Etude bibliographique
Les surfaces d’oxydes de métaux de transition
Transition metal oxides represent a vast family of materials with highly diverse properties For instance, in terms of electronic characteristics, NiO, CoO, and FeO act as insulators, while TiO2, ZnO, and SnO2 are classified as semiconductors, and YBaCuO is recognized as a superconductor These materials find applications in various fields, including magnetic detectors, transparent transistors, anti-corrosive coatings, and photocatalysis, particularly with TiO2 Additionally, transition metal oxide-based catalysts exhibit significant selectivity in photocatalytic degradation processes, such as the CoO-CuO-MnO2 catalyzed oxidation of CO.
The properties of materials are influenced by their composition, crystalline structure, and notably, the nature of their surface Surface atoms and ions are often under-coordinated, leading to relaxation in atomic positions across several layers The density and variety of surface defects directly affect surface properties, including reactivity and electronic structure For instance, oxygen vacancies on the surface of TiO2 impact the bandgap width Consequently, the physical and chemical properties of oxides are significantly determined by their surface characteristics and interfaces Advances in techniques like scanning tunneling microscopy (STM) and X-ray photoelectron spectroscopy have enabled extensive theoretical and experimental studies on the surface properties of metal oxides, particularly transition metal oxides.
The crystalline surface of a transition metal oxide is defined by its physicochemical properties and geometry Crystal surfaces are inherently uneven and exhibit various types of surface defects.
The actual surface of a crystal is made up of parallel atomic planes, known as terraces, which are separated by steps caused by atomic dislocations per square centimeter In metallic or ionic crystals, the density of these defects is typically on the order of
The defect density in crystalline networks can vary significantly, with values ranging from 10^6 to 10^8 cm^-2, compared to approximately 10^4 to 10^6 cm^-2 for insulators or semiconductors These defects may manifest as two-dimensional flaws, such as stacking faults or twin boundaries, which can lead to the development of linear defects on the surface.
Figure I- 1 : La surface réelle d’un cristal
Surface defects can include steps measuring between 0.5 to 1 nm, vacancies, adatoms, and point defects such as substitutions or insertions, where ions or atoms exhibit lower coordination than those on the terraces In the case of TiO2 (110), cations on the terraces have a coordination number of 5, while those at the steps have a reduced coordination of 4 Additionally, notches can be observed, with their structure influenced by their position within the lattice and the orientation of the steps The coordination of atoms at the outer corners of these notches is even lower, creating highly unsaturated coordination sites.
When transition metal oxides undergo thermal treatment or are exposed to radiation, oxygen vacancies can form on their surface These vacancies are found on nearly all oxides and alter their properties significantly.
Vacance coordinences et l’état d’oxydation des ions adjacents, conduisant à des changements significatifs de la structure ộlectronique et influenỗant fortement la rộactivitộ de la surface.
Le dioxyde de titane
Titanium was discovered in 1791 by William Gregor in England while studying ilmenite (FeTiO3) from Helford River sands Four years later, Martin H Klaproth identified the element in rutile and named it titanium Pure titanium cannot be extracted from these minerals and exists only in compounds like titanium dioxide (TiO2) This oxide is relatively abundant, inexpensive, and low in toxicity, making it a well-known white pigment in paints, papers, and plastics, as well as an excipient in medications Recently, titanium dioxide has gained renewed interest due to its properties in the field of photocatalysis.
II.1 Structures cristallines du TiO 2
Titanium dioxide (TiO2) crystallizes in eleven polymorphs, with seven stable at room temperature and pressure: rutile, anatase, brookite, hollandite (TiO2(H)), TiO2(B), ramsdellite (TiO2(R)), and columbite isotype α-PbO2 (TiO2-II, columbite) The other four polymorphs are stable only under high pressure In all TiO2 varieties, titanium is in octahedral coordination (TiO6), and the structure varies based on the arrangement of these octahedra, which can be connected by edges and/or vertices Among these, rutile and anatase are the most common crystalline phases and are widely used in photocatalytic applications.
The structure of anatase is characterized by a three-dimensional arrangement of TiO6 octahedra linked at the edges, forming a zig-zag chain The connection of these chains through shared vertices creates sheets in the (001) plane, resulting in a three-dimensional network The crystallographic parameters of anatase are summarized in Table I-1.
Figure I- 2 : Reprộsentation de la structure cristalline de l’anatase : (a) chaợne d’octaốdres TiO 6 en zigzag selon l’axe c ; (b) association suivant l’axe b de deux chaợnes en zigzag ; (c) développement d’un réseau 3D d’octaèdres TiO 6
The structure of rutile is characterized by a compact hexagonal arrangement of oxygen atoms, with titanium occupying half of the octahedral sites Each TiO6 octahedron shares two opposite edges with adjacent octahedra, creating infinite chains along the c-axis Additionally, each chain is connected to four neighboring chains through the sharing of vertices.
The transformation of anatase to rutile occurs at a temperature of 823 K, with rutile being the more thermodynamically stable structure The parameters of the rutile lattice are detailed in Table I-1, and the presence of square-section tunnels is illustrated in Figure I-3b.
Figure I- 3 : Reprộsentation de la structure cristalline du rutile : (a) chaợne d’octaốdres TiO6 selon l’axe c ; (b) développement d’un réseau 3D d’octaèdres TiO6
Système cristallin Tétragonal Tétragonal Groupe d’espace D 4h
Nombre de motifs par maille Z 4 2
Distance Ti-Ti (Ǻ) 3,790 et 3,040 3,570 et 2,960
Tableau I- 1 : Paramètres cristallins des deux phases cristallographiques anatase et rutile [32, 33, 34]
Le dioxyde de titane est un semi-conducteur A la différence d’un conducteur métallique qui possède un continuum d’états électroniques, les semi-conducteurs comme TiO2 a c b
(b) valence et la bande de conduction La bande de valence est composée majoritairement des orbitales 2p des atomes d’oxygène et la bande de conduction des orbitales 3d des atomes de titane
Rutile has a bandgap of 3.0 eV, while anatase has a bandgap of 3.2 eV When energy is supplied (through temperature increase, magnetic fields, or light absorption) that meets or exceeds the bandgap, some valence electrons can be promoted to the conduction band The wavelengths of photons that rutile and anatase can absorb are 400 nm and 380 nm, respectively, falling within the ultraviolet radiation spectrum The vacancy left in the valence band is referred to as an electronic vacancy or hole, leading to the formation of an electron-hole pair If the conduction electron returns to its original position in the valence band, it is known as direct electron-hole recombination.
II.3 Les surfaces de TiO 2
Titanium dioxide (TiO2) is extensively utilized in photocatalysis, where surface properties play a crucial role Therefore, a comprehensive understanding of the TiO2 surface is essential for grasping the phenomena involved in reactions occurring at this surface.
II.3.1 Surface de TiO 2 rutile
Among the three crystalline forms of TiO2, rutile is the most thermodynamically stable phase Rutile TiO2 powder predominantly exhibits three crystallographic orientations: (110) at 60%, and (100) and (101) at 20% each Although the (001) and (111) faces are less common, they have also been experimentally observed, and oriented single crystals can be stabilized The (110) surface is the most stable, while the (001) surface has the highest energy.
The surface of rutile (110) can be characterized by a 1×1 unit cell with dimensions of 2×c,54Ų This unit cell features titanium atoms in both fivefold and sixfold coordination.
Ti(5) and Ti(6) are arranged along the [001] direction to form two distinct rows, separated by a row of surface oxygen atoms (coordination 3), referred to as O S, which are bonded to two Ti(5) atoms and one Ti(6) atom Additionally, there is another type of undercoordinated oxygen atom, denoted as O b, which acts as a bridging atom between two Ti(6) atoms.
Figure I- 4 : Représentation de la surface (110) du rutile
Studies using atomic force microscopy (AFM) and scanning tunneling microscopy (STM) reveal that oxygen vacancies are present at a rate of 7 ± 3% per unit area These vacancies can diffuse within the crystal lattice, leading to defects such as volume oxygen vacancies or the presence of titanium atoms in interstitial sites Additionally, the color of the crystal can range from light blue to black depending on the concentration of these vacancies.
Figure I-5 illustrates the stoichiometric surface of TiO2 (100), characterized by a wavy appearance due to bridging oxygen atoms positioned above the main surface plane The titanium cations are in a coordination state of five, denoted as Ti(5).
Figure I- 5 : Représentation de la surface (100) du rutile [43]
II.3.2 Surface de TiO 2 anatase
The surfaces of anatase are significantly less studied compared to those of rutile Existing literature indicates theoretical studies that suggest greater stability for the (101) surface Additionally, Hebenstreit utilized tunneling microscopy to describe the geometric structure of the surface.
The structure consists of terraces made up of titanium atoms with five and six coordination, along with titanium atoms with four coordination at the edges There are two types of oxygen present: one with two coordination and the other with three coordination.
The (001) surface of anatase is stable and nonpolar, featuring titanium atoms coordinated with five oxygen atoms To date, there has been limited research published on this surface.
II.4 Adsorption sur les surfaces de TiO 2
La photocatalyse hétérogène
With a bandgap value of 3.0 – 3.2 eV, titanium dioxide (TiO2) is the leading material for photocatalytic applications Since the mid-1990s, Fujishima's team has explored the photocatalytic properties of TiO2 for building applications, particularly in the development of self-cleaning glass, pollution-reducing materials, and anti-fog solutions Photocatalysis is also crucial for air treatment and has gained recent attention from the scientific community for improving indoor air quality Additionally, the use of photocatalysis for domestic water treatment and industrial effluent management is a rapidly growing field in North America, exemplified by the work of Purifics, a Canadian company.
Historically, TiO2 powder has been utilized as a white pigment, but its light-sensitive properties have led to issues such as paint peeling and canvas degradation Scientific studies on the photoactivity of this pigment have been conducted since the early 20th century In 1929, Kiedel discovered that certain additives in TiO2-based paints experienced degradation Additionally, it has been reported that UV absorption generates reactive oxygen species, causing photobleaching of dyes without degrading TiO2.
There are no specific references regarding the early research on the use of TiO2 for photocatalytic applications, although initial studies began in the 1950s A significant comparison of the photocatalytic activities of the two allotropes, TiO2 anatase and rutile, was conducted by Mashio et al., highlighting the superior self-oxidation activity of anatase.
The oil crisis of 1960 highlighted the urgent need for alternative energy sources The photocatalytic electrolysis of water into hydrogen and oxygen using TiO2, demonstrated by Honda and Fujishima in 1972, was pivotal in advancing research on the photocatalytic properties of TiO2 This led to extensive studies on the use of TiO2 for the photocatalytic degradation and mineralization of organic pollutants In 1977, Frank and Bard reported the photocatalytic degradation of cyanide using TiO2 powder in aqueous solutions By the 1980s, the potential of TiO2 for detoxifying water and air was established, particularly in the degradation of various pesticides and organophosphates.
III.1 Principe de la photocatalyse
Photocatalysis is an advanced oxidation process that accelerates chemical reactions through electronically excited species generated by photon absorption This technique enhances the speed of chemical reactions in the presence of light radiation and catalysts known as photocatalysts.
Various semiconductor materials can serve as photocatalysts, including TiO2, ZnO2, and chalcogenides like CdS and ZnS Table I-2 summarizes the bandgap (Eg) values and excitation wavelengths (λ) of key semiconductor solids Chalcogenides such as CdS and CdSe are notable for their low bandgap values but are prone to corrosion under irradiation Similarly, oxides like ZnO and α-Fe2O3 (hematite) are unsuitable for photocatalysis due to degradation during photocatalytic cycles.
Titanium dioxide (TiO2) is recognized as the most suitable semiconductor for photocatalysis, particularly in pollution remediation It offers several advantages over other photocatalysts: it is biologically, chemically, and photolytically inert, cost-effective, and capable of efficiently utilizing a portion of sunlight.
Tableau I- 2 : Valeurs des gaps et des longueurs d’onde d’excitation pour différents matériaux semi-conducteurs
The photocatalysis of TiO2 involves a five-step electronic process First, the adsorption of organic molecules occurs on the surface of light-activated particles, which is essential for photocatalytic degradation, unlike typical photochemical reactions Next, when the TiO2 semiconductor is exposed to photon radiation with energy equal to or greater than its band gap, an electron transitions from the highest occupied molecular orbital (HOMO) in the valence band to the lowest unoccupied molecular orbital (LUMO) in the conduction band, creating a hole (h+).
TiO 2 → h ν e - BC + h + BV (1) c Séparation et transfert des charges photogénérées vers la surface : la durée de vie des espèces chargées (e - , h + ) est de quelques nanosecondes [77] Leur recombinaison peut avoir lieu à l’intérieur du volume ou à la surface de TiO 2 et s’accompagne d’un dégagement de chaleur Elle peut être évitée par le transfert, le piégeage, la séparation des charges vers des niveaux d’énergies intermédiaires (défauts de structure, de surface…) ou grâce à la présence de photosensibilisateurs d Réaction d’oxydation et de réduction : dès que les charges migrent à la surface du catalyseur, elles peuvent réagir avec les substances adsorbées susceptibles d’accepter ou de donner des électrons à condition que ces substances possèdent un potentiel redox thermodynamiquement compatible avec les niveaux énergétiques des bandes de valence et conduction (Figure I- 6) Ceci conduit à des réactions d’oxydation et de réduction
En solution aqueuse (aq), les électrons peuvent réagir avec l’oxygène dissous pour former les radicaux superoxydes O2 ●-
(2) Il peut s’en suivre la formation d’un radical hydroperoxyde puis de peroxyde d’hydrogène (3), qui va ensuite être décomposé à la surface de TiO 2 en radical hydroxyle OH ● (4) e - BC + O 2, ads → O 2 ●- aq (2)
Figure I- 6 : Schéma des potentiels pour de différents processus redox à la surface de
TiO 2 (SHE : électrode standard d’hydrogène)[9]
Les trous vont être piégés par les donneurs d’électrons tels que H 2 O, OH - pour former les radicaux hydroxyles OH ● (5 et 6) Ils peuvent également directement oxyder les molécules organiques R en radicaux R ● (7)
H 2 O ads + h + BV → H + aq + OH ● ads (5)
OH - ads + h + BV → OH ● ads (6)
Hydroxyl radicals (OH●) are highly reactive and can oxidize organic compounds adsorbed on surfaces, leading to their complete mineralization into CO2 and H2O This process also involves the desorption of the reaction products, namely CO2 and H2O.
Toutes les étapes peuvent être résumées dans la Figure I- 7 h + h + e - e -
OH ● /H2O(+2,27) Potentiel redox (V) (/SHE) pH = 7
Figure I- 7 : Schéma des étapes de la photocatalyse sur TiO 2
Hypothèse du modèle cinétique de la photocatalyse sur TiO2
In general, the photodegradation rate of low concentrations of pollutants in water adheres to the Langmuir-Hinshelwood kinetic law According to this model, photocatalytic reactions occur on the surface of the catalyst through two mechanisms: the pollutant and water may compete for adsorption on the catalyst's active sites.
Dans le cas I-2, la vitesse de dégradation est proportionnelle à la fraction de surface recouverte par le polluant s s r x r KC K C
The article discusses key variables in a chemical reaction involving O2, including v, the degradation rate; θx, the fraction of the surface covered; kr, the reaction rate constant; K, the solvent adsorption constant; C, the concentration at time t; Ks, the solvent adsorption constant; and Cs, the solvent concentration.
Puisque C s >> C, on peut considérer que la concentration du solvant est constante Alors, les deux expressions précédentes peuvent être intégrées pour obtenir :
Ln(C o )+ ( o − )= r (I-2) (11) avec C0, la concentration initiale
Si la concentration du polluant initiale Co est très faible, on peut simplifier les équations
Ln(C o )= ' (12) avec k’, la constante de vitesse apparente
According to this expression, if the degradation follows the Langmuir-Hinshelwood model, the variation of Ln(C₀/C) will be proportional to the reaction time, which is characteristic of a pseudo-first-order reaction This allows us to determine the apparent rate constant k' for the reactions.
III.3 Facteurs affectant l’activité photocatalytique
The photocatalytic activity of TiO2 is primarily influenced by its crystalline structure Sclafani demonstrated that anatase exhibits higher activity than rutile in the treatment of aqueous and gaseous effluents through pollutant degradation This enhanced activity is attributed to greater oxygen adsorption and hydroxylation on the surface of anatase Additionally, anatase features a wider band gap of 3.20 eV (387.8 nm) compared to rutile's 3.0 eV (413.6 nm), which further contributes to its superior photocatalytic performance.
II.3.1 Influence de la taille de particule
The particle size of materials significantly influences photocatalytic activity, which increases with a reduction in particle size due to enhanced specific surface area However, excessively small particle sizes can lead to a higher recombination density of photogenerated species There is no linear relationship between photocatalytic efficiency and particle size; nonetheless, the photocatalytic activity of TiO2 is optimal for sizes ranging from 3.8 to 40 nm, depending on experimental conditions such as the organic compounds being treated and the light source used.
II.3.2 Influence de la température
Techniques de caractérisation
La diffraction des rayons X sur poudre
X-ray diffraction (XRD) was utilized in this study to determine the crystallographic structure and phase composition of the TiO2 precursors, including anatase, rutile, and P25 Additionally, the evolution of the catalysts' structure was monitored in relation to fluorination and thermal shock temperature.
X-ray powder diffraction is a non-destructive method for the structural characterization of crystalline materials, applicable only to crystalline powders with periodic atomic arrangements This technique involves the acquisition of diffracted X-rays from a sample when exposed to a monochromatic X-ray beam with a wavelength ranging from 0.5 to 2 Å According to Bragg's law, diffraction occurs when the angle θ between the incident X-rays and the crystal planes satisfies the equation: 2d sin θ = nλ, where d is the distance between the Miller indices (h,k,l) planes in Å, θ is the Bragg angle in degrees, n is the order of diffraction (an integer, with n = 1), and λ is the wavelength of the incident X-ray in Å.
X-ray diffraction patterns of materials were recorded using a Debye-Scherrer type INEL CPS 120 diffractometer, equipped with a copper anticathode and a germanium (111) monochromator, generating K α1 radiation with a wavelength of λ = 1.5406 Å The recording conditions were specified as follows:
The refinement of the structure using the Rietveld method was conducted in this study with the Fullprof software This software requires a data file from the experimental diffractogram and a command file where initial values, the number of parameters to refine, and constraints are inputted to calculate and adjust structural data It subsequently generates a results file containing the correlation factors from the calculations and additional information.
To ensure the convergence of software calculations, it is essential to first refine the most influential parameters affecting the diffractogram shape, including scale, background noise, and then gradually adjust less critical structural parameters such as asymmetry and thermal factors The sequence for parameter refinement is as follows: i Scale factor ii Background noise parameters iii Sample angular offset iv Lattice parameter v Peak width parameter W vi Atomic positions x, y, z vii Pseudo-Voigt profile parameter a viii Peak width parameters U and V ix Pseudo-Voigt profile parameter b x Atom occupancy rates in crystallographic sites xi Peak asymmetry and preferred orientation xii Thermal agitation factors (isotropic or anisotropic).
La spectroscopie photoélectronique à rayonnement X
To better understand the reactivity and reaction mechanisms involved in catalytic reactions, it is essential to establish relationships between the structure, composition, and nature of active sites In liquid or gaseous reaction environments, the activity of heterogeneous catalysts primarily depends on their surface properties.
In our study, we characterized the surface of our samples using X-ray photoelectron spectroscopy This technique is a preferred method for surface analysis of materials, providing a wealth of information.
La composition élémentaire de surface
L’environnement chimique des atomes et la structure électronique des solides
X-ray photoelectron spectroscopy (XPS) is a surface analysis technique that identifies all elements in a material, excluding hydrogen and helium, within a depth of 5 to 10 nm This method analyzes the kinetic energy of photoelectrons emitted when the sample's surface is irradiated with low-energy monochromatic X-rays By applying the law of conservation of energy, the binding energy of the photoelectron can be determined using the relationship: hν + E_k_initial(N) = E_k_final(N-1) + E_k_kinetic, leading to the formula E_k_binding = E_k_final(N-1) - E_k_initial(N), or E_k_binding = hν - E_k_kinetic, where E_k_binding represents the binding energy of the photoelectron.
E cinétique k est l’énergie cinétique du photoélectron k hν est l’énergie du photon incident (Al Kα = 1486,6 eV)
E initial k (N) et E final k (N-1) sont les énergies totales du système à l’état fondamental initial (neutre à N électrons) et à l’état final (ionisé à N-1 électrons)
Le niveau de Fermi est utilisé comme référence
After photoexcitation, the ionized system can return to its ground state through two processes: Auger effect and X-ray fluorescence In the Auger effect, an electron from a higher energy orbital fills the vacancy left by the photoelectron, transferring the released energy to another outer electron, which is then ejected from the ion This process is common in light atoms (Z < 20) In contrast, X-ray fluorescence occurs when an electron from a higher orbital fills the vacancy left by the photoelectron, resulting in the emission of X-rays, which is more frequently observed in heavy atoms.
On distingue généralement les potentiels d’ionisation de valence (≈ 0 à 30 eV) et de coeur (≈
Figure II- 1 : Exemple de spectre XPS obtenu sur un monocristal MgO avec une source d'Al
The significance of determining binding energies associated with core peaks lies in their sensitivity to the atomic environment Variations in binding energy or chemical shifts can be interpreted based on properties related to the initial state and are primarily influenced by the charge of the considered atom Therefore, high-resolution analysis of core ionization potentials, which is the main application of this technique, provides insights into the local atomic environments, electronic transfers, and oxidation states.
Il convient toutefois de souligner qu’il existe deux ô sondes ằ complộmentaires dont l’exploitation bien que moins courante n’en est pas moins riche d’informations :
- les pics satellites (pics secondaires d’ionisation) : ils se révèlent dans certains cas mieux adaptés que les déplacements chimiques pour différencier des degrés d’oxydation,
- les bandes de valence (visualisation expérimentale des densités d’états du matériau) : leur analyse requiert fréquemment le support de modélisations théoriques
Par ailleurs, l’exploitation des résultats nécessite un traitement des spectres qui s’effectue sur la base de différents critères déduits de l’analyse de composés de référence
It is essential to recognize that discussions are typically based on oxidation states represented by whole formal charges, despite the fact that many subtle nuances exist that are challenging to quantify.
II.2 Appareillage et conditions d’analyse
XPS analyses of TiO2 samples, both fluorinated and non-fluorinated, in single-phase and bi-phase forms were conducted using a Thermo K Alpha spectrometer This spectrometer is integrated with a glove box equipped with a powerful regulation system for controlling oxygen levels.
The system operates with water levels below 10 ppm and utilizes an immersion lens to enhance the solid angle for photoelectron collection It employs a monochromatic Kα line of aluminum (1486.6 eV) focused to a spot size of 200 x 200 μm², functioning at 450 W under a chamber pressure of approximately 10^-8 mBar.
Figure II- 2 : Spectromètre Thermo K-alpha
The spectrometer was calibrated using the photoemission line of gold Ag 3d 5/2 (binding energy of 368.3 eV, full width at half maximum of 0.61 eV) as a reference The hemispherical analyzer operates at a pass energy of 20 eV for high-resolution spectra and 200 eV for low-resolution spectra All binding energies were calibrated according to the C 1s peak at 285 eV due to carbon contamination.
La nanosonde Auger
The morphological characterization of the catalysts synthesized in this study was conducted using a JEOL JAMP 9500F Auger nanosonde This method is based on detecting secondary electrons emitted from the surface after interaction with a scanning electron beam It enables the acquisition of topographic contrast micrographs, allowing for the visualization of surface morphology and particle size of solid materials at the microscopic or sub-microscopic scale.
The images were captured using the JEOL JAMP 9500F Auger Microprobe (Figure II-3) Samples are ground and placed in grooves on a graphite sample holder, which is essential to prevent charging effects when working with poorly conductive or insulating solids The recording conditions are detailed in the following table.
Paramètre de travail du microscope électronique en balayage – JAMP 9500F
Type d’échantillon à analyser Poudre broyée
Tableau II- 1 : Conditions d’enregistrement des micrographies
Figure II- 3 : Microsonde Auger JEOL JAMP 9500F
La mesure de la surface spécifique et porosimétrie
The adsorption of an inert gas onto the surface of a powdered sample is a technique used to characterize the sample by measuring various parameters, including specific surface area, pore size distribution, pore shape, and mesopore volume This method relies on the analysis of the adsorption isotherms of an inert gas on the sample's surface, following the physical adsorption model of multiple gas monolayers at low temperatures.
The specific surface area of TiO2 powders and fluorinated samples was determined using adsorption isotherms measured with an ASAP 2010 Micropore device (Micromeritics) Prior to measurement, the samples were ground and placed in a measurement cell, then heated to 110°C for 24 hours to remove residual gases and adsorbed water from their surfaces After cooling to 77.2 K (the boiling point of N2), the powders underwent adsorption analysis Additionally, the dead volume, or the volume of the measurement cell not occupied by the sample itself, was also determined.
The adsorption isotherm is determined using a volumetric method, where a specific amount of inert gas, typically nitrogen (N2), is introduced to a known quantity of sample The adsorption temperature is set to the boiling point of nitrogen, and both the pressure (P) before and at equilibrium are measured to calculate the corresponding amount of gas adsorbed (V) by the sample The isotherm is then established by plotting the volume of gas adsorbed (V) against the relative pressures (P/P0), where P0 represents the saturated vapor pressure of the adsorbed gas.
Techniques spectrophotométriques
The principle of spectrophotometry involves measuring the amount of light absorbed by a sample as a function of wavelength The experimental setup includes a monochromatic light source achieved either through a dispersive system (prism) or a diffractive system (grating) The light beam is split, with one component passing through the sample and the other serving as a reference A photomultiplier records the light intensity, and by varying the wavelength over an appropriate range, an electronic spectrum can be obtained.
UV-Visible absorption spectroscopy in transmission allows for the direct measurement of species concentration by assessing the amount of light absorbed, following Beer-Lambert's law This technique is essential for the quantitative analysis of transition metal ions, organic compounds, and biological macromolecules in solution that can absorb photons within the wavelength range of 200 to 800 nm It will be utilized to monitor the concentration of methylene blue over time during photocatalytic degradation The principle relies on Beer-Lambert's law, which states that when radiation of intensity Io passes through a solution, some intensity of the light is absorbed by solutes, resulting in a transmitted light intensity I that is less than Io Absorbance is defined by the following expression:
A= o avec A : absorbance de la solution à analyser
The Beer-Lambert law states that the absorbance of a solution is directly proportional to the concentration of its solutes and the optical path length through which the radiation passes.
A λ =ε λ avec Aλ : l’absorbance de la solution à la longueur d’onde donnée λ
C : la concentration de la solution, en mol.m 3
L : la longueur de trajet optique, en m ε λ : le coefficient d’extinction molaire de l’espèce absorbante en solution à la longueur d’onde λ, en mol -1 m 2
The absorbance of methylene blue solutions was measured after equilibrium and photocatalytic tests using a Thermo Scientific Helios Omega spectrophotometer with an optical fiber and a tungsten lamp The absorbance measurements were taken at a wavelength of 664 nm, which corresponds to the maximum absorbance of methylene blue.
Figure II- 4 : Spectre d’absorption UV-Visible du BM
We validated the Beer-Lambert law for our study by measuring the absorbance of a series of known concentrations of BM in aqueous solution The data obtained is presented in Table II-12, confirming that the Beer-Lambert law holds true across the entire concentration range examined (Figure II-5).
Volume de solution BM 10 -4 mol.L -1 (ml) 0,00 1,00 2,00 3,00 4,00 5,00 Volume d’eau distillée (ml) 50,0 49,0 48,0 47,0 46,0 45,0 Concentration de BM (10 -6 mol.L -1 ) 0,0 2,0 4,0 6,0 8,0 10,0
Tableau II- 2 : Absorbances des solutions du BM
Concentration du bleu de méthylène (10 -6 M)
Figure II- 5 : Courbe d’étalonnage de la solution de BM
For both quantitative and qualitative analysis of solids, diffuse reflectance spectroscopy (R) is the appropriate method It can be viewed as the result of multiple reflections, refractions, and diffractions occurring on randomly oriented particles, with the specular component eliminated This technique is defined by a specific equation.
R I avec I R l’intensité du faisceau réfléchi et I I l’intensité du faisceau incident
L’allure du spectre de réflexion diffuse de l’échantillon dépend majoritairement de sa structure électronique, mais également de sa morphologie et de la taille des particules
Several models describe the absorption and scattering of light in inhomogeneous media, with the Kubelka-Munk model being the most widely used This model considers two opposing light fluxes propagating through an infinitely thick sample with a non-luminescent chromophore For analysis, we consider a solid of thickness X with an infinite surface to disregard edge effects The downward light flux (incident) is denoted as i_T, while the upward light flux (reflected) at height x is represented as i_R.
Figure II- 6 : Illustration du phénomène de réflectance
K represents the fraction of light flux absorbed per unit thickness (absorption coefficient, in cm -1), while S denotes the fraction of light flux scattered per unit thickness (scattering coefficient, in cm -1), which is influenced by the surface texture of the material and the particle size in the case of powders Both parameters depend on the wavelength λ The model assumes that the layer is homogeneous in depth, meaning that the absorption and scattering coefficients remain constant throughout the studied layer The analysis of the fluxes traversing an infinitesimal layer of thickness dx at altitude x leads to a system of linear differential equations with constant coefficients for the variations of flux in both directions As the thickness of the solid approaches infinity, solving the differential equation yields the Kubelka-Munk remission relation F(R), which indicates that the infinite diffuse reflectance R ∞ is solely dependent on the ratio of K to S.
If the dependence of Sλ on λ is gradual, S can be treated as constant In the absence of molecular interactions, K is expressed as K = ελ × C, where ελ is the molar extinction coefficient and C is the concentration of the adsorbate Consequently, F(R) becomes a linear function of the concentration C.
Diffuse UV-Visible reflectance is a common technique for measuring the properties related to the Kubelka-Munk function and the energy E_g of the band gap The most frequently employed method, known as the Tauc plot, involves plotting the variation of the absorption coefficient F(R) against photon energy using a specific equation.
The equation F(R) hν = α0 (hν-Eg) γ describes the relationship between photon energy (hν) and absorption in semiconductors, where α0 is a constant ranging from 10^5 to 10^6 cm^-1 eV^-1 The parameter γ varies depending on the type of electronic transition In solids, there are two primary types of electron transitions from the valence band to the conduction band: direct transitions and indirect transitions.
Figure II- 7 : Représentation des transitions de bande : (a) transition directe autorisée, (b) transition directe interdite, (c) transition indirecte
A direct (allowed) transition occurs when the momenta of electrons and holes are identical in both the conduction band and the valence band This allows for the absorption of a photon with energy Eg, corresponding to a direct gap, to happen without the need for a phonon to conserve momentum In contrast, a forbidden direct transition involves a minimal momentum of photons, making it less likely to occur.
An indirect transition occurs when an electron moves through an intermediate state during the transition, requiring a phonon to transfer its momentum to the crystal lattice.
La variable γ vaut 1/2 pour une transition directe autorisée, 3/2 pour une transition directe interdite, 2 pour une transition indirecte autorisée et 3 pour une transition indirecte interdite
Semiconductors are classified as materials with either direct (γ = 0.5) or indirect (γ = 2) allowed transitions Most authors agree that at the nanoscale, semiconductors exhibit indirect transitions Consequently, they derive the energy of the band gap using the relationship between F(R) and the photon energy hν, by extrapolating the value for F(R)=0 in the linear portion of the curve.
Figure II- 8: Spectrophotomètre UV-Visible Perkin Elmer Lambda 850
Our diffuse reflectance measurements were conducted using a PerkinElmer UV 850 double-beam spectrophotometer, covering a wavelength range from 250 to 800 nm with a 1 nm step and a scanning speed of 100 nm/min The measurements utilized an integrating sphere (Labsphere PELA-1001, 15 cm in diameter) with a horizontal sample holder positioned below the sphere For each wavelength, the device sequentially measures the intensity reflected by the sphere and the intensity reflected by the sample, providing the ratio I/I To ensure accuracy, all spectra are compared to a certified Spectralon standard (Labsphere, North Sutton, USA), whose spectrum is recorded on the computer and must be corrected prior to interpretation It is essential to know the values of R0, the reflectance measured in the absence of light.
Etudes des propriétés acido-basiques des oxydes fluorés
The coupling of gas probe chemisorption with X-ray photoelectron spectroscopy (XPS) has found extensive applications in characterizing the reactivity of solids In this study, ammonia (NH3) was selected as a basic gas probe, while sulfur dioxide (SO2) was used as an acidic probe The adsorption of NH3 and SO2 on the surface of samples is monitored through XPS analysis, allowing us to determine the nature and quantity of the adsorbed probe molecules This information provides insights into the acid-base characteristics of the studied oxide surfaces.
The study of the surface reactivity of fluorinated TiO2 catalysts was conducted using a combination of a chemisorption instrument, the Autochem 2920 (Micromeritics), and the X-ray photoelectron spectrometer Kalpha (Thermo).
L’appareil Autochem est composé de quatre voies principales (Figure II- 11):
The gas probe on Loop ằ enables the automatic injection of known volume gas pulses into the sweep line or allows for mixing with the sweep gas.
Toutes ces voies sont confinées dans une enceinte thermostatée dont la température a été fixée à 110°C
Figure II- 11 : Synopsis de l’Autochem 2920
La détection à conductibilité thermique est constituée de deux voies de gaz (référence et mesure) et de quatre filaments qui forment un pont de Wheatston
Figure II- 12 : Détecteur à conductibilité thermique (catharomètre)
A variable intensity current, depending on the vector gas used, powers the filaments at points A in Figure II-12 This current causes the filaments to heat up due to the Joule effect As each pathway is swept by a gas, it exchanges a specific amount of heat with a pair of filaments based on its composition and flow rate These thermal exchanges alter the temperature and, consequently, the resistivity of the filaments, allowing for the measurement of a signal at terminals S that reflects the flow rate and nature of the gas used.
Le four de l’appareil permet d’obtenir une température maximale de 1000°C avec une rampe de montée de température de 1 à 90 °C/min
Préalablement à chaque adsorption, les échantillons ont été broyés finement, puis déposés sur un fritté en acier inoxydable situé à l’intérieur d’un réacteur en quartz (Figure II-
The analyzed sample quantity remains consistent across all tests, set at 100 ± 1 mg This reactor is linked to the Autochem 2920 and has been modified to ensure that the sample, following chemisorption, is maintained under a controlled atmosphere during its transfer to the glove box of the spectrophotometer.
Figure II- 13 : Réacteur utilisé pour la chimisorption
Les poudres ont été tout d’abord prétraitées à 30°C pendant 1 heure sous flux d’argon
(50 cc/min) L’adsorption a ensuite été effectuée sous un flux continu d’un mélange de gaz constitué de 80% du gaz sonde (SO 2 ou NH 3 à 5 % dans l’He) et de 20% d’hélium (débit :
The samples were subjected to a flow of helium at 50 cc/min and a temperature of 30°C for 10 minutes to saturate the active sites on their surfaces They were then maintained at 30°C for an additional hour under the same helium flow It is crucial to keep working temperatures below 35°C, as higher temperatures can alter the surface characteristics of the samples, particularly affecting the fluorine environment After adsorption, the reactor was transferred to a glove box connected to the Thermo K-alpha spectrometer, ensuring no exposure to the atmosphere The conditions for recording XPS spectra remained consistent with those outlined in section II-2.
Tests photocatalytiques
Les tests photocatalytiques ont été effectués à l’aide du réacteur représenté à la Figure II- 14 Ce réacteur consiste en :
- un bécher, contenant la suspension de catalyseur, placé sous agitation et au centre d’un système de circulation d’eau afin de maintenir une température constante lors des tests,
- une lampe visible (420 nm) ou ultra-visible (350 nm) placée à 10 cm de la surface de la suspension pour permettre son irradiation
Tous ces éléments sont placés dans un cadre en verre et couvert de papier aluminium
Figure II- 14 : Le réacteur photocatalytique
Before assessing the activity of photocatalysts, we ensured that the evaporation of the solution is negligible during testing and that the temperature is maintained at 30°C The design of this reactor effectively minimizes external influences on photocatalytic reactions.
We initially studied the adsorption of methylene blue (MB), a commercial dye used in this research as a model for organic pollutants, at equilibrium on the surface of the synthesized catalysts Subsequently, we assessed the photocatalytic activity of the materials through the degradation of MB.
The adsorption of BM over time was monitored using UV-Visible spectrophotometry Catalyst suspensions with a mass concentration of 0.5 g.L -1 were prepared in 250 mL of a 10 -5 mol.L -1 aqueous BM solution These suspensions were then homogenized using an ultrasonic probe for 10 minutes and kept in the dark under agitation Samples of 5 mL were taken every 30 minutes to track the BM concentration in the suspension until adsorption equilibrium was reached.
BM à la surface du catalyseur soit atteint Le temps nécessaire à l’établissement de l’équilibre du BM à la surface des catalyseurs a ainsi été déterminé
The photocatalytic activity of the catalysts was evaluated following a specific protocol we established Catalyst suspensions were prepared as previously described Once adsorption equilibrium was achieved, the suspension containing the BM and the catalyst was irradiated using a visible light lamp (420 nm) or an ultraviolet lamp (350 nm) for 4 hours The monitoring of photocatalytic degradation was conducted through UV-Visible spectrophotometric analysis of samples (5 mL every 30 minutes) taken from the suspension after centrifugation.
[2] J Rodriguez–Carvajal, Laboratoire Léon Brillouin, http://www.ill.eu/sites/fullprof (2011)
[3] P Kubelka and F Munk, Z Tech Phys 11a, 593 (1931)
[4] V Weidner and J Hsia, J Opt Soc Am 71, 856 (1981)
[6] J Tauc, R Grigorovici and A Vancu, Phys Stat Sol 15, 627 (1966)
[7] K.A Michalow, D Logvinovich, A Weidenkaff , M Amberg, G Fortunato, A Heel, T Graule and M Rekas, Catal Today 144, 7 (2009)
[8] A.B Murphy, Sol Energ Mat Sol C 91, 1326 (2007)
Catalyseurs à base de TiO 2 issus des phases cristallographiques anatase ou rutile
In the field of photocatalysis, current research and applications primarily focus on the use of semiconductors, particularly titanium dioxide (TiO2) This material offers several advantages, including chemical stability, low cost, minimal toxicity, and strong oxidizing power at room temperature However, the photocatalytic activity of titanium dioxide is constrained for practical applications due to its bandgap width (Eg = 3.0 – 3.2 eV, depending on its crystalline or amorphous nature), necessitating ultraviolet (UV) radiation as an excitation source Consequently, applications under visible light irradiation are limited.
This chapter evaluates the potential of photoactivable titanium oxide (TiO2) materials under UV and visible light by examining the effects of fluorine atoms chemisorbed on the surface of both anatase and rutile forms of TiO2 The synthesis method employed is known as thermal shock The first section describes the fluorination process of TiO2 anatase and rutile, while the second section investigates the impact of synthesis temperature on the structure and morphology of the catalysts The final sections focus on their surface and optical properties, which are crucial for understanding the kinetics and mechanisms of photocatalytic reactions Additionally, the photocatalytic activity of these modified oxides under UV and visible irradiation is reported.
Synthèse et caractérisation des matériaux
This study investigated various fluorination temperatures for two crystallographic forms of titanium dioxide: anatase and rutile The commercial phases obtained from Sigma Aldrich have purities of 99.5% to 99.7% and particle sizes of less than 100 nm.
Potassium fluoride (KF), utilized in the thermal shock method, is sourced from Sigma Aldrich with a purity of 99.99% All chemical substances used in this study are applied without further purification.
Figure III- 1 : Représentation schématique du processus de synthèse des catalyseurs fluorés
La mộthode du ô choc thermique ằ (CT) consiste en deux ộtapes principales et est décrite schématiquement à partir de la Figure III- 1:
• L’échange simple de ligand en solution aqueuse entre les ions fluorures et les groupements hydroxyles de surface de TiO 2
Thermal shock is assessed at various temperatures (400, 500, 600, 700, 800, and 950°C) The ligand exchange in aqueous solution begins with the addition of 10 mL of 0.625 mol.L -1 KF solution to 0.5 g of TiO2 powder, either anatase or rutile, achieving a 1:1 atomic ratio of titanium to fluorine The mixture is then stirred magnetically for 15 minutes to ensure proper dispersion of TiO2 particles During this agitation, ligand exchange occurs as described in reaction (1) [2, 3, 4].
10 mL de solution du KF 0,625 mol.l -1
Lavage et Filtrage sur Büchner Sécher à T = 150°C/ 1h
The suspension is dried at 150°C for 3 hours to evaporate residual water The resulting powders are subjected to a thermal treatment at temperatures of 400, 500, 600, 700, 800, or 950°C for 5 minutes, followed by air cooling They are then washed with distilled water through a filtration membrane to remove potassium and fluoride ions that may be physisorbed on the surface Finally, the powders are dried again at 150°C for 1 hour The fluorinated samples derived from the anatase and rutile phases are designated as AFTO-X and RFTO-X, where X represents the thermal treatment temperature: 400, 500, or 600.
I.2 Caractérisation des catalyseurs monophasés anatase et rutile
I.2.1.1 Catalyseurs issus de la phase anatase
X-ray diffraction (XRD) analysis enables the examination of the effects of fluorination through the thermal shock method on the structure and composition of the crystalline phase of TiO2 Figure III-2 displays the X-ray diffractograms obtained for both raw anatase phase TiO2 samples and fluorinated samples This study determined the crystalline phase composition and lattice parameters for all synthesized samples, as detailed in Table III-1.
The diffractogram of the raw sample reveals a single phase of TiO2 anatase (space group I41/amd, a = 3.7826 Å, c = 9.498 Å, JCPDS No 21-1272) The characteristic diffraction peaks are marked as "a" on the diffractogram, corresponding to specific diffraction lines.
Up to temperatures of 500°C, the diffractogram shows no alteration in the structure, and the lattice parameters listed in Table III-1 are nearly identical to those of the raw anatase phase.
At higher temperatures, additional peaks emerge from the anatase phase In the AFTO-600 sample, two types of additional peaks are identified, labeled as "r" and "*" The first type corresponds to the rutile crystallographic phase (space group P4/mnm, a = 4.532 Å, c = 2.983 Å), which typically forms around 700°C The second type is linked to the presence of another phase.
The structure is a zig-zag formation made up of TiO6 octahedra connected at their vertices, creating rectangular tunnels along the b-axis that are occupied by potassium ions The lattice parameters of this phase, determined from recorded X-ray diffractograms, align with existing literature.
The K2Ti6O13 structure belongs to the family of alkali titanates, characterized by the general formula A2TixO2x+1, where A represents either K or Na and n can be 2, 4, or 6 Titanates with n = 6, such as K2Ti6O13, exhibit a tunnel-type structure, while those with n = 2 and 4, including K2Ti2O5 and K2Ti4O9, display a lamellar structure.
Research indicates that the synthesis of K2Ti6O13 can be achieved through a solid-state reaction between KF and TiO2 at high temperatures The formation of this phase at a critical temperature of 600°C is attributed to the synthesis conditions of fluorinated TiO2 Increasing the temperature to 950°C leads to the nearly complete transformation of anatase and rutile phases into K2Ti6O13.
Figure III- 2 : Diffractogrammes RX des échantillons non fluorés et fluorés issus de la phase anatase (a, r, * respectivement pour anatase, rutile et K2Ti6O13)
Tableau III- 1 : Composition de phase et paramètres de mailles des échantillons non fluorés et fluorés issus de la phase anatase
Figure III- 3 (a): Schéma de la structure de K2Ti6O13 : Structure de la maille de K2Ti6O13
Figure III- 3 (b): Schéma de la structure de K 2 Ti 6 O 13 : Tunnels rectangulaires occupés par les ions potassium
I.2.1.2 Catalyseurs issus de la phase rutile
Figure III-4 compares the X-ray diffractograms of raw TiO2 rutile samples with those of fluorinated TiO2 samples Their phase composition and lattice parameters are detailed in Table III-2 For the raw sample, all peaks ((110), (101), (111), and (220)) can be indexed and align with literature data Up to 500°C, the structure of the fluorinated catalysts remains similar to the TiO2 rutile phase.
The K2Ti6O13 phase is first observed at 600°C, and as temperatures increase, the concentration of K2Ti6O13 rises until the complete transformation of the rutile phase into the titanate phase occurs, similar to the behavior seen in fluorinated catalysts derived from the anatase phase.
Figure III- 4 : Diffractogrammes des échantillons non fluorés et fluorés issus de la phase rutile (r, * respectivement pour rutile et K2Ti6O13)
Tableau III- 2 : Composition des phases et paramètres de mailles des échantillons non fluorés et fluorés issus de la phase rutile
La caractérisation des catalyseurs synthétisés a été complétée par une étude microscopique en détection d’électrons secondaires, possible à partir d’un équipement de type nanosonde Auger JEOL JAMP 9500F
I.2.2.1 Morphologie des catalyseurs issus de la phase anatase
The image obtained from the raw TiO2 anatase sample at a magnification of ×100,000 reveals an aggregation of spherical particles The average particle size distribution, based on ten particles, ranges from 25 to 45 nm, which aligns with the narrow full width at half maximum of the diffraction peaks Additionally, the morphology and size of the fluorinated sample subjected to a calcination temperature of 500°C remain unchanged.
The diameter of crystallites can be assessed using the full width at half maximum (FWHM) of the diffraction peaks, applying the Debye-Scherrer formula The most intense diffraction peaks for anatase (101) and rutile (110) were chosen for this evaluation.
Etude par Spectroscopie Photoélectronique à rayonnement X
X-ray photoelectron spectroscopy (XPS) enables the analysis of the chemical composition of surfaces and the local environment of various elements present Additionally, it allows for monitoring the effects of CT fluorination on catalyst surfaces.
II.1 Analyses XPS des matériaux de références
Prior XPS analysis of reference compounds (TiO2 anatase, TiO2 rutile, KF, and TiF4) is essential for identifying the chemical environment and oxidation state of elements in fluorinated catalysts derived from anatase or rutile phases.
II.1.1 Analyse des précurseurs TiO 2 issus des phases anatase et rutile
The precursors TiO2 anatase and TiO2 rutile were analyzed using X-ray photoelectron spectroscopy (XPS) The findings from these XPS analyses are detailed in Table III-5 Figures III-7 and III-8 display the spectra corresponding to the core peaks Ti 2p and O 1s.
The C 1s peak for anatase and rutile can be divided into three components: the main component at 285.0 eV, indicative of carbon contamination (carbon-carbon bond), and two minor components at 286.6 eV and 289.1 eV, which are associated with carbon atoms in CO and CO2 environments, respectively.
Figure III- 7 : Pics de cœur Ti 2p des échantillons TiO2 anatase et TiO2 rutile
Due to spin-orbit coupling, the titanium 2p core level appears as a doublet separated by an energy of 5.7 eV, characteristic of the Ti 2p3/2 and Ti 2p1/2 components This doublet exhibits a slight asymmetry, commonly observed in transition metals The associated binding energies are 459.0 eV and 464.7 eV, which align with the XPS data obtained for TiO2 in our laboratory.
Additionally, each main peak of the Ti 2p 3/2 and Ti 2p 1/2 components is accompanied by a satellite peak located 13 eV higher in binding energy than the main peak The origin of the Ti 2p satellite peaks remains a topic of debate, with one explanation being the strong covalent hybridization between the metal's d orbital and the p orbitals of the oxygen atoms.
Figure III- 8: Pics de cœur O 1s des échantillons TiO 2 anatase et TiO 2 rutile
The heart spectra O 1s exhibit two main components: the predominant component (O I) at 530.3 eV is linked to O 2- ions within the crystalline lattice of the oxide, while the minor component (O II) at a binding energy of 531.6 eV corresponds to surface hydroxyl groups.
Les pics Ti 2p et O 1s des deux oxydes sont très similaires, en accord avec des environnements très proches au sein des deux structures Néanmoins, le calcul du rapport
The rutile phase is slightly oxygen-rich, while the anatase phase maintains a precise 2:1 ratio of oxygen to titanium Calculations of OII/Ti ratios have been conducted to monitor changes in surface hydroxyl group content during fluorination These hydroxyl groups are crucial for photocatalytic properties, as they can act as hydroxyl radicals upon irradiation.
Table III-5 presents the bond energy (eV), full width at half maximum (eV), and atomic percentage of the constituent elements in anatase TiO2 and rutile TiO2, along with the atomic percentage ratio between the O 1s components and the Ti 2p core peaks.
The XPS analysis of TiF4 was conducted under specific experimental conditions, utilizing a sample holder and analysis chamber maintained at approximately -140°C with a liquid nitrogen cooling system to prevent potential sublimation and degradation of the compound, which is highly hygroscopic Titanium tetrafluoride is a polymeric compound composed of dimers or trimers of TiF6 octahedra linked at the vertices The results of this XPS analysis are presented in Table III-6.
The C 1s core spectrum consists of several components, with peaks at 285.0, 286.4, and 289.8 eV attributed to carbon atoms from C-C contamination and those in CO and CO2 environments Additionally, components at 295.7, 297.3, 298.5, and 300.3 eV are linked to carbon atoms in a fluorine environment.
Figure III-9 illustrates the Ti 2p3/2-1/2 core spectrum of the TiF4 compound, revealing two main components labeled A and B The high binding energy component B, ranging from 463.0 to 469.0 eV, is accompanied by three satellite peaks at 3.7, 6, and 12.9 eV, which correspond to titanium atoms surrounded by fluorine, similar to TiF Additionally, it features two satellite peaks at 471.1 and 477.4 eV, indicating an intermediate binding energy between that of oxygen-rich environments like TiO2 (Ti 2p3/2: 459.1 eV) and fluorine-rich environments such as TiF4 This suggests that titanium exists in a mixed oxygen/fluorine environment The energy positions align with a first-order interpretation of a chemical shift related to an initial state effect, based on the net charge of the atoms influenced by the electronegativity of their surroundings.
Figure III-9 illustrates the Ti 2p heart peaks of the TiF4 sample, highlighting the main Ti B 2p 3/2 peak and its blue satellites, along with the Ti B 2p 1/2 peak and its gray satellites Additionally, it features the primary Ti A 2p peaks and their corresponding red satellites.
The F 1s spectrum of the TiF 4 sample reveals an asymmetric peak that can be divided into two components The first peak, F B, located at 685.7 eV, is associated with the fluorine atoms in TiF 4, which is further validated by the F B/Ti B ratio of 4.35.
Le deuxième pic FA localisé à 686,8 eV est attribué aux atomes F du TiF4 oxygéné Le rapport
The ratio F A/Ti A is 4.15, which is lower than that of F B/Ti B This observation aligns with the mixed fluorine/oxygen environment of titanium associated with component A of the Ti 2p core peak, confirming the presence of oxygen atoms on the surface of TiF4, likely due to its highly hygroscopic nature.
Figure III- 10 : Pics de cœur F 1s de l’échantillon TiF4
Propriétés optiques
In our study, TiO2 oxides were fluorinated using the CT method to create new photocatalysts designed to operate under UV and visible light irradiation The impact of fluorination on the optical properties of TiO2 is essential and will be analyzed through UV-visible diffuse reflectance spectra The bandgap values were determined from these spectra by plotting the variation of the absorption coefficient F(R) as a function of photon energy using the following equation:
(F(R)hν) 1/2 = α 0 (hν-Eg) (3) [33, 34] α 0 est une constante (10 5 à 10 6 cm -1 eV -1 pour des oxydes semi-conducteurs), hν l’énergie du photon, et E g est l’énergie du gap
III.1 Absorption UV-Vis des échantillons fluorés issus de la phase anatase
Figure III-20 (a) illustrates the UV-Vis diffuse reflectance spectra of fluorinated samples derived from the anatase phase, while the Tauc plots and calculated gap values for these samples are presented in Figure III-20 (b) and Table III-10 The TiO2 anatase sample exhibits an absorption band located in the UV range below 385 nm, corresponding to a gap value of 3.19 eV The AFTO samples prepared at temperatures between 400 and 600°C show gap values nearly identical to that of the raw anatase (approximately 3.16 – 3.18 eV) Fluorination of anatase results in a minimal shift of the absorption band towards the visible region for thermal shock temperatures below 600°C However, fluorination at higher temperatures (700 to 950°C) causes a shift of the absorption band towards the UV region, leading to an increase in gap values from 3.21 to 3.31 eV This increase is attributed to the formation of rutile and K2Ti6O13 phases observed from 600°C onwards The AFTO-950 sample, with K2Ti6O13 structure, exhibits the highest gap value of 3.31 eV, consistent with Wang et al.'s findings, which indicate a gap value of 3.4 eV for K2Ti6O13.
Pour ces échantillons, la fluoration et la présence de lacune d’oxygène n’affectent pas considérablement les propriétés optiques de ces catalyseurs
Figure III- 20 (a) : Spectres de réflectance diffuse UV-visible
Figure III- 20 (b) : (F(R)hν) 0.5 en fonction de hν pour TiO 2 et les échantillons fluorés issus de la phase anatase
Anatase AFTO-400 AFTO-500 AFTO-600 AFTO-700 AFTO-800 AFTO-950
Tableau III- 10 : Valeurs de gap et longueurs d’onde d’absorption des échantillons brut et fluorés issus de la phase anatase
III.2 Absorption UV-Vis des échantillons fluorés issus du rutile
Figure III-21 (a) displays the UV-Vis diffuse reflectance spectra of fluorinated materials derived from the rutile phase The raw rutile shows localized absorption in a wavelength range below 410 nm, which corresponds to a band gap of 2.97 eV, as determined by the Tauc plot (Figure III-21 b).
La fluoration provoque un gap réduit de 0,1 eV pour les échantillons RFTO-400, RFTO-
The displacement observed in the RFTO-600 sample supports the earlier hypothesis regarding the formation of titanium vacancies at fluorination temperatures of rutile TiO2 below 600°C However, the energy gap significantly increases for samples fluorinated above 700°C The RFTO-950 sample exhibits an absorption band shifted into the UV region, corresponding to the energy gap of the K2Ti6O13 phase, which is 3.35 eV, similar to the AFTO-950 sample.
Figure III- 21 (a) : Spectres de réflectance diffuse UV-visible
Figure III- 21 (b) : (F(R)hν) 0.5 en fonction de hν pour TiO 2 et les échantillons fluorés issus de la phase rutile
Rutile RFTO-400 RFTO-500 RFTO-600 RFTO-700 RFTO-800 RFTO-950
Tableau III- 11 : Valeurs de gap et longueurs d’onde d’absorption des échantillons brut et fluorés issus de la phase rutile
The UV-Vis diffuse reflectance spectra reveal the varying effects of fluorination on the optical properties of anatase and rutile Within the temperature range of 400°C to 600°C, fluorination does not significantly alter the optical characteristics of TiO2 in both anatase and rutile forms.
The minimal fluctuation in the value of rutile can be attributed to the presence of Ti 3+ ions, which arise from titanium vacancies created during the fluorination process.
Thermal shock fluorination within a specific temperature range may enhance the photocatalytic activity of rutile TiO2 under visible light irradiation Additionally, the formation of oxygen vacancies in fluorinated TiO2 catalysts derived from the anatase phase could account for the minimal decrease in bandgap values at fluorination temperatures below 600°C In contrast, fluorination conducted above 600°C leads to a significant increase in the bandgap due to the formation of the K2Ti6O13 phase.
Fluorination plays a minor role in the optical properties of catalysts and does not alter the band gap, as confirmed by Yamaki et al [36] Their research indicates that the F 2p levels of fluorine are distinct within the valence band, showing minimal interaction with the 2p and 3d levels of oxygen and titanium atoms in TiO2.
Tests photocatalytiques
After preparing and characterizing fluorinated catalysts derived from anatase and rutile phases, we assessed their photocatalytic properties To evaluate the performance of the synthesized catalysts, we investigated the kinetics of the photodegradation of a model organic dye, methylene blue (MB).
The study of photocatalytic properties was conducted in two phases Initially, the adsorption of BM on the surface of the catalysts was performed to assess the effects of fluorination on the interaction between the oxide surfaces and BM molecules The results are detailed in section IV.1.
Puis, nous avons évalué l’activité photocatalytique de ces oxydes sous irradiation UV et sous irradiation visible
IV.1 Adsorption du bleu de méthylène
Most photocatalytic reactions occur on the surface of catalysts, making the adsorption of organic molecules a crucial process in photocatalysis Figures III-22 and III-23 illustrate the percentage of methylene blue (MB) adsorbed on the surfaces of anatase and rutile phase catalysts at adsorption equilibrium The raw TiO2 samples show MB adsorption percentages of 12.3% for anatase and 10.4% for rutile The results indicate that all fluorinated samples exhibit higher MB adsorption capacities compared to the raw samples Furthermore, the percentage of adsorbed MB correlates with the F I/Ti ratio, peaking at 67.5% for AFTO-600 and 82.8% for RFTO-700.
Anatase AFTO-400 AFTO-500 AFTO-600 AFTO-700 AFTO-800 AFTO-950 0
Figure III- 22 : Comparaison du pourcentage de BM adsorbé sur la surface (colonnes) et du rapport FI/Ti (point) des catalyseurs TiO2 issus de la phase anatase
Several hypotheses can explain the increase in adsorption, including changes in the crystalline structure, morphology, and surface polarity of the catalyst However, within the temperature range of 400 to 500°C, the first two hypotheses can be dismissed, as no structural or particle size modifications are observed, despite the rising percentage of adsorbed BM The enhanced adsorption of BM on catalysts prepared at 400 and 500°C is attributed to the presence of fluorinated species detected on the surface via XPS Adsorption processes are influenced by electrostatic interactions between organic molecules and the oxide surface BM, a dye with electron-rich π aromatic cycles, is easily attracted to Lewis acid-type sites on the surface When TiO2 is fluorinated, the chemisorbed F- ions on the oxide surface, being highly electronegative, elevate the charge of neighboring titanium sites.
Ti 4+ jouent alors le rôle de sites acides de Lewis et peuvent contribuer à augmenter l’adsorption du BM à la surface des TiO2 fluorés
At temperatures exceeding 600°C, the crystalline phase K2Ti6O13 forms, leading to the agglomeration of anatase and rutile particles, which reduces the surface area accessible for BM catalysts and their adsorption capacity Despite this, the dye remains strongly adsorbed on the surfaces of AFTO-600 and RFTO-600 samples, which exhibit a high concentration of chemisorbed F- anions This indicates that structural and morphological changes at these temperatures are insufficient to mitigate the impact of surface F- species The highest BM adsorption is observed in AFTO-600 and RFTO-700 samples, both of which have elevated levels of surface F- anions (FI/Ti = 0.21 for AFTO-600 and FI/Ti = 0.18 for RFTO-700).
Rutile RFTO-400 RFTO-500 RFTO-600 RFTO-700 RFTO-800 RFTO-950 0
Figure III- 23 : Comparaison du pourcentage de BM adsorbé sur la surface (colonnes) et du rapport FI/Ti des catalyseurs TiO2 issus de la phase rutile
For TiO2 samples derived from fluorinated anatase phases obtained above 600°C, the adsorbed BM content begins to decrease This trend is even more pronounced in TiO2 samples from fluorinated rutile phases above 700°C, where the BM content decreases despite an increasing F I/Ti ratio This indicates that the significant growth of K2Ti6O13 particles and particle agglomeration lead to a reduction in the percentage of adsorbed BM.
IV.2 Activités photocatalytiques sous irradiation UV
IV.2.1 Activités des catalyseurs issus de la phase anatase
Figure III-24 illustrates the degradation of BM over time using fluorinated catalysts at varying temperatures derived from the anatase phase under UV irradiation The decomposition of BM in aqueous solution adheres to a pseudo-first-order kinetic model of the Langmuir-Hinshelwood type, represented by the general equation (4): t ' k.
C ) ln(C 0 = (4) ó C 0 et C (mol.L -1 ) sont respectivement la concentration initiale et la concentration du BM au moment t (h), k’ (h -1 ) représentant la constante de vitesse apparente
Anatase AFTO-400 AFTO-500 AFTO-600 AFTO-700 AFTO-800 AFTO-950 L n (C 0 /C )
Figure III- 24 : Ln(C 0 /C) en fonction du temps : Détermination de la constante de vitesse apparente de dégradation du BM sous irradiation UV sur les échantillons AFTO-X
Les constantes de vitesse k’ sont calculộes en traỗant Ln(C 0 /C) en fonction du temps et présentées sur la Figure III- 25 Pour l’anatase brut, la constante de vitesse k’ atteint 0,31 h -1
The degradation rate of BM significantly increases during the fluorination of TiO2 at temperatures of 400°C and 500°C The optimal photocatalytic activity is achieved with the AFTO-500 sample, which has a rate constant of k' = 2.36 h^-1.
La vitesse de dégradation du BM commence à diminuer pour des températures supérieures à 600°C La constante de vitesse de l’échantillon AFTO-950 est environ quatre fois inférieure à celle du meilleur photocatalyseur AFTO-500
Anatase AFTO-400 AFTO-500 AFTO-600 AFTO-700 AFTO-800 AFTO-950 0,0
Figure III- 25 : Comparaison des constantes de vitesse k’ de la dégradation du BM sous irradiation UV pour les catalyseurs AFTO-X et du rapport F/Ti (évalué en XPS)
IV.2.2 Activités des catalyseurs issus de la phase rutile
Similar to the fluorinated catalysts derived from anatase, the degradation of BM on oxides from the rutile phase follows a pseudo-first-order kinetic model of the Langmuir-Hinshelwood type As shown in Figure III-27, the degradation rate constants for BM decomposition on these catalysts are compared Under UV irradiation, the degradation rate of BM for raw rutile (k' = 0.36 h⁻¹) is comparable to that of raw anatase (k' = 0.31 h⁻¹) The rate constant k' significantly increases for the RFTO-400 and RFTO-500 samples, highlighting the effect of fluorination on photocatalytic activity, as no structural or morphological changes are observed.
Néanmoins, l’activité photocatalytique chute au-delà de 500°C La constante de vitesse passe de 0,94 h -1 pour le RFTO-500 à 0,72 h -1 pour le RFTO-600, puis atteint 0,30 h -1 pour le RFTO-950
Rutile RFTO-400 RFTO-500 RFTO-600 RFTO-700 RFTO-800 RFTO-950
Figure III- 26 : Ln(C0/C) en fonction du temps : Détermination de la constante de vitesse apparente de dégradation du BM sous irradiation UV sur les échantillons RFTO-X
Rutile RFTO-400 RFTO-500 RFTO-600 RFTO-700 RFTO-800 RFTO-950 0,0
Figure III- 27 : Comparaison des constantes de vitesse k’ de la dégradation du BM sous irradiation UV pour les catalyseurs RFTO-X et du rapport F/Ti (évalué en XPS)
IV.2.3 Influences de la fluoration par choc thermique
Under UV irradiation, the photocatalytic activity of fluorinated catalysts from both anatase and rutile phases increases, with the maximum reaction rate achieved at a fluorination temperature of 500°C This enhancement in photocatalytic performance can be attributed to the presence of F- species on the surface and an increased concentration of surface hydroxyl groups The chemisorption of fluoride ions leads to the production of surface hydroxyl groups to maintain electroneutrality These hydroxyl groups can react with photogenerated holes during irradiation to form adsorbed hydroxyl radicals (OH●), which act as oxidizing agents for organic compounds Consequently, the degradation of BM occurs through the oxidation of BM molecules adsorbed on the oxide surface by the OH● radicals.
≡Ti-OH + h vb + → ≡Ti-OH ●+ (5)
De plus, la formation des espèces F - à la surface de TiO2 augmente la disponibilité des trous photogénérés sous irradiation UV [3, 38], ceux-ci ne pouvant pas réagir avec les sites
Ti-F exhibits a high redox potential due to the F ● /F - couple (3.6 eV), enabling interactions with water molecules or OH- ions in solution, leading to the generation of mobile OH ● radicals These radicals are effective in degrading organic matter in solution, highlighting that surface fluorination significantly enhances photocatalytic performance.
≡Ti-F + H 2 O (ou OH - ) + h vb + → ≡Ti-F + OH ● mobile + H + (6)
The presence of fluoride ions on the surface due to fluorination enhances the adsorption of the organic molecule through stronger electrostatic interactions This dual effect of fluorination and the formation of surface hydroxyl groups promotes the degradation of the organic molecule, thereby improving the photocatalytic activity of fluorinated catalysts at temperatures of 400°C and 500°C.
Above 600°C, the activity of fluorinated TiO2 decreases despite a significant amount of chemisorbed F- Crystallographic and morphological studies highlight the emergence of the K2Ti6O13 phase, the transformation of anatase to rutile, and an increase in particle size.
The anatase phase exhibits higher photocatalytic activity under UV irradiation compared to the rutile phase, aligning with the findings of Ranjith G Nair et al Furthermore, the presence of the K2Ti6O13 phase significantly impacts the photocatalytic activity of our catalysts, as its degradation rate constant is considerably lower.
L’augmentation de la taille des particules a également des conséquences sur la diminution de la constante de vitesse de dégradation : réduction du nombre de groupements
OH disponibles en surface (cf Tableau III- 8 et Tableau III- 9) et recombinaisons électrons- trous
In summary, photocatalytic test results under UV irradiation reveal the effects of fluorination using the CT method on the activity of anatase and rutile TiO2 Key parameters to consider include:
- la transformation de phase anatase-rutile et la formation de la phase K 2 Ti 6 O 13
- la teneur en ions fluorures et groupements hydroxyles en surface
• de 400 à 500°C : l’activité photocatalytique augmente en raison de la présence des espèces F - chimisorbées et de l’augmentation de la teneur en OH de surface