Open Access
Yoriya, Sorachon
Graduate Program:
Materials Science and Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
September 10, 2010
Committee Members:
  • Prof Craig A Grimes, Dissertation Advisor/Co-Advisor
  • Prof Craig A Grimes, Committee Chair/Co-Chair
  • Elizabeth C Dickey, Committee Member
  • Joan Marie Redwing, Committee Member
  • Thomas E Mallouk, Committee Member
  • Titanium dioxide
  • titania nanotubes
  • anodization
  • anodic oxidation
  • fabrication
  • porosity
  • organic electrolytes
  • in-situ crystallization
  • photoelectrochemical properties
  • biomedical applications
Self-organized, vertically oriented TiO2 nanotube arrays are synthesized by anodic oxidation or electrochemical anodization of titanium foil generally in fluoride ion containing electrolytes. The titanium dioxide nanotube arrays have attracted considerable due to their large surface area, high photocatalytic activity, and photoelectrochemical behavior. TiO2 nanotubes are currently used in a variety of applications such as photocatalysis, biomolecular immobilization, chemical gas sensors, and dye-sensitized solar cells. The ability to precisely control the architecture of TiO2 nanotubes, including wall thickness, pore diameter, wall topology, length, and tube-to-tube spacing is necessary to further enhance device performance and application. In order to repeatedly and precisely control the tube morphology, the inter-related effects of the various anodizing variables including fluoride concentration, anodization voltage, time, concentration of additive species, electrolyte composition, reusing of the electrolyte, and the nature electrolyte medium must be understood to achieve nanotube arrays with desired tube morphology. Anodization in aqueous electrolyte limits the nanotube length to less than 10 &#61549;m, while the polar organic electrolytes are capable of producing the nanotube length relatively longer, to date, up to approximately 1000 &#61549;m using ethylene glycol. Different organic electrolytes commonly produce TiO2 nanotube arrays having different tube morphologies, however, very similar tube geometries can be achieved from different organic electrolytes through optimization of the anodizing parameters and the electrolyte composition. From these findings, it was hypothesized that the fundamental principles controlling the formation process could be similar for different electrolytes, when the effect of the anodizing parameters on the resulting morphology should be clarified in each case to fundamentally understand the formation mechanism of nanotubes as well as the effects of the anodizing parameters on the tube morphology. This dissertation focuses on fabrication and improvement of morphological features of TiO2 nanotube arrays in the selected organic electrolytes including dimethyl sulfoxide (DMSO; see Chapter 4) and diethylene glycol (DEG; see Chapter 5). Using a polar dimethyl sulfoxide containing hydrofluoric acid, the vertically oriented TiO2 nanotube arrays with well controlled morphologies, i.e. tube lengths ranging from few microns up to 101 &#61549;m, pore diameters from 100 nm to 150 nm, and wall thicknesses from 15 nm to 50 nm were achieved. Various anodization variables including fluoride ion concentration, voltage, anodization time, water content, and reuse of the anodized electrolyte could be manipulated under proper conditions to control the nanotube array morphology. Anodization current behaviors associated with evolution of nanotube length were analyzed in order to clarify and better understand the formation mechanism of nanotubes grown in the organic electrolytes. Typically observed for DMSO electrolyte, the behavior that anodization current density gradually decreases with time is a reflection of a constant growth rate of nanotube arrays. Large fluctuation of anodization current was significantly observed probably due to the large change in electrolyte properties during anodization, when anodizing in high conductivity electrolytes such as using high HF concentration and reusing the anodized electrolyte as a second time. It is believed that the electrolyte properties such as conductivity and polarity play important role in affecting ion solvation and interactions in the solution consequently determining the formation of oxide film. Fabrication of the TiO2 nanotube array films was extended to study in the more viscous diethylene glycol (DEG) electrolyte. The arrayed nanotubes achieved from DEG electrolytes containing either HF or NH4F are fully separated, freely self-standing structure with open pores and a wide variation of tube-to-tube spacing ranging from < 100 nm to ~2 &#61549;m. In comparison to DMSO electrolyte, the electrochemical anodization rates are relatively slower in DEG electrolyte; as a result, the nanotube length is typically less than 10 &#61549;m. Pore size of nanotubes grown in DEG has been extended from 150 nm up to approximately 400 nm. The approach to pore widening could be achieved by using a specific condition of low HF concentration and prolonged anodization time. The study of evolution of nanotubes grown in DEG electrolytes showed that a fibrous layer was formed in the early growth stages and then was chemically and gradually removed after a long duration, leaving behind the nanotubes with large pore size. In DEG electrolyte, the closer spacing between Ti and Pt electrodes resulted in the larger nanotube morphological parameters due to the enhanced electrode kinetics facilitating the electrode reactions. The cation choice of fluoride bearing species was found to enhance the nanotube growth rate; with larger cation size leading to the increased tube length of about 20 &#61549;m to be obtained. The relatively larger cation may influence ion solvation behavior in electrolyte, which directly affects the oxide growth and the tube formation. Also the inhibitory effect of larger cation was believed to restrict the barrier layer thickness thus enhancing the nanotube growth rate. The average growth rate in DEG electrolyte is about 0.1-0.3 µm h-1, which is slowest among DMSO (1 µm h-1), FA (2 µm h-1), and EG (15 µm h-1) electrolytes. For a given fluoride ion concentration, pore size (Dinner) and outer diameter (Douter) was found to depend on the anodization voltage as these following equations; Dinner = 0.86 V + 102.9 and Douter = 2.65 V + 137.1. The current behavior observed for DEG electrolyte is different from those observed in aqueous electrolytes, EG, FA, or DMSO. Typically observed in the DEG electrolyte without additives, the anodic current showed a graduate increase with anodization time. When high polar solvents were incorporated, the typical DEG anodization current was also observed in the early period of nanotube growth where the oxidation process dominates. Then the current density starts to decrease when the chemical dissolution begins to dominate. In DEG electrolyte, the effect of electrolyte properties on the resulting nanotube array morphology was quantitatively investigated. Important anodizing parameters including anodizing voltage, time, HF concentration, inter-electrode spacing, and solvent additives were found to have a strong influence on the electrolyte conductivity and the titanium ion concentration dissolving in electrolyte. This work has confirmed that the electrolyte conductivity has a linear relationship with the titanium concentration under a variety of anodizing conditions. Using a fixed Ti sample area of 3.0 cm2, a nominal electrolyte conductivity of 250 µS cm-1 and titanium concentration values of 1200 ppm were achieved. The conductivity-titanium concentration relation was established as a function of anodization voltage, in which the self ordering regime was also built upon this relation based on the construction architecture of nanopores. Using DEG&#61485;HF electrolyte as the model system, this work showed that the degree of self ordering was found to strongly depend on the electrolyte properties; particularly the electrolyte conductivity is a critical factor in the control of nanotube array morphological features. In addition, porosity of titania nanotube films was calculated and porosity regime was created as a function of conductivity of anodized electrolyte. The aim of this part was to provide a new insight into the electrolyte properties-related porosity and the self ordering of pore arrangement affected by the synthesis conditions. From the achieved results, the alumina-like porous structure was found to have porosity values between 18 % and 70 %, when the films were grown in the electrolytes having low conductivity (< 100 &#61549;S cm-1). Whereas the more separated, well-ordered titania nanotube structure, the films grown in higher conductivity electrolytes have % porosity in a lower range of 10 %-30 %. The % porosity also tends to decrease with the increased voltage. A combination of anodization voltage and conductivity was expected to be a self-ordering determining factor that consequently controls the tube formation and porosity of the titania film. The proposed porosity regimes could also be applied to the titania nanotube array films fabricated in other organic electrolytes such as DMSO and EG. Furthermore, this dissertation showed possibilities to crystallize the titania nanotube array films at room temperature via anodization in either DMSO or DEG electrolytes. The partially crystallized films could be achieved specifically in the optimum slow growth process conditions. Due to partial crystallization of the as-anodized samples, the high temperature annealing study revealed that the temperatures of phase transformation are 260 ºC and 430°C for respectively amorphous to anatase and anatase to rutile, which are accounted as the lowest phase transformation temperatures reported to date (2010). Finally, the photoelectrochemical properties of the DMSO fabricated nanotubes were investigated. The maximum photocurrent density of ~ 11 mA cm-2 was achieved by using the 46-&#61549;m long nanotube array sample with completely open pores, and photoconversion efficiencies of 5.425 % (&#61617; 0.087) (under UV light) and 0.197 % (&#61617; 0.001) (under solar spectrum AM 1.5) have been demonstrated. Biomedical applications of the DEG fabricated nanotube arrays films such as blood clotting, hemocompatibility, and drug delivery were investigated. The titania nanotube arrays showed a significant platelet adhesion and activation, a higher viability, and a greater capability in blood clotting compared to a smooth Ti surface. In drug delivery application, the drug elution kinetics, behavior and diffusion of drug molecules were most profoundly affected by the nanotube architectures such as the pore packing density and the gap or separation between the tubes, the nanotube length, and especially the nanotube pore diameter.