Fast electrodeposition, influencing factors and catalytic properties of dendritic Cu–M (M = Ni, Fe, Co) microstructures

Abstract

The rapid electrochemical deposition of dendritic Cu–M (M = Ni, Fe, Co) microstructures with excellent catalytic activity is reported here. Simple Cu²⁺ and M²⁺ salts were employed as the initial metal sources and boric acid was used as the buffer solution. The electrodeposition process was carried out at a deposition current of 10 mA for 5 min in air at room temperature. The phase and morphology of the as-prepared products were characterized by field emission scanning electron microscopy (FESEM), powder X-ray diffraction (XRD), energy dispersive spectrometry (EDS) and transmission electron microscopy (TEM). It was found that the formation of dendritic Cu–M microstructures could be affected by some factors including the amounts of M²⁺ salts and boric acid, and the deposition current and time. Cu–Ni dendrites were used as a model and their performance was studied. The investigations showed that the as-deposited Cu–Ni dendrites exhibited good electrochemical responses in 0.1 M KOH solution and could be used as an electrochemical catalyst for the reduction of nitrate and the oxidation of glucose. Also, the as-deposited Cu–M dendrites exhibited excellent catalytic activities for the reduction of 4-nitrophenol (4-NP) to 4-aminophenol (4-AP) in excess NaBH₄ solution.

---

1. Introduction

Since the properties of nanomaterials can be tuned by their sizes and shapes, realizing their controllable syntheses is significant. Compared with other structures, branched nanostructures generally have the larger surface areas, allow for heterostructures and can easily form continuous networks. Hence, the preparation of dendritic micro/nanostructures has been paid much attention over the past decade.¹ Many methods have been developed, including the hydrothermal method,² the template-assisted route,³ laser-driven growth,⁴ rapid circumfluence synthesis,⁵ electrochemical deposition,⁶ the wet chemical approach,⁷ and so on. Among them, electrodeposition technology offers a simple, fast and energy-efficient method. Also, the electrodeposition process for the synthesis of dendrites can be carried out at ambient pressure and temperature, requiring relatively inexpensive equipment compared with other methods. By employing this technology, some elemental and compound dendrites have been successfully obtained in our groups.⁸

Transition metal elements, especially iron, cobalt and nickel, always attract extensive research interest in materials science due to their outstanding physicochemical, mechanical, magnetic and catalytic performances, and wide applications in many fields. With the development of nano-science/technology, Fe, Co, Ni and their alloys with various morphologies have been obtained through a variety of methods.⁹ As the element following nickel, copper also draws much attention owing to its wide commercial applications in a variety of fields in heat transfer materials, antimicrobial materials, super strong materials, sensors and catalysts, etc.¹⁰ Furthermore, Cu can easily form alloys with many metals including Fe, Co and Ni, and Cu–M (M = Fe, Co and Ni) micro/nanostructures exhibit special magnetic or magnetoresistive properties.¹¹ Recently, nanostructural Cu–Ni alloys with various atomic ratios have drawn much attention because of their excellent catalytic activity for nitrate reduction, furfural hydrogenation, and methane reforming with CO₂.^12–18 For example, Mattarozzi and coworkers deposited Cu–Ni alloys from a single citrate bath through a potentiostatic mode and found that the as-obtained Cu–Ni alloys presented superior performances in the reduction of nitrate and nitrite when compared to pure Cu and Ni.¹⁹ Subsequently, a hydrogen evolution assisted potentiostatic electrodeposition route was also employed for the preparation of porous Cu–Ni alloy foams, which exhibited outstanding performances including large stable currents, and fast and selective nitrate reduction to ammonia.¹⁸ Similarly, Zhang et al. potentiostatically deposited porous Cu–Ni foam films and investigated their magnetic properties and superhydrophobicity. Experiments showed that Cu–Ni porous films displayed superior stability and enhanced electrocatalytic activity for hydrogen evolution reduction when compared to pure Cu and Ni porous films.²⁰

In contrast to the above potentiostatic deposition route, in this work, we designed a simple galvanostatic deposition route to prepare dendritic Cu–M (M = Fe, Co and Ni) microstructures from a H₃BO₃ solution under ambient conditions, employing CuCl₂ and MCl₂ as the initial metal sources. Some factors affecting the formation of dendritic Cu–M microstructures were systematically investigated, including the deposition time, the amounts of M²⁺ salts and boric acid, and the deposition current. Experiments showed that the presence of M²⁺ salts played a crucial role in the formation of dendritic Cu–M microstructures. Moreover, the performance tests showed that the deposited dendritic Cu–M microstructures presented excellent catalytic activities for the reduction of 4-nitrophenol (4-NP) to 4-aminophenol (4-AP) in excess NaBH₄ solution. It was found that dendritic Cu–Ni microstructures still presented outstanding electrocatalytic abilities for the reduction of nitrate ions (NO₃⁻) to ammonia (NH₃) and the oxidation of glucose in KOH solution.

2. Experimental

All reagents and chemicals were analytically pure, bought from Shanghai Chemical Company and used without further purification.

2.1 The deposition of dendritic Cu–M microstructures

In a typical experiment, a conventional three-electrode cell was used, employing a Pt wire as the counter electrode, a saturated Ag/AgCl electrode as the reference electrode, and a conductive ITO (indium–tin oxide) glass with a size of 1 × 3 cm² as the working electrode. Before the experiment, the ITO electrode was treated by ultrasound for 30 min in turn in acetone, ethanol and twice-distilled water, and then dried at room temperature. To obtain dendritic Cu–M microstructures, 1 mmol of CuCl₂ was firstly dissolved in a small volume of twice-distilled water. Then, 9 mmol MCl₂ (M = Fe, Co and Ni) and 1 mmol of H₃BO₃ were introduced to the above solution. Subsequently, the system was diluted to 30 mL by adding twice-distilled water. Here, the concentrations of Cu²⁺, M²⁺ and H₃BO₃ were in turn 0.033, 0.3 and 0.033 mol L⁻¹. Finally, the electrodeposition experiments were conducted at a deposition current of 10 mA for 5 min in air at room temperature.

2.2 Characterization

The X-ray diffraction (XRD) patterns of the electrodeposited products were recorded on a Shimadzu XRD-6000 X-ray diffractometer (Cu Kα radiation, λ = 0.154060 nm), employing a scanning rate of 0.02° s⁻¹ and 2θ ranges from 30° to 80°. Field emission scanning electron microscopy (FESEM) images and energy dispersive spectra (EDS) of the final products were taken on a Hitachi S-4800 field emission scanning electron microscope, employing an accelerating voltage of 5 or 15 kV (15 kV for EDS). High resolution transmission electron microscopy (HR/TEM) images were recorded on a FEI Tecnai G²⁰ transmission electron microscope, employing an accelerating voltage of 200 kV. X-ray photoelectron spectroscopy (XPS) of the products was conducted on a Thermo ESCALAB 250 instrument, employing monochromatic Al Kα (hν = 1486.6 eV) at a power of 150 W.

2.3 Performance tests

To investigate the catalytic properties of the as-deposited dendritic Cu–M microstructures for the reduction of 4-NP to 4-AP, a series of solutions was freshly prepared. Firstly, a certain amount of catalyst, 0.012 mg, was dispersed into small amounts of 4-NP solution. Then, NaBH₄ solution was introduced into the above system. The total volume of the system was adjusted to 3 mL. Here, the concentrations of 4-NP, NaBH₄ and catalyst were 1.0 × 10⁻⁴ mol L⁻¹, 2 × 10⁻² mol L⁻¹ and 4 mg L⁻¹, respectively. The reduction processes were monitored with a Metash 6100 UV-vis spectrophotometer.

In parallel, to investigate the electrochemical properties of Cu–Ni dendrites for the reduction of NO₃⁻ to NH₃ and the oxidation of glucose to gluconolactone in KOH solution, a Ag/AgCl (in saturated KCl, aq.) reference electrode, a platinum coil (0.5 mm × 4 cm) counter electrode and a modified glass carbon working electrode (GCE) were employed. The working electrode was prepared as follows: firstly, a GCE was polished in 1700# diamond paper with 1 mm, 0.3 mm, 0.05 mm alumina in turn, and washed successively with 0.1 M HNO₃ and ethanol in an ultrasonic bath. Secondly, 1 mg of the dendrite was dispersed into 1 mL of twice distilled water under sonication. Next, 5 μL of the dendrite solution was dropped onto the surface of the GCE. Subsequently, 3 μL of 0.5 wt% Nafion solution was dropped on the surface to tightly attach the dendrite onto the surface of the electrode. Thus, the GCE modified by Cu–Ni dendrites was obtained. The reduction process of NO₃⁻ to NH₃ and the oxidation process of glucose to gluconolactone were monitored with a computer-controlled CHI660D electrochemical workstation.

3. Results and discussion

3.1 Morphology and structure characterization

Fig. 1 shows the representative FESEM images of the products deposited from CuCl₂–H₃BO₃ systems without and with M²⁺ salts (M = Fe, Co or Ni) under the stated experimental conditions. Obviously, when the system did not contain M²⁺ salts, no dendrite was deposited (see Fig. 1a). After introducing M²⁺ salts into the system, dendritic microstructures were obtained (see Fig. 1b–d). Namely, the presence of M²⁺ salts caused the growth of dendritic microstructures under the given electrodeposition conditions. Since the deposited products could be moved by a magnet, it was clear that magnetic material could be produced in the present electrodeposition process. Considering the system composition and the electro-reduction conditions, it was possible that the deposited products contained a magnetic metal, Fe, Co or Ni. Fig. 2a–c depicts the EDS analyses of the products deposited from CuCl₂–H₃BO₃ systems with M²⁺ salts (M = Fe, Co or Ni). Besides the Cu peak, strong Fe, Co or Ni peaks are also detected in the respective samples, indicating that the final products indeed contain Fe, Co or Ni. However, according to the calculations of the peak areas, the Cu/M molar ratio is almost 4/1 for Cu/Co, 2/1 for Cu/Fe, and 1/3 for Cu/Ni (see the insets in Fig. 2a–c). The above facts imply that the deposited amounts of magnetic metals are different under the same electrodeposition conditions owing to the difference in the original metal sources. Furthermore, the amounts of elemental C are similar in the three samples, which could be attributed to the double-sided-tape. However, the amounts of elemental O differ in the samples. The amount of elemental O in the Cu–Fe sample is the highest, that in the Cu–Co sample is intermediate, and the Cu–Ni sample contains the least. The above phenomena are understandable. In the light of the standard potentials, the reductive abilities of Fe (E^θ = −0.44 V NHE), Co (E^θ = −0.28 V NHE) and Ni (E^θ = −0.25 V NHE) weaken in turn. Nanosized Fe is easily oxidized in air whereas Ni is the most stable. Hence, the amount of elemental O is different in the three samples. Fig. 2d depicts the XRD patterns of the products. According to the order Co, Fe, Ni, the intensities of cubic Cu peaks gradually weaken and those of M peaks increase, indicating that the amount and the crystallinity of M increase in turn. The above result is in good agreement with the Cu/M molar ratio (see the insets in Fig. 2a–c). However, no obvious Co peak is detected in the Cu–Co sample, implying that a small amount of Co was deposited. In 1998, Bakkaloǧlu et al. found that the diffraction peak of Co could not be detected from Cu–Co alloy films with a cobalt content from 6% to 26%.²¹ Based on the EDS calculation, ∼20% Co existed in the present Cu–Co sample. Thus, no obvious Co peak is detected in the Cu–Co sample.

Fig. 1 FESEM images of the samples deposited from CuCl₂–H₃BO₃ systems without and with M²⁺ (M = Fe, Co and Ni) salts at the deposition current of 10 mA for 5 min in air at room temperature: (a) without M²⁺ ions, (b) with Fe²⁺ ions, (c) with Co²⁺ ions and (d) with Ni²⁺ ions.

Fig. 2 (a–c) EDS analyses and (d) XRD patterns of the products deposited by CuCl₂–H₃BO₃ systems with M²⁺ (M = Fe, Co and Ni) salts at the deposition current of 10 mA for 5 min in air at room temperature.

To further ascertain the state of M in the surface of Cu–M alloys, XPS technology was employed. As shown in Fig. 3a and b, no Fe(0) and Co(0) are detected.^22,23 However, a Ni(0) peak can still be detected in the Cu–Ni sample (see Fig. 3c).²⁴ The above facts indicate that Fe and Co in the surfaces of Cu–Fe/Co samples have been oxidized but Ni is only partially oxidized.

Fig. 3 High resolution XPS of (a) Fe, (b) Co and (c) Ni in the three samples.

Here, Cu–Ni dendrites are selected as the representative model for further investigation since Ni is more stable than Fe and Co. Fig. 4a depicts a high-magnification FESEM image of Cu–Ni dendrites, from which some sub-branches of dendrites are clearly visible. TEM observations further prove the dendritic structure of the Cu–Ni sample. As shown in Fig. 4b, all stems are constructed by nearly-spherical nanoparticles with a mean size of ∼50 nm, like knotted straight strings. The sub-branches grow from these knots and are almost parallel to each other. Similar phenomena were found during the electrodeposition process of PbTe.²⁵ The right inset in Fig. 4b shows a HRTEM image of the tip of a stem. Clear lattice fringes imply good crystallinity of the dendritic structures. The d-spacing value of neighbouring planes is calculated to be ∼0.21 nm, which is close to the (111) plane of Cu(Ni). The corresponding SAED pattern confirms the single crystal nature of the sample (see the left inset in Fig. 4b).

Fig. 4 (a) High magnification FESEM image and (b) TEM image of Cu–Ni dendrites deposited from the system with 1 mmol CuCl₂, 9 mmol NiCl₂ and 1 mmol H₃BO₃ at the deposition current of 10 mA for 5 min. The insets in (b) show the SAED pattern (left) and the HRTEM image (right) of the stem.

3.2 Influencing factors

3.2.1 The original amounts of NiCl₂ and boric acid. In the present system there were only three components besides the solvent: CuCl₂, NiCl₂ and H₃BO₃. Keeping the amount of CuCl₂ constant, experiments uncovered that the original amounts of NiCl₂ and boric acid could markedly affect the morphology of the Cu–Ni microstructures.

Fig. 5 exhibits typical FESEM images of the products deposited from systems with different original amounts of NiCl₂. Dendritic structures can always be obtained under the same deposition conditions but with the increase of the original NiCl₂ amount, the sizes of the dendrites gradually increase. Furthermore, the dendrites gradually become perfect with an increase in the amount of NiCl₂ from 0.1 mmol to 9 mmol; then, retrograde from 9 mmol to 15 mmol. Namely, 9 mmol of NiCl₂ is the optimum amount for the formation of perfect Cu–Ni dendrites.

Fig. 5 FESEM images of Cu–Ni microstructures deposited at the deposition current of 10 mA for 5 min from systems containing different amounts of NiCl₂: (a) 0.1 mmol, (b) 2 mmol and (c) 15 mmol.

As a buffering reagent, boric acid is often used during electrodeposition. In the present system, when no boric acid was used, the deposited product consisted of abundant irregular particles and few immature dendrites under the same experimental conditions (see Fig. 6a). After 0.5 mmol of boric acid was introduced into the system, the product presented mostly dendritic microstructures (see Fig. 6b) except for a few sheet-like particles (see the red circle in Fig. 6b). After 1 mmol of boric acid was employed, a great number of perfect dendrites were obtained (see Fig. 1d and 4). Upon further increasing the amount of boric acid, for example to 3 mmol, the dendrites continued to grow; however, the sub-structures of the dendrites were gradually lost (see Fig. 6c). Namely, perfect dendritic-structures retrograded with increasing amounts of boric acid. The above facts clearly showed that the amount of boric acid also plays an important role in the formation of perfect Cu–Ni dendrites. Some studies have shown that boric acid can lower over-potential and increase current efficiency, which promotes the electrodeposition of metals.^26,27 Recently, Graham et al. investigated the role of boric acid in the formation of Ni nanotubes under anodic aluminium oxide (AAO) template-assisted electrodeposition conditions, and believe that a nickel–borate complex, formed between Ni²⁺ and borate ions, plays the crucial role in the formation of Ni nanotubes.²⁷ Also, their research discovered that the concentration of boric acid could tune the wall thickness of nanotubes.²⁷ Obviously, in the present work, metal–borate complexes could be formed, too. Under the same electrochemical conditions, the deposition of metal (Fe, Co and Ni) was promoted. Within the same deposition time, more metal particles were produced. The previously-produced metal nuclei became the seeds for further growth of freshly deposited particles. Since borate ions (or boric acid molecules) still existed in the surroundings of the deposited metal nuclei, it was possible that they affected the nucleation and growth of the product. Namely, boric acid probably acted as the structure-directing reagent. As a result, dendritic Cu–Ni microstructures were obtained.

Fig. 6 FESEM images of Cu–Ni microstructures deposited at the deposition current of 10 mA for 5 min from systems containing different amounts of H₃BO₃: (a) 0 mmol, (b) 0.5 mmol and (c) 3 mmol.

3.2.2 The deposition time. To further ascertain the growth process of Cu–Ni dendrites, a time-dependent shape evolution experiment was designed. As shown in Fig. 7a and its inset, a few flowerlike microstructures were scattered in abundant small particles after deposition for 30 s. After 1 min, certain petals of micro-flowers rapidly grew to form dendritic microstructures. Simultaneously, more small particles grew into new flowerlike microstructures (see Fig. 7b). Dendritic microstructures continuously grew up with the prolongation of the deposition time. After deposition for 3 min, dendritic microstructures became the main product (see Fig. 7c). When the deposition time of 5 min was employed, plenty of perfect dendrites were obtained (see Fig. 1d and 4). Hereafter, the shape of the product gradually retrograded with the expansion of the deposition time. For example, Fig. 7d displays a typical FESEM image of the product deposited at the current of 10 mA for 10 min. Dendritic microstructures are still clearly visible, but the sizes have become bigger and the sub-branches are coarser compared with those of microstructures deposited for 5 min.

Fig. 7 FESEM images of Cu–Ni microstructures deposited under the same conditions for different durations: (a) 30 s, (b) 1 min, (c) 3 min and (d) 10 min.

The above shape evolution experiment clearly shows the formation process of dendritic microstructures. In the initial stage, abundant small particles were obtained and acted as the seeds for further growth. Subsequently, flowerlike microstructures were rapidly formed. With the expansion of the deposition time, the flowerlike microstructures grew preferentially along certain directions owing to their high surface energies. Thus, dendritic microstructures started to appear and gradually became more perfect. With the elongation of the deposition process, however, perfect dendrites gradually retrograded, which should probably be attributed to the growth rate change of various crystal planes.

3.2.3 The deposition current. Furthermore, it was found that the deposition current could affect the morphology of the product. Fig. 8a and b shows the representative FESEM images of the products deposited from the same system for 5 min at deposition currents of 5 mA and 20 mA. Compared with the FESEM images shown in Fig. 1d (deposition current of 10 mA), it is clear that although dendritic microstructures can be still obtained, the dendrites obtained at 20 mA are bigger in size and markedly retrograde. Namely, the higher deposition current is unfavourable to the growth of perfect dendrites. This is understandable when the other conditions are kept constant, the deposition rate increases with the increase of the deposition current. Since the nucleation and growth rates of the product are determined mainly by the deposition rate, more products will be deposited under the higher deposition current within the same time. Simultaneously, the nucleation and growth rates increase, which lead to the morphology change of the final product observed in the present work.

Fig. 8 FESEM images of Cu–Ni microstructures deposited from the system with 1 mmol CuCl₂, 9 mmol NiCl₂ and 1 mmol H₃BO₃ at different currents for 5 min: (a) 5 mA and (b) 20 mA.

In 1963, Brenner pointed out that the electrodeposition of alloys containing one or more of iron, cobalt or nickel often takes place in an anomalous codeposition mode, which is characterized by the anomaly that the less noble metal deposits preferentially.²⁸ In the present system, the concentration of M²⁺ ions (M = Fe, Co or Ni) was 9 times higher than that of Cu²⁺ ions. The redox potential of the M²⁺/M pair was calculated to be in turn 0.296 V for Cu²⁺/Cu, −0.265 V for Ni²⁺/Ni, −0.295 V for Co²⁺/Co and −0.455 V for Fe²⁺/Fe. The Cu²⁺/Cu pair bears a markedly higher redox potential than other M²⁺/M pairs. Fewer Cu²⁺ ions should be deposited prior to M²⁺ ions. It follows that the electrodeposition of the present Cu–M systems should comply with the above anomalous codeposition mode. Thus, Cu–M dendrites were finally deposited.

3.3 Properties of Cu–M microstructures

3.3.1 Catalytic activities for the reduction of 4-nitrophenol. To investigate the catalytic activities of Cu-M (M = Fe, Co and Ni) dendrites, the reduction of 4-NP by excess NaBH₄ to 4-AP in aqueous solution was used as the model reaction. Generally, 4-NP solution has a strong absorption peak at 317 nm and a weak shoulder peak at 400 nm in the region of 250–550 nm (see the thick curve in Fig. 9a). However, the absorption peak at 317 nm will disappear after alkali NaBH₄ is added into the 4-NP solution. Here, only one peak at 400 nm exists and the peak intensity markedly increases, which should be attributed to the production of the intermediate state, corresponding to the 4-nitrophenolate ion.²⁹ If a catalyst is introduced into the above system, the peak at 400 nm gradually decreases and concomitantly, a new peak at ∼305 nm appears due to the production of 4-AP.³⁰ Fig. 9a–c shows the UV-vis absorption spectra of the 4-NP–NaBH₄ system in the presence of 4 mg L⁻¹ Cu–M dendrites at various reaction times. With the prolonging of the reaction time, the peak at 400 nm decreases, indicating the good catalytic activities of the as-deposited Cu–M dendrites for the reduction of 4-NP to 4-AP in excess NaBH₄ solution. Fig. 9d depicts the linear relationships between ln(C_t/C₀) and the reaction time in the presence of the as-deposited Cu–M dendrites. Here, C₀ and C_t stand for the initial concentration of 4-NP and the concentration of 4-NP at reaction time t, respectively. The corresponding rate constants are calculated to be 0.83 min⁻¹ for Cu–Ni, 0.69 min⁻¹ for Cu–Fe and 0.48 min⁻¹ for Cu–Co.

Fig. 9 (a–c) The UV-vis absorption spectra of the 4-NP–NaBH₄ system in the presence of 4 mg L⁻¹ Cu–M dendrites at various reaction times: (a) Cu–Ni, (b) Cu–Fe and (c) Cu–Co. (d) The linear relationships between ln(C_t/C₀) and the reaction time in the presence of 4 mg L⁻¹ Cu–M dendrites.

Fig. 10a depicts a histogram of the catalytic efficiencies of Cu–M catalysts vs. cycle times. After cycling for 5 times, the catalytic efficiency of Cu–Fe dramatically decreases, that of Cu–Co evidently reduces and that of Cu–Ni slightly decreases. Namely, Cu–Ni microstructures display better catalytic stability than Cu–Fe/Co ones. Fig. 10b–d displays FESEM images of Cu–M catalysts after 5 cycles. By comparing these images with those shown in Fig. 1b–d, one can find that the shapes of Cu–Fe/Co obviously retrograde and the shape of Cu–Ni hardly changes. The above facts indicate that the catalytic efficiencies of Cu–M catalysts are related to the morphologies of the Cu–M microstructures.

Fig. 10 (a) Histogram of the catalytic efficiencies of Cu–M catalysts vs. cycle times. (b–d) FESEM images of Cu–M catalysts after 5 cycles: (b) Cu–Fe, (c) Cu–Co and (d) Cu–Ni.

3.3.2 Electrocatalytic activity. It was found that the present Cu–Ni dendrites also displayed excellent electrochemical activities for the catalytic reduction of NO₃⁻ ions and the oxidation of glucose. Fig. 11a depicts CVs of the Cu–Ni dendrite-modified GCE in 1 M KOH solution and a mixed solution of 1 M KOH + 1 M KNO₃ at a scan rate of 100 mV s−1. A pair of weak redox peaks can be seen in 1 M KOH solution, and a pair of strong redox peaks appears in the mixed solution of 1 M KOH + 1 M KNO₃. Also, the peak sites differ in the two solutions. Obviously, the presence of KNO₃ causes the change in the redox peaks. Mattarozzi's studies have confirmed that Cu–Ni alloys can catalyze the reduction of NO₃⁻ ions to NH₃ in an alkaline medium.^18,19 Since the current catalytic experiments were carried out under the same conditions as used in Mattarozzi's studies, the present Cu–Ni dendrites also exhibited excellent catalytic activity for the reduction of NO₃⁻ ions to NH₃ in the alkaline medium. Fig. 11b shows the CVs of Cu–Ni/Nafion/GCE in the mixed solution of 1 M KOH + 1 M KNO₃ at scan rates of 10–100 mV s⁻¹. It can be seen that the Cu–Ni/Nafion/GCE presents well-defined redox peaks at the different scan rates, which should be assigned to the Ni^III/Ni^II redox couple in the alkaline medium.^18,19,31 Moreover, with the increase of scan rate from 10–100 mV s⁻¹, both cathodic and anodic peak currents linearly increase with the square root of the scan rate (see the inset in Fig. 11b), which reveals the diffusion controlled electrochemical process of the electron transfer on the Cu–Ni/Nafion/GCE. Mattarozzi et al. considered that the electrocatalytic activity of Cu–Ni alloys for the reduction of NO₃⁻ ions should be attributed to the synergistic effect between Cu and Ni.^18,19 In alloys, Cu sites adsorb NO₃⁻ ions well and Ni sites efficiently adsorb H atoms produced by the discharge of H₂O molecules. Subsequently, the reductive reaction is initiated through the transfer of adsorbed H atoms. The reductive reaction can be simply described as follows:^18,19

NO₃⁻ + 6H₂O + 8e⁻ → NH₃ + 9OH⁻ (1)

Fig. 11 (a) Cyclic voltammograms (CVs) of Cu–Ni dendrite-modified GCE in a 1 M KOH solution and the mixed solution of 1 M KOH + 1 M KNO₃ at a scan rate of 100 mV s⁻¹; (b) CVs of Cu–Ni dendrite-modified GCE in the mixed solution of 1 M KOH + 1 M KNO₃ at different scan rates (from inside to outside: 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 mV s⁻¹). The inset in (b) shows the linear relationships between the peak currents and the square root of the scan rate.

Fig. 12a depicts CVs of the bare GCE and the Cu–Ni modified GCE in 0.1 M KOH solution before and after adding 10 mM glucose. No obvious redox peak can be observed either before or after adding the 10 mM glucose when the bare GCE is used as the working electrode at a scan rate of 100 mV s⁻¹. If the Cu–Ni modified GCE is used as the working electrode instead of the bare GCE, however, a pair of marked redox peaks can be seen. The different experimental phenomena observed before and after adding 10 mM glucose, indicate an electrocatalytic oxidation process.^30,31 Namely, the as-obtained dendritic Cu–Ni microstructures can be used as a catalyst for the electrochemical oxidation of glucose in alkaline medium. Fig. 12b displays the current–time plots for the CuNi/Nafion/GCE at the oxidative potential of ∼0.56 V with successive addition of glucose in 0.1 M KOH solution. The CuNi/Nafion/GCE rapidly responds to the changes of glucose concentration and a steady state signal is obtained within 3 s. The linear relationship between the glucose concentration and the catalytic current is shown in the inset of Fig. 12b. The detection limit is 0.098 M.

Fig. 12 (a) CVs of the bare GCE and the Cu–Ni modified GCE in 0.1 M KOH solution before and after adding 10 mM glucose. (b) Amperometric response of the CuNi/Nafion/GCE electrode upon the successive addition of glucose into 0.1 M KOH solution at 0.56 V. The inset shows the linear relationship between the glucose concentration and the catalytic current.

Recently, Wang et al. reported the preparation of reduced graphene oxide–chitosan (RGO–CHIT) nanocomposite-modified Cu–Co/GCE and its application in the detection of glucose.³² They believe that Cu(0) and Co(0) were firstly transformed into Cu(OH)₂ and Co(OH)₂ in the alkaline conditions, and further electrochemically oxidized into CuOOH and CoO₂. Then, glucose was oxidized by CuOOH and CoO₂ into gluconolactone. In the present work, a similar redox process could also take place: in alkaline solution, Cu–Ni microstructures are firstly oxidized into Cu(OH)₂ and Ni(OH)₂, then further electrochemically oxidized into CuOOH and NiOOH. The produced CuOOH and NiOOH present strong oxidative ability. Thus, glucose can be oxidized into gluconolactone, simultaneously reforming Cu(OH)₂ and Ni(OH)₂. The above process is simply illustrated in Scheme 1.

Scheme 1 Illustration of the glucose electrocatalytic reaction mechanism.

4. Conclusions

In summary, dendritic Cu–M (M = Fe, Co and Ni) microstructures have been successfully obtained via a facile, environmentally-friendly, and rapid electrodeposition route in a boric acid system at room temperature. It was found that the presence of M²⁺ salts caused the growth of dendritic microstructures under the current electrodeposition conditions. As a model case, Cu–Ni dendrites were studied in detail to ascertain the factors affecting the formation of perfect Cu–M dendrites. Experiments uncovered that the original amounts of NiCl₂ and boric acid were two main parameters affecting the morphology of dendritic Cu–Ni microstructures. Moreover, a higher deposition current and longer deposition time were unfavourable to the growth of perfect dendrites. More importantly, the as-deposited Cu–M dendrites exhibited excellent catalytic activity for the reduction of 4-nitrophenol in excess NaBH₄ solution. The corresponding rate constants were calculated to be 0.83 min⁻¹ for Cu–Ni, 0.69 min⁻¹ for Cu–Fe and 0.48 min⁻¹ for Cu–Co. Furthermore, Cu–Ni dendrites exhibited excellent catalytic activity for the electrochemical reduction of NO₃⁻ ions to NH₃ in alkaline medium, and were prepared into a sensor for the electrochemical detection of glucose in KOH solution with a detection limit of 0.098 M.

Acknowledgements

The authors thank the National Natural Science Foundation of China (21171005 and 21571005) for the fund support.

Notes and references

1. (a) Y. M. Zhang, Y. H. Ni, X. M. Wang, J. Xia and J. M. Hong, Cryst. Growth Des., 2011, 11, 4368–4377; (b) Y. H. Ni, X. X. Wang and J. M. Hong, RSC Adv., 2012, 2, 546–551; (c) J. J. Xi, Y. H. Ni and A. M. Liu, New J. Chem., 2014, 38, 1738.
2. M. Hu, J. S. Jiang and X. D. Li, Cryst. Growth Des., 2009, 9, 820–824.
3. (a) M. Sanles-Sobrido, M. A. Correa-Duarte, S. Carregal-Romero, B. Rodríguez-González, R. A. Álvarez-Puebla, P. Hervés and L. M. Liz-Marzán, Chem. Mater., 2009, 21, 1531–1535; (b) J. X. Fang, H. J. You, P. Kong, Y. Yi, X. P. Song and B. J. Ding, Cryst. Growth Des., 2007, 7, 864–867.
4. H. Basu, K. M. Kolwankar, A. K. Dharmadhikari, J. A. Dharmadhikari, K. Bambardekar, S. Sharma and D. Mathur, J. Phys. Chem. C, 2012, 116, 11480–11485.
5. Z. P. Peng, Y. S. Jiang and Y. H. Song, Chem. Mater., 2008, 20, 3153–3162.
6. (a) Q. M. Shen, L. P. Jiang, H. Zhang, Q. H. Min, W. H. Hou and J. J. Zhu, J. Phys. Chem. C, 2008, 112, 16385–16392; (b) C. L. Yan and D. F. Xue, Cryst. Growth Des., 2008, 8, 1849–1854.
7. L. Wang and Y. Yamauchi, Chem. Mater., 2009, 21, 3562–3569.
8. (a) X. M. Wang and Y. Ni, RSC Adv., 2012, 2, 2340–2345; (b) Y. H. Ni, Y. M. Zhang and J. M. Hong, CrystEngComm, 2011, 13, 934–940; (c) Y. H. Ni, Y. M. Zhang and J. M. Hong, CrystEngComm, 2011, 13, 1910; (d) Y. H. Ni, Y. M. Zhang, L. Zhang and J. M. Hong, CrystEngComm, 2011, 13, 794–799.
9. (a) R. Qiu, J. Y. Zheng, H. G. Cha, M. H. Jung, K. J. Lee and Y. S. Kang, Nanoscale, 2012, 4, 1565; (b) R. K. Shervedani, M. Karevan and A. Amini, Sens. Actuators, B, 2014, 204, 783–790; (c) K. Rumpf, P. Granitzer, N. Koshida, P. Poelt and M. Reissner, Nanoscale Res. Lett., 2014, 9, 412; (d) K. L. Wu, X. W. Wei, X. M. Zhou, D. H. Wu, X. W. Liu, Y. Ye and Q. Wang, J. Phys. Chem. C, 2011, 115, 16268–16274.
10. (a) D. Q. Zhang, R. R. Wang, M. C. Wen, D. Weng, X. Cui, J. Sun, H. X. Li and Y. F. Lu, J. Am. Chem. Soc., 2012, 134, 14283–14286; (b) A. R. Rathmell, S. M. Bergin, Y. L. Hua, Z. Y. Li and B. J. Wiley, Adv. Mater., 2010, 22, 2558; (c) J. C. Yu, F. G. Zhao, W. Shao, C. W. Ge and W. S. Li, Nanoscale, 2015, 7, 8811–8818.
11. S. Kang, M. Takeda, Z. Hiroi, G. W. Kim, C. G. Lee and B. H. Koo, J. Phys. D: Appl. Phys., 2010, 43, 415001.
12. I. E. Stewart, A. R. Rathmell, L. Yan, S. R. Ye, P. F. Flowers, W. You and B. J. Wiley, Nanoscale, 2014, 6, 5980.
13. A. Ungureanu, B. Dragoi, A. Chirieac, S. Royer, D. Duprez and E. Dumitriu, J. Mater. Chem., 2011, 21, 12529.
14. T. Wu, W. Y. Cai, P. Zhang, X. F. Song and L. Gao, RSC Adv., 2013, 3, 23976.
15. Q. X. Wu, W. L. Eriksen, L. D. L. Duchstein, J. M. Christensen, C. D. Damsgaard, J. B. Wagner, B. Temel, J.-D. Grunwaldtae and A. D. Jensen, Catal. Sci. Technol., 2014, 4, 378.
16. S. H. Pang, N. E. Love and J. W. Medlin, J. Phys. Chem. Lett., 2014, 5, 4110–4114.
17. H. Liu, C. Y. Guan, X. Li, L. Y. Cheng, J. B. Zhao, N. H. Xue and W. P. Ding, ChemCatChem, 2013, 5, 3904–3909.
18. L. Mattarozzi, S. Cattarin, N. Comisso, A. Gambirasi, P. Guerriero, M. Musiani, L. Vázquez-Gómez and E. Verlato, Electrochim. Acta, 2014, 140, 337–344.
19. L. Mattarozzi, S. Cattarin, N. Comisso, P. Guerrierob, M. Musiania, L. Vázquez-Gómez and E. Verlato, Electrochim. Acta, 2013, 89, 488–496.
20. J. Zhang, M. D. Baró, E. Pellicer and J. Sort, Nanoscale, 2014, 6, 12490.
21. Ö. F. Bakkaloǧlu, İ. H. Karahan, H. Efeoǧlu, M. Yıldırım, U. Çevik and Y. K. Yoǧurtcu, J. Magn. Magn. Mater., 1998, 190, 193–198.
22. R. Pattanayak, S. Panigrahi, T. Dash, R. Muduli and D. Behera, Phys. B, 2015, 474, 57–63.
23. I. S. Zhidkov, N. A. Skorikov, A. V. Korolev, A. I. Kukharenko, E. Z. Kurmaev, V. E. Fedorov and S. O. Cholakh, Carbon, 2015, 91, 298–303.
24. R. Souchet, F. Danoix, A. D'Huysser and M. Lenglet, Appl. Surf. Sci., 1995, 87/88, 271–278.
25. (a) Y. H. Ni, Y. M. Zhang and J. M. Hong, Cryst. Growth Des., 2011, 11, 2142–2148; (b) G. R. Li, C. Z. Yao, X. H. Lu, F. L. Zheng, Z. P. Feng, X. L. Yu, C. Y. Su and Y. X. Tong, Chem. Mater., 2008, 20, 3306–3314.
26. M. Supicova, R. Rozik, L. Trnkova, R. Orinakova and M. Galova, J. Solid State Electrochem., 2006, 10, 61.
27. L. M. Graham, S. Cho, S. K. Kim, M. Noked and S. B. Lee, Chem. Commun., 2014, 50, 527.
28. A. Brenner, Electrodeposition of alloys: principles and practice, Academic press, New York and London, 1963, vol. 1, pp.77–78.
29. (a) C. L. Yan and D. F. Xue, Cryst. Growth Des., 2008, 8, 1849–1854; (b) W. Che, Y. H. Ni, Y. X. Zhang and Y. Ma, J. Phys. Chem. Solids, 2015, 77, 1–7.
30. F. F. Yuan, Y. H. Ni, L. Zhang, S. M. Yuan and J. D. Wei, J. Mater. Chem. A, 2013, 1, 8438–8444.
31. L. Zheng, J. Q. Zhang and J. F. Song, Electrochim. Acta, 2009, 54, 4559.
32. L. Wang, Y. L. Zheng, X. P. Lu, Z. Li, L. L. Sun and Y. H. Song, Sens. Actuators, B, 2014, 195, 1–7.

---