Elsevier

Catalysis Today

Volume 248, 15 June 2015, Pages 101-107
Catalysis Today

Numerical analysis of mass transport effects on the performance of a tubular catalytic membrane contactor for direct synthesis of hydrogen peroxide

https://doi.org/10.1016/j.cattod.2014.05.048Get rights and content

Highlights

  • We adopted a model for direct hydrogen peroxide synthesis in a membrane reactor.

  • The simulations describe the experimental data for various catalytic membranes.

  • Mass transfer limitations govern the liquid phase H2O2 direct synthesis.

  • Future systems benefit from combination of membrane and microreactor technology.

Abstract

Despite intensive research carried out by academia and industry over the past decades, the liquid phase direct hydrogen peroxide synthesis still faces a lot of challenges including safety, selectivity and productivity of the reaction. One facet of this multiphase reaction is the mass transport of the gaseous reactants, hydrogen and oxygen, in the liquid phase and inside of the catalyst. However, this has not been receiving much attention although, in many of the experiments reported in literature it seems to have affected to a large extent the results obtained. In this paper we report an analysis of performance data from a catalytic membrane reactor system published by Pashkova et al. in 2010 [1]. A rigorous two-dimensional model based on ANSYS Fluent was adopted to study the effects of an inert packing installed in the membrane reactor. The simulation results reveal a massive influence of mass transport limitation on the performance of the experimental system and also indicate significant potential for optimization of productivity and selectivity by adopting micro process technology combined with membranes for reactant dosage.

Introduction

Hydrogen peroxide is a powerful, clean and very versatile oxidant with water being the only oxidation byproduct. The worldwide production exceeds 2.2 Mt/a with an annual growth rate prediction of 4% [2], [3]. Most commonly hydrogen peroxide is used as a bleaching agent in the pulp/paper industry or as a disinfectant in wastewater treatment. Currently hydrogen peroxide is produced by the anthraquinone auto-oxidation (AO) process, where the product is formed via a two-step process on an organic carrier molecule. The AO process is only viable for large scale production greater than 40 kt/a [3], which is accompanied by high transportation costs to the consumer. The liquid phase direct synthesis of hydrogen peroxide offers the advantage that hydrogen peroxide is directly formed from its elements, hydrogen and oxygen, and the only byproduct is water. This technology would enable small scale on-site production, thus reducing transportation costs and expanding the market even further by making hydrogen peroxide available for innovative oxidation processes.

However, the direct synthesis of hydrogen peroxide still faces a lot of challenges, which hinders the process from being commercially applicable. One main problem is to achieve high selectivity [4]. Hydrogen peroxide is a thermodynamically unfavoured intermediate of the considered reaction network, which includes parallel water formation and consecutive decomposition and hydrogenation of hydrogen peroxide besides the actual synthesis itself [5]. In addition, the materials which catalyze the synthesis are also active for the unwanted side reactions. Another major drawback is that hydrogen and oxygen form an explosive gas mixture over a wide range of concentrations. This has to be avoided by dilution of the gases, which leads to low concentrations in the liquid phase and thereby reduces the overall rate of hydrogen peroxide production [4], [6].

Especially at these low concentrations mass transfer of the gaseous reactants in the liquid phase plays a huge role. The gas–liquid mass transfer is limited by the solubility of hydrogen and oxygen in the solvent, whereas the liquid–liquid, liquid–solid and the mass-transfer inside the porous catalyst is limited by slow liquid phase diffusion [7], [8], [9], [10], [11]. All these effects have to be taken into account when assessing the performance of different catalysts and reactor concepts or when trying to obtain further inside on the reaction mechanism and kinetics. The mass transfer influences the reactant concentrations in the liquid phase and at the active sites and thereby not only the overall production rate, but also the selectivity by e.g. determining the hydrogen to oxygen ratio inside the catalyst.

Membrane reactors are promising systems to overcome the challenges mentioned above [1], [12], [13], [14]. The membrane technology enables a separated dosage of the reactant gases into the liquid reaction medium, for example by having one reactant dissolved in the liquid feed, while the other reactant is fed over the membrane. This allows the usage of high partial pressures without risking the formation of an explosive gas mixture. So far the sometimes delicate handling of the membranes as well as mass transfer limitations have hindered these types of reactors from being commercially applicable [3]. Recently microreactors gained interest for the direct synthesis of hydrogen peroxide [15], [16], [17], [18]. They offer the possibility to operate within the explosive regime without causing safety issues [16], [17], [19]. This also allows the usage of higher reactant pressures resulting in higher liquid concentrations and higher production rates. Another advantage is that microreactors show excellent mass transfer properties due to the small inner dimensions. Hence microreactors appear to be very attractive to overcome the challenges in the direct synthesis of hydrogen peroxide.

In the present work the influence of the mass transfer on the direct synthesis of hydrogen peroxide has been examined. A two-dimensional model of the catalytic membrane reactor for ANSYS Fluent was adopted. To prove the reliability of the model the simulations are compared to experimental data generated by Pashkova et al. [1], [20].

Section snippets

Experimental setup of the membrane contactor

Pashkova et al. concluded from their experiments that the performance of their membrane contactor for the liquid phase direct synthesis of hydrogen peroxide strongly depends on the diffusive transport of hydrogen to the catalytic zone [1]. They showed this by filling their membrane reactor tube with inert glass beads to enhance the radial dispersion of hydrogen, which led to a ca. 100 fold increase of produced hydrogen peroxide.

A single channel tubular membrane reactor with a fine porous,

Two-dimensional simulation of the membrane reactor

The simulation of the tubular catalytic membrane contactor for direct hydrogen peroxide synthesis were performed for water and methanol as solvent, respectively with and without the implementation of the radial dispersion coefficient in the glass bead bed (for glass beads with 0.5 mm diameter). The reaction conditions were chosen according to the experiments with the Pd/C-Al2O3 membrane (mpd 6.9 mg, pO2 54 bar, pSaturator 50 bar, H2/N2-ratio 30/70) reported by Pashkova et al. [1]. The liquid flow

Conclusions

The calculated hydrogen peroxide concentration at the reactor outlet was in good agreement with the available experimental data for different catalytic membranes produced by Pashkova et al., even though only estimated kinetics for the Pd/C-Al2O3 catalyst was at hand. Since there was no significant influence of the catalyst on the overall system performance in contrast to the productivities and selectivities from their semi batch experiments, a strong mass transfer limitation has to be

References (25)

  • A. Pashkova et al.

    Chem. Eng. J.

    (2010)
  • C. Samanta

    Appl. Catal., A

    (2008)
  • N. Gemo et al.

    Chem. Eng. J.

    (2012)
  • J.H. Lunsford

    J. Catal.

    (2003)
  • G. Centi et al.

    Catal. Today

    (2003)
  • A. Pashkova et al.

    Chem. Eng. J.

    (2008)
  • T. Inoue et al.

    Fuel Process. Technol.

    (2013)
  • V. Paunovic et al.

    J. Catal.

    (2014)
  • Y. Voloshin et al.

    Catal. Today

    (2007)
  • Z. Anxionnaz-Minvielle et al.

    Chem. Eng. Process.

    (2013)
  • R. Hage et al.

    Angew. Chem., Int. Ed.

    (2006)
  • J.K. Edwards et al.

    Angew. Chem., Int. Ed.

    (2008)
  • Cited by (10)

    • Electrochemical multisensor system for monitoring hydrogen peroxide, hydrogen and oxygen in direct synthesis microreactors

      2018, Sensors and Actuators, B: Chemical
      Citation Excerpt :

      From the perspective of heterogeneous catalysis, selective catalysts with a minimized water formation reaction path are still subject to recent research activities. Moreover, the wide explosive range of H2/O2 gaseous mixtures [18] as well as reduced catalyst performance due to mass transfer effects [19] impede the potential of this process route in conventional multiphase reactors. Against this background, various advanced reaction engineering concepts have been employed in the recent decade to overcome these limitations, such as catalytic membrane reactors [18,20], microstructured reactors [7] and the rising electrochemical synthesis of hydrogen peroxide [21–24] that has recently become commercially available [25].

    • A flow-through tubular catalytic membrane reactor using zirconium sulfate tetrahydrate-impregnated carbon membranes for acidified oil esterification

      2017, Journal of the Energy Institute
      Citation Excerpt :

      Microreactor could provide excellent heat and mass transfer and shorter residence time [40]. The reaction could take place in the fine pores [41]. The esterification was conducted at the methanol/oil mass ratio of 1.0:1.0, reaction temperature of 65.0 °C and catalyst loading of 1.0% in the TCMR system [14].

    • Review: Microstructured reactors as efficient tool for the operation of selective oxidation reactions

      2016, Catalysis Today
      Citation Excerpt :

      It was concluded that a high oxygen excess is required to gain high hydrogen peroxide yields [18]. Recently Selinsek et al. proposed the application of microreactor technology to overcome the mass transfer limitations observed [19] for their membrane reactor concept applied for the direct synthesis of hydrogen peroxide [19]. Concluding the literature discussed, the feasibility of direct synthesis of hydrogen peroxide at high conversion and selectivity has been proven and suitable catalyst technology has been identified.

    • Towards a new membrane micro reactor system for direct synthesis of hydrogen peroxide

      2016, Catalysis Today
      Citation Excerpt :

      The fluxes of the dissolving gases are calculated by Fick’s first law with a concentration gradient determined by the saturation concentration calculated from Henry’s law at the wall representing the membrane and the concentration in the cell center calculated by the software as well as the cell dimensions. Depending on the cell concentrations calculated by ANSYS Fluent, the liquid phase diffusion coefficients were estimated either using temperature dependent correlations from DETHERM [26] or the Wilke–Chang correlation [27]. The reaction was approached as a pseudo-homogenous volumetric reaction at this stage to model the catalyst suspension.

    View all citing articles on Scopus
    View full text