Au@Pt Nanotubes within CoZn-Based Metal-Organic Framework for Highly Efficient Semi-hydrogenation of Acetylene.

Designing nanocatalysts with synergetic functional component is a desirable strategy to achieve both high activity and selectivity for industrially important hydrogenation reaction. Herein, we fabricated a core-shell hollow Au@Pt NTs@ZIFs (ZIF, zeolitic imidazolate framework; NT, nanotube) nanocomposite as highly efficient catalysts for semi-hydrogenation of acetylene. Hollow Au@Pt NTs were synthesized by epitaxial growth of Pt shell on Au nanorods followed with oxidative etching of Au@Pt nanorod. The obtained hollow Au@Pt NTs were then homogeneously encapsulated within ZIFs through in situ crystallization. By combining the high activity of bimetallic nanotube and gas enrichment property of porous metal-organic frameworks, hollow Au@Pt NT@ZIF catalyst was demonstrated to show superior catalytic performance for the semi-hydrogenation of acetylene, in terms of both selectivity and activity, over those of monometallic Au and solid bimetal nanorod@ZIF counterparts. This catalysts design idea is believed to be inspirable for the development of highly efficient nanocomposite catalysts.

Introduction

Metal-organic frameworks (MOFs), constructed by the self-assembly of metal ions and organic linkers in an appropriate solvent (Altaf et al., 2018, Cook et al., 2013, Tsuruoka et al., 2018, Yaghi et al., 2003), have permanent microporous structures and are typically characterized by extraordinarily large surface areas, well-defined pore structures, and tailorable chemistries (Corma et al., 2010, Farrusseng et al., 2009, Feng et al., 2015, Lee et al., 2009, Zhang et al., 2018). These properties make MOFs potential candidates for various applications such as gas storage and separation (Alezi et al., 2015, Wang and Yuan, 2014; Furukawa et al., 2013, Suh et al., 2012), ion exchange (Gao et al., 2014), sensing (Li et al., 2018, Ma et al., 2012), drug delivery (Wu et al., 2014), and catalysis (Feng et al., 2012, Liu et al., 2014). In particular, the combination of functional guest species and MOFs can impart new functionalities owing to the guest-host synergistic effect (Dhakshinamoorthy and Garcia, 2012). Fabricating multicomponent nanostructures either by embedding nanoparticles (NPs) in MOF matrices (Jang et al., 2017, Zhao et al., 2018, Zlotea et al., 2010) or incorporating preformed NPs into MOFs (Buso et al., 2011, Lohe et al., 2011, Ma et al., 2018) is an efficient route for extending MOF applicability. In the latter case, preformed NPs with diverse sizes and structures can be encapsulated into MOFs to generate core-shell nanocomposites. Although a few researchers have successfully synthesized NP@MOF composites through constructing MOFs on the surface of functional NPs (Du et al., 2017, He et al., 2013, Shang et al., 2016, Sugikawa et al., 2013, Zhao et al., 2014), effective control over the dispersibility of NPs within the MOFs as well as the morphology and size of the shell materials remains significant challenges. For example, it is difficult to encapsulate NPs with random shapes and compositions owing to the competition between the homogeneous nucleation and the heterogeneous growth of MOFs on the NP surface. In the present work, we attempted to synthesize a core-shell nanocomposite, in which preformed NPs with well-defined morphologies were entirely confined within the MOF layers, as advanced nanocatalysts for selective hydrogenation of acetylene.

Ethylene is the basic raw material in the industrial manufacturing of polymers and fine chemicals. The catalytic semi-hydrogenation of acetylene, C2H2 + H2 → C2H4, is one of the most important reactions in the ethylene industry, since even trace amount of C2H2 in C2H4 can deactivate the downstream Ziegler-Natta catalyst for further polymerization. The main challenge of this reaction is to selectively hydrogenate acetylene into ethylene and meanwhile prevent the excess reduction of ethylene to ethane (Wang et al., 2019, Yang et al., 2015). In this process, metal NPs play a very important role in controlling the overall catalytic performance. Recently, Au nanomaterials have been found to be suitable catalysts for selective hydrogenation of acetylene (Sun et al., 2019). Apart from monometallic Au, bimetallic nanostructures have also attracted significant attention owing to their synergistic effects improving the physical and chemical properties of their monometallic counterpart (Yasukawa et al., 2012, Zhang et al., 2019). Given the enhanced properties of bimetallic nanostructures, controlling the synthesis of noble metal nanocrystals with well-defined morphology and size becomes very significant. Hollow Pt nanotubes (NTs) reportedly exhibited a much better catalytic activity than Au/Pt nanorods (NRs) and 5 nm Pt/C, which can be attributed to the unique tubular characteristic of the Pt nanostructure (Huang et al., 2016, Peng et al., 2010). However, Au NRs, Au@Pt NRs, and Au@Pt NTs were prone to agglomerate. A suitable carrier is important to maintain its morphology and dispersion for preventing its agglomeration and inactivation. Furthermore, inspired by the work wherein the encapsulation of Au NRs into porous MOFs led to enhanced stability and activity in applications such as catalysis or sensing (Chen et al., 2014, Zheng et al., 2016), we designed hollow Au@Pt NTs through etching interior Au of Au@Pt NRs and tried to encapsulate these NPs into zeolitic imidazolate framework (ZIF) nanocrystals in a core-shell-type configuration, forming NPs@ZIFs core-shell structures to maintain the morphology and dispersion of NPs and avoid agglomeration.

ZIFs were chosen as the host material owing to their porosity, high surface area, and relatively high structural robustness. The porous-robust ZIF-67 or ZIF-8 can be synthesized at mild temperature (room temperature), which is suitable for the supporter. Au NRs, Au@Pt NRs, and Au@Pt NTs were encapsulated into ZIFs forming well-dispersed NPs@ZIFs core-shell structures. Here, we report a facile and general encapsulation strategy to synthesize catalysts that incorporates different structures of NPs into ZIFs (Figure 1).

Figure 1

Schematic Illustration of the Formation of Au, Au@Pt NR, and Hollow Au@Pt NTs and Their ZIF Nanocomposites

ZIF-67 is constructed by the assembly of cobalt ions and methyl imidazole. It has the same structure as that of ZIF-8, with sod topology, in which a large sod cage (11.6 Å) is accessible through a narrow six-ring pore (3.4 Å). By combining the catalytic properties of metal NPs with the large internal surface area, tunable crystal porosity, and unique chemical properties of ZIFs, the hydrogenation of acetylene was studied to elucidate the role of encapsulated nano-sized metal structures in catalytic activities.

Results and Discussion

Au NRs with an average length of 53.5 nm (Figure 2A) were prepared by a previously reported procedure (Feng et al., 2008). Au@Pt NRs and hollow Au@Pt NTs were synthesized according to a previously reported method (Lee et al., 2016), with some modifications to obtain NPs with desired structures (Figures 2D and 2G). The lengths of Au@Pt NRs and hollow Au@Pt NTs were 56.2 and 52.0 nm, respectively. The surfaces of these NPs were functionalized with polyvinylpyrrolidone (PVP) after synthesis. The ZIF incorporation process was conducted with Au NRs, Au@Pt NRs and hollow Au@Pt NTs, respectively. In a typical procedure, cobalt nitrate hexahydrate (291 mg) and 2-methylimidazole (369.5 mg) were each dissolved in methanol (25 mL). Afterward, the ligand solution was poured into the pink solution containing Co2+ ions. After allowing the mixture to stand for 5 min, a PVP-Au@Pt NT solution, concentrated to 2 mL in methanol by centrifugation (12,000 rpm, 5 min), was injected into the above solution.

Figure 2

TEM Images of Different Samples

(A) Au NRs, (B) Au NRs@ZIF-67, (C) Au NRs@ZIF-8, (D) Au@Pt NRs, (E) Au@Pt NRs@ZIF-67, (F) Au@Pt NRs@ZIF-8, (G) Au@Pt NTs, (H) Au@Pt NTs@ZIF-67, and (I) Au@Pt NTs@ZIF-8.

Thereafter, the resulting mixture was allowed to rest at room temperature for 24 h without mechanical perturbation. For this synthesis of NPs@ZIF-67 with well-defined morphology and size, the amount of PVP was a key factor. The encapsulation procedure is based on the successive adsorption of PVP-modified NPs on the surface of the growing ZIF-67 crystals until the NPs are exhausted (Du et al., 2017). According to the investigation of the adsorption of amphiphilic PVP on solid surfaces, the polar and apolar groups in PVP are believed to promote the adsorption process (Mdluli et al., 2011). It can be seen that, without PVP, severe aggregation of Au@Pt NTs occurred during the preparation of the core-shell structure (Figure S13A). However, when excessive PVP was used in the synthesis, ZIF-67 rhombic dodecahedral shells of different sizes were observed, followed by agglomeration (Figure S13D), which can be attributed to the binding of PVP on the crystal surface. Thus, an appropriate amount of PVP is required to guarantee the surface functionalization of metal NPs and maintain the morphology of the MOFs. Scanning electron microscopy (SEM) images showed that Au@Pt NTs@ZIF-67 crystals with a uniform rhombic dodecahedral shape had a narrow size distribution with an average size of 767.9 nm (Figure S11). Transmission electron microscopy (TEM) images demonstrated that each ZIF-67 crystal contained multiple Au@Pt NTs that were fully confined within the MOF. Freestanding metal NTs were not observed (Figure 2H). The distance between the NTs varied from 10 to 450 nm, suggesting that Au@Pt NTs were well isolated from each other during the assembly process.

There were no remarkable deviations in the morphology and size among the encapsulated Au@Pt nanotubes. The size distribution of Au@Pt NTs@ZIF-67 nanocomposites ranged from 600 to 1,000 nm. To gain a better understanding of the formation of multi-core-shell structures, the encapsulation process was then investigated by conducting time-dependent experiments. Upon reacting for 30 min, the obtained products were spherical multi-Au@Pt NTs-ZIF-67 hybrids that possessed an average diameter of 600 nm. The spherical shells grew larger with an average size of 750 nm upon further extending the incubation time. After 8 h, the encapsulation process was nearly completed, and the hybrid spheres gradually evolved into rhombic dodecahedral shapes (Figure S14). The same strategy could be applied to encapsulate other nanostructured particles with different compositions (Figures 2B and 2E). Similarly, ZIF-8 can be selected as the host for embedding metal NPs. When Au@Pt nanotubes were used as seeds in the procedure, TEM revealed (Figure 2I) that the nanocomposites were of the same core-shell structure as Au@Pt NTs@ZIF-67, in which multiple Au@Pt nanotubes, ∼51.9 nm in length, were coated with a ZIF-8 shell having a thickness of ∼250 nm (Figure S12). The incorporated Au@Pt NTs maintained their integrity in the presence of a hollow interior, but the crystals were observed to be multilayered (Figure S12).

The crystal structure analysis of ZIF-67 and ZIF-8 composites from powder X-ray diffraction (XRD) patterns indicated that the incorporation of metal nanoparticles did not affect the crystalline structure of MOF. The two additional peaks originated from the presence of Au and Pt (Figures S16 and S17). Energy-dispersive X-ray spectroscopy (EDS) elemental mapping of Au@Pt NTs@ZIF-67 further revealed that Au and Pt were located only in the hollow Au@Pt nanotubes, whereas C and Co were homogeneously distributed throughout the entire nanoparticle assembly (Figure 3). The obvious double peaks in the elemental line scan along the radial direction also confirmed the hollow nanostructure feature of Au@Pt NTs, and the content of Pt was more than that of Au (Figure 3H). After these particles were exposed to the acetylene reduction reaction, their powder XRD patterns confirmed the stability of MOFs (Figure S19). The Brunauer-Emmett-Teller (BET) results showed that the surface areas of Au@Pt NTs@ZIF-67 and Au@Pt NTs@ZIF-8 were 1,543.6 and 888.5 m2g−1, respectively. These values are less than those of ZIF-67 and ZIF-8 possibly because the nanoparticles occupied a part of the micropores in the MOFs.

Figure 3

High-Angle Annular Dark-Field Scanning TEM Images and EDS Elemental Mapping Images

High-angle annular dark-field scanning TEM images of (A) Au@Pt NTs@ZIF-67 and (G) Au@Pt NTs. EDS elemental mapping of (B–F) Au@Pt NTs@ZIF-67 marked in (A) and (I) Au@Pt NT marked in (A). EDS line scan of (H) Au@Pt NTs.

The hydrogenation reaction of acetylene, conducted in a fixed-bed reactor, was used as a probe reaction to compare the catalytic activities of these different samples.

Gas component concentrations were analyzed at specific times by gas chromatography (GC). The catalytic efficiencies of ZIFs, Au NRs@ZIFs, Au@Pt NRs@ZIFs, and Au@Pt NTs@ZIFs are summarized in Figure 4. At an operating temperature of 175°C, gas flowrate of 40 mL min−1, and catalyst dosage of 50 mg, the acetylene conversion was 12.9% for Au NRs@ZIF-67, 43.5% for Au@Pt NRs@ZIF-67, and 69.1% for Au@Pt NTs@ZIF-67; and ethylene yields were 9.1% for Au NRs@ZIF-67, 35.1% for Au@Pt NRs@ZIF-67, and 49.3% for Au@Pt NTs@ZIF-67, respectively. The above results reveal that Au@Pt NRs@ZIF-67 showed better catalytic performance than Au NRs@ZIF-67, which can be ascribed to the highly active Pt nanoparticles loaded on the Au NRs. More interestingly, hollow Au@Pt NTs@ZIF-67 exhibited the best catalytic activity with slightly compromised selectivity, owing to the unique tubular structure that allowed the existence of more active sites on the surface of the Au@Pt nanotube wall. Comparing the conversion of the catalysts with different nanoparticles, the bimetallic Au@Pt NTs or Au@Pt NRs in ZIFs greatly facilitated the reaction of semi-hydrogenation of acetylene than monometallic Au NRs. The existence of highly active Pt around Au promotes the conversion of C2H2 hydrogenation to C2H4 owing to the synergistic effects of bimetallic nanostructures. The reaction was then studied over Au NRs@ZIF-8, Au@Pt NRs@ZIF-8, and Au@Pt NTs@ZIF-8. Results similar to those for NPs@ZIF-67 were obtained. Au@Pt NTs@ZIF-8 showed the best catalytic performance with 88.1% conversion at 175°C. Compared with the Au@Pt NTs@ZIF-67 catalyst, acetylene conversion can be further improved by using ZIF-8 as the host material. After catalysis, Au@Pt nanotubes in the core maintained their tubular features, and the particles retained their original morphologies, and the morphology of the Au@Pt NTs encapsulated in ZIFs was well dispersed, showing the core-shell structure was intact and not damaged, albeit with a roughened surface (Figure S15), suggesting their robust structural stability so that their activities would not decrease and the catalyst could be reused. Compared with other similar catalysts reported for the acetylene semi-hydrogenation reaction, these M@ZIFs catalysts exhibited higher acetylene conversion or ethylene selectivity at a lower temperature (Gonçalves et al., 2020, Osswald et al., 2008, Redfern et al., 2018, Shun et al., 2015, Yang et al., 2015); the catalytic effects of this work and other similar materials have been listed in Table S2. Obviously, some catalysts showed high acetylene conversion but low ethylene selectivity, whereas some catalysts showed high ethylene selectivity but low acetylene conversion. In terms of both selectivity and activity, the catalysts Au@Pt NTs@ZIF-67 and Au@Pt NTs@ZIF-8 were better than that of the other catalysts, showing their superior catalytic performance for the semi-hydrogenation of acetylene. Encapsulation of hollow metal NTs with excellent catalytic properties imparts functionality to the porous MOFs, leading to an even higher catalytic activity for the semi-hydrogenation reaction of acetylene.

Figure 4

Acetylene Conversion and Ethylene Selectivity Over the Catalysts NPs@ZIF-67 and NPs@ZIF-8

To further explain the results, theoretical insights were explored by DFT calculations (Figure 5). Based on the aforementioned catalysts analyzed, models of Au@Pt NTs and Au@Pt NRs have been constructed in Figures 5 and S22. TS of each step of C2H2 hydrogenation to C2H4 on previous constructed configuration was searched by Dimer method. And energetics of forming C2H3∗ (∗ represents adsorbate) and C2H4∗ on two catalysts were shown in Figures S23 and S24. Energy barriers of C2H2∗ converted to C2H3∗ and then to C2H4∗ on Au@Pt NTs is 0.85 and 0.64 eV, respectively. For catalyst Au@Pt NRs, energy barriers of two steps are 0.65 and 1.10 eV. Thus, the highest energy barriers of Au@Pt NTs and Au@Pt NRs are 0.85 and 1.10 eV. Obviously, Au@Pt NTs have a lower energy barrier compared with Au@Pt NRs, which suggests that Au@Pt NRs are favorable in thermodynamics. This result is consistent with experimental analysis.

Figure 5

DFT Calculations for Energetics Diagram of C2H2 Hydrogenation to C2H4

Energetics diagram of C2H2 hydrogenation to C2H4 (A1) and (A2) for Au@Pt NTs and (B1) and (B2) for Au@Pt NRs. Brown is carbon atom, white is hydrogen, orange is gold, and gray is platinum.

Conclusions

In summary, multifunctional nanocatalysts, Au@Pt NTs@ZIFs for which multiple hollow Au@Pt NTs are fully incorporated in the ZIFs (ZIF-67 and ZIF-8) matrix, can be successfully assembled by a facile in situ encapsulation strategy. To the best of our knowledge, this is the first report on a synthesis of multicore-shell nanoparticles with hollow Au@Pt NTs as cores. This synthesis strategy allows the encapsulation of hollow metal nanotubes in a non-aggregated pattern and the fabrication of size-controlled MOFs by adjusting the amount of added ligand. In combination with the unique tubular structure of hollow Au@Pt NTs and porosity of ZIFs, the as-prepared Au@Pt NTs@ZIFs exhibited high catalytic performance and ethylene selectivity in the semi-hydrogenation of acetylene. We believe this nanocomposite encapsulation design, which involves the trapping of hollow bimetallic NPs in MOFs, will promote their potential applications in catalysis.

Limitations of the Study

This study provides a strategy for encapsulating metal nanotubes into ZIFs to form multifunctional composite materials. However, the multi-step synthesis of Au@Pt NTs and Au@Pt NTs@ZIFs makes the material costly. Limited by the cost, it is currently difficult to achieve industrialization. Hence, a simpler and cheaper method for preparing multifunctional composites should be developed in future research.

Resource Availability

Lead Contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Haitao Xu (xuhaitao@ecust.edu.cn).

Materials Availability

This study did not generate new unique reagents.

Data and Code Availability

The published article includes all datasets/code generated or analyzed during this study.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.