Elsevier

Applied Catalysis A: General

Volume 468, 5 November 2013, Pages 160-174
Applied Catalysis A: General

Pd-Au and Pd-Pt catalysts for the direct synthesis of hydrogen peroxide in absence of selectivity enhancers

https://doi.org/10.1016/j.apcata.2013.07.057Get rights and content

Highlights

  • Performance of Pd, Au/Pd and Pt/Pd catalysts supported on K2621.

  • Catalysts tested with no selectivity enhancer in the DS under batchwise condition.

  • The productivity of hydrogen peroxide is very little sensitive to the amount of gold.

  • Addition of a small amount of Pt and Au makes the catalysts less active than 1% Pd.

  • Selectivity raises up to 60% with the addition of the second metal to the catalyst.

Abstract

Organic polymers are suitable candidates as catalytic materials for the direct synthesis of hydrogen peroxide (DSR) from dihydrogen and dioxygen, in view of the low temperature applied. In particular, strongly acidic ion-exchange resins have been already proposed for this reaction, either as carriers of PdII ions or Pd0 nanoparticles, but not for bimetallic Pd/Pt and Pd/Au catalysts, which are also active in this reaction. The introduction of PdII, PtII and AuIII precursors into the commercial ion-exchangers Lewatit K2621 and their subsequent reduction with an aqueous formaldehyde solution produced a library of bimetallic catalysts with 1% Pd (w/w) and a variable content of either Pt (0.1–1%, w/w) or Au (0.25–1%, w/w). The catalysts were tested with no selectivity enhancer in the DSR under batchwise condition at 2 °C in MeOH at 2 MPa, with a CO2/O2/H2 (72/25.5/2.5) mixture. In the Pd-Pt catalysts the addition of the smallest amount of Pt (0.1%, w/w) gave the best results in terms of the initial selectivity (43%) and the top concentration of H2O2 achieved in solution increased to 14.3 mM from 10.8 mM with the monometallic catalyst. With higher platinum amount the initial selectivity was lower, but in all these catalysts the addition of platinum limited the decrease of selectivity with time in comparison with the monometallic one. The addition of platinum in increasing amount seems also to favour poisoning of the catalysts by the products (most likely by H2O2). The presence of small amount of gold is also effective in achieving a comparatively high initial selectivity. With the addition of only 0.25% Au (w/w), the initial selectivity towards H2O2 and the top concentration of the product were boosted up to 61% and 17 mM, respectively. Gold in relatively little amount is also useful to limit the decrease of selectivity with time. In particular, it seems to limit the combustion of hydrogen and to be much more effective than platinum in this respect. Preliminary TEM analysis of the size and size distribution of the metal nanoparticles in the Au-Pd bimetallic catalysts shows that relatively large nanoparticles are preferable to achieve higher and stable selectivity, but also that size is likely not the most important parameter in controlling the catalyst performance.

Introduction

Hydrogen peroxide is a strong and environmental friendly oxidizing agent with many applications, such as paper and pulp bleaching, water treatment, chemical synthesis and electronics [1]. More than 95% of the worldwide production of H2O2 currently comes from the two-step Riedl–Pfleiderer process. In the first step an alkyl-anthraquinone is hydrogenated over a Ni or Pd catalyst to the corresponding alkyl-anthraquinol. Hydrogen peroxide is formed in the second step upon oxidation of the anthraquinol with air to the starting alkyl-anthraquinone, which is eventually returned to the hydrogenation unit. The major drawbacks of this process are the use of extensive amounts of organic solvents, loss of the alkyl-anthraquinone with formation of large amounts of by-products, the need of several steps of separation and concentration, large energy input. Moreover it is affordable only for large scale productions, which implies high investments costs and concentration of the production in a limited number of sites, from where hydrogen peroxide must be transported as (dangerous) concentrated solutions. On the other hand it allows the safe production of H2O2 at moderate temperature (40–60 °C), with no direct contact of H2 and O2. Although it has been known since the beginning of the 20th century, the direct synthesis of hydrogen peroxide (DSR) [2], only recently has started to raise interest as an alternative to the anthraquinone route [1], [3], [4] and its appeal even for potential industrial application is demonstrated by a pretty large patent literature [5], [6], [7], [8], [9], [10]. It is a conceptually simple reaction, where H2 and O2 dissolved in a liquid react over a solid catalyst in a single step.

Palladium is the best suited metal for this reaction, but unfortunately it is also active for the decomposition, the hydrogenation of hydrogen peroxide and the combustion of hydrogen to water. This can greatly lower the selectivity towards H2O2, because of the formation of water, which is by far the most thermodynamically favoured (but undesired) product (Scheme 1).

Another important issue to be coped with in the DSR is the safety in handling mixtures of H2 and O2. They must be kept out of the explosive (4–96%, v/v, for H2/O2 mixtures) and flammability ranges [11], [12], which are very large. Moreover the problem of flammability can concern the solvent too, for instance when methanol (as it is often the case) is used [13]. Being the most expensive reagent dihydrogen is usually used in the least possible amount, so that the DSR is generally carried out in the low H2 concentration safety range. Therefore the catalysts must ensure conversion of dihydrogen and selectivity towards hydrogen peroxide high enough to obtain useful concentrations of the product in the liquid phase. For all these reasons the DSR is a very demanding reaction, and in spite of great efforts from both the Academia and the Industry no technical breakthrough has been achieved yet.

The design of the catalyst has been so far the main tool employed to attain the required level of catalytic performance and a lot of studies on this topic have been continuously appearing in the literature. Most of them are devoted to palladium or bimetallic palladium/gold catalysts supported onto a number of different solids, such as SiO2 [14], TiO2 [15], ZrO2 and CeO2 [16], [17], carbon [18] and ion-exchange resins (sulfonated polystyrene-divinylbenzene) [19], [20], [21]. The colloidal palladium described by Liu and Lunsford [22] and gold–palladium systems [23] developed by Ishihara et al. were are also highly active in the DSR.

The bimetallic supported catalysts showed particularly promising results.

Pd/Au systems were carefully investigated, showing the benefits of the gold promotion of palladium [17], [24], [25]. Although bimetallic platinum–palladium systems have been reported in the patent literature [6], [7], [8] more often than the gold–palladium ones, they received comparatively much smaller attention in the open literature [16], [26], [27]. A 0.5% (w/w) Pd/SiO2 catalyst containing a small platinum amount (Pt/Pd = 5/95, mol/mol) showed good results, with a 2.5-fold increase in the production of H2O2 and only a slight decrease of the selectivity in comparison with a monometallic catalyst [26]. However the addition of higher amounts of platinum resulted in a significant decrease of the selectivity, which dropped from ca 60% to ca 50 and 20% when the molar Pt/Pd ratio was increased from 5/95 to 50/50 and 95/5 respectively [26].

It has been already shown in our laboratories that reduced palladium catalysts (1%, w/w) supported by the macroreticular ion-exchange resin Lewatit K2621% are excellent catalysts for the DSR in absence of acids and halides, performing better than Pd/SiO2 catalysts [20]. We report herein on the performance of Au/Pd and Pt/Pd catalysts supported on the same K2621 resin in the DSR. All the reported catalysts contain 1% (w/w) palladium, but the content of either gold or platinum differs from one to another (for gold 0.25%, 0.5% and 1%, w/w; for platinum 0.1%, 0.25%, 0.5% and 1%, w/w). Under halide-free conditions the best results in terms of selectivity towards hydrogen peroxide and its top concentration achieved were obtained with small additions of either Pt and Au, with the second being more effective.

Section snippets

Materials and apparatus

Unless otherwise stated, all the reagents and the materials were used as received from the supplier. A batch of Lewatit K2621 (sulfonated polystyrene-divinylbenzene macroreticular ion-exchange resin; exchange capacity = 1.92 mmol/g) was kindly provided by Lanxess. [Au(en)2]Cl3 was synthesized according to ref. [28]; [Pd(NH3)4]SO4 and [Pt(NH3)4](NO3)2 were purchased from Alfa Aesar; sodium thiosulfate pentahydrate (99.5%), potassium iodide, starch, concentrated sulphuric acid, 37% aqueous

Results and discussion

Lewatit K2621 is a macroreticular, sulfonated polystyrene-divinylbenzene (S-PSDVB) resin. It possesses permanent meso- and macropores both in the dry and in the swollen state and is a strongly acidic ion-exchanger, with an exchange capacity of 1.92 mmol/g [20].

S-PSDVB resins have several features making them attractive supports for DSR catalysts. The introduction of the metal precursors in the support can be accomplished with a simple ion-exchange of the counter-ions of the sulfonic groups (H+

Conclusions

Two sets of bimetallic Pt-Pd (Pd: 1.0; Pt: 0.1–1.0%, w/w) and Au-Pd (Pd: 1.0; Au: 0.25–1.0%, w/w) catalysts supported on the macroreticular ion-exchange resin Lewatit K2621 were prepared by simple ion-exchange in water of suitable precursors and reduction thereof with a refluxing aqueous solution of formaldehyde. It was found that the addition of a small amount of metal for both platinum (0.1%) and gold (0.25%) makes the catalysts less active and more selective than a 1% palladium catalyst on

Acknowledgments

S.S. is grateful to the University of Padova for financial support under the “Progetti di Ateneo 2010” initiative-project number CPDA100928. P.B. is grateful to the Otto A. Malm Foundation for financial support. This work is part of the activities at the Åbo Akademi Process Chemistry Centre (PCC) within the Finnish Centre of Excellence Programmes (2000–2005 and 2006–2011) by the Academy of Finland. The TEM pictures in Fig. 10 are courtesy of J.P. Mikkola, Umeå University.

References (35)

  • C. Samanta

    Appl. Catal. A: Gen.

    (2008)
  • G. Bernardotto et al.

    Appl. Catal. A: Gen.

    (2009)
  • F. Menegazzo et al.

    J. Catal.

    (2008)
  • C. Burato et al.

    Appl. Catal. A: Gen.

    (2009)
  • Q. Liu et al.

    J. Catal.

    (2006)
  • P. Biasi et al.

    Chem. Eng. J.

    (2011)
  • Q. Liu et al.

    Appl. Catal. A: Gen.

    (2008)
  • V.R. Choudhary et al.

    Appl. Catal. A: Gen.

    (2006)
  • F. Menegazzo et al.

    J. Catal.

    (2009)
  • A. Villa et al.

    Appl. Catal. A: Gen.

    (2009)
  • J.M. Campos-Martin et al.

    Angew. Chem. Int. Ed.

    (2006)
  • H. Henkel, W. Weber, US Patent 1,108,752...
  • J.K. Edwards et al.

    Angew. Chem. Int. Ed.

    (2008)
  • B. Bertsch-Frank, I. Hemme, S. Katusic, J. Rollmann, US Patent 6,387,364...
  • L.W. Gosser, US Patent 4,681,751...
  • L.W. Gosser, J.T. Schwartz, US Patent 4,772,485...
  • G. Paparatto, R. D’Aloisio, G. De Alberti, R. Buzzoni, US Patent 6,630,118...
  • Cited by (52)

    • Using black carbon modified with NbMo and NbPd oxide nanoparticles for the improvement of H<inf>2</inf>O<inf>2</inf> electrosynthesis

      2020, Journal of Electroanalytical Chemistry
      Citation Excerpt :

      Several studies published in the literature have sought to improve the capacity of hydrogen peroxide generation. Among the studies that deserve mentioning are those involving the modification of carbon with either organic compounds [10–14] or inorganic compounds [15–19]. Valim et al. (2013) studied the modification of Printex L6 carbon with tert-butyl-anthraquinone, which is an organic molecule that contains a quinine group [20].

    View all citing articles on Scopus
    View full text