Alkaline hydrothermal treatment of titanate nanostructures

ALKALINE HYDROTHERMAL TREATMENT OF TITANATE NANOSTRUCTURES Submitted by DANA LEE MORGAN Bachelor of Applied Science (Honours, Chemistry) Associate Degree of Applied Science A thesis presented to the Queensland University of Technology, in fulfillment of the requirements of the degree of Doctor of Philosophy August 2010 ABSTRACT Since its initial proposal in 1998, alkaline hydrothermal processing has rapidly become an established technology for the production of titanate nanostructures This simple, highly reproducible process has gained a strong research following since its conception However, complete understanding and elucidation of nanostructure phase and formation have not yet been achieved Without fully understanding phase, formation, and other important competing effects of the synthesis parameters on the final structure, the maximum potential of these nanostructures cannot be obtained Therefore this study examined the influence of synthesis parameters on the formation of titanate nanostructures produced by alkaline hydrothermal treatment The parameters included alkaline concentration, hydrothermal temperature, the precursor material‘s crystallite size and also the phase of the titanium dioxide precursor (TiO2, or titania) The nanostructure‘s phase and morphology was analysed using X-ray diffraction (XRD), Raman spectroscopy and transmission electron microscopy X-ray photoelectron spectroscopy (XPS), dynamic light scattering (non-invasive backscattering), nitrogen sorption, and Rietveld analysis were used to determine phase, for particle sizing, surface area determinations, and establishing phase concentrations, respectively This project rigorously examined the effect of alkaline concentration and hydrothermal temperature on three commercially sourced and two self-prepared TiO2 powders These precursors consisted of both pure- or mixedphase anatase and rutile polymorphs, and were selected to cover a range of phase concentrations and crystallite sizes Typically, these precursors were treated with 5– 10 M sodium hydroxide (NaOH) solutions at temperatures between 100–220 °C Both nanotube and nanoribbon morphologies could be produced depending on the combination of these hydrothermal conditions Both titania and titanate phases are comprised of TiO6 units which are assembled in different combinations The arrangement of these atoms affects the binding energy between the Ti–O bonds Raman spectroscopy and XPS were therefore employed in a preliminary study of phase determination for these materials The change in binding energy from a titania to a titanate binding energy was investigated in this i study, and the transformation of titania precursor into nanotubes and titanate nanoribbons was directly observed by these methods Evaluation of the Raman and XPS results indicated a strengthening in the binding energies of both the Ti (2p3/2) and O (1s) bands which correlated to an increase in strength and decrease in resolution of the characteristic nanotube doublet observed between 320 and 220 cm–1 in the Raman spectra of these products The effect of phase and crystallite size on nanotube formation was examined over a series of temperatures (100–200 °C in 20 °C increments) at a set alkaline concentration (7.5 M NaOH) These parameters were investigated by employing both pure- and mixed- phase precursors of anatase and rutile This study indicated that both the crystallite size and phase affect nanotube formation, with rutile requiring a greater driving force (essentially ―harsher‖ hydrothermal conditions) than anatase to form nanotubes, where larger crystallites forms of the precursor also appeared to impede nanotube formation slightly These parameters were further examined in later studies The influence of alkaline concentration and hydrothermal temperature were systematically examined for the transformation of Degussa P25 into nanotubes and nanoribbons, and exact conditions for nanostructure synthesis were determined Correlation of these data sets resulted in the construction of a morphological phase diagram, which is an effective reference for nanostructure formation This morphological phase diagram effectively provides a ‗recipe book‘ for the formation of titanate nanostructures Morphological phase diagrams were also constructed for larger, near phase-pure anatase and rutile precursors, to further investigate the influence of hydrothermal reaction parameters on the formation of titanate nanotubes and nanoribbons The effects of alkaline concentration, hydrothermal temperature, crystallite phase and size are observed when the three morphological phase diagrams are compared Through the analysis of these results it was determined that alkaline concentration and hydrothermal temperature affect nanotube and nanoribbon formation independently through a complex relationship, where nanotubes are primarily affected by temperature, whilst nanoribbons are strongly influenced by alkaline concentration Crystallite size and phase also affected the nanostructure formation Smaller ii precursor crystallites formed nanostructures at reduced hydrothermal temperature, and rutile displayed a slower rate of precursor consumption compared to anatase, with incomplete conversion observed for most hydrothermal conditions The incomplete conversion of rutile into nanotubes was examined in detail in the final study This study selectively examined the kinetics of precursor dissolution in order to understand why rutile incompletely converted This was achieved by selecting a single hydrothermal condition (9 M NaOH, 160 °C) where nanotubes are known to form from both anatase and rutile, where the synthesis was quenched after 2, 4, 8, 16 and 32 hours The influence of precursor phase on nanostructure formation was explicitly determined to be due to different dissolution kinetics; where anatase exhibited zero-order dissolution and rutile second-order This difference in kinetic order cannot be simply explained by the variation in crystallite size, as the inherent surface areas of the two precursors were determined to have first-order relationships with time Therefore, the crystallite size (and inherent surface area) does not affect the overall kinetic order of dissolution; rather, it determines the rate of reaction Finally, nanostructure formation was found to be controlled by the availability of dissolved titanium (Ti4+) species in solution, which is mediated by the dissolution kinetics of the precursor KEYWORDS Alkaline hydrothermal treatment, Soft-chemical, Morphological phase diagram, Titanium dioxide, Titania, TiO2, Titanate, Nanoparticle, Nanotube, Nanoribbon, Nanostructure, Anatase, Rutile, Degussa P25, Crystallite size, Alkaline concentration, Hydrothermal temperature, Raman spectroscopy, Transmission electron microscopy, X-ray diffraction, X-ray photoelectron spectroscopy, Dissolution, Kinetics iii PUBLICATIONS ARISING FROM THIS PROJECT Morgan, D.L.; Triani, G.; Blackford, M.G.; Raftery, N A.; Frost, R.L.; Waclawik, E.R Submitted to Journal of Materials Science, 30th July 2010 Morgan, D.L.; Liu, H-W.; Frost, R.L.; Waclawik, E.R Journal of Physical Chemistry C 2010, 114, 101–110 Morgan, D L.; Zhu, H-Y.; Frost, R.L.; Waclawik, E.R Chemistry of Materials 2008, 20, 3800–3802 Morgan, D L.; Waclawik, E R.; Frost, R L Advanced Materials Research (Zuerich, Switzerland) 2007, 29–30, 211–214 Morgan, D L.; Waclawik, E R.; Frost, R L In International Conference on Nanoscience and Nanotechnology; Jagadish, C., Lu, G Q M., Eds.; IEEE Publishing Company: Brisbane, 2006, p 60–63 iv TABLE OF CONTENTS ABSTRACT i KEYWORDS iii PUBLICATIONS ARISING FROM THIS PROJECT iv TABLE OF CONTENTS v LIST OF FIGURES x LIST OF TABLES xvii ABBREVIATIONS xviii STATEMENT OF ORIGINALITY xix ACKNOWLEDGEMENTS xx CHAPTER INTRODUCTION 1 Introduction 2 Description of the Scientific Problem Investigated Overall Objectives of this Study Specific Aims of the Study Account of Scientific Progress Linking the Scientific Papers CHAPTER LITERATURE REVIEW Introduction Hydrothermal Processing Crystallographic Properties of Titanium Oxides 3.1 Titanium Dioxide 3.2 Titanate 13 Synthesis of Titanium Oxide Nanotubes 16 Alkaline Hydrothermal Method 18 Variations to the Alkaline Hydrothermal Method 19 6.1 Hydrothermal Liquor 19 6.2 Starting Material 20 6.3 Method of Treatment 21 Nanostructure Morphology 22 v 7.1 Nanotubes 23 7.2 Nanoribbons 24 Nanostructure Functionalisation 25 8.1 Nanostructure Doping 25 8.2 Decoration (or Adhesion) of Nanoparticles onto the Nanostructure Surface 26 8.3 Nanostructure Composite Materials 27 Nanostructure Applications 28 9.1 Catalysis and Photocatalysis 28 9.2 Hydrogen Sensing and Storage 29 9.3 Lithium Batteries 29 9.4 Biomedical Applications 30 10 Mechanisms of Nanostructure Formation 30 10.1 Nanotube Formation Mechanisms 30 10.2 Nanoribbon Formation Mechanisms 35 11 Nanostructure Phase and Composition 36 11.1 Nanotubes 36 11.2 Nanoribbons 40 12 Conclusion 41 13 References 41 CHAPTER RELATIONSHIP OF TITANIA NANOTUBE BINDING ENERGIES AND RAMAN SPECTRA 49 Statement of Contribution 50 Synopsis 51 Abstract 52 Keywords 52 Introduction 52 Experimental 55 Reagents and Synthesis 55 Characterisation 55 Results and Discussion 56 Conclusions 60 vi Acknowledgment 60 References 60 CHAPTER SYNTHESIS AND CHARACTERISATION OF TITANIA NANOTUBES: EFFECT OF PHASE AND CRYSTALLITE SIZE ON NANOTUBE FORMATION 63 Statement of Contribution 64 Synopsis 65 Keywords 66 Abstract 66 Introduction 66 Experimental 68 Reagents, Synthesis and Characterisation 68 Results and Discussion 68 Conclusions 72 References 73 CHAPTER DETERMINATION OF A MORPHOLOGICAL PHASE DIAGRAM OF TITANIA/TITANATE NANOSTRUCTURES FROM ALKALINE HYDROTHERMAL TREATMENT OF DEGUSSA P25 74 Statement of Contribution 75 Synopsis 76 Manuscript Text 77 Supporting Information Available 83 References 84 Electronic Supporting Information 86 CHAPTER IMPLICATIONS OF PRECURSOR CHEMISTRY ON THE ALKALINE HYDROTHERMAL SYNTHESIS OF TITANIA/TITANATE NANOSTRUCTURES 90 Statement of Contribution 91 Synopsis 92 vii Abstract 93 Introduction 94 Experimental Procedure 96 2.1 Synthesis of Nanostructures 96 2.2 Characterization of Nanostructures 97 Determination of Morphological Phase Diagrams 97 3.1 Qualitative X-Ray Diffraction Investigation of Nanostructures 97 3.1.1 Anatase Precursor 99 3.1.2 Rutile Precursor 101 3.1.3 Degussa P25 Precursor 101 3.2 Identification of Nanostructures by Raman Spectroscopy 102 3.2.1 Anatase Precursor 103 3.2.2 Rutile Precursor 103 3.2.3 Degussa P25 Precursor 105 Construction of the Morphological Phase Diagrams 105 Confirmation of the Morphological Phase Diagrams 106 5.1 Confirmation of Nanostructure Morphologies by Transmission Electron Microscopy Investigations 106 5.2 Brunauer–Emmet–Teller Surface Area Measurements 108 Interpretation of the Morphological Phase Diagram 111 Conclusions 113 Acknowledgement 114 Supporting Information Available 114 References 115 Electronic Supporting Information 117 CHAPTER ALKALINE HYDROTHERMAL KINETICS IN TITANATE NANOSTRUCTURE FORMATION 123 Statement of Contribution 124 Synopsis 125 Abstract 126 Introduction 127 Experimental Procedure 130 viii Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation confirmed by the absence of anatase in the XRD pattern As the treatment duration was increased, there were no discernable changes within the nanotubes produced Although X-ray diffraction indicated that no nanotube reflections were present in the rutile-treated powder after hours, small quantities of nanotubes and nanosheets were observed by TEM investigation After hours hydrothermal treatment, considerable quantities of well-formed nanotubes were observed for the rutile-treated powder as can be seen in Fig 3d and 3e Similarly to the anatase-treated samples, these nanotubes appeared to cluster around nanoparticle aggregates Rutile-produced nanotubes had an average internal diameter of 3.6 ± 0.7 nm, external diameter of 9.8 ± 1.5 nm and lengths of up to several hundred nanometers A high resolution image of a rutile-produced nanotube is presented in Fig 3f, whose asymmetric walls consisted of 4–5 shells All rutile-treated powders contained considerable quantities of nanoparticles as indicated by the presence of unreacted rutile in the XRD analysis When the specific surface areas of the products were analyzed, an increase in specific surface area was always observed in products corresponding to greater nanotube yields (as determined by XRD and Raman spectroscopy) for both the anatase- and the rutile-treated powders Fig presents the specific surface areas analyzed with nitrogen sorption using the Brunauer-Emmet-Teller (BET) calculations The specific surface areas of the anatase-treated powders increased rapidly until a maximum of ca 250 m2 g–1 was reached for hours treatment before leveling out to ca 220 m2 g–1 between 16 and 32 hours The maximum in surface area measured for the hour treated powder suggests that nanosheets and partially Fig BET surface areas of materials produced after hydrothermal treatment of commercial anatase and commercial rutile 136 Submitted to Journal of Materials Science Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation formed nanotubes were present within the sample Both of these nanostructures have greater specific surface areas than the fully formed nanotubes and can contribute up to a possible 25% increase in surface area compared to a sample containing fully formed nanotubes This scenario is the probable cause of the increase in specific surface area as this sample contained ca 6% unreacted anatase, which has a comparatively small surface area compared to the nanotubes The presence of small quantities of nanosheets was observed within the hour treated anatase sample during the TEM investigation An increase in specific surface area was also observed for the rutile-treated powders between and 16 hours of treatment after which no substantial increases were observed These surface areas were significantly smaller compared to the anatase-treated powders The origin of this lower surface area probably originated from the high concentrations of low surface area rutile particles remaining unreacted within the powder In accordance with the currently accepted mechanism of nanotube formation, the intermediate nanosheet morphology is believed to be constructed through the condensation and polymerization of mono- and polytitanic ions (TiO32–, TiO2(OH)22-, and TinO2n+ m2m–) [42] Through the reassembly of these titanic ions, it can be presumed that smaller, intermediate nanosheet precursors may be produced within the hydrothermal liquor To confirm this hypothesis, dynamic light scattering using non-invasive back scattering technology was performed on each hydrothermally treated sample to determine the particle size distributions of the nanostructures and their intermediates formed during hydrothermal treatment To minimize the concentration of large particles and aggregates that could interfere with the detection of smaller particles, the samples were filtered through 0.45 and 0.1 µm Teflon filters This process removed unreacted precursor particles >100 nm in diameter and large nanosheets whose dimensions exceeded 100 nm in height and width In Fig multiple particle size distributions can be observed ranging between 0.6 and 6500 nm Since the filtering process removed particles with a hydrodynamic radius of 100 nm or greater, these large particle size distributions could be produced by elongated structures with a width or diameter significantly less than 100 nm We note that with diameters between 5–15 nm, some nanotubes could theoretically pass longitudinally through the filter pores, however their lengths can reach several hundred nanometers and it is more likely that the largest particle size distributions 137 Submitted to Journal of Materials Science Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation Fig Particle sizes of nanostructures formed after hours hydrothermal treatment of commercial rutile after filtering through 0.45 and 0.1 m filters Original traces of the three runs and their average are presented were generated by titanate nanotubes, possibly titanate nanotube aggregates Over the duration of the analysis, the smaller particle size distributions were observed to increase in intensity and thus decrease in concentration whilst the large particle size distributions increased in concentration This suggests that these intermediates were actively reassembling into larger intermediates and nanotubes These observations are consistent with the currently accepted mechanism of nanotube formation through the dissolution of the TiO2 precursor and subsequent recrystallisation and scrolling of titanate sheets into nanotubes [18, 19, 42, 43] Assuming the nanotubes form following the dissolution and subsequent recrystallisation of the particulate precursor [18], the decrease in the observed crystalline concentration over time could be attributed to precursor dissolution The precursor dissolution products are Ti4+ species mainly in the form of mono- or polytitanic ions [42], where the concentration of dissolved Ti4+ species in solution is a function of temperature [19] Nanosheets and nanotube formation occurs spontaneously [19, 44] when Ti4+ species are liberated from the precursor crystallites, and dynamic equilibrium is rapidly established between the Ti4+ species and the nanosheet intermediate [44] As the nanosheet‘s surface charge is imbalanced because of undercoordinated sites [20], the nanosheets scroll or fold into nanotubes to neutralize this charge Since nanotube formation is favored by this mechanism the reaction is constantly driven towards product formation It can therefore be assumed that precursor dissolution is the rate determining step in the formation of nanotubes 138 Submitted to Journal of Materials Science Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation Rietveld analysis of the XRD peaks was performed on each sample to quantify the extent of precursor consumption and nanotube production after hydrothermal treatment This was achieved through evaluating the residual crystalline content As the exact phase of the nanotube titanate is not currently known, modeling of the nanotube XRD reflections for Rietveld analysis could not be performed, and the absolute amorphous and nanotube concentrations could not be determined However the TEM measurements did confirm that nanotube formation occurred in quantities commensurate with the quantities of precursor consumed With these considerations, Fig Percent compositions of anatase, rutile and unmodelled phases (nanotubes and amorphous) content after hydrothermal treatment of commercial anatase (a) and commercial rutile (b); and their respective zero (c) and second order fittings (d) 139 Submitted to Journal of Materials Science Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation all Rietveld analyses for crystalline (precursor) and unmodeled phases (nanotube and amorphous content) are presented as wt % of the total sample (m/m %) The kinetic order of precursor consumption was calculated for each precursor using the mass fraction of crystallite determined by Rietveld analysis over time (hrs) As suggested by interpretation of the intensity of the XRD patterns, anatase reacted quickly as evidenced by the rapid decrease of crystalline material with increasing treatment duration from to hours in Fig 6a The observed rate of dissolution corresponds to zero-order kinetics (Linear regression of mass fraction vs time, R2 = 0.986, see Fig 6c) suggesting that the reaction was independent of the reactant concentration In contrast to this result, a slow, non-linear decrease in crystalline concentration was observed for the rutile precursor with a ca 67% decrease in crystalline content over 32 hours treatment duration (Fig 6b) This corresponds to a second-order dissolution rate (Linear regression of mass fraction–1 vs time, R2 = 0.999, see Fig 6d) These different kinetic orders of dissolution of anatase and rutile precursor materials are of primary interest because they highlight significant differences manifest in the mechanism of precursor consumption To further examine the dissolution process, the effect of specific surface area over time was also examined The specific surface areas of the precursors were calculated through equation SA  Where 3000 rm (1) equals density (m2 g–1), r equals radius (nm, calculated by applying the Scherrer equation to the [101] and [110] reflections of anatase and rutile, respectively) and m equals the mass of unreacted precursor In these calculations it is assumed that the particles are spherical in shape Although these calculated values overestimated the specific surface area compared to the experimental results, the theoretical calculations are considered viable for the comparative estimation of surface area changes through the dissolution process The relationship between specific surface area and time was determined to be first-order for both anatase and rutile precursors (see Fig 7) Since the relationship of specific surface area to the observed dissolution kinetic orders varied (first-order compared to zero- and second140 Submitted to Journal of Materials Science Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation Fig First-order relationship of specific surface area to time for anatase (a) and rutile (b) order), the shrinking spheres approximations could not be applied The shrinking spheres approximation assumes that the rate of reaction is dependent only on the surface area of the dissolving solid [45] This indicates that the increase in the dissolving precursor‘s specific surface area did not mediate the dissolution process Although the anatase and rutile precursors have different surface areas and crystallite sizes (9.89 ± 0.03 m2 g–1, 112 ± nm and 2.33 ± 0.02 m2 g–1, 320 ± 10 nm, respectively), these parameters not affect the overall kinetic order of dissolution, rather, they determine the rate constant of the reaction As the TiO2 crystallites were completely immersed in concentrated NaOH, it is reasonable to assume that the TiO2 surfaces become wetted to form a hydroxylated surface through the dissociative chemisorption of water The OH– ions in solution would then compete with the surface hydroxyl groups as ligands towards the metal ions at crystallite surface sites to weaken the critical metal-oxygen bonds [46] Cleavage of the metal-ion centers would then occur once the surface site was sufficiently disturbed The speciation of these hydroxylated Ti4+ metal ion complexes would be dependent on the degree of hydroxylation within the hexa-coordinate environment (e.g Ti(OH)x4–x) Through this process, cleavage of mono- and polytitanic ions is also feasible (e.g TiO2(OH)22– and TinO2n+ m2m–) The Na+ ion in solution could also influence the dissolution of the metal oxide through ion-exchange with the proton of the surface hydroxyls This could result in the formation of partially substituted hydroxylated Ti4+ ion complexes Reassembly of these ions and free Na+ ions through condensation and polymerization processes would most likely produce a sodium titanate material, as proposed previously [18, 19] Since the dissolved Ti4+ complexes and reassembles into a solid product (nanosheet), the hydrothermal liquor would never supersaturate with dissolved Ti4+ This suggests 141 Submitted to Journal of Materials Science Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation that the dissolution of TiO2 proceeds unhindered as nanotube product formation was observed to be favored under the conditions examined in this study The observed dissolution kinetics for the anatase precursor was determined to be zero order whilst the dissolution process of rutile was second order when treated with mol L–1 NaOH at 160 °C Zero order kinetics indicates the dissolution process of anatase was independent of precursor concentration, and that dissolution was not restricted by the solubility of anatase [47] The conversion of rutile has a linear relationship to time when applied to the parabolic diffusion model [48] This suggests that the dissolution of rutile is affected by the diffusion of products away from, and the replenishment of reactant to the surface The variation of surface energies and atomic configurations could influence the dissolution kinetics of anatase and rutile Anatase crystals are dominated by exposed (101) surfaces which can constitute greater than 94% of the crystal surface [41], whereas the predominantly exposed surface of rutile crystals are (110) planes, which can comprise between 50–80% of the crystals surface [41, 49] Although the energies of these surfaces vary, the surface energies of rutile are generally greater than those of anatase [50] This suggests that a greater driving force is required to liberate Ti4+ species from the rutile surface As both anatase and rutile are composed of TiO6 octahedra it is conceivable that the mechanism of detachment of the hydroxylated metal centre species during dissolution from the metal oxide surface would be identical This implies that formation of nanotube structures occurs through the reassembly of similar intermediate building blocks, whether originating from anatase or rutile Thus it is plausible that nanotube formation occurs through the same kinetic process irrespective of the starting material Considering that nanotubes and nanosheets have been observed to form after short treatment durations [51] and from low dissolved Ti4+ concentrations [52], this suggests that nanotubes form readily once Ti4+ species are liberated into solution In this study, nanotubes were observed after all treatment durations for both anatase and rutile precursors After hours hydrothermal treatment ca 20 and 13% of the total crystalline component had dissolved for the anatase and rutile precursors respectively In both samples commensurate quantities of nanotubes and nanosheets were observed in the TEM investigation These results 142 Submitted to Journal of Materials Science Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation agree with the supposition that nanotubes form readily once liberated Ti4+ species are available in solution This then implies that the observed latency in nanotube formation from rutile precursors is caused by the reduced availability of dissolved Ti4+ species in solution due to the slower second order dissolution kinetics compared to the rapid zero order dissolution of anatase This variation in dissolution kinetics significantly influences nanotube syntheses from both natural and synthesized titania precursors as the slower second order dissolution of rutile requires longer processing times than anatase to achieve 100% conversion of precursor to nanotubes This has considerable ramifications towards the commercial viability of high-yield, largescale processing of titanate nanostructures CONCLUSIONS The dissolution kinetics of anatase and rutile and their effects on nanotube formation have been determined by examining products formed following hydrothermal reaction in mol L–1 NaOH at 160 °C for a series of treatment durations Anatase crystallites were observed to be consumed through zero order kinetics, whereas the rutile precursor was consumed through a second order process The dissolution kinetics strongly influenced the formation of nanotubes which formed rapidly, in high yields from the fast, zero order dissolution process of anatase, whilst reduced conversion was observed for the slower, second order process of rutile This dependence of the precursor dissolution on nanotube formation was related to the liberation of Ti4+ species into the solution It was determined through a shrinking spheres approximation that the specific surface area of the precursor did not mediate the kinetic order of dissolution, indicating that the dissolution process was directly related to the phase of the precursor ACKNOWLEDGEMENT The authors gratefully acknowledge financial support from the Australian Institute of Nuclear Science and Engineering through the provisioning of an award (AINGRA07051P) for access to research equipment at the Australian Nuclear Science and Technology Organisation The financial and infrastructure support of the Queensland University of Technology Inorganic Materials Research Program of the 143 Submitted to Journal of Materials Science Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation School of Physical and Chemical Sciences is gratefully acknowledged The Australian Research Council (ARC) is thanked for funding the instrumentation Drs L Rintoul and D Cassidy are thanked for their assistance and expertise with the instrumentation used in this study REFERENCES 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Reyes-Coronado D, Rodríguez-Gattorno G, Espinosa-Pesqueira ME, Cab C, de Coss R, Oskam G (2008) Nanotechnology 19:145605 Fujishima A, Zhang X, Tryk DA (2008) Surf Sci Rep 63:515–582 Maldotti A, Molinari A, Amadelli R (2002) Chem Rev 102:3811–3836 Maira AJ, Yeung KL, Lee CY, Yue PL, Chan CK (2000) J Catal 192:185– 196 Triani G, Evans PJ, Attard DJ, Prince KE, Bartlett JR, Tan S, Burford RP (2006) J Mater Chem 16:1355–1359 Kasuga T, Hiramatsu M, Hoson A, Sekino T, Niihara K (1998) Langmuir 14:3160–3163 Kasuga T, Hiramatsu M, Hoson A, Sekino T, Niihara K (1999) Adv Mater 11:1307–1311 Elsanousi A, Elssfah EM, Zhang J, Lin J, Song HS, Tang C (2007) J Phys Chem C 111:14353–14357 Morgado E, Jr., De Abreu MAS, Moure GT, Marinkovic BA, Jardim PM, Araujo AS (2007) Chem Mater 19:665–676 Lan Y, Gao X, Zhu H, Zheng Z, Yan T, Wu F, Ringer SP, Song D (2005) Adv Funct Mater 15:1310–1318 Morgan DL, Liu H-W, Frost RL, Waclawik ER (2010) J Phys Chem C 114:101–110 Niu HY, Wang JM, Shi YL, Cai YQ, Wei FS (2009) Microporous Mesoporous Mater 122:28–35 Prasad GK, Mahato TH, Singh B, Ganesan K, Srivastava AR, Kaushik MP, Vijayaraghavan R (2008) AIChE J 54:2957–2963 Nian J-N, Chen S-A, Tsai C-C, Teng H (2006) J Phys Chem B 110:25817– 25824 Xu J-C, Lu M, Guo X-Y, Li H-L (2005) J Mol Catal A 226:123–127 Kavan L, Kalbáč M, Zukalová M, Exnar I, Lorenzen V, Nesper R, Grätzel M (2004) Chem Mater 16:477–485 Li J, Tang Z, Zhang Z (2006) Chem Phys Lett 418:506–510 Wang W, Varghese OK, Paulose M, Grimes CA (2004) J Mater Res 19:417– 422 Bavykin DV, Parmon VN, Lapkin AA, Walsh FC (2004) J Mater Chem 14:3370–3377 Saponjic ZV, Dimitrijevic NM, Tiede DM, Goshe AJ, Zuo X, Chen LX, Barnard AS, Zapol P, Curtiss L, Rajh T (2005) Adv Mater 17:965–971 Stumm W, Furrer G (1987) In: Stumm W (ed) Aquatic surface chemistry: Chemical processes at the particle-water interface, Wiley, New York Zhang Y, Walker D, Lesher CE (1989) Contrib Mineral Petrol 102:492–513 Lasaga AC, Luttge A (2001) Science 291:2400–2404 144 Submitted to Journal of Materials Science Chapter 7: Alkaline Hydrothermal Kinetics in Titanate Nanostructure Formation 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Knauss KG, Dibley MJ, Bourcier WL, Shaw HF (2001) App Geochem 16:1115–1128 Finnegan MP, Zhang HZ, Banfield JF (2008) Chem Mater 20:3443–3449 Oskam G, Nellore A, Penn RL, Searson PC (2003) J Phys Chem B 107:1734–1738 Chen Y-F, Lee C-Y, Yeng M-Y, Chiu H-T (2003) Mater Chem Phys 81:3944 Seo D-S, Lee J-K, Kim H (2001) J Cryst Growth 229:428-432 Meng X-D, Wang D-Z, Liu J-H, Zhang S-Y (2004) Mater Res Bull 39:2163– 2170 Yuan Z-Y, Su B-L (2004) Colloids Surf, A 241:173–183 Nakahira A, Kato W, Tamai M, Isshiki T, Nishio K, Aritani H (2004) J Mater Sci 39:4239–4245 Kolen'ko YV, Kovnir KA, Gavrilov AI, Garshev AV, Frantti J, Lebedev OI, Churagulov BR, Van Tendeloo G, Yoshimura M (2006) J Phys Chem B 110:4030–4038 Schmidt J, Vogelsberger W (2006) J Phys Chem B 110:3955–3963 Hill RJ, Howard CJ (1987) J App Crystallogr 20:467–474 O'Connor BH, Raven MD (1988) Powder Diffr 3:2–6 Taylor JC (1991) Powder Diffr 6:2–9 Deng Q, Wei M, Ding X, Jiang L, Ye B, Wei K (2008) Chem Commun 3657–3659 Swamy V, Gale JD, Dubrovinsky LS (2001) J Phys Chem Solids 62:887–895 Burdett JK, Hughbanks T, Miller GJ, Richardson Jr JW, Smith JV (1987) J Am Chem Soc 109:3639–3646 CRC handbook of chemistry and physics (2009) vol 89th CRC Press Inc., New York Lazzeri M, Vittadini A, Selloni A (2001) Phys Rev B: Condens Matter 63:155409/1–155409/9 Wu D, Liu J, Zhao X, Li A, Chen Y, Ming N (2006) Chem Mater 18:547– 553 Bavykin DV, Friedrich JM, Walsh FC (2006) Adv Mater 18:2807–2824 Bavykin DV, Cressey BA, Light ME, Walsh FC (2008) Nanotechnology 19:275604/1–275604/5 Marabi A, Mayor G, Burbidge A, Wallach R, Saguy IS (2008) Chem Eng J 139:118–127 Schindler PW, Stumm W (1987) In: Stumm W (ed) Aquatic surface chemistry: Chemical processes at the particle-water interface, Wiley, New York Chida T, Niibori Y, Tochiyama O, Mimura H, Tanaka K (2004) Mater Res Soc Symp Proc 824:467–472 Tikhov SF, Sadykov VA, Ratko AI, Kouznetsova TF, Romanenkov VE, Eremenko SI (2007) React Kinet Catal Lett 92:83–88 Mendive CB, Bredow T, Feldhoff A, Blesa MA, Bahnemann D (2008) Phys Chem Chem Phys 10:1960–1974 Zhang HZ, Banfield JF (1998) J Mater Chem 8:2073–2076 Yang J, Jin Z, Wang X, Li W, Zhang J, Zhang S, Guo X, Zhang Z (2003) Dalton Trans 3898–3901 Bavykin DV, Walsh FC (2007) J Phys Chem C 111:14644–14651 145 Submitted to Journal of Materials Science CHAPTER CONCLUSIONS 146 Chapter 8: Conclusions CONCLUSIONS This PhD project has expanded the knowledge in the area of nanostructure synthesis through the alkaline hydrothermal method This thorough, fundamental investigation into the influence of synthesis parameters on nanostructures has provided significant insight into the mechanism and influence of competing factors that determine the final titanate nanostructure formation over a broad range of synthesis conditions The findings have contributed to the literature in this field through published works, and augment the overall understanding of the alkaline hydrothermal method This study explored a comprehensive range of hydrothermal conditions to determine the effects of alkaline concentration, hydrothermal temperature, crystallite size and phase on the formation of alkaline hydrothermally produced nanostructures The information gained from this study yielded a map of nanostructure type as a function of these hydrothermal treatment conditions in the form of a morphology-phase diagram It was found that a complex relationship existed between alkaline concentration and hydrothermal temperature, which affected the formation of nanotubes and nanoribbons independently Nanotube formation was primarily affected by hydrothermal temperature, with a critical concentration threshold observed when nanotubes were synthesised at reduced temperatures (100–140 °C) and low alkaline concentrations (5 M NaOH), for Degussa P25 Nanoribbons exhibited a greater dependence on alkaline concentration than nanotubes once a critical temperature threshold was surpassed (>160 °C) This temperature threshold was observed for all the commercial TiO2 precursors treated with alkaline concentrations between and 10 M Crystallite size of the precursors was observed to significantly affect the yields of both nanotubes and nanoribbons at reduced temperatures Degussa P25, the commercial titania precursor with the smallest crystallite size, was always observed to produce considerable quantities of nanostructures compared to when larger near phase-pure particulate anatase and rutile precursors were employed Likewise, greater yields were produced from smaller anatase and rutile crystallites when compared to their larger counterparts This variance in conversion indicated a greater driving force was required when forming nanostructures from larger crystallites, due to the increased solubility of smaller crystallites; which can be explained by the 147 Chapter 8: Conclusions Kelvin effect Also, since larger crystallites have smaller inherent surface areas, this provided fewer surface sites available for dissolution; a process governed by surface interactions and diffusion Thus, a greater rate of reaction is expected, and was observed for smaller crystallites compared to larger crystallites for both anatase and rutile Precursor phase strongly influenced the nanostructure formation with nanotubes forming more readily from anatase than rutile over the full range of conditions explored The influence of phase was also observed to a lesser degree for the formation of nanoribbons unless high temperatures (220 °C) and strong alkaline concentrations (10 M NaOH) were employed To elucidate the effect of phase on nanostructure formation the dissolution kinetics of anatase and rutile were examined These results indicated that the order of dissolution varied considerably between anatase and rutile, which exhibited zero- and second-order kinetics, respectively The zero-order reaction of anatase with NaOH suggested that the dissolution was not restricted by the solubility of anatase, and was independent of precursor concentration Whereas, the second-order dissolution of rutile suggested the process was mediated by diffusion The difference in kinetic order cannot be simply explained by the crystallite sizes of the two precursors as the influence of inherent surface area was determined to have a first-order relationship with time Therefore, the latency in nanostructure formation from rutile precursors was phase related, and was not strictly affected by the crystallite size The data sets of alkaline hydrothermal conditions generated in this study were compiled into morphological phase diagrams which portrayed nanotube and nanoribbon formation through a matrix of alkaline concentration and hydrothermal temperature Morphological phase diagrams were assembled for all three commercial precursors employed in this study These simple, visual interpretations are an effective reference for nanostructure formation, as they indicate the conditions required to form both nanotubes and nanoribbons from anatase and rutile precursors with different crystallite sizes The morphological phase diagram of Degussa P25 effectively contributed to the published literature in this field, obtaining eight citations since publication 148 Chapter 8: Conclusions Within this study, the hydrothermal conditions employed were also selected to help elucidate the phase and formation mechanism of the nanostructures A preliminary study utilised the combination of Raman spectroscopy and XPS for phase identification This study indicated a correlation between the strength and resolution of the characteristic nanotube doublet (between 320 and 220 cm–1) and an increase in binding energy of both the Ti (2p3/2) and O (1s) bands This transition and strengthening of the binding energy related to the transformation of titania precursor into nanotubes and titanate nanoribbons, and the shift from a titania to a titanate binding environment Although definitive phase elucidation by XPS was not achieved (minor variations in binding energies (≤0.7 eV)), the nanotubes were observed to exhibit both titania- and titanate-like characteristics, indicating that variation in phase could occur The results obtained in this study provide strong experimental support for the proposed dissolution/recrystallisation mechanism of nanostructure formation This process satisfactorily explains the influence of phase and crystallite size on nanostructure formation, where it was observed that nanostructures formed more readily from smaller crystallites and from anatase in particular This mechanism was specifically examined in ―Alkaline hydrothermal kinetics on titanate nanostructure formation‖, where dynamic-light scattering was employed to examine the particle size distribution of the product shortly after treatment These results indicated that small particles (300 nm) during analysis, which suggests that nanotubes actively assembled within the filtered hydrothermal liquor In summary, this PhD has systematically examined the influence of synthesis parameters on titanate nanostructure formation through the alkaline hydrothermal method Comprehensive understanding of these parameters and their effects on nanostructure formation is essential for the future application of these structures Advancement of this chemical technology towards commercial viability will require refinement of the synthesis process, to obtain high yields of high quality nanostructures Understanding the effect of alkaline concentration, hydrothermal temperature, crystallite size and phase are fundamental towards attaining this goal This thesis presents a thorough investigation into the effects of these parameters to 149 Chapter 8: Conclusions the wider research community through the three morphological phase diagrams, essentially providing a ‗recipe book‘ for the production of titanate nanostructures Furthermore, the thorough understanding of the kinetic effect of precursor phase dissolution determined in this study will result in the application of cheaper, natural titania sources for high-yield, large scale processing of titanate nanostructures Therefore this thesis contributes significantly towards the greater understanding of titanate nanostructures and their formation 150 ... thesis, Alkaline hydrothermal treatment of titanate nanostructures , investigated synthesis parameters on the formation of nanostructures from titanium dioxide powders through the alkaline hydrothermal. .. 73 CHAPTER DETERMINATION OF A MORPHOLOGICAL PHASE DIAGRAM OF TITANIA /TITANATE NANOSTRUCTURES FROM ALKALINE HYDROTHERMAL TREATMENT OF DEGUSSA P25 74 Statement of Contribution ... hours (E–F) of hydrothermal treatment Figure BET surface areas of materials produced after hydrothermal 136 treatment of commercial anatase and commercial rutile Figure Particle sizes of nanostructures