* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Experimental details
Woodward–Hoffmann rules wikipedia , lookup
Marcus theory wikipedia , lookup
Deoxyribozyme wikipedia , lookup
Electrochemistry wikipedia , lookup
Nanofluidic circuitry wikipedia , lookup
Determination of equilibrium constants wikipedia , lookup
Equilibrium chemistry wikipedia , lookup
Chemical thermodynamics wikipedia , lookup
George S. Hammond wikipedia , lookup
Chemical equilibrium wikipedia , lookup
Physical organic chemistry wikipedia , lookup
Hydrogen-bond catalysis wikipedia , lookup
Ultraviolet–visible spectroscopy wikipedia , lookup
Stability constants of complexes wikipedia , lookup
Enzyme catalysis wikipedia , lookup
Ene reaction wikipedia , lookup
Rate equation wikipedia , lookup
Industrial catalysts wikipedia , lookup
Surface properties of transition metal oxides wikipedia , lookup
Radiation Induced Processes at SolidLiquid Interfaces MIAO YANG 杨淼 KTH Royal Institute of Technology School of Chemical Science and Engineering Department of Chemistry Applied Physical Chemistry SE-100 44 Stockholm, Sweden Copyright © Miao Yang All rights reserved Paper I © 2013 Elsevier B. V. Paper II © 2013 Elsevier B. V. Paper III © 2014 American Chemical Society Paper IV © 2015 Elsevier B. V. Paper VIII © The Royal Society of Chemistry 2015 TRITA-CHE Report 2015:39 ISSN: 1654-1081 ISBN: 978-91-7595-656-5 Akademisk avhandling som med tillstånd av KTH Kungliga Tekniska Högskolan i Stockholm framlägges till offentlig granskning för avläggande av teknologie doktorsexamen fredagen den 18 September 2015, kl 10.00 i sal F3, KTH, Lindstedtsvägen 26, Stockholm. Avhandlingen försvaras på engelska. Opponent: Prof. Simon Pimblott från The University of Manchester, UK. ii To my family If we knew what it was we were doing, it would not be called research, would it? Albert Einstein iii iv Abstract In the thesis, the reactions between water radiolysis products—H2O2, HO• and O2—with metals and metal oxides utilized in nuclear industry are studied. The reactions include not only surface reactions, e.g. redox reactions and catalytic decomposition of H2O2, but also solution reactions (Haber-Weiss reactions). To study the interfacial reactions, it is crucial to monitor the dissolution of the solid material, reactivity of H2O2 and formation of the intermediate hydroxyl radicals. Hydroxyl radicals are captured by probe (Tris or methanol) to generate CH2O which can be quantified by the modified Hantzsch method. The results from γ-irradiation experiments on homogeneous system show that the conversion yield of CH2O from hydroxyl radicals is affected by O2 and pH. A mechanism of CH2O production from Tris is proposed. Besides, the consumption rate of H2O2 in the H2O2/ZrO2/Tris system is found to be influenced by Tris. A mechanism for the catalytic decomposition of H2O2 upon ZrO2 surface is proposed which includes independent surface adsorption sites for H2O2 and Tris. Moreover, it is demonstrated that the deviation of detected CH2O concentration by the modified Hantzsch method from actual concentration increases with increasing [H2O2]0/[CH2O]0. The inhibition of sulfide on the radiation induced dissolution of UO2 is confirmed and is dependent on sulfide concentration. And the inhibition of sulfide is independent to that of H2/Pd. It is found that the reactivity of H2O2 and dynamics of CH2O formation are different for the studied materials in the H2O2/MxOy/Probe system. The kinetic parameters, such as rate constant, activation energy, frequency factors are determined. Both surface and solution reactions are observed in the aqueous W(s)/H2O2/Tris system. It is also demonstrated that Haber-Weiss reactions which produce HO• continuously are dominating. Furthermore, it is found that hydroxyl radicals are formed simultaneously during the dissolution of W in aerobic aqueous system. The knowledge conveyed by the thesis is relevant to nuclear technology, as well as the applications related in photocatalysis, biochemistry, corrosion science, catalysis and optics/electronics. v Sammanfattning I denna avhandling presenteras experimentella studier av reaktioner mellan radiolysprodukterna H2O2, HO• och O2 och metaller och metalloxider, förekommande i kärntekniska tillämpningar. Reaktionerna inkluderar ytreaktioner, t.ex. redoxreaktioner och katalytiskt sönderfall av H2O2, samt reaktioner i lösning (Haber-Weiss). För att kunna studera ovanstående ytreaktioner är det viktigt att studera upplösning av det fasta materialet, försvinnandet av H2O2 och produktion av hydroxylradikaler. Genom att tillsätta Tris eller metanol kan hydroxylradikalen fångas in under bildande av en stabil produkt som lätt kan detekteras (formaldehyd) genom en metod som kallas Hantzschmetoden. Resultat från homogena gammabestrålningsexperiment visar att utbytet av formaldehyd påverkas av såväl O2 som pH. En mekanism för produktionen av CH2O från Tris föreslås. Dessuton påverkas konsumtionshastigheten av H2O2 på ZrO2 av Tris. En modifierad mekanism för katalytisk sönderdelning av H2O2 på ZrO2ytor föreslås, vilken omfattar oberoende ytadsorptionsplatser för H2O2 och Tris. Dessutom är det visat att avvikelsen i CH2O-koncentration, uppmätt med den modifierade Hantzsch metoden, från den verkliga CH2O-koncentrationen ökar med ökande [H2O2]0/[CH2O]0. Detta arbete bekräftar även att sulfid inhiberar strålningsinducerad upplösning av UO2. Effekten beror på sulfidkoncentration. Det har också visats att denna inhibering är oberoende av H2/Pd. Det har visat sig att reaktiviteten för H2O2 och dynamiken för bildning av formaldehyd vid ytkatalys beror på vilken oxid som studeras. De kinetiska parametrarna, såsom hastighetskonstanter, aktiveringsenergier och frekvensfaktorer, har bestämts. Både reaktioner på ytor och reaktioner i lösning har observerats i W(s)/H2O2/Tris-systemet. Det har visats att Haber-Weissreaktioner dominerar systemet. Även i system där O2 används för att oxidera W bildas hydroxylradikaler. Detta beror på att H2O2 bildas som en intermediär. Den nya kunskap som denna avhandling förmedlar är relevant inom kärnteknik såväl som inom tillämpningar av fotokatalys, biokemi, korrosionsvetenskap, katalys och optik/elektronik. vi List of Papers I. Inhibition of Radiation Induced Dissolution of UO2 by Sulfide – A Comparison with the Hydrogen Effect. Miao Yang, Alexandre Barreiro Fidalgo, Sara Sundin, Mats Jonsson, Journal of Nuclear Materials, 434(2013)38-42, DOI: 10.1016/j.jnucmat.2012.10.050 II. Catalytic Decomposition of Hydrogen Peroxide on Transition Metal and Lanthanide Oxides. Claudio M. Lousada, Miao Yang, Kristina Nilsson, Mats Jonsson. Journal of Molecular Catalysis A: Chemical, 379(2013)178-184. DOI: 10.1016/j.molcata.2013.08.017 III. Evaluation of the O2 and pH Effects on Probes for Surface Bound Hydroxyl Radicals. Miao Yang, Mats Jonsson. The Journal of Physical Chemistry C, 118(2014)7911-7979, DOI: 10.1021/jp412571p IV. Surface Reactivity of Hydroxyl Radicals Formed upon Catalytic Decomposition of H2O2 on ZrO2. Miao Yang, Mats Jonsson. Journal of Molecular Catalysis A: Chemical, 400(2015)49-55, DOI: 10.1016/j.molcata.2015.02.002 V. The Reactivity and Mechanism of Reactions between H2O2 and Tungsten Powder. Miao Yang, Xian Zhang, Alex Grosjean, Inna Soroka, Mats Jonsson. The Journal of Physical Chemistry C, under review. VI. Influence of H2O2 on the Quantitative Determination of Formaldehyde using Acetoacetanilide as Detection Reagent in the Hantzsch Method. Miao Yang, Inna Soroka, Mats Jonsson. Manuscript, to be submitted. vii VII. Hydroxyl Radical Production in Aerobic Aqueous Solution containing Metallic Tungsten. Miao Yang, Inna Soroka, Communications, under review. VIII. Mats Jonsson. Catalysis Kinetics and Mechanisms of Reactions between H2O2 and Copper and Copper Oxides. Åsa Björkbacka, Miao Yang, Claudia Gasparrini, Christofer Leygraf, Mats Jonsson. Dalton Transactions, Accepted Manuscript, DOI: 10.1039/C5DT02024G. http://pubs.rsc.org/en/Content/ArticleLanding/2015/DT/C5DT02024G #!divAbstract My contribution to the papers Paper I: I have performed the major part of the experiments, participated to a large extent in the data evaluation, and written the major part of the manuscript. Paper II: I have performed the major part of the experiments, participated in the data evaluation and the preparation of the manuscript. Paper III IV VI VII: I have performed the experiments, participated to a large extent in the data evaluation, and written the manuscript. Paper V: I have performed the major part of the experiments, participated to a large extent in the data evaluation, and written the major part of the manuscript. Paper VIII: I have performed the experiments for the mechanistic study, participated in the data evaluation and written the mechanistic part in the manuscript. viii List of abbreviations SNF – Spent nuclear fuel G(x) – Radiation chemical yield for the species x Ċ(x) – Amount of species x produced Ḋ – Radiation dose rate ρ – Solvent density Gy – Gray, equal to 1 J kg−1 k – Reaction rate constant k1 – First-order rate constant k2 – Second-order rate constant t – Time SA – Surface area of solid V – Volume of solution Ea – Arrhenius activation energy A – Arrhenius pre-exponential or frequency factor R – Gas constant T – Absolute temperature UV-Vis – UV/Vis spectrophotometry; HPLC – High performance liquid chromatography; ESR – Electron spin resonance Tris – Tris(hydroxymethyl) aminomethane TAPS – N-Tris(hydroxymethyl)methyl-3-aminopropanesulfonic acid DMPO – 5,5-Dimethyl-1-pyrroline-N-oxide 4-CP – 4-Chlorophenol DDL – 3,5-Diactyl-1,4-dihydrolutidine CHD – 1,3-Cyclohexanedione MA – Methyl acetoacetate AAA – Acetoacetanilide AA – Ammonium acetate ix Y – Yield of formaldehyde B.E.T. – Brunauer-Emmett-Teller XRD – X-ray diffraction SEM – Scanning electron microscopy XPS – X-ray photoelectron spectroscopy ICP-OES – Inductively coupled plasma optical emission spectroscopy BE – Binding energy K – Adsorption equilibrium constant kf – Forward reaction rate constant of equilibrium reaction kr – Reverse reaction rate constant of equilibrium reaction Sa – Surface sites for H2O2 Sb – Surface sites for Tris b2 – Intercept at zero coordinate Abs – Absorbance x Table of Contents 1 2 Introduction ................................................................... 1 1.1 Background ............................................................................... 1 1.2 Water radiolysis ......................................................................... 3 1.3 Chemical kinetics of heterogeneous system ............................. 5 1.4 Oxidative dissolution of metals and metal oxides ..................... 8 1.5 Catalytic decomposition of H2O2 at solid-liquid interfaces ....... 10 1.6 Haber-Weiss reactions ............................................................ 12 1.7 Methodology for the detection of hydroxyl radicals ................. 13 1.8 The scope of this thesis........................................................... 14 Experimental details ................................................... 16 2.1 Instrumentation ........................................................................ 16 2.2 Materials and reagents ............................................................ 17 2.3 Basic methodology .................................................................. 18 2.4 Oxidative dissolution of metals and metal oxides ................... 19 2.4.1 Oxidative dissolution of W by H2O2 .................................... 19 2.4.2 Oxidative dissolution of W by O2 ........................................ 19 2.4.3 Inhibition of radiation induced dissolution of UO2 by sulfide20 2.5 Catalytic decomposition of H2O2 on metal and metal oxide surfaces ............................................................................................... 20 2.5.1 Metals ................................................................................. 20 2.5.2 ZrO2 .................................................................................... 21 2.5.3 Other lanthanide and transition metal oxides ..................... 22 2.6 Haber-Weiss reactions ............................................................ 22 2.7 Methodology for the detection of hydroxyl radicals ................. 23 2.7.1 The O2 and pH effects on probes for hydroxyl radicals ...... 23 xi 2.7.2 The interference of H2O2 on the modified Hantzsch method using AAA as derivatization reagents .............................................. 23 3 Results and discussion .............................................. 24 3.1 Methodology for the detection of hydroxyl radicals ................. 24 3.1.1 The effects of O2 and pH on probe chemistry III ................. 24 3.1.2 The effect of Tris on the catalytic decomposition of H2O2 on ZrO2 surface IV ................................................................................ 32 3.1.3 The interference of H2O2 on the modified Hantzsch method using AAA as derivatization reagents VI ......................................... 38 3.1.4 3.2 Summary ............................................................................. 39 Inhibition of radiation induced dissolution of UO2 by sulfide I . 40 3.3 Kinetics, mechanistic and simulation study of catalytic decomposition of H2O2 on metal and metal oxide surfaces ................ 42 3.3.1 Cu, Cu2O and CuO VIII ....................................................... 42 3.3.2 Other lanthanide and transition metal oxides II .................. 43 3.3.3 Summary ............................................................................. 48 3.4 Kinetics and mechanistic study of reactions between W and H2O2 or O2 ............................................................................................ 48 3.4.1 W and H2O2 V ..................................................................... 48 3.4.2 W and O2 VII ....................................................................... 57 3.4.3 Summary ............................................................................. 60 4 Conclusions ................................................................ 61 5 Future work ................................................................. 63 6 Acknowledgements .................................................... 64 7 References................................................................... 66 xii 1 Introduction 1 Introduction 1.1 Background Nuclear power has been and will be an increasingly significant contributor to energy production in several countries due to its outstanding performance in Greenhouse gas emission, efficiency and safety. One of the major drawbacks of nuclear power is the generation of radioactive waste. Regardless of the future of nuclear power, nuclear waste already exists. The major nuclear waste is the used nuclear fuel, normally called spent nuclear fuel (SNF) which is highly radioactive. The radioactivity of SNF will remain at a high level for around 100, 000 years.1 Hence, the SNF must be handled safely for the present and future generations. In light water cooled reactors, the nuclear fuel normally consists of ceramic UO2. After use in a reactor, UO2 is still the main constituent of the fuel (95 wt.%) and the remaining 5% comprises of highly radioactive fission products and transuranium elements.2, 3 Globally, about 10, 000 tonnes of SNF are produced every year and so far, it has been accumulated to approximately 300, 000 tonnes in pools or dry casks onsite.1 There are different strategies for disposal of SNF, e.g. on-site storage, space disposal, subseabed disposal, ice-sheet disposal, reprocessing and geological repository.4 To fulfill the goal of SNF disposal safety, Sweden chose the KBS-3 model, developed by SKB (Swedish Nuclear Fuel and Waste Management Company), for direct disposal of SNF in a deep geological repository.5 As shown in Figure 1,5 there are multiple barriers to protect the SNF, i.e. copper canister, bentonite clay and crystalline bedrock. In the worst case scenario, e.g. canister failure, the highly radioactive SNF will induce radiolysis of the intruding groundwater, thereby initiating a potentially aggressive chemistry at the surface of materials used in the repository. The chemistry at the interface of solid materials and water upon exposure to ionizing radiation is highly important in nuclear reactors as well as in reprocessing technology.6 1 1 Introduction Figure 1. Scheme of the SNF deep geological repository.5 There are two fundamental issues for the safety assessment of these systems (geological repository, nuclear reactor and reprocessing): the stability of materials and the release rate of radioactive elements.7 Both processes are to a large extent interfacial processes. However, in-situ studies are problematic mainly due to the extreme operating temperature, pressure as well as the very intense ionizing radiation emitted. The relevant materials in nuclear technological applications are required to be stable and durable after having withstood these extreme conditions. The materials stability under high temperature and pressure is primarily controlled by the chemistry of the system and can be predicted from experimental data obtained under the much less extreme conditions based on the tools of thermodynamics.8 The radionuclide release process can be divided into two stages: instantaneous release from the SNF surface and long-term release due to dissolution of the fuel matrix.1 The study of the release process is not only important for geological repository, but also for reprocessing technology.7 The study of heterogeneous systems is not restricted to nuclear technological applications. In heterogeneous catalysis, biochemistry, photocatalysis, corrosion science, surface polymerization and many industrial applications, e.g. water purification, interfacial processes are 2 1 Introduction also of utmost importance. The ionizing radiation present in nuclear technological applications makes the problem more intriguing. The knowledge of the radiation chemistry of water is essential to be able to control the chemistry of nuclear technological applications. The homogeneous radiation chemistry of water is well known since many decades. However, radiation induced processes at solid-liquid interfaces remain to be fully understood, although they have been investigated to some extent by both experimental and theoretical studies.6, 9-11 Consequently, better understanding of these processes is a prerequisite for the development of materials used in nuclear technology. 1.2 Water radiolysis The interaction between ionizing radiation and water leads to ionization and electron excitation of water. Electrons emitted can lead to further ionizations which are also called secondary ionizations. Upon absorbing energy from fast charged particles or high energy photons, water is decomposed into its radiolysis products. Within 10−7 s after energy deposition, the primary radiolysis products are completely formed in a spur. The products will have diffused away from the spur and the interactions between the formed products in the spur become negligible.12 It should be noted that other oxidants (e.g. O2•−, HO2• and O2) are also formed from further reactions between the primary products.13 In pure water, the sequence of events that follows the deposition of energy is summarized in Figure 2.12 The radiation chemical yield, also called G-value, is defined as the amount of specified radiolysis product formed per 100 eV of energy deposited from ionizing radiation.12 The G-value is expressed in S.I. units as 𝐺(𝑥) = 𝑛𝑥 (1) 𝛿𝛿 where G(x) is the radiation chemical yield for the radiolysis species x and nx is the number of moles of x formed per unit of energy (δE) in Joules (J) deposited in the medium. 3 1 Introduction Time/s H 2O Excitation Ionization - H2O+ + H2O* e 10-16 HO + H3O+ H2 + O H 10-14 + OH 10-13 - eaq Formation of molecular products, diffusion of molecular species and radicals out of spur H2 HO H 2O 2 H 3O + HO - - H eaq 10-7 Figure 2. Time scale of events in water radiolysis resulting in the primary products. The radiolysis products of water and their chemical yields are well determined under a multitude of conditions. Generally accepted G-values for the primary products after 10−7 s are listed in Table 1. Table 1. Chemical yields (G-values) of radiolysis products in irradiated neutral water.12 Radiation G(H2) µmol J−1 G(H2O2) µmol J−1 G(eaq−) µmol J−1 G(H•) µmol J−1 G(HO•) µmol J−1 G(H3O+) µmol J−1 γ and fast e− 0.047 0.073 0.28 0.062 0.28 0.28 The G-value for a specific radiolysis product can be used to calculate its formation rate via the following equation Ċ(x) = Ḋ × G × ρ (2) where Ċ(x) is the production rate in mol dm–3 s–1, Ḋ is the dose rate in Gy s–1 and ρ is the density of the solvent. The S.I. unit Gray (Gy) represents the absorbed dose and 1 Gy corresponds to 1 J kg−1. 4 1 Introduction The powerful oxidizing (HO•) and reducing (eaq−) radiolysis products have approximately the same chemical yields. The interconversion between oxidizing and reducing condition can be simply performed by adding a reactant to the media. By this means, the media can be alternated to primarily oxidizing or reducing condition. For instance, the most readily and effective interconversion is that of eaq− to HO• through the following reactions in N2O-saturated solution.14 − eaq + N2O → N2 + O •− •− −1 9 O + H2O → HO + HO • k3 = 9.1 × 10 L mol − 6 −1 k4 = 1.8 × 10 L mol s s −1 −1 (3) (4) where k is the rate constant for the respective reaction. Under these conditions, G(HO•) is almost doubled to 6.0 x 10–7 mol J–1 and the system is alternated to oxidizing media.15 Another type of reagent is secondary alcohol which gives reducing condition upon reacting with HO• and H•. For instance, 2-propanol reacts with both radicals forming a strong reducing radical as follows.16 The rate constant k5 for this reaction is equal to 3.3 × 1012 L mol−1 s−1.16 H3C-CHOH-CH3 + HO (H ) → (CH3)2-C OH + H2O (H2) • • • (5) One of the main long-lived products of radiolysis of water is H2O2 which can react with metal/metal oxide surfaces both via redox reactions and catalytic decomposition.17-19 The relative impact of several radiolytic oxidants on the dissolution of UO2 was studied and H2O2 was determined to be the only radiolytical oxidant needed to be accounted for under geological repository conditions.20 Therefore, the interactions between H2O2 and metal/metal oxide surfaces have been treated as key reactions in the radiation induced processes at solid-liquid interfaces and have been investigated in a number of studies.1, 6, 21-26 By focusing on the key reactions, the study of the chemistry occurring at the interfaces can be understood. 1.3 Chemical kinetics of heterogeneous system The heterogeneous system, involving a solute and a solid phase, has more steps than the homogeneous system, e.g. diffusion and adsorption. 5 1 Introduction The reactions in the heterogeneous system can be divided into several steps as shown in Figure 3. Figure 3. Schematic representation of reactions in the heterogeneous system. Rsol and Rads are the reactants in solution and adsorbed on the surface, respectively; TSads is the transition state for the surface reaction; Pads and Psol represent the products of reaction adsorbed on the surface and in solution, respectively; kads, kSr and kdes denote the rate constants for the adsorption, surface reaction and desorption, respectively. For a non-catalytic process, the surface will undergo alterations, e.g. corrosion, dissolution, formation of complexes etc. On the other hand, for catalytic process, the catalyst surface will exhibit no alterations during the whole process. As can be seen in Figure 3, the process includes several steps: diffusion of the free reactant to the surface, adsorption onto the surface followed by the surface reaction, desorption of products into solution and products in solution (Psol) may react with free reactant (Rsol). Any of these steps may be rate-limiting.27 In general, the rate of a chemical reaction is dependent on the concentration of reactants and the rate constant for the reaction. In the case of a solute and a solid reactant, the reaction rate expression can be derived as Eq. (6) regarding the consumption rate of the solute and solid, respectively. − d[Solute] d𝑡 SA = 𝑘 � � [Solute] or − V d𝑛𝑠𝑠𝑠𝑠𝑠 d𝑡 = 𝑘(𝑆𝑆)[Solute] 6 (6) 1 Introduction where SA, V and k denote the surface area of solid, the solution volume and the reaction rate constant, respectively. In the excess of solid material, the kinetics of solute reactant consumption becomes pseudo first order. The pseudo first order rate constant can be obtained from the slope when plotting the logarithm of the solute reactant concentration versus reaction time. By repeating this procedure for a series of different SA/V, the second order rate constant can be obtained by plotting the pseudo first order rate constants against SA/V. The unit of the second order rate constant expressed in this way is m s−1. Powder suspensions are preferential for studying this heterogeneous system due to its high surface area to volume ratio and large flexibility in SA/V. The activation energy of the reaction can be obtained by determining the rate constants as a function of temperature according to the Arrhenius equation 𝑘 = 𝐴𝑒 −𝐸𝐸 ⁄𝑅𝑅 (7) where Ea represents the activation energy for the reaction; A is the preexponential factor. If the activation energy is low, the rate constant is mainly determined by the pre-exponential factor. Under this condition, the reaction rate is dominated by the rate constant of the diffusion of the reacting species in the system. Therefore, the pre-exponential factor is sometimes referred to as the diffusion controlled rate constant and can be considered as the highest possible rate of reaction. The diffusion rate for a heterogeneous system is significantly smaller than its homogeneous counterpart for the same reaction.28 In some systems, both catalytic and non-catalytic reactions may take place. The products from the non-catalytic process, e.g. corrosion, dissolution, formation of complexes, may react with the reactants in the solution, thereby affecting the kinetics of the whole process. 7 1 Introduction 1.4 Oxidative oxides dissolution of metals and metal The matrix of UO2-based spent nuclear fuel is UO2 (95%) while the remaining 5% are highly radiotoxic fission products and higher actinides. The release of these is dependent on the dissolution of the UO2 matrix after the initial dissolution of readily soluble radionuclides.29 Besides UO2, the dissolution behavior of other metals, metal oxides utilized in nuclear technological application is also of importance, such as copper which is used for the canister in the deep geological repository.5 Hence, it is worthwhile to understand the mechanism and kinetics of the oxidative dissolution of these metals/ metal oxides. H2O2, one of the key long-lived molecular radiolysis products of water, can react with metal/metal oxide surfaces both via redox reaction and catalytic decomposition.17-19 If the oxidized metals/metal oxides are soluble, the redox reaction is oxidative dissolution. For example, metallic tungsten has been proven to be readily dissolved in aqueous H2O2 solution30, 31 via a step-wise oxidation process from metallic state to W(VI) via W(III), W(IV) and W(V).32-34 Except for metals, H2O2 also leads to the dissolution of some metal oxides. For instance, insoluble UO2 can be oxidized to soluble UO22+ via a two-step single electron oxidation. It should be noted that hydroxyl radicals can be produced during the single electron oxidation from U(IV) to U(V) and this step is the rate limiting step of the oxidative dissolution of UO2.35 The dissolution process can be greatly enhanced in the presence of carbonate (HCO3-) via formation of soluble complexes.36 The carbonate concentration in Swedish groundwater is approximately 2 mM.37 It was proven that the oxidative dissolution rate of UO2 by H2O2 increases linearly with [HCO3-] from 0 to 1 mM, but becomes independent of [HCO3-] above 1 mM since the reaction becomes oxidation-limited.23 H2 is another major long-lived molecular radiolysis product of water12 and H2 is also expected to be produced in a large amount as a result of anaerobic corrosion of the iron insert38 in case of canister damage in the deep geological repository. H2 was proven to inhibit the radiation induced 8 1 Introduction dissolution of SNF and the inhibition process is significantly enhanced by the presence of noble metal nanoparticles of fission products in the SNF which are normally referred to as ε-particles (containing Mo, Ru, Tc, Pd and Rh).39-41 The scheme of the oxidative dissolution of UO2 and its inhibition by H2 is shown in Figure 4.42 Figure 4. Oxidative dissolution of UO2 and its inhibition by H2.42 In the groundwater environment of the SNF geological repository, sulfide is another common species, the concentration of which is between 1 and 2.1 μM.43 H2S is slightly soluble in water with pKa1 = 7.05, pKa2 = 14.0 at 293 K for HS− and S2−, respectively.44 Therefore, the main sulfide species in water are H2S and HS- and denote as sulfide throughout the whole thesis. Sulfide may inhibit the radiation induced dissolution of UO2 due to its reducing ability for U(VI).45, 46 The reaction stoichiometry determined in the literature46 is shown as follows: UO2 + HS → UO2 + S + H 2+ - 0 + (8) However, HS− can also scavenge radiolytic oxidants.29 Palladium is one of the fission products39-41 and sometimes used as a model for ε- 9 1 Introduction particles.25, 47 Palladium can catalyze the inhibition of oxidative dissolution of UO2 by H2 (see Figure 4).47 It was revealed by previous studies that palladium can be poisoned by sulfur compounds in aqueous solutions.48-50 Hence, it is necessary to investigate not only the inhibition of sulfide but also to study if the poisoning effect exists in the H2/Pd system. 1.5 Catalytic decomposition of H 2 O 2 at solid-liquid interfaces Except for redox reactions, the catalytic decomposition is the other type of interfacial reaction between H2O2 and metal/metal oxide surface which has been studied for decades both experimentally and theoretically.9, 10, 18, 51 Both reactions occur if the metal in its oxide, such as U in UO2,21-23, 51 is not in its highest oxidation state.51 In contrast, it was shown that when the metal is in its highest oxidation state only catalytic decomposition occurs at the interface9, 51, 52 with the exception of CeO2 which can be reduced to Ce(III) by H2O2.53 54 One of the most studied metal oxides regarding the catalytic decomposition of H2O2 is ZrO2.10, 11, 51, 52, 55 The 9, 109, 109, 109, 109, 109, 109, 109, 109, 109, 109, 109, 1010, 119, 1010, 1110, 11mechanism is summarized as follows9, 10: H2O2(ads) → 2HO (ads) • (9) H2O2(ads) + HO (ads) → H2O(ads) + HO2 (ads) (10) 2HO2 (ads) → H2O2 + O2 (11) • • • where ads represents the adsorbed state of the species. The overall reaction is H2O2 → H2O + 1/2O2 (12) It was demonstrated both experimentally and theoretically that the hydroxyl radical is the primary intermediate product during the catalytic decomposition of H2O2 at oxide interfaces.10, 51, 52 The half-life of the produced intermediate surface bound radicals could be sustained significantly to several hours or even up to few days.56, 57 The stability of 10 1 Introduction the materials in the presence of H2O2 may be affected by the strong reactive HO•(ads) at the interface. Hydroxyl radicals cannot be detected or quantified directly due to its short half-life and high reactivity. Hence, several probes for HO•, such as TAPS (N-Tris(hydroxymethyl)methyl-3-aminopropanesulfonic acid), methanol, Tris (Tris(hydroxymethyl) aminomethane), DMPO (5,5dimethyl-1-pyrroline-N-oxide), benzoic acid and 4-CP (4-chlorophenol)51, 52, 58-61 have been applied to detect the adsorbed hydroxyl radicals via different reactions and detection techniques (see Table 2). Table 2. The reaction types and detection techniques of common probes for the hydroxyl radical.24, 34-38 Probe Tris/Methanol/TAPS DMPO Benzoic acid/4-CP Reaction type Detection technique Oxidation UV-Vis/HPLC Addition ESR Hydroxylation HPLC UV-Vis: UV/Vis spectrophotometry; HPLC: high performance liquid chromatography; ESR: electron spin resonance Tris was selected as hydroxyl radical scavenger in a number of applications, such as interfacial radiation chemistry,52 catalytic reaction,51 photocatalysis62 and biological processes.63 As previously described,52 • HO abstracts hydrogen from Tris, yielding formaldehyde as one of the • products. The rate constant of the reaction between Tris and HO in homogeneous aqueous solution is 1.1 x 109 M–1 s–1.64 Formaldehyde can be detected and quantified by a modified version of the Hantzsch method.65, 66 Similar to Tris, methanol is another scavenger that also yields formaldehyde upon hydrogen abstraction by HO• with a rate constant of 9.7 x 108 M–1 s–1 in homogeneous solution.14 However, the rate constants for Tris and methanol reacting with surface bound hydroxyl radicals in the heterogeneous system are yet not determined. Although the kinetic data and mechanism for H2O2 decomposition on ZrO2 were elucidated, the reaction needs to be studied for a diversity of metals/metal oxides. 11 1 Introduction 1.6 Haber-Weiss reactions Besides the redox reactions and catalytic decomposition introduced in above sections, the reactions between H2O2 and one-electron redox couple of metal ion(s) become important when the amount of ion resulted from the oxidative dissolution is sufficient. The reaction is called HaberWeiss reactions and the mechanism is shown as follows67, 68 Mox + H2O2 → Mred + HO2 + H • + Mred + H2O2 → Mox + HO + HO • (13) − (14) Mox and Mred represents the oxidized and reduced forms of the oneelectron redox couple in metal complexes (e.g., Fe(III)/Fe(II), W(VI)/W(V)). Except in the aqueous phase, the reactions R(13) and R(14) can also occur in heterogeneous system at the solid-liquid interfaces, 35, 69, 70 e.g. hydroxyl radical is produced in the oxidative dissolution of UO2 by H2O2 as shown in Figure 4. The rate limiting step is R(13),67 the rate of which can be influenced by binding to ligands or the formation of polyperoxometalate.68, 71, 72 R(14) is also called a Fenton-like reaction. The Haber-Weiss reactions (e.g. H2O2/Fe(III)/Fe(II)) show the behavior of producing hydroxyl radical continuously without changing total concentration of iron ions.73, 74 The generation rate of hydroxyl radicals is governed by the steady-state concentration of Fe(II) which is mainly controlled by the total iron concentration.73 In the heterogeneous system in the presence of H2O2 and highly soluble metal/metal oxide, all the three mentioned reactions: oxidative dissolution, catalytic decomposition and Haber-Weiss reactions, may occur. A thorough study is needed on a single system including all of these reactions. 12 1 Introduction 1.7 Methodology for the detection of hydroxyl radicals As mentioned in the above section, some probes, e.g. Tris, TAPS and methanol, can be used to determine the amount of hydroxyl radicals by quantifying the amount of formed product (CH2O).51, 52 A number of methods for the determination of formaldehyde have been proposed, many of which are based on the Hantzsch reaction including the cyclization of amine, aldehyde and β-diketon to form a dihydropyridine derivative (Figure 5a).65 A modified version of the Hantzsch method was introduced by Nash66 using acetylacetone or 2,4-pentanedione to react with formaldehyde in the presence of ammonia to form a yellow derivative of 3,5-diactyl-1,4-dihydrolutidine (DDL). Both spectrophotometry66 and fluorometry75 can be utilized to measure DDL with high sensitivity. Though, this method is time-consuming and requires elevated temperature. Several new β-diketon based derivatization reagents of formaldehyde have been presented, such as 5,5dimethyl-1,3-cyclohexanedione (dimedone),76 4-amino-3-pentene-2-one (Fluoral-P),77 1,3-cyclohexanedione (CHD),76 methyl acetoacetate (MA)78 and acetoacetanilide (AAA) (Figure 5b)65. R2 CH + NH2 O R3 O C H H + +2 H +2 N H H O R C CH2 C O R1 H O N C H 3C O R C R1 N O C R + 2 H2O + 1/2 O2 R1 (a) R3 CH2 C R2 NH H 3C O O C O C N H NH + 2 H2O + 1/2 O2 (b) CH3 Figure 5. Scheme for the Hantzsch reaction (a) Using β-diketon as derivatization reagent; (b) Using acetoacetanilide (AAA) as derivatization reagent. As an excellent derivatization reagent, AAA has been adopted in a number of studies.10, 11, 51, 52, 79-81 In many of these studies, CH2O is formed in the presence of relatively high initial concentration of H2O2 (mM level). The similar reagent CHD is known to suffer positive interference from 13 1 Introduction H2O2 at H2O2 concentration on µM level.82-85 However, the interference from H2O2 on CH2O determination using AAA is still unknown. The yield (Y) of CH2O in a radiolysis experiment is defined (Eq. (15)) as the ratio between the concentration of scavenged hydroxyl radicals (equal to the measured formaldehyde concentration) and the accumulated concentration of hydroxyl radicals formed during the radiolysis of water which can be calculated based on Eq. (2). • Y = [CH2O]/C(HO ) (15) In N2O-saturated Tris solution, Y was measured to be 35% by using γ radiation of water as the source of hydroxyl radicals.52 However, the influencing factors (e.g. O2 and pH) for the yield of formaldehyde as well the mechanism of hydroxyl radical scavenging Tris were still not fully understood. The kinetics of H2O2 decomposition will be changed in the presence of Tris according to the abovementioned competition.52 However, the mechanism of catalytic decomposition of H2O2 at interfaces in the presence of hydroxyl radical scavenger remains to be elucidated. 1.8 The scope of this thesis The goal of this thesis is for further understanding of the radiation induced processes at solid-liquid interfaces. The kinetics and mechanism of the processes are mainly investigated by studying the reactions between the key water radiolysis product—H2O2 and metals/metal oxides. The reactions include oxidative dissolution of solid material, catalytic decomposition of H2O2 and Haber-Weiss reactions. The scheme is shown in Figure 6. Specifically, the present thesis is to answer the following questions which are crucial for better understanding of the interfacial processes. 1. Is the kinetics and mechanism of catalytic decomposition of H2O2 at interfaces the same for different metal oxides? 2. Does the dissolution of metals/metal oxides affect the mechanism of the studied heterogeneous system? 14 1 Introduction 3. Do the presence of hydroxyl radical probe, O2 and pH affect the mechanism of catalytic decomposition of H2O2 at interfaces? 4. What is the limitation of the methodology for quantifying the production of hydroxyl radicals in the studied heterogeneous system? The knowledge is not only relevant to nuclear technological applications, but also for applications related in photocatalysis, biochemistry, corrosion science, catalysis and optics/electronics. Ultimately it is my wish to provide a better vision for material scientists regarding the reactivity of radiolysis products upon the surface of materials on the basis of the present thesis. Figure 6. Scheme of radiation induced processes at solid-liquid interfaces. 15 2 Experimental details 2 Experimental details 2.1 Instrumentation The specific surface areas of the powders were determined by using the Brunauer-Emmett-Teller (B.E.T.) method of isothermal adsorption and desorption of a gaseous mixture consisting of 30% N2, 70% He in a Micrometrics Flowsorb II 2300 instrument. The samples were weighed using a Mettler Toledo AT261 Delta Range microbalance. Structural studies were done by powder XRD (X-ray diffraction) method with PANalytical X'Pert PRO diffraction system using BraggBrentano geometry in the 2θ angle from 10° to 100° and CuKα irradiation (λ = 1.54 Å). Surface morphology of samples was obtained using SEM (Scanning electron microscopy) analysis with an FEI-XL 30 Series instrument. The accelerating voltage was 20 kV. Elemental analysis was done using XPS (X-ray photoelectron spectroscopy). The spectra of samples were recorded with a Kratos Axis Ultra electron spectrometer with a delay line detector. The depth of analysis of metal oxides/hydroxides was about 6 nm. The element detection limit was typically 0.1 at. %. Trace elemental analysis on all solutions were conducted by using Thermo Scientific iCAP6000 series ICP-OES (Inductively coupled plasma optical emission spectroscopy). The analysis for tungsten sample was performed at wavelengths of 207.9, 209.8 and 224.8 nm using ICP element standard IV from Sigma Aldrich. Gamma-irradiation was conducted in a MDS Nordion 1000 Elite Cs137 γ-source with a dose rate of 0.15 Gy s–1, quantified by Fricke dosimetry.12 The powder suspensions were stirred using a Heidolph MR3001K magnetic stirrer. The temperature was controlled throughout the 16 2 Experimental details experiments using a Lauda E100 thermostat, calibrated against a Therma 1 Thermometer coupled to a submersible K-type (Ni-Cr-Ni) temperature probe, with a precision of ± 0.1 K. UV/Vis spectra were collected by a Thermo Scientific Genesys 20 UV/Vis spectrophotometer. The pH of solution was measured by Metrohm 713 pH Meter with an accuracy of ± 0.1, calibrated by standard pH reagents. The experiments were performed either in open air or in different gaseous atmosphere, N2O (AGA Gas AB), 20% O2 in N2O (N2O 80 ± 1.0%, Strandmøllen A/S), N2 (≥ 99.999%, Strandmøllen A/S) and H2 (AGA Gas AB). 2.2 Materials and reagents All the solutions used in the thesis were prepared using Millipore MilliQ water (18.2 MΩ cm). The major information about the powders used is listed in Table 3. All the powders were used without further purification. The B.E.T. specific surface area of the powders is the average value of three measurements obtained by averaging the measured sorption and desorption isotherm values. Table 3. The B.E.T. specific surface areas for the powders. Powder Information B.E.T surface area/m2 g−1 ZrO2 CeO2 HfO2 Gd2O3 Fe2O3 CuO Cu2O Cu W WO3 CAS[1314-23-4], Aldrich 99% CAS[1306-38-3], Alfa Aesar 99.99% CAS[12055-23-1], Alfa Aesar 99.95% CAS[12064-62-9], Aldrich 99.9% CAS[1309-37-1], Aldrich 99% CAS[1317-38-0], Aldrich 99.99% CAS[1317-39-1], Aldrich, 99,9 % CAS[7440-50-8], Alfa Aesar, 99,9 % CAS[7440-33-7], Aldrich 99.9%, CAS[1314-35-8], Aldrich 99.9% 5.0 ± 0.2 14.3 ± 1.0 10.0 ± 0.1 1.7 ± 0.1 9.0 ± 1.0 15.3 ± 0.1 0.46 ± 0.01 0.089 ± 0.002 0.11 ± 0.01 3.51 ± 0.01 17 2 Experimental details A UO2 pellet supplied by Westinghouse Atom AB was washed by 10 mM NaHCO3 prior to use in the experiments. Pd powder (Aldrich, average particle size 0.625 μm) was used as the catalyst for H2 in inhibiting dissolution of UO2. H2O2 solutions were prepared from a 30% standard solution (Merck). Reference formaldehyde solution was prepared from a stock solution (CAS[50-00-0], Aldrich 37 wt% in H2O). 2.3 Basic methodology The concentration of U(VI) was determined by the Arsenazo III method86, 87. Arsenazo III (1,8-dihydroxynaphtalene-3,6-disulphonic acid-2,7-bis[(azo-2)-phenylarsonic acid]) and U(VI) can form a colored complex which can be quantified by a UV-Vis spectrophotometer at wavelength 653 nm. The detection limit is 5 x 10−7 M.88 The concentration of H2O2 was determined using the Ghormley triiodide method89, 90 by mixing samples with the detection reagents ammoniumdimolybdate (Alfa Aesar, 4% w/v) and potassium iodide (VWR BDH Prolabo, 99.0%). The absorbance of the product I3- was measured at the wavelength of 350 nm using a UV-Vis spectrophotometer. Two linear calibration curves were obtained by plotting the H2O2 concentration against the absorbance of I3– in the concentration range 0.01 – 1.0 mM (sample volume 0.2 mL) and 1.0 – 5.0 mM (sample volume 0.02 mL) respectively. The uncertainty associated with the determination of the concentration of H2O2 from the actual concentration was < 2%. Produced hydroxyl radicals in the experiments were scavenged by Tris(hydroxymethyl) aminomethane (CAS[77-86-1], BDH Chemicals, 99%) or methanol (CAS[67-56-1], Aldrich, 99.9%) generating formaldehyde. Formaldehyde was quantified by the modified version of the Hantzsch method65, 66 using derivatization reagents acetoacetanilide (AAA) (CAS[102-01-2], Aldrich, ≥98%) and ammonium acetate (AA) (CAS[631-61-8], Aldrich, ≥98%). Specifically, 1 mL AAA (0.2 M, dissolved in ethanol (99.5%)), 2.5 mL AA (4 M) and 1.5 mL sample were mixed and 18 2 Experimental details incubated at 40°C for 15 min in a testing tube so as to form the DDL derivative (Figure 5b) which can be measured at the wavelength of 368 nm by a UV-Vis spectrophotometer. A linear calibration curve was obtained by plotting the absorbance of derivative against formaldehyde concentration (in bulk),52 ranging from 0.15 µM to 1 mM. As reported earlier, the uncertainty of this method is < 2%.52 All samples in suspensions were filtered through a 0.2 μm cellulose acetate filter and pH of solutions was adjusted to 7.5 by HCl/NaOH in the presence of Tris unless otherwise specified. 2.4 Oxidative oxides 2.4.1 dissolution of metals and metal Oxidative dissolution of W by H2O2 To study the reaction between W and H2O2, W powder (SA/V = 550 – 8800 m–1) was submerged into H2O2 (5 mM) and/or Tris (100 mM) solution at room temperature. In parallel, the dissolution behavior of WO3 was also examined as a comparison to metallic W powder. The experiments were conducted in the dark at ambient temperature, N2purged solution and continuous magnetic stirring at a rate of 600 rpm. The pH of suspension was adjusted to 7.5 by HCl/NaOH after mixing Tris and W powder. In the case without Tris, no pH adjustment was done for the suspensions. After 30 min pre-mixing, H2O2 was added to the suspension to initialize the reaction. The amount of dissolved tungsten in the samples filtered at selected time points was measured via ICP-OES. 2.4.2 Oxidative dissolution of W by O2 Similar to the experiments for the oxidative dissolution of W by H2O2, the air-saturated suspension of W powder and Tris or methanol (“free” pH) were also undertaken the ICP-OES measurements for the concentration of dissolved tungsten. 19 2 Experimental details 2.4.3 Inhibition of radiation induced dissolution of UO2 by sulfide The inhibition of radiation induced dissolution of UO2 by sulfide was studied in experiments performed at room temperature by using a UO2 pellet exposed to γ irradiation in aqueous solution containing sulfide and NaHCO3. The U(VI) concentration of samples taken at selected time points was measured. Besides the pellet experiments, the inhibition of sulfide was studied further by mixing U(VI) solution and sulfide directly. The mixture of palladium powder (11 mg/100 mL) and 4 mM NaHCO3 was purged with N2 for 20 minutes before adding to the autoclave. After purging with N2 for another 10 minutes and then pressurizing with pure N2 or H2 (20 bar), the autoclave was left overnight with continuous stirring (60 rpm). UO2(NO3)2 solution and H2S saturated solution were added to initialize the reaction. The obtained initial concentrations of U(VI) and sulfide (represents total H2S/HS− throughout the whole thesis) were 150 μM and 1 or 2 mM, respectively. The inhibition of sulfide was compared to H2, the present experiments were divided into several groups: 1) N2 and sulfide, 2) N2, sulfide and Pd, 3) H2 and Pd, 4) H2 and sulfide, 5) H2, Pd and sulfide. The U(VI) concentrations of filtered samples taken at selected time points were determined. 2.5 Catalytic decomposition of H 2 O 2 on metal and metal oxide surfaces 2.5.1 Metals Copper and tungsten powder were used to study the catalytic decomposition of H2O2 at solid-liquid interfaces. In the thesis, the mechanistic study of such reaction on metal surface is the primary task. Both the H2O2 and CH2O concentrations were measured throughout the whole reaction (room temperature) in the suspension containing Cu (SA/V = 4800 m−1)/H2O2 (5 mM)/methanol (100 mM) or W (SA/V = 550 - 8800 m−1)/H2O2 (5 mM)/Tris (100 mM). 20 2 Experimental details 2.5.2 2.5.2.1 ZrO2 Effect of Tris The experiments were performed in N2-saturated H2O2/ZrO2/Tris suspensions in sealed vials with magnetic stirring at a rate of 750 rpm at room temperature. SA(ZrO2)/V was fixed at 4.5 x 105 m−1 which is in line with the previous study.10 The initial concentration of Tris varied from 20 mM to 400 mM while that of H2O2 varied from 0.5 mM to 5 mM. The sample was only taken and filtered when the reaction was complete (after 16 h). The concentration of CH2O in the filtered sample was measured. The model for H2O2 decomposition on the surface of ZrO2 in the presence of Tris was built up using the kinetic simulation software Gepasi 3 which has been used in kinetic simulation studies.91, 92 All the relevant kinetic parameters in the model were adapted to the reported experimental works.52, 93 2.5.2.2 Effect of O2 For investigating the O2 effect on the catalytic decomposition of H2O2 on ZrO2 surface, the experiments were performed with N2-purged (O2 free) or air-saturated suspensions for methanol (20 mM, pH 7.5), Tris (20 mM, pH 7.5) and a mixture of the two scavengers (20 mM methanol + 5 mM Tris, pH 7.5). The initial concentration of H2O2 and SA(ZrO2)/V were fixed at 5 mM and 4.5 x 105 m−1, respectively. Both the concentrations of H2O2 and CH2O were monitored during the reaction. 2.5.2.3 Effect of pH For the pH effect, the experiments were only performed in the presence of Tris under the conditions: pH 7.0 – 9.0 (within buffering range) and air-saturated. The other procedures were the same as the above section on the effect of O2. 21 2 Experimental details 2.5.3 2.5.3.1 Other lanthanide and transition metal oxides Kinetic study The reaction suspension contains 0.5 mM H2O2 and different oxides with different masses at different temperatures (298 – 353 K): Fe2O3 (0.2 – 1.5 g), HfO2 (0.1 - 0.75 g), Gd2O3 (0.25 - 1.0 g), CuO (0.0025 - 0.1 g). A volume of 0.2 ml of filtered sample at selected time point was used for the measurement of H2O2 concentration. 2.5.3.2 Mechanistic study The reactions between the oxides and H2O2 were performed at room temperature in the presence of 50 mL N2-purged Tris solution (20 mM). The suspension contains H2O2 (5 mM), Tris (20mM) and different amount of oxides: Fe2O3 (1.5 g) or CeO2 (1.6 g) or HfO2 (2.25 g) or Gd2O3 (3.0 g) or CuO (0.06 g). Both the concentrations of H2O2 and CH2O were measured in the filtered samples at selected time points. 2.6 Haber-Weiss reactions For the study of the homogeneous Haber-Weiss reactions between H2O2 and dissolved tungsten, the experiment procedure was the same as above (Section 2.5.1) in the heterogeneous system before the filtration. At predetermined time point, a volume (10 or 20 mL) of sample was extracted and then stored in a dark vial. The concentrations of H2O2 and CH2O in the filtered solution as well as in the original suspension were monitored and compared. In order to further investigate the homogeneous reactions, extra H2O2 was added to the abovementioned filtered solution when H2O2 was consumed completely. The concentrations of H2O2 and CH2O were monitored. Besides, the amount of dissolved tungsten prior to and after the addition of H2O2 was examined. 22 2 Experimental details 2.7 Methodology for the detection of hydroxyl radicals 2.7.1 The O2 and pH effects on probes for hydroxyl radicals The O2 and pH effects on the scavenging yield (Eq. (2)) of the probes (Tris and/or methanol) for hydroxyl radicals were investigated by varying dissolved O2 concentration and pH value of the solution. The experiments were performed in N2O- or (80% N2O + 20% O2)- saturated aqueous Tris (20 mM) or methanol (20 mM) solution which were subjected to γ irradiation. The pH effect was only examined in the Tris case and the pH was varied from 7.0 to 9.0. The G-value of hydroxyl radical is doubled to 6.0 x 10–7 mol · J–1 since all the solvated electron was converted to hydroxyl radical by reacting with N2O according to R(3)and R(4).14, 15 The accumulated concentration of CH2O in the sample taken at selected time point was measured, which can be used to calculate the corresponding yield based on Eq. (2). 2.7.2 The interference of H2O2 on the modified Hantzsch method using AAA as derivatization reagents The interference of H2O2 on the modified version of the Hanztsch method52 was investigated by adding concentrated H2O2 solution to a reference CH2O solution with known concentration. The CH2O concentration in the mixed solution (H2O2 + CH2O) was measured by the procedure described in section 2.3. Additional experiments were conducted by adding concentrated H2O2 to the mixed solution (H2O2 + CH2O) after the incubation time (15 min). The absorbance of the formed derivative after the incubation was monitored after the addition of H2O2. 23 3 Results and discussion 3 Results and discussion A number of metals and metal oxides (W, Cu, UO2, ZrO2, CeO2, HfO2, Gd2O3, CuO, Cu2O and Fe2O3) were investigated regarding the interaction with H2O2 in the aqueous system which includes oxidative dissolution, catalytic decomposition and Haber-Weiss reactions. Depending on the oxidation state as well as the solubility of the materials, one or more of these processes can occur in the heterogeneous system containing H2O2 and metal/metal oxide powder. The kinetics, mechanism and the factors influencing the processes were investigated. Furthermore, since the hydroxyl radical is the intermediate product of all of these three processes, the studies on the probes for hydroxyl radicals are crucial. Here, the focus was on factors influencing the scavenging yield, the scavenging mechanism, the influence of the probes on the mechanism of the studied system and the influence of H2O2 on the accuracy of the Hanztsch method for the quantification of hydroxyl radical. 3.1 Methodology for the detection of hydroxyl radicals 3.1.1 3.1.1.1 The effects of O2 and pH on probe chemistry III The effect of O2 The experiments for studying the O2 effect on the homogeneous aqueous system were performed under different conditions. The dissolved O2 concentration (DO) in this work is kept constant at 0.26 mM (Henry’s law constant is 769 atm dm3 mol–1 at 298 K).94 Hydroxyl radicals are produced by using γ-irradiation on the solution. The concentration of produced formaldehyde as a function of reaction time in the presence and absence of O2 for both methanol and Tris is presented in Figure 7. 24 3 Results and discussion 0.6 [CH2O] / mM 0.5 0.4 0.3 0.2 0.1 0 0 1000 2000 3000 4000 5000 6000 Time / s Figure 7. Concentration of CH2O as a function of reaction time for the γ-radiolysis of water in different reacting solutions: methanol (20 mM, N2O/O2, unadjusted pH) ( mM, N2O, unadjusted pH) ( ), Tris (20 mM, N2O/ O2, pH 7.5) ( ), methanol (20 ), Tris (20 mM, N2O, pH 7.5) ( ). Hydroxyl radicals formed in the radiolysis of water calculated from Eq.(2) ( American Chemical Society ). © As can be seen in Figure 7, the concentrations of CH2O are higher in the presence of O2 both in the methanol and Tris cases. Specifically, the presence of O2 leads to a five-fold (14% to 68%) and two-fold (16% to 29%) increase in the formaldehyde yield in methanol and Tris, respectively. After evaluating the O2 effect on the homogeneous system, the effect on the heterogeneous system was also investigated under different conditions. The production of CH2O as a function of reaction time is shown in Figure 8. 25 3 Results and discussion 0.14 0.12 [CH2O] / mM 0.1 0.08 0.06 0.04 0.02 0 0 5000 10000 15000 20000 25000 30000 35000 40000 Time / s Figure 8. Concentration of CH2O as a function of reaction time for the reaction of H2O2 (initial concentration 5 mM) with ZrO2 (4.5 g) under different conditions: methanol (20 mM, air, pH 7.5) ( ), methanol (20 mM, N2, pH 7.5) ( mM, N2, pH 7.5) ( ), Tris (20 mM, air, pH 7.5) ( ), mixture (20 mM methanol + 5 mM Tris, air, pH 7.5) ( mM methanol + 5 mM Tris, N2, pH 7.5) ( American Chemical Society ), Tris (20 ), mixture (20 ) and Tris (5 mM, air, pH 7.5) ( ). © 2014 As can be seen in Figure 8, the production of CH2O in the Tris case is enhanced in the presence of O2. However, this enhancement is insignificant in the methanol case and not detectable in the mixture. As compared to the O2-dependence experiments in the homogeneous solution (see Figure 7), the O2 effect observed in the heterogeneous system is much weaker, especially for methanol. The reason for this inconsistence in the methanol case should be that H2O2 in the present system competes with O2 to react with •CH2OH forming CH2O.95, 96 The hydroxymethyl radical is an O2-sensitive precursor for CH2O and plays an important role in both the Tris and methanol case. However, the O2-free system is not oxidant-free. Therefore, addition of O2 only has marginal effect on the production of CH2O. In general, the production of CH2O increases in the order: methanol < Tris/methanol mixture < Tris. The production of CH2O from the mixture is approximately the sum of the 26 3 Results and discussion individual contributions from its constituents. Hence, stabilizing the pH by addition of Tris, does not appear to influence the yield of formaldehyde from methanol. The higher production in the presence of Tris than methanol may be attributed to the difference in affinity towards ZrO2 surface. The difference between their affinities towards TiO2 has already been illustrated in a photocatalysis system.62, 97 Firstly introduced in Paper II, a plot was employed to distinguish the effect of different oxides on the production of CH2O as a function of H2O2 conversion. A slight revision was made by using normalized evolution of CH2O ([CH2O]t/[CH2O]f) instead. [CH2O]t and [CH2O]f represent the concentration of CH2O at a selected time point and the final concentration, respectively. [CH2O]t/[CH2O]f is plotted against H2O2 conversion for different scavengers and the results are shown in Figure 9. % of [CH2O]t/[CH2O]f 100 80 60 40 20 0 0 20 40 60 80 100 % of H2O2 converted Figure 9. [CH2O]t/[CH2O]f as a function of conversion of H2O2 under different conditions (all with ZrO2 (4.5g): methanol (20 mM, air, pH 7.5) ( Tris (20 mM, air, pH 7.5) ( ), methanol (20 mM, N2, pH 7.5) ( ), Tris (20 mM, N2, pH 7.5) ( mM Tris, air, pH 7.5) ( ) and mixture (20 mM methanol + 5 mM Tris, N2, pH 7.5) ( 2014 American Chemical Society 27 ), ), mixture (20 mM methanol + 5 ). © 3 Results and discussion It can be seen from Figure 9 that there is no clear oxygen-dependent variation for any of the scavengers. This similarity indicates that O2 has minor influence on the evolution of formaldehyde as a function of the consumption of H2O2 3.1.1.2 The effect of pH A thorough study regarding the pH effect in homogeneous aqueous solution was performed only for Tris since methanol cannot buffer the solution pH. The experiment was performed under the conditions: 20 mM Tris, pH = 7.0 – 9.0 (pH buffering range of Tris), 20% O2 in N2O and room temperature. The accumulated concentrations of CH2O are plotted as a function of reaction time in Figure 10. 0.3 [CH2O] / mM 0.25 0.2 0.15 0.1 0.05 0 0 1000 2000 3000 4000 5000 6000 Time / s Figure 10. Concentration of CH2O as a function of reaction time for the γ-radiolysis of water in Tris (20 mM) with pH 7.0 ( ), 7.5 ( ), 8.0 ( ), 8.5 ( ) and 9.0 ( ). All the experiments were purged with mixed N2O/O2. © 2014 American Chemical Society It is clearly seen from Figure 10 that the yield of formaldehyde increases with increasing pH and is doubled by raising pH from 7.0 to 9.0. This fairly strong pH-dependence may indicate that the reaction yielding formaldehyde is to some degree base-catalyzed. 28 3 Results and discussion In addition, the pH effect on the heterogeneous system was investigated with varied pH. The initial and final pH of each experiment was measured and it was observed that the pH values were stable in all cases except of the lowest pH. However, the changes in pH during the course of the experiments are negligible. The production of CH2O is plotted as a function of reaction time in Figure 11. 0.4 0.35 [CH2O] / mM 0.3 0.25 0.2 0.15 0.1 0.05 0 0 5000 10000 15000 20000 25000 30000 Time / s Figure 11. Concentration of CH2O as a function of reaction time for the reaction of H2O2 (initial concentration 5 mM) with ZrO2 (4.5g) in Tris (20 mM) with pH: 7.0 ( ( ), 8.5 ( ) and 9.0 ( Chemical Society ), 7.5 ( ), 8.0 ). All the experiments were exposed to air. © 2014 American It can clearly be seen from Figure 11 that the production of CH2O increases with increasing pH but the pH effect is more pronounced than in the homogeneous solution (see Figure 10). It should be noted that a 4-fold increase in the production of CH2O is displayed when increasing pH from 7.0 to 7.5. The CH2O production at pH 9.0 is almost one order of magnitude higher than the lowest one (pH 7.0). Solution pH can not only influence the yield of CH2O as in the homogeneous system, but also change the surface charge and the adsorption of the scavenger in the 29 3 Results and discussion heterogeneous system, thereby altering the relative coverage of reactants on the surface. The normalized evolution of CH2O as a function of the conversion of H2O2 in the pH range under study is shown in Figure 12. % of [CH2O]t/[CH2O]f 100 80 60 40 20 0 0 20 40 60 80 100 % of H2O2 converted Figure 12. [CH2O]t/[CH2O]f as a function of conversion of H2O2 (initial concentration 5 mM) in Tris (20 mM) with pH: 7.0 ( Chemical Society ), 7.5 ( ), 8.0 ( ), 8.5 ( ) and 9.0 ( ). © 2014 American As can clearly be seen in Figure 12, pH does not induce a change in the relative dynamics of the formation of CH2O. 3.1.1.3 Formaldehyde production mechanism by Tris and hydroxyl radical Based on the experimental results, a formaldehyde production mechanism by Tris and hydroxyl radical is proposed as in Figure 13. The scheme consists of two independent routes: O2- and pH- dependent routes. Each route contains several possible reacting pathways. 30 3 Results and discussion HO O (a) + HO HO HO HO NH HO HO H O NH C H HO NH2 + H HO HO NH2 HO (b) β-scission O + HO HO (Tris) NH2 O + C NH2 H OH HO H 1, 2 -H sh ift + HO HO (c) HO +O2 HO HO HO NH2 HO O -HO2 CHOO NH2 NH2 HO NH2 HO HO (d) Cβ-Cγ cleavage HO OH + +O2 CH2 OH -HO2 O OOCH2OH C H H Figure 13. Scheme for formaldehyde production mechanism by Tris. © 2014 American Chemical Society As shown in the scheme, aminyl, alkoxyl and hydroxyalkyl radical are formed via hydrogen abstraction from Tris by hydroxyl radicals. The alkoxyl radical can undergo 1, 2-hydrogen shift reaction and be converted to a hydroxyalkyl radical.98, 99 Formaldehyde can be formed from the aminyl radical via a fivemembered cyclic transition state and this route is base-catalyzed and O2independent (Reaction (a)).100 In addition, formaldehyde can be formed directly from the alkoxyl radical via an O2-independent β-scission reaction (Reaction (b)). Furthermore, the hydroxyalkyl radical can undergo two O2-dependent pathways (Reaction (c) and (d)) producing aldehyde and formaldehyde, respectively. To produce formaldehyde, the radical may undergo a fragmentation reaction by the cleavage of the Cβ-Cγ bond (Reaction (d)) yielding •CH2OH—the precursor for formaldehyde in O2-containing systems. The Cβ-Cγ cleavage reaction of alkyl radicals has been reported101, 102 and may also take place in the present system. 31 3 Results and discussion The effect of Tris on the catalytic decomposition of H2O2 on ZrO2 surface IV 3.1.2 The two competing reactions: O2 and H2O2 oxidizing •CH2OH and H2O2 and Tris reacting with HO•, were evaluated under different conditions. The final production of CH2O is plotted against [H2O2]0 in Figure 14. 0.40 0.35 2 f [CH O] / mM 0.30 0.25 0.20 0.15 0.10 0.05 0.00 0 1 2 3 [H2O2]0 / mM 4 5 Figure 14. Final production of CH2O as a function of initial concentration of H2O2 under different conditions: (N2 and 100 mM Tris) (black squares), (Air and100 mM Tris) (red circles), (N2 and 20 mM Tris) (blue up triangles), (Air and 20 mM Tris) (cyan down triangles). © 2015 Elsevier B. V. In Figure 14, it can be seen that at a given [H2O2]0, higher [Tris]0 leads to higher [CH2O]f both in the presence and absence of O2. With a given [Tris]0, [CH2O]f increases with increasing [H2O2]0. Nevertheless, it is apparent that a 5-fold increase in [Tris]0 or [H2O2]0 does not result in a 5-fold increase in [CH2O]f (only by a factor of 2~3 in the studied concentration range). This obvious inconsistency is due to the fact that formaldehyde is formed via a surface reaction and the competition between hydroxyl radicals reacting with Tris and H2O2 is governed by relative surface coverage rather than relative bulk concentrations. 32 3 Results and discussion Based on the assumption that ZrO2 behaves similarly as TiO2,62 i.e., H2O2 has higher affinity than Tris to ZrO2, a modified mechanism (R16)R20)) must be employed which includes two different surface sites: Sa and Sb for H2O2 and Tris, respectively. This mechanism constitutes the basis for numerical simulation of the kinetics of the H2O2/ZrO2/Tris system. All kinetic parameters are assigned on the basis of experimental data or reasonable estimations and listed in Table 4. The detailed deduction process of some parameters is presented in Paper IV.11 Table 4. Modified mechanism of H2O2 decomposition on ZrO2 surface in the presence of Tris and the related parameters. © 2015 Elsevier B. V. No. Reaction Constant Initial Conc./mM -2 -1 -1 R16 H2O2 + 2Sa ⇌ H2O2(Sa) k16f = 5.4 x 10 M s k16r = 9 x 10-7 s-1 K16 = 6 x 104 M-1 R17 H2O2(Sa) →2HO•(Sa) K17 = 3 x 10-4 s-1 R18 R19 R20 Tris + Sb ⇌ Tris(Sb) k18f = 5 x 10-5 M-1 s-1 k18r = 5 x 10-7 s-1 K18 = 1 x 102 M-1 HO•(Sa) + H2O2(Sa) → H2O + HO2• + 3Sa K19 = 1 x 108 M-1 s-1 HO•(Sa) + Tris(Sb) → CH2O + Sa + Sb [H2O2]0 = 0.5-5 [Sa]0 = 2 [Tris]0 = 20, 100 [Sb]0 = 1 K20 = 1 x 108 M-1 s-1 As shown in Table 4, the sensitivity of the systems at the reactant concentrations and adsorption capacity used in the simulations can be summarized as: the surface is saturated with H2O2 and Tris under the present conditions. In this case, increasing the adsorption equilibrium constants (K16 and K18) has negligible effect on the dynamics of the system. However, decreasing K16 results in higher final production of formaldehyde whereas decreasing K18 results in lower final production of formaldehyde. This is directly resulted from the competition between H2O2 and Tris for surface bound hydroxyl radicals. In addition, the ratio between k19 and k20 also has a direct impact on the final production of formaldehyde. Additional simulations were performed by using the kinetic simulation software Gepasi 3.091, 92 to further evaluate the model. The initial Tris and 33 3 Results and discussion 5 0.10 4 0.08 3 0.06 2 0.04 1 0.02 0 0 [CH2O] / mM [H2O2] / mM H2O2 concentration in the simulation were kept the same as in the experiments. The experimental dynamics of H2O2 decomposition is presented in Figure 15, consisting of the time-resolved H2O2 concentration and accumulated CH2O concentration. The dynamics of simulation is shown in Figure 16a and Figure 16b. In addition to [H2O2] and [CH2O], the time-resolved [H2O2(Sa)] and [Tris(Sb)] in the simulations are also presented. 0.00 5000 10000 15000 20000 25000 30000 35000 40000 Time / s Figure 15. Concentration of H2O2 (black squares) and the production of CH2O (red circles) as a function of reaction time in suspension with H2O2 (initial concentration 5 mM), ZrO2 (SA/V = 4.5 x 105 m-1) and 20 mM Tris at pH 7.5. © 2015 Elsevier B. V. By comparing Figure 16a to Figure 15, it can be observed that the dynamics in the simulation resembles the experimental counterpart. More specifically, a fast decrease of the concentration of H2O2 in the initial period can be seen in both cases indicating that H2O2 adsorbs onto ZrO2 surface rapidly. The adsorption step is more readily observed in Figure 16c and Figure 16d where the logarithmic axis is applied. 34 3 Results and discussion 4 [CH2O] [H2O2] [H2O2](Sa) [Tris](Sb) b 4 4 5 4 3 3 3 3 2 2 2 2 1 1 1 1 0 0 5 Concentration / mM 5 5 5000 10000 0 0 20000 0 15000 [CH2O] [H2O2] [H2O2](Sa) [Tris](Sb) c 4 5 5 5000 10000 0 20000 15000 [CH2O] [H2O2] [H2O2](Sa) [Tris](Sb) d 4 4 5 4 3 3 3 3 2 2 2 2 1 1 1 1 0 100 101 102 103 104 0 0 0 100 Time / s Concentration / mM Concentration / mM [CH2O] [H2O2] [H2O2](Sa) [Tris](Sb) Concentration / mM a 5 101 102 103 104 Time / s Figure 16. The simulated concentrations of different species as a function of reaction time: [CH2O] (black), [H2O2] (red), [H2O2(Sa)] (blue), [Tris(Sb)] (cyan). (a) 20 mM Tris, (b) 100 mM Tris, (c) 20 mM Tris and logarithmic x axis, (d) 100 mM Tris and logarithmic x axis. © 2015 Elsevier B. V. According to the higher adsorption constant (K16) and forward reaction rate constant (k16f), H2O2 can reach the maximum surface coverage instantly (~ 10 s) and remain as the dominating species for approximately 6000 s. Although slower than the adsorption of H2O2, Tris can saturate the surface in approximately 500 s at a concentration of 100 mM and the saturated surface concentration ([Tris(Sb)]) is approximately equal to the saturated surface concentration of H2O2 ([H2O2(Sa)] = 1 mM). In general, the H2O2 concentration exhibits a fast initial drop due to the adsorption upon ZrO2 surface. Then the decomposition of H2O2 becomes the rate limiting step, following first-order kinetics. Not surprisingly, in the last period when [H2O2] in solution approaches zero, CH2O is still produced until H2O2 on the surface diminishes completely. The ratio between final production of formaldehyde and initial concentration of H2O2 in the simulations as well as in the experiments 35 3 Results and discussion 0.12 1.2 0.10 1.0 0.08 0.8 0.06 0.6 0.04 0.4 0.02 0.2 [CH2O]f/[H2O2]0 [CH2O]f/[H2O2]0 (under N2 atmosphere) is plotted against [H2O2]0 and the results are presented in Figure 17. 0.0 0.00 0 1 2 3 [H2O2]0 / mM 4 5 Figure 17. The experimental and simulation [CH2O]f/[H2O2]0 as a function of [H2O2]0 under different conditions: 100 mM Tris, experimental (black squares), 20 mM Tris, experimental (black circles), 100 mM Tris, simulation (red up triangles), 20 mM Tris, simulation (red down triangles). © 2015 Elsevier B. V. It can be seen from Figure 17 that, the trend in the simulation case is similar to the trend in the experimental case. [CH2O]f/[H2O2]0 decreases with increasing [H2O2]0 which is due to the competition between Tris and H2O2 in the reaction with the surface bound hydroxyl radical. Nonetheless, it should be noted that the experimental results are approximately one order of magnitude lower than the CH2O productions in the simulation (8% and 12% for 20 and 100 mM Tris, respectively). This difference can be attributed to the low experimental yield in converting hydroxyl radical to CH2O. As being determined in Paper III, 16% is the yield in homogeneous system.55 In the simulations the yield is set as 100%. However, [CH2O]f/[H2O2]0 is higher at 100 mM than 20 mM Tris in both cases. 36 3 Results and discussion [CH2O]t/[CH2O]f is plotted against H2O2 conversion for the simulation data and compared to the corresponding experimental curve55 (as presented in Figure 18). 100 [CH2O]t/[CH2O]f / % 80 60 40 20 0 0 20 40 60 Converted H2O2 / % 80 100 Figure 18. The experimental (black) 55 and simulation (red) normalized evolution of CH2O as a function of conversion of H2O2. [Tris]0 in both cases are set as 20 mM. © 2015 Elsevier B. V. As can be seen, the curve based on simulation data is similar to the experimental one. In the initial adsorption-governed period, the reaction between hydroxyl radicals and the adsorbed H2O2 (R(19)) is the primary reaction which is attributed to the slower adsorption of Tris. As a result, the initial rate of CH2O-formation is relatively low. However, the rate in the experimental case is non-zero which is attributed to the fact that Tris was added 30 min prior to the addition of H2O2. As H2O2(Sa) and Tris(Sb) both reaches a steady-state concentration on the surface, the formation rate of CH2O increases and remains constant until the steady-state concentrations diminish. 37 3 Results and discussion 3.1.3 The interference of H2O2 on the modified Hantzsch method using AAA as derivatization reagents VI Experiments were performed regarding the influence of H2O2 on the detection of CH2O at different [H2O2]0/[CH2O]0 ratios. The results are presented in Figure 19. The experimental deviation (Dexp) is defined in Eq. (21). Dexp = 100% x ([CH2O]0 – [CH2O])/[CH2O]0 (21) 40 Deviation / % 30 20 10 0 1 10 100 1000 [H2O2]0/[CH2O]0 Figure 19. The experimental deviation (Dexp, black squares) as a function of [H2O2]0/[CH2O]0. As shown in Figure 19, Dexp [H2O2]0/[CH2O]0 and the deviation [H2O2]0/[CH2O]0 reaches 20. increases becomes with increasing significant after In order to figure out the reason for the observed deviation, additional experiments were performed by adding concentrated H2O2 to the mixed solution (H2O2, CH2O and derivazation reagents) after the incubation 38 3 Results and discussion time (15 min). The absorbance of the formed derivative is monitored before and after the addition of extra H2O2 (represented as Abs0 and Abs). The result is shown in Figure 20 where ln(Abs/Abs0) is plotted as a function of reaction time. 0.0 -0.2 -0.4 ln(Abs/Abs0) -0.6 -0.8 -1.0 -1.2 -1.4 -1.6 0 200 400 600 800 1000 Time / s Figure 20. ln(Abs/Abs0) of the derivative as a function of reaction time (black squares). It can clearly be seen from Figure 20 that ln(Abs/Abs0) is linearly correlated to the reaction time. Therefore, it is worthwhile to point out that H2O2 reacts with the produced derivative and the reaction follows first order kinetics under the present conditions. The observed deviation of CH2O in Figure 19 should mainly be due to this irreversible reaction consuming the produced derivative during the incubation. 3.1.4 1. Summary The yields of formaldehyde in the reaction between the hydroxyl radical and methanol and Tris are both enhanced by increasing 39 3 Results and discussion O2 concentration. For Tris, the yield also increases with increasing solution pH. This is not observed for methanol. 2. The production of CH2O in the heterogeneous H2O2/ZrO2/Probe system is affected by O2 and solution pH. Though, the dynamics of the whole process is not likely influenced. Based on the results, a mechanism for the CH2O production by Tris and HO• is proposed including O2- and pH-dependent pathways. 3. Tris competes with H2O2 on reacting with hydroxyl radical, thereby affecting the mechanism of the catalytic decomposition of H2O2 on ZrO2 surface. A modified mechanism is proposed including two independent surface adsorption sites for H2O2 and Tris on ZrO2 surface. The simulated results based on the model are in good agreement with the experimental results. 4. H2O2 reacts with the formed derivative by the modified Hantzsch method, thereby lowering the accuracy of the method. 3.2 Inhibition of radiation induced dissolution of UO 2 by sulfide I The experiments were performed by using a UO2 pellet exposed to γ irradiation in aqueous solution containing sulfide and NaHCO3. The concentration of U(VI) is plotted against the reaction time as presented in Figure 21. As can be seen in Figure 21, the rate of radiation induced dissolution of UO2 is reduced by the addition of sulfide and the reduction is dependent on the sulfide concentration. Specifically, the inhibition lasts for a given time after which the formation of U(VI) accelerates. The period is longer with higher sulfide concentration and the length of the periods roughly corresponds to the times required to produce equal amounts of radiolytic oxidants under the present conditions (dose rate). 40 3 Results and discussion 10.0 [U(VI)] / µM 8.0 6.0 4.0 2.0 0.0 0 5000 10000 15000 20000 25000 30000 35000 Time / s Figure 21. Concentration of U(VI) plotted against irradiation time with different sulfide concentrations: 0 mM ( ), 1 mM ( ), 2 mM ( ). © 2013 Elsevier B. V. The inhibition of sulfide was confirmed. Furthermore, the effect was compared to that of H2/Pd by studying their capacity on the reduction of U(VI) in aqueous solution. The results are shown in Figure 22. From Figure 22, it is observed that the reduction of U(VI) by H2 in the absence of Pd is negligible in the present work which is in line with previous observations.25, 47 Moreover, the reduction of sulfide appears to be completely independent of the presence of H2 without a catalyst. The reduction rate in H2/Pd system is much higher than sulfide within the concentrations used in the present work. The reduction rate in the H2/Pd system in the absence and presence of 1 mM sulfide are identical which indicates that there is no poisoning effect of sulfide on Pd catalyst in the present work. On the other hand, there is no synergy of sulfide and H2/Pd on the reduction of U(VI). 41 3 Results and discussion 180 160 [U(VI)] / μM 140 120 100 80 60 40 20 0 0 5000 10000 15000 20000 25000 30000 35000 Time / s Figure 22. Concentration of U(VI) plotted against reaction time at 20 bar H2 or 1.5 bar N2 under different conditions: 20 bar H2 ( mM sulfide ( sulfide + Pd ( ); 20 bar H2 and 1 mM sulfide ( ); 1.5 bar N2 and 2 mM sulfide ( ); 20 bar H2 and Pd ( ); 20 bar H2 and 2 ); 20 bar H2 + 1 mM ). © 2013 Elsevier B. V. 3.3 Kinetics, mechanistic and simulation study of catalytic decomposition of H 2 O 2 on metal and metal oxide surfaces 3.3.1 Cu, Cu2O and CuO VIII Cu, Cu2O and CuO were investigated under the same conditions ([H2O2]0 = 5 mM, [CH3OH]0 = 100 mM) and similar SA/V (4600 ± 200 m-1). Since the solubilities of Cu2O and CuO under the present condition are relatively low (3 and 10 µM, respectively),103 the solution reactions are assumed to be negligible. In Figure 23, the H2O2 consumption and the CH2O production as function of reaction time is plotted for the three materials. 42 3 Results and discussion Not surprising, as can be seen in Figure 23, the final production of formaldehyde in the CuO case is the highest since catalytic decomposition is the dominating surface reaction pathway for consuming H2O2 in this system. While the lower CH2O production in the Cu and Cu2O case is because smaller fractions of H2O2 are decomposed to hydroxyl radical as a result of the competition between surface oxidation and catalytic decomposition on consuming H2O2. 0.20 0.08 (a) (b) 5 0.07 4 0.06 4 0.12 3 [CH2O] / mM [H2O2] / mM [H2O2] / mM 0.16 0.05 3 0.08 2 1 0.04 0 0.00 5000 0 1000 2000 3000 4000 0.03 0.02 1 0.01 0 0 Time / s 500 1000 1500 2000 2500 0.00 3000 Time / s 0.7 (c) 5 0.04 2 [CH2O] / mM 5 0.5 3 0.4 0.3 2 0.2 1 [CH2O] / mM [H2O2] / mM 0.6 4 0.1 0.0 0 0 5000 10000 15000 20000 Time / s Figure 23. Concentration of H2O2 (black squares) and CH2O (red squares) as a function of reaction time for (a) Cu, (b) Cu2O, and (c) CuO powder. © The Royal Society of Chemistry 2015 3.3.2 Other lanthanide and transition metal oxides II A number of transition and lanthanide oxides were studied, including CeO2, Fe2O3, HfO2, Gd2O3 and CuO powder. The kinetic studies were performed by following the concentration of H2O2 with reaction time. The results are shown in Figure 24. 43 3 Results and discussion 1.0 [H2O2]/[H2O2]0 0.8 0.6 0.4 0.2 0.0 0 5000 10000 15000 20000 25000 Time / s Figure 24. Normalized H2O2 concentration ([H2O2]0 = 0.5mM in 50 mL) as a function of reaction time for the reaction with CuO ( ), CeO2 ( ), HfO2 ( ), Gd2O3 ( ) and Fe2O3 ( ) at T = 298.15 K SA: Fe2O3 (4.5 m2), CeO2 (7.5 m2); CuO (0.3 m2); HfO2 (7.5 m2); Gd2O3 (1.7 m2). © 2013 Elsevier B. V. It can be seen from Figure 24 that, except of Gd2O3, the other oxides show only slightly different reactivity towards H2O2 under the present conditions. The first-order rate constants k1 obtained from the data of Figure 24 are given in Table 5. The second order rate constants and the intercepts at the zero coordinate (b2) of the least squares fits for the plots of k1 as a function of SA/V are given in Table 5. Table 5. Obtained k1, k2 and b2 for decomposition of H2O2 (0.5mM; 50 mL) catalyzed by different oxides at T = 298 K. k1 was obtained with the following SA of oxides: Fe2O3 (4.5 m2), CeO2 (7.5 m2); CuO (0.3 m2); HfO2 (7.5 m2); Gd2O3 (1.7 m2). © 2013 Elsevier B. V. Material Fe2O3 CeO2 CuO HfO2 Gd2O3 k1 (s-1) k2 (m·s-1) -4 (2.1 ± 0.4) × 10 (1.7 ± 0.5) × 10-4 (1.90 ± 0.05) × 10-4 (4.3 ± 0.9) × 10-4 (3.6± 0.3) × 10-5 b2 (s-1) -9 (3.0 ± 0.06) × 10 (2.80 ± 0.07) × 10-8 (1.23 ± 0.06) × 10-9 (2.78 ± 0.02) × 10-9 (9.4 ± 1) × 10-10 44 2 × 10-4 5 × 10-6 6 × 10-6 1 × 10-5 6 × 10-6 3 Results and discussion As can be seen in Table 5, the values of k2 deviate significantly from the expected diffusion controlled reaction rate constants for particles of this range of sizes (in the order of 10-5 m·s-1).104 In addition, the value of b2 for Fe2O3 is significantly higher than for the other oxides. This indicates that for Fe2O3, a homogeneous Fenton reaction in the bulk is also occurring due to the presence of dissolved Fe2+. This extra reaction has increased significance on the overall reaction kinetics when compared with the other materials.105 The temperature intervals for determining the Arrhenius activation energies were set as T = (298–334) K for Fe2O3, CuO, HfO2 and T = (298– 353) K for CeO2 and Gd2O3. The obtained Arrhenius energies are shown in Table 6. Table 6. Arrhenius activation energies (Ea) and frequency factors (A) for the decomposition of H2O2 catalyzed by different oxide materials. © 2013 Elsevier B. V. Material Ea (kJ·mol-1) A (s-1) Particle size Fe2O3 CeO2 CuO HfO2 Gd2O3 47 ± 1 40 ± 1 76 ± 1 60 ± 1 63 ± 1 2.2 × 103 1.4 × 103 3.5 × 109 1.1 × 107 3.4 × 106 < 5 μm 14 μm < 50 nm 44 μm 12 nm It can be seen from Table 6 that the obtained Ea values differ for the different oxides studied. This indicates that the activation energies are most likely determined by microstructural properties of the particles, e.g. the type of atoms present at the catalytically active surface sites and the extent of hydroxylation at these sites. The differences in the values of frequency factors indicate that surface catalytic capacities of the oxides are very different. The differences may be attributed to several reasons. The most obvious one is that the number of active surface sites (e.g. adsorption sites) differs for different oxides. Other possible reasons are the different numbers of catalytically active sites at the interfaces and the strength of bonding between decomposition products upon H2O2 and these active sites. The dynamics of formation of the hydroxyl radicals during catalytic decomposition of H2O2 on the oxides were studied by using the probe Tris. 45 3 Results and discussion The concentration of formed CH2O is plotted against reaction time in Figure 25. 0.6 [CH2O] / mM 0.5 0.4 0.3 0.2 0.1 0 0 10000 20000 30000 40000 Time / s Figure 25. Concentration of CH2O as a function of reaction time under the conditions: 5 mM H2O2, 20 mM Tris, 50 mL for different oxides. Fe2O3( CuO ( ), CeO2 ( ), HfO2 ( ), Gd2O3 ( ), ). © 2013 Elsevier B. V. It can be seen from Figure 25 that the dynamics of formation of CH2O for the different oxides vary considerably. In Figure 26 the concentration of CH2O as a function of relative conversion of H2O2 is shown. However, this can be optimized by using normalized CH2O concentration ([CH2O]t/[CH2O]f) instead. (see Paper III and IV) The initial reaction conditions are the same in all experiments ([H2O2]0= 5 mM and [Tris]0= 20 mM). As can be seen in Figure 26, it is obvious that the CH2O production is very low up to a certain conversion of H2O2 for a few of the oxides. These curves resemble the one shown in above section (Figure 18) and the low CH2O production period is the time needed for H2O2 and Tris to reach a steady state on the oxide surface. When the steady state for both H2O2 46 3 Results and discussion and Tris is obtained, the formation rate of CH2O starts to increase, displayed as an inflection point in the figure. 0.6 0.5 [CH2O] / mM 0.4 0.3 0.2 0.1 0 0 20 40 60 80 100 % of H2O2 consumed from solution Figure 26. Concentration of CH2O present in the reaction system as a function of the percentage of H2O2 consumed from solution during reaction with different oxide materials. CeO2 ( ), Gd2O3 ( ), Y2O3 ( ), ZrO2 ( UO2 ( ). © 2013 Elsevier B. V. ), Fe2O3 ( ), HfO2 ( ), CuO ( ), TiO2 ( ) and For some oxides, e.g. TiO2 and Gd2O3, a clear inflection point is never reached under the present conditions and very small amounts of formaldehyde are formed. It may be attributed to that the relative surface coverage of H2O2 is much higher than Tris, thereby most of H2O2 are consumed via reacting with HO• other than reacting with Tris to produce CH2O. For other more extreme cases such as HfO2, an inflection point appears when the H2O2 is almost completely consumed form the solution. However, it is not possible to attribute the position of the inflection points as the result of a single effect for all the materials. In the UO2 case, the production of CH2O reaches a plateau corresponding to 20% of H2O2 consumed. The CH2O concentration does not increase afterwards regardless of the continuous consumption of 47 3 Results and discussion H2O2 from solution. H2O2 can react with UO2 both via catalytic decomposition and oxidative dissolution.35 The hydroxyl radicals produced in the oxidative dissolution of UO2 cannot be scavenged by Tris.106 The observed plateau of CH2O may be attributed to that catalytic decomposition is hindered since the regenerated fresh surface is dissolved according to the oxidative dissolution. 3.3.3 1. Summary The competition between surface oxidation and catalytic decomposition on consuming H2O2 is the main reason for the lower production of CH2O in the Cu and Cu2O cases than that in the CuO case. 2. H2O2 can be catalytically decomposed on the surface of a number of oxides. The reactivity of H2O2 and dynamics of CH2O formation is not the same for all the studied materials. 3.4 Kinetics and mechanistic study of reactions between W and H 2 O 2 or O 2 3.4.1 W and H2O2 V The XRD, SEM and XPS measurements were conducted to investigate the crystalline structure and composition of the tungsten powder before and after treatment. A typical X-ray diffraction pattern is shown in Figure 27a. It can be seen from Figure 27a that there are XRD peaks attributed to two crystalline phases present in the W powder. One group of peaks belongs to metallic W which possess bcc (base-center-cubic) structure,107 the other group belongs to the so-called A15-W phase where the tungsten unit cell is stabilized by oxygen.108, 109 Note that, the XRD patterns recorded from the W powder after different treatments, considered in the current study, look quite similar to those for the untreated powder. This observation is supported by our SEM studies (see Figure 27b and 48 3 Results and discussion Figure 27c) where the surface morphology of W powder before and after the reaction with H2O2/Tris is presented. Figure 27. (a) X-ray diffraction pattern of untreated tungsten powder. Two types of crystalline structures of tungsten and their crystalline phases are labelled on the graph. SEM image of (b) untreated tungsten powder; (c) tungsten powder after reaction with Tris/H2O2. It can be seen from Figure 27b and Figure 27c that the W powder consists of two types of particles (faceted crystal and porous). These two phases are most probably corresponding to metallic W and A-15 WOx determined by XRD analysis. Moreover, no significant changes in the surface morphology of W powder before and after treatment were 49 3 Results and discussion observed. Also, no other phases than metallic tungsten and A-15 have been detected by XRD and SEM. The composition of tungsten powder before and after reaction with Tris and H2O2 was studied by means of XPS. Since the penetration depth of photoelectrons is about 6 nm, mostly surface composition was studied. The results are shown in Table 7. Table 7. XPS results about atomic concentration (AC) of different W species on the surface. C 1s O 1s W4f7/2 BE, eV W ref., at % W (2g) and H2O2, at. % W (2g) and Tris, at % W (2g) and Tris/H2O2 , at % W (1g) and Tris/H2O2 , at % Chemical binding, compounds 284.5 16.3 11.5 15.2 11.9 19.9 C-(C,H) 286.3 2.9 5.2 6.7 5.5 7.9 C-OH 289.0 1.1 2.5 2.9 2.2 4.7 COOH 530.9 49.4 47.2 45.1 46.2 33.6 W-O, 531.9 5.8 8.0 7.1 6.6 6.3 W-OH, C=O 533.1 3.2 3.8 3.1 3.6 5.0 C-OH 31.4 3.9 5.2 3.7 7.7 8.7 W metal110 32.0 1.6 1.9 1.8 2.7 4.0 W-O, W-C111 35.9 15.8 14.8 14.5 13.6 10.0 WO3112 BE: Binding energy As seen in Table 7, the surface of W powder consists mostly of metallic tungsten and WO3. No significant changes on the amount of metallic tungsten and WO3 were observed after being treated with H2O2 and Tris. Generally, XPS, XRD and SEM studies reveal that W powder does not undergo significant changes in the composition and structure when react with Tris and H2O2. The kinetics of H2O2 decomposition and formation of CH2O in the W(s)/H2O2/Tris system at varied amount of tungsten powder was investigated. The results are shown in Figure 28. 50 3 Results and discussion 6 (a) [H2O2] / mM 5 4 3 2 1 0 0 4000 8000 12000 16000 Time / s 20000 24000 8000 12000 16000 Time / s 20000 24000 0.35 (b) 0.30 [CH2O] / mM 0.25 0.20 0.15 0.10 0.05 0.00 0 4000 Figure 28. (a) Concentration of H2O2 and (b) CH2O as a function of reaction time with varied amounts of tungsten powder: SA/V = 8800 m–1 (black squares), SA/V = 4400 m–1 (red circles), SA/V = 2200 m–1 (green up triangles), SA/V = 1100 m–1 (blue down triangles), SA/V = 550 m–1 (cyan diamonds). All cases are under the condition: [H2O2]0 = 5 mM, [Tris] = 100 mM, N2 and pH 7.5. It can be seen from Figure 28a that the consumption of H2O2 appears to be almost independent of the amount of tungsten powder except in the case of the lowest amount (550 m–1). Generally, the H2O2 concentration appears as a slow initial decrease followed by accelerated consumption. This acceleration indicates that the reaction is autocatalyzed. As be seen in Figure 28b, the final production of formaldehyde decreases with increasing the amount of tungsten powder. Moreover, an initial slow phase followed by an acceleration of CH2O evolution can be observed in all cases. This slow process producing CH2O (Haber-Weiss reactions) will be discussed below and the period corresponds to the time required to build up a sufficiently high concentration of W(V). The increasing CH2O production rate indicates that H2O2 is initially consumed through oxidation of W. The produced dissolved tungsten species is capable of catalytically decomposing H2O2 to hydroxyl radicals. 51 3 Results and discussion 6 (a) [H2O2] / mM 5 4 3 2 1 0 0 4000 8000 12000 16000 20000 24000 Time / s [CH2O] / mM 0.30 (b) 0.25 0.20 0.15 0.10 0.05 0.00 0 4000 8000 12000 16000 20000 24000 Time / s 0.25 (c) [W] / mM 0.20 0.15 0.10 0.05 0.00 0 4000 8000 12000 16000 20000 24000 Time / s Figure 29. Concentration of (a) H2O2, (b) CH2O and (c) dissolved tungsten species as a function of reaction time in non-filtered (black squares) and filtered (red circles) system. Filtration was done at 1860 s. The experiment was done under the condition: SA/V = 4400 m−1, [H2O2]0 = 5 mM, [Tris] = 100 mM, N2 and pH 7.5. Considering the possibility of reactions between H2O2 and tungsten ions in homogeneous solution, the solid was removed by filtration from the W(s)/H2O2/Tris system. The concentrations of H2O2 and CH2O in the filtered solution as well as in the original non-filtered solution were monitored and compared. The concentration of dissolved tungsten species was also measured at the same sampling time points. W(VI) is the dominating stable ionic tungsten species in solution under basic condition according to the reported Eh-pH diagram of the W-H2O system.113 The results are presented in Figure 29. The standard redox 52 3 Results and discussion potential of W(VI)/W(V) couple is -0.03 V which is much lower than that of Fe(III)/Fe(II) (+0.77V).67 Hence, it is more difficult for H2O2 to reduce W(VI) to W(V) via R(13) than to reduce Fe(III) to Fe(II). In the H2O2/Fe(III) system, the concentration of Fe(II) is insignificant as compared to that of Fe(III).28 Similar as for Fe(III)/Fe(II), it can be assumed that [W(V)] << [W(VI)] ≈ [W]total. It can be seen from Figure 29a that in the non-filtered solution the consumption of H2O2 is accelerated with reaction time while in the filtered solution no acceleration is observed. Nevertheless, H2O2 is also consumed in the filtered solution which implies that the product responsible for the autocatalysis is only formed in the presence of solid W. Moreover, the major part of the H2O2 is consumed by the soluble Wspecies. From Figure 29b, it can be observed that H2O2 is catalytically decomposed in both the filtered and the non-filtered samples. The formaldehyde production accelerates more in the non-filtered sample than in the filtered sample. This is consistent with the observed acceleration in the H2O2 consumption for the non-filtered sample. The initial slow phase producing CH2O exactly corresponds to the period to build up a sufficiently high concentration of W(V), while in the nonfiltered case the phase is shorter. Similar as in the H2O2/Fe(III)/Fe(II) system where the Fe(II) concentration is determined by the total iron concentration,73 sufficient W(V) is readily obtained in the non-filtered samples due to the continuous dissolution of W. It is observed in Figure 29c that the W concentration in solution increases almost linearly with time during the initial phase of the reaction and then reaches a maximum level. By comparison of Figure 29c and Figure 29b, it can be seen that the production of formaldehyde remains fairly low during the initial phase when W is being released to solution. This indicates that the catalytic decomposition of H2O2 is mainly attributed to solution reactions. The kinetics of H2O2 consumption at different W-concentrations in solution was studied in filtered samples extracted from the 53 3 Results and discussion W(s)/H2O2/Tris system. The experimental details can be seen in Paper V. The results are presented in Figure 30. 0.0 (a) ln([H2O2]/[H2O2]0) -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 0 5000 10000 15000 20000 25000 Time/s 7.0x10-4 6.0x10-4 5.0x10 kobs / s-1 (b) y= (1.50 +/- 0.01)x + 9.96E-6 R2 = 97.1% -4 4.0x10-4 3.0x10-4 2.0x10-4 1.0x10-4 0.0 0.0 1.0x10-4 2.0x10-4 3.0x10-4 4.0x10-4 [W] / M Figure 30. (a) ln([H2O2]/[H2O2]0) as a function of reaction time in different experiments: experiment NO. 1 (black squares), 2 (red circles), 3 (green up triangles), 4 (blue down triangles), 5 (cyan diamonds), 6 (magenta left triangles), 7 (yellow right triangles), 8 (dark yellow hexagons) and 9 (navy stars). (b) Observed first-order rate constants of experiments NO. 1- 9 as function of [W] (black squares). From Figure 30a, it is clear that H2O2 is consumed by W species in solution. This reaction follows first order kinetics. It is also shown in Figure 30b that the first order rate constant is linearly dependent on the W-concentration corresponding to a second order rate constant of (1.50 ± 0.01) M−1 s−1. The intercept obtained is close to zero which implies that the only reaction responsible for the consumption of H2O2 is the reaction with W in solution. Given the concentrations of the reactants, first order kinetics would not be observed if the solute W-species was consumed as well. Hence, the solute Wspecies must be a catalyst for the decomposition of H2O2. 54 3 Results and discussion It has been proven in Paper V that the addition of Tris can significantly improve the decomposition rate of H2O2 in the presence of dissolved tungsten. 1.0 (a) [W] / mM 0.8 0.6 0.4 0.2 0.0 [W] / mM 0 0.30 0.25 0.20 0.15 0.10 0.05 0.00 4000 6000 8000 10000 12000 14000 16000 Time / s (b) 0 rdiss. / 10-8 M s-1 2000 8 7 6 5 4 3 2 1 0 2000 4000 6000 8000 100001200014000160001800020000 Time / s (c) 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 SA/V / m-1 Figure 31. Concentration of dissolved tungsten species ([W] is used throughout the thesis to represent total concentration of dissolved tungsten species) as a function of the reaction time under different conditions: (a) WO3 + H2O (black squares), WO3 + H2O + Tris (red circles), WO3 + H2O + Tris + H2O2 (green up triangles), SA/V = 140400 m–1. (b) W + H2O (black squares), W + H2O + Tris (red circles), W + H2O + H2O2 (green up triangles), W + H2O + Tris + H2O2 (blue down triangles), SA/V = 4400 m–1. (c) Dissolution rate (rdiss.) of 55 3 Results and discussion tungsten species (black squares) as a function of tungsten powder amount (SA/V) under the condition: [H2O2]0 = 5 mM, [Tris] = 100 mM, N2 and pH 7.5. Experiments on oxidative dissolution of W powder by H2O2 (5 mM) were performed in anoxic aqueous solution with H2O2 and/or Tris (100 mM). XPS result of W powder (Table 7) shows that trace amount of WO3 was found on the surface of tungsten powder, so parallel experiments were also performed for WO3. The results are shown in Figure 31. As can be seen in Figure 31a, the dissolution of W from WO3 in water is enhanced by Tris. In this process, the presence of H2O2 is insignificant. It is clearly seen from Figure 31b that both H2O2 and Tris influence the dissolution of W. The combined effect between H2O2 and Tris is most significant which indicates that oxidative dissolution of W in the presence of Tris is the key-process. In Figure 31c, it is also clear that the dissolution rate of W depends on the amount of solid W. Based on all results, a mechanism (Figure 32) is proposed regarding H2O2 reacts with metallic tungsten powder in the presence of anoxic Tris solution, including both heterogeneous and homogeneous reaction pathways. H 2O 2 W(s) OH H 2O 2 OH (ii) (i) W(n)(s) (iv) W(VI)(aq) (0 < n < 6) HO2 H 2O 2 W(V)(aq) (Tris) (v) H 2O 2 (iii) OH H 2O 2 OH (vi) OH Tris CH2O Figure 32. Reactions between H2O2 and metallic tungsten in anoxic Tris solution. The whole process is initiated when H2O2 is added to the suspension of tungsten powder and Tris. Firstly, H2O2 adsorbs on the surface of 56 3 Results and discussion tungsten powder triggering the oxidative dissolution of W (i) and (ii) and catalytic decomposition of H2O2 on surface (iii). The production of hydroxyl radicals is accompanying these two surface reactions, although the production is probably negligible as compared to that in the solution reactions (see Figure 28). Hydroxyl radicals produced during the oxidative dissolution has been reported in previous studies.69, 70 Though, it is impossible to distinguish among the surface reactions, (i), (ii) and (iii). The dissolved tungsten (mostly W(VI)) reacts with H2O2 via the Haber-Weiss reactions in the solution ((iv) and (v)) which form hydroxyl radicals continuously. This process is facilitated by the presence of Tris. Regardless if produced via surface reactions ((i), (ii) and (iii)) or solution reactions ((iv) and (v)), the hydroxyl radicals will be scavenged by probes to produce formaldehyde (vi). In the W(s)/H2O2/Tris system, the overall consumption rate of H2O2 can be defined as: − 𝑑[𝐻2 𝑂2 ] 𝑑𝑑 = 𝑘2𝑠𝑠𝑠𝑠 𝑆𝑆 𝑉 [𝐻2 𝑂2 ] + 𝑘2𝑠𝑠𝑠 [𝑊][𝐻2 𝑂2 ] (22) where k2surf and k2sol represents the second-order rate constant for the surface and solution reactions, respectively. The latter has already been determined to (1.50 ± 0.01) M−1 s−1 from the slope of Figure 30b. Using this rate constant and the concentration of W and H2O2, the rate of H2O2 consumption in the homogeneous phase can be calculated. These rates are very close to the corresponding slopes of Figure 28a, again proving that the H2O2 consumption is dominated by the solution reactions. 3.4.2 W and O2 VII The experiments were performed in the aerobic or anaerobic Tris and methanol solution. The results are presented in Figure 33. As can be seen from Figure 33a and Figure 33d, both concentrations of dissolved tungsten and CH2O are significantly enhanced by O2 in both the Tris and methanol cases. This indicates that hydroxyl radicals are produced during the oxidative dissolution of W. This is in line with the similar systems where H2O2 is produced and then decomposed to 57 3 Results and discussion hydroxyl radicals via the Haber-Weiss reactions.71, 114-117 (Paper V). The detected concentration of H2O2 is marginal which should be due to the fact that H2O2 is consumed rapidly in the present system. Furthermore, a slight delay in the production of CH2O is observed in Figure 33c and 33d for both Tris and methanol. This may be attributed to the delay in producing surface catalytic sites for H2O2 decomposition as was shown in Paper V. 500 500 400 400 300 300 200 200 100 100 0 0 4000 300 16000 0 (c) 250 [CH2O] / uM 8000 12000 Time / s 4000 (d) 0 8000 12000 16000 20000 24000 Time / s 300 250 200 200 150 150 100 100 50 50 0 0 4000 8000 12000 Time / s 16000 0 4000 [W] / uM (b) [CH2O] / uM [W] / uM (a) 0 8000 12000 16000 20000 24000 Time / s Figure 33. Concentration of dissolved tungsten ([W]) and CH2O in aerobic (Air, black squares) and anaerobic (N2, red circles) as a function of reaction time under the condition: (a) and (c) Tris (100 mM, pH 7.5), (b) and (d) methanol solution (100 mM, unadjusted pH). The concentration of formed CH2O was plotted against the concentration of dissolved tungsten species in Figure 34. Every point in the figure corresponds to [CH2O] and [W] measured from the same sample (same reaction time). The figure is based on the data already presented in Figure 33. 58 3 Results and discussion 250 y = (1.02 +/- 0.04)x - 181.36 R2 = 99.1% [CH2O] / uM 200 150 Tris Methanol 100 y = (0.33 +/- 0.01)x - 4.83 R2 = 99.4% 50 y = 0.16x - 1.20 R2 = 100% 0 0 100 200 300 [W] / uM 400 500 Figure 34. Concentration of formaldehyde in W (s)/air/Tris (black squares) and W(s)/air/methanol system (red circles) as a function of the amount of dissolved tungsten. It is observed in Figure 34 that the slope of [CH2O]/[W] in the Tris case is half of that in the methanol case for low [W] (< 200 µM). During this period, the difference in the slope is virtually identical to the difference in the conversion yield Y reported in Paper III (29% and 68% for Tris and methanol in the presence of O2, respectively).55 This implies that the production rate of HO• is the same for both systems at fairly low [W] where surface reactions are expected to dominate. Interestingly, the slope of [CH2O]/[W] increases at higher [W] for Tris. The slope becomes three times higher than that in the methanol case. A plausible explanation for this is that the homogeneous Haber-Weiss reactions become dominating at high [W] in the presence of Tris. This is in line with the observations on the impact of Tris and methanol. The homogeneous Haber-Weiss reactions are facilitated by Tris but not by methanol. (Paper V and VII). 59 3 Results and discussion 3.4.3 1. Summary Homogeneous Haber-Weiss reactions are mainly responsible for the consumption of H2O2 and formation of CH2O in the aqueous W(s)/H2O2/Tris system. Another two reactions in this system are oxidative dissolution of W and catalytic decomposition of H2O2 occurring at the interface. A mechanism including these reactions in the present system was proposed. 2. A slow phase of CH2O formation in both filtered and non-filtered system corresponds to the time to build up sufficient W(V) to accelerate the homogeneous Haber-Weiss reactions to produce hydroxyl radicals continuously. 3. Oxidative dissolution of W by H2O2 is evident and can be facilitated in the presence of Tris. 4. Oxidative dissolution of W by O2 is also evident and hydroxyl radicals are formed during the dissolution. The key step is the formation of H2O2. 60 4 Conclusions 4 Conclusions The aim of the thesis was to further investigate radiation induced processes in heterogeneous systems including surface reactions (between H2O2 and metal/metal oxide surfaces, i.e. redox reactions and catalytic decomposition) as well as solution reactions (Haber-Weiss reactions). Based on the studies presented in the thesis, the following conclusions can be drawn: • H2O2 oxidatively dissolves W. The reaction is facilitated by the presence of Tris. In the W(s)/H2O2/Tris aqueous system, after a short initial period of dissolution, solution reactions (homogeneous catalysis) become the dominating reactions consuming H2O2. The later reaction is accompanied by formation of hydroxyl radicals (monitored via the formation of CH2O). • Hydroxyl radicals are formed also during the O2-induced dissolution of W. This is the result of the Haber-Weiss reactions between the formed intermediate H2O2 and dissolved W producing HO• continuously. • The competition between the two surface reactions on consuming H2O2 is the main reason for the lower production of CH2O in the Cu and Cu2O cases than that in the CuO case. The catalytic decomposition of H2O2 is the only surface reaction in the CuO case. • Sulfide inhibits the radiation induced dissolution of UO2 and the inhibition is weaker than H2/Pd in the present work. However, sulfide does not interfere with the inhibition process by H2/Pd. • The mechanism of catalytic decomposition of H2O2 on ZrO2 surface is not likely affected by changing the dissolved O2 concentration and solution pH. However, the consumption rate of H2O2 is affected by the competition between Tris and H2O2 reacting with surface bound hydroxyl radical. • CH2O production by probes (Tris and methanol) and HO• include both O2-dependent and O2-independent pathways. There is a pH- 61 4 Conclusions dependent pathway for Tris however not for methanol. The accuracy of the modified Hanztsch method for detection of CH2O is affected by the presence of H2O2 via the reaction with the formed derivative. 62 5 Future work 5 Future work To the best of my knowledge, this thesis presents some of the first systematic studies of the kinetics and mechanism of radiation induced processes in heterogeneous system. In addition, the methodology for the detection of hydroxyl radicals is also studied to a large extent. Based on the results, the experimental design of kinetics and mechanistic studies on the heterogeneous system can be further developed. For some metals and metal oxides (e.g. Fe2O3, UO2) which can be oxidatively dissolved by H2O2, filtrations at selected time points are desirable. The modification is in order to compare the reactivity of H2O2 and production of CH2O in both filtered and non-filtered samples. The time-resolved dissolution of studied materials can also be monitored. Consequently, the mechanism which may include both surface and solution reactions can be elucidated. The kinetics, the dependence on the concentration of dissolved species and the effect of ligands on the solution reactions in the filtered homogeneous system could be interesting to study. The radiation induced processes on the aqueous W(s)/Tris system may be studied by using γ-irradiation. The surface structure and composition of tungsten powder and the formation of CH2O after irradiation would be interesting to study. Furthermore, the study on the temperature dependence for the formation of CH2O in both homogeneous and heterogeneous H2O2/MxOy/Probe systems would give valuable information for understanding the interfacial reactions. 63 7 Acknowledgements 6 Acknowledgements The PhD journey would not have been so wonderful without the help and support on both academic and daily life from many enthusiastic individuals in the last four years. I would like to express my most sincere gratitude and appreciation to all of you: Prof. Mats Jonsson, my supervisor. I really appreciate you for providing me the opportunity to work in this project, leading me the way into the academia and orienting me during the journey with full support, endless patience and continuous encouragement. Associate Prof. Susanna Wold, my co-supervisor. Thank you so much for sharing your professional knowledge and support. Prof. An Lin, Prof. Fuxin Gan, Prof. Dihua Wang, my former supervisors in Wuhan University, China. Thank you for introducing me into the scientific world and supporting me to study abroad. Special thanks for Prof. An Lin for immense help and suggestions on my career. Prof. Daqing Cui at Stusvik. I really appreciate you for sharing professional knowledge and your immense help on my future career. Prof. Lars Kloo. I really appreciate you for sharing professional knowledge and your invaluable suggestions on my thesis. Sara Sundin, Claudio Miguel Lousada. Thank you very much for guiding me into my doctoral study, teaching me the basic knowledge and the experimental methodology. Also thanks for the collaboration and invaluable discussions. Åsa Björkbacka, Karin Norrfors, Alexandre Barreiro Fidalgo, Veronica Diesen, Björn Dahlgren, Mats Jansson, Prof. Gábor Merényi, Prof. Johan Lind, Majid Safdari, Annika Maier, Elin Toijer, Joakim Halldin Stenlid, Camilla Gustafsson. I am sincerely grateful for your kindly help and for sharing the wonderful moments with me. For Åsa and Alex, special thanks for your immense scientific input and invaluable discussions on our collaborations. 64 7 Acknowledgements Inna Soroka, Prof. Christofer Leygraf, Kristina Nilsson, Alex Grosjean, Claudia Gasparrini, my co-authors. Thanks for the scientific input and invaluable discussions. Special thanks Inna for the kindly help on the thesis. Bo Xu, Jing Huang, Hong Chen, Yinan Yang, Lei Wang, Jiajia Gao, Haining Tian, Qian Gao, Yong Hua, Juan Sun, Ming Cheng, Cheng Chen, Qingpeng Meng, Wen Zhang, Wangzhong Mu, Lan Hu, Yifeng Yun. Great thanks for all of you for making my everyday life in Sweden enjoyable and unforgettable. China Scholarship Council (CSC) and Swedish Nuclear Fuel and Waste Management Co (SKB) are gratefully acknowledged for financial support. My family, both in Yang’s and Zhang’s. Thank you so much for your endless support and understanding all these years. My parents, thank you for your endless love and full support. My wife Xian Zhang, I would not have become a better man and a Ph.D without your immense encouragement, remarkable consideration and endless love. My daughter Ruoyu Alice Yang, you are the most amazing gift in my life, thank you. 65 8 References 7 References 1 R. C. Ewing, Nat. Mater., 2015, 14, 252-257. 2 H. Kleykamp, J. Nucl. Mater., 1985, 131, 221-246. 3 R. J. M. Konings, T. Wiss and O. Benes, Nat. Mater., 2015, 14, 247252. 4 L. McDonald, Environmental Impact Statement for A Geologic Repository for the Disposal of Spent Nuclear Fuel and High-Level Radioactive Waste at Yucca Mountain, Nye County, Nevada, U.S. Department of Energy, 2002. 5 SKB, Long-term safety for the final repository for spent nuclear fuel at Forsmark. Main report of the SR-Site project. TR-11-01, Swedish Nuclear Fuel and Waste Management Co. SKB, Stockholm, Sweden, 2011. 6 M. Jonsson, in Recent Trends in Radiation Chemistry, eds. J. F. Wishart and B. Rao, World Scientific Publishing Co. Pte. Ltd., Singapore, 2010. 7 B. F. Myasoedov and Y. M. Kulyako, J. Radioanal. Nucl. Chem., 2013, 296, 1127-1131. 8 R. B. Adamson and P. Rudling, Dimensional Stability of Zirconium Alloys, 2002. 9 A. Hiroki and J. A. LaVerne, J. Phys. Chem. B, 2005, 109, 33643370. 10 C. M. Lousada, A. J. Johansson, T. Brinck and M. Jonsson, J. Phys. Chem. C, 2012, 116, 9533. 11 M. Yang and M. Jonsson, J. Mol. Catal. A: Chem., 2015, 400, 4955. 66 8 References 12 G. Choppin, J. O. Liljenzin and J. Rydberg, Radiochemistry and Nuclear Chemistry, Second edn., Reed Educational and Professional Publishing Ltd, Oxford, 1995. 13 J. W. T. Spinks and R. J. Woods, An Introduction to Radiation Chemistry, Third edn., John Wiley & Sons Inc, New York, 1990. 14 G. V. Buxton, C. L. Greenstock, W. P. Helman and A. B. Ross, J. Phys. Chem. Ref. Data, 1988, 17, 513-886. 15 R. H. Schuler, L. K. Patterson and E. Janata, J. Phys. Chem, 1980, 84, 2088-2089. 16 R. Overend and G. Paraskevopoulos, J. Phys. Chem, 1978, 82, 1329-1333. 17 E. B. Spear, J. Am. Chem. Soc., 1908, 30, 195-209. 18 C. Satterfield and T. Stein, Ind. Eng. Chem., 1957, 49, 1173-1180. 19 D. W. McKee, J. Catal., 1969, 14, 355-364. 20 E. Ekeroth, O. Roth and M. Jonsson, J. Nucl. Mater., 2006, 355, 38-46. 21 L. Wu and D. W. Shoesmith, Electrochim. Acta, 2014, 137, 83-90. 22 R. Pehrman, M. Trummer, C. M. Lousada and M. Jonsson, J. Nucl. Mater., 2012, 430, 6-11. 23 M. M. Hossain, E. Ekeroth and M. Jonsson, J. Nucl. Mater., 2006, 358, 202-208. 24 R. Pehrman, M. Amme, O. Roth, E. Ekeroth and M. Jonsson, J. Nucl. Mater., 2010, 397, 128. 25 S. Nilsson and M. Jonsson, J. Nucl. Mater., 2008, 372, 160. 26 F. Nielsen and M. Jonsson, J. Nucl. Mater., 2008, 374, 281-285. 27 D. P. Woodruff, Chemical Bonding at Surfaces and Interfaces, Elsevier, Amsterdam, 2008. 28 J. De Laat and H. Gallard, Environ. Sci. Technol., 1999, 33, 27262732. 29 D. W. Shoesmith, J. Nucl. Mater., 2000, 282, 1-31. 67 8 References 30 D. J. Stein, D. Hetherington, T. Guilinger and J. L. Cecchi, J. Electrochem. Soc., 1998, 145, 3190-3196. 31 G. Lim, J.-H. Lee, J.-W. Son, H.-W. Lee and J. Kim, J. Electrochem. Soc., 2006, 153, B169-B172. 32 D. Dermatas, W. Braida, C. Christodoulatos, N. Strigul, N. Panikov, M. Los and S. Larson, Environmental Forensics, 2004, 5, 5-13. 33 E. Lassner and W.-D. Schubert, Tungsten Properties, Chemistry, Technology of the Element, Alloys, and Chemical Compounds, Kluwer Academic/Plenum Publishers, Boston, 1999. 34 G. S. Kelsey, J. Electrochem. Soc., 1977, 124, 814-819. 35 E. Ekeroth and M. Jonsson, J. Nucl. Mater., 2003, 322, 242-248. 36 S. M. Peper, L. F. Brodnax, S. E. Field, R. A. Zehnder, S. N. Valdez and W. H. Runde, Ind. Eng. Chem. Res., 2004, 43, 8188-8193. 37 J. A. T. Smellie, M. Laaksoharju and P. Wikberg, Journal of Hydrology, 1995, 172, 147-169. 38 N. R. Smart, D. J. Blackwood and L. Werme, The anaerobic corrosion of carbon steel and cast iron in artificial groundwaters TR-01-22, Swedish Nuclear Fuel and Waste Management Co. SKB Stockholm, Sweden, 2001. 39 H. Kleykamp, Nucl. Technol., 1988, 80, 412-422. 40 R. Håkansson, Beräkning av nuklidinnehåll, resteffekt, aktivitet samt doshastighet för utbränt kärnbränsle R-99-74, Swedish Nuclear Fuel and Waste Management Co. SKB, Stockholm, Sweden, 1999. 41 D. Cui, J. Low, C. J. Sjöstedt and K. Spahiu, Radiochim. Acta, 2004, 92, 551-555. 42 M. Trummer, S. Nilsson and M. Jonsson, J. Nucl. Mater., 2008, 378, 55-59. 43 E.-L. Tullborg, J. Smellie, A.-C. Nilsson, M. J. Gimeno, L. Auqué, V. Brüchert and J. Molinero, SR-Site-sulphide content in the 68 8 References groundwater at Forsmark TR-10-39, Swedish Nuclear Fuel and Waste Management Co. SKB, Stockholm, Sweden, 2010. 44 D. D. Perrin, in Ionisation Constants of Inorganic Acids and Bases in Aqueous Solution (Second Edition), ed. D. D. Perrin, Pergamon, 1982, pp. 1-138. 45 A. V. Kochenov, K. G. Korolev, V. T. Dubinchuk and Y. L. Medvedev, Geochem. Int., 1978, 14, 82-87. 46 B. Hua, H. F. Xu, J. Terry and B. L. Deng, Environ. Sci. Technol., 2006, 40, 4666-4671. 47 S. Nilsson and M. Jonsson, J. Nucl. Mater., 2008, 374, 290-292. 48 W. W. McNab, R. Ruiz and M. Reinhard, Environ. Sci. Technol., 2000, 34, 149-153. 49 C. Schüth, N.-A. Kummer, C. Weidenthaler and H. Schad, Appl. Catal., B, 2004, 52, 197-203. 50 G. V. Lowry and M. Reinhard, Environ. Sci. Technol., 2000, 34, 3217-3223. 51 C. M. Lousada, M. Yang, K. Nilsson and M. Jonsson, J. Mol. Catal. A: Chem., 2013, 379, 178-184. 52 C. u. M. Lousada and M. Jonsson, J. Phys. Chem. C, 2010, 114, 11202-11208. 53 L. Ge, T. Chen, Z. Liu and F. Chen, Catal. Today, 2014, 224, 209215. 54 N. P. Sardesai, D. Andreescu and S. Andreescu, J. Am. Chem. Soc., 2013, 135, 16770-16773. 55 M. Yang and M. Jonsson, J. Phys. Chem. C, 2014, 118, 7971-7979. 56 E. Giamello, L. Calosso, B. Fubini and F. Geobaldo, J. Phys. Chem, 1993, 97, 5735-5740. 57 C. J. Rhodes, Prog. React. Kinet. Mech., 2005, 30, 145-213. 58 X. Zhou and K. Mopper, Mar. Chem., 1990, 30, 71-88. 69 8 References 59 S.-K. Han, K. Ichikawa and H. Utsumi, Water Res., 1998, 32, 32613266. 60 J. K. Kim and I. S. Metcalfe, Chemosphere, 2007, 69, 689-696. 61 M. E. Lindsey and M. A. Tarr, Chemosphere, 2000, 41, 409-417. 62 V. Diesen and M. Jonsson, J. Adv. Oxi. Technol., 2012, 15, 392-398. 63 K. S. Haveles, A. G. Georgakilas, V. Sophianopoulou and E. G. Sideris, Radiat. Res., 1997, 148, 500-501. 64 M. Hicks and J. M. Gebicki, FEBS Lett., 1986, 199, 92-94. 65 Q. Li, P. Sritharathikhun and S. Motomizu, Anal. Sci., 2007, 23, 413-417. 66 T. Nash, Biochem. J., 1953, 55, 416. 67 R. A. Sheldon and J. K. Kochi, in Metal-catalyzed Oxidations of Organic Compounds, eds. R. A. Kochi and K. Sheldonjay, Academic Press, 1981, pp. 33-70. 68 I. C. M. S. Santos, F. A. A. Paz, M. M. Q. Simões, M. G. P. M. S. Neves, J. A. S. Cavaleiro, J. Klinowski and A. M. V. Cavaleiro, Appl. Catal., A, 2008, 351, 166-173. 69 F. C. C. Moura, M. H. Araujo, R. C. C. Costa, J. D. Fabris, J. D. Ardisson, W. A. A. Macedo and R. M. Lago, Chemosphere, 2005, 60, 1118-1123. 70 R. C. C. Costa, F. C. C. Moura, J. D. Ardisson, J. D. Fabris and R. M. Lago, Appl. Catal., B, 2008, 83, 131-139. 71 C. R. Keenan and D. L. Sedlak, Environ. Sci. Technol., 2008, 42, 6936-6941. 72 C. Lee, C. R. Keenan and D. L. Sedlak, Environ. Sci. Technol., 2008, 42, 4921-4926. 73 D. Nichela, L. Carlos and F. G. Einschlag, Appl. Catal., B, 2008, 82, 11-18. 70 8 References 74 D. Nichela, M. Haddou, F. Benoit-Marquié, M.-T. Maurette, E. Oliveros and F. S. García Einschlag, Appl. Catal., B, 2010, 98, 171179. 75 S. Belman, Anal. Chim. Acta, 1963, 29, 120-126. 76 E. Sawicki and R. A. Carnes, Microchim. Acta, 1968, 56, 148-159. 77 B. J. Compton and W. C. Purdy, Anal. Chim. Acta, 1980, 119, 349357. 78 Q. Li, P. Sritharathikhum, M. Oshima and S. Motomizu, Anal. Chim. Acta, 2008, 612, 165-172. 79 K. Matsuno and S. Suzuki, Anal. Biochem., 2008, 375, 53-59. 80 T. Salthammer, S. Mentese and R. Marutzky, Chemical Reviews, 2010, 110, 2536-2572. 81 V. Diesen, C. W. Dunnill, E. Osterberg, I. P. Parkin and M. Jonsson, Dalton Trans., 2014, 43, 344-351. 82 J. Li, P. K. Dasgupta and W. Luke, Anal. Chim. Acta, 2005, 531, 5168. 83 C. de Serves, J. Geophys. Res. Atmos., 1994, 99, 25391-25398. 84 T. Staffelbach, A. Neftel, A. Blatter, A. Gut, M. Fahrni, J. Stähelin, A. Prévôt, A. Hering, M. Lehning, B. Neininger, M. Bäumle, G. L. Kok, J. Dommen, M. Hutterli and M. Anklin, J. Geophys. Res. Atmos., 1997, 102, 23345-23362. 85 J. Li, P. K. Dasgupta, Z. Genfa and M. A. Hutterli, Field Analytical Chemistry & Technology, 2001, 5, 2-12. 86 S. B. Savvin, Talanta, 1961, 8, 673-685. 87 I. K. Kressin, Anal. Chem., 1984, 56, 2269-2271. 88 S. Nilsson and M. Jonsson, J. Nucl. Mater., 2011, 410, 89-93. 89 J. A. Ghormley and A. C. Stewart, J. Am. Chem. Soc., 1956, 78, 2934-2939. 90 C. J. Hochanadel, J. Phys. Chem., 1952, 56, 587-594. 71 8 References 91 H. Hein, A. Hoffmann and R. Zellner, Phys. Chem. Chem. Phys., 1999, 1, 3743-3752. 92 S. Zhang, R. Strekowski, L. Bosland, A. Monod and C. Zetzsch, Phys. Chem. Chem. Phys., 2011, 13, 11671-11677. 93 V. Diesen and M. Jonsson, J. Adv. Oxid. Technol., 2013, 16, 16-22. 94 D. R. Lide and H. P. R. Frederikse, CRC Handbook of Chemistry and Physics, 76th ed., CRC Press, Boca Raton, FL, 1995. 95 W. J. McElroy and S. J. Waygood, J. Chem. Soc., Faraday Trans., 1991, 87, 1513-1521. 96 A. Monod, A. Chebbi, R. Durand-Jolibois and P. Carlier, Atmos. Environ., 2000, 34, 5283-5294. 97 J. Chen, D. F. Ollis, W. H. Rulkens and H. Bruning, Water Res., 1999, 33, 1173-1180. 98 B. C. Gilbert, R. G. G. Holmes, H. A. H. Laue and R. O. C. Norman, J. Chem. Soc. Perkin Trans. 2, 1976, 1047-1052. 99 B. C. Gilbert, P. D. R. Marshall, R. O. C. Norman, N. Pineda and P. S. Williams, Journal of the Chemical Society, Perkin Transactions 2, 1981, 1392-1400. 100 А. А. Sladkova, A. G. Lisovskaya, А. А. Sosnovskaya, I. P. Edimecheva and О. I. Shadyro, Radiat. Phys. Chem., 2014, 96, 229-237. 101 T. Song, C.-Y. Ma and I. K. Chu, J. Phys. Chem. A, 2013, 117, 10591068. 102 G. Hao and S. Gross, J. Am. Soc. Mass Spectrom., 2006, 17, 17251730. 103 M. Nakahara, 中原勝, I. C. o. t. P. o. Water and Steam, Water, Steam, and Aqueous Solutions for Electric Power: Advances in Science and Technology : Proceedings of the 14th International Conference on the Properties of Water and Steam : Kyoto 72 8 References International Conference Hall Kyoto, Japan August 29-September 3, 2004 : ICPWS14, Maruzen Company, 2005. 104 M. Jonsson, in Recent Trends in Radiation Chemistry eds. J. F. Wishart and B. S. M. Rao, World Scientific Singapore 2010, pp. 301-323. 105 F. Haber and J. Weiss, Naturwissenschaften, 1932, 20, 948-950. 106 C. M. Lousada, M. Trummer and M. Jonsson, J. Nucl. Mater., 2013, 434, 434-439. 107 H. E. Swanson and E. Tatge, Natl. Bur. Stand. (U.S.) Circ., 1953, Ι, 23-33. 108 P. Petroff, T. T. Sheng, A. K. Sinha, G. A. Rozgonyi and F. B. Alexander, J. Appl. Phys., 1973, 44, 2545-2554. 109 M. J. O’Keefe, J. T. Grant and J. S. Solomon, J. Electron. Mater., 1995, 24, 961-967. 110 C. D. Wagner and G. E. Muilenberg, Handbook of X-Ray Photoelectron Spectroscopy: a Reference Book of Standard Data for use in X-Ray Photoelectron Spectroscopy, Physical Electronics Division, Perkin-Elmer Corp., 1979. 111 L. Leclercq, M. Provost, H. Pastor, J. Grimblot, A. M. Hardy, L. Gengembre and G. Leclercq, J. Catal., 1989, 117, 371-383. 112 R. J. Colton, A. M. Guzman and J. W. Rabalais, J. Appl. Phys., 1978, 49, 409-416. 113 M. Anik and K. Osseo-Asare, J. Electrochem. Soc., 2002, 149, B224-B233. 114 C. R. Keenan and D. L. Sedlak, Environ. Sci. Technol., 2008, 42, 1262-1267. 115 A. D. Bokare and W. Choi, Environ. Sci. Technol., 2009, 43, 71307135. 116 G. Wen, S.-J. Wang, J. Ma, T.-L. Huang, Z.-Q. Liu, L. Zhao and J.-L. Xu, J. Hazard. Mater., 2014, 275, 193-199. 73 8 References 117 H. Lee, H.-j. Lee, H.-E. Kim, J. Kweon, B.-D. Lee and C. Lee, J. Hazard. Mater., 2014, 265, 201-207. 74