III. RESULTS AND DISCUSSION

The x-ray diffraction (XRD) patterns of TiO₂ doped with different concentrations of Al (0–5.0%) and calcined at 900 °C are shown in the Fig. 1. The peaks indexed to the reflection from (110), (101), (200), (111), (210), (211), (220), (002), and (310) planes, at 2θ values 27.54°, 36.04°, 39.20°, 41.12°, 44.14°, 54.33°, 56.89°, 62.82°, and 64.17°, correspond to rutile phase of TiO₂ (JCPDS Card No. 86-0148), as well as the reflection from (101), (004), (200), (105), (211), and (204) planes, at 2θ values 25.44°, 38.02°, 48.19°, 53.94°, 54.98°, and 63.06°, correspond to anatase phase of TiO₂ (JCPDS Card No. 86-1157). Figure 1 reveals that undoped sample exhibits characteristic peaks of only rutile phase, while the Al-doped samples (0.1%, 1.0%, 5.0% doped Al) exhibit peaks of both anatase and rutile phases. From the XRD patterns, it is clear that with the increase in Al concentration, the peaks get broadened and phase transformation from anatase to rutile registers decline. Absence of characteristic peaks of Al indicates the good dispersity of the incorporated Al.Reference Yang, Lu, Xu, Yuan, Chang, Zhou and Pan¹⁷

FIG. 1.

X-ray diffraction patterns of (a) 0%, (b) 0.1%, (c) 1.0%, and (d) 5.0% Al-doped TiO₂nanoparticles.

Anatase to rutile phase transformation for undoped TiO₂ mainly occurs at 650–800 °C.Reference Baiju, Shukla, Sandhya, James and Warrier⁴ We have earlier reported the complete transformation from anatase to rutile phase at the transition temperature of 900 °C.Reference Kumar, Verma and Singla¹⁸ In the present work, Al was used as dopant in the TiO₂ nanoparticles. It acts as an inhibitor to phase transformation.Reference Grzmil, Rabe, Kic and Lubkowski¹⁴ The influence of dopant on TiO₂ depends on its concentration, properties of dopant in relation to TiO₂, experimental conditions, and the manner of distribution of dopant. It has been reported earlier that growth of grains of TiO₂ particles is limited if the dopants are present on their surfaces, and because of this, the phase transition temperature from anatase to rutile increases.Reference Periyat, Baiju, Mukundan, Pillai and Warrier¹⁹ In our case, anatase to rutile phase transition temperature is observed at 900 °C due to Al doping; this shows that the dopant is mainly present on the surface of TiO₂.

Table I shows the crystallite sizes of the samples, calculated using the Debye Scherrer’s equation.Reference Mingce, Weimin, Heng and Jun²⁰ The full width at half maximum has been calculated from the most intense peak (Fig. 1), using the Lorentz best fit curve (R ² > 0.98). It is clear from Table I that with the increase in Al concentration, the crystallite size decreases. The weight fraction (W _R) of rutile phase content calcined at 900 °C (Fig. 1) was calculated from the below equation [Eq. (1)]Reference Grzmil, Rabe, Kic and Lubkowski¹⁴:

(1)$$W_{\rm{R}} = {1 \over {(1 + I_{\rm{A}} /KI_{\rm{R}} )}},$$

TABLE I.

Crystallite size, particle size, and band gap energy variation with dopant concentration.

where I _A and I _R are the intensity of most intense peaks of anatase and rutile phase, respectively. K (0.882) is the ratio of the most intense peaks of the complete anatase and rutile phase. The Eq. (1) gives W _R, for rutile phase, as 80%, 20%, and 18%, respectively, for samples doped with 0.1%, 1.0%, and 5.0% of Al.

EDX is used to find elemental composition of Al-doped TiO₂. Figure 2 shows EDX spectrum of the TiO₂ doped with 5.0% of Al, which reveals that the atomic percentage of Ti, O, and Al are 60.77, 34.10, and 5.13, respectively. This confirms that incorporation of Al is quite good in the TiO₂. The EDX spectrum for other Al concentrations, 0.1% and 1.0%, also corroborates the presence of Al in TiO₂.

FIG. 2.

EDX spectra of TiO₂doped with 5.0% of Al.

Figure 3 shows the TEM micrographs of TiO₂ nanoparticles doped with different concentration of Al. The micrographs reveal that the particles are nearly spherical. The average size (Table I) of synthesized particles has been calculated from TEM micrographs by plotting histograms. As observed from the Table I, with increase in the dopant concentration, the size of TiO₂ nanoparticles calcined at 900 °C decreases. It has been well reported in the literature that the size of pure TiO₂nanoparticles increases due to the growth of grain boundary at 900 °C.Reference Kumar, Verma and Singla¹⁸

FIG. 3.

TEM micrograph of (a) 0%, (b) 0.1%, (c) 1.0%, and (d) 5.0% Al-doped TiO₂nanoparticles.

Figure 4 shows the SEM micrographs of TiO₂ particles. As seen in the micrographs, the particles are nearly spherical and in aggregated state. With increase in the doping concentration (0–5.0%), there is a decrease in the aggregation and particles are more spherical. Micrographs reveal that with increase in the doping concentration, the particle size also decreases. This might be due to the presence of dopant. Al is mainly present on TiO₂ surface, which limits the growth of grain boundaries of TiO₂ particles and hence acts as a surfactant; this helps to control the morphology of TiO₂nanoparticles.Reference Grzmil, Rabe, Kic and Lubkowski¹⁴

FIG. 4.

SEM micrograph of (a) 0%, (b) 0.1%, (c) 1.0%, and (d) 5.0% Al-doped TiO₂nanoparticles.

Figure 5 shows the absorption spectra of the TiO₂ nanoparticles, which reveal two facts—first, with increase in Al concentration, the absorption edge gets shifted to shorter wave length (blue shift) and, second, there is an increase in absorption from undoped TiO₂ to 5.0% Al-doped TiO₂. This might be due to larger crystallite size of the undoped nanoparticles than the doped nanoparticles as well as their smoother surface with less number of grain boundaries resulting in less absorption. The appearance of large number of grain boundaries on doping nanoparticles enhance their absorption, which further increases with increase in the dopant concentration.Reference Mingce, Weimin, Heng and Jun²⁰ The incident photons are scattered and lost by absorption; the rate at which the loss takes place will increase as the crystallite size decreases. The band gap energies (E _g) of TiO₂ nanoparticles were calculated using the below equation [Eq. (2)]Reference Zachariah, Baiju, Shukla, Deepa, James and Warrier²¹:

(2)$$$E_{\rm{g}} = {{hc} \over {\bar\mkern-10mu\lambda _{{\rm{int}}} }},$$$

FIG. 5.

Absorption spectra of TiO₂nanoparticles doped with different concentration of Al.

where h is Planck’s constant (4.135 × 10⁻²⁴ eV ns), c the velocity of light (3 × 10⁸ m/s), and $\,\,\,\mathchar\bar\mkern-10mu\lambda _{{\rm{int}}} $ the wave length (in nm) corresponding to the intersection of extension of linear parts of the spectrum of y-axis and x-axis. From Fig. 5, the energy band gap values for undoped and doped TiO₂ nanoparticles are in the range of 3.28–3.42 eV as shown in Table I.

Figure 6 shows the PL spectra of TiO₂ nanoparticles with an excitation wave length of 370 nm. As seen from the spectra, a broad peak at 430 nm is observed for all the samples, irrespective of the doping. This emission band originated from charge recombination on the shallow-trap surface states.Reference Das, Sharma, Kumar and De²² These surface states originate from the oxygen vacancies, which act as radiative centers. The peak around 470 nm for undoped TiO₂ nanoparticles is due to the oxygen vacancy with two trapped electrons, i.e., F center.Reference Li, Haneda, Hishita and Ohashi²³ The peaks around 500 and 505 nm attribute, respectively, to the Al ions in the case of 0.1% and 1.0% doped samples because no such peaks are found for undoped samples. It is difficult to believe that these peak attributed by Al ions due to their low luminescence intensity, but these results are well repeated and checked. No luminescence peak is observed for sample doped with 5.0% of Al; this might be because at higher concentration, Al fills the interstitial sites.Reference Grzmil, Rabe, Kic and Lubkowski¹⁴ Further, the PL intensity decreases with Al doping. Therefore, TiO₂doped with 0.1% of Al shows the least number of surface defects (oxygen vacancies).

FIG. 6.

PL emission spectra of (a) 0%, (b) 0.1%, (c) 1.0%, and (d) 5.0% Al-doped TiO₂nanoparticles, at an excitation wave length of 370 nm.

Figure 7 shows the diffuse reflectance spectra of TiO₂ nanoparticles. As seen from the spectra, undoped TiO₂ nanoparticles show reflectance from 100% to 97.00% in the visible range (400–700 nm). Samples doped with 0.1% and 1.0% of Al show reflectance in the range of 97.00–98.00%, and those, doped with 5.0%, show reflectance in the range of 94.00–97.00%. Thus, it is concluded that with increase in Al concentration from 0% to 5.0%, there is a decrease in the reflectance. These results are also observed in absorption spectra (Fig. 5) as Al doping altered the absorption characteristics of TiO₂ nanoparticles. Figure 7 also shows that doped samples have one reflectance edge in the range of 650–680 nm, which might be due to the Al states on the TiO₂surface.

FIG. 7.

Diffuse reflectance spectra of (a) 0%, (b) 0.1%, (c) 1.0%, and (d) 5.0% Al-doped TiO₂nanoparticles.

Doping of Al (0.1% and 1.0%) improves the crystal structure of TiO₂ by reducing surface defects as well as generating energy states in the TiO₂ energy band as observed in PL (Fig. 6). These changes have not been observed in 5.0% Al concentration, and also, no energy state corresponding to Al was found (Fig. 6). Depending upon concentration, Al can be doped in TiO₂ lattice in two ways.Reference Gesenhues²⁴ At low concentration of Al³⁺, it substitutes Ti⁴⁺ into the cation sublattice and also creates vacancies in the anion sublattice, whereas at higher concentration, Al³⁺ is additionally set on interstitial sites, which corresponds to the clustering of Al defect states.Reference Gesenhues and Rentschler²⁵ Therefore, the vacancy at lower concentration and strong segregation at higher concentrations of Al on the surface of TiO₂lead to decrease in reflectance.

TiO₂ nanoparticles (rutile phase) are mainly used as reflective pigment in reflective coatings, which is an important material for enhancing reflectivity of the light reflectors as well as the reflectance of the laser cavity mirrors.Reference Vurgaftman and Singh^10, Reference Wang, Watson, Sung, Tseng, Bouis and Fernando^26, Reference Prasad, Pinjari, Pandit and Mhaske²⁷ The photostability of nanoparticles has been found to be more when doped; this ensures longer life of the above devices–in fact, an extremely pertinent result.

Figure 8 shows the diffuse reflectance spectra of undoped and doped TiO₂ nanoparticles exposed to light of 20,000 lux for 100 h. The reflectance spectra of 0.1% and 1.0% doped Al show same reflectance spectra (Fig. 8) from 425 to 700 nm. However, in the wave length region from 400 to 425 nm, absorption is observed for all concentrations–not found in unexposed samples (Fig. 7).

FIG. 8.

Diffuse reflectance spectra of (a) 0%, (b) 0.1%, (c) 1.0%, and (d) 5.0% Al-doped TiO₂nanoparticles exposed to light radiation.

On exposing an undoped sample, the reflectance changes from 91.00% to 92.00% (Fig. 8), which, otherwise, had the reflectance from 97.00% to 100% (Fig. 7). This drastic decrease in reflectance is due to generation of surface defects in TiO₂nanoparticles due to exposure to strong illumination, which degrades the photostability of nanoparticles.Reference Jaroenworaluck, Sunsaneeyametha, Kosachan and Stevens^28, Reference Mager²⁹

Pure TiO₂ gives off atomic oxygen when exposed to visible light and thus serves as an oxidant.Reference Mager^29, Reference Whately³⁰ This photooxidation leads to migration of the electrons in the conduction band and holes in the valance band at the surfaceReference Eastaugh, Walsh, Chaplin and Siddall⁵ along with the formation of free radicals.Reference Grzmil, Rabe, Kic and Lubkowski¹⁴ These factors result in increasing absorption and decreasing the reflectance of pure TiO₂ nanoparticles on exposure to light. In doped TiO₂, Al creates trap states for both electrons and holes (Fig. 6), which were generated under the visible light. This makes TiO₂ photostable and prevents deterioration of the reflectance of doped TiO₂ nanoparticles. Hence, Al-doped TiO₂ is better than pure TiO₂. It has also been well reported that mixed-phase nanocrystalline TiO₂ with varying rutile content exhibits synergistic effect between anatase and rutile phase.Reference Zachariah, Baiju, Shukla, Deepa, James and Warrier²¹ This leads to an increase in the photocatalytic activity of the mixture. However, when Al is doped in this mixture, it acts as a quencher of photocatalytic activity but the photostability of mixed-phase TiO₂ increases (Fig. 8). High photostability of TiO₂nanoparticles ensures longer life of the polymeric coatings; in fact, an extremely pertinent result.

A small peak around 550 nm has been observed for the pure TiO₂ nanoparticles in the diffuse reflectance spectra (Fig. 8). This peak is either due to the instrumental disturbance (low signal-to-noise ratio, alignment problems, spatial nonuniformity of the detector, etc.) or characteristics of the sample surface.Reference Hanssen³¹ Former case can be excluded as the experiment was repeated a number of times to check the instrumental disturbance. Sample surface is the factor most strongly affecting reflectance. Therefore, this small peak in the diffuse reflectance spectra is attributed to the surface inhomogeneity, which alters the reflectance of pure TiO₂nanoparticles.

The TiO₂ particles doped with 0.1% Al show high diffuse reflectance (Fig. 7) and high photostability (Fig. 8). Therefore, these nanoparticles are used as reflective pigment for the preparation of coating material. Coatings of these nanoparticles have been applied to plastic substrate. Figure 9 shows the diffuse reflectance of the reflectors with various coating thicknesses. The diffuse reflectance has been found to vary from 93.39% to 93.46%, 96.16% to 96.16%, and 98.17% to 98.29%, respectively, for coating thickness 0.15, 0.20, and 0.25 mm showing that the increase in reflectance is due to high coating thickness, as higher the number of nanoparticles on the substrate, more the reflectance. With further increase in thickness, i.e., from 0.25 to 0.30 mm, there is practically no change in the reflectance (98.17–98.29% to 98.29–98.29%). Thus, the coating thickness, 0.25 mm, has been found to be the optimum one. At the optimum coating thickness, the critical volume pigment concentration plays an important role in determining reflectance. The critical volume is the volume above which no substantial difference in the behavior and appearance of the coating is observed.

FIG. 9.

Diffuse reflectance spectra of designed reflectors with different coating thickness.