Synergistic Enhancement in Catalytic Performance of Superparamagnetic Fe3O4@Bacilus subtilis as Recyclable Fenton-Like Catalyst

Novel well-defined superparamagnetic Fe3O4@Bacilus subtilis composite (Fe3O4@B. subtilis SPMC) was synthesized through a facile electrostatic attraction method and used as a recyclable heterogeneous Fenton-like catalyst. With the presence of H2O2, Fe3O4@B. subtilis SPMC can remove nearly 87% of the doxycycline at the initial concentration of 50 mg L−1, exhibiting enhanced Fenton-like catalytic performance than pristine Fe3O4. The mechanism study demonstrates the synergistic effect between Bacilus subtilis adsorption and Fenton-like ability of Fe3O4 dominates the enhancement for Fenton-like catalytic efficiency of Fe3O4@B. subtilis SPMC. The obtained composite shows excellent recycling ability, reusability, and stability, which pave a new way for future design on highly efficient Fenton-like catalyst for degradation of organic pollutants.


Introduction
Magnetite (Fe 3 O 4 ) has attracted growing research interests in various fields, such as Li ion batteries [1], catalysts [2], drug delivery and targeting [3], etc., due to its distinguished physical and chemical properties [4,5]. However, these magnetite Fe 3 O 4 nanoparticles (MNPs) may aggregate into large clusters because of the anisotropic dipolar interactions, reducing their dispersibility and other specific properties and ultimately diminishing their activity [6]. In addition, another challenge that impedes the practical application of Fe 3 O 4 is the production of iron-containing waste sludge, which introduces secondary pollution [7,8]. Hence, to overcome these disadvantages, significant efforts have been made to immobilize the Fe 3 O 4 particles onto various support materials, simultaneously preserving their unique magnetic property [9][10][11][12].
Lately, microorganisms employed as support materials has being attracting great interest because of their major advantages such as ample resources, environmentally friendliness, and abundant functional groups [13]. Bacillus subtilis is a kind of genus Bacillus and can be easily found in water, soil, air, and decomposing plant matter [13][14][15]. Thus, owing to its distinguished physicochemical/biological properties, B. subtilis has been selected as an ideal candidate for the synthesis of composite materials for organic pollutants removal [13][14][15][16]. However, until now, no one has been focused on the preparation and application of integrated Fe 3 O 4 @B. subtilis heterogeneous catalyst.
In this study, we present a facile electrostatic attraction method for the coating of Fe 3 O 4 nanoparticles onto the surfaces of B. subtilis to form superparomagnetic recyclable Fe 3 O 4 @B. subtilis heterogeneous Fenton-like catalyst. The morphology, crystal structure, functional groups, magnetic property, and Fenton-like catalytic performance of Fe 3 O 4 @B. subtilis superparomagnetic composite (Fe 3 O 4 @B. subtilis SCP) are systematically characterized and evaluated. The catalytic performance of Fe 3 O 4 @B. subtilis SCP is significantly enhanced compared to the pristine Fe 3 O 4 . The prepared Fe 3 O 4 @B. subtilis SCP exhibits excellent recycling ability due to its superparamagnetic feature. The outstanding reusability was evaluated by successive batches of Doxycyline (DC) degradation, and the good chemical stability was also investigated by X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS). The leaching of iron ions from Fe 3 O 4 @B. subtilis was evaluated by inductively coupled plasma mass spectrometry (ICP-MS). The mechanism for the enhanced Fenton-like catalytic activity of Fe 3 O 4 @B. subtilis SCP is proposed and investigated through photoluminescence (PL) study.

Characterization
The pristine B. subtilis cells (Figure 1a) are rod-shaped with smooth surface and the length and width if which are approximately 1.4 ± 0.2 and 0.6 ± 0.1 µm, respectively (Supplementary Figure S1a,b). The synthesized Fe 3 O 4 @B. subtilis SPMC maintain the bacilliform morphology of B. subtilis cells, but with much rougher surfaces, indicating successful coating of Fe 3 O 4 magnetic nanoparticles ( Figure 1b). The rougher surfaces may be beneficial for the photocatalytic ability by providing a higher specific surface area (Supplementary Figure S2). Moreover, the slight increase in diameter (length = 1.45 ± 0.5 µm; width = 0.65 ± 0.3 µm (Supplementary Figure S1c,d)) in comparison with B. subtilis cells provides assertive evidence that the Fe 3 O 4 MNPs are attached onto the surfaces of B. subtilis. Furthermore, the magnification image (the insert image in Figure 1b) shows that some residual bare area on the surface of B. subtilis still remain, which might be used as adsorption sites for pollutants. The elemental composition of the prepared samples was studied by Energy Dispersive X-ray Spectroscopy (EDX) analysis. Comparing Figure 2c  heterogeneous Fenton-like catalyst. The morphology, crystal structure, functional groups, magnetic property, and Fenton-like catalytic performance of Fe3O4@B. subtilis superparomagnetic composite (Fe3O4@B. subtilis SCP) are systematically characterized and evaluated. The catalytic performance of Fe3O4@B. subtilis SCP is significantly enhanced compared to the pristine Fe3O4. The prepared Fe3O4@B. subtilis SCP exhibits excellent recycling ability due to its superparamagnetic feature. The outstanding reusability was evaluated by successive batches of Doxycyline (DC) degradation, and the good chemical stability was also investigated by X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS). The leaching of iron ions from Fe3O4@B. subtilis was evaluated by inductively coupled plasma mass spectrometry (ICP-MS). The mechanism for the enhanced Fenton-like catalytic activity of Fe3O4@B. subtilis SCP is proposed and investigated through photoluminescence (PL) study.

Characterization
The pristine B. subtilis cells (Figure 1a) are rod-shaped with smooth surface and the length and width if which are approximately 1.4 ± 0.2 and 0.6 ± 0.1 μm, respectively (Supplementary Figure S1a,b). The synthesized Fe3O4@B. subtilis SPMC maintain the bacilliform morphology of B. subtilis cells, but with much rougher surfaces, indicating successful coating of Fe3O4 magnetic nanoparticles (Figure 1b). The rougher surfaces may be beneficial for the photocatalytic ability by providing a higher specific surface area (Supplementary Figure S2). Moreover, the slight increase in diameter (length = 1.45 ± 0.5 μm; width = 0.65 ± 0.3 μm (Supplementary Figure S1c,d)) in comparison with B. subtilis cells provides assertive evidence that the Fe3O4 MNPs are attached onto the surfaces of B. subtilis. Furthermore, the magnification image (the insert image in Figure 1b) shows that some residual bare area on the surface of B. subtilis still remain, which might be used as adsorption sites for pollutants. The elemental composition of the prepared samples was studied by Energy Dispersive X-ray Spectroscopy (EDX) analysis. Comparing Figure 2c  The crystallographic structure of the as-obtained products was determined by XRD analysis (Figure 2a). The broad peak around 2θ = 20 • shows that B. subtilis is amorphous (black curve) [13,15]. All the diffraction peaks of Fe 3 O 4 MNPs (blue curve) could be readily indexed as face-centered cubic structured Fe 3 O 4 , which coincides well with the standard data (JCPDS card No. 19-0629) [17]. It is noted that the diffraction peaks corresponding to Fe 3 O 4 are also presented in the XRD pattern of The crystallographic structure of the as-obtained products was determined by XRD analysis (Figure 2a). The broad peak around 2θ = 20° shows that B. subtilis is amorphous (black curve) [13,15]. All the diffraction peaks of Fe3O4 MNPs (blue curve) could be readily indexed as face-centered cubic structured Fe3O4, which coincides well with the standard data (JCPDS card No. 19-0629) [17]. It is noted that the diffraction peaks corresponding to Fe3O4 are also presented in the XRD pattern of Fe3O4@B. subtilis SPMC (red curve), confirming the successful coating of Fe3O4 nanoparticles onto the surfaces of B. subtilis.  [18]. In Figure 2c, two main peaks with satellite peaks between 705.0 and 735.0 eV are resolved at 710.9 eV and 724.5 eV, which are assigned to Fe 2p1/2 and Fe 2p3/2, respectively, and are consistent with the standard Fe3O4 XPS spectrum [18,19]. In addition, the elemental concentration of C, N, O and Fe in atomic % calculated according to relative sensitivity factors (RSFs) and spectra intensities were 69.02%, 4.0%, 22.87%, and 4.13%, respectively.
The magnetically controllable aggregation behavior of Fe3O4@B. subtilis was investigated by VSM study. Figure 2d shows the magnetic hysteresis loops of Fe3O4 and Fe3O4@B. subtilis SPMC with an extra magnetic field of 20,000 Oe at 300 K. The negligible coercivity (Hc) of hysteresis loop (71.1 Oe) and consequently no remanence (Mr, 1.23 emu/g) indicate the superparamagnetic nature of the Fe3O4@B. subtilis SPMC [20], exhibiting excellent redispersion stability (Figure 2d, inset). The saturation magnetization of the Fe3O4@B. subtilis SPMC (16.8 emu g −1 ) is lower than the pristine Fe3O4 because of the non-magnetic properties of B. subtilis. In their homogeneous dispersion, Fe3O4@B. subtilis SPMC show quick movement under the extra magnetic field and re-disperse quickly with a  [18]. In Figure 2c, two main peaks with satellite peaks between 705.0 and 735.0 eV are resolved at 710.9 eV and 724.5 eV, which are assigned to Fe 2p 1/2 and Fe 2p 3/2 , respectively, and are consistent with the standard Fe 3 O 4 XPS spectrum [18,19]. In addition, the elemental concentration of C, N, O and Fe in atomic % calculated according to relative sensitivity factors (RSFs) and spectra intensities were 69.02%, 4.0%, 22.87%, and 4.13%, respectively.
The magnetically controllable aggregation behavior of Fe 3 O 4 @B. subtilis was investigated by VSM study. Figure 2d shows the magnetic hysteresis loops of Fe 3 O 4 and Fe 3 O 4 @B. subtilis SPMC with an extra magnetic field of 20,000 Oe at 300 K. The negligible coercivity (Hc) of hysteresis loop (71.1 Oe) and consequently no remanence (Mr, 1.23 emu/g) indicate the superparamagnetic nature of the Fe 3 O 4 @B. subtilis SPMC [20], exhibiting excellent redispersion stability (Figure 2d, inset). The saturation magnetization of the Fe 3 O 4 @B. subtilis SPMC (16.8 emu g −1 ) is lower than the pristine Fe 3 O 4 because of the non-magnetic properties of B. subtilis. In their homogeneous dispersion, Fe 3 O 4 @B. subtilis SPMC show quick movement under the extra magnetic field and re-disperse quickly with a slight shake after removing the magnetic field, suggesting the prepared Fe 3 O 4 @B. subtilis SPMC presents excellent magnetic responsivity and redispersibility [21].

Fenton-Like Catalytic Degradation of Doxycycline
To test the catalytic performance of the as-prepared Fe 3 O 4 @B. subtilis SPMC, Fenton-like reactions were conducted for DC degradation. Comparable dosages of Fe 3 O 4 @B. subtilis, B. subtilis and Fe 3 O 4 were used. Direct oxidation (blank) by H 2 O 2 without any particles was also performed as a reference. Solution pH values were not controlled during the degradation process because they decreased very slightly (∆pH < 0.5).
As depicted in Figure 3a, direct oxidation of DC molecules (blank) by H 2 O 2 is negligible. B. subtilis shows very low removal rate for DC within 30 min. When the Fe 3 O 4 @B. subtilis SPMC were used as catalyst, DC was nearly degraded completely in 30 min, showing a superior high catalytic activity in the Fenton-like system. Contrastively, under the same reaction condition, only 68.5% of DC was degraded by the as-synthesized Fe 3 O 4 MNPs. The pseudo-first-order reaction was used to describe the kinetics of these catalytic reactions, and the rate constants k for DC degradation are calculated according to the regression curves of −ln(C/C 0 ) vs. time (t) (Figure 3b). The Fe 3 O 4 @B. subtilis SPMC shows a highest k value of 0.129 min −1 , followed by Fe 3 O 4 (0.0374 min −1 ) and B. subtilis (0.00222 min −1 ), indicating the faster catalytic degradation kinetics and stronger catalytic ability of Fe 3 O 4 @B. subtilis SPMC.

Fenton-Like Catalytic Degradation of Doxycycline
To test the catalytic performance of the as-prepared Fe3O4@B. subtilis SPMC, Fenton-like reactions were conducted for DC degradation. Comparable dosages of Fe3O4@B. subtilis, B. subtilis and Fe3O4 were used. Direct oxidation (blank) by H2O2 without any particles was also performed as a reference. Solution pH values were not controlled during the degradation process because they decreased very slightly (ΔpH < 0.5).
As depicted in Figure 3a, direct oxidation of DC molecules (blank) by H2O2 is negligible. B. subtilis shows very low removal rate for DC within 30 min. When the Fe3O4@B. subtilis SPMC were used as catalyst, DC was nearly degraded completely in 30 min, showing a superior high catalytic activity in the Fenton-like system. Contrastively, under the same reaction condition, only 68.5% of DC was degraded by the as-synthesized Fe3O4 MNPs. The pseudo-first-order reaction was used to describe the kinetics of these catalytic reactions, and the rate constants k for DC degradation are calculated according to the regression curves of −ln(C/C0) vs. time (t) (Figure 3b). The Fe3O4@B. subtilis SPMC shows a highest k value of 0.129 min −1 , followed by Fe3O4 (0.0374 min −1 ) and B. subtilis (0.00222 min −1 ), indicating the faster catalytic degradation kinetics and stronger catalytic ability of Fe3O4@B. subtilis SPMC. The enhanced catalytic activity of Fe3O4@B. subtilis SPMC compared to pristine B. subtilis and Fe3O4 MNPs can be attributed to the synergistic effect between adsorption by the B. subtilis bodies and Fenton-like oxidation by the Fe3O4 particles. The isoelectric point of Fe3O4@B. subtilis SPMC was determined to be 4.1 according to the zeta potentials curves (Supplementary Figure S4). Thus, the negative charge of Fe3O4@B. subtilis SPMC surfaces can absorb cationic form of DC via electrostatic interaction at neutral pH, resulting in a higher reactant concentration around the composite surfaces. In turn, the adsorption sites on the B. subtilis surface can be refreshed because of the decomposition of the DC molecules by Fe 2+ /Fe 3+ /H2O2 Fenton-like system.

Effect of H2O2 Dosage on Degradation of DC
The degradation efficiency of DC increased with increasing H2O2 dosage from 5.0 to 20.0 mmol, and then slightly decreased beyond 20.0 mmol (Figure 4a). The decreased DC removal under higher H2O2 dosage may be because that excessive H2O2 can induce OH• radicals scavenging effect (HO• + H2O2 → HOO• + H2O) [15,22]. It is noteworthy that the generation of another radical HOO•, whose oxidation ability is much lower than that of the HO• radicals, shows much less contribution to DC  Figure S4). Thus, the negative charge of Fe 3 O 4 @B. subtilis SPMC surfaces can absorb cationic form of DC via electrostatic interaction at neutral pH, resulting in a higher reactant concentration around the composite surfaces. In turn, the adsorption sites on the B. subtilis surface can be refreshed because of the decomposition of the DC molecules by Fe 2+ /Fe 3+ /H 2 O 2 Fenton-like system.

Effect of H 2 O 2 Dosage on Degradation of DC
The degradation efficiency of DC increased with increasing H 2 O 2 dosage from 5.0 to 20.0 mmol, and then slightly decreased beyond 20.0 mmol (Figure 4a) [15,22]. It is noteworthy that the generation of another radical HOO·, whose oxidation ability is much lower than that of the HO· radicals, shows much less contribution to DC degradation [23]. Therefore, the initial H 2 O 2 dosage was selected as 20.0 mmol to achieve the highest removal of DC in the present work.
Catalysts 2017, 7, 349 5 of 10 degradation [23]. Therefore, the initial H2O2 dosage was selected as 20.0 mmol to achieve the highest removal of DC in the present work.

Iron Ion Leaching
To exclude the possibility that the observed catalytic activity of Fe3O4@B. subtilis SPMC in Fenton-like system is caused by the leaching ions, similar batch reactors with solutions (100 mL) containing Fe3O4@B. subtilis SPMC (0.5 g L −1 ) at a neutral pH were mechanically stirred for 30 min at room temperature. Then, the Fe3O4@B. subtilis SPMC were removed by centrifugation to obtain leaching solution. 0.1 mL of the above supernatant was diluted with 4.9 mL HNO3 (5%) to analyze the dissolved ions with ICP-MS. The result shows a Fe ions concentration of 0.612 ppm, which correspond to about 0.68% of the total Fe content in Fe3O4@B. subtilis SPMC. The remaining part of the leaching solution was employed to the homogeneous Fenton-like degradation of DC, and was initiated by the addition of fresh H2O2. By adding 2.5 mg DC and 20 mmol H2O2, only 7.5% of DC was removed within 30 min, which was much less than the removal of 98.1% in the Fe3O4@B. subtilis SPMC-H2O2 Fenton-like system at neutral pH (Figure 4a). Since the homogeneous degradation is negligible, the DC degradation at neutral pH mainly originates from the heterogeneous Fe3O4@B. subtilis SPMC-H2O2 Fenton-like system instead of the leached ions.

Stability and Reusability
The reusability and stability of Fe3O4@B. subtilis SPMC was evaluated by successive batches of DC degradation. After each run, Fe3O4@B. subtilis SPMC was re-collected, rinsed with DI water, and tested again. Fe3O4@B. subtilis SPMC can be reused for at least five consecutive runs, and the reused ones nearly retained the catalytic activity of the fresh catalyst (Figure 4b). The reusability of Fe3O4 was not investigated in this work because of two reasons: (1) the removal efficiency of DC, used for evaluation of reusability of Fe3O4@B. subtilis in Figure 4b, can be attributed the synergistic effect of adsorption by B. subtilis support and degradation by generated radicals, which is more complicated than bare Fe3O4; (2) Fe3O4 MNPs may present higher removal efficiency of DC after six cycles than that of Fe3O4@B. subtilis because they may leach much more iron ions for involving the Fenton-like process. However, in turn, more leaching ions means worse stability. Therefore, few publications have reported on the comparison between the repeatability of Fe3O4 and Fe3O4@support.
The good chemical stability of Fe3O4@B. subtilis SPMC was further confirmed by conducting XRD and XPS measurements after reaction. No considerable changes in XRD patterns (Figure 5a) were observed after being reused for five cycles, indicating the stable chemical and crystalline structure of the Fe3O4@B. subtilis SPMC. XPS measurement (Figure 5b) demonstrates that the at % of C, N, O, and Fe are almost the same as those before reaction. These results indicate that Fe3O4@B.

Iron Ion Leaching
To exclude the possibility that the observed catalytic activity of Fe 3 O 4 @B. subtilis SPMC in Fenton-like system is caused by the leaching ions, similar batch reactors with solutions (100 mL) containing Fe 3 O 4 @B. subtilis SPMC (0.5 g L −1 ) at a neutral pH were mechanically stirred for 30 min at room temperature. Then, the Fe 3 O 4 @B. subtilis SPMC were removed by centrifugation to obtain leaching solution. 0.1 mL of the above supernatant was diluted with 4.9 mL HNO 3 (5%) to analyze the dissolved ions with ICP-MS. The result shows a Fe ions concentration of 0.612 ppm, which correspond to about 0.68% of the total Fe content in Fe 3 O 4 @B. subtilis SPMC. The remaining part of the leaching solution was employed to the homogeneous Fenton-like degradation of DC, and was initiated by the addition of fresh H 2 O 2 . By adding 2.5 mg DC and 20 mmol H 2 O 2 , only 7.5% of DC was removed within 30 min, which was much less than the removal of 98.1% in the Fe 3 O 4 @B. subtilis SPMC-H 2 O 2 Fenton-like system at neutral pH (Figure 4a). Since the homogeneous degradation is negligible, the DC degradation at neutral pH mainly originates from the heterogeneous Fe 3 O 4 @B. subtilis SPMC-H 2 O 2 Fenton-like system instead of the leached ions.

Stability and Reusability
The reusability and stability of Fe 3 O 4 @B. subtilis SPMC was evaluated by successive batches of DC degradation. After each run, Fe 3 O 4 @B. subtilis SPMC was re-collected, rinsed with DI water, and tested again. Fe 3 O 4 @B. subtilis SPMC can be reused for at least five consecutive runs, and the reused ones nearly retained the catalytic activity of the fresh catalyst (Figure 4b). The reusability of Fe 3 O 4 was not investigated in this work because of two reasons: (1) the removal efficiency of DC, used for evaluation of reusability of Fe 3 O 4 @B. subtilis in Figure 4b, can be attributed the synergistic effect of adsorption by B. subtilis support and degradation by generated radicals, which is more complicated than bare Fe 3 O 4 ; (2) Fe 3 O 4 MNPs may present higher removal efficiency of DC after six cycles than that of Fe 3 O 4 @B. subtilis because they may leach much more iron ions for involving the Fenton-like process. However, in turn, more leaching ions means worse stability. Therefore, few publications have reported on the comparison between the repeatability of Fe 3 O 4 and Fe 3 O 4 @support.
The good chemical stability of Fe 3 O 4 @B. subtilis SPMC was further confirmed by conducting XRD and XPS measurements after reaction. No considerable changes in XRD patterns (Figure 5a) were observed after being reused for five cycles, indicating the stable chemical and crystalline structure of the Fe 3 O 4 @B. subtilis SPMC. XPS measurement (Figure 5b) demonstrates that the at % of C, N, O, and Fe are almost the same as those before reaction. These results indicate that Fe 3 O 4 @B. subtilis SPMC are durable and can be reused without great loss of catalytic activity in a long term Fenton-like reaction system at a neutral pH.
Catalysts 2017, 7, 349 6 of 10 subtilis SPMC are durable and can be reused without great loss of catalytic activity in a long term Fenton-like reaction system at a neutral pH.

Degradation Mechanism
The dramatic Fenton-like catalytic activity of the as synthesized Fe3O4@B. subtilis SPMC motivated us to further investigate the mechanism of the Fenton-like degradation process. Although the Fenton-like reaction mechanism has not been fully understood, it is generally accepted that a series of reactive radical species (RRS)-such as HOO•, hydroxyl radical (OH•), superoxide radical (O2• − ), or singlet oxygen ( 1 O2)-are supposed to be involved in the Fenton-like process [24][25][26]. However, which one plays the most important role in the Fenton-like process is still unclear, and is quite different from various systems. Luo et al. suggested that OH• radicals make a major contribution in the Fenton-like degradation of organic dyes at weak acidic conditions [27]. Li and co-workers demonstrated that the singlet oxygen 1 O2 produced from HOO• and HO• directly participates in the degradation of organic pollutants [25]. Therefore, on the basis of all the information obtained above and observations in the literatures [28][29][30], we propose that the mechanism of the H2O2 activation by Fe3O4@B. subtilis SPMC under neutral condition may involve the initial formation of complex intermediates between ≡Fe II , ≡Fe III , and H2O2, being marked as H2O2≡Fe II Equation (1) and H2O2≡Fe III Equation (2), where ≡Fe II and ≡Fe III stands for Fe(II) and Fe(III) sites on the surface of Fe3O4@B. subtilis SPMC. The intermediate H2O2≡Fe Ⅲ can convert to ≡Fe II species and HO2• (Equation (3). The generated HO2• can further react with ≡Fe III to produce ≡Fe II species and O2 Equation (4), giving an explanation of those bubbles generated in the reaction process. All the formed ≡Fe II species Equations (3) and (4) and the initially generated ≡Fe II species Equation (1) activate H2O2 to generate hydroxyl radicals HO• Equation (5), which is a very strong oxidative radical to degrade DC molecules Equation (6).

Degradation Mechanism
The dramatic Fenton-like catalytic activity of the as synthesized Fe 3 O 4 @B. subtilis SPMC motivated us to further investigate the mechanism of the Fenton-like degradation process. Although the Fenton-like reaction mechanism has not been fully understood, it is generally accepted that a series of reactive radical species (RRS)-such as HOO·, hydroxyl radical (OH·), superoxide radical (O 2 · − ), or singlet oxygen ( 1 O 2 )-are supposed to be involved in the Fenton-like process [24][25][26]. However, which one plays the most important role in the Fenton-like process is still unclear, and is quite different from various systems. Luo et al. suggested that OH· radicals make a major contribution in the Fenton-like degradation of organic dyes at weak acidic conditions [27]. Li and co-workers demonstrated that the singlet oxygen 1 O 2 produced from HOO· and HO· directly participates in the degradation of organic pollutants [25]. Therefore, on the basis of all the information obtained above and observations in the literatures [28][29][30] (4), giving an explanation of those bubbles generated in the reaction process. All the formed ≡Fe II species Equations (3) and (4) and the initially generated ≡Fe II species Equation (1) activate H 2 O 2 to generate hydroxyl radicals HO· Equation (5), which is a very strong oxidative radical to degrade DC molecules Equation (6).
Based on our hypothesis, HO· radicals make a great contribution to DC degradation by Fe 3 O 4 @B. subtilis SPMC. Thus, the formation of HO· was confirmed by the PL spectra using terephthalic acid (TA) as a probe molecule (Supplementary Figure S5). Figure 5 shows the fluorescence spectral changes observed during Fenton-like process with Fe 3 O 4 @B. subtilis suspended in a basic TA solution. As can be seen, with reaction time prolongs the PL intensity of 2-hydroxy terephthalic acid around 428 nm gradually increased, demonstrating the production of the hydroxyl radicals. Therefore, we can conclude that the enhanced degradation performance of DC by Fe 3 O 4 @B. subtilis under neutral pH can be due to the synergistic effect between the excellent adsorption ability and hydroxyl radicals generated from Fenton-like activation of H 2 O 2 by Fe 3 O 4 @B. subtilis SPMC.

Materials
B. subtilis powder was purchased from Guangzhou lvhui biological Company. All chemical reagents including FeCl 3 ·6H 2 O, N 2 H 4 ·H 2 O, CH 2 O, H 2 SO 4 , and Doxycyline were purchased from Sigma-Aldrich (Sigma-Aldrich, St. Louis, MO, USA) and used directly without further purification. The chemical structure of doxycyline is presented in Figure 6. Absolute ethanol and double-distilled water were used through this work.
Based on our hypothesis, HO• radicals make a great contribution to DC degradation by Fe3O4@B. subtilis SPMC. Thus, the formation of HO• was confirmed by the PL spectra using terephthalic acid (TA) as a probe molecule (Supplementary Figure S5). Figure 5 shows the fluorescence spectral changes observed during Fenton-like process with Fe3O4@B. subtilis suspended in a basic TA solution. As can be seen, with reaction time prolongs the PL intensity of 2-hydroxy terephthalic acid around 428 nm gradually increased, demonstrating the production of the hydroxyl radicals. Therefore, we can conclude that the enhanced degradation performance of DC by Fe3O4@B. subtilis under neutral pH can be due to the synergistic effect between the excellent adsorption ability and hydroxyl radicals generated from Fenton-like activation of H2O2 by Fe3O4@B. subtilis SPMC.

Materials
B. subtilis powder was purchased from Guangzhou lvhui biological Company. All chemical reagents including FeCl3•6H2O, N2H4•H2O, CH2O, H2SO4, and Doxycyline were purchased from Sigma-Aldrich (Sigma-Aldrich, St. Louis, MO, USA) and used directly without further purification. The chemical structure of doxycyline is presented in Figure 6. Absolute ethanol and double-distilled water were used through this work.

Synthesis of Fe3O4 MNPs
Fe3O4 MNPs were synthesized through a hydrothermal method using FeCl3•6H2O as a single iron source [31]. Specifically, 1.20 g of FeCl3•6H2O was placed in a dry beaker, and then 5 mL of N2H4•H2O was added into the beaker dropwise with continuous stirring. After that, 2 mL of CH2O and 33 mL deionized water were added into the baker, stirred for 10 min, and then the mixture was transferred to a 50 mL Teflon-lined stainless-steel autoclave and heated at 120 °C for 5 h. After cooling down, the black precipitate was collected with a magnet and washed with water and absolute ethanol for three times and then dried at 80 °C for 10 h.

Preparation of Fe3O4@B. subtilis Superparamagnetic Composite (SPMC)
Fe3O4@B. subtilis SPMC were fabricated based on an electrostatic attraction method (Supplementary Figure S6). Typically, dried magnetic Fe3O4 MNPs and 1.0 g of B. subtilis dry cell powder were dispersed in 100 mL of distilled water separately and the pH values was adjusted to

Synthesis of Fe 3 O 4 MNPs
Fe 3 O 4 MNPs were synthesized through a hydrothermal method using FeCl 3 ·6H 2 O as a single iron source [31]. Specifically, 1.20 g of FeCl 3 ·6H 2 O was placed in a dry beaker, and then 5 mL of N 2 H 4 ·H 2 O was added into the beaker dropwise with continuous stirring. After that, 2 mL of CH 2 O and 33 mL deionized water were added into the baker, stirred for 10 min, and then the mixture was transferred to a 50 mL Teflon-lined stainless-steel autoclave and heated at 120 • C for 5 h. After cooling down, the black precipitate was collected with a magnet and washed with water and absolute ethanol for three times and then dried at 80 • C for 10 h.

Preparation of Fe 3 O 4 @B. subtilis Superparamagnetic Composite (SPMC)
Fe 3 O 4 @B. subtilis SPMC were fabricated based on an electrostatic attraction method (Supplementary Figure S6). Typically, dried magnetic Fe 3 O 4 MNPs and 1.0 g of B. subtilis dry cell powder were dispersed in 100 mL of distilled water separately and the pH values was adjusted to 5.0 by adding 1 M H 2 SO 4 . The isoelectric points of Fe 3 O 4 MNPs and B. subtilis were determined to be pH 2.2 and 6.6, respectively. Therefore, according to the zeta potential-pH curve (Supplementary Figure S4), at an experimental pH of 5.0, B. subtilis is negatively charged and Fe 3 O 4 MNP is positively charged. Next, the two suspensions were sonicated for 20 min to facilitate deaggregation and then centrifuged. After that, the Fe 3 O 4 MNPs and B. subtilis were re-dispersed in 100 mL distilled water separately and mixed under continuous mechanical stirring for 1 h at room temperature. The mixture was then left for 3.0 h at room temperature to form the Fe 3 O 4 @B. subtilis SPMC. The resulting particles were centrifuged, washed by ethanol and water for three times. The actual Fe content (18%, i.e., Fe 3 O 4 content of 25%) in Fe 3 O 4 @B. subtilis SPMC was detected using ICP-MS. For the ICP-MS measurements, 10 mg Fe 3 O 4 @B. subtilis SPMC were first dissolved with 5 mL 5% concentrated HNO 3 , and then diluted 50 times using HNO 3 (5%).

Characterizations
Scanning electron microscopy (SEM) images were recorded using a microscope (6300F, JEOL Ltd., Tokyo, Japan). The elemental composition was carried out by energy dispersive spectroscopy (EDX) analysis and energy dispersive X-ray (EDX) mapping. XRD patterns were acquired over a diffraction angle range (2θ) 5-80 • using an X'Pert X-ray diffraction spectrometer with a Cu Kα X-ray source (MiniFlex 600, Rigaku, Tokyo, Japan). XPS spectra were obtained on an electron spectrometer (ESCALab220i-XL, VG Scientific, Waltham, MA, USA) using 300 W Al-Ka radiation. Magnetization measurements at room temperature were obtained using a vibrating sample magnetometer (Lake Shore Cryotronics, Inc., Carson, CA, USA).

Catalytic Tests
The catalytic performance of Fe 3 O 4 @B. subtilis SPMC was tested by the degradation of DC in semi-batch operation mode at neutral pH 0.05 g catalysts were dispersed into 100 mL aqueous solution of DC (25 mg L −1 ) at room temperature. The suspensions were mechanically stirred in dark for 30 min to achieve the adsorption/desorption equilibrium. The DC concentrations after equilibration were measured and taken as the initial concentration (C 0 ). Then, the degradation reaction was initiated by the addition of H 2 O 2 under mechanical stirring conditions at room temperature. At regular time intervals, about 5 mL of solution were taken out and immediately centrifuged at 10,000 rpm for 3 min, and the supernatant was determined by using a UV-vis spectrophotometer. Each experiment was conducted in triplicate.

Detection of Radical Species
Terephthalic acid (TA) was used to detect OH·. Briefly, 0.1 g Fe 3 O 4 @B. subtilis SPMC was dispersed in a 100 mL of the TA (5 × 10 −4 mol L −1 ) aqueous with an addition of 20.0 mmol H 2 O 2 at room temperature and without any DC. The intensity of the PL signal at 428 nm was investigated on a fluorescence spectrophotometer (Shimadzu RF-5301PC, Tokyo, Japan) at an excitation wavelength of 315 nm.

Conclusions
In this study, novel superparamagnetic Fe 3 O 4 @B. subtilis Fenton-like catalyst was synthesized through a facile electrostatic attraction process. The obtained Fe 3 O 4 @B. subtilis exhibits significantly enhanced Fenton-like catalytic ability than pristine Fe 3 O 4 and excellent recycling ability, reusability, and stability. The detailed doxycyline degradation mechanism was explored by employing photoluminescence technology. The result clearly demonstrates that the synergistic effect of Bacillus subtilis adsorption and Fenton-like ability of Fe 3 O 4 dominates the enhancement of Fenton-like catalytic efficiency of Fe 3 O 4 @B. subtilis.
Supplementary Materials: The following are available online at www.mdpi.com/2073-4344/7/11/349/s1, Figure S1: Size histogram of B. subtilis (a,b) and Fe 3 O 4 @ B. subtilis (c,d), Figure S2: N2 adsorption-desorption isotherm and pore size distribution (inset) of (a) the bare B. subtilis and (b) Fe 3 O 4 MNP and (c) Fe 3 O 4 @B. subtilis, Figure S3: Selected zones of Fe 3 O 4 @B. subtilis SPMC (a) and corresponding X-ray mapping, for C (b), O (c), and Fe elements (d), Figure S4: Zeta potential of B. subtilis, Fe 3 O 4 and Fe 3 O 4 @B. subtilis suspensions as a function of pH, Figure S5: Formation of hydroxyl products as the result of reaction between terephthalic acid (TA) and OH·, Figure S6: Schematic illustration of the synthesis process of Fe 3 O 4 @B. subtilis.