Bismuth vanadate, bismuth oxyiodide and bismuth oxybromide thin films growth via Aerosol-Assisted CVD as the main process and their photoelectrochemical properties

UNIVERSIDAD DE VALLADOLID ESCUELA DE INGENIERIAS INDUSTRIALES Máster en Ingeniería Química Bismuth vanadate, bismuth oxyiodide and bismuth oxybromide thin films growth via Aerosol-Assisted CVD as the main process and their photoelectrochemical properties Autor: Vicente Matas, Javier Mato Chaín, Rafael Bartolome Imperial College London Londres, Enero 2022. TFM REALIZADO EN PROGRAMA DE INTERCAMBIO TÍTULO: Bismuth vanadate, bismuth oxyiodide and bismuth oxybromide thin films growth via Aerosol-Assisted CVD as the main process and their photoelectrochemical properties ALUMNO: Javier Vicente Matas FECHA: 07 de Enero de 2021 CENTRO: Faculty of Engineering, South Kensington Campus UNIVERSIDAD: Imperial College London TUTOR: Dr. Salvador Eslava Content 1 INTRODUCTION ..................................................................................................... 1 1.1 Background................................................................................................................................... 1 1.2 Solar Fuels .................................................................................................................................... 4 1.2.1 Current methods of hydrogen production ............................................................................... 6 2 PHOTOELECTROCHEMICAL WATER SPLITTING (PEC) .............................................. 9 2.1 Introduction.................................................................................................................................. 9 2.2 General explication of the process .............................................................................................. 10 2.3 Semiconductor properties .......................................................................................................... 11 2.4 Space charges and band bending in the semiconductor-electrolyte interface ............................. 12 2.5 Fundamentals of photoelectrochemical water splitting .............................................................. 14 2.6 PEC cell configurations20,16 .......................................................................................................... 16 2.7 Photoelectrode materials in PEC devices .................................................................................... 18 3 BISMUTH VANADATE AS PHOTOANODE IN PEC DEVICES ...................................... 20 3.1 Description ................................................................................................................................. 20 3.2 Crystal and electronic structures of BiVO4 .................................................................................. 20 4 AA(CVD) PROCESS ................................................................................................ 23 4.1 Fundamentals of the CVD process .............................................................................................. 23 4.2 Historical evolution .................................................................................................................... 24 4.3 Fundamentals of AACVD process ................................................................................................ 25 4.4 Mass-transport mechanisms ....................................................................................................... 26 4.4.1 The boundary layer ................................................................................................................. 26 4.5 Rate-limiting steps ...................................................................................................................... 29 4.5.1 Control of the limiting step33 .................................................................................................. 30 5 MATERIALS AND METHODS ................................................................................. 31 5.1 Research objectives .................................................................................................................... 31 5.2 AACVD set up ............................................................................................................................. 31 5.3 Materials .................................................................................................................................... 32 5.4 Methodology .............................................................................................................................. 33 5.4.1 Synthesis of BIVO4 in two steps process.................................................................................. 33 5.4.2 Synthesis of BIVO4 in one AACVD step process ....................................................................... 33 5.4.3 Final experiments studied ....................................................................................................... 34 5.5 Security and flammability limits of the solution ......................................................................... 35 5.6 Characterization ......................................................................................................................... 36 5.6.1 X-ray diffraction (XRD) ............................................................................................................. 36 5.6.2 UV Vis Spectroscopy ................................................................................................................ 36 5.6.3 Scanning electron microscope (SEM) ...................................................................................... 36 5.6.4 Thermogravimetric Analysis (TGA) .......................................................................................... 36 5.6.5 Raman spectroscopy ............................................................................................................... 36 5.6.6 PEC measurements .................................................................................................................. 37 5.7 Results and discussion ................................................................................................................ 37 5.7.1 Flow rates ranges and carrier gas ............................................................................................ 37 5.7.2 Temperature............................................................................................................................ 39 5.7.3 Structures ................................................................................................................................ 41 5.7.4 Physical Characterisation ........................................................................................................ 43 5.7.5 PEC measurements .................................................................................................................. 55 6 CONCLUSIONS ..................................................................................................... 58 7 BIBLIOGRAPHY ..................................................................................................... 60 RESUMEN Y PALABRAS CLAVE Este trabajo propone la síntesis de fotoelectrodos de vanadato de bismuto (BiVO4), óxido de bromuro de bismuto (BiOBr) y óxido de yoduro de bismuto (BiOI) en sustratos de flúor dopado de vidrio de óxido de estaño (FTO) a trasvés de un proceso AACVD de una etapa o un proceso AACVD de dos etapas para su utilización en PEC devices. Los fotoelectrodos se caracterizan por difracción de rayos X (XRD), Raman espectroscopía (RS), espectroscopía UV-VIS y Microscopia electrónica de rastreo (SEM). Las propiedades fotoelectroquímicas (PEC) de los fotoelectrodos se estudiaron en Na2SO4 acuoso 0.5 M y muestran que el mejor valor fue el BiOI depositado durante 5 horas y 350 ºC (1,1 mA cm-2 a 1,23 V vs RHE). Aunque el objetivo inicial era obtener BiVO4, tanto en el proceso de dos pasos como en el de un paso se obtuvieron valores inferiores con respecto a la deposición de BiOI. En el proceso de dos pasos, el mejor valor fue la deposición para 5h y 350 ºC (0.7 mA cm-2 a 1.23V vs RHE). En el proceso de un solo paso, la mala adhesión con el FTO produjo valores bajos. Palabras clave: AACVD, BiVO4 BiOI BiOBr, División de agua fotoelectroquímica, Fotoelectrodo, Dispositivos PEC. ABSTRACT AND KEYWORDS This work proposes the synthesis of photoelectrodes of bismuth vanadate (BiVO4), bismuth bromide oxide (BiOBr), and bismuth iodide oxide (BiOI) on tin oxide glass doped fluorine substrates (FTO) through a one-step AACVD process or a two-step AACVD process for use in PEC devices. Photoelectrodes were characterised by X-ray diffraction (XRD), Raman spectroscopy (RS), UV-VIS spectroscopy, and Scanning electron microscopy (SEM). The photoelectrochemical properties (PEC) of the photoelectrodes were studied in 0.5 M aqueous Na2SO4 and show that the best value was the BiOI deposited during 5 hours and 350 ºC (1.1 mA cm-2 at 1.23 V vs RHE). Although the initial goal was to obtain BiVO4, both the two-step and one-step processes will have lower values regarding BiOI deposition. In the two-step process, the best value was deposition for 5h and 350 ºC (0.7 mA cm-2 at 1.23V vs RHE). In the one-step process, poor adhesion with the FTO produced low values. Keywords: AACVD, BiVO4 BiOI BiOBr, Photoelectrochemical Water splitting, Photoelectrode, PEC Devices. 1 1 Introduction 1.1 Background In the last decades, environmental problems have become one of the principal challenges in humanity. In 2015, 17 global goals agreed by world leaders known as Sustainable Development Goals or SDGs were created. Among these goals, several related to environmental problems are clean water and sanitation, affordable and clean energy, climate action, and responsible consumption and production. Many of the ecological problems are due to the increase of the population, global economic and human well-being standards that create an immediate consumption of natural resources and the gains of the greenhouse gas emissions, principally CO2. Figure 1.1 Sustainable Development Goals. Extracted from the Sustainable Development Goals Report.1 Moreover, after the United States formed the Environmental Protection Agency (EPA) in 1979, Green chemistry took the EPA´s mandate developing different principles by asking chemists and engineers to design chemicals, chemical processes, and commercial products to protect and benefit the economy, people and the planet by finding creative and innovative ways to reduce waste, conserve energy, and discover replacements for hazardous substances. The following figure shows the 12 principles of Green chemistry: 2 Figure 1.2 12 Principles of Green Chemistry. Extracted from ACS chemistry for life.2 However, the reality of the production system is far away from these principles. The companies must compete against themselves with the principal objective of increasing their production to eliminate or absorb other competitors. In this competitive system, the main aim is economical, and generally, the environmental changes are so slow if the economic impact is harmful to the companies. Sustainable campaigns are to manipulate public opinion. Meanwhile, principal governments in the world try to have the best geographical agreements to defend the interest of their companies. 2020 will be the year remembered for the crisis of the COVID-19 and how it turned our daily lives upside down. The measures taken by the governments to control the propagation of the virus as the imposition of lockdowns around the world produced a reduction of everyday human activities. The COVID-19 pandemic dramatically impacted energy markets producing that the primary energy and carbon emissions had the most significant recession since World War II (-4.5% and -6.3%, respectively)3. The drop in energy consumption was driven mainly by oil, contributing almost threequarters of the net decline, although natural gas and coal saw significant reductions. The pandemic also led to enormous economic loss. Global GDP is estimated to have fallen by over 3.5% last year – the most considerable peacetime recession since the Great Depression. Despite this strange year, the amount of energy consumed by the world has increased throughout the last 50 years. The trend of the energy consumption during the 3 following years is to grow continually. The increase in the amount of energy consumption in the world is shown in Figure 1.3: Figure 1.3 Total primary energy consumed from 1965 to 2020. Energy comprises fossil fuels (such as oil, gas and natural gas) and renewable sources of energy. Data extracted from the BP 2020 Statistical Review of World Energy.3 The central part of this energy comes from fossil fuels sources of energy, such as oil, natural gas, and coal. Historically the most significant energy source is oil, with 33% of the energy produced in 2019. Around the late 1990s, gas consumption overtook coal, and it was not until the 2000’s that the consumption of energy from renewable sources (solar, wind, geothermal and biomass) started to be necessary. In 2020 the renewable energy, led by wind and solar energy, has increased the amount of energy produced, wind and solar capacity increased 238 GW in 2020 – 50% larger than at any time in history.3 Figure 1.4 Wind and solar capacity in the last years.3 The use of fossil fuels and coal is the unavoidable release of greenhouse gas emissions, principally CO2, which harms global warming and thus climate change that has an - 100.00 200.00 300.00 400.00 500.00 600.00 700.00 1960 1970 1980 1990 2000 2010 2020 2030 Exajoule Year Total world energy comsumption 10 advantages of this process is that oxygen and hydrogen are produced simultaneously in different electrodes. Separate production allows greater control of the process avoiding safety problems and subsequent treatment to separate the gases that make the process more expensive. The purity of the hydrogen produced is an essential requirement for fuel cells. Furthermore, the process can be carried out at room temperature, so photochemical water splitting devices could have applications on a large scale. Another advantage of the process is that the materials in the system can be constructed from inorganic products, so an improvement in the resistance of the materials, more excellent durability and less degradation is achieved.13,9 2.2 General explication of the process There 2 approaches for photocatalytic water splitting: - Basic methods where a powder photocatalyst is dissolved in an aqueous solution - Photoelectrochemical methods (PEC) Basic photocatalyst systems are heterogeneous processes with 2 active phases, a photocatalyst solid and an aqueous solution. This configuration is simple, and the water splitting to produce hydrogen occurs irradiating the photons to the solution with the solute dissolved. However, this process presents problems as pH changes, mainly when a scavenger is used, problems with light absorbance by the suspended particles, generation of H2 and O2 together and secondary reactions such as the reduction of O2 to form superoxide (O2-).14,15 On the other hand, PEC systems do not present these limitations, although this method has other challenges like bad stability of the materials, low efficiency. Some studios on a large scale in both configurations have demonstrated that particle systems have a low cost of hydrogen production, but this cost is higher due to the safety concerns involved in the process and the posttreatment to separate the hydrogen from the oxygen.9 Alternatively, PEC systems can be easier to scale up with fewer safety concerns despite the principal limitations are the low efficiencies and the reduced lifetime of the materials used. This is because some of the main properties of the photoelectrodes depend on factors that are either difficult to control or not well understood. An example of this problem is the creation of impurities during material synthesis. Optimising the synthesis is crucial for developing materials on a large scale, primarily when low-temperature and low-cost methods are used.13 11 2.3 Semiconductor properties Semiconductors, according to molecular bonding theory, are materials with a valance band (VB) (highest occupied molecular orbital) and conduction band (CB) (lowest unoccupied molecular orbital), which correspond to the bonding and anti-bonding energy levels separated by an energy gap (Eg). The bandgap energy is defined by the energy difference between CB and VB, and its value is usually in the range of 1-5 eV.11,16 There are two types of optical band gap transitions depending on the nature of the electronic properties of the semiconductor: Direct transition: The lowest point of the CB is situated at the same k-vector as the highest point in the VB. In this case, the crystal momentum is maintained when a change of energy occurs. Indirect transition: The lowest point of the CB is situated at a different k-vector to the highest point in the VB. Crystal momentum appears with the change of energy. These transitions require the absorption or emission of a photon, and the probability of it occurring is less. Figure 2.1 Simplified band diagram of a semiconductor. (A) Direct band gap transition. (B) Indirect band gap transition. Adapted from ref.13 At zero kelvin, the energy states of the CB are mostly empty of electrons, whereas the energy levels of the VB are filled up to maximum energy, the Fermi level (Ef). The excitation by an external energy source with energy equal to or greater than the band gap energy (Eg) produces a jump of electrons from the levels of the VB to the levels of the CB, leaving the same number of holes in the VB. This is the case for intrinsic or un-doped semiconductors where the Ef is placed at the mid-gap position between VB and CB. On the other hand, doped semiconductors with impurity atoms have additional energy levels in their electronic configurations. Suppose the A B 12 semiconductor has impurities atoms with a more significant number of valence electrons than the standard configuration. In that case, it results in an electron-rich semiconductor (n-type, donor-doped semiconductor). In contrast, if the impurities atoms of the semiconductor have fewer valance electrons than the normal material, it results in a hole-rich semiconductor (p-type, acceptordoped semiconductors). In some cases, the defects present in the configuration define the electronic character of the material. For doped semiconductors, the number of free electrons and holes depends on the level of doping rather than the thermal excitation of electrons and holes in the band gap producing a change in the position of the Fermi level. Figure 2.2 Band diagram of an intrinsic, p-type and n-type semiconductor. Adapted from ref11 2.4 Space charges and band bending in the semiconductor-electrolyte interface The semiconductor in contact with an electrolyte creates an electric current that initially flows across the junction semiconductor-electrolyte until electronic equilibrium is reached where the Fermi energy of the electrons in the semiconductor (Ef) is equal to the redox potential of the electrolyte (Eredox). This transfer of electronic transfers produces a region at the semiconductor-electrolyte with a charge distribution different from the bulk material, known as the space-charge layer, at the semiconductor side, and electrolytic double layer at the electrolyte side, which is divided in the Helmholtz layer and the diffuse Gouy-Chapman layer. The diffuse Gouy- Chapman layer is the region where water molecules surround the ions present in the solution. The Helmholtz layer is the region closest to the semiconductor, formed principally for adsorbed ions and low ions in the solution.13,17 13 Figure 2.3 Schematic diagram of the semiconductor-electrolyte interface. Space-charge layer and electrolytic double layer. Adapted from ref.13 A band bending of the CB and VB is necessary to achieve an electronic equilibrium in the interface of the semiconductor-electrolyte. On the junction region of the semiconductor, the band bending depends on the position of the Fermi level in the semiconductor. There are 4 situations possible: Flat-Band potential (Ufb): When there is no excess of charges on either side of the junction, and the bands are flat, a space charger is not formed, and an electric field is not generated. Accumulation layer: when an excess of positive (negative) charges is adsorbed at the surface of an n-type (or p-type) semiconductor, it produces a downward (upward) bending of the CB and VB towards the interface. Free majority carriers create the accumulation layer near the surface to compensate for these surface charges. Depletion layer: If there is a transfer of an excess of positive (negative) charges from the surface of the n-type (p-type) semiconductor to the electrolyte, produce an upward (downward) bending of the CB and VB towards the interface. Opposite charges species must compensate for the number of charges in the depletion layer at the material's surface. Inversion layer: n-type (p-type) can become p-type (n-type) if the number of negative (positive) surface charges adsorbed increases the Fermi level crosses the middle of the band gap and a downward (upward) band bending of the CB and VB is formed.13,17 14 Figure 2.4 Electronic energy levels of an n-type semiconductor in contact with an electrolyte. (A) flat band potential, (B) accumulation layer, (C) depletion layer and (D) inversion layer. Adapted from ref.17 Copyright 2001 Springer Nature. In addition, the effect on the bad edges of the semiconductor of an applied bias or a change in the pH must be considered. An external bias can be applied to obtain the potentials required to undertake the water reactions through the photoelectrode and the counter electrode. The charges in the space layer are modified when an external bias is applied, and the semiconductor's band bending is also limited. The position of the band edges depends on the pH of the solution, about 59 mV per pH unit concerning the redox potential in the electrolyte. However, this is not a helpful property due to the reduction and oxidation potentials of water also depend on the pH for the Nest equation: 2-1 2.5 Fundamentals of photoelectrochemical water splitting The mechanism of the PEC water splitting begins with the absorption of a photon in the semiconductor material; this energy must be equivalent to or greater than its band gap to form excited electron-hole pairs. The energy requires a minimum Gibbs free energy of 237 kJ mol-1.16 The electrons (e-), which are in the valence band (VB), are exited into the conduction band (CB), producing the creation of holes (h+) in the VB. At this moment, the system lost the condition of equilibrium, especially at the spacecharge region, and quasi-Fermi levels appear for e- and h+, instead of having a unique A B C D 15 Fermi level in the semiconductor-electrolyte junction, which moves its position downward or upward for a p-type or n-type semiconductor, respectively.18,19,13 Regardless of the configuration used in the process, the overall mechanism of the water splitting is the next18,9,10: 1) The material's absorption of light with an energy equivalent to or greater than the band gap of the semiconductor material to form excited electron-hole pairs. 2) Separation and migration of these photocarriers to the photocatalyst surface trying to avoid recombination in the bulk or surface, which can appear if there are defects on the surface of the semiconductor avoiding the migration of the photocurrents within the space charge layer. In basic systems, both photocarriers migrate to the surface, but in PEC systems, band bending at the semiconductor/electrolyte interface is the driving force to separate photogenerated carriers; one of the photocarriers migrates to the surface-electrolyte interface of the photoelectrode while the other carrier migrates to the counter electrode in the external circuit. The efficiency of the generation and separation of photocurrents depends on the penetration depth of the light on the surface, the width of the space charger region and the diffusion length of the minority carrier. 3) The holes drive the oxygen evolution reaction (OER) in one of the surfaces while the electrons produce the hydrogen evolution reaction (HER) at a separate surface or electrode. In an n-type semiconductor photoanode, electrons go to the cathode, where the HER is produced. Meanwhile, the OER occurs on the photoelectrode photoanode. In a p-system semiconductor photoanode, the holes flow to the anode, and the HER occurs on the photoelectrode surface. Associated with driving the kinetics of the HER and OER, there are overpotentials at the solid-liquid interface. Minimising these overpotentials is a crucial step to creating efficient water splitting devices. In addition, overpotential, entropic losses with the generation of electrons and holes and other parasitic losses must be considered in the band gap requirement. Considering these losses, the driving force for water splitting is always higher than the material's band gap. The bandgap needs to be enough to provide the energy necessary to split the water and must be as small as possible to have the capability to absorb a significant portion of the solar spectrum. 16 Figure 2.5 photocatalytic overall water splitting on a semiconductor particle. Adapted from ref.18 Copyright (2007) American Chemical Society. To inject h+ and einto the solution, the energy levels must exceed the electrochemical redox potentials for the OER and HER. For this reason, the potential of CB of the semiconductor at the semiconductor/liquid junction must be more negative than the reduction potential of H+ to H2 (0 VNHE at pH=0); meanwhile, the VB must have a more positive potential than the oxidation potential of H2O to O2 (1.23 VNHE at pH=0).9 Figure 2.6 depicts a schematic illustration of band structures of some photocatalysts, which presents the above described: Figure 2.6 Band edge positions for a range of semiconductors. Adapted from ref.21 2.6 PEC cell configurations20,16 A PEC water splitting system mainly is formed for an electrode or photoelectrode, a cathode or photocathode and electrolyte. In the PEC systems exist 2 types different of configurations depending on the number of electrodes used: Two-electrode systems form a counter electrode, usually platinum and the working electrode, which contains the photocatalyst of interest. 17 Three-electrode systems with a reference electrode coupled with the other two electrodes described in the two-electrodes system. Different Setups for the PEC water splitting exist depending on the electrode configuration: Figure 2.7 PEC devices configurations: (A) type I, (B) type II, (C) type III, (D) type IV, (E) type V, (F) type VI. It is reproduced from ref.16 Published by The Royal Society of Chemistry. The simplest configuration is the type I. This configuration only contains a metal counter electrode and one semiconductor light absorber, which can be used either as a photoanode or photocathode to carry the water splitting oxidation or reduction reaction. An external bias is needed for one semiconductor light absorber to trigger the separation of the photocarriers throughout the cell. Type II or heterojunction photoelectrode Pec device. Two semiconductors are coupled in only one electrode in this system to create a heterojunction. For instance, an n-type semiconductor couple with a p-type semiconductor is a heterojunction system. A space-charger layer is formed at the interface of both surfaces producing the diffusion of photocarriers between them. Depending on the alignment of the bands in the ntype and p-type semiconductors, e- and h+ can diffuse the different forms. In the example shown in Figure 2.7 B, holes migrate from the n-type semiconductor's VB to the p-type semiconductor's VB. Meanwhile, the electrons migrate from the CB of the p-type semiconductor to the CV of the n-type semiconductor. The advantages of this setup are the better separation of the charges, a bigger lifetime of the charges, higher 18 reaction rates and an increase in the light absorption when the alignment of the bands have the correct positions.24 Type III and type IV utilise a photocathode and photoanode connected in series in a tandem configuration. A conductive metal wire is used for the setups of Figure 2.7 C, and the setups of Figure 2.7 D, a wireless setup, is used in which a transparent conductive substrate is used as an ohmic contact. To ensure a complimentary light absorption where the second material should absorb the photons from the first material, the CV minimum of the photoanode must be more negative than the VB maximum of the photocathode. The principal advantages of these configurations are higher photovoltage and a significant range of materials used in the tandem. However, both parts of the tandem must maintain a similar current density when an external bias is not applied. Type V device configuration combines a PEC system with a photovoltaic cell. The additional energy necessary to split the water is supplied by the photovoltaic cell, increasing the range of materials used in these systems. A transparent oxide layer protection between them is necessary to provide a recombination layer of the holes from the photovoltaic cell and the electrons from the PEC system (only for n-type configurations). Type VI configuration is a combination of a photovoltaic cell with an electrolyser. The photovoltaic cell supplies the energy, and the electrolyser performs the water splitting. Even though it is not a PEC system in practice, this configuration can be considered because the reactions occur inside the cell. 2.7 Photoelectrode materials in PEC devices Despite 50 years of investigation from the discovery of PEC water splitting, creating a large-scale process with enough efficiency using visible light is still a long way, so the investigation is essential. The principal problem is to find a semiconductor material able to satisfy all the requirements that need to be accomplished in a large-scale system:13,16,23 - Position of the band gap (Eg): The position of the band gap should be in the range of the visible spectrum to maximise the absorption of light. Natural light is formed by 5% of UV (399-400nm), 43% visible (400-700 nm), and 52% infrared radiation (700- 2500nm).16 The optimal band gap in PEC devices is in the range of 1.9 and 3.1 eV, which corresponds to the range of 400-650 nm. The minimum band gap requires enough energy to split the water (1.23 eV) and all the losses and overpotentials 19 generate during the process.16 The theoretical maximum solar-to-hydrogen (STH) efficiency is determined by the position of the band gap of the semiconductor. The highest STH efficiency corresponds with the lower value of the band gap due to the larger absorption of the solar light. - Charge transport efficiency in the semiconductor: Photocathode materials: Recombination is the major problem in the low STH efficiency. Thus, it is necessary to investigate better structures (crystallinity, nanostructure and hole electron mobility) to avoid recombination and promote efficient charge carrier separation and transport.13,16 - Stability in an aqueous electrolyte in the dark and under light illumination: The material must be electrochemically, photoelectrochemically and chemically stable in the solution. Many non-oxide semiconductors dissolve or form a thin oxide layer at the surface in contact with the electrolyte solution producing problems in the charge transfer across the junction interface. In these conditions, metal oxides are more stable, but they can suffer from photocorrosion, particularly metal sulphides, by the holes generated rather than the water. This photocorrosion depends upon the relative positions of the band edges and the respective decomposition potentials. - The band edges position must straddle the redox potentials of the photocatalytic reaction for the hydrogen and oxygen half-reactions: A limited number of semiconductors meet this requirement, which often requires large band gap positions. One advantage of the PEC devices is that only one of the half-reactions for water splitting is needed; therefore, a bigger range of materials can be used. The conduction band must be more negative than the H2 generation potential, and the valence band more positive than the O2 generation potential to produce the overall water splitting. - Low overpotentials for the reactions: The overpotentials should be as low as possible, so the charge transfer should be fast to avoid recombination by accumulating charges at the surface. The combination with other materials has been extensively studied to minimise these problems. - Low-cost and sustainable: To meet the global sustainable development, the materials utilised should be cheap, abundant, and ideally prepared via a green synthetic route. 26 evaporation, descomposition, and homogeneous or/and heterogeneous chemical reactions to form the desired products. Homogeneous reaction processes occur when the temperature of the reactor chamber is too high. Meanwhile, the heterogeneous reactions process occurs when the reactor chamber's temperature is not too high. Typically, the vaporised precursor in a heterogeneous reaction system suffers from preliminary decomposition in the gas phase. The heated surface adsorbs the resultant intermediate products on the heated surface undergoing chemical reactions/decomposition to form the desired film. In a homogeneous reaction system, vaporised precursor decomposes/reacts primarily in the gas phase giving rise to films with very fine particles and structures. Finally, diffusion of by-products to the exhaust system occurs35,41 A schematic image of an AACVD process is shown in Figure 4.2. Figure 4.2 Schematic of the AACVD setup used during the experiments. 4.4 Mass-transport mechanisms 4.4.1 The boundary layer The behaviour of the flow of the gas is controlled by fluid mechanics and is characterised by the Reynolds number (Re). In general, the gas flow is laminar, but at some points, the flow may be disturbed by convective forces and may become turbulent. 27 The boundary layer can be defined as the region of the tube in which the flow velocity changes from a velocity of zero at the wall of the tube to that of the bulk gas away from the wall. The thickness of the boundary layer increases along the tube from the tube's inlet until the flow becomes stabilised, as in Figure 4.3 The reactant gases must diffuse through it to reach the heated surface. The thickness of the boundary layer ∆, is inversely proportional to the square root of the Reynolds number, so the thickness of the boundary layer decreases with higher gas-flow velocity and increases with the distance from the inlet of the tube as the following equation:33,38,39 ∆=�𝑥𝑥 𝑅𝑅𝑅𝑅 4 −1 Where: 𝑅𝑅𝑅𝑅=𝜌𝜌𝑢𝑢𝑥𝑥 µ ρ = mass density u = flow density x = distance from the inlet in the flow direction µ = viscosity We can divide the tube into 3 zones where the boundary layer is different. The first zone exists a “plug flow” regime, where the real flow has a nearly constant velocity with a relatively thin boundary layer, which can be ignored. In the second zone, 2 different boundary layers grow until finally connected, creating an indistinct region in the third zone of the tube with a Poiseuille flow. In the second zone, the velocity, temperature, concentrations, and other parameters are relatively fixed out the boundary layers.38,39 Figure 4.3 increase of the boundary layer along the tube. Adapted from.39 28 Across the boundary layer, different zones delimit where homogeneous or heterogeneous reactions occur. Figure 4.4 shows the different reaction zones. In reaction zone 5 and in the gas stream, homogeneous reactions may occur in the vapour, producing undesirable homogeneous nucleation with bad characteristics for the deposition as flaky and non-adherent coating. In some cases, these reactions are favourable when there is no homogeneous nucleation. In reaction zone 4, heterogeneous reactions occur between the phase boundary vapour/coating. These types of reactions control the deposition rate and the characteristics of the film. If the temperature is relatively high in the process, other types of solid-state reactions can occur in zones 3-1 as phase transformations, precipitation, or grain growth). In zone 1, some intermediate phases can be formed, and the reactions in this part are important to achieve good adhesion of the film on the substrate.33 Equal to the velocity, there are a temperature boundary layer and a concentration boundary layer. In the case of the temperature, it is very similar to the velocity layer. The gases that flow along the tube are heated quickly when they enter into contact with the tube's hot surface, producing a steep temperature gradient. Exist an increase in the average temperature toward downstream. The concentration along the tube becomes gradually depleted for material deposition, and the increase of the by-products gas increases in the boundary layer. Figure 4.4 reaction zones in CVD. Adapted from ref.33 29 Reactor. Adapted from ref.38 4.5 Rate-limiting steps In a CVD some several rate-limiting steps can be identified. There are usually ordered and discussed in five principal categories:33,38 - Surface reactions kinetics: This rate-limiting step occurs when the deposition rate is lower than the mass rate that enters the tube and the mass transport rate in the vapour to or from the substrate. At low temperatures, the reactions occur slowly, and there is an accumulation of reactants at the surface for the low-pressure, creating a thin boundary layer, and the reactants can reach the deposition surface easily. If the boundary layer is thin, the diffusion coefficient is not essential, and the components reach the surface with ease. - Thermodynamic control is produced when the mass input into the reactor and the deposition rate is equal. This is produced at extreme deposition conditions as high temperatures and low flow rates. - Mass transport control usually happens when the pressure and the temperature are high and produce a low flow that creates a thicker boundary layer. In this case, it is the phenomenon that controls the diffusion of the reactants through the boundary layer and the diffusion of the by-products in the desapsortion process. In addition, the decomposition reactions are very quickly at a high temperature. - Nucleation control is produced at low supersaturation - Homogeneous reactions control is produced for some species that the formation rate of key species in the vapour zone can control the deposition rate. Figure 4.5 Temperature boundary layer and temperature changes in a tubular Reactor. Adapted from ref. 38 30 Although these five categories can appear and influence the result, mass control and surface kinetics are usually the rate-limiting steps that control the different processes. Surface kinetics led to achieving better results with uniform thickness on the substrate. 4.5.1 Control of the limiting step33 To predict the rate-limiting step in CVD, reaction resistances are usually used. Diffusion flux JD in the boundary layer can be expressed as: 𝐽𝐽𝐷𝐷=𝐷𝐷 𝑅𝑅𝑅𝑅𝑃𝑃𝑏𝑏−𝑃𝑃𝑠𝑠 ∆ 4 −2 Where R= gas constant, T=absolute temperature (K), D= Diffusion coefficient, ∆= Boundary layer thickness The mass flux JM towards the surface of the substrate can be expressed as: 𝐽𝐽𝑀𝑀=1 𝑅𝑅𝑅𝑅𝑃𝑃𝑏𝑏−𝑃𝑃𝑒𝑒𝑒𝑒 ∆ 𝐷𝐷+1 𝐾𝐾𝑚𝑚 4 −3 Where δ/D and 1/km are the reaction resistances and Pb – Peq is the driven force. If δ/D is much bigger than 1/km, the mass transport in the vapour controls the process; meanwhile, the surface reaction kinetic control the process if 1/Km is much bigger than δ/D. As expressed in equation 4-1, the thickness ∆ of the boundary layer is inversely proportional to the square root of the Reynolds number. The density, ρ, depends on the temperature and the pressure, while the viscosity, 𝝁𝝁, and flow density, v, depend only on temperature, so the dependence of the ∆ cab is expressed as: δ = const (Tm/2 /p1/2) 4−4 Meanwhile for the diffusion coefficient D is affected by the pressure and temperature according to the equation: q D = Di,o (pi/p)(T/To)1.75 4−5 Where Di,o is the value of reference of the diffusion coefficient, and pi is the partial pressure of the specie i. The value of the mass transfer coefficient follows the Arrhenius equation: Km=AexpEa/RT 4 −6 31 Where A is the factor preexponential and Ea is the activation energy. The surface reaction increases with decreasing temperature, in the case of the δ/D, the rise in value with decreasing temperature and increasing pressure. This increase with the decreasing of the temperature is more significant than for the case of the surface reaction. 5 Materials and methods 5.1 Research objectives 1. Study the impact of different parameters in the deposition on FTO in AACVD processes. 2. Synthesis of BIOI and BIOBr using the AACVD process and their transformation in BIVO4 to study different properties and viability in PEC solar-driven water splitting systems. 3. Synthesis of BIVO4 thin film on FTO in one-step AACVD process to study different properties and viability in PEC solar-driven water splitting systems. 5.2 AACVD set up Figure 5.1 Schematic of the AACVD setup used during the experiments. The aerosol droplets were generated using a TSI Model 3076 Constant Output Atomizer using different solutions and air or nitrogen as carrier gases at different flow rates. FTO-ABS glasses placed inside a tube furnace were chosen to be the film depositing substrate due to its good conductivity and wide range of use. The tube furnace used during the experiments was a ZS-3G Carbolite tube furnace with 3 heaters separated with unheated zone barriers and a quartz tube. Different parameters and conditions were performed to assess the growth mechanism and 32 optimisation. A solvent trap was situated at the end of the line to collect the byproducts generated. Figure 5.2 Image of the ZS-3G Carbolite tube furnace and a quartz tube. 5.3 Materials Chemicals used in this study have been listed in Table 5.1 Table 5.1 Table 3.1: List of chemicals used in this study. Chemicals Supplier Purity Bismuth (III) bromide Sigma-Aldrich 99.998% Bismuth (III) iodide Sigma-Aldrich 99% N, N-Dimethylformamide (DMF) Sigma-Aldrich 99.80% Bismuth (III) nitrate pentahydrate Fisher scientific 99.999% Vanadyl acetylacetonate Sigma-Aldrich 98% Acetone Fisher scientific >95% Isopropyl alcohol Sigma-Aldrich >95% Hellmanex III solution Hellma - Fluorine doped tin oxide coated glass slide (FTO) Sigma-Aldrich 100% Nitrogen Zero Grade BOC Ltd 99.998% Air Zero Grade BOC Ltd - 33 5.4 Methodology 5.4.1 Synthesis of BIVO4 in two steps process First, BiOBr/BiOI were deposited onto FTO substrates using an AACVD process. Depositions were carried out under air. BiBr3/BiI3 was dissolved in N, N- dimethylformamide (DMF). The resulting solution was stirred for 30 min to dissolve BiBr3/BiI3 and then atomised. Aerosol droplets were generated using a TSI Model 3076 Constant Output Atomizer and air as carrier gas. The TSI Model 3076 Constant Output Atomizer was used in the recirculation mode, and larger droplets and condensed precursors were recycled in the precursor solution bubbler. Different times, temperatures, concentrations and flow rates of precursor were carried out to study the impact of these variables in the deposition and characteristics of the final film. At the end of the deposition time, the substrates were left to cool down under the carrier gas until 100ºC. The ZS-3G Carbolite tube furnace is divided into three 15 mm heated zones with two 75 mm unheated zones barriers. Each heated zone has its temperature controller and thermocouple, so different configurations were used in the experiments to study the impact in the film deposition. FTO substrates were cleaned by ultrasonication in a 2% aqueous Hellmanex III solution, deionised water, acetone, and isopropyl alcohol (each step for 10 min) dried with compressed air followed by an oxygen plasma treatment for 20 min to enhance surface energy and finally stored in deionised water. Different positions and configurations of the substrates in the tube were studied. Secondly, BiOBr/BiOI were transformed into BiVO4 using a process described by Tae Woo Kim and Kyoung-Shin Choi.42 Briefly, 0.15-0.2 mL of a dimethyl sulfoxide (DMSO) solution containing 0.2 M vanadyl acetylacetonate (VO(acac)2) was placed on the BiOI/BiOBr substrates. The substrates were annealed at 450 ºC with a ramping rate of 2 ºC/min for 2 h. Finally, the excess of V2O5 present in the BiOV4 was removed by soaking the substrates in 1 M NaOH solution with gentle stirring until all the V2O5 was eliminated. The final substrates were rinsed with deionised water, dried with compressed air and stored at room temperature. 5.4.2 Synthesis of BIVO4 in one AACVD step process Bismuth(III) nitrate pentahydrate and vanadyl acetylacetonate were dissolved in DMF. The resulting solution was stirred for 30 min to dissolve the solutes and then atomised. The droplets were generated using the same TSI Model 3076 Constant Output 34 Atomizer as the 2 steps process, and air or nitrogen as carrier gas was used. Different times, temperatures, concentrations and flow rates of precursor were carried out to study the impact of these variables in the deposition of the film. At the end of the deposition time, the substrates were left to cool down under the carrier gas until 100ºC. FTO was cleaned with the same process as the 2 steps process. The V2O5 present in the BiOV4 after the deposition was removed by soaking the substrates in 1 M NaOH solution with gentle stirring until all the V2O5 was eliminated. The final substrates were rinsed with deionised water, dried with compressed air and stored at room temperature. 5.4.3 Final experiments studied The final experiments performed were as follows: BiOBr two-step process It was only studied under the conditions of 100 ºC in the first heater, 300 ºC in the second and third heater and 3 hours of deposition (BiOBr). The transformation in BiVO4 was carried out as explained in section 1.4.1. BiOI two-step process Different times of deposition were studied with 100ºC in the first heater, 300 ºC in the second and third heater: 3 hours (BiOI 3h), 5 hours (BiOI 5h) and 7 hours (BiOI 7h). In addition, the influence of the temperature change in the process was also studied, for 5h of deposition a configuration of 100 ºC in the first heater, 350 ºC in the second and third heater (BiVO4 from BiOI 5h 350 ºC) and a configuration of 100 ºC in the first heater, 400 ºC in the second and third heater (Component 400ºC) were studied. Meanwhile, for 7h of deposition, a configuration of 100 ºC in the first heater, 350 ºC in the second and third heater (BiVO4 from BiOI 7h 350 ºC) was studied. The transformation in BiVO4 was carried out as explained in section 1.4.1. In both cases (BiOI/BiOBr), FTO samples were placed in the second heater. The deposition was not studied at a time of 7h and 400 ºC because at this temperature for the 5h process, a yellowish compound was formed instead of the BiOI (reddish). This yellowish compound is also present in the rest of the experiment, but it appeared in a residual form in the last FTO located in the tube. When the temperature was 35 increased to 350 ºC, the yellow part deposited increased in size in the last FTO, and at the temperature of 400 ºC, most of the deposition was of the yellow compound. Figure 5.3 image of the yellow compound generated at 400 ºC and image of the tube after the experiment BiVO4 in one-step process For the deposition of BiVO4 in one step, a temperature of 400 ºc was used in the 3 heaters. The deposition of the BiVO4 in the FTO occurred between the last insulator and the last heater. 5.5 Security and flammability limits of the solution AACVD involves atomising a precursor solution fine, sub-micrometre-sized aerosol droplets, which are delivered to a heated reaction zone. Although the process temperature is usually lower than in the CVD, it is crucial to avoid security problems due to the possible solution accumulation within the closed system. In general, if the carrier gas is an inert gas like nitrogen, the only issue to consider is air entrance from the outside through leaks in the connections. On the other hand, if the carrier gas is air, the limits of exposition of the mixture air-solution must be calculated to ensure the correct operation of the system. The solvent's autoignition temperature is another crucial factor since it marks the temperature limit of operating the process. However, it is essential to consider the percentage of solvent in the air that enters the tube furnace. It is generally shallow and allows this temperature to be increased as long as it is checked that there is no type of obstruction at the outlet system. Finally, check that the solutes and solvents used do not create more reactive mixtures or produce more dangerous by-products. DMF was used as a unique solvent in all the experiments and the solutes used do not create dangerous mixtures with it. The autoignition temperature of DMF is 440ºc; this temperature can be considered as the limit temperature in the experiments. 42 fragile material that broke very easily. The result of the experiment was favourable since the deposition was improved. Figure 5.11 Dimensions of the rectangular structure created in Auto-Cad. Another idea was to use another tube with a smaller diameter inside the big tube. The problem with this structure was controlling the temperature as the layer in the top of the small tube created a more significant temperature gradient. The experiments were better for the deposition of BiOBr, but for BiOI and BiOV4, similar results were achieved concerning the standard tube with the metal structure. This configuration was used in most experiments to synthesise BiVO4 in one step. Figure 5.12 Structure with the two tubes, comparative of the diameters. The final structure was another new metal piece with the length of the heaters with a flat top to perfectly support the FTO and the shape of the tube on the bottom. This configuration was used in the important experiments to synthesise BiOI in the twostep process. In this case, the temperature gradient is mainly concentrated in the heater producing a better deposition as the surface of the FTO is in complete contact with the metal piece. 43 Figure 5.13 Dimensions of the last metal structure and the position in the tube. 5.7.4 Physical Characterisation X-ray diffraction (XRD) Different XRD analyses were made to compare the distribution: Figure 5.14 XRD patterns of standard FTO, standard bismuth oxide bromide and bismuth oxide bromide deposited. 44 Figure 5.15 XRD patterns of standard FTO, standard bismuth oxide Iodide and bismuth oxide iodide deposited. Figure 5.16 XRD patterns of standard FTO, standard bismuth vanadate and bismuth vanadate deposited in one-step and two-step processes. The XRD patterns confirmed the presence of BiOI, BiOBr and BiVO4 in the two-step process and the one-step process. BiOI and BiOBr were phased pure and crystallised in the tetragonal matlockite structure. BiVO4 patterns confirm the monoclinic scheelite. The peaks of BiVO4 in the two-step process and the one-step process are very similar, ensuring that BiVO4 could be generated in different ways. 45 However, a complete analysis of the changes generated under different conditions was not carried out since the XRD was only used to confirm the presence of the other compounds. UV-Vis Figure 5.17 Absorbance of BiOBr and BiVO4 from BiOBr. Figure 5.18 Absorbance of BiOI in different times. The absorbance of BiOI at different times shows that the absorption edge increases when the time of deposition is longer. In addition, the absorption edge of all of the BiOI deposition at different times have an absorption edge with a bigger value than BiOBr. 46 Figure 5.19 Absorbance of BiVO4 from BiOI at different times. However, when the BiOI/BiOBr is transformed into BiVO4, the absorption edge is in the same position in all the cases. This could indicate that there is homogenisation after its transformation. Figure 5.20 Absorbance of BiOI in different temperatures and the formation of the component at 400 ºC. In the case of BiOI at different temperatures, the absorption edge is slightly displaced, so when the temperature increases, it could indicate that there is not much difference in the deposition in 5h or 7h. However, the component formed at 400 ºC presents a different absorption edge respect BiOI. 47 Figure 5.21 Absorbance of BiVO4 from BiOI in different temperatures. The compound formed at 400 ºC presents an absorption edge similar to the BiVO4. Although it is slightly displaced to the left with respect to the other two BiVO4 absorbance edges, it could indicate that it is BiVO4, but It can also be another compound that could be interesting to study. The production of this compound is possible due to secondary reactions that take place at higher temperatures. Again, after transformation, the produced BIVO4 present the same absorption edge. This indicates that the second stage has a more significant impact on absorption Figure 5.22 Absorbance of BiVO4 of the one-step process compared with Figure 5.20. In this case, the BiVO4 produced in a one-step process presents a different absorption edge compared with the BiVO4 made in a two-step process. 48 Scanning electron microscope (SEM) The BiOBr sample consists of clusters of crystallites contributing to the particles resembling nanoflowers. The clusters appear to be composed of smaller crystallites. However, when transformed into BiVO4, these structures lose shape and become more rounded. The structures leave a lot of gaps between them, creating porous, so it could be beneficial in the problem of electron recombination since there is an increase in the surface in contact with the electrolyte Figure 5.23 SEM micrographs at different magnifications of BiOBr. Figure 5.24 SEM micrographs at different magnifications of BiVO 4 from BiOBr. Figure 5.25 SEM micrographs at different magnifications of BIOI 3h. 49 In the case of BiOI, the structure consists of a nanoplatelet morphology. As the deposition time increases, these structures become more extensive, but the width decreases with the time becoming almost flat by the 7h process. For the case of 3h, the structures are practically joined rectangles, while when the deposition time is increased, they unite more and more until they form chains and there is no defined shape. Both for the case of BiOBr and BiOI, the electronic properties of these materials are anisotropic, and this faceted growth may help charge carrier mobility along with certain facets. Figure 5.26 SEM micrographs at different magnifications of BIOI 5h. Figure 5.27 SEM micrographs at different magnifications of BIOI 7h Figure 5.28 SEM micrographs at different magnifications of BiVO 4 from BiOI 3h. 50 When it becomes BiVO4, it can be seen that for the 3h deposition, there is still some separation in the rounded structures that are formed. However, as the deposition time increases, the forms become less and less appreciated. In the 5h deposition, some separation between the structures is still appreciated. In comparison, in the 7h deposition, the structures form a more homogeneous mass with almost no gaps and no porous. The morphology of the structures formed is a fundamental factor in improving semiconductors’ performance for PEC services. BiVO4 suffers from rapid photocarrier recombination as a result of short electron diffusion length (with only 10 nm) that affects its apparent quantum efficiency directly; photocurrent stability is not very good due to form a dissolution of V5+ ions in the solution and reactions in the surface of the photoelectrode such as O2 and H2O2 that act as centres where there is recombination and is a material that suffers for a slow hole transfer kinetic producing a poor surface water oxidation. Some studies reveal that increasing the contact surface could improve electron recombination and perform better. In this sense, the homogenisation of the structures with increasing time could be a negative factor because the number of pores is drastically reduced. Porosity gives the films a large effective surface area and an increased electrode/electrolyte interfacial area, reducing electron-hole recombination as the Figure 5.29 SEM micrographs at different magnifications of BiVO 4 from BiOI 5h. Figure 5.30 SEM micrographs at different magnifications of BiVO 4 from BiOI 7h. 51 photogenerated hole travels through less material before being collected at the electrode/electrolyte interface. In the one-step process, it is observed that there are areas of the surface of the FTO where there is no BiVO4 deposited. In addition, sphere-shaped structures are formed, but there is no good contact with each other or with the FTO. This is in accordance with the idea that its deposition was through homogeneous reactions, and the adhesion is due to this type of reaction, so it is not an adequate deposition. Thermogravimetric Analysis (TGA) The thermogravimetric analysis of BiBr3 BiI3 and the solution of the precursor for the one-step process were studied to see the effect in experiments: Figure 5.32 Thermogravimetric analysis (TGA) in Air of BiBr3 The BiBr3 has a decomposition temperature between 300 ºC and 400 ºC. Therefore, a higher temperature could favour the decomposition and deposition of a greater quantity of BiOBr. However, secondary reactions could affect the primary process in this range. There is a small decomposition zone between 25ºc and 300ºC, the temperature range where BiOBr deposition occurs in the experiments. Figure 5.31 SEM micrographs at different magnifications of BiVO 4 one-step process. 58 compared to the data in Figure 5.39 photocurrent potential curves of BiVO4 from BiOI at different times. Therefore, it reinforces the hypothesis that increasing the temperature improves the results obtained. Although the values obtained for the component deposition at 400 ºC are low, it is a material that presents photocatalytic activity. A complete study of the material in question and the improvement of the conditions in its deposition could considerably improve the values obtained in this study. Due to the bad adhesion between the FTO and the BiVO4 deposited in the one-step process, the values obtained from photocurrent potential curves are too low to be considered. Figure 5.42 Degradation of the samples of BiOI after PEC analysis. Something to consider is the stability of the BiOI samples in all the pec analyses since they rapidly degraded, forming a black film on the part in contact with the electrolyte. This may negatively influence the results obtained as it is likely that better results would be obtained if the BiOI samples were more stable. 6 Conclusions This work demonstrated the production of BiVO4, BiOI and BIOBr for photoelectrochemical devices. In addition, a compound (probably BiVO4) was generated at higher temperatures; it is a compound with a photocatalytic activity that could be investigated in more detail in the future. These photoanodes were grown by a two-step process or one-step process being the AACVD the primary process. Structures with different morphologies and characteristics were obtained. Secondly, a study of the impact of different parameters in the deposition on FTO in AACVD processes was carried out. Factors such as flow rate and temperature 59 significantly impact the process. Its effects were studied by modifying its values to improve the experiments. SEM micrographs showed nanoplatelet morphology for BiOI, nanoflower like particles were observed for the BiOBr film. However, when transformed into BiVO4, these structures lose shape and become more rounded. In the one-step process, sphereshaped structures are formed, but there is no good contact with each other or with the FTO. In the one-step synthesis process of BiVO4, in the first part of the tube, the principal reactions produce V2O4, but along the tube, other types of reactions appear with the increase of the temperature of the carrier gas. The deposition of V2O4 is principally heterogeneous; meanwhile, the deposition of BiVO4 is homogenous, producing a very bad adhesion with the FTO. In the Raman spectroscopy, all BiVO4, both in the two-step process and one-step process, show the peaks in the same positions. The most substantial peak at 827 cm-1 corresponds to antisymmetric stretching modes of the VO4 tetrahedra, while the shoulder peak at 718 cm-1 is attributed to the same symmetric stretching modes. The peaks at 366 and 326 cm-1 are due to the bending modes of the VO4 tetrahedra, and the peaks at 211 and 129 cm-1 correspond to the vibration of the crystal lattice (external modes). The component deposited at 400 ºC shows peaks similar to the rest of BiVO4 but with other stranger peaks. In addition, the images of the Raman spectroscopy show differences between the different experiments and the experiments performed at the exact times and temperatures. Photoelectrochemical measurements revealed that the best value was BiOI deposited for 5 hours and 350 ºC (1.1 mA cm-2 at 1.23V vs RHE). Although the initial objective was to obtain BiVO4 as photoanodes, both in the two-step process and in the one-step process, lower values were obtained with respect to the BiOI deposition. In the twostep process, when it was transformed into BiVO4, the maximum value reached in the BiOI before transformation into BiVO4, was reduced being the best value the deposition for 5h and 350 ºC (0.7 mA cm-2 at 1.23V vs RHE), while in the one-step process, poor adhesion with the FTO produced very low values. However, BiOI rapidly degraded, forming a black film on the part in contact with the electrolyte. 60 7 BIBLIOGRAPHY UNITED NATIONS. Sustainable Development Goals Report,2020. ACS chemistry for life. 12 Principles of Green Chemistry, 2021. BP, BP Statistical Review of World Energy, 2020. International Energy Agency, Global Energy Status Report, 2020. Royal Society of Chemistry, Solar Fuels and Artificial Photosynthesis, 2012. Furat Dawood, Martin Anda, G.M. Shafiullah, Hydrogen production for energy: An overview, International Journal of Hydrogen Energy, 2020, vol. 45, Issue 7, pp. 3847- 3869. El-Shafie, M., Kambara, S. and Hayakawa, Y., Hydrogen Production Technologies Overview. Journal of Power and Energy Engineering, 2019 vol. 7, pp. 107-154. R. J. Detz, J. N. H. Reek and B. C. C. Van Der Zwaan, Energy Environ. Sci., 2018, 11, 1653–1669. B. A. Pinaud, J. D. Benck, L. C. Seitz, A. J. Forman, Z. Chen, T. G. Deutsch, B. D. James, K. N. Baum, G. N. Baum, S. Ardo, H. Wang, E. Miller and T. F. Jaramillo, Energy Environ. Sci., 2013, 6, 1983–2002. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38. S. Giménez and J. Bisquert, Photoelectrochemical Solar Fuel Production: From Basic Principles to Advanced Devices, Springer International Publishing, 2016. T Bak, J Nowotny, M Rekas, C.C Sorrell, Photo-electrochemical hydrogen generation from water using solar energy. Materials-related aspects, International Journal of Hydrogen Energy, Volume 27, Issue 10, 2002, Pages 991-1022, ISSN 0360-3199. R. Van De Krol and M. Grätzel, Photoelectrochemical Hydrogen Production, Springer US, 2012. H. Ahmad, S. K. Kamarudin, L. J. Minggu and M. Kassim, Renew. Sustain. Energy Rev., 2015, 43, 599–610. S. Zhu and D. Wang, Adv. Energy Mater., 2017, 7, 1700841. 61 C. Jiang, S. J. A. Moniz, A. Wang, T. Zhang and J. Tang, Chem. Soc. Rev., 2017, 46, 4645– 4660. M. Grätzel, Nature, 2001, 414, 338–344. K. Maeda and K. Domen, J. Phys. Chem. C, 2007, 111, 7851–7861. H.-J. Lewerenz and L. Peter, Eds., Photoelectrochemical Water Splitting: Materials, Processes and Architectures, The Royal Society of Chemistry, 2014. Z. Chen, H. N. Dinh and E. Miller, Photoelectrochemical water splitting: standards, experimental methods, and protocols, Springer, New York, 2013. A. G. Tamirat, J. Rick, A. A. Dubale, W.-N. Su and B.-J. Hwang, Nanoscale Horiz., 2016, 1, 243–267. J. Li and N. Wu, Catal. Sci. Technol., 2015, 5, 1360–1384. A. Kudo, K. Ueda, H. Kato and I. Mikami, Catal. Letters, 1998, 53, 229–230. S. J. A. Moniz, S. A. Shevlin, D. J. Martin, Z.-X. Guo and J. Tang, Energy Environ. Sci., 2015, 8, 731–759. Rayees Ahmad Rather, Akansha Mehta, Youming Lu, Matjaz Valant, Ming-Fang, Wenjun Liu, Influence of exposed facets, morphology and hetero-interfaces of BiVO4 on photocatalytic water oxidation: A review, International Journal of Hydrogen Energy, Volume 46, Issue 42, 2021, Pages 21866-21888, ISSN 0360-3199. Y. Park, K. J. McDonald and K.-S. Choi, Chem. Soc. Rev., 2013, 42, 2321–2337. Trinh, D.T.T., Khanitchaidecha, W., Channei, D. et al. Synthesis, characterization, and environmental applications of bismuth vanadate. Res Chem Intermed 45, 5217–5259 (2019). Aron Walsh, Yanfa Yan, Muhammad N. Huda, Mowafak M. Al-Jassim, and Su-Huai Wei Chemistry of Materials 2009 21 (3), 547-551. Kudo, A.; Omori, K.; Kato, H. A Novel Aqueous Process for Preparation of Crystal Form- Controlled and Highly Crystalline BiVO4 Powder from Layered Vanadates at Room Temperature and Its Photocatalytic and Photophysical Properties. J. Am. Chem. Soc. 1999, 121, 11459−11467. Z. Zhao, Z. Li and Z. Zou, Phys. Chem. Chem. Phys., 2011, 13, 4746–4753. 62 D.J. Payne, M.D.M. Robinson, R.G. Egdell, A. Walsh, J. McNulty, K.E. Smith, et al., Appl. Phys. Lett. 98 (2011), 212110. Carlsson J-O, Martin PM. Chemical vapour deposition. Handbook of Deposition Technologies for Films and Coatings (Third Edition): Elsevier; 2010. p. 314–63. K. L. Choy, Prog. Mater. Sci., 2003, 48, 57–170. P. Marchand, I. A. Hassan, I. P. Parkin and C. J. Carmalt, Dalt. Trans., 2013, 42, 9406– 9422. A. C. Jones and M. L. Hitchman, Chemical Vapour Deposition: Precursors, Processes and Applications, The Royal Society of Chemistry, 2009. M. D. Allendorf and A. M. B. van Mol, Precursor Chem. Adv. Mater., 2005, 9, 1–48. Pierson HO. Handbook of chemical vapor deposition. Park Ridge (NJ): Noyes; 1992. Dobkin DM, Zuraw MK. Principles of chemical vapourdeposition. Dordrecht, The Netherlands: Kluwer AcademicPress; 2003. A. C. Jones and M. L. Hitchman, Chemical Vapour Deposition: Precursors, Processes and Applications, The Royal Society of Chemistry, 2009. X. Hou and K.-L. Choy, Chem. Vap. Depos., 2006, 12, 583–596. T. W. Kim and K.-S. Choi, Science, 2014, 343, 990–994 Salvador Eslava, PhD Senior Lecturer (Assoc. Prof.) Department of Chemical Engineering Imperial College London London, SW7 2AZ, United Kingdom Phone: +44(0)20 7594 8977 Email: [email protected] London, 9/1/2022 To Whom It May Concern, I, Salvador Eslava, hereby declare that the thesis titled Bismuth vanadate, bismuth oxyiodide and bismuth oxybromide thin films growth via Aerosol- Assisted CVD as the main process and their photoelectrochemical properties reports the research carried out and written by Javier Vicente Matas under my supervision at Imperial College of London. Dr Salvador Eslava Senior Lecturer (Assoc. Prof.) in Chemical Engineering and EPSRC Early Career Fellow Salvador Eslava Digitally signed by Salvador Eslava Date: 2022.01.09 13:21:45 Z