Synthesis of nanospindle-akaganéite and its photocatalytic degradation for methyl orange

Iron oxyhydroxides (FeOOH), widely present in soils and aquatic sediments in several different polymorphs such as goethite, akaganéite and lepidocrocite, are a major scavenger for sequestration of pollutants (Wu et al. 2016). Synthesized FeOOH minerals have been used as adsorbent or catalyst materials for environmental remediation and studied by many investigators. Their environmental functions are closely related to the synthesis methods, morphology and interface properties of particles (Sudakar et al. 2003; Zhang & Jia 2014; Rahimi et al. 2015).

It is well known that the morphology and properties of FeOOH precipitates prepared by ferric hydrolysis can be affected by factors such as pH, temperature, and additives (ions, surfactants and polymers) (Yang et al. 2008; Bashir et al. 2009). For example, Tong et al. (2011) used glucose as a structure-directing agent to prepare the precursors of sea urchin-like goethite (α-FeOOH) and obtained hematite (α-Fe₂O₃) hierarchical nanostructures by adjusting the annealing temperature. It was reported that the composition, structure and size of the α-FeOOH products obtained from ferric solutions can be tailored by the presence of surfactants such as cetyltrimethylammonium bromide (CTAB) and ethylenediaminetetraacetic disodium salt (Na₂EDTA) (Wei et al. 2012; Li et al. 2015; Ristić et al. 2015). The morphology and size of the akaganéite (β-FeOOH) products can be adjusted by varying the concentrations of iron(III) chloride (FeCl₃) and urea, and affected significantly by anionic surfactant cetylpyridinium (SDS) compared to other surfactants (Wei & Nan 2011). It was also documented that under the action of various surfactants, such as cetylpyridinium (CPCI), SDS and Brij56 [C₁₆EO₁₀] (PEO) as structure directors, the mesoporous crystalline β-FeOOH nanoparticles can be synthetized in FeCl₃ solutions at 40 °C, and hydronium jarosite was prepared in the presence of SDS at 80 °C (Yuan & Su 2003; Yuan et al. 2004).

Azo dyes are recalcitrant and refractory pollutants that constitute a significant burden on the Methyl orange (MO) is a typical compound of a series of common water soluble azo dyes widely used in chemistry, textiles and paper industries (Fan et al. 2009). It is essential to select the efficient treatment methods to deal with MO wastewater. Heterogeneous photocatalysis is a discipline that includes a large variety of reactions: mild or total oxidations, dehydrogenation, hydrogen transfer, O₂¹⁸–O₂¹⁶ and deuterium-alkane isotopic exchange, metal deposition, water detoxification, gaseous pollutant removal, etc. (Herrmann 1999). It has several advantages, including the wide-operating pH range and controllable iron leaching into solution, compared with the homogeneous Fenton oxidation process. FeOOH and Fe-based metal-organic frameworks as the potential heterogeneous catalysts have been increasingly used in the Fenton-like advanced oxidation for effective removal of organic pollutants from wastewater (Plata et al. 2010; Pouran et al. 2014; Wang et al. 2018). For instance, α-FeOOH with high catalytic activity can be used for effective removal of organic dye of Orange II by Fenton degradation process (Tiya-Djowe et al. 2015). Han et al. (2016) prepared bismuth oxychloride (BiOCl)/β-FeOOH composite with a glucose-assisted hydrothermal method and found it had enhanced photocatalytic activity for Rhodamine B degradation. The higher photocatalytic activity for heterogeneous Fenton-like advanced oxidation has been mainly attributed to the joint effects of structure, crystallinity and surface areas of the FeOOH catalysts (Shen et al. 2012).

Based on the structural aspect and enhanced magnetic property for collecting and recycling, an efficient heterogeneous catalytic system (ZnO/Fe₃O₄@ pumice nanophotocatalyst), which was constructed of natural pumice clay, iron oxide nanoparticles and zinc oxide nano-rods, was prepared in composite form. This system was applied for photocatalytic degradation of methylene blue dye in aqueous samples under green and white light exposure (Taheri-Ledari et al. 2020). These analytical methods have also been investigated and introduced for calculation of the photocatalytic efficiency. The convenient Cr(VI) removal from aqueous samples by the HNT/Fe₃O₄-HA magnetic nanoabsorbent has also been monitored. This is a promising clay-based catalytic system, magnetized by Fe₃O₄ nanoparticles and functionalized with humic acid, and this desired product with the Cr(VI) adsorption removal efficiency of 98% could be used at least seven times (Hajizadeh et al. 2020). Rabbani et al. (2016) reported under visible light irradiation, that zinc(II) tetrakis (4-carboxyphenyl)porphyrin (Zn-TCPP) has a higher activity than Ag-doped mesoporous TiO₂ in photocatalytic degradation of methylene blue and p-nitrophenol, but its activity was lower under UV light irradiation. 5,10,15,20-tetrakis(4-N,N,N-trimethylanilinium)-porphyrin (TAPP) is also found to be more effective than the Zn-TAPP in the photooxidation of phenol and photo-inactivation of the bacteria (Fayyaz et al. 2015).

Comparatively, β-FeOOH, as one of the FeOOH polymorphs with the special tunnel structure, has drawn much attention for applications in degradation of organic pollutants. In this work, we examined the crystal structures and particle morphologies of akaganéite precipitates obtained from FeCl₃ solutions containing surfactant Brij30 [(C₂₀H₄₂O₅)_n] or glucose at different temperatures (40, 60 and 80 °C). In addition, we evaluated the catalytic performance of akaganéite nanospindles produced at 60 °C in the heterogeneous Fenton removal of MO. This study will enable us to better control the structures of FeOOH and offer opportunities for practical applications in heterogeneous photocatalytic process.

Synthesis experiments of FeOOH by ferric hydrolysis were carried out in 40 mL of 0.1 mol•L⁻¹ FeCl₃ solution divided into three group A (CK) with no additives, group B with Brij30 and group C with glucose additives, and their detailed conditions are shown in Table 1. The reactions were allowed to proceed at pH 1.7 for 3 days with water bath aging. After the reactions ended, the iron precipitates were collected and pretreated by centrifugation to eliminate the remaining Brij30, glucose and impurity ions, and then dried at 40 °C for characterizations.

Characterization methods of FeOOH products were described as below. X-ray diffraction pattern (XRD) was examined by a German Bruker AXS D8 Advance Model diffractometer with a Cu target at 40 kV and 200 mA. Fourier-transform infrared spectrum (FTIR) was monitored by a Nicolet 740 Fourier-transform spectrometer at a resolution of 4 cm⁻¹. Morphology was determined by a Philips Tecnai-12 transmission electron microscope (TEM), and Hitachi S-3400N (II) (or S-4800) field emission-scanning electron microscope (FESEM). The particle diameter distribution was obtained by laser diffraction.

In the three groups of batch experiments, influences of catalyst dosage, H₂O₂ concentration and pH value on degradation of MO were studied. Based on preliminary experiments, akaganéite product C60 was selected as the candidate catalyst. An initial concentration of 80 mg·L⁻¹ for MO was used due to its maximum pre-adsorption efficiency. Under UV irradiation at pH 4.5, the level ranges of 50–400 mg·L⁻¹ for C60 and H₂O₂ were selected to keep an optimum catalytic efficiency. B60 as a reference catalyst was also researched.

Finally, the recycle experiments were also carried out to test the stability and recyclability of the C60 catalyst. After each cycle, the C60 catalyst was separated from the suspension, washed several times to remove residual MO using distilled water and ethanol, and dried for the next cycle. To identify the active species formed during the photocatalytic process, the hole and free radical trapping experiments were determined using scavengers of excess potassium iodide (KI), isopropanol (IPA) and benzoquinone (BQ), respectively.

XRD and IR characterization results of all iron precipitate products obtained from groups A and B at various temperatures (40, 60 and 80 °C) are shown in Figure 1. All the characteristic peaks occurring in their XRD patterns (Figure 1(a)) can be indexed to pure tetragonal β-FeOOH with a good crystallinity, according to JCPDS card no. 34-1266. However, compared with the diffraction peaks for the XRD pattern of the standard β-FeOOH, the diffraction peaks at the d value of 2.55 Å for (211) crystal plane were slightly higher than that of 3.34 Å for (310) crystal plane. XRD results indicated that akaganéite product had the slightly better crystallinity in presence of Brij30 at a higher reaction temperature above 60 °C. Moreover, iron precipitates were not observed in group A at 40 °C. It was suggested that the presence of Brij30 could promote akaganéite formation. Further, their corresponding IR spectra proved that the obtained precipitates were akaganéite (Figure 1(b)) because the bands at 831, 698 and 635 cm⁻¹ were assigned to the in-plane bending vibrations of hydroxyl groups of OH•••OH₂ and OH•••Cl hydrogen bonds (Murad & Bishop 2000; Song & Boily 2012) and the other two absorption bands at 490 and 431 cm⁻¹ were assigned to the Fe–O vibrational modes in β-FeOOH (Musić et al. 2004).

TEM morphologies of the aforementioned β-FeOOH products are depicted in Figure 2. All products were spindle nanoparticles with a length of about 200–600 nm and an axial ratio of about 5. An FESEM morphology extra displayed for the B40 products clearly showed that the spindle nanoparticles had a rough surface. Moreover, in their TEM images (Figure 2) the approximate mean particle sizes (nm) of the spindle nanoparticles were in a decreasing order of A80 (600) > B80 (400) > B60/A60 (350) > B40 (250). As seen from Table 2, the average particle sizes (μm) of the agglomerated particles in their suspensions were 0.62 (A80), 0.41 (B80), 0.35 (B60), 0.34 (A60) and 0.28 (B40). These results suggest the optimum temperature of 60 °C for production of akaganéite precipitates obtained from FeCl₃ solutions. Wei & Nan (2011) also reported that in the absence of any additives, the reaction temperature had a great impact on the morphologies of β-FeOOH formed from FeCl₃ solutions, where the spindle and rod particles were produced at a range from 70 °C to 80 °C. At 60 °C or 80 °C, the particle sizes of the β-FeOOH products from group B were not larger than those from group A.

XRD patterns (Figure 3(a)) and IR spectra (Figure 3(b)) of the precipitate products obtained from group C are shown in Figure 3. According to the JCPDS card no. 34–1266, it was found that the whole-angle of akaganéite phase was exhibited in their XRD patterns. There were some strong and sharp diffraction peaks for the crystal planes of (110), (310), (211), (301) and (521). Their infrared spectra exhibited bands at 840, 698 and 645 cm⁻¹ and were attributed to the vibration of the characteristic O–H•••Cl hydrogen bonds for chloride-containing akaganéite (Schwermann & Cornell 2000), and the symmetric Fe–O–Fe stretching vibration present in the intense band at 490 and 431 cm⁻¹ (González-Calbet et al. 1981). Thus, the XRD and IR results jointly identified the produced iron minerals to be akaganéite. XRD results showed that the crystallinity of β-FeOOH products was enhanced as the temperature increased. Compared with the results in Figure 1 for the products from FeCl₃ solutions without additives in group A, it is suggested that the presence of glucose facilitated the formation of β-FeOOH precipitates.

FESEM images in Figure 4 show that the β-FeOOH products were spindle-shaped particles with a rough surface and a length of about 250–500 nm with an axial ratio of about 5. The size of β-FeOOH products slightly increased as the temperature increased. The average sizes of their agglomerated particles (Table 2) were 0.52 (C60), 0.51 (C80) and 0.46 μm (C40).

To summarize, the surfactant in the reaction mixture reduces the particle sizes of β-FeOOH crystals in nanometer scale, and also regulates the growth and aggregation of the nanoparticles through hydrogen bond formation with surface −OH groups of β-FeOOH (Ren et al. 2003). Glucose as a hydroxyl-enriched molecule could facilitate the crystallization of β-FeOOH and the formation of rod-like structure due to its abundant −OH groups (Zhu et al. 2011), and also interact with FeOOH surface sites via hydrogen bonding formation (Olsson et al. 2011). Therefore, the presence of surfactant Brij30 or glucose in FeCl₃ solutions could promote akaganéite formation.

Considering size and aggregation of the prepared akaganéite particles and the above characterized results, the akaganéite product of C60 could be selected as the candidate catalyst to explore its MO degradation mechanism.

The catalytic action of C60 akaganéite to the dye MO is displayed in Figure 5 (three solid lines). Under H₂O₂ addition and UV irradiation, the MO removal efficiencies could reach nearly 100%. However, when only one of the two conditions was provided, MO degradation of about 4% (or 7%) was observed in the dark (or in the absence of H₂O₂). The combined actions of UV irradiation and the presence of H₂O₂ significantly enhanced the photo-Fenton removal efficiency of MO. Compared to the dark condition, UV irradiation can accelerate the heterogeneous Fenton reaction due to the enhancement of •OH production (Xu et al. 2013c).

Under 100 mg·L⁻¹ of H₂O₂ and pH 4.5, removal efficiencies of MO (initially at 80 mg·L⁻¹) by C60 akaganéite are shown in Figure 6(a). There was little difference in the efficiencies of MO removal or degradation. It has been recognized that the increase of active sites on the catalyst surface could lead to the enhanced production of hydroxyl and superoxide radicals (Akpan & Hameed 2009). In this study, with the addition of H₂O₂ at pH 4.5 and the removal efficiencies for C60 akaganéite at 100 mg·L⁻¹ degrading 80 mg·L⁻¹ of MO were examined and the results are shown in Figure 6(b). Within 20 min of reaction, the MO removal efficiencies rapidly reached nearly 100, 90, 80 and 60% for the treatments of H₂O₂ concentrations at 400, 200, 100 and 50 mg·L⁻¹, respectively. After about 35 min of reaction, the MO removal efficiencies increased slowly to a stable state of nearly 100%, except for the treatment with 50 mg·L⁻¹ H₂O₂ (about 80% observed after 60 min). It was documented that the amount of •OH produced by the photo-Fenton reaction was directly related to the concentration of H₂O₂ (Li et al. 2016). The MO removal efficiency insignificantly increased when the concentration of H₂O₂ exceeded a certain range, which could be caused by the competitive adsorption of excess H₂O₂ and MO on the surface of β-FeOOH and by quenching of •OH by excess H₂O₂ (Li & Zhang 2010). Therefore, it suggested that 100 mg·L⁻¹ of H₂O₂ should suffice for akaganéite (100 mg·L⁻¹) degrading 80 mg·L⁻¹ of MO.

The initial pH also plays a key role in photo-Fenton reaction and influences the action of catalysts. In the current work, effects of the reaction pH values (4.5, 7.0 and 9.5) on the degradation of MO by C60 akaganéite were also explored and the results of MO removal are shown in Figure 7(a). Removal efficiencies of MO gradually decreased with an increase in initial pH values, which indicated that the akaganéite catalysts could overcome the drawback of a narrow pH range of homogeneous Fenton reaction. Moreover, under the same reaction conditions, the removal or photocatalytic efficiencies of MO by C60 were slightly higher, compared with B60 (as shown in Figure 7(b)). It has been suggested that iron oxides typically have a point of zero charge in the pH range of 7–9, and the initial pH value could affect the charge state of the akaganéite surface (Zhao & Hu 2008). In the case of pH from neutral to alkaline, the surface of akaganéite was gradually deprotonated to the negative ion FeO₂⁻, causing repulsion with the negatively charged MO molecules. On the contrary, under acidic conditions, the surfaces of the iron minerals had more positive charges, facilitating adsorption of MO molecules on the catalyst surfaces and production of more •OH to attack MO molecules (Xu et al. 2013b). This was why the optimum pH of 4.5 was chosen for the photocatalytic process of MO degradation. Li & Zhang (2010) used an amorphous FeOOH as a catalyst to degrade 50 mg·L⁻¹ of MO in a heterogeneous Fenton system, and found that after 80 min the degradation efficiency was about 96% for the reaction system with 2.5 g•L⁻¹ of catalyst and 15.8 mmol•L⁻¹ of H₂O₂ under UV irradiation at pH 7.0. At pH 5, the highest photocatalytic activity was reached for β-FeOOH/TiO₂ degrading 80 mg·L⁻¹ of MO in the system with 200 mg·L⁻¹ of catalyst, and 3 mL of H₂O₂ under a 300 W halogen tungsten lamp irradiation for 45 min (Chowdhury et al. 2015). However, in this work, a high MO removal efficiency was achieved at a wider pH range for the system with a smaller catalyst dosage and H₂O₂ concentration.

Moreover, according to the calculation methods in the literature (Najafian et al. 2015), it is deduced in the photocatalytic reaction that the turnover number (TON) of MO is 2,174 and its turnover frequency (TOF) is 2,174 h⁻¹. Generally, a high TON and TOF indicates a stable and long-lived catalyst. The results showed that each 1 mol akaganéite converts the 2,174 mol of MO molecules within their reaction time.

Finally, stability and recyclability of C60 akaganéite were evaluated by four successive runs for the oxidation of MO under standard photocatalytic conditions. As displayed in Figure 8, after four cycles of repeated use, 86% of MO could still be removed, indicating that the reuse of C60 was cost-effective. The reduced MO degradation efficiency probably attributed to unavoidable leaching of iron and possible effect of contaminant adsorption on the reactive sites on the catalyst surfaces.

This research showed that the presence of organic glucose or surfactant Brij30 in FeCl₃ solutions significantly promoted the formation of nanospindle-akaganéites at 40 °C. The particle sizes of akaganéite products formed at the higher temperature (60 °C or 80 °C) were larger than those obtained at 40 °C. Results indicated that the performance of akaganéite products formed at 60 °C for degradation of MO dye was affected by the initial solution pH values (4.5–9.5) and the higher removal efficiencies of MO (80 mg·L⁻¹) could be achieved at pH 4.5, in the presence of the catalyst and H₂O₂ at the optimal levels (100 mg·L⁻¹). The free radical trapping experiment revealed a possible catalytic mechanism of akaganéite catalyst, which indicated degradation of MO was mainly ascribed to •O₂⁻, •OH and h⁺ active species. These results have an implication for the potential applications of FeOOH catalysts in the treatment of organic pollutants in wastewaters.

The authors acknowledge the National Natural Science Foundation of China (no. 41472034 and 31372133), Natural Science General Fund of Jiangsu Province (no. BK20191444) and Jiangsu Provincial Key Laboratory of Environmental Materials and Engineering (no. 14022 and K13058) for supporting for the present study.

All relevant data are included in the paper or its Supplementary Information.