High-throughput methods offer a powerful strategy to accelerate the development of electro- and photoelectrocatalytic materials by enabling the rapid exploration of compositional, structural, and processing parameters. In this work, we present an integrated approach based on the fabrication of materials libraries with controlled gradients, combined with spatially resolved electrochemical and photoelectrochemical screening. This methodology allows multiple material configurations to be investigated within a single sample, reducing experimental time while providing direct structure–performance correlations.

The versatility of the approach is demonstrated through selected application examples. Ti-modified hematite photoanodes are investigated as model photoanodes for water oxidation, with high-throughput screening used to correlate annealing temperature, Ti concentration, and deposition sequence with photocurrent response and charge-transfer behavior (1). In a second example, Cu-based cathode architectures are explored for electro- and photoelectrochemical CO₂ reduction, including Cu₂O-based multilayers and mixed Cu-containing systems designed to tune charge transport, interfacial properties, and catalytic activity. Finally, Si-based photocathodes coated with insulating layers of gradually increasing thickness are used to study thickness-dependent effects at the semiconductor–electrolyte interface, to quantitatively monitor the photovoltage in the operating photoelectrode and the associated potential losses (2). This example highlights how high-throughput sample design can support systematic mechanistic investigations, beyond simple activity screening.

Overall, the proposed high-throughput strategy provides a flexible platform to connect synthesis, interface engineering, and functional performance in complex photoelectrochemical systems. By guiding targeted characterization and accelerating the identification of optimal materials architectures, it supports the rational design of improved catalysts and photoelectrodes for solar fuel production, water splitting, and CO₂ conversion.

Operational durability is poorly characterized by traditional (photo)electrocatalyst discovery workflows, creating a barrier to scale-up and deployment. Corrosion is a prominent degradation mechanism whose thermodynamics depend on the concentration of corrosion products in electrolyte. We presented an automated system for characterizing the equilibration of (photo)electrodes with dissolved metals in electrolyte for a given electrode, pH, and electrochemical potential, the accelerated durability screening system (ADSS).[1]  Automation of electrode selection, electrolyte preparation, and electrolyte aliquoting enables rapid identification of self-passivating electrodes and estimation of the equilibrium dissolved metals concentrations. The technique was demonstrated for metal oxide photoanodes in alkaline electrolyte displaying continuous corrosion or self passivation (eg. BiVO₄ and an amorphous Ni−Sb−O material). However, material properties often vary with the sequence of environments in an irreversible manner, producing path-dependency in performance, such as those resulting from different break-in procedures applied to Li-ion batteries or fuel cells. While sequential learning techniques have been effectively deployed for accelerating learning of state properties of materials, they often use a consistent environment path in all experiments. To elevate such techniques for experimental investigations of path-dependent properties, we introduced an iterated expected information gain acquisition function that optimizes over entire experimental trajectories.[2] This approach was implemented within a cloud-based Materials Acceleration Platform architecture utilizing an event-driven stateful broker coupled with remote HELAO (Hierarchical Experimental Laboratory Automation and Orchestration) instances [3,4] and an AI science manager.[2] The platform's efficacy was demonstrated through a case study optimizing multi-step spectro-electrochemical experiments to identify optically stable potential windows in (Co–Ni–Sb)O_z compositions. The system successfully integrated AI-driven experiment design, remote laboratory automation, and cloud-based data infrastructure, validating the platform's capability for managing complex, adaptive, path-dependent workflows in materials discovery. Current effort is focused upon implementing the AI-driven science manager with the automated ADSS to enable a self-driving laboratory for path-dependent and electrolyte engineering approaches to understanding and controlling (photo)electrochemical corrosion.

Since the discovery of photo-assisted water splitting on TiO₂ by Honda and Fujishima in 1972, solar-driven H₂ production has been a major research focus[1]. However, this approach suffers from the thermodynamic barrier of water oxidation and rapid charge recombination, issues typically addressed by introducing sacrificial hole scavengers. In this context, saccharides, the major components of lignocellulosic biomass, represent a dual advantage: beyond acting as hole scavengers, they can undergo photoreforming reactions producing up to 12 moles of H₂ per monosaccharide unit, far exceeding the 3 moles obtained from methane steam reforming. In previous studies, we demonstrated that such carbohydrates can be oxidized under realistic sunlight conditions (AM 1.5G) using Au nanoparticles deposited on a semiconducting support (Au/SC)[2]. Building on this, the present study reports the synthesis of Au/TiO₂ (P25) nanocatalysts and evaluates their performance for solar hydrogen production from glucose under simulated sunlight (AM 1.5G). Wavelength-selective experiments clarify the origin of the photocatalytic excitation and confirm the respective roles of the semiconductor, including the specific contributions of the anatase and rutile polymorphs, and gold nanoparticles in the reaction mechanism. This work further investigates the photoreforming of carbohydrates ranging from mono to polysaccharides, aiming to elucidate the underlying photocatalytic mechanisms and maximize H₂ production yield.

The design and implementation of complete, operational photo-electrochemical devices for solar fuels and materials generation require careful consideration of material requirements and their interplay in an integrated device. While efficiency has become the predominant metric for the assessment of the performance and viability of such devices, degradation has emerged as a limiting factor. Typically, performance-optimized material systems and devices do not necessarily also result in the longevity-optimized version, and vice versa. In this talk, I will discuss how multi-scale and multi-physics degradation modeling can help in guiding in the design, material and operating condition choice for improved longevity while also ensuring high performance. I will discuss lumped parameter models [1], multi-scale models [2], and detailed interface models that are all required for a thorough degradation assessment. I will also discuss how accelerated degradation testing can be developed based on these insights [3]. I will also highlight how degradation plays into techno-economic and life cycle assessments of photo-electrochemical devices [4,5].

Water pollution remains a critical global challenge, demanding efficient and scalable treatment technologies. While photocatalysis is a promising advanced oxidation strategy, conventional powder-based systems are hindered by poor catalyst recovery, limited reusability, and inefficient pollutant-catalyst interactions, restricting their practical application.

Here, we demonstrate how three-dimensional (3D) printing enables the design of photocatalytic materials with enhanced performance. By combining complementary 3D printing techniques, we fabricate structured photocatalytic supports that improve light harvesting, mass transport, and pollutant accessibility to the active surface. Fused deposition modelling (FDM) offers a versatile and scalable route for fabricating robust structures through the direct use of composite filaments, while digital light processing (DLP) provides superior resolution and greater design freedom for resin-based systems. Together, these techniques constitute a powerful and flexible toolbox for the development of advanced photocatalytic systems.

In this context, we first exploit FDM to fabricate silicon carbide (SiC) structures using composite printable filaments. This approach provides SiC architectures with high mechanical robustness and an interconnected porous network, enabling improved photocatalytic performance and long-term reusability. The resulting 3D-SiC exhibits efficient capture and degradation of polystyrene (PS) microplastics under simulated solar irradiation, achieving up to 85% removal within 8 h. Prolonged irradiation induces significant morphological and structural changes in the microplastics, confirming progressive degradation, while the photocatalytic structure retains excellent mechanical stability and performance over repeated cycles.

Building on these results, we extend this approach using DLP, enabling higher-resolution architectures and alternative pathways for the fabrication of functional materials. Multicomponent systems such as Cu₂O/g‑C₃N₄/ZnGa₂O₄:Cr³⁺ are obtained through post-printing modification, where thermal treatment and surface coating are applied to form well-coupled semiconductor interfaces. This strategy enables the integration of multiple functional materials within complex 3D microarchitectures, enhancing light harvesting, charge separation, and interfacial charge transfer for efficient photocatalytic water remediation.

Overall, this work establishes complementary 3D printing technologies such as FDM and DLP as powerful tools for the design of scalable, reusable, and high-performance photocatalytic systems for advanced water remediation.

While major strides have been made in deploying and scaling technologies for renewable electricity generation, this represents only half of what is required to build a truly sustainable economy. The remaining bottleneck lies in establishing renewable feedstocks and developing sustainable pathways to produce petrochemical analogues, which continue to be the critical missing link for decarbonising materials and chemical manufacturing. Lignin, one of the principal constituents of lignocellulosic biomass, is the largest renewable reservoir of aromatic carbon in nature; thus, enabling routes to selectively depolymerize it could unlock access to a broad portfolio of petrochemical‑relevant molecules [1].

At present, thermo‑chemical and thermo‑catalytic processes dominate lignin valorisation. However, their reliance on external oxidants or reductants increases operational costs and complicates scale‑up, while the harsh reaction conditions typically required undermine selectivity and complicate downstream separation. In recent years, alternative routes based on photocatalysis and electrocatalysis have emerged with the promise of depolymerizing lignin under ambient conditions using only light as the driving force. Despite the significant progress in this catalytic space, monomer yields, and the mechanistic understanding of how these transformations proceed, still lag far behind those of thermocatalytic approaches.

In this talk, we will introduce a series of novel photocatalytic strategies to deconstruct lignin into specific, value‑added chemicals. We will present photocatalytic systems based on anthraquinone derivatives [2] and semiconductor quantum dots, together with a comprehensive mechanistic picture supported by complementary computational and experimental evidence that rationalizes their reactivity and product distribution. We will also discuss the prospects for holistic lignin utilization by exploring photocatalytic flow‑reactor platforms and showcasing additional photocatalytic transformations designed to edit and upgrade the resulting bio‑derived products.

Transition-metal oxide and nitride semiconductors offer considerable promise for photoelectrochemical energy conversion. Yet their efficiency and stability are often limited by surface and interface processes, where atomic structure, electronic defects, and chemical reactions are strongly coupled. In such systems, disorder, point defects, and dynamic surface transformations govern charge transport, interfacial charge transfer, and degradation. Here, we investigate Ta₃N₅ thin film photoanodes as a model system to determine how defect chemistry, structural disorder, and surface composition influence photoelectrochemical performance and stability.

Using controlled synthesis from different precursors, we prepared Ta₃N₅ photoelectrodes with varied oxygen contents, structural order, and shallow- and deep-level defect concentrations. Reduced oxygen incorporation improves long-range structural order but increases deep-level defect densities, whereas higher oxygen concentrations introduce shallow donor states and passivate detrimental mid-gap defects. Depth-sensitive characterization reveals oxygen-enriched, structurally disordered surface regions with elevated defect densities relative to the bulk. The extent and nature of these surface layers depend strongly on precursor chemistry: Ta₃N₅ derived from tantalum oxide forms an extended amorphous oxide-rich surface, while Ta₃N₅ from tantalum nitride or metallic tantalum exhibits thinner, more crystalline surfaces with higher mid-gap state densities.

Photoelectrochemical stability measurements show that under water oxidation conditions, Ta₃N₅ photoanodes degrade through formation of a surface oxide layer that suppresses hole injection and enhances recombination. In contrast, under ferrocyanide oxidation conditions, oxygen-rich films exhibit improved long-term stability, whereas oxygen-poor films with high deep-level defect densities degrade rapidly. Specifically, shallow oxygen donors can kinetically stabilize the interface, whereas deep-level defects facilitate rapid photocarrier trapping and surface oxidation[1]. Brief hydrofluoric acid treatment removes the disordered surface layer, improves crystallinity and hydrophilicity, and enhances performance and stability. Overall, this work highlights the importance of defect, surface, and interface engineering for improving charge transport, corrosion resistance, and catalyst integration in durable photoelectrodes for solar fuel generation.

In this presentation, I will describe recent advances from our laboratory on cuprous oxide (Cu₂O) photoelectrodes for solar-driven chemical transformations. Our research encompasses several complementary material architectures—thin-film photocathodes, thin-film photoanodes, and particle-based photocatalysts—all built on Cu₂O as the light-absorbing semiconductor. On the reductive side, our efforts focus on the photoelectrochemical hydrogen evolution reaction (HER) from water, as well as value-added reductions of organic molecules, in which solar-generated electrons drive the synthesis of useful chemical products. On the oxidative side, we investigate the oxygen evolution reaction (OER) from water together with selective oxidations of organic substrates, pointing toward opportunities for solar-powered chemical manufacturing that extend beyond fuel production alone.

I will first outline our strategies for fabricating and modifying Cu₂O thin-film electrodes, including approaches that allow the material to operate as either a photocathode or a photoanode despite its intrinsic electronic properties. I will then highlight representative catalytic reactions that showcase the versatility of this semiconductor platform. Finally, I will present our recent progress in translating thin-film photoelectrode concepts into particle-based systems, with the long-term aim of realizing scalable photocatalytic architectures for solar-driven chemical synthesis.

The mimicry of photosynthesis to produce sustainable fuels and chemicals has long inspired scientists, but fully functional and scalable systems that fully replicate natural photosynthesis remain rare, and viable routes to commercialisation are uncertain. Recent advances in the assembly of photosynthesis-inspired architectures have enabled the construction of prototype solar devices for direct CO₂ fixation. Artificial leaves combine semiconductor light absorbers with immobilised (bio)molecular catalysts to drive solar-powered CO₂ reduction, producing fuels, alongside oxygen evolution from water oxidation. These products can be further upgraded via integrated catalytic processes, for example converting formate into enantioenriched organics through enzymatic cascades or into sugars using engineered microorganisms. The replacement of water oxidation by the valorisation of waste substrates provides a possible path towards commercialisation. This “solar reforming” approach offers favourable thermodynamics and kinetics while improving economic viability by coupling fuel production with waste upcycling. Notably, outdoor solar plastic reforming is currently being tested at the kilogram/square meter scale. This presentation will outline the emerging paradigm of integrated solar chemistry with a focus on solar reforming. It will also highlight strategies and frontiers such as atmospheric CO₂ utilisation, advanced light management in integrated devices, and solar-driven cascade catalysis for high-value chemical synthesis. 

More information: http://www-reisner.ch.cam.ac.uk/

Replacing water oxidation with alternative anodic oxidation reactions in photoelectrochemical (PEC) systems offers a powerful route to enhance the value of solar-driven fuel production.^[1] By coupling hydrogen evolution with selective oxidation of organic substrates, PEC devices can convert low-value feedstocks into valuable chemicals. However, practical implementation is governed by the intrinsic complexity of multi-electron, multi-pathway oxidation chemistry, where controlling reaction selectivity remains a central challenge. In this talk, we will present our recent work on understanding and steering selectivity in PEC oxidation reactions using oxide-based systems. Using glycerol oxidation as a model reaction, we show that electrolyte composition plays a critical role in governing reaction kinetics and product distribution, where ion-specific interactions and interfacial buffering strongly influence glycolaldehyde formation.^[2] We further demonstrate that reaction selectivity can be dynamically tuned through light intensity, which modulates surface hole accumulation and shifts reaction pathways from partial oxidation products toward deeper oxidation to formic acid via identifiable intermediates.^[3] In parallel, interfacial engineering strategies, including heterojunction and doped overlayer design, enable suppression of non-selective radical pathways and improved control of C–C bond scission. Finally, extending these concepts beyond biomass-derived feedstocks, we demonstrate a coupled photoelectrochemical–thermal approach for solar-driven upcycling of PET waste, where oxygen generated at the photoanode is utilized to selectively convert ethylene glycol into glycolic acid while simultaneously producing hydrogen. Together, these studies establish selectivity control as a key design principle for enabling PEC oxidation chemistry beyond water splitting.

References

1. K. Zhu; X. Zhang; L. Wen; S. Zhou; D. S. Achilleos; R. Amal; Y. H. Ng; F. F. Abdi, Nat. Rev. Clean Tech. 2025, 1, 621-637
2. H. Kong; S. Gupta; A. F. Pérez-Torres; C. Höhn; P. Bogdanoff; M. T. Mayer; R. van de Krol; M. Favaro; F. F. Abdi, Chem. Sci., 2024, 15, 10425-10435.
3. L. Wen; K. Zhu; X. Zhang; H. Y. Chung; S. Qu; H. Wu; F. F. Abdi, Cell Rep. Phys. Sci., accepted.

Solar-driven CO₂ reduction is a key route toward sustainable chemical and fuel production, yet persistent challenges in catalyst durability, interfacial stability, and reactor engineering continue to hinder practical deployment. Within the EU H2020 SunCOChem project, we pursue an integrated materials-to-device-to-system approach, targeting CO-rich synthesis gas streams via (photo)electrochemical CO₂ reduction, coupled to a downstream hydroformylation unit for value-added chemical synthesis at scale.

Novel core-shell Cu₂O/SnO₂-based gas diffusion electrodes (GDEs) were designed, fabricated, and validated in a continuous-flow photo-anode/dark-cathode photoelectrochemical (PEC) cell operated under simulated photovoltaic-coupled conditions. Spray-coated catalytic layers were optimized for spatial uniformity, triple-phase boundary density, and mechanical robustness across electrode areas from 10 to 120 cm², with a total validated area of 0.24 m². The optimized electrodes achieved 84–90% Faradaic efficiency for CO production at current densities of −20 to −100 mA cm⁻².¹ The 120 cm² PEC cell was operated for over 200 hours under intermittent current profiles mimicking renewable energy supply, delivering CO-rich streams (CO:H₂ = 9:1) and syngas at total currents up to 1.5 A — directly relevant to industrial hydroformylation and carbonylation feedstock specifications.

An operando diagnostic framework was exploited by combining electrochemical impedance spectroscopy (EIS) with accelerated stress testing protocols. Moving beyond conventional equivalent-circuit EIS modeling, we employ distribution of relaxation times (DRT) analysis, giving insights into overlapping physicochemical processes such as charge transfer, ionic transport, flooding, and interfacial passivation, without imposing a priori circuit assumptions. This enables early, quantitative identification of degradation pathways and their kinetic evolution, directly linking impedance signatures to productivity-relevant parameters such as CO partial current density and selectivity retention in the 120 cm² prototype cell, thereby demonstrating a novel approach for real-time, in situ monitoring of failure modes.

Post-operation and operando physicochemical characterization (SEM, TEM, XRD, ICP-MS)² systematically informed degradation mechanisms identified under operating conditions, revealing catalyst phase restructuring and photocatalyst delamination/metal leaching as dominant failure modes, and guiding rational regeneration strategies for extended device lifetime.

Complementing the experimental framework, multiphysics modeling was deployed as a digital twin of the reactor system. Model-guided flow-field optimization addressed CO₂ starvation, inhomogeneous current-density distributions, and bubble-induced mass-transfer limitations at the cathode interface.

This work demonstrates that operando DRT-EIS diagnostics, post-test materials analysis, and multiphysics digital twins can be tightly integrated at the device level, yielding actionable design principles for next-generation PEC and hybrid solar fuels reactors. The results have direct implications for understanding interface stability, guiding CO-productivity recovery strategies, benchmarking solar fuels devices, and advancing the long-term techno-economic viability of solar-to-chemical conversion systems.

Scaling up photoelectrochemical (PEC) devices to commercially viable industrial systems requires a comprehensive understanding of coupled transport phenomena across multiple length scales, which can be enabled by numerical modelling to guide system design and optimization. This is especially important for systems using unconventional substrates like recently developed transparent porous conductive substrates (TPCS) [1]. These novel substrates offer significant increases in electrochemically active surface area but also pose challenges with light scattering and recombination losses. This work presents a holistic, bottom-up multi-scale and multi-physics modelling framework for a PEC reactor featuring porous photoelectrodes made by depositing photocatalysts on TPCS fibers. The system is evaluated for hydrogen evolution combined with glycerol oxidation (an industrial waste byproduct), promising a waste-valorisation pathway.

The presented modelling framework uniquely bridges microscale optics with macro-scale reactor engineering. First, the interaction of individual multilayered microfibers with the incident radiation was resolved and subsequently used to compute the radiation characteristics of a thick homogeneous fibrous medium. A Monte Carlo ray tracing method was developed to simulate light propagation through the medium, accounting for fiber orientations and multiple scattering. The local photon absorption profile along the photoelectrode thickness was computed precisely and coupled with charge transport and heat transfer physics to determine local photocurrent and heat generation rates.

To capture device-level behaviour, localized data was integrated into a reactor-level homogenized continuum model. This macro-scale model simultaneously simulates light transport, fluid flow, heat transfer, charge transport, and species transport. After verifying predictions against experimental measurements on equivalent systems, the model was used to quantify performance metrics such as photocurrent density, temperature, and local overpotentials under varying solar concentrations and flow rates to identify limiting regimes and optimization pathways. Additionally, model predictions with and without glycerol were compared.

Results demonstrated that performance was limited by significant ohmic losses at large current densities and non-uniform light absorption inside the porous sample. As solar concentration C_sun increased, the performance gap between the ideal uniform case and real cases widened further due to escalating optical and ohmic losses. Convective heat transfer from fluid flow maintained reactor temperatures below 45°C, even under high irradiation (C_sun = 50). Concentration gradients confirmed successful glycerol oxidation and hydrogen evolution, while the impact of different material and operating parameters on the J-V curves was quantified.

This model offers a powerful framework for the rational design of next-generation PEC reactors incorporating porous photoelectrodes, accelerating their transition to real-world solar-fuel deployment.

Photocatalytic water-splitting is widely regarded as a promising route to solar hydrogen and decarbonization, yet most investigations remain at laboratory scale, and the scarcity of large-scale demonstrations limits our understanding of how photocatalyst performance translates to larger systems. Addressing this requires both scalable material synthesis and validation under realistic operating conditions.

In this work, we present a scaled-up hybrid solar/artificial light-driven photoreactor, from laboratory screening to a continuously operated pilot plant. The reaction mixture consists of a COF-gC₃N₄ photocatalyst as the primary active material, synthesized at the tens-of-grams scale [1], polyvinylpyrrolidone (PVP)-encapsulated platinum nanoparticles as co-catalyst, and ascorbic acid as sacrificial electron donor. Laboratory experiments first established optimal catalyst and co-catalyst loadings, sacrificial-agent concentration, temperature, and light intensity; elevated temperatures (up to 70 °C) markedly enhanced activity, motivating a reactor design that can recover UV-A LED waste heat to potentially increase hydrogen output.

To enable continuous generation, we developed a helical photoreactor (4 L total capacity, 1.5 L illuminated volume), with dimensions determined using the optical depth of the reaction mixture as the guiding metric. Compact linear Fresnel reflectors (CLFPs) concentrate sunlight by ~5×, while UV-A LED strips inside the reactor helix provide supplemental irradiation, allowing operation under natural sunlight, artificial illumination, or both in tandem. The spectral overlap between the UV-A emission and the catalyst absorption sustains activity when sunlight is intermittent or absent.

Pilot trials achieved a peak hydrogen evolution rate of 14.8 mmol/hr/g.cat under ~2.5 suns, observed on days 2–3 of a four-day continuous run on a single reaction batch, compared to 22.3 mmol/hr/g.cat under ~1.5 suns at laboratory scale. The solar-to-hydrogen (STH) efficiency reached 0.282%, exceeding the 0.087% reported in [2] for a slurry-type pilot reactor. These results demonstrate that the pilot-scale reactor retains the high activity of the COF-gC3N4 photocatalyst, with room for further gains through optimization of slurry composition and solar tracking.

Waste carbon and nitrogen vectors can serve as fundamental building blocks for the synthesis of important chemicals for the energy, manufacturing, and agricultural industries. Their charged reactive character (e.g., in bi/carbonates and nitrates) offers a wide cross-reactivity landscape and access to high-value, complex carbon-nitrogen chemicals through co-electrolysis. Realizing this potential is challenged by overlapping thermodynamics and kinetics involving long sequences of multielectron/proton transfers. I will present our recent mechanistic insights on carbonate and nitrate supported reactions, including coelectrolysis, based on operando probes. I will further show how active control of the electrochemical interface and environment in these reactions enables advances in selectivity and other performance metrics. To conclude, I will offer a brief broader overview of different Solar Fuel technologies.                                                               

Photoelectrochemical (PEC) cells offer a direct route to solar-driven fuels and chemicals, yet practical devices still operate far below their thermodynamic limits owing to coupled losses spanning light absorption, carrier recombination and transport, interfacial charge transfer, and semiconductor–electrolyte matching. These losses are intrinsically entangled, and a general framework for quantifying and comparing these losses across real devices remains lacking, leaving PEC optimization largely empirical.

Here we introduce, to our knowledge, the first unified analytical model that treats both built-in junction (BIJ) and semiconductor–electrolyte junction (SEJ) photoelectrodes within a single, physically meaningful set of parameters while retaining the distinct interfacial physics of each architecture. By fitting experimental current–voltage data, the model decomposes device performance into thermodynamic, optical, recombination/transport, interfacial charge-transfer, and parasitic loss channels, and maps each directly onto concrete optimization strategies such as surface passivation, co-catalyst integration, contact optimization, and nanostructuring. Energy flows are visualized through Sankey diagrams, giving an intuitive picture of how incident solar energy is absorbed, dissipated, or converted into chemical output.

We validate the model against state-of-the-art devices spanning solar water splitting, CO₂ reduction, NH₃ synthesis, and solar redox flow batteries, achieving high-quality fits and physically consistent loss attributions. The analysis reveals a systematic divergence between material classes: photovoltaic-grade absorbers such as Si and perovskite in BIJ devices are dominated by bulk recombination and transport losses, whereas non-photovoltaic materials such as BiVO₄ and Ta₃N₅ in SEJ devices are limited primarily by interfacial reaction kinetics. We further propose a semiconductor–electrolyte matching criterion beyond simple energy-level alignment and construct efficiency maps that define material-selection windows, identifying a moderate reaction-potential window of approximately 0.7–1.1 V as a universal operating regime for high-efficiency PEC systems.

Together, these capabilities help researchers see where energy is lost and turn that insight directly into better device designs, supporting a shift from empirical optimization toward mechanism-informed rational design of viable solar fuels technologies. The model is accompanied by an open, ready-to-use toolkit for efficiency-map generation, J–V fitting with loss decomposition, and interface-matching analysis.

Heterogeneous photocatalytic CO2 reduction has attracted significant attention as a promising route for solar fuel and chemical production [1, 2]. Among the possible reaction pathways, gas-phase CO2 hydrogenation remains comparatively underexplored despite its potential to produce methanol (CO2 + 3 H2 -> CH3OH + H2O) and other valuable chemicals under mild conditions [3]. By avoiding the energetically-demanding water oxidation half-reaction, CO2 hydrogenation represents a unique opportunity for single-bandgap photocatalysis by relaxing the requirement of a large photovoltage which then significantly limits the fraction of solar spectrum absorbed [4]. CO2 photo-hydrogenation therefore offers an alternative carbon utilization technology to energy-intensive thermocatalysis that can be coupled with renewable hydrogen generation and carbon capture to realize a more sustainable future.

Despite its potential, significant challenges continue to hinder progress. Photocatalytic activities remain low due to challenges in simultaneously achieving suitable optoelectronic properties, catalytic performance and material stability [1, 3]. Equally important, but often overlooked, are the experimental challenges associated with photocatalytic CO2 reduction [1]. Low product formation rates necessitate highly sensitive analytical methods and rigorous experimental protocols, while variations in reactor configurations, light sources, and the lack of standardised testing procedures hinder reproducibility. Together, these factors increase the risk of experimental artefacts complicating the reliable evaluation and comparison of photocatalyst performance between different laboratories.

This presentation identifies the common pitfalls encountered during photocatalytic CO2 hydrogenation and examines experimental factors that influence the measured performance. Firstly, the effect of reactor configuration, light source used, operating conditions and sample deposition methodology on the final photocatalytic performance is quantified. The challenge of standardization demonstrated and the distinction between surface area normalized and apparent photocatalytic activity is highlighted [5]. We next examine the erroneous sources of products species with particular attention on contamination, photocatalyst decomposition, and degradation of reactor/system components (such as polymer seals and recirculating pump materials). Based on empirical evidence, guidelines for good photocatalysis experimental design are outlined and the our photocatalytic experimental setup is presented. Furthermore, to address the issue of contamination-free gas recirculation, we present a novel custom-made hermetically-sealed piston pump designed to meet the challenging requirements of high-pressure contamination-free gas recirculation in batch experiments. The open-source design provides a low-cost and reproducible platform for the community. Finally, we consider the role of control experiments in establishing product origin demonstrating how the source of artefacts can be identified through systematic control experiments.

By identifying common sources of experimental error and outlining practical recommendations for reactor design, analytical protocols, and validation strategies, this work aims to support more robust, reproducible, and comparable photocatalytic CO2 hydrogenation research.

Using solar energy to split water and produce hydrogen has the potential to become a key technology to sustainably meet our rising energy demands [1]. Techno-economic studies revealed that economically viable solar water splitting should produce hydrogen at a cost of less than 2 USD per kg. To achieve this economic target, particle-based photocatalytic systems need to reach a solar-to-hydrogen (STH) efficiency of 10 % and a lifetime of at least 5 years [2].

Among the material classes currently being investigated for application in solar water splitting, oxynitrides are considered promising candidates [3]. They are typically synthesized by nitridation of oxidic materials introducing the N 2p orbitals that form the valance band edge, resulting in smaller band gaps and therefore in the absorption of a larger part of the visible light in the solar spectrum. A promising example is LaTiO₂N, which has a band gap of about 2.1 eV, and is composed of earth abundant elements [4]. It has already achieved competitive oxygen evolution rates and even overall water splitting [5, 6]. The deposition of cocatalysts such as CoO_x to enhance the performance is thereby essential [7].

Although improving the lifetime of oxynitride photocatalysts is as important as their efficiency, significantly less research efforts have been dedicated to stability investigations [8]. Fully understanding the degradation mechanisms that deactivate the involved photo- and cocatalyst materials is therefore essential. In general, water splitting systems deactivate over time either via charge related degradation induced by photogenerated electrons and holes, via chemical degradation by the electrolyte environment or via mechanical degradation, such as the detachment of cocatalyst particles. Investigating these processes is important to develop photocatalytic systems that are competitive with other technologies.

In this work we investigate the degradation mechanisms of LaTiO₂N particles, both with and without deposited CoO_x cocatalysts. The materials are therefore subjected to well-defined conditions, involving exposure to an electrolyte and light irradiation. The particles are characterized before and after the procedure, to reveal changes caused by degradation. Structural and morphological investigations are carried out by X-ray diffraction (XRD), as well as scanning and (scanning) transmission electron microscopy (SEM/(S)TEM) coupled to electron energy loss spectroscopy (EELS) to investigate the chemical composition. Optical properties are assessed by UV-vis spectroscopy and the photocatalytic activity is investigated by measuring the oxygen evolution rate via gas chromatography.

The development of artificial photosynthetic reactions that harness solar energy to drive the synthesis of organic compounds using CO₂ and H₂O as feedstocks holds significant potential for mitigating CO₂ emissions and enabling the environmentally sustainable conversion of abundant solar energy into chemical energy carriers. Our primary strategy for achieving artificial photosynthesis involves the integration of combinatorial technologies that leverage the advantageous properties of solid semiconductor photocatalysts and molecular metal complex catalysts. ^{1 2}

We have achieved a low-overpotential CO₂ reduction reaction (CO₂RR) coupled with the H₂O oxidation reaction (WOR), a key reaction pair for artificial photosynthesis, within a single aqueous solution at near-neutral pH. The simultaneous operation of CO₂RR and WOR in a unified solution represents a crucial concept for the development of a simplified artificial photosynthetic system, operating via a two-step photoexcitation (Z-scheme) mechanism in a self-organized manner. The system utilizes an aqueous suspension of particulate (CuGa)_0.3Zn_1.4S₂(CGZS), BiVO₄, and a water-soluble cobalt complex ([Co(4,4’-dimethyl-2,2’-bipyridine)₃]²⁺, [Co-dmbpy]). Upon irradiation with visible light (λ > 420 nm), the photocatalytic reaction facilitates O₂ evolution while achieving a CO production selectivity of 62-98% by effectively suppressing competitive H₂ generation. This reaction occurs in an aqueous NaHCO₃ solution bubbled with gaseous CO₂, demonstrating efficient and selective photocatalytic CO₂ reduction.^{3, 4} Experimental investigations combined with density functional theory (DFT) calculations indicate that the cobalt complex exhibits dual functionality in synergy with CGZS and BiVO₄. Specifically, it serves as an efficient ionic electron mediator while also functioning as a highly selective cocatalyst for CO₂RR on CGZS. The system continuously generates CO and H₂ accompanied by O₂ evolution, with an electron/hole ratio close to the stoichiometric value of unity.

In the meeting, I will explain a compact photovoltaic-cell (PV)-powered electrolyzer system using catalysts composed of earth-abundant elements, specifically a Mn(I) complex polymer and β-FeOOH, together with a Si photoabsorber, as well as a large-scale PV-powered electrolyzer system employing 1 m² electrodes for the highly selective conversion of CO2 to CO or formate ions. ^5-8 I will also discuss the long-term stability, high energy efficiency, and potential applicability of molecular metal complex catalysts for CO2 electrolysis and the synthesis of higher-carbon chemicals. ^9,10

 Compared to thermal or electrochemistry, photocatalytic processes are much less well studied. In my talk I will compare the three types of systems and present an adequate formalism for the better screening of materials for photocatalytic applications.

Photoelectrochemical (PEC) water splitting is a direct route to solar H₂, but scaling to cm² areas is limited by ohmic drops, current collection and photoelectrode durability. We report a 10 cm² zero-gap tandem PEC cell built only from earth-abundant oxides (α-Fe₂O₃ photoanode/CuO photocathode) and an anion-exchange membrane. The work quantifies module-scale losses via EIS and distribution-of-relaxation-times analysis and assesses 240 h stability and gas purity.

Nanostructured α-Fe₂O₃ was electrodeposited on FTO and annealed; Cu was electrodeposited on carbon paper, chemically oxidised to Cu(OH)₂ and converted to CuO. A PiperION AEM (OH⁻ form) and matching ionomer layers formed a glass electrode membrane assembly in a PMMA/steel zero-gap prototype (10 cm²).  J–V curves, EIS+DRT, 240 h potentiostatic test (−1.3 V) and Micro-GC were performed under AM 1.5G.

Under AM 1.5G the tandem delivers a photoresponse already near 0 V and reaches −25 mA cm⁻² at −1.3 V, giving a maximum hydrogen-production efficiency of 5.1%. EIS shows an area-specific ohmic resistance of ~18 Ω cm², while illumination lowers the interfacial polarization (R_ct ~32 Ω cm² at −1.3 V). DRT resolves a dominant millisecond relaxation assigned to charge transfer/double-layer dynamics, with slower contributions emerging at 0 V, consistent with transport/ionic redistribution in the membrane–electrode assembly. A 240-h durability test at −1.3 V maintains nearly constant current; Micro-GC confirms ~99.95% H₂ with O₂ at 0.04–0.06% and no detectable CO/CO₂. Post-operando XRD/SEM show stable hematite, whereas CuO partially reduces and delaminates, identifying the photocathode and series resistance as the main scaling bottlenecks.

A bias-assisted 10 cm² α-Fe2O3/CuO tandem PEC cell in zero-gap AEM architecture achieves −25 mA cm−2 and 5.1% H2 efficiency at −1.3 V with 99.95% H2 purity over 240 h. Performance is governed by ~18 Ω cm² ohmic loss and CuO instability: future work will focus on reducing ohmic losses and on implementing protected photocathodes [1].