Synthesis and Characterization of Enhanced Photocatalytic Activity with Li+-Doping Nanosized TiO2 Catalyst.

The photocatalytic efficiency of TiO2 is reduced by rapid electron-hole recombination. An effective approach to address this limitation is to have TiO2 doped with various metal ions or heteroatoms. Herein, we prepared a series of Li+-doped TiO2 nanoparticles showing high photocatalytic activities through the sol-gel method. The samples were characterized by X-ray diffraction (XRD) and surface area analyses. Effects of Li+ doping on the Brunauer-Emmett-Teller (BET) surface area, crystallite size, phase transformation temperature, and phase composition were studied. The results showed that Li+ doping can promote the generation of the rutile crystal phase in TiO2, lower the anatase-to-rutile transformation temperature, and generate the mixed-crystal effect. The photocatalytic degradation of methyl orange (MO) was used as a probe reaction to evaluate the photoactivity of the nanoparticles. Parameters affecting the photocatalytic efficiency, including the Li+ doping amount, calcination temperature, and catalyst amount, as well as the kinetics of the photocatalytic process toward the degradation of MO, were investigated. The mixed-crystal TiO2, which was doped with 1.0 mol % Li+ and calcined at 550 °C containing 27.1% rutile and 72.9% anatase phase, showed a 2.2-fold increase in the photoactivity on the basis of the rate constant of MO decomposition as compared with the undoped TiO2. The existence of a definite quantity of rutile phase could effectively inhibit the recombination of the electron-hole pairs, thus promoting photocatalytic activity.

Introduction

Titanium dioxide (TiO2) is considered to be the most efficient and environmental-friendly photocatalyst[1−3] because of its particular combination of low budget, chemical stability, nonpoisonous, high reactivity, favorable stability against photocorrosion, and further amelioration of functionalization.[4] Hence, it has been widely applied in all sorts of applications such as sewage treatment, gas purification, and environmental protection.[3,5,6] There are four natural polymorphs of TiO2, namely, anatase (tetragonal), rutile (tetragonal), brookite (orthorhombic), and TiO2 (monoclinic),[7] of which rutile is more stable than others while three other forms are metastable.[6] It also confirmed that anatase TiO2 has higher photocatalytic activity than rutile TiO2.[6,8] When irradiated with UV light, TiO2 shows strong oxidation property and reducing capacity. The photocatalytic process results from the generation of charge carriers, specifically electrons (e–) in the conduction band and holes (h+) in the valence band. However, the newly photogenerated holes can easily recombine with conduction band electrons, which lowers the efficiency of photocatalytic reactions.[9]

To enhance the photocatalytic efficiency, numerous modification strategies of TiO2 have been carried out.[10−13] Metal-ion doping is considered to be one of the most effective methods.[14−17] Khairy et al.[18] found that doping ions contribute to the increase in the absorption edge wavelength but decrease the band gap energy of TiO2 nanoparticles. Furthermore, doping ions promote the growth of TiO2 anatase, leading to a higher polymorph purity of the TiO2 anatase phase. On the other hand, studies showing detrimental effects of metal-ion doping on the photocatalytic activity of TiO2 have also been reported.[10,19] A study of the influence of transition-metal-doped TiO2 nanoparticles on the environment reveals that the doped nanoparticles can produce acute toxicity to zebrafish in the order of Mn– < Cu– < Ni– ≤ Fe–TiO2.[20] Photocatalytic activity was found to be correlated with the TiO2 phase,[21,22] and the synergistic effect with higher catalytic activities could be obtained using mixed-phase TiO2.[23−27]

We have previously demonstrated that rubidium- and potassium-doped TiO2 catalysts can certainly enhance the photocatalytic activities as compared with undoped one.[19,28] In our present work, the study of Li+-doped TiO2 nanoparticles prepared with the sol–gel method and their photocatalytic activities toward the degradation of methyl orange (MO) under UV irradiation was reported. Specifically, the effects of Li+ doping on the Brunauer–Emmett–Teller (BET) surface area, crystal size, anatase-to-rutile transformation temperature, and phase composition is investigated. Besides, parameters affecting the photodegradation process of MO, including the Li+ doping amount, calcination temperature, and catalyst quantity, as well as the kinetics of the photocatalytic process, will be examined.

Results and Discussion

Characterization of Photocatalysts

X-ray diffraction (XRD) has been commonly used for the identification of the crystal phases of TiO2 and the estimation of the phase composition and crystallite size of each phase present. The XRD patterns of the samples with different Li+ doping concentrations and calcination temperatures are shown in Figure .

Figure 1

XRD patterns of (a) pure TiO2, (b) Li1, and (c) Li2.

“A” and “R” in Figure denote the anatase (diffraction peak of the 101 plane) and rutile (diffraction peak of 110 plane) phases, respectively.[30] The phase content, crystallite size, and BET surface area of the samples are summarized in Table .[31]

Table 1

Phase Content, Crystallite Size, and BET Surface Area of the Samples

	crystallite size (nm)
samples	anatase	rutile	rutile phase content (%)	BET surface area (m2/g)
Li0-400	11.5			100.01
Li0-500	13.5			76.07
Li0-600	24.9			57.84
Li0-650	31.5	16.2	25.7	39.61
Li1-400	13.0			82.56
Li1-500	15.1			73.92
Li1-550	21.8	20.7	27.1	65.95
Li1-600	25.2	40.5	31.5	53.45
Li2-400	13.2			72.15
Li2-500	21.8	28.9	12.4	64.74
Li2-550	26.8	33.7	31.6	52.43
Li2-600	27.8 (28.8)	44.9	45.7	49.26

Only the diffraction peaks of the rutile and anatase phases are observed for these samples. The lack of the characteristic peak of lithium oxide in the XRD patterns implies that either Li+ was incorporated into the crystal lattice of TiO2, or lithium oxide was in very small quantity and homogeneously dispersed.[32,33] The XRD patterns of the samples with different Li+ concentrations present a clear evolution of the anatase-to-rutile ratio of TiO2.

For the Li0 (undoped TiO2) samples, only the anatase phase is generated when the thermal treatment is below 650 °C, as shown in Figure a. The characteristic peak of the rutile crystal phase appears upon increasing the calcination temperature to 650 °C, at which the phase content of rutile is 25.7%. Thus, the phase transformation from anatase to rutile occurs at 650 °C.

For the Li1 (Figure b) and Li2 (Figure c) samples, the peak of the rutile phase appears at the calcination temperatures of 550 and 500 °C, respectively. The corresponding rutile phase contents are 27.1 and 12.4%, respectively. These results indicate that Li+ doping promotes rather than inhibits the generation of the rutile crystal phase. The anatase-to-rutile transformation temperature has been reduced because of Li+ doping; furthermore, the larger the doping amount of Li+, the greater are the temperature drops. These results reveal that Li+ doping has an effect different from that of other alkali metal ions. Our previous experimental results showed that Na+, K+, Rb+, and Cs+ had a significant inhibitory effect on the formation of the rutile crystal phase.[19,28] In addition, Na+, K+, Rb+, or Cs+ doping elevated the phase transformation temperature. However, in the present case, there is no definitive explanation for why Li+ doping promotes the generation of the rutile crystal phase. Nevertheless, a comparison of the radii of Na+ (0.106 nm), Li+ (0.07 nm), K+ (0.138 nm), Rb+ (0.152 nm), and Cs+(0.167 nm) with that of Ti4+ (0.074 nm) shows that only Li+ has a smaller radius than that of Ti4+. Consequently, it appears that Li+ could enter the crystal lattice and replace Ti4+, which presumably lowers the temperature of the rutile phase formation.

On the other hand, Figure and Table show that the relative intensity of the 110 peak and the phase content of rutile increase with the increase of Li+ amount and calcination temperature, whereas the BET surface area decreases. Thus, it is clear that the BET surface area and the crystallite size depend on both the doping amount of Li+ and the calcination temperature.

Figure shows the transmission electron microscopy (TEM) images of (a) undoped TiO2 and (b) Li+-doped TiO2 samples. All of these samples consist of spherical nanoparticles. An average particle size of 30–35 nm for pure TiO2 (i.e., Li0-550) is decreased to 20–25 nm for Li1-550 and other doped TiO2 samples. Figure shows the histogram of the particle size statistics, which shows better statistics than the average particle size. Figure a indicates that the most particle counts were focused on the 40–45 nm, while Figure b illustrates the frequency of all of the particle sizes from Figure , and the result shows that the particle size ranging from 40–45 nm has the largest frequency.

Figure 2

TEM images of the patterns of (a) Li0-550 and (b) Li1-550.

Figure 3

Histogram of particle size statistics: (a) the counts and (b) frequency.

The adsorption–desorption curves and the pore distributions of TiO2 with and without Li+ doping are separately shown in Figures and 5, respectively. As shown in Figure , the average quantity of absorption/desorption at equilibrium is increased from 50.7 cm3/g for undoped to 80.7 cm3/g for Li+-doped TiO2 (Figure a,b). In other word, a 1.6-fold increase is found with Li+ doping. This observation is consistent with the pore radius changes, in which a maximum pore radius of 25 nm for undoped sample is increased to 65 nm for Li+-doped one (Figure a,b).

Figure 4

Adsorption and desorption profiles of TiO2 before and after Li+ doping: (a) Li-500 and (b) Li2-500.

Figure 5

Pore radius distributions of TiO2 with and without Li+ doping: (a) Li-500 and (b) Li2-500.

Figure shows the diffuse reflection of TiO2 samples with different amounts of Li+ doping. For Lix-600, with a gradual red shift in the adsorption with increase in x from 0 to 7%, which is favorable to the absorption and excitation in longer wavelength especially the visible-light region. Further increase in Li+ doping (e.g., x = 7%), however, causes a decrease in red shift. As a result, 5% Li+ doping is considered to be the optimal amount with the absorption band edge lying in the visible region. To make the comparison of the band gap and band edge, the band gap as a function of the amount of Li+ doping for Lix-600 (x = 0–7%) is shown in Figure . From Figure , Eg was increased with the increasing Li+ amount, but the maximum Eg was 2.77 eV, which is larger than 2.73 (x = 7% doping). The result is consistent with the diffuse reflection results. When the doping amount is high enough, the band gap becomes narrow because of the appearance of impurity band and band tail. The result indicates that a higher doping amount is not necessarily better. Thus, doping with suitable amounts of Li+ ions can significantly increase the photocatalytic activity of TiO2.

Figure 6

Diffuse reflection as a function of the amount of Li+ doping for Lix-600 (x = 0–7%).

Figure 7

Band gap as a function of the amount of Li+ doping for Lix-600 (x = 0–7%).

Photocatalytic Activity

Effect of Li+ Doping Amount

Our results above, as well as previous studies,[19,28] have shown that the amount of metal doping and calcination temperature can change the crystal structure, particle size, specific surface area, phase transformation temperature, and phase content. Therefore, they are also expected to be important factors that determine the photocatalytic activity of TiO2.[34−36] The photocatalytic efficiencies corresponding to different doping amounts of Li+ on TiO2 calcined at 550 °C (27.1% rutile phase content) are shown in Figure .

Figure 8

Effect of Li+-doping amount (x) on the photocatalytic activity of Lix-550.

The photocatalytic efficiency improves with the increase in the Li+ doping amount from 0 to 1%, which is followed by a slight decrease from 1 to 2%, and then largely decreases from 2 to 5%. The maximum efficiency at 1% doping level could be mainly attributed to the mixed-crystal effect,[37] in which a mixture of anatase and rutile could provide a greater catalytic activity than a single TiO2 phase. Because the energy band structures of anatase and rutile differ, there is an overlapping band in the mischcrystal grains. Conduction band electrons transfer from the rutile to anatase phase, whereas the photogenerated holes transfer from the anatase to rutile phase, and this transfer of charge carriers inhibits the recombination between electrons and holes. Thus, the photocatalytic activity of the mischcrystal TiO2 is enhanced as compared with those of pure phases. It is interesting to note that only Li+-doped TiO2 samples exhibit the mixed-crystal effect, which was not observed with Na+-, K+-, Rb+-, and Cs+-doped TiO2.[19,28] Thus far, the reason for this difference is unclear.

Although the photocatalytic efficiency starts to decrease when Li+ doping amount is larger than 1%, it keeps holding a value that is higher than that from the undoped TiO2 until x is beyond 3%. This suggests that the rutile phase must be present to a certain extent in the sample with x between 1 and 3% so that charge separation is promoted. However, excessive doping could increase the rate of the electron–hole pair recombination, causing a decrease in the photocatalytic activity,[38] as demonstrated in the present case when x is approaching 5%. Note that, as shown in Table , Li+ doping results in the decrease in the BET surface area, which is supposed to be unfavorable to the catalytic activity. The fact that low amounts of Li+ doping can enhance the photocatalytic activity proves that the catalytic activity of TiO2 depends not only on the specific surface area; more often, it depends on the crystalline structure, crystalline size, and phase composition.

The spectra and effect picture of MO degradation with time are shown in Figures and 10 when the doping amount is 1%. The UV spectra in Figure show that the beginning of 464 nm absorbance value falls faster, while the degradation rate constant of MO is large. The MO almost finished degradation when time reaches 30 min. From the effect picture in Figure , the decoloring effect of the MO solution can be seen obviously when time reaches 20 min.

Figure 9

Spectra of the MO degradation for different times.

Figure 10

Effect picture of the MO degradation (photograph courtesy of Xiaohui Jing. Copyright 2020).

Effect of Calcination Temperature

As mentioned previously, the calcination temperature has a significant effect on TiO2 activity. To find the optimal calcination temperature for the Li+-doped TiO2 catalyst, Li1 samples calcined at different temperatures were used to degrade MO. The results are illustrated in Figure , where the photocatalytic activity of Li1 increases with the increase in the calcination temperature from 400 to 550 °C before it decreases at 600 °C. The XRD patterns shown in Figure reveal that as the calcination temperature increases, the peaks associated with anatase- and rutile-phase TiO2 become sharper, indicating that high calcination temperature promotes the crystal formation and growth. The decrease in the photocatalytic activity at 600 °C could be attributed to reduced surface area and excessive increase in the rutile phase content. Although both Li+-doped and undoped TiO2 have the same optimal calcination temperature of 550 °C, Li1-550 displays a much higher photocatalytic activity probably due to the mixed-crystal effect as a result of Li+ doping.

Figure 11

Effect of calcination temperature on the photocatalytic activity.

Effect of Catalyst Amount

Figure shows the degradation curves of MO at different amounts of Li1-550. It can be seen that as the photocatalyst amount increases from 0.8 to 1.4 g/L, the photocatalytic efficiency of the process increases and reaches the maximum of 99.4% at an irradiation time of 35 min. The enhancement of the efficiency is probably due to (i) the increase in the amount of catalyst, which increases the number of dye molecules adsorbed, and (ii) the increase in the density of particles in the area of illumination.[39]

Figure 12

Effect of catalyst amount on the photocatalytic activity.

The photocatalytic efficiency, however, decreases with a further increase in the catalyst amount. This may be due to the light blocking effect, in which excessive amounts of catalyst could prevent TiO2 from illumination. In addition, the agglomeration and sedimentation of catalyst particles may also have a negative effect on the photocatalytic efficiency.[21]

Kinetic Analysis

Generally, photocatalytic degradation of organic substrates follows the Langmuir–Hinshelwood model.[40] At low substrate concentrations, the dependency of the photocatalytic reaction rate on the concentration of the organic pollutant can be simplified to a pseudo-first-order equation, which can be expressed as follows[41]where C0 is the initial concentration of the reactant (mg/L), C is the concentration of the reactant (mg/L) at time t (min), and kapp is the apparent rate constant (min–1). To examine whether the reaction rate is congruent with a first-order reaction and to determine the extent to which the efficiency of the photocatalytic process could be increased with the optimum doping amount of Li+, ln (C/C0) is plotted as a function of the irradiation time t (Figure ) for the Li0-550 and Li1-550 samples.

Figure shows a good linear relationship between ln (C/C0) and t at the reaction stage for Li0-550 (R2 = 0.9974) and Li1-550 (R2 = 0.9963), indicating that eq can be used to describe the photocatalytic rate of MO. Therefore, the apparent rate constant kapp can be obtained from the slope of the ln(C/C0) vs t curve, which produces a kapp value of 0.057 min–1 for Li0-550 and 0.124 min–1 for Li1-550, respectively. In other words, the photocatalytic activity of TiO2 toward MO degradation is increased by about 120% after Li+ doping.

Figure 13

Relationship between ln(C/C0) and irradiation time for Li0-550 and Li1-550.

Conclusions

Li+ doping promotes the generation of the rutile crystal phase in TiO2 and lowers the anatase-to-rutile transformation temperature. The doping amount of Li+ and calcination temperature have a strong influence on the surface area, crystalline structure, crystalline size, and phase composition. Under the present experimental conditions, the optimal amount of Li+ doping and calcination temperature are found to be 1.0 mol % and 550 °C, respectively.

The mixed-crystal TiO2, containing 27.1% rutile and 72.9% anatase, shows a much higher performance than pure anatase in photocatalytic experiments (mixed-crystal effect). The existence of a definite quantity of the rutile phase inhibits electron–hole recombination and promotes the photocatalytic activity. The photocatalytic degradation of MO follows the pseudo-first-order kinetic model well. The apparent rate constants for pure TiO2 and Li1-550 are 0.057 and 0.124 min–1, respectively. Thus, doping with suitable amounts of Li+ ions can significantly increase the photocatalytic activity of TiO2.

Materials and Methods

Chemicals and Materials

Tetrabutyltitanate (CP), anhydrous ethanol (AR), Triton X-100 (CP), glacial acetic acid (AR), polyethylene glycol 600 (AR), Li2CO3 (AR), methyl orange (MO, AR), and HCl (AR) were purchased from Sinopharm Chemical Reagent Co., Ltd. (China). Doubly distilled water was used throughout this study.

Preparation of Photocatalysts

The Li+-doped TiO2 materials were prepared using a modified sol–gel method. The samples with Li+ mol % α and calcination temperature (°C) β were labeled as Liα-β. For example, Li1-600 means that the doping amount of Li+ is 1.0 mol % and the calcination temperature is 600 °C.

The crystal structure and phase purity of the prepared samples were verified with X-ray diffraction (XRD) using a D8 Advance X-ray diffractometer (Bruker, Germany) with Cu Kα radiation. The crystal size was evaluated using Scherrer’s formula. The BET surface areas were measured on an ASAP 2020 apparatus (Micromeritics) and calculated using the accompanying software. The diffuse reflection was verified with a UV-3600 spectrometer (Shimadzu, Japan), while the total organic carbon (TOC) was characterized with a TOC analyzer (Multi N/C3100 (Analytikjena, German)). The transmission electron microscopy (TEM) images of undoped and Li+-doped TiO2 samples were taken with GeminiSEM 300 (Zeiss, German).

Measurement of Photocatalytic Activity

Photocatalytic degradation was conducted in a thermostatic, cylindrical Pyrex reactor containing 200 mL of MO (initial concentration of 10 mg/L and initial pH of 6.3) and operated at 25 °C. A 300 W mercury lamp (Philips) was used as the light source. Before irradiation, the solution was stirred continuously in the dark for 1 h to achieve an adsorption equilibrium of MO on the catalyst. The absorbance of the MO solution was detected with a TU-1800SPC UV–vis spectrometer (Beijing Purkinje General Instrument Co., Ltd., China) at 464 nm, the maximum absorbance wavelength of MO. The dosage of the catalyst was 1.0 g/L. The photocatalytic efficiency φ was calculated using the following formulawhere A0 and A are the absorbances of the MO solution before irradiation and at time t of irradiation, respectively.

The weight fraction of rutile was calculated using the following equation[29]where WR represents the weight fraction of rutile and IA and IR are the integrated intensities of the anatase (101) and the rutile (110) peaks of X-ray diffraction, respectively.