The design and implementation of complete, operational photo-electrochemical devices requires a careful consideration of material requirements and their interplay in an integrated device. Multi-physical, multi-dimensional and multi-scale modeling tools are essential for the investigation of feasibility, optimization and implementation of such devices. Modeling activities however always have to be fed and, ultimately, validated by experimental data of material and device characteristics. In this talk, I will discuss the development of predictive modeling tools for photo-electrochemical water and CO2 reduction. These models account for the atomistic-scale via density functional theory, molecular-scale by micro-kinetic and molecular dynamics models, nm-scale by double layer transport models, meso-scale by tomography-based direct numerical simulations, and device scale by volume-averaged device models. I will review each of these modeling steps and highlight the importance of interfacing them. I will then show what experimental data is needed to run these simulations and how these models can be validated. I will end with some device-level demonstrations of photo-electrochemical devices that follow the design guidelines developed by the multi-scale models.

Unlimited access to clean, sustainable and renewable source of energy seems to be a dream scenario, however with an assistance of machine learning (ML) approach the experimental work which actually reveals the right descriptors may gain a powerful tool allowing to omit a time consuming trials path. As a matter of fact, the ML has already shown the potential to significantly reduce the time and effort required for developement of new and efficient photocatalysts. 

However, precise definition of descriptors will require an identification of the bottleneck materials parameters, features and processes, thus different architectures and working arrangements will be shown and disscused in this presentation in order to test their usfulness for prediction of novel and better arrangements for hydrogen and other carbon based solar fuels production. 

Over 400 samples based on the semiconductor oxides have been prepared and measured in order to bulid a pre-base for ML tests. Different materials and working systems, their compositions, geometry, properties will be disscused in view of their impact on the final prediction. In this presentation we  will discuss the strengths and weaknesses of different architectures and working arrangements for hydrogen and carbon-based solar fuels production.

Nowadays, the existence of high entropy perovskite oxides is well established and, since their discovery, their fields of application have been continuously studied. The main advantage provided by these types of materials with general formula ABO₃ is the high possibility of tuning their properties by acting on the chemical compositions. The study of stability, fields of existence, and solubility limits in a systematic way remains lacking in comparison with continuous application research, despite the presence of large datasets and computational studies.¹ In our work, we investigated the structure and solubility limits using a chemometric approach for two families of perovskite oxides with the following compositions: on the site A we chose to keep lanthanum due to its stability, while, on the B site, we explored different cation mixtures based on Cr, Mn, Fe, Co, Ni and Zn, to determine an experimental domain by including various experimental data, i.e. crystal structure, oxygen vacancies amount and temperature-dependance, and composition.² The synthesized samples were subjected to x-ray diffraction (XRD) analysis followed by Rietveld refinement to determine crystallographic parameters. Through multivariate analysis, elemental concentrations were correlated to phase stability and nature, and cell parameters. The structural study aims at highlight possible correlations between the crystal symmetry and the catalytic characteristics of the material. Finally, through thermogravimetric analysis, it is possible to study non-stoichiometry and phase transitions to be integrated within the previously designated experimental domains.³ The overall dataset will allow to define the best compositions to be used for the desired applications such as heterogeneous catalysis, solid-oxide fuel cells, and oxygen transport membranes.⁴

Luminescent solar concentrators (LSCs) are a promising technology to help integrate photovoltaic (PV) cells as functional architecture in the urban environment.[1] LSCs typically consist of a waveguide slab, which is doped or coated with a luminescent species, or luminophore.[2] Incident light is absorbed by the luminophores and re-emitted at longer wavelengths through photoluminescence. A large portion of the emitted light is concentrated by the waveguide via total internal reflection to the edges of the LSC, where a strip of PV cells may be retrofitted attached to collect the spectrally converted light. While LSCs have traditionally been cuboidal in geometry, in recent years, new architectures have been explored for example, circular, curved, polygonal, wedged, and leaf-shaped designs.[3,4] These new
designs can help reduce optical losses, enable integration within existing infrastructure, or unlock new applications.

However, fabrication of complex LSC geometries can be both time- and resource-intensive. LSCs are commonly made using conventional manufacturing methods such as casting, which either limits possible architectures to the availability of suitable moulds or leads to material waste as designs are cut to shape. Moreover, it is not possible to introduce secondary internal structure within LSCs by casting. To tackle these limitations, in this talk the use of 3D printing (specifically fused deposition modelling) as a tool to fabricate complex prototype LSC architectures for rapid optical evaluation will be described. It will be shown that the print pattern (e.g. line thickness, build-up of layers) introduces internal structure within the LSCs, affecting the optical pathways for light absorption and emission. Moreover, we will demonstrate how we have implemented upgrades to the Monte Carlo ray-trace software pvtrace, to enable advance prediction of the optical efficiency of 3D-printed LSCs.[5] The more versatile computational workflow afforded by our upgrades, coupled with 3D-printed prototypes, will enable rapid screening of more intricate LSC architectures, while reducing experimental waste.

The search for new photoactive materials able to efficiently produce solar fuels is a matter of growing interest due to the current global energetic crisis. In response to this situation, the generation of solar fuels has appeared as a sustainable alternative. In last years, extensive efforts have been made to develop efficient catalytic systems capable of harvesting light absorption and reducing CO₂ especially when using water as the electron donor.

Herein, we report different strategies and modifications of photocatalysts to increase process performance. Among them, an interesting approach to improve charge separation in photocatalytic systems is the use of heterojunctions. In this line, the combination of different semiconductors with noble metal nanoparticles or organic semiconducting polymers leads to a separation of the photogenerated charge carriers to increasing their life time, facilitating charge transfer to adsorbed molecules.

Organo-inorganic hybrid materials show a dramatic reactivity improvement in CO₂ photoreduction, enhancing methane selectivity. Reaction pathways are not well defined for this reaction and several uncertains are still unsolved. To explain this behavior a combination of in-situ NAP-XPS, FTIR, TAS spectroscopies and theoretical tools has been used, showing a more efficient light absorption and charge transfer in the hybrid photocatalyst compared with bare materials.

In photocatalytic systems for solar energy conversion, a key challenge for efficient operation is the efficient separation and stabilisation of photogenerated charges, and thereby the minimisation of undesired recombination losses. This challenge is particularly great for photocatalytic systems because of the long charge lifetimes required to drive catalysis A further consideration is to minimise the energy losses required to drive this charge separation. My talk will cover examples of recent work from my group addressing aspects of this challenge. I will start by considering the role of polaron localisation / relaxation in driving charge separation in metal oxides, and quantification of the energetic loss associated with this polaron formation. Kinetic competion between polaron formation and ligand field state mediated charge recombination will be highlighted as the key challenge limiting the performance of many visible light absorbing metal oxides. The roles of metal oxide dielectric constant, surface facet energies, traps, ‘photocharging’ and heterojunctions in aiding charge separation in metal oxides will be discussed, drawing on examples from BiVO₄ and SrTiO₃. I will then go on to discuss charge separation and stabilisation in alternative photocatalyst materials, including organic semiconductor nanoparticles and metal-organic frameworks, highlighting the observation of remarkably long lived charge photogeneration in both systems and their relevance to the efficiency photocatalytic performance.

Hydrogen and other solar fuels have been highlighted as one of the future energy vectors. Having natural photosynthesis as inspiration, we can develop a device capable to split water using sunlight, obtaining oxygen and hydrogen. [1], [2] Different strategies can be used to achieve this: from separated light harvesting and catalytic systems to all integrated devices able to transform directly sunlight into fuels. These systems can be built with a variety of (photo)electrodes such as organic based material, chalcogenides or metal oxides. In most of them the addition of a co-catalyst layer it is pivotal to improve their efficiency. Although rapid progress is being made in the field, understanding of the limiting factors of these catalysts has allowed remarkable improvements in their performance.

In this talk I will focus on the use of organometallic method to prepare nanoparticles as catalysts for solar fuels production. This is a versatile method able to grow NPs in bulk or in the presence of a support.[3], [4] In here I will present the use of this method to prepare different metal NPs on different organic based light absorbers and their activity as photocatalysts for hydrogen production. Interestingly in this work the photocatalysts containing Ni NP present a similar performance than the Pt ones. In order to correlate the performance with the nature of each photocatalyst, we have characterised of the materials before and after catalysis. The latter is a key information to unravel the main deactivation pathways, and consequently, systematically designing new and more efficient photocatalytic systems.

Bismuth vanadate (BiVO₄) is a nontoxic and medium band gap (2.4 eV) n-type semiconductor that has attracted worldwide attention as a visible-light active photocatalyst. Particularly, scheelite monoclinic BiVO₄ is one of the most active polymorphs in photocatalytic oxygen evolution. However, the oxidation capability of bare BiVO₄ is hindered by weak charge carrier mobilities, short charge carrier diffusion lengths, and a high charge recombination rate at the surface. To circumvent these inconveniences, heterostructuring by contacting BiVO₄ with other materials such as noble metals and metal oxides has been proposed. 

Usually, contact materials are deposited in a random manner on particulate BiVO₄-based photocatalysts. In this case, no distinct flow path for photogenerated holes and electrons can exist which makes bulk and surface recombination still prominent due to insufficient charge carrier separation. Since the pioneer works on anatase and rutile TiO₂ particles which have revealed the key role of exposed facets in the separation of photoinduced electron–hole pairs, photodeposition is now considered as a sort of general route for synthesizing heterostructures. Indeed, photogenerated electrons and holes preferentially accumulate on different crystalline facets which renders cocatalyst deposition regioselective depending on whether the process involves an oxidation or reduction reaction. Systematic studies on single crystalline TiO₂ (anatase) exposing different facets have shown that the different surfaces (100) vs (110) provide indeed different electronic surface properties including different work functions that promote charge carrier separation into different directions.

Anisotropic photodeposition on BiVO₄ microcrystals has been extensively studied. It has been shown that metallic deposits were preferentially found on {010} facets after metal ion photoreduction, while metal oxide deposits were found on the {110} facets after metal ion photo-oxidation, evidencing that electrons and holes accumulate on different sites on anisotropic BiVO₄ crystals. Besides providing proof of enhanced charge carrier separation, the preferential photodeposition was used to construct BiVO₄ photocatalysts with even better photocatalytic oxygen evolution efficiencies by depositing reduction cocatalysts on the more reducing {010} facets and oxidation cocatalysts on the more oxidative {110} facets.

On the other hand, BiVO₄ microcrystals generally reported in the literature present compromised colloidal stability and low surface activity due to size effects. Thus, dedicated experiments need to be tackled to control and decrease the size.

In this context, we herein report for the first time a universal photodeposition of noble metal [Au, Ag, Pd] and/or metal oxide [CoO_x, MnO_x, FeO_x] cocatalysts, via the original use of laser light, on BiVO₄ nanocrystals exposing well-defined {010} and {110} facets, being the main advantage of laser photodeposition the shorter procedure times due to high intensity effects. The monoclinic BiVO₄ nanocrystals here exhibited are one of the smaller reported ever, showing more than an 8-fold edge length and 4-fold thickness reduction compared to the common microcrystals generally reported in the literature. Both the chemistry and regioselectivity of the deposition were investigated by employing X-ray photoelectron spectroscopy and various electron microscopy techniques. Then, the photocatalytic efficiencies of the structures prepared were tested for water oxidation reactions and were compared with previously studied systems.

Metal oxide photoelectrodes tend to be cheap, easy to fabricate, and show relatively good (photo)chemical stability in aqueous solutions. This makes them attractive candidates as light absorbers in a variety of photoelectrochemical and photocatalytic applications. However, the efficiencies of these absorbers are poor compared to photovoltaic-grade materials. This is usually attributed to polaron formation and recombination or trapping at defects. To determine the quality of metal oxide absorbers, time-resolved photoconductance measurements are often used as a convenient and contact-free method to determine the carrier diffusion length. Here, one of the main challenges is to determine whether the decay in photoconductivity is due to a decay in carrier concentration (recombination), a decay in carrier mobility (trapping, polaron formation), or both. I will present a general analysis method for determining the diffusion length which is valid for time-dependent mobilities as well as time-dependent lifetimes [1]. We have applied this method to a range of metal oxides and validated the method by applying it also to crystalline silicon and a halide perovskite. We find a carrier diffusion length of only 15 nm for BiVO₄, which is significantly shorter than previously reported values by us and others. I will discuss how this value can be reconciled with the relatively high photocurrents that many groups have reported for this material. For several other oxides that we studied, we find evidence for nm-scale carrier localization [2], which is likely due to nano-sized crystallites or defect clusters. Mitigation of this strong localization may be possible and would offer a promising strategy for designing more efficient metal oxide-based photoelectrodes.

Electro and photochemical CO₂ reduction (CO₂R) is the quintessence of modern-day sustainable research. We report our studies on the electro and photoinduced interfacial charge transfer occurring in nanocrystalline mesoporous TiO₂ film and two TiO₂/Iron porphyrin hybrid films (meso-aryl and b-pyrrole substituted porphyrins, respectively) under CO₂R conditions. We used Transient Absorption Spectroscopy (TAS) to demonstrate that under 355 nm laser excitation and an applied voltage bias (0 to -0.8 V vs. Ag/AgCl), the TiO₂ film exhibited a diminution in the transient absorption (at -0.5 V by 35%), as well as a reduction of the lifetime of the photogenerated electrons (at -0.5 V by 50%) when the experiments were conducted under CO₂ atmosphere changing from inert N₂. The TiO₂/Iron porphyrin films showed faster charge recombination kinetics, featuring 100-fold more rapid transient signal decay than the TiO₂ film. The electro, photo, and photoelectrochemical CO₂R performance of the TiO₂ and TiO₂/Iron porphyrin films are evaluated within the bias range of -0.5 to -1.8 V vs. Ag/AgCl. The bare TiO₂ film produced CO and CH₄, as well as H_2, depending on the applied voltage bias.
In contrast, the TiO₂/Iron porphyrins films showed the exclusive formation of CO (100% selectivity) under identical conditions. During the CO₂R, gain in the overpotential values is obtained under the light irradiation conditions. This finding indicated a direct transfer of the photogenerated electrons from the film to absorbed CO₂ molecules and a decreased decay of the TAS signals. In the TiO₂/Iron porphyrin films, we identified the interfacial charge recombination processes between the oxidized iron-porphyrin and the electrons of the TiO₂ conduction band. These competitive processes are considered responsible for the diminution of direct charge transfer between the film and the adsorbed CO₂ molecules, explaining the moderate performances of the hybrid films for the CO₂R.

The increasing levels of carbon dioxide (CO₂) in the atmosphere and the growing global need for energy necessitate the development of renewable methods for electricity production that can help reduce carbon emissions and of green energy-based CO₂ utilization technologies. In this context, sustainable and clean energy sources have attracted a lot of attention. In particular, solar energy-based techniques can be used to convert CO₂ into valuable chemicals and fuels, such as CO, formate or methanol [1]. Currently, one established way to utilize CO₂ is to use the coenzyme-dependent activity of formic acid dehydrogenase (FDH) to catalyze the reduction of CO₂ to formate [2]. Nevertheless, the high cost of the coenzymes (i.e., NADH and MV^•+) have so far limited the application of this technology [3].

Therefore, efficient and inexpensive coenzymes regeneration methods using economic energy sources would provide a greener and sustainable pathway for CO₂ reduction.

In recent years, different groups have been focused their activity on the improvement of visible-light-driven photoredox reactions to implement artificial photosynthesis systems for the efficient regeneration of enzymatic cofactors, in order to lower the cost of the CO₂ reduction process [4-7]. These systems are usually formed by a photosensitizer, an electron donor, and an electron acceptor [4-7]. Upon illumination and light absorption, the photosensitizer is converted from the ground state to the excited state. The photo-excited electrons from the triplet state are then transferred to the electron acceptor (i.e., NAD⁺ or MV²⁺), and at the same time, the triplet state of the photosensitizer regains the electrons from the electron donor (usually, triethanolamine or TEOA). Moreover, for NADH regeneration, rhodium-based compounds are commonly added to the reaction and used as electron mediators to guarantee the regiospecific reduction of NAD⁺ to the enzyme active form of NADH (i.e., 1,4-NADH) [8, 9].

So far, different materials have been used as photosensitizers, from inorganic species [10] to organic dyes such as porphyrins and their derivatives [11]. Compared to the inorganic materials, the organic molecules are cheaper and easier to obtain and have a wide variety of chemical and physical properties [12]. Nevertheless, there are some practical limitations to their application, including instability, slightly high cost and difficulty of recovery [12].

Here we report about the use of conjugated-polymer-based water-processable nanoparticles (WPNPs) as photosensitizer in a photoredox system for the regeneration of both NADH and MV^•+. Such materials show excellent thermo- and photo-stability, good visible absorption properties, low cost/low temperature solution processing and easy removal [13]. Moreover, the use of non-toxic solvents (i.e., water) makes them appealing in biomedical and biocompatible applications [14].

For this purpose, we synthetized a derivative of the semiconducting polymer P3HT, i.e., poly[2,2''''-bis[[(2-butyloctyl)oxy]carbonyl][2,2':5',2'':5'',2'''-quaterthiophene]-5,5'''-diyl] (PDCBT), and prepared the corresponding WPNPs through a miniemulsion approach [15-17]. Then, the PBDTTPD WPNPs preparation was employed in a visible-light-driven NADH/MV^•+ regeneration system consisting of TEOA as electron donor and NAD⁺ or MV²⁺ as electron acceptors and enzyme-catalyzed redox reactions were used to validate the production of the regenerated cofactor.

Our results show a successful example of conjugated-polymer WPNPs used as a photosensitizer for selective coenzyme regeneration in an artificial photosynthesis system, which is easy to build and usable under mild conditions for the facile regeneration of different coenzymes. These results could pave the way for other photoredox reaction systems that enable the selective and sustainable production of chemicals and fuels through the use of solar light.

This study demonstrates the disruptive potential of organic photoabsorbing blends in overcoming a critical limitation of metal oxide photoanodes: insufficient photogenerated current. Various organic blends, including PTB7-Th:FOIC, PTB7-Th:06T-4F, PM6:Y6, and PM6:FM, were systematically tested. When coupled with state-of-the-art electron transport layer (ETL) contacts, these blends exhibit exceptional charge separation and extraction, with PM6:Y6 achieving saturation photocurrents up to 17 mA/cm² at 1.23 V_RHE (oxygen evolution thermodynamic potential). Employing a novel architecture, we highlight the remarkable attributes of PM6:Y6 blends, showcasing their ability to be fabricated as semitransparent organic photoanodes in tandem structures. The implementation in a double PM6:Y6 photoanode/photovoltaic structure resulted in photogenerated currents exceeding 7 mA/cm² at 0 V_RHE (hydrogen evolution thermodynamic potential) and cathodic onset potentials as low as -0.5 V_RHE.

To our knowledge, this work represents the first time a tandem structure utilizing organic photoanodes has been computationally designed and fabricated, achieving unprecedented performance levels for both the standalone photoanode and the tandem configuration. The structural design has been modeled using the transfer matrix method and experimentally tested, determining the optimal thickness of the first semitransparent light-absorbing element, the photoanode, to be 75 nm. This integrated approach holds promise for advancing the field of photoelectrochemical solar conversion to fuels and paves the way for further exploration of blend combinations and oxidative reactions targeting.

Photoelectrochemical (PEC) artificial leaves can lower the costs of sustainable solar fuel production by integrating light harvesting and catalysis within a compact panel.^[1] However, most prototypes can only perform water splitting over a few hours on a cm² scale, whereas conventional light absorbers limit solar-to-fuel efficiencies and product rates.^[2,3] Here, we explore alternative routes to expand the performance and functionality of PEC devices, by designing integrated systems which benefit from unconventional materials and complementary energy harvesting. To this end, we first demonstrate the fabrication of lightweight artificial leaves by employing thin, flexible substrates and carbonaceous protection layers,^[4,5] which are compatible with modern fabrication techniques.^[6] These materials allow 100 cm² perovskite-BiVO₄ artificial leaves to float along River Cam (UK), showcasing the potential of floating solar fuel farms.^[5] The same carbonaceous protection layers can be employed to increase the moisture stability of an underexplored BiOI photocathode from minutes to >500 h, whereas a pixelated design provides the additional photovoltage required for unassisted water and CO₂ splitting.^[7] Product rates can be significantly boosted by integrating PEC devices onto thermoelectric (TE) generators, which harvest waste heat from thermalisation and IR photons.^[8,9] Accordingly, a Pt-TE-BiVO₄ device can already yield unassisted water splitting under 2 sun irradiation, while the photocurrent of a Pt-perovskite-TE-Fe₂O₃ device is boosted 30-fold under 5 sun irradiation.^[9] Further up-scaling of PEC systems towards m² areas can be performed by taking advantage of the modularity of artificial leaves,^[10] however, manual fabrication must be replaced by high-throughput techniques for large scale applications.^[6]

Organic semiconductors are gaining prominence as promising materials for heterogeneous photocatalytic (PC) solar fuel production due to their molecular tunability and scalable processability. Notably, bulk heterojunction (BHJ) nanoparticles (NPs), synthesized through a mini-emulsion approach, have proven to be high-performance and cost-effective photocatalysts for solar hydrogen production under sacrificial conditions. Despite their success, the factors crucial for optimizing the performance of these BHJ NPs remain poorly understood. This presentation delves into the intricacies of BHJ-based systems, drawing insights from photoelectrochemical measurements and model photocatalysts. It highlights the significance of co-catalyst loading and particle formation methods, unraveling their impact on the stability and overall water-splitting ability of BHJ NPs. In particular, the importance of controlling the nucleation and growth of the Pt co-catalyst on the BHJ surface is highlighted. By elucidating this critical factor and others, the presentation paves the way for a more comprehensive understanding of BHJ-based NP systems, aiming to enhance their efficiency and durability in the realm of solar fuel production.  

To fight against climate change, the production of carbon-neutral energy is critical. Among the solar fuels, hydrogen stands out, allowing the storage of the sun’s energy as chemical bonds, by the photoreduction of protons. Pinaud et al. ^[1] have shown that using particle slurries can be a competitive and green way to produce hydrogen (Fig. 1).

Figure 1 Scheme of the cycle of Photocatalytic Hydrogen Generation, from the harvesting of the photons to the use of the solar fuel

In 2016 Tian et al.^[2] managed to produce hydrogen using organic semiconductor nanoparticles. A hydrogen evolution rate (HER) of 8 mmol.h^-1.g^-1 was achieved and showed how the surfactant is essential to increase the catalytic activity, by creating pdots.^[2] Moreover, the surfactant’s charge can potentially increase the surface photon affinity and the surfactant itself can lead to nanoparticles with a more suitable morphology, as demonstrated by Kosco et al.^[3]: by changing the surfactant from SDS to TEBS, the morphology of the nanoparticles shifted from core-shell to intermixed, which resulted in a higher HER. It appears clear that it is fundamental to understand the role of the surfactant, in order to find the best one to increase the hydrogen evolution rate.

In this communication, it will be discussed how changing the surfactant used for nanoparticle stability affects the HER. A first study on the system P3HT:PC₆₁BM was conducted, as a reference, with nanoparticles prepared via nanoprecipitation. For instance, a Janus morphology was recently reported for this system, which can be of great interest in the case of photocatalytic applications.^[4]. From this, a state-of-the-art donor:acceptor couple will be introduced, with a promising HER of 13 mmol.h^-1.g^-1. To characterize the particles before and after the hydrogen evolution Dynamic Light Scattering, Cryo-TEM, and UV-Visible spectroscopy were used.

Photoelectrochemical (PEC) water reduction is one of the key reaction routes to harness solar energy for the production of green hydrogen (H₂). Copper oxides are promising semiconductor materials for PEC applications as it possesses a narrow bandgap and favourable energy band positions. Owing to its severe photocorrosion, poor charge separation and rapid recombination of photogenerated electron-hole pairs, however, the use of conventional pristine Cu₂O photocathodes is not technically feasible for PEC reduction of water. Moreover, it is widely known that the morphology and structure of copper oxide-based photocathodes are pivotal to their PEC performance as they heavily influence the intrinsic properties, such as the density of active sites and specific surface areas. Modulation of these properties is crucial in increasing the available active sites for surface redox reactions and improving the separation efficiency of photogenerated charge carriers. Therefore, it is highly desirable to construct an efficient nanostructured heterojunction of Cu₂O-based photocathode to improve its photoconversion efficiency. Herein, we report a carefully engineered multi-component cuprous-cupric oxide heterostructure with distinct 2D-nanoflake morphologies (Cu₂O-CuO-NF) to overcome the technical bottlenecks encountered in conventional Cu₂O photocathodes. This was achieved through hot alkali treatment and subsequent thermal oxidation on the electrodeposited Cu₂O/FTO, which partially oxidises Cu₂O into 2D-CuO nanoflakes. Positive characteristics brought upon through this synthesis technique resulted in an enhanced specific surface area, which could facilitate a greater photocathode-electrolyte interfacial contact and higher water reduction activity during the PEC application. Enhanced charge separation efficiency is achieved as a result of the band bending at the Cu₂O-CuO interface, which provides a strong driving force for electron-hole separation and therefore, reduces the electron-hole pairs’ recombination process. Through morphological and compositional analyses using FE-SEM, XRD, and XPS, the formation of CuO nanoflakes on the surface of the Cu₂O photocathode was proven and validated. From the optical and electrochemical measurements, the charge carrier density was enhanced while the interfacial charge transfer resistance was reduced. This confirmed that the formation of heterojunction coupled with the surface nanoflake morphology of CuO plays a pivotal role in facilitating charge separation of photogenerated charge carriers. A high photocurrent density of -2.96 mA/cm² at -0.6 V vs Ag/AgCl under AM 1.5 G solar irradiation (100 mW/cm²) was achieved, which was approximately 27 times, and 1.5 times higher than that of the pristine Cu₂O (-0.11 mA/cm²) photocathode and Cu₂O-CuO (-1.96 mA/cm²) photocathode without 2D surfaces nano-structuration, respectively. Ultimately, experimental-derived information and data were deployed to construct a theoretical band diagram that elucidates the heterojunction band alignment of Cu₂O and CuO over the PEC water reduction mechanism.

Transition metal nitride semiconductors are rapidly emerging as a promising class of materials that provide a combination of properties well suited for advanced optoelectronic and energy conversion applications. Compared to oxides, nitride semiconductors offer narrower bandgaps, stronger bond covalency, and improved carrier transport properties that make them well suited for harvesting sunlight in photovoltaic and photoelectrochemical systems. Despite this considerable promise, dramatically fewer nitrides than oxides have been experimentally investigated due to their synthetic complexity. Consequently, a broad range of new compounds remain to be explored as functional semiconductors. Furthermore, synthesis challenges have led to poorly controlled defect and impurity properties within this class of materials. In this work, we overcome these limitations using reactive co-sputtering as a non-equilibrium deposition approach to synthesize thin film nitride semiconductors with precisely controlled compositions, exploring both dopants and new compounds in the Ti-Ta-N and Zr-Ta-N composition spaces. Starting with orthorhombic Ta₃N₅, which stands as the best performing photoanode material within this class, we investigate the critical roles of nitrogen vacancy (v_N), substitutional oxygen (O_N), and reduced Ta centers (Ta³⁺) on thin film photoconversion efficiencies [1]. Using in situ photoelectrochemical transient absorption spectroscopy, we identify spectral signatures of photogenerated holes and assess the competition between recombination and chemical reaction at the interface, which is directly impacted by point defect concentrations. Armed with these insights, we show that substitutional Ti doping on Ta sites (TiTa) can be used to improving photoconversion efficiencies [2]. While the structural properties and optical bandgap are only minimally affected by the incorporation of Ti impurities, we observe substantial changes in surface photovoltage, band bending, and recombination dynamics with Ti content. At low Ti content, Ti doping substantially improves the photoelectrochemical performance, manifesting in increased photocurrent densities and favorably reduced onset potentials. While high Ti contents lead to the precipitation of a secondary TiN phase, different behavior is observed for the case of Zr incorporation. In particular, introduction of a 1:1 Zr:Ta ratio leads to formation of a new ternary nitride compound, bixbyite-type ZrTaN₃. Comprehensive investigation of the properties of this material reveal that it is a strong visible light absorber and is an active photoanode material. Complementary DFT calculations indicate a direct bandgap that is tunable based on cation site occupancy. Thus, this material offers exciting prospects not only for solar energy conversion but also for optoelectronics applications. Overall, these results highlight the promise of both established and new transition metal nitride semiconductors for solar energy harvesting, as well as the importance of precise composition engineering to tune optoelectronic and charge transport characteristics.

Photocatalytic systems for the reduction of CO2 into fuels and platform chemicals are multi-variable multi-metric systems. As such, long-established optimisation approaches are poorly suited maximising system performance because they aim to optimise one performance metric while sacrificing the others and thereby limit overall system performance. I will present work to address this multi-metric challenge by defining a metric for holistic system performance that takes all figures of merit into account, and employing a machine learning algorithm to efficiently guide our experiments through the large parameter matrix to make holistic system optimisation accessible for human experimentalists. The test platform is a five-component system that self-assembles into photocatalytic micelles for CO2-to-CO reduction, which we experimentally optimised to simultaneously improve yield, quantum yield, turnover number, and frequency while maintaining high selectivity. Leveraging the dataset with machine learning algorithms allows quantification of each parameter’s effect on overall system performance, revealing that the buffer concentration is the dominating parameter for optimal performance with nearly four times more importance than the catalyst concentration. The expanded use and standardisation of this methodology to define and optimise holistic performance will accelerate progress in different areas of catalysis by providing unprecedented insights into performance bottlenecks, enhancing comparability, and taking results beyond comparison of subjective figures of merit.

Simultaneously harvesting, converting and storing solar energy in a single device represents an appealing technological approach for the next generation of power sources. Herein, we propose a device consisting of a single piece of (rhenium, Re; niobium, Nb) doped WSe2 photocathode and zinc foil anode system capable of harvesting solar energy (and converting it into electricity) and a rechargeable aqueous zinc metal cell. The proposed cell demonstrated an improvement in the specific capacitance of around 45% under visible light irradiation. Also, Re doped WSe2 exhibited a loss of 4.44 % at its initial specific capacitance after 500 galvanostatic charge-discharge cycles at 50 mA g-1. As a result, the device delivered a specific energy density of 574.21 mWh Kg-1 at a power density of 5906 mW Kg-1 and at 0.015A g-1. Furthermore, the notable enhancement in capacitance achieved through the Re-WSe2 photo-zinc ion capacitor (photo-ZIC) offers a straightforward and constructive avenue to establish a cost-effective and highly efficient solar-electrochemical capacitor system.

Polymeric carbon nitride (CN) has emerged as a promising photoanode material in photoelectrochemical cells (PEC) thanks to its cost-effectiveness, lack of toxic or rare elements, high chemical stability, and suitable band edge positions (for common redox reactions as water-splitting and a band gap in visible range.^1–3 However, CN faces several challenges as an active photoanodic material due to its inherent moderate light-harvesting capabilities and inadequate exciton separation properties. These intrinsic properties, combined with the difficulty in establishing intimate contact with the substrate, hinder the successful application of CN in PEC systems. Controlling the properties of a CN layer during its in–situ growth on a conductive substrate, including its directionality, morphology, surface area, and defects, is challenging because of the high temperature of the reaction and the substrate properties. Here, we report the growth of defect–rich, porous 1D CN with enhanced optical properties and photocatalytic activity by utilizing sophisticated supramolecular assemblies composed of melamine and HCl as starting precursors. The supramolecular assembly composition directs the 1D growth, and the presence of protonated amines in the precursor leads to partial condensation and defect generation. The 1D configuration, high surface area, and abundance of defects of the photoanode result in high photoelectrochemical activity for water and benzylamine oxidation. The new design of the CN photoanode leads to very low overpotential, good photocurrent density of 183 ± 8 μA cm−2 at 1.23 V vs. RHE, with an incident photon to current efficiency (IPCE) of up to 10.5% and enhanced stability (retaining ~62% activity) for 10 h. Moreover, the conversion of benzylamine into benzaldehyde and imine N–benzylideneaniline in the presence of O2 reaches 56% compared to 16% for the reference CN photoanode.

Graphitic carbon nitride (gC₃N₄) has been deployed in various applications, including photocatalysis. Among photocatalytic reactions studied, H₂O splitting for the production for H₂ is the most common one, and carbon nitride can serve as pure catalyst, cocatalyst, catalyst support, or part of a heterojunction. Photocatalytic CO₂ reduction is another reaction of interest, combining utilisation of CO₂ emissions and production of sustainable fuels and chemicals. Research on the use of carbon nitride for this purpose is less extensive, and almost always includes the use of dopant materials or heterojunctions. For instance, the role of boron (B) as dopant for gC₃N₄ has started to be explored but mostly for application in zinc batteries, photodegradation of organics, and photocatalytic H₂O splitting/H₂ production. Only a couple of studies have investigated the role of B-doping on CO₂ photoreduction and B-gC₃N₄ seems superior to the pristine material, for all reactions studied. Yet, the relationship between the structure/chemistry of B-doped gC₃N₄ on its chemical, sorptive and optoelectronic properties, as well as CO₂ photoreducing activity remains largely unknown. If understood, a greater control of and more efficient B-doped gC₃N₄ could be reached.

In our study, we aim to bridge this knowledge gap in (photo)chemistry of B-gC₃N₄. We produced two sets of B-doped gC₃N₄ samples through calcination of melamine mixture with varying amount of either amorphous boron, or boric acid.  Once synthesised, we characterized our samples using: XPS, XRD, N₂ sorption (77 K), CO₂ sorption (288, 298, 308 K), DRS UV-Vis, steady-state PL, TAS and EPR. We confirmed the successful B-doping of gC₃N₄ using both B precursors (from 0.5 to 11 at% B). We could control better the amount of doping using boric acid, owing to its greater reactivity with melamine. Introducing B causes oxygen (O) to be also included in the structure in analogous amounts and B-O bonds. High B content results in increased BET area and enhanced CO₂ adsorption. B-doping lowers the band edges, without changing the bandgap of the material. All samples show similar tri-s-triazine structure and light absorbance, however different relaxation patterns and creation of mid-gap states. The samples share similar charge carrier lifetimes and kinetics, even though B-doping up to 5 at% increases the amount of excitons. We noticed differences in the amount of unpaired electrons, which are potentially connected to the chemical structure changes caused by B integration from different precursors. Most of the samples show change in EPR signal intensity before and after irradiation, an indication of excited electrons. Our study provides for the first time a comparison between B precursors for B-doping of C₃N₄, and thorough investigation of their effect on the material’s sorptive and optoelectronic properties. We will next study the B-gC₃N₄ materials’ photocatalytic properties.

Wastewater generated in gold leaching processes, particularly in cyanidation treatments, still contains harmful amounts of cyanide (free and complexed with different heavy metals). In most cases, the cyanide content in these effluents exceeds the minimum concentration limits established for their discharge. Therefore, the highly toxic nature of these effluents requires adequate treatment prior to disposal. In this respect, a novel advanced oxidation technique was evaluated for the elimination of cyanide compounds, called here as photo-electrocatalytic-peroxone (PEPX), consisting of the coupling of two processes: photoelectrocatalysis (PEC) and electro-peroxone (EPX).

Different parameters and operating conditions were evaluated such as the electrogeneration of hydrogen peroxide, pH, anodic and cathodic overpotential, photon flux, the presence of substrates on the photocurrent and cathodic reduction of oxygen. To interpret the interfacial phenomena that were presented, electrochemical techniques such as linear and cyclic voltammetry, chronoamperometry, open-circuit chronopotentiometry, rotating disk electrode (RDE) and electrochemical impedance spectroscopy were used.

Likewise, through a response surface analysis, the effect of variables such as current density and initial substrate concentration on the free cyanide degradation capacity and on the specific energy consumption of the process was determined.

With the proposed process, 100% of the CN^- ion was degraded using 0.086 mA/cm² and 94 ppm initial concentration, with the additional degradation of CNO^- (29.4%) and a specific energy consumption of 4.68 kW-h L^-1. Furthermore, in the treatment of cyanide wastewater from mining activities, the complete degradation of the metal cyanide complexes (copper and iron) and the total precipitation of their metals as hydroxides and oxy-hydroxides were evident. Finally, it was evident that the PEPX process could not only degrade cyanide substrates but could also provide, through photovoltaic effects, part of the electrical energy necessary for the operation of the cell.

Unique insights into the synthesis-structure-properties dependencies of α-SnWO₄ photoanodes will be presented, enabled by synthesis conditions far from thermodynamic equilibrium (i.e., non-equilibrium, NE) based on plasma deposition processes combined with rapid thermal processing (RTP).

Recently, we have shown that a highly controlled NE synthesis approach based on pulsed laser deposition (PLD) combined with RTP had significant effects on the crystallinity, morphology, crystallographic orientation, and sulfite oxidation performances of α-SnWO₄, culminating in photocurrents ~ 70 % higher than PLD-grown photoanodes annealed via conventional furnaces (furnace heating, FH).^[1] In a follow-up study of the structural and electronic properties of α-SnWO₄ films annealed via RTP and FH, we utilized X-ray diffraction texture analysis, electron backscatter diffraction in scanning electron microscopy, and atomic force microscopy modes of Kelvin probe, electrical conductivity, and tapping mode for surface morphology.

Our results show that the local (micrometer-scale) crystallographic orientation is very homogenous in the RTP-treated films, both in-plane and perpendicular to the substrate. Considering their reported higher crystallinity and the anisotropic charge transport nature of orthorhombic α-SnWO₄,^[1–3] the RTP-treated films demonstrate significant changes in contact potential difference (CPD) and conductivity. The CPD between α-SnWO₄ and platinum-coated silicon tip was ~ 0.35 eV lower than the CPD of the FH-treated films, indicating increased band bending. This suggests that the RTP treatment strongly contributes to an increased driving force for charge injection. In addition, the local conductivity of the RTP-treated films was higher by more than two orders of magnitude than that of the FH-treated films.

Our results can lead toward broader tunable materials synthesis and design pathways and more scalable discovery and development of new chemical spaces of multinary metal-oxide photoelectrodes inaccessible through conventional solid-state reactions.^[4]

Wide bandgap semiconductors are well-known active materials in photo-driven processes. Howbeit, a specific class of nanocrystals (NCs) made of doped metal oxides stands out among the others for an additional degree of manipulation of the light–matter interaction.[1] Such nanomaterials can increase their charge density upon absorbing photons above the bandgap.[2] This photo-doping effect occurs primarily thanks to illumination and can induce an accumulation of electrons, as for the case of Tin doped Indium Oxide (ITO) nanoparticles.[3] The possibility to access such excess of charges can open the way for many different applications in light-based technologies like solar batteries or photo-electro-catalysis.[4] In order to promote such a phenomenon, the material must be able to delocalize the charges and neutralize those of opposite sign. In the work herein presented, we investigated the ITO NCs photo-doping with the aim of supporting and enhancing it with opportune counterparts acting as hole-scavenger.[5,6] By means of spectroscopic analysis we explored reversible charge transfer with redox mediators and demonstrated improved stability thanks to electron donating graphene quantum dots. We further studied the ITO NCs in the form of thin films and in particular the photo-electrochemical response under the influence of the light-driven charging of the ITO-based electrodes. With this paper, we give an overview over our recent achievements in the field.

To master the energy transition, green hydrogen is heavily discussed as a supplement to fossil energy carriers such as oil and natural gas. Photocatalysis enables the direct conversion of solar power to hydrogen. In this work, the photocatalytic application of aqueous organic semiconductor dispersions is investigated for photocatalytic hydrogen generation. The nanoparticles comprise bulk-heterojunctions of donor/acceptor-materials and are synthesized by nanoprecipitation. Using colloidal particles warrants largest photoactive interfaces in the dispersion. To further maximize the photoactive interface, this work deliberately omits the use of surfactants for steric stabilization of the dispersion. Instead, for the first time, electrical dopants are used to charge the nanoparticles and thus to implement electrostatic repulsion of the nanoparticles. By attaching the co-catalyst platinum to the nanoparticle surface, the overpotential of the hydrogen evolution reaction is reduced. The result is an efficient generation of hydrogen. These aqueous dispersions exhibit great stability of more than 10 weeks on the shelf and are also stable during operation.