Concentrated irradiation enables the production of solar fuels using smaller and inexpensive devices, resulting in decreased costs for solar fuel production. Additionally, by carefully selecting optimal operating conditions, it might be possible to enhance the performance of photoelectrochemical (PEC) devices [1]. A large proportion of the research on semiconducting materials for solar-driven water splitting has focused on their behaviour at room temperature and under 1 sun illumination. Hence, the utilization of high photon fluxes presents new possibilities for solar fuel production, as it has been demonstrated for integrated PV+EC devices [2]. Nevertheless, many obstacles hinder the design and testing of PEC cells intended for operation under concentrated irradiation. The potentially advantageous synergy between higher temperatures and irradiation makes the system complex and difficult to study.

In that regard, the effect of temperature alone on the properties of photoelectrodes has been reported recently for α-Fe₂O₃ [3], showing increased photocurrent densities at high potentials and increased recombination rates at low electrode potentials. In a similar study, BiVO₄ photoelectrodes have shown substantial increase in photocurrent densities across all electrode potentials [4]. The behaviour of photoelectrodes under high irradiations has been performed under relatively low irradiances (<30 kW m^‑2, c.f. 1 sun ≈ 1 kW m^-2) and usually for very stable photoelectrodes, e.g. Fe₂O₃ [5].

The combined effects of temperature and high irradiations on the properties of photoelectrode materials has been rarely investigated experimentally. With the aim to tackle this challenge, we designed and developed a PEC cell for operation under high irradiations (30 – 360 kW m^-2) complemented with a multiphysics model to predict the surface temperature, and the effects of ohmic drop due to bubble evolution, substrate resistance, and current density distributions. The High-Flux PEC cell (HF-PEC) comprises a stainless body steel acting as counter electrode, a high transmission quartz window and a modified Luggin capillary for the reference electrode. To avoid extreme temperatures at the surface of the photoelectrode, the electrolyte was recirculated at the back of the substrate to keep it at lower temperatures. Two photoelectrode materials were investigated: Sn-doped α-Fe₂O₃ and BiVO₄, both deposited on FTO or Ti substrates via spray pyrolysis.

In the case of Ti|α-Fe₂O₃ photoelectrodes, it was found that at high temperatures under 1 sun illumination, the charge transfer efficiencies decreased with temperature at low potentials (< 1.2 V vs. RHE), but increased at higher potentials; in contrast, impedance spectroscopy under high irradiations indicated that charge transfer efficiencies increased for all electrode potentials, even when the surface of the photoelectrode was at higher temperatures. This suggests, that although e-h recombination kinetics increased with temperature, the increased charge transport kinetics and higher exciton concentration could ultimately favour the charge transfer at the semiconductor-electrolyte interface. For FTO|BiVO₄ photoelectrodes, it was found that photocurrent densities do not scale proportionately with irradiance, and that degradation rates for this material are the result of complex interactions between increased transfer efficiencies, high temperatures and high photocurrent densities; it was found that the highest (irradiance-normalized) current density for BiVO₄ was achieved at ca. 120 kW m^-2.

Photoelectrochemical water splitting is a promising way of solar hydrogen generation. However, in order to achieve inexpensive, efficient, and stable photoelectrodes significant further improvements are needed. Perovskite and organic photoactive materials have attracted great scientific interest by reaching record high single-junction solar cell efficiencies, but their photoelectrode performance is currently limited by their instability in an aqueous environment. In this presentation, we will show a cost-effective way of protecting halide perovskite, as well as organic photoactive layers used for solar water splitting to reach both stable (>100 hours) and remarkably high photocurrents (>8 mA cm^‑2 and >25 mA cm^‑2 at 1.23 V_RHE, respectively). The perovskite photoelectrodes were developed using a low annealing temperature carbon paste with tuned energy level CsPbBr₃ photoactive layer. The effect of 2-dimensional (CsPb₂Br₅) and 0-dimensional (Cs₄PbBr₆) perovskite phases on efficiency and stability will be discussed including ways to control their formation. The organic photoanodes applied a ternary bulk heterojunction blend (non-fullerene acceptor and polymer donor) and the interlayers were optimised for stability under 1 sun illumination. Importantly, the lifetime of the photoelectrodes was increased by applying a protective graphite sheet functionalised with electrodeposited oxygen evolution catalyst (NiFeOOH), allowing to reach remarkable stability of several days – instead of the often-reported few hours – at high photocurrent densities, far beyond the current state-of-the art levels.

Transition metal (TM) nitrides are an emerging class of catalytic and photoelectrocatalytic materials.[1],[2] In general, nitrogen-poor TM nitrides are usually refractory materials with metallic character while nitrogen-rich nitrides often possess semiconducting character.[3] Hence, while the former are potential electrocatalyst candidates, the latter may qualify as photoelectrode absorber materials. For example, ZrN has recently been proposed as an electrocatalyst for both the electrochemical nitrogen[4] and oxygen[5] reduction reactions (ORR and NRR), while Zr₃N₄ and also Zr₂N₂O have been suggested as potential photoanode materials in photoelectrochemical water oxidation.[2] In this contribution, we test these hypotheses regarding the (photo)electrochemical characteristics of Zr-based (oxy)nitrides by experiment. To this end, we investigate reactively sputtered thin films for the electrochemical NRR/ORR and the photoelectrochemical oxygen evolution reaction (OER). Previous experiments on Ta-based nitrides have shown that addition of oxygen during the reactive sputter process is necessary to access higher metal oxidation states.[6] Indeed, as we introduce controlled amounts of oxygen at otherwise fixed deposition conditions, we observe a transition from metallic ZrN to a disordered nitrogen-rich Zr_xN_y to a crystalline bixbyite-type Zr₂N₂O to nitrogen-doped cubic ZrO₂. Crystalline Zr₃N₄ was not accessible under the used experimental conditions. While we observe a lack of electrocatalytic activity for ZrN in NRR and ORR and instabilities of the disordered nitrogen-rich Zr_xN_y in the photoelectrochemical OER, introducing more oxygen into the structure leads to a more stable crystalline structure (Zr₂N₂O), the opening of a band gap in the visible range, and the emergence of photoelectrochemical activity for oxidation reactions. Based on chopped linear sweep voltammetry measurements, we show that Zr₂N₂O films are photoactive for the OER in alkaline electrolyte with low onset potentials, indicating an overall favorable band alignment of the material with respect to the water oxidation and reduction potentials. While the observed photocurrents are still significantly lower than for the benchmark oxynitride TaON, further material optimization could potentially close this gap and provide a materials system functioning as sustainable photoanode.

Harnessing solar energy through photoelectrochemical (PEC) water splitting approach offers a promising and sustainable pathway to produce green hydrogen (H₂). However, to ensure the viability, cost-effectiveness, and long-term stability of this approach, fabricating an abundant and durable photoelectrode is crucial. Barium stannate (BaSnO₃) has not been considered promising material for PEC water splitting applications due to its wide band gap (˃3 eV) and limited solar absorption characteristics.[1] In this study, we optimized a spray pyrolysis method to fabricate phase pure BaSnO₃ photoanodes with a smaller optical gap of ∼2.2 eV. By annealing these photoanodes in a mild H₂ gas environment, we observed a significant decrease in the optical gap to ~1.5 eV. This led to an enhanced photoelectrochemical performance, with a 5-fold increase in the photocurrent density reaching ~0.5 mAcm^-2 at 1.23 V vs. RHE and an improved onset potential of ~0 V vs. RHE. To understand the reason behind this enhancement, we have used a combination of spectroscopy techniques, including photoluminescence (PL), time-resolved surface photovoltage analysis (TR-SPV), and time-resolved microwave photoconductivity (TRMC) measurements. We find that mild H₂ annealing of BaSnO₃ generates a set of mid-gap defect states associated with oxygen vacancies and Sn²⁺ centres.[2] Increasing the population of these mid-gap states, shifts the optical onset of photocurrent collection and mobile charge carrier generation to photon energies as low as ~1.5 eV. Furthermore, the transient signals of charge carriers show the extended lifetime after H₂ treatment and provides the evidence that carriers are transported via tunnelling through delocalized defect states. We show here that BaSnO₃ is a promising material with tunable optical properties and electronic structure.

The recent emergence of ultrathin oxide layers in catalysis has led to more corrosion-resistant photoelectrodes and enhanced catalytic selectivity – suppressing side reactions and catalyst poisoning [1]  Silica nanolayers for example demonstrated enhanced proton transfer [2] and electron transfer [3] across solid-solid interfaces when embedding molecular wires. The nanoscale integration of incompatible catalytic environments of CO₂ reduction and H₂O oxidation – similar to a thylakoid membrane – can potentially be achieved by such ultrathin oxide layers, provided we can maintain directional electronic and protonic communication across the layer.

Here, we investigate the proton conductivity and O₂ impermeability of ultrathin, dense, and amorphous Al₂O₃ layers and find that the diffusivity of protons is strongly dependent on the flux due to induced morphological changes in the alumina layer over time. Dense amorphous alumina layers were prepared via pulsed laser deposition (PLD; 2.5 nm, 3 nm, and 5 nm) and the porosity was varied via the oxygen background pressure of the chamber. We used electrochemical impedance spectroscopy to determine diffusion coefficients and charge transfer resistances, and complementary electrochemical FT-Infrared reflection-absorption spectroscopy was applied to analyze the dynamic structural changes of alumina over time in-situ (Figure 1b). In fact, we observe that initially, 2.5 nm of alumina can block both protons and oxygen, whereas over time the proton permeation improves while we can exclude dissolution of the Al₂O₃ layer (Figure 1c)

These discoveries facilitate the integration of incompatible catalytic functions on the nanoscale, thereby opening a new design space for developing macroscale systems. The core–shell nanotube array geometry for developing an artificial photosystem with the goal of placing CO₂ reduction inside the tubes and H₂O oxidation outside of them is just one among many opportunities to pursue.

During my talk I will discuss four topics, over which I have felt concern over the past few years:

- The concept of applied bias photon-to-current efficiency (ABPE).

The ABPE is described as ‘a diagnostic measurement in materials development’. However, the widely accepted ABPE equation is not formulated as a true energetic efficiency. Furthermore, negative ABPEs are obtained for applied potentials > 1.23 V, resulting in increasingly negative efficiencies for increasingly high photocurrents, which is non-sensical. Both of these issues invalidate the ABPE equation for the very condition it is designed to represent. It is also noticeable that a lot of recent reviews and systematic studies of photoelectrode materials in biased systems do not use the ABPE as an effective way to compare their performances. I will discuss the issues of the currently accepted ABPE formulation and suggest a more rigorous formulation.

- The perpetual reporting of incident radiation on photoelectrochemical cells as AM 1.5 G and 100 mW cm^-2

The accepted basic requirement for the characterisation of solar water splitting systems is the same as for benchmarking the performances of solar cells: the incident spectral irradiance has to match that of the Air Mass 1.5 Global (AM 1.5 G) spectrum of natural sunlight, amounting to 100 mW cm^-2 when integrated. The spectra of light sources is rarely an exact match to the AM 1.5 solar spectrum, even if an AM 1.5 filter is used.  I will discuss the importance rigorous calibration and the need to report the actual spectra of light sources used during the characterization of photoactive materials and assessment of photoelectrochemical devices.

- The construction of Pourbaix diagrams to predict photoelectrode and catalyst stability – example of BiVO₄ degradation

There are some puzzling aspects in the Pourbaix diagrams being used for demonstrating the reactions responsible for light-induced BiVO₄ degradation. I will present the published Pourbaix diagrams for the combined aqueous bismuth and vanadium systems and discuss why some postulated reactions that are presented as electrochemical (i.e. involving a change in oxidation state) cannot actually be so. I will also consider why chemical speciation in different electrolytes can contribute to electrode degradation.  

- Effect of temperature on (photo)electrolyser performance

Solar cells must be maintained at 25 °C during characterisation and here the characterisation of water splitting systems ceases to follow the same standards. The question is whether this matters. The deviation of photoabsorber temperature from 25°C in solar water splitting systems is not necessarily unreasonable or preventable, since the presence of electrolyte makes it harder to fix the photoabsorber temperature to one specific constant value; photoelectrochemical cells are not usually jacketed or designed to be immersed in water baths and so the electrolyte (and hence photoabsorber) temperature could readily vary by > 5 °C, depending on the ‘room temperature’ of a laboratory at the start of the experiment, and can increase by > 10 °C relative to the starting temperature, depending on the photocurrent density, size of the absorber relative to the volume of electrolyte and the duration of the experiment. It therefore seems more important for the electrolyte temperature to be measured, rather than controlled, since temperature affects Gibbs free energy of formation of the water splitting reaction and the volumetric molar densities of evolving gases. I will show, from the thermodynamic viewpoint, the importance of electrolyte temperature in (photo)electrolysis.

Millions of families around the world remain vulnerable to water scarcity and have no access to drinking water. Advanced oxidation processes (AOPs) are an effective way towards water purification with qualified reactive oxygen species (ROSs) while are impeded by the high-cost and tedious process in either input of consumable reagent, production of ROS, and the pre-treatment of supporting electrolyte. Herein, we couple solar light-assisted H2O2 production from water and photo-Fenton-like reactions into a self-cyclable system by using artificial leaf, achieving an unassisted H2O2 production rate of 0.77 μmol/(min·cm2) under 1 Sun AM 1.5 illumination. Furthermore, a large (70 cm2) artificial leaf was used for an unassisted solar-driven bicarbonate-activated hydrogen peroxide (BAP) system with recycled catalysts for real-time wastewater purification with requirements for only water, oxygen and sunlight. This demonstration highlights the feasibility and scalability of photoelectrochemical technology for decentralized environmental governance applications from laboratory benchtops to industry.

Nitriles are interesting class of organic compounds included in a wide variety of natural products and industrial processes, that have been explored as potential Liquid Organic Hydrogen Carriers (LOHCs) due to their high hydrogen capacity and stability.  Conventional methods for nitrile synthesis suffer from low atom economy, the use of toxic reagents, and limited selectivity. An alternative approach involves the oxidation of primary amines through organic electrosynthesis, which offers a safer and more environmentally friendly alternative by utilizing electric current and reducing hazardous solvents. Our research focuses on developing a green, selective, and gentle method for synthesizing nitriles in aqueous media, based on electrochemical procedures that allow eliminating the need for harsh reaction conditions. To that end, we have devised a facile alternative process for synthesizing electrodes through the electrodeposition of nickel on various supports for the electrooxidation of amines to nitriles.

In the electrooxidation of primary amines to nitriles with Ni based electrodes, hydrogen in produced at the cathode, whereas de amine is oxidized at the anode due to the redox par Ni²⁺/Ni³⁺ formed at the anode. To evaluate the effectiveness of different nickel-based electrodes, we conducted screening tests involving the electrooxidation of different aromatic and aliphatic primary amines. The electrodes tested include Ni foil (NiF), Ni45^[1], Ni foam (NF), Ni electrodeposited on NF (NieNF), and Ni electrodeposited on a graphite pencil rod (NieGPR). Among these, the NieNF electrode demonstrated the most favorable results in terms of high yields (>95%) and substrate tolerance. Notably, our proposed electrodes exhibit a significant advantage over Ni/Se electrodes reported in the literature^[2] , enabling a reduction of the reaction time for complete amine oxidation by half.

To sum up, the current project paves the way to produce both nitriles and H₂ with low cost and high   efficiency, thus grasping great potentials in the future energy conversion and storage technologies, sustainable environmental issues and beyond.

Solar hydrogen generation by photoelectrochemical (PEC) water splitting allows for direct photon-to-molecule conversion in a standalone system. Although it is one of the most promising approaches to generate renewable hydrogen, it is also the most challenging because of the multitude of processes occurring concurrently and at very short time scales.

Bismuth vanadate (BiVO₄) is an ideal photoanode material because of its favourable band gap of 2.4 eV [1], which allows it to absorb photons in the visible region of the solar spectrum and has a theoretical solar to hydrogen (STH) efficiency of ~ 9.2%. It has suitable band edges for the water redox process, is durable in aqueous solution, good crystallinity and made up of earth abundant materials. However, BiVO₄ photoanodes suffer from poor charge separation in the bulk of the material and at the electrolyte interface during the PEC water splitting process [2]. Doping of BiVO₄ is a common strategy used to improve charge separation [3]. Doping by electron donors can improve the electrical conductivity by increasing the electron density of BiVO₄ [4]. Doping of BiVO₄ with hexavalent Molybdenum (Mo⁶⁺) has been shown to enhance photocurrent generation due to more efficient charge separation [5].

In this work, we compare the charge separation efficiency of a homogenous doping concentration profile to a gradient doping concentration profile of Mo-doped BiVO₄. Gradient doping induces an upward band bending which amplifies charge separation within the bulk of the BiVO₄ film. Mo-doped BiVO₄ thin films are deposited on FTO-coated glass substrates by ultrasonic spray pyrolysis (USP). Morphological characterisations reveal controlled successive layered films, wherein the number of layers correspond to the number of cycles using our USP deposition process. Gradient doping is achieved by depositing controlled successive Mo-doped BiVO₄ layers that contain increasing dopant concentrations using this method. Their improvement in optical absorption and PEC performance is elucidated by material, optical, Mott-Schottky characterisations and electrochemical impedance spectroscopy.

Gas diffusion electrodes are essential components of common fuel and electrolysis cells that use gas phase reactants and products, but are typically made from graphitic carbon or metallic materials, which do not allow light transmittance, and thus limit the development of photoelectrode membrane assemblies for gas-phase solar fuel production. In this presentation, the application of solar H₂ production from humid air is motivated and the simple and scalable preparation of transparent gas-diffusion layers using F-doped SnO₂ (FTO) coated SiO₂ interconnected fiber felt substrates is introduced[1]. The resulting substrates have a porosity of 90 %, a roughness factor of 15.8 and a Young’s Modulus of 0.2 GPa. A sheet resistivity of 20 ± 3 Ω sq⁻¹ and a loss of incident light of 41% at an illumination wavelength of 550 nm is found with FTO coating. The application of various semiconductors on the substrates was established including Fe₂O₃ (chemical bath deposition), CuSCN and Cu₂O[2] (electrodeposition ), and semiconducting polymers[3] (dip coating ) and the liquid-phase photoelectrochemical performance commensurate with flat FTO substrates was confirmed. Finally, gas phase H₂ production was demonstrated with a polymer semiconductor photocathode membrane assembly at 1-Sun photocurrent density on the order of 1 mA cm^–2 and Faradaic efficiency of 40%.

A fundamental understanding of processes occurring at the electrode-electrolyte interface is essential for the development of optimized materials in sustainable energy production. In this talk, I will present two examples that highlight the relevance of atomistic studies on model electrocatalytic systems in understanding existing structure-function relationships and reaction-mechanism.

Firstly, I will introduce our custom experimental setup, which enables a comprehensive characterization of structural, chemical, and electrochemical properties on the same sample. This setup facilitates the transfer of electrodes from ultra-high vacuum conditions, ideal for surface science tools, to an electrochemical cell in an inert atmosphere. By employing this approach, we investigated potential-induced changes on Au(111) surfaces in acidic media. We monitored chemical transformations using ex-situ X-ray photoemission spectroscopy (XPS), providing insights into the electrochemical response, including oxide formation during the onset of the oxygen evolution reaction. Additionally, I will discuss the influence of various electrolyte anions on surface electro-oxidation.

Secondly, I will provide an overview of the local properties of point defects in transition metal dichalcogenide semiconductors (MX₂, M = transition metal, X = chalcogen) that have been proposed as active sites for the hydrogen evolution reaction (HER). By combining 4K scanning tunneling microscopy and spectroscopy (STM/STS), and non-contact atomic force microscopy (nc-AFM), we achieved atomic-scale correlation between the morphology and electronic properties of different types of structural defects in MoSe₂ and WS₂ monolayers. Our investigations unveiled the predominance of substitutional oxygen as the point defect in these semiconductors, effectively suppressing deep in-gap states associated with chalcogen vacancies [1, 2], which have been discussed to mediate the HER. These findings highlight the critical role of local properties in determining the functionality of materials for hydrogen production, ultimately governing the underlying reaction mechanism.

Photoelectrochemical (PEC) water splitting is a promising approach for generating solar hydrogen. To realize inexpensive PEC devices, TiO₂ has been considered a potential photoanode material because of its abundance, stability, and suitable valence-band energy edge for water oxidation. In this work, we will present results on post-treatment strategies to extend the high efficiencies of these photoanodes to the visible region of the solar spectrum. Our research paper details the successful introduction of different point defects (V_O and V_Ti) in TiO₂ photoanodes using a chemical treatment process. We utilized a range of advanced techniques, including surface photovoltage signal measurement (SPV), electron paramagnetic resonance measurement (EPR), and proton magic angle spinning nuclear magnetic resonance (¹H MAS NMR) to investigate the intrinsic band energy structure and surface state of synthesized TiO₂ photoanodes. Our finding indicates that V_O and V_Ti have favorable effects on photoanodes, leading to enhanced PEC performance. The enhancement is attributed to the interplay of collective and localized effects of point defects on TiO₂. The increased point defects narrow the bandgap and enhance donor density in TiO₂, which increases light absorption, conductivity, and photovoltage, as well as reducing the flat-band potential of TiO₂ photoanodes. Additionally, our results show that the presence of localized surface V_Ti surrounded by O^- increases the amount of surface hydroxyls and creates a more basic environment for Ti-OH species. This facilitates the water oxidation process by serving as trapping sites and self-reduction sites during OER. Our study on OER kinetics, using photoelectrochemical impedance spectroscopy (PEIS) and intensity modulated photocurrent spectroscopy (IMPS), demonstrates that the interaction between collective and localized effects of point defects can decrease the charge transfer resistance on the TiO₂-electrolyte junction and increase the concentration of interface hole flux on TiO₂ photoanodes. This is consistent with the observed higher photovoltage after chemical treatment. Our study has revealed that the interaction between V_O and V_Ti has a significant impact on the PEC performance of TiO₂ photoanodes. We believe that these findings will make a valuable contribution to the development of efficient and stable TiO₂ photoanodes and provide valuable insights into the development of higher performance photoanodes for sustainable energy conversion applications.

Water oxidation photoanodes based on earth-abundant metal oxide semiconductors are actively studied due to their limited environmental impact, the stability to photo-corrosion and the tunable optoelectronic properties [1]. BiVO₄/WO₃ heterostructures are now emerging as one of the most promising photoanode, but the activity is limited by surface recombination and sluggish charge transfer kinetics. Cobalt-iron mixed compounds, such as CoFeOx and CoFe hexacyanoferrate, a Prussian blue analogue, proved to effectively enhance photocurrent when coupled to BiVO₄/WO₃ photoanodes as water oxidation co-catalyst. The origin of the increased efficiency is still debated, but recent studies suggest an enhancement of charge separation efficiency within the semiconductor [2], rather than improved water oxidation kinetics, underlying possible charge accumulation on the co-catalyst leading to transient modification of the local structure. X-ray Absorption Spectroscopy (XAS) provides a unique view of the structural features of ultrathin catalyst layers, however often limited to a steady state ex-situ characterization of the catalyst local structure and valence state. To achieve structural understanding of the catalyst activity in the actual PEC operating conditions, a custom PEC-XAS cell was specifically designed to allow for operando monitoring of Co-Fe based co-catalysts deposited on  BiVO₄/WO₃ photoanodes, upon bias and illumination stimuli fully replicating the ex-situ conditions. The developed setup (Fig 1a), providing limited x-ray photons interaction path with the controlled-flow electrolyte (<100 μ m), prevents bubbles formation and enhances mass transfer (Fig 1a), allowing for stable, real-time correlation of the structural features with the photoelectrochemical response. The activity of CoFeOx and CoFe-PB  was investigated by either performing fixed potential spectroscopy (Fig 1b) at Co K-edge, or fixed X-ray energy absorption voltammograms [3] (FEXRAV, Fig 1c). The resulting characterization displayed a drastic modification of the structure between operando and ex-situ analysis, as well as a dependence on bias and illumination. In addition, despite the apparent similarities between CoFeOx and CoFe-PB in terms of composition and PEC activity, their operando-XAS behaviour displays a different enhancement mechanism, further correlated to the charge dynamics characteristics, investigated by IMPS and PEIS.

Organic semiconductors heterojunctions have recently gained significant interest in solar fuel production. In contrast to widely studied inorganic photocatalysts, the synthetic tunability of organic semiconductors, such as conjugated polymers and covalent organic framework (COF), allows the design of materials with tuned bandgaps for photocatalysts that can absorb a wider proportion of solar spectrum. Recently, heterojunctions nanoparticles prepared from a blend of conjugated polymer and non-fullerene small molecules, as those used in the organic photovoltaic field, have shown promising efficiency for hydrogen production in the visible. [1,2]  Our previous transient and operando studies have shown remarkably long-lived charge generation within the donor:acceptor heterojunction nanoparticles on timescales useful for photocatalysis (microsecond-millisecond).[3] Using a similar approach, 2D templated polymer heterojunctions were explored as a photocatalyst for hydrogen production, prepared by combining donor and acceptor 2D polymers in a templated growth method. Transient absorption spectroscopy on timescales ranging from picoseconds to seconds were employed to monitor the kinetics of photogenerated charges in the 2D heterojunction and single 2D polymer components and their correlation with hydrogen production.[4] In this talk, the charge carrier dynamics of the novel 2D templated polymer photocatalyst with improved hydrogen evolution rates will be discussed, including the effect of charge trapping and metal cocatalyst on charge extraction.

The pursuit of efficient, earth-abundant and stable photocatalysts has been the holy grail since TiO₂ was discovered as a water-splitting photocatalyst more than 50 years ago.[1,2] Ever since, wide bandgap absorbers such as TiO₂, SrTiO₃ or Ga₂O₃ have dominated the field of particulate photocatalyst development predominantly due to their deep valence bands driving water oxidation and remarkable stability.[3] However, with less than 5% UV light in our solar spectrum, the overall solar energy conversion efficiency of such wide-bandgap systems is by default limited to efficiencies below 1%. Even with the most outstanding Al:SrTiO₃ photocatalyst achieving one of the highest quantum yields reported, the overall solar-to-hydrogen conversion efficiency was only 0.65%.[4]

Tremendous efforts have been undertaken to extend the absorption into the visible to maximise the light-harvesting efficiency. On this quest, countless oxide and nitride as well as some oxynitride and oxysulfide materials with smaller bandgaps have been intensively studied with varying success.[3,5] In the novel visible light absorber (La,Sr)(Rh,Ti)O₃ employed in the Z-scheme photocatalyst sheet device from Profs. Wang and Domen, the controlled A and B site substitutions led to a record 1% solar-to-hydrogen efficiency.[6]

Substitution on A and B-sites of the perovskite ABX₃ crystal structure creates interesting defect chemistry widely studied in photo- and electrocatalysis. However, as our recent work on the La,Rh-co-doped SrTiO₃ showed, attention to the formation of undesired trap states is particularly important when attempting to induce visible light absorption into wide-bandgap semiconducting photocatalysts.[7]

In this talk, I will dive into defect engineering and chemistry of a small series of oxide perovskites that we employ in photo- and electrocatalysis and discuss challenges and ambiguities in such defect engineering approaches as well as considerations when operating under reducing conditions.

A key challenge in photocatalysis is to generate sufficiently long-lived excited states capable of driving the desired transformations but with a minimal energy cost. In Photosystem II, extending lifetimes is achieved with a series of redox co-factors. This enables the oxygen evolution reaction to take place, but at a high energy penalty (approx. loss of 40% of the energy of a red photon). Similarly, in artificial systems recombination reactions are suppressed by applying external biases or by using sacrificial reagents. This enables carriers to live long enough to react but lowers the overall energy efficiency of the system. While many photo-absorbers can be prepared, there is still no clear blueprint for the design of materials with intrinsically long-lived carriers that maximise photo-conversion efficiencies. In this talk, I will present an experiment systematically characterising the excited state lifetime and recombination dynamics in a series of energy conversion materials. Our data reveals a correlation between the achievable lifetime and the solid’s electronic configuration pointing towards intrinsic factors limiting performance. I will discuss the implications and opportunities that such intrinsic behaviour has in our ability to generate long-lived excited states for catalysis and other energy conversion systems.

High performance, light absorbing semiconductors are of intense interest for the solar-driven generation of sustainable fuels in photocatalytic and photoelectrochemical systems. Whilst metal oxides are the most-studied photoactive materials for the evolution of hydrogen from water and methanol from carbon dioxide and water, organic semiconductors can advantageously be synthesised from abundant sources; offer control over microstructure, light harvesting properties and redox potentials; and have the potential to be used in technologically simple devices. Despite substantial increases in the activities of organic photo-driven systems over the last five years, few studies[1] have managed to deconvolute the many factors (structural, optical, and electronic) that define the performance of these materials.

In this talk, I will primarily use transient absorption spectroscopy to demonstrate that generating charges which can live for microseconds is critical to the performance of a wide variety of different carbon nitride and linear conjugated polymer photocatalyst and photoelectrode systems [2-5]. However, generating large yields of these long-lived charges is a serious challenge in organic materials as excitons and charges typically recombine on the picosecond-nanosecond timescale. I will present several design strategies which can be employed to retard these processes, with particular focus on the timescales at which sacrificial agents and water influence the photophysics of the polymer system.

I will first use a series of novel sulfone-containing linear conjugated polymer photocatalysts to demonstrate that excitonic hole transfer from the polymer to the sacrificial donor can occur on the picosecond timescale, strongly competing with exciton recombination [2]. Crucially, the ability of the best-performing polymer to generate long-lived electrons is dependent on the thermodynamic driving force for hole transfer to the sacrificial donor.

Alternatively, excitons can be separated using an organic heterojunction. However, whilst a PM6:Y6 photoanode can efficiently separate excitons on the picosecond timescale with no applied bias, long-lived active charges are only observed when spatially separated by an applied bias [3]. We similarly find that low charge mobility in carbon nitride photoanodes limits their ability to physically separate charges, and can be improved by introducing a conductive network into the electrode [4]. Building on these observations, we demonstrate that the lifetime and yield of PM6 hole polarons can be improved in the PM6:Y6 photoanode by adding a PM6 overlayer on top of the bulk heterojunction. The PM6 overlayer aids the spatial separation of charges, retarding bimolecular recombination [3]. Finally we show that the inclusion of oligo(ethylene glycol) side chains to a polymer photocatalyst also extends the lifetime and yield of the active electrons relative to its alkylated analogue [5]. In this case, the hydrophilicity of the side chains causes substantial swelling, with the resulting spatial separation again reducing recombination.

Overall, I aim to show that a wide range of photo(electro)chemical organic materials and device structures can be improved through several common design themes.

Intensity Modulated Photocurrent Spectroscopy (IMPS) has been largely employed in semiconductors characterization, to probe the operando behavior with widely available facilities. However, the implementation of IMPS data analysis to complex structures, such as the semiconductor-electrolyte interface or buried heterojunctions, is generally limited to qualitative evaluation of the steady-state conversion efficiency. In this work, a new algorithm for the analysis of IMPS data is developed, providing unprecedented insights into the time resolved charge transfer/recombination paths in semiconductors for solar energy conversion, from photoelectrochemical (PEC) water splitting to photovoltaics. The algorithm, based on the previously developed Distribution of Relaxation Times (DRT) analysis, here modified with Lasso regression method, is validated on different PEC systems, such as Ti modified hematite photoanode for PEC water oxidation, identified as a standard platform in the field, but also CoPi modified hematite for valorization of biomass derivated and CIGS based photocathodes for water reduction. The application of the L-DRT algorithm to IMPS data provides several advantages compared to the conventional kinetic analysis, including straightforward identification of multiple charge transfer and recombination paths, improved time resolution, providing precise and reliable description of the characteristic lifetime of each process and extension of the potential window suitable to kinetic analysis due to the deconvolution of subtle recombination peaks [1], [2].