Integral structured Co–Mn composite oxides grown on interconnected Ni foam for catalytic toluene oxidation

Considering the three-dimensional ordered network of Ni foam-supported catalysts and the toxicity effects of volatile organic compounds (VOCs), the design of proper active materials for the highly efficient elimination of VOCs is of vital importance in the environmental field. In this study, a series of Co–Mn composite oxides with different Co/Mn molar ratios grown on interconnected Ni foam are prepared as monolithic catalysts for total toluene oxidation, in which Co1.5Mn1.5O4 with a molar ratio of 1 : 1 achieves the highest catalytic activity with complete toluene oxidation at 270 °C. The Co–Mn monolithic catalysts are characterized by XRD, SEM, TEM, H2-TPR and XPS. It is observed that a moderate ratio of Mn/Co plays significant effects on the textural properties and catalytic activities. From the XPS and H2-TPR characterization results, the obtained Co1.5Mn1.5O4 (Co/Mn = 1/1) favors the excellent low-temperature reducibility, high concentration of surface Mn3+ and Co3+ species, and rich surface oxygen vacancies, resulting in superior oxidation performance due to the formation of a solid solution between the Co and Mn species. It is deduced that the existence of the synergistic effect between Co and Mn species results in a redox reaction: Co3+–Mn3+ ↔ Co2+–Mn4+, and enhances the catalytic activity for total toluene oxidation.


Introduction
With increasing energy consumption and emission of air pollutants from industrial processes and transport vehicles, the large amounts of volatile organic compounds (VOCs) in the atmosphere are harmful to human health and the living environment. [1][2][3][4] Catalytic oxidation is one of the low-cost and most promising technologies for VOC degradation in recent decades. [5][6][7][8] In addition, the commercial catalysts are designed by dispersing the active components into honeycomb ceramics (2MgO$2Al 2 O 3 $5SiO 2 ). Unfortunately, this method always causes inhomogeneous active components in impregnation processes, which results in lower catalytic efficiency and limited application. 9 Meanwhile, some studies have been performed on the development of metallic substrates or modular catalysts in the environmental eld. [9][10][11][12][13][14][15] Among the metallic substrates, Ni foam has special advantages such as a low cost, high porosity, rich accessible electroactive sites, high thermal conductivity, and mass transfer ability, 13,15,16 so may be considered as a better alternative to honeycomb ceramics.
Various transition metal oxides as active components in reaction processes have been extensively studied to replace noble metals over the past several years, because they have low cost, unique structural morphology, adequate catalytic activity and high thermal stability. 3,17-21 Co 3 O 4 , a transition metal oxide, has been proven to excellent catalytic activity in numerous reactions duo to its superior physical-chemical properties. [22][23][24][25][26][27] Additionally, extensive efforts have revealed that the synergistic effect of Co species and other transition metal oxides has dramatically enhanced the redox properties and catalytic activities due to the formation of a solid solution, compared with single oxides. 11,[28][29][30][31] For example, Jiang et al. 30 reported the preparation of Mn-Co-O x nanocubes with different Mn/Co molar ratios derived from metal-organic frameworks. It is shown that hierarchical Mn-Co-O x mixed metal oxides exhibited the better redox properties, more exposed active sites and superior oxidation performances of total toluene oxidation. Chen et al. 11 synthesized mesoporous CoMnAl mixed metal oxides from layered double hydroxide (LDH) for total benzene oxidation. Results showed that CoMn 2 AlO-550 displayed rich oxygen vacancies and optimal catalytic activity due to the formation of a solid solution of cobalt-manganese oxides. Therefore, one effective strategy on introducing Mn species into Co 3 O 4 phase to develop multi-element mixed oxides, can be successfully used for removal of VOCs and improved redox properties.
Herein, a series of binary Co-Mn oxides with different Co/ Mn molar ratios embedded in interconnected Ni foam were well prepared via a simple hydrothermal reaction, and their catalytic performances were investigated in the total toluene oxidation (a model reaction). Furthermore, integral structured Co-Mn composite oxides grown on interconnected Ni foam were characterized by numerous techniques, such as XRD, SEM, TEM and XPS, to understand the correlation between physicalchemical properties and reactivity of Co-Mn composite oxides. This study is to provide guidelines for the rational fabrication of integral structured Co-based composite oxides grown on interconnected Ni foam for effectively catalytic toluene oxidation.

Preparation of catalysts
An aqueous salt solution (with Co/Mn molar ratios of 3/0, 2/1, 1/ 1, 1/2 and 0/3, Co 2+ + Mn 2+ ¼ 3 mmol) was prepared by dissolving Co(NO 3 ) 2 $6H 2 O and Mn(NO 3 ) 2 into 40 mL of deionized water. The 12 mmol of solid urea was then added to the homogeneous salt solution. The cleaned Ni foam (4 cm Â 6 cm) was immersed in the precursor solution, and were then transferred into in a 50.0 mL Teon-lined autoclave at 95 C for 12 h in an electric oven. Aer cooling down to indoor temperature, the covered Ni foams with array precursors were washed several times with deionized water and ethanol. Finally, the covered Ni foams were heated at 400 C in air for 2 h to form the composite oxide phase. These calcined Ni foams were designated as Co 3Àx Mn x O 4 (x ¼ 0, 1, 1.5, 2 and 3), respectively.
The pore size distribution, pore volume and the Brunauer-Emmett-Teller (BET) surface areas of as-prepared structured catalysts were measured by using a Micromeritics ASAP2020 at À196 C. Before the tests, all structured catalysts were degassed at 120 C for 2.5 h.
The surface morphology and microstructure of as-prepared structured catalysts were carried out by using scanning electron microscopy (SEM, SU-8020) and transmission electron microscope (TEM, JEOL 2100F), respectively.
The H 2 temperature programmed reduction (H 2 -TPR) measurements were carried out on an Automated Catalyst Characterization System (Autochem 2920, MICROMERITICS). Prior to H 2 -TPR, the structured catalyst (1 cm Â 3 cm) were heated under a gas ow of 5% O 2 /He (25 mL min À1 ) from indoor temperature to 300 C. Aer cooling to room temperature, the structured catalyst was reduced under a gas ow of 10% H 2 /Ar (30 mL min À1 ) with at a heating rate of 10 C min À1 .
X-ray photoelectron spectroscopy (XPS) measurements were recorded by using an XLESCALAB 250Xi electron spectrometer from VG Scientic with monochromatic Al Ka (1486.6 eV) radiation, and the peak positions were calibrated by the C 1s peaks at 284.6 eV.

Catalytic performance test
The catalytic activities of total toluene oxidation over the Co 3Àx Mn x O 4 composite oxides were performed in a continuous-ow quartz tube (i.d. ¼ 10 mm) using about 0.24 g covered Ni foams (2 cm Â 4 cm), the covered Ni foams were bent into the reaction tube. The test was carried out in the temperature range of 180-300 C under 1000 ppm toluene balanced with air at a total ow rate of 120 mL min À1 . Thus, a weight hourly space velocity (WHSV) of 30 000 mL g À1 h À1 or a gas hour space velocity (GHSV) of 12 000 h À1 was applied for the whole experiment. The toluene conversion was persistently measured, and was reached typically at the nal temperature for 1 h in each testing temperature. The concentrations of toluene and products (CO or CO 2 ) in the effluent gas was analyzed by using an online GC-2014 with two ame ionization detector (FID). The catalytic activities over the Co 3Àx Mn x O 4 composite oxides were calculated as following equation: where h, C toluene,in and C toluene,out are the total VOCs conversion, toluene in the inlet and outlet gas, respectively. The apparent activation energy (E a ) values of were calculated by the equations:

Results and discussion
As shown in XRD patterns of as-synthesized Co-Mn composite oxides, the strong peaks at 2q ¼ 44. 8  . According to the XRD results, it could be found that the Co 1.5 Mn 1.5 O 4 catalyst with the molar ratio of Co/Mn ¼ 1/1 has the lower intensity of the diffraction peak, inferring that there is a low crystallinity which will result in abundant crystal defects (Fig. 1). The nitrogen adsorption/desorption isotherms and pore size distributions of as-prepared Co-Mn composite oxides grown on interconnected Ni foam are tested to further research the porous structure, as shown in Fig. 2. With the increased Mn species doped into Co 3 O 4 , the specic surface area and pore volume gradually decreases, as summarized in Table 1. There is a type IV sorption isotherms with a type H2 hysteresis loop for each of the samples (Fig. 2a), indicating the presence of homogeneous mesopores. The average pore sizes of as-prepared Co-Mn samples keep in a narrow size distribution (6.8-8.8 nm), as described in Fig. 2b.
Aer hydrothermal reactions, the surface of Ni foams is orderly covered via uniform coverage of vertical Co/Mn oxide arrays, as demonstrated in Fig. 3k. The morphologies of asprepared Co-Mn arrays in Ni foam have changed with the increased Mn species doping into Co 3 O 4 arrays. The Co 3 O 4 sample mainly exhibits a series of intertwined and hexagonal nanosheets with the epitaxial nanowires in a parallel fashion, as shown in Fig. 3a and b. When the atomic ratio of Co : Mn is about 2 : 1, the nanowires grow signicantly at the edge of the nanosheets in Fig. 3c Fig. 3i and j, it could be observed that many hexagonal nanosheets with a diameter of 1-2 mm aggregated on the margins of sample, and the morphology of Mn 3 O 4 sample is not completely uniform. In addition, EDX mapping ( Fig. 3l-o) shows that Co, Ni, Mn elements are uniformly distributed on the surface of Ni foam.
The micro-structure of as-prepared Co-Mn composite oxides was investigated via TEM analysis. As shown in Fig. 4a and b, a series of nanowires are uniformly distributed on the extremely rough surface of nanosheets with tiny pores inside, which is consistent with SEM results. A HRTEM image of Co 3 O 4 is shown in Fig. 4b, the high-resolution lattice fringes calculated to be 0.445 nm could be indexed to the (111) lattice plane of Co 3 O 4 phase. TEM image in Fig. 4c further reveals that the morphology of Co 1.5 Mn 1.5 O 4 sample is composed of a large number of nanowires with a diameter of 50-70 nm, the corresponding HRTEM image in Fig. 4d shows that the lattice fringes with an interplanar spacing of 0.286 nm is assigned to the (220) lattice plane of Co 3 O 4 phase. No lattice streaks of manganese oxides are observed, indicating the formation of a solid solution between Co and Mn species. The TEM image of Mn 3 O 4 sample in Fig. 4e exhibits hexagonal nanosheets with a diameter of 2 mm, some lattice fringes belong to the (101) lattice planes of manganese oxides (lattice fringes ¼ 0.48 nm) can be observed. The formation of a solid solution is favorable for low temperature reduction, electron transfer, abundant surface oxygen vacancy and oxidation-reduction reaction, which will be further conrmed via other characterization analysis.
The reducibility of as-prepared Co-Mn composite oxides grown on interconnected Ni foam was carried out by the H 2 -TPR test. The H 2 -TPR curves of Co-Mn composite oxides are showed in Fig. 5, there are mainly two to three reduction peaks over the Co-Mn catalysts. The reduction step of Co 3 O 4 phase is Co 3+ / Co 2+ / Co 0 , and the reduction step of manganese oxides is Mn 4+ / Mn 3+ / Mn 2+ . 2,33 For Co 3 O 4 sample with three reduction peaks, the one peak in the low temperature range of 200-300 C is associated with the reduction of surface Co 3+ into Co 2+ , the overlapping peaks in the low temperature range of 300-400 C is attributed to the further reduction of bulk Co 3+ into Co 2+ and metallic cobalt. 19,34 The TPR proles of Mn 3 O 4 sample exhibits two separated reduction peaks at 303 C and 354 C, which are assigned to the following two reduction processes: Mn 4+ / Mn 3+ and Mn 3+ / Mn 2+ , respectively. In addition, compared to single Co 3 O 4 and Mn 3 O 4 catalysts, the rst reduction peaks of Co-Mn mixed phase catalysts are gradually shied to low temperature regions, suggesting that their low temperature reducibility is improved via the synergistic effect of Co and Mn species. All the results reveal that the Co 1.5 Mn 1.5 O 4 is easier to be reduced in the low temperature range of 200-300 C than other monolithic array catalysts, meaning a facilitated redox process and a better catalytic performance.    Table 2. The Ni 2p XPS spectra in Fig. 6d shows four peaks at the binding energy of 856.4, 861.5, 873.1 and 880.2 eV attributed to Ni 2p 3/2 , satellite peak of Ni 2p 3/2 , Ni 2p 1/2 and the satellite peak of Ni 2p 1/2 , respectively. 35,36 The Ni  2p XPS spectra primarily come from Ni foams. As for Co 2p XPS spectra (Fig. 6a), the peaks at the binding energy of 779.9, 781.5 and 785.5 eV are related to surface Co 3+ , Co 2+ and satellite peak of Co species, respectively. 17,18,31 In addition, the Mn 2p 3/2 XPS spectrum in Fig. 6b ). According to the Co 2p and Mn 2p results, surface oxygen vacancies are benecial for maintaining electrostatic balance and participating in the redox reaction. Therefore, it is reasonable that the Co 1.5 Mn 1.5 O 4 will exhibit an excellent catalytic activity for total toluene oxidation.
The toluene catalytic activity of as-prepared Co-Mn composite oxides as a function of temperature is shown in Fig. 7A. It could be observed that the catalytic combustion of toluene over catalysts increases with the increased temperature. As presented, the Mn species doping into Co 3 O 4 arrays has a signicant inuence on the catalytic activity of toluene oxidation. The optimal molar ratio of Co/Mn resulted in the large increase of catalytic activity for toluene oxidation. The reaction temperatures T 10 , T 50 and T 90 at which the toluene conversions of 10%, 50% and 90% is converted to CO 2 , are summarized in Table 1, which are used to compare the catalytic activities for toluene oxidation. For Co 3 O 4 sample, T 10 , T 50 and T 90 values are 255, 273 and 277 C, respectively, and toluene could be completely converted into CO 2 and H 2 O at about 280 C. The T 10 an T 50 of Mn 3 O 4 sample are superior to those of Co 3 O 4 sample, whose 10% and 50% toluene conversions could be obtained at 240 and 255 C, whereas the T 90 value over  Mn 3 O 4 obviously shis to high temperature region. It can be seen that adding manganese to cobalt improved the toluene conversions. According to the temperature values of complete toluene oxidation (T 99 ), the catalytic activity over the ve samples follows the sequence of Co 1.5 Mn 1.5 O 4 (270 C) z Co 2 MnO 4 (270 C) > Co 3 O 4 (280 C) > CoMn 2 O 4 (290 C) > Mn 3 O 4 (300 C), suggesting the Co 1.5 Mn 1.5 O 4 exhibits the highest catalytic activity for total toluene oxidation. CO 2 concentration is also detected during catalytic toluene reaction, and CO 2 yield is shown in Fig. 7b. It can be concluded that toluene is completely converted into CO 2 . The specic toluene reaction rates are calculated at 270 C, as shown in Fig. 7c. The Co 2 MnO 4 and Co 1.5 Mn 1.5 O 4 catalysts exhibits higher reaction rate with 1.34 mmol g À1 h À1 , which is four point two times the reaction rate of Co 3 O 4 catalyst with 0.32 mmol g À1 h À1 . Furthermore, the activation energies (E a ) over the catalysts are calculated by the Arrhenius plots in Fig. 7d. The E a decreases from 154.78 to 115.31 kJ mol À1 in the sequence of Co 3 O 4 , Mn 3 O 4 and Co 1.5 Mn 1.5 O 4 , which is related to the catalytic performance for toluene oxidation.
In addition, the stability of Co 1.5 Mn 1.5 O 4 with the best catalytic activity is also tested by a long-term time at different temperatures, as shown in Fig. 8. Under the complete removal temperature (270 C), the toluene conversions with    to provide a redox reaction: Co 3+ -Mn 3+ 4 Co 2+ -Mn 4+ . In addition, abundant surface oxygen vacancies are benecial for maintaining electrostatic balance and participating in the redox reaction. These results indicates the insignicant effect of lowtemperature reducibility, the synergistic effect of Co and Mn species and surface oxygen vacancies on the catalytic activity for total toluene oxidation.

Conclusions
A series of unique Co-Mn oxides with different Co/Mn molar ratios and morphology grown on interconnected Ni foam were well prepared via an ordinary hydrothermal reaction, in which Co 1.5 Mn 1.5 O 4 with the molar ratios of 1 : 1 displayed the highest catalytic activity for total toluene oxidation. It is observed that the Co/Mn molar ratio played signicant inuence on the textural properties and catalytic activities of obtained catalysts. Pure Co 3 O 4 sample mainly exhibited a series of intertwined and hexagonal nanosheets with the epitaxial nanowires, pure Mn 3 O 4 sample showed many hexagonal nanosheets with a diameter of 1-2 mm, the Co-Mn composite oxides mainly appeared urchin shapes self-assembled via a large number of nanowires. Based on the temperature values of complete toluene oxidation, the activity of toluene catalytic oxidation follows Co 1.5 Mn 1.5 O 4 z Co 2 MnO 4 > Co 3 O 4 > CoMn 2 O 4 > Mn 3 O 4 . From the characterization results of XPS and H 2 -TPR, introducing Mn element into Co 3 O 4 sample resulted in the formation of a solid solution between Co and Mn species, improved the low-temperature reducibility, concentration of surface Mn 3+ and Co 3+ species, and surface oxygen vacancies. It is deduced that he synergistic effect of Co and Mn species provided a redox reaction: Co 3+ -Mn 3+ 4 Co 2+ -Mn 4+ and enhanced the catalytic activity for total toluene oxidation.

Conflicts of interest
There are no conicts to declare.