Preparation and characterization of composite magnetic photocatalyst Mn_xZn_1−xFe₂O₄/β-Bi₂O₃

Abstract

Composite magnetic photocatalyst Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ was synthesized by a dip-calcination method using manganese zinc ferrite as a magnetic substrate. The effects of composite mass ratio, reaction time and calcination temperature on the degradation of Rhodamine B (RhB) under the simulated sunlight were observed with various investigations. The as-prepared Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ was characterized by XRD, FTIR, VSM, UV-vis DRS and SEM. The photodegradation rate of RhB in Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ was higher (99.1%) than that (83.6%) in pure β-Bi₂O₃ within 2.5 h. XRD spectra revealed that the composites presented a tetragonal type, which was similar to that of β-Bi₂O₃. FTIR spectra exhibit peaks for the absorption of both Bi–O bonds and Mn_xZn_1−xFe₂O₄. The saturation magnetization (M_s) and coercivity (H_c) of the composite photocatalyst were 7.01 A m² kg⁻¹ and 25.38 A m⁻¹, respectively. DRS analysis revealed that the optical band gap of this composite was 2.31 eV, which was lower than that of β-Bi₂O₃ (2.45 eV). Moreover, the photocatalytic activity was still maintained at 82.7% after five cycles. The magnetic property with an appropriate amount of manganese zinc ferrite inhibited the recombination of photo-produced electrons (e⁻) and holes (h⁺), and enhanced the photocatalytic property of β-Bi₂O₃.

---

1. Introduction

TiO₂ has been widely utilized as a promising catalyst for photochemical oxidation since Fujishima et al. applied the catalyst for electrochemical proteolysis of water for the first time in 1972.¹ Carey et al. (1972) suggested that TiO₂ showed promise as a semiconductor photocatalyst for the degradation of organics in water.² This indicated that TiO₂ could be employed in the field of environmental pollutants control. Frank et al. (1977) found further applications of photocatalytic oxidation technology for degrading pollutants from industrial waste water.^3–5 These contributions laid the theoretical foundation and proved the considerable serviceability of TiO₂ as a promising photocatalyst for pollution control.^6–8

Due to the wide band gap and weak response to visible light of TiO₂, intensive attention has recently been directed towards exploring novel photocatalysts such as Ag₃PO₄ and their composites,^9,10 sulfides,¹¹ bismuth compounds,^12–14 and doping metals or nonmetals.¹⁵ Bismuth compounds of Bi₂O₃ have recently been considered as a preferable option for the synthesis of optical catalysts due to its narrower band gap and larger absorption wavelength. Bi₂O₃ has four polymorphs, namely, α, β, γ, and δ. Their band gaps are 2.85, 2.58, 2.68 and 2.75 eV, respectively.^16–18 β-Bi₂O₃ is superior to α-Bi₂O₃ in photocatalytic properties. However, most traditional and new photocatalysts in the photocatalytic reaction system are present in a suspension state, which makes separation of the materials rather difficult, let alone their recycling and reuse. The problem seriously restricts the effective utilization of photocatalysts and consumes more energy and is cost-intensive in the photocatalytic process. It would be noteworthy to develop a composite magnetic photocatalyst to be recycled by an external magnetic field.^19,20 The magnetic photocatalyst is expected to overcome the difficulty of recycling the photocatalyst from the suspension state.

Mn_xZn_1−xFe₂O₄, which is a typical magnetic material, has several advantages, including saturation magnetization and permeability, large production efficiency, stable performance and excellent corrosion resistance,^21,22 compared with traditional metallic soft magnetic materials (e.g. Fe₃O₄). There have been few studies about Mn_xZn_1−xFe₂O₄ involved in composite magnetic photocatalysis to date. Wang et al. prepared a magnetic photocatalyst of TiO₂/Mn_xZn_1−xFe₂O₄/AC by a sol–gel method, and they reported that the as-synthesized catalyst presented superior photocatalytic activity over pure TiO₂ under ultraviolet light.²³ In our previous study, we prepared magnetic solid acid catalyst S₂O₈²⁻/ZrO₂–Mn_xZn_1−xFe₂O₄ based on a magnetic matrix Mn_xZn_1−xFe₂O₄.¹⁹ It was found that the as-prepared magnetic solid acid catalyst exhibited a higher catalytic activity for the transesterification with respect to the pure substance. In the present study, Mn_xZn_1−xFe₂O₄ was chosen as the magnetic substrate, and the composite magnetic photocatalyst Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ was synthesized by a dip-calcination method. The photocatalytic activity of the as-prepared catalyst was then tested under simulated sunlight. Further insights have been focused on the mechanism and recycling feasibility with an external magnetic field. The results in this study will lay the groundwork for further application of the photocatalyst in pollutant removal.

2. Experimental

Analytical reagents of Bi(NO₃)₃, Na₂CO₃, FeCl₃, ZnSO₄ and HNO₃ were used as raw materials for sample preparation, and provided by Shanghai Jiuyi Chemical Ltd. MnSO₄ solution was leached and refined from a manganese ore.²⁴

2.1 Preparation of β-Bi₂O₃

3.12 g Bi(NO₃)₃·5H₂O was dissolved in 20 mL diluted HNO₃ solution and stirred for 0.5 h. The Bi(NO₃)₃–HNO₃ solution was slowly added to 0.6 mol L⁻¹ Na₂CO₃ solution and stirred for 2 h at ambient temperature.²⁵ The obtained precipitates were filtered and washed in deionized water several times and the precipitates were dried at 60 °C. Subsequently, the dry blocks were crushed into fine powder and calcined at 380 °C in a pure N₂ atmosphere for 10 minutes and then bright yellow β-Bi₂O₃ was obtained.

2.2 Preparation of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃

The molar ratio of n(ZnO) : n(MnO) : n(Fe₂O₃) = 13.3 : 32.8 : 53.9, the defined dosage of ZnSO₄, MnSO₄ and FeCl₃·6H₂O, was dissolved in deionized water, and the solution was stirred with a mechanical stirrer for 0.5 h at ambient temperature. Subsequently, the as-prepared ZnSO₄ and FeCl₃ solutions were added to a MnSO₄ solution with continuous stirring to obtain a mixed solution. A definite amount of (NH₄)₂C₂O₄·H₂O was dissolved in deionized water, and a (NH₄)₂C₂O₄ solution was obtained. The mixed solution was slowly added to the (NH₄)₂C₂O₄ solution after the mixed solution and the (NH₄)₂C₂O₄ solution were heated to 80 °C. The pH value of the system was adjusted to 7, and a great amount of precipitate was observed. The mixtures were filtered, and the precipitate was washed with the deionized water and dried at 80 °C for 12 h. The residues were sintered at 1200 °C for 3 h and magnetic Mn_xZn_1−xFe₂O₄ was obtained.

Mn_xZn_1−xFe₂O₄ with the mass ratio of 15 wt% was added into the above mentioned Bi(NO₃)₃–HNO₃ solution and stirred for 30 minutes. The mixture was then added dropwise to 0.6 mol L⁻¹ Na₂CO₃ solution, and stirred mechanically for 2 h at ambient temperature. After filtration, the residue was washed several times with deionized water and the precipitate was dried at 60 °C. The Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) composite was produced by sintering the dried filter residue at 380 °C for 10 minutes. A series of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ composites was synthesized by adjusting the mass ratios of Mn_xZn_1−xFe₂O₄ and β-Bi₂O₃ (5 wt%, 10 wt%, 15 wt%, 20 wt%, and 25 wt%). In addition, the Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ composites were prepared at the various reaction times (1 h, 1.5 h, 2 h, 2.5 h, and 3 h) and calcination temperatures (360 °C, 370 °C, 380 °C, 390 °C, and 400 °C).

2.3 Characterization of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃

Phase identification of the as-prepared products was performed with an X-ray diffractometer (Shimadzu XRD-6000) at a scanning rate of 4° min⁻¹ in the 2θ range of 10° to 80°. Fourier transform infrared (FTIR) spectra of the samples were recorded on a 5DX FTIR spectrometer (5DX, Nicolet. Co., USA). The morphologies and microstructures of the products were observed by scanning electron microscopy (SEM, Hitachi S-4800, Japan). The UV-vis diffuse reflectance spectra (DRS) and the magnetic properties of the tested samples were studied using a UV-vis spectrophotometer (TU1901, China) and a vibrating sample magnetometer (MPMS-XL-7 Quantum Design, USA), respectively. Element contents in the composite were analyzed by ICP (inductively coupled plasma spectrometer, ICP 6300 series, Thermo scientific Co., USA). The concentration of the composite solution was maintained at 500 μg mL⁻¹ to meet the detection range of ICP spectrometer.

2.4 Measurement of photocatalytic property

The photocatalytic activity of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ was evaluated by the degradation of Rhodamine B (RhB) under simulated sunlight. 100 mL of RhB aqueous solution (10 mg L⁻¹) and 0.2 g of photocatalyst were placed into a quartz container and stirred for 1 hour in the dark to reach adsorption and desorption balance. Then, a xenon lamp (CEL-HXF3000, AULTT) of 300 W was employed as the light source and the solution was irradiated for 2.5 hours with continuous stirring. 5 mL RhB solution was withdrawn at set time intervals. Then, the mixtures were centrifuged at 3800 rpm for 10 minutes to obtain the supernatant. The photocatalytic degradation process of RhB was monitored by measuring its characteristic absorption at 554 nm with a UV-vis spectrophotometer.

3. Results and discussion

3.1 Optimization condition of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ preparation

The effects of mass ratio, reaction time and calcination temperature on the degradation rate of RhB were investigated. It can be observed in Table 1 that the degradation rate increased at first and then decreased with the increase in the Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ mass ratio from 5 wt% to 25 wt%. The maximum degradation rate was reached when the mass ratio was 15 wt%. The reason is that the photo-produced electrons (e⁻) and holes (h⁺) are not easy to separate effectively when the loading amount of Mn_xZn_1−xFe₂O₄ is low. The magnetic field generated by an appropriate amount of Mn_xZn_1−xFe₂O₄ reduced the recombination of photo-produced electrons (e⁻) and holes (h⁺), and extended the lifetime of photo-produced electrons. Thereby, the photocatalytic activity of the composite was improved. However, the photocatalytic activity did not appear to be enhanced with the ever-increasing loading amounts of Mn_xZn_1−xFe₂O₄. Because Mn_xZn_1−xFe₂O₄ itself is a semiconductor, excessive Mn_xZn_1−xFe₂O₄ loaded in β-Bi₂O₃ resulted in a significantly large number of recombination centers of photo-produced electrons (e⁻) and holes (h⁺). The photocatalytic active property of this catalyst declined.

Table 1 The influence of mass ratio and reaction time as well as calcination temperature on the degradation rate of RhB

Mass ratio (wt%)	Degradation rate (%)	Reaction time (h)	Degradation rate (%)	Calcination temperature (°C)	Degradation rate (%)
5	82.8	1	80.8	360	63.6
10	93.9	1.5	86.4	370	90.6
15	99.1	2	98.7	380	99.5
20	78.7	2.5	77.7	390	76.6
25	73.8	3	70.2	400	72.5

---

It can be observed in Table 1 that the photocatalytic activity of 15 wt% Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ increased first and decreased later with increasing time within 3 h. The largest photocatalytic activity of β-Bi₂O₃ was reached at a reaction time of 2 h. The effect of calcination temperature on the photocatalytic activity of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) was investigated within the optimal time of 2 h. It was found that the photocatalytic activity increased with increasing calcination temperature from 360 °C to 380 °C, and then the activity dropped with a further increase of calcination temperature. The excellent photocatalytic activity was achieved at the calcination temperature of 380 °C. The result was mostly caused due to α-Bi₂O₃ at a higher calcination temperature, because the photocatalytic activity of α-Bi₂O₃ was considerably lower than that of β-Bi₂O₃. At the same time, the low activity of β-Bi₂O₃ at a lower calcination temperature was related to the incomplete conversion of the crystalline content.

The orthogonal experiments with three factors and three levels were carried out to obtain the optimal synthesis conditions of the photocatalyst. The results are shown in Table 2, which proved again that the large photocatalytic activity is presented under the condition of the mass ratio of 15 wt%, a reaction time of 2 h and a calcination temperature of 380 °C. The experiments demonstrated that calcination temperature is the key factor to impact the photocatalytic activity. The importance order of impact factors in the photocatalytic activity are calcination temperature > mass ratio > reaction time. The conclusion of the orthogonal experiments was consistent with the result of the single-factor experiment.

Table 2 The orthogonal experimental design

	Factor 1	Factor 2	Factor 3
	Mass ratio (wt%)	Reaction time (h)	Calcination temperature (°C)
Level 1	10	1	370
Level 2	15	2	380
Level 3	20	3	390

---

3.2 Structure characteristics

X-ray diffraction peaks of β-Bi₂O₃ and Mn_xZn_1−xFe₂O₄ as well as those of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) were sharp and symmetrical, as shown in Fig. 1. It was deduced that the as-prepared products exhibit good crystallinity. The characteristic spectra of β-Bi₂O₃ are well indexed with the standard card (JCPDS card number: 27-0050), corresponding to the diffraction phases of (201), (220), (222) and (421). It is reasonable to assume that the as-prepared β-Bi₂O₃ belongs to the tetragonal type. The diffraction peaks of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) consisted each diffraction phase of β-Bi₂O₃, as shown in Fig. 1(c). It is reasonable to conclude that the crystallinity of β-Bi₂O₃ ​is integral.

Fig. 1 XRD spectra of pure β-Bi₂O₃ and Mn_xZn_1−xFe₂O₄ as well as Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) composite.

This result indicates that Mn_xZn_1−xFe₂O₄ did not change the structure of β-Bi₂O₃. There were no diffraction peaks for Mn_xZn_1−xFe₂O₄ in the XRD pattern of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%). Very less Mn_xZn_1−xFe₂O₄ was detected by X-ray diffraction. In addition, the peaks of Mn_xZn_1−xFe₂O₄ were covered by strong diffraction peaks of β-Bi₂O₃. The existence of Mn_xZn_1−xFe₂O₄ was confirmed by ICP, FTIR and SEM measurements.

The qualitative analysis of the as-prepared product was performed with ICP, and its principle is that each element creates its own unique spectrum when it is activated, and therefore an element can be confirmed according to the existence of its corresponding spectral lines. Three groups of data are listed in Table 3. Element Bi was observed in Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) as the signal average of Bi was 8 : 1 : 16. In the same way, the existence of other elements, including Mn and Zn as well as Fe, were determined in the tested sample. This indicated that the Mn, Zn, Fe and Bi elements were present in the composite catalyst. The wavelengths corresponding to the minimum detectability for Mn, Zn, Fe and Bi were 257.6 nm, 213.8 nm, 259.9 nm and 223.0 nm, respectively, which are consistent with the data reported in the literature.^26–28

Table 3 ICP analysis of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%)

Elements	Bi			Mn			Zn			Fe
Wavelength/nm	190.2	195.4	223.0	257.6	293.9	348.2	213.8	328.2	481	234.3	259.9	322.7
Signal intensity	117 100	14 810	242 200	804 000	188 600	22 300	103 300	1197	16 600	177 200	364 500	3438
	117 100	14 830	239 700	793 500	187 600	22 230	101 900	1183	16 500	178 500	365 200	3455
	117 200	14 810	239 900	787 000	182 700	21 720	101 900	1181	16 110	178 700	351 700	3404
Average signal intensity	117 133	14 817	240 600	794 833	186 300	22 083	102 367	1187	16 403	178 133	360 467	3432
Ratio	8 : 1 : 16			36 : 8 : 1			86 : 1 : 14			52 : 105 : 1

---

FTIR spectra of Mn_xZn_1−xFe₂O₄, β-Bi₂O₃, and Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) as well as those of recovery composites (1#, 2#, 3#, and 4#) are shown in Fig. 2. The characteristic peaks at 3448.3 cm⁻¹ and 1637.0 cm⁻¹ (green lines) are attributed to the stretching vibration and deformation vibration of the hydroxyl group (–OH) of water from wet atmosphere. Sample 1 shows that three characteristic peaks of Mn_xZn_1−xFe₂O₄ were at 1118.1 cm⁻¹, 556.6 cm⁻¹ and 436.6 cm⁻¹ (blue lines),²⁴ respectively. The intensive signals around 1394.1 cm⁻¹ and 846.4 cm⁻¹ as well as 521.6 cm⁻¹ (red lines) appearing in the FTIR spectrum of β-Bi₂O₃ refer to the stretching vibration of Bi–O bonds.²⁹ Moreover, sample 4 shows the FTIR pattern of the recovered composite that was washed with ethanol. It was noticed that the characteristic peaks of both β-Bi₂O₃ and Mn_xZn_1−xFe₂O₄ were observed for the recovery sample, and no impurity peaks appeared. This result illustrates that RhB dye molecules were not adsorbed inside the composite particles after the photocatalytic degradation tests of RhB were completed. This dye did disappear from the solution with the photocatalyst under the light illumination. There was a little wave shift in the characteristic peaks in the FTIR spectra of both β-Bi₂O₃ and Mn_xZn_1−xFe₂O₄ for the vibration coupling between the peaks of Mn_xZn_1−xFe₂O₄ and Bi₂O₃ bonds. It was important that the characteristic peaks were fairly obvious in this pattern.

Fig. 2 The FTIR spectra of pure β-Bi₂O₃ and composite Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%).

Fig. 3 shows the SEM images of Mn_xZn_1−xFe₂O₄ and β-Bi₂O₃ as well as that of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%). It can be observed that Mn_xZn_1−xFe₂O₄ has particles with irregular shape in Fig. 3(a). As can be seen from Fig. 3(b), β-Bi₂O₃ displayed nanoscale irregular granular flakes and rod-like and spherical particles. The irregular β-Bi₂O₃ particle and Mn_xZn_1−xFe₂O₄ were complexed together successfully, as shown in Fig. 3(c) and (d). In detail, β-Bi₂O₃ grains in the composite were smaller than that in pure Bi₂O₃, but the favored growth direction did not alter in the composite system. The flat surface did not have significant defects from the loading process of Mn_xZn_1−xFe₂O₄. As we all know, the defects on the surface of photocatalyst are usually regeneration centers for photo-produced electrons (e⁻) and holes (h⁺). Thus, the morphology structure of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ was conducive to improve the photocatalytic activity.

Fig. 3 SEM images of Mn_xZn_1−xFe₂O₄ (a) and pure β-Bi₂O₃ (b) and composite Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) ((c) and (d)) (at 10 μm and 2 μm).

3.3 Absorption light ability and magnetic property

Light absorption is an important property of semiconductors. The optical band gap energy (E_g) is directly related to the light absorbance. E_g was estimated from eqn (1):

αhν = C(hv − E_g)^1/2 (1)

where α is the absorbance, h is Planck constant, ν is the photonic frequency, and C is an experience constant.

Fig. 4 shows diffuse reflectance spectra, corresponding to the band gap of pure β-Bi₂O₃ (a) and composite magnetic photocatalyst Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) (b). β-Bi₂O₃ is a type of direct transition,^25,30 as shown by the curve fitted by eqn (1) in Fig. 4. It is clearly seen from Fig. 4 that E_g of β-Bi₂O₃ and Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) are 2.45 eV and 2.31 eV, respectively. The result is in accordance with the literature.^31,32 It is clear that both pure β-Bi₂O₃ and the composite exhibit strong absorption in the UV light region (wavelength: 200–400 nm). However, it is worth noting that the composite showed a strong absorption in a wide wavelength range from UV to visible light, compared to that of pure β-Bi₂O₃. Moreover, the maximum absorption wavelength of the composite increased to 537 nm. Therefore, the visible light absorption ability of 15 wt% Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ was enhanced, and the composite was adapted to the employed visible catalyst.

Fig. 4 Diffuse reflectance spectra of pure β-Bi₂O₃ (a) and the composite magnetic photocatalyst Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) (b). The inset shows the corresponding band gaps fitted by eqn (1).

The magnetic hysteresis loops of Mn_xZn_1−xFe₂O₄ and the Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) composite are displayed in Fig. 5. It can be determined from Fig. 5 that the saturation magnetization (M_s) and remanent magnetization (M_r) of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) were 7.01 and 0.39 A m² kg⁻¹, respectively. Larger M_s was conducive for the separation and the recovery of the composite after the photocatalytic tests ended. Compared to Mn_xZn_1−xFe₂O₄, the M_s and M_r of 15 wt% Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ declined by 91.6% and 88.4%, respectively. The declination was due to the decrease of the magnetic substance per unit mass. Some β-Bi₂O₃ attached on the surface of Mn_xZn_1−xFe₂O₄ also exhibited an adverse role on the magnetic property.

Fig. 5 The magnetic hysteresis loops of Mn_xZn_1−xFe₂O₄ and the Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) composite.

3.4 Photocatalytic activity

The photocatalytic mechanism of the semiconductor is stated as follows: the photocatalytic process of β-Bi₂O₃ was stimulated by the photons with sufficient energy. Electronics with valence band of semiconductor excited energy grade transition for much photo-induced electron–hole pairs as β-Bi₂O₃ can absorb enough energy. Photo-electrons and holes migrated to the surface of the composite catalyst. Thus, the dissolved oxygen was restored to the superoxide anion, and at the same time hydroxide ions and water were oxidized to hydroxyl radicals (˙OH) by electrons and holes. With very strong oxidation resistance, superoxide anions and hydroxyl radicals (˙OH) oxidized the organics to the final products CO₂ and H₂O,^33–35 and broke down some inorganics completely to achieve the degradation of the pollutants.

The photocatalytic degradation rate of RhB with pure β-Bi₂O₃ reached to 83.6% at 2.5 hours, as shown in Fig. 6. The self-degradation of RhB was very weak in the comparative experiment. The degradation rates of RhB were significantly different under the Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (10 wt%, 15 wt% and 20 wt%) composites. The degradation rate with Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) reached 99.1% under the same conditions, while excessive Mn_xZn_1−xFe₂O₄ resulted in degradation rate lower than that of pure β-Bi₂O₃. Namely, the photocatalytic activity of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) was the highest. Furthermore, the degradation rate in the irradiation period of 2–2.5 h was considerably higher than before. We explained this phenomenon as follows. First, the photocatalyst and the RhB solution reached absorption equilibrium entirely after stirring under irradiation for 2 h. Photo-electrons and holes migrating to the surface of the composite catalyst also reached a balance. Second, the phenomenon occurred (especially for 10 wt%, 15 wt% and 20 wt%). This is because a constant magnetic field causes electrons to undergo a directional movement at a constant speed. The valence band of stimulated electronics (composites of 10 wt%, 15 wt% and 20 wt%) did not accumulate in a place completely. Thus, the phenomenon of the above mentioned three types of photocatalyst was more obvious.

Fig. 6 Photocatalytic properties of pure β-Bi₂O₃ and Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (5 wt%, 10 wt%, 15 wt%, 20 wt%, and 25 wt%).

It was important to note that the M_r of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) was 7.01 A m² kg⁻¹, indicating that the composite produced a stable magnetic field. As is well known, a constant magnetic field causes electrons to undergo a directional movement at a constant speed. Some of the electrons underwent a uniform motion along the direction of magnetic field, but the others underwent a uniform circular motion perpendicular to the plane of the magnetic field.^36,37 Combined with the movement of electrons in a constant magnetic field, photo-generated electrons brought directional movement when a constant magnetic field was applied around the photocatalyst. Furthermore, a constant magnetic field promoted a bidirectional shunt (as shown in Fig. 7) of photo-generated electrons (e⁻).³⁸ So, the stimulated electronics with valence band did not accumulate in a place. Based on these effects, the utilization efficiency of light energy (namely, the photoelectric conversion efficiency) increased and the photocatalytic activity was improved.

Fig. 7 The photocatalytic mechanism of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%).

In addition, Mn_xZn_1−xFe₂O₄ was able to absorb the most of the solar photons and increase the light response of β-Bi₂O₃.³⁹ The action helped to absorb more incident photons and produced more photo-produced electrons (e⁻) and holes (h⁺), improving the photocatalytic activity. The magnetic field yielded by Mn_xZn_1−xFe₂O₄ promoted the charged particles to move by specific roles, inhibiting the recombination of photo-generated electron–hole pairs and thus extending the lifetime of the photo-generated pairs. As a result, the photocatalytic activity was increased. We cannot ignore the fact that Mn_xZn_1−xFe₂O₄ is a semiconductor; thus, it would become the recombination center of photo-produced electrons (e⁻) and holes (h⁺) and lower the photocatalytic activity if loaded too heavily.

3.5 Recycling ability and stability

0.2 g of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) was weighed and added into RhB solution at a concentration of 10 mg L⁻¹ (100 mL); then, the photocatalytic experiments were carried out under simulated sunlight. After the reaction was completed, Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ and supernatant liquid were separated by an external magnetic field. The recovered composites were washed by anhydrous ethanol several times and then activated at a temperature of 60 °C for 2 h. The weight of the dried solid was 0.1786 g; thus, the recovery rate was 89.3%.

The stability and recyclability of Mn_xZn_1–xFe₂O₄/β-Bi₂O₃ (15 wt%) was studied by recycling experiments. It can be seen in Fig. 8 that the degradation rate of RhB with the composite magnetic photocatalyst was 99.1% and 82.7% at the 1st and 5th cycles. The experimental results revealed that the composites could be reused several times without a significant decrease in activity. After five recycles, the magnetic property parameter, as illustrated in Fig. 9, is in contrast with the original magnetic data. M_s and M_r of the recycled composite in Fig. 9 were 5.37 and 0.39 A m² kg⁻¹, respectively, after five reuses. Compared to the original magnetic data, the saturation magnetization is a little lower. This was due to a loss in the recovery process, resulting in a decrease of magnetic substance per unit mass. The coercivity and residual magnetization were not changed, indicating that the composites exhibited relatively stable magnetic properties. The above mentioned data showed that Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) has high stability and good repeatable property.

Fig. 8 The recycle experiments of degrading RhB on the Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) composite under simulated sunlight irradiation.

Fig. 9 The magnetic hysteresis loops of Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) and composites after five cycles.

In Fig. 9, we can conclude that M_s, M_r, coercivity and residual magnetization did not change. So, Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) demonstrated high stability and a good reusable property.

4. Conclusions

The composite magnetic photocatalyst Mn_xZn_1−xFe₂O₄/β-Bi₂O₃ (15 wt%) was synthesized by a dip-calcination method. The synthesis was adapted to facilitate a large-scale production with a simple preparation technology and a low-cost procedure. Photocatalytic activity was evaluated with the degradation rate of RhB under simulated sunlight. VSM test results indicated that the composite possesses good magnetic properties, which is conducive for the separation and reuse. Recycling experiments showed that the recovery rate was as high as 89.3%, and the degradation rate of RhB in the composite magnetic photocatalyst decreased from 99.1% to 82.7% after five cycles. The composite exhibited excellent catalytic activity and stability. The photocatalytic mechanism of the composite was discussed, and the magnetic photocatalyst produced a more stable magnetic field and increased the utilization efficiency of light, prompting the bidirectional shunt of photo-generated e⁻–h⁺ pairs. Mn_xZn_1−xFe₂O₄ is able to absorb a great amount of solar photons and increase the intensity of the light response of β-Bi₂O₃.

Acknowledgements

The authors are grateful for the support from the National Natural Science Foundation of China (NSFC. 51374259), the Fundamental and advanced research projects of Chongqing Science and Technology Commission (2013jjB20001), and Fundamental and Chongqing Graduate Student Research Innovation Project (CYB14040).

References

1. A. H. K. Fujishima, Nature, 1972, 238, 37–39.
2. J. H. Carey, J. Lawrence and H. M. Tosine, Bull. Environ. Contam. Toxicol., 1976, 16, 697–701.
3. S. N. Frank and A. J. Bard, J. Am. Chem. Soc., 1977, 99, 303–308.
4. S. N. Frank and A. J. Bard, J. Chem. Phys., 1977, 81, 1484–1489.
5. S. N. Frank and A. J. Bard, J. Am. Chem. Soc., 1977, 14, 4667–4675.
6. C. Ooka, S. Akita and Y. Ohashi, J. Mater. Chem., 1999, 9, 2943–2952.
7. A. N. Murashkevich, O. A. Alisienok, I. M. Zharskiy and E. K. Yukhno, J. Sol-Gel Sci. Technol., 2013, 65, 367–373.
8. L. G. Devi and R. Kavitha, RSC. Adv., 2014, 4, 28265–28299.
9. M. S. Zhu, P. L. Chen and M. H. Liu, Chin. Sci. Bull., 2013, 58, 84–91.
10. Y. H. Yan, H. Y. Guan, S. Liu and R. Y. Jiang, Ceram. Int., 2014, 40, 9095–9100.
11. W. Y. Shan and X. F. Bai, Chem. Adhesion., 2009, 31, 35–38.
12. X. J. Lu, P. Yin and Z. Z. Han, Front. Optoelectron., 2012, 5, 439–444.
13. X. L. Dong, Y. Shao, X. X. Zhang, H. C. Ma, X. F. Zhang, F. Shi, C. Ma and M. Xue, Res. Chem. Intermed., 2014, 40, 2953–2961.
14. Y. Wang, Z. Q. Shi, C. M. Fan, X. G. Hao, G. Y. Ding and Y. F. Wang, Int. J. Miner., Metall. Mater., 2012, 19, 467–472.
15. T. T. Lea, M. S. Akhtara, D. M. Parka, J. C. Leec and O. B. Yang, Appl. Catal., B, 2012, 111–112, 397–401.
16. Y. J. Li, T. P. Cao, C. L. Shao and C. H. Wang, Int. J. Inorg. Mater., 2012, 27, 687–692.
17. M. Drache, P. Roussel and J. P. Wignacourt, Chem. Rev., 2007, 107, 80–96.
18. A. Cabot, A. Marsal and J. Arbiol, et al., Sens. Actuators, B, 2004, 99, 74–89.
19. L. J. Xu, L. Y. Huang, C. L. Liu and T. P. Xie, Chinese Patent, CN103447054A, 2013.
20. A. A. Aziz, Y. H. Yau, G. L. Puma, C. Fischer and S. Pichiah, Chem. Eng. J., 2014, 235, 264–274.
21. K. S. Chee, M. Daimu, A. Nobuyasu and O. Toshitaka, J. Am. Ceram. Soc., 2009, 45, 34–39.
22. G. Thirupathi and R. Singh, AIP Conf. Proc., 2013, 1591, 470–471.
23. S. H. Wang and S. Q. Zhou, Appl. Surf. Sci., 2010, 256, 6191–6198.
24. C. Xie, L. J. Xu, Y. J. Ye, X. Y. Li and S. Y. Wang, Chin. J. Geochem., 2015, 34, 219–223.
25. H. F. Cheng, B. B. Huang, J. B. Lu, Z. Y. Wang, B. Xu, X. Y. Qin, X. Y. Zhang and Y. Dai, Phys. Chem. Chem. Phys., 2010, 12, 15468–15475.
26. Q. H. Yan, L. Yang and L. J. Yang, Chin. J. Pharm. Anal., 2011, 31, 1393–1396 (chinese).
27. H. F. Shi, Spectrosc. Spectral Anal., 2004, 24, 229–232 (chinese).
28. X. M. Tang, X. R. Ren and Y. Q. Fang, Environ. Res. Eng. Manag., 2013, 27, 831–834 (chinese).
29. T. P. Xie, C. L. Liu, L. J. Xu, J. Yang and W. Zhou, J. Phys. Chem. C, 2013, 117, 24601–24610.
30. N. Pugazhenthiran, P. Sathishkumar, S. Murugesan and S. Anandan, Chem. Hem. Eng. J., 2011, 168, 1227–1233.
31. Y. F. Qiu, M. L. Yang, H. B. Fan, Y. Z. Zuo and Y. Y. Shao, Mater. Lett., 2011, 65, 780–782.
32. O. Monnereau, L. Tortet, P. Llewellyn, F. Rouquerol and G. Vacquier, Solid State Ionics, 2003, 157, 163–169.
33. T. Sakata, J. Photochem., 1985, 29, 205–215.
34. Y. Yang, H. Zhong and C. X. Tian, Res. Chem. Intermed., 2011, 37, 91–102.
35. T. Sakata, J. Photochem., 1985, 29, 205–215.
36. H. Kisch, Angew. Chem., Int. Ed., 2013, 52, 812–847.
37. C. F. Li and Q. Wang, Phys. B, 1999, 269, 22–27.
38. M. Torres and A. Kunold, Phys. Lett. A, 2004, 323, 290–297.
39. B. Neppolian, Y. Kim, M. Ashokkumar, H. Yamashita and H. Choi, J. Hazard. Mater., 2010, 182, 557–562.

---