Introduction

Literature has shown that TiO2 is an excellent candidate for photocatalytic processes [1]. Specifically, these processes include the photoelectrochemical splitting of water and global environmental issues such as treatment of polluted air and wastewater [2]. The principal photochemical event in this system is the production of holes and electrons upon irradiation. Several research groups have reported higher photocatalytic activity for the anatase phase of TiO2 when compared to the rutile phase [3]. Both the anatase and rutile TiO2 have a tetragonal structure, however rutile is thermodynamically more stable than anatase. Titania nanoparticles can be produced in a variety of ways. Particles produced via sol gel process offer the advantage of obtaining large amounts of powder with high purity [4].

It is always been argued that decreasing particle size improves the catalytic performance due to an increase in surface area. In general, nanoparticles have been under intense examination due to their ability to possess properties that vary from the bulk material. Specifically, titanium dioxide nanoparticles have been shown to possess highly efficient photocatalytic activity [5]. The initial process for heterogeneous catalysis of organic or inorganic compounds by TiO2 is the generation of electron-hole pairs in the particles by UV light. Ultimately resulting in a highly active species that reacts with the adsorbed molecules [6]. The results reported in the literature consistently show the observation of ESR signals due to radical formation upon irradiation. Herein, we have observed ESR signals at room temperature on TiO2 particles prepared in our laboratory without irradiation. This is an important result indicating these ESR signals obtained on these particles may be due to the presence of radical species.

In the present work, TiO2 nanoparticles were prepared by simple hydrolysis method followed by calcination. Results obtained using XRD, PAS/FTIR, Raman and ESR are reported. The ESR signals obtained without irradiation is intriguing and its possible origin is discussed.

Experimental

Synthesis of titanium dioxide was achieved by adding 10 ml of water; drop wise, to a 0.15 M solution of titanium isopropoxide (Aldrich) in ethanol (Aldrich) while stirring. The solution was then dried in an oven at 80 °C. The resulting powder was then calcinated at 200, 250, 300, and 350 °C for 3 h to obtain particle sizes around 4, 5, 6 and 10 nm respectively. Room temperature XRD measurements were done with a Rigaku diffractometer using Cu Kα source (λ = 0.14518 nm). The temperature (4 ≤ T ≤ 300) dependence of the electron spin resonance (ESR) was also performed. The ESR measurements were carried out on a standard reflection-type x-band (9.24 GHz) spectrometer employing a varian cavity magnet system and a variable temperature cryostat obtained from oxford instruments. PAS/FTIR was carried out using a Gold Infinity Mattson FTIR. Micro Raman measurements were performed using a Jobin–Yovin Raman spectrometer.

Results and Discussion

There are several methods reported in the literature for the preparation of TiO2 nanocrystals [3]. One of the simplest methods is the hydrolysis of a precursor such as titanium isopropoxide undergoing hydrolysis to produce titanium oxide nanopowders. The remaining solution was then dried and heated to various temperatures to produce different nanosized TiO2. Below 160 °C the powders were amorphous and begin to form crystalline phases at higher temperatures. This is evident from the XRD pattern indicating various crystalline peaks. Heating the powders to temperatures of 200, 250, 300, and 350 °C transform the amorphous powders to the anatase phase of TiO2 with crystallite sizes of 4, 5, 6, 10 nm respectively. From our previous work utilizing this technique we have determined that the particle size has a fairly narrow distribution typically ±2 nm [7–9]. Figure 1 shows the XRD patterns for TiO2 samples annealed at different temperatures. At higher temperatures exceeding 500 °C the anatase phase transforms into rutile phase. The particle sizes were determined via peak analysis (Sheerer equation) of the XRD patterns and confirmed with TEM (not shown).

Fig. 1
figure 1

XRD patterns for TiO2 samples annealed at different temperatures

Full size image

PA/FTIR spectra were investigated in the 500–4,000 cm−1 range as shown in Fig. 2. Three specific bands are observed around 1,600 cm−1 and a broad band appeared at 3,400 cm−1. The interpretation of these bands for various bulk and surface radical species is important to understand the structural nature of these nanoparticles. It has been previously reported for TiO2 as well as other nanomaterials such as FeOH [10] that the band corresponding to 3400 cm−1 is due to the presence of –OH groups present in the bulk and surface of the particles. However, this band completely disappears when the temperature was raised above 500 °C. The bands appearing around 1,400, 1,600, 1,700 cm−1, respectively may be also be explained due to Ti–O and Ti–OH modes. When heated to 800 °C only one band is observed at 1,000 cm−1 due primarily to Ti–O. Again this indicates the removal of all the hydroxyl species and compares well with the previously reported results [11].

Fig. 2
figure 2

PA/FTIR spectra for different TiO2 samples

Full size image

Raman spectroscopic examinations also show interesting results. Figure 3 shows the Raman spectra for all the heat-treated TiO2 samples. All the observed bands were indexed as anatase phase [12]. There are six possible modes for anatase TiO2 in which three modes A1g, 2B1g, 3Eg are Raman active, two modes A2u + 2Eu are infrared active and one vibration mode B2u will be inactive in both IR and Raman [13]. The Raman spectrum taken for 200 and 250 °C annealed samples show bands at 154, 220, 260, 420, 613 cm−1. The bands appearing at 420, 613 cm−1 are broad and begin to narrow and shift for 300 and 350 °C annealed samples. The band appearing at 516 cm−1 and the disappearance of the 254.7 cm−1 are unique to the 300 and 350 °C annealed samples. Samples annealed above 500 °C showed structural features or the rutile phase.

Fig. 3
figure 3

Raman spectra for all the heat-treated TiO2 samples

Full size image

The most interesting properties are revealed from our research focused on the ESR investigations. Previously, there are several ESR studies on the effects of irradiation on both anatase and rutile TiO2 solid phases [6, 13]. In the present work, we studied the temperature (4 ≤ T ≤ 300) dependence of the ESR. We detected the derivative spectrum and determined the temperature dependence of the three ESR parameters: resonance field H r, peak to peak linewidth ΔH, and the intensity I ≅ (ΔH)2 × h where h is the height between peaks of the derivative spectrum.

Figure 4 shows the ESR spectra (derivative) at 4 K for the TiO2 nanoparticles with sizes of 4, 5, 6, 10 nm. It is evident that the intensity (I) decreases as the particle size increases. The variation of the ESR signal with increasing temperature for 4-nm particle is shown in Fig. 5 specifically showing the decrease in ESR line intensity. As expected for paramagnetic species following the Curie-law, the intensity of the ESR intensity versus temperature decreased as the temperature approaches 300 K. For all the samples, we have carried out a detailed temperature dependence of ESR data. The intensity, which is proportional to the susceptibility (I ∝ χ) is plotted against T for all the samples as shown in Fig. 6. Since I ∝ χ = C/(T  − θ), C and θ can be determined by plotting the intensity data (1/I) versus temperature with the calculated values given in Table 1. The ESR parameters (ΔH, H r, g-value) do not change significantly for various particle sizes. The number of spins versus temperature is plotted in Fig. 7 and indicates that the number of spins is higher for lower particle size. Literature has shown that when TiO2 is irradiated in solution, OH radicals are produced due to photo generated electron-hole pairs as reported in literature [13]. However, the presence of –OH species in these particles confirmed by PAS/FTIR spectra does not appear to contribute to ESR signals. Thus, the ESR signals observed may be attributed to the presence of free electrons associated with the dangling bonds in the nanophase TiO2 rather than the –OH species similar to our previously reported ESR studies on activated carbon [14]. Essentially, the OH radical, with an electron, cannot exist in solid particles. Even in solution the OH radicals can only be detected by ESR if a spin trapping agent is present to form a stable complex with the radical allowing detection [15]. Therefore, the ESR signal is due to the dangling bonds in the nanophase.

Fig. 4
figure 4

ESR spectra at temperature 4 K for the four TiO2 samples with particle size 4, 5, 6 and 10 nm

Full size image
Fig. 5
figure 5

The ESR lines at different temperatures for the sample of 4-nm particle size

Full size image
Fig. 6
figure 6

Plots of ESR intensity (I ≅ (ΔH)2 × h) versus temperature for all four samples

Full size image
Table 1 ESR experimental results
Full size table
Fig. 7
figure 7

Plot of the number of spins/gram versus temperature for all four samples

Full size image

In summary, the TiO2 particles prepared by hydrolysis of titanium isopropoxide following heat treatment at different temperatures produced nanoparticles of various sizes. XRD and Raman confirmed that the anatase phase of TiO2 nanocrystals was achieved. ESR signals in TiO2 are usually observed upon irradiation in solution, however when compared to the previous reports in the literature, the present TiO2 nanoparticles showed ESR signals at g = 2 at room temperature without irradiation. The presence of –OH species in these particles supported by PAS/FTIR spectroscopy does not seem to induce ESR signals and the observed ESR signals are attributed to the presence of dangling bonds in the TiO2 nanoparticles. Thus, the investigation on the controllable nanosized TiO2 prepared by hydrolysis method can be of valuable information for basic and industrial scientific research and applications.