http://dx.doi.org/10.1016/j.apcata.2007.04.004
1. Introduction
Gold catalysts are useful for ablating pollutants, cleaning H2 streams, and synthesizing chemicals [1], [2], [3] and [4]. The supports of choice are metal oxides, such as TiO2, CeO2, and Fe2O3 [5]. Although carbon materials are useful as adsorbents, electrodes, and catalyst supports owing to their high surface area, stability in acid and basic media, and the ease of recovering precious metals [6], carbon-based gold catalysts are seldom reported. Recently, the use of Au/C in liquid-phase oxidation of organics has been explored [7], [8], [9], [10], [11], [12], [13], [14] and [15]. This major advance may trigger a new wave of research into Au/C-based catalysts. The activity of Au/C in CO oxidation is very low, however [14], [15], [16], [17], [18] and [19] F. Wang and G.X. Lu, Catal. Lett. 115 (2007), p. 46. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (0)[19].
Attempts have been made to modify carbon supports before loading gold. In a 3M patent, the inventors impregnated carbon with K2CO3 and loaded gold through physical vapor deposition [18]. They observed enhanced activity in CO oxidation, but the origin of this promotion was not clear, and the leaching of K2CO3 in aqueous reactions was a problem. Kiwi-Minsker et al. [20] impregnated carbon fibers with Fe(NO3)3 followed by precipitation, and loaded gold using Au(en)2Cl3 precursor. The Au/FeOx/C exhibited some activity in CO oxidation but quickly deactivated [20]. Rønning and coworkers [21] and [22] modified carbon nanofibers by TiO2 nanoparticles and loaded gold by two methods. The deposition–precipitation method led to large gold particles (>50 nm), whereas the colloidal dispersion led to smaller gold particles (6 nm). The Au/TiO2/C synthesized via dispersion of gold colloids showed activity in the water–gas shift reaction, but no data on CO oxidation were reported [22].
To design surface-modified carbon-based gold catalysts, it is essential to properly introduce the modifier so as to avoid the mechanical segregation of the modifier and carbon and achieve the catalytic synergistic effect of gold and the modifier. Recently, there has been great interest in the electroless deposition of MnOx on carbon electrodes achieved by immersing carbon in an aqueous KMnO4 or NaMnO4 [23], [24], [25], [26], [27], [28] and [29]. This practice results in the conformal coating of MnOx on carbon with the sacrificial oxidation of carbon surface (4KMnO4 + 3C + 2H2O → 4MnO2 + 3CO2 + 4KOH) [29]. The objective of these studies is to develop high-performance MnOx/C capacitors as energy-storage devices; however, to the best of our knowledge, these novel materials have not yet been used for preparing supported gold catalysts. Herein we report the preparation of gold particles on MnOx/C, along with the characterization, activity, and stability of Au/MnOx/C in CO oxidation. The promotional effect of MnOx is established.
========================
http://dx.doi.org/10.1016/j.jcat.2007.08.013
1. Introduction
Since Haruta and coworkers demonstrated that supported gold nanoparticles could be highly active for low-temperature CO oxidation [1], many different gold catalyst formulations have been reported. Deposition-precipitation, coprecipitation, and impregnation are common synthetic approaches utilized to prepare gold catalysts on a variety of supports including TiO2, Al2O3, Fe2O3, and CeO2 [2], [3], [4], [5] and [6]. Among the various combinations of supports and synthetic methods utilized, the deposition-precipitation of Au(OH)xCl4−x− complexes onto TiO2 leads to highly active Au/TiO2 that is the most studied gold catalyst in the literature [2], [3], [4], [5] and [6]. However, despite the wide use of SiO2 as a support for a variety of metal catalysts owing to its high surface area, thermal stability, mechanical strength, and non-reducibility [7] and [8], it is often deemed unsuitable for loading gold. Indeed, the activity of Au/SiO2 in CO oxidation is generally much lower than that of Au/TiO2 [9], [10], [11], [12], [13] and [14].
Several possibilities exist which can explain the failures in obtaining active Au/SiO2 catalysts. First, non-reducible and inherently “inert” SiO2 support does not supply reactive oxygen for CO oxidation. In contrast, TiO2, Fe2O3, and CeO2 supports are reducible, inherently “active”, and are thought to activate and store oxygen [14], [15], [16] and [17] U.R. Pillai and S. Deevi, Appl. Catal. A 299 (2006), p. 266. Article | PDF (496 K) | View Record in Scopus | Cited By in Scopus (17)[17]. Second, agglomeration of gold nanoparticles can more easily occur if the interaction between gold and SiO2 is inherently weak [6] and [13]. Third, failures due to the use of conventional preparation methods could inadvertently mask the real value of SiO2 as a support for gold particles. For instance, the shortcoming with using deposition-precipitation methods lies in the mismatch between the isoelectric point of SiO2 (IEP 2) and the pH range needed to sufficiently hydrolyze the HAuCl4 precursor to Au(OH)3 or Au(OH)4− (pH 8–10) [6] and [13]. Regardless of the reasons, many attempts have been made to prepare Au/SiO2 (mostly Au/mesoporous SiO2) via alternative methods [12], [14], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29] and [30], but the activity for CO oxidation (which is a sensitive probe reaction to compare the performance of gold catalysts [2], [3], [4], [5] and [6]) was either not reported [19], [20], [21], [22], [23], [24], [26], [29] and [30] or found to be very low [12], [14], [18], [25], [27] and [28].
The assumption that Au/SiO2 is not active for CO oxidation has been challenged since Okumura et al. reported that Au/SiO2 prepared via gas-phase grafting of dimethyl gold acetylacetonate exhibited high activity in CO oxidation [31] and [32]. Others showed that the grafting of alkylammonium [33] or aminosilane [34] onto mesoporous SiO2 could facilitate the interaction between the gold complex and the grafted SiO2 surface, thus resulting in active catalysts. Alternatively, gold particles capped with alkanethiol and alkoxysilane groups could polymerize with tetraethyl orthosilicate to form a metal-organic-inorganic composite active for CO oxidation after calcination [35]. We recently reported the preparation of highly active and stable Au/mesoporous SiO2 (Au/SBA-15) using Au(en)2Cl3 (en = ethylenediamine) as the precursor [36]. One of our key observations was that the catalytic activities of our Au/SBA-15 catalysts were highly dependent on the pH value of deposition solutions [36]. A similar deposition method was simultaneously developed by Zanella et al. to prepare gold particles supported on Aerosil fumed SiO2 [29]. They systematically studied the influence of solution pH value and adsorption time on gold loading and gold particle size of the resulting Au/SiO2 samples [29]. However, the catalytic activities of these Au/SiO2 samples were not investigated.
Because examples of highly active Au/SiO2 catalysts for CO oxidation are scarce [31], [32], [33], [34], [35] and [36], the current research extends the synthetic methods developed to produce Au/mesoporous SiO2 [36] to that of Au/Cab-O-Sil fumed SiO2, and surveys several important catalytic characteristics associated with the catalyst pretreatment, the effect of gold loading, post-treatments in acidic or basic media, catalyst deactivation, storage, regeneration, and the effect of metal oxide additives. All of these parameters were found to subtly influence the catalytic performance. Our results can furnish fresh perspective on the activation and promotion of Au/SiO2-based catalysts that have gone unaddressed, and provide new grounds for the following fundamental and applied research using such easy-to-synthesize and highly active Au/SiO2 catalysts.
- 