Hierarchically-Structured TiO 2 / MnO 2 Hollow Spheres Exhibiting the Complete Mineralization of Phenol

Although TiO2 or MnO2-based materials have been widely used for the degradation of phenolic compounds, complete mineralization is still a challenge, especially for TiO2-based materials. Here, we devise a hierarchically-structured TiO2/MnO2 (HTM) hollow sphere, in which hollow TiO2 acts as a skeleton for the deposition of MnO2 in order to prevent the aggregation of MnO2 nanoparticles and to maintain its hollow structure. During the oxidation reaction, the as-synthesized HTM can fully exert their respective advantages of the TiO2 and MnO2 species to realize the first stage of the rapid oxidation degradation of phenol and the second stage of the complete photo-mineralization of residual phenol and its intermediates, which efficiently overcomes the incomplete mineralization of phenolic compounds. The degradation mechanism and pathway of phenol are also proposed according to the analysis of Mass Spectrometry (MS). Therefore, this work provides a new insight for exploring hierarchically-structured materials with two or more species.


Introduction
As one of the persistent organic pollutants in surface water sources, phenol and its derivatives are widely used as raw materials in many chemical, petrochemical, and pharmaceutical fields [1,2].Clearly, the continuous addition of phenolic compounds to the environment has led to a significantly enhanced toxic contamination because they are hard to be decomposed even at low concentrations [3,4].Therefore, effective strategies for completely removing phenolic compounds from wastewater are urgently required.
Membrane-based separation methods are widely used for wastewater treatment because of the good performance of the reverse osmosis and nanofiltration process [5,6].However, the method often shows a low retention of many organic micropollutants and suffers from severe fouling problems.By contrast, considerable efforts have been focused on the design of TiO 2 -based photocatalysts for the removal of phenolic compounds [7,8] due to their powerful oxidizing ability, chemical stability, and low cost [9][10][11].For example, Makrigianni et al. [12] confirmed that pure TiO 2 could only remove 65% of phenol within five hours under simulated natural sunlight irradiation.Wang et al. [13] prepared a composite hollow TiO 2 photocatalyst and found that it degraded 73% of phenol under sunlight irradiation after 4 h.Zheng et al. [14] synthesized electrochromic titania nanotube arrays and removed 78.2% of phenol after 3 h under simulated sunlight irradiation.Unfortunately, an efficient method for the complete degradation of phenolic compounds is extremely difficult for TiO 2 -based photocatalysts even under simulated sunlight irradiation [15,16].Accordingly, the design of novel and efficient degradation systems that can mineralize phenol completely is highly desirable.
Based on the above considerations, we devise a hierarchically-structured TiO 2 /MnO 2 (HTM) for the first time, in which the outer MnO 2 can achieve the fast and efficient oxidation of phenol by activating PMS, while the inner TiO 2 not only acts as a skeleton to sustain MnO 2 , but also further mineralizes a small amount of undegraded phenol and intermediates under simulated sunlight irradiation.Composites comprised of two metal oxides possess improved physicochemical properties compared to the use of pure oxides [26].Moreover, the hollow/porous nature of the HTM boosts the diffusion of reactants by lowering diffusion resistance [27].More importantly, the HTM can fully exert their respective advantages in the TiO 2 and MnO 2 species, realizing the first stage of chemical oxidation degradation and the second stage of complete photo-mineralization. Therefore, the as-synthesized HTM manifests the complete mineralization of phenol, efficiently overcoming the respective disadvantages of TiO 2 and MnO 2 for the incomplete mineralization of phenolic compounds in wastewater.This work provides a new insight for constructing hierarchically-structured materials with two or more species.

Results and Discussion
2.1.The Preparation and Structure of TiO 2 /MnO 2 (HTM) Spheres HTM hollow spheres were prepared via several steps: Firstly, a CPS@TiO 2 composite was synthesized by referring our previous works [9,10,28].Secondly, TiO 2 hollow spheres could be achieved after calcination at 450 • C. Thirdly, the resulting TiO 2 hollow spheres were used as templates for the deposition of manganese nitrate due to the electrostatic interactions.Finally, the TiO 2 /MnO 2 (HTM) was formed through calcination at 450 • C.
Figure 1 indicates the TEM and SEM images of hollow TiO 2 and HTM spheres.Compared with TiO 2 hollow spheres (average size: 510 ± 10 nm) (Figure 1a), each HTM hollow sphere has a thicker shell (average size: 523 ± 15 nm), suggesting a successful deposition of manganese dioxide.More importantly; some crystalline floccules located on the outer surface of the HTM can be clearly observed (Figure 1b), which are further confirmed by the SEM image of the HTM (Figure 1d).Unlike the smooth surface of hollow TiO 2 spheres (Figure 1c), the surface of the HTM hollow spheres become rougher due to the existence of many small crystalline MnO 2 nanoparticles (Figure 1d), further confirming the successful loading of manganese dioxide.The hollow structure of TiO 2 /MnO 2 is ascertained by a STEM mapping measurement.As shown in Figure 2, energy dispersive x-ray (EDX) spectral region exhibits clear signals of O, Mn and Ti elements in the HTM sample, indicating the presence of MnO 2 and TiO 2 .Figure 2d indicates that the Ti signal keeps the shape of the sphere, which fairly agrees with the results measured by the SEM and TEM.Interestingly, the Mn signal (Figure 2e) is mostly dispersed throughout the HTM spheres, forming a sphere similar to the Ti signal.EDX mapping results confirm that MnO 2 has been homogeneously anchored on the surface of the hollow TiO 2 spheres.Figure 3 shows that the HTM exhibits a representative type IV curve with a clear hysteresis loop [29,30], which is clearly different from the porous/hollow TiO 2 spheres [31] (Figure 3a) with a high surface area (~90.2283m 2 /g).By contrast, the capillary condensation of the HTM in N 2 adsorption and desorption is completely ascribed to the formation of MnO 2 particles on the surface of the porous/hollow TiO 2 , leading to a decreased surface area (53.7827 m 2 /g).In addition, thermogravimetric measurement of the HTM shows that only a 5.69% mass loss is observed between 100 and 450 • C (Figure S1), indicating that the HTM sample has a good structural stability and a low carbon amount in the HTM spheres.This is further confirmed by Raman spectra because no significant peaks are found to be attributed to the D-bands and G-bands of carbon materials (Figure S2) [32].All these results strongly confirm that the hierarchical TiO 2 /MnO 2 hollow spheres have been successfully prepared.

Crystal Structure and Bonding Environment of HTM Spheres
Figure 4 shows that the identified peaks of the HTM can be clearly attributed to anatase TiO 2 (JCPDS: 21-1272) [33,34], while four distinctive MnO 2 reflection peaks at 37.12 • (100), 42.4 • (101), 56.03 • (102), 66.76 • (110) are in line with the standard spectrum for MnO 2 (JCPDS: 30-0820) [35,36].In addition, some weak peaks from MnTiO 3 are seen.According to the Scherrer equation (D = kλ β cos θ ), the crystallite size of TiO 2 is ~3.7 nm resulting from XRD, while the crystallite size of MnO 2 is ~11.96nm.The well-defined diffraction rings in the selected electron area diffraction (SEAD) can be indexed as (100), ( 101), (102), (110) planes of MnO 2 (Figure S3), which is consistent with the XRD results.The XRD results show the formation of manganese dioxide and a small amount of MnO 2 doping in the HTM spheres [37].XPS analysis is used to obtain the chemical state and bonding environment of HTM, helping us understand basic information about the interaction between TiO 2 and MnO 2 .As shown in Figure 5a, the elements of C, Ti, O and Mn can be evidently seen according to the binding energies of C1s, Ti2p, O1s and Mn2p, which fairly agrees with the measurement in the EDX spectral region (Figure 2a).In the case of the Ti2p, Figure 5b shows that the peak at 464.4 eV is attributed to Ti2p 1/2 , and the other peak at 458.75 eV is indexed to Ti2p 3/2 .The positive shift (~0.2 eV) of Ti2p 3/2 and (~0.17 eV) of Ti2p 1/2 can be observed for the HTM compared with the corresponding binding energies of hollow TiO 2 spheres (HT) (458.55 eV and 464.23 eV) [31].The increases in the binding energies may be ascribed to the combined actions between MnO 2 , carbon and host TiO 2 .In detail, the Ti2p peak of HTM can be resolved into two Gaussian peaks, as demonstrated in Figure 5c.The peaks at 464.84 and 459.05 eV can be indexed to the Ti2p 1/2 and Ti2p 3/2 core levels of Ti 4+ , respectively.While two peaks located at 463.62 and 458.05 eV are ascribed to the Ti2p 1/2 and Ti2p 3/2 peaks of Ti 3+ , respectively, revealing the existence of oxygen vacancies [38].Figure 5d shows that the C1s spectrum of the HTM sphere can be fitted to three different peaks of 284.77, 286.1 and 288.73 eV, indicating that there are three carbon chemical environments in the HTM spheres.The peak at 284.77 eV is assigned to the existence of vestigial carbon, which serves as an adsorbent for organic pollutants and as a photo-sensitizer to drive the faster degradation of pollutants [9].As demonstrated in Figure 5e and Table S1, two distinct peaks located at 642 and 653.8 eV can be ascribed to Mn2p 3/2 and Mn2p 1/2 , respectively [39].The energy separation between Mn2p 3/2 and Mn2p 1/2 is about 11.8 eV, which is a typical value for MnO 2 [39,40].Figure 5f shows the XPS spectrum of O1s, there is a peak at 529.97 eV related to the bulk oxygen bound of TiO 2 .In addition, the peak at 532.1 eV is responsible for hydroxyl groups adsorbed onto the surface of TiO 2 [9,10].

Chemical Oxidation Activity of HTM
Under the existence of oxone (PMS), the chemical oxidation performance of as-synthesized HTM was investigated by degrading phenol.The commercial active MnO 2 (CAM) and hollow TiO 2 spheres (HT) were used as reference materials.Figure 6 suggests that the removal rate of phenol is negligible after the addition of PMS, suggesting that phenol is extremely hard to be degraded under only the existence of an oxidant because there is no thermal activation of Oxone for the production of sulfate radicals [2,24].Similarly, hollow TiO 2 spheres (HT) with a high surface area (~90.2283m 2 /g) also cannot degrade phenol by adding PMS without light irradiation.For commercial active MnO 2 (CAM), the degradation rate of phenol is as low as 18% under PMS.By contrast, the removal rate of phenol is as high as 92% under the same oxidation conditions for the as-synthesized HTM hollow spheres because its large surface-to-volume ratio (HTM: 53.7827 m 2 /g; MnO 2 (Figure S4): 25.7805 m 2 /g) and hollow/porous characteristics facilitate the access of reactant phenol molecules [30,37].Unfortunately, even if the PMS is added again after 60 min (Figure S5), the removal of phenol does not increase accordingly.In addition, a longer oxidation times does not result in a higher degradation rate of phenol for the as-synthesized HTM, indicating the incomplete removal of phenol, as reported in many previous publications [2,19,23,25].Therefore, further degradation is very necessary for the complete removal of phenol.

Photo-Degradation Activity of HTM
As afore-mentioned, the complete removal of phenol is very difficult in using only the chemical oxidation ability of MnO 2 in the HTM sample.Gratifyingly, the as-synthesized HTM also holds strong photocatalytic activity due to the modification of the electronic energy band structure of TiO 2 through the coating of MnO 2 [30].Figure S6 shows that HTM exhibits strong light absorption in the region of 200-750 nm due to MnO 2 coverage, narrowing the band gap (~2.86 eV) of TiO 2 , as shown in Figure S6b [30].After the chemical oxidation (60 min), the second stage of the photo-degradation experiment was further carried out under simulated solar light radiation.Figure 7 indicates that the removal rate of phenol further increases with the increase in irradiation time, suggesting that HTM exhibits a photo-degradation activity for phenol.Excitedly, the complete degradation of phenol can be achieved after 180 min irradiation, effectively overcoming the incomplete degradation of phenol for hollow TiO 2 and MnO 2 (Figure S7).

Total Organic Carbon (TOC) Measurement
Total organic carbon (TOC) measurement has been widely used to assess the mineralization of organic pollutants [2,19].TOC removal profiles for phenol degradation on HTM hollow spheres are presented in Figure 8.As seen, the as-synthesized HTM with or without simulated sunlight irradiation exhibits TOC reduction of phenol under the presence of oxidant PMS in a short time.Only under PMS, the HTM provides excellent phenol mineralization capabilities and more than 91% of TOC (Table S2) is removed and transfers into inorganic carbon forms within 60 min.However, 100% of TOC reduction is not able to be achieved even when the oxidation time is prolonged to 180 min, indicating the incomplete mineralization of phenol for HTM.Excitedly, an enhanced TOC reduction of phenol is observed after simulated solar light irradiation, and 100% of TOC reduction is obtained when the irradiation time is prolonged to 180 min, strongly suggesting that a small quantity of residual phenol and intermediates can be completely removed by the photo-degradation stage.The order of the TOC removal fairly agrees with the order of the two stages of degradation reaction rates on HTM hollow spheres.

Mineralization Mechanism and Pathway of Phenol
To expound the complete mineralization of phenol for HTM hollow spheres, a probable degradation-mechanism is proposed based on the hierarchically-structured TiO 2 /MnO 2 , as shown in Figure 9a.In the first stage, a MnO 2 shell in the HTM sample can activate PMS to generate lots of sulfate radicals (SO 4− •) [19,24], subsequently, the generated sulfate radicals rapidly react with phenol in a short time (60 min), leading to the fast degradation of phenol (91%).However, the higher removal of phenol cannot be obtained even when the oxidation time is further prolonged to 180 min, which is mainly ascribed to the low concentration of phenol after oxidation degradation and extremely low adsorption amount of phenol (~3%) obtained by referring to our previous method [10], significantly decreasing the reaction with phenol and resulting in a negligibly enhanced removal of phenol.This is why the incomplete mineralization of phenolic compounds is reported in many previous publications [2,19,23,25].In the second stage, inner TiO 2 in the HTM sample is easy to activate under simulated solar light irradiation and generates numerous electron/hole pairs [9].The dissolved O 2 easily traps photo-generated electrons [41], forming superoxide radical (•O 2− ), in which •O 2− is a crucial active specie for mineralizing residual phenol and its intermediates because dissolved O 2 can participate in the photo-degradation process through the porous channels and reaction medium (water) [42], resulting in the complete mineralization of phenol.
To obtain the possible degradation pathway of phenol, the transformation intermediates were detected by MS during the degradation process, as shown in Figure 9b.The product with m/z = 94.09 is attributed to the molecular weight of phenol.Two possible degradation pathways of phenol are proposed in this study.After sulfate radicals and hydroxyl radicals are generated by activating PMS under MnO 2 , phenol is attacked by the radicals, generating phenol radicals.Subsequently, the phenol radical is further attacked to produce p-hydroxybenzoic acid with m/z = 139.96or a small amount of pyrocatechol with m/z = 109.05.Finally, these intermediates undergo further degradation under radicals or light irradiation to produce CO 2 and H 2 O.

Recyclability of HTM Spheres
Recyclability is an important factor for the practical application of HTM.After each degradation test, the HTM sample was centrifuged, rinsed, dried, and reused for the next run.Figure 10a illustrates the performance of HTM towards the degradation of phenol.As seen, the HTM can still be used with the order of the two stages of phenol removal even after three recycling runs, implying that the HTM possess a good stability (Figure S8) for the mineralization of phenol in wastewater.Clearly, the enhanced catalytic performance of the HTM has been associated with the changes in their structural and optical properties [26].To further understand the action of TiO 2 and MnO 2 , we prepared MnO 2 without a hollow TiO 2 template by using the same method [43].Accordingly, MnO 2 and hollow TiO 2 spheres (HT) were used for the degradation of phenol.Figure 10b indicates that MnO 2 and HT both exhibit a lower removal rate of phenol than the HTM.Particularly, MnO 2 does not enhance the degradation rate of phenol even under simulated solar light irradiation; by contrast, an enhanced removal of phenol is observed by HT, yet below the mineralization of phenol for as-synthesized HTM.Moreover, poor recyclability is found for MnO 2 and HT.The results strongly confirm that the as-synthesized HTM exhibits a complete removal of phenol and good recyclability, substantially broadening its practical potential.

Preparation of TiO 2 Spheres
Hollow TiO 2 spheres (HT) were synthesized via the following steps: firstly, cationic polystyrene spheres (CPS) were developed by referring to our previous works [10,28,44].Secondly, the CPS spheres were dispersed in absolute ethanol at 0 • C and then TBT ethanol solution was slowly added into the reaction system.After that, the reaction medium was further maintained for 24 h in order that that TBT hydrolysate be adsorbed on the surface of CPS via electronic attraction.Thirdly, a mixture of deionized water and ethanol was dropped into the system to form CPS@TiO 2 particles.Finally, the resulting CPS@TiO 2 was filtered, washed repeatedly and calcined at 450 • C to prepare hollow TiO 2 spheres.

Preparation of TiO 2 /MnO 2 (HTM) Spheres
The as-synthesized hollow TiO 2 spheres (1.0 g) were further dispersed in ethanol solution (50 mL).Manganese nitrate (2.0 g) ethanol solution (20 mL) was slowly added into the reaction system via a syringe.The system was kept at 0 • C for 12 h.Subsequently, the solution was evaporated at 70 • C by referring to previous publication [43].Finally, TiO 2 /MnO 2 (HTM) spheres were obtained after calcination at 450 • C for 6 h.

Characterization
The internal and external morphology of TiO 2 and HTM spheres were measured by a field emission scanning electron microscope (JEOL, Tokyo, Japan, JSM-7001F, 10.00 KV) and a high-resolution transmission electron microscope (JEOL JEM-2100, 200 KV).Specific surface area and pore size distribution of the TiO 2 and HTM spheres were determined by Tristar 3020II (Micromeretics, Norcross, GA, USA).The crystal structure of the HTM spheres were investigated by X-ray diffractometer (Bruker D8 Advance) using Cu-Kα radiation at 40 kV and 40 mA (20 • -80 • ).In addition, the bond energy of the HTM spheres were measured by X-ray photoelectron spectroscopy (Thermo ESCALAB 250Xi).UV-vis spectroscopy was tested by UV2600 (Hitach, Tokyo, Japan).Total organic carbon (TOC) was determined using a Shimadzu TOC-L cph analyzer for selected samples.

Figure 1 .
Figure 1.TEM and SEM images of hollow TiO 2 spheres (a,c) and HTM spheres (b,d).

Figure 4 .
Figure 4. XRD patterns of TiO 2 /MnO 2 and standard XRD patterns of anatase and manganese dioxide.

Figure 6 .
Figure 6.Chemical oxidation activity of different materials for phenol removal.

Figure 9 .
Figure 9. Proposed mechanism (a) and pathway (b) of phenol degradation for as-synthesized HTM hollow spheres.

Figure 10 .
Figure 10.Recyclability of HTM spheres (a) and HT and MnO 2 (b) for phenol degradation.