Effect of Heat Treatment Process on the Structure and Properties of Nano-TiO2

Nano-TiO2 was prepared with butyl titanate as a precursor by sol-gel method. The samples were analysed by TG-DTA, X-ray diffraction, TEM and so on to assess the effects of different temperatures on the crystal structure, grain size, and microstructure of nano-TiO2. Meanwhile, the catalytic effect of heat treatment temperature on the degradation performance of TiO2 to methyl orange was investigated. The dynamic process of grain growth was preliminarily analysed by Eastman’s particle growth theory. The result shows that TiO2 particle size gradually increases with the heat treatment temperature. At 450 to 550°C, the grain is mainly anatase phase, a mixture of anatase and rutile phase was found at 650°C (mass ratio A:R = 9:1), and the degradation rate of nano-TiO2 on methyl orange reaches 97.75%. When the calcination temperature exceeds 850°C, TiO2 particles almost entirely are composed of rutile phase, and the photocatalytic activity decreases significantly. At 730°C, half of the crystalline TiO2 is transformed from anatase to rutile form. The apparent activation energies of the anatase and rutile crystals of nano-TiO2 are 18.15 kJ/mol and 42.56 kJ/mol, and the fastest grain growth occurs at 546°C and 1280°C respectively.


INTRODUCTION
Photocatalytic oxidation technology has the prominent characteristics of low energy consumption, strong oxidation ability, mild reaction conditions, and reduced secondary pollution when removing pollutants (Liu et al. 2013, Chekem et al. 2019, Varnagiris et al. 2019). It has the potential for broad application in sewage treatment and air purification , Al-Mamuna et al. 2019, Srikanth et al. 2017, and the development of nano-TiO 2 has become an active area of scientific research (Zhang et al. 2014). The sol-gel method can be used to prepare nanometric catalysts with high purity, uniform particle size, and high chemical activity at low temperature, and the reaction conditions are easy to control and the preparation process is relatively simple (Hu et al. 2010, Hakki et al. 2019, Wei et al. 2018. Therefore, the sol-gel method has attracted more attention due to its significant advantages in film formation and doping modification; however, to prepare TiO 2 with high photocatalytic activity, powders with uniform grain size and a low degree of agglomeration are required. Among them, the heat treatment process is one of the important means to promote the transition from amorphous to the crystalline state and adjust the grain size (Liao et al. 2011, Tong et al. 2018, Sun et al. 2017. In this study, TiO 2 particles were synthesised by sol-gel method, and the effects of heat treatment on the crystal structure, particle size, and crystal morphology of TiO 2 particles were analysed by XRD and TEM, and the reasons for TiO 2 phase transition during heat treatment were preliminarily analysed from the perspective of reaction kinetics, and the effect of heat treatment temperature on the photocatalytic performance of TiO 2 has also been discussed.
Sample preparation and characterization: At room temperature, 10 mL of butyl titanate was fully mixed with 10 mL of anhydrous ethanol as Solution A. 4 mL of deionised water, 20 mL of glacial acetic acid, 2 mL of concentrated nitric acid, and 10 mL of anhydrous ethanol were fully mixed Solution B. Under vigorous stirring, Solution B was added (drop-wise) into Solution A, then 0.5 g polyethylene glycol-4000 was added (drop-wise), the hydrolysis temperature was maintained at 40°C, and then stirred for 2 h to ensure the sufficient chelation. The product was dried at a constant temperature 60°C for 24 h and ground into a fine powder. After calcination at different temperatures (450 Vol. 20, No. 1, 2021 • Nature Environment and Pollution Technology to 850°C) for 2 h, TiO 2 powder can be obtained by cooling in the furnace to room temperature.
Using A1 2 O 3 powder as reference material, the composite was subjected to TG-DTA thermal analysis by NETZSCH STA 499C integrated thermal analyser at a rate of 20°C/min in the air; we then used a JEOL 2100 transmission electron microscope (Shimazu, Japan) to observe the microstructure of the composite. A Bruker D8-advance X-ray diffractometer (Germany) was used for phase analysis of the samples; and ASAP 2010 specific surface area and pore size analyser (Micromeritics Co., USA) was employed to determine the specific surface area and total pore volume.
Photocatalytic activity test: The prepared 100 mg TiO 2 (calcined sample at different temperatures and P 25 ) was added to the 200 mL methyl orange solution with a mass concentration of 10 mg/L. After magnetic stirring in the dark for 30 min, the adsorption and desorption of methyl orange on the surface of TiO 2 particles reached equilibrium. The specimen was then irradiated with UV light for 2 h (the main wavelength is 254 nm, the distance between the light source and the liquid surface is 80 mm), taking 10 mL every 0.5 h. After centrifugation, the supernatant was taken and the absorbance measured at the maximum absorption wavelength of methyl orange at 464 nm to allow calculation of the rate of degradation rate h: temperatures and P25) was added to the 200 mL methyl orange solution with a mass concentration of 10 mg/L. After magnetic stirring in the dark for 30 min, the adsorption and desorption of methyl orange on the surface of TiO2 particles reached equilibrium.
The specimen was then irradiated with UV light for 2 h (the main wavelength is 254 nm, the distance between the light source and the liquid surface is 80 mm), taking 10 mL every 0.5 h. After centrifugation, the supernatant was taken and the absorbance measured at the maximum absorption wavelength of methyl orange at 464 nm to allow calculation of the rate of degradation rate : Where, A 0 and A are the absorbances of the reaction solution before and after a certain illumination time. Fig.1 shows the TG-DTA curve of TiO 2 after drying at 105℃ . It can be seen from the DTA curve that there is an endothermic peak between about 50°C and 100℃ , and the mass loss on the TG curve is about 5%. This peak represents the process of water absorption and drainage within the material. There is an exothermic peak between 300°C and 400℃ , and the mass loss on the TG curve is about 12%. This peak is caused by the exotherm caused by the combustion of organic matter in TiO 2 prepared by the sol-gel method. It has been reported in the literature that the powders prepared from titanium alcohols exhibit a strong exothermic peak at 400 to 500℃ , which is the result of the transformation from amorphous TiO 2 to anatase-type TiO 2 ; however, the TiO 2 prepared with butyl titanate in this study only shows a weak similar exothermic peak on the DTA curve. There is a mass loss of 3% on the TG curve between 750°C and 950°C (while the exothermic peak on the DTA curve is not obvious), as attributed to the phase transition of anatase TiO 2 to rutile TiO 2 , indicating that the crystalline transition (anatase-rutile phase) is a slow, gradual process.

…(1)
Where, A 0 and A are the absorbances of the reaction solution before and after a certain illumination time. Fig.1 shows the TG-DTA curve of TiO 2 after drying at 105°C. It can be seen from the DTA curve that there is an endothermic peak between about 50°C and 100°C, and the mass loss on the TG curve is about 5%. This peak represents the process of water absorption and drainage within the material. There is an exothermic peak between 300°C and 400°C, and the mass loss on the TG curve is about 12%. This peak is caused by the exotherm caused by the combustion of organic matter in TiO 2 prepared by the sol-gel method. It has been reported in the literature that the powders prepared from titanium alcohols exhibit a strong exothermic peak at 400 to 500°C, which is the result of the transformation from amorphous TiO 2 to anatase-type TiO 2 ; however, the TiO 2 prepared with butyl titanate in this study only shows a weak similar exothermic peak on the DTA curve. There is a mass loss of 3% on the TG curve between 750°C and 950°C (while the exothermic peak on the DTA curve is not obvious), as attributed to the phase transition of anatase TiO 2 to rutile TiO 2 , indicating that the crystalline transition (anatase-rutile phase) is a slow, gradual process. Fig. 2 illustrates the X-ray diffraction patterns of nano-TiO 2 prepared at different temperatures. It can be seen from the figure that the samples treated at 450°C and 550°C have diffraction peaks near 2θ = 25.3°, 37.7°, 48.0°, 53.8°, 55.1°, and 62.8°, corresponding to the (101), (004), (200), (105), (211) and (204) planes of anatase phase, respectively, indicating that the product is mainly composed of anatase phase; the diffraction peaks are wider than normal, indicating that the particles are small, and the grain development is irregular. Quite sharp diffraction peaks and weak rutile phases appear at 650°C (Daniyal et al. 2019, Zarhri et al. 2020). The characteristic diffraction peaks of the samples treated at 850°C and 950°C were 2θ = 27.4°, 36.0°, 41.3°, 54.3°, and 69.1° corresponding to the (110), (101), (111), (211), and (301) planes of the rutile TiO 2 , respectively, indicating that the products are mainly rutile. The phase change temperature of XRD is the heat treatment temperature when half of the measured at the maximum absorption wavelength of m calculation of the rate of degradation rate :

XRD Analysis
Where, A 0 and A are the absorbances of the reaction solution time. Fig.1 shows the TG-DTA curve of TiO 2 after drying at 105℃ that there is an endothermic peak between about 50°C and 100 is about 5%. This peak represents the process of water absorp There is an exothermic peak between 300°C and 400℃ , and t 12%. This peak is caused by the exotherm caused by the c prepared by the sol-gel method. It has been reported in the lite titanium alcohols exhibit a strong exothermic peak at 400 transformation from amorphous TiO 2 to anatase-type TiO 2 ; h titanate in this study only shows a weak similar exothermic pe loss of 3% on the TG curve between 750°C and 950°C (wh curve is not obvious), as attributed to the phase transition of that the crystalline transition (anatase-rutile phase)  Fig. 2 illustrates the X-ray diffraction patterns of nano-TiO can be seen from the figure that the samples treated at 450° 2θ = 25.3°, 37.7°, 48.0°, 53.8°, 55.1°, and 62.8°, correspo (211) and (204)   anatase changes to rutile. As can be seen from Fig. 2, the phase change temperature is about 730°C.

XRD Analysis
For this TiO 2 the average particle size was estimated using the X-ray width method and the Scherrer formula. The mass percentage X A of anatase phase in the samples was calculated thus (Zhang et al. 2017, Li et al. 2018): X A = I A / (I A + 1.265I R ) × 100%, where, I A is the diffraction intensity of the strongest characteristic peak of the anatase phase on the crystal plane (101), and I R is the diffraction intensity of the strongest characteristic peak of the rutile phase on the crystal plane (110). Table 1 gives the anatase and TiO 2 contents at different heat treatment temperatures and the respective grain sizes: with the increase of the heat treatment temperature, in nano-TiO 2 , the anatase phase gradually changes to rutile phase. In the temperature range from 450 to 650°C, due to the random grain orientation, there are more grain boundary defects and greater activity, structural relaxation was first produced in the grain boundary during the heating process, and the atomic rearrangement occurred, thus making the structure tend to become more orderly thus reducing the free energy of the grain boundary. Therefore, the apparent grain growth is slow within this temperature range; however, when the temperature rises to 650°C, the structural relaxation process ends, the grain boundary migration process begins, and the grain size increases rapidly after entering the phase of grain growth and merger of large grains with small grains. Fig. 3 shows the micro-morphology of nano-TiO 2 at different calcining temperatures. The sample size at 450°C is small (about 10 nm) and the particle boundaries are fuzzy, indicating that TiO 2 has not yet formed complete grains at this calcination temperature, and most of the particles are in an amorphous state. The sample particles at 650°C are intact, the boundary is clear, and the particle size is about 20 nm. The sample particle size at 950°C is coarse (at about 80 nm), the boundary is clear, and the shape is not very regular. The test results are consistent with the analytical results of TiO 2 grain size at different heat treatment temperatures. Table 2 provides the specific surface area and pore structure of the samples at different heat treatment temperatures. As seen from Table 2, the specific surface area and pore volume of the samples prepared at different calcination temperatures decreases with increasing calcination temperature because the increased calcination temperature leads to the gradual growth and refinement of TiO 2 grains and the gradual  Fig. 3 shows the micro-morphology of nano-TiO 2 at different calcining temperatures. The sample size at 450°C is small (about 10 nm) and the particle boundaries are fuzzy, indicating that TiO 2 has not yet formed complete grains at this calcination temperature, and most of the particles are in an amorphous state. The sample particles at 650°C are intact, the boundary is clear, and the particle size is about 20 nm. The sample particle size at 950°C is coarse (at about 80 nm), the boundary is clear, and the shape is not very regular. The test results are consistent with the analytical results of TiO 2 grain size at different heat treatment temperatures.  Table 2 provides the specific surface area and pore structure of the samples at different heat treatment temperatures. As seen from Table 2, the specific surface area and pore volume of the samples prepared at different calcination temperatures decreases with increasing calcination temperature because the increased calcination temperature leads to the gradual growth and  increase in grain size. When treated at 450°C, nano-TiO 2 had its maximum specific surface area and pore volume and, at thereat, the nano-TiO 2 had not yet formed crystals, remaining in an amorphous state with small particles and uniform pores. In the heat-treatment temperature range from 550°C to 850°C, the specific surface area and pore volume of TiO 2 samples gradually decreased with the increase of temperature after roasting. This is because the increase in heat treatment temperature resulted in obvious growth of TiO 2 grains. At the same time, because the small pores were subjected to greater pressure than the large pores during heat treatment, the pores formed by the accumulation of the original TiO 2 particles collapsed or disappeared, gradually forming larger holes. When the temperature reached 950°C, TiO 2 particles grew to 78.6 nm, the small particles merged to form larger particles, and the pore structure decreased in volume, so the specific surface area of the sample reached 3.4 m 2 g -1 with a pore volume of 0.03 cm 3 .g -1 .

Phase Change Kinetic Analysis
According to Eastman particle growth theory: Where, D t is the particle size of the grain at time t (nm); D 0 is the grain size at the initial time (nm); k is constant; n is the reaction index; E represents the apparent activation energy (J/mol); R is the ideal gas constant, and T is the reaction temperature (K); because the grain size of the initial stage of the preparation is very small, D 0 ≈ 0, and t n can also be treated as a constant when the heat treatment time is fixed. Let K = kt n , then formula (2) can be transformed into:

Phase Change Kinetic Analysis
According to Eastman particle growth theory: Where, D t is the particle size of the grain at time t (nm); D 0 is the grain size at the initial time (nm); k is constant; n is the reaction index; E represents the apparent activation energy (J/mol); R is the ideal gas constant, and T is the reaction temperature (K); because the grain size of the initial stage of the preparation is very small, D 0 ≈ 0, and t n can also be treated as a constant when the heat treatment time is fixed. Let K = kt n , then formula (2) can be transformed into: For (3) formula dD t /dT Find the maximum value (d 2 D t /Dt 2 = 0) for equation (5), and we obtain the temperature at which the fastest grain growth occurred T fast Plotting 2lnD v. 1/T gives the straight line shown in Fig. 4, 2lnD and 1/T do not obey a single linear relationship because the TiO 2 crystal type changes from anatase to rutile at around 730 ℃ , and Section AB of the curve is dominated by the anatase phase, while Section BC is dominated by the rutile phase. The lines of best-fit are y = -4.3669x + 10.7697 (R 2 = 0.9252), and y = -10.2403x + 17.2735 (R 2 = 0.9667), respectively. From this, the apparent activation energies E a and E r of anatase and rutile grain growth can be calculated (Table 3). The apparent activation energy of rutile TiO 2 grain growth is larger than that of anatase-type growth. On the one hand, due to the early-stage growth, the TiO 2 particle size is small, lattice defects are large, the particle growth is almost spontaneous, and the energy required is very small. Increased heat treatment temperature, fewer lattice defects (lattice development tends to be complete), and TiO 2 grain growth difficulties, all …(5) Find the maximum value (d 2 D t /Dt 2 = 0) for equation (5), and we obtain the temperature at which the fastest grain growth occurred T fast

Phase Change Kinetic Analysis
According to Eastman particle growth theory: Where, D t is the particle size of the grain at time t (nm); D 0 is the grain size at the initial time (nm); k is constant; n is the reaction index; E represents the apparent activation energy (J/mol); R is the ideal gas constant, and T is the reaction temperature (K); because the grain size of the initial stage of the preparation is very small, D 0 ≈ 0, and t n can also be treated as a constant when the heat treatment time is fixed. Let K = kt n , then formula (2) can be transformed into: D t 2 = kt n exp (-E/RT)= Kexp (-E/RT) …(3) For (3) formula dD t /dT Find the maximum value (d 2 D t /Dt 2 = 0) for equation (5), and we obtain the temperature at which the fastest grain growth occurred T fast Plotting 2lnD v. 1/T gives the straight line shown in Fig. 4, 2lnD and 1/T do not obey a single linear relationship because the TiO 2 crystal type changes from anatase to rutile at around 730 ℃ , and Section AB of the curve is dominated by the anatase phase, while Section BC is dominated by the rutile phase. The lines of best-fit are y = -4.3669x + 10.7697 (R 2 = 0.9252), and y = -10.2403x + 17.2735 (R 2 = 0.9667), respectively. From this, the apparent activation energies E a and E r of anatase and rutile grain growth can be calculated (Table 3). The apparent activation energy of rutile TiO 2 grain growth is larger than that of anatase-type growth. On the one hand, due to the early-stage growth, the TiO 2 particle size is small, lattice defects are large, the particle growth is almost spontaneous, and the energy required is very small. Increased heat treatment temperature, fewer …(6) Plotting 2lnD v. 1/T gives the straight line shown in Fig.  4, 2lnD and 1/T do not obey a single linear relationship because the TiO 2 crystal type changes from anatase to rutile at around 730°C, and Section AB of the curve is dominated by the anatase phase, while Section BC is dominated by the rutile phase. The lines of best-fit are y = -4.3669x + 10.7697 (R 2 = 0.9252), and y = -10.2403x + 17.2735 (R 2 = 0.9667), respectively. From this, the apparent activation energies E a and E r of anatase and rutile grain growth can be calculated (Table 3). The apparent activation energy of rutile TiO 2 grain growth is larger than that of anatase-type growth. On the one hand, due to the early-stage growth, the TiO 2 particle size is small, lattice defects are large, the particle growth is almost spontaneous, and the energy required is very small. Increased heat treatment temperature, fewer lattice defects (lattice development tends to be complete), and TiO 2 grain growth difficulties, all cause the activation energy to increase; on the other hand, as the heat treatment temperature increases further, more of the smaller anatase TiO 2 grains are transformed into rutile phase grains, and some fully-grown grains even undergo a phase transition. Fewer small grains are available for the continuous growth of anatase grains in the samples, so the activation energy of grain growth is greatly increased (Amora et al. 2019, Jithin et al. 2017).

Effect of Heat Treatment Temperature on Photocatalytic Performance of Samples
Fig. 5 demonstrates the heat treatment temperature and its effect on photocatalytic activity: when the heat treatment temperature is between 450 and 650°C, a pairing effect occurs. When the heat treatment temperature is higher than 650°C, the photocatalytic activity of nano-TiO 2 decreases significantly. When the heat treatment temperature reaches 950°C, the removal rate of TiO 2 to the methyl orange aqueous solution is almost zero. This is because, when the calcination temperature is low, the degree of crystallisation degree of anatase is low and the particle size of TiO 2 is small, and the shorter the time required for the photogenic carrier to diffuse from the grain to the surface, the better the photogenic charge separation, and the lower the recombination probability of the electron and hole. On the other hand, the particle size of nano-TiO 2 is smaller, the proportion of surface atoms increases and the light absorption efficiency is higher. In addition, due to the large specific surface area of the nanoparticles and the large contact area upon which the reaction can occur, it is also beneficial to the adsorption of the reactants, which leads to its high photocatalytic activity (Cheng et al. 2016,  XRD analysis shows that P 25 also contained two crystal types (with 650°C calcined TiO 2 being similar therewith). This kind of mixed crystal TiO 2 is not a simple combination of anatase and rutile, but a thin rutile crystal layer grows on the surface of anatase crystal, forming the so-called "mixed crystal effect", which not only does not hinder the absorption of light by the "nucleus", but also because of the rutile TiO 2 having a narrow bandgap (E g = 3.0 eV), it is easily excited by lower-energy light (which has a longer wavelength and is closer to ultraviolet light), owing to the difference in the rutile and anatase band gaps, the hole migrates from the valence band of anatase to the surface rutile valence band, and the photogenerated electron migrates from the rutile conduction band of the surface to the anatase phase, which effectively promotes the separation of photogenerated electron-hole separation in anatase and rutile crystals and reduces the recombination of electron-hole pairs, thus improving the photocatalytic activity of the catalyst (Kuang et al. 2020, Arbuj et al. 2010, Yu et al. 2003. As the temperature continues to rise, the amount of rutile phase continues to increase, TiO 2 particles grow rapidly, and the specific surface area decreases. Furthermore, rutile-anatase high-temperature transition occurs, treatment factors caused by the difference in the surface state of the material arise, surface-active group reduction occurs (on the anatase-type TiO 2 surface with catalytically active groups OHand Ti 3+ ), the TiO 2 surface undergoes a dehydrogenation reaction and the surface hydroxyl groups are lost, whereas hydroxyl groups can capture the photogenerated holes to form highly-oxidised hydroxyl radicals (·OH). In addition, rutile TiO 2 has poor adsorption capacity for O 2 , a lower specific surface area, and photogenic electrons and holes are easily compounded. Moreover, the rutile type material has a low ability to adsorb oxygen, and photo-generated electrons are difficult to remove therefrom, which significantly reduces its photocatalytic performance (Xiao et al. 2010).

CONCLUSIONS
TiO 2 was prepared by sol-gel method, and the effects of heat treatment temperature on TiO 2 crystal structure, particle size, and photocatalytic performance were investigated. In addition, the phase transition of grains was analysed from the perspective of reaction kinetics. The results show that, with increasing calcination temperature, TiO 2 crystallisation was gradually completed, the grains gradually grew in size, and at 450 to 550°C, a thermally stable region was reached, the energy was mainly used for atomic rearrangement, reducing both the number and size of grain boundary defects and reducing the grain boundary free energy, so the particle size increase was small; beyond 650°C, the energy was used for grain boundary migration, and the grains grew rapidly: some grains in the sample were transformed from an anatase phase to a rutile phase. The rate of degradation of methyl orange solution reached a maximum when calcined at 650°C for 2 h. At this time, TiO 2 contained mixed anatase and rutile phases; this indicated that the mixed crystal effect of TiO 2 can effectively inhibit photogenic electron-hole recombination and that the sample has a better photocatalytic performance than single anatase-TiO 2 .
the sample prepared in this study, under the same conditions, had a P 25 value for methyl oval of up to 98.63%). analysis shows that P 25 also contained two crystal types (with 650°C calcined TiO 2 being ewith). This kind of mixed crystal TiO 2 is not a simple combination of anatase and rutile, utile crystal layer grows on the surface of anatase crystal, forming the so-called "mixed ct", which not only does not hinder the absorption of light by the "nucleus", but also the rutile TiO 2 having a narrow bandgap (E g = 3.0 eV), it is easily excited by lowert (which has a longer wavelength and is closer to ultraviolet light), owing to the generated electrons are difficult to remove therefrom, which si performance (Xiao et al. 2010).

CONCLUSIONS
TiO2 was prepared by sol-gel method, and the effects of crystal structure, particle size, and photocatalytic performanc phase transition of grains was analysed from the perspective o that, with increasing calcination temperature, TiO2 crystallis grains gradually grew in size, and at 450 to 550°C, a thermally was mainly used for atomic rearrangement, reducing both the defects and reducing the grain boundary free energy, so the par 650°C, the energy was used for grain boundary migration, and in the sample were transformed from an anatase phase to a rut methyl orange solution reached a maximum when calcined a contained mixed anatase and rutile phases; this indicated that effectively inhibit photogenic electron-hole recombination