Improving the Performance of Zinc Oxide Photocatalysts for Phenol Degradation through Addition of Lanthanum Species

One green approach to degrade organic pollutants, such as phenol, is through the photocatalytic reaction. Despite having large band gap energy, which is enough for phenol degradation, zinc oxide (ZnO) has low photocatalytic efficiency. In this study, ZnO was modified by lanthanum (La) species


Introduction
Phenol is an organic pollutant widely used in various industries such as paint, pesticide, polymer synthesis, and petroleum. As phenol is highly soluble in water and has toxicity in low concentrations [1], it is crucial to treat phenol in wastewater. Various methods, including physical and chemical treatments, have been developed to treat phenol. Some physical methods are distillation [2], adsorption [3], and extraction [4], while some chemical methods are Fenton [5], ozonation [6], and photochemical processes [7]. However, these methods still have low efficiency [8], and thus, alternative methods are still required. Recently, many types of research have been extensively carried out to develop phenol removal methods through heterogeneous photocatalysis using semiconductors such as titanium dioxide (TiO2) [9,10,11] and zinc oxide (ZnO) [12,13,14]. This method has several advantages, including efficient mineralization of various organic pollutants such as phenol and its potential to use visible and UV rays from sunlight [15,16].
As a semiconductor with the band gap of 3.37 eV [17], ZnO shall have enough energy for phenol degradation. However, the electrons and holes produced through photon irradiation often undergo rapid recombinations, thus decreasing the photocatalytic activity of ZnO [18]. One way to reduce the rate of recombination and improve charge separation is through doping, which creates a new donor or acceptor energy level between the semiconductor bandgap [19]. The addition of rare earth elements to ZnO for photocatalytic degradation of organic wastes has caused lower recombination rates on ZnO as a result of electron trapping in the acceptor energy levels created through the doping process [20,21].
It has been reported that trivalent lanthanide (Ln 3+ ) dopants in ZnO assisted the creation of superoxide anions and were responsible for the creation of hydroxyl radicals [22]. These two species have been proposed as the primary causes of organic matter mineralization. Special attention has been put into the use of lanthanum (La) as a rare earth element and trivalent lanthanide, which addition gave a red-shift on the ZnO absorption spectrum, leading to the better activity of ZnO for Rhodamine B degradation [22]. The La 3+ has shown the ability to occupy the ZnO crystal lattice as dopant through pyrolysis [22] and electrospinning-calcination method [23]. However, as both methods use high-temperature calcination, the nucleation and crystal growth of ZnO is rapid, resulting in a disruption of ZnO crystal lattice [22]. In addition, the methods often require either expensive equipment or high growth temperatures to achieve quality and crystallinity [24]. Another option of the synthesis method is low-temperature synthesis via a hydrothermal method, which has gained attention for its simplicity, reproducibility, and cost-effectiveness [25]. In this study, the effect of La species addition on ZnO synthesized using hydrothermal method towards its photocatalytic activity under visible and UV light irradiation was studied.

Synthesis of Photocatalysts
The syntheses of undoped and La-doped ZnO photocatalysts were done based on the method reported previously [26]. The Zn(CO2CH3)2.2H2O and NaOH were each dissolved in 75 mL of water at a concentration of 0.1 and 0.2 M, respectively. The Zn(CO2CH3)2.2H2O solution was then added dropwise to the NaOH solution, and the solution was stirred vigorously at 700 rpm for an hour. The mixture was subsequently poured into a Teflon lined autoclave and heated using an oven at 150°C for 6 hours.
The obtained ZnO suspension was separated through decantation. The ZnO slurry was then washed repeatedly using distilled water until the suspension pH reached 7 to get rid of any leftover precursor and byproducts before the drying process using the oven at 60°C for overnight.
Synthesis of La-doped ZnO photocatalysts was carried out similarly to the synthesis of ZnO, except that the synthesis involved the addition of the La(NO3)3.6H2O as the La source. The La source was added into the Zn(CO2CH3)2.2H2O solution to give the La/Zn composition of 0.01, 0.1, 1, and 2 wt%. The solution was then added into the NaOH solution while stirring the solution vigorously at 700 rpm. Similar steps to those mentioned for the synthesis of ZnO were then used to obtain the La-doped photocatalysts.

Characterizations
The undoped ZnO and La(1%)/ZnO samples were characterized using XRD at room temperature. Crystallite size (D) was determined by a Scherrer equation, using the highest diffraction peak of the (1 0 1) plane. The lattice parameter was determined by employing the Bravais-Miller system and the d spacing calculated by Bragg's law. The functional group analysis of all undoped ZnO and Ladoped ZnO samples were investigated using FT-IR by preparing a potassium bromide (KBr) pellet. In order to study the optical and fluorescence properties, the materials were also characterized by the DR UV-vis spectrophotometer, and spectrofluorometer, respectively. All the spectra were recorded at room temperature.

Photocatalytic Tests
To evaluate the photocatalytic activity of ZnO and La-doped ZnO, the photocatalysts were tested for phenol degradation. The photocatalyst (50 mg) was suspended into 25 mL of 50 ppm phenols. The suspension was then stirred for 2 hours under the dark condition to achieve the adsorption-desorption equilibrium, and followed by irradiation using a 150 W halogen lamp (I = 326 μmol/m 2 s) for 3 hours or using an 8 W UV lamp (I = 2 μmol/m 2 s) for 6, 12, 15, and 24 hours at the distance of 12 cm from the photocatalytic reactor. The suspension was then filtered using a filter membrane prior to the measurement of phenol content using HPLC.

Structural Analysis
In order to determine the structure of ZnO, the diffraction pattern of the synthesized ZnO was measured. As shown in Figure 1 1 2), and (2 0 1) planes, respectively. Based on the diffraction pattern, the crystalline ZnO was successfully synthesized, and the structure was in the wurtzite form.
As the representative sample, the diffraction pattern of La(1%)/ZnO was measured and also shown in Figure 1 (a). After the addition of 1 wt% of La, no additional peaks were detected, indicating that no new structure such as lanthanum oxide was formed. Moreover, the doped ZnO still showed high crystallinity, and this result again confirmed that lanthanum doping did not disrupt the crystal lattices of ZnO. The successful doping of La into ZnO crystal lattice was further confirmed from the shifting of diffractogram angles of doped ZnO, as shown in Figure 1 (b). After the La addition, the diffraction angles were shifted to a larger angle due to the replacement of Zn 2+ atom by La 3+ atom in the ZnO crystal lattice. Since the atomic radius size of La 3+ (1.16 Å) is larger than the atomic radius of Zn 2+ (0.74 Å) [23], the diffraction angle of La(1%)-doped ZnO was slightly shifted to larger diffraction angle. Also, the bond distances in the crystal lattice became slightly shorter, which could be seen through the slight decrease in the lattice parameters shown in Table 1. The reasonable decrease in the lattice constants was in good agreement with the previous result [22], which reported that the decrease was due to the mismatch atomic radius size of the La 3+ and Zn 2+ , as well as the reduced oxygen vacancy. On the other hand, the values of α, β, γ were remained to be the same before and after La addition, suggesting that the La-doping did not affect the wurtzite structure of ZnO. The effect of La addition on the crystallite size of the ZnO was investigated by using the Scherrer equation using the main peak at (1 0 1) plane. The addition of La to the ZnO also decreased the crystallite size of the synthesized ZnO from 32.8 to 29.2 nm after 1 wt% addition of La. It was reported that during the ZnO synthesis, La could form La-O-Zn entities on the surface of ZnO, which slowed down the growth of ZnO [27]. The addition of La also distorted the lattice crystal of ZnO, which could inhibit the ZnO growth [23].

Functional Group Analysis
To determine the functional groups in the synthesized ZnO and La-doped ZnO samples, the samples were measured using an FT-IR spectrophotometer. Figure 2 shows the FT-IR spectra of synthesized materials. The undoped ZnO showed the absorption at 3400 and 1639 cm -1 that corresponded to the stretching and bending vibrations of O-H and H-O-H of water [28]. COO-and C-C vibrational peaks from acetic acid at 1415 and 896 cm -1 could also be observed [29]. Acetic acid, as a byproduct in the ZnO synthesis, may have been adsorbed at the surface of ZnO due to acid-base interactions between the acetic acid and the ZnO. On the other hand, the stretching vibration of Zn-O could be detected at 423 cm -1 [30,31]. After the La addition, similar peaks to those of the undoped ZnO could be observed, again confirming that the functional groups in the ZnO were not much affected with the La addition. The vibrational peaks from the acetic acid were found to be increased in intensity after La addition since the basicity of the lanthanum oxides could increase the basicity of ZnO and thus, increased the interaction with acetic acid. This acid-base interaction might prevent acetic acid desorption from the material surfaces despite repeated washing with distilled water after the materials were synthesized. After the addition of 2 wt% lanthanum, the La-O vibrational peak could be detected at 510-530 cm -1 [32]. This could be attributed to the formation of lanthanum oxide on the surface of ZnO, which happened at excessive lanthanum doping concentrations [23].

Optical Properties
To determine the optical characteristic changes of ZnO after the addition of lanthanum, synthesized materials were measured using the DR UV-vis spectrophotometer and spectrofluorometer. The DR UVvis spectra of the materials were measured and shown in Figure 3 (a) as a plot of Kubelka-Munk (KM) function vs. wavelength. The ZnO and La-doped ZnO samples were shown to have strong absorption in the UV region with a maximum peak at 313 nm. The addition of La did not affect the absorption of the maximum peak. However, it gave a blue shift on the ZnO band edge. The blue shift suggested the effect of particle size reduction in the La-doped ZnO samples, creating a quantum confinement effect on the doped material [26]. This result was in good agreement with the crystallite size determination, where La(1%)-ZnO sample showed a smaller crystallite size than the undoped ZnO. In order to determine the band gap energy of the undoped ZnO and La-doped ZnO samples, the Tauc plot was used, as shown in Figure 3 (b). La doping resulted in an increase of band gap energy from 3.22 eV in undoped ZnO to 3.29, 3.29, 3.26, and 3.26 eV after the addition of 0.01, 0.1, 1, and 2 wt% of La respectively. The increase of the band gap again showed the effect of particle size reduction of doped ZnO, as discussed above.
Fluorescence spectra of the undoped and La-doped ZnO are shown in Figure 4. The excitation spectra were measured at an emission wavelength of 558 nm, while the emission spectra were measured at excitation wavelengths of 372-384 nm based on each respective excitation peak, which changed after the addition of La. The maximum excitation wavelength in Figure 4 (a) showed a red shift from 372 to 378, 380, and 384 nm after 0.01, 0.1, and 1 wt% lanthanum addition, which suggested the exchange interaction of sp-d orbitals in ZnO band electrons and localized d electrons of La doping in the semiconductor matrix [33]. This result showed the successful creation of the acceptor level by La in the ZnO band gap. However, after 2 wt% addition of La, the excitation peak was blue-shifted to 356 nm. This could be due to the influence of surface lanthanum oxide at excessive addition of La [23], in which its excitation wavelength was at 300 nm [34].
The yellow emission of ZnO at 558 nm is attributed to the electron transition from the conduction band (CB) of ZnO to the oxygen vacancy (Vo) [35], indicating the presence of oxygen vacancy defect in the synthesized ZnO. By the addition of 0.01 and 0.1 wt% of La, the emission intensity was reduced, which indicated the lower Vo concentration on ZnO after the La addition. The suppression of oxygen vacancies was due to the bond strength differences in Zn-O and La-O bonds. Doping of La 3+ ions into ZnO would create the La-O bonds, which have a higher diatomic bond dissociation energy than Zn-O bonds. The reported values for Zn-O bonds and La-O bonds were ≤ 250 and 798 kJ mol -1 [36], respectively. This difference indicated that the La-O bonds enhanced the average bond strength between oxygen and neighboring cations and decreased the possibility of oxygen-cation dissociation during synthesis, thus decreased the Vo concentration [37]. However, when the addition of La was 1 wt%, the emission intensity was not much affected as it gave a similar intensity to the ZnO, implying an abrupt increase of Vo concentration. This result was in good agreement with the work reported by Park et al. [37]. In contrast, at 2 wt% La addition, the emission intensity was at its lowest one due to the formation of lanthanum oxide at the ZnO surface [23], which could hinder the photon absorption of material, thus decreased the emission intensity as well.

Photocatalytic Activities
Undoped ZnO and La-doped ZnO samples were tested for their photocatalytic phenol degradation under visible and UV light. The photocatalytic activity of synthesized materials under visible light for 3 hours is depicted in Figure 5. After La addition, photocatalytic activity enhancement of ZnO could be achieved, from 0.1% phenol degradation using undoped ZnO to 1.4, 2.8, 6.1, and 5.6% phenol degradation using La(0.01%)/ZnO, La(0.1%)/ZnO, La(1%)/ZnO, and La(2%)/ZnO, respectively. The improved photocatalytic activity could be due to several reasons. The smaller crystallite size of the La-doped ZnO samples could be a sign of smaller particle size, which could give a larger specific surface area of the La-doped ZnO [38]. Another reason was the creation of acceptor energy level by La in the ZnO bandgap, which could inhibit electron-hole recombination in the ZnO [23]. The photocatalytic activity of doped ZnO and Laundoped ZnO photocatalysts under UV light for 6, 12, 15, and 24 hours is depicted in Figure 6. After the addition of 0.01 and 0.1 wt% La, the photocatalytic activity of ZnO declined from 93.6% to 13.1 and 14.1% phenol degradation, respectively, after 24 hours-reaction. This is caused by the decrease of Vo concentration after La doping. As a donor, the existence of Vo can improve electron excitation and the production of hydroxyl radical, which was the primary species responsible for phenol photomineralization [39]. After 1 wt% addition of La, the photocatalytic activity of ZnO gave an increase to 100% phenol degradation after 24 hours-reaction. Referring to its Vo concentration in Figure 4, it was clear that the photocatalytic activity of ZnO under UV light has a strong relation with Vo concentrations. The Vo as donors could act as a recombination center, which improved electron and hole separation [35], resulting in higher hydroxyl radical production and, thus, higher phenol degradation. After 2 wt% addition of La, the photocatalytic activity of ZnO decreased again to 35.2% due to the formation of lanthanum oxide at the ZnO surface [23], which could hinder ZnO photon absorption and limit contact between phenol and ZnO.
Based on the photocatalytic activity tests under visible and UV light, the optimum lanthanum doping concentration on ZnO was 1 wt%. At 1 wt% addition, Vo concentration increased along with a red shift of maximum excitation wavelength due to acceptor energy level introduction by lanthanum [33]. The low doping concentrations (0.01, 0.1 wt%) decreased the Vo concentration on ZnO that resulted in low photocatalytic activity. On the other hand, the excess La doping (2 wt%) limited photon absorption and phenol-ZnO surface contact due to the formation of surface lanthanum oxide [23].

Conclusion
Undoped ZnO and La-doped ZnO samples having La/Zn composition of 0.01, 0.1, 1, and 2 wt% were successfully synthesized by the hydrothermal method. XRD patterns showed that ZnO has a crystalline wurtzite structure, and the structure was not disturbed by the addition of La. However, the diffraction of the ZnO was shifted to a larger angle, indicating the successful substitution of Zn by the La dopant. The crystallite size was also found to be decreased with the addition of La. FTIR spectra detected La-O stretching vibration due to the formation of surface lanthanum oxide at 2 wt% La addition. The fluorescence spectra of La-doped ZnO showed the successful introduction of acceptor energy level by lanthanum doping. La(1%)/ZnO exhibited a higher emission intensity as compared to other doped samples. Photocatalytic tests showed that the best sample having the highest photocatalytic activity for phenol degradation under both visible and UV light was La(1%)/ZnO. Under visible light for 3 hours, the activity of undoped ZnO increased more than sixty times from 0.1% to 6.1%. On the other hand, under UV light for 24 hours, the activity increased from 93.6% to 100%. These improvements of photocatalytic activity were contributed from several parameters, including the reduced crystallite size, introduction of acceptor energy level by lanthanum, and high Vo concentration.