Volatile organic compounds (VOCs) are chemicals that are defined by a saturated vapor pressure that exceeds 70 Pa at room temperature or a boiling point that is lower than 260 ℃ at standard pressure. They are present in haze, photochemical smog, and contribute to the greenhouse effect. They are also involved in the depletion of atmospheric ozone and the production of ground-level ozone [1, 2]. In particular, chlorinated volatile organic compounds (CVOCs), which are highly toxic, are difficult to decompose in air because of the occurrence of polychlorinated byproducts, being increasingly one of the VOCs which handle hardly [3]. The CVOCs, as semi-finished chemical products, are widely used as solvents for dry cleaning and degreasing processes in industrial manufacturing [4, 5].

Generally, the techniques for the destructive removal of CVOCs include high-temperature direct combustion, catalytic hydrodechlorination, catalytic steam reforming, photocatalytic oxidation, and catalytic combustion. Compared with other techniques, catalytic combustion is applied in a wide range of applications because of its lower energy consumption, lower handling temperature, and higher purification efficiency.

For the catalytic combustion of CVOCs, catalyst support materials with a large specific surface area and pore volume, which are beneficial for the dispersion of the active substances, are always prevalent. They usually consist of molecular sieves (such as γ-Al₂O₃, HFAU, HZSM-5, MCM-41), metal oxides, and honeycomb ceramics [3]. Compared with other two supports, molecular sieves not only have better thermal stability, but also possess higher specific surface areas and more acidic sites [6]. However, the problem of carbon deposition is yet to be solved. As a nanostructured material, MCM-41 with a hexagonal ordered pore arrangement and large adsorption quantity was first discovered in 1992 [7]. It has been widely applied in resid catalytic cracking, heavy oil catalytic hydrogenation, alkyl isomerization, and macromolecular fine chemicals. Furthermore, it is completely composed of silica and has no acidity, effectively avoiding the issue of carbon deposition.

Currently, the catalysts used for the destruction of CVOCs can be classified into two types according to their active components, including noble metals catalysts [8-10] and non-noble metal catalysts. In practice, noble-metal-based catalysts show higher activity. However, they are more expensive and undergo deactivation because of chlorine poisoning and the formation of polychlorinated compounds. Therefore, many efforts have been devoted to the synthesis of transition-metal oxide catalysts [11-16] because they resist deactivation to a larger extent and are cheaper. Usually, metal oxides (such as V₂O₅, Cr₂O₃, MnO₂, Co₃O₄ and NiO) or composite metal oxides are used as active components of transition metal oxide catalysts. In particular, manganese and cobalt oxides possess better activity, lower cost, and create little or no environmental pollution. Therefore, they have been widely used both domestically and internationally [17, 18]. However, when use only one component, the loss of active sites (which could adsorb HCl and/or Cl₂) or the formation of volatile species (metal chloride or metal chloride oxide) at lower temperature may cause partial deactivation of these transition metal oxides catalysts, which may restrict the applications of transition metal oxides catalysts [19]. Thus, composite metal oxides catalysts have been reported, which improve the mobility of the surface-active oxygen and increase the surface acidic sites through the synergistic effect of the metal-metal and metal-support interactions.

There have been numerous reports of molecular-sieves-supported transition-metal oxides as catalysts that have been used to eliminate CVOCs [20-27]; for example, Cu-ZSM-5 [26] and Cr₂O₃-CuO/HZSM-5 [27]. Furthermore, the MCM-41 as a catalyst support, its applications for the destruction of CVOCs have also been reported, including Pt/P-MCM-41, La/MCM-41, Ce/MCM-41, and Pt/MCM-41, which showed a high catalytic performance [28-30]. In the catalytic combustion field of CVOCs, there are rarely reports about MCM-41 supported MnCo catalysts for destruction of chlorobenzene (CB). Therefore, we synthesized MCM-41-supported MnCo catalysts for CB combustion.

In this paper, we prepared MCM-41-supported MnCo catalysts, and explored their catalytic performance for combustion of CB in air. Furthermore, the synergistic effect between Mn (the main active phase), Co (the second active phase) and MCM-41 was studied. The catalysts were analyzed by X-ray diffraction (XRD), high-resolution transmission electron microscopy (HRTEM), and energy dispersive X-ray spectroscopy (EDS), nitrogen (N₂) adsorption-desorption, temperature-programmed reduction of hydrogen (H₂-TPR), and temperature-programmed desorption of CB (CB-TPD). Furthermore, the activity and durability of the materials was evaluated to further investigate the performance of the catalysts.

The MCM-41 was prepared according to the following procedure. Tetrabutylammonium silicate (TBAS) was obtained from tetrabutylammonium hydroxide (TEAOH, 40 wt%, Aldrich) and fumed silica (Sigma) in a 10:1 ratio. Then, 12.21 g of TBAS was combined with 20.3 g of cetyltrimethylammonium chloride (CTAC, 25 wt%, Aldrich) and 5.94 g of H₂O under stirring, followed by the addition of 5.91 g of fumed silica. The resulting gel was placed in a Teflon bottle and heated for 120 h at 100 ℃ to perform the crystallization. After cooling to room temperature, the solid product was recovered by filtration, washed thoroughly with deionized water, and dried in an oven at 80 ℃ overnight. Finally, the template CTAC was removed by calcination in a stream of air at 550 ℃ for 5 h [31]. A series of MnCo/MCM-41 catalysts with different molar ratios of Mn/Co (3:1, 6:1, and 9:1) were prepared by co-impregnation of MCM-41 with an aqueous solution of Mn(NO₃)₂ (50%) and Co(NO₃)₃·6H₂O overnight. Then, the samples were stirred in the electrothermal setup until desiccation. Later, calcination was performed at 500 ℃ for 2 h. The particles with a 40–60 mesh size were sieved to obtain the products. The total content of the single metal (Mn or Co) and mixed MnCo (the sum of both Mn and Co) for all the catalysts was10 wt%.

The activity evaluation of the catalysts (350 mg) was performed in a WFS-3010 microreactor (Xianquan, Tianjin, China), in which the space velocity was 20000 h^–1 and the concentration of CB was 1000 ppm. The concentration of O₂ in the gas flow stream was 21%. The on-line measurements were conducted using a Shimadzu GC-14 (Japan), which had a flame ionization detector (FID). The temperatures of the vaporizer and column were both 120 ℃. The data were obtained and analyzed using a N2000 chromatography data workstation. The degradation products were detected by mass spectrometry (MS, QGA, Hiden, UK). No byproducts other than H₂O, CO₂, and HCl were detected. Thus, the conversion was calculated based on CB consumption. The durability of the catalysts for CB combustion was also investigated under the same conditions.

The composition of the catalyst phase was determined by XRD, which was performed with a PANalytical EMPYREAN powder diffractometer operated with a Cu K_α source (λ = 0.15406 nm). The experimental conditions were as follows: tube current of 40 mA, tube voltage of 40 kV, scanning rate of 0.02°/s and scanning limitation of 10°–80°.

HRTEM was carried out using a JEOL-2010 to investigate the morphologies of MCM-41 and 10% MnCo (6:1)/MCM-41. The working voltage was 200 kV. The elements present in the catalyst were determined by EDS using an OXFORD INCA instrument (Oxford Instruments, UK).

The specific surface area and pore volume of the catalysts were measured by a Tristar Ⅱ 3020 apparatus (Micromeritics Company, USA), by using the N₂ adsorption-desorption technique (–195.8 ℃). The sample was pretreated at 200 ℃ for 4 h. The specific total surface area (A_BET) was calculated using the Brunauer-Emmett-Teller (BET) equation, whereas the total pore volume (V_p) was evaluated from the nitrogen uptake at a relative N₂ pressure of p/p₀ = 0.99. The Barrett-Joyner-Halenda method was used to calculate the average pore size.

The oxidation properties of the catalysts were determined by H₂-TPR experiments. 50 mg of the catalyst was pretreated at 450 ℃ in argon for 1 h, after which the temperature was decreased to 50 ℃. When the adsorption of the organic compounds was at a state of dynamic balance, in other words, the base line was stable, and then the temperature was programmed to rise at a constant rate of 7.5 ℃/min up to 450 ℃. Hydrogen uptake was measured with a thermal conductivity detector, and the compounds were detected by gas chromatography-mass spectrometry (GC-MS) (QGA, HIDEN, UK) on-line.

The CB-TPD experiments were performed in a quartz tube. Before the adsorption of CB, 350 mg of catalyst was pretreated in dry air at 300 ℃ for 0.5 h. After it was cooled down to 50 ℃, the adsorption of CB was carried out under a flow of CB/N₂ until adsorption saturation, which was symbolized by a stable CB signal in the mass spectrometer. Then, pure N₂ was blown for 1 h to clean the CB of the tubing in the system. Finally, desorption of CB was implemented by flowing pure N₂ at a temperature step of 7.5 ℃/min from 50 to 500 ℃. The concentration of CB was detected on-line by GC-MS [32].

Fig. 1 shows the catalytic oxidation activity of different catalysts. The activity of the catalysts for CB combustion with 1000 ppm decreased in the order: MnCo (6:1)/MCM-41 > MnCo (9:1)/MCM-41 > MnCo (3:1)/MCM-41 > Mn/MCM-41 > Co/MCM-41. The 10% MnCo (6:1)/MCM-41 catalyst exhibited the highest activity, and the temperature for complete CB degradation was 270 ℃. The 10% MnCo (9:1)/MCM-41 catalyst exhibited a relatively low activity; this was attributed to the higher amount of Co. Too much Co loading could override the MnO_x sites (main active sites). The catalyst activity was also lower when the Mn/Co molar ratio was 3, as a consequence of the lack of oxygen vacancies. Thus, the appropriate loading of Mn and Co played an important role in improving the MnCo/MCM-41 activity. The catalytic activity of similar catalysts reported in the literature for the catalytic combustion of CB are listed in Table S1 [33-37], which shows that 10% MnCo (6:1)/MCM-41 had higher activity.

Apart from catalytic activity, the durability of a catalyst is also very important in determining the performance of catalysts. The MnCo (6:1)/MCM-41 catalyst showed good durability in a continuous 1000 h reaction under dry air at 260 ℃ (Fig. 2). However, the conversion of CB dropped in the presence of benzene because of competitive adsorption. Notably, the conversion of CB decreased more when the water was also added, indicating that water also inhibited CB oxidation, because some active sites were overlapped. Nonetheless, the conversion of CB recovered without the presence of benzene and water.

The XRD patterns are shown in Fig. 3. A big diffraction peak attributed to silica was observed at approximately 2θ = 22.8°. Meanwhile, the characteristic peaks of MnO₂ and Co₃O₄ were detected in 10% Mn/MCM-41 and 10% Co/MCM-41. The peak of MnCoO_xwas observed in 10% MnCo (6:1)/MCM-41. In the 10% MnCo (6:1)/MCM-41 catalyst, two weakly intense peaks appeared at a diffraction angle 2θ of 36.0° and 54.2°, which corresponded to the XRD patterns of MnCo₂O₄ and CoMn₂O₄, respectively [38]. Accordingly, we deduced the presence of MnCo₂O₄ and CoMn₂O₄. The presence of Co₃O₄, CoO, MnO, Mn₃O₄, Mn₂O₃, and MnO₂, or other species was not detected in the XRD pattern of 10% MnCo (6:1)/MCM-41, indicating that no Co₃O₄, CoO, MnO, Mn₃O₄, Mn₂O₃, and MnO₂ crystal phases existed in the MnCo/MCM-41 composite oxide system, or that they existed in highly dispersed or amorphous forms. Thus, Mn⁴⁺ and Mn³⁺ were the main crystallite phase species that existed in the prepared MnCo/MCM-41 oxide catalysts.

Fig. 4 shows the HRTEM images and the EDS spectra of the samples. Fig. 4(a) shows that MCM-41 had a well-ordered hexagonal array structure. Fig. 4(b) shows that the particles of MnO₂ and Co₃O₄ dispersed uniformly. The EDS spectrum clearly confirmed the existence of Mn, Co, and O, which indicated the successful loading of the active components on the surface of MCM-41.

Fig. 5 shows the N₂ adsorption-desorption isotherms of the samples. The A_BET, V_p, and average pore size of the samples are summarized in Table 1. From the figure and table, a distinct capillary condensation phenomenon could be observed at relative pressures between 0.45 and 0.85, indicating the presence of uniform mesopores. The existence of micropores was also confirmed by the N₂ adsorption-desorption isotherms rising sharply giving 0~0.3 (p/p₀). And it could be found, for those supported active components materials, all their A_BET, V_p, pore size and adsorption quantity reduced compared with MCM-41. It was probably attributed to that the oxides of Mn and Co blocked the pores of MCM-41. With the addition of Mn and Co, the relative content of MCM-41 decreased, this also makes the A_BET, V_p, pore size and adsorption quantity of the samples smaller.

Fig. 6 shows the H₂-TPR profile of the MnCo/MCM-41 catalysts which supported different molar ratios of Mn and Co. The reduction of the 10% Mn/MCM-41 catalyst occurred at 267, 288, and 510 ℃. The 10% Co/MCM-41 was reduced in two steps [39]: the first peak centered at 270 ℃ was attributed to the reduction of Co³⁺ to Co²⁺, whereas the second step reduction proceeded at 341 ℃ and was attributed to the subsequent reduction of Co²⁺ to Co. Compared with above two catalysts, the reduction peaks of 10% MnCo (6:1)/MCM-41 catalyst were shifted to a lower temperature. This shift indicated that the oxidation property of catalyst was improved because of the interaction between MnO_x, CoO_x, and MCM-41. The Mn-O chemical bonds that existed in the MnCoO_x composite compounds could be weakened by the strong interaction; the weakened Mn-O chemical bonds could react with hydrogen molecules to achieve reduction at a low temperature. Therefore, the supported MnCo composite compounds showed high oxidability at a low reduction temperature. The high low-temperature oxidability could produce reactive oxygen species at a low reaction temperature. Thus, the supported MnCo composite compound catalysts provided high catalytic combustion activity for organic compounds [40]. The catalytic activity of the mixed-oxides-based catalysts depended on the redox properties of the Mn⁴⁺/Mn³⁺ system that were present on the surface of the oxides. That was probably modulated by the quantity of the manganese employed in the synthesis, as well as the use of cobalt with the activity to promote redox cycles [41]. These were maybe the synergistic effect.

Fig. 7 shows the CB adsorption-desorption profiles of the support and the catalysts in the TPD test. For supported catalysts, some Mn or Co particles entered into the pores of the supports, which caused the adsorption capacity of CB reducing than MCM-41. However, bimetal-oxide-based catalyst had larger adsorption capacity of CB than that of single metal oxide catalysts. The order of CB adsorption capacity was as follows: 10% Co/MCM-41 < 10% Mn/MCM-41 < 10% MnCo (6:1)/MCM-41 < MCM-41. The desorption temperatures of the supported catalysts increased, which indicated the increased adsorption ability of CB. Higher desorption temperature implied higher adsorption strength for CB. The order of CB desorption temperatures was as follows: MCM-41 < 10% Co/MCM-41 < 10% Mn/MCM-41 < 10% MnCo (6:1)/MCM-41. Surely, larger adsorption capacity and stronger adsorption ability of the catalysts are both beneficial for combustion of CB. In other words, 10% MnCo (6:1)/MCM-41 catalyst had the best combustion activity for CB. This conclusion was consistent with the result of the activity evaluation experiments.

In this paper, mixed oxides with active binary phases based on MCM-41 catalysts were prepared. The performance of CB combustion over MnCo/MCM-41 catalysts was studied. The activity tests showed that the 10% MnCo (6:1)/MCM-41 catalyst had the highest catalytic activity. The complete conversion of CB was achieved at 270 ℃. The texture, structure, morphology, dispersion of the active components, and the oxidation properties of catalysts were characterized by a series of techniques. The catalytic behavior of the catalysts was significantly affected by the molar ratios of the metal oxides and the interaction between the supported components. The XRD indicated the formation of MnCoO_x compounds for 10% MnCo (6:1)/MCM-41 catalyst. The HRTEM-EDS indicated a better dispersion of Mn or Co oxides on MCM-41 for 10% MnCo (6:1)/MCM-41 catalyst. The N₂ adsorption-desorption and CB-TPD results showed that some particles of the supported components could enter into the pores, resulting in a smaller specific surface area, pore volume, pore size, and lower adsorption quantity of CB. Furthermore, the durability test demonstrated that there was no deactivation for 10% MnCo (6:1)/MCM-41 catalyst during a 1000 h continuous reaction. This research provides valuable information for the development of better catalysts with wide-spread applications.