Graphene stabilized ultra-small CuNi nanocomposite with high activity and recyclability toward catalysing the reduction of aromatic nitro-compounds

Abstract

Nowadays, it is of great significance and a challenge to design a noble-metal-free catalyst with high activity and a long lifetime for the reduction of aromatic nitro-compounds. Here, a 2D structured nanocomposite catalyst with graphene supported CuNi alloy nanoparticles (NPs) is prepared, and is promising for meeting the requirements of green chemistry. In this graphene/CuNi nanocomposite, the ultra-small CuNi nanoparticles (∼2 nm) are evenly anchored on graphene sheets, which is not only a breakthrough in the structures, but also brings about an outstanding performance in activity and stability. Combined with a precise optimization of the alloy ratios, the reaction rate constant of graphene/Cu₆₁Ni₃₉ reached a high level of 0.13685 s⁻¹, with a desirable selectivity as high as 99% for various aromatic nitro-compounds. What's more, the catalyst exhibited a unprecedented long lifetime because it could be recycled over 25 times without obvious performance decay or even a morphology change. This work showed the promise and great potential of noble-metal-free catalysts in green chemistry.

---

Introduction

As a kind of toxic organic compound, aromatic nitro-compounds are hard to be degraded in the environment.¹ So conversion from aromatic nitro-compounds into aromatic amino-compounds conforms with the demands of green chemistry. Due to the inertness of the nitro group, reduction of it can hardly take place spontaneously with a reductant at room temperature, so a catalyst is required for the reaction.² Based on these considerations, fabricating a desirable catalyst which can combine high activity, favorable selectivity, good recyclability and low cost is of great significance, but remains a challenge.

Nowadays, in order to elevate the activity of catalysts, alloys involving noble metals are usually employed due to their high intrinsic activities,^3,4 but their applications are limited due to their selectivity and rising cost.^5–7 As an alternative, the synergistic effect of alloying between different metals is considered as a promising strategy to achieve high activity, selectivity and low cost.^8,9 Besides, the lifetime of catalysts is another crucial parameter. To extend the lifetime and recyclability of catalysts while keeping their activity, one of the ideal strategies is to anchor catalyst nanoparticles (NPs) onto a certain support,^10,11 which can effectively prevent the aggregation of the NPs and further maintain their activity and lifetime. Benefiting from its tremendous surface area and conductivity, graphene is believed to be a good candidate as a catalyst support.^12–14 Due to its lamellar structure, enormous amount of NPs can distribute discretely thus generating much more catalytic active sites and contacting surfaces with substrates. Therefore, based on performance optimizing through delicate adjustments of the alloy ratios and cooperation with the graphene support, it is reasonable to believe that an ideal catalyst can be fabricated, which combines high activity, high selectivity, long lifetime and much lower costs. The nanocomposite promises to exert fantastic catalytic performance in the reduction of aromatic nitro-compounds and even other applications.

Herein, a high performance graphene/CuNi nanocomposite catalyst was synthesized via electrostatic–adsorption interactions^15,16 followed by an in situ simultaneous reduction process. CuNi alloy NPs with a diameter of ∼2 nm are evenly anchored on the reduced graphene oxide. The ultra-small CuNi NPs in the nanocomposite catalyst can remarkably enhance the catalytic activity due to its huge surface area. Together with delicate adjustments of the alloy molar ratios, graphene/Cu₆₁Ni₃₉ manifested the highest activity with favorable selectivity to catalytically reduce aromatic nitro-compounds. What's more, this nanocomposite catalyst has an outstanding recyclable lifetime compared to previous reports.^17,18 This catalyst exhibited a recycle lifetime of 25 times, which was an unprecedented record compared to previous literature. So it puts forward a promising orientation for constructing superior nanocomposite catalysts.

Results and discussion

Characterization of the catalyst

Ahead of other characterizations, TEM images were required to investigate the morphologies of the product. Fig. 1A and B display different magnifications for an overall and detailed image of the as-designed graphene/CuNi nanocomposite catalyst. It can be seen that the monodispersed CuNi NPs are distributed closely and evenly on the reduced graphene oxide sheets. The inset of Fig. 1B plots the size distributions of the NPs. The average diameter is ∼2 nm with a narrow size variation. HRTEM was used to further investigate the structures. The image in Fig. 1C shows the nature of the single crystals of all the NPs presented. Clear lattice fringes of 0.206 nm are just right between the characteristic d-spacings of Ni (111) planes of 0.195 nm and those of Cu (111) planes of 0.201 nm. It proves that the alloy solid solution constitutes both Cu and Ni. The inset in Fig. 1C provides the SAED pattern of the polycrystalline diffraction rings, showing the tiny particle sizes relative to the electron beam spot. The brighter inner diffraction ring and dimmer outer ring were indexed as the (111) and (200) planes of the face-centered cubic (fcc) structure of the CuNi NPs. Fig. 1D shows the XRD patterns of the graphene/CuNi nanocomposite, the reduced graphene oxide and the CuNi alloy NPs. The peaks at 43.8, 51.2 and 75.1 degrees in the pattern of the nanocomposite can be indexed to the (111), (200) and (220) planes of the fcc structure for the anchored CuNi NPs, which are located between the standard sites of Cu (JCPDS no. 04-0836) and Ni (JCPDS no. 04-0850), confirming the formation of a CuNi alloy. The broad peak centered around 25 degrees at the left end in the nanocomposite corresponded to the characteristic peak for reduced graphene oxide.¹³ Peaks in the nanocomposite match well with the patterns of the CuNi alloy NPs and reduced graphene oxide, demonstrating that the catalyst was a composite of the two components. In addition, EDS was used to analyze the elemental profiles of the graphene/CuNi nanocomposite with an optimal ratio (Fig. 2A). The molar ratio of Cu to Ni is about 60 : 40, which agrees well with the initial target ratio. Furthermore, elemental mappings were carried out to illustrate the space distributions of the component elements. Fig. 2B gives a comprehensive image of the element mappings of Ni, Cu and C, overlapped with the corresponding TEM image. Fig. 2C–E exhibit the individual mappings of the three component elements in distinct colors.

Fig. 1 Graphene/CuNi nanocomposite. (A) TEM image at low magnification. (B) TEM image at high magnification with inset of size distribution plot of the CuNi NPs. (C) HRTEM image with lattice fringes. Inset is the SAED pattern. (D) XRD patterns of graphene/CuNi nanocomposite, reduced graphene oxide and CuNi alloy NPs.

Fig. 2 Element analysis of the graphene/CuNi nanocomposite with optimal alloy ratio. (A) EDS spectrum. (B) Comprehensive image of element mappings overlapped with TEM image. (C–E) Individual mappings of Ni, Cu and C.

Additionally, XPS was performed to acquire further insight into the valence information of the catalyst. Fig. 3 plots the full spectrum and detailed portions of the elements Ni and Cu. In the full spectrum (Fig. 3A), peaks of C 1s, O 1s, Ni 2p and Cu 2p can be found in the binding energy region from 0 to 1200 eV. Fig. 3B, C show magnified spectra of Ni 2p3/2 and Cu 2p3/2 respectively. In Fig. 3B, the peak at 853 eV corresponded to the Ni(0), and the binding energies of 855 eV and 855.8 eV proved to be the Ni(II) and Ni(III).¹⁷ The peaks located at the binding energies of 932.25 eV and 934 eV in Fig. 3C were related to the Cu(0) and Cu(II) respectively.¹⁹ The existence of other valence states on the surface were due to the tiny particle sizes. But changes have been found in the spectra taken after 25 catalytic recycles (Fig. S1, in ESI†). A decrease in the proportions for Cu(II) and Ni(III) revealed the catalyst had recovered during the catalysis process. It also proved the recyclable stability of the nanocomposite catalyst.

Fig. 3 XPS analysis of the graphene/CuNi nanocomposite. (A) Full spectrum in the binding energy region from 0 to 1200 eV. (B) Detailed spectrum of Ni 2p3/2. (C) Detailed spectrum of Cu 2p3/2.

Fabrication mechanism

Scheme 1A presents the fabrication mechanism of the graphene/CuNi nanocomposite by electrostatic-adsorption between the graphene oxide and the metal precursors followed by an in situ simultaneous reduction process. Driven by the electrostatic attraction forces of the graphene oxide, Cu²⁺ ions and Ni²⁺ ions were adsorbed by –OH, –COOH and other oxygen-containing groups on the surface of graphene oxide (Fig. S2†).^15,16 Then an in situ simultaneous reduction of both the graphene oxide and the metal precursors was triggered by a heat of 180 °C with ethylene glycol, resulting in the formation of the nanocomposite with CuNi NPs anchored on graphene sheets. Monitored by Raman spectroscopy (Fig. S3†), the curves of (a) graphene/CuNi nanocomposite and (b) reduced graphene oxide have similar area ratios for peak D to G, but differ from that of (c) graphene oxide. The D to G ratio of the nanocomposite altered distinctly during the reduction (from 1.074 to 0.724), indicating the sp² hybridized structure of graphene obtained effective restoration in the in situ simultaneous reduction process.

Scheme 1 (A) Mechanism for the fabrication of the graphene/CuNi nanocomposite. (B) Illustration of the catalysis mechanism for the reduction of p-nitrophenol by the graphene/CuNi nanocomposite catalyst.

Catalytic performances

For an ideal catalyst, it is best to meet the requirements of high selectivity, high activity, long lifetime and low cost. The as-prepared graphene/CuNi nanocomposite catalyst has reached the above demands although free of noble metals. Detailed investigations were carried out according to the above criteria to examine the performance in the absence of any noble metals. For initiating our evaluations on selectivity and conversion rate, a series of aromatic nitro-compounds were designated as the catalytic substrates. Herein, the conversion rate was defined as the ratio of substrates that are involved in the reaction to the whole number of substrates, while selectivity describes the ratio of desired products to the whole number of products obtained. Specifically for the aromatic nitro-compounds, the desired products should be aromatic amino-compounds. Table 1 and Fig. S4† gather the data for the conversion rates and selectivities of the seven designated aromatic nitro-compounds as determined by gas chromatography. The seven substrates represent various species differing in the groups opposite the nitro group. It can be concluded from the table that most substrates have been converted to the desired products. Some of them were converted as much as 99%, even too thoroughly to detect the substrates. Moreover, the selectivity of the above reactions reached a favourable level. In most cases, there were no detectable by-products in the solution, that is to say almost all substrates were converted to the desired products. It revealed the excellent specificity of the as-prepared graphene/CuNi nanocomposite for the reaction. As a typical reaction, the reduction of p-nitrophenol to p-aminophenol is often used in catalyst evaluations. Both the conversion rate and selectivity of the reaction catalyzed by the graphene/CuNi nanocomposite achieved were higher than 99%, proving the superb performance. In addition, the data in Table 1 also declares the wide scope of application for the catalyst, including nitro substituted ketones, aldehydes, alcohols, ethers, etc., with a high conversion rate and selectivity.

Table 1 Conversion rate and selectivity data of the designated aromatic nitro-compounds catalyzed by the graphene/CuNi nanocomposite, with the general formula of the reduction reaction

Aromatic nitro-compounds		Conversion (%)	Selectivity (%)
p-Nitroacetophenone		72.54	>99.99
p-Nitrobenzyl alcohol		77.59	>99.99
p-Nitrotoluene		79.53	64.67
p-Nitrobenzaldehyde		86.56	>99.99
p-Nitroaniline		98.74	>99.99
p-Nitroanisole		>95.47	>99.99
p-Nitrophenol		>99.99	>99.99

---

As the most crucial feature, the activity of the catalysts is the emphasis of our investigations. For convenience of horizontal comparison, p-nitrophenol was chosen as a benchmark to evaluate the activity of the as-prepared catalysts. Besides, reactions catalyzed by CuNi NPs and reduced graphene oxide were also examined by UV-Vis absorption spectroscopy under identical conditions, and the spectra are gathered in Fig. S5.† Before starting the reaction, sodium borohydride was added. The color of the solution changed from light to dark yellow, implying that the p-nitrophenol molecules (whose characteristic absorption peak is at 317 nm) dissociated into p-nitrophenolate ions (characteristic peak at 400 nm). Decay of this peak could reflect the concentration reduction of the p-nitrophenol over time, and the normalized data were extracted in Fig. 4A. The plot clearly exhibits that when the reaction was triggered, the peak at 400 nm dropped drastically under the catalysis of the graphene/CuNi nanocomposite, and the p-nitrophenol was completely extinct in seconds. Meanwhile, a new weak peak at 280 nm emerged and increased with the reduction of p-nitrophenol (Fig. S4A†), suggesting the generation of p-aminophenol. From Fig. 4A we can also see that when the sample is catalyzed by the CuNi alloy NPs, the reaction rate is much lower than that of the graphene/CuNi nanocomposite, thus complete conversion of the p-nitrophenol takes a much longer time. Hardly any effect is noted in the sample catalyzed by graphene, indicating that the reaction cannot be catalyzed by it, and further implying that the activity of the nanocomposite catalyst originates from the CuNi alloy NPs anchored on the graphene. Because the concentration of the sodium borohydride greatly exceeded that of p-nitrophenol, it can be regarded as a constant during the catalytic process. The reaction can be considered as a pseudo-first-order reaction for the p-nitrophenol. Consequently, there is a linear relationship between ln(C_t/C₀) vs. time, as shown in Fig. 4B (where C_t refers to the concentration of p-nitrophenol at time t, and C₀ refers to the initial concentration). From the lines fit using the raw data, rate constants k can be obtained according to the slopes. To our delight, for the graphene/CuNi nanocomposite, k = 0.13685 s⁻¹, while for the CuNi NPs the rate constant is 0.02284 s⁻¹. The former is much higher than the latter, manifesting the obvious advantages of the nanocomposite structure.

Fig. 4 Plots of (A) C_t/C₀ and (B) ln(C_t/C₀) vs. time for the reduction of p-nitrophenol using the different catalysts. Plots of (C, E) C_t/C₀ and (D, F) ln(C_t/C₀) for the reduction of p-nitrophenol by CuNi NPs with different ratios of Cu to Ni.

For optimizing the catalytic performances of the CuNi alloy NPs, the molar ratios of the two metals were discussed in detail. The optimal molar ratio of the CuNi NPs was obtained by calculating the rate constants from the UV-Vis absorption spectra (Fig. 4C–F, Table S1 and Fig. S5–S7†). Fig. 4C exhibits a normalized concentration reduction of p-nitrophenol over time catalyzed by the CuNi alloy NPs of various molar ratios, while Fig. 4E gives more precise optimizations via minor ratio variations based on the results of Fig. 4C. The plots suggest that the CuNi alloy NPs manifested better activities than the pristine Cu or Ni. The catalytic activity reached the highest point with the rate constant k = 0.02284 s⁻¹ when the Cu to Ni ratio was 61 : 39. Therefore, the molar ratio of 61 : 39 for Cu to Ni was used to synthesize the graphene/CuNi nanocomposite.

In considering that the size of the NPs plays an important role on their catalytic performance, the activities of two graphene/CuNi nanocomposites with different CuNi NP sizes were also studied in this work. As Fig. S8† shows, a graphene/CuNi nanocomposite with an average NP size of ∼200 nm was prepared to show the size-dependency of the catalytic activity. From the plots of C_t/C₀ and ln(C_t/C₀) (Fig. S8D,E†), the rate constant of the nanocomposite with the larger NP size is 0.09119 s⁻¹, which was lower than that of the nanocomposite with the smaller NP size (k = 0.13685 s⁻¹). To acquire further insight into these two nanocomposites, TG analyses were conducted in air (Fig. S8C†). It can be found that the total amounts of CuNi alloy loaded on the graphene differed a lot. The nanocomposite with the larger NP size had a larger loading amount, while the nanocomposite with the smaller NP size loaded less. But the results of the catalytic activities are just the opposite. It is revealed that the nanocomposite with the smaller NP size harvested a higher activity but cost less in the amount of metals, due to the larger surface area generated from the numerous tiny sized CuNi NPs. Many more atoms were at the surface but not inside, leading to a much higher activity than the nanocomposite with the larger NP size.

As an important index of catalysts, the lifetime is undoubtedly a crucial aspect needing to be evaluated. To explore the lifetime and related performances of the graphene/CuNi nanocomposite catalyst, 25 successive catalytic tests were carried out. During the process, both the activity and conversion rate of each cycle were determined to investigate whether the catalyst performed steadily in this process. For obtaining the catalytic activities data, normalized C_t/C₀ of each recycle and ln(C_t/C₀) of the 5th, 10th, 15th, 20th and 25th recycle (every 5 recycles for concision) were plotted in Fig. 5A and the inset of Fig. 5B . The rate constants of each recycle were calculated and gathered in Table S2.† From the table it can be seen that although the catalyst has been retrieved many times, there was no obvious indication of activity decay. Moreover, the conversion rates of the recycles are always higher than 99.99% (Fig. 4B), all of the above data confirm the extraordinary stability and superiority of the nanocomposite structure. For obtaining further insight into the origin of the long lifetime of the catalyst, TEM images were acquired after the 25 catalytic recycles. Unexpectedly, after such many recycles, the CuNi alloy NPs anchored on the graphene sheets almost kept their original sizes and shapes, and the distribution density hardly changed (Fig. S9†). Owing to the supporting effect of the graphene, the CuNi alloy NPs in the nanocomposite can maintain their sizes and shapes without agglomeration for a long period. However, the CuNi alloy NPs without support exhibited a distinct contrast (Fig. S10†). Suffering from the lack of support, apparent activity decay appeared at the 6th recycle, resulting in a much shorter lifetime compared with the nanocomposite. It can be demonstrated from the TEM image (Fig. S11†) that the CuNi alloy NPs will quite readily agglomerate without the support, causing the initially dispersive NPs to sinter to bulk. The reduction of surface area and the covering of the catalytic active sites eventually made the catalyst invalid.

Fig. 5 Recyclability tests of the graphene/CuNi nanocomposite: (A) Catalytic recycles of the reduction of p-nitrophenol. (B) Conversion rates of each recycle (inset: plot of ln(C_t/C₀) of the 5th, 10th, 15th, 20th and 25th recycle. (C) Hysteresis loop with a photograph of the catalyst responding to a magnet.

As a magnetic retrievable catalyst, the graphene/CuNi nanocomposite can be retrieved and separated from the reaction solution by magnets. Fig. 5C exhibits the magnetic hysteresis loop obtained at room temperature. It definitely indicates the superparamagnetic nature of the as-prepared catalyst, which coincides with the sizes of the NPs as observed by TEM. The hysteresis loop manifests a saturated magnetization (M_s) of 3.2 emu g⁻¹ due to its tiny particle sizes.²⁰ It is sufficient to be collected in a few seconds by magnets (inset of Fig. 5C), and more importantly, will readily redisperse in solution due to its low coercivity (H_c) (11.4 Oe) and retentivity (M_r) (0.06 emu g⁻¹), which also originates from the superparamagnetism of the NPs that are smaller than its critical size.²¹ The desirable magnetic properties of the catalyst make it promising in magnetic retrievable catalysis.

Catalysis mechanism

Based on the aforementioned characterizations and analyses, a catalysis mechanism was elaborated and illustrated in Scheme 1B. Because the reaction takes place on the surface of the catalyst, it obeys the classical Langmuir–Hinshelwood model.² Together with the great excess of borohydride ions, p-nitrophenolate ions can be adsorbed onto the surface of the CuNi alloy NPs of the nanocomposite. Both the adsorptions are reversible and can be modelled by the Langmuir isotherm.² Under the catalysis of the CuNi NPs, the borohydride ions would dissociate, active surface hydrogen species are thus generated and bonded to the surfaces of the CuNi NPs. The adsorbed p-nitrophenolate ion can be reduced by these active hydrogen species derived from the numerous borohydride ions, and hence be converted into p-aminophenol.^22,23 Because of the much weaker adsorbability of the amino groups than that of the nitro groups, p-aminophenol would then desorb readily from the catalyst surface once generated.²⁴ So the reaction can proceed spontaneously. Given that the rates of diffusion and adsorption/desorption are quite fast, the overall rate is controlled by the determining step of the nitro group reduction. Therefore, the reaction can be considered as a pseudo-first-order reaction.

Experimental section

Chemicals

Natural flake graphite was purchased from Qingdao Guyu graphite Co., Ltd. Concentrated sulfuric acid, sodium nitrate, potassium permanganate, hydrogen peroxide, ethylene glycol, sodium hydroxide, nickel chloride, copper chloride, sodium borohydride, ethanol, and hydrazine hydrate were all of analytical purity and purchased from Sinopharm Chemical Reagent Co., Ltd (SCRC). p-Aminobenzyl alcohol, p-nitrobenzyl alcohol, p-nitrobenzaldehyde, p-aminobenzaldehyde, p-aminoacetophenone, p-nitroacetophenone, p-nitrotoluene, p-toluidine, p-aminophenol, p-nitrophenol, p-nitroaniline, p-phenylenediamine, p-anisidine, p-nitroanisole, and methanol were of analytical purity and purchased from Aladdin Reagents, Shanghai. All reagents were used without further purification.

Preparation of the graphene/CuNi nanocomposites

For the synthesis of the typical graphene/CuNi nanocomposite, a 0.5 mL 20 mM CuCl₂ and NiCl₂ mixed ethylene glycol solution of a certain Cu/Ni molar ratio was mixed with 1.4 mL 0.5 g mL⁻¹ graphene oxide ethylene glycol solution in a glass vessel. After magnetic stirring for 10 min, 0.2 mL sodium hydroxide ethylene glycol solution (0.2 M) and another 6 mL ethylene glycol were added followed by another stirring for 10 min. Then, the glass vessel was put in an oil bath at 180 °C for 15 min. The product was collected by centrifuge and washed with ethanol and deionized water three times, then the product was dried under vacuum at 60 °C for 4 hours before further characterization.

To investigate the influence of particle size on the catalytic activity, graphene/CuNi nanocomposites with CuNi particle sizes of ∼200 nm were also prepared to contrast with the typical product. In this synthesis, a mixed salt of 0.1 mmol CuCl₂ and NiCl₂ of certain Cu/Ni molar ratios was dissolved in 14 mL 0.5 g·mL⁻¹ graphene oxide ethylene glycol solution in a glass vessel. After magnetic stirring for 10 min, 1 mL sodium hydroxide ethylene glycol solution (0.2 M) and another 7 mL ethylene glycol were added followed by another 10 min stirring. The glass vessel was put in an oil bath at 180 °C. Then 400 μL hydrazine hydrate was added to the vessel carefully under magnetic stirring. The product was collected by centrifuge and washed with ethanol and deionized water three times, then was dried under vacuum at 60 °C for 4 hours before further characterizations.

Preparation of the CuNi alloy NPs

In a typical synthesis, 3.05 mL 20 mM CuCl₂ ethylene glycol solution and 1.95 mL 20 mM NiCl₂ ethylene glycol solution were mixed in a glass vessel and magnetically stirred for 10 min. Then the glass vessel was put into an oil bath at 180 °C, followed by the dropwise addition of 0.5 mL hydrazine hydrate carefully under magnetic stirring. The product was collected by centrifuge and washed by ethanol and deionized water three times, then it was dried under vacuum at 60 °C for 4 hours for further characterizations. For preparing alloy NPs of different Cu to Ni ratios, different volumes of the above CuCl₂ and NiCl₂ solutions were adopted according to the target molar ratios, but the total volume of 5 mL was kept constant. For example, taking 2.0 mL 20 mM CuCl₂ ethylene glycol solution and 3.0 mL 20 mM NiCl₂ ethylene glycol solution for a Cu₄₀Ni₆₀ ratio, and taking 2.9 mL 20 mM CuCl₂ ethylene glycol solution and 2.1 mL 20 mM NiCl₂ ethylene glycol solution for a Cu₅₈Ni₄₂ ratio, and so on.

Preparation of the graphene oxide

Graphene oxide was synthesized from natural flake graphite using a modified Hummers method. 15 mg of as-obtained product and 30 mL ethylene glycol were added to a 100 mL beaker which was transferred into an ultrasonic instrument to exfoliate for more than 8 h. Well-dispersed homogeneous graphene oxide solution was formed. The graphene oxide was collected by centrifuge and washed by ethanol and deionized water three times, then it was dried under vacuum at 60 °C for 4 hours for further characterizations.

Characterizations and apparatus

A JEOL JEM-2100 high-resolution transmission electron microscope was used to observe the morphology and determine the microstructures at 200 kV. Element analyses were carried out on an Oxford TN5400 EDS accessory on the HRTEM. The X-ray powder diffraction patterns were obtained on a Bruker D8 Focus diffractometer with Cu Kα radiation. Raman spectra were obtained on a Renishaw Invia under a laser wavelength of 514.5 nm. The X-ray photoelectron spectra were obtained on an AXIS Ultra DLD from Shimadzu Kratos systems using Al Kα radiation. All the spectra were calibrated by the C 1s peak located at 284.8 eV of the contaminated carbon. Thermogravimetric analyses were carried out on Netzsch STA409 at a heating rate of 10 °C min⁻¹ in air from 30 to 800 °C. The catalytic kinetics were investigated on an Agilent 8453 UV-Vis spectrophotometer at a constant temperature of 25 °C. An Agilent 8453 gas chromatograph with a thermal conductivity detector (TCD) was used to determine the selectivity and conversion rate of the products. The magnetic measurements were performed on a Lakeshore 735 vibrating sample magnetometer.

Catalytic tests for the reduction of aromatic nitro-compounds

For evaluating the performances of the as-prepared catalysts, the reductions of the aromatic nitro-compounds were monitored using a UV-Vis spectrophotometer at a constant temperature of 25 °C. In a typical experiment, 1 mL 0.1 mM aromatic nitro-compounds solution and 1 mL 0.1 M sodium borohydride solution were added to a quartz cuvette, which was subsequently put into the spectrophotometer to determine the original absorbance of the substrates. Then 100 μL 3 mg mL⁻¹ catalyst suspension was fast injected into the cuvette to start the reaction under magnetic stirring. The catalytic kinetics was reflected by the fading of the characteristic peaks until the absorbance became constant. After every test, the obtained solutions were reserved for determining the selectivity and conversion rate using a gas chromatograph. The lifetime tests were similar to the method above, but the catalysts would be retrieved by a magnet after each catalytic test.

Conclusions

In conclusion, a high performance noble metal free nanocomposite catalyst of graphene and ∼2 nm CuNi alloy NPs was synthesized through electrostatic-adsorption and an in situ simultaneous reduction process. The synergistic effect of the CuNi alloy endows the graphene/Cu₆₁Ni₃₉ nanocomposite with outstanding catalytic activity and selectivity with long-term recyclability for a range of aromatic nitro-compounds. It puts forward a promising orientation for constructing superior composite catalysts.

Acknowledgements

We acknowledge the funding support from the National Natural Science Foundation of China (NSFC) (Nos: 21171130, 51271132, 21471114 and 91222103) and 973 Project (No: 2011CB932404).

Notes and references

1. L. Yang, S. Luo, Y. Li, Y. Xiao, Q. Kang and Q. Cai, Environ. Sci. Technol., 2010, 44, 7641.
2. P. Hervés, M. Pérez-Lorenzo, L. Liz-Marzán, J. Dzubiella, Y. Lu and M. Ballauff, Chem. Soc. Rev., 2012, 41, 5577.
3. W. Niu and G. Xu, Nano Today, 2011, 6, 265.
4. S. Guo and E. Wang, Nano Today, 2011, 6, 240.
5. J. Clarke, Chem. Rev., 1975, 75, 291.
6. I. Lee, F. Delbecq, R. Morales, M. Albiter and F. Zaera, Nat. Mater., 2009, 8, 132.
7. K. An and G. Somorjai, ChemCatChem, 2012, 4, 1512.
8. A. Singh and Q. Xu, ChemCatChem, 2013, 5, 652.
9. A. Allen and D. MacMillan, Chem. Sci., 2012, 3, 633.
10. L. Rogatis, M. Cargnello, V. Gombac, B. Lorenzut, T. Montini and P. Fornasiero, ChemSusChem, 2010, 3, 24.
11. W. Wu, M. Lei, S. Yang, L. Zhou, L. Liu, X. Xiao, C. Jiang and V. A. L. Roy, J. Mater. Chem. A, 2015, 3, 3450.
12. C. Huang, C. Li and G. Shi, Energy Environ. Sci., 2012, 5, 8848.
13. B. Machado and P. Serp, Catal. Sci. Technol., 2012, 2, 54.
14. X. Huang, X. Qi, F. Boey and H. Zhang, Chem. Soc. Rev., 2012, 41, 666.
15. D. Li, M. Müller, S. Gilje, R. Kaner and G. Wallace, Nat. Nanotechnol., 2008, 3, 101.
16. A. Bagri, C. Mattevi, M. Acik, Y. Chabal, M. Chhowalla and V. Shenoy, Nat. Chem., 2010, 2, 581.
17. S. Cai, H. Duan, H. Rong, D. Wang, L. Li, W. He and Y. Li, ACS Catal., 2013, 3, 608.
18. J. Yang, X. Shen, Z. Ji, H. Zhou, G. Zhu and K. Chen, Appl. Surf. Sci., 2014, 316, 575.
19. L. Li, X. Chen, Y. Wu, D. Wang, Q. Peng, G. Zhou and Y. Li, Angew. Chem., Int. Ed., 2013, 52, 11049.
20. T. Sato, T. Iijima, M. Seki and N. Inagaki, J. Magn. Magn. Mater., 1987, 65, 252.
21. D. Leslie-Pelecky, Chem. Mater., 1996, 8, 1770.
22. G. Guella, B. Patton and A. Miotello, J. Phys. Chem. C, 2007, 111, 18744.
23. B. Liu and Z. Li, J. Power Sources, 2009, 187, 527.
24. J. Lv, A. Wang, X. M. R. Xiang, J. Chen and J. Feng, J. Mater. Chem. A, 2015, 3, 290.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed SEM and TEM images, XRD patterns, XPS, EDS, Raman spectra, gas chromatograms, TG analyses, UV-vis spectra, and reaction rate constant tables. See DOI: 10.1039/c5nr05016b

---