Au is a precious metal that retains its metallic state in nature as its spontaneous oxidation is thermodynamically unfavorable. Hence, Au has been regarded as a chemically inert metal. Accordingly, the classical uses of Au were in jewelry, coins, and ornaments, while the industrial use was in thin wires. Colloidal Au (particle diameter:~50-100 nm) imparts a red color to glass and is hence used in stained glass windows of cathedrals and churches. In 1987, Haruta and co-workers [1, 2] reported that Au nano clusters (Au NCs) with diameters of 2-5 nm deposited on certain metal oxides, such as Fe2O3, Co3O4, and NiO, were remarkably active for CO oxidation at temperatures as low as 200 K. Hutchings [3, 4] also showed the catalytic activity of cationic Au for the hydrochlorination of acetylene. Consequently, Au NC catalysts have attracted considerable attention in the field of catalysis [5].
Rossi et al. [6] demonstrated that small Au clusters showed an intrinsic catalytic activity for aerobic oxidation. In this reaction, bare Au clusters ephemerally generated in solution can act as catalysts for the oxidation of glucose. These results suggested that the small Au clusters confined in some aggregation inhibitors would act as oxidation catalysts. Therefore, colloidal Au would be an ideal model for oxidation reactions. In the case of colloidal Au, the Au NCs are partially stabilized by multipoint weak interactions with organic stabilizers, so that a part of the cluster surface is exposed and can contribute to the catalytic reaction [7]. Tsukuda and coworkers [8, 9] reported that Au NCs stabilized by poly(N-vinyl-2-pyrrolidone) (PVP, (C6H9ON)n), abbreviated as Au:PVP, could selectively oxidize p-hydroxybenzyl alcohol to the corresponding aldehyde in water. Following this report, Au:PVP systems were systematically investigated in terms of size-controlled synthesis, electronic structure and activity correlation, and applications in organic transformations. Thus, this article mainly summarizes our progress regarding the interplay of theoretical calculations and experiments on the reactions catalyzed by polymer-stabilized Au NPs.
The technique for preparing polymer-stabilized nano-sized metal clusters has been commonly used in the preparation of colloidal nanocluster catalysts. In 1970s, Toshima and coworkers [10] reported the use of fine colloidal dispersions of transition metals as catalysts for the selective hydrogenation of olefins and dienes. The basic concept underlying the preparation of polymer-stabilized nanoclusters is the reduction of metal ions in the presence of an organic polymer. The main role of the polymer is to inhibit the aggregation of neutral metal atoms at the initial stage to give nano sized metal clusters. The polymers on the Au NC prevent aggregation of the Au NCs and facilitate their dispersion in liquid media. PVP, polyvinylalcohol (PVA) and polyallylamine (PAA), as well as their derivatives, are used in the production of Au NCs. Oxygen, nitrogen, or sulfur-containing functional groups are usually introduced into the polymers to allow coordination to the Au NCs. In particular, PVP has been typically used for stabilizing NCs of precious metals such as Pt, Pd, and Rh [7, 11-15], and it has been proposed that the oxygen moiety of PVP interacts with the metal clusters [7]. A conventional synthesis of polymer-stabilized Au NCs, such as Au:PVP, entails rapid reduction of Au(III) ions with a strong reducing agent (NaBH4) in the presence of PVP [16, 17]. Tsukuda and coworkers [17] used a microfluidic reactor to create a homogeneous mixture of AuCl4- and BH4- solutions for the preparation of Au:PVP. The resulting Au:PVP catalysts were characterized by optical spectroscopy, transmission electron microscopy (TEM), etc. TEM observations of Au:PVP prepared by using the microfluidic reactor indicated that both the mean diameter and size distribution of the Au nanoclusters were smaller than those prepared by the conventional method.
To elucidate the size effect of the polymer-stabilized Au clusters on catalytic activity, a series of polymer-stabilized Au NCs were prepared by growing the small NCs as seeds. Additional AuCl4- ions were reduced in the presence of Au:PVP [9, 18, 19]. In this process, a weak reducing agent (SO32-) instead of NaBH4 was used for the reduction of Au(III) present on the Au cluster surfaces. A series of Au:PVP catalysts, with mean diameters ranging from 1.1 to 9.5 nm, were obtained [9].
The Au:PVP catalyst prepared from Au NCs with 1.3 nm diameter catalyzed the oxidation of several types of alcohols in water, as shown in Scheme 1. The key feature of these reactions is that they proceed only under aerobic conditions, which means that molecular oxygen dissolved in water acts as the oxidant [18].
Initially, these reactions did not proceed in the absence of a base, in contrast to those in the presence of Pd-or Pt-based catalysts [20-24]. For the first step of the aerobic oxidation of alcohols, the hydrogen elimination from the OH group of the alcohols must be needed to form the corresponding oxidation reaction products. However, Au NCs show the low catalytic activity for hydrogen elimination from the OH group of the alcohols adsorbed on Au:PVP. Thus, a base is necessary for the reaction to proceed, i.e., for the production of RO- and H+ from ROH in the liquid phase. A similar trend for behavior of the hydrogen elimination from an allylic alcohol on a Au NC supported catalyst was elucidated theoretically [25, 26].
The effects of cluster size on the catalytic activity for the aerobic oxidation of p-hydroxybenzyl alcohol have examined (Fig. 1). This reaction is commonly used as a test reaction because the clusters do not aggregate throughout the reaction and hydroxybenzaldehyde is selectively obtained. The turnover frequency (TOF) of this reaction was found to appear below 5 nm, and it increased dramatically with a decrease in the size of the Au NCs (Fig. 1) [9, 19].
For a rational explanation of the aerobic oxidation of alcohols on polymer-stabilized Au NC catalysts, it is considered that the active molecular oxygen is present on the catalysts. In order to investigate the possibility of activation of molecular oxygen, several experiments were conducted. The results of XPS measurements [19] revealed that the Au NCs were negatively charged. Theoretical studies also demonstrated that Au13 clusters acquired a negative charge because of electron donation from the carbonyl group of the PVP units to Au13 [27]. The gross negative charge of Au13 increased with an increase in the number of PVP units. O2- is produced on Au13 protected with four PVP units. These results indicated that PVP not only worked as a protecting material to prevent the aggregation of Au NCs, but also affected the characteristics of the Au NC surface. Further, O2 is activated to form superoxo-like species (O2-) on the negatively charged Au clusters protected by PVP to promote the oxidation. The importance of the negatively charged Au atoms in the clusters are also confirmed by the Ag-doping on the catalytic activity of polymer stabilized Au clusters in aerobic oxidation of alcohol experimentally [28]. In this case, the doped Ag atom also acts as an electron donor to Au atoms in the cluster. In addition, it has been reported that the presence of water enhances O2 activation on the negatively charged Au10 cluster [29]. Therefore, water is beneficial for Au:PVP catalysis. Fig. 2 shows a plausible mechanism for alcohol oxidation in the presence of Au:PVP [30]. Theoretical calculations were carried out in order to investigate the mechanism of the aerobic oxidation of the alcohol over polymer-stabilized Au NC catalysts. The results suggested that the production of p-hydroxybenzaldehyde was triggered via both C-H bond dissociation of 1-position carbon of p-hydroxybenzyl alcohol by the oxygenated Au NC and hydride elimination from p-hydroxybenzyl alcohol on the Au NC surface. Accordingly, H2O2 was produced in all the reaction pathways investigated. Finally, H2O2 decomposed to afford two OH- ions. Ehara and coworkers [31, 32] also showed theoretical investigations for the aerobic oxidation of methanol to formic acid, catalyzed by Au8- and Au20-. From these research results, the fundamental mechanism of oxidation was elucidated.
It is well known that the catalytic activities of Au supported catalysts, such as Au/TiO2, Au/Al2O3, change depending on the support chosen [33]. In the case of the polymer-stabilized Au NC catalysts, the selection of the polymers is a crucial step. For example, Tsukuda et al. [19] reported that the Au NCs stabilized by poly(allylamine) (PAA; (C3H5NH3)n), abbreviated as Au:PAA, could also selectively oxidize p-hydroxybenzyl alcohol to the corresponding aldehyde under the condition used for Au:PVP; however, its activity is much lower than that of Au:PVP. To elucidate the reason for this difference, Okumura and coworkers [35] investigated the characteristics of these two types of polymer-stabilized Au NC catalysts using density functional theory (DFT) and molecular dynamics (MD) calculations. From the DFT calculations, it was found that the O2 activation abilities for both Au:PVP and Au:PAA were comparable, while the adsorption energies between the Au NC and the polymers were slightly different. Next, the surface coverage of the Au NC by the stabilizing polymers was investigated by MD calculations. From the trajectories of the MD simulations, the radial occupancy distribution of the polymers in the polymer-stabilized Au NC model systems which was closely related to the cluster surface coverage was calculated based on the van der Waals (vdW) radii of the atoms constituting the model systems [35]. The results are summarized in Fig. 3. It appears that the surface coverage in the vicinity of the Au NC by PAA is larger than that of the Au NC by PVP. From these results, it could be concluded that the difference in the catalytic activities of Au:PVP and Au:PAA is mainly due to the difference in substrate accessibility on the Au NC surface, which in turn is caused by the difference in the Au NC surface coverage by the polymers, while O2 activation abilities for both Au:PVP and Au:PAA are almost same. These results suggest that essential roles for the polymer used for the polymer-stabilized Au NCs are the inhibition of Au NCs aggregation, the electron donation to Au NCs, and less strong passivation of Au NC surface by the polymer adsorption. Consequently, both the essential nature of the stabilizing polymers and the mechanism of aerobic oxidation of alcohols by polymer stabilized Au NCs were elucidated in detail by both theoretical and experimental investigations.
The polymer stabilized Au NCs in aqueous solution catalyze the oxidation of alcohols into the corresponding acids under basic conditions. This suggests that aldehydes can also be transformed into acids under similar conditions. In particular, the aerobic oxidation of glucose to gluconic acid which is adopted in the pharmaceutical and food industries is an important reaction. There is a strong demand for the use of metal catalysts to achieve large-scale productions of gluconic acid, although biochemical transformations are employed in practice. From this point of view, Au has considerable potential for use as a catalyst because of its high activity for the aerobic oxidation of the alcohols. Haruta and co-workers have developed Au catalysts on various organic supports, such as cellulose [36] or ion-exchange resins [37], the TOF values of which exceed 32000 h-1.
Toshima and co-workers [38] attempted to tune the catalytic performance of PVP-protected Au NPs by introducing additional metals for the aerobic oxidation of glucose. Generally, Au exhibits superior catalytic activity than other noble metals such as Pd and Pt. However, Au-Ag (9/1), Au-Pt (8/2) bimetallic, and Au-Pt-Ag (7/2/1) trimetallic NPs showed much higher catalytic activities (TOF of 14000 molglucose molmetal-1 h-1 for bimetallic [38] and 20000 molglucose molmetal-1 h-1 for trimetallic NPs [39], respectively) than monometallic Au NPs (TOF: 4000 molglucose molAu-1h-1), although these NPs had similar mean diameters ranging from 1.4 to 1.6 nm. Toshima and co-workers [40] also prepared Au-Ag bimetallic NPs by physical mixing of colloidal dispersions of Au and Ag monometallic NPs. In addition, core-shell structured core(Ag)-shell(Au) (2/8) NPs showed an increased TOF of 16900 molglucose molmetal-1 h-1 [41]. This result was explained by an electron transfer from the Ag core to the Au shell, which increased the electron density of the Au shell, making it suitable for O2 activation. DFT calculations performed using an Ag43Au12 cluster supported the fact that Au atoms were negatively charged, which was in a good agreement with the XPS results [42, 43]. Further, Toshima and coworkers [44-46] reduced the amount of Au in inverse Pd core-Au shell bimetallic NPs, and succeeded to obtain crown-jewel clusters, where Au atoms were located at the top positions of a Pd147 cluster (Fig. 4). This synthetic strategy is based on the difference in the surface free energies of Au and Pd, and the replacement of Pd with Au occurs upon the introduction of Au(III) ions onto the Pd147 clusters. The obtained crown-jewel clusters showed an extremely high TOF of 195000 molglucose molAu-1 h-1, exceeding that of Au/ZrO2 [44]. This value was much larger than other cluster catalysts. Since the catalytic activity was decreased with an increase in Au loading, it could be presumed that the activities of the Au atoms positioned at other sites, such as the edge sites, were inferior to that of the top Au atoms for glucose oxidation [45]. DFT studies using a Pd43Au12 cluster, in which all the top atoms were replaced with Au, showed that the negative charge density of the top atoms significantly increased as compared to that in the case of a Pd55 cluster, indicating electron transfer from the Pd atoms to the top Au atoms. The negatively charged top Au atoms are responsible for the glucose oxidation. DFT results also revealed that the adsorption energy of one Au atom on the Pd(111) facet was larger than that on Pd(100), suggesting that Au was preferentially replaced at the top of the Pd(111) facets in a Pd55 cluster [46].
The polymer-stabilized Au and Au containing NCs catalyze numerous interesting reactions [47-50]. In particular, polymer-stabilized Au-Pd bimetalic cluster catalysts also play an important role in C-C bond formation reaction. Dhital et al. [51] and Boekfa et al. [52] showed the unique catalytic activity of Au-Pd alloy NCs for Ullmann coupling of chloroarenes in aqueous solution at low temperature. The Ullmann coupling product, 4, 4’-dimethylbiphenyl, was not obtained when monometallic Au:PVP or Pd:PVP was used as the catalyst. Furthermore, this reaction could not be realized using a physical mixture of monometallic Au and Pd NCs. Consequently, it could be concluded that the Au-Pd bimetalic NC was essential for the generation of Ullman coupling of chloroarenes on the cluster surface. The calculation results also confirmed that the Au10Pd10 NC surrounded by four n-ethylpyrrolidone molecules could activate the substrate as electron donors and stabilize it.
We showed that the polymer-stabilized Au and Au containing NC catalysts exhibited unique catalytic activities in various types of reactions. In each case, the active sites are the surfaces of small Au NC and Au containing NCs. Therefore, it was found that the characteristics of the polymer-stabilized Au NC catalysts were different from those of conventional metal oxide supported Au NC and Au containing NC catalysts. This is mainly due to the result of the electron donation from PVP to the Au NCs. These results suggested that we could control the reactivity of the polymer-stabilized Au and Au containing NCs by appropriate design of the polymers used. Although there are unsolved problems related to the polymer-stabilized Au NC catalysts, we are convinced that the polymer-stabilized Au NCs would show features different form those of heterogeneous catalysts, owing to the interplay between theoretical and experimental works. In the near future, the interplay between the highly organized model catalysts experiments and the more reliable theoretical calculations is expected to be useful for elucidating the activities and characteristics of the polymer-stabilized Au NC and Au containing NC catalysts.
2016,
Vol. 37 Issue (10): 1588-1593 
PDF
2016,
Vol. 37 Issue (10): 1588-1593 PDF