The Efficient Photocatalytic Degradation of Organic Pollutants on the MnFe2O4/BGA Composite under Visible Light

The MnFe2O4/BGA (boron-doped graphene aerogel) composite was prepared by hydrothermal treatment of MnFe2O4 particles, boric acid, and graphene oxide. When applied as a photo-Fenton catalyst for the degradation of rhodamine B, the MnFe2O4/BGA composite yielded a degradation efficiency much higher than the sum of those of individual MnFe2O4 and BGA under identical experimental conditions, indicating a strong synergetic effect established between MnFe2O4 and BGA. The catalytic degradation of rhodamine B was proved to follow pseudo first-order kinetics, and the apparent reaction rate constant on the MnFe2O4/BGA composite was calculated to be three- and seven-fold that on BGA and MnFe2O4, respectively. Moreover, the MnFe2O4/BGA composite also demonstrated good reusability and could be reused for four cycles without obvious loss of photocatalytic activity.


Introduction
Recently, the spinel-type bimetal oxide MnFe 2 O 4 nanoparticles have been intensively investigated as photo-Fenton catalysts for the degradation of organic pollutants in wastewater [1], mainly owing to their perfect chemical stability, efficient visible light response, and good catalytic performance [2,3]. However, pure MnFe 2 O 4 does not demonstrate enough efficiency under visible light irradiation, because of its low conductivity, easy aggregation, and quick recombination of photo-generated electron-hole pairs [4]. To overcome these disadvantages, researchers have paid much attention to the composites of MnFe 2 O 4 nanoparticles and other materials, including metal/semiconducting nanoparticles and twodimensional sheets. Qin et al. synthesized a composite of MnFe 2 O 4 and gold nanoparticles for photo-Fenton degradation of tetracycline (TC) under neutral pH [5]. It was found that the synergistic effect between MnFe 2 O 4 and gold nanoparticles endowed the MnFe 2 O 4 /Au composite with quite good photo-Fenton catalytic performance. Zhao et al. prepared flower-like SnS 2-loaded MnFe 2 O 4 nanocomposites and demonstrated that SnS 2 could effectively inhibit electron-hole pair recombination [6]. Later, they also reported a ternary MnFe 2 O 4 /CeO 2 /SnS 2 photocatalyst [7], which exhibited much higher photocatalytic activity than MnFe 2 O 4 particles toward the degradation of methylene blue (MB) under visible light irradiation. As for the two-dimensional sheets composited with MnFe 2 O 4 nanoparticles, they are only limited to C 3 N 4 and graphene-based materials. Vignesha et al. applied the nanocomposites (MnFe 2 O 4 /g-C 3 N 4 /TiO 2 ) to the photo-degradation of methyl orange (MO) and ascribed their much-enhanced photocatalytic activity to the synergistic effect between TiO 2 , g-C 3 N 4 , and MnFe 2 O 4 [8]. As a matter of fact, MnFe 2 O 4 nanoparticles were commonly reported to be immobilized on graphene-based matrices to accelerate the transfer of photo-induced carriers and to reduce the electron-hole pair recombination rate. Fu et al. first reported the improved photocatalytic activity of MnFe 2 O 4 /graphene composite for the degradation of MB in the absence of hydrogen peroxide [9]. Subsequently,

Preparation of the MnFe 2 O 4 /BGA Composite
The MnFe 2 O 4 nanoparticles were prepared by a co-precipitation procedure [24]. Briefly, MnCl 2 ·4H 2 O (0.3919 g) and FeCl 3 ·6H 2 O (1.0931 g) were dissolved in deionized water (50 mL), and then adjusted with 6 mol/L NaOH until pH = 12 under vigorous mechanical stirring. MnFe 2 O 4 nanoparticles precipitated while the above suspensions were stirred at 100 • C for 4 h. After washing alternately with water and ethanol 3 times to remove some adsorbed impurities and drying in a vacuum oven at 60 • C for 12 h, the product was ground to fine powder in a mortar.
Graphene oxide (GO) was prepared from specpure graphite powder by a modified Hummers method [25]. In brief, KMnO 4 (3.0 g) was gradually added to a mixture of concentrated H 2 SO 4 (23 mL) and graphite powder (1.0 g) cooled in an ice bath (to keep the temperature in the range of 0-10 • C), and then magnetically stirred for 0.5 h at 50 • C. When 18 mL of deionized water was slowly added, the mixture was stirred for another 10 min at 95 • C. After 35 mL deionized water and 3 mL H 2 O 2 (30 wt%) were successively added, the reactant mixture was centrifuged (DT5A, Hunan Kaida Scientific Instruments Co., Ltd., Changsha, China) at 5000 rpm for 10 min. The collected centrifugal precipitate was dispersed in 100 mL 10% HCl and then centrifuged again. After the above dispersioncentrifugation process was repeated 3 times, the centrifugal precipitate was collected and dispersed in 100 mL deionized water and subjected to the recycle of dispersioncentrifugation process until pH = 6.0. The exfoliated GO was obtained after the centrifugal precipitate was dried at 50 • C for 24 h under vacuum.
For the synthesis of MnFe 2 O 4 /BGA composite, MnFe 2 O 4 (65 mg) and H 3 BO 3 (230 mg) were added into 25 mL aqueous dispersion of GO (2 mg/mL). After stirring for 30 min at room temperature, the mixture was transferred to a Teflon-lined stainless steel autoclave and hydrothermally treated at 180 • C for 12 h. When the autoclave cooled down to ambient temperature, the product was collected and soaked in an appropriate amount of aqueous solution of ethanol (20%) for 12 h. Finally, the MnFe 2 O 4 /BGA composite was obtained after freeze-drying treatment of the product for 48 h. Analogously, BGA and the MnFe 2 O 4 /GA composite were also synthesized in the same procedure without the addition of MnFe 2 O 4 or H 3 BO 3 , respectively.

Photo-Fenton Degradation of Organic Pollutants on the MnFe 2 O 4 /BGA Composite
The photo-Fenton activity of MnFe 2 O 4 /BGA composite was evaluated by photocatalytic degradation of such organic dyes as RhB, crystal violet, or MB under visible-light irradiation. Typically, 2 mg catalyst was dispersed in 50 mL aqueous solution of organic dye (10 mg/L) and stirred for 90 min in the dark. A 300 W xenon lamp (CEL-HXF-300, Beijing, China) coupled with a 400 nm cut-off filter was used as a visible light source. The photo-Fenton degradation was initiated by injection of 1.0 mL 30% H 2 O 2 under visible light. At certain time intervals, 3 mL aliquots were withdrawn by pipette and filtered with 0.22 µm membrane, and the concentration of residual organic dyes in the filtrates was detected by their UV-vis spectra.

Characterization Methods
Morphologies were observed by scanning electron microscopy (SEM, QUANTA 200S, FEI, Eindhoven, The Netherlands) and a high-resolution transmission electron microscope (TEM, JEM2100, JEOL, Tokyo, Japan). Structure and composition were investigated by X-ray diffraction (XRD, D8 Advance, Bruker, Berlin, Germany) and X-ray photoelectron spectroscopy (XPS, KRATOS, Stretford, UK). Brunauer-Emmett-Teller (BET) surface area was collected by N 2 adsorption-desorption method at 77 K on a Micromeritics ASAP 2010M analyzer. Fourier transform infrared spectra (FTIR) were recorded by a Perkin Elmer (Waltham, MA, USA) Spectrum 100FT-IR spectrometer. The concentration of organic dyes was measured by UV-vis absorption spectrophotometer (UV-3600, Shimadzu, Kyoto, Japan). Thermal stability of the sample was estimated using thermo-gravimetric analysis (TGA, SDT2960). Raman scattering was carried out on a Jobin Yvon (Palaiseau, France) HR 800 micro-Raman spectrometer with 458 nm excitation from a 20 mW air-cooled argon ion laser. The magnetic behavior of MnFe 2 O 4 /BGA was recorded with an MPMS-3 superconducting quantum interference device (SQUID) magnetometer in the applied field range of ±20 kOe at room temperature. The UV-vis diffuse reflectance spectra (DRS) of the samples were determined using a UV-vis spectrophotometer combined with the powerful operating software UVProbe (Shimadzu/UV-2550). Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were conducted on an SP-300 electrochemical workstation (Bio-Logic, Seyssinet-Pariset, France) in a three-electrode system with 1 mol/L Na 2 SO 4 as electrolyte, and the working, reference, and counter electrodes were MnFe 2 O 4 (BGA, MnFe 2 O 4 /GA or MnFe 2 O 4 /BGA), Ag/AgCl electrode, and Pt foil, respectively. Electron spin resonance (ESR) spectra were recorded on a Bruker A300 ESR spectrometer. The recombination rate of electron-hole pairs was detected with the Edinburgh FLS1000 fluorescence and phosphorescence spectrometer.

Materials Characterizations
The morphology of MnFe 2 O 4 /BGA composite was explored through SEM and TEM observations. From Figure 1a,b, it can be seen that the as-prepared MnFe 2 O 4 is composed of irregular particles with an average size around 60 nm, and the bare BGA forms a welldefined and interconnected 3D porous network structure. The average pore size is around 4 micrometers, and the pore walls consist of thin layers of stacked graphene nanosheets. Figure 1c shows that a large number of MnFe 2 O 4 nanoparticles are evenly anchored on the 3D hierarchical networks of BGA. SEM mapping technology was also applied to the analysis of element distribution. As demonstrated in Figure 1e-    Transmission electron microscope observation also confirmed the formation of MnFe2O4/BGA composite. From Figure 2a, it can be seen that the MnFe2O4 nanoparticles with an average size around 16 nm are randomly distributed in the hierarchical networks of BGA. From the HRTEM image shown in Figure 2b, a MnFe2O4 nanoparticle is observed. It is obvious that the typical d-spacing of 0.245 nm for those well-resolved lattice fringes corresponds to the (400) plane of MnFe2O4 crystal.   XPS analyses were employed to elucidate the surface chemical bonding states of the MnFe2O4/BGA composite. As shown in Figure 4a, the survey spectrum evidences the existence of Mn 2p, Fe 2p, C 1s, O 1s, and B 1s regions. It is should be noted that the B 1s peak locates at 192.5 eV, which is larger than that of elemental B (187.1 eV), indicating that B atoms from H3BO3 are successfully doped into the carbon networks of graphene after hydrothermal processing [23]. Based on the XPS quantification analysis, the molar content of boron in the MnFe2O4/BGA composite was determined to be 0.93%. In the deconvoluted Mn 2p spectrum (Figure 4b), the two peaks appearing at binding energies of 642.2 and 653.6 eV are attributed to the Mn 2p3/2 and Mn 2p1/2, respectively, confirming the presence of Mn 2+ ions in the MnFe2O4/BGA composite. In the deconvoluted Fe 2p spectrum (Figure 4c), the two main peaks at 710.9 and 724.6 eV are assigned to the Fe 2p3/2 and Fe 2p1/2, respectively, and the two satellite peaks detected at 716.2 and 732.1 eV offer further support for the presence of the Fe 3+ ions in the MnFe2O4/BGA composite [34,35]. In the deconvoluted B 1s spectrum (Figure 4d), the intense peak at 192.2 eV corresponds to in-plane -BC3 type bond, which was formed by the substitution of B for C in the hexagonal lattice of graphene [21]. However, the weak peak at 192.98 eV was ascribed to the presence of boric acid ester  Figure 3b are the Raman spectra of MnFe 2 O 4 , BGA, and MnFe 2 O 4 /BGA composite. The characteristic D and G scattering bands of graphite are clearly observed for BGA and MnFe 2 O 4 /BGA composite. It is worth noting that the G-band shifts from 1599 cm −1 for GO to around 1584.5 cm −1 for BGA and MnFe 2 O 4 /BGA composite after hydrothermal processing, evidencing that GO was largely reduced. Furthermore, the D band of BGA or MnFe 2 O 4 /BGA is wider than that of GA, confirming the successful incorporation of B atoms into the carbon matrix of graphene [21]. In addition, the intensity ratios between D and G bands (I D /I G ) are nearly same for BGA and MnFe 2 O 4 /BGA composite, implying that the loading of MnFe 2 O 4 nanoparticles has little effect on the porous network structure of BGA. On the other hand, the characteristic scattering peak of MnFe 2 O 4 is also observable at 612 cm −1 in MnFe 2 O 4 and MnFe 2 O 4 /BGA, offering additional evidence for the formation of MnFe 2 O 4 /BGA composite.

Displayed in
The thermo-stability of MnFe 2 O 4 /BGA composite was estimated by thermogravimetric analysis (TGA). It can be observed from Figure 3c that the greatest weight loss occurs in the temperature range of 400-600 • C, corresponding to the burning of the carbon skeletons of BGA. On the other hand, the weight ratio of MnFe 2 O 4 to GO deduced from the TGA curve is consistent with the mixing weight ratio of MnFe 2 O 4 to GO before hydrothermal processing, indicating that the doped boron element in BGA is not high enough to be detectable by TGA. Figure 3d displays the FT-IR spectra of as-synthesized MnFe 2 O 4 , BGA, and MnFe 2 O 4 /BGA composite. It is clear that the bands ascribed to the sp 2 -hybridized C=C in-plane, the B-C, and O-B-O bond stretching vibrations are observed at 1573, 1030, and 655 cm −1 , respectively [27][28][29], indicative of the successful doping of boron in the graphene frameworks both in BGA and in MnFe 2 O 4 /BGA composite [30]. The peak at about 1721 cm −1 is due to the C=O stretching vibrations from carbonyl. Additionally, the peak at 1195 cm −1 is attributed to the infrared vibration of the C-O bond. Obviously, the distribution of unreduced oxygen-containing groups on BGA facilitates the loading of MnFe 2 O 4 nanoparticles. Furthermore, the characteristic intense peak observed at 571 cm −1 is due to the lattice absorption of Fe-O in MnFe 2 O 4 , which shifted to 541 cm −1 in the MnFe 2 O 4 /BGA composite [31,32]. The shifting in the Fe-O bond vibrations is indicative of the presence of a relatively strong interaction between MnFe 2 O 4 and BGA in the MnFe 2 O 4 /BGA composite, which might be a result of the competitive effect between the electron-deficient boron element and the positively charged iron ions toward the valence electrons of oxygen atoms in the lattice of MnFe 2 O 4 nanoparticles [33].
XPS analyses were employed to elucidate the surface chemical bonding states of the MnFe 2 O 4 /BGA composite. As shown in Figure 4a, the survey spectrum evidences the existence of Mn 2p, Fe 2p, C 1s, O 1s, and B 1s regions. It is should be noted that the B 1s peak locates at 192.5 eV, which is larger than that of elemental B (187.1 eV), indicating that B atoms from H 3 BO 3 are successfully doped into the carbon networks of graphene after hydrothermal processing [23]. Based on the XPS quantification analysis, the molar content of boron in the MnFe 2 O 4 /BGA composite was determined to be 0.93%. In the deconvoluted Mn 2p spectrum (Figure 4b), the two peaks appearing at binding energies of 642.2 and 653.6 eV are attributed to the Mn 2p 3/2 and Mn 2p 1/2 , respectively, confirming the presence of Mn 2+ ions in the MnFe 2 O 4 /BGA composite. In the deconvoluted Fe 2p spectrum (Figure 4c), the two main peaks at 710.9 and 724.6 eV are assigned to the Fe 2p 3/2 and Fe 2p 1/2 , respectively, and the two satellite peaks detected at 716.2 and 732.1 eV offer further support for the presence of the Fe 3+ ions in the MnFe 2 O 4 /BGA composite [34,35]. In the deconvoluted B 1s spectrum (Figure 4d), the intense peak at 192.2 eV corresponds to in-plane -BC 3 type bond, which was formed by the substitution of B for C in the hexagonal lattice of graphene [21]. However, the weak peak at 192.98 eV was ascribed to the presence of boric acid ester (-BC 2 O) and boronic acid (-BCO 2 ) moieties [23]. Therefore, the deconvoluted B 1s spectrum confirms the successful doping of B atoms in the skeleton of graphene during hydrothermal processing. The incorporation of the B atom into the carbon skeleton can alter the original sp 2 -hybridized structure, induce uneven charge distribution, and form new active regions in favor of activating reactions. Particularly, the B atom with positive charge polarization substitution position (-BC 3 ) was the dominant species, which could serve as the activation region to migrate electrons rapidly, and thereby can facilitate the efficient separation of photogenerated carriers. In the deconvoluted C 1s spectrum (Figure 4e), the peak at 284.7 eV is ascribed to the sp 2 structure of C-C/C=C in graphene, and the peaks at 282.3, 285.6, and 287.7 eV could be assigned to C-B, C-O, and C=O, respectively [36]. In the deconvoluted O 1s spectrum (Figure 4f and carboxylic acid functionalities on its surface, which is consistent with the results of its FT-IR spectrum (Figure 3d). 282.3, 285.6, and 287.7 eV could be assigned to C-B, C-O, and C=O, respectively [36]. In the deconvoluted O 1s spectrum (Figure 4f), the four fitted peaks centered at 530.5, 531.7, 532.5, and 533.3 eV are assigned to the Mn-O-Mn, Fe-O-C, Fe-OH, and C-O/C=O bands, respectively. Notably, the presence of Fe-O-C illustrates the strong interaction between the iron species and graphene, which is beneficial for the rapid transfer of electrons during the photo-Fenton reaction. In addition, both the deconvoluted C 1s and O 1s spectra evidence that the MnFe2O4/BGA composite is rich in hydroxyl, carbonyl, and carboxylic acid functionalities on its surface, which is consistent with the results of its FT-IR spectrum ( Figure 3d).  Figure 5 are the N2 adsorption-desorption isotherms of MnFe2O4, BGA, and MnFe2O4/BGA. It can be seen that all of them belong to the typical type IV isotherm with an H3 hysteresis loop between the adsorption and desorption curves at higher relative pressure, indicating that these samples were composed of either flaky granular or fractured pore materials with flat slit-, crack-, or wedge-shaped mesoporous structures [37]. Moreover, the calculated specific surface area of MnFe2O4/BGA (136.7 m 2 /g) is much larger than that of BGA (108.2 m 2 /g) or pure MnFe2O4 (57.9 m 2 /g). Therefore, the  Figure 5 are the N 2 adsorption-desorption isotherms of MnFe 2 O 4 , BGA, and MnFe 2 O 4 /BGA. It can be seen that all of them belong to the typical type IV isotherm with an H3 hysteresis loop between the adsorption and desorption curves at higher relative pressure, indicating that these samples were composed of either flaky granular or fractured pore materials with flat slit-, crack-, or wedge-shaped mesoporous structures [37]. Moreover, the calculated specific surface area of MnFe 2 O 4 /BGA (136.7 m 2 /g) is much larger than that of BGA (108.2 m 2 /g) or pure MnFe 2 O 4 (57.9 m 2 /g). Therefore, the MnFe 2 O 4 /BGA composite might offer more reactive sites to accelerate the generation of free radicals. MnFe2O4/BGA composite might offer more reactive sites to accelerate the generation of free radicals.  Figure 6a are the UV-vis DRS spectra of pure MnFe2O4 and MnFe2O4/BGA, respectively. It is obvious that the MnFe2O4/BGA composite exhibits a much higher absorption band than pure MnFe2O4 in the visible and near-IR regions, evidencing its enhanced absorption property for visible and near-IR light. Furthermore, in line with Figure 6b, the optical band gap of MnFe2O4/BGA composite is estimated to be 1.75 eV, which is smaller than that of pure MnFe2O4 (2.09 eV). Thus, with the combination of BGA, the energy gap of MnFe2O4 in the MnFe2O4/BGA composite narrows, which is beneficial to the generation and separation of the photo-induced electron-hole pairs, and eventually to the improvement in photocatalytic performance under visible light irradiation.

The Optimized Experimental Conditions of Photo-Fenton Catalytic Degradation
For the optimization of photo-Fenton experimental conditions, the catalytic degradation of MnFe2O4/BGA composites on RhB was investigated under visible light irradiation. Shown in Figure 7a is the degradation efficiency dependence on the mass ratio of MnFe2O4 to BGA in MnFe2O4/BGA composite. With the increase in the mass ratio from 1.0 to 1.3, MnFe2O4/BGA composite might offer more reactive sites to accelerate the generation free radicals.  Figure 6a are the UV-vis DRS spectra of pure MnFe2O4 and MnFe2O4/BGA respectively. It is obvious that the MnFe2O4/BGA composite exhibits a much higher a sorption band than pure MnFe2O4 in the visible and near-IR regions, evidencing its enhance absorption property for visible and near-IR light. Furthermore, in line with Figure 6b, the o tical band gap of MnFe2O4/BGA composite is estimated to be 1.75 eV, which is small than that of pure MnFe2O4 (2.09 eV). Thus, with the combination of BGA, the energy ga of MnFe2O4 in the MnFe2O4/BGA composite narrows, which is beneficial to the generatio and separation of the photo-induced electron-hole pairs, and eventually to the improv ment in photocatalytic performance under visible light irradiation.

The Optimized Experimental Conditions of Photo-Fenton Catalytic Degradation
For the optimization of photo-Fenton experimental conditions, the catalytic degrad tion of MnFe2O4/BGA composites on RhB was investigated under visible light irradiatio Shown in Figure 7a is the degradation efficiency dependence on the mass ratio of MnFe2O to BGA in MnFe2O4/BGA composite. With the increase in the mass ratio from 1.0 to 1   Figure 7a is the degradation efficiency dependence on the mass ratio of MnFe 2 O 4 to BGA in MnFe 2 O 4 /BGA composite. With the increase in the mass ratio from 1.0 to 1.3, the degradation efficiency was greatly improved, from 39.6% to 92.3%. However, with the further increase in the mass ratio from 1.3 to 1.5, the degradation efficiency was largely reduced to 62.7%. The influence of the dosage of MnFe 2 O 4 /BGA composite on the degradation efficiency of RhB is shown Figure 7b. With the increase in the catalyst dosage from 1 to 2 mg, the degradation efficiency increased from 33.7% to 92.3% in 90 min. Nevertheless, with a further increase in the catalyst dosage from 2 to 4 mg, the degradation efficiency slightly declined to 85%. Shown in Figure 7c is the degradation efficiency dependence on the dosage of H 2 O 2 . With the increase in the dosage of H 2 O 2 from 0.5 to 1.0 mL, the degradation efficiency increased rapidly from 62% to 92.3%. However, with the further increase of H 2 O 2 from 1.0 mL to 1.5 mL, the degradation efficiency decreased to 86.1%. The effect of RhB initial concentration on its photo-degradation on the MnFe 2 O 4 /BGA composite was also examined. As demonstrated in Figure 7d, the degradation efficiency within 75 min was monotonically decreased with the increase in the initial RhB concentration from 5 to 10 and to 20 mg/L. Nonetheless, the differences in degradation efficiency among the three cases became smaller and smaller after 45 min, and the degradation efficiency ultimately reached 91.7%, 92.3%, and 91.1%, respectively, at 90 min. Therefore, the optimized parameters in photo-Fenton catalytic degradation of RhB on MnFe 2 O 4 /BGA composite were summarized as follows: 2 mg MnFe 2 O 4 /BGA−1.3, 1.0 mL H 2 O 2 , 10 mg/L RhB in 50 mL aqueous solution.
Nanomaterials 2021, 11, x FOR PEER REVIEW 10 of the degradation efficiency was greatly improved, from 39.6% to 92.3%. However, with further increase in the mass ratio from 1.3 to 1.5, the degradation efficiency was larg reduced to 62.7%. The influence of the dosage of MnFe2O4/BGA composite on the deg dation efficiency of RhB is shown Figure 7b. With the increase in the catalyst dosage fr 1 to 2 mg, the degradation efficiency increased from 33.7% to 92.3% in 90 min. Nevert less, with a further increase in the catalyst dosage from 2 to 4 mg, the degradation e ciency slightly declined to 85%. Shown in Figure 7c is the degradation efficiency depen ence on the dosage of H2O2. With the increase in the dosage of H2O2 from 0.5 to 1.0 m the degradation efficiency increased rapidly from 62% to 92.3%. However, with the f ther increase of H2O2 from 1.0 mL to 1.5 mL, the degradation efficiency decreased to 86. 1 The effect of RhB initial concentration on its photo-degradation on the MnFe2O4/BG composite was also examined. As demonstrated in Figure 7d, the degradation efficien within 75 min was monotonically decreased with the increase in the initial RhB conc tration from 5 to 10 and to 20 mg/L. Nonetheless, the differences in degradation efficien among the three cases became smaller and smaller after 45 min, and the degradation e ciency ultimately reached 91.7%, 92.3%, and 91.1%, respectively, at 90 min. Therefore, optimized parameters in photo-Fenton catalytic degradation of RhB on MnFe2O4/BG composite were summarized as follows: 2 mg MnFe2O4/BGA−1.3, 1.0 mL H2O2, 10 mg RhB in 50 mL aqueous solution.

Degradation of Rhodamine B on Relevant Photo-Fenton Catalysts
For the comparison of degradation efficiency of relevant photo-Fenton catalysts, degradation of RhB was investigated under the above optimized conditions on MnFe2 /BGA follows the pseudo firstorder kinetics, and the calculated reaction rate constants are 3.59 × 10 −3 , 7.9 × 10 −3 , 1.11 × 10 −2 , and 2.495 × 10 −2 min −1 , respectively. In line with the definition of synergistic index (SI) [38], SI was calculated to be 2.16, i.e., the synergistic effect yielded an additional 116% efficiency to the degradation of RhB. Thus, it was concluded that the photocatalytic potency of MnFe 2 O 4 /BGA composite is much higher than pure MnFe 2 O 4 , pure BGA, and MnFe 2 O 4 /GA composite. Nanomaterials 2021, 11, x FOR PEER REVIEW 11 of BGA, MnFe2O4/GA, and MnFe2O4/BGA, respectively. It can be seen from Figure 8a th the degradation efficiency of MnFe2O4/BGA to RhB is much larger than that MnFe2O4/GA, indicative of the importance of boron doping in GA. Moreover, the deg dation efficiency of RhB within 90 min was 28.1%, 53.1%, and 92.3% on MnFe2O4, BG and MnFe2O4/BGA, respectively. Because the degradation efficiency on MnFe2O4/BG composite is much larger than the sum of the degradation efficiencies on individu MnFe2O4 and BGA under identical experimental conditions, it was concluded that a sy ergistic effect must be established between MnFe2O4 and BGA during the photo-Fent degradation of RhB. It could be discovered from Figure 8b that the catalytic degradati of RhB by MnFe2O4, BGA, MnFe2O4/GA, and MnFe2O4/BGA follows the pseudo first-ord kinetics, and the calculated reaction rate constants are 3.59 × 10 −3 , 7.9 × 10 −3 , 1.11 × 10 −2 , a 2.495 × 10 −2 min −1 , respectively. In line with the definition of synergistic index (SI) [38], was calculated to be 2.16, i.e., the synergistic effect yielded an additional 116% efficien to the degradation of RhB. Thus, it was concluded that the photocatalytic potency MnFe2O4/BGA composite is much higher than pure MnFe2O4, pure BGA, a MnFe2O4/GA composite. In order to compare with the photo-Fenton degradation, the degradation of RhB the dark using various photocatalysts and H2O2 was also carried out, and the results a displayed in Figure 9. It can be seen from Figure 9 that the degradation rates of RhB a 8.3%, 49.2%, 56.16%, and 78.8% for MnFe2O4, BGA, MnFe2O4/GA, and MnFe2O4/BGA, spectively, which are uniformly lower than observed with the corresponding photo-Fe ton degradation processes. Nonetheless, a synergistic effect was also observed betwe MnFe2O4 and BGA during the Fenton degradation of RhB. In order to compare with the photo-Fenton degradation, the degradation of RhB in the dark using various photocatalysts and H 2 O 2 was also carried out, and the results are displayed in Figure 9. It can be seen from Figure 9 that the degradation rates of RhB are 8.3%, 49.2%, 56.16%, and 78.8% for MnFe 2 O 4 , BGA, MnFe 2 O 4 /GA, and MnFe 2 O 4 /BGA, respectively, which are uniformly lower than observed with the corresponding photo-Fenton degradation processes. Nonetheless, a synergistic effect was also observed between MnFe 2 O 4 and BGA during the Fenton degradation of RhB. The broad application of as-synthesized MnFe2O4/BGA composite was studied for

Degradation of Crystal Violet and Methylene Blue on the MnFe 2 O 4 /BGA Composite
The broad application of as-synthesized MnFe 2 O 4 /BGA composite was studied for degradation of organic pollutants. The organic pollutants applied in textile industry, including crystal violet (C 25 H 30 N 3 Cl) and methylene blue (C 16 H 18 N 3 ClS), were selected as degradation targets. All are listed as carcinogenic substances with molecular structures containing multiple benzene rings. Shown in Figure 10 are the UV-vis spectra of the organic pollutants in aqueous solutions recorded after catalytic degradation on the MnFe 2 O 4 /BGA composite at certain reaction intervals. It is obvious that the absorbance for the typical absorption peaks of the organic pollutants decreases gradually with prolonged time, and all of the solutions become colorless at 90 min, as displayed in the insets of Figure 10. Thus, the Fenton-like process on the MnFe 2 O 4 /BGA composite is non-selective for the degradation of these organic pollutants, indicative of the wide applicability of the MnFe 2 O 4 /BGA composite.

Degradation of Crystal Violet and Methylene Blue on the MnFe2O4/BGA Composite
The broad application of as-synthesized MnFe2O4/BGA composite was studied fo degradation of organic pollutants. The organic pollutants applied in textile industry, in cluding crystal violet (C25H30N3Cl) and methylene blue (C16H18N3ClS), were selected a degradation targets. All are listed as carcinogenic substances with molecular structure containing multiple benzene rings. Shown in Figure 10 are the UV-vis spectra of the or ganic pollutants in aqueous solutions recorded after catalytic degradation on th MnFe2O4/BGA composite at certain reaction intervals. It is obvious that the absorbance for th typical absorption peaks of the organic pollutants decreases gradually with prolonged time and all of the solutions become colorless at 90 min, as displayed in the insets of Figure 10 Thus, the Fenton-like process on the MnFe2O4/BGA composite is non-selective for the deg radation of these organic pollutants, indicative of the wide applicability of th MnFe2O4/BGA composite.  Figure 11 are the catalytic degradation performance curves of MB and crystal violet on the MnFe2O4/BGA composite. Nonetheless, the degradation efficiency o MB and crystal violet reached 90.6% and 88.7% within 90 min, respectively, confirmin the wide applicability of the MnFe2O4/BGA composite for these organic pollutants. More over, as shown in Figure 11b, these organic pollutant degradation kinetics based o pseudo first-order fit well with the experimental data under visible light irradiation. Th

Stability and Reusability of the MnFe 2 O 4 /BGA Composite
Effective magnetic separation of the MnFe 2 O 4 /BGA nanocomposite is important for its recyclable use in wastewater treatment. As shown in Figure 12a, the MnFe 2 O 4 /BGA composite displays a symmetrical S-shaped magnetization curve, evidencing the superparamagnetic behavior of this composite. The saturation magnetization was measured to be 7.25 emu g −1 , which is high enough for the magnetic separation of the catalyst with an external magnet in the degradation solution of RhB (see the inset of Figure 12a). when MB and crystal violet were photo-Fenton degraded on the MnFe2O4/BGA composite, respectively, which further supports the very good photocatalytic performance of the MnFe2O4/BGA composite for photodegradation of these organic dyes under visible light irradiation. In comparison to the data reported previously, the calculated reaction rate constant of MnFe2O4/BGA toward MB is 4.6 times that of MnFe2O4/rGO (5.26 × 10 −3 min −1 ) [12] and 1.7 times that of 10% MnFe2O4/rGO (1.443 × 10 −2 min −1 ) [39], evidencing the much improved photo-Fenton degradation activity of the MnFe2O4/BGA composite.

Stability and Reusability of the MnFe2O4/BGA Composite
Effective magnetic separation of the MnFe2O4/BGA nanocomposite is important for its recyclable use in wastewater treatment. As shown in Figure 12a, the MnFe2O4/BGA composite displays a symmetrical S-shaped magnetization curve, evidencing the superparamagnetic behavior of this composite. The saturation magnetization was measured to be 7.25 emu g −1 , which is high enough for the magnetic separation of the catalyst with an external magnet in the degradation solution of RhB (see the inset of Figure 12a).
Undoubtedly, the stability of the catalyst is quite crucial for practical applications. Batch experiments were carried out to evaluate the photocatalytic stability of the MnFe2O4/BGA composite for RhB degradation. It can be seen from Figure 12b that no apparent deactivation was observed after 4 cycles, indicative of the high stability and repeatability of the MnFe2O4/BGA composite for visible-light driven photocatalytic degradation of RhB. In line with the definition of "turnover number" proposed by Gomathi Devi and Shyamala [40], the turnover number after the 4th cycle was calculated to be 0.2611. Moreover, the recovered MnFe2O4/BGA composite (after the 4th cycle) was investigated by FT-IR spectroscopy. Shown in Figure 12c are the FT-IR spectra of the fresh and recovered MnFe2O4/BGA composite. It is obvious that the characteristic IR absorption peaks of the initial MnFe2O4/BGA composite observed in Figure 3d are still retained in the recovered sample. Nonetheless, two weak bands detected at 1074 and 1179 cm −1 in the recovered sample (see the inset of Figure 12c) were assignable to the bending vibration of aromatic C-H bonds in an intermediate product formed during the degradation process. Meanwhile, the lattice absorption of Fe-O detected in the recovered MnFe2O4/BGA composite was blue-shifted, indicative of the change in its chemical environment after reusing 4 times. Such blue shift might be due to the effect of Verway hopping (Mn 2+ + Fe 3+ →Mn 3+ + Fe 2+ ) [41], which leads to the increase in electron cloud density around iron ions. Thus,

Possible Degradation Mechanism
It is well accepted that hydroxyl radical (•OH) is the specific active species in Fenton or Fenton-like systems. To identify the major reactive species formed in the present system, trapping experiments were performed with ammonium oxalate (AO), isopropanol (IPA), and benzoquinone (BQ) as the scavengers to quench h + , •OH, and •O2 − , respectively. As shown in Figure 13a, the removal rate of RhB was decreased by 70.6% and 68.1% in the presence of IPA and OA, respectively, whereas the removal rate was only decreased 10.4% in the presence of BQ. Therefore, the photocatalytic degradation of RhB takes place mainly via photo-generated eand h + on the MnFe2O4/BGA composite under visible light illumination.
On the other hand, the EPR spin-trap method, with 5,5-dimethy-1-pyrroline N-oxide (DMPO) as spin trapping active species, was performed to evaluate the radicals produced in the photodegradation of RhB under visible light illumination. As shown in Figure 13b, Undoubtedly, the stability of the catalyst is quite crucial for practical applications. Batch experiments were carried out to evaluate the photocatalytic stability of the MnFe 2 O 4 /BGA composite for RhB degradation. It can be seen from Figure 12b that no apparent deactivation was observed after 4 cycles, indicative of the high stability and repeatability of the MnFe 2 O 4 /BGA composite for visible-light driven photocatalytic degradation of RhB. In line with the definition of "turnover number" proposed by Gomathi Devi and Shyamala [40], the turnover number after the 4th cycle was calculated to be 0.2611. Moreover, the recovered MnFe 2 O 4 /BGA composite (after the 4th cycle) was investigated by FT-IR spectroscopy. Shown in Figure 12c are the FT-IR spectra of the fresh and recovered MnFe 2 O 4 /BGA composite. It is obvious that the characteristic IR absorption peaks of the initial MnFe 2 O 4 /BGA composite observed in Figure 3d are still retained in the recovered sample. Nonetheless, two weak bands detected at 1074 and 1179 cm −1 in the recovered sample (see the inset of Figure 12c) were assignable to the bending vibration of aromatic C-H bonds in an intermediate product formed during the degradation process. Meanwhile, the lattice absorption of Fe-O detected in the recovered MnFe 2 O 4 /BGA composite was blueshifted, indicative of the change in its chemical environment after reusing 4 times. Such blue shift might be due to the effect of Verway hopping (Mn 2+ + Fe 3+ →Mn 3+ + Fe 2+ ) [41], which leads to the increase in electron cloud density around iron ions. Thus, both the emerged weak bands and the strengthening in the lattice absorption of Fe-O were not results of the change in material structure of the MnFe 2 O 4 /BGA composite.

Possible Degradation Mechanism
It is well accepted that hydroxyl radical (•OH) is the specific active species in Fenton or Fenton-like systems. To identify the major reactive species formed in the present system, trapping experiments were performed with ammonium oxalate (AO), isopropanol (IPA), and benzoquinone (BQ) as the scavengers to quench h + , •OH, and •O 2 − , respectively. As shown in Figure 13a, the removal rate of RhB was decreased by 70.6% and 68.1% in the presence of IPA and OA, respectively, whereas the removal rate was only decreased 10.4% in the presence of BQ. Therefore, the photocatalytic degradation of RhB takes place mainly via photo-generated eand h + on the MnFe 2 O 4 /BGA composite under visible light illumination. The following electrochemistry and photoluminescence investigations strongly su port that the MnFe2O4/BGA composite possesses higher electronic conductivity and ele tron transfer efficiency than MnFe2O4, BGA, and the MnFe2O4/GA composite. Hence, t photo-generated electrons and holes on the MnFe2O4/BGA composite are more effective separated under visible light irradiation, which is mainly responsible for the excelle photocatalytic performance of the MnFe2O4/BGA composite in the degradation of organ pollutants under visible light irradiation.
It can be seen from Figure 14a that the MnFe2O4/BGA composite demonstrates t highest response current among the measured materials, indicative of its rapid electr transfer. The CV cures of BGA, MnFe2O4/GA, and MnFe2O4/BGA are of quasi-rectangu shapes, whereas that of MnFe2O4 is largely deviated from a quasi-rectangular shape. should be noted that the loop area within the CV curve follows the sequence MnFe2O4/BGA > MnFe2O4/GA > BGA > MnFe2O4, implying the highest electrochemic activity and electron transfer efficiency of MnFe2O4/BGA among these four materia which is consistent with the sequence of their surface areas (see Figure 5). Moreover, displayed in Figure 14b, the semicircle of the MnFe2O4/BGA composite in the high-fr quency region is the smallest among the tested materials. Because a smaller semicircle the electrochemical impedance spectroscopy (EIS) Nyquist plot means an overall smal charge transfer resistance and a faster interfacial charge transfer, the electron mobility On the other hand, the EPR spin-trap method, with 5,5-dimethy-1-pyrroline N-oxide (DMPO) as spin trapping active species, was performed to evaluate the radicals produced in the photodegradation of RhB under visible light illumination. As shown in Figure 13b Figure 5). Moreover, as displayed in Figure 14b, the semicircle of the MnFe 2 O 4 /BGA composite in the high-frequency region is the smallest among the tested materials. Because a smaller semicircle in the electrochemical impedance spectroscopy (EIS) Nyquist plot means an overall smaller charge transfer resistance and a faster interfacial charge transfer, the electron mobility is greatly accelerated and the recombination rate of photo-generated electrons and holes is effectively inhibited when the MnFe 2 O 4 /BGA composite is applied as a visible light photocatalyst. The following electrochemistry and photoluminescence investigations strongly su port that the MnFe2O4/BGA composite possesses higher electronic conductivity and ele tron transfer efficiency than MnFe2O4, BGA, and the MnFe2O4/GA composite. Hence, t photo-generated electrons and holes on the MnFe2O4/BGA composite are more effective separated under visible light irradiation, which is mainly responsible for the excelle photocatalytic performance of the MnFe2O4/BGA composite in the degradation of organ pollutants under visible light irradiation.
It can be seen from Figure 14a that the MnFe2O4/BGA composite demonstrates t highest response current among the measured materials, indicative of its rapid electr transfer. The CV cures of BGA, MnFe2O4/GA, and MnFe2O4/BGA are of quasi-rectangul shapes, whereas that of MnFe2O4 is largely deviated from a quasi-rectangular shape. should be noted that the loop area within the CV curve follows the sequence MnFe2O4/BGA > MnFe2O4/GA > BGA > MnFe2O4, implying the highest electrochemic activity and electron transfer efficiency of MnFe2O4/BGA among these four materia which is consistent with the sequence of their surface areas (see Figure 5). Moreover, displayed in Figure 14b, the semicircle of the MnFe2O4/BGA composite in the high-fr quency region is the smallest among the tested materials. Because a smaller semicircle the electrochemical impedance spectroscopy (EIS) Nyquist plot means an overall small charge transfer resistance and a faster interfacial charge transfer, the electron mobility greatly accelerated and the recombination rate of photo-generated electrons and holes effectively inhibited when the MnFe2O4/BGA composite is applied as a visible light ph tocatalyst.  It is well-known that the photoluminescence spectrum (PL) is an effective tool to estimate the recombination probability of photo-generated charge carriers [42]. As the emission signal is derived from the recombination of photo-induced e − , h + pairs, the weaker the emission signal is, the higher the separation efficiency of charge carriers. From As a matter of fact, the catalytic degradation of organic pollutants involves their adsorption on photocatalysts and subsequent oxidation by reactive oxidation species (ROSs). Because pre-adsorption can generate high concentrations of organic pollutants around photocatalysts, the contact probability between organic pollutants and ROSs is naturally enhanced, which benefits the following photocatalytic degradation of organic pollutants. In contrast to pure MnFe 2 O 4 , the MnFe 2 O 4 nanoparticles in MnFe 2 O 4 /BGA composite become smaller and are homogeneously loaded on the inner surface of BGA ( Figure 1). Meanwhile, the 3D interconnected porous structure of the MnFe 2 O 4 /BGA composite provides numerous channels for the quick diffusion and adsorption of organic pollutant molecules, mainly due to the strong π-π bonding between 3D graphene aerogel and aromatic organic pollutants. Moreover, the introduction of boron atoms also brings some defects to the networks of reduced graphene oxide. The specific surface area of the MnFe 2 O 4 /BGA composite is much larger than those of BGA and MnFe 2 O 4 ( Figure 5), and its adsorption rate in the dark is much higher than those of BGA, MnFe 2 O 4 , and the MnFe 2 O 4 /GA composite (Figure 8a). Thus, it is concluded that many more organic pollutant molecules were absorbed in the MnFe 2 O 4 /BGA composite than in BGA, MnFe 2 O 4 , or the MnFe 2 O 4 /GA composite, because of its much larger number of active sites and the stronger organic pollutant adsorption on each active site. It is well-known that the photoluminescence spectrum (PL) is an effective tool to estimate the recombination probability of photo-generated charge carriers [42]. As the emission signal is derived from the recombination of photo-induced e − , h + pairs, the weaker the emission signal is, the higher the separation efficiency of charge carriers. From Figure 15, it can be seen that the emission signal trended lower and lower in the order of pure MnFe2O4, BGA, MnFe2O4/GA, and MnFe2O4/BGA, i.e., the recombination rate of the photo-induced e − , h + pairs decreased in the same order. Therefore, the combination of BGA and MnFe2O4 can effectively inhibit the recombination of electron-hole pairs, which is mainly responsible for the excellent photo-Fenton performance of the MnFe2O4/BGA composite. It should be noted that the results extracted from Figure 15 are highly consistent with the photo-Fenton performance of pure MnFe2O4, BGA, MnFe2O4/GA, and MnFe2O4/BGA (Figure 8). As a matter of fact, the catalytic degradation of organic pollutants involves their adsorption on photocatalysts and subsequent oxidation by reactive oxidation species (ROSs). Because pre-adsorption can generate high concentrations of organic pollutants around photocatalysts, the contact probability between organic pollutants and ROSs is naturally enhanced, which benefits the following photocatalytic degradation of organic pollutants. In contrast to pure MnFe2O4, the MnFe2O4 nanoparticles in MnFe2O4/BGA composite become smaller and are homogeneously loaded on the inner surface of BGA ( Figure 1). Meanwhile, the 3D interconnected porous structure of the MnFe2O4/BGA composite provides numerous channels for the quick diffusion and adsorption of organic pollutant molecules, mainly due to the strong π-π bonding between 3D graphene aerogel and aromatic organic pollutants. Moreover, the introduction of boron atoms also brings some defects to the networks of reduced graphene oxide. The specific surface area of the MnFe2O4/BGA composite is much larger than those of BGA and MnFe2O4 ( Figure 5), and its adsorption rate in the dark is much higher than those of BGA, MnFe2O4, and the MnFe2O4/GA composite ( Figure 8a). Thus, it is concluded that many more organic pollutant molecules were absorbed in the MnFe2O4/BGA composite than in BGA, MnFe2O4, or the MnFe2O4/GA composite, because of its much larger number of active sites and the stronger organic pollutant adsorption on each active site.
Because of the semiconducting or semimetallic properties of such photocatalysts as MnFe2O4, MnFe2O4/GA, BGA, and MnFe2O4/BGA, light irradiation is indispensable for the photo-Fenton degradation of organic pollutants, and it is an important approach for the improvement in photo-Fenton performance to maximize the utilization efficiency of photocatalysts to visible light. On one hand, incident light reflectance and scattering multiplies in the 3D interconnected networks and channels (Figure 1c) of the MnFe2O4/BGA

Conclusions
The hydrothermally prepared MnFe2O4/BGA composite exhibited much enhanced photo-Fenton catalytic activity in the degradation of organic dyes owing to the synergistic effect between MnFe2O4 and BGA, which could be accounted for by the evenly distributed MnFe2O4 nanoparticles and the interconnected 3D porous network as well as the modulated electric charge distribution of BGA. Furthermore, because of the strong anchoring of MnFe2O4 nanoparticles on BGA, the MnFe2O4/BGA composite also demonstrated quite good stability for visible-light driven photocatalytic organic dye degradation. Moreover, it was found that the initial organic dye concentration, catalyst dosage, different content of MnFe2O4, and H2O2 dosage could significantly affect organic dye photodegradation on the MnFe2O4/BGA composite.  − radicals. Thus, the recombination of photo-excited electrons and holes will be greatly depressed, leading to enhanced visible light photoactivity to the degradation of organic pollutants on the MnFe 2 O 4 /BGA composite. This is consistent with the above trapping results that confirmed the important roles of the •OH and •O 2 − radicals.

Conclusions
The hydrothermally prepared MnFe 2 O 4 /BGA composite exhibited much enhanced photo-Fenton catalytic activity in the degradation of organic dyes owing to the synergistic effect between MnFe 2 O 4 and BGA, which could be accounted for by the evenly distributed

Conflicts of Interest:
There are no conflict to declare.