NOVEL PREPARATION OF NANOSTRUCTURED TITANIUM DIOXIDE PHOTOCATALYTIC PARTICLES, FILMS, MEMBRANES, AND DEVICES FOR ENVIRONMENTAL APPLICATIONS

Transcription

1

2 NOVEL PREPARATION OF NANOSTRUCTURED TITANIUM DIOXIDE PHOTOCATALYTIC PARTICLES, FILMS, MEMBRANES, AND DEVICES FOR ENVIRONMENTAL APPLICATIONS A dissertation submitted to the Division of Research and Advanced Studies of the University of Cincinnati in partial fulfillment of the requirements for the degree of DOCTORATE OF PHILOSOPHY (Ph.D.) in the Department of Civil and Environmental Engineering of the College of Engineering 2007 by Hyeok Choi Diploma (B.S./M.S.) in Civil and Environmental Engineering, Sungkyunkwan University in Korea, 2000 Committee Chair: Dr. Dionysios D. Dionysiou i

3 ABSTRACT Precise manipulation of matter at the nanoscale will enhance our potential to synthesize materials with tailor-designed properties and functionalities for their environmental applications. This dissertation explores the development of innovative nanotechnological procedures for the preparation of highly efficient visible light-activated nanostructured TiO 2 photocatalytic particles, films, membranes, and devices for environmental applications. Nanocrystalline TiO 2 particles and immobilized films and membranes with mesoporous inorganic network were prepared via a sol-gel method modified with surfactants as pore-directing agents. Not only did we manipulate the physicochemical properties of TiO 2 such as crystallographic structure, particle size, and defect structure but also tailor-design its structural properties such as surface area, pore volume, and pore size distribution. Asymmetric mesoporous multilayer TiO 2 photocatalytic membranes exhibiting hierarchical changes in pore diameter and materials porosity were also fabricated. These TiO 2 films and membranes inherently possessed multiple and simultaneous functions including photocatalytic decomposition of organic pollutants, inactivation of pathogenic microorganisms, physical separation of contaminants, and anti-biofouling action. In addition, for the design of solar-driven treatment technologies, highly efficient visible light-activated TiO 2 photocatalysts with mesoporous structure and narrowed band gap energy were synthesized by introducing nitrogen-containing surfactant as a pore templating material as well as a nitrogen dopant in the sol-gel method of TiO 2. For the ii

4 development of highly sensitive and stable electrochemical sensors to detect a neurotransmitter, catechol, sonogel carbon electrodes were modified with the nanostructured TiO 2 acting as an adsorbent for catechol and a redox mediator for electron transfer. We also elucidated the formation of nanocrystalline TiO 2 particles at ambient synthesis conditions via sol-gel method employing water immiscible room temperature ionic liquid as reaction medium and modified with surfactant as pore template. Detail information on the preparative method, synthesis route and mechanism, crystallographic and structural properties, and photocatalytic activity of the nanocrystalline TiO 2 particles with thermal stability was investigated. From a scientific point of view, this study will provide new nanotechnological and materials chemistry procedures to synthesize highly efficient photocatalytic TiO 2 particles, films, and membranes that can be used for the treatment and disinfection of water and wastewater under even visible light irradiation, and highly sensitive TiO 2 - based devices for the development of new type of sensors. iii

5 EXTENDED SUMMARY Nanotechnology has recently become one of the most active research areas in a variety of fields including basic science and engineering. Controlling materials at the nano-level can accelerate the development of new types of products with improved properties and functionalities for environmental applications. In particular, nanostructured TiO 2 photocatalysts have been extensively researched for the development of highly efficient water treatment processes. In this dissertation, we explore new nanotechnological and materials chemistry procedures for the synthesis of tailor-designed TiO 2 photocatalysts via sol-gel methods modified with surfactants as pore templating materials and/or ionic liquids as reaction media. The motivation, proposed idea, general background, and objectives and challenges of this study are described in Chapter 1. It has been reported that addition of certain inorganic materials into TiO 2 sol would enhance the thermal stability of resulting TiO 2 material and thus inhibit its crystal phase transformation. In Chapter 2, the effect of adding amphiphilic organic molecules in TiO 2 sol on the physicochemical properties and photocatalytic activity of crystalline TiO 2 nanoparticles prepared via a simple sol-gel route at high temperatures from 400 to 800 C was studied. The addition of polyoxyethylene sorbitan surfactant and polyethylene oxide and polypropylene oxide triblock copolymer as particle growth inhibitors and pore directing agents into a stable TiO 2 sol affected the properties of TiO 2 nanoparticles such as their crystallographic structure, morphology, and defect structure. With the addition of the surfactants, the ratio of anatase and rutile crystal phases of TiO 2 was controlled and an active anatase crystal phase was maintained during heat treatment up to 800 ºC. iv

6 Decrease in the sintering rate and crystal growth was also observed, which resulted in prevention of crystallite aggregation and thus maintenance of high surface area. Bulk defects in TiO 2 decreased while surface defects increased as a result of the addition of surfactants. The high crystallinity, anatase crystal phase, high specific surface area, surface defects, and segregated morphology of TiO 2 nanoparticles, which were induced by the addition of surfactants, were more advantageous for enhancing photocatalytic destruction of 4-chlorophenol. Another important aspect for widespread use of TiO 2 photocatalyst in a variety of applications is its immobilization onto various substrates as the form of thin films and membranes, where it is critical to improve and control the photocatalytic and structural properties of TiO 2 material. In Chapter 3, mesoporous TiO 2 photocatalytic particles, films, and membranes were synthesized via a simple method that involves dip-coating of borosilicate glass substrate for TiO 2 films or porous alumina substrate for TiO 2 composite membranes into an organic/inorganic sol composed of isopropanol, acetic acid, titanium tetraisopropoxide, and various surfactants followed by calcination of the coating at 500 C. Controlled hydrolysis and condensation reactions were achieved through in-taking of water released from the esterification reaction of acetic acid with isopropanol. The subsequent stable incorporation of the Ti O Ti network onto self-assembled surfactants resulted in TiO 2 with enhanced structural and catalytic properties. The effect of surfactant type and concentration on the homogeneity, morphology, light absorption, dye adsorption and degradation, and hydrophilicity of TiO 2 films as well as on the structural properties of the corresponding TiO 2 particles was investigated. The uniform transparent TiO 2 thin films were hydrophilic and possessed thermally stable bicontinuous mesoporous v

7 inorganic network. The porous TiO 2 films prepared with polyethylene glycol sorbitan monooleate surfactant exhibited four times higher photocatalytic activity for the decoloration of methylene blue than the nonporous control TiO 2 films. Their properties included high surface area of 147 m 2 /g and porosity of 46%, narrow pore size distribution ranging from 2 to 8 nm, active anatase crystal phase, and small crystallite size of 9 nm. High water permeability and sharp polyethylene glycol retention of the TiO 2 /Al 2 O 3 composite membranes evidenced the good structural properties of the TiO 2 coatings. In addition, repeating the coating procedure made it possible to control the physical properties of the TiO 2 layer such as the coating thickness, catalyst amount, photocatalytic activity, water permeability, and organic retention. These photocatalytic TiO 2 thin films and membranes have great potential in developing highly efficient water treatment and reuse systems because of their multiple and simultaneous functions such as decomposition of organic pollutants (methylene blue, creatinine, microcystin-lr), inactivation of pathogenic microorganisms (Escherichia coli), physical separation of contaminants (polyethylene glycol, activated sludge), and anti-biofouling action. In order to further enhance the performance of TiO 2 /Al 2 O 3 membranes, an advanced sol-gel dip-coating process was introduced in Chapter 4 for the fabrication of a hierarchical mesoporous multilayer structure within TiO 2 skin layer. The ideal membrane structure would be composed of a thin asymmetric multilayer with increasing pore size and porosity from the top to the bottom of TiO 2 skin layer. Titania sols containing different concentrations of polyethylene glycol sorbitan monooleate surfactant were dipcoated on the top of porous alumina sublayer, dried and calcined. This procedure was repeated with varying sols from high to low surfactant concentration in succession to vi

8 tailor-design the structural properties of the TiO 2 material in each layer. The resulting asymmetric mesoporous TiO 2 membrane with thickness of 0.9 μm exhibited hierarchical changes in pore diameter and porosity from 2 6 nm and 46.2%, 3 8 nm and 56.7%, to 5 11 nm and 69.3% from the top to the bottom layer of TiO 2 skin. Compared to a conventional repeated-coating process for the same mesoporous overlayers prepared from a single sol, the hierarchical multilayer process improved water permeability significantly, maintaining the same organic retention and photocatalytic activity as the repeated-coating process. Since the TiO 2 photocatalysts described so far in Chapters 2 4 are activated when irradiated by UV light with above the band gap energy of TiO 2, it is imperative to activate them under visible light radiation. This can have tremendous impact on the design of solar-driven treatment technologies. In Chapter 5, we demonstrate the development of highly efficient visible light-activated nitrogen doped TiO 2 (N-TiO 2 ) photocatalysts and their environmental applications in destroying an emerging water contaminant, the cyanobacterial toxin microcystin-lr (MC-LR) under visible light. In a nanotechnological sol-gel synthesis method, a nitrogen-containing surfactant (dodecylammonium chloride) is introduced as a pore templating material for tailordesigning the structural properties of TiO 2 and as a nitrogen dopant for its visible light response. The resulting N-TiO 2 exhibits significantly enhanced structural properties including 2-8 nm mesoporous structure (porosity 44%) and high surface area of 150 m 2 /g. Red shift in light absorbance up to 468 nm, 0.9 ev lower binding energy of electrons in Ti 2p state, and reduced inter-planar distance of crystal lattices prove nitrogen doping in the TiO 2 lattice. Due to its narrow band gap at 2.65 ev, N-TiO 2 efficiently degrades MC- vii

9 LR under visible spectrum above 420 nm. Acidic condition (ph 3.0) is more favorable for the adsorption and photocatalytic degradation of MC-LR on N-TiO 2 due to electrostatic attraction forces between negatively charged MC-LR and +6.5 mv charged N-TiO 2. Even under UV light, MC-LR is decomposed 3~4 times faster using N-TiO 2 than control TiO 2. The degradation pathways and reaction intermediates of MC-LR are not directly related with energy source for TiO 2 activation (UV and visible) and nature of TiO 2 (neat and nitrogen-doped). This study implies a strong possibility for the in-situ photocatalytic remediation of contaminated water with cyanobacterial toxins and other toxic compounds using solar light, a sustainable source of energy. In Chapter 6, the nanostructured TiO 2 coatings developed in Chapter 3 are also applied in developing highly sensitive and stable sonogel carbon electrodes (SGC/TiO 2 electrode) to detect catechol electrochemically using cyclic voltammetry. Recently, electrochemical detection of neurotransmitters using smart sensors has attracted neuroscientists attention since their altered levels have been associated with mental and behavioral disorders. The deliberate chemical modification of carbon electrode surface with the nanostructured TiO 2 controlled the rates and selectivities of electrochemical reactions at the solid/liquid interface. The SGC/TiO 2 electrode developed here meets the profitable features of electrode including quantitative and qualitative detection of catechol, mechanical stability, physical rigidity, and enhanced catalytic properties. A possible rationale for the stable catechol detection of SGC/TiO 2 electrode is attributed to the adsorption of catechol onto highly porous TiO 2 and the formation of bonding between catechol and SGC/TiO 2 electrode, C 6 H 4 (OTi) 2. Catechol absorbed onto TiO 2 rapidly viii

10 reaches the SGC surface, then is oxidized. As a result, the surface of TiO 2 acts as a redox mediator for the electron transfer between the SGC electrode and catechol. In addition to the use of surfactants as templating materials, it was attempted in Chapter 7 to utilize room temperature ionic liquids (ILs) as an additional solvent in conjunction with alcoholic solvents used in the traditional sol-gel methods as well as a template material. Recently, sol-gel methods employing ILs have shown significant implications for the synthesis of well-defined nanostructured inorganic materials. Herein, we synthesized nanocrystalline TiO 2 particles via an alkoxide sol-gel method employing a water immiscible room temperature IL (1-butyl-3-methylimidazolium hexafluorophosphate, [bmim][pf 6 ]) as a new reaction medium and further modified with nonionic surfactant (polyoxyethylene sorbitan monooleate) as a pore templating material at ambient synthesis condition. Detail information on the preparative method, synthesis route and mechanism, crystallographic and structural properties, and photocatalytic activity of the TiO 2 particles was described. The possible rationale for the formation of nanocrystalline TiO 2 particles with high surface area and activity was discussed with respect to the special characteristics of [bmim][pf 6 ] as well as the role of the surfactant self-assembly in the sol-gel network. Due to its capping effect and water immiscibility, the use of [bmim][pf 6 ] in sol-gel synthesis of TiO 2 induced controlled hydrolysis of titanium alkoxide precursor, resulting in stable sol-gel network with an ordered array, and localized water-poor conditions, resulting in the formation of completely condensed and directly crystalline systems at ambient condition. The low surface energy and adaptability of [bmim][pf 6 ] facilitated the generation of very small nanocrystalline TiO 2 particles and then it also acted as a particle aggregation inhibitor. The ensuing TiO 2 particles had good ix

11 thermal stability to resist pore collapse and anatase-to-rutile crystal phase transformation during thermal treatment. The TiO 2 from as-synthesized to calcined at high temperatures even up to 800 ºC exhibited promising photocatalytic activity to degrade 4-chlorophenol. These results are the outcome of applying novel concepts of nanoscience and nanotechnology, inspired by the potential of this emerging field to provide new directions in the synthesis of advanced catalytic materials with unique hierarchical structures and functionalities. These materials are an inherent component of efficient water purification systems, which can be used as stand-alone technologies or as supplementary and complementary to existing treatment technologies. In Chapter 8, future studies and potential applications of these nanostructured TiO 2 photocatalytic particles, films, and membranes are described, and we also emphasize the Green Engineering and Green Chemistry features of this study. x

12 ACKNOWLEDGEMENTS I would like to express my sincere gratitude to my research advisor, Prof. Dionysios D. Dionysiou, for his continuous support, motivation and outstanding scientific guidance. I would also like to thank Prof. Chong H. Ahn, Prof. Paul L. Bishop and Prof. Margaret J. Kupferle for serving in my Ph.D. committee willingly as well as for their advice, criticism and encouragement. Special thanks also go to the following organizations and individuals for supporting my graduate research and studies through assistantship, scholarship, fellowships, and awards: University of Cincinnati (Research Council, Graduate Student Governance Association, Department of Civil and Environmental Engineering, Jacob D. and Lillian Rindsberg, Richard C. Wigger, John D. Eye), American Chemical Society (Division of Environmental Chemistry), National Science Foundation (Funded Research, CAREER Award to Dr. Dionysiou, REU Program), National Aeronautics and Space Administration (Funded Research). I really appreciate the collaborations with the following individuals in various aspects in environmental research: Prof. George A. Sorial and Prof. Daniel B. Oerther (University of Cincinnati), Prof. Gregory V. Lowry (Carnegie Mellon University), Prof. Suzanne Lunsford (Wright State University), Prof. Elias Stathatos (University of Patras, Greece) and Dr. Armah A. de la Cruz, Dr. Rajender S. Varma, Dr. Souhail R. Al-Abed, Dr. Gautham Jegadeesan and Dr. Vasudevan V. Namboodiri (US Environment Protection Agency, Cincinnati). xi

13 DEDICATION All to My Wonderful Coworkers and Group Members, Friends, and Loving Relatives, Family, and Wife xii

14 TABLE OF CONTENTS COVER PAGE ABSTRACT EXTENDED SUMMARY ACKNOWLEDGEMENTS DEDICATION i ii iv xi xii TABLE OF CONTENTS 1 LIST OF TABLES 9 LIST OF FIGURES 10 LIST OF CHEMICAL REACTIONS 16 LIST OF SCHEMES 17 LIST OF SYMBOLS AND ABBREVIATIONS 18 CHAPTER 1 Introduction Motivation TiO 2 Photocatalysis Proposed Idea I: Modification of Sol-Gel Methods with Surfactants Proposed Idea II: Modification of Sol-Gel Methods with Ionic Liquids Background TiO 2 Photocatalysis Sol-Gel Methods 28 1

15 1.2.3 Self-Assembling of Surfactants Ionic Liquids Environmental Nanotechnology Objectives and Challenges Preparation of Photocatalytic TiO 2 Particles, Films and Membranes Using Sol-Gel Methods Modified with Surfactants Objective I: Role of Surfactants Objective II: TiO 2 Thin Films and Membranes Objective III: Versatile Applications of TiO Preparation of Photocatalytic TiO 2 Particles Using Sol-Gel Methods Modified with Ionic Liquids Objective IV: Exploring Ionic Liquids as Reaction Media in Sol-Gel Methods Characterization and Evaluation References 38 CHAPTER 2 Effect of Surfactant in a Modified Sol on the Physicochemical Properties and Photocatalytic Activity of TiO 2 Nanoparticles Introduction Experimental Sol Preparation Formation of Crystalline TiO 2 Nanoparticles Material Characterization Measurement of Photocatalytic Activity 49 2

16 2.3 Results and Discussion Crystallographic Properties Inhibition of Anatase to Rutile Crystal Phase Transformation Physical Structure Morphology and Defect Structure Photocatalytic Activity References 64 CHAPTER 3 Sol-Gel Preparation of Mesoporous Photocatalytic TiO 2 Particles, Films and Membranes Using Surfactant-Assisted Sol-Gel Methods and Their Environmental Applications Introduction Experimental Sol Synthesis Formation of TiO 2 Particles and Films and TiO 2 /Al 2 O 3 Membranes Materials Characterization Dye Adsorption and Photocatalytic Activity of TiO 2 Films Water Permeability and Organic Retention of TiO 2 Composite Membranes Environmental Applications Results and Discussion Structural Characteristics of TiO 2 Particles Thickness, Mass, UV Light Absorption, and Dye Adsorption of TiO 2 Films Hydrophilicity of TiO 2 Films 88 3

17 3.3.4 Porosity and Morphology of TiO 2 Films Photocatalytic Activity of TiO 2 Films Effect of Number of Coatings on TiO 2 Films Properties Photocatalytic Degradation of Creatinine by TiO 2 Films Morphology and Elemental Composition of TiO 2 /Al 2 O 3 Membranes Properties of TiO 2 /Al 2 O 3 Composite Membranes Environmental Applications References 112 CHAPTER 4 Nanocrystalline TiO 2 Photocatalytic Membranes with a Hierarchical Mesoporous Multilayer: Synthesis, Characterization, and Multifunction Introduction Experimental Sol Preparation Dip-Coating and Heat Treatment Materials Characterization Evaluation of TiO 2 Membranes Results and Discussion Surfactant Effect on the Properties of TiO 2 Material TiO 2 Multicoating Water Permeability and Organic Retention Photocatalytic Activity and Antifouling Properties 139 4

18 4.4 References 142 CHAPTER 5 Mesoporous Nitrogen-Doped TiO 2 for the Photocatalytic Destruction of the Cyanobacterial Toxin Microcystin-LR under Visible Light Irradiation Introduction Experimental Synthesis of Mesoporous N-TiO Properties and Characterization of N-TiO Photocatalytic Degradation of Microcystin-LR Results and Discussion Optical Band Gap Electronic Structure Crystallographic and Physicochemical Properties Formation of Mesoporous N-TiO Photocatalytic Destruction of MC-LR under Visible Light Photocatalytic Destruction of MC-LR under UV Light Conculsions References 167 CHAPTER 6 Voltammetric Determination of Catechol Using a Sonogel Carbon Electrode Modified with Nanostructured TiO Introduction 172 5

19 6.2 Experimental Sono-Gel Carbon Electrode Modification with Nanostructured TiO Characterization of SGC/TiO Detection of Catechol Results and Discussion Catechol Detection in the Presence of Ascorbic Acid Comparison with Conducting Polymer-Modified Carbon Electrode Properties of SGC/TiO 2 Electrodes Detection Route and Mechanisms Calibration Curve Conclusions References 187 CHAPTER 7 Thermally Stable Nanocrystalline TiO 2 Photocatalysts Prepared by Sol-Gel Method Modified with Water Immiscible Room Temperature Ionic Liquids Introduction Experimental Preparation Procedure Materials Characterization NMR and FTIR Analyses Measurement of Photocatalytic Activity 197 6

20 7.3 Results and Discussion Physicochemical Properties of [bmim][pf 6 ]-Templated TiO Photocatalytic Activity of [bmim][pf 6 ]-Templated TiO Effect of [bmim][pf 6 ]/TTIP and H 2 O/TTIP Molar Ratio Synthesis Mechanisms of [bmim][pf 6 ]-Templated TiO Heat Treatment of [bmim][pf 6 ]-Templated TiO Modification of [bmim][pf 6 ]-Assisted Sol-Gel Method with Surfactant Self-Assembly as Pore Templates Removal of [bmim][pf 6 ] and Surfactant and Properties of TiO Thermal Stability of [bmim][pf 6 ]- and Surfactant-Templated TiO Photocatalytic Activity References 223 CHAPTER 8 Recommendations and Potential Applications Recommendations Fundamentals on Synthesis Mechanism Use of Other Ionic Liquids Incorporation of TiO 2 Layer with Support Materials Flow-Through TiO 2 Photocatalytic Reactors Testing Real Water and Wastewater Potential Applications Compact Water and Wastewater Treatment Systems Space Wastewater Treatment and Reuse 231 7

21 8.2.3 Detoxification of Biological Toxins Air Purification Adsorbent Electrode and Sensor Other Applications Green Engineering Considerations References 235 8

22 LIST OF TABLES 1.1 Properties of amphiphilic organic molecules Elemental composition of TiO 2 nanoparticles calcined at 600 ºC Surface area of TiO 2 nanoparticles calcined at 600 C Structural properties of TiO 2 nanoparticles calcined at 600 ºC Structural characteristics of TiO 2 particles Structural characteristics of TiO 2 film control at R = 0.0 and film T80 at R = Properties of TiO 2 /Al 2 O 3 composite membranes Physicochemical properties of TiO 2 materials Properties of multi-coating TiO 2 /Al 2 O 3 membranes Crystallographic and physicochemical properties of control TiO 2 and N-TiO 2 prepared at different calcination conditions Summarized IR results showing possible bondings between TiO 2 and catechol at ph BET specific surface area of TiO 2 particles prepared at different conditions Physicochemical properties of TiO 2 particles prepared under various synthesis conditions 213 9

23 LIST OF FIGURES 2.1 (a) XRD patterns and (b) fraction of anatase and rutile phases of TiO 2 nanoparticles calcined at 600 C upon addition of T Fraction of anatase and rutile phases of TiO 2 nanoparticles upon heat treatment XPS spectra of TiO 2 nanoparticles prepared with (a) control at R = 0.0, (b) T20 at R = 0.242, and (c) P105 at R = at 600 ºC Effect of calcination temperature on (a) surface area and (b) crystallite size of TiO 2 nanoparticles HR-TEM images of anatase crystalline TiO 2 nanoparticles prepared with (a) control at R = 0.0 and (b) T20 at R = at 600 C Adsorption and photocatalytic degradation of 4-CP by TiO 2 nanoparticles prepared with T20 at different concentrations at 600 C (a) adsorption and (b) photocatalytic degradation of 4-CP by TiO 2 nanoparticles prepared with T20 at different calcination temperatures XRD patterns of TiO 2 particles prepared with T Thickness and mass of TiO 2 films UV-visible absorption spectrum of TiO 2 film UV light absorption of TiO 2 films at 365 nm Visible light absorption of MB-adsorbed TiO 2 films at 564 nm (a) water contact angle and (b) wettability of TiO 2 films 90 10

24 3.7 (a) nitrogen adsorption-desorption isotherms and (b) pore size distribution of TiO 2 film control at R = 0.0 and film T80 at R = Morphology and pore structure of (a-b) film control at R = 0.0 and (c-d) film T80 at R = 1.0 at different magnifications Photocatalytic decoloration of MB by TiO 2 films T XRD patterns of TiO 2 films (a) Visible light absorbance of TiO 2 films with adsorbed MB and (b) decoloration of MB solution by TiO 2 films UV absorbance at 365 nm and thickness of TiO 2 films Creatinine degradation by TiO 2 films ESEM images of TiO 2 /Al 2 O 3 composite membranes XPS spectrum of TiO 2 /Al 2 O 3 composite membrane (a) Permeate water flux and (b) MWCO of TiO 2 /Al 2 O 3 composite membranes Destruction of MC-LR by photocatalytic TiO 2 films Inactivation of E. coli by photocatalytic TiO 2 films Anti-biofouling property of TiO 2 /Al 2 O 3 composite membrane under static condition contacting dissolved organic solution Permeate water flux and organic retention of TiO 2 /Al 2 O 3 composite membrane treating dissolved organic solution Degradation of methylene blue dye and creatinine by photocatalytic TiO 2 membranes

25 4.1 TEM images of TiO 2 material prepared at: (a) R = 0, (b-d) R = 1 at different magnifications, (e) R = 2, and (f) R = (a) N 2 adsorption/desorption isotherms and (b) pore size distribution of TiO 2 material XRD patterns of TiO 2 material ESEM image of TiO 2 membrane TEM images of TiO 2 films with a multicoating layer: (a) three layers prepared at R = 3, 2, and 1 in succession, (b) boundary between R = 3 and 2, and (c) boundary between R = 2 and (a) Water permeability and (b) PEG retention of TiO 2 membranes (a) Photocatalytic degradation of methylene blue dye and (b) photocatalytic inactivation of pathogenic microorganism, E. coli by TiO 2 membranes Antifouling properties of TiO 2 membranes Optical UV-Visible absorption spectra of control TiO 2 and N-TiO 2 calcined at 350 C Effective band gap energy, E eff g and nitrogen content of control TiO 2 and N-TiO 2 prepared at different calcination temperatures XPS spectrum of N-TiO 2 calcined at 350 C for 5 h High resolution XPS of control TiO 2 and N-TiO 2 calcined at 350 C (a) at ev for Ti 2p, (b) at ev for O 1s, and (c) at ev for N 1s (a) scanning-tem image and (b) line analysis results for Ti, O, N, and C elemental mapping of N-TiO 2 particles calcined at 350 C

26 5.6 XRD patterns of control TiO 2 and N-TiO 2 prepared at different calcination temperatures (101) plane d space for control TiO 2 and N-TiO 2 prepared at different calcination temperatures (a) N 2 adsorption-desorption isotherms and (b) pore size distribution of control TiO 2 and N-TiO 2 prepared at 350 C HR-TEM morphology of (a) control TiO 2 and (b) N-TiO 2 (1) as-synthesized and (2) and (3) calcined at 350 C Adsorption followed by photocatalytic degradation of MC-LR using N-TiO 2 prepared at 350 C under visible light (>420 nm) at different ph conditions MC-LR degradation efficiency of control TiO 2 and N-TiO 2 at ph 3.0 after visible light (>420 nm) irradiation for 30 and 120 min MC-LR degradation by control TiO 2 and N-TiO 2 at ph 3.0 under UV-365 nm irradiation Cyclic voltammograms of (a) 5 mm catechol and 5 mm ascorbic acid in 10 mm sulfuric acid (ph 1.7), (b) 1 mm dopamine in 0.1 M phosphate buffer M NaCl (ph 7.4), and (c) 10 mm catechol and10 mm acetaminophen in 0.1 M phosphate buffer M NaCl solution (ph=7.4) Cyclic voltammograms of 5 mm catechol in 10 mm sulfuric acid during 25 scans ESEM images of SGC/TiO 2 electrode: (a) the tip and (b) the surface of the area highlighted in (a)

27 6.4 Elemental analysis of SGC/TiO 2 electrode HR-TEM image of TiO 2 with mesoporous structure Relation between current in cyclic voltammetry and catechol concentration (a) wide angle and (b) low angle XRD patterns of [bmim][pf 6 ]-templated TiO 2 particles (S PF6 ) calcined at 100 C Nitrogen adsorption-desorption isotherms and pore size distribution of [bmim][pf 6 ]-templated TiO 2 particles (S PF6 ) calcined at 100 C HR-TEM images of [bmim][pf 6 ]-templated TiO 2 particles (S PF6 ) calcined at 100 C HR-TEM images of [bmim][bf 4 ]-templated TiO 2 particles (S BF4 ) calcined at 100 C CP adsorption and photocatalytic degradation by [bmim][pf 6 ]-templated TiO 2 particles (S PF6 ) calcined at 100 C over time NMR spectrum of [bmim][pf 6 ]/TTIP mixture during the synthesis of [bmim][pf 6 ]-templated TiO 2 (S PF6 ) FTIR spectra of [bmim][pf 6 ]/TTIP mixture during the synthesis of [bmim][pf 6 ]-templated TiO 2 (S PF6 ) Particle size of agglomerated TiO 2 during sol-gel synthesis without IL (S control ) and with [bmim][pf 6 ] (S PF6 ) and [bmim][bf 4 ] (S BF4 ) Crystal phase evolution upon heat treatment: (a) control TiO 2 (S control ) and (b) [bmim][pf 6 ]-templated TiO 2 (S PF6 ) Structural evolution of the TiO 2 particles upon heat treatment: (a) pore volume, (b) surface area, (c) pore size, and (d) crystallite size

28 7.11 Particle size of as-synthesized and calcined TiO 2 prepared without IL (S control ) and with [bmim][pf 6 ] (S PF6 ) and [bmim][bf 4 ] (S BF4 ) (a) 4-CP adsorption and (b) photocatalytic degradation by control TiO 2 (S control ) and [bmim][pf 6 ]-templated TiO 2 (S PF6 ) Size measurement of Tween 80 surfactant self-assembly FTIR spectra of TiO 2 (S PF6,T80 ) prepared via [bmim][pf 6 ]- and surfactantassisted sol-gel method: (a) as-synthesized before [bmim][pf 6 ] extraction, (b) as-synthesized, and (c) after heat treatment at 500 ºC TGA/DSC pattern of as-synthesized TiO 2 (S PF6,T80 ) via [bmim][pf 6 ]- and surfactant-assisted sol-gel method HR-TEM morphology of nanostructured anatase crystalline TiO 2 (S PF6,T80 ) prepared via [bmim][pf 6 ]- and surfactant-assisted sol-gel method at 500 ºC (a) N 2 adsorption-desorption isotherms and (b) pore size distribution of TiO 2 (S PF6,T80 ) prepared via [bmim][pf 6 ]- and surfactant-assisted sol-gel method at 500 ºC Powder XRD patterns of TiO 2 (S PF6, T80 ) prepared via [bmim][pf 6 ]- and surfactant-assisted sol-gel method upon heat treatment Structural evolution of TiO 2 (S PF6,T80 ) prepared via [bmim][pf 6 ]- and surfactant-assisted sol-gel method upon heat treatment: (a) surface area and pore volume and (b) average pore size and crystallite size Photocatalytic degradation of 4-chlorophenol by commercially available TiO 2 (P-25), [bmim][pf 6 ]-templated TiO 2 (S PF6 ), and [bmim][pf 6 ]- and surfactant-templated TiO 2 (S PF6, T80 )

29 LIST OF CHEMICAL REACTIONS 3.1 iproh + AcOH iproac + HOH Ti-OiPr + AcOH Ti-OAc + iproh Ti-OAc + iproh iproac + Ti-OH Ti-OiPr + Ti-OAC iproac + Ti-O-Ti Ti[OCH(CH 3 ) 2 ] H 2 O + [Bmim]PF 6 TiO (CH 3 ) 2 CHOH + [Bmim]PF

30 LIST OF SCHEMES 1.1 Structure of amphiphilic organic molecules Surfactant micelle or reverse micelle formation followed by the formation of TiO 2 porous network or TiO 2 nanoparticle Structure of water immiscible room temperature ionic liquid, 1-butyl-3-methylimidazolium hexafluorophosphate Structure of organic contaminants used in this study Incorporation of Ti O Ti network onto self-organized DDAC surfactant micelles to form an organic core/inorganic shell composite, followed by the removal of the organic templates to form N-TiO 2 with mesoporous structure Bonding and electron transfer in sonogel carbon electrode modified with titanium dioxide Synthesis route of porous anatase crystalline TiO 2 particles using sol-gel method modified with [bmim][pf 6 ]

31 LIST OF SYMBOLS AND ABBREVIATIONS AcOH AOTs APS ASAP BET BJH [bmim][pf 6 ] [bmim][bf 4 ] Cyano-HABs CMC CMT Creatinine CS XRD CS BET CS TEM CV D BJH Da DDA DDAC DEA Acetic acid Advanced oxidation technologies Agglomerated particle size Accelerated surface area and porosymetry Brunauer, Emmett, and Teller Barrett, Joyner and Halenda 1-butyl-3-methylimidazolium hexafluorophosphate 1-butyl-3-methylimidazolium tetrafluoroborate Harmful algal blooms of cyanobacteria Critical micellar concentration Critical micelle temperature 2-imino-1-methylimidazolidin-4-one Crystal size estimated from ZRD Crystal size calculated from BET Crystal size measured from TEM Cyclic voltammetry BJH average pore size Dalton Dodecylamine Dodecylammonium chloride Diethanolamine 18

32 DOC DS DSC DTA E g Dissovled organic carbon Dissolved solids differential scanning calorimetry Differential thermal analysis Band gap energy E. coli Escherichia coli EDX ESEM Film surfactant Film control Flux FT-IR HAB HPLC HR-TEM IL i-proh L p MB MC-LR MSTs MTMOS MW Electron dispersive X-ray spectroscopy Environmental scanning electron microscopy TiO 2 film prepared with surfactants TiO 2 film prepared without surfactants Permeate water flux Fourier transform infrared spectroscopy Hazardous algal bloom High performance liquid cromatopgraphy High lesolution-transmission electron microscopy Ionic liquid Isopropanol Permeability coefficient Methylene blue Microcystin-LR Membrane separation technologies methyltrimethoxysilane Molecular weight 19

33 MWCO NMR Molecular weight cut off Nuclear magnetic resonance spectroscopy N-TiO 2 Nitrogen-doped TiO 2 P-25 Degussa P-25 TiO 2 powder P105 P123 P3MT PEG PEO PPO PS PV PZC R R m Range BJH RTIL S BET, SA S control Pluronic P-105, Block copolymer of PEO-PPO Pluronic P-123, Block copolymer of PEO-PPO Poly(3-methylthiophene) Polyethylene glycol Polyethylene oxide Polypropylene oxide Pore size Pore volume Point of zero charge Molar ration of surfactant to TTIP Filtration resistance BJH pore size distribution Room temperature ionic liquid BET specific surface area TiO 2 particles prepared without either surfactants or ILs S BF4 TiO 2 particles prepared with [bmim][bf 4 ] S PF6 TiO 2 particles prepared with [bmim][pf 6 ] S PF6,T80 TiO 2 particles prepared with [bmim][pf 6 ] and Tween 80 S surfactant TiO 2 particles prepared with surfactant 20

34 SAXS SEM SGC T T20 T80 TEM TGA TiO 2 TMP TOC TTIP UV Vis V pore WIRTIL Small angle X-ray scattering Scanning electron microscopy Sonogel carbon Numer of coatings Tween 20, Polyoxyethylene sorbitan monolaurate Tween 80 Polyoxyethylene sorbitan monooleate Transmission electron microscopy Thermogravimetric analysis Titanium dioxide, Tatania Trnasmembrane pressure Total organic carbon Titanium tetraisopropoxide Ultraviolet Visible Pore volume Water immiscible room temperature ionic liquid X-100 Triton X-100, Polyethylene glycol tert-octylphenyl ether XPS XRD 4-CP X-ray photoelectron spectrometer X-day diffraction 4-chlorophenol 21

35 CHAPTER 1 Introduction 22

36 Nanocrystalline TiO 2, one of the most important semiconductor oxide materials, has been extensively researched for the development of photocatalytic water treatment processes. The TiO 2 photocatalytic process is characterized by high reaction rates and short treatment times due to rapid oxidation reactions by hydroxyl radicals. However, the catalytic activity should be increased significantly before the process can become competitive for full-scale applications in water treatment. In addition to the intrinsic photocatalytic properties of TiO 2, which are directly related to its crystal phase, crystallite size, and defect structures, its structural properties, including surface area, porosity, morphology, and pore size distribution, are of great importance. Another important aspect for widespread use of TiO 2 photocatalysts in a variety of applications is their immobilization onto various substrates in the form of films and membranes since they are usually used in the form of nanoparticles in suspension, which need to be removed after their application. In this dissertation, we explore nanotechnological and new materials chemistry procedures for the synthesis of TiO 2 photocatalysts with the desirable properties mentioned above. Controlling materials at the nano-level can accelerate the development of new types of products with improved properties and functionalities for environmental applications. 23

37 1.1 Motivation TiO 2 Photocatalysis The inability of conventional treatment processes to decompose organic chemicals and inactivate pathogenic microorganisms in water resources has propelled the development of modern water treatment processes. Among them, advanced oxidation technologies (AOTs) and membrane separation technologies (MSTs) have received great attention for environmental remediation [1 3]. Along with the ability of MSTs to reject most contaminants in water, TiO 2 /UV photocatalysis is among the most promising AOTs due to the effectiveness of TiO 2 to generate hydroxyl radicals and its environmentally benign properties [4 8]. However, TiO 2 photocatalysis is limited for its widespread applications by: (i) the need of photo-excitation energy beyond the bandgap of TiO 2, (ii) the low light utilization efficiency, (iii) the low specific surface area and porosity, and (iv) the requirement of removal of TiO 2 catalysts after their applications in case of TiO 2 particles in suspension. In order to commercialize the process as a full-scale technology, it is critial to increase the catalytic activity of TiO 2 significantly and to fabricate TiO 2 films and membranes immobilized onto support materials [9]. In addition to the intrinsic photocatalytic activity of TiO 2, which is directly related to its crystallographic properties, its structural properties such as surface area, pore volume and pore size distribution are also of importance due to the accessibility of the reactants to/from the active sites in TiO 2 catalyst. As a result, the development of TiO 2 particles and immobilized films and membranes with enhanced catalytic activity and structural properties is urgently needed. 24

38 1.1.2 Proposed Idea I: Modification of Sol-Gel Methods with Surfactants Sol-gel methods refer to wet chemistry-based synthesis of solid inorganic materials from liquid molecular precursors [9 13]. Following the first work for the synthesis of a welldefined mesoporous molecular sieve material, the preparation of nanocomposite structures has been of interest for advanced materials with functionalities [14,15]. Among the synthetic routes, sol-gel methods employing pore directing agents, including surfactants and block copolymers have been extensively researched so far to tailor-design the properties of inorganic materials [15 19]. Most research studies have focused on the synthesis of silica with well-defined mesoporous structure [20]. However, not many studies have been done for the fabrication of TiO 2 using this surfactant-template approach because the poor reactivity of TiO 2 precursor with surfactants makes it difficult to form strong organic-inorganic composite and thus the initial mesoporous sturcutre of TiO 2 is easily collapsed during heat treatment. In this study, we attempt to utilize the sol-gel methods employing surfactants as pore directing agents in controlling the structural properties of mesoporous TiO 2 [21]. We address the following crucial aspects: (i) high photocatalytic activity by enhancing surface area and materials crystallinity, (ii) uniform immobilization of TiO 2 onto substrates in the form of thin films and membranes, (iii) controlled pore size and porosity for enhanced permeation flux and selectivity in case of TiO 2 membranes, and (iv) versatile applications of TiO 2 thin films and membranes in photocatalysis, disinfection, separation, and other promising areas including development of new sensors. 25

39 1.1.3 Proposed Idea II: Modification of Sol-Gel Methods with Ionic Liquids The presence of water in sol-gel methods is crucial to the formulation of the sol containing alkoxide TiO 2 precursor, alcohol solvent and other additives. However, hydrolysis and condensation reactions of TiO 2 precursors are too fast in the presence of excessive amounts of water, resulting in immediate precipitation of TiO 2 particles with uncontrolled and undesired properties including low surface area and amorphous phase. In addition, during ageing of the sol and drying of the gel, solvent evaporation can cause pore shrinkage and collapse. In this study, we explore a more effective synthesis route for TiO 2 catalysts with controlled properties, an innovative modification of the traditional sol-gel methods with ionic liquids (ILs) as reaction media. ILs are exceptional type of solvents that have practically no vapor pressure and possess tunable solvent properties. The low vapor pressure of ILs can eliminate the problem of gel shrinkage during ageing. In addition, water immiscible room temperature ionic liquids (WIRTILs) containing small amounts of water can reduce the rate of hydrolysis reaction of the titanium alkoxide precursor. 1.2 Background TiO 2 Photocatalysis The photocatalytic activity of TiO 2 (titanium dioxide, titania) along with its excellent physicochemical properties has propelled the development of new technologies for environmental and other applications [22,23]. Titanium dioxide is found in four different phases: amorphous, brookite, anatase, and rutile [24 26]. The last three are crystalline 26

40 phases used in photocatalytic processes. Anatase is more stable and active but it is transformed to rutile after heating at higher temperatures (i.e., C). TiO 2 is a semiconductor exhibiting a discrete energy region between its valence and conduction bands, called the band gap (E g ) [5]. When the TiO 2 catalyst is provided with sufficient energy (usually by UV radiation with above the band gap energy of TiO 2, 3.2 ev for anatase and 3.0 ev for rutile), electrons and holes are generated in the conduction and valence bands, respectively [27 33]. Redox reactions at the catalytic surface involve the formation of superoxide radical anion (O - 2 ) via the reduction of oxygen adsorbed at catalytic sites and the formation of hydroxyl radicals ( OH) via the interaction of holes with surface hydroxyl groups [29,31]. The hydroxyl radicals readly attack organic compounds to transform their molecular structure to simpler organics, then finally to carbon dioxide, water and other inorganic species [30 32]. Two types of photocatalytic reactors were used for studies of water treatment; those that utilize TiO 2 nanoparticles in suspension and those that utilize TiO 2 films immobilized on substrate [34 39]. Although characterized by higher catalytic surface area and less mass transfer limitations, systems with TiO 2 nanoparticles in suspension require an additional processs to remove TiO 2 particles from the effluent [40]. On the other hand, utilizing immobilized TiO 2 films for water treatment is more practical and it is a prerequisite key feature for the development of continuous-mode water treatment systems [40]. Moreover, when TiO 2 films immobilized onto porous substrates are used as a skin layer of membranes, the TiO 2 membranes can achieve multiple actions including separation function to retain water contaminants and photocatalysis to degrade them, which is one of the focus aspects in this study [41,42]. 27

41 1.2.2 Sol-Gel Methods Sol-gel methods refer the formation of solid inorganic materials from liquid molecular precursors via room-temperature wet chemistry-based procedures without melting inorganic raw materials at elevated temperatures. Sol-gel processes involve a series of processing steps: i) preparation of a polymeric or particulate sol containing inorganic materials from the hydrolysis and condensation reactions of alkoxide inorganic precursors, ii) deposition of substrate into the sol for the formation of a uniform coating, iii) evaporation of the solvent and other volatile compounds for the solidification of the gel, iv) low temperature drying for more condensed state of inorganic network, and v) high temperature heat treatment for the removal of organics and crystallization of the solid material [43 48]. Sol-gel technology is also a popular method for the fabrication of inorganic materials: powders, coatings, thin films, membranes, adsorbents, fiber-optics, and sensors [9 13]. It has major advantages over other deposition technologies: i) there is a wide range for selection of precursor compounds and support materials, ii) the precursors can be used at selected concentration for the precise control of the final film, iii) dopants or other metals with catalytic properties can be easily incorporated in the sol, and iv) the porosity of the material can be finely tuned by controlling certain parameters in the preparation steps [49-51]. 28

42 1.2.3 Self-Assembling of Surfactants Following the pioneering work for the first synthesis of mesoporous molecular shieves (zeolite MCM-41), the preparation of well-defined nanocomposite structures has been considered as one of the synthetic challenges in fabricating advanced functional materials [14,18]. Among the synthetic routes, modified sol-gel methods using pore templating agents including micelles, block copolymers, and simple surfactant molecules, have been attaractive methods during the last decade as promising approaches for the tailordesigning of the inorganic structure at the nano level [52 54]. Scheme 1.1 shows the molecular structures of amphiphilic organic molecules used in this study, and their properties are summarized in Table 1.1. Above a certain concentration (critical micellar concentration, CMC), surfactants and some other organic compounds with long hydrocarbon chain are known to selfassemble into cylindrical, spherical or plannar micelles held together by van der Waals forces and hydrogen bonds in water-rich conditions. These micelles subsequently transform into liquid crystals of hexagonal, cubic or lamellar structures. During thermal treatment, the surfactants and other organic residues are pyrolyzed, leaving a pore structure. On the other hand, in case of water-poor conditions, the stufactants molecules start to self-organize to form reverse micellar structures. Then titanium alkoxide precursors are hydrolyzed inside the reverse micelles containing water, forming an inorganic core/organic shell composite. After removing the surfactant shell frame, welldefined TiO 2 nanoparticles are obtained. Examples of such organic/inorganic composites are shown in Scheme

43 (a) (b) (c) (d) and (e) CH3 HO-(CH2CH2O)X-(CH2CHO)Y-(CH2CH2O)X'-H Scheme 1.1: Structure of amphiphilic organic molecules: (a) polyoxyethylene sorbitan monolaurate (Tween 20), (b) polyoxyethylene sorbitan monooleate (Tween 80), (c) polyethylene glycol tert-octylphenyl ether (Triton X-100), and (d) and (e) polyethylene oxide-polypropylene oxide-polyethylene oxide triblockcopolymers (Pluronic 105 and Pluronic 123). The source of the chemical structures is Sigma-Aldrich Co. 30

44 Table 1.1: Properties of amphiphilic organic molecules. Surfactant Hydrophobic group MW a (g/mole) CMC a (mg/l) HLB a Tween 20 Lauric acid 1228 ~ Tween 80 Oleic acid 1310 ~13 15 X-100 Phenyl ether 625 ~160 Pluronic 105 Propylene (50%) Pluronic 123 Propylene (70%) a MW: molecular weight, CMC: critical micelle concentration, and HLB: hydrophilelipophile balance. Micelle in water-rich condition Ti O O O Ti Ti-O-Ti- O O Ti O Ti O O Ti O Ti Ti Ti O O Ti O O Ti Organic/inorganic composite -Ti-O-Ti O Ti O O Ti O Porous network O Ti O Ti O Ti O O Ti O O Ti O Ti Ti-O-Ti- -Ti-O-Ti O O Ti O Ti O Ti O O O Ti Ti Ti O O O O O Ti O O Ti Ti Ti Ti O O O Ti Ti Ti O O O Ti O O Ti Ti Ti Ti O O O Ti Ti Ti Reverse micelle in water-poor condition Nanoparticle Scheme 1.2: Surfactant micelle or reverse micelle formation followed by the formation of TiO 2 porous network or TiO 2 nanoparticle. 31

45 1.2.4 Ionic Liquids Ionic liquids (ILs) are liquids that are comprised entirely of ions, usually one type of anion and one type of cation with melting point lower than 100 ºC [55 59]. ILs are clearly different from molten salts. For example, molten salts, such as molten sodium chloride at 812 ºC, are obtained at much higher temperatures. In addition to relatively large anion, the larger size of the organic cation in ILs and its asymmetric structure are the main reasons for the low melting points of ILs. Room temperature ILs (RTILs) refer to ILs at or below room temperature. The benign properties of ILs along with their thermal, chemical, electrochemical, and photochemical stability have captured the interest of chemists, chemical engineers and more recently environmental engineers. ILs are currently explored as a New Generation of Solvents in electrochemistry, catalysis, separations, chemical synthesis, and photochemistry. The approach introduced in this study is to use water immiscible RTILs (WIRTILs), 1-butyl-3-methylimidazolium hexafluorophosphate shown in Scheme 1.3 as an additional reaction medium with alcohol used in traditional sol-gel methods to control the hydrolysis rate of titanium precursors. Scheme 1.3: Structure of water immiscible room temperature ionic liquid, 1-butyl-3- methylimidazolium hexafluorophosphate. The source of the chemical structure is Sigma- Aldrich Co. 32

46 1.2.5 Environmental Nanotechnology Nanotechnology has tremendous potential to profoundly change current science and engineering and to improve our standard of living. Nanotechnology is any technology which exploits phenomena or structures that can only occur at the nanometer scale. The United States National Nanotechnology Initiative defines it as follows: nanotechnology is the understanding and control of matter at dimensions of roughly 1 to 100 nanometers, where unique phenomena enable novel applications [60]. Such phenomena include quantum confinement, which can result in different electromagnetic and optical properties of a material between nanoparticles and the bulk material. For example, TiO 2 nanoparticles (quantum dots) exhibit attractive optical, electrical, chemical and catalytic properties, which are different from those of bulk TiO 2 [61]. Nanotechnology is an extension of the field of materials science in conjunction with physics, chemistry, mechanical engineering, bioengineering, chemical engineering, and environmental engineering. More broadly, nanotechnology includes the many techniques used to create structures at a size scale below 100 nm, including those used for fabrication of nanowires and nanotubes, and further including molecular self-assembly techniques employing block copolymers and surfactants. Specifically, when this concept is associated with environmental science and engineering to synthesize nanostructured functional materials which are used for environmental remediation, the technology is known with the term environmental nanotechnology. Consequently, controlling materials at the nano-level can accelerate the development of new types of products with improved properties and functionalities for environmental applications [62]. 33

47 1.3 Objectives and Challenges Preparation of Photocatalytic TiO 2 Particles, Films, and Membranes Using Sol-Gel Methods Modified with Surfactants Objective I: Role of Surfactants The objective of this study is to investigate the effect of addition of various surfactants into TiO 2 sol on the physicochemical and structural properties of TiO 2 materials. Specifically, we are interested in (i) surfactant self-assembly and its transformation to pore structure in TiO 2 inorganic network, and (ii) heat treatment conditions for the removal of surfactant templates and crystallization of the TiO 2 catalysts. This part also includes fundamental studies on the physicochemical phenomena that take place during the sol-gel synthesis of TiO 2. The following specific tasks are investigated: surfactant type and concentration, titanium alkoxide precursor, viscosity control agent, other organic additives (e.g., acetic acid), sol ageing and drying, and heat treatment Objective II: TiO 2 Thin Films and Membranes Recently, many research studies have been carried out to immobilize TiO 2 catalyst onto various substrates as the form of thin films and membranes for versatile applications. Moreover, if the TiO 2 material is immobilized onto porous support, the TiO 2 photocatalytic membrane reactors may gain tremendous popularity because of their multiple functions including photocatalysis and separation. This study aims at elucidating synthesis and dip-coating conditions for the fablication of uniform TiO 2 thin films and 34

48 membranes on various substrates. The prepared nanocomposite TiO 2 /inorganic membranes will have improved photocatalytic properties and enhanced selectivity and water permeability. These features are crucial for the fabrication of robust integrated advanced oxidation reactors and membranes for the purification and disinfection of water. The following specific tasks are investigated: support material, dip-coating, and heat treatment Obje ctive III: Versatile Applications of TiO 2 These TiO 2 films and membranes inherently possess multiple and simultaneous functions including photocatalytic decomposition of organic pollutants, destruction of biological toxins, inactivation of pathogenic microorganisms, physical separation of contaminants, and anti-biofouling action. These photocatalytic purification systems can be used as stand-alone technologies or as supplementary and complementary to existing treatment technologies. The TiO 2 coatings can also be applied in developing highly sensitive electrodes and sensors to detect organic moleculaes of interest in water due to the high catalytic activity of nanostructured TiO 2. The following specific tasks are investigated: catalytic activity, organic decomposition, disinfection, separation, water permeability, anti-biofouling and sensor development Preparation of Photocatalytic TiO 2 Particles Using Sol-Gel Methods Modified with Ionic Liquids 35

49 Objective IV: Exploring Ionic Liquids as Reaction Media in Sol-Gel Methods In this study, we explore RTILs as an additional reaction medium in conjunction with alcoholic solvents used in the traditional sol-gel methods as well as a template material. This approach not only expands the applications of Green Chemistry in the processing of advanced nanomaterials but has also the potential to design materials with improved properties and functionalities. Here, considering the special characteristics of WIRTILs such as water immscibility and zero vapor pressure, we can cotroll the hydrolysis of titanium precursors and prevent the shrinkage of inorganic network during the sol-gel synthesis of TiO 2. Here, ILs as green solvents are explored for the preparation of TiO 2 photocatalytic nanoparticles. The following specific tasks are investigated: RTIL type, molar ratio of RTIL/titanium alkoxide and water/titanium alkoxide, chemical reaction and bondings between RTIL and titanium precursors, and RTIL removal Characterization and Evaluation The characterization of TiO 2 photocatalysts prepared in this study is performed using state of the art instrumentation to unveil crystallographic structure, morphology, nanostructure, elemental composition, optical properties, film thickness, and other properties. The TiO 2 photocatalysts are also tested for the destruction and elimination of emerging water contaminants such as cyanobacterial toxins (e.g., microcystin-lr) and other contaminants of concern such as organics found in human perspiration and waste (e.g., creatinine) and in industrial wastewater (e.g., methylene blue dye and chlorinated phenolics) shown in Scheme 1.4. Moreover, various functions of TiO 2 membranes are evaluated for environmental applications in terms of water permeability and selctivities. 36

50 (a) (b) (c) (d) Scheme 1.4: Structure of organic contaminants used in this study: (a) methylene blue, (b) creatinine, (c) 4-chlorophenol, and (d) microcystin-lr. The source of the chemical structures is Sigma-Aldrich Co. 37

51 1.4 References [1] H.Zhou, D.W. Smith, Can. J. Civil. Eng. 28 (2001) 49. [2] B.E. Reed, M.R. Matsumoto, R. Viadero, R.L. Segar, R. Vaughan, D. Masciola, Wat. Environ. Res. 71 (1999), 584. [3] AWWA, Water Treatment Membrane Processes; McGraw Hill: New York, [4] A. Fujishima, K. Honda, Nature 238 (1972) 37. [5] M.R. Hoffmann, S.T. Martin, W. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995) 69. [6] R. Molinari, M. Mungari, E. Drioli, A. Di Paola, V. Loddo, L. Palmisano, M. Schiavello, Catal. Today 55 (2000) 71. [7] S.Y. Kwak, S.H. Kim, S.S. Kim, Environ. Sci. Technol. 35 (2001) [8] C. Wei, W.-Y. Lin, Z. Zainal, N.E. Williams, K. Zhu, A.P. Kruzic, R.L. Smith, K. Rajeshwar, Environ. Sci. Technol. 28 (1994) 934. [9] H. Al-Ekabi, Proceedings of the Workshop on Advanced Oxidation Technologies: Fundamental and Environmental Applications; Albuquerque, New Mexico, [10] C. Agrafiotis, A. Tsetsekou, J. Europ. Cer. Soc. 20 (2000) 825. [11] C.J. Brinker, G.W. Scherer, Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing; Academic Press: San Diego, [12] L.L. Hench, J.K. West (Eds.), Chemical Processing of Advanced Materials; John Wiley: New York, [13] J.H. Fendler, Nanoparticles and Nanostructured Films; Wiley-VCH: Weinheim, Germany,

52 [14] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992) 710. [15] C.J. Brinker, Current Opinion in Solid State and mater. Sci. 1 (1996) 798. [16] J.Y. Ying, C.P. Mehnert, M.S. Wong, Angew. Chem. Int. Ed. 38 (1999) 56. [17] S. Mann, G.A. Ozin, Nature 382 (1996) 313. [18] P. Alexandridis, L. Yang, Macromolecules 33 (2000) [19] R.L. Putnam, N. Nakagawa, K.M. McGrath, N. Yao, I.A. Aksay, S.M. Gruner, A. Navrotsky, Chem. Mater. 9 (1997) [20] C.Y. Tsai, S.Y. Tam, Y.F. Lu, C.J. Brinker, J. Membr. Sci. 169 (2000) 255. [21] Y. Lu, G. Cao, R.P. Kale, S. Prabakar, G.P. Lopez, C.J. Brinker, Chem. Mater. 11 (1999) [22] R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni, E. Kojima, A. Kitamura, M. Shimohigoshi, T. Watanabe, Nature 388 (1997) 431. [23] M. Gratzel, Nature 414 (2001) 338. [24] J.S. Reed, Principles of Ceramic Processing; John Wiley: New York, [25] A.K. Cheetham, P. Day (Eds.), Solid State Chemistry; Oxford University Press: New York, [26] C.M. Freeman, J.M. Newsam, S.M. Levine, C.R.A. Catlow, J. Mater. Chem. 3 (1993) 531. [27] N.H. Ince, G. Tezcanli, R.K. Belen, I.G. Apikyan, Appl. Cat. B: Environ. 29 (2001) 167. [28] A. Fujishima, K. Honda, Nature 238 (1972) 37. [29] M.A. Fox, M.T. Dulay, Chem. Rev. 93 (1993)

53 [30] O. Legrini, E.Oliveros, A.M. Braun, Chem. Rev. 93 (1993) 671. [31] A.L. Linsebigler, G.Q. Lu, J.T. Yates, Chem. Rev. 95 (1995) 735. [32] D.F. Ollis, H. Al-Ekabi (Eds.) Photocatalytic Purification and Treatment of Water and Air; Elsevier Science Publishers BV: Amsterdam, [33] D.F. Ollis, E. Pelizzetti, N. Serpone, Environ. Sci. Technol. 25 (1991) [34] A.E. Cassano, O.M. Alfano, Catal. Today 58 (2000) 167. [35] A.E. Cassano, O.M. Alfano, J. Phys. Chem. 213 (1999) 17. [36] A.E. Cassano, C.A. Martin, R.J. Brandi, O.M. Alfano, Ind. Eng. Chem. Res. 34 (1995) [37] H.Y. Ha, M.A. Anderson, J. Environ. Eng. 122 (1996) 217. [38] N.J. Peill, M.R. Hoffmann, Environ. Sci. Technol. 29 (1995) [39] A.K. Ray, A.A.C.M. Beenackers, AIChE Journal 43 (1997) [40] M. Bideau, B. Claudel, C. Dubien, L. Faure, H. Kazouan, J. Photochem. Photobiol. A 91 (1995) 137. [41] R.L. Pozzo, M.A. Baltanas, A.E. Cassano, Catal. Today 39 (1997) 219. [42] G. Balasubramanian, M.T Suidan, D.D. Dionysiou, V. Subramanian, I. Baudin, J.-M. Laine, J. Mater. Sci. 38 (2003) 823. [43] C.J. Brinker, G.W. Scherer, Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing; Academic Press: San Diego, [44] L.L. Hench, J.K. West (Eds.), Chemical Processing of Advanced Materials; John Wiley: New York, [45] J.H. Fendler, Nanoparticles and Nanostructured Films; Wiley-VCH: Weinheim, Germany,

54 [46] C.J. Brinker, Current Opinion in Solid State and Mater. Sci. 1 (1996) 798. [47] C.J. Brinker, A.J. Hurd, G.C. Frye, P.R. Schunk, C.S. Ashley, In: L.L. Hench, J.K. West (Eds.), Chemical Processing of Advanced Materials; John Wiley: New York, [48] L.F. Francis, Materials and Manufacturing Processes 12 (1997) 963. [49] U.S. Bjorkert, R. Mayappan, D. Holland, M.H. Lewis, J. Eur. Ceram. Soc. 19 (1999) [50] Y. Tanaka, J. Cer. Soc. Jap. 107 (1999) 47. [51] L.-Q. Wu, P. Huang, N. Xu, J. Shi, J. Membr. Sci. 173 (2000) 263. [52] M. Singh, M.K. Trivedi, J. Bellare, J. Mat. Sci. 34 (1999) [53] F. Chen, L. Huang, Q. Li, Chem. Mater. 9 (1997) [54] de R.S.A. Lange, J.H.A. Hekkink, K. Keizer, A.J. Burggraaf, J. Membr. Sci. 99 (1995) 57. [55] R.D. Rogers, K.R. Seddon (Eds.), Ionic Liquids-Industrial Applications to Green Chemistry, ACS Symposium Series 818; American Chemical Society: Washington, D.C., [56] K.R. Seddon, J. Chem. Tech. Biotechnol. 68 (1997) 351. [57] K.R. Seddon, Kinet. Catal. 37 (1996), 693. [58] J.D. Holbrey, K.R. Seddon, Clean Products and Processes 1 (1999) 223. [59] P. Wasserscheid, W. Keim, Angew. Chem. Int. Ed. 39 (2000) [60] National Nanotechnology Initiative: [61] F, Caruso, Adv. Mater. 13 (2001) 11. [62] Y. Liu, H. Choi, D.D. Dionysiou, G.V. Lowry, Chem. Mater. 17 (2005)

55 CHAPTER 2 Effect of Surfactant in a Modified Sol on the Physicochemical Properties and Photocatalytic Activity of TiO 2 Nanoparticles 42

56 This study describes the effect of amphiphilic organic molecules (surfactants) in a TiO 2 sol on the physicochemical properties and photocatalytic activity of crystalline TiO 2 nanoparticles prepared via a simple sol-gel route at high temperatures from 400 to 800 C. The addition of polyoxyethylene sorbitan surfactant and polyethylene oxide and polypropylene oxide triblock copolymer as particle size inhibitors and pore directing agents into a stable titania sol affected the physicochemical properties of TiO 2 nanoparticles such as their crystallographic structure, morphology, and defect structure. With the addition of the surfactants, the ratio of anatase and rutile crystal phases of TiO 2 was controlled and an active anatase crystal phase was maintained during heat treatment up to 800 ºC. Decrease in the sintering rate and inhibition in crystal growth were also observed, which resulted in higher surface area and inhibition of crystallite aggregation. Bulk defects in TiO 2 decreased while surface defects increased as a result of the addition of surfactants. These physicochemical properties of TiO 2 nanoparticles were correlated with photocatalytic degradation of 4-chlorophenol in water. The results revealed that high crystallinity, anatase crystal phase, high specific surface area, surface defects, and segregated morphology of TiO 2 nanoparticles, which were induced by the addition of surfactants, were more advantageous for enhancing photocatalytic destruction of the model organic compound tested in the study. 43

57 2.1 Introduction Nanocrystalline TiO 2, one of the most important semiconductor oxide materials, has been extensively researched for the development of photocatalytic water treatment processes [1 8]. Although the TiO 2 photocatalytic process is characterized by high reaction rates and short treatment times due to rapid oxidation reactions by hydroxyl radicals, it is believed that the catalyst activity should be increased by at least one order of magnitude before the process can become competitive for full-scale applications in water treatment [9]. In addition to the surface area and morphology of the TiO 2 catalyst, its crystallographic properties such as crystal phase and crystallite size, as well as defect structures on its surface and in bulk largely affect its catalytic activity. For example, the anatase crystal phase has been generally known for having a far higher activity than the other known phases (brookite and rutile) and amorphous TiO 2 particles have no considerable photocatalytic activity due to many defects in the bulk [10 13]. So, controlling the crystallographic and morphological properties of TiO 2 material via alternative synthesis procedures could be one approach to develop such catalyst with activity that can be high enough to make the process attractive for large scale applications for the destruction of harmful organic contaminants in water. In order to obtain high photocatalytic activity, the reduction of bulk defects is generally achieved by heat treatment at high temperature. Unfortunately, the surface area of the TiO 2 material is reduced dramatically and the metastable anatase phase transforms to rutile during calcination [14]. Considering bulk defects, porosity, surface area, and active anatase crystal phase, it is expected that TiO 2 calcined at high temperatures in a 44

58 certain range without transforming anatase to rutile and collapsing the porous structure could exhibit higher photocatalytic activity [15,16]. It has been reported that addition of certain inorganic materials, including alumina, silica, and activated carbon would inhibit the crystal phase transformation of TiO 2 material and enhance its thermal stability [16 19]. Although amphiphilic organic molecules (i.e., surfactants) have been extensively used in sol-gel methods to control structural properties of materials, there has been no attempt to investigate the effect of adding organic molecules into TiO 2 sol, considering beneficial aspects such as those in the case of inorganic additives [20 25]. For example, in the sol-gel methods modified with surfactants, the as-synthesized amorphous TiO 2 material has high surface area but weak inorganic network walls. During heat treatment for removing organic templates and increasing materials crystallinity, this material undergoes thermal nucleation and crystal growth of the anatase phase. This results in the destruction of pore structure, reduction of surface area, and formation of the rutile phase [24,25]. As a result, such studies have mainly focused on the low temperature preparation of mesoporous TiO 2 inorganic network with high surface area but low crystallinity [23]. There are only a few studies on the properties and photocatalytic activity of nanocrystalline TiO 2 photocatalyst prepared with surfactants at high temperature [24]. In this study, nanocrystalline TiO 2 particles were prepared via a simple sol-gel method modified with surfactants at elevated temperatures to manipulate the physicochemical properties of TiO 2 photocatalysts such as their crystallographic structure, morphology, and defect structure [26]. These properties were correlated with the photocatalytic activity of the prepared catalyst for the degradation of 4-chlorophenol (4- CP). 45

59 2.2 Experimental Sol Preparation Several studies have focused on the synthesis of a stable sol. In this work, we adapted a method for titania sol preparation already described in the literature, which consists of titanium tetraisopropoxide (TTIP), isopropanol (i-proh), diethanolamine (DEA) and water because the sol is very stable and has versatile applications in preparing photocatalytic TiO 2 films [6,27,28]. First, a suitable amount of DEA (Aldrich) as a sol stabilizer was added to 0.5 M TTIP (Aldrich) solution dissolved in i-proh (Fisher) at a molar ratio of DEA/TTIP = 4. Then, water was added drop by drop under vigorous stirring at a molar ratio of H 2 O/TTIP = 2. A clear yellowish stable sol was obtained. Two types of surfactants were selected: Tween surfactant (T20; polyoxyethylene sorbitan monolaurate; molecular weight of 1228 g/mol; Aldrich) and Pluronic block copolymer (P105; triblock copolymers of polyethylene oxide (PEO)-polypropylene oxide (PPO)-polyethylene oxide (PEO); molecular weight of 6500 g/mol; BASF). Compared to other commonly used toxic additives (i.e., long-chain ammonium salts), these organics are relatively inexpensive, biodegradable, non-toxic, and easily-removable and exhibit the ability to control large inorganic clusters [29 31]. According to quasi elastic light scattering data and a fluorescent probe study for the Pluronic block copolymer, further addition of EO/PO block copolymer after monolayer saturation on particle surface does not lead to classical micellization as in Tween surfactant [32]. This indicates a selfassociation of the molecules on the particle surface, forming a multilayer. Surfactant 46

60 solutions were prepared by dissolving the surfactants in i-proh at various concentrations, R (molar ratio of surfactant/ttip) Formation of Crystalline TiO 2 Nanoparticles A modified titania sol was prepared by adding the titania sol into the surfactant solution drop by drop under vigorous stirring. In this sol-gel method, synthesis of TiO 2 follows two possible mechanisms: (1) the formation of an inorganic network on self-organized organic molecules, and (2) the formation of an organic layer around an inorganic network in a sol, mainly depending on the concentration and nature of the organics and mixing order of the ingredients [33,34]. From the different mechanisms of self-organization (i.e., classical micellization of T20 and multilayer formation of P105), Tween surfactant tends to follow mechanism 1 while Pluronic copolymer favors mechanism 2. No matter which mechanism is the most predominant or which pathway is most desirable for enhancing material properties, it is usually observed that the surfactants can significantly affect the physicochemical properties of the final TiO 2 catalyst. Among various preparation strategies using modified sol-gel methods, the procedure mentioned above is the most successful in controlling the structural properties of TiO 2 nanoparticles. The modified sol was dried and aged at room temperature for 12 h, and heattreated using a multi-segment programmable furnace (Paragon Model HT-22-D, Thermcraft) [35]. The furnace temperature was increased at a ramp rate of 3.0 C/min to 100 C and held for 1 h to completely remove water and solvent. The temperature was further increased to a certain temperature in the range from 400 to 800 C and held for 1 h. The product was ground into a powder for further material characterization and 47

61 photocatalytic activity test. For convenience, the abbreviations S surfactant and S control denote TiO 2 prepared with surfactants and TiO 2 prepared without surfactants, respectively Material Characterization A Kristalloflex D500 diffractometer (Siemens) with Cu Kα (λ = Å) radiation was used for the X-ray diffraction (XRD) analysis to determine the crystal structure and crystallinity of the TiO 2 catalyst. For each scan, 2θ was incremented from 20.0 to 60.0 degrees with a step of 0.1 and time to step ratio of 1.0. Porosimetry analyzers (ASAP 2020 and Tristar 3000; Micromeritics) were used to measure Brunauer, Emmett, and Teller (BET) specific surface area, pore size distribution, and porosity of the materials, using nitrogen adsorption and desorption isotherms. Before the measurement, the samples were purged with nitrogen gas for 2 h at 150 C using Flow prep 060 (Micromeritics). In order to visually examine the morphology of TiO 2 at the nano-level, a JEM-2010F high resolution transmission electron microscope (HR-TEM, JEOL) was used. The samples were dispersed in methanol (HPLC grade, Pharmco) using an ultrasonicator (2510R-DH, Bransonic) for 10 min and fixed on a carbon-coated copper grid (LC200-Cu, EMS). The elemental composition analysis of the TiO 2 particles was performed using an energy dispersive X-ray spectroscope (EDX, Oxford Isis) connected to HR-TEM. For accurate measurement of trace elements on the material surface and the oxidation states of Ti, an X-ray photoelectron spectroscope (XPS, Perkin-Elmer 5300) with Mg Kα X-rays was used with a step size of 0.5 ev for survey spectrum and 0.05 ev for high resolution spectrum at a take-off angle of 45º. Thermogravimetric analysis (TGA, TA instruments 48

62 2050) was performed in the presence of air at a ramp rate of 3 ºC/min to measure the weight change of the TiO 2 gel network during calcination Measurement of Photocatalytic Activity An experimental set up was constructed for determining the adsorption capacity and photocatalytic activity of TiO 2 nanoparticles. The reactor was comprised of a cylindrical cell made of borosilicate glass (i.d. 5.5 cm). Four 15 W integrally filtered low pressure mercury UV tubes (Spectronics) emitting near UV radiation ( nm) with a peak at 365 nm were positioned at two opposite sides of the reactor. The incident light intensity was 1.39 mw/cm 2 measured at the center of the reactor using a UV radiation meter (IL1700, International Light). The TiO 2 nanoparticles were thoroughly ground and predispersed in water by sonication (Model B-22-4, Branson Ultrasonic Cleaner). Aeration of the solution was achieved by dispersing fine air bubbles through a ceramic diffuser in an additional flask to avoid any hindrance to UV penetration caused by air bubbles into the reaction solution. Prior to its addition to the flask, the air was passed through an activated carbon column to remove any contaminants and through a humidifier column to saturate the air. A cooling fan maintained the solution temperature at around 25 ± 3 C. The photocatalytic activity was evaluated using a 4-chlorophenol (4-CP, Aldrich) solution with an initial concentration of 50 mg/l (dissolved organic carbon (DOC) = 28 mg/l). The following experimental conditions were kept constant: initial volume of reaction solution = 0.6 l; initial ph = 7.0 ± 0.1 without buffer; solution recirculation rate = 0.3 l/min; air flow rate = 0.5 l/min; and TiO 2 loading = 500 mg/l. The experiments were initially performed in the dark without UV radiation for 2 h to allow 4-CP adsorption 49

63 onto the TiO 2 particles. Then, UV radiation was applied for 4 h in order to degrade 4-CP by the photocatalytic activity of TiO 2 particles. The concentration of 4-CP was determined using a high performance liquid chromatography (HPLC; Agilent, Series 1100) equipped with a C-16 Discovery column (Supelco). The mobile phase was a mixture of acetonitrile (HPLC grade, Fisher) and 0.01 N sulfuric acid (96%, Fisher) in a ratio of 30:70 (v/v) with a flow rate of 1.5 ml/min. Total organic carbon (TOC or DOC) was measured with a TOC analyzer (TOC-VCSH with ASI-V, Shimadzu) after filtering the water samples with a 0.1 μm nylon membrane (Magna). 2.3 Results and Discussion Crystallographic Properties An EDX elemental analysis showed that the TiO 2 particles calcined at temperatures greater than 500 ºC were composed of mainly Ti and O elements. Nonstoichiometry of TiO 2 occurred at around 1:1.89 possibly due to the presence of a defect structure in the TiO 2 and oxidation states other than TiO 2 [36]. Results from a typical crystallographic analysis of TiO 2 prepared with T20 at 600 C are shown in Fig

64 (a) (101) Intensity (a.u.) (110) θ (degrees) R, [T20]/[TTIP] 1.00 (b) 0.00 Anatase fraction, F A Rutile fraction, F B 0.75 (0.0) Molar ratio of T20 to TTIP, R Fig. 2.1: (a) XRD patterns and (b) fraction of anatase and rutile phases of TiO 2 nanoparticles calcined at 600 C upon addition of T20 ( : anatase and : rutile). The fraction of anatase (F A ) and rutile (F R ) was estimated by assuming a linear relation between the XRD peak intensity and the fraction of given crystallite. The crystallite size was calculated from the XRD peak broadening analysis at (101). 51

65 The peaks indicate that the addition of T20 results in a mixture of anatase and rutile crystal phases at a controlled ratio. The maximum rutile fraction, F R, which was estimated according to the method of Spurr and Myers, was about 0.21 at R = 0. [37]. As the concentration of T20 increased, the rutile peak at (110) was significantly reduced and the crystallite size calculated using Scherrer s equation also decreased from 19 to 16 nm [38]. Similar behavior was also observed in the case of P105. The controllability of the TiO 2 crystal phases between anatase and rutile has important implications since anatase TiO 2 catalysts containing a small amount of rutile phase under an optimum ratio has been reported to have better photocatalytic activity than pure anatase for certain contaminants including phenol, salicylic acid, and p-coumaric acid [13,39] Anatase fraction, F A Increasing R T20, R=0.242 P105, R= T20, R= Rutile fraction, F R 0.0 Control, R= Calcination temperature ( o C) Fig. 2.2: Fraction of anatase and rutile phases of TiO 2 nanoparticles upon heat treatment. 52

66 The anatase-rutile phase transformation inhibition by the surfactants was clearly observed when varying the calcination temperature, as summarized in Fig For S control, the anatase-rutile phase transformation occurred at around 550 C, and it was accelerated between 600 to 650 C, and completed at below 700 C. However, the addition of surfactants suppressed the phase transformation significantly. Even at 600 C, only anatase was present. Rutile appeared at around 650 C. In the case of high dosage of surfactant (i.e., T20 at R = 0.242), a significant fraction of anatase phase was still present even at 800 C. The degree of crystallinity, defined as the relative ratio of crystalline fraction to amorphous fraction in the material, was approximately calculated based on the XRD peak analysis with background and noise removed [40]. Comparing the TiO 2 calcined at 400 C where the anatase crystal phase started to appear, the crystallinity apparently increased in the order of: S surfactant prepared with T20 at R = (68%) > S surfactant prepared with P105 at R = (63%) > S control at R = 0.0 (54%) Inhibition of Anatase to Rutile Crystal Phase Transformation It was previously reported that the addition of Triton X-100 surfactant (polyethylene glycol tert-octylphenyl ether) in a sol resulted in a mixture of anatase and rutile (up to 35%) at 550 ºC, compared to the pure anatase structure of the control TiO 2 [41]. However, the role of the surfactant in facilitating the anatase-rutile phase transformation was not studied in detail. The mechanism by which the additives either inhibit or enhance the phase transformation has been studied in relation with the defect structures [42]. Although it was suggested that increasing oxygen vacancies would tend to enhance the phase transformation while increasing titanium interstitials would retard the 53

67 transformation, the mechanism has not been fully understood [43]. The inhibition mechanisms for phase transformation observed in this study turned out to be very complicated. A significant portion of heat energy applied for crystallization and solidification of the TiO 2 material was utilized for the surfactant removal. However, the effect was considered negligible since the slow temperature ramp rate and long holding time were enough to remove most organics before reaching the target temperature below 500 C, which was evidenced by the TGA analysis of the prepared TiO 2 material, where no significant weight loss of samples was observed during heat treatment above 450 C. Min: 0, Max: (c) (b) F 1s O 1s Ti 2p N 1s C 1s N(E) (a) Binding Energy (ev) Fig. 2.3: XPS spectra of TiO 2 nanoparticles prepared with (a) control at R = 0.0, (b) T20 at R = 0.242, and (c) P105 at R = at 600 ºC. 54

68 In order to investigate impurities in TiO 2, which might affect the phase transformation by causing point defect structure such as interstitial and substitutional impurity atoms, an XPS analysis was conducted at the surface of the TiO 2 particles within a few nanometers. Fig. 2.3 shows prominent peaks of Ti 2p at ev and O 1s at ev for TiO 2 material and traces of C 1s at ev, F 1s at ev, and N 1s at ev which were considered impurities in the sol ingredients. The elemental composition is summarized in Table 2.1. Nonstoichiometry of TiO 2 at around 1: was slightly different from 1:1.89 measured using EDX possibly due to the presence of many hydroxyl groups at the very surface of TiO 2 nanoparticles. Although fluoride in TiO 2 has been reported to inhibit the phase transformation, it was difficult to fully explain the inhibition mechanism with the small increase in fluoride content [44]. The photoelectron peak for Ti 2p originates mainly from TiO 2 (IV) and Ti 2 O 3 (III). In the high resolution XPS analysis used for determining the Ti oxidation states at the 2p 3/2 level corresponding to binding energy of ev, the peaks of Ti(IV), Ti(III) and other oxidation states in S surfactant were almost identical to those in S control. Table 2.1: Elemental composition of TiO 2 nanoparticles calcined at 600 ºC. Element (atomic %) Control R = 0.0 T20 R = P105 R = O Ti Ti(IV) Ti(III) C F N S Na

69 2.3.3 Physical Structure As shown in Table 2.2, the addition of surfactants resulted in a slightly decreased surface area possibly due to pore coalescence and multi-micellar interactions at too high concentration of templating materials (i.e., DEA and surfactants), which was evidenced by TiO 2 particles prepared with a comparative sol containing no DEA. The surface area of m 2 /g was significantly high, compared to other reported TiO 2 materials prepared in conventional sol-gel methods at 600 ºC [45,46]. The structural properties of some representative TiO 2 particles at 600 C are summarized in Table 2.3. In spite of the longer chain length of P105 than that of T20, P105 produced TiO 2 materials with small pore, narrow pore size distribution, and low porosity, while T20 generally resulted in large pore size, wide pore size distribution, and high porosity. This could be attributed to the dependency of the final structure after calcination on the thermal stability of the material rather than the initial pore structure affected by the size, shape, and mechanism of self-organization of the templating material. The crystallite size estimated from the surface area was always larger than those from the XRD peak analysis and TEM image, implying that most particles were more or less aggregated [47]. Table 2.2: Surface area of TiO 2 nanoparticles calcined at 600 C. R S BET (m 2 /g) T20 P (10.4) a 69.8 (10.4) (12.5) 68.6 (11.4) (15.1) 65.5 (15.8) (24.3) 64.7 (21.3) (31.2) 59.4 (30.4) (36.3) - a The values in parenthesis were obtained from comparative sol without DEA. 56

70 Table 2.3: Structural properties of TiO 2 nanoparticles calcined at 600 ºC. Surfactant S BET V pore Porosity a D BJH (nm) Range BJH (nm) Crystal size b (nm) (m 2 /g) (cm 3 /g) (%) Ads. Des. Ads. Des. XRD TEM S BET Control T20, R= P105, R= T20, R= a Barrett, Joyner and Halenda pore sizes from N 2 adsorption and desorption branches. b XRD: calculated from XRD peak broadening analysis, TEM: measured from TEM image, and S BET : estimated from BET surface area, assuming all crystallites are spherical and present separately without aggregation. The effect of the calcination temperature on the surface area and crystallite size is presented in Fig The decrease in the surface area and the increase in the crystallite size of the calcined particles are due to the growth and sintering of crystallites. The surface area of S control decreased rapidly when the calcination temperature increased from 600 to 700 C. Subsequently, S surfactant started to have a slightly higher surface area than S control after calcination at high temperatures, indicating that the material prepared in the presence of surfactant was more thermally stable than that of the control. The thermal stability also prevented crystallite size growth Morphology and Defect Structure The HR-TEM images in Fig. 2.5 show the morphology of nanocrystalline anatase TiO 2 particles prepared at 600 C. Compared to S control with too much aggregated structure and no distinct boundaries between primary particles (see Fig. 2.5(a)), S surfactant was almost exclusively composed of well-defined small size spherical nanoparticles agglomerated 57

71 into large clusters (see Fig. 2.5(b)). This implies that the surfactant isolated the grains and thus acted as a particle size inhibitor. The TEM image at high magnification shown in Fig. 2.5(c) shows many randomly oriented nanocrystallites in size of 15 to 20 nm with sets of clearly resolved lattice fringes giving evidence that the material is highly crystalline. These results were more pronounced with an increase in surfactant concentration (a) BET surface area (m 2 /g) T20, R=0.242 P105, R= Control, R= Calcination temperature ( o C) 50 (b) Crystallite size (nm) T20, R=0.242 P105, R= Control, R= Calcination temperature ( o C) Fig. 2.4: Effect of calcination temperature on (a) surface area and (b) crystallite size of TiO 2 nanoparticles. 58

72 Fig. 2.5: HR-TEM images of anatase crystalline TiO 2 nanoparticles prepared with (a) control at R = 0.0 and (b) T20 at R = at 600 C. Many defect structures were observed in the regions highlighted with white circles in (b). (c) Highly crystallized area of the sample shown in (b). (d) Imperfectly rearranged crystallographic planes to adjacent ones as illustrated with two parallel lines, forming unique patterns due to the different contrast. 59

73 It is interesting to note that defect structures in the TiO 2 surface such as screw and mixed dislocations and interfacial defects such as grain boundaries and stack faults were more frequently observed in S surfactant with increasing surfactant concentration. Fig. 2.5(d) shows one of such imperfection structures (stack faults) in TiO 2 surface. Anatase particles are formed by crystallization of amorphous particles and grow by three possible pathways as follows: (1) direct aggregation of amorphous particles, (2) solid-state aggregation of anatase particles, and (3) atom-by-atom-recrystallization of anatase particles [48,49]. During heat treatment, the remaining organics are believed to inhibit the natural stages for crystallization and solidification of TiO 2 material by surrounding TiO 2 grains. This can also explain how the surfactants can act as a crystallite size inhibitor by obstructing growth and nucleation of adjacent anatase and rutile grains [17] Photocatalytic Activity The photocatalytic degradation of 4-CP by TiO 2 nanoparticles in suspension was tested to investigate the effect of the physicochemical properties of TiO 2 modified with the surfactants on its photocatalytic activity. Here, we report representative results obtained using TiO 2 prepared with T20 since similar results were obtained with P105. Fig. 2.6 shows adsorption and photocatalytic degradation of 4-CP by the TiO 2 particles prepared at different concentrations of T20 (see properties of the material in Fig. 2.1 and Table 2.2). In general, with increasing R, the adsorption capacity decreased due to the decrease in surface area while the photocatalytic activity was slightly improved mainly due to the retaining of the active anatase phase and partially due to the increased materials crystallinity. 60

74 Removal efficiency (%) (0.0) Photocatalytic degradation Adsoprtion Molar ratio of T20 to TTIP, R 4-CP DOC 4-CP (DOC) Fig. 2.6: Adsorption and photocatalytic degradation of 4-CP by TiO 2 nanoparticles prepared with T20 at different concentrations at 600 C. Error bars indicate the standard deviation of triplicated results. The heat treatment of TiO 2 particles in a certain temperature range has two opposite actions: enhancing materials crystallinity but reducing catalytic surface area. The effect of calcination temperature on adsorption and photocatalytic activity of the TiO 2 particles (see properties in Figs. 2.2 and 2.4) is shown in Fig The 4-CP adsorption capacity of TiO 2 nanoparticles was a function of their surface area. The results shown in Fig. 2.7(b) demonstrate the effect of the physicochemical properties of TiO 2 nanoparticles. The photocatalytic activity increased upon calcination temperature until the anatase phase began to transform to the rutile phase. In the case of S control, the photocatalytic activity continuously decreased because of the anatase-rutile phase transformation at above 500 ºC as well as the decreased surface area. 61

75 Removal efficiency (%) (a) R=0 R= Calcination temperature ( o C) 60 (b) Removal efficiency (%) R=0.242 R=0 R=0.242 R=0 4-CP DOC Calcination temperature ( o C) Fig. 2.7: (a) adsorption and (b) photocatalytic degradation of 4-CP by TiO 2 nanoparticles prepared with T20 at different calcination temperatures. 62

76 On the other hand, S surfactant had the highest activity at 600 C since better material crystallinity (i.e., anatase), approximately 97% at 600 C compared to only 91% at 500 C, overcame the decrease in surface area from 83.6 to 54.3 m 2 /g. In all cases, the S surfactant had higher photocatalytic activity than the S control, especially at 600 C where 20% enhanced activity was observed. These results imply that the crystal structure (i.e., crystal phase and crystallinity) of the TiO 2 particles is very important factor with respect to photocatalytic activity and that the former property sometimes could have a more dramatic effect than the latter. Comparing pure anatase TiO 2 particles prepared at 500 ºC showed that the photocatalytic activity of S surfactant with lower surface area was slightly better than that of S control possibly due to the better crystallinity of S surfactant. The surface and volume (inside) of TiO 2 photocatalysts have defect structures which are related to the type and concentration of vacancies and interstitials of the atoms in TiO 2 [49,50]. Bulk defects in S control with relatively low crystallinity, where the photogenerated electrons and holes are recombined, lower the photocatalytic activity. Electron spin resonance spectroscopic studies have revealed that the photogenerated electrons and holes are captured at titanium atoms and oxygen atoms within the bulk defects, respectively [51,52]. Along with higher crystallinity of S surfactant, many surface defects observed in S surfactant can act as active sites, where the electron acceptor or donor is adsorbed, and thus can be beneficial for enhancing the photocatalytic activity of TiO 2. Moreover, comparison of the agglomerated shapes of S control and S surfactant shown in Fig. 2.5(a) and (b) shows that the relatively high surface area of S control has originated from the N 2 -accessible porous inner space of the highly agglomerated clusters where the UV energy cannot be utilized effectively. On the other hand, in spite of its low N 2-63

77 accessible surface area, S surfactant had higher surface existing in the outer space of the clusters due to the shape of the agglomerates exclusively composed of well-defined small size nanoparticles, which is advantageous for absorbing UV radiation energy. 2.4 References [1] A. Fujishima, K. Honda, Nature 238 (1972) 37. [2] M.A. Fox, M.T. Dulay, Chem. Rev. 93 (1993) 341. [3] O. Legrini, E. Oliveros, A.M. Braun, Chem. Rev. 93 (1993) 671. [4] E. Stathatos, P. Lianos, F. Del Monte, D. Levy, D. Tsiourvas, Langmuir 13 (1997) [5] S.-Y. Kwak, S.H. Kim, S.S. Kim, Environ. Sci. Technol. 35 (2001) [6] D.D. Dionysiou, A.A. Burbano, M.T. Suidan, I. Baudin, J.M. Laîné, Environ. Sci. Technol. 36 (2002) [7] K. Yoo, H. Choi, D.D. Dionysiou, Chem. Comm. (2004) [8] K. Yoo, H. Choi, D.D. Dionysiou, Catal. Comm. 6 (2005) 259. [9] H. Al-Ekabi, Proceedings of the Workshop on Advanced Oxidation Technologies: Fundamental and Environmental Applications, Albuquerque, New Mexico, [10] Y.H. Hsien, C.F. Chang, Y.H. Chen, S. Cheng, Appl. Catal. B 31 (2001) 241. [11] K.M. Schindler, M. Kunst, J. Phys. Chem. B 94 (1990) [12] M.A. Fox, M.T. Dulay, Chem Rev. 93 (1993) 54. [13] G.P. Fotou, S. Pratsinis, Chem. Eng. Comm. 151 (1996) 251. [14] L.-Q. Wu, P. Huang, N. Xu, J. Shi, J. Membr. Sci. 173 (2000)

78 [15] B. Ohtani, Y. Ogawa, S.-I. Nishimoto, J. Phys. Chem. B 101 (1997) [16] K.Y. Jung, S.B. Park, J. Photochem. Photobiol. A 127 (1999) 117. [17] J. Yang, J.M.F. Ferreira, Mater. Lett. 36 (1998) 320. [18] B. Tryba, A.W. Morawski, M. Inagaki, Appl. Catal. B 46 (2003) 203. [19] S. Vemury, S.E. Pratsinis, J. Am. Ceram. Soc., 78 (1995) [20] P. Yang, D. Zhao, D.I. Margolese, B.F. Chmelka, G.D. Stucky, Nature 396 (1998) 152. [21] E.L. Crepaldi, G.J.d.A.A. Soler-Illia, D. Grosso, F. Cagnol, F. Ribot, C. Sanchez, J. Am. Chem. Soc. 125 (2003) [22] E.L. Crepaldi, G.J.d.A.A. Soler-Illia, D. Grosso, P.-A. Albouy, C. Sanchez, Chem. Commun. (2001) [23] F. Bosc, A. Ayral, P.-A. Albouy, C. Guizard, Chem. Mater. 15 (2003) [24] F. Bosc, A. Ayral, P.-A. Albouy, L. Datas, C. Guizard, Chem Mater. 16 (2004) [25] N. Idrissi-Kandri, A. Ayral, M. Klotz, P.-A. Albouny, A.E. Mansouri, A. Van der Lee, C. Guizard, Mater. Lett. 50 (2001) 57. [26] H. Choi, E. Stathatos, D.D. Dionysiou, Topics in Catalysis (In Press). [27] Y. Takahashi, Y. Matsuoka, J. Mater. Sci. 23 (1988) [28] E. Prouzet, F. Cot, G. Nabias, A. Larbot, P. Kooyman, T.J. Pinnavaia, Chem. Mater. 11 (1999) [29] M. Klotz, N. Idrissi-Kandri, A. Ayral, C. Guizard, Mater. Res. Soc. Symp. Proc. (2000) 628. [30] V.M. Nace (Eds.), Nonionic Surfactants: Polyoxyalkylene Block Copolymers; Marcel Dekker: New York,

79 [31] Technical brochure on Surfactants: Pluronic and Tetronic, BASF, New Jersey. [32] S. Günter (Eds.), Nanoparticles; Wiley-VCH: Weinheim, [33] J.Z. Zhang, Z.-L. Wang, J. Liu, S. Chen, G.-Y. Liu, Self-Assembled Nanostructures; Kluwer/Plenum: New York, [34] X. Ju, P. Huang, N. Xu, J. Shi, J. Membr. Sci. 202 (2002) 63. [35] D. Eder, R. Kramer, Phys. Chem. Chem. Phys. 5 (2003) [36] R.A. Spurr, W. Myers, Anal. Chem. 29 (1957) 760. [37] S. Nakade, M. Matsuda, S. Kambe, Y. Saito, T. Kitamura, T. Sakata, Y. Wada, H. Mori, S. Yanagida, J. Phys. Chem. B 106 (2002) [38] R.R. Bacsa, J. Kiwi, Appl. Catal. B 16 (1998) 19. [39] R.J. Young, Introduction to polymers; Chapman and Hall: New York, [40] E. Stathatos, P. Lianos, C. Tsakiroglou, Microporous and Mesoporous Mater. 75 (2004) 255. [41] R.D. Shannon, J.A. Pask, J. Am. Ceram. Soc. 48 (1965) 391. [42] S. Hishita, M. Takata, H. Yanagida, Yogyo-Kyokai-Shi 86 (1978) 621. [43] J.C. Yu, J. Yu, W. Ho, Z. Jiang, L. Zhang, Chem. Mater. 14 (2002) [44] T. Van Gestel, C. Vandecasteele, A. Buekenhoudt, C. Dortremont, J. Luyten, R. Leysen, B.V. der Bruggen, G. Maes, J. Membr. Sci. 207 (2002) 73. [45] M.M. Yusuf, H. Imai, H. Hirashima, J. Sol-Gel Sci. Technol. 28 (2003) 97. [46] M. Niederberger, M.H. Bartl, G.D. Stucky, Chem. Mater. 14 (2002) [47] A. Mersmann (Eds.), Crystallization Technology Handbook; Marcel Dekker, Inc.: New York,

80 [48] D.C. William, Materials Science and Engineering: An Introduction; John Wiley and Son, Inc.: New York, [49] T. Torimoto, R.J. Fox III, M.A. Fox, J. Electrochem. Soc. 143 (1996) [50] R.F. Howe, M. Gratzel, J. Phys. Chem. 89 (1985) [51] T. Nakaoka, Y. Nosaka, J. Photochem. Photobiol. A 110 (1997) 299. [52] G. Balasubramanian, D.D. Dionysiou, M.T. Suidan, V. Subramanian, I. Baudin, J.-M. Laine, J. Mater. Sci. 38 (2003)

81 CHAPTER 3 Sol-Gel Preparation of Mesoporous Photocatalytic TiO 2 Particles, Films and Membranes Using Surfactant- Assisted Sol-Gel Methods and Their Environmental Applications 68

82 This study describes the application of novel chemistry methods for the fabrication of robust nanostructured TiO 2 photocatalysts. Such materials can be applied in the development of efficient photocatalytic systems for the treatment of water and wastewater. Mesoporous photocatalytic TiO 2 particles, films, and membranes were synthesized via a simple method that involves dip-coating of appropriate substrates into an organic/inorganic sol composed of isopropanol, acetic acid, titanium tetraisopropoxide, and various surfactant molecules followed by calcination of the coating at 500 C. Controlled hydrolysis and condensation reactions were achieved through in-taking of water molecules released from the esterification reaction of acetic acid with isopropanol. The subsequent stable incorporation of the Ti-O-Ti network onto self-assembled surfactants resulted in TiO 2 with enhanced structural and catalytic properties. The effect of surfactant type and concentration on the homogeneity, morphology, light absorption, dye adsorption and degradation, and hydrophilicity of TiO 2 films as well as on the structural properties of the corresponding TiO 2 particles was investigated. The method resulted in the synthesis of mesoporous TiO 2 material with enhanced structural and catalytic properties including high surface area, large pore volume, pore size controllability, small crystallite size, enhanced crystallinity, and active anatase crystal phase. The prepared TiO 2 thin films were super-hydrophilic and possessed thermally stable spherical bicontinuous mesopore structure with highly interconnected inorganic network. The highly porous TiO 2 films prepared with polyethylene glycol sorbitan monooleate surfactant possessed high surface area of 147 m 2 /g and porosity of 46%, narrow pore size distribution ranging from 2 to 8 nm, homogeneity without cracks and pinholes, active anatase crystal phase, and small crystallite size of 9 nm. They also 69

83 exhibited four times higher photocatalytic activity for the decoloration of methylene blue dye than the nonporous control TiO 2 films prepared without the surfactant. High water permeability and sharp polyethylene glycol retention of the prepared photocatalytic TiO 2 /Al 2 O 3 composite membranes evidenced the good structural properties of the TiO 2 coatings. In addition, the repeated-coating procedure made it possible to effectively control the physical properties of TiO 2 layer onto the porous Al 2 O 3 support such as the coating thickness, catalyst amount, photocatalytic activity, water permeability, and organic retention. Finally, in order to develop efficient photocatalytic TiO 2 films and membranes for application in water and wastewater treatment and reuse systems, simultaneous photocatalytic, disinfection, separation, and anti-biofouling properties of TiO 2 thin films and TiO 2 /Al 2 O 3 composite membranes fabricated here were demonstrated. These TiO 2 materials were highly efficient in the decomposition of methylene blue and creatinine, destruction of biological toxin (microcystin-lr), and inactivation of pathogenic microorganism (Escherichia coli). Moreover, the photocatalytic TiO 2 membranes exhibited less adsorption fouling tendency when tested with dissolved organic carbon solution obtained from activated sludge. 70

84 3.1 Introduction Nanocrystalline titanium oxide (TiO 2, titania) has been extensively studied owing to its outstanding physical and chemical properties in photocatalytic applications for environmental remediation [1 4]. Titania is usually used in the form of nanoparticles in suspension for high catalytic surface area and activity [5]. However, nanosize TiO 2 particles in suspension are difficult to handle and remove after their application in water and wastewater treatment [6]. Since the pioneering work on TiO 2 photocatalytic membranes by Anderson and his coworkers [7,8], many research studies have been carried out to immobilize TiO 2 catalyst onto various substrates as thin films and membranes for use in a variety of applications in spite of their lower catalytic surface area than that of nanoparticles in suspension [9 12]. In addition to crystal structure (crystal phase and degree of crystallinity) of TiO 2, which is among the parameters determining its intrinsic catalytic properties, its structural properties are also important. This is because catalyst morphology can affect the transport of reactants and products to or from the catalytic active sites as well as the light absorbance for the photo-excitation of the catalyst and the generation of electron-hole pairs. In the case of TiO 2 membranes, the structural properties can determine, to a large extent, their permeability and selectivity. Consequently, it is critical to control and improve the photocatalytic and structural properties of TiO 2 material for application in catalysis and separations. In common sol-gel methods for preparing TiO 2 material, highly reactive alkoxide titanium precursors are violently hydrolyzed and further condense to form the Ti-O-Ti network. Unfortunately, this route can lead to the precipitation of amorphous particles 71

85 with uncontrolled structure. To overcome many of the specific problems of sol-gel methods employing water as the hydrolysis agent and to control hydrolysis and polycondensation reactions, nonhydrolytic methods [13], addition of ionic liquids and organic additives [2,14], and coordination chemistry [15] have been attempted. Recently, some research groups have utilized the water molecules coming from moisturized air [16] and sol ingredients [17], and even forming from certain chemical reactions during synthesis procedures, instead of externally added water [18,19]. Prior work on the use of acetic acid (AcOH) as a titania sol modifier in alcohol solvent showed a successful synthesis route of the Ti-O-Ti inorganic network with controlled properties [20 22]. In this approach, water molecules are formed from the esterification reaction of AcOH with isopropanol (iproh) as follows: iproh + AcOH iproac + HOH (3.1) The source of iproh can be either its direct addition as a solvent or as an intermediate produced from the following reaction: Ti-OiPr + AcOH Ti-OAc + iproh (3.2) where Ti-OiPr is titanium tetraisopropoxide (TTIP) in this study. It is generally accepted that alkoxy groups bonded to titanium can be replaced by acetate groups, forming Ti- OAc and iproh. As a result, Ti-O-Ti inorganic network can be formed from the 72

86 hydrolysis and condensation reactions. Another formation route of Ti-O-Ti network is the following: Ti-OAc + iproh iproac + Ti-OH (3.3) Ti-OiPr + Ti-OAC iproac + Ti-O-Ti (3.4) In this case, titanium-bonded acetate ligands can be hydrolyzed and participate in a direct condensation reaction, resulting in Ti-O-Ti condensed bridge. Whatever is the mechanism with major contribution and no matter which reaction pathway is the fastest, the final product is titanium atoms linked together with oxygen atoms. Most attention in previous studies was focused on elucidating the reaction mechanisms and applying this simple methodology. However, little work has been devoted to the fabrication of highly porous TiO 2 materials through a new synthesis method employing surfactant molecules as a pore-directing agent along with the acetic acid-based sol-gel route. Various template-based sol-gel methods using a variety of surfactants and block copolymers have been applied in the past to synthesize tailordesigned TiO 2 catalytic materials [23 25]. Nanocrystalline TiO 2 films were also prepared in the past using the acetic acid-based sol modified with Triton X-100 surfactant (polyethylene glycol tert-octylphenyl ether) [22]. However, no attempt to fabricate TiO 2 composite membranes using this new methodology has been reported so far. Especially, ionic surfactants such as alkyl phosphate, dodecylamine, and cetyltrimethylammonium chloride were initially used in the synthesis due to the strong and well-organized incorporation of titania inorganic framework onto surfactant micelles by electrostatic 73

87 interactions [24,26,27]. However, the use of ionic surfactants showed limited potential for such applications since the strong electrostatic binding force makes it difficult to remove the templates completely through extraction methods and even heat treatment at high temperature. In order to overcome these concerns and challenges, this research deals with the preparation of highly efficient nanostructured photocatalytic TiO 2 particles, thin films and membranes via an acetic acid-assisted sol-gel method employing nonionic surfactants [28 30]. A systematic study is conducted herein to elucidate the effect of surfactant type and concentration on the structural and catalytic properties of TiO 2 material as well as on the physical properties of TiO 2 thin films including homogeneity, reproducibility, thickness, and hydrophilicity. Furthermore, the TiO 2 thin films and membranes were characterized for their material crystal structure, morphology, nanostructure, and elemental composition, and evaluated for their dye adsorption capacity, photocatalytic activity, water permeability, and polyethylene glycol retention. Finally, the TiO 2 films and membranes were tested in the destruction of organic pollutants in water such as methylene blue dye, creatinine, and biological toxin (microcystin-lr, MC-LR), as well as to inactivate pathogenic microorganisms such as Escherichia coli (E. coli). The TiO 2 membranes were further tested with a dissolved organic carbon solution obtained from an activated sludge for their water permeability, organic retention, and anti-biofouling properties. 74

88 3.2 Experimental Sol Synthesis The organic molecules used as pore directing agents were representative nonionic long chain surfactants including Tween 20 (T20, polyethylene glycol sorbitan monolaurate), Tween 80 (T80, polyethylene glycol sorbitan monooleate), and Triton X-100 (X100, polyethylene glycol tert-octylphenyl ether) purchased from Aldrich. Compared to other commonly used toxic and ionic templating agents, these organics are relatively inexpensive, biodegradable, non-toxic, and easily removable. Such large amphiphilic molecules exhibit the existence of ordered mesophase and the ability to synthesize tailordesigned porous TiO 2 catalytic materials [31,32]. Each surfactant was dissolved in isopropanol (i-proh, Fisher). Before adding alkoxide precursor, acetic acid (Fisher) was added into the solution for the esterification reaction with alcohol. Then, titanium tetraisopropoxide (TTIP, Aldrich) was added under vigorous stirring. The molar ratio of surfactant:i-proh:acetic acid:ttip was R:45:6:1, where the surfactant concentration R was varied in the range from 0.0 to 3.0. Regardless of surfactant addition, the sol was transparent, homogeneous, and stable. For the fabrication of TiO 2 /Al 2 O 3 composite membranes, the molar ratio of the ingredients was optimized at R = 1 with respect to film homogeneity and porous structure Formation of TiO 2 Particles and Films and TiO 2 /Al 2 O 3 Membranes For the synthesis of immobilized TiO 2 thin films, borosilicate glass (Micro slide, Gold Seal) substrate with an effective surface area of 10 cm 2 was cleaned with water followed 75

89 by acetone. The substrate was dip-coated with the sol using a dip-coating apparatus at a withdrawal rate of 12.8 cm/min. After coating, the films were dried at room temperature for 1 h, calcined at a ramp rate of 3 ºC/min in a programmable furnace (Paragon HT-22-D, Thermcraft), stayed at 500 ºC for 15 min, and cooled down naturally. This dip-coating procedure was repeated until the desired thickness and homogeneity of the film were obtained. The number of dip-coatings (denoted as T) was usually three unless otherwise specified. For convenience, the abbreviations, film surfactant and film control denote TiO 2 film prepared with surfactant and control TiO 2 film prepared without surfactant, respectively. Because of the difficulty in direct characterization of the porosity and crystal structure of small quantity of TiO 2 films immobilized on glass support, material characterization was carried out on the corresponding TiO 2 particles obtained from thick films. Even though the properties of TiO 2 particles are not exactly the same as those of TiO 2 films, this technique is useful for quickly examining and comparing the effect of sol conditions on the structural properties of the final material [33,34]. For TiO 2 powder synthesis, the sol was spread on the glass substrate, dried, and heat-treated at 500 ºC for 1 h (instead of 15 min as in making thin films) to thoroughly remove all the organics, resulting in the formation of a thick film. The TiO 2 particles were collected by scraping the thick films and ground for further characterization. In order to fabricate photocatalytic TiO 2 composite membranes, the sol was coated on a home-made porous alumina substrate. The substrate membrane was prepared from A16-SG alumina powder (Alcoa Chemicals) with a particle size of 0.48 μm. The powder was mixed with a suitable amount of water, put into a mold, and pre-pressed at 76

90 5000 lb for 30 sec and pressed at 7500 lb for 3 min using a hydraulic unit (Model 3925, Carver). The pressed disks were calcined and sintered at up to 1220ºC for 7 days in a furnace (Vulcan HTA, Neytech) with programmed multi-stage temperatures and holding times. The pore size and pure water permeability coefficient of the Al 2 O 3 support membrane were 0.1 μm and 11.0 l/m 2 /hr/bar, respectively. The one side of the alumina substrate with diameter of 21 mm (effective D=18 mm) and thickness of 2.2 mm was dipped into the sol for 20 seconds and taken out with extreme caution. After coating procedure, the other drying and calcination procedures were the same as for the TiO 2 thin films on glass substrates Materials Characterization X-ray diffraction (XRD) analysis using a Kristalloflex D500 diffractometer (Siemens) with Cu Kα (λ = Å) radiation was employed to determine the crystal structure and crystallinity of the TiO 2 catalyst. A porosimetry analyzer (Tristar 3000, Micromeritics) was used to investigate structural characteristics of TiO 2 material including Brunauer, Emmett, and Teller (BET) surface area, porosity, and pore size and distribution after purging samples with nitrogen gas for 2 h at 150 C using Flow prep 060 (Micromeritics). UV-visible light absorbance of TiO 2 films was measured using a UV-Vis spectrophotometer (Hewlett Packard 8452A) to determine their band gap energy and evaluate their light utilization characteristics. For the film morphology, crystallite size, and nanocrystal planes orientation, a JEM-2010F (JEOL) high resolution-transmission electron microscope (HR-TEM) with field emission gun at 200 kv was used. The samples were dispersed in methanol (HPLC grade, Pharmco) using an ultrasonicator 77

91 (2510R-DH, Bransonic) for 5 min and fixed on a carbon-coated copper grid (LC200-Cu, EMS). An environmental scanning electron microscope (ESEM, Philips XL 30 ESEM- FEG) was used at accelerating voltage of 10K to measure the thickness of coatings and examine the homogeneity of TiO 2 films and membranes. The samples were cautiously fractured into small pieces and fixed into sample holder without polishing or pretreatment for quick and convenient analysis. An elemental composition analysis of the TiO 2 was performed using energy dispersive X-ray spectroscope (EDX, Oxford Isis) connected to the HR-TEM and ESEM. For the accurate measurement of trace elements in the material surface and oxidation states of Ti, an X-ray photoelectron spectroscope (XPS, Perkin- Elmer model 5300) with Mg Kα X-rays was used at a take-off angle of 45º. For measuring the contact angle of the film surface, the sessile drop method was employed using a static angle goniometer (Ramé-Hart). In order to measure weight change of the material due to desorption of water and solvent and decomposition of organics in TiO 2 films, thermogravimetric analysis (TGA, TA instruments 2050) was performed in the presence of air at a ramp rate of 3 ºC/min Dye Adsorption and Photocatalytic Activity of TiO 2 Films In order to determine their dye adsorption characteristics, the prepared films were soaked into 3.0 mm methylene blue (MB, Riedel-deHaën) for several seconds and dried thoroughly. The visible light absorbance of the TiO 2 films with adsorbed MB was measured using a UV-Vis spectrophotometer (Hewlett Packard 8452A). During the procedure, the initial visible light absorbance of MB solution at 664 nm was shifted to 564 nm for the MB-absorbed TiO 2 films. Prior experiments to examine the photocatalytic 78

92 activity of TiO 2 films were conducted by immersing a glass substrate with 10 cm 2 area covered by TiO 2 into borosilicate glass dish containing 8 ml of 30 μm MB solution. The initial solution ph was controlled at 3.0 without buffer by adding H 2 SO 4 solution. Two 15 W low pressure mercury UV tubes (Spectronics) emitting near UV radiation with a peak at 365 nm were used at a light intensity of 3.48 mw/cm 2 measured near the film surface using a UV radiation meter (IL1700, International Light). The concentration of MB solution was determined by measuring the visible light absorbance at 664 nm. In order to apply the TiO 2 films in destructing recalcitrant organic contaminants in water, creatinine (2-imino-1-methylimidazolidin-4-one, Aldrich), which is usually found in urine and perspiration, was selected as a target compound. The photocatalytic degradation of 8 ml of 0.2 mm creatinine solution (total organic carbon (TOC) of 9.6 mg/l) was monitored at different ph. The initial solution ph was adjusted at 7.0 by adding NaOH solution in 10 mm phosphate buffer and 3.0 by adding H 2 SO 4 without buffer. The concentration of creatinine was determined using a high performance liquid chromatograph (HPLC; Agilent, Series 1100) equipped with a C-16 Discovery column (Supelco). TOC was measured with a TOC analyzer (TOC-VCSH with ASI-V, Shimadzu). The water used in all experiments was double-distilled (Corning Mega-Pure system) Water Permeability and Organic Retention of TiO 2 Composite Membranes The water permeability and organic retention of the prepared TiO 2 /Al 2 O 3 composite membranes were tested with a home-fabricated membrane chamber of dead-end filtration type fitting with the size and shape of the substrate membrane. The permeate water flux 79

93 was determined by direct measurement of the permeate flow at 20ºC. The water in the chamber was pressurized with a nitrogen gas at a transmembrane pressure (TMP) from 0 to 6 kg/cm 2. In order to evaluate molecular weight cut-off (MWCO) and integrity of the composite membranes, experiments on organic retention of the prepared membranes were conducted using 200 mg/l of polyethylene glycol (PEG, Fluka) solutions with different molecular weights and TMP of 1 kg/cm 2. Organic carbon concentrations in the raw PEG solution and permeate water through the membranes were measured using the TOC analyzer Environmental Applications TiO 2 films or TiO 2 /Al 2 O 3 composite membranes with surface area of approximately 10 cm 2 were placed into borosilicate glass dish with 8 ml of reaction solution containing: (1) 1 mg/l of MC-LR (Calbiochem Cat #: ) released from cyanobacteria as a biological toxin, which is known to cause skin irritations and liver damage or affect the nervous systems [35], (2) cfu/ml of E. coli (ATCC 11229) as a model pathogenic microorganism, (3) 30 μm methylene blue (MB, Riedel-deHaën), which is a commonly used dye in photocatalytic tests, or (4) 0.2 mm of creatinine (2-imino-1- methylimidazolidin-4-one, Aldrich), which is usually found in urine and perspiration. Destruction of MC-LR and inactivation of E. coli were tested for the TiO 2 films while decomposition of MB and creatinine was examined for the TiO 2 /Al 2 O 3 membranes. The following analytical instruments were used to determine the concentration of the contaminants over time: the HPLC for MC-LR and creatinine, the TOC analyzer for dissolved organic carbon (DOC), colony forming test for E. coli, and the UV-Vis 80

94 spectrophotometer for MB. The initial solution ph was controlled at 3.0 without buffer by adding H 2 SO 4 solution except for the E. coli inactivation test. Two 15 W low-pressure mercury UV tubes (Spectronics) emitting near UV radiation at 365 nm were used at a light intensity of 3.48 mw/cm 2. For the treatment of real wastewater using the TiO 2 /Al 2 O 3 membranes, dissolved organic carbon solution was obtained by filtering an activated sludge treating municipal wastewater (Mill Creek Wastewater Treatment Plant in Cincinnati, Ohio, USA) using 1.2 μm glass fiber filter. The solution contained dissolved solids (DS) of around 800 mg/l (170 mg/l organics and 630 mg/l inorganics) and DOC of around 120 mg/l. The organic solution was further diluted to investigate the effect of feed strength on membrane performance including adsorption fouling under UV radiation (anti-biofouling properties) and variation of permeate water flux and organic retention over time. 3.3 Results and Discussion Structural Characteristics of TiO 2 Particles In order to investigate the effect of surfactant type and concentration on the structural properties of TiO 2, porosimetry and crystallographic analyses of TiO 2 particles were employed, and the results are summarized in Table 3.1. The structural properties were significantly improved with the addition of surfactant. In the absence of surfactant, the materials were almost nonporous with surface area of m 2 /g, pore volume of cm 2 /g, and porosity of 11.6%. Increasing R from 0.0 to 1.0 resulted in times increase in surface area and times increase in pore volume without significant 81

95 change in pore size. The structural characteristics were consistent with the properties of the surfactants. It is well known that increasing the surfactant chain length has a similar effect as increasing surfactant concentration [36]. The higher porosity and larger pore of TiO 2 prepared with Tween surfactants compared to those prepared with X100 can be ascribed to the fact that Tween surfactants have a chain of 20 ether groups while X100 bears a chain of only 10 ether groups. TiO 2 prepared with T80 had the highest porosity and largest pore size due to longer hydrophobic tail length of T80 (C17) than T20 (C11). In addition, T80 had better controllability of pore size in the mesoporous range. Table 3.1: Structural characteristics of TiO 2 particles. Surfactant R S BET V pore Porosity D BJH (nm) D BJH range (nm) CS, crystallite size a (nm) (m 2 /g) (cm 3 /g) (%) Ads. Des. Ads. Des. CS BET CS XRD CS TEM None T T X a CS BET : calculated from BET surface area, CS XRD : estimated from XRD peak, and CS TEM : measured from TEM image. 82

96 Intensity (arb. units) (101) (004) (200) (105) (211) R= θ (degrees) Fig. 3.1: XRD patterns of TiO 2 particles prepared with T80. Inserted numbers are Miller indices. The crystallite size (CS BET ) was calculated from the BET surface area, assuming all crystallites are spherical and separate without aggregation [37]. The result was compared with the crystallite size (CS XRD ) estimated using Scherrer s equation from the XRD peak broadening analysis [38]. The CS BET was always larger than the CS XRD, implying that most primary particles were more or less aggregated. The actual crystallite size (CS TEM ) measured from TEM image slightly decreased with increasing R, but it was within a similar range from 7 to 12 nm. In the case of R = 0.0, the large discrepancy between CS TEM and CS BET suggests the prepared particles were highly aggregated. As a result, the addition of surfactants as pore directing agents also inhibited the growth of crystallites as well as the aggregation of adjacent primary particles, and thus made the materials highly porous. The representative XRD patterns of the TiO 2 materials prepared 83

97 with T80 are shown in Fig All the peaks are designated to anatase phase with size less than 16 nm. Interestingly, more distinct peaks with higher intensities were observed with increasing R, suggesting the material crystallinity was enhanced by the addition of surfactants [39]. The same trend was observed in the case of the other two surfactants Thickness, Mass, UV Light Absorption, and Dye Adsorption of TiO 2 Films According to ESEM observation, the prepared transparent TiO 2 thin films were very reproducible and homogeneous without cracks and pin-holes. As shown in Fig. 3.2, the thickness of films increased as surfactant concentration R increases due to an increase in the viscosity of the sol by the surfactants. The mass of TiO 2 immobilized on the glass substrates was calculated from the thickness and porosity of films. In spite of the increased film thickness, the mass of TiO 2 decreased over R because of the high porosity of the materials, which implies that the TiO 2 inorganic structure in the films became less dense upon addition of surfactants Thickness (μm) T T20 X X100 T20 T Molar ratio of surfactant to TTIP, R Fig. 3.2: Thickness and mass of TiO 2 films Mass of TiO 2 in film (μg/cm 2 ) 84

98 Absorbance (a.u.) λ = nm Wavelength (nm) Fig. 3.3: UV-visible absorption spectrum of TiO 2 film. The UV-Vis absorption band edge is a strong function of the crystallite size of nanosize TiO 2 catalyst. Usually the band gap between the valance band and the conduction band of semiconductor increases with decreasing particle size [40]. A representative UV-visible absorption spectrum of the TiO 2 films (e.g., with T = 3 at R = 1) is shown in Fig The onset of absorption of wavelength (λ) and the corresponding band gap energy (E g ) of TiO 2 material are well known to be λ = 385 nm and E g = 3.23 ev for anatase phase, respectively [41]. Extrapolating the spectral curve, the band gap energy of the TiO 2 catalyst was measured to be E g = 3.34 ev corresponding to λ = nm which is in agreement with other research results [12,42]. Regardless of surfactant type and concentration, the UV absorption edges of the TiO 2 films were within the range of nm, corresponding to band gap energy of around ev. The ΔE g = 0.11 ev 85

99 of this blue shift compared to bulk anatase TiO 2 particles indicated a size of TiO 2 crystallites smaller than 10 nm due to so called quantum size effect [40]. A particle size of 6 nm was estimated by the effective mass approximation based on the observed band gap shift [43]. However, there was a slight discrepancy between the particles sizes determined by XRD analysis (or TEM analysis as discussed later) and UV-visible spectroscopy. This could be due to the fact that the effective mass approximation is relatively less correct for small nanoparticles and statistical effects of spatial confinement also influence the optical properties of nanocrystalline semiconductors [44]. As shown in Fig. 3.4, the UV light absorption spectra of TiO 2 films prepared under different conditions were studied at 365 nm as an indirect evaluation of the photocatalytic activity under near UV radiation ( nm). Increasing surfactant concentration up to R = 1.0 resulted in a significant increase in the UV light absorption of TiO 2 films, and then increase in the UV light absorption became less significant over R. This might be because the specific surface area of the materials increased while the mass of TiO 2 immobilized on the substrate decreased upon the addition of surfactants. Moreover, the UV light absorption properties of TiO 2 film is a complex function of physicochemical properties of the TiO 2 material itself such as crystallinity, crystal phase, crystal size, and purity as well as the structural properties of the film such as thickness, mass, and surface area. In spite of their smallest mass as shown in Fig. 3.2, films T80 had the highest UV light absorption properties among films prepared with other surfactants. 86

100 Absorbance at 365 nm (arb. units) T T20 X Molar ratio of surfactant to TTIP, R Fig. 3.4: UV light absorption of TiO 2 films at 365 nm. The visible light absorption of TiO 2 films with adsorbed MB at 564 nm as an indication of the porosity and homogeneity of TiO 2 films is shown in Fig MB adsorption of the films increased with increasing surfactant concentration up to R = In spite of the increased surface area of TiO 2 material and thickness of the film with increasing surfactant concentration, further addition of surfactant slightly decreased the adsorption capacity due to a lower TiO 2 mass immobilized on the substrates. Another reason is perhaps associated with the fact that the homogeneity of films prepared at too high concentration of surfactants decreased significantly, most probably, due to the high viscosity of the sol, pore coalescence, and multi-micellar interactions. 87

101 Absorbance at 564 nm (arb. units) T T X Molar ratio of surfactant to TTIP, R Fig. 3.5: Visible light absorption of MB-adsorbed TiO 2 films at 564 nm Hydrophilicity of TiO 2 Films Hydrophilicity (or wettability) of TiO 2 films was investigated by measuring the water contact angle. Surface hydrophilicity is important for the accessibility of organic compounds to catalytic sites [45]. It has been well observed that the surface of TiO 2 exhibits super-hydrophilicity under UV irradiation due to hydrophilic groups introduced [46,47]. Prior to coating with TiO 2 films, the pre-cleaned glass substrate showed water contact angle of around 16.1º, corresponding to relatively hydrophilic surface. As shown in Fig. 3.6(a), the contact angles after TiO 2 coating significantly decreased to the reliable detection limit of 4º (actually the contact angle reached 0º) without any UV radiation, indicative of super-hydrophilic surface. The decreased contact angle may be associated with the porous structure of TiO 2 films and the increase in surface hydrophilic groups, 88

102 which was supported by TGA analysis of TiO 2 films. For the TGA analysis, approximately 10 mg of particles were collected from the films T80 and the weight change was monitored during thermal treatment at a ramp rate of 3 ºC/min. The weight loss of the samples within the temperature range ºC increased significantly from 1.98% for film prepared at R = 0.0 to 2.31% for that at R = 0.5. Further increase in concentration up to R = 3.0 caused slight increase in weight loss of 2.64%. This result indicated that the TiO 2 films prepared at high surfactant concentration entrapped a large amount of water in the pores. In addition, according to the Laplace equation, the wettability of porous materials is known to be proportional to the surface tension of liquid and inversely proportional to the pore size and surface energy (contact angle) of the material [48]. The increase in surfactant concentration resulted in decreased contact angles as observed in Fig. 3.6(a) but increased pore size as summarized in Table 3.1. Considering the same surface tension of water used, Fig. 3.6(b) suggests that the optimum surfactant concentration for preparing TiO 2 films with better hydrophilicity was at around R = Films T20 and films X100 had a higher wettability than films T80 because more organics were entrapped in films prepared with T80, which has longer hydrocarbon chain length. According to TGA analysis of films prepared with surfactants at R=1.0, negligible weigh loss was observed after 100 ºC in the case of film T20 and film X100. On the other hand, film T80 still had a considerable weight loss of 1.7% between 100 ºC and 500 ºC due to the decomposition of organics remaining in the materials. Moreover, EDX results showed that the carbon content of film T80 was around 1.2%, significantly higher than in others films, which was in the range %. 89

103 Contact anlge (degrees) (a) T80 X100 T Molar ratio of surfactant to TTIP, R Wettability, Cosθ / r (b) T20 T80 X Molar ratio of surfactant to TTIP, R Fig. 3.6: (a) water contact angle and (b) wettability of TiO 2 films. The contact angle and pore size of TiO 2 material are denoted as θ and r, respectively. 90

104 3.3.4 Porosity and Morphology of TiO 2 Films Based on the obtained results so far, film T80 at R = 1.0 among all conditions investigated was considered the most effective. For determining the structural properties of film T80, the TiO 2 material was collected by scrapping the thin film surface cautiously and analyzed using a porosimetry analyzer and TEM. Fig. 3.7(a) shows the nitrogen adsorption-desorption isotherms of film T80. Compared to N 2 isotherms of the film control, which represents a nonporous material, these type IV isotherms of film T80 were typical of those of a well-developed mesoporous material. A hysteresis loop in the isotherms was observed with dissimilar shapes for the adsorption and desorption branches, implying a different size of pore throat diameter. The sharp drop on the desorption branch can be assigned to the presence of mesopore constrictions at the boundaries between the ordered domains and of smaller pores in the titania walls [49]. The pore size distribution shown in Fig 3.7(b) was relatively narrow ranging from 2 to 8 nm. The BJH pore diameters measured from the adsorption and desorption branches were 4.04 nm and 3.72 nm, respectively. These results imply good homogeneity of the pores. The main structural characteristics deduced from the isotherms are reported in Table 3.2. In spite of the high heat treatment temperature of 500 ºC, the BET surface of 147 m 2 /g and porosity of 46.2% were significantly high, compared to other research results [22,23]. The film thickness of 0.31 μm was measured using ESEM, and its mass of 62.2 μg/cm 2 was calculated from the pore volume and density of the anatase phase. Even though the thickness of film T80 was much larger than that of film control, the amount of TiO 2 catalyst in film T80 was smaller due to its high porosity. 91

105 Volume Adsorbed (cm 3 /g STP) 160 (a) Desorption 100 Adsorption 80 R= R= Relative Pressure (P s /P o ) Pore volume (cm 3 /g) (b) R=1.0 R= Pore diameter (nm) Fig. 3.7: (a) nitrogen adsorption-desorption isotherms and (b) pore size distribution of TiO 2 film control at R = 0.0 and film T80 at R = 1.0. Inserted image is side view of the film T80 at R =

106 Table 3.2: Structural characteristics of TiO 2 film control at R = 0.0 and film T80 at R = 1.0. Parameter R=0.0 R=1.0 S BET (m 2 /g) V pore (cm 3 /g) Porosity (%) D BJH from adsorption branch (nm) D BJH from desorption branch (nm) CS TEM (nm) Film thickness (μm) TiO 2 mass (μg/cm 2 ) As shown in Fig. 3.7(b), the cross-section of film T80 on glass substrate revealed that film T80 was homogeneous and well-incorporated to the glass substrates without cracks and pin-holes. EDX elemental analysis of collected films showed that the films were composed of Ti and O elements without any impurities. Fig. 3.8 shows the morphology of the nanostructured anatase TiO 2 thin films. For film control, no distinct mesopore structure was observed and even the lattice fringes were not clear. On the other hand, film surfactant were highly porous and exhibited distinct pore structure. The films had slightly collapsed spherical bicontinuous structure with highly interconnected network [50]. The image at high magnification showed many randomly oriented nanocrystallites with size of nm and sets of clearly resolved lattice fringes giving evidence that the TiO 2 material was highly crystalline, which was in good agreement with the XRD results. The porous structure was very strong and stable since the structure still remained at large extent until heat treatment at 700 ºC. 93

107 Fig. 3.8: Morphology and pore structure of (a-b) film control at R = 0.0 and (c-d) film T80 at R = 1.0 at different magnifications. 94

108 3.3.5 Photocatalytic Activity of TiO 2 Films Photocatalytic activity of the TiO 2 films was measured in terms of MB decoloration. As shown in Fig. 3.9, there was no direct photolysis of MB in the absence of TiO 2 photocatalysts. In spite of its relatively high catalyst mass of 88.4 μg/cm 2, the film control at R = 0.0 did not show significant decoloration of MB due to the almost nonporous properties of the film, suggesting the photocatalytic reaction occurred only at the very surface of TiO 2 film. Increasing the amount of surfactant template up to R = resulted in a significant improvement of the photocatalytic activity of the films mainly due to the porous structure of the film with high surface area of 147 m 2 /g and porosity of 46%, and partially the enhanced material crystallinity as observed in the XRD and TEM analyses. However, further addition of surfactant beyond R = 2.0 caused adverse effect on the photocatalytic activity of films, which was consistent with the results on MB adsorption. It was because the homogeneity of the films prepared under too high concentration of surfactants (R > 2.0) decreased and the amount of TiO 2 photocatalyst immobilized TiO 2 films also decreased. These results show the importance of preparing highly porous and homogeneous TiO 2 thin films, which might facilitate MB adsorption and UV light utilization. Moreover, considering the small amount of TiO 2 catalyst immobilized on the glass substrate, the TiO 2 films were highly efficient to decolorize the dye. 95

109 Normalized MB absorbance, I/I o R=3.0 R=1.0 R= Reaction time (h) Without TiO 2 R=0.0 R=0.5 Fig. 3.9: Photocatalytic decoloration of MB by TiO 2 films T Effect of Number of Coatings on TiO 2 Films Properties Considering overall results so far especially on film homogeneity and porous structure, the molar ratio of the ingredients was optimized at Tween 80:iPrOH:acetic acid:ttip = 1:45:6:1 for the fabrication of TiO 2 thin film and TiO 2 /Al 2 O 3 composite membrane. The number of dip-coating layers is one of the most critical parameters in the synthesis of thin films and membranes, from both scientific and engineering prospectives. Increasing film thickness can bring better results with respect to adsorption capacity and photocatalytic activity of the films at some extent but it also increases the preparation time and fabrication cost. 96

110 (101) Intensity (a.u) (004) (200) (105)(211) θ (degrees) Powder Number of coatings, T Fig. 3.10: XRD patterns of TiO 2 films. Inserted numbers are Miller indices at the corresponding crystallographic planes. Fig shows XRD spectra of TiO 2 films. Below a certain number of coatings, the XRD peaks were too weak to be identified due to the small amount of TiO 2 material immobilized onto the substrate. After at least seven coatings, the anatase crystal peak at (101) could be clearly seen. For more accurate crystallographic analysis, a large amount of TiO 2 particles was collected by scrapping the thin films and analyzed as powder in the XRD instrument. The last line (upper) in Fig shows the XRD spectrum of the powder. All the peaks are assigned to anatase crystal phase. The relatively wide width of the peaks indicates small crystallite size of 9 nm, which is known to be optimum for high catalytic activity. This is because a very small crystallite size causes a blue shift in the 97

111 light absorption spectrum and favors surface recombination of the photo-exited holes and electrons while a larger crystallite size exhibits lower surface area and thus a smaller number of catalytic active sites per unit mass of catalyst [51,52]. It should also be noted that films made of nanosize TiO 2 particles coated on substrates may exhibit resistance to abrasion and good mechanical stability [53]. As shown in Fig. 3.11(a), increasing the number of dip-coating layers caused an increase in the absorbance at 564 nm, indicating larger amount of MB adsorbed on the films. According to ESEM observation of the film cross-section, the film thickness was proportional to the number of coatings. This is consistent with the results on MB adsorption since the number of coatings correlates with the amount of TiO 2 catalyst immobilized on the substrate and thus the amount of MB adsorbed onto the TiO 2 catalyst. Moreover, these results showed that MB molecules could move freely into the TiO 2 multi-coating layers, implying that the TiO 2 films have highly porous and interconnected structure, which is an important property considering adsorption and photocatalytic degradation of organic contaminants at the TiO 2 catalytic sites. However, the experimental results on the photocatalytic decoloration of MB by the TiO 2 films under UV radiation at 365 nm were slightly different than those on MB adsorption. As shown in Fig. 3.11(b), the photocatalytic activity was significantly improved with the number of coatings only up to three coatings. After that, the activity was not much affected in spite of increased mass of TiO 2 catalyst immobilized on the support. This result suggests that the UV radiation incident to the TiO 2 films was utilized to photo-generate electrons and holes mainly from the outer three coatings of the films corresponding to surface depth of approximately 0.3 μm. 98

112 Absorbance at 564 nm (a.u.) (a) Absorbance (a.u.) T= Wavelength (nm) Number of coatings, T Normalized concentration, C/C o (b) T=3 T=5 T=7 T= Reaction time (hr) T=2 T=0 T=1 Fig. 3.11: (a) Visible light absorbance of TiO 2 films with adsorbed MB and (b) decoloration of MB solution by TiO 2 films. 99

113 This observation was further evidenced by comparing UV absorbance at 365 nm and thickness of TiO 2 films, as shown in Fig The film thickness was linearly increased over the number of coating layers at a rate of approximately 0.1 μm per coating and the corresponding mass of TiO 2 catalysts in the films is also linearly increased. Initially, there was a steep increase in the rate of UV absorbance of the films with a few initial coating layers. This rate was significantly reduced after 3-5 coating layers, which was in good agreement with the observed results shown in Fig. 3.11(b) Absorbance at 365 nm (a.u.) Film thickness (μm) Number of coatings, T 0.0 Fig. 3.12: UV absorbance at 365 nm and thickness of TiO 2 films Photocatalytic Degradation of Creatinine by TiO 2 Films Experiments were conducted to evaluate the photocatalytic activity of TiO 2 films with three coating layers (i.e., T=3) for the degradation of creatinine in water and the results 100

114 are presented in Fig Creatinine, a compound with chemical structure containing an imidazole ring with amide bonds and double bonds of C=N (615 KJ/mol) and C=O (745 KJ/mol), was efficiently degraded by the TiO 2 films. Approximately 60 % of the initial amount of creatinine degraded after 14 hours of photocatalytic reaction using a very small amount of catalyst (62.2 μg/cm 2 ). However, most organic carbon remained due to the formation of organic intermediates and their slow mineralization. In addition, decreasing solution ph to the acidic range slightly improved the photocatalytic degradation of creatinine. It should be emphasized that the TiO 2 films prepared in this study showed great potential for application in the preparation of photocatalytic TiO 2 membranes. This is also in consideration that their high photocatalytic activity per unit mass of catalyst can compensate the disadvantage of the small amount of TiO 2 photocatalyst immobilized onto ultra thin membranes, since the latter (i.e., small film thickness) is also a requirement for high water permeability. Normalized concentration, C/C o Without TiO 2 TOC ph = 7 ph = 3 ph = 7 Creatinine ph = 3 Creatinine Time (hr) Fig. 3.13: Creatinine degradation by TiO 2 films. 101

115 3.3.8 Morphology and Elemental Composition of TiO 2 /Al 2 O 3 Membranes The integrity of the TiO 2 active layer and its incorporation with the porous Al 2 O 3 substrate are crucial factors in the fabrication of defect-free TiO 2 /Al 2 O 3 composite membranes since even a few cracks and pin-holes and breakage of weak joints can cause failure of the separation function. Examining the ESEM images (plain view) of TiO 2 /Al 2 O 3 composite membranes shown in Fig. 3.14, we observe that at least three coating layers are necessary using this method to fabricate TiO 2 skin layer with good integrity and without significant cracks and pin-holes. The thickness of the TiO 2 skin layer was uniform at approximately 0.9 μm and the TiO 2 /Al 2 O 3 composite membranes were relatively well-incorporated without the use of an intermediate layer. In common sol-gel methods for the synthesis of micro and mesoporous membranes, mechanical polishing and an intermediate layer are required to reduce the coarse pore structure of the support material, large surface irregularities or defects which cause cracking of the overlying sol-gel derived membranes because of uneven stress development [31,51]. However, the polishing is tedious, repetitious, and impractical for some types of support configurations, and the formation of intermediate layer itself also needs multi-coatings of an additional sol containing relatively large size TiO 2 particles. This results in increased preparation cost and time and reduced water permeability. It is very interesting in this study that the first several coatings of sol with submicron thickness acted as the intermediate layer for modifying the coarse support surface to facilitate the deposition of overlying TiO 2 skin layer and the thickness was still in the submicron range. As a result, the composite membranes are expected to have both high water permeation flux and selectivity. 102

116 T=1, plain view T=2, plain view 10 micron T=3, plain view 10 micron T=3, side side view view TiO2 skin layer Alumina substrate 10 micron 2 micron 1 micron 10 micron Fig. 3.14: ESEM images of TiO 2 /Al 2 O 3 composite membranes. In order to investigate elemental composition at the outer surface of TiO 2 /Al 2 O 3 composite membranes within several nanometers, which can significantly affect the photocatalytic activity, XPS analysis was employed. Fig shows prominent photoelectron peaks for Ti 2p at ev and O 1s at ev, and negligible peaks for C 1s at ev, F 1s at ev, and N 1s at ev that were considered either contaminants from air during XPS analysis or impurities of the sol ingredients. The photoelectron peak for Ti 2p originates from Ti-O bond mainly in the form of TiO 2 (IV) and Ti 2 O 3 (III). In the high resolution XPS analysis for determining the Ti oxidation states at 2p 3/2 level corresponding to binding energy of ev, the peak of Ti(IV) 103

117 was predominantly curve-fitted compared to that of Ti(III). Moreover, no significant diffusion of aluminum from the Al 2 O 3 substrate was observed. Min: 0, Max: O 1s 66.6% N(E) F 1s 0.9% Ti 2p 30.5% 27.1 Ti(IV) 3.4 Ti(III) N 1s 0.7% C 1s 1.3% Binding energy (ev) Fig. 3.15: XPS spectrum of TiO 2 /Al 2 O 3 composite membrane Properties of TiO 2 /Al 2 O 3 Composite Membranes Water permeability and PEG retention of TiO 2 /Al 2 O 3 composite membrane were measured to evaluate the integrity and properties of the membranes, as shown in Fig The properties of TiO 2 /Al 2 O 3 composite membranes are summarized in Table 3.3. Considering membrane function, increasing the skin layer thickness through multicoatings can enhance organic retention at a large extent but can also increase the filtration 104

118 resistance and the possibility for crack formation [53]. The permeate water flux of the membranes with one or two coating layers was not significantly different from that of the porous Al 2 O 3 substrate itself since the coating layers did not cover the substrate thoroughly, as observed in the ESEM images. After three coatings, the TiO 2 skin layer was almost homogeneous and had a thickness of 0.9 μm, exhibiting a significant decrease in the permeate water flux. The permeability coefficient of the membrane decreased from 11.0 l/m 2 /hr/bar for the alumina support to 3.55 l/m 2 /hr/bar for 5 coating layers. The corresponding hydraulic filtration resistance increased from m -1 for the alumina support to m -1 for 5 coating layers. Results on organic retention of the membrane using PEG solutions with molecular weights of 1,000 to 20,000 Dalton, approximately corresponding to molecular sizes of 2 to 12 nm, were in good agreement with those on water permeation test and ESEM images. As increasing the number of coating layers, the PEG retention efficiency was increased and the molecular weight of PEG rejected by the membranes was shifted to lower size. At least three coating layers were needed to ensure the removal of small molecules. The molecular weight cut off (MWCO) of the membranes was approximately 12,000 Dalton, which is equivalent to approximately 7 nm. The surface of the membranes was completely wetted by water droplet [48]. The permeability coefficient of the membrane with three coating layers was at 6.71 l/m 2 /hr/bar, which was significantly high considering the low MWCO of the TiO 2 membrane and the low permeability coefficient (11.0 l/m 2 /hr/bar) of the thick Al 2 O 3 support [56]. 105

119 Flux (l/m 2 /hr) (a) Substrate T=0 T=1 T=2 T=3 20 T= Pressure (kg/cm 2 ) 100 (b) T=5 T=3 Retention (%) T=2 T=1 20 Substrate T= Molecular weight of PEG (g/mol) Fig. 3.16: (a) Permeate water flux and (b) MWCO of TiO 2 /Al 2 O 3 composite membranes. 106

120 Table 3.3: Properties of TiO 2 /Al 2 O 3 composite membranes. Parameter Substrate Number of coating layers, T Filtration resistance, R m ( m -1 ) Permeability coefficient, L p (l/m 2 /hr/bar) L p /L p,substrate Thickness (μm) 22 mm na >0.5 (>0.25) a 0.9 (0.30) 1.4 (0.28) Molecular weight cut-off (Dalton) 0.1 μm >20K >20K ~12K ~7K Mass of TiO 2 (μg/cm 2 ) - na >103 (>51.6) 186 (62.0) 289 (57.8) Contact angle (º) completely wetted a per unit coating layer Environmental Applications As shown in Figs and 3.18, destruction of MC-LR and inactivation of E. coli were examined to evaluate the photocatalytic activity of the prepared highly porous TiO 2 films. A significant amount of MC-LR disappeared in the dark possibly by adsorption due to the high surface area of TiO 2 film and small initial MC-LR concentration. In the case of TiO 2 /UV photocatalytic experiments, all MC-LR was completely destroyed within 3 h. In order to achieve 4-log removal of E. coli, the TiO 2 /UV photocatalysis process needed 1.5 h, which is significantly fast considering the small amount of TiO 2 catalyst used (622 μg TiO 2 per 8 ml of E. coli solution). These results indicate that the TiO 2 photocatalytic film is highly effective in destroying biological toxins and inactivating microorganisms in water. The high efficiency of TiO 2 films is ascribed to their porous structure and high surface area. A TiO 2 film with nonporous structure (surface area of 22.7 m 2 /g and pore volume of cm 3 /g) was prepared without surfactant and tested. The results denoted 107

121 as [TiO 2 /UV] in Figs and 3.18 indicate that the photocatalytic activity of the control was much lower than that of porous TiO 2 film and similar (or slightly improved) to photolytic activity of UV radiation only. Compared to the porous TiO 2 film, which is able to facilitate adsorption of water contaminants and utilize UV light within several layers of the film (i.e., bulk), the nonporous TiO 2 film limits adsorption of water contaminants and absorption of UV radiation only within the outer layers of the film. The TiO 2 membranes were tested for their anti-biofouling property, which is of high interest in membrane research and industry [57 61]. Pure water flux was measured at TMP of 3 kg/cm 2 after contacting the membrane with dissolved organic solution in static condition at TMP of 0 kg/cm 2 in the presence or absence of UV radiation. As shown in Fig. 3.19, decrease in pure water flux was observed with increasing organic concentration and contact time. Interestingly, TiO 2 membranes irradiated by UV exhibited less flux decline and no significant fouling formation over time, compared to the control experiment without UV. This difference is attributed to the photocatalytic activity of TiO 2. While organic contaminants and microorganisms attach at the TiO 2 membrane surface and interact to form an adsorption fouling layer, they are also attacked by the photocatalytic action of TiO 2, and thus they are decomposed or their attachment strength is weakened. 108

122 Normalized concentration, C/Co TiO 2, dark [TiO 2 / UV] 0.2 TiO 2 / UV Reaction time (h) Fig. 3.17: Destruction of MC-LR by photocatalytic TiO 2 films. 0 TiO 2, dark Log (N/N 0 ) TiO 2 / UV [TiO 2 / UV] UV Reaction time (h) Fig. 3.18: Inactivation of E. coli by photocatalytic TiO 2 films. 109

123 20.1 Permeate water flux (l/m 2 /hr) with UV No UV Contact time (min) Fig. 3.19: Anti-biofouling property of TiO 2 /Al 2 O 3 composite membrane under static condition contacting dissolved organic solution obtained from activated sludge (Circle: DS 200 -DOC 30, square: DS 400 -DOC 60, and triangle: DS 800 -DOC 120 ). Filtration of the dissolved organic solution was performed in the absence of UV illumination, and the results are shown in Fig With increasing organic strength, the flux decreased due to higher fouling potential of the organic solution. In spite of the deadend filtration mode and relatively high concentration of DS, the permeate water flux decreased only slightly over time from 20.1 to 17.9 l/m 2 /hr for the high strength solution. In addition, much higher water permeability may be achieved in scale-up applications of such UV-TiO 2 /Al 2 O 3 composite membrane reactors considering the anti-biofouling properties of the TiO 2 membrane in the presence of UV radiation and the use of thinner porous substrates instead of the 2 mm thick alumina substrates used in this study. DOC rejection over time increased significantly with increasing organic strength due to the 110

124 secondary dynamic membrane effect [58]. The high and reliable DOC rejection efficiency of the membrane beyond 65% after 5 min filtration evidenced the small MWCO of the TiO 2 membrane and indicated the absence of any significant defects Permeate water flux (l/m 2 /hr) Permeate water flux DOC rejection DOC rejection (%) Filtration time (min) 50 Fig. 3.20: Permeate water flux and organic retention of TiO 2 /Al 2 O 3 composite membrane treating dissolved organic solution (Circle: DS 200 -DOC 30, square: DS 400 -DOC 60, and triangle: DS 800 -DOC 120 ). In addition to water permeability, organic retention, and anti-biofouling properties of TiO 2 /Al 2 O 3 membranes, the TiO 2 membrane could effectively decolorize MB dye and decompose creatinine. As shown in Fig. 3.21, the initial blue color of MB dye was completely decolorized within 4 h. Approximately 14 h were required to completely decompose creatinine. However, most organic carbon remained due to the formation of organic intermediates and their slow mineralization. Considering that % m/v 111

125 (TiO 2 mass/reaction volume) of TiO 2 photocatalyst is usually used in suspension for the decomposition of contaminants in water, the high photocatalytic activity per unit TiO 2 mass in this film (0.0078% m/v) and membrane (0.0234% m/v) shows great promise, especially in preparing ultra-thin photocatalytic TiO 2 membranes that can possess even higher permeation water flux at the same TMP. Normalized concentration, C/Co Without TiO2 films TOC (creatinine) Creatinine Methylene blue Reaction time (h) Fig. 3.21: Degradation of methylene blue dye and creatinine by photocatalytic TiO 2 membranes. 3.4 References [1] W. Wang, B. Gu, L. Liang, W.A. Hamilton, D.J. Wesolowski, J. Phys. Chem. B 108 (2004) [2] K. Yoo, H. Choi, D.D. Dionysiou, Chem. Commun. (2004)

126 [3] C. Burda, Y. Lou, X. Chen, A.C.S. Samia, J. Stout, J.L. Gole, Nano Letters 3 (2003) [4] A. Mills, G. Hill, S. Bhopal, I.P. Parkin, S.A.J. O Neil, J. Photochem. Photobiol. A 160 (2003) 185. [5] A.D. Modestov, O. Lev, J. Photochem. Photobiol. A 112 (1998) 261. [6] W. Xi, S.V. Geisen, Wat. Res. 35 (2001) [7] M.D. Moosemiller, C.G. Hill Jr., M.A. Anderson, Sep. Sci. and Technol. 24 (1989) 641. [8] M.A. Anderson, M.J. Gieselmann, Q. Xu, J. Membr. Sci. 39 (1988) 243. [9] F. Bosc, A. Ayral, P.-A. Albouy, L. Datas, C. Guizard, Chem. Mater. 16 (2004) [10] G. Balasubramanian, D.D. Dionysiou, M.T. Suidan, V. Subramanian, I. Baudin, J.M. Laîné, J. Mat. Sci. 38 (2003) 823. [11] P.C.A. Alberius, K.L. Frindell, R.C. Hayward, E.J. Kramer, G.D. Stuchy, B.F. Chmelka, Chem. Mater. 14 (2002) [12] S.-Y. Kwak, S.H. Kim, S.S. Kim, Environ. Sci. Technol. 35 (2001) [13] J.N. Hay, H.M. Raval, Chem. Mater. 13 (2001) [14] Y. Takahashi, Y.Matsuoka, J. Mater. Sci. 23 (1988) [15] T.J. Trentler, T.E. Denler, J.F. Bertone, A. Agrawal, V.L. Colvin, J. Am. Chem. Soc. 121 (1999) [16] O. Dag, I. Soten, O. Helik, S. Polarz, N. Coombs, G.A. Ozin, Adv. Funct. Mater. 13 (2003) 30. [17] E. Stathatos, P. Lianos, P. Falaras, Progr. Colloid Polym. Sci. 118 (2001) 96. [18] S. Barboux-Doeuff, C. Sanchez, Mater. Res. Bull. 29 (1994)

127 [19] C. Sanchez, J. Livage, M. Henry, F. Babonneau, J. Non-Cryst. Solids 100 (1988) 65. [20] D.P. Birnie III, N.J. Bendzko, Mater. Chem. Phys. 59 (1999) 26. [21] C. Wang, Z.X. Deng, Y. Li, Inorg. Chem. 40 (2001) [22] E. Stathatos, P. Lianos, C. Tsakiroglou, Microporous and Mesoporous Mater. 75 (2004) 255. [23] F. Bosc, A. Ayral, P.-A. Albouy, C. Guizard, Chem. Mater. 15 (2003) [24] P. Yang, D. Zhao, D.I. Margolese, B.F. Chmelka, G.D. Stucky, Chem. Mater. 11 (1999) [25] D.M. Antonelli, J.Y. Ying, Angew. Chem. Int. Ed. Engl. 34 (1995) [26] N. Idrissi-Kandri, A. Ayral, M. Klotz, P.-A. Albouny, A.E. Mansouri, A. Van der Lee, C. Guizard, Mater. Lett. 50 (2001) 57. [27] D. Trong On, Langmuir 15 (1999) [28] H. Choi, E. Stathatos, D.D. Dionysiou, Appl. Catal. B 63 (2006) 60. [29] H. Choi, E. Stathatos, D.D. Dionysiou, Thin Solid Films 510 (2006) 107. [30] H. Choi, M.G. Antoniou, A.A de la Cruz, E. Stathatos, D.D. Dionysiou, Desalination 202 (2007) 199. [31] E. Prouzet, F. Cot, G. Nabias, A. Larbot, P. Kooyman, T.J. Pinnavaia, Chem. Mater. 11 (1999) [32] M. Klotz, N. Idrissi-Kandri, A. Ayral, C. Guizard, Mater. Res. Soc. Symp. Proc. (2000) 628. [33] R.S.A. de Lange, J.H.A. Hekkink, K. Keizer, A.J. Burggraaf, J. Non-Crystalline Solids 195 (1996) 203. [34] C.-Y. Tsai, S.-Y. Tam, Y. Lu, C.J. Brinker, J. Membr. Sci. 169 (2000)

128 [35] G.A. Codd, Wat. Sci. Technol., 32 (1995) 149. [36] J. Zhang, Z.-L. Wang, J. Liu, S. Chen, G.-Y. Liu, Self-Assembled Nanostructure; Kluwer Academic/Plenum Publishers: New York, [37] M. Ivanda, S. Music, S. Popovic, M. Gotic, J. Mol. Struc. 480 (1999) 645. [38] S. Nakade, M. Matsuda, S. Kambe, Y. Saito, T. Kitamura, T. Sakata, Y. Wada, H. Mori., S. Yanagida, J. Phys. Chem. B 106 (2002) [39] R.J. Young, Introduction to polymers; Chapman and Hall: New York, [40] A. Henglein, Ber. Bunsenges. Phys. Chem. 86 (1982) 241. [41] C. Kormann, D.W. Bahnemann, M.R. Hoffmann, J. Phys. Chem. 92 (1988) [42] Y. Li, T.J. White, S.H. Lim, J. Solid State Chem. 177 (2004) [43] L.E. Brus, J. Chem. Phys. 80 (1984) [44] S.V. Gaponenko, Optical Properties of Semiconductor Nanocrystals; Cambridge University Press: Cambridge, [45] S. Hata, Y. Kai, I. Yamanaka, H. Oosaki, K. Hiroto, S. Yamagishi, JSAE Rev. 21 (2000) 97. [46] T. Wanatabe, A. Nakajima, R. Wang, M. Minabe, S. Koizumi, A. Fujishima, K. Hashimoto, Thin Solids Films 351 (1999) 260. [47] H. Irie, S. Washizuka, N. Yoshino, K. Hashimoto, Chem. Comm. (2003) [48] M. Mulder, Basic Principles of Membrane Technology; Kluwer Academic Publishers: Netherlands, [49] P.I. Ravikovitch, A.V. Neimark, Langmuir 18 (2002) [50] P.A.A. Alberius, K.L. Frindell, R.C. Hayward, E.J. Kramer, G.D. Stucky, B.F. Chmelka, Chem. Mater. 14 (2002)

129 [51] Z. Zhang, C.-C. Wang, R. Zakaria, J.Y. Ying, J. Phys. Chem. B 102 (1998) [52] M. Anpo, T. Shima, S. Kodamo, Y. Kubokawa, J. Phys. Chem. 91 (1987) [53] A. Mills, S. L. Hunte, J. Photochem. Photobiol. A 108 (1997) 1. [54] T.V. Gestel, C. Vandecasteele, A. Buekenhoudt, C. Dortemont, J. Luyten, R. Leysen, B.V.D. Bruggen, G. Maes, J. Membr. Sci. 207 (2002) 73. [55] A.J. Burggraaf, L. Cot (Eds.), Fundamentals of Inorganic Membrane Science and Technology; Elsevier: Netherlands, [56] F. Bosc, A. Ayral, C. Guizard, In: Proceedings of Eighth International Conference on Inorganic Membranes, Cincinnati, [57] H. Choi, H.-S. Kim, I.-T. Yeam, D.D. Dionysiou, Desalination 172 (2005) 281. [58] H. Choi, K. Zhang, D.D. Dionysiou, D.B. Oerther, G.A. Sorial, J. Membr. Sci. 248 (2005) 189. [59] H. Choi, K. Zhang, D.D. Dionysiou, D.B. Oerther, G.A. Sorial, Sep. Purif. Technol. 45 (2005) 68. [60] H. Choi, K. Zhang, D.D. Dionysiou, D.B. Oerther, G.A. Sorial, Chemosphere 63 (2006) [61] K. Zhang, H. Choi, D.D. Dionysiou, G.A. Sorial, D.B. Oerther, Environ. Microbiol. 8 (2006)

130 CHAPTER 4 Nanocrystalline TiO 2 Photocatalytic Membranes with a Hierarchical Mesoporous Multilayer: Synthesis, Characterization, and Multifunction 117

131 A novel sol-gel dip-coating process to fabricate nanocrystalline TiO 2 photocatalytic membranes with a robust hierarchical mesoporous multilayer and improved performance was studied. Various titania sols containing polyoxyethylene sorbitan monooleate (Tween 80) surfactant as a pore directing agent to tailor-design the porous structure of TiO 2 materials at different molar ratios of Tween 80:isopropanol:acetic acid:titanium tetraisopropoxide = R:45:6:1 were synthesized. The sols were dip-coated on top of a home-made porous alumina sublayer to make TiO 2 /Al 2 O 3 composite membranes, dried and calcined, and this procedure was repeated with varying sols in succession. The resulting asymmetric mesoporous TiO 2 membrane with thickness of 0.9 μm exhibited a hierarchical change in pore diameter from 2 6 nm, 3 8 nm, to 5 11 nm from the top to the bottom layer. Moreover, the corresponding porosity was incremented from 46.2 %, 56.7 % to 69.3 %. Compared to a repeated-coating process using a single sol, the hierarchical multilayer process improved water permeability significantly, without sacrificing organic retention and photocatalytic activity of the TiO 2 membranes. The prepared TiO 2 photocatalytic membrane has great potential in developing highly efficient water treatment and reuse systems because of its multifunction such as decomposition of organic pollutants, inactivation of pathogenic microorganisms, physical separation of contaminants, and anti-biofouling action. 118

132 4.1 Introduction Along with new and increasingly stringent regulations for the discharge of effluents, which require a more reliable and sustainable treatment process, the development of costeffective membrane manufacturing has accelerated the use of membrane technology in wastewater treatment and reuse [1,2]. Membrane-based separations are energy-efficient and cost-effective when optimized [3]. They represent promising alternatives to conventional water and wastewater treatment processes as well as air purification and gas separation processes [1,2,4,5]. Microporous and mesoporous inorganic membranes have attracted considerable attention for water treatment due to their excellent thermal, chemical, and mechanical stability and reusability after burning over conventional polymeric membranes (e.g., cellulose derivatives, polysulfone, polyamide, PVDF, and PTFE) [6,7]. Among the materials used for the preparation of inorganic membranes, TiO 2 photocatalysts have gained tremendous popularity since a membrane skin layer composed of TiO 2 material allows only small molecules and water to penetrate through membrane pores (i.e., separation function) and has photocatalytic activity to decompose recalcitrant organic pollutants in water [8,9]. The photocatalytic activity is also expected to affect antibiofouling properties of the membrane, which is one of the main goals of scientific research and industrial progress in membrane-based water and wastewater treatment applications [8,10]. TiO 2 /UV systems can generate hydroxyl radicals efficiently. In fact, TiO 2 photocatalysis has been found extremely effective for the complete mineralization 119

133 of virtually all organic compounds and for the inactivation of pathogenic microorganisms present in contaminated water [11 14]. An inorganic membrane generally consists of a macroporous substrate providing mechanical strength for an overlying thin active layer [15,16]. It is known that the quality of the underlying support layer determines, to a large extent, the quality of the top skin mesoporous titania membrane. A support surface with large pore irregularity would cause cracking and pin-holes in the overlying skin layer because of uneven stress development on the coatings during the sol-gel process. In order to reduce surface roughness, conventional mechanical polishing could be applied, but it is a time consuming procedure impractical for some membrane configurations. Therefore, it is important to develop a simple and effective strategy to modify the support surface to facilitate deposition of a defect-free overlying TiO 2 membrane. The use of an intermediate layer between macroporous sublayer and mesoporous (or microporous) skin layer has been introduced to overcome flow resistance, surface roughness, and inherent support defects [16,17]. However, the formation of an intermediate layer itself also needs to be multicoated with an additional sol composed of a different material with relatively large size particles. This results in increased preparation cost and time and reduced water permeability. Alternatively, repetition of the dip-coating procedure using a single sol, until the desired thickness and homogeneity of the coating layer are obtained, has been suggested [18,19]. In this method, the thick symmetric TiO 2 coating layer consists of a number of single coating layers, which results in increase of hydraulic resistance. The ideal membrane structure would be composed of a thin asymmetric multilayer with increasing pore size and porosity from the top to the bottom of TiO 2 skin layer (i.e., hierarchical 120

134 multilayer). Compared to the usual repeated-coating prepared with only the same mesoporous overlayers, a hierarchical multilayer can improve water permeation flux significantly, maintaining the same organic retention and photocatalytic activity as the typical repeated-coating process. In this study, we report the preparation of photocatalytic TiO 2 membranes with hierarchical mesoporous multilayer structure using a sol-gel method modified with surfactant [20]. It is well known that the presence of templates such as surfactants in solgel chemistry plays a crucial role in creating the porous structure of the TiO 2 inorganic network, which reduces the hydraulic resistance of TiO 2 membranes and enhances their photocatalytic activity [21 24]. In this method, the surfactant concentration largely affects the structural properties of the TiO 2 material [25,26]. Controlling materials at the nano level makes it possible to develop new types of catalytic membrane products with tailor-designed properties exhibiting a hierarchical change in the diameter of the mesopore from the top to the bottom of the TiO 2 skin layer [27,28]. The multifunction of the prepared TiO 2 photocatalytic membrane including decomposition of organic pollutants, inactivation of pathogenic microorganisms, physical separation of contaminants, and self-antifouling action are investigated. 4.2 Experimental Sol Preparation The common sol-gel methods employing direct water addition lead to the precipitation of amorphous particles with uncontrolled structure and then the TiO 2 particles are 121

135 peptized in acidic conditions to make a homogeneous sol [29]. Alternatively, in an acetic acid-based sol-gel method, water molecules released from the esterification reaction of acetic acid with isopropanol during synthesis was utilized instead of directly adding water [30]. In addition, various template-based sol-gel methods using a variety of surfactants and block copolymers have been applied in the past to synthesize tailor-designed TiO 2 catalytic materials [31,32]. In this study, the acetic acid-based sol-gel method was modified with surfactants. Polyoxyethylene sorbitan monooleate (Tween 80, Aldrich) surfactant was selected as a pore directing agent in a sol. A suitable amount of Tween 80 was homogeneously dissolved in isopropanol (iproh, Fisher). Acetic acid (Fisher) was added into the solution for the esterification reaction. When adding titanium tetraisopropoxide (TTIP, Aldrich) into the solution, hydrolysis and condensation reactions occurred through in-taking of water molecules released from the esterification reaction. The molar ratio of the ingredients was optimized at Tween 80:iPrOH:acetic acid:ttip = R:45:6:1, where the surfactant concentration R was varied in the range from 0 for the control to 1 3 for target structural properties. At surfactant concentrations above R = 3, film homogeneity suddenly deceased Dip-Coating and Heat Treatment Fabrication of the porous substrate made of fine alumina powder was performed by modification of well established procedures [33]. A16-SG alumina powder (Alcoa Chemicals) was mixed with a suitable amount of water, placed into a mold, and pressed at 5000 lb for 30 sec followed by 7500 lb for 3 min using a hydraulic unit (Model 3925, Carver). The pressed disks were calcined and sintered up to 1220 ºC for 7 d in a furnace 122

136 (Vulcan HTA, Neytech) with programmed multistage temperatures and holding times. The diameter and thickness of the alumina substrate were 21 mm and 2.2 mm, respectively. The pore size and pure water permeability coefficient were 0.1 μm and 11.0 L/m 2 /h/bar, respectively. One side of the alumina substrate was dipped into the sol for 20 sec and taken out with extreme caution using a home-made dip-coating device. After coating, the membranes were dried at room temperature for 1 h, calcined in a multisegment programmable furnace (Paragon HT-22-D, Thermcraft) at a ramp rate of 3 ºC min -1 up to 500 ºC, maintained at this temperature for 15 min, and cooled down naturally. This dip-coating procedure was repeated with different sols at various surfactant concentrations, R, in succession. The number of dip-coatings (denoted as T) was usually one or three. Due to the difficulty to directly characterize the properties of TiO 2 materials on an alumina support layer, easy-to-remove TiO 2 films were prepared by dip-coating borosilicate glass (Micro slide, Gold Seal) with the same sols used for the preparation of the TiO 2 membrane at a withdrawal rate of 12.8 cm/min. After the coating procedure, the other drying and calcination procedures were the same as for the TiO 2 membranes. Only TiO 2 material was collected by scrapping the thin film surface cautiously Materials Characterization In order to determine the crystal structure and crystallinity of the TiO 2 material, X-ray diffraction (XRD) analysis using a Kristalloflex D500 diffractometer (Siemens) with Cu Kα (λ = Å) radiation was employed. A porosimetry analyzer (Tristar 3000, Micromeritics) was used to determine structural characteristics of TiO 2 materials 123

137 including Brunauer, Emmett, and Teller (BET) specific surface area, porosity, and pore size distribution in the mesoporous range, using nitrogen adsorption and desorption isotherms. In order to determine the band gap energy of TiO 2, its UV-visible light absorbance was measured using a UV-Vis spectrophotometer (Hewlett Packard 8452A). The structure of TiO 2 materials at the nano-level was visualized using a JEM-2010F (JEOL) high resolution-transmission electron microscope (HR-TEM) with a field emission gun. An environmental scanning electron microscope (ESEM, Philips XL 30 ESEM-FEG) was used to measure the thickness of coatings and ensure the homogeneity of TiO 2 membranes. An elemental composition analysis of the TiO 2 was performed using an energy dispersive X-ray spectroscope (EDX, Oxford Isis) connected to the HR-TEM and ESEM. An X-ray photoelectron spectroscope (XPS, Perkin-Elmer model 5300) with Mg Kα X-rays was used at a take-off angle of 45º for the accurate measurement of trace elements in a few nanometers of the material surface and oxidation states of titanium. An organic component in the TiO 2 during heat treatment was monitored using Fourier transform infrared spectroscope (FTIR, Nicolet Magna-IR 760) and thermogravimetric analysis instrument (TGA, TA instruments 2050) Evaluation of TiO 2 Membranes The water permeability and organic retention of the prepared TiO 2 /Al 2 O 3 composite membranes were evaluated with a home-fabricated membrane chamber. The pure water was pressurized with a nitrogen gas at a transmembrane pressure (TMP) from 0 to 6 kg cm -2, and the permeate water flux was measured. Instead of pure water, 200 mg/l of polyethylene glycol (PEG, Fluka) solutions with different molecular weights were 124

138 pressurized at TMP of 1 kg/cm to determine the molecular weight cut-off (MWCO) of the TiO 2 membranes and investigate their integrity. Organic carbon concentrations in the raw PEG solution and permeate water were measured using a total organic carbon (TOC) analyzer (Shimadzu, TOC-V CSH). Experiments to examine the photocatalytic activity of TiO 2 membranes were conducted by immersing TiO 2 membranes with 10 cm 2 surface area into a borosilicate glass reactor containing 8 ml of 30 μm methylene blue dye (MB, Riedel-deHaën), which is a commonly used dye in photocatalytic tests or 8 ml of cfu/ml E. coli (ATCC 11229), which is a model pathogenic microorganism for microbial inactivation tests. Two 15 W low pressure mercury UV tubes (Spectronics) emitting near UV radiation with a peak at 365 nm were used at a light intensity of 3.48 mw/cm. The concentration of MB solution was determined by measuring the visible light absorbance at 664 nm using a UV-Vis spectrophotometer (Hewlett Packard 8452A). The number of viable E. coli was counted using a colony forming test where membrane filtration technique and ColiBlue24 broth (enrichment medium, HACH) were employed. The antifouling properties of TiO 2 membranes were investigated with a high strength organic solution obtained by filtering an activated sludge treating municipal wastewater using a 1.2 μm glass fiber filter. The organic solution contained around 800 mg/l dissolved solids (170 mg/l organics and 630 mg/l inorganics) and around 120 mg/l dissolved organic carbon. Pure water flux was measured at TMP of 3 kg/cm after contacting the membrane with the organic solution in static condition (i.e., TMP of 0 kg/cm) in the presence or absence of UV radiation. 125

139 4.3 Results and Discussion Surfactant Effect on the Properties of TiO 2 Material Regardless of surfactant addition, significant organic content and further weight loss of the TiO 2 material calcined at temperatures greater than 500 ºC were not observed during FTIR and TGA analyses. A bulk EDX elemental analysis showed that the TiO 2 material was composed of mainly Ti and O elements at a stoichiometry of around 1:1.89 ~ 1.95 possibly due to the presence of defect structures in the TiO 2 and oxidation states other than TiO 2. In order to determine its structural and crystallographic properties, the TiO 2 material was collected by scrapping the thin film surface and was analyzed. The structure and porosity of TiO 2 material at the nano-level should not be affected by such an experimental procedure, and the nanostructure of TiO 2 material in the membrane is considered identical to that in the film [15,18]. Fig. 4.1 shows the morphology of the nanostructured anatase TiO 2 material. Control TiO 2 material shown in Fig. 4.1(a) demonstrated no distinct mesoporous structure. On the other hand, TiO 2 materials prepared with Tween 80 were highly porous and exhibited a distinct pore structure, as shown in Figs. 4.1(b), 4.1(e) and 4.1(f). Compared to Fig. 4.1(b), Figs. 4.1(e) and 4.1(f) presented relatively larger pore sizes, which suggests that the pore size increased slightly with increasing surfactant concentration. As illustrated in Fig. 4.1(c), all the TiO 2 materials have slightly collapsed spherical (cubic) bicontinuous structure with highly interconnected network, inducing high materials porosity from 46% to 69%, as surfactant concentration increases [33]. 126

140 Fig. 4.1: TEM images of TiO2 material prepared at: (a) R = 0, (b-d) R = 1 at different magnifications, (e) R = 2, and (f) R =

141 This interconnected pore network phase would be much more attractive than a two-dimensional hexagonal phase for applications in photocatalysis requiring diffusion of species into and out of the pore network and in membrane separation requiring water permeability. Fig. 4.1(d) at high magnification showed obvious 4 6 nm pores and many randomly oriented 5 10 nm nanocrystallites with sets of clearly resolved lattice fringes giving evidence that the TiO 2 material was highly crystalline. The porous structure was thermally stable since the structure still largely remained until heat treatment at 600 ºC. The pore size controllability of TiO 2 materials was evidenced by measuring N 2 adsorption-desorption isotherms, as shown in Fig Compared to N 2 isotherms of the control TiO 2 representing a nonporous material, those of TiO 2 prepared with Tween 80 exhibited IV type isotherms typical for a well-developed mesoporous material. A hysteresis loop in the isotherms was observed with dissimilar shapes for the adsorption and desorption branches. The sharp drop on the desorption branch can be assigned to the presence of smaller pores in the titania walls [35]. These results implied different pore throat diameters and slightly disordered pores, which are in agreement with TEM images. Fig. 4.2(b) demonstrates that the pore size distribution of TiO 2 materials is controllable by varying surfactant concentration. The pore size distribution was shifted from 2 6 nm, 3 8 nm, to 5 11 nm with increasing R from 1 to 3. Moreover, the corresponding structural properties of TiO 2 materials were enhanced from 147, 155, to 159 m 2 /g for BET specific surface area, and from 46.2, 56.7, to 69.3% for materials porosity. 128

142 Volume Adsorbed (cm 3 g -1 ) (a) R = 3 R = 2 R = 1 R = Relative Pressure (P s /P o ) Pore volume (cm 3 g -1 ) (b) R=0 R=1 R=2 R= Pore diameter (nm) Fig. 4.2: (a) N 2 adsorption/desorption isotherms and (b) pore size distribution of TiO 2 material. 129

143 The underlying mechanism for the formation of the highly porous structure with controlled pore size is associated with the role of surfactant that can act as a pore structure-forming agent [26,31]. Amphiphilic molecules such as surfactants exist in a wide range of ordered structures in their condensed states [36]. Titanium alkoxide precursors, when added into the self-organized surfactant solution, are hydrolyzed and condense around the surfactant templates. By increasing the surfactant concentration, water molecules available for association with the hydrophilic head group of the surfactant decreases, resulting in a decrease in the degree of hydration of the surfactant head group and thus a decrease in the effective head group area. Based on the critical packing parameters, which determine the packing geometry of surfactant-assembly, a reduction of the head group area increases the critical packing parameters, which favors a less curved geometry (e.g., flat lamellar phase) rather than the more curved spherical micelles [24]. During thermal treatment, the surfactants and other organic residues are removed, leaving a pore structure that mimics the properties of the surfactant packing geometry. However, in this study, it was difficult to identify a change in such pore geometry since the high calcination temperature of 500 ºC is believed to be high enough to destroy the initial ordered pore structure of TiO 2 material [22]. Moreover, the high surfactant concentrations used are expected to cause more pore coalescence and multimicellar interactions during heat treatment, resulting in high porosity and large pores. Long-range ordering of TiO 2 pore network is not a definite proof for better materials for applications in photocatalytic membranes. This is because the enhanced crystallinity of TiO 2 material from 67% at 350 ºC, where the ordered pore structure started to collapse, to 94% at 500 ºC and its highly interconnected pore network 130

144 during heat treatment at high temperatures can overwhelm disadvantages caused by the decrease in pore ordering of TiO 2 material [22]. As a result, in this sol-gel method, the controlled hydrolysis of titania precursor is achieved through indirectly taking water, and the stable incorporation of the titania inorganic network onto self-organized surfactant molecules could induce the formation of TiO 2 photocatalysts with tailor-designed structural properties. In addition, as the concentration of templating materials in a sol increases, higher porosity and relatively larger pore size could be created. (101) (004) (200) (105) (211) Intensity (a.u.) θ (degrees) R, [Tween 80]/[TTIP] Fig. 4.3: XRD patterns of TiO 2 material. 131

145 Fig. 4.3 shows XRD spectra of TiO 2 materials. All peaks are assigned to the anatase crystal phase, which is known to be the most active phase. The relatively wide width of the peaks indicates small crystallite size, which was in good agreement with the TEM results. The crystallite size was estimated to be approximately 12.4 nm at R = 0, 9.20 nm at R = 1, 8.85 nm at R = 2, and 9.46 nm at R = 3 using Scherrer s equation from the XRD peak broadening analysis at (1 0 1) [37]. The 8 10 nm crystallite size of TiO 2 has been reported to be optimum for high catalytic activity [38]. Table 4.1: Physicochemical properties of TiO 2 materials. a Parameter Control Surfactant concentration R=0 R=1 R=2 R=3 BET specific surface area (m 2 /g) Pore volume (cm 3 /g) Porosity (%) BJH adsorption pore diameter (nm) BJH desorption pore diameter (nm) Crystal phase Anatase Anatase Anatase Anatase Crystallite size (nm) Film thickness (nm) TiO 2 mass (μg/cm 2 ) Band gap energy (ev) Ti:O stoichiometry in bulk 1:1.94 1:1.92 1:1.89 1:1.95 Ti:O stoichiometry in surface 1:2.06 1:2.18 1:2.19 1:2.17 TiO 2 (IV):Ti 2 O 3 (III) 1:0.15 1:0.13 1:0.13 1:0.12 Traces (Atomic %) C: 0.8 F: < 0.2 N: < 0.2 P: <0.2 C: 1.3 F: 0.7 N: 0.7 P: 0.3 C: 1.5 F: 0.8 N: 0.7 P: < 0.2 C: 1.8 F: 0.9 N: 0.9 P: 0.3 a Scraped from TiO 2 thin film on glass substrate. 132

146 The main structural characteristics obtained from the analyses so far are summarized in Table 4.1. In spite of the high heat treatment temperature of 500 ºC for the TiO 2 material prepared with Tween 80, a BET surface area of above 147 m 2 /g, pore volume of above cm 3 /g and porosity of above 46% were significantly high, compared to other research results reported and to the control TiO 2 material with surface area of 22.7 m 2 g, pore volume of cm 3 /g, and porosity of 12.6% [21,32]. The Barrett, Joyner and Halenda (BJH) pore diameter based on the adsorption branch was similar to that based on the desorption branch, implying good homogeneity of pores. The film thickness of around 100 nm per coating was measured using ESEM, and it increased slightly due to the increased viscosity of the sols as increasing surfactant concentration. The amount of TiO 2 mass was calculated from the pore volume and density of the anatase crystal phase. Even though the film thickness increased with increasing surfactant concentration, the amount of TiO 2 catalyst decreased due to the high materials porosity. The results so far show that this sol-gel method has good potential for application in fabricating hierarchical mesoporous multilayer TiO 2 membranes with less hydraulic resistance. The band gap energy, E g of the TiO 2 materials was slightly larger than 3.23 ev for typical bulk anatase TiO 2 particles. This blue shift was caused by small crystallites less than 10 nm due to the quantum size effect [39]. The crystallite size effect on the band gap energy was consistent with the observed result that smaller crystallites in TiO 2 prepared with Tween 80 have higher band gap energies. TiO 2 stoichiometry in the material surface measured using XPS was slightly different from that in the bulk measured using EDX due to the presence of many hydroxyl groups at the material surface. High resolution 133

147 XPS analysis for determining the Ti oxidation states at 2p 3/2 level corresponding to binding energy of ev showed that Ti(IV) is predominant with a small fraction of Ti(III). It should be noted that the surfactant addition in this sol-gel method did not affect the TiO 2 stoichiometry and titanium oxidation state. However, compared to relatively pure control TiO 2 at R = 0, TiO 2 materials prepared with Tween 80 had trace amount of impurities including carbon, fluoride, nitrogen, and phosphorous TiO 2 Multicoating The integrity of the TiO 2 skin layer and its incorporation with the support layer are crucial factors in the fabrication of defect-free TiO 2 membranes. As mentioned in Chapter 3, at least three coating layers in this sol-gel method are necessary to fabricate a TiO 2 skin layer with good integrity and without significant cracks and pin-holes. A multicoating with three layers was fabricated with sols at different concentrations in the order of R = 3, R = 2, and R = 1 in succession or a repeated-coating with three layers was fabricated with a single sol at R =1 (otherwise specified, R = 2 or R = 3). The top layer of TiO 2 membrane for both cases was prepared with the sol at R = 1 to achieve the same extent of organic retention. Fig. 4.4 shows the ESEM image of the multicoating TiO 2 /Al 2 O 3 composite membrane. The thickness of the TiO 2 skin layer was uniform at approximately 0.9 μm, which is three times thicker than the TiO 2 coating on smooth glass substrate due to the surface irregularity and porous structure of the alumina substrate, and the TiO 2 membranes were relatively well-incorporated with alumina substrate. 134

148 Fig. 4.4: ESEM image of TiO 2 membrane. In order to visually investigate how the multicoating layer is developed and connected to each other at the nano-level, the sols were coated onto glass substrates in the same manner as the TiO 2 /Al 2 O 3 membranes since the TiO 2 multicoating layer on the alumina substrate does not satisfy sample conditions required for the TEM analysis. The multicoating layer was cautiously scratched from the film and sonicated for 1 h to thoroughly disturb the multilayer in order to observe a contrast between the layers. In general, it is very difficult to find the multilayer since a side-view section of the film hardly appears under TEM analysis condition. In this study, such a multilayer structure is illustrated in Fig Each layer with a thickness of around 100 nm was well-developed, and the hierarchical mesoporous structure previously explained in Figs. 4.1 and 4.2 is also shown. Considering the long time sonication of 1 h, the multilayer structure is considered robust. 135

149 Fig. 4.5: TEM images of TiO 2 films with a multicoating layer: (a) three layers prepared at R = 3, 2, and 1 in succession, (b) boundary between R = 3 and 2, and (c) boundary between R = 2 and

150 4.3.3 Water Permeability and Organic Retention Fig. 4.6 shows water permeability and PEG retention of some TiO 2 /Al 2 O 3 composite membranes prepared using multicoating and repeated-coating processes, and the properties are summarized in Table 4.2. The permeate water flux of the membranes with one or two coating layers was not significantly different from that of the porous substrate itself. After three coatings for both the multicoating membrane and the repeated-coating membrane, the TiO 2 skin layer was almost homogeneous and had a thickness of 0.9 μm, exhibiting a significant decrease in the permeate water flux. The permeability coefficient decreased from 11.0 L/m 2 /h/bar for the alumina support to 7.69 L/m 2 /h/bar for the multicoating membrane and to 6.71 L/m 2 /h/bar for the repeated-coating membrane prepared with R = 1. The corresponding hydraulic filtration resistance increased from m -1 for the alumina support to m -1 for the multicoating membrane and to m -1 for the repeated-coating membrane. The water permeability of the multicoating layer was significantly increased up to 15%, compared to the repeatedcoating membrane. Table 4.2: Properties of multi-coating TiO 2 /Al 2 O 3 membranes. Parameter Substrate Multicoating Repeated-coating T=1 R=3 T=2 R=3-2 T=3 R=3-2-1 T=3 R=1 (T=3) (R=2) (T=3) (R=3) R m ( m -1 ) L p (L/m 2 /h/bar) L p /L p,substrate Thickness (μm] 22 (mm) na ~0.6 ~0.9 ~0.9 ~0.9 ~0.9 MWCO (Dalton) 0.1 (μm) >>20k >>20k ~13k ~12k ~17k >20k Mass (μg/cm 2 ) na na >

151 80 (a) Flux (L m -2 hr -1 ) Substrate T=0 T=1, R=3 T=2, R=3-2 T=3, R=3-2-1 T=3, R= Pressure (kg cm -2 ) Retention (%) (b) T=3, R=1-1-1 T=3, R=3-2-1 T=2, R=3-2 T=1, R=3 20 Substrate T= Molecular weight of PEG (g mol -1 ) Fig. 4.6: (a) Water permeability and (b) PEG retention of TiO 2 membranes. Note that T is the number of dip-coatings. 138

152 The water permeability of the TiO 2 membranes was in good agreement with their PEG retention. As the number of coating layers increased, lower molecular weight PEG was rejected by the membranes. At least three coating layers were needed for both the multicoating membrane and the repeated-coating membrane to ensure the removal of small molecules. The molecular weight cut off (MWCO) of the multicoating membrane was 13,000 Dalton, which was similar to 12,000 Dalton for the repeated-coating layer. This result implies that most of the organic retention occurred at the very surface of membranes, the top layer prepared at R = 1. The MWCO of the membranes, which is equivalent to approximately 6 7 nm, is relatively consistent with the pore size distribution of the TiO 2 material prepared at R = 1 (note Fig. 4.2). As a result, compared to a repeated-coating process using a single sol, the hierarchical multilayer process prepared with the various sols improved water permeability significantly, without sacrificing organic retention of the TiO 2 membranes Photocatalytic Activity and Antifouling Properties The photocatalytic activity of TiO 2 membranes was examined in terms of methylene blue (MB) dye decomposition and E. coli inactivation, as shown in Fig Compared to no direct photolysis of MB in the absence of TiO 2 photocatalyst, the TiO 2 /UV photocatalytic system effectively decolorized the initial blue color of MB within 3 h. The photocatalytic activity of the multicoating membrane was similar or even slightly higher than that of the repeated-coating membrane. 139

153 Normalized concentration, C/C o (a) UV without TiO 2 TiO 2 without UV T=3, R=1-1-1 T=3, R= Reaction time (hr) Log inactivation of E. coli (N/N 0 ) (b) TiO 2 / UV TiO 2 without UV UV without TiO Reaction time (hr) Fig. 4.7: (a) Photocatalytic degradation of methylene blue dye and (b) photocatalytic inactivation of pathogenic microorganism, E. coli by TiO 2 membranes. 140

154 In order to achieve 4-log inactivation of E. coli, the TiO 2 /UV photocatalytic process needed less than 1.5 h while the UV disinfection required more than 3 h. Considering that % m/v (TiO 2 mass/reaction volume) of TiO 2 photocatalyst is usually used in suspension for the decomposition of contaminants in water, the high photocatalytic activity per TiO 2 mass in these membranes ( % m/v) shows a great promise [40]. The high photocatalytic efficiency of the TiO 2 membrane is attributed to their porous structure with high surface area, which is able to facilitate adsorption of water contaminants and effective utilization of UV light Permeate water flux (L m -2 hr -1 ) with UV without UV Contact time (min) Fig. 4.8: Antifouling properties of TiO 2 membranes. Inherent antibiofouling in a membrane is highly important in membrane research and industry. As shown in Fig. 4.8, interestingly, TiO 2 membranes irradiated by UV exhibited less flux decline and no significant fouling formation over time, compared to 141

155 the control experiment without UV. This difference is contributed to the photocatalytic and photolytic activity of TiO 2 /UV system. While organic contaminants and microorganisms attach at the TiO 2 membrane surface and interact to form an adsorption fouling layer in static condition, they are also attacked by the photocatalytic and photolytic action and thus the foulants are decomposed or their attachment strength is weakened [41]. 4.4 References [1] H. Choi, H.-S. Kim, I.-T. Yeam, D.D. Dionysiou, Desalination 172 (2005) 281. [2] H. Choi, K. Zhang, D.D. Dionysiou, D.B. Oerther, G.A. Sorial, Sep. Purif. Technol. 45 (2005) 68. [3] G. Owen, M. Bandi, J.A. Howell, S.J. Churchouse, J. Membr. Sci. 102 (1995) 77. [4] X. Tan, K. Li, W.K. Teo, AICHE Journal 51 (2005) [5] R.M. de Vos, W.F. Maier, H. Verweij, J. Membr. Sci. 158 (1999) 277. [6] L.G.A. van de Water, T. Maschmeyer, Top. Catal. 29 (2004) 67. [7] Y.S. Lin, Sep. Purif. Technol. 25 (2001) 39. [8] S.-Y. Kwak, S.H. Kim, S.S. Kim, Environ. Sci. Technol. 35 (2001) [9] D.-S. Bae, K.-S. Han, S.-H. Choi, Solid State Ionics 100 (1998) 239. [10] H. Choi, K. Zhang, D.D. Dionysiou, D.B. Oerther, G.A. Sorial, J. Membr. Sci. 248 (2005) 189. [11] K. Yoo, H. Choi, D.D. Dionysiou, Catal. Commun. (2005) 259. [12] J.-M. Herrmann, Top. Catal. 34 (2005)

156 [13] C. Chen, P. Lei, H. Ji, W. Ma, J. Zhao, H. Hidako, N. Serpone, Environ. Sci. Technol. 38 (2004) 329. [14] D.D. Dionysiou, A.A. Burbano, M.T. Suidan, I. Baudin, J.M. Laîné, Environ. Sci. Technol. 36 (2002) [15] T.V. Gestel, C. Vandecasteele, A. Buekenhoudt, C. Dortemont, J. Luyten, R. Leysen, B.V.D. Bruggen, G. Maes, J. Membr. Sci. 207 (2002) 73. [16] C.-Y. Tsai, S.-Y. Tam, Y. Lu, J. Brinker, J. Membr. Sci. 169 (2000) 255. [17] A.J. Burggraaf, L. Cot, In: A.J. Burggraaf, L. Cot (Eds.), Fundamentals of Inorganic Membrane Science and Technology; Elsevier: Netherlands, [18] L.-Q. Wu, P. Huang, N, Xu, J. Shi, J. Membr. Sci. 173 (2000) 263. [19] R.M. De Vos, H. Verweij, J. Membr. Sci. 143 (1998) 37. [20] H. Choi, A.C. Sofranko, D.D. Dionysiou, Adv. Funct. Mater. 16 (2006) [21] E. Stathatos, P. Lianos, C. Tsakiroglou, Micropor. Mesopor. Mat. 75 (2004) 255. [22] F. Bosc, A. Ayral, P.-A. Albouy, L. Datas, C. Guizard, Chem. Mater. 16 (2004) [23] A.S. Deshpande, D.G. Shchukin, E. Ustinovich, M. Antonietti, R.A. Caruso, Adv. Funct. Mater. 15 (2005) 239. [24] H. Choi, K. Yoo, D.D. Dionysiou, In: F.T. Akin, Y.S. Lin (Eds.), Proceedings of Eighth International Conference on Inorganic Membranes, Cincinnati, [25] J.Z. Zhang, Z.-L. Wang, J. Liu, S. Chen, G.-Y. Liu, Self-Assembled Nanostructures; Kluwer Academic / Prenum Publishers: New York, [26] J.Y. Ying, C.P. Mehnert, M.S. Wong, Angew. Chem. Int. Ed. 38 (1999) 56. [27] Y. Chen, J.C. Crittenden, S. Hackney, L. Sutter, D.W. Hand, Environ. Sci. Technol. 39 (2005)

157 [28] K. Yoo, H. Choi, D.D. Dionysiou, Chem. Comm. (2004), [29] D.G. Shchukin, R.A. Caruso, Adv. Funct. Mater. 13 (2003) 789. [30] C. Wang, Z.X. Deng, Y. Li, Inorg. Chem. 40 (2001) [31] P. Yang, D. Zhao, D.I. Margolese, B.F. Chmelka, G.D. Stucky, Chem. Mater. 11 (1999) [32] F. Bosc, A. Ayral, P.-A. Albouy, C. Guizard, Chem. Mater. 15 (2003) [33] J. Kim, Y.S. Lin, J. Am. Ceram. Soc. 82 (1999) [34] N.K. Raman, M.T. Anderson, C.J. Brinker, Chem. Mater. 8 (1996) [35] P.I. Ravikovitch, A.V. Neimark, Langmuir 18 (2002) [36] Ö. Dag, I. Soten, Ö. Çelik, S. Polarz, N. Coombs, G.A. Ozin, Adv. Funct. Mater. 13 (2003) 30. [37] S. Nakade, M. Matsuda, S. Kambe, Y. Saito, T. Kitamura, T. Sakata, Y. Wada, H. Mori., S. Yanagida, J. Phys. Chem. B 106 (2002) [38] Z. Zhang, C.-C. Wang, R. Zakaria, J.Y. Ying, J. Phys. Chem. B 102 (1998) [39] A. Henglein, Ber. Bunsenges. Phys. Chem. 86 (1982) 241. [40] D.D. Dionysiou, M.T. Suidan, I. Baudin, J.-M. La, T.L. Huang, Wat. Sci. Technol.: Wat. Supply 1 (2001) 139. [41] E.J. Wolfrum, J. Huang, D.M. Blake, P.-C. Maness, Z. Huang, J. Fiest, W.A. Jacoby, Environ. Sci. Technol. 36 (2002)

158 CHAPTER 5 Mesoporous Nitrogen-Doped TiO 2 for the Photocatalytic Destruction of the Cyanobacterial Toxin Microcystin-LR under Visible Light Irradiation 145

159 The presence of the harmful cyanobacterial toxins in water resources worldwide drives the development of an innovative and practical water treatment technology with great urgency. This study deals with two important aspects: the fabrication of mesoporous nitrogen doped TiO 2 (N-TiO 2 ) photocatalysts and their environmental application for the destruction of microcystin-lr (MC-LR) under visible light. In a nanotechnological solgel synthesis method, a nitrogen-containing surfactant (dodecylammonium chloride) was introduced as a pore templating material for tailor-designing the structural properties of TiO 2 and as a nitrogen dopant for its visible light response. The resulting N-TiO 2 exhibited significantly enhanced structural properties including 2-8 nm mesoporous structure (porosity 44%) and high surface area of 150 m 2 /g. Red shift in light absorbance up to 468 nm, 0.9 ev lower binding energy of electrons in Ti 2p state, and reduced interplanar distance of crystal lattices proved nitrogen doping in the TiO 2 lattice. Due to its narrow band gap at 2.65 ev, N-TiO 2 efficiently degraded MC-LR under visible spectrum above 420 nm. Acidic condition (ph 3.0) was more favorable for the adsorption and photocatalytic degradation of MC-LR on N-TiO 2 due to electrostatic attraction forces between negatively charged MC-LR and +6.5 mv charged N-TiO 2. Even under UV light, MC-LR was decomposed 3~4 times faster using N-TiO 2 than control TiO 2. The degradation pathways and reaction intermediates of MC-LR were not directly related with energy source for TiO 2 activation (UV and visible) and nature of TiO 2 (neat and nitrogen-doped). This study implies a strong possibility for the in-situ photocatalytic remediation of contaminated water with cyanobacterial toxins and other toxic compounds using solar light, a sustainable source of energy. 146

160 5.1 Introduction The increasing occurrence of cyanobacteria harmful algal blooms (cyano-habs) in water resources worldwide is alarming the environmental and health authorities because of their ability to release toxic metabolites, cyanotoxins [1]. The presence of these toxins in various aquatic environments including fresh, recreational, processed, and reclaimed water, has both environmental and socioeconomic impacts [2-5]. Specifically, the piscary of Atlantic salmon is an important industry in USA and Canada, and the salmon has been reported to suffer from liver disease due to the chronic exposure to a naturally occurring hepatotoxin [2]. Studies in the Mediterranean region showed that the duration of cyano- HABs in some Greek lakes lasts up to eight months while algal blooms (which can lead to cyano-habs) occur throughout the year because of the favorable weather conditions for their growth [4]. A well-documented incident of cyanobacterial water contamination occurred in a dialysis clinic in Brazil in 1996, resulting in human fatalities [5]. The aforementioned incidences propelled research efforts on the detection and treatment of cyanotoxins. Most of the studies utilized a hepatotoxin from the group of microcystins, microcystin-lr (MC-LR) due to its high toxicity and frequent appearance in cyano- HABs [1,6-13]. Many technologies, including coagulation/sedimentation, activated carbon adsorption, and membrane separation, have been tested for the treatment of MC-LR [1,6-13]. Recently, titanium dioxide (TiO 2 ) photocatalysis, one of the most effective advanced oxidation technologies (AOTs), has demonstrated high decomposition and detoxification efficiency for cyanobacterial toxins [9-13]. Accordingly, extensive studies on the effect of TiO 2 loading, initial MC-LR concentration, light intensity and other chemical additives 147

161 on MC-LR degradation kinetics and reaction pathways have been conducted. However, in order to generate the reactive oxidizing species (hydroxyl radicals), which are responsible for the destruction of the toxins, the essential requirement in TiO 2 based AOTs is the use of ultraviolet (UV) light irradiation with high photon energy above the high band gap of TiO 2 (E g 3.2 ev) to photoexcite the catalyst. This inhibits the utilization of solar light as a sustainable energy source for TiO 2 activation because only 5% of the incoming solar energy on the earth s surface is in the UV range. Consequently, activation of TiO 2 under visible light can facilitate the development of promising processes for the remediation of contaminated water resources using solar light without complicated facilities for generating and introducing UV light. In order to utilize visible light for TiO 2 excitation, dye-sensitized or metal ion-doped TiO 2 has been developed and showed promising results for the degradation of perchlorinated compounds and nitrogen oxides [14,15]. Introduction of anionic dopants, especially nitrogen, to TiO 2 also makes it possible to achieve TiO 2 band gap narrowing [16-18]. In general, nitrogen-doped TiO 2 (N-TiO 2 ) is synthesized through two timeconsuming consecutive steps: synthesis of TiO 2 and then nitrogen doping of the TiO 2 from various nitrogen-containing chemicals (e.g., urea, ethylamine, gaseous nitrogen) at high temperature. In this approach, nitrogen atoms should be limited in the exterior of TiO 2 lattice [16]. In TiO 2 synthesis methods, one of the ways to increase nitrogen content in the TiO 2 lattice is by using titanium precursors combined with a nitrogen-containing ligand, such as Ti 4+ -bipyridine or Ti 4+ -amine complexes [19,20]. Herein, we developed a simple sol-gel method to synthesize visible light activated N-TiO 2 by employing a nitrogen-containing surfactant (dodecylammonium chloride, 148

162 DDAC) [21]. In our previous studies, surfactant templating strategies were introduced in the sol-gel synthesis of TiO 2 to tailor-design the structural properties of TiO 2 and to fabricate TiO 2 nanomaterials with unique reactivities and functionalities for environmental applications [22,23]. In this current study, we utilized DDAC surfactant as a pore templating material to tailor-design the structural properties of TiO 2 and as a nitrogen dopant to narrow its band gap. The synthesis of mesoporous TiO 2 and in-situ nitrogen-doping of the TiO 2 were concurrently achieved. The visible light activated mesoporous N-TiO 2 was evaluated for the photocatalytic destruction of MC-LR using visible and UV light. The synthesis route, physicochemical properties, and electronic structures of N-TiO 2 were investigated to clarify the mode and nature of action of N-TiO 2 under visible and UV light to degrade MC-LR. 5.2 Experimental Synthesis of Mesoporous N-TiO 2 DDAC aqueous solution was prepared by reacting a stoichiometric amount of dodeylamine (DDA, Aldrich) with HCl in water as solvent [24]. In such a method, the DDAC molecules self-assemble to form micelles [25]. The DDAC micellar solution was stirred vigorously, then titanium tetraisopropoxide (TTIP, Aldrich), a titanium alkoxide precursor dissolved in isopropanol (i-proh) was added into the DDAC solution. Hydrolysis and condensation reactions of TTIP occurred, forming a white turbid precipitate. The molar ratio of DDA:HCl:H 2 O:i-PrOH:TTIP was 0.5:0.5:100:30:1, where control TiO 2 was also synthesized without DDA. The precipitate was condensed during 149

163 drying at 20 C for a day, finally forming a white solid TiO 2 /organic composite. The composite was heat-treated in a furnace (Paragon HT-22-D, Thermcraft) to increase the crystallinity of TiO 2 and remove the DDAC surfactant template. The heat-treatment temperature was increased at a ramp rate of 60 ºC/h to 100 ºC, maintained at this temperature for 1 h. Then the temperature was increased again to 350 ºC or up to 550 ºC, maintained at this temperature for 2 h except for some, specifically 5 h for 350 ºC, and cooled down naturally. Finally yellowish white N-TiO 2 was obtained Properties and Characterization of N-TiO 2 A UV-Vis spectrophotometer (Shimadzu 2501 PC) mounted with an integrating sphere attachment (ISR1200) for diffuse reflectance measurement was used to investigate the optical band gap of TiO 2. X-ray diffraction (XRD) analysis of TiO 2 using a Kristalloflex D500 diffractometer (Siemens) was employed to study crystallographic properties of TiO 2. A Tristar 3000 (Micromeritics) porosimetry analyzer was used to determine the structural properties of TiO 2 including Brunauer, Emmett, and Teller (BET) surface area, pore volume, and Barrett-Joyner-Halenda (BJH) pore size and distribution, based on nitrogen adsorption and desorption isotherms. For the nano-level morphology of TiO 2, a JEM-2010F (JEOL) high resolution-transmission electron microscope (HR-TEM) was used after the samples were dispersed and fixed on a carbon-coated copper grid (LC200- Cu, EMS). The elemental composition and electronic structure of TiO 2 were investigated using energy dispersive X-ray spectroscope (EDX, Oxford Isis) and X-ray photoelectron spectroscope (XPS, Perkin-Elmer Model 5300). The particle size of agglomerated TiO 2 150

164 (APS) and its point of zero charge (PZC) were measured using a Zetasizer (Malvern Instruments) Photocatalytic Degradation of Microcystin-LR TiO 2 particles dispersed in water using a sonicator (2510R-DH, Bransonic) for 2 h were transferred to borosilicate reactors containing MC-LR (Calbiochem Cat #: ). The solution ph was adjusted at 3.0 without buffer by adding H 2 SO 4 solution. The reactors were completely sealed and stirred vigorously to reduce mass transfer limitation, and covered with aluminum foil to make dark condition for adsorption of MC-LR onto TiO 2 for 2 h. The reactors were then irradiated for up to 4 h by two 15 W low pressure mercury UV tubes (Spectronics) emitting near UV radiation with a peak at 365 nm or two 15 W fluorescent lamps (Cole-Parmer) mounted with UV block filter (UV420, Opticology) to cut spectral range below 420 nm. The experimental conditions were as follows: volume of solution = 10 ml, ph = 3.0 ± 0.1 (for a comparative study, ph = 5.7 ± 0.2 with super quality (SQ) water), MC-LR concentration = 5.0 ± 0.1 mg/l (5 μm), TiO 2 loading = 0.5 g/l, and temperature = 25 ± 3 C. Sample of 0.2 ml was taken and filtered with syringeless 0.45 μm glass microfiber vial (L815, Whatman) containing 0.2 ml of methanol as a quenching agent to stop further reaction of hydroxyl radicals generated. The samples were equally split in 0.2 ml inserts placed in vials, and analyzed with high performance liquid chromatography (HPLC, Agilent Series 1100) for the quantification of MC-LR and mass spectrometry (MS) for the determination of reaction intermediates. For the HPLC, the injection volume to a C-18 Discovery column (Supelco) at 40 C was 50 μl. The mobile phase in isocratic 151

165 mode with a flow rate of 1 ml/min was a mixture of 0.05% trifluoroacetic acid (TFA) in water and 0.05% TFA in acetonitrile at a 60:40 ratio. MC-LR was eluted at 5.4 min and measured with a photo-diode array detector at 238 nm. 5.3 Results and Discussion Optical Band Gap As shown in Fig. 5.1, the absorption spectrum shoulder of N-TiO 2 calcined at 350 C was significantly extended toward the visible light range, indicating a red shift effect of nitrogen doping. Instead of intrinsic wavelength (λ int g ), the linearly extrapolated wavelength was used as effective wavelength (λ eff g ) for the measurement of the effective optical band gap, E eff eff g [26]. The λ g of N-TiO 2 was approximately 468 nm (E eff g = 2.65 ev) while that of control TiO 2 was 420 nm (E eff g = 2.95 ev). As shown in Fig. 5.2, the eff E g of N-TiO 2 steeply increased from 2.65 ev to 2.80 ev upon heat treatment to 400 C eff and then close to 2.89 ev after further heat treatment, compared to the stable E g of eff control TiO 2 at around 2.91 ev. This decrease in E g of N-TiO 2 was caused by the loss of nitrogen atoms incorporated in TiO 2 lattice since nitrogen atoms tend to be replaced with oxygen atoms in the air at high temperature [27]. N-TiO 2, especially that prepared at 350 C, had nitrogen content up to around 1.7% while the control TiO 2 had negligible nitrogen content at below 0.2%, most probably originated from the impurities of the ingredients used. The decrease in the nitrogen content in N-TiO 2 over calcination eff temperature is in agreement with the increase in the E g of N-TiO 2. This implies the effect of nitrogen doping in TiO 2 on the visible light response of N-TiO

166 Absorbance (a.u.) int λ g eff λ g Absorbance (a.u.) Bandgap Narrowing N-TiO 2 TiO Photon energy (ev) N-TiO 2 Control TiO Wavelength (nm) Fig. 5.1: Optical UV-Visible absorption spectra of control TiO 2 and N-TiO 2 calcined at 350 C. (ev) Bandgap energy, E g eff Bandgap Control TiO 2 N-TiO 2 N-TiO 2 Control TiO 2 N content Nitrogen content (atomic %) Heat treatment temperature ( o C) Fig. 5.2: Effective band gap energy, E g eff and nitrogen content of control TiO 2 and N- TiO 2 prepared at different calcination temperatures (E g eff = /λ g eff ). Note 5 h calcination time for 350 C while 2 h for others. 153

167 5.3.2 Electronic Structure To exclude the effect of carbon in N-TiO 2 on visible light response, we traced and completely removed carbon in N-TiO 2 by increasing either the heat treatment temperature or the calcination time, as shown in Fig The electrons of nitrogen 2p states in TiO 2 are known to contribute to band gap narrowing by mixing with those of oxygen 2p states and thus N-TiO 2 not only reduces the photoexcitation energy but also facilitates the transport of photocarriers to the surface for photodecomposition [16]. However, the role of nitrogen in TiO 2 on its visible light response is still being debated, depending on the model applied for calculating the band structures and charge densities of N-TiO 2 [28,29]. Max: N(E) O 1s Ti 2p N 1s C 1s Binding energy (ev) Fig. 5.3: XPS spectrum of N-TiO 2 calcined at 350 C for 5 h. 154

168 (a) Ti 2p (b) O 1s (C) N 1s 2p 3/2 N-TiO 2 N-TiO 2 Intensity (a.u.) 2p 1/2 N-TiO 2 Intensity (a.u.) Intensity (a.u.) Control TiO Binding Energy (ev) Control TiO Binding Energy (ev) Control TiO Binding Energy (ev) Fig. 5.4: High resolution XPS of control TiO 2 and N-TiO 2 calcined at 350 C: (a) at ev for Ti 2p, (b) at ev for O 1s, and (c) at ev for N 1s. Fig. 5.4 shows the XPS spectra of control TiO 2 and N-TiO 2 prepared at 350 C for Ti 2p, O 1s and N 1s core levels. Ti 2p 3/2 core levels for control TiO 2 and N-TiO 2 are shown at ev and ev, respectively. The 0.9 ev shift in the binding energy of electrons towards lower energy indicates that the electronic interaction of Ti with anions in N-TiO 2 is different from that in control TiO 2 [28]. This implies that TiO 2 crystal lattices are modified with substitutional and/or interstitial N atoms (e.g., Ti N Ti, Ti O N Ti), where the electron density around Ti atoms increases since the tendency of nitrogen (i.e., Pauling s electronegativity 3.04) to attract the bonding electrons in a chemical bond toward itself is lower than that of oxygen (i.e., 3.44). Oxygen 1s core level appears at ev for both samples, suggesting substitutional N atoms in TiO 2 lattice (i.e., Ti N Ti). However, a little bit wide shoulder at around 532 ev in the spectrum of N-TiO 2 indicates the possibility of different oxygen environment, Ti O N Ti (interstitial 155

169 N in TiO 2 lattices) than TiO 2 (Ti O Ti). N 1s core level for N-TiO 2 shows a single peak at around ev. Nitrogen from chemisorbed N or TiN appears at below 398 ev, nitrogen from NO and NO 2 appears above 400 ev, and nitrogen in Ti O N Ti environment shows higher binding energy above 400 ev [30]. As a result, even though the presence of interstitial N atoms in TiO 2 lattices cannot be ruled out, the lower binding energy of N-TiO 2 is likely to result from anion (N - ) like species. This leads to N 2p states on the top of the valence band of TiO 2 and narrows the overall band gap of TiO 2. (a) Z Y X Counts (b) X O Ti Y Z N 5060 nm nm C 100 Position (nm) Position (nm) 150 Fig. 5.5: (a) scanning-tem image and (b) line analysis results for Ti, O, N, and C elemental mapping of N-TiO 2 particles calcined at 350 C. As shown in Fig. 5.5 for the elemental mapping of Ti, O, N, and C in scanning- TEM mode, the N-TiO 2 was composed of mainly titanium and oxygen and partially nitrogen and carbon. Even though DDA consists of mainly carbon (77.7%) and partially 156

170 nitrogen (7.5%), nitrogen signal was quite stronger than carbon signal that was originated most probably from the TEM grid with carbon mashes. The preferential nitrogen doping to TiO 2 over carbon from DDAC is due to the formation of a chemical bond between N in the amine group of DDAC and the Ti metal center in TTIP (i.e., formation of N- containing Ti precursors) during sol-gel synthesis of TiO 2 [20]. Table 5.1: Crystallographic and physicochemical properties of control TiO 2 and N-TiO 2 prepared at different calcination conditions. Sample Calcination Crystallographic b Physicochemical b Temp. ( C) a Phase CS CS% D (101) SA PV PV% PS APS PZC (nm) (%) (Å) (m 2 /g) (cm 3 /g) (%) (nm) (nm) TiO Anatase Anatase Anatase N-TiO Anatase Anatase Anatase Anatase Anatase a Note 5 h calcination time for 350 C while 2 h for others. b CS: crystallite size, CS%: Crystallinity, D (101) : d space in (101) plane, SA: BET surface area, PV: pore volume, PV%: porosity, PS: BJH average pore size from adsorption isotherm branch, APS: agglomerated particle size, and PZC: point of zero charge Crystallographic and Physicochemical Properties Some important properties of TiO 2 are summarized in Table 5.1. The crystal phase of all TiO 2 samples was anatase. In general, high crystallinity and small crystallite size are desirable for high catalytic activity of TiO 2. During calcination, materials crystallinity 157

171 increased while crystallite size also increased. As shown in Fig. 5.6, a clear peak broadening at (101) plane for N-TiO 2 suggested small size N-TiO 2 crystallites. Fig. 5.7 indicates that N-TiO 2 has smaller (101) plane d space compared to control TiO 2. Due to higher atomic radius of N atom (0.65 Å) compared to O atom (0.60 Å), substitutional N for O in TiO 2 lattice results in a decrease in the inter-planar distance. Once heat treatment temperature increased to 550 C, d space was restored to Å, typical d space for (101) of pure anatase due to the loss of N. N-TiO 2 Intensity (a.u.) A R A A AA 500 o C 450 o C 400 o C 350 o C Control TiO o C 450 o C 400 o C 350 o C θ (degrees) Fig. 5.6: XRD patterns of control TiO 2 and N-TiO 2 prepared at different calcination temperatures (A: anatase and R: rutile). 158

172 3.54 d space (Angstrom) Control TiO 2 N-TiO Heat treatment temperature ( o C) Fig. 5.7: (101) plane d space for control TiO 2 and N-TiO 2 prepared at different calcination temperatures. Note 5 h calcination time for 350 C while 2 h for others. N 2 adsorption-desorption isotherms of N-TiO 2 particles, shown in Fig. 5.8, were type IV typical for mesoporous materials. The BJH pore size distribution of N-TiO 2 calcined at 350 C was narrow ranging from 2 to 8 nm. Its BET surface area and pore volume are significantly high at 150 m 2 /g and cm 3 /g, respectively. During calcination, the overall structural properties, except for crystallinity, were deteriorated due to the collapse of initial porous structure and growth of crystallites. However, the structural properties of N-TiO 2 are much better than those of control TiO 2, suggesting that the DDAC surfactant effectively acted as a pore template. As shown in Fig. 5.9, nanocrystalline N-TiO 2 clusters of average size of 152 nm exhibited highly porous interconnected inorganic network, yielding high porosity of 44.0%, compared to dense control TiO 2 clusters with low porosity of 25.1% and surface area of 72.5 m 2 /g and large aggregate size of 226 nm. 159

173 Pore volume (cm 3 /g) (b) Volume adsorbed (cm 3 /g) (a) N-TiO 2 40 Control TiO Relative pressure (P s /P o ) Pore diameter (nm) Fig. 5.8: (a) N 2 adsorption-desorption isotherms and (b) pore size distribution of control TiO 2 and N-TiO 2 prepared at 350 C. Fig. 5.9: HR-TEM morphology of (a) control TiO 2 and (b) N-TiO 2 : (1) as-synthesized and (2) and (3) calcined at 350 C. 160

174 5.3.4 Formation of Mesoporous N-TiO 2 The critical micelle concentration of DDAC aquatic solution at ph 4 7 is around 10-2 mol/l, above which the DDAC has been reported to form well-defined self-assembled structure [25,31]. In this study, the DDAC concentration in water is around mol/l. When the liquid molecular precursor of titanium is added to DDAC solution, TTIP is hydrolyzed and condensed around the self-assembled surfactants, forming a surfactant organic core/tio 2 inorganic shell composite, as demonstrated in Scheme 5.1. The effect of DDAC in the TiO 2 network is clearly seen even before calcination (note Fig. 5.9). The inorganic network of control TiO 2 is continuous phase while that of the composite exhibits much less condensed phase. During thermal treatment, the surfactant templates are removed, leaving a porous structure. In addition, nitrogen atoms in DDAC surfactant are diffused and incorporated into the crystal lattice of TiO 2 as either Ti N Ti or Ti N O Ti environment. O Ti N O Ti Ti O O O Ti N Ti-O-Ti- N O Ti N O Ti O N O Ti N -Ti-O-Ti Ti O O Ti Ti-O-Ti- N N O Ti N Void N O Ti O O Ti N N -Ti-O-Ti O Ti O N Ti O O Ti N Ti O N O O O Ti O Ti O Ti N N O Ti O O Ti O N Ti O O O O Ti O Ti N Scheme 5.1: Incorporation of Ti O Ti network onto self-organized DDAC surfactant micelles to form an organic core/inorganic shell composite, followed by the removal of the organic templates to form N-TiO 2 with mesoporous structure. 161

175 Due to the formation of nitrogen-containing Ti precursors as discussed previously, nitrogen atoms are already incorporated with the TiO 2 nuclei consisting of several unit cells during sol-gel synthesis of TiO 2 [20]. When the nuclei crystallize and grow during calcination, nitrogen atoms can easily diffuse into TiO 2 lattices in the whole grain. This even distribution of N in TiO 2 was proved by tracking the amount of nitrogen in N-TiO 2 prepared at 350 C using XPS depth profiling. Nitrogen content at X-ray incident angles at 15, 45, and 75, corresponding to surface depth of approximately 2-3, 4-5, and 6-7 nm, was relatively stable at 1.88±0.35, 1.81±0.26, and 1.51±0.31%, respectively Photocatalytic Destruction of MC-LR under Visible Light As shown in Fig. 5.10, no photolysis of MC-LR under visible light in the absence of TiO 2 was observed. The dark adsorption of MC-LR with molecular size of 2-3 nm on 2-8 nm mesoporous N-TiO 2 occurred immediately and reached equilibrium within 2 h [1,6]. Afterwards, MC-LR was effectively destroyed under visible light above 420 nm using N- TiO 2 with narrowed band gap energy, E eff g = 2.65 ev (468 nm). Acidic ph condition was favorable for both the adsorption and photocatalytic degradation of MC-LR. Based on the surface chemistry of MC-LR and TiO 2 at ph 3.0, the surface of MC-LR is overall negatively charged by the deprotonation of its free carboxylic groups while that of N- TiO 2 with PZC of 6.2 is positively charged at +6.5 mv, resulting in electrostatic attraction forces between MC-LR and N-TiO 2 [7,10]. On the other hand, at ph 5.7, adsorption of negatively charged MC-LR onto relatively neutral N-TiO 2 (+0.9 mv) was negligible. The difference in the degradation kinetics at the ph conditions is obvious, when considering that the first step of photocatalytic oxidation is adsorption of MC-LR 162

176 on TiO 2 surface since the lifetime of hydroxyl radicals is very short and thus they react with mainly adsorbed species. The degradation kinetics followed pseudo first order reaction and the apparent reaction rate constant, k at ph 3.0 (k=0.026 min -1 ) was 2.5 times higher than that at ph 5.7 (k=0.010 min -1 ). Concentration of MC-LR, C/C o Dark adsorption Visible light > 420 nm 1.0 (Photolysis without TiO 2 ) 0.8 ph k=0.010 min ph k=0.026 min Reaction Time (hr) Fig. 5.10: Adsorption followed by photocatalytic degradation of MC-LR using N-TiO 2 prepared at 350 C under visible light (>420 nm) at different ph conditions. The summarized results of MC-LR degradation under visible light are shown in Fig The concentration of MC-LR, after 2 h dark adsorption where approximately 10-15% of MC-LR disappeared, was reset as its initial concentration for measuring the photocatalytic degradation efficiency. MC-LR degradation under visible light well proved the obvious role of N doping, band gap narrowing and thus visible light response of TiO 2. MC-LR degradation using control TiO 2 and even P25 (well-defined 30 nm TiO 2 163

177 nanoparticles, Degussa, Germany), which is a benchmark TiO 2 photocatalyst with high activity, was negligible because of their higher E eff g than the photon energy provided by the visible light. As expected from its E eff g shown in Fig. 5.2, N-TiO 2 calcined at 350 C could destroy 50% of MC-LR within 30 min and almost completely within 2 h and the other N-TiO 2 photocatalysts also exhibited visible light activation for the degradation of MC-LR. However, a further increase in calcination temperature resulted in a significant decrease in MC-LR degradation efficiency of N-TiO 2. As a result, the activity of TiO 2 under visible light is in agreement with mainly its nitrogen content demonstrated in Fig. 5.2, rather than any other properties including crystallinity and surface area. MC-LR Degradation Efficiency (%) Control TiO 2 P-25 Control 350 o C TiO 2 Sample N-TiO o C 30 min 120 min 500 o C Fig. 5.11: MC-LR degradation efficiency of control TiO 2 and N-TiO 2 at ph 3.0 after visible light (>420 nm) irradiation for 30 and 120 min. The initial concentration of MC- LR was reset as its concentration after 2 h adsorption equilibrium in dark condition. Control in the X-axis represents all control TiO 2 samples calcined at 350, 400, and 500 C. 164

178 Concentration of MC-LR, C/C N-TiO 2 Control TiO o C k=0.088 min o C k=0.064min Reaction time (min) 350 o C k=0.018 min o C k=0.025 min -1 Fig. 5.12: MC-LR degradation by control TiO 2 and N-TiO 2 at ph 3.0 under UV-365 nm irradiation. The initial concentration of MC-LR was reset as its concentration after 2 h adsorption equilibrium in dark condition Photocatalytic Destruction of MC-LR under UV Light No photolysis of MC-LR even under UV-365 nm irradiation was observed since the maximum light absorbance of MC-LR is at 238 nm. As shown in Fig. 5.12, the advantage of utilizing mesoporous N-TiO 2 with high surface area in degrading MC-LR under UV light was studied in comparison with nonporous control TiO 2. As expected, both control TiO 2 and N-TiO 2 effectively destroyed MC-LR and the MC-LR degradation kinetics under UV light was much faster than under visible light. In case of N-TiO 2, MC-LR was almost completely decomposed within 30 min. The apparent first order reaction rate constants of MC-LR using N-TiO 2 prepared at 350 C and 400 C were min -1 and min -1, respectively, which are 3~4 times higher than those using the control TiO 2. As demonstrated in Table 5.1, this is due to much enhanced structural properties of N- 165

179 TiO 2 (i.e., high surface are, small crystallite size, less agglomeration). Based on the results, the activity of TiO 2 under UV light seemed to be dependent on mainly its structural properties since no significant difference in UV-365 nm light absorbance between N-TiO 2 and control TiO 2 was observed in Fig Increase in the MC-LR decomposition rate by TiO 2 calcined at elevated temperature can be ascribed to the slight increase in crystallinity from 82% to 85% for control TiO 2 and from 83% to 86% for N- TiO Conclusions Titanium dioxide (TiO 2 ), that can be photoexcited under visible light irradiation, is of great interest for the design of solar-driven water treatment technologies. This study demonstrated the development of highly efficient visible light-activated nitrogen doped TiO 2 (N-TiO 2 ) photocatalysts and their environmental applications in destroying an emerging water contaminant, the cyanobacterial toxin, microcystin-lr (MC-LR) under visible light. The nanotechnological sol-gel method employing nitrogen-containing surfactant (dodecylammonium chloride) as a pore templating material for tailor-designing the structural properties of TiO 2 and as a nitrogen dopant for its visible light response was promising. Due to its enhanced structural properties and nitrogen doping, N-TiO 2 efficiently degraded MC-LR under visible spectrum above 420 nm and even under UV light, N-TiO 2 decomposed MC-LR 3~4 times faster than control TiO 2. The degradation pathways and reaction intermediates of MC-LR were not directly related with energy source for TiO 2 activation (UV and visible) and nature of TiO 2 (neat and nitrogen-doped). 166

180 For the design of solar-driven treatment technologies, this study implies a strong possibility for the in-situ photocatalytic remediation of contaminated water with cyanobacterial toxins and other toxic organic compounds using solar light, a sustainable source of energy. 5.5 References [1] M.G. Antoniou, A.A. de la Cruz, D.D. Dionysiou, J. Environ. Eng. 131 (2005) [2] R.J. Andersen, H.A Luu, D.Z.X Chen, C.F.B. Holmes, M.L. Kent, M. le Blanc, F.J.R. Taylor, D.E. Williams, Toxicon 31 (1993) [3] Z.A. Mohamed, W.W. Carmichael, A.A. Hussein, Environ. Toxicol. 18 (2003) 137. [4] C.M. Cook, E. Vardaka, T. Lanaras, Acta Hydrochim. Hydrobiol. 32 (2004) 107. [5] S. Pouria, A. de Andrade, J. Barbosa, R.L. Cavalcanti, V.T.S. Barreto, C.J. Ward, W. Preiser, G.K. Poon, G.H. Neild, G.A. Codd, Lancet 352 (1998) 21. [6] L.A. Lawton, P.K.J. Robertson, Chem. Soc. Rev. 28 (1999) 217. [7] W. Song, A.A. de la Cruz, K. Rein, K.E. O shea, Environ. Sci. Technol. 40 (2006) [8] J. Lee, H.W. Walker, Environ. Sci. Technol. 40 (2006) [9] L.A. Lawton, P.K.J. Robertson, B.J.P.A. Cornish, M. Jaspers, Environ. Sci. Technol. 1999, 33, 771. [10] A.J. Feitz, D.T. Waite, G.J. Jones, B.H. Boyfen, P.T. Orr, Environ. Sci. Technol. 33 (1999) 243. [11] I. Lui, L.A. Lawton, P.K.J. Robertson, Environ. Sci. Technol. 37 (2003)

181 [12] L.A. Lawton, P.K.J. Robertson, B.J.P.A. Cornish, I.L. Marr, M. Jaspars, J. Catal. 213 (2003) 109. [13] B. Yuan, Y. Li, X. Huang, H. Liu, J. Qu, J. Photochem. Photobiol. A 178 (2006) 106. [14] E. Bae, W. Choi, Environ. Sci. Technol. 37 (2003) 147. [15] Y.-M. Lin, Y.-H. Tseng, J.-H. Huang, C.C. Chao, C.-C. Chen, I. Wang, Environ. Sci. Technol. 40 (2006) [16] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Science 293 (2001) 269. [17] H. Irie, Y. Wanatabe, K. Hashimoto, J. Phys. Chem. B 107 (2003) [18] C. Burda, Y. Lou, X. Chen, A.C.S. Samia, J. Stout, J.L. Gole, Nano Lett. 3 (2003) [19] T. Sano, N. Negishi, K. Koike, K. Takeuchi, S. Matsuzawa, J. Mater. Chem. 14 (2004) 380. [20] C. Belver, R. Bellod, A. Fuerte, M. Fernandez-Garcia, Appl. Catal. B 65 (2006) 301. [21] H. Choi, M.G. Antoniou, M. Pelaez, A.A. de la Cruz, J.A. Shoemaker, D.D. Dionysiou. Environ. Sci. Technol. (Submitted for Publication). [22] H. Choi, A.C. Sofranko, D.D. Dionysiou, Adv. Funct. Mater. 16 (2006) [23] H. Choi, E. Stathatos, D.D. Dionysiou, Appl. Catal. B 63 (2006) 60. [24] W. Loh, L.A.C. Teixeira, L.-T. Lee, J. Phys. Chem. B 108 (2004) [25] Q. Dai, J.S. Laskowski, Langmuir 7 (1991) [26] B. Sun, P.G. Smirniotis, P. Boolchand, Langmuir 21 (2005)

182 [27] M. Sathish, B. Viswanathan, R.P. Viswanath, S.G. Chinnakonda, Chem. Mater. 17 (2005) [28] N.C. Saha, H.G. Tompkins, J. Appl. Phys. 72 (1992) [29] T.-H. Xu, C.-L. Song, Y. Liu, G.-R. Han, J. Zhejiang Univ. Sci. 7 (2006) 299. [30] C.D. Wagner, W.M. Riggs, L.E. Davis, J.F. Moulder, G.E. Muilenberg, Handbook of X-Ray Photoelectron Spectroscopy; Perkin-Elmer Corporation: Eden Prairie, MN, [31] B.S. Aksoy, Hydrophobic Forces in Free Thin Films of Water in the Presence and Absence of Surfactants; Ph.D. Thesis: Virginia Polytechnic Institute and State University,

183 CHAPTER 6 Voltammetric Determination of Catechol Using a Sonogel Carbon Electrode Modified with Nanostructured TiO 2 170

184 Recently, electrochemical detection of neurotransmitters using smart sensors has attracted neuroscientists attention since their altered levels have been associated with mental and behavioral disorders. The deliberate chemical modification of electrode surface with a suitable reagent has been attempted to control the rates and selectivities of electrochemical reactions at the solid/liquid interface due to the real challenges of achieving successful analysis of clinical and environmental samples in the presence of common interferents. In this study, we investigate highly efficient sonogel carbon electrode (SGC/TiO 2 ) modified with nanostructured titanium dioxide synthesized via solgel method employing surfactant template for tailor-designing the structural properties of TiO 2. The stable SGC/TiO 2 electrode detects catechol, a neurotransmitter, in the presence of ascorbic acid, a common interferent, using cyclic voltammetry. A possible rationale for the stable catechol detection of SGC/TiO 2 electrode is attributed to most likely the adsorption of catechol onto highly porous TiO 2 (surface area of 147 m 2 g -1 and porosity of 46.2%), and the formation of C 6 H 4 (OTi) 2 bond between catechol and TiO 2. The catechol absorbed onto TiO 2 rapidly reaches the SGC surface, then is oxidized, involving two electrons (e - ) and two protons (H + ). As a result, the surface of TiO 2 acts as an electron-transfer accelerator between the SGC electrode and catechol. In addition to the quantitative and qualitative detection of catechol, the SGC/TiO 2 electrode developed here meets the profitable features of electrode including mechanical stability, physical rigidity, and enhanced catalytic properties. We anticipate this new modified sonogel carbon electrode with nanostructured titanium dioxide could be used in several chemical sensor applications for environmental and biological molecules of interest. 171

185 6.1 Introduction Recently, electrochemical detection of neurotransmitters using smart sensors has attracted neuroscientists attention since the altered levels of such chemicals have been associated with mental and behavioral disorders such as schizophrenia, attention deficient disorder, Alzheimer s disease, Parkinson s disease, eating disorders, epilepsy, amphetamine addiction, and cocaine addiction [1]. One of the major challenges in assessing their environmental fate and distribution in the human body is to measure them in-situ in a more effective way. As a result, there is an urgent need to develop innovative sensors to detect catechol and its derivatives, a class of neurotransmitters (1,2 dihydroxybenzenes) [2]. In synthesizing inorganic materials with reactivity and functionality, sol-gel process is among the most promising methods. Sol-gel method refers to the formation of solid inorganic material from its liquid molecular precursors through room temperature wet chemistry-based procedures [3-7]. The application of sol-gel chemistry in fabricating carbon-based electrode has attracted great attention in developing a new type of sensors [8 10]. Recently, solid graphite composite electrode was synthesized via sonocatalysis, where high energy ultrasonic cavitation is applied directly to the silica alkoxide precursors for their prompt hydrolysis [11]. Furthermore, the deliberate chemical modification of the electrode surface with a suitable reagent results in the control of the rates and selectivities of electrochemical reactions at the solid/liquid interface [12,13]. One of popular approaches for the chemical modification is to use electroactive polymeric films onto the electrode surface [14 16]. Chitosan-modifed electrode has been 172

186 also reported to be effective for the detection of phenolic compounds [17]. However, due to the real challenges of achieving successful analysis of clinical and environmental samples in the presence of common interferents, it is believed that the properties of the electrode, such as mechanical stability, physical rigidity, surface renewability, and selectivity, should be significantly improved before the electrode can become competitive for full-scale applications in developing such sensors. In this study, we have synthesized a sonogel-carbon electrode modified with highly active nanostructured titanium dioxide (denoted as SGC/TiO 2 ) as a new class of sensors [18]. The synthesis of nanostructured TiO 2 is of great interest because of its attractive optical, electrical, chemical and catalytic properties [19,20]. Fabrication of well-defined nanostructured TiO 2 or other inorganic oxides has been achieved via sol-gel methods employing self-assembly of surfactant molecules as pore templates in the inorganic networks with high surface area [21 23]. Surfactants added in the TiO 2 sol can self-assemble into micelles, which can incorporate the titanium alkoxide around the micellar corona, forming surfactant organic core/tio 2 inorganic shell composites. After heat treatment to remove the surfactant templates, the final TiO 2 has continuous phase porous inorganic network and tunable pore size similar to the micellar size. Applying such a nanotechnological approach, we can fabricate TiO 2 material with tailor-designed porous structure, which can enhance adsorption of molecules of interest at the TiO 2 surface. The nanostructured TiO 2 shows high adsorption capacity and catalytic activity toward organic molecules [21]. In addition to new nanotechnological and new materials chemistry procedures for the synthesis of SGC/TiO 2 electrode, this paper investigates the detection of catechol in 173

187 the presence of ascorbic acid, a common interferent, with the SGC/TiO 2 electrode using cyclic voltammetry (CV), where the oxidation peaks for catechol and ascorbic acid are too close to allow their resolution [24,25]. A possible rationale for the stable catechol detection of SGC/TiO 2 electrode and the role of TiO 2 in facilitating adsorption of catechol and thus in accelerating charge transfer between catechol and SGC electrode are discussed. The SGC/TiO 2 electrode developed here meets the profitable features of electrode including the quantitative and qualitative detection of catechol, mechanical stability, physical rigidity, and enhanced catalytic properties. 6.2 Experimental Sono-Gel Carbon Electrode For the fabrication of sonogel-carbon (SGC) electrode, 1.5 ml of methyltrimethoxysilane (MTMOS, Fluka) was added into 0.3 ml of 0.2 M HCl solution [11]. This mixture was sonicated in an ultrasonicator (2510R-DH, Bransonic) for 15 sec. Then, 3 g of graphite carbon powder (Alfa Aesar, %) was added into the MTMOS solution and mixed thoroughly for 10 min, where the total volume of the reactants was significantly reduced up to 20%, condensing the SGC. A 0.25 mm copper wire (Alfa Aesar) was installed inside 0.69 mm I.D. capillary glass tube (Sutter Instrument) used as the bodies of the SGC electrode. The glass tube was filled with the SGC, and dried at 40 C for 24 h. Finally, the tip of the SGC electrode was polished with a fine sand paper, followed by wiped with a soft tissue. Adherence between copper wire and SGC was stable at more than 180 g/cm of tensile strength. 174

188 6.2.2 Modification with Nanostructured TiO 2 The TiO 2 was synthesized via sol-gel method. In order to control the structural properties of TiO 2 at the nano-level, the approach introduced in this study involved the use of a surfactant for the precise orchestration of the titanium precursor in the sol and the synthesis of the final TiO 2 material with high porosity [21-23]. Polyoxyethylenesorbitan monooleate (Tween 80, Aldrich) surfactant was selected as a pore directing agent in TiO 2 sol [21]. A suitable amount of Tween 80 was homogeneously dissolved in isopropanol (iproh, Fisher). Acetic acid (Fisher) was added into the solution for the esterification reaction with iproh to generate water [26]. When adding titanium tetraisopropoxide (TTIP, Aldrich) as TiO 2 precursor into the solution, hydrolysis and condensation reactions of TTIP occurred, forming a stable TiO 2 sol. The molar ratio of the ingredients was optimized at Tween 80:iPrOH:acetic acid:ttip = 1:45:6:1. The tip of SGC electrode was dipped into the TiO 2 sol for 3 sec and taken out. After coating, the SGC/TiO 2 electrode was dried at room temperature for 1 h and calcined in a programmable furnace (Paragon HT-22-D, Thermcraft) to remove the surfactant templates and obtain a desirable crystal phase of TiO 2. The temperature was increased at a ramp rate of 3 ºC min -1 to 500 ºC, maintained at this temperature for 20 min, and cooled down naturally. Due to the difficulty to directly characterize the properties of TiO 2 materials on the SGC electrode, easy-to-remove TiO 2 coating was prepared on borosilicate glass (Micro slide), assuming that if the substrate changes, the porous structure and crystal phase of TiO 2 reported here are similar since those properties originate from the surfactant addition and heat treatment rather than the substrate 175

189 properties [27]. For comparison, control SGC electrode without TiO 2 was synthesized and SGC electrode modified with conducting polymer, poly(3-methylthiophene) (denoted as P3MT) was also fabricated, based on the method described elsewhere [1,28] Characterization of SGC/TiO 2 In order to determine the crystallographic structure of TiO 2, X-ray diffraction (XRD) analysis using a Kristalloflex D500 diffractometer (Siemens) with Cu Kα (λ = Å) radiation was employed. A porosimetry analyzer (Tristar 3000, Micromeritics) was used to determine structural characteristics of TiO 2 including Brunauer, Emmett, and Teller (BET) specific surface area, porosity, and pore size distribution in the mesoporous range, using nitrogen adsorption and desorption isotherms. The structure of TiO 2 materials at the nano-level was visualized using a JEM-2010F (JEOL) high resolution-transmission electron microscope (HR-TEM) with a field emission gun at 200 KV. The TiO 2 sample scratched from the easy-to-remove TiO 2 coating on glass substrate was dispersed in methanol and fixed on a carbon-coated copper grid (LC200-Cu, Electron Microscopy Sciences). An environmental scanning electron microscope (ESEM, Philips XL 30 ESEM-FEG) was used to investigate the electrode surface at the micro-level. An elemental composition analysis of TiO 2 was performed using an energy dispersive X-ray spectroscope (EDAX, Oxford Isis) connected to the HR-TEM and ESEM Detection of Catechol Electrochemical measurement of catechol was carried out with an Electrochemical Workstation (Epsilon, Bioanalytical Systems), based on CV employing three electrodes: 176

190 Pt auxiliary electrode (Bioanalytical Systems), Ag/AgCl reference electrode (Bioanalytical Systems), and SGC/TiO 2 working electrode developed here. Catechol of 5 mm solution (C 6 H 4 (OH) 2, Fluka) as a target compound to detect and ascorbic acid of 5 mm solution (C 6 H 8 O 6, Aldrich) as an interferent compound to resolve were prepared in 10 mm sulfuric acid (Aldrich) at ph 1.7 with deionized water. The scan rate of CV was 100 mv s -1. In order to investigate the relation between current in CV and catechol concentration, catechol concentration was varied from 0.1 to 1.0 mm. The chemical bondings between TiO 2 and catechol during the experiment were monitored using Fourier transform infrared spectroscope (FTIR, Perkin Elmer 1610). In comparison to the conditions above, i) 1 mm dopamine (C 8 H 11 NO 2 HCL, Aldrich), another neurotransmitter in 0.1 M phosphate buffer with 0.1 M NaCl at ph 7.4, and ii) 10 mm catechol and 10 mm acetaminophen (CH 3 CONHC 6 H 4 OH, Aldrich), another common interferent in 0.1 M phosphate buffer and 0.1 M NaCl solution at ph 7.4 were tested. 6.3 Results and Discussion Catechol Detection in the Presence of Ascorbic Acid Cyclic voltammograms shown in Fig. 6.1(a) illustrate the selectivity principle in the electrochemical detection of catechol in the presence of ascorbic acid, a common interferent. Ascorbic acid exhibits irreversible behavior while catechol does reversible behavior. The oxidation peaks of catechol at (i) E pa = 518 mv (anodic potential) and ascorbic acid at (iii) E pa = 246 mv are exceptionally resolved and the reduction peak of catechol at (ii) E pc = 445 mv (cathodic potential) is detected with the SGC/TiO 2 electrode. 177

191 Current, I (ma) (a) SGC SGC/TiO 2 (ii) E pc, catechol (iii) E pa, ascorbic acid (i) E pa, catechol Potential, E (mv) 0 Current, i (ma) (b) E pc, dopamine E pa, dopamine Current, i (ma) (c) E pc, acetaminophen E pa, catechol E pc, catechol E pa, acetaminophen Potential, E (mv) Potential, E (mv) 0 Fig. 6.1: Cyclic voltammograms of (a) 5 mm catechol and 5 mm ascorbic acid in 10 mm sulfuric acid (ph 1.7), (b) 1 mm dopamine in 0.1 M phosphate buffer M NaCl (ph 7.4), and (c) 10 mm catechol and 10 mm acetaminophen in 0.1 M phosphate buffer M NaCl solution (ph=7.4). In (a), line in the middle is for bare sonogel-carbon electrode (SGC) and line in the outer is for titanium dioxide-modified sonogel carbon electrode (SGC/TiO 2 ) developed in this study. 178

192 However, the bare SGC electrode does not respond to the presence of catechol and ascorbic acid, making flat line at current of around zero. These results imply the presence of TiO 2 onto SGC electrode surface is crucial to detect catechol electrochemically. Another neurotransmitter, dopamine at a neutral ph 7.4 was also detected with the SGC/TiO 2 electrode, as shown in Fig. 6.1(b). Detection of neurotransmitters can be inhibited by the presence of common inteferents such as acetaminophen and ascorbic acid since they are also oxidized at around the same potential as the target compound catechol. However, with the SGC/TiO 2 electrode, catechol and acetaminophen at ph 7.4 were simultaneously detected at different positions, as demonstrated in Fig. 6.1(c) Comparison with Conducting Polymer-Modified Carbon Electrode Typically, conducting polymers such as P3MT are an alternative to the detection of catechol in the presence of common interferents [1, 28]. The P3MT electrode improves the reversibility of catechol oxidation and thus detects catechol and ascorbic acid simultaneously. However the problem with the P3MT electrode like other modified electrodes is poor reproducibility, mechanical instability and electrode fouling. CV signal stability of the SGC/TiO 2 electrode is compared with that of the P3MT electrode during over 25 scans to detect catechol, as shown in Fig CV signal for the P3MT electrode started to be more and more diverged from the initial, which indicates P3MT conducting polymer comes off from the electrode surface and thus the CV signal is unstable. On the other hand, the SGC/TiO 2 shows good stability and reproducibility for the detection of catechol. In addition, the SGC/TiO 2 electrode has a more positive potential shift to detect 179

193 catechol in CV via oxidation, allowing better resolution for the ascorbic acid peak and thus simultaneous analysis without prior separation of ascorbic acid. 0.2 P3MT SGC/TiO 2 Current, I (ma) Potential, E (mv) 0 Fig. 6.2: Cyclic voltammograms of 5 mm catechol in 10 mm sulfuric acid during 25 scans. Dotted line is for P3MT electrode and sold line is for SGC/TiO 2 electrode Properties of SGC/TiO 2 Electrodes The SGC/TiO 2 electrode exhibits so far improved electrocatalysis and selectivity towards catechol compared to the neat SGC and P3MT modified electrodes, indicating that the TiO 2 in the electrode played a crucial role in detecting catechol electrochemically. In order to investigate what properties of TiO 2 affected the stable detection of catechol, various materials characterization techniques are introduced to the SGC/TiO 2 electrode. Fig. 6.3 shows ESEM images of the surface of the SGC/TiO 2 electrode. The tip of the electrode was packed well with condensed graphite powder due to the chemical bonding 180

194 of carbon in the graphite and silicon in MTMOS. Fig. 6.3(b) shows the smooth surface of the SGC electrode incorporated with nanostructured TiO 2. There were no serious microcracks and defect structures. Even though it was difficult to identify TiO 2 inorganic spots in the ESEM images as expected, it is believed that TiO 2 is uniformly coated and distributed on the graphite powders present at the tip of the SGC electrode since EDX analysis at various spots showed the relatively similar elemental composition for Ti. The representative EDX elemental composition at the tip of the SGC/TiO 2 electrode is shown in Fig Carbon, silicon, titanium, and oxygen were identified as major elements, and their sources were obviously graphite powder, MTMOS, TTIP, and oxides and surface hydroxyl groups, respectively. Fig. 6.3: ESEM images of SGC/TiO 2 electrode: (a) the tip (scale bar = 500 μm) and (b) the surface of the area highlighted in (a) (scale bar = 100 μm). 181

195 C O Eleme Wt At% CK OK Si K Ti K Si Ti Fig. 6.4: Elemental analysis of SGC/TiO 2 electrode using EDX connected to ESEM. XRD analysis showed that the TiO 2 heat-treated at 500 C is active anatase crystal phase with crystallite size of approximately 9 nm. It should be noted that the nanosize TiO 2 particles exhibit good mechanical stability on substrate and resistance to abrasion [29]. In order to investigate structural properties of TiO 2, collected powder from the easyto-remove TiO 2 on glass substrate was analyzed using HR-TEM and porosimetry analyzer. Fig. 6.5 shows the morphology of the nanostructured TiO 2. The TiO 2 was highly porous and a distinct pore structure was observed with average diameter of 5 nm. The synthesized films had bicontinuous structure with highly interconnected network, indicating the surfactant template used in this synthesis method effectively acted as a pore directing agent [27,30]. The mesoporous structure is important for the accessibility of the reactants to or from the active sites in TiO 2 [21]. Nitrogen adsorption-desorption isotherms were typical IV type, characteristic of well-developed mesoporous materials. The pore size distribution was significantly narrow, ranging from 2 to 8 nm. The BET 182

196 surface and porosity were significantly high at 140 m 2 g -1 and 45 %, respectively. The porous structure was reported to exhibit high adsorption ability for the reactants and following enhanced chemical reaction [21]. Fig. 6.5: HR-TEM image of TiO 2 with mesoporous structure, taken from TiO 2 particles scrapped from the SGC/TiO 2 electrode Detection Route and Mechanisms A possible rationale for the stable catechol detection of SGC/TiO 2 electrode is attributed to most likely the adsorption of catechol onto highly porous TiO 2 and the formation of bonding between catechol and SGC/TiO 2 electrode [31-33]. The important IR peaks and their designation are summarized in Table 6.1 and compared with findings by Martin et al. [31]. The main bands and their assignment are as follows: 1491 cm -1, stretching (C C); 183

197 1451 cm -1, stretching (C=C); 1258 cm -1, stretching (C O); 1208 and 1095 cm -1, bending (C H). The 1198 cm -1 OH wag is absent for the doubly deprotonated species and the 1353 cm -1 in-plane OH bending is not seen (only a weak and broad feature centered at 1208 cm -1 was seen). As a result, a possible mechanism for the efficient detection catechol on SGC/TiO 2 electrode is suggested in Scheme 6.1 [32]. The TiO 2 acts as an electron-transfer accelerator between SGC and catechol [33]. This nanostructured TiO 2 exhibits the electrocatalytic properties for the redox of catechol. Catechol absorbed onto TiO 2 rapidly reaches the SGC surface, then is oxidized, involving two electrons (e - ) and two protons (H + ). Similar observation on the enhanced adsorption of the electroactive species and acceleration of the proton transfer step was reported [34,35]. As a result, the surface of TiO 2 acts as a redox mediator for the electron transfer between the SGC electrode and catechol [36,37]. Table 6.1: Summarized IR results showing possible bondings between TiO 2 and catechol at ph 7.0. a Frequencies (cm -1 ) Assignment Catechol + TiO 2 [Ref. 30] Catechol adsorbed on nanostructured TiO 2 in this study γc H in plane ν( C C ) ν( C=C ) ν(c O) 1215; ; 1095 γ(c H) in plane a This test was performed with nanostructured titanium dioxide thin film immobilized on glass substrate. The titanium dioxide film was dipped into the catechol solution (ph=7.0) and washed with super quality water several times. 184

198 SGC electrode 2e- TiO2 coating Ti Ti OH OH O O 2H+ Scheme 6.1: Bonding and electron transfer in sonogel carbon electrode modified with titanium dioxide (SGC/TiO 2 ) Calibration Curve For quantitative analysis of catechol at low concentration using the SGC/TiO 2 electrode, ΔI (difference in reduction current, I pc and oxidation current, I pa in unit of ma) in CV was monitored upon varying concentration of catechol ranging in 0 to 1.0 mm and presented in Eq. (6.1). As shown in Fig. 6.6, linear relation between ΔI and catechol concentration was observed with R 2 = This result implies that the SGC/TiO 2 electrode can detect catechol both quantitatively and qualitatively. Catechol Concentration (mm) = (I pc I pa ) (6.1) 185

199 y = 0.021x R 2 = ΔΙ (ma) Catechol concentration (mm) Fig. 6.6: Relation between current in cyclic voltammetry and catechol concentration. Error bars indicate the standard deviation of cyclic voltammetry result for three SGC/TiO 2 electrodes. 6.4 Conclusions The deliberate chemical modification of carbon electrode surface with nanostructured TiO 2 made it possible to control the rates and selectivities of electrochemical reactions at the solid/liquid interface. The SGC/TiO 2 electrode detected catechol efficiently in the presence of ascorbic acid, a common interferent, using cyclic voltammetry. Due to the high surface area of porous TiO 2, catechol was easily adsorbed onto TiO 2 and formed a chemical bonding, C 6 H 4 (OTi) 2, where the surface of TiO 2 acted as an electron-transfer accelerator between the SGC electrode and catechol. The SGC/TiO 2 electrode developed here met the profitable features of electrode including quantitative and qualitative 186

200 detection, mechanical stability, physical rigidity, and enhanced catalytic properties. We anticipate this new modified sonogel carbon electrode with nanostructured titanium dioxide could be used in several chemical sensor applications for environmental and biological molecules of interest. 6.5 References [1] H.B. Mark, N.F. Atta, Y.L. Ma, L.L. Petticrew, H. Zimmer, Y. Shi, S.K. Lunsford, J.F. Rubinson, A. Galal, Bioelectrochem. Bioenerg. 38 (1995) 229. [2] N. Atta, I. Marawi, K. Pettricrew, H. Zimmer, H.B. Mark, Jr., A. Galal, J. Electroanal. Chem. 408 (1996) 47. [3] S. Alegret, Analyst 121(1996) [4] P. Ugo, M. Moretto, Electroanalysis 10 (1998) [5] O. Lev, M. Tionsky, L. Rabinovich, V. Glezer, S. Sampath, I. Pankratov, J. Gun, Chem. Mater. 9 (1997) [6] F. Tian, G. Zhu, Sens. Actuat. B 86 (2002) 266. [7] M. Ahkim, W.Y. Lee, Anal. Chim. Acta 479 (2003) 143. [8] L. Rabinovich, O. Lev, Electroanalysis 13 (2001) 265. [9] M.M. Collinson, A.R. Howells, Anal. Chem. 72 (2000) 702A. [10] J. Wang, P.V.A. Pamidi, K.R. Rogers, Anal. Chem. 70 (1998) [11] M. del Mar Cordero-Ro, J.L.H.-H. Cisneros, E. Blanco, I.I. Naranjo-Rodríguez, Anal. Chem. 74 (2002) [12] R.F. Lane, A.T. Hubbard, J. Phys. Chem. 77 (1973)

201 [13] P.R. Moses, L. Wier, R.W. Murray, Anal. Chem. 47 (1975) [14] L. Su, L. Mao, Talanta, 70 (2006) 68. [15] L. M. Zen, P.J. Chen, Anal. Chem. 69 (1997) [16] J. Wang, S.P. Chen, M.S. Lin, J. Electroanal. Chem. 273 (1989) 231. [17] A. Liu, I. Honma, H. Zhou, Biosens. Bioelectron. 21 (2005) 809. [18] S.K. Lunsford, H. Choi, J. Stinson, A. Yeary, D.D. Dionysiou, Talanta (In Press). [19] K. Yoo, H. Choi, D.D. Dionysiou, Chem. Comm. (2004) [20] F. Caruso, Adv. Mater. 13 (2001) 11. [21] H. Choi, E. Stathatos, D.D. Dionysiou, Thin. Solid Films 510 (2006) 107. [22] M. Antonelli, J.Y. Ying, Angew. Chem. Int. Ed. Engl. 34 (1995) [23] H. Yoshitake, T. Sugihara, T. Tatsumi, Chem. Mater. 14 (2002) [24] R.C.S. Luz, F.S. Damos, A.B. Oliveira, J. Beck, L.T. Kubota, Sens. Actuators B: Chem. 117 (2006) 274. [25] R.M. Carvalho, C. Mello, L.T. Kubota, Anal. Chim. Acta 420 (2000) 109. [26] C. Wang, Z.X. Deng, Y. Li, Inorg. Chem. 40 (2001) [27] H. Choi, A.C. Sofranko, D.D. Dionysiou, Adv. Funct. Mater. 16 (2006) [28] J.G. Redepenning, Anal. Chem. 6 (1987) 18. [29] A. Mills, S. L. Hunte, J. Photochem. Photobiol. A 108 (1997) 1. [30] F. Bosc, D. Edward, N. Keller, V. Keller, A. Ayral, Thin Solid Films 495 (2006) 272. [31] S.T. Martin, J.M. Kesselman, D.S. Park, N.S. Lewis, M.R. Hoffmann, Environ. Sci. Technol. 30 (1996) [32] S. Mu, Biosens. Bioelectron. 21 (2006)

202 [33] F.S. Damos, M.D.T. Sotomayor, L.T. Kubota, S.M.C.N. Tanaka, A.A. Tanaka, Analyst, 128 (2003) 225. [34] S. Dong, T. Kuwana, J. Electrochem. Soc. 6 (1994) 617. [35] J. Wang, P. Pamidi, and M. Jiang, Analytica Chemica Acta. 360 (1998) 171.H. [36] Razmi, M. Agazadeh, B. Habibi, J. Electroanal. Chem. 547 (2003) 25. [37] H. Qi, C. Zhang, Electroanalysis 17 (2005)

203 CHAPTER 7 Thermally Stable Nanocrystalline TiO 2 Photocatalysts Prepared by Sol-Gel Method Modified with Water Immiscible Room Temperature Ionic Liquids 190

204 Recently, sol-gel methods employing ionic liquids (ILs) have shown significant implications for the synthesis of well-defined nanostructured inorganic materials. Herein, we synthesized nanocrystalline TiO 2 particles via an alkoxide sol-gel method employing a water immiscible room temperature IL (1-butyl-3-methylimidazolium hexafluorophosphate, [bmim][pf 6 ]) as a new reaction medium and further modified with nonionic surfactant (polyoxyethylene sorbitan monooleate) as a pore templating material. Detail information on the preparative method, synthesis route and mechanism, crystallographic and structural properties, and photocatalytic activity of the TiO 2 particles is described. The possible rationale for the formation of nanocrystalline TiO 2 particles with high surface area and activity is discussed with respect to the special characteristics of [bmim][pf 6 ] as well as the role of the surfactant self-assembly in the sol-gel network. Due to its capping effect and water immiscibility, the use of [bmim][pf 6 ] in sol-gel synthesis of TiO 2 induced controlled hydrolysis of titanium alkoxide precursor, resulting in stable sol-gel network with an ordered array, and localized water-poor conditions, yielding in the formation of completely condensed and directly crystalline systems at ambient condition. The low surface energy and adaptability of [bmim][pf 6 ] facilitate the generation of small nanocrystalline TiO 2 particles and then it also acts as a particles aggregation inhibitor. The TiO 2 particles have good thermal stability to resist pore collapse and anatase-to-rutile crystal phase transformation during thermal treatment. The TiO 2 particles from as-synthesized to calcined at high temperatures even up to 800 ºC exhibit promising properties for the degradation of 4-chlorophenol. 191

205 7.1 Introduction Due to its high reaction rates and short treatment times to decompose recalcitrant organic chemicals in water, TiO 2 -based advanced oxidation technology has been extensively researched in environmental remediation [1-8]. In order to enhance the activity and widespread application of TiO 2 photocatalyst for water and air treatment, several previous studies have focused on controlling the physicochemical properties of TiO 2, such as grain size, morphology (i.e., surface area, pore volume, pore size distribution), and crystal phase [9,10]. Special attention was given to the synthesis of nanostructured TiO 2 because of its attractive optical, electrical, chemical and catalytic properties [11]. In several previous studies, fabrication of well-defined nanostructured TiO 2 or other inorganic oxides was achieved via sol-gel methods modified with self-assembly of amphiphilic organic molecules (i.e., surfactants, block copolymers) as pore forming agents in the inorganic network [2,7,8,12-16]. The sol-gel method is one of the most convenient processes for the preparation of TiO 2 due to its ability to tailor-design the structural properties of TiO 2 with mesoporous inorganic matrix as well as its versatile applications in fabricating TiO 2 thin films and membranes, over other synthesis methods such as flame synthesis, hydrolysis precipitation, and hydrothermal synthesis [2,6-8, 16,17]. However, in conventional solvents used in sol-gel synthesis of TiO 2, the hydrolysis and condensation reactions of titanium alkoxide precursors are too fast, resulting in uncontrolled physicochemical properties of TiO 2 [16]. The synthesis method, regardless of hydrolytic or non-hydrolytic reaction pathways, typically yields amorphous 192

206 TiO 2, which needs further heat treatment for crystallization [17]. However, the TiO 2 inorganic network, like some other non-silica oxides, is not thermally stable enough to resist breakage of the initial nanostructured Ti O Ti inorganic matrix and anatase-torutile crystal phase transformation during thermal treatment [18]. The last two are more critical in case of employing pore templating strategies. As a result, it is beneficial to synthesize nanocrystalline TiO 2 particles with high surface area and activity at ambient condition (e.g., room temperature), allowing for a wider selection of support materials to be coated as well as avoiding heat treatment at high temperature. If possible, substituting inorganic supports with polymeric supports could be an important breakthrough in the fabrication of inorganic membranes and films, considering the fragility of inorganic materials, their lack of elasticity, and limited options for shaping and forming. Recently, alkoxide sol-gel methods employing ionic liquids (ILs) have been introduced for the synthesis of inorganic materials with unique shape and structure since ILs posses tunable solvent properties so that they can easily interact with various surface and chemical reaction environments [17,19-28]. In particular, room temperature ILs (RTILs), which are usually composed of a large organic cation and a weakly coordinating anion, are of great interest due to their advantageous physical and chemical properties, such as low melting point, negligible vapor pressure, and high thermal stability [21-28]. In addition, RTILs with hydrophobic regions and a high directional polarizability form extended hydrogen bond systems in the liquid state, resulting in a highly structured selfassembly of RTILs without the formation of ordered micelle structure of hydrophilic and hydrophobic chains as in long chain surfactants [21,29,30]. This organized assembly was reported to be able to act as a template for the preparation of well-defined nanoparticles, 193

207 nanorods, and other nanostructured inorganic materials [25-27]. In previous studies, nanocrystalline TiO 2 particles were also synthesized at ambient condition via sol-gel methods employing RTILs most probably (i.e., still not clear) due to the self-assembly of RTILs and consequent formation of Ti O Ti inorganic network around the RTILs templates [21,22]. However, few studies were focused on understanding the synthesis route and mechanism of TiO 2 in this sol-gel method, beyond the role of RTILs as a template, since the growth of TiO 2 nanocrystals in RTILs at room temperature is a very complex process. In addition, limited experimental information on the effect of RTILs on the physicochemical properties and photocatalytic activity of TiO 2 has been reported. Herein, we synthesized nanocrystalline TiO 2 particles via an alkoxide sol-gel method employing water immiscible RTIL as a solvent medium and later modified with nonionic surfactant as a pore templating material [22,28,31]. Detail information on the preparative method, crystallographic and structural properties, and photocatalytic activity of the TiO 2 particles is described. The possible synthesis route and mechanism of the TiO 2 particles and advantages of using water immiscible RTILs in this sol-gel method are discussed in comparison with the challenges of conventional solvent systems. 7.2 Experimental Preparation Procedure Titanium tetraisopropoxide (TTIP, Ti(OCH(CH 3 ) 2 ) 4, Aldrich) was added to isopropanol (i-proh, (CH 3 ) 2 CHOH, Fisher) at an i-proh/ttip molar ratio of 30. A water immiscible RTIL (1-butyl-3-methylimidazolium hexafluorophosphate, [bmim][pf 6 ], Sachem) was 194

208 then added into the mixture at a [bmim][pf 6 ]/TTIP molar ratio of 3 and vigorously stirred for 10 min. When the RTIL-templated titania sol was added into water drop by drop up to a H 2 O/TTIP molar ratio of 100 at room temperature, hydrolysis and condensation reactions occurred, forming a precipitate of white TiO 2 particles. The solution was stirred vigorously for 30 min, and the TiO 2 particles were recovered by filtration with a fine filter paper (P2, Fisher), washed thoroughly with water, and dried at 100 ºC for 2 h in an oven (Model , Cole-Parmer). The entrapped [bmim][pf 6 ] and organic residues were removed by extraction with acetonitrile (C 2 H 3 N, Fisher) for 12 h. Then, the TiO 2 particles were recovered again by filtration, washed with acetonitrile and water several times, and dried at room temperature for 24 h. The obtained product was a white fine powder. For comparison using a relatively water miscible RTIL, 1-butyl-3- methylimidazolium tetrafluoroborate ([bmim][bf 4 ], Aldrich) was added into the solution instead of the water immiscible [bmim][pf 6 ]. For abbreviations, S control, S PF6 and S BF4 represent TiO 2 particles prepared without RTIL, with [bmim][pf 6 ], and with [bmim][bf 4 ], respectively. For further modification of [bmim][pf 6 ]-templated sol-gel method with surfactant templates, polyoxyethylene sorbitan monooleate (Tween 80, Aldrich) as a nonionic surfactant was dissolved in i-proh. Then, [bmim][pf 6 ] followed by TTIP was added into the solution. The mixture was added into water. The molar ratio of Tween 80 to TTIP was 1. The corresponding TiO 2 particles are denoted as S PF6, T80. The TiO 2 particles were subsequently heat-treated at elevated temperatures (up to 1000 C) to increase materials cyrstallinity and to remove any impurities and surfactant templates using a multi-segment programmable furnace (Paragon Model HT-22-D, Thermcraft) in the presence of air. The temperature of the furnace was increased at a 195

209 ramp rate of 3.0 C/min to a final temperature and held for 1 h. Then, the furnace was allowed to cool down gradually to room temperature Materials Characterization A Kristalloflex D500 diffractometer (Siemens) with Cu Kα (λ = Å) radiation was used for X-ray diffraction (XRD) analysis to determine the crystal phase of the TiO 2 catalyst. For each scan, 2θ was increased from 20.0 to 60.0 degrees with a step of 0.1 and time to step ratio of 1.0. Accelerated surface area and porosimetry (ASAP model 2020 and Tristar 3000, Micromeritics) was applied to measure Brunauer, Emmett, and Teller (BET) surface area, pore size distribution, and porosity of the micro- and mesoporous materials using nitrogen adsorption and desorption isotherms with approximately 30 values of relative pressure ranging from 0.01 to 0.99 at 77 K. Before the measurement, the samples were purged with helium gas for 2 h at 150 C or 12 h at 40 C using Flow prep 060 (Micromeritics). The morphology of the TiO 2 particles was examined using a JEM-2010F (JEOL) high resolution transmission electron microscope (HR-TEM) with field emission gun at 200 kv after dispersing samples in methanol using an ultrasonicator (2510R-DH, Bransonic) for 5 min followed by fixing them on a carbon-coated copper grid (LC200-Cu, EMS). Compositional analysis of the TiO 2 particles was conducted using an energy dispersive X-ray spectroscope (EDX, Oxford Isis) NMR and FTIR Analyses In order to investigate the synthesis route of TiO 2 using the sol-gel method modified with [bmim][pf 6 ], changes in chemical structure and bonding during the synthesis were 196

210 tracked with nuclear magnetic resonance (NMR, Bruker AMX300) and Fourier transform infrared (FTIR, Nicolet Magna-IR 760) spectroscopy. For 1 H-NMR analysis, approximately 0.1 ml of sample was transferred to the NMR holder containing 1.9 ml of deuterated chloroform (CDCl 3, Aldrich). Along with the NMR analysis, drops of the samples were put onto potassium bromide (KBr, Sigma) thin films for FTIR analysis and scanned 64 times at a resolution of 4 cm -1 in transmission mode with a KBr background. In order to study the C-H stretching vibrational infrared spectra of the [bmim][pf 6 ] around 2900 cm -1 and other residual organics, the prepared TiO 2 particles were thoroughly ground, mixed with KBr powder at a ratio of sample:kbr = 1:9, shaped into a thin film, and analyzed Measurement of Photocatalytic Activity The photocatalytic activity of TiO 2 particles was measured in terms of 4-chlorophenol (4- CP, Aldrich) degradation. After dispersing the TiO 2 particles in water by sonication (Model B-22-4, Branson Ultrasonic Cleaner) for 1h, the suspension was added into a cylindrical borosilicate reactor with inner diameter of 5.5 cm and recirculated continuously using a pump. Pre-treated air passing through an activated carbon column and a humidifier was supplied into an additional flask. The following conditions were kept constant: initial volume of reaction solution = 0.6 L, solution recirculation rate = 0.3 L/min, air flow rate = 0.5 L/min, solution temperature = 25 ± 3 C, ph = 7.0 ± 0.1 in 10 mm phosphate buffer, initial concentration of 4-CP = 50 mg/l, and TiO 2 dosage = 500 mg/l. Four 15 W low pressure mercury UV tubes (Spectronics) emitting near UV radiation ( nm) with a peak at 365 nm were used at a light intensity of

211 mw/cm 2 at the center of the reactor. The concentration of 4-CP was determined using a high performance liquid chromatography (HPLC; Series 1100, Agilent) equipped with a C-16 Discovery column (Supelco). The mobile phase was a mixture of acetonitrile and 0.01 N sulfuric acid at a ratio of 30:70 (v/v) with a flow rate of 1.5 ml/min. 7.3 Results and Discussion Physicochemical Properties of [bmim][pf 6 ]-Templated TiO 2 The [bmim][pf 6 ]-templated TiO 2 particles (S PF6 ) were initially heat-treated at 100 C for 1 h. The EDX analysis clearly showed that only Ti and O elements were present in the final product and the C H stretching vibrational infrared spectra of the IL around 2900 cm -1 disappeared after solvent extraction, meaning that most of [bmim][pf 6 ] and other organics were removed by extraction with acetonitrile. Fig. 7.1(a) shows wide angle XRD pattern of S PF6. All the peaks were relatively wide owing to the presence of the small size nanocrystals and they were assigned to anatase crystalline phase. The crystallite size of around 5 nm was calculated using Scherrer s equation based on the XRD peak broadening analysis at (101). It is notable that S PF6 contained anatase crystalline phase in spite of the low heat treatment temperature of 100 C. Crystallinity of around 37 % was approximately calculated from the mass fraction of crystal, based on the XRD peak area and base line [32]. The low angle diffraction pattern in Fig. 7.1(b) showed a wide single peak at 2θ = 1.73º corresponding to a d-spacing of 5.1 nm, indicating disordered mesostructure without long-range order in the pore arrangement. 198

212 (101) d=3.52 (a) Relative intensity (a.u.) (004) d=2.38 (105) d=1.70 (200) d=1.89 (211) d= θ (deg) (b) Relative intensity (a.u.) θ (deg) Fig. 7.1: (a) wide angle and (b) low angle XRD patterns of [bmim][pf 6 ]-templated TiO 2 particles (S PF6 ) calcined at 100 C. Inserted numbers are Miller indices (hkl) and d- spacing of the corresponding crystallographic planes to the peaks. The d-spacing was calculated from Bragg s law: d-spacing = λ/(2 sinθ) where λ is Å. 199

213 Volume N2 adsorbed (cm 3 /g STP) Adsorption Desorption dv/dd (cm3/g) Relative pressure (P/P o ) Pore diameter (nm) Fig. 7.2: Nitrogen adsorption-desorption isotherms and pore size distribution of [bmim][pf 6 ]-templated TiO 2 particles (S PF6 ) calcined at 100 C. The N 2 adsorption-desorption isotherms of S PF6 shown in Fig. 7.2 exhibit type IVlike behavior, which is a typical characteristic of mesoporous materials [33]. A sharp inflection of adsorbed volume at P/P o = 0.45 (hysteresis loop) and a relatively steep desorption branch indicated a distribution of various sized cavities but the same entrance diameter [34]. This means the existence of mesoporosity with similar sized entrances, being in agreement with the low angle diffraction peak in Fig. 7.1(b). The pore size distribution was extremely narrow, implying good homogeneity of the pores. The average pore diameter of 4.5 nm was in a good approximation of d-spacing of 5.1 nm measured from the low angle XRD pattern. The BET surface area and pore volume of the material were also high, about 273±7.4 m 2 g and 0.296±0.011 cm 3 g, respectively. These structural 200

214 characteristics including anatase phase, high surface area, and narrow pore size distribution imply that S PF6 have a great potential for environmental applications in the area of fabrication of TiO 2 photocatalytic films and membranes. The HR-TEM images in Fig. 7.3(a) clearly show the morphology of anatasecontaining nanostructured S PF6 (note Fig. 7.4 for the HR-TEM images of S BF4 ). It is observed that S PF6 was exclusively composed of small sized nanocrystallites. S PF6 was highly porous and all areas of the samples showed more or less similar pore morphology, a disordered mesostructure without long-range order in the pore arrangement, which was consistent with the low angle XRD result. The micrograph in Fig. 7.3(b) shows several randomly oriented nanocrystallites, ranging in size from 4.5 to 6.0 nm with sets of clearly resolved lattice fringes thereby indicating that S PF6 is highly crystalline. (b) (a) (b) Fig. 7.3: HR-TEM images of [bmim][pf 6 ]-templated TiO 2 particles (S PF6 ) calcined at 100 C. 201

215 (a) (b) 100 nm 20 nm Fig. 7.4: HR-TEM images of [bmim][bf 4 ]-templated TiO 2 particles (S BF4 ) calcined at 100 C. Many tiny pores with size of 2-3 nm are observed but the density of the pores is low, resulting in low specific surface area of 141 m 2 /g and pore volume of cm 3 /g, compared to those of [bmim][pf 6 ]-templated TiO 2 (S PF6 ). The approximate d-spacing values of 3.61 Å at (1 0 1) plane, 2.48 Å at (0 0 4), 1.95 Å at (2 0 0), 1.75 Å at (1 0 5) and 1.62 Å at (2 1 1) measured from the major lattice fringes in the TEM image were almost identical with their theoretical values presented in Fig. 7.1(a). The crystallinity of S PF6 was additionally confirmed by the selected area electron diffraction pattern inserted in Fig. 7.3(a), revealing diffraction rings typical for a crystalline powder due to the absence of an amorphous halo. 202

216 7.3.2 Photocatalytic Activity of [bmim][pf 6 ]-Templated TiO 2 The experiment was conducted in the dark for 2 h to allow 4-CP adsorption, followed by UV radiation for 4 h to photocatalytically degrade 4-CP, as shown in Fig The initial ph of 7.0 ± 0.1 without buffer solution decreased slightly to after 4-CP adsorption and decreased significantly to after UV radiation, depending on the extent of 4-CP degradation. Due to its high surface area, S control had a higher adsorption capacity. After 2 h in the dark, 4-CP adsorption on TiO 2 particles reached a pseudo steady state. As expected from the crystallographic analysis of both TiO 2 particles, compared to the amorphous S control, the anatase crystal phase-containing S PF6 had a significant activity. It is well known that no considerable photocatalytic activity is observed for the amorphous TiO 2 due to large numbers of bulk defects where the photogenerated electrons and holes are recombined simultaneously. 4-CP concentration (mg/l) Without TiO 2 (photolysis) S control S PF6 Dark UV irradiation Reaction time (h) Fig. 7.5: 4-CP adsorption and photocatalytic degradation by [bmim][pf 6 ]-templated TiO 2 particles (S PF6 ) calcined at 100 C over time. Error bars indicate the standard deviation of triplicated results. 203

217 Table 7.1: BET specific surface area of TiO 2 particles prepared at different conditions. Entry [bmim][pf 6 ]/TTIP H 2 O/TTIP BET surface area (m 2 /g) ± ± A B ±8.0 (165±6.0) a C (186) D ±9.7 (198±7.9) E ±7.4 (130±11.9) F a The values in parenthesis were obtained from comparative experiments where the water was added into the mixed sol Effect of [bmim][pf 6 ]/TTIP and H 2 O/TTIP Molar Ratio The effect of sol composition on the BET surface area was investigated, as shown in Table 7.1. [bmim][pf 6 ]/TTIP molar ratio was initially studied at a constant H 2 O/TTIP molar ratio of 100. It can be seen that the surface area decreased steeply with addition of [bmim][pf 6 ] up to [bmim][pf 6 ]/TTIP ratio of 3. Above this ratio, the surface area was not changed significantly and anatase peaks started to appear. The surface area of S control (sample 1) was only 112 m 2 /g after 400ºC heat treatment at which S contol possessed comparable anatase crystallinity with S PF6 calcined at 100ºC (sample 4). It should be noted that the surface area of S PF6 is much higher than other anatase-containing TiO 2 particles obtained by conventional sol-gel method in which the prepared TiO 2 particles are amorphous and then calcined to yield anatase crystalline phase with a much reduced 204

218 surface area. In this modified sol-gel method using [bmim][pf 6 ], the surface area and crystallinity of the samples were hardly affected by the amount of H 2 O, as shown in Table 7.1 where H 2 O/TTIP molar ratio was studied at a constant [bmim][pf 6 ]/TTIP molar ratio of Synthesis Mechanisms of [bmim][pf 6 ]-Templated TiO 2 The underlying mechanism for the formation of the highly porous crystalline TiO 2 particles in this study is depicted in Scheme 7.1. In order to better understand the formation of TiO 2 sol-gel network, a series of NMR/FTIR studies were carried out. The formation of TiO 2 sol in preparation step I shown in Scheme 7.1 was evidenced from a mixture of [bmim][pf 6 ] and TTIP in CDCl 3. TTIP and RTIL in i-proh RTIL in i-proh and H2O Void H2O i-proh H2O RTIL Step I: TiO2 particles in sol Step II: TiO2 particles in sol-gel network Step III: Porous TiO2 material Scheme 7.1: Synthesis route of porous anatase crystalline TiO 2 particles (S PF6 ) using solgel method modified with [bmim][pf 6 ]. Note steps I, II, and III upon chemical addition into the solution. 205

219 Fig. 7.6: NMR spectrum of [bmim][pf 6 ]/TTIP mixture during the synthesis of [bmim][pf 6 ]-templated TiO 2 (S PF6 ). As shown in Fig. 7.6, along with resonance peaks for [bmim][pf 6 ] marked as 1 6, TTIP as 7 and 8, and CDCl 3 as 11, the concommitant generation of new resonance peaks marked as 9 and 10, which correspond to free i-proh, was observed. The in-situ formation of i-proh is because of the reaction of isopropyl groups in TTIP with hydroxyl groups in [bmim][pf 6 ] containing a small quantity of water up to % [29,35]. No direct chemical bonding between [bmim][pf 6 ] and TTIP was observed. After stabilizing the [bmim][pf 6 ]/TTIP mixture for 12 h, an obvious phase separation was observed due to the water immiscibility of [bmim][pf 6 ]. FTIR analysis shown in Fig. 7.7 verified the Ti O bond formation. The upper layer was composed of TTIP and the broad band in cm -1 indicated bonded hydroxyl groups of the newly formed i-proh. 206

220 On the other hand, the bottom layer was found to be [bmim][pf 6 ], and new peaks at around 824 cm -1 can be assigned to the characteristic Ti O Ti stretching frequency. These results suggest that the small quantity of water in the [bmim][pf 6 ] would initiate the formation of Ti O Ti bond and the TiO 2 particles would be homogeneously dispersed only in the hydrophobic [bmim][pf 6 ] phase (Equation 7.1). However, the role of water in the [bmim][pf 6 ] is complex and its structure and chemical reactivity seem to be different from those of bulk water as it is tightly bound and activated in the hydrogen bond system of the [bmim][pf 6 ] [29,36,37]. Ti[OCH(CH 3 ) 2 ] H 2 O + [Bmim]PF 6 TiO (CH 3 ) 2 CHOH + [Bmim]PF 6 (7.1) In preparation step II shown in Scheme 7.1, where the non-reacted TTIP is completely hydrolyzed, the [bmim][pf 6 ] acts as a capping agent for preventing direct hydrolysis of TTIP due to its water immiscibility. The reduced hydrolysis rate makes it possible to achieve a longer ageing time for the formation of a stable sol-gel network possibly with an ordered array. The stability of TiO 2 sol-gel network was further verified during heat treatment (see the next section). The capping effect of [bmim][pf 6 ] also makes localized anhydrous or water-poor area between [bmim][pf 6 ] and inorganic starting material, TTIP. It is usually observed that this water-poor condition can suppress the formation of hydroxide and oxihydrate and generation of amorphous species as limited amounts of water drive the mass balance to completely condensed and directly crystalline systems [17]. 207

221 Absorbance (a.u.) bonded OH Upper layer 824 Lower layer Wavenumbers (cm -1 ) Fig. 7.7: FTIR spectra of [bmim][pf 6 ]/TTIP mixture during the synthesis of [bmim][pf 6 ]-templated TiO 2 (S PF6 ). Also, in conventional sol-gel solvent systems employing excessive amount of water (surface tension, γ = 72 mj/m 2 at 293 K), the nucleation of TiO 2 is slow while low interface energy (γ = 46 mj/m 2 at 293 K) and adaptability of the [bmim][pf 6 ] result in high nucleation rates, generating very small nanocrystalline particles [17,30,38]. Homogeneity of TiO 2 nanoparticles dispersed in sol-gel network was retained throughout the entire ageing process and no phase separation was observed. In a previous study, a reaction-limited aggregation theory where a small portion of particle collisions results in limited adherence of particles in the presence of IL was proposed to explain the formation of these highly porous nanocrystals [17]. In order to verify how the presence of 208

222 [bmim][pf 6 ] initially affected the particle adherence, particle size of agglomerated TiO 2 was measured when the [bmim][pf 6 ]/TTIP mixture in i-proh was added into water, as shown in Fig In the absence of any RTILs or in the presence of [bmim][bf 4 ], large TiO 2 clusters were formed in average size of 856 nm and 690 nm, respectively. However, the presence of [bmim][pf 6 ] significantly reduced the cluster size to 291 nm, resulting from inhibition of TiO 2 particle aggregation. These results imply that the water immiscibility of [bmim][pf 6 ] successfully controlled hydrolysis reaction of TTIP and thus made it possible to homogeneously disperse the particles in the [bmim][pf 6 ] phase, resulting in the formation of highly porous TiO 2 particles. 30 S BF4 Intensity (%) S PF6 S control Size (nm) Fig. 7.8: Particle size of agglomerated TiO 2 during sol-gel synthesis without IL (S control ) and with [bmim][pf 6 ] (S PF6 ) and [bmim][bf 4 ] (S BF4 ). 209

223 In addition, a highly structured frame from the extended hydrogen bond system in [bmim][pf 6 ] is well-known to act as a template for the well-defined nanostructured particles [21,29,39]. The [bmim][pf 6 ] template was just entrapped in the growing covalent titania network rather than being chemically bound to the inorganic matrix, which was evidenced by the NMR and FTIR analyses. Since the [bmim][pf 6 ] does not form any chemical bonds with TiO 2, the [bmim][pf 6 ] can be easily removed through extraction [17,40]. In preparation step III shown in Scheme 7.1, where solvents and [bmim][pf 6 ] were removed, the [bmim][pf 6 ] provides an attractive method for achieving longer ageing time without shrinkage of the gel network [41]. Compared to conventional solvents which evaporate quickly before formation of a stable sol-gel network during the ageing process, the negligible vapor pressure of the [bmim][pf 6 ] can prevent gel shrinkage and reduction in surface area Heat Treatment of [bmim][pf 6 ]-Templated TiO 2 Figs. 7.9 and 7.10 summarize crystal phase transformation and structural evolution of S control and S PF6 upon heat treatment. For S control, anatase phase started to appear after heat treatment at 250 C. The anatase crystallinity of S control calcined at C was comparable to that of S PF6 at 100 C. The anatase phase of S control started to transform to rutile at above 600 ºC [42]. On the other hand, S PF6 was thermally very stable. Even without further heat treatment, an anatase peak existed. Only the anatase phase was present at up to 900 C [43]. The thermal stability of S PF6 was in good agreement with their structural characteristics. 210

224 Intensity (a.u.) Intensity (a.u.) (a) (b) θ (degrees) Temperature ( o C) θ (degrees) Temperature ( o C) Fig. 7.9: Crystal phase evolution upon heat treatment: (a) control TiO 2 (S control ) and (b) [bmim][pf 6 ]-templated TiO 2 (S PF6 ) ( : anatase and : rutile). The pore volume and surface area of S control decreased rapidly from to cm 3 /g and from 537 to 3.76 m 2 /g, respectively, while those of S PF6 decreased gradually from to cm 3 /g and from 282 to 47.9 m 2 /g upon heat treatment from 20 C to 800 C. Consequently, S PF6 still kept its highly porous structure, which suppressed solidstate aggregation of the anatase crystallites. The pore size and crystallite size of S PF6 even at 800 C were less than 15 and 20 nm, respectively. On the other hand, the structural properties of S BF4 from as-synthesized and calcined at 500 C were relatively poor, compared to those of S PF6 and even those of S control (note Table 7.2). Although the use of [bmim][bf 4 ] also affected anatase-rutile crystal phase transformation and thermal stability during calcination, the magnitude of change in such properties was not considerable. The size of as-synthesized and calcined TiO 2 particles was also investigated, 211

225 as shown on Fig As-synthesized S PF6 was less aggregated in size of 136 nm, compared to 260 nm for S control and 230 nm for S BF4. After heat treatment at 500 ºC, size of S control and S BF4 significantly increased to 708 nm and 567 nm, respectively while S PF6 had more or less small size of 433 nm. These results also support the role of [bmim][pf 6 ] explained in the section of synthesis route. Pore volume (cm 3 /g) S control S PF6 600 (a) (b) Surface area (m 2 /g) S control S PF Calcination temperature ( o C) Calcination temperature ( o C) Pore size (nm) (c) S control S PF6 Crystallite size (nm) S control (d) S PF Calcination temperature ( o C) Calcination temperature ( o C) Fig. 7.10: Structural evolution of the TiO 2 particles upon heat treatment: (a) pore volume, (b) surface area, (c) pore size, and (d) crystallite size. 212

226 Table 7.2: Physicochemical properties of TiO 2 particles prepared under various synthesis conditions. Synthesis condition Temp. (ºC) Crystal phase V pore (cm 3 /g) S BET (m 2 /g) D BJH (nm) CS XRD (nm) No addition as-syn. Amorphous (S control ) 500 Anatase [bmim][pf 6 ] as-syn. Anatase (S PF6 ) 500 Anatase [bmim][bf 4 ] as-syn. Amorphous (S BF4 ) 500 Anatase [bmim][pf 6 ] and T80 as-syn. Anatase (S PF6,T80 ) 500 Anatase 0.312± ± T80 a as-syn. Amorphous Anatase a TiO 2 particles were prepared with T80 surfactant only in the same methodology, which is not properly designed for the use of the surfactant. 30 As-synthesized Calcined at 500 o C Intensity (%) S PF6 S BF4 S control Size (nm) Fig. 7.11: Particle size of as-synthesized and calcined TiO 2 prepared without IL (S control ) and with [bmim][pf 6 ] (S PF6 ) and [bmim][bf 4 ] (S BF4 ). 213

227 The physicochemical properties of TiO 2 particles were correlated with their adsorption ability and photocatalytic activity, as summarized in Fig The adsorption capacity of the TiO 2 particles was a strong function of the surface area upon calcination temperature. It is shown that S PF6, even prepared at low temperatures, had considerable photocatalytic activity due to its crystal content and high surface area. Indeed, the photocatalytic activity of S PF6 at 100 C was comparable to that of S control prepared at C. Moreover, the photocatalytic activity of S PF6 with around 37% anatase crystallinity at 100 C was approximately one third of that of S PF6 at C, where the anatase crystallinity further increased. The photocatalytic activity of S control increased with increasing calcination temperature up to 600 C, above which the activity decreased due to the reduction in surface area and the anatase-rutile phase transformation [44]. On the other hand, the activity of S PF6 was relatively stable, even at up to 800 C, because of the retention of active anatase phase and the relatively high surface area [45,46] (a) S control (b) Removal (%) S PF6 Removal (%) S control S PF Calcination tmeperature ( o C) Calcination temperature ( o C) Fig. 7.12: (a) 4-CP adsorption and (b) photocatalytic degradation by control TiO 2 (S control ) and [bmim][pf 6 ]-templated TiO 2 (S PF6 ). 214

228 7.3.6 Modification of [bmim][pf 6 ]-Assisted Sol-Gel Method with Surfactant Self- Assembly as Pore Templates Although many self-assembling strategies have been developed using a variety of surfactants as a templating material to improve the structural and catalytic properties of TiO 2 materials, collapse of the initial porous TiO 2 inorganic network is accompanied during heat treatment for removing the organic templates and ensuring catalytic activity [2,18]. In this study, the ability of [bmim][pf 6 ] to make TiO 2 photocatalysts thermally stable implies that [bmim][pf 6 ]-assisted sol-gel method, once further modified with surfactant templates, can result in the formation of highly porous crystalline TiO 2 particles after heat treatment at even high temperatures. In order to investigate the formation of surfactant self-assembly in [bmim][pf 6 ]-assisted sol-gel method, the assembly size of Tween 80 in i-proh was measured, as shown in Fig The intensity average size of surfactant self-assembly was approximately 3.5 nm. No direct chemical bonding between nonionic Tween 80 and ionic [bmim][pf 6 ] was observed, based on FTIR and NMR analyses. The surfactant self-assembly is used as a template during hydrolysis condensation reactions of TTIP and cross-linking of Ti-O-Ti network, resulting in the formation of inorganic-organic nanocomposite Removal of [bmim][pf 6 ] and Surfactant and Properties of TiO 2 Before [bmim][pf 6 ] extraction, S PF6,T80 had low surface area of 15.0 m 2 /g and pore volume of cm 3 /g since [bmim][pf 6 ] and surfactant added were embedded into the TiO 2 network as shown in Fig. 7.14(a). The [bmim][pf 6 ] was extracted with acetonitrile 215

229 as shown in Fig. 7.14(b) where most of [bmim][pf 6 ] peaks, especially the C-H stretching vibrational IR spectrum of the [bmim][pf 6 ] around 2900 cm 1, disappeared. Then, S PF6,T80 had slightly increased surface area of 58.9 m 2 /g and pore volume of cm 3 /g. Fig shows result on TGA/DSC analysis for determining a heat treatment temperature required to completely remove all the organic residues embedded in S PF6,T80. The as-synthesized S PF6,T80 undergoes endothermic desorption of solvents, including water and i-proh, at low temperature below 120 ºC. Exothermic peaks localized at mainly 269 ºC and partially 420 ºC reflect the processes of oxidation of the organic residuals and anatase crystallization of amorphous structure. Due to the nonionic characteristic of Tween 80, after calcination of S PF6,T80 at 500 ºC, no significant weight change was observed and the resulting IR spectrum shown in Fig. 7.14(c) and EDX study of S PF6,T80 showed that S PF6,T80 was composed of only Ti and O elements Intensity (%) Size (nm) Fig. 7.13: Size measurement of Tween 80 surfactant self-assembly. 216

230 Absorbance (a.u.) (c) (b) (a) Wavenumber (cm -1 ) Fig. 7.14: FTIR spectra of TiO 2 (S PF6,T80 ) prepared via [bmim][pf 6 ]- and surfactantassisted sol-gel method: (a) as-synthesized before [bmim][pf 6 ] extraction, (b) assynthesized, and (c) after heat treatment at 500ºC ( : [bmim][pf 6 ] and : Tween 80). 269 o C Exo 4 Weight (%) TGA 420 o C DSC Heat flow (W/g) Calcination temperature ( o C) Fig. 7.15: TGA/DSC pattern of as-synthesized TiO 2 (S PF6,T80 ) via [bmim][pf 6 ]- and surfactant-assisted sol-gel method. 217

231 Fig shows the highly porous inorganic network of S PF6,T80 after heat treatment at 500 ºC. S PF6,T80 has slightly collapsed spherical bicontinuous structure with highly interconnected network with porosity of 54.8%, indicating that the surfactant used in this synthesis method effectively acted as a pore directing agent. Obvious 5 8 nm pores and many randomly oriented 5 10 nm nanocrystallites with sets of clearly resolved lattice fringes were observed. Fig. 7.16: HR-TEM morphology of nanostructured anatase crystalline TiO 2 (S PF6,T80 ) prepared via [bmim][pf 6 ]- and surfactant-assisted sol-gel method at 500 ºC. 218