Artificial photosynthesis is considered a promising method for achieving carbon-neutral targets. The hydrogen evolution reaction (HER) from the photoelectrolysis of water and the photoelectrochemical (PEC) CO2 reduction have gathered significant attention as an effective way to store intermittent solar energy in fuels and chemicals, as well as closing the chemical carbon cycle. Unfortunately, the photoelectrode materials used in these reactions are often unstable or exhibit insufficient activity or selectivity for the CO2 reduction reaction (CO2RR). To show the technological path of this approach, we need to focus on addressing stability and efficiency of these systems.

In this context, we present a few examples of how light-absorbing materials can be utilized in integrated photoelectrochemical cells or when directly interfaced with the electrolyte for HER and CO2RR. Specifically, we demonstrate how we can analyze and enhance the stability and performance of various photoelectrode materials used in these reactions focusing on different classes of materials.

Electrocatalysis research has gained momentum recently owing to its crucial role in the conversion and storage of renewable energy. Numerous material science advancements have been demonstrated in fuel cells, water electrolysis, carbon dioxide, and nitrogen reduction technologies. Materials with high intrinsic catalytic activity and selectivity have been looked for in such research, aiming at improved performance and maximal product yields. However, approaching the commercialization stage, the research focus has shifted to the stability of electrocatalysts. Only materials that can demonstrate reliable operation over thousands to tens of thousands of hours are considered at this stage. Since typical testing time ranges are much shorter, a question arose – how can we quantify the degree of degradation and use it to predict stable operation over the years?

Numerous accelerated stress tests (AST) have been proposed to address this issue. Such tests imply that the underlying degradation governing mechanisms and their dependence on the device's operational conditions changes are well-known. Unfortunately, this is not the case for many technologies, sparking advancements in testing instrumentation development. Thus, the introduction of online inductively coupled plasma mass spectrometry (online ICP-MS) in the electrocatalysis research has assisted in resolving the mechanism of Pt dissolution in fuel cells and Ir dissolution in water electrolysis [1]. The application of identical location transmission electron microscopy (IL-TEM) has been crucial in understanding the morphological and compositional changes in the catalyst during the operation [2]. These techniques can also be combined with complementary surface science tools in in-situ and in-operando modes. When the catalyst layer rather than catalyst properties are to be studied, e.g., mass transport effects, gas diffusion electrode (GDE) half-cell setups are more suitable [3].

Despite the demonstrated benefits of such tools in electrocatalysis, their penetration in photoelectrocatalysis (PEC) research is still limited. This talk aims to change this status quo and motivate PEC researchers to adopt existing electrocatalysis techniques. To this end, a short overview will be given of how online ICP-MS, IL-TEM, GDE, and related methods have been used in fuel cell and water electrolysis research. A summary of our recent studies on the dissolution of representative photoabsorbers, such as Fe₂O₃, WO₃, BiVO₄, etc, and co-catalysts will follow this [4, 5]. The talk will be summarized by comparing and contrasting the state of the art in electrocatalysis and PEC research and discussing further tentative directions for the latter.

The development of photoelectrochemical (PEC) systems for solar fuel production has been focused mainly on the fabrication and study of highly efficient materials. However, when the main target is to increase performance in terms of photocurrent densities, the selection of operating conditions becomes inconsistent due to ad hoc approaches preferred by researchers to maximise photocurrent densities at the expense of stability, scalability and practical application. The operating conditions for increased stability and for improved solar-to-fuel efficiency are usually at odds. For example, BiVO₄ photoelectrodes report higher stability when used in near-neutral electrolytes, although photocurrent densities are higher when alkaline electrolytes are used [1]. Similarly, chronoamperometries performed at low electrode potentials usually register more stable, although lower, photocurrent densities compared to those at high electrode potentials. For BiVO₄ photoelectrodes, the effect of different operating condition has been studied separately, e.g. irradiance [2], temperature [3] and pH [4], although with a focus on performance instead of stability. More recently, the authors decoupled the effect of increased irradiance and temperature on the dissolution of BiVO₄ by examining the competing behaviour between these two factors [5]. However, the effect of other relevant operating conditions such as electrolyte concentration, pH, electrode potential and electrolyte flow, and their interactions, has yet to be reported.

In this study, we realised a systematic experimental study of different factors, including irradiance, temperature and electrolyte nature (pH, concentration and flow) and its effect on the degradation of spray pyrolyzed BiVO₄ photoelectrodes. For this, a 4-cell array was developed to perform simultaneous tests under different conditions. Fresnel lenses were used to achieve irradiances between 1 and 110 suns, and an external water bath to control the electrolyte temperature (25 to 50 °C). KPi buffer solution was used as electrolyte at different pHs (5 to 11) and concentrations (0.1 to 1.0 M). Chronoamperometries were performed for long term tests, while electrochemical impedance spectroscopy was used to assess in-operando the charge transfer processes at different stages during the degradation and to quantify the effect of operating conditions on the properties of the semiconducting material, i.e. flat band potential and donor density. The array allowed to investigate a wide range of operating conditions, and its combinatorial effect on the dissolution of BiVO₄. Additionally, seldom reported reproducibility and sensitivity studies were also performed.

It was found that increased temperature impacts negatively the stability of the photoelectrodes, while increasing the irradiance limits the amount of charge towards the dissolution process. However, increased irradiance also induces concomitant effects, that is, increased temperature and lowered pH at the surface of the oxygen-evolving photoanode. These effects were successfully decoupled by using the above-described experimental design. It was also found that the electrolyte concentration, and respective buffer capacity, have a significant effect on the stability of the photoelectrode, although to a lesser extent compared to pH, while the electrolyte flow has a minor but noticeable effect on the of the photoelectrochemical performance.

This study contributes to the broader field of solar fuel production by highlighting the importance of stability alongside efficiency in the development of new photoelectrode materials. Understanding the complex interaction between operating conditions and degradation mechanisms can prove useful in future research to tailor these conditions to optimize the longevity and efficiency of PEC systems. Additionally, the methodology discussed here can serve as a framework for upcoming systematic studies on photoelectrode materials.

The generation of hydrogen by renewable processes is a current challenge for researchers around the world [1, 2]. Photocatalytic water splitting is a method in which photocatalytically active particles are dispersed in an aqueous solution and illuminated by sunlight, resulting in the evolution of gaseous hydrogen and oxygen [3]. It is considered a promising approach, especially because of the low complexity of its technological implementation, which holds the promise of cost-competitive hydrogen [4].
To be suitable as a photocatalyst a material is required to fulfill criteria, such as a suitable band gap size, band edges that straddle the water splitting reaction potentials and good stability under reaction conditions. The n-type semiconductor BiVO₄, with a band gap of 2.4 V, is considered to fullfil these criteria for the oxygen evolution reaction.
One of the most investigated methods to modify the optical and electronic properties to enhance the photocatalytic performance is doping [5]. For example Wang et al. have shown an increased photocatalytic performance for cationic substitution of BiVO₄ with Mo [6] and Rohloff et al. have shown an increased photocurrent density for BiVO₄ photoelectrodes by anionic doping with fluorine [7].
In this study the influence of the band gap shift due to anionic substitution of chlorine into BiVO₄ is investigated. Therefore, BiVO₄ particles are prepared via hydrothermal synthesis with the addition of halide salts, leading to partial substitution of oxygen in the lattice. The crystal structure is characterized by X-ray diffraction and the morphology by scanning electron microscopy. The chemical composition is analyzed by energy dispersive x-ray spectroscopy and electron energy loss spectroscopy. UV-vis spectroscopy is performed to investigate the optical properties. The oxygen evolution rate is determined by gas chromatography during the photocatalytic reaction in the presence of a sacrificial reagent.

Life cycle assessment (LCA) studies clearly indicate that a minimum 10% solar to hydrogen (STH) conversion efficiency and a 10-year lifetime are required for a positive energy return on energy invested (ERoRI) for photoelectrochemical (PEC) water splitting [1-3]. Analysis of published reports reveals that many lab-scale and larger demonstrations satisfy the STH criterion, but no demonstration comes within an order of magnitude of the lifetime requirement. A drastic improvement in the demonstrated/projected lifetime of PEC water splitting schemes is urgently needed if this approach is to compete with hydrogen generation via renewables (wind/solar) coupled to electrolyzers. Although the LCA and technoeconomic analysis (TEA) of PEC CO₂ reduction (CO₂R) are less numerous, examination of the related electrocatalytic literature points to minimum lifetimes of five years or more [4].

Two experimental systems which address the PEC CO₂R stability challenge will be presented.

Many promising light absorbers are not stable under the conditions of PEC CO₂R reduction. Use of a transparent electron transport layer (ETL) can address this challenge, but the ETL must have good electronic transport, be stable under CO₂R conditions, and, ideally, be catalytically inert for the competing hydrogen evolution reaction (HER). Oxygen-deficient TaO_x satisfies these criteria, and p-Si/TaO_x/Cu photocathodes have CO₂R faradaic efficiencies (FE) >50% at photocurrent densities up to 8 mA cm^-2 under 1 sun illumination with multi-hour stability [5]. Dissolution of the Cu co-catalyst appears to be the dominant degradation mechanism, suggesting that redeposition of the catalyst could extend the lifetime.

Are there any materials which have intrinsic activity/stability (that is, without the aid of electron transport/protection layers and/or co-catalysts) as photocathodes for CO₂R? This question is especially pertinent if operation in aqueous media is considered. Reports of photocatalytic CO₂R using metal sulfides are suggest a starting point for materials discovery [6]. More specifically, the Cu(In,Ga)(S,Se)₂ (CIGS) alloy family is interesting due to the extensive study of its properties as photovoltaic materials and its wide bandgap tuning range. Indeed, co-catalyst free Cu(In,Ga)S₂ (CIGS) thin-film photocathodes (Eg ~1.8 eV) reduce CO₂ to CO and HCOO^- in aqueous media at FEs of 28-32% and 14%, respectively. Extensive structural characterization (Raman, ambient pressure XPS, XAS) shows that Cu (In,Ga)S₂ photocathodes are stable for at least a few hours. Interestingly, as would be predicted by considerations of equilibrium (Pourbaix) stability, Se-alloyed photocathodes corrode rapidly. Additionally, Cu(In,Ga)S₂ films with lower bandgaps also appear to be unstable. These findings suggest that the previously unexplored Cu-deficient surface composition and specific surface defects, especially deep anti-site defects, might be playing a key role in governing the PEC CO₂R stability of CIGS-based photocathodes..

Among the various strategies proposed for enhancing the photoelectrochemical (PEC) performance of state-of-the-art photoelectrodes, the construction of heterostructures based on semiconductors and co-catalysts is widely acknowledged as one of the most versatile and reliable. This approach is often associated with higher quantum efficiency for photoinduced redox processes, due to improved interfacial charge transfer dynamics, and reduced degradation due to photocorrosion, as the photoinduced charge is less susceptible to this less competitive pathway. However, numerous studies[1-3] have demonstrated that this simplistic view conceals a much more complex microscopic nature, often significantly dependent on the specific characteristics of the interfaces. Therefore, understanding the redox dynamics in such complex systems is essential to direct the carriers along the desired pathways. Several indirect methods are available for this purpose, such as electrochemical impedance spectroscopy (EIS) and transient absorption spectroscopy (TAS), but operando X-ray absorption spectroscopy (XAS) stands out as the most promising technique[4,5]. Operando XAS can provide direct, element-selective information about changes in the redox state and the local environment of the co-catalyst metal centers.

In line with this aim, we report our recent advancements in the development of operando PEC-XAS experimental techniques to understand the redox dynamics of various semiconductor/co-catalyst assemblies, fully replicating the operating conditions of a PEC cell. This tool is employed to investigate potential and light-induced processes in i) BiVO4/WO3 heterostructures coated with Co-based co-catalysts, providing evidence for specific electronic interactions with the semiconductor, and ii) Ni-based co-catalysts on hematite photoanodes, offering insights into the rate-determining step of PEC oxidation of biomass derivatives by such systems. Additionally, operando PEC-XAS has proven to be an invaluable tool for monitoring degradation processes, as it enables structural understanding and real-time monitoring of the dissolution processes of the catalysts' active elements.

Transition metal nitrides and oxynitrides are a promising class of functional materials with tunable electronic and optical properties based on anion and cation composition. As such, they offer significant potential for custom-designed applications in photoelectrochemical (PEC) energy conversion. This contribution summarizes our group’s recent advancements and insights into transition metal nitride and oxynitride thin films, emphasizing their potential for solar fuels synthesis. Despite their promise, transition metal nitrides and oxynitrides have been less explored than their oxide counterparts due to complex synthesis requirements. We address these synthetic challenges using non-equilibrium reactive sputter deposition, allowing us to deposit well-controlled thin films within the Ti-Ta-N, Zr-Ta-N, and Zr-O-N composition spaces. Our investigations start with orthorhombic Ta₃N₅, the most established nitride-based photoanode material. With a highly controlled synthesis approach and tailored defect concentrations, we examine the impact of shallow and deep trap states on the operational stability of Ta₃N₅.[1] By comparing water and ferrocyanide oxidation conditions, we observe that shallow oxygen donors within Ta₃N₅ can kinetically stabilize the photoelectrode|electrolyte interface. In contrast, deep-level defects are generally detrimental to its PEC performance. We demonstrate, though, that the concentration of such deep-level defects can be dramatically reduced by controlled Ti doping.[2] While Ti doping minimally affects the structure and band gap of Ta₃N₅, it significantly impacts surface photovoltage, band bending, and photogenerated charge carrier lifetimes. Comprehensive characterization reveals that Ti⁴⁺ ions substitute for Ta⁵⁺ lattice sites, introducing compensating acceptor states and reducing concentrations of deleterious nitrogen vacancies and Ta³⁺ states. This reduction of deep-level defects suppresses trapping and recombination, leading to enhanced PEC activity. Beyond orthorhombic Ta₃N₅, we have explored novel photoanodes with cubic bixbyite-type structures, such as ZrTaN3 and Zr₂ON₂, featuring multiple cation and anion identities. We have recently identified the ternary nitride ZrTaN₃ as a strong visible light absorber and functional photoanode thin film.[3] Complementary density functional theory (DFT) calculations indicate that ZrTaN₃ has a direct band gap that can be tuned by modulating the elemental occupancy of inequivalent cation sites. Finally, changing the anion composition in the Zr-O-N space allows us to tune the band gap in the UV-visible range, achieving PEC activity for oxidation reactions.[4] For crystalline Zr₂ON₂, we observe mild surface oxidation beneficially passivates the surface, while too-thick surface oxides suppress charge transfer and PEC activity. This observation highlights an appealing feature of oxynitrides as photoanodes: the formation of self-passivating surface oxide layers. This passivation makes them suitable for integration with ultrathin atomic layer deposition protection layers, whose major drawbacks are pinholes that can potentially be tolerated by (oxy)nitride semiconductors through the passivating surface oxide layers. Overall, our results underscore the potential of the defect engineering of established and the development of novel transition metal nitride and oxynitride semiconductors for achieving robust and stable materials in solar energy conversion.

The conversion of energy from renewable and CO₂ free energy sources into chemical energy, has the potential to contribute significantly to cover our energy needs [1]. Photoelectrochemical (PEC) water splitting, with its simpler setup compared to combining a photovoltaic cell and an electrolyser, shows promise as a cost-effective method for producing green hydrogen in the future [2]. A PEC system consists of two separate electrodes connected by an ohmic contact, each providing a site for one of the two water-splitting half-reactions [2].

Research over the past decades has primarily focused on improving the photoanode efficiency, leading to a thorough understanding of the material properties required for high conversion efficiency [3-5]. However, for PEC systems to be viable on an industrial scale, the long-term stability of the photoelectrode materials is essential. These materials must withstand degradation  by the electrolyte and resist photo corrosion [2, 6] Investigating the degradation processes of photoelectrodes is therefore important from both scientific and practical perspectives, as it is crucial for developing photoelectrodes with longer lifespans [3, 7].  

In photoanodes, photogenerated holes  are known to participate either in the oxygen evolution reaction, to be lost to recombination, or to cause the oxidation of anions in the photocatalyst [6]. This oxidation process can lead to the dissolution of metal ions into the electrolyte, as the crystal lattice becomes destabilized due to changes in the local structure around the oxidized ions [8, 9].

Since oxynitrides are known for their good performance with respect to efficiency, LaTiO₂N based photoanodes were selected for this study on degradation. The LaTiO₂N was synthesized via solid state synthesis followed by thermal ammonolysis. The photoanodes were prepared by electrophoretic deposition followed by TiCl₄ necking and cocatalyst application via dip coating. 

The degradation of the LaTiO₂N based photoanodes was investigated as a function of the illumination conditions and the electrolyte temperature using chronoamperometry in combination with linear scan voltammetry. IPCMS was used to detect potential dissolution of cations into the electrolyte. The photoanodes were investigated postmortem with respect to compositional, morphological and optical changes using STEM-EDX, SEM and UV-Vis-Spectroscopy.

A large body of literature exists on photocatalytic H2 generation using a wide variation of semiconductors, morphologies, and strategies to split water using the semiconductors suspended in an aqueous solution (with or without sacrificial agents). Many semiconductors have in common that for an efficient transfer of photogenerated charge carriers, a co-catalyst is required. For electron transfer and H2 generation mostly Pt nanoparticles are used that are deposited onto the semiconductor surface by various techniques. Due to the precious nature of Pt, over the years, numerous efforts have been devoted to the shrinkage of the particle size and thus to enhance the utilization of the noble metal – in the most extreme case down to an insulated single atom of Pt.

In the presentation we discuss the use of Pt dispersed and anchored as single atoms (SAs) on TiO2 surfaces and the activation to a most efficient use for photocatalytic H2 generation. We discuss various trapping and stabilization approaches of SAs on photocatalysts that prevent agglomeration (and accordingly the deactivation of SA Pt). Moreover, we show that only a very small amount of Pt (loading density of SAs ) is needed to achieve a maximum activity of a semiconductor surface.

The climate crisis is one of the most pressing challenges of our time. Global and coordinated action to reduce greenhouse gas emissions and mitigate their impacts is indispensable. Proposed solutions to address this crisis include the adoption of renewable energy, increased energy efficiency, reforestation, and the use of innovative technologies. In the technology sector, photocatalysis is attracting growing interest, fueling intense research into new materials.¹ In our study, we examined the photocatalytic activity of heterojunctions between Bi₂WO₆ and Cs₃(Bi_1-xSb_x)₂Br₉. Bi₂WO₆ is a widely recognized graphitic material in the field of photocatalysis, while perovskites, known for their excellent tunability, prove to be outstanding co-catalysts.^2,3 Lead-free inorganic perovskites have been the subject of numerous studies for years.^4,5 Of particular interest is the introduction of different metals into the B-site, introducing a high entropy approach which is still in its infancy.⁶ In our study, we analyzed how the structural and optical properties change with increasing doping rates. The formation of a heterojunction between the two materials could create a more complex charge carrier transport pattern than with single materials, increasing their lifetime and thus improving photocatalytic results.⁷ The heterojunction is facilitated by the innovative in-situ synthesis adopted, in which perovskite is formed within a solvent in which Bi₂WO₆ has been dispersed. The aim of the project is the photocatalytic splitting of water and CO₂, underlining the importance of property-structure relationships also in the photocatalytic field. The project aims to highlight the differences in various set-ups, whether liquid-liquid or gas-solid. The discovery of new materials and the acquisition of new knowledge about photocatalysts are key to the development of advanced technologies for CO₂ reduction. Photocatalytic methane evolution paves the way for a circular ‘waste to fuel’ economy in which waste molecules such as CO₂ regain value.

Photoelectrochemical (PEC) devices can mimic photosynthesis and show great promise for sustainable fuel production. These artificial leaves integrate light absorbers with suitable catalysts to directly harness, convert, and store abundant solar energy in the form of value-added chemical fuels.^[1] However, most conventional prototypes employ wide bandgap semiconductors, moisture-sensitive inorganic light absorbers, expensive materials or corrosive electrolytes. Here, we introduce the design and assembly of PEC devices that contain an organic π-conjugated donor-acceptor bulk heterojunction (PCE10:EH-IDTBR) with sufficient photovoltage for both proton reduction and CO₂-to-syngas conversion.^[2] The rational combination of design strategies from organic photovoltaic (OPV) and inorganic PEC fields, coupled with a carbon-based encapsulant, promoted long-term H₂ production over 12 days in benign aqueous media. Given the modular nature of our device design, interfacing the devices with a molecular cobalt porphyrin catalyst allowed for tunable and selective CO production under 0.1 sun. Further assembly of these OPV photocathodes with BiVO₄ in a standalone artificial leaf demonstrated unassisted concurrent CO₂ reduction and water oxidation over 4 days. This establishes a new path for organic semiconductors, as we approach the composition, function, and efficiency of natural leaves.

The pursuit of sustainable green fuel production through photocatalysis is a central theme in modern scientific exploration. The focus lies in developing an active photocatalytic architecture capable of efficiently converting light into energy. Plasmonic metals, especially gold, have garnered attention due to their unique optical properties, known by tuneable localized surface plasmon resonance (LSPR).¹ However, a significant trouble in achieving overall efficiency is the limited lifetime of photogenerated charge carriers, known as hot carriers.²

Addressing this challenge involves strategically incorporating a metal or semiconductor near the plasmonic entity. This approach aims to extend the recombination time of hot carriers, enhancing the overall efficiency of the photocatalytic system. While considerable developments have been made, a systematic exploration into the deposition of metals, semiconductors, and hybrid systems remains crucial for advancing plasmonic photocatalysis.³

This research depicts the key strategies and motivations driving the engineering of efficient architectures in plasmonic photocatalysis, including metal-metal, metal-semiconductor, and metal-semiconductor-metal configurations. The presented research involves photocatalytic applications of diverse plasmonic nanostructures, such as gold@gold-silver alloy nanostructures, gold@manganese oxide core-shell nanostructures, and gold@platinum core-shell nanorods deposited on ceria particles. These nanostructures showcase the potential for light-enhanced photocatalysis, offering valuable insights into sustainable energy solutions.^4-6 Through a detailed understanding of plasmonic-metal-semiconductor interactions, this research contributes to the ongoing efforts to harness renewable energy sources and advance green fuel production via innovative photocatalytic architectures.

Polymeric carbon nitride (CN) materials have emerged as metal-free, low-cost, and environmentally friendly semiconductors in various applications, including photoelectrodes in photoelectrochemical water-splitting. Unlike CN powder, which is used as a dispersed photocatalyst, for applications such as photoelectrochemical cells, light-emitting diodes, and solar cells, the deposition of the CN on a conductive substrate is required. The deposition of CN layers on different substrates can be divided into two main categories: 1) ex-situ deposition of prepared CN powder and 2) in-situ growth of a CN layer directly on the substrate. Generally, the in-situ methods comprise two steps: the first one is the deposition or growth of nitrogen-rich monomers (such as melamine, urea, thiourea, dicyanamide, etc.), forming a film of the monomers on the substrate. The following step is the calcination process, in which these monomers polymerize, forming the final CN electrode. Most in situ preparation techniques differ in the deposition or growth method of the monomers' film but maintain a similar 'standard' calcination process of slowly heating this film to 500–550 °C and keeping the final temperature for several hours, usually under an inert atmosphere. A significant drawback of a long heating process is the possible sublimation and decomposition of the CN monomers and final layer, which may lead to a less uniform film. In addition, from an economical point of view, a long calcination process at a high temperature consumes much energy.

In this talk, I will introduce a facile, scalable, energy-saving, and reproducible synthesis of CN layers on conductive substrates using a fast heating method. In this method, the predesigned CN monomer films are subjected for several minutes (5-20 min) to higher temperatures than the 'standard' calcination procedure (650–680 °C). The high-temperature process enables fast condensation of the monomers, and negligible degradation is obtained thanks to the short time at the target temperature. As a result, the formation of a uniform CN layer with excellent contact with the substrate and good activity as a photoanode in PEC is achieved. The optimal CN photoanode reaches photocurrent densities of ~200 μA cm⁻² at 1.23 vs. RHE in neutral and acidic solutions and 120 μA cm⁻² in a basic solution.^[1]

While III-V semiconductors have achieved the highest photo-electrochemical solar-to-hydrogen conversion efficiencies, they are remarkably unstable during operation in a harsh electrolyte. The first half of this talk will focus on the degradation mechanism of III-V cells and surface modification strategies aimed at protecting them from photocorrosion. We applied noble metal catalysts, oxide coatings by atomic layer deposition, and MoS₂ in an effort to protect the GaInP₂ surface that was in contact with acidic electrolyte. We also grew epitaxial capping layers from III-V alloys that should be more intrinsically stable than GaInP₂. The ability of the various modifications to protect semiconductor surfaces was evaluated by operating each photoelectrode at short circuit for extended periods of time.

In the second part of this talk, I’ll discuss potential pitfalls to consider when characterizing semiconductor materials for durability. Measurements of photoelectrode durability in three-electrode (half-cell) configurations are typically not representative of the results obtained for nominally the same electrodes/materials when tested in a two-electrode (full-cell) configuration. While full-cell measurements are the best proxy to predict performance in a deployed photo-electrochemical water-splitting system, there are few materials that can drive both the water reduction and oxidation half-reactions. Component-level or half-cell testing is the only option for materials unable to perform unassisted water splitting. During this talk the results of multiple durability tests, some with and some without a reference electrode, will be used to evaluate key differences between the two types of testing in an effort to elucidate what leads to the incongruity between half- and full-cell measurements. It is anticipated that with this understanding, the photoelectrochemical water-splitting community can begin to move towards an accepted protocol for long-term durability testing that will have more predictive relevance to realistic device configurations.

In this talk, I will present recent advances of artificial photosynthesis utilizing gallium nitride (GaN), which is the second most produced semiconductors next only to silicon. Through nanoscale, quantum, and catalyst engineering, conventional GaN-based semiconductors can be transformed to be efficient and stable photocatalyst materials for a broad range of artificial photosynthesis reactions, including solar water splitting, carbon dioxide reduction, methane oxidation, and nitrogen reduction to ammonia. By growing GaN nanostructures under N-rich conditions, the nonpolar surfaces can be transformed to be gallium oxynitride during harsh photocatalysis reaction, which not only protects the surfaces of the light absorber but leads to significantly enhanced photocatalytic and photoelectrochemical performance. Moreover, we have developed unique photocatalytic processes wherein high efficiency solar hydrogen can be produced utilizing tap water, or seawater, without any wire connection, or electricity input. The demonstration of large-scale solar water splitting systems and the performance will be discussed and reported, together with advances in carbon dioxide and nitrogen reduction to clean chemicals and fuels.

One of the current issues in the electrochemical CO2 reduction reaction (CO2RR) is the stability of the catalyst. [1] Copper is the most promising material for producing desirable C2+ chemicals at reasonable rate, yet this metal reconstructs during operation. [1] A solution is to either block this reconstruction or to learn how to direct it towards structures with the desired selectivity. [2]  

Herein, we propose to explore a different class of materials. These materials are liquid metal Ga-based nanoparticles (NPs). We propose them as alternative to traditional solid catalysts with great potential for stable CO2RR thanks to their self-regenerating dynamic surface. [3,4] Yet, we are still learning about their chemistry, which is at its infancy compared to other NPs. [5,6] Developing their chemistry is important to further explore them for selectivity in CO2RR and eventually for other reactions.

In this talk, we will focus on the most recent work where we synthetize tunable and monodisperse Ga-based NPs incorporating a variety of different metals through colloidal chemistry. In particular, we prepare solid-liquid-solid Ga-M NPs (M= Ag, Cu, Au, Pd) wherein a solid metal is encapsulated within a liquid Ga NP confined by its oxide skin. We elucidate the formation mechanism of these unique nanostructures through a combination of state-of-the-art in-situ techniques, including electron microscopy and X-ray absorption spectroscopy. Finally, we demonstrate the functionality of these NPs as CO2RR electrocatalysts.

Coupled multi-physics modeling can reveal new insights into the processes that govern the performance and degradation of photoelectrodes under varying operation conditions [1]. While many subprocesses have been studied to a great extent, a unified model for PEC device operation is still a work in progress [2]. For example, recent studies use simplified models for the charge transport in the semiconductor [3] or neglect the electrolyte species transport [4]. Also, charge transfer theory at the semiconductor-electrolyte junction (SCEJ) coupled to the solid and liquid has been seldom investigated [5]. Furthermore, the modeling of (time-dependent) degradation through (electro-)chemical corrosion has been initiated recently yet with simplified charge transfer and neglected species transport [6].

Herein, new physics relevant for photoelectrode operation have been included to predict the performance and stability of photoelectrode material systems. Transport of electrons and holes in the semiconductor were modeled with Poisson-drift-diffusion equations including carrier generation and recombination terms. Special attention was put on the surface recombination by trap states. Band edge unpinning was introduced through surface state (dis-)charging, which allowed to study a surface state-passivating co-catalyst. Transport of species in the electrolyte was modeled in a multi-modal manner including diffusion, drift in electric field and drift in boundary layer flow. Charge transfer modeling at the SCEJ based on Marcus Theory was implemented to study simultaneously multiple redox reactions of different energetics and kinetics. Degradation was modeled as competition between charge transfer to electrolyte species and charge capture by surface bonds with time-dependent photoelectrode dissolution.

A BiVO₄ photoanode with CoPi co-catalyst for water splitting served as case study for model validation and demonstration. Parameters of the SCEJ and charge transfer were determined through first principle calculations and with dedicated experiments. Voltammograms were used for validation by variation of operational conditions (pH, temperature and irradiance) and its effect on, otherwise, inaccessible material properties was assessed by sensitivity analysis. Voltammograms showed a mild pH-dependence resulting from changes in the surface state charging and electrolyte species concentrations. Under concentrated sunlight (> 10 kW m^-2), a cathodic shift in the onset potential was observed, dependent on surface state charging and recombination. The effect of mass transport limitation on the photocurrent caused by sluggish species near the SCEJ was also quantified. A photocurrent dependence on the illumination direction (front vs. back), typical for BiVO4 due to limited majority carrier transport, was confirmed in the model. The dissolution-based degradation model was able to capture characteristics of the photocurrent decrease experimentally observed in BiVO4 photoanodes.

In-depth parametric studies on photoelectrode operation (including degradation) were conducted with a coupled multi-physics model. The model has capabilities to be expanded to include multiple dimensions and complex photoelectrode structures to provide more insights into the heterogeneity and local variation of operation condition in a photoelectrochemical device or component. Moreover, additional degradation mechanisms, e.g. surface passivation, will permit stability studies of a broad range of photoelectrode materials.

The atomistic understanding of complex reaction mechanisms in (photo-)electrocatalysis aids not only the discovery of improved catalytic materials but also the choice of ideal reaction environments for tailored products.

In my talk, I will present density functional theory-based studies on electrocatalytic reaction mechanisms with a special focus on electrochemical CO₍₂₎ reduction and biomass valorization. I will describe how the combination of constant-potential DFT approaches and transition state theory-based considerations allow us to explicitly study the potential, pH and electrolyte dependence of multistep reaction networks relevant for the green transition [1].

Further, I will discuss general trends in the thermodynamic and kinetic preferences of the competing elementary reactions in electrocatalytic reductions. Here, I will show how the potential and pH response of specific reaction pathways can be exploited for tuning product selectivity [2]. Finally, I will show the generality of the found trends by extrapolating from electrochemical CO₍₂₎ reduction to the electrochemical reduction of Furfural [3].

In most photo-electrochemical devices the kinetic phenomena does not only imply the reactions we are aiming at but also some others that are linked to the dynamics of the material acting as catalysts. Some of them are positive for the performance of the material increasing the number or improving the nature of active sites. Still, some others are detrimental for the system and can cause failure. In the talk I will address what can we do from the point of view of atomistic simulations to start understanding this phenomena so that we can control the nature and prolong the stability of the active sites under true reaction conditions. I will elaborate on the gaps in our understanding and the role of accelerators to investigate this long-term effects in the active materials. Finally,  I will indicate the most relevant challenges in the field highlighting what is needed to progress further in this area.

Oxide and oxynitride oxygen-evolution (OER) photoelectrodes in contact with an electrolyte and under an applied bias and/or light irradiation undergo often irreversible changes in surface structure and composition. Computational determination of the energetically most favored changes as a function of experimental parameters constitutes a powerful complementary technique to experimental spectroscopy and local-probe studies. Perovskite tantalate photoelectrocatalysts exist as bulk perovskite oxides, layered oxides, and bulk perovskite oxynitrides. Based on density functional theory calculations, we determine likely alterations under application conditions and examine their effect on the catalytic activity. For the layered Sr₂Ta₂O_7, we predict a partial dissolution of the surface that deactivates the catalyst. A similar structural change on the NaTaO₃ surface, however, leads to enhanced catalytic activity by enabling an alternative OER mechanism. For SrTaO₂N we predict a loss of nitrogen from the surface layers, the resulting electron doping of the surface also leading to a deactivation of the OER. These results show that compositionally similar materials can undergo very diverse changes under OER application conditions that, while mostly detrimental, can - in select cases - also improve the catalytic activity.

Understanding the dynamics at catalyst-liquid interfaces is crucial for gaining insights into degradation/deterioration of catalysts and complex reaction such as the oxygen evolution reaction (OER). Notwithstanding LaTiO₂N and BiVO₄ are gaining attention as active photocatalysts for OER, a lot of questions are still elusive such as the identification of active catalytic surface sites, their catalytic properties as well as the organization of water at the catalyst-water interface. Spin-polarized density functional theory-based molecular dynamics (DFT-MD) enables extensive investigations of various catalytic systems, explicitly considering environmental factors such as solvents under ambient conditions. For instance, possible degradation mechanisms of LaTiO₂N (100) surface have been investigated by DFT-MD. Notably, we identified the transfer of oxygen atoms from the sub-layer to the surface as a potential degradation process for LaTiO₂N. The generated oxygen vacancy in the sublayer can potentially propagate into the bulk and break the lattice. For BiVO₄ we investigated the desorption process of surface vanadium atoms from surface towards the solvent employing enhanced sampling techniques, which is powerful tool for the understanding of reaction mechanisms and calculation of free energy surfaces.  Pourbaix diagrams have been calculated in order to understand the stable LaTiO₂N and BiVO₄ surface configuration at different electrochemical conditions. Preliminary catalytic activity of LaTiO₂N and BiVO₄ in catalyzing the OER have been elucidated by DFT-static calculations for free-energy change, obtaining a good agreement with the available experimental results. Our research has also focused on the implementation of grand-canonical-ensemble approaches in the CP2K simulation software. Testing and applications have been applied on LaTiO₂N and BiVO₄ systems as well.

Halide perovskites and organic bulk heterojunction semiconductors have attracted significant interest for photovoltaic applications due to their excellent optoelectronic properties, such as strong solar absorption, wide defect tolerance, and long charge diffusion lengths. These properties are also crucial in the development of photoelectrochemical devices for solar fuels and chemicals. However, the instability of these devices in aqueous environments must be addressed. In this talk, I will present our team's recent progress in protecting CsPbBr₃ halide perovskite and PM6:D18:L8-BO and PTQ10:GS-ISO organic bulk heterojunctions with various carbon allotrope layers and sheets. These include mesoporous carbon, graphite, glassy carbon, and boron-doped diamond, which are decorated with electrocatalysts. These layers and sheets offer protection and catalytic activity but can degrade under harsh conditions, such as those required for oxygen evolution—a bottleneck reaction in many solar fuel and chemical productions. The addition of Ni and NiFeOOH is crucial to ensuring photocurrent stability, maintaining photocurrents above 6 mA cm^-2 with projected stability of months under harsh +1.23 V vs RHE applied bias on CsPbBr₃ photoanodes. A similar approach provides PM6:D18:L8-BO photoanodes that achieve 25 mA cm^-2 at +1.23 V_RHE and monolithic tandem organic photoanodes with PM6:D18:L8-BO and PTQ10:GS-ISO with 5% unassisted solar-to-hydrogen efficiency, both showing days-long stability. In these cases, the stability is mainly limited by the morphological instability of organic bulk heterojunctions. Oxygen bubble accumulation on the surface of these devices is also a limiting factor for photocurrent stability. These and other challenges in achieving stable photocurrents in perovskite and organic bulk heterojunction photoanodes will be discussed.

Gas diffusion electrodes, crucial for fuel and electrolysis cells using gas-phase reactants, are often made from materials like graphitic carbon or metals, which block light and hinder photoelectrode membrane assemblies for solar fuel production. This presentation introduces transparent gas-diffusion layers using F-doped SnO₂ (FTO) coated SiO₂ fiber felt substrates for solar H₂ production and water splitting from humid air. The initially demonstrated substrates exhibited a porosity of 90%, a roughness factor of 15.8, and a Young’s Modulus of 1 GPa.[1] FTO coating provides a sheet resistivity of 20 ± 3 Ω sq⁻¹ and 50% transmittance in the 300nm-800nm range for the transparent conductive porous substrates (TPCSs). Various semiconductors, including Fe₂O₃, BiVO₄, Cu₂O, and semiconducting polymers, were deposited on the TPCSs, showing superior photoelectrochemical performance compared to flat FTO photoelectrodes. Moreover, strategies to improve the robustness, tune the porosity, and scale up the production of the TPCSs are herein presented, and the challenges and limitations of gas-phase PEM-PEC water splitting are discussed. Finally, unassisted gas-phase water splitting is demonstrated with a PEM-PEC cell at 1-sun, achieving a photocurrent density on the order of 1 mA cm^–2 with a system integrating a BiVO₄ photoanode and a polymer semiconductor photocathode for complementary light absorption together with a polymer electrolyte membrane.

Membrane-photoelectrode assemblies, i.e. photoelectrodes on porous substrates coated with an ionomer and directly integrated with a polymeric ion-exchange membrane, are a promising design configuration to perform photoelectrochemical water splitting [1]. This compact and modular design inspired by commercial electrolyzers allows to feed the system with pure water (in liquid or gaseous phase) minimizing the products crossover.

The most common proton-exchange membranes (PEMs) absorb light only in the UV region (for λ< 400 nm) but they impose an acidic pH which causes the corrosion of the majority of the catalysts. The most used anion-exchange membranes (AEMs) are partially opaque in the visible range but the alkaline environment assures the chemical stability of a large variety of materials. Although multiple demonstrations of proton-exchange and anion-exchange membrane-photoelectrode assemblies have been reported [2], the effects of (photo)corrosion of the semiconductors and/or of the co-catalyst used were not investigated in-depth.

Here, we studied the dynamics and the effects of photocorrosion of molybdenum-doped bismuth vanadate (Mo:BiVO₄) photoanodes with cobalt phosphate (CoPi) co-catalyst integrated in proton-exchange and anion-exchange membrane-photoelectrode assemblies with commercial Pt/C cathodes. BiVO₄ suffers from (photoelectro)chemical instability in neutral and alkaline solutions, leading to homogeneous film dissolution [3]. We investigated for the first time the (photo)corrosion of the Mo:BiVO₄ photoanode with CoPi co-catalyst in contact with hydrated ionomers.

The photoanodes were deposited on porous metallic felts through a dip coating technique followed by annealing and the photoelectrodeposition of CoPi co-catalyst. The preliminary optimization of the dopant concentration, semiconductor loading and co-catalyst deposition time was done testing the performance of the coated felts in a simplified cell configuration with liquid electrolyte. The optimal photoelectrodes on metallic felts were placed in thin layers of ionomer solutions obtained by a doctor blade method. After the evaporation of the solvents, the ionomer-coated felts were hot pressed to form membrane-photoelectrode assemblies.        

Chronoamperometric durability tests of the membrane-photoelectrode assemblies with liquid water and humid air at different temperatures were interspersed with cyclic voltammetries and impedance spectroscopies to determine the degradation rates in these different operating conditions. Scanning electron microscopy, energy dispersive x-ray spectroscopy and x-ray photoelectron spectroscopy of the aged samples with inductively-coupled plasma mass spectroscopy of the liquid solutions were used to observe the effects of Mo:BiVO₄ and CoPi photocorrosion in PEM and AEM assemblies.

The methodology introduced and the setup developed allow to investigate the photostability of membrane photoelectrode-assemblies with other semiconductors and co-catalysts with the aim to develop efficient, sustainable but also durable materials and devices for the conversion of solar light into fuels.

BiOBr is used as a promising material in various photoelectrocatalytic reactions, but its application in hydrogen evolution reaction (HER) is not commonly reported, because the instability under photoelectrochemical environment has become a drawback for BiOBr as a photoelectrocatalyst for reduction reactions.¹ To solve this problem, MoS2 is induced to form a van der Waals heterojunction to stabilize the crystal structure of BiOBr in HER.² By DFT calculation, we state that the active sites in the heterojunction are on sulfur centers, which also combines BiOBr by van der Waals force. Moreover, different MoS2/BiOBr ratios can result in different structure tolerance of BiOBr: heterojunction with 1% MoS2 can increase the stability of BiOBr while 50% MoS2 even accelerate the reduction of BiOBr. By performing the in situ wide-angle X-ray diffraction (WAXD) on MoS2/BiOBr with 1% and 50% of MoS2, respectively, we monitored the phase transfer speed of BiOBr during the HER. Interestingly, when UV light is induced, there is less amount of BiOBr reduced under negative potential due by the photogenerated holes that could react with extra electrons from negative bias or photo energy and prevent BiOBr to be further reduced.³