Elsevier

Catalysis Today

Volume 164, Issue 1, 30 April 2011, Pages 451-456
Catalysis Today

Performance of bifunctional Pd/MxNyO (M = Mg, Ca; N = Zr, Al) catalysts for aldolization–hydrogenation of furfural–acetone mixtures

https://doi.org/10.1016/j.cattod.2010.11.032Get rights and content

Abstract

The performance of different bi-functional Pd catalysts (supported on Mg–Zr, Mg–Al and Ca–Zr mixed oxides) for the aqueous-phase aldol-condensation of acetone and furfural was studied in this work. Experiments were carried out in a batch slurry reactor at 323–393 K. Activity trends, as well as selectivity for the formation of the different adducts (C8 and C13), were correlated with the physico-chemical properties of the materials, mainly with the distribution of acid and basic sites. In general, it was observed that the catalyst with the lowest concentration of basic sites (Pd/Ca–Zr) presents the poorest performance. By contrast, the presence of medium-strength acid sites (as the observed in the case of one the Pd/Mg–Al catalyst) seems to enhance the catalytic activity of the material. Although selectivities trends were similar for the catalysts supported on Mg–Zr and Mg–Al catalysts, the ability of the materials for catalyzing the formation of the heaviest adducts depends mainly on the total concentration of basic sites.

Introduction

Renewable energies are nowadays a major technological challenge because of both the fossil fuels depletion and the environmental concerns about greenhouse gases emissions [1]. Among these renewable energy sources, biomass-based processes are the only able to produce organic fuels at relevant amounts, because of the carbon content of the biomass. The main first generation biofuels are bioethanol, used as a gasoline substitute, and biodiesel, produced from vegetable oils after conversion into the corresponding fatty acid methyl esters. These processes have several drawbacks, such as the competition with food crops, and the low efficiency in terms of fraction of biomass carbon atoms transformed into fuels [2].

In order to overcome these problems, efforts are focused on developing the so-called second generation biofuels, which are expected to be better in terms of energy balances and greenhouse gas emission reduction; as well as for avoiding the problems of competition with food and for water resources. There is still much work to be done to improve the existing processes and for the development of new efficient technologies. The most effective and efficient utilization of renewable biomass resources is through the development of an integrated biorefinery, in which the energy requirements of each process are balanced with those of other processes analogous to petroleum refinery [3]. Among the proposed processes (thermochemical, biological, chemical processes), those consisting of low-temperature hydrolysis of cellulosic materials (both chemically or enzymatically promoted) are the most promising. The final product of these hydrolyses are both single sugars (C5–C6) and oxidized/dehydrated derivatives of these sugars (furfural, hydroximethylfurfural, etc.). The transformation of the resulting sugars onto these derivatives involves dehydration reactions which can be easily carried out [4]. Direct hydrogenation of all these compounds will lead to C6 hydrocarbons of low cost [5], but performing a condensation (because of the carbonyl groups of the dehydrated molecules) of some of these units prior the deep hydrogenation step will lead to C9–C15 hydrocarbons, very valuable as diesel fuels. A pioneer work of Prof. Dumesic group [6] explored this possibility, summarizing the advantages expected from this process (low energy consumption, high carbon atom efficiency, easiness of purification steps, etc.). In subsequent papers, these authors proposed as model reaction for catalyst and process development the system furfural–acetone mixtures (furfural can be obtained by sugar dehydration, whereas acetone can also be obtained from biomass) [7]. In summary, these authors proposed a process consisting of the aldol condensation of both species (acetone can condense with one or two furfural molecules, yielding C8 or C13 species), followed by the deep hydrogenation of formed adducts. However, as the process is carried out in aqueous phase, it can be limited by the low solubility of adducts. Thus, a subsequent hydrogenation under mild conditions has been proposed [8]. During this hydrogenation, carbonyl functional groups are transformed into hydroxyl groups, the resulting adducts increasing the solubility.

Therefore, the proposed process requires catalysts with basic properties (required for the aldol condensation) and metal phases (mainly Pd). Among basic materials, basic mixed oxides are the first choice for this kind of applications. Although different oxides have been proposed (Mg–Al, Mg–Zr, Mg–Ti) [9], [10], [11], there are not, to the best of our knowledge, systematic studies relating the activity of these catalysts with their basic properties. In other applications of chemical reactions catalyzed by basic materials, such as conversion of 2-propanol in propanone [12], the decomposition of 2-methyl-3-bunil-2-ol [13], transesterification reactions [14], Knoevenagel condensation [15] or aldol condensation [16], it was observed that aspects as the concentration of basic sites and the distribution of the basicity strength of these materials play a key role on the catalytic performance.

The present work tries to fill this gap. For accomplishing this purpose, three different basic solids were used to prepare bi-functional basic-metal catalysts. The first one is a Mg–Zr mixed oxide, the best support among the tested in the published works specifically devoted to study this reaction [7], [8], [9]. A hydrotalcite-derived Mg–Al support was also considered, based on a previous work of our research group, in which the preparation procedure (saturation conditions, ultrasonication, etc.) has been optimized in order to obtain enhanced basic properties of the material [17]. The third considered support is a Ca–Zr support, prepared according to the method proposed by Liu et al., which leads to a mesoporous material with enhanced basicity [18], [19]. Concerning to the palladium addition, the conventional incipient wetness impregnation was compared to the isoelectric point impregnation method, proposed by Lambert et al. [20] for the preparation of highly dispersed catalysts. Comparison was done both in terms of catalytic performance and on the effect of palladium addition on the physico-chemical properties of the material.

Section snippets

Catalysts preparation

Mg–Al layered double hydroxides (LDH) with Mg/Al ratio of 3 were synthesized by co-precipitation at low super-saturation conditions (constant pH), according to the previously optimized procedure [17]. The material was prepared mixing 1 M solutions of Mg(NO3)2·6H2O (Fluka, >99%) and Al(NO3)3·9H2O (Panreac, 98%) in 3/1 molar ratio. A volume of 150 mL of this solution was added drop wise to 100 mL of K2CO3 (Panreac, 99%) 0.2 M under sonication at 298 K. The pH was kept at 10 by adding appropriate

Characterization of fresh catalysts

Crystalline structures of the four catalysts studied in this work, as well as the parent mixed oxides, were studied by XRD. If the bulk mixed oxides are considered, similar trends to those reported in the literature were observed. Periclase (2θ = 37°, 43° and 62°) and cubic MgO (2θ = 29° and 38°) are the only crystalline phases observed for the hydrotalcite-derived Mg–Al mixed oxides (Al is expected to be either or very disperse or in amorphous phases), whereas in the case of Mg–Zr mixed oxide both

Conclusions

The proposed catalysts are active for the aqueous-phase aldol-condensation of acetone–furfural mixtures. The activity of these materials is closely related to the distribution of both basic and acid sites. Pd/Ca–Zr catalyst presents the worst performance, related to its lower concentration of basic sites, as well as the weakness of these sites.

Pd/MgO–Al2O3 catalyst shows the highest furfural conversion, although its basic sites are not the strongest or in a higher concentration. Thus, an

Acknowledgments

This work was supported by the Spanish Government (contract CTQ2008-06839-C03-02). L. Faba thanks the Government of the Principality of Asturias for a Ph.D. fellowship (Severo Ochoa Program).

References (27)

  • C. Perego et al.

    Chem. Eng. J.

    (2010)
  • J.N. Chheda et al.

    Catal. Today

    (2007)
  • K. Karimi et al.

    Biomass Bioenergy

    (2006)
  • R.M. West et al.

    Catal. Today

    (2009)
  • G.W. Huber et al.

    Catal. Today

    (2006)
  • J.I. Di Cosimo et al.

    Appl. Catal. A

    (1996)
  • R. Zeng et al.

    J. Mol. Catal. A: Chem.

    (2005)
  • B.M. Choudary et al.

    J. Mol. Catal. A

    (1999)
  • J.C.A.A. Roelofs et al.

    J. Catal.

    (2001)
  • Y. Shigemasa et al.

    Tetrahedron Lett.

    (1994)
  • A.L. McKenzie et al.

    J. Catal.

    (1992)
  • M.A. Aramendía et al.

    J. Mol. Catal. A: Chem.

    (2004)
  • R. Sree et al.

    Fuel Process. Technol.

    (2009)
  • Cited by (42)

    View all citing articles on Scopus
    View full text