Graphene hydrogel supported palladium nanoparticles as an efficient and reusable heterogeneous catalysts in the transfer hydrogenation of nitroarenes using ammonia borane as a hydrogen source

Introduction

Amines are valuable building blocks for manufacturing of pharmaceuticals, pigments, agrochemicals and numerous types of other chemicals [1]. Among the amine family, anilines occupy a special place because they are important raw materials in various fields with the production over 4000000 tonnes per year [2]. Due to their industrial importance, many methods have been developed for the preparation of amines. However, in synthetic organic chemistry, the reduction of nitroarenes is the common route for the synthesis of aniline derivatives [3, 4]. In this route, classical hydrogenation reactions catalyzed by a noble metal catalyst are generally applied to yield desired aniline derivative [5]. However, this methodology have several well-known safety and technical handicaps such as the need of high pressure supplied by a molecular hydrogen tube in the laboratory at high temperature and the hardly dissolution of hydrogen gas in the solution [6]. Additionally, over-hydrogenation products are generally observed by this type of reduction process [7]. Therefore, the scientists have recently focused on development of a greener yet efficient method for the large scale synthesis of anilines. In this respect, examples of the reusable and recyclable heterogeneous catalysts in water-like green solvents for the synthesis of anilines are on the rise every day [8]. In these eco-friendly reduction protocols, the role of heterogeneous catalyst is key for both efficiency and selectivity of the system. In this regard, transition metal-based nanoparticles (TMNPs) have recently attracted increase attention owing to their exclusive physical and chemical properties compared to the bulk heterogeneous catalysts [9]. Moreover, the TMNPs provide higher catalytic performance and selectivity than their bulk analogs owing to their high surface to volume ratio and more energetic surface atoms [10]. However, the TMNPs are kinetically unstable to agglomerate into bulk metals and therefore they have to be stabilized against to agglomeration by using surfactants either electrostatically or sterically [11]. Upon stabilization of TMNPs, a certain amount of the catalytically active surface atoms are blocked by the surfactants, which lowers their catalytic activity and selectivity. To eliminate these problems, TMNPs can be supported on a high surface are materials. The selection of the support material is very important because the TMNPs-support material interaction affects the catalytic performance dramatically. Therefore, hitherto a variety of support materials including commercially available carbon blacks [12], carbon nanotubes [13], alumina [14], zeolites [15], polymers [16] and graphitic carbon nitride [17] have been used for supporting various metal nanoparticles. In recent years, reduced graphene oxide (rGO) have been reported as a versatile support for the TMNPs owing their advantageous properties [18]. However, the 2D layered structure of rGO sometimes creates stability problem for the TMNPs in liquid phase catalysis because the considerable π–π stacking interactions between rGO nanosheets, which results in aggregation and restacking of graphene nanosheets, and following serious decrease their specific surface area and block active catalytic sites of TMNPs on graphene [19]. In this respect, after the pioneer work of Shi et al. [20], the self-assembled graphene hydrogel (GHJ) has recently appeared as an interesting support material for the TMNPs [21, 22] because it both comprises self-assembled rGO networks and water besides having large surface area with porous structure. Moreover, GHJ is a three-dimensional (3D) material eliminating the problems associated with 2D rGO nanosheets and considered to be a different support materials compared to conventional carbon-based materials. Besides these advantageous properties, the GHJ–TMNPs nanocomposites can easily be prepared by using one-pot hydrothermal approach comprising concurrently formation of GHJ from graphene oxide and TMNPs from the reduction of metal precursor. Up to date, GHJ supported Au [23], Ag [24], Cu [25] and Pt NPs [26] have been reported but to the best our knowledge there is no report on one-pot synthesis Pd NPs supported on GHJ for catalytic applications.

In this study, we report for the first time a facile one-pot synthesis of Pd NPs supported on graphene hydrogel (Pd-GHJ) as a reusable and recyclable heterogeneous catalysts for the transfer hydrogenation of nitroarenes using ammonia borane (AB) as a hydrogen source at room temperature in methanol/water mixture. It should be noted that the presented synthesis of Pd NPs supported on GHJ and their application in the transfer hydrogenation reaction using AB as a hydrogen storage material are the first in the literature.

Results and discussion

Pd NPs supported on GHJ (Pd-GHJ) were synthesized by using a novel one-pot synthetic protocol comprising the combination of hydrothermal treatment and polyol reduction methods in water. In this protocol, Shi’s [20] pioneer hydrothermal treatment protocol was applied to the graphene oxide (GO)-potassium tetrachloropalladate(II) (K₂PdCl₄) mixture in aqueous dispersion. Three consecutive reactions, namely (i) the reduction of GO into the rGO nanosheets, (ii) the self-assembly of rGO nanosheets into the GHJ and (iii) the reduction of [PdCl₄]⁻² complex by ethylene glycol under hydrothermal conditions, were occurred simultaneously. Figure 1a shows a representative TEM image of Pd-GHJ nanocomposites clearly indicating the formation of well-dispersed Pd NPs with an average particle diameter of 10 nm over the GHJ nanosheets. It should be noted that there was no agglomerated Pd NPs observed by the examination of almost entire TEM grid. Figure 1b shows a HR-TEM image of a single Pd NP on GHJ. It indicates the formation of highly crystalline face centered cubic (fcc) Pd NPs with a measured adjacent lattice fringe distance of 0.23 nm corresponds well to the (111) lattice spacing of the fcc Pd (JCPDS card, file no. 05-0681) [27].

Fig. 1:

(a) TEM (b) HR-TEM image of Pd-GHJ nanocomposites.

The crystal structure of Pd-GHJ nanocomposites was further studied by powder XRD and the typical PXRD pattern was depicted in Fig. 2. In the XRD pattern, all of the peaks can readily be indexed to fcc-Pd (JCPDS card, 05-0681) and no characteristic peaks were observed for neither K₂PdCl₄ precursor nor PdO, which indicates the formation of phase purity and high crystallinity of the as-prepared Pd NPs over GHJ. Besides the peaks belong to fcc-Pd, another peak appeared at 2θ=24.3° is attributed to the GHJ support [20]. In order to determine the Pd content of Pd-GHJ nanocomposites for the catalytic applications, inductively coupled plasma mass spectroscopy (ICP-MS) analysis were performed on several batch of the samples and Pd content was found to be 6.3 wt.%. This value was used to determine the loading ratio of Pd-GHJ nanocomposites for the catalytic applications (vide infra).

Fig. 2:

XRD pattern of Pd-GHJ nanocomposites.

X-ray photoelectron spectroscopy (XPS) is one of the indispensable tools for the characterization of surface properties of metal NPs, especially the ones to be used for catalytic applications. In this regard, we studied XPS for the Pd-GHJ nanocomposites. Figure 3 shows both survey and high-resolution XP spectra for Pd3d core-level of Pd-GHJ nanocomposites. All the expected elements C, O, Pd are available in the XP survey spectrum (Fig. 3a). In the case of high-resolution XP spectrum for Pd3d core-level (Fig. 3b), two major assymetric-shape peaks appeared at binding energies of 335.1 and 340.4 eV with a spin-orbit component (Δ=5.3 eV) that are readily assigned to metallic Pd. Moreover, there is no characteristic peaks observed for neither K₂PdCl₄ precursor nor PdO indicating the successful reduction of Pd precursor into the Pd(0) by the presented synthesis protocol.

Fig. 3:

(a) Survey (b) Pd3d core-level high-resolution XP spectra of Pd-GHJ nanocomposites.

Raman spectroscopy gives important information about the carbon-based nanostructures including graphene-based materials. As the presented nanocomposite includes GHJ as a support material and it is prepared hydrothermally starting from GO, we studied the structure of Pd-GHJ nanocomposites, GO and rGO with Raman spectroscopy. Figure 4 shows a typical Raman spectrum for the Pd-GHJ nanocomposites, GO and rGO. For the Pd-GHJ spectrum, two major peaks are observable at 1342 and 1585 cm⁻¹ that are attributed to D and G bands of GHJ, respectively. Compared to the Raman spectra of GO and rGO, the position and relative intensities of D and G bands are different. Moreover, the intensity ratio of D/G (1.07) is higher than those of GO (1.03) and rGO (0.693). These results indicates the successful conversion of GO into GHJ upon the hydrothermal treatment.

Fig. 4:

Raman spectra of Pd-GHJ (a) GO (b) and rGO (c).

As the Pd-GHJ nanocomposites are planned to be used as catalysts, the analysis of their surface area is an important criteria. The surface area and pore analysis of Pd-GHJ were done by studying N₂ adsorption-desorption isotherm (Fig. 5). Pd-GHJ shows a type-IV isotherm indicating a mesoporous structure of the material with BET surface area of 112.5 m²/g. This value is slightly than the pristine GHJ (135 m²/g) which is due to the Pd NPs supported on the GHJ.

Fig. 5:

N₂ adsorption-desorption isotherm of Pd-GHJ nanocomposites.

After the catalyst preparation and characterization, we turned our attention to the application of the catalyst in a reduction of nitroarenes. Firstly, the reaction conditions were optimized by studying various reaction parameters including solvent type, AB amount, and reaction time over nitrobenzene as a model compound. The reaction optimization data are summarized in Table 1. As a result of examining the effect of AB amounts, it was found that the best yield was obtained by using 1 mmol of AB (Table 1, entries 1–3). It should be noted that the reaction did not occur without catalyst (Table 1, entry 4) and also using ethanol instead of methanol resulted in reduced product yield (Table 1, entry 5). As can be seen from the results depicted in Table 1, the hydrogenation time less than 20 min was not enough for the full conversion of the substrate (Table 1, entry 6). As a summary of the optimization experiments, it was concluded that the optimized reaction conditions was obtained by using aqueous methanol mixture (H₂O/MeOH, v/v=7/3), 1 mmol of AB and 15 mg Pd-GHJ catalyst in 20 min reaction time (Table 1, entry 3).

With the optimized reaction conditions in hand, the Pd-GHJ catalyzed transfer hydrogenation reactions were applied over various substituted nitrobenzenes (total 9 examples) and the results were depicted in Table 2. As clearly be resulted by the Table 2, all the tested nitroarenes were efficiently converted to the related nitrobenzene derivatives with the yields reaching up to ≥99 within only 20 min reaction times at room temperature. As the first example, the catalytic transfer hydrogenation reaction proceeded successfully on nitrobenzene and aniline was yielded 99% in 20 min (Table 2, entry 1). Except for three reactions (Table 2, entries 2, 5 and 8), all other reactions were completed within 20 min reaction times. Moreover, several electron withdrawing and electron donating nitroarene derivatives have been also successfully reduced by the presented hydrogenation protocol, indicating the high substituent tolerance of the presented catalytic transfer hydrogenation protocol. On the other hand, the presented Pd-GHJ catalyzed transfer hydrogenation showed selectivity towards the reduction of only –NO₂ group on the aromatic ring bearing another reducible group such as –Cl and –CN (Table 2, entry 8, 9). In this respect, p-chloroaniline and p-aminobenzonitrile yielded by 81% and ≥99%, respectively, with the presented catalytic transfer hydrogenation protocol.

Table 2:

Substrate scope of Pd-GHJ catalyzed tandem dehydrogenation of AB and hydrogenation of nitroarenes.^a

Entry	Substrate	Product	t (min)	Yield (%)b
1			20	≥99
2			20	95
3			20	≥99
4			20	≥99
5			20	77
6			20	≥99
7			20	≥99
8			20	81
9			20	≥99

1. ^aReaction conditions: 0.35 mmol nitro compound, 1 mmol NH₃BH₃, 15 mg of Pd-GHJ nanocomposites 5 mL of water/methanol mixture (v/v=7/3) and room temperature. ^b1H-NMR yield.

After presenting applicability of the Pd-GHJ nanocomposites catalyzed tandem dehydrogenation of AB and hydrogenation reaction for the reduction of a variety of nitroarenes, we then studied the reusability of Pd-GHJ nanocomposites, which is another important criteria for showing the practicability of a heterogeneous catalyst. Figure 6a shows a five runs reusability tests of Pd-GHJ nanocomposites in the transfer hydrogenation of nitrobenzene as a model compound. As can be easily concluded by the reusability test, Pd-GHJ nanocomposites provided 99% yield after five consecutive runs in the transfer hydrogenation of nitrobenzene, indicating their high durability. Moreover, we analyzed the morphology and the structure Pd-GHJ nanocomposites after the reusability test by using TEM and XRD. As clearly be seen by the TEM image given in Fig. 6b, Pd NPs preserved almost their initial morphology and particles size over the GHJ without any significant agglomeration. Besides morphology preserving, there is no change on the crystal structure of Pd NPs and GHJ after the catalysis, which is concluded by the XRD pattern given in Fig. 6c.

Fig. 6:

(a) The reusability test of Pd-GHJ nanocomposites in the tandem dehydrogenation of AB and hydrogenation of nitrobenzene (b) a representative TEM image and (c) XRD pattern of the Pd-GHJ nanocomposites after the reusability test.

Conclusions

In summary, we reported a facile synthesis and structural characterization of Pd-GHJ nanocomposites and their catalysis in the tandem dehydrogenation of AB and hydrogenation of nitroarenes. Pd-GHJ nanocomposites were synthesized for the first time by one-pot protocol including the combination of hydrothermal treatment and polyol reduction in aqueous solution. The stable and highly-dispersed 10 nm Pd NPs were generated on GHJ by the presented synthesis protocol. Then, Pd-GHJ nanocomposites as a novel and reusable catalysts showed high performance in the reduction of various nitroarenes with the yields reaching up to ≥99% and high substituent tolerance. Moreover, the Pd-GHJ catalyzed transfer hydrogenation protocol provided a selectivity towards the reduction of only –NO₂ group on the aromatic ring bearing other reducible groups such as –Cl and –CN. We believe that this work will open a new perspective for the development of more environmental and efficient catalytic strategies in the organic synthesis for the reduction of other functional groups in the organic synthesis via a transfer hydrogenation.

Experimental