grown both in size and frequency, reaching now 8–9 annual meetings across the world.

This present meeting was set to cover a large range of topics and so was divided into four main themes, with a session devoted to each theme. Namely, (1) the dynamics of hot electron generation in metallic nanostructures; (2) the theory of hot electrons; (3) new materials for hot electron generation and (4) applications of hot electrons in catalysis, photochemistry and photodetection. With great anticipation delegates arrived throughout the morning for registration, followed by lunch and a quick chance to browse through some of the posters being presented later that evening. The meeting was then officially opened by both Stefan Maier (Ludwig Maximilian-Universitat Munich, Germany) and Anatoly Zayats (Kings College London, UK), Chairs of the scientific committee, with a welcoming address. The delegates were reminded of the outline and format of the discussions from the Royal Society of Chemistry's Publishing Editors. Having been officially welcomed and ready for the meeting to begin Stefan Maier retook the stage to introduce the introductory lecture, given by Naomi Halas (Rice University, USA), winner of the prestigious 2019 Spiers Memorial Award (Fig. 1).

Fig. 1 Delegates for the 300th Faraday Discussion seated down in the library of the Royal Society of Chemistry, listening to the opening remarks to the Faraday Discussion by Naomi Hallas (Rice University, USA). Image courtesy of RSC.

Opening remarks

As delegates had come from a variety of different scientific backgrounds and fields, Halas opened the meeting by reminding the attendees of a brief history of hot electrons. Using high photon energies, it has long been known that electrons can be removed from their respective materials. For metals, this is commonly referred to as the photoelectric effect. Over the last almost 80 years this knowledge has been applied through photoemission spectroscopic techniques to probe the binding energies within molecules or the band structures of materials. Reducing the photon energy below the work function can allow an internal photoemission process, hot carrier production. However, the efficiency of this process is often very low – without the addition of an electron or hole acceptor to help separate the carriers. Using plasmonic structures to induce hot carriers, often metal nanostructures, significantly increases the generation of hot carriers, following the plasmon resonance optical absorption. Halas showed how, although more hot carriers can be induced using larger nanoparticles (NPs), the distribution of carriers is cooler than with smaller NPs. In addition, the size of the NPs can also affect the efficiency of hot carrier generation.

With a greater understanding of hot carrier generation, Halas then moved on to discuss two of the main applications for plasmonic devices, photodetection and photocatalysis (using hot carriers). Through the use of a Schottky barrier, generated hot electrons via plasmon decays can be emitted through the metal–semiconductor junction, generating a photocurrent. Halas presented that using a nanoscale hot electron antenna as part of a complete circuit, with only one Schottky barrier, produced what they described as the “world's smallest spectrometer”. A novel approach to using hot carrier generation for industrial applications, with higher efficiency. Another way to control hot carrier emission was shown through controlling the directionality of hot electrons. Halas showed how embedding gold antenna into Si can increase the quantum yield in addition to directional hot electron emission, due to the polarization of EM fields. Halas showed how a graphene antenna “sandwich” with gold NP heptamers can be used to create an interesting plasmonic photodetector device. The heptamer Fano resonance allowed efficient hot electron production alongside the efficient hot electron collection due to the conformal graphene layers resulting in a 22% internal quantum yield, at 600 nm.

In contrast electron emission over a metal–semiconductor barrier, transferring hot electrons from a surface to anti-bonding orbitals of adsorbed molecules can induce chemical reactions. Plasmonic decays, at longer time scales, result in thermalization of the plasmonic particle. The photothermal heating due to plasmonic excitation can also be used to drive chemical reactions. Halas presented how metal nanostructures can act as a plasmonic photocatalyst, such as H₂ splitting on gold NPs. Coinage metals, which are the traditional choice for plasmonic devices, although great harvesters of light, are generally very weak catalysts. Conversely, transition metals are well known to be excellent catalysts but are very poor plasmonic materials. Halas showed how by coating plasmonic antennae with nanoparticle transition metal “reactors” excites a “forced plasmon in the catalytic reactor”. Meaning hot carriers are then generated for catalysis due to the plasmon decay. Halas argued that this style of photo- or thermal-catalytic device, based on plasmonic nanoantennae, are superior due to hot electron activation causing a change in the reaction pathway.

As such, it was clear to see hot carrier processes have an important role to play in industrial applications from photodetection to plasmon induced catalysis.

Session 1: Dynamics of hot electrons generation in metallic nanostructures

The need for experimental studies of hot electron generation has led to many different approaches to study hot carriers within plasmonic nanostructures. For example, colloidal systems and top-down fabrication are often used. The material's nanostructure has long been known to affect hot carrier generation, transport and hot carrier extraction processes. However, the efficiency of each respective process, for a specific application, is often much lower than desired. One of the key challenges present today is the need to develop methods of analyzing the ultrafast time scale processes occurring during carrier excitation and extraction. Hence, the first theme of topics discussed at this Faraday Discussions meeting was devoted to an understanding of the dynamics of electron processes.

A simple quantum theory based method using a physically transparent treatment of plasmonic hot carrier generation processes was presented by Jacob Khurgin (Johns Hopkins University, USA), with estimations of the generation rates for hot carriers and their locations through analytical expressions. By considering four main absorption mechanisms Khugin presented how perhaps the most useful process to produce hot carriers for practical photocatalytic and photodetection applications would be through the use of a surface-collision assisted absorption mechanism – commonly referred to as Landau damping. This was argued based on how the majority of the carriers would be generated at the surface at an average energy of ħω/2.

Plasmonic nanoparticles are well known to be excellent harvesters of light, where the general understanding of this mechanism relies on an induced charge transfer across a metal–semiconductor or metal–adsorbate interface. It is important to note that the charge transfer process mentioned does not refer to a typical electronic transfer, but rather a resonant condition commonly referred to as a charge transfer within the plasmonic community. Phillip Christopher (University of California, USA) presented a novel approach to quantitatively measure the influence of molecular adsorption on a variety of parameters, shedding light on how surface plasmon dephasing is impacted by chemical interfaces. From these results it was suggested that using materials with well-defined electronic structures and designing chemical interfaces hot carrier distributions could be engineered, introducing a way to perhaps control hot carrier transfer processes.

The use of single-particle techniques can facilitate a nanoscopic approach to monitoring hot carrier reactivity. An understanding of the underlying mechanism of hot carrier generation, transport and injection can also aid in the design of plasmonic systems, such as plasmonic photocatalysis. Emiliano Cortes (Ludwig Maximilian-Universitat Munich, Germany) presented three different possible approaches used to monitor hot carrier reactions at a single particle level, through the use of sub-particle special resolution imaging, at a single particle or molecule level. The meeting presented much debate over the origins of hot carriers’ results, in contrast to the use of a thermal analysis to describe the experimental results. However, through the use of induced chemical reactions, Cortes was able to disentangle the role of temperature increases arguing the products detected can only result from the enhanced near-fields as a result of plasmonic hot carrier injection.

The afternoon continued after a short break with a unique approach presented by John C. Polanyi (University of Toronto, Canada) to gain information about the dynamics of chemical reactions from crossed molecular beam projectiles. An effective molecular beam of highly collimated reactive molecules was projected across a surface at chemisorbed stationary target molecules, observed by scanning tunneling microscopy (STM). Using molecular dynamics in conjunction with the experimental results, Polanyi presented how a selected impact parameter could be used by situating the target molecule at a specific site along the projectile path. Polanyi also showed how they were able to determine a pattern corresponding to reactive and non-reactive scattering in each case.

The final work presented in this session of the meeting took a computational approach to study the effect of hot electrons on state-to-state scattering probabilities of H₂ on a silver surface (111). Reinhard J. Maurer (University of Warwick, UK) presented results from their Molecular Dynamics with Electronic Friction (MDEF) simulations for H₂ on silver surfaces suggesting that plasmon driven MDEF studies can provide a systematic improvement on current state-of-the-art Local Density Friction Approximation work, with electron frication calculated directly from Density Functional Theory (DFT) orbitals.

Celebrating the 300th Faraday Discussion – poster presentation

Before the start of the poster presentations, the delegates were treated to a surprise. Claire Vallance (University of Oxford, UK), the current chair of the Faraday Division Awards Committee, made a guest appearance explaining to the delegates a little more about the Faraday Discussions and how to get the most out of them. She also gave a brief introduction to her role on the Awards Committee, explaining to the meeting the importance and history behind the Spiers Memorial Award. Just before the delegates could leave to attend the poster presentation, Claire presented one more surprise. To mark and celebrate the 300th Faraday Discussion the delegates were treated to two delicious looking cakes, decorated in RSC colors with a 300 in the center. A sweet way to start off the poster presentation reception. Javier Aizpurua (Materials Physics Center (CSIC-UPV/EHU, PS), Spain) and Alexandra Boltasseva (Purdue University, USA) had accepted the difficult jobs of judges to find a winner for the poster prize at this meeting. After a long discussion, due to the high caliber of posters presented this year, Michael Hartelt (TU Kaiserslautern, Germany) was announced as the winner of the poster prize at the conference dinner, Fig. 2.

Fig. 2 Winner of the poster prize, Michael Hartelt (TU Kaiserslautern, Germany) – center, receiving his certificate at the conference dinner with judges Javier Aizpurua (Materials Physics Center (CSIC-UPV/EHU, PS), Spain) and Alexandra Boltasseva (Purdue University, USA), left and right respectively. Image courtesy of Susan Weatherby.

Discussions day 2

Session 2: Theory of hot electrons

Javier Aizpurua (Materials Physics Center (CSIC-UPV/EHU, PS), Spain) began the session by presenting his work on “Dynamics of electron-emission currents in plasmonic gaps induced by strong fields”. The authors demonstrated the role of the quiver motion of the photoemitted electrons in the plasmonic gap region and its role in the electron transfer between the nanoparticles using the combination of quantum calculations (Time-Dependent Density Functional Theory) and classical modelling (Simpleman's model). They studied the complex dynamics of electron-emission currents in plasmonic gaps under ultra-short illumination (induced by strong fields) and showed the importance of the Carrier Envelope Phase (CEP) of an incident pulse to control the mutual coherence between electron and photon dynamics in specific metallic junction.

The second speaker, Priyank V. Kumar (ETH Zurich, Switzerland), talked about direct hot carrier transfer at a model metal–adsorbate interface formed by metallic Ag₁₄₇ nanoparticles and CO molecules in plasmonic catalysis, demonstrating an approach for the capture and quantification of the direct transfer process at a given metal–molecule interface by employing an RT-TDDFT and proposing that CO molecules adsorbed at bridge and hollow sites are more likely to undergo a hot electron induced photochemical reaction, given all other steps are similar across the three configurations. This approach will help to understand and enhance direct transfer transition and thus pave the way for efficient plasmonic catalysis.

Following the morning tea break, Alexander O. Govorov (Ohio University, USA) opened the second part of the Session 2 with a lecture on generation of hot electrons in nanostructures incorporating conventional and unconventional plasmonic materials. They investigated different plasmonic nanostructures, such as strongly-plasmonic materials (Au, Ag, Cu and Al) and crystals with damped plasmonic resonances (Pt, TiN and ZrN) with two types of nanostructure: single nanocrystals and metamaterial absorbers. Using a theoretical model, they presented that the hot electron generation rate are strongly dependent on their geometry and material, showing that metastructures with a plasmonic mirror generate hot electrons much more efficiently than single nanocrystals. According to these results, the proposed material designs seem very interesting for experimental realisation, in terms of possible experimental detection methods for high-energy and hot electron generation, and application in photochemistry and opto-electronics.

Yonatan Sivan's paper (Ben-Gurion University, Israel) reported the assistance of metal nanoparticles in photocatalysis, claiming that it can accurately be understood as nothing more than a classical heat source. The authors suggested that the faster chemical reactions originate from a purely thermal effect and as a result are unlikely to be due to high-energy non-thermal electrons, such as hot carriers. They demonstrated a theory for the photo-generation of non-thermal energetic carriers in metallic nanostructures based on a quantum-like version of the Boltzmann equation. Thus presenting a question to many of experimentalists present; are the results reported primarily due to hot electron effects or thermal effects?

The second session was concluded by Andrea Mini (University of L’Aquila, Italy) with an out-of-equilibrium electron dynamics study of silver driven by ultrafast electromagnetic fields. To be able to describe the ultrafast nonlinear dynamics of electrons in this material upon excitation by ultrashort pulses of electromagnetic radiation with a duration of a few femtosecond, the authors developed a novel set of hydrodynamical equations by solving the Boltzmann equation through the methods of moments. Their results open up new possibilities for the suppression of ohmic losses in metals and thus for the development of novel, low-loss plasmonic waveguides and interconnects with enhanced operational functionalities.

Session 3: New material for hot electron generation

After a much-desired lunch the afternoon brought about a new theme of discussion. Alexandra Boltasseva (Purdue University, USA) presented how the promising candidate for solar water splitting hematite, a form of iron oxide, can be improved through utilizing gold nanostructures together with an ultra-thin hematite layer. Increases in oxidation photocurrent at near-infrared wavelengths were shown, due to hot carrier generation and decay from gap-plasmon resonances. Boltasseva concluded by suggesting the plasmonic community could use their knowledge of plasmonic materials and nanostructure geometry to shift plasmon resonances above the hematite band gap, resulting in even greater photocurrent enhancements.

Another plasmonic metal nanostructure–semiconductor hybrid material for plasmon driven photocatalysis was presented by Ping Xu (Harbin Institute of Technology, China). Using a uniform silver–molybdenum disulfide (Ag–MoS₂) hybrid film Xu presented how changes in the crystallinity and thickness of the MoS₂ can change the efficiency of the hot carrier transfer process. The results presented suggest that a thinner more crystalline MoS₂ film would give greater enhancements. In addition, this study adds to a consensus that hybrid metal–semiconductor systems show great promise for the future of plasmonic driven reactions and sensors.

The greatest plasmonic enhancements are known to occur where the electric field is confined within nanoscopic gaps. Preparing an array of gold nanodisks on a MoS₂ surface, Mahfujur Rahaman (Chemnitz University of Technology, Germany) presented how unprecedented enhancements and resolution through tip enhanced Raman spectroscopy (TERS), through a coupled tip-nanodisk system. Matching theoretical calculations to the experimental data, with good agreement, Rahaman presented how the strongest fields were distributed at the rims of the nanodisks.

After a quick break for afternoon tea the delegates returned to the library for the last few discussions of the session. Plasmonic resonance gratings can be used to differentiate bulk absorption processes from plasmonic resonant excitation. Stephen B. Cronin (University of Southern California, USA) presented how this can be achieved using polarized light perpendicular to the grating. Relatively small photocurrents could be measured due to tuning a plasmon resonance of the grating, demonstrating hot carrier induced enhanced photoelectrochemical performance.

Nanostructured titania with gold nanoparticles often appears in the literature as a photocatalytic platform for more efficient solar energy conversion and water splitting. Laura Fabris (Rutgers University, USA) presented how the design parameters for the gold nanostructures and the metal–semiconductor interface are the main drivers behind these processes. Optimized hot carrier photocatalysis for increased H₂ production rates was presented along with a different additional possible mechanism for hot carrier generation.

Using Co₃O₄ nanocubes with a Pt/TiO₂ nanodiode, as an alternative metal oxide–metal interface, Jeong Young Park (Institute for Basic Science, Republic of Korea) presented a mechanistic understanding of hot carrier dynamics. The newly designed catalytic nanodiode clarified the effect of the oxide–metal interface on catalytic properties. Park also showed how UV treatment significantly improved both the chemicurrent and turnover frequency.

When an incoming EM wave propagating normal to the surface interacts with bilayers of colloidal gold nanoparticles, a dark mode is excited due to the field retardation, leading to hot carriers. Stephanie Reich (Freie Univeristat Berlin, Germany) presented the optical excitation of dark interlayer plasmons with a strong absorption peak which was only found to be present in the bilayer, as opposed to the monolayer. In comparison to bright plasmons, the damping of dark modes is strongly reduced in addition to the potential to produce large scale substrates with low costs. At the end of the session the delegates left the RSC, excitedly on their way to the conference dinner.

Conference dinner

After two long days of intense discussions a general feeling of excitement and anticipation could be felt in the RSC's library as time ticked ever closer to the start of the conference dinner. For this meeting, delegates had the pleasure of a short walk towards the famous Royal Society, where the dinner was to be held. During the dinner the poster prize, discussed, above was awarded, Fig. 2. In addition to being treated to a delicious meal, many of the delegates, being their first time at a Faraday Discussion, were also introduced to the Faraday's famous Loving Cup ceremony. The attendees pass around an 18th century silver cup filled with port at the end of the meal, toasting a former officer of the Faraday Society (Fig. 3). With the traditional ceremony over, and for many delegates a rather full stomach after such a great feast, the delegates left to get some rest before the final day of the discussions.

Fig. 3 Top: Conference chair Anatoly Zayats (Kings College London, UK) performing the famous Loving Cup ceremony. Bottom: Delegates enjoying their time at the conference dinner.

Discussions day 3

Session 4: Applications in catalysis, photochemistry and photodetection

The last session of the Faraday Discussions, chaired by Prof. Hongxing Xu (Chinese Academy of Sciences, China), covered the most challenging and fascinating topic, applications of hot electron research. Recently, hot electron generation in various plasmonic nanostructures has opened up a new “route” for photocatalysis due to their optical and catalytic properties. Under visible illumination at their plasmonic frequency, the excitation of localised surface plasmon resonances can promote chemical reactions on their surfaces, which can be tuned across the visible spectrum.

The first half of the morning session began with a presentation from Jorge U. Salmon-Gamboa (King's College London, UK), focusing on the optimisation of hot carrier effects in plasmonic heterostructures. Their study investigated hot electron generation and extraction from Pt decorated SiO₂–Au nanoparticles using the degradation of methylene blue dye. Designed plasmonic heterostructures have been found to be very attractive for hot electron production, due to the tunability of their plasmonic resonance in the visible spectrum and enhanced catalytic activity.

In addition, the photocatalytic efficiency can be dramatically increased through the combination of a semiconductor with plasmonic nanostructures, which was the subject of the next speaker of the second lecture of the morning, Dr Madasamy Thangamuthu (EPFL, Switzerland). In this work, the authors demonstrated an alternative route for ammonia synthesis and the impact of plasmonic effects of aluminium nanotriangles on photocatalytic ammonia synthesis. Using the electrochemical photocurrent measurements, they studied plasmonic near field coupling to semiconductor materials (such as TiO₂) and hot electron generation under 365 nm illumination. Moreover, according to the presented results, the use of plasmonic aluminium structures remarkably improves the ammonia production rate.

Following that, Valérie Caps (CRNS, France) showed the impact of the interface nanostructuration of gold–titania composites for solar and visible photocatalytic gas phase reduction of CO₂. The high surface area 1 wt% Au/TiO₂-UV100, prepared by adsorption of a NaBH₄-protected 3 nm gold sol, readily catalysed the photoreduction of carbon dioxide with water into methane under both solar and visible-only irradiation with a CH₄vs. H₂ selectivity of 63%. This showed how plasmonic nanoparticle structures might come to play a significant role in preferential CO₂ reduction and CH₄ selectivity for industrial processes.

Junyang Huang (University of Cambridge, UK) opened the second part of Session 4 with a talk on plasmon-induced direct optical control over dithionite-mediated chemical redox reactions in aqueous environments. The authors presented a novel hybrid colloidal system for light-driven reversible reduction of chemical species. Using sub-nm molecular spacers (CB) or salt induced gold nanoparticles (AuNP) aggregates, the reduction process can be turned on and off repeatedly by controlling irradiation intensities, thus making it possible to observe the change in redox state of the reported molecules.

The last speaker of the session was Prof. Zachary D. Schultz (Ohio State University, USA) who shed further light on the impact of optically rectified fields on plasmonic electrocatalysis. The excitation of electrons in localised surface plasmon resonances can drive catalytic reactions by tunnelling though nanojunctions and giving rise to optically rectified fields. The combination of Stark spectroscopy with electrochemical measurements showed evidence that the optical rectification of plasmonic fields can alter the electrochemical reactivity on plasmonic surfaces by modifying the overpotential. In addition, he suggested that understanding the impact and optimizing this nonlinear process might help to design and improve plasmonic catalysts.

Concluding remarks

After three days of fruitful discussions, Jeremy J. Baumberg (University of Cambridge, UK) closed the Faraday discussions on hot electron science in plasmonics and catalysis by giving an inspirational concluding remarks lecture. He summarised how hot electron photochemistry has made strong progress towards increased control of chemical reactions and demonstrated how key issues can be solved through the implementation of various experimental and theoretical models (Fig. 4).

Fig. 4 Jeremy Baumberg (University of Cambridge, UK) addressing the conference with his concluding remarks.

He highlighted many concepts that utilised hot electrons for different applications by presenting models of hot electron production at plasmonic surfaces, experimental progress of nanostructured materials with plasmonic components and several photocatalytic probes, with the aim of studying photocatalytic processes in greater detail.

In addition, many overlapping terminologies were discussed, of which the focus was ‘what is a hot electron?’ He pushed for a more concise definition for the term ‘hot electrons’; one which will allow the community to either discretise between catalytic mechanisms, such as non-thermal electronic excitations and local thermal lattice fluctuations (‘superheating’), or to generalise the nomenclature so that it encompasses some/all of these energetic mechanisms.

Baumberg concluded that we have reached the crucial stage where we need to consider new constructs that allow for full control over hot electrons, in precise nano-geometries which can be easily probed. In order to achieve this goal, there is a requirement for single nano-geometries in which all the electromagnetic and thermal configurations are fully controlled. This would provide a widely-trusted tool for probing many of the issues he raised, including the influence of inhomogeneous temperatures, thermal transport, gradient forces and potentials, direct and indirect electrons, shape, diffusion of species, reconstruction and degradation of activity, as well as investigating their influence on a range of reactions.

Throughout the event, it was noted that the majority of the research was focused on optical field enhancements in plasmonics, rather than on the phonon transport in such nanostructures. Baumberg emphasized the need to understand nanoscale thermal hotspots, thermal transport and gradient, as well as the conventional optical confinement in plasmonics. His concluding remarks lecture provided a list of common goals for the whole interdisciplinary community which would enlighten everyone on what is really happening inside the photocatalytic (nano)box.

---