Min
He
a,
Defa
Li
a,
Yu
Liu
a,
Taohai
Li
ab,
Feng
Li
*ab,
Javier
Fernández-Catalá
bc and
Wei
Cao
*b
aCollege of Chemistry, Key Lab of Environment Friendly Chemistry and Application in Ministry of Education, Xiangtan University, China. E-mail: fengli_xtu@hotmail.com
bNano and Molecular Systems Research Unit, University of Oulu, P.O. Box 3000, FIN-90014, Finland. E-mail: wei.cao@oulu.fi
cInorganic Chemistry Department, Materials Science Institute, University of Alicante, Ap. 99, Alicante 03080, Spain
First published on 11th March 2024
FeNbO4 sonocatalysts were successfully synthesized by a simple hydrothermal route at pH values of 3, 5, 7, 9 and 11. The catalysts were characterized by XRD, XPS, TEM, SEM, N2 adsorption and DRS to analyse the effect of pH parameters on the physicochemical properties of the materials during hydrothermal synthesis. The sonocatalytic activity of FeNbO4 microspheres was evaluated by using acid orange 7 (AO7) as the simulated contaminant. The experimental results showed that the best sonocatalytic degradation ratio (97.45%) of organic dyes could be obtained under the conditions of an initial AO7 concentration of 10 mg L−1, an ultrasonic power of 200 W, a catalyst dosage of 1.0 g L−1, and a pH of 3. Moreover, the sonocatalysts demonstrated consistent durability and stability across multiple test cycles. After active species capture experiments and calculation of the energy band, a possible mechanism was proposed based on the special Fenton-like mechanism and the dissociation of H2O2. This research shows that FeNbO4 microspheres can be used as sonocatalysts for the purification of organic wastewater, which has a promising application prospect.
Advanced oxidation processes (AOPs) have been proved to be effective in degrading various pollutants.10–12 Sonocatalytic degradation technology is a type of AOP that relies on the simultaneous presence of an appropriate semiconductor catalyst and ultrasonic radiation.13 The hydroxyl radicals (˙OH) generated during the sonocatalytic process can more effectively convert organic pollutants in water into carbon dioxide, water, and inorganic ions, or less toxic and more easily degradable substances.14–17 However, the degradation of organic wastewater by single ultrasonic irradiation is not satisfactory. Studies have shown that the combination of two or more treatment techniques is more economical and effective than the single ultrasonic degradation technology.18,19 The combination of heterogeneous Fenton-like catalysis and sonocatalysis is currently being investigated and is considered another possible way to increase free radical production in the system, with a great enhancement in terms of improving the degradation.17,18
So far, a series of metal complex oxide niobates (EuNbO4, BiNbO4, AgNbO3, and FeNb2O6) have been discovered and applied in the fields of photocatalysis, electrocatalysis, and electrolysis.20–22 However, there are few reports on the sonocatalytic activity of niobates.23–25 FeNbO4 is a metal oxide type of ABO4 with good magnetic and electrical properties, unique narrow band gaps, and excellent photocatalytic properties.26,27 It has recently attracted attention for its potential applications in photocatalysis, photodetectors, and gas sensors.28–30 However, the conditions for the synthesis of niobates are relatively harsh, and the traditional high-temperature solid-phase method for the synthesis of FeNbO4 requires temperatures of more than 1000 °C.31 The samples obtained by this method are characterized by large particle size, low crystallinity, and phase mixing, which can lead to the low catalytic activity of niobates, and greatly limit their application in catalytic water treatment.30,32 Therefore, the scientific community has been developing new synthesis strategies in recent years. Up to now, several attempts to reduce the synthesis temperature of FeNbO4 have been reported. Recently, Devesa et al.33 reported the synthesis of FeNbO4 with monoclinic and orthorhombic structures by the sol–gel method. The nanomaterials prepared by this method can be mixed up to the molecular level, but the secondary phase of Fe2O3 has been present with the increase of the heat treatment temperature from 400 °C to 1200 °C. Wang et al.34 prepared FeNbO4 nanorods with a diameter of 100 nm and a length of 1–3 μm by electrostatic spinning. The advantage of this method is that adjustable fibers can be prepared, but the yield is low. Shim et al.35 produced FeNbO4 nanopowder with a smaller particle size and a larger specific surface area under hydrothermal reaction conditions at 250 °C. This method avoids the problems of oversized grains, crystal defects, and the introduction of heteroatoms caused by the calcination method at high temperatures. However, a high hydrothermal temperature places stringent requirements for the preparation equipment. Ahmed et al.36 obtained the finer and uniformly distributed FeNbO4 by co-precipitation of reaction regents of ferric nitrate and ammonium niobate oxalate hydrate. The process was followed by annealing at 1100 °C for 6 h. Besides the niobate, the accompanied NbO2 phase was also detected. In addition, almost all of the above syntheses of FeNbO4 are based on the more expensive niobium pentachloride and requirements of heat treatments above 900 °C. Therefore, the study of new methods to achieve simple, economical and controllable synthesis reactions at low temperatures or the improvement of existing methods is an urgent need for the application of this material. At present, there is no report on the sonocatalytic activity of FeNbO4. The FeNbO4 metal complex oxide semiconductor is expected to be applied to degrade pollutants in the field of acoustic catalysis.
Based on the above discussion, FeNbO4 microspheres were prepared in this study by a simple hydrothermal method (200 °C) under different pH conditions, using low temperatures and non-toxic reagents. The FeNbO4 microspheres were used as Fenton-type catalysts for the first time to degrade acid orange 7 (AO7) under ultrasonication (US) with H2O2 as the oxidant. In this work, the effects of catalyst, catalyst dosage, pollutant pH, and pollutant concentration on the performance of degradation of AO7 were investigated. The results show that FeNbO4 particles have an effective synergistic effect in the presence of ultrasonic radiation and H2O2. In addition, the possible sonocatalytic degradation mechanism of AO7 in the presence of FeNbO4 is preliminarily proposed.
To explore the role of active species in the sonocatalysis process, the experiments of radical capture were carried out. The influence of active species to the photocatalytic degradation of tetracycline was investigated by adding 1 mM of isopropyl alcohol (IPA), potassium iodide (KI), and 1,4-benzoquinone (BQ) as scavengers to detect the hydroxyl radical (˙OH), hole (h+) and superoxide radical (˙O2−), respectively, before the photodegradation experiment.
![]() | ||
Fig. 1 XRD patterns of FeNbO4 samples. • Indicates the (012) and (220) crystallographic plane positions of Fe2O3 (JCPDS No. 33–0664). |
The morphology of the samples prepared in this work by hydrothermal synthesis (FN3, FN5, FN7, FN9, and FN11) has been studied by SEM and TEM analyses. Fig. 2a–e shows the SEM images of FN3, FN5, FN7, FN9, and FN11 samples, respectively. The SEM images of FN3, FN5 and FN7 samples show that under basic and neutral conditions the samples present an agglomeration of a laminar structure to micro-sphere particles (∼1 μm) due to their self-assembly.40Fig. 2a–e show that the particle size of the sample tends to increase with the increase in pH of the reaction medium under acidic conditions, and the surface tends to be smooth and sharp-edged. The size of 20 individual FNO particles was measured statistically using the Nano Measure 1.2 software. The average particle sizes of FN3 and FN5 are 0.92 μm and 1.14 μm, respectively. The FN7 (Fig. 2c) sample presents an average size of 1.53 μm, indicating an increase in the size of the particles when the pH in the medium of reaction was increased until pH = 7. However, when the reaction solution became alkaline, the samples have a heterogeneous morphology based on flocculated balls with a particle size of about 200 nm (Fig. 2d) for the FN9 sample and approximately 100 nm for the FN11 sample (Fig. 2e). These SEM results show a clear change in the morphology of FeNbO4 samples prepared in this work, indicating that the pH parameter in the hydrothermal synthesis affects the final morphology of the materials.41,42
The TEM images and HRTEM images of FN3 (Fig. 3a and b) and FN9 (Fig. 3c and d) samples are shown in Fig. 3. TEM analysis reveals that the FN3 sample presents a micro-spheric morphology and a particle size of around 1 μm (see Fig. 3a). The FN9 sample has a heterogeneous morphology based mainly on the spheric morphology with a particle size of about 200 nm (Fig. 3c). The results of TEM are in agreement with the SEM results. The HRTEM images (Fig. 3b and d) show the lattice fringes of the FN3 and FN9 samples, indicating high crystallinity of the prepared samples. The sample presents a lattice spacing can of 0.332 nm, which corresponds to the (110) crystal plane of the tetragonal FeNbO4 (JCPDS No. 16-0357). These results are in agreement with the XRD results obtained for this material.
The porous texture of the synthesized samples was investigated by N2 adsorption measurements (Fig. 4). The N2 adsorption–desorption isotherms of the samples prepared in this work (FN3, FN5, FN7, FN9) are shown in Fig. 4. Fig. 4 shows the isothermal adsorption curves of the samples; in the low-pressure region, an increase in the amount of N2 adsorbed is not observed. At higher relative pressures, an increase in the N2 adsorption is observed, showing pore filling. This effect is the characteristic of isotherms of type-III and hysteresis of type H3,30 indicating that FeNbO4 is a non-porous material. The FN9 sample presents the highest uptakes at low and high relative pressures than the FN3, FN5, and FN7 sample, indicating that the alkaline media in the synthesis favour the porosity in FeNbO4 materials. This fact was also evidenced by SEM and TEM analyses because the FN9 sample presents lower size particles with respect to FN3, FN5, and FN7 samples.
Concerning the textural properties shown in Table 1, the increase of alkaline conditions in the synthesis medium increases the surface area of the materials. In this sense, the specific surface area of FN9 was slightly larger than that of the other samples. This fact may be caused due to the small particle size of FeNbO4 under alkaline conditions as it was observed by SEM and TEM analyses. The pore sizes of FN3, FN5, and FN7 are decreased due to a smoother surface of the samples, as evidenced by the SEM analyses. However, the FN9 sample under alkali conditions presents a higher pore size. This fact evidences that the pH parameter affects the porosity and pore size of FeNbO4, as was observed by N2 adsorption–desorption isotherms.
Samples | BET surface area (m2 g−1) | Total pore volume (cm3 g−1) | Average porous size (nm) |
---|---|---|---|
FN3 | 12 | 0.036 | 12.7 |
FN5 | 31 | 0.043 | 5.9 |
FN7 | 29 | 0.022 | 3.9 |
FN9 | 44 | 0.12 | 17.1 |
The optical absorption properties of all samples were evaluated by UV-vis diffuse reflectance spectroscopy (Fig. 4c). In the figure, all samples showed a wide range of optical absorption. The intensity of optical absorption by the samples decreases with the preparation pH, with the FN3 sample exhibiting the maximum optical absorbance. It is also observed that the optical absorption becomes stronger when the pH increases to 11 (the FN11 sample). This may be due to the presence of a large amount of the Fe2O3 phase. The bandgap was calculated by the Tauc plot analysis in Fig. 4c, obtaining a value of 2.13 eV. This fact indicates that FeNbO4 is a semiconductor with an intermediate band gap, which is a very interesting property for sonocatalysis. The MS curve in Fig. 4e has a positive slope, indicating an n-type semiconductor characteristic.43 The flat band potential of the sample was determined to be −0.06 V (vs. Ag/AgCl) by extrapolating the linear part of the MS curve. Since the conduction band edge is usually close to the flat band potential, the band gap of the UV-vis absorption spectrum can be used to deduce the conduction band position of the FN3 sample. The bottom of the n-type semiconductor conduction band was typically about 0.2 V negative than the flat band potential, so the minimum value of the sample conduction band (CB) is estimated to be −0.26 V, and the maximum value of the valence band (VB) is estimated to be 1.87 V.44
X-ray photoelectron spectroscopy (XPS) analysis was used to determine the chemical composition and valence estates of the elements present in the prepared samples. The XPS survey showed that the main compositions of the FN3 sample were Fe, Nb, O and C elements (Fig. 5a), which contains 47.37% of O, 13.07% of Nb, 6.33% of Fe, and 33.23% of C. The high-resolution Fe 2p spectra (Fig. 5b) showed the presence of Fe2+ and Fe3+ in FeNbO4, two characteristic peaks located at 709.7 eV and 723.3 eV corresponded to 2p3/2 and 2p1/2 of Fe2+, respectively, and the two peaks at 715.6 eV and 729.1 eV were satellite peaks of Fe2+. In addition, two peaks attributed to 2p3/2 and 2p1/2 of Fe3+ were observed at 711.9 eV and 725.5 eV. Similarly, two peaks at 719.8 eV and 732.5 eV were satellite peaks of Fe3+.18,45 The results suggest that an oxidation/reaction process like-Fenton reaction may be occurring between Fe3+ and Fe2+.46 Nb 3d XPS spectra (Fig. 5c) show two peaks at 207.1 eV and 209.8 eV matching with Nb 3d5/2 and Nb 3d3/2, and this fact confirms the presence of Nb5+ in the samples,47 indicating that FeNbO4 was synthesized successfully. The O 1s XPS spectrum shows two peaks at 529.8 eV and 531.2 eV in Fig. 5d. The first one corresponds to O2− ions at the lattice region and the second peak indicate oxygen deficiency regions.43
Fig. 6c shows the effect of different amounts of FN3 samples for catalytic degradation of AO7 under ultrasonic irradiation. In Fig. 6c, the catalytic conversion by three different concentrations was roughly the same, and the degradation rates were 61.55%, 65.59%, and 66.48%, respectively for 0.5 g L−1, 1 g L−1, and 1.5 g L−1. The increase in the catalyst concentration promotes the increase in the number of active factors on the catalyst surface, which will produce more active substances to participate in the degradation of AO7 solution and improve the degradation rate. 1 g L−1 of the catalyst is sufficient to provide sufficient active sites for promoting the formation of free radicals.48 The reaction rate constant of AO7 degradation is fitted as shown in Fig. 6d. The kinetics of the degradation reaction considering the effect of the catalyst concentration shows first-order kinetics.
To study the effect of the pH under ultrasonic catalysis performance, the pH values of the AO7 solutions were adjusted to 3, 7, and 11 by adding hydrochloric acid and sodium hydroxide solutions (catalyst dosage 1 g L−1 and dye concentration 10 mg L−1), respectively. As shown in Fig. 6e, the degradation rates of the dye solutions at pH = 3, 7, and 11 were 97.45%, 65.59%, and 56.08%, respectively, after 180 min of reaction. Fig. 6f shows that this parameter using the FN3 sample presents first-order kinetics. The high activity for the sample at acidic pH (pH = 3) is due to the high H+ concentration present in the solution. Then, the sulfo group in the azo dye is more easily removed, because of the high generation of radicals under this condition. Moreover, the electron-withdrawing properties of H+ make the unsaturated bond in the azo dye more easily broken, and thus the dye molecule is more easily decomposed. Therefore, the degradation effect of AO7 under acidic conditions is significantly higher than that under neutral or alkaline conditions.
Fig. 6g depicts the experimental results of ultrasonic catalysis when varying dye concentrations. The degradation rates of dye concentrations of 5 mg L−1 and 10 mg L−1 show high catalytic activity after 180 min of reaction, which are 93.84% and 97.45%, respectively. In this regard, the highest degradation rate was achieved when the dye concentration was 10 mg L−1. When the dye concentration was 20 mg L−1, the degradation rate decreased until 61.80% after 180 min of ultrasound irradiation. This effect might be due to a higher concentration, and the removal of AO7 was blocked due to the saturation of the adsorption and active sites. As it was mentioned for another parameters, the degradation kinetics of different dye concentrations (Fig. 6h) are consistent with the first-order kinetics.
Fig. 6i shows the comparison degradation effects under different processing conditions, such as (1) blank, (2) only H2O2, (3) Fenton condition, and (4) H2O2 + catalyst. The amount of the catalyst FN3 was 1 g L−1, the pH of the AO7 solutions was 7, and the solution concentration was 10 mg L−1. Without the presence of the catalyst (FN3) even using the H2O2 or Fenton process, there is no degradation of the AO7 dye. The incorporation of the catalyst (FN3) in the reaction media under sonification shows an increase in the degradation of AO7. When subjected to ultrasonic radiation, FeNbO4 and H2O2 were present together, the degradation effect was greatly improved. This effect may be due to the Fenton-like reaction introduced in the reaction system after the addition of H2O2.49 The addition of H2O2, which reacts with holes to form ˙OH, initiates the reaction process of FN3 as a semiconductor catalyst and reduces the electron–hole rate recombination.
The recoverability and reusability of the sonocatalysts are important indicators for practical applications. Since the FN3 sample had the best sonocatalytic activity for AO7, the FN3 sample was chosen for the recycling experiment. The FN3 sample was recovered for recyclability experiments using the optimal experimental conditions because this catalyst exhibits the best sonocatalytic activity for AO7 and it is a pure material. The experimental results are shown in Fig. 7a and b. After 5 cycles, the degradation rate of the FN3 sample was still high, indicating that the FN3 sample has high recyclability.
In this study, the prepared catalysts were compared with those reported in the literature for the removal of AO7 at different types of energy sources. The results are shown in Table 2. First, FeNbO4 catalyst applications in ultrasonic catalysis are still rare. Second, in the case of ultrasonic energy, the removal efficiency from pure FeNbO4 was higher than other composites used in this application. The catalyst prepared in this study has a high removal efficiency of the azo dye under the excitation of ultrasound similar to another energy sources such as light, with makes this catalyst and this technology interesting for a future application.
Catalysts | Ultrasonic power (W) | Concentration (mg L−1) | Catalyst dosage (g L−1) | Degradation (%) | Time (min) | Ref. |
---|---|---|---|---|---|---|
FeNbO4 | 200 | 10 | 1 | 92 | 120 | This work |
CoFe2O4/rGO | 350 | 10 | 0.08 | 90.5 | 120 | 50 |
CaMoO4 | 200 | 5 | 1 | 97.02 ± 0.65 | 120 | 51 |
CPP/TiO2/Ag | 300 | 10 | 5 | 20 | 120 | 52 |
Ce2Sn2O7 | 240 | 10 | 1 | 58 | 120 | 6 |
Dy2Sn2O7/Sepiolite | 240 | 10 | 1 | 84 | 120 | 53 |
The contribution of oxidative species in the AO7 removal process was studied by quenching experiments. In quenching experiments, KI, BQ (1,4-benzoquinone), and IPA (isopropanol) were used as scavengers of holes, ˙O2− and ˙OH, respectively. Fig. 7c shows a slight inhibition with the addition of BQ in the removal of AO7, implying that ˙O2− had a small influence on the catalytic efficiency. The degradation of AO7 was suppressed by KI and IPA, indicating that holes and ˙OH contribute to the elimination of AO7. The addition of IPA reduced the degradation rate to 28.41%, which had a great influence on the degradation effect. This indicates that the active species that play a leading role in the ultrasonic catalysis process are ˙OH and h+.
Based on the above experimental results, a possible ultrasonic catalysis mechanism may be proposed, as shown in Fig. 7d. The mechanical energy of ultrasonic wave acts on the liquid system and generates many extremely small cavitation bubbles. In the process of its formation to breaking, a large amount of energy released will generate local hot spots and be accompanied by sonoluminescence.54 The energy generated by ultrasound (e.g., in the form of an optical cavity) causes an electron excitation in the semiconductor from the conduction band (CB) to the valence band (VB), generating the electron–hole pairs. The conduction band position of FN3 is −0.26 V, not close to the −0.33 V of O2/˙O2−, which means that it is impossible to produce ˙O2−, as it is in agreement with the results of quenching experiments. Indeed, the valence band position of FeNbO4 is 1.87 V, which is more negative than the potential of ˙OH/H2O (2.28 V).55 These results indicate that the h+ generated in the semiconductor FeNbO4 is not capable to generate the radical ˙OH. As shown in Fig. 6i, H2O2 contributed to the efficient degradation of AO7 by the sonocatalyst, which indicates that H2O2 plays a vital role in the degradation process. Quenching experiments confirmed the participation of ˙OH species in the reaction mechanism. The generation of the radical ˙OH could be explained because the presence of Fe2+ in the catalyst, as it is showed in XPS, might be reduced to Fe3+ by the e− generated in the semiconductor. Then, the species Fe3+ could react with H2O2 to produce the ˙OH radical with enough oxidant capability to degrade the AO7 contaminant, following a Fenton-like reaction.
Fe3+ + e− → Fe2+ |
Fe2+ + H2O2 → Fe3+ + ˙OH + OH− |
Fe3+ + H2O + hν → Fe2+ + ˙OH + H+ |
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2024 |