Introduction
Titanium dioxide (TiO2) has been used in various fields such as pharmaceuticals and cosmetics [1-4]. More specifically, TiO2 were used for dye-sensitized solar cells, gas sensors, electrochromic devices, catalyst, light sensitizers, electro chromophores, optoelectronics, and ceramic membrane [1,4,5]. For dental fields, TiO2 was used for surface modification of implant to reduce the bacterial adhesion [6] and also as an ingredient of toothpaste, dental resin or amalgam to improve the tensile strength and bioactivity [7-10].
Various forms of TiO2 nanoparticles (NPs) were generated via the temperature-controlled transformation: amorphous - anatase - rutile [11], which requires dehydration and structural rearrangement of the Ti - O lattice at high temperature [12]. This was achieved via two-stage process: amorphous to anatase conversion by dehydration and rearrangement of the Ti - O lattice, and the precipitation of rutile from the solution of the anatase [13,14]. For the thermal stability, the anatase of TiO2 is at metastable with lower surface energy while the rutile is at stable phase [11,15]. Structurally both anatase and rutile phases have octahedral structure but the crystal units are different from each other with four or two-edge sharing connectivity [12].
Recent studies showed that TiO2 has various properties in optical band gap [16], surface charge, and photocatalytic reactivity [17-20] and these were dependent on the phase composition of TiO2. Thus, understanding the physicochemical properties of individual crystalline phase of TiO2 is important to extend the application of TiO2 NPs into the biomedical field. In this study, we prepared the TiO2 NPs (amorphous, amTiO2; anatase, anTiO2; a mixture of anatase and rutile, arTiO2 and rutile, ruTiO2) with a specific structure and phase composition by using the controlled calcination method. The physicochemical properties of each crystalline phase of TiO2 NPs were analyzed and compared by using scanning electron microscopy (SEM), X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FT-IR), ultraviolet-visible (UV-Vis) spectroscopy, capillary electrophoresis (CE), and photocatalytic dye degradation assay. Here we present the results that different phases of crystalline TiO2 have characteristic properties in crystallinity, particle size, surface chemical property, UV absorption, and photocatalytic activity.
Materials and Methods
1. Chemicals
TiCl4 (99.9%), (NH4)2SO4, NH4OH and bromophenol blue (BPB) were purchased by Sigma-Aldrich (St. Louis, MO, USA).
2. Preparation of TiO2
Various forms of TiO2 NPs were prepared by the hydrolysis reaction of TiCl4 and calcination process. Briefly, 1 mL of cold TiCl4 was slowly added into the vial containing 10 mL of cold water while vigorously stirring. After 30 minutes, the mixture was refluxed at 100℃ for 2 hours. After cooling to room temperature, the mixture was diluted with 30 mL of (NH4)2SO4 (100 mM) and adjusted pH to 7.0 by adding NH4OH (2.5 M). The particles formed were centrifuged at 12,000 rpm for 10 minutes and washed with distilled water (× 5) to remove excess chlorine and sulfate. Finally, the particles were dried at 100℃ for 12 hours to get amTiO2 and further calcinated at 400, 600, and 900℃ for 2 hours (heating rate of 100℃/h) to produce aTiO2, arTiO2 and rTiO2, respectively.
3. FT-IR and UV-Vis spectroscopy
FT-IR spectrum in the range of 400–4,000 cm –1 was measured using a NICOLET 6700 spectrophotometer (Thermo Scientific, Madison, WI, USA). The UV-Vis absorbance was measured by running U-1900 UV-Vis spectrophotometer (Hitachi, Tokyo, Japan). The sample was prepared by suspending TiO2 NPs in water (0.05 mg/mL) and scanned in the wavelength range of 250 to 550 nm.
4. XRD analysis
XRD patterns were recorded by an X’Pert PRO Alpha-1 diffraction system (PANalytical B.V., Almelo, Netherlands) with Cu Kα method (λ = 1.54056 Å). The diffraction was observed in the range of 20° to 80° with a step size of 0.02° and a dwell time of 1 second. The fraction of crystallinity was estimated from the specific peak intensity. The crystallite size was calculated by using the following equation:
where D is the crystal size (nm), K is Scherrer’s constant (K = 0.89), λ is the X-ray wavelength of Cu Kα radiation (λ = 1.54056 Å), β is the line width of half maximum (radian), θ is the Bragg’s diffraction angle.
5. SEM imaging
SEM was performed by S-4800 SEM instrument (Hitachi). The SEM specimens were prepared by loading TiO2 NPs on the SEM-Stub (2.5 cm) with double-stick conductive carbon tape (TED PELLA Inc., Redding, CA, USA), and sputter-coating with Pt ion in a Polaron system (Quorum Technologies Ltd., East Sussex, UK). SEM imaging was obtained under the accelerating voltage of 15 kV with ultrahigh resolution mode (emission current of 20 μA; scanning time of 40 seconds; working distance of 5 mm).
6. CE analysis
CE analyses were carried out by a P/ACE 5500 system (Beckman Coulter, Brea, CA, USA) using a background electrode (BGE) of 10 mM sodium borate pH 9.1 and a fused-silica capillary (60 cm length, 100 µm inner diameter) (Polymicro Technologies, Phoenix, AZ, USA). Electrophoresis was carried out at constant positive voltage (+20 kV), current (50 ± 2 μA) and temperature (25℃). The sample (0.1 mg/mL of TiO2), which was suspended in BGE by sonication for 10 minutes, was injected into capillary by the pressure of 0.5 psi for 3 seconds. The detection was carried out by a photodiode array detector at the wavelength of 319 nm.
7. Dye degradation assay
A 20 mg of TiO2 NPs was added into 50 mL of BPB (0.05 mM) solution (100 mM of Tris-HCl, pH 8.5) and sonicated for 10 minutes. The suspension was stirred for 30 minutes in the dark. Dye degradation assay was carried out by UV irradiation (wavelength at 366 nm; distance of 30 cm) with 15W G30T8 Hg-LAMP (Sankyo Denki Co., LTD, Kanagawa, Japan). After appropriate period of irradiation, 200 uL of suspension was centrifuged (10 minutes at 12,000 rpm), and then the supernatant was used to determine the absorbance at 595 nm by a Multiskan EX spectrophotometer (Thermo Fisher Scientific, Shanghai, China). The percent dye degradation was calculated as follows:
where Ao and At are the absorbance at 0 min and t min of UV irradiation.
Results
1. SEM images
The phase transformation of amorphous-anatase-rutile was carried out by adjusting the calcination temperature. The calcination at 400℃ or 900℃ for 2 hours generated pure anatase and rutile TiO2 while the metaphase TiO2 was formed at 600℃ (Fig. 1). In SEM images, amTiO2 and aTiO2 showed homogeneous shape with estimated particle size at 9 and 13 nm, respectively. On the contrary, TiO2 of metaphase or rutile phase (arTiO2 and rTiO2) showed large particle size (20 nm for arTiO2 and 50 to 60 nm for rTiO2) with heterogeneous morphology.
2. XRD analysis
XRD patterns of prepared TiO2 NPs are shown in Fig. 2A. XRD of amTiO2 was in flat-curve with weak peaks, indicating the amorphous state, not crystalline. The phase transformation of amorphous to anatase and anatase to rutile were confirmed by the change of the peak shapes at 25.2 and 27.1 indicating for anatase and rutile phase, respectively. Also estimated particle size by XRD increased as the calcination temperature was raised (Fig. 2B). The relative size of prepared TiO2 NPs was in the order of amTiO2 < aTiO2 < arTiO2 < rTiO2, which was consistent with the particle size estimated by SEM.
3. Surface chemical property
To understand the surface chemical property of prepared TiO2 NPs, FT-IR spectra were taken (Fig. 3A). All spectra showed the Ti-O-Ti absorption bands at 400–900 cm –1 along with stretching bands of Ti-OH at 3,600 – 3,100 cm –1 and 1,640 – 1,620 cm –1. However, the peak intensity of hydroxyl group was gradually decreased during the phase transformation of amorphous to anatase and eventually disappeared at rutile phased TiO2 NPs.
4. UV-Vis spectrum and CE detection
The UV-Vis spectroscopy was used to analyze the surface property of TiO2 in UV-light absorption. Fig. 3B shows UV-Vis spectra of different crystal types of TiO2 NPs. The maximum absorbance of amTiO2, aTiO2 and arTiO2 was at 309 nm while rTiO2 was blue-shifted to the wavelength of 290 nm with lower absorbance. Interestingly, amTiO2 had higher maximum absorbance than aTiO2 and arTiO2 with the absorbance order of amTiO2 > aTiO2 > arTiO2 > rTiO2. In the CE electropherogram, amTiO2 and aTiO2 were detected as a broaden peak while rTiO2 was not, suggesting that the surfaces of amTiO2 and aTiO2 are electrically active and physically small enough to migrate in capillary (Fig. 3C).
5. Photocatalytic activity
To further understand the photocatalytic activity of different phases of TiO2 NPs, the dye degradation assay was carried out by irradiating UV light to organic dye (BPB) solution in the presence of TiO2 NPs. As shown in the Fig. 4, aTiO2 irradiated with UV light decreased BPB concentration dramatically by the active photocatalytic action while rTiO2 decreased slowly. The relative photocatalytic activity was in the order of aTiO2 > ar- TiO2 > rTiO2.
Discussion
In this study, we investigated the physicochemical properties of TiO2 NPs in different phases (amTiO2, aTiO2, arTiO2, and rTiO2) by analyzing the results of SEM, XRD, FT-IR, UV-Vis spectroscopy, CE, and photocatalytic dye degradation assay. The phase transformations of TiO2 crystal structure, amTiO2 to aTiO2 and aTiO2 to rTiO2 via the metaphase of arTiO2, were carried out by adjusting the specific calcination temperature. Namely, pure aTiO2, metaphase arTiO2 and pure rTiO2 were prepared by setting specific calcination temperature at 400, 600, and 900℃ for 2 hours. Prepared TiO2 crystal phases showed characteristic physicochemical properties in particle size, surface chemical property, UV-Vis absorption, CE migration, and photocatalytic activity. In amorphous - anatase - rutile transformation, XRD data revealed the characteristic rearrangement in structure by the changes of peak position, shape, and intensity. In FT-IR spectrum, the surface of rTiO2 was in dehydration with the weak hydroxyl bands at 3,600 – 3,100 cm –1 and 1,640 – 1,620 cm –1. The UV absorbance of TiO2 phase was blue-shifted with the absorbance order of amTiO2 > aTiO2 > arTiO2 > rTiO2. The small sized aTiO2 (13 nm), which would have larger surface area than rTiO2 (57 nm), showed good mobility under the electric field of CE. For rTiO2, surface dehydration and structural rearrangement caused the aggregation of nanoparticles leading to the large-sized heterogeneous shape with different physicochemical properties. The photocatalytic activity of TiO2 NPs demonstrated that the anatase has the highest catalytic activity with the order of aTiO2 > arTiO2 > rTiO2. The low UV absorbance of rTiO2, probably due to the small surface area, caused poor photocatalytic activity since the small surface generates less reactive species to catalize BPB degradation. Also increased size and poor surface charge of rTiO2 prevented the migration in CE. Overall, we demonstrated that different phases of TiO2 NPs have characteristic physicochemical properties. Understanding the physicochemical properties of individual crystalline phase of TiO2 NPs could be important for the future application of TiO2 NPs in biomedical field.
In conclusion, different phases of TiO2 NPs (amTiO2, aTiO2, arTiO2, and ruTiO2), generated by the hydrolysis reaction of TiCl4 and calcination at a specific temperature, showed characteristic physicochemical properties in size, shape, surface chemical property, UV-Vis absorption, CE migration, and photocatalytic activity.