Elsevier

Catalysis Today

Volume 342, 15 February 2020, Pages 178-186
Catalysis Today

Immobilization of a selective Ru-pincer complex for low temperature methanol reforming–Material and process improvements

https://doi.org/10.1016/j.cattod.2018.12.005Get rights and content

Highlights

  • Homogeneous Ru-pincer complex immobilized in supported liquid phase (Ru-SLP).

  • Ru-SLP catalyzed low temperature methanol steam reforming (T < 160 °C, p < 2 bar).

  • Promising Catalyst with high activity (TOF > 150 h-1), high CO2 selectivity (> 99.9 %) and reasonable lifetime (TOS > 70 h).

  • Addition of ionic liquids improves activity, but leads to less stable SLP catalysts.

  • Potential combination with HT-PEM-FC for decentralized energy systems feasible.

Abstract

A homogeneous Ru-pincer complex was immobilized by the supported liquid phase (SLP) technique and tested in continuous gas-phase steam reforming of methanol. Variations of catalyst composition (liquid phase, base and catalyst loading) as well as process parameters (temperature, flow rate) were carried out. High activity was obtained when ionic liquids were used for catalyst immobilization. However, better stability was obtained in the absence of ionic liquids using pure KOH.

Introduction

Recently, a range of homogeneous catalysts for aqueous phase methanol reforming have successfully been employed. They facilitate the release of hydrogen and the concomitantly carbon dioxide in the ratio 3 to 1 from liquid mixtures of methanol and water according to Scheme 1 already at temperatures below 100 °C. Remarkable activities of up to 4719 h−1 are achievable while the amounts of carbon monoxide as an undesired side product are minimal. [1]

Most of the published catalysts are Ruthenium-based and need high amounts of base, commonly KOH, to catalyze the reaction efficiently [[1], [2], [3]]. Currently one of the most promising systems is the Ruthenium-based PNP-Pincer catalyst [RuH(CO)Cl(HN(C2H4Pi-Pr2)2)] (1 in Fig. 1) published by Beller and co-workers. Mechanistic [4,5] and DFTbased [[6], [7], [8]] investigations on this specific molecular catalyst revealed the vital role of the ligand and basic additive, and thereby proposing an overall mechanistic picture. Notably a bi-catalytic Ru-based system can catalyze the methanol reforming reaction without the need of a basic additive, albeit at highly reduced activities of only 90 h−1 [9]. A different approach is the use of a Lewis-acidic additive and thereby circumventing the implication of the basic additive like solids (carbonates) formation [10]. Other catalysts based on iron, manganese and iridium have been investigated, although not reaching the activities of the [RuH(CO)Cl(HN(C2H4Pi-Pr2)2)] while also being more prone to oxidation (or sensitive to oxygen) [[11], [12], [13]]. Comprehensive reviews of homogeneous catalysts for CO2 hydrogenation and methanol reforming can be found in the literature [[14], [15], [16]].

Applying 1 the aqueous-phase methanol dehydrogenation and reforming was performed at low temperature around 90 °C. The reaction proceeds in three steps from methanol via formaldehyde/gem-diolate and formic acid to carbon dioxide and hydrogen in a 1:3 ratio (Scheme 1). Thus, full conversion of all hydrogen atoms present in the substrates to hydrogen is achieved. Besides an excellent productivity (> 350,000 turn overs) and stability (> 3 weeks) also a very good selectivity was obtained as only less than 10 ppm CO and CH4 were detected. Detailed mechanistic studies revealed that the C-H cleavage steps occur via an inner-sphere coordination of the methanolate, the gem-diolate and the formate. The presence of the base turned out to be essential for the reaction. The main role of the base is to enable the energetically favorable anionic, inner-sphere pathway. Further roles are the dehydrochlorination of the precatalyst 1 and the sequestration of the dehydrogenation products. The latter renders the key dehydrogenation step thermodynamically more feasible by providing a significant driving force (Fig. 1). The dehydrogenation of the formate was identified to be the least facile step of the overall process This is in agreement with the previously reported accumulation of formate during the reaction [4,5].

Aiming at the combination of methanol steam reforming for hydrogen generation and its use in PEM fuel cells (see Fig. 2), we became interested in the immobilization of such homogeneous catalyst complexes for continuous operation. Liquid-phase operation is not feasible, since the high amount of formed carbonates and the corrosive nature of the KOH constitute major obstacles. Ideally, a continuous vapor-phase process is envisioned, allowing direct use of the hydrogen in the fuel cell. A promising technology to enable homogeneous catalysis in vaporphase mode is the so called supported liquid phase (SLP) approach. The technology involves surface modification of a porous solid by dispersing a thin film of liquid onto the large inner surface area. Dedicated transition metal complexes dissolved in supported organic films (SLP) have been studied by Schoulten in the 1980s in great detail, mainly for the application in gas-phase hydroformylation reactions [[17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32]]. Supported aqueous phase (SAP) materials have been applied as suspended catalysts by the groups of Davis and Hanson in the 1990s [[33], [34], [35], [36], [37]]. Ionic liquids were independently introduced as novel coating materials by Mehnert et al. and Riisager et al. in 2003 [38,39]. Due to the extremely low vapor pressure of ionic liquids under the conditions of commercial gas-phase reactions, these supported ionic liquid phase (SILP) materials represent a fundamentally new approach to realize long-term stable, surface coated catalytic materials and absorbents [40].

Here we immobilize the homogeneous Ru-pincer complex RuH(CO)Cl(HN(C2H4Pi-Pr2)2) together with KOH in two ionic liquids dispersed onto alumina. In addition, a pure KOH deposited on alumina was also tested. The latter catalyst became a SLP system under reaction conditions, when the substrates water and methanol formed a liquid layer inside the pore system of alumina. Process parameters were identified and a preliminary optimization was carried out in order to achieve the highest catalyst activity. These data will be the basis for the implementation of continuous vapor- phase methanol reforming in direct combination with a PEM-fuel cell as depicted in Fig. 2.

Section snippets

Experimental

Chemicals were purchased from Strem and VWR and used without further purification. [P1444][MeSO4] and [P1444][NTf2] were purchased from IoLiTec. Nitrogen (99.9990 vol.-%) was purchased from Linde Gas.

Results and discussion

The Ru-pincer complex reported by Beller and co-workers is highly active in low temperature methanol reforming at 90 °C, but requires rather high concentrations of base, in this case KOH. Particularly challenging for this reaction is the presence of water and a strong base at elevated temperatures. Most if not all ionic liquids do not cope well with these harsh conditions and quickly decompose. Numerous reviews concerning IL stability were published over the last two decades [[41], [42], [43]].

Conclusion

In the present work we have for the first time immobilized an active Ru-pincer complex in the supported liquid phase for the continuous vapor phase methanol steam reforming using fixed bed reactors. By employing a basic and hygroscopic coating in the form of KOH deposited onto alumina support, the ruthenium-based PNPPincer complexes can efficiently be immobilized and catalyze the selective decomposition of methanol and water to hydrogen and carbon dioxide, with only trace amounts of CO under

Acknowledgements

The authors gratefully acknowledge funding from the German Federal Ministry of Economic Affairs and Energy (BMWi) within the project Metha-Cycle (grant number 03ET6071F).

References (47)

  • H.L. Pelt et al.

    The thermal and chemical stability limits of supported liquid phase rhodium catalysts in the hydroformylation of propene

    J. Mol. Catal.

    (1985)
  • J.M. Herman et al.

    The industrial hydroformylation of olefins with a rhodium-based supported liquid phase catalyst: I: description of the catalyst system, catalyst characterization and preparation, together with some orienting hydroformylation experiments with propyl

    Chem. Eng. J.

    (1986)
  • J.M. Herman et al.

    The industrial hydroformylation of olefins with a rhodium-based supported liquid phase catalyst (SLPC): II. Influence of the degree of liquid loading on the catalytic performance of the rhodium-based supported liquid phase catalyst

    Chem. Eng. J.

    (1987)
  • J.M. Herman et al.

    The industrial hydroformylation of olefins with a rhodium-based supported liquid phase catalyst (SLPC): III. The kinetics of propylene hydroformylation catalyzed by the rhodium-based supported liquid phase catalyst

    Chem. Eng. J.

    (1987)
  • J.M. Herman et al.

    The industrial hydroformylation of olefins with a rhodium-based supported liquid phase catalyst (SLPC): IV: heat-transfer measurements in a fixed bed containing alumina SCS9 particles

    Chem. Eng. J.

    (1987)
  • J.M. Herman et al.

    The industrial hydroformylation of olefins with rhodium-based supported liquid phase catalyst (SLPC): part V. The formation of 2-ethyl-2-hexenal and propane: extrinsic catalyst poisoning

    Chem. Eng. J.

    (1987)
  • J.M. Herman et al.

    The industrial hydroformylation of olefins with rhodium-based supported liquid phase catalyst (SLPC): part VI. General predesign of a large-scale SLPC Plant for the hydroformylation of propylene

    Chem. Eng. J.

    (1987)
  • J.P. Arhancet et al.

    Supported aqueous-phase catalysts

    J. Catal.

    (1990)
  • J. Arhancet

    Supported aqueous-phase, rhodium hydroformylation catalysts I. New methods of preparation

    J. Catal.

    (1991)
  • J.P. Arhancet et al.

    Supported aqueous-phase, rhodium hydroformylation catalysts II. Hydroformylation of linear, terminal and internal olefins

    J. Catal.

    (1991)
  • I. Guo et al.

    Bis[tris(m(sodium sulfonato)phenyl)phosphine] hexacarbonyl dicobalt, Co2(CO)6 (P(m-C6H4SO3Na)3)2, in a supported aqueous phase for the hydroformylation of 1-hexene

    J. Organomet. Chem.

    (1991)
  • A. Riisager et al.

    Continuous fixed-bed gas-phase hydroformylation using supported ionic liquid-phase (SILP) Rh catalysts

    J. Catal.

    (2003)
  • M. Nielsen et al.

    Low-temperature aqueous-phase methanol dehydrogenation to hydrogen and carbon dioxide

    Nature

    (2013)
  • Cited by (9)

    • On-board methanol catalytic reforming for hydrogen Production-A review

      2021, International Journal of Hydrogen Energy
      Citation Excerpt :

      Despite the reaction temperatures are already below 100 °C, most of the published catalysts need high amounts of base, commonly KOH, to catalyse the reaction efficiently [48,148,149]. Therefore, aiming at the on-board application of this efficient catalyst for SRM, Schwarz et al. [138] adopted the supported liquid phase (SLP) approach to enable homogeneous catalysis in vapour-phase mode, as shown in Fig. 7. By employing a basic and hygroscopic coating in the form of KOH deposited onto Al2O3 support, the Ru-based PNP-Pincer complexes can efficiently catalyse the reforming reaction to give H2 and CO2, with only trace amounts of CO under suitable conditions (130–170 °C, ambient pressure).

    • Spectroscopy, lifetime, and charge-displacement of the methanol-noble gas complexes: An integrated experimental-theoretical investigation

      2021, Spectrochimica Acta - Part A: Molecular and Biomolecular Spectroscopy
      Citation Excerpt :

      Among the complexes that are governed by weak non-covalent interactions, compounds formed by methanol have been used in a wide variety of studies such as those focused on hydrogen bond [18], vibrational analysis [19], influence of solvation effects on vibrational frequencies, hydrogen bonds, proton-transfer [20], and physicochemical properties like viscosity, density, and refractive index [21]. Methanol is also a conventional fuel in industrial processes and it appears in several other applications of relevance for biotechnology [22,23], materials science [24] and chemical processes [25]. In particular, aggregates formed by methanol with unsaturated molecules, in the matrix of Ng atoms, received attention to isolate the selectivity of interactions between the CH3 group of methanol and π-electrons [26] and to identify the role of long lifetime complexes in sensors applications [27,28].

    View all citing articles on Scopus
    View full text