Elsevier

Journal of Catalysis

Volume 344, December 2016, Pages 38-52
Journal of Catalysis

Synthesis, characterization, and catalytic performances of potassium-modified molybdenum-incorporated KIT-6 mesoporous silica catalysts for the selective oxidation of propane to acrolein

https://doi.org/10.1016/j.jcat.2016.08.014Get rights and content

Highlights

  • A novel Mo-incorporated KIT-6 catalyst was synthesized by a one-pot co-assembly method.

  • Mo-KIT-6 gave a high concentration of framework-isolated MoOx and anchored them firmly.

  • In situ Raman results showed that Mo-KIT-6 has a good redox ability to regenerate the reduced MoOx.

  • In situ Raman results showed that Mo-KIT-6 has strong carbon deposition resistance and stability.

  • The maximal yield of acrolein reached 25.9% over 0.25 K/0.1Mo-KIT-6 catalyst.

Abstract

A series of novel molybdenum-incorporated mesoporous silica catalysts (Mo-KIT-6) were successfully synthesized by a one-pot co-assembly method. For comparison, corresponding mesoporous KIT-6-supported molybdena catalysts (Mo/KIT-6) were also prepared by the impregnation method. For Mo-KIT-6 catalysts, the molybdenum was substituted into the framework of the KIT-6 support, which contributed to obtaining high concentrations of highly dispersed and isolated active sites and to anchoring the active sites firmly. We determined the identity of the active sites of Mo-KIT-6 catalysts as Mo oxide units with more anchoring Mosingle bondOsingle bondSi bonds than in the corresponding Mo/KIT-6 at elevated temperature. The Mo-KIT-6 catalysts possess appropriate redox properties, high stability, and a strong ability to resist carbonaceous species formation, which was confirmed by in situ UV Raman results. Furthermore, the addition of K to Mo-KIT-6 catalyst further promoted the formation of acrolein, and the maximum single-pass yield of acrolein reached 25.9%.

Introduction

The selective oxidation of light alkanes is of fundamental and industrial significance because it provides a potential route to effectively transform alkanes instead of olefins to value-added and useful chemicals for the effective utilization of abundant natural gas and shale gas resources. The direct transformation of propane to olefins or oxygenates such as acrolein and acrylic acid has attracted considerable attention in the literature [1], [2], [3], [4], [5], [6]. However, activation of the Csingle bondH bonds of light alkanes and combustion pathways to COx limit selective oxidation products yields. The design of selective oxidation catalysts appears critical for the efficiency of the selective oxidation process. The development of new catalysts, which can activate propane in the presence of oxygen and promote the formation of product with high yield, is the key to success [7], [8]. In the past decades, a wide variety of attempts and efforts at selective oxidation of propane have been made, but the single-pass yield of product is comparatively low (Table S1 in the Supplementary Material).

Supported molybdenum-based catalysts are one class of efficient catalysts employed in the selective oxidation of low alkanes [9], [10], [11], [12], [13]. The performance of these catalysts is strongly related to the nature of the active species. It has been reported that isolated vanadyl or molybdyl species with low aggregation and/or coordination structure favor the selective formation of oxygenates, whereas polymeric or crystalline species are suitable for deep oxidation of light alkanes [14], [15], [16]. Moreover, the activity and the selectivity of the catalysts also depend on the nature of the support [17]. Mesoporous silica molecular sieves, such as MCM-41 and SBA-15, which have high surface area, regular structure, and relatively high thermal stability, are preferred supports for catalyst preparation. The three-dimensional (3D) mesoporous silica KIT-6, possessing large and tunable pores with thick pore walls and complementary pores between the two main channel systems, has been widely applied in catalysis [18], [19], [20], [21]. This material is expected to be superior to mesoporous structures with one- or two-dimensional channels due to better dispersion of active species and faster diffusion of reactants and products in the 3D interconnected mesopores during reaction.

The incorporation of certain transition metal ions into the frameworks of mesoporous molecular sieves can contribute to achieving high concentrations of highly dispersed active sites, which may remarkably promote catalytic activity for reactions [22], [23], [24]. For example, Weng et al. [23] found that a mesoporous silica catalyst with incorporated V and Te species (VTeOx-MS) was more effective than one prepared by the conventional impregnation method for the oxidative conversion of propane to acrolein: the yield of acrolein was improved from 4.1% to 7.0%. Haller et al. [24] investigated comparison of vanadium-impregnated and incorporating MCM-41 catalysts in the partial oxidation of methane to formaldehyde. The results showed that V-MCM-41 was a superior catalyst, with high catalytic stability compared to the impregnated catalyst, and the authors proposed that the advantage of the vanadium-incorporated catalyst would likely increase with vanadium content, because aggregation becomes an increasing problem, especially for impregnated catalysts, at higher loadings. The framework-incorporated catalysts are different from supported catalysts: the active metal ions are substituted into the silica matrix, and high concentrations and high stability of active sites can be achieved. As for supported catalysts, the active phases are only present on the surface or in the pores of the support and may aggregate further. However, the structure of active sites, especially for Mo-incorporated catalysts, is ambiguous from molecular level in the literature.

The use and development of in situ techniques for the characterization of catalysts have gained remarkable interest. In situ studies ensure that the observed time dependence and structural changes are related to the changes in catalytic activity. Therefore, direct correlations can be established between the structural and chemical features and the performance of catalysts under different process conditions. In situ Raman spectroscopy serves as a unique tool to investigate catalysts and to get structural information about active sites. In situ Raman spectroscopy can study the molecular structures of the catalytic active sites dynamically responding to reactive environments (reactant concentration and reaction temperature), which can provide information on the structure of the dispersed surface metal oxide species by probing the vibrational properties of metal–oxygen bonds under in situ conditions [25], [12], [26], [27], [28], [29], [30]. Unfortunately, it is hard to obtain definite information on the transition metal ions in supports characterized by conventional Raman spectroscopy, due to their low concentration or strong fluorescence interference. UV Raman spectroscopy is a useful technique for addressing these challenges [31], [32] because fluorescence interference can be avoided in UV Raman spectra. In addition, the excitation source is shifted to the UV region close to an electronic absorption. A resonance-enhanced signal can be improved by several orders of magnitude compared with the normal Raman signal [33]. Resonance Raman spectroscopy is particularly useful when applied to a sample containing a mixture of individual species [34]. Therefore, in situ UV Raman spectroscopy will play an important role in studies of catalytic structure–activity relationships. However, in situ UV Raman characterization of transition-metal-incorporated mesoporous silica in the selective oxidation of propane remains relatively unexplored in the literature.

Alkali metals, such as potassium, are an important component in catalysts and generally act as promoters [16], [35], [36], [37], [38], [39]. The effect of alkali metal commonly involves acid–base interaction, adsorption/desorption behavior, and the reducibility of the catalysts, which are related to structural changes of the active species. In the selective oxidation of light alkanes, the addition of alkali can reduce the acidity of the catalyst and increase the dispersion of the active sites. As a result, these functions can inhibit the excessive oxidation of the reaction intermediate and enhance the formation of oxygenates [40]. Thus, alkali metal modification may be a good method for creating effective catalysts for the selective oxidation of low alkanes.

In this work, we report the synthesis of a novel type of Mo-incorporated KIT-6 catalysts by a one-pot co-assembly method. As a comparison for introducing the Mo after synthesis of the KIT-6 support, KIT-supported molybdena catalysts were also prepared by impregnation. The catalytic behavior of these two series of catalysts for the selective oxidation of propane was comparatively evaluated. The Mo-incorporated KIT-6 catalysts were further modified by the potassium to promote acrolein formation. Furthermore, we study the Mo oxide nanostructure on mesoporous silica KIT-6 by combining in situ Raman spectroscopy with a variety of other analytical techniques, and the origins for the large difference in catalytic performance between Mo-framework incorporated KIT-6 and KIT-6 supported molybdena catalysts are discussed.

Section snippets

Catalyst preparation

Mo-incorporated mesoporous molecular sieves (Mo-KIT-6) were synthesized by a one-pot co-assembly method using the nonionic triblock copolymer surfactant Pluronic P123 (EO20PO70EO20, MW = 5800, Aldrich) and n-butanol (BuOH) as a structure-directing mixture, tetraethyl orthosilicate (TEOS) as silica source, and ammonium heptamolybdate (AHM) as molybdenum source in hydrochloric acid. A typical procedure was as follows: a quantity of 2 g of Pluronic P123 was added into 19 g HCl (2 mol L−1) with stirring

Catalytic performance

The catalytic performance of the samples prepared by different synthesis methods for the selective oxidation of propane was evaluated in a fixed-bed quartz reactor, and the results are listed in Table 1. These results show that Mo-incorporated KIT-6 catalyst possesses higher activity and selectivity than the KIT-6-supported molybdena catalysts, which favor deep oxidation to COx. Considering the yield sum of acrolein and olefins, 8Mo-KIT-6 is the best catalyst; the highest yield of acrolein and

The origins of different catalytic performance between framework-incorporated and supported catalysts

Compared with the supported catalyst prepared by the impregnation method, the framework-incorporation strategy can control the degree of polymerization of MoOx species and achieve high concentrations of the highly dispersed and isolated active sites. Further, the framework-incorporation strategy can obtain a special structure of active sites, which usually show better properties than the supported catalysts [10], [19], [55].

There is a general consensus in the literature that Csingle bondH bond breaking is

Conclusions

A series of Mo-incorporated KIT-6 mesoporous molecular sieve catalysts were successfully synthesized by a one-pot co-assembly method and applied in the selective oxidation of propane. For comparison, corresponding mesoporous KIT-6-supported molybdena (Mo/KIT-6) catalysts were also prepared by the impregnation method. The catalytic performance shows that Mo-incorporated KIT-6 catalysts possess higher activity and selectivity than the KIT-6-supported molybdena catalysts. The highest yields of

Acknowledgments

This work was supported by the National Natural Science Foundation of China (91545117), the National Basic Research Program of China (Grant 2012CB215001), and the Scientific Research Foundation of China, University of Petroleum, Beijing (Grant 2462013YJRC016).

References (71)

  • Y. Liu et al.

    J. Catal.

    (2004)
  • B. Schimmoeller et al.

    J. Catal.

    (2010)
  • X. Yi et al.

    Catal. Commun.

    (2009)
  • L. Chen et al.

    Catal. Commun.

    (2004)
  • M.M. Bettahar et al.

    Appl. Catal. A

    (1996)
  • C. Batiot et al.

    Appl. Catal. A

    (1996)
  • K. Chen et al.

    J. Catal.

    (2000)
  • J. Li et al.

    J. Energy Chem.

    (2014)
  • Y. Lou et al.

    J. Catal.

    (2007)
  • A. Christodoulakis et al.

    J. Catal.

    (2008)
  • Z. Zhao et al.

    J. Catal.

    (2000)
  • Z. Zhao et al.

    Catal. Today

    (2004)
  • K. Soni et al.

    Appl. Catal. B

    (2009)
  • Z. Liu et al.

    Appl. Catal. B

    (2012)
  • H. Wu et al.

    J. Catal.

    (2014)
  • J.A. Melero et al.

    Appl. Catal. A

    (2007)
  • L. Shi et al.

    J. Catal.

    (2013)
  • G. Du et al.

    Appl. Catal. A

    (2006)
  • A. Christodoulakis et al.

    J. Catal.

    (2006)
  • G. Xiong et al.

    J. Catal.

    (1999)
  • C. Li

    J. Catal.

    (2003)
  • K. Chen et al.

    J. Catal.

    (2000)
  • F. Ivars et al.

    Catal. Today

    (2009)
  • R.B. Watson et al.

    J. Catal.

    (2000)
  • J. Liu et al.

    J. Catal.

    (2012)
  • L. Kong et al.

    Appl. Catal. A

    (2016)
  • D. Liu et al.

    Appl. Catal. A

    (2009)
  • L. Wang et al.

    J. Mol. Catal. A

    (2005)
  • M.R. Smith et al.

    Catal. Lett.

    (1993)
  • M.A. Smith et al.

    J. Catal.

    (2014)
  • C. Zhao et al.

    Catal. Today

    (2006)
  • G. Du et al.

    J. Catal.

    (2008)
  • K. Alexopoulos et al.

    J. Catal.

    (2012)
  • C. O’Neill et al.

    Catal. Today

    (2010)
  • A. Kaddouri et al.

    Appl. Catal. A

    (1999)
  • Cited by (0)

    1

    This author contributed equally as the first author.

    View full text