Electrospun Foamlike NiO/CuO Nanocomposites with Superior Catalytic Activity toward the Reduction of 4-Nitrophenol.

Foamlike NiO/CuO nanocomposites were prepared using a simple electrospinning technique combined with appropriate calcination. By tuning the Ni/Cu molar ratio (1:2, 1:1, and 2:1) in the initial material, different NiO/CuO nanocomposites were obtained. X-ray diffraction (XRD), field-emission scanning electron microscopy (FE-SEM), X-ray photoelectron spectroscopy (XPS), and nitrogen adsorption-desorption isotherms were used to characterize the composites. Furthermore, they were investigated as catalysts for the reduction of 4-nitrophenol (4-NP) in the presence of NaBH4. The test results demonstrate that the nanocomposite with Ni/Cu = 1:1 presents the best catalytic activity for its high content of surface oxygen vacancy and specific surface area.

Introduction

4-Nitrophenol (4-NP) is one of the major class of toxic pollutants extensively existing in agricultural and industrial effluent.[1] It has been marked as one of the most hazardous materials by the U.S. Environmental Protection Agency (EPA) since short-term inhalation of 4-NP would lead to headache, nausea, cyanosis, and drowsiness in humans.[2,3] Various methods have been reported to remove 4-NP efficiently, such as electrochemical oxidation methods,[4] photocatalytic reactions,[5] microorganism decomposition,[6] and microwave-assisted catalytic degradation.[7,8] The most direct and environment-friendly method is to use an appropriate catalyst to reduce 4-NP by NaBH4.[9−11] Moreover, the reaction product, 4-aminophenol (4-AP), is a useful intermediate to prepare other chemicals, such as analgesic drugs, antipyretics, polymers, and dyes.[12−14] Thus, developing an efficient material to reduce 4-NP to 4-AP is of great significance.

Noble metal nanoparticles (NPs), such as Ag, Au, Pd, and Pt, are major catalysts used due to their excellent activity.[15−17] However, they tend to aggregate and then decrease the catalytic activity. Loading the noble metal nanoparticles on a substrate is one route to avoid aggregation.[18,19] For example, Ishida et al. reported that Au nanoparticles deposited on poly(methyl methacrylate) showed high catalytic activity for the reduction of 4-NP with a rate constant in the range of (7.2–7.9) × 10–3 s–1.[18] Minami et al. prepared cellulose/Ag particles, whose rate constant for the catalytic reduction of 4-NP was 2.83 × 10–2 s–1, while that of cellulose-free Ag nanoparticles was 3.66 × 10–5 s–1.[19] In this way, aggregation is avoided and catalytic activity can effectively be improved, but the preparation is generally complex and noble metals are rare and very expensive.

Recently, doping or compositing on metal oxides was developed to substitute normal noble metal catalysts, such as Co-doped CuO, Cu@MnO2, CuO/TiO2, NiS–NiCo2O4@C, and so on, which generally have enhanced catalytic abilities.[10,20−23] For example, the rate constant for the reduction of 4-NP of 5.1% Co-doped CuO nanoparticles (k = 43.8 × 10–3 s–1) was fivefold higher than that of pure CuO NPs.[10] NiS–NiCo2O4@C composites exhibit better catalytic activity for the reduction of 4-NP and higher specific capacitance than the monometallic Co or Ni composites.[23] Therefore, formulating nanocomposites is an operative and efficient route to improve the catalytic properties. However, to prepare composite materials, multistep processes are generally employed.[24−28] Depositing one component on the as-prepared material is the most common one, and appropriate surface modification is necessary to ensure homogeneous deposition, which will induce some shortcomings, such as time consumption, high cost, and difficulty in controlling the amount of the loading component.

Electrospinning the metal/polymer precursor combined with appropriate thermal treatment can overcome the above problems. Electrospinning combined calcination is a simple and direct route to synthesize metal oxide, while the composition and its content can be easily tuned in the metal/polymer precursor.[29,30] In addition, the as-prepared product owns a porous structure. Generally, the catalytic performances could be enhanced by improving the interfacial properties between the catalyst and the reactants. The porosity and interconnectivity of the catalyst structures would be beneficial for the interface, for it could adsorb the maximum reactants, promote the electron transportation, and then facilitate an effectual catalytic performance. The nanofibers always own high specific surface area, porous structure, and strong interconnectivity, which favor the efficient adsorption of the reactant and effective electron transportation.[29−31]

Although NiO and CuO are easily available metal oxides, their preparation still has some disadvantages, such as big size of particles and easy aggregation.[32,33] Hence, nickel nitrate, copper nitrate, and poly(vinylpyrrolidone) (PVP) were selected as raw materials to prepare the Ni(NO3)2/Cu(NO3)2/PVP precursor, and the electrospinning technique was used to design porous NiO/CuO composites as the effectual catalyst to reduce 4-NP under mild conditions.

Results and Discussion

For the preparation of the electrospinning solution, Ni(NO3)2·6H2O and Cu(NO3)2·3H2O were used as raw materials with PVP as the spinning additive and ethanol and deionized water as solvents. After the electrospinning and calcination treatments, the product was obtained. As shown in Figure , the diffraction peaks are indexed to cubic NiO (JCPDS no. 47-1049) and monoclinic CuO (JCPDS no. 45-0937), and no other diffraction peaks indexed to impurities were detected. According to the peak shape parameter obtained by Rietveld refinement, the crystallite sizes for CuO and NiO can be estimated to be ca. 23.6 and 15.3 nm using the Debye–Scherrer formula. Meanwhile, a little large lattice parameter is estimated for cubic NiO (a = 4.181 Å) compared with that of bulk NiO (a = 4.177 Å) and the lattice parameter for CuO changes to a = 4.689 Å, b = 3.424 Å, and c = 5.119 Å in comparison with bulk CuO (a = 4.685 Å, b = 3.426 Å, and c = 5.130 Å). Further analysis of X-ray diffraction (XRD) data by the Rietveld method showed that the relative weight contents of NiO and CuO are 48.1 and 51.9%, so their corresponding molar ratio is very close to 1:1.

Figure 1

XRD patterns of the as-prepared product.

Besides, X-ray photoelectron spectroscopy (XPS) measurement was applied to detect the surface state of the as-prepared nanocomposite. The survey scan spectrum of the product (Figure a) shows Cu, Ni, O, and C elements on the surface. The O 1s spectrum (Figure b) contains three components, namely, lattice oxygen (OL, 529.6 eV), oxygen vacancy (OV, 531.6 eV), and surface adsorbed oxygen (OA, 533.6 eV), according to the literature.[10,34] The peak area ratios for OL, OV, and OA are 53.57, 46.04, and 0.39%, respectively. The two peaks located at 953.6 and 933.7 eV (Figure c) are ascribed to Cu 2p1/2 and Cu 2p3/2, respectively.[35] The peak of Cu 2p3/2 at 933.7 eV and the shake-up peak appearing at 938–945 eV were assigned to CuO.[36−38] As depicted in Figure d, the characteristic edges of Ni 2p1/2 and Ni 2p3/2 corresponded to NiO.[39] The main peaks at 872.2 eV (Ni 2p1/2) and 853.7 eV (Ni 2p3/2) with their satellite peaks locating at 879.4 and 860.9 eV indicate the existence of Ni2+ in NiO. The shoulder peak (855.5 eV) may be ascribed to the presence of the surface Ni2+ species.[40] The energy difference (18.5 eV) between Ni 2p3/2 and Ni 2p1/2 splitting demonstrates the well-defined symmetry of Ni2+ in the oxide form. Besides, the vibrations at 527 and 429 cm–1 in IR spectra (Figure S1) further confirm the existence of Cu–O and Ni–O bonds in the composite.

Figure 2

XPS spectra of survey scan (a), O 1s (b), Cu 2p (c), and Ni 2p (d) of the as-prepared products.

The morphology of the mixture was characterized by scanning electron microscopy (SEM). As can be seen from Figure a, the gel fibers exhibit homogeneous beltlike structures and their widths are in the range of 2.8 ± 0.6 μm. By observing the magnified image of single fibers (Figure b), the thickness was found to be lower than 1 μm (ca. 0.7 μm). After the calcination treatment, as shown in Figure c, the single beltlike structure disappeared and a three-dimensional foamlike structure with obvious pores formed, resulting from the removal of organics. By magnifying the observing area, it can be seen that the skeleton of the foamlike structure is coarse and composed of nanoparticles (Figure d). The cross-sectional image also confirms the presence of the porous foamlike structure (Figure e,f). The observation of the product treated for a shorter time reveals that pores are obvious on the beltlike structure, indicating that gas bubbles may form during the calcination. It is speculated that the decomposition of PVP and the nitrate in the gel precursor fiber leads to the destruction of the connected beltlike structure and the gas bubbles lead to the appearance of macropores during heat treatment.

Figure 3

SEM images of the electrospun gel fibers (a, b); top-view (c, d) and cross-sectional view of the calcined product (e, f).

Transmission electron microscopy (TEM) was also used to characterize the morphology of the product. From Figure a, irregular macroporous skeletons were observed, which is consistent with the SEM image (Figure c). The magnified image (Figure b) of the peripheral area shows that the nanoparticles are of irregular shapes with a wide size range (10–30 nm). They connected loosely with each other, resulting in plenty of small pores. The high-resolution TEM (HR-TEM) image (Figure c) presents continuous parallel lattice fringes on a single particle, indicating its good crystallinity. The lattice spacing of 0.23 nm corresponds to the {111} plane of CuO, and the other (0.20 nm) is consistent with the {200} interplanar spacing of NiO. By SEM and TEM characterization, the product presents a porous foamlike structure and is composed of small nanoparticles, which may own a high specific surface area and many active sites when used as a catalyst.

Figure 4

TEM (a, b) and HR-TEM (c) images of the product.

SEM–energy-dispersive spectrometry (EDS) mapping was performed to clarify the content and homogeneity of the component in the mixture. The EDS spectrum in Figure a presents Ni, Cu, and O elements in the area of Figure b. The elemental content table (Table S1) shows that the molar ratio of Ni/Cu is nearly 1:1, which is very close to the initial additive content. The mapping result (Figure c) demonstrates that every element is homogeneously distributed. From the above analysis, it can be found that the as-prepared foamlike product is the NiO/CuO composite with a molar ratio of 1:1, and every element dispersed well in the product.

Figure 5

EDS spectrum (a); SEM image (b); and EDS mapping results of all elements (c), oxygen element (d), nickel element (e), and copper element (f). The scale bars are 200 nm.

To investigate the effect of the Ni/Cu molar ratio on the morphology, the initial Ni/Cu molar ratio was varied from 1:2, 1:1 to 2:1. After the electrospinning process, it was observed that beltlike structures could be obtained although the molar ratio of the raw material was tuned (Figure a,c,e). As to Ni/Cu = 1:2, the image of which is shown in Figure a, the width of the beltlike fibers is in the range of 2.0 ± 0.5 μm, but the corresponding calcined products did not retain the beltlike structure and are fused together with small amount of pores (Figure b). Upon increasing the amount of NiO to an equimolar amount of CuO, the width of the fiber increased to 2.8 ± 0.6 μm (Figure c) and after calcination treatment, a three-dimensional foamlike structure formed with more pores than that of Ni/Cu = 1:2 (Figure d). When further increasing the amount of NiO to Ni/Cu = 2:1, the width of the fiber is 3.6 ± 1.3 μm (Figure e), while the calcined product also presents a three-dimensional structure with large pores (Figure f). By tuning the initial Ni/Cu molar ratio, it is found that upon increasing the amount of NiO, the beltlike structures of the gel fibers were not altered, but their widths increased, while the corresponding calcined product exhibits a three-dimensional porous structure, but the pore sizes differ obviously. The Brunauer–Emmett–Teller (BET) specific surface areas for Ni/Cu = 1:2, 1:1, and 2:1 are 25.3, 45.1, and 29.1 m2/g, respectively (Figure S3), which are higher than that reported for the electrospinned CuO/NiO composite nanofibers.[32] Therefore, they may exhibit different catalytic activities when used as catalysts.

Figure 6

SEM images of the gel fibers and the corresponding calcined products: Ni/Cu = 1:2 (a, b), Ni/Cu =1:1 (c, d), and Ni/Cu =2:1 (e, f).

The catalytic activities of the products were evaluated with the reduction of 4-NP into 4-AP by excess NaBH4, which was monitored by UV–vis absorption for a time interval of 1 min. As shown in Figure a, the mixture reaction solution without a catalyst presents a typical absorption at 400 nm and the absorbance did not weaken for 30 min, indicating that the reaction occurred slowly without a catalyst. When the prepared catalyst (Ni/Cu = 1:1) was introduced into the system, the reaction mixture changed from bright yellow to colorless within 3 min and a new absorption peak appeared at 300 nm (Figure b), which is the characteristic absorption of 4-AP. Because of the high kinetic barrier between 4-NP and BH4–, the reduction of 4-NP to 4-AP by NaBH4 generally occurs slowly.[41,42] The color of the reaction mixture fades fast, and the appearance of a new absorption peak indicates that the reduction of 4-NP by NaBH4 is catalyzed in the present reaction system. When the catalysts (Ni/Cu = 2:1 and 1:2) were introduced into the reaction, similar spectral changes were observed (Figure c,d). With NiO/CuO (Ni/Cu = 1:2) as a catalyst, the reaction was complete in 8 min (Figure c), and as to NiO/CuO (Ni/Cu = 2:1), the total time is 12 min (Figure d). The turn over frequencies (TOFs) of NiO/CuO (Ni/Cu = 1:1, 1:2, and 2:1) were calculated to be 2.0 × 1019, 7.5 × 1018, and 5.0 × 1018 molecules/g/s, respectively, which is higher than those for the reported Au and Ag nanocatalysts.[15] Pure NiO and CuO (Figure S4) were also prepared using the same procedure as the NiO/CuO composites for comparison. When pure NiO was added, the reduction was complete in ca. 25 min (Figure e), while when CuO was used, the reaction nearly did not change in 16 min (Figure f). In all, the catalytic activity of the samples follows the order: Ni/Cu = 1:1 > Ni/Cu = 1:2 > Ni/Cu = 2:1 > NiO > CuO. As the concentration of NaBH4 largely exceeds that of 4-NP, the reaction was of zero order with NaBH4 and followed a pseudo-first-order rate to the concentration of 4-NP.[43,44] Thus, the kinetic information can be calculated from the following equationsC is the concentration of 4-NP, t denotes the reaction time, C0 and C denote the concentration of 4-NP at the initial and that at time t, and kapp stands for the apparent rate constant. From eq , kapp can be easily calculated from the slope of the fitting curve of ln(C/C0)–t. The higher kapp is, the faster the reaction proceeds. The (C/C0)–t plots of all of the samples and the corresponding ln(C/C0)–t plots are shown in Figure g,h. Based on the fitted curves, their apparent rate constants were calculated and are shown in Table . The catalytic activity of the prepared NiO/CuO sample (Ni/Cu = 1:1) is better than that reported in the literature, even including some noble metal catalysts.[44−48] The superior catalytic properties might result from its unique structure and high specific surface area.

Figure 7

Catalytic activity for the reduction of 4-NP. (a) UV–vis spectra of the reaction solution before and after the catalytic reaction. (b–f) Time-resolved UV–vis spectra of the catalytic reaction in the presence of different catalysts. (g) Plot of C/C0 versus time for the catalytic reduction of 4-NP. (h) Plot of ln(C/C0) versus time for the catalytic reduction of 4-NP.

Table 1

Apparent Rate Constants (kapp) of Different Catalysts for the Reduction of 4-NP

catalyst	concentration of catalyst (mg/mL)	concentration of 4-NP (mmol/L)	kapp (min–1)	reference
NiO/CuO (Ni/Cu = 1:1)	0.017	0.1	1.12	this work
NiO/CuO (Ni/Cu = 1:2)	0.017	0.1	0.45	this work
NiO/CuO (Ni/Cu = 2:1)	0.017	0.1	0.30	this work
NiO	0.017	0.1	0.11	this work
CuO	0.017	0.1	0.0033	this work
cabbage type CuO nanostructure	not noted	0.0083	0.371	(44)
cabbage type Cu2O nanostructure	not noted	0.0083	0.587	(44)
CuO nanosheet	0.16	0.02	0.264	(45)
Pd@Au CSNTPs	0.0017	0.0067	0.14	(46)
NiO/TiO2/SiO2/Au	0.0076	0.38	0.66	(47)
NiO/C	0.5	2.5	2.52	(48)
CuO/NiO@C hollow sphere	0.017	0.1	1.5	(43)

XPS is an effective route to detect the surface state and evaluate the amount of oxygen vacancies. Figure presents the XPS spectra of O 1s of NiO/CuO (Ni/Cu = 1:2 and 2:1) and they were also deconvoluted into three peaks according to the literature.[41] The atomic ratio of OL/M (OL, lattice oxygen, M = Cu + Ni) was used to determine the surface oxygen vacancies.[49] For the three samples, the metal molar mass is the same, so the higher the lattice oxygen (OL), the lower the ability to form oxygen vacancies. As shown in Figures b and 8, the OL content follows the order of Ni/Cu = 2:1 > Ni/Cu = 1:2 > Ni/Cu = 1:1, indicating that the ability to form surface oxygen vacancies increases in the order Ni/Cu = 2:1 < Ni/Cu = 1:2 < Ni/Cu = 1:1.

Figure 8

O 1s XPS spectra of NiO/CuO: (a) Ni/Cu = 2:1 and (b) Ni/Cu = 1:2.

The Langmuir–Hinshelwood model is generally accepted for the reduction mechanism of 4-NP.[50,51] First, BH4– adsorbs on the surface of the foamlike NiO/CuO nanocomposite and reacts with the surface to generate active hydrogen species. Then, 4-NP diffused into the surface of the catalyst and reacted with the produced hydrogen species to form 4-AP. Finally, the reduction product, 4-AP, desorbs from the foamlike NiO/CuO nanocomposite into the reaction solution. So the surface state of the catalyst is a critical factor to affect its activity. It is reported that the reduction of 4-NP in the presence of a catalyst is greatly dependent on its oxygen vacancies, for the oxygen vacancies generally formed as VO2+ for its lowest formation energy.[52,53] Therefore, the existence of oxygen vacancies would result in the formation of the positive surface charge, which could enable the catalyst to adsorb 4-NP and BH4– anions easily, and thus promote the catalytic reduction process. The more the oxygen vacancies on the catalyst, the higher the catalytic activity.[53] Moreover, the special foamlike structure also contributes to its catalytic activity. On one hand, the porous structure results in a high specific surface area, which can supply many active sites. On the other hand, the porosity and interconnectivity of the foamlike catalyst would be beneficial for adsorbing the maximum reactants and promoting electron transportation. Therefore, the highest catalytic activity of the nanocomposite (Ni/Cu = 1:1) mainly stems from its highest content of the surface oxygen vacancy and special structure.

The stability is another factor to evaluate the catalyst in the practical application, so the recyclability test was performed for the NiO/CuO composite (Ni/Cu = 1:1), and the conversion was determined for every cycle. From Figure , it can be seen that 10 cycles of the catalytic reduction of 4-NP were conducted, and the corresponding conversion rate for every cycle is higher than 97.4%. Besides, the SEM image of the nanocomposite in Figure S5 shows that nearly no changes occurred on its porous structure, indicating that the catalyst owns high stability. Although the catalyst needs recycling by centrifugation, its superior catalytic activity and stability make it potential for practical application.

Figure 9

Recyclability of the catalyst for the reduction of 4-NP.

Conclusions

Three-dimensional foamlike NiO/CuO nanocomposites were easily prepared by the electrospinning technique combined with appropriate calcination. The Ni/Cu molar ratio in the nanocomposites can be easily tuned by changing the amount of the raw materials in the precursor. Furthermore, the effect of the Ni/Cu molar ratio on their morphologies and catalytic activities was systematically studied. The nanocomposite (Ni/Cu = 1:1) has the highest specific surface area and surface oxygen vacancies, which make it present the best catalytic reduction of 4-NP. Besides, the nanocomposite (Ni/Cu = 1:1) also has superior stability for its special structure.

Experimental Section

Synthesis of NiO/CuO Nanocomposites

All chemicals were purchased and used directly, and the information of all of the chemicals are listed in Table S2. For the synthesis of the NiO/CuO nanocomposite with a molar ratio of 1:1 (MNi/MCu = 1:1), 1.0 g of poly(vinylpyrrolidone) (PVP, Mw = 1 300 000) was dissolved in 10 mL of ethanol with magnetic stirring to dissolve PVP completely, and then, 2 mL of an aqueous solution containing 0.45 g of Ni(NO3)2·6H2O and 0.36 g of Cu(NO3)2·3H2O was added. The mixture was magnetically stirred overnight to make it homogeneous and then transferred to a needle tube with an inner diameter of 0.5 mm. The electrospinning voltage was set as 20 kV, the receiving distance between the needle and the collector was 20 cm, and the pumping speed was 0.003 mm/s. After the electrospinning process, the matlike product was collected and heated at 80 °C overnight. Finally, the dried matlike product was heated in a muffle furnace to 450 °C with a heating rate of 1 °C/min and was kept at that temperature for 2 h. In addition, by varying the amount of Ni(NO3)2·6H2O or Cu(NO3)2·3H2O with the other constant, pure NiO, CuO, and NiO/CuO nanocomposites with other molar ratios (MNi/MCu = 1:2 and 2:1) were obtained using the same synthesis procedure.

Material Characterization

A scanning electron microscope (SEM, Hitachi, Regulus 8220) with an accelerating voltage of 5 kV and a transmission electron micrcoscope (TEM, JEOL, JEM-2100) with an accelerating voltage of 200 kV were employed to characterize the morphology and structure of all samples. The sample phase was determined using a Bruker D8 Advance X-ray diffractometer with graphite monochromated Cu Kα (λ = 0.15418 nm) radiation, and the scan range is from 30 to 80°. X-ray photoelectron spectroscopy (XPS) was performed on a Thermo XPS ESCALABXi+ instrument using Al Kα as the X-ray source and corrected with the C 1s line at 284.8 eV. V-Sorb2800p was employed to measure the nitrogen adsorption–desorption isotherms of the nanocomposites at liquid nitrogen temperature (77 K). The absorbance was detected by a Shimadzu UV-2450PC ultraviolet–visible spectrometer at room temperature. Fourier transform infrared (FT-IR) spectra were collected on a Nicolet iS 20 spectrometer (ThermoFisher Scientific) in the range of 400–4000 cm–1.

Catalytic Activity Measurements

The reduction of 4-NP by NaBH4 was employed to estimate the catalytic activity of the sample. It was conducted in a quartz cuvette and detected by UV–vis spectroscopy at room temperature. The NiO/CuO nanocomposite was ultrasonicated in deionized water to obtain a stock suspension (1.0 mg/mL). Typically, 9.5 mg of NaBH4 (0.25 mmol) was added to an aqueous solution of 4-NP (3.0 mL, 0.10 mM), while the color of the solution changed to deep yellow. Then, 50 μL of NiO/CuO stock solution was added. The reaction process was monitored by UV–vis absorption spectroscopy. For the recyclability test, 10 recycles of the activity were measured for the NiO/CuO nanocomposite. After the first run, 0.25 mmol of NaBH4 (9.5 mg) and 10 μL of 4-NP (30 mM) were directly added into the reaction solution, and the same procedure was repeated for the next eight runs. Besides, the TOF was calculated from the total molecule number of 4-NP converted per weight of the catalyst per second.[54]