Wu Genghuang; Rong Junfeng

(SINOPEC Research Institute of Petroleum Processing, Beijing 100083)

Abstract: Replacement of precious noble metal catalysts with cost‐effective, non‐noble heterogeneous catalysts for chemoselective hydrogenation of nitroarenes holds tremendous promise for the clean synthesis of nitrogen‐containing chemicals.Graphitic carbon layers encapsulated Ni catalysts (Ni@CN) are generated by a facile, scalable and straightforward strategy via the pyrolysis of 2,5‐pyridinedicarboxylic acid coordinated Ni‐MOF acting as the precursor.Physicochemical properties of the Ni@CN catalysts have been investigated by X‐ray diffraction, scanning electron microscopy, transmission electron microscopy, elemental analysis and N2 adsorption‐desorption analysis.The Ni@CN catalysts were found to be highly efficient in the chemoselective hydrogenation of various nitroarenes with other functional groups towards corresponding anilines under mild reaction conditions (85 °C, 1.0 MPa of H2 pressure).Based on the results of controlled tests, the catalytic activity can be attributed to the Ni NPs, while the presence of graphitic carbon layers favors the preferential adsorption of the nitro groups.The recyclability and anti‐sulfur poisoning capability of Ni@CN were also investigated.

Key words: chemoselective hydrogenation; carbon encapsulated; metal‐organic frameworks; nickel nanoparticles; nitroarenes

1 Introduction

Aromatic amines, with a total production amounting to more than 4 million tons per year, are key building blocks and central intermediates for the production of bulk and fine chemicals, such as dyes, pharmaceuticals,agrochemicals, and polymers[1‐3].Conventional Bechamp(using Fe/HCl) or sulfide reduction (with H2S or NaHS as reducing agents) processes are widely applied to produce aromatic amines in industry[4].Unfortunately,such processes show a low conversion rate and produce large amounts of waste acids and residues, resulting in severe environmental problems[5].Among the alternative methods used for their synthesis, catalytic reduction of nitroarenes using molecular hydrogen is the most environmentally friendly and cost‐effective method.Currently, hydrogenation of nitroarenes is mainly performed on noble metal‒based catalysts[2].Despite high activity achieved by noble metals, the expensive price and rare storage limit their applications.Besides, the selective hydrogenation of a nitro group in the presence of other reducible functional groups, e.g., halogen, cyano,benzyloxy, and carbonyl species, is challenging[6‐7].

As a response, tremendous efforts have been devoted to designing heterogeneous catalysts based on cheaper first‐row transition‐metals (i.e.Fe, Co, and Ni)[8].However, most of these supported nonprecious‐metal catalysts are prone to oxidation, sintering, and leaching problems under the reaction conditions.For example,commercially available Raney Ni catalysts are pyrophoric when dry and are readily deactivated when they are exposed to air.Therefore, different strategies have been proposed to develop stable, easy‐to‐handle non‐precious metal nanoparticles (NPs) based catalysts, including coating with organic molecules, polymers, oxides, or carbon[4,9].Especially, graphitic carbon coating can serve as a passivation layer to prevent non‐precious metal NPs from agglomeration and oxidation due toits diffusion suppression and chemical inertness to oxidizing gases, leading to high activity in a variety of catalytic reactions[10].Recently, the application of Co NPs embedded in N-doped carbon matrix[11]and 3D N‐doped graphene/carbon nanotube hybrids with encapsulated Ni NPs[12]acting as catalysts displayed exciting selectivity for numerous structurally diverse nitroarenes, respectively.However, the catalysts require relatively expensive nitrogen‐containing precursors or graphene as well as high H2pressure (50 bar and 20 bar, respectively).In this respect, it is necessary to develop highly efficient and cost‐effective non‐noble metal catalysts for the reduction of nitroarenes.

Metal‐organic frameworks (MOFs) have recently gained particular attention as a new class of ordered nanoporous materials mainly because of their designable framework structures that are modularly built from transition‐metal nodes and organic linkers[13].A new family of graphitic carbon encapsulating nonprecious‐metal nanocomposites can be created by pyrolysis of MOFs as the precursors, which may find promising applications in the related energy, environmental protection and chemical engineering areas[14].Tang, et al.reported the MOF‐derived porous graphitic carbon layers encapsulated Ni NPs (Ni@C) for the chemoselective hydrogenation of nitroarenes under relatively mild conditions[15].Pan,et al.[16]have lately constructed the N doped Ni@C(Ni@CN) derived from di‐ligands Ni‐MOF to catalyze mild hydrogenation of nitroarenes.The conversion of nitrobenzene can reach 99.1%, and the selectivity of aniline is over 99%.However, most MOF solids are solvothermally prepared using organic solvents (e.g.,dimethylformamide) and acid/base additives such as HCl or amines, which would limit their production in commercial scale[17].It is therefore of interest to develop a robust Ni@C catalyst derived from MOF that can be directly assembled by using water as the sole solvent without use of any additives.

Based on the above consideration, herein, we report the facile synthesis of 2,5‐pyridinedicarboxylic acid coordinated Ni‐MOF by one‐step hydrothermal crystallization, which is then subject to pyrolysis under N2atmosphere to generate Ni@CN materials.The resultant Ni@CN materials exhibit excellent catalytic hydrogenation activity and selectivity for the conversion of various nitroarenes with other functional groups to corresponding anilines under mild reaction conditions(85 °C, 1.0 MPa of H2pressure).Based on the results of controlled tests, the catalytic activity can be attributed to the Ni NPs while the presence of graphitic carbon layers favors preferential adsorption of the nitro groups.The unique Ni@CN materials also exhibit excellent catalytic ability against thiophene poisoning.Moreover, these Ni@CN catalysts are easily separated and can be conveniently reutilized, which is desirable for developing a practical and economical aromatic amine synthesis process.

2 Experimental

2.1 Materials

Ni(NO3)2·6H2O (A.R.) was purchased from the Beijing Yili Fine Chemical Company Co., Ltd.; NaOH (G.R.) and 2‐propanol were purchased from the Damao Chemical Reagent Factory; 2,5‐pyridinedicarboxylic acid (98%), 4‐chloronitrobenzene (>99.5% (GC grade)), 4‐nitrophenol (A.R.), 4‐nitroanisole (98%),4‐nitrophenylacetonitrile (98%), 4‐nitrobenzenamine(A.R.), and 1,3‐dinitrobenzene (≥99% (GC grade))were purchased from the Aladdin Reagent Co., Ltd.;nitrobenzene (A.R.) was purchased from the Sinopharm Chemical Reagent Co., Ltd.(Shanghai, China); and thiophene (99%) was purchased from the J&K Scientific Ltd.(China).

2.2 Catalysts synthesis

2.2.1 Synthesis of Ni-MOF

Ni(NO3)2·6H2O (29.08 g, 100 mmol) was dissolved in 100 mL of deionized water, 2,5‐pyridinedicarboxylic acid(16.71 g, 100 mmol) and NaOH (8.00 g, 200 mmol) were dissolved in 150 mL of deionized water in a beaker at 90 °C.The two solutions were mixed and stirred for 10 min at 90 °C.The resulting homogeneous solution was transferred into a stainless steel autoclave, followed by static hydrothermal processing at 150 °C for 7 h.After being cooled to room temperature, the generated green precipitates were filtered and washed with deionized water several times, followed by drying at 70 °C in air for 6 h to derive the Ni‐MOF material.

2.2.2 Synthesis of Ni@CN

In a typical synthesis experiment, a ceramic boat with 4 g of Ni‐MOF powder was placed in a tubular furnace, and was heated to 500 °C (or 600 °C, 700 °C, and 800 °C,respectively) for 2.5 h hours at a temperature increase rate of 5 °C/min under N2flow.The furnace was then cooled down naturally to room temperature to obtain a series of Ni@NC‐T catalysts, where T represents the pyrolysis temperature.

For comparison, the Ni@CN‐800 was further treated in 1 mol/L HCl solution for 4 h at 90 °C to dissolve the exposed and insecure Ni NPs, followed by a repeatedly washing process with deionized water until the pH value of discharged water reached 7.The as‐prepared catalyst was denoted as Ni@CN‐800‐acid.On the other hand, the carbon shell of Ni@CN‐800 was eliminated by heating Ni@CN‐800 at 380 °C in air for 2 h and was subsequently kept at that temperature under a 20% H2/N2flow for another 2 h to obtain the Ni@CN‐800‐air sample.

2.3 Catalysts characterization

The X‐ray diffraction (XRD) patterns were recorded on a Panalytical Empyrean X‐ray diffractometer, using CuKα radiation at a scanning angle (2θ) range of 5° ‒80°.Thermogravimetric/differential thermal analysis(TG/DTA) was performed on a TA SDTQ600 instrument in the N2flow by heating at a rate of 5 °C/min from room temperature to 800 °C.The surface area and pore volume were measured on a Micromeritics ASAP 2420 apparatus according to the BET method.Scanning electron microscopy (SEM) images were acquired using a Hitachi S‐4800 i8 instrument.Transmission electron microscopy (TEM) and high‐resolution TEM (HRTEM)images were taken by a field emission transmission electron microscope (FEI Tecnai G2 F20) operated at 200 kV.X‐ray photoelectron spectroscopy (XPS) analysis was performed on an ESCALAB 250 X‐ray photoelectron spectrometer.The C, H, N, and O elemental composition was quantified by CHNS/O analysis using a Vario MICRO cube and rapid OXY cube elemental analyzer(Elementar Analysensysteme GmbH), respectively.

2.4 Evaluation of the catalytic performance for hydrogenation reactions:

The catalytic hydrogenation reactions were carried out in a 50 mL cylindrical stainless steel high‐pressure reactor(Parr 4848, Instrument Company, USA).The reactant,solvent and catalyst were added into the reactor.The reactor was sealed, purged two times with nitrogen under a pressure of 1 MPa, and then was pressurized with H2to a setting point.The reactor was then heated to the target temperature under continuous stirring at 600 r/min.After cooling down with cold water, the gaseous phase was released, and the catalyst was collected with a magnet.The liquid product was identified by gas chromatography‐mass spectrometry (GC‐MS, Agilent 7200 GC QTOF MS) and was quantitatively analyzed by GC (Agilent 7890A).

3 Results and Discussion

3.1 Characterization of the catalysts

Figure 1(a) shows the XRD pattern of the as‐synthesized Ni‐MOF precursor, which matched well with previously reported pyridinedicarboxylate‐containing Co‐MOF[18],indicating that the Ni‐MOF had good phase purity.TG‐DTA of the Ni‐MOF was performed under the N2atmosphere (Figure 1b).According to previous work[19],the weight loss before 200 °C was attributed to the loss of the physically absorbed and coordinated water.The endothermic peak at 425 °C and the weight loss between 350 °C and 450 °C signified the collapse of the framework[1,20].During this step, the carbonization of the pyridinedicarboxylate unit and the reduction of Ni nodes were completed.Based on the TG‐DTA curve, four representative pyrolysis temperatures (500 °C, 600 °C,700 °C, and 800 °C) were adopted for preparation of the catalysts.As shown in Figure 1(c), the XRD patterns of the Ni@CN‐T exhibited three diffraction peaks at around 44.5°, 51.8° and 76.3°, corresponding to the characteristic diffractions from (111), (200), and (220) planes of face‐centered‐cubic (fcc)‐structured Ni, respectively (JCPDS 04‐0850).In addition, the diffraction peak at 26.1°(Ni@CN‐700 and Ni@CN‐800) can be indexed to the diffraction from (002) plane of graphite[20].The mean size of Ni NPs in Ni@CN‐500, Ni@CN‐600, Ni@CN‐700,and Ni@CN‐800 is calculated to be 11.2 nm, 13.9 nm,28.0 nm, and 32.0 nm, respectively, using the Debye‐Scherer equation based on the XRD results.The particlesize of the NPs increased with the rise in pyrolysis temperature, especially when the pyrolysis temperature was above 600 °C.

Figure 1 Powder XRD patterns and TG-DTA curve of Ni-MOF and Ni@CN-T (a) and (b) Ni-MOF; (c) Ni@CN-T

To evaluate the morphologies and microstructural change of Ni‐MOF after the pyrolysis treatment, representative SEM, TEM, and HRTEM images of Ni‐MOF precursor and Ni@CN‐T samples are presented in Figure 2 and Figure 3, respectively.As shown in Figure 2, the Ni‐MOF precursor is composed of irregularly shaped crystals with smooth surfaces and a wide size distribution.After pyrolysis, the Ni@CN‐T retains the shape of the original Ni‐MOF but the surfaces of the samples turn to much rougher and there are some spherical particles formed on the bulk.As the pyrolysis temperature rises, the particles tend to agglomerate and the size become larger.The change of the morphology of the samples is in agreement with the XRD results shown in Figure 1, which clearly indicates that as the pyrolysis temperature increased to higher than 500 °C, the Ni‐MOF precursor turned into Ni/carbon composites.

The TEM and HRTEM images in Figure 3 reveal that Ni@CN‐T samples are made of 1‒3 nm thick carbon nanoshells encapsulated Ni NPs, indicating that Ni@CN samples have been successfully derived from Ni‐MOF and can be directly assembled by using water as the sole solvent.As for Ni@CN‐500 and Ni@CN‐600, most Ni NPs are dispersed effectively with a mean size of less than 20 nm.However, for Ni@CN‐700 and Ni@CN‐800,there are lots of Ni NPs with a size of over 50 nm.For Ni@CN‐800, some hollow carbon nanoshells of 10‒15 nm in size can be observed, suggesting that the adjacent Ni NPs tend to agglomerate to form larger particles at a pyrolysis temperature of above 700 °C[22].In can be seen from the HRTEM images that carbon nanospheres have the typical distance value of graphite (~0.34 nm), and are tightly wrapped around Ni NPs to prevent Ni NPs from being in contact with oxygen, thus avoiding the oxidation of the Ni NPs.However, there are still some cracks in the carbon layers that can be found.In a recent work, Deng,et al.prepared a sample with CoNi alloy NPs totally encapsulated by several layers of graphene.Those CoNi NPs were not soluble in a strong acid environment due to the protection provided by the carbon layers.In our case, lots of Ni NPs can be dissolved in 1 mol/L of HCl solution, further suggesting that some Ni NPs are not totally covered by carbon layers.This fact meant that H2or even reactants could directly have access to the Ni nanoparticles through the carbon layers, when the hydrogenation process occurred.

The textural properties of these catalysts are evaluated by the N2adsorption‐desorption analysis.As shown in Figure 4, the isotherms of all the catalysts are of type IV in the IUPAC classification, which is characteristicof mesoporous materials.A hysteresis loop at above P/P0≈ 0.8 is observed for Ni@CN‐500, suggesting the presence of intergranular mesoporosity.At the same time, the N2adsorption‐desorption isotherms of Ni@CN‐600, Ni@CN‐700, and Ni@CN‐800 change to type IV with type H4hysteresis that closes atp/p0≈ 0.4,suggesting the presence of mesoporous structure inside the nanoparticles[23].It is obvious that the porosity structure allows easy penetration of reactants and helps increase the exposed active sites for hydrogenation[24].The BET area (SBET), pore volume (Vpore), and elemental compositions of the Ni@CN‐T are summarized in Table 1.TheSBETandVporeof Ni@CN‐500 (122 m2/g, 0.105 cm3/g) are about 1.3 times and 1.4 times those of Ni@CN‐800 catalyst (93 m2/g, 0.073 cm3/g).Both theSBETandVporedecrease with the rise in pyrolysis temperature,indicating that some pores are collapsed in the catalysts at high pyrolysis temperature and thus decrease the BET surface.As for the elemental compositions, the contents of C remain almost the same for all Ni@CN‐T samples.At the same time, the contents of Ni increase with the rise in pyrolysis temperature, while the contents of H, N and O decrease with the rise in pyrolysis temperature.These results indicate a better graphitization degree of carbon at higher pyrolysis temperature, which is in agreement with the emerging of diffraction peak at 26.1°in XRD patterns for Ni@CN‐700 and Ni@CN‐800 as shown in Figure 1.

Figure 2 SEM images of Ni-MOF and Ni@CN-T samples

Figure 3 TEM and HRTEM images of Ni@CN-T samples

Figure 4 N2 adsorption-desorption isotherms of Ni@CN-T samples

3.2 Evaluation of the chemoselective catalytic hydrogenation performance

In the commercial process, the widely used noble catalysts (Pd, Pt, etc.) undergo serious deactivation,which mainly arises from the serious metal leaching due to the unselectively hydrogenated dehalogenation ofvarious halogenated nitroarenes.Therefore, the catalytic performance of the freshly prepared nanocatalysts for the selective hydrogenation of nitro groups was first investigated by using 4‐chloronitrobenzene (denotedas 4‐CNB) as a benchmark substrate.As shown in Table 2, the parent Ni‐MOF gave no conversion of 4‐CNB, suggesting that nickel ions coordinated with 2,5‐pyridinedicarboxylic acid were not active for reduction of 4‐CNB under the investigated conditions(Table 2, entry 1).Interestingly, the Ni‐MOF derived Ni@CN‐T catalysts showed good performance in this transformation under mild reaction conditions (85 °C,1.0 MPa of H2pressure, Table 2, entries 2‒5).Among which, Ni@CN‐800 exhibited the highest catalytic performance, achieving an 100% conversion of 4‐CNB and a 98.3% selectivity to 4‐chloroaniline (4‐CAN) in 100 min.Aniline is the only detectable by‐product.In addition, a small amount of 1,2‐bis(4‐chlorophenyl)diazene as the intermediate has been observed by GC‐MS over Ni@CN‐600 (Table 2, entry 3).Therefore, it can be concluded that the reaction mechanism probably works through the condensation route[25].

Table 1 Physicochemical properties of Ni@CN-T samples

Table 2 Hydrogenation of 4-chloronitrobenzene using different Ni catalysts1)

According to some previous researches, the introduction of nitrogen into the carbon matrix plays an important role to provide non-noble metal nanoparticles with high activity[26‐27].However, in our cases, though the bulk N content varies from catalyst to catalyst, all the Ni@CN‐T achieved similar catalytic performance for the chemoselective hydrogenation of 4‐CNB (Table 2).In order to understand the catalytic performance of Ni@CN‐T, controlled tests were further implemented.When the Ni NPs of Ni@CN‐800 were etched by HCl solution, the residual Ni@CN‐800‐acid did not exhibit legible activity for the hydrogenation of 4‐CNB(Table 2, entry 6), indicating that the H2could not be activated by hollow carbon nanoshells or Ni NPs totally encapsulated by several layers of graphene, thus the Ni NPs covered with defective carbon shells should be the active sites for hydrogenation reactions.To further investigate if the carbon layers played an intrinsic role in the chemoselective hydrogenation reaction, the Ni@CN‐800 was subjected to heating at 380 °C in air for 2 h and subsequently under a 20% H2/N2flow for another 2 h to remove the carbon layers on the Ni NPs and obtain a Ni@CN‐800‐air sample.Similar to Ni@CN‐800, Ni@CN‐800‐air also exhibited great activity toward the hydrogenation of 4‐CNB, but the selectivity dropped from 98.3% of Ni@CN‐800 to 95.4% (Table 2, entry 7).Basedon these results, it can be proposed that the catalytic activity was attributed to the Ni NPs while the presence of graphitic carbon layers favored the preferential adsorption of the nitro groups, leading to the chemoselective hydrogenation of 4‐CNB.

Table 3 Recycling of the Ni@CN-800 catalyst

Recyclability of a catalyst is critical for practical applications.In this sense, the stability and reusability of the Ni@CN‐800 were investigated in the chemoselective hydrogenation of 4‐CNB to 4‐CAN.Since metallic Ni NPs are paramagnetic, the separation of the solid catalyst from the liquid was very facile with the help of a magnet.After being washed 2 times with ethanol and being dried at 90 ℃ in air, the catalyst can be reused without further treatment.As shown in Table 3, Ni@CN‐800 can be conveniently reused at least six times without significant deactivation for the hydrogenation of 4‐CNB.In the sixth cycle, a high selectivity of 4‐CAN (99.0%) was still achieved at the full conversion of 4‐CNB.These results strongly emphasized the fact that Ni@CN‐T manufactured by the strategy in the present study displayed impressive stability and recyclability, thus showing its potential application in the chemical industry.

Inspired by the performance of Ni@CN‐800 for 4‐CNB hydrogenation, the general applicability of Ni@CN‐800 in the chemoselective hydrogenation of a variety of substituted nitroarenes was further investigated.As shown in Table 4, the Ni@CN‐800 catalyst exhibits a high activity for the hydrogenation of industrially relevant nitroarenes such as nitrobenzene, 4‐nitrobenzenamine and 1,3‐dinitrobenzene toward their corresponding anilines with excellent selectivity (Table 4, entries 1‒3).It is well known that the most‐challenging substrates are those which contain other easily reducible groups.Interestingly,substrates containing sensitive groups or reducible groups,such as ‒OH, ‐OCH3, or C≡N can be as well smoothly and chemoselectively hydrogenated into the correspondinganilines without competitive hydrogenation on these substituted moieties (Table 4, entries 4‒6).In addition,the existence of trace sulfur‐containing contaminants in the industrial‐grade nitroarenes, primarily extracted from the coal tar, can cause serious catalyst poisoning[28‐29].Therefore, the catalytic performance for Ni@CN‐800 against sulfur poisoning was further investigated.Surprisingly, when 0.5 mL of thiophene acting as poison to Ni NPs was added, the catalytic activity of the Ni@CN‐800 catalyst toward the hydrogenation of 4‐CNB is maintained at the same level with the selectivity to 4‐CAN being even higher than that case without the use of thiophene (Table 4, entry 7).According to previous research, the anti‐sulfur poisoning of Ni@CN‐800 could be attributed to the presence of surface/subsurface carbon at the Ni NPs[27].All these results demonstrate again that this Ni@CN‐800 catalyst displays an excellent activity and chemoselectivity in the hydrogenation of substituted aromatic compounds in general.

Table 4 Scope of the hydrogenation of nitroarenes catalyzed by Ni@CN-800 catalyst1)

4 Conclusions

In summary, a highly efficient method has been developed for the chemoselective hydrogenation of nitroarenes into corresponding anilines over a non‐noble Ni@CN‐T catalyst.The Ni@CN‐T catalysts are made by a facile,scalable and straightforward strategy via the pyrolysis of 2,5‐pyridinedicarboxylic acid coordinated Ni‐MOF serving as the precursor, which is synthesized by using one‐step hydrothermal crystallization method.As revealed by different physical characterization techniques,the pyrolysis temperature markedly affected the structure of the Ni@CN‐T catalysts.Among which, Ni@CN‐800 exhibited a highest catalytic performance, achieving an 100% conversion of 4‐CNB and a 98.3% selectivity to 4‐CAN under mild reaction conditions (85 °C, 1.0 MPa of H2pressure).Full conversion of various nitro substrates with excellent selectivity (>99%) toward the corresponding amines was also achieved.Based on the comparison of catalytic performance of Ni@CN‐800, Ni@CN‐800‐acid, and Ni@CN‐800‐air, it can be proposed that the catalytic activity of Ni@CN‐T is attributed to the Ni NPs, while the presence of graphitic carbon layers can favor preferential adsorption of the nitro groups, leading to the chemoselective hydrogenation of nitroarenes.Owing to the carbon graphitic carbon layers encapsulated Ni NPs structure, Ni@CN‐800 also exhibited high anti‐sulfur poisoning capability.Due to their facile operation on synthesis, storage, and utilization,the stable and cost‐effective Ni@CN‐T catalysts would have the potential for broad application in green and economical synthesis of fine chemicals.

Acknowledgement:This work was financially supported by the China Petroleum & Chemical Corporation (SINOPEC 420043‐4, 420043‐10)