Influence of copper content on the electrocatalytic activity toward methanol oxidation of CoχCuy alloy nanoparticles-decorated CNFs

In this study, CoCu alloy nanoparticles-incorporated carbon nanofibers are introduced as effective non precious electrocatalyst for methanol oxidation in alkaline medium. The introduced electrocatalyst has been synthesized by simple and effective process; electrospinning. Typically, calcination, in nitrogen atmosphere, of electrospun nanofibers composed of cobalt acetate, copper acetate and poly (vinyl alcohol) leads to form carbon nanofibers decorated by CoCu nanoparticles. The nanofibrous morphology and alloy structure have been confirmed by SEM, TEM and XRD analyses. Investigation of the electrocatalytic activity indicates that copper content has strong influence, the alloy nanoparticles having the composition Cu5%Co95% showed distinct high performance; 100 times higher than other formulations. Overall, the introduced study revealed the veil about the distinct role of copper in enhancing the electrocatalytic activity of cobalt-based materials.


Results and Discussion
Characterization of CoCu-decorated CNFs. Figure 1(A) shows the XRD patterns for the sintered nanofibers prepared from electrospun solution having 0.05 and 0.95 g CuAc and CoAc, respectively (Cu 5% Co 95% -CNFs); the symbols show the positions of XRD peaks of Co-FCC (JCPDS #15-0806), Co-HCP (JCPDS# 05-0727) and Cu (JCPDS# 3-2838). As shown in XRD pattern the strong diffraction peaks of 2θ values of 44.35°, 51.65° and 75.95° corresponding to (111), (200) and (220) crystal planes, respectively indicate formation of Co-FCC crystals moreover Co-HCP could be also detected as the standard peaks indicating that the formation of this phase can be seen at 41.72°, 44.73° and 47.63° corresponding to crystal planes of (100), (002) and (101), respectively. The aforementioned two phases of Co usually coexist at room temperature and difficult to be separated from each other and these two are most common phases of Co 40 . The XRD analysis in Fig. 1(A) specifies that one solid peak and two much weaker ones exist at 43.3°, 50.4° and 74° (marked by black circles & arrows along with high magnification for the marked area) which are indexed as the (111), (200) and (220) crystal planes, respectively of Cu-FCC (JCPDS # 3-2838). It was freshly stated that the (111) and (110) crystals planes of copper had boosted catalytic activity to methanol oxidation in contrast to the (100) crystal plane of copper 41 . Figure 1(A) shows that the (111) crystal plane exposed completely, which is promising for the catalytic Scientific RepoRts | 5:16695 | DOI: 10.1038/srep16695 activity to methanol oxidation. It is noteworthy mentioning that the same XRD data were obtained with the other formulations.
Beside the polycondensation property of acetate ions, during calcination under inert environment this ion abnormally decomposes and producing strong reducing gases (CO and H 2 ) which play key role in complete reduction for the salt resulting in formation pristine metals rather than metal oxides 42,43 Figure 1(B-D) shows the FESEM images of powder obtained from the electrospun mats containing 0.05 and 0.95 g CuAc and CoAc, respectively (Cu 5% Co 95% -CNFs). As shown, smooth and good morphology nanofibers were obtained. Moreover, the obtained nanofibers are decorated by nanoparticles which are expected to be Cu and/or Co metals. Figure 2 displays FE SEM images for nanofibers obtained from sol-gels having different CoAc/CuAc ratios, as shown almost all the powders demonstrate the nanofibrous morphology; however the best morphology was corresponding to CuAc content of 10 ( Fig. 2B) and 15% (Fig. 2C). To examine the hypothesis of formation of bimetallic nanoparticles decorating carbon nanofibers; TEM analysis was carried. Figure 3(A,B) displays TEM images of the obtained sintered nanofibers from the electrospun mats containing 0.05 g CuAC (Cu 5% Co 95% − CNFs). As shown, the nanofibers are ornamented by crystalline metallic nanoparticles distributed along with nanofibers which also supports the FESEM image ( Fig. 1(B-D). Figure 3(C) displaying HR TEM image shows that the attached nanoparticles have good crystallinity (red arrow) which indicates that these nanoparticles compose of Co and/or Cu. Moreover the main nanofiber has an amorphous structure which refers to carbon. The selected area electron diffraction pattern (SEAD) demonstrated in panel D indicated good crystallinity for the metallic nanoparticles and simultaneously supports the XRD results.
To investigate the elemental distribution of the obtained nanofibers, TEM-EDX has been carried out; as shown in the Fig. 3(E), the attached metallic nanoparticles have different in sizes and also possess good crystallinity. Line EXD analysis was carried out at randomly selected line, the observed concentration profiles reveal that Co, Cu and C are detected along with chosen line. Interestingly, Co and Cu have almost the same elemental dissemination along with the nominated line which indicates alloy structure.
Electrocatalytic activities. Influence of copper content. In contrast to the noble metals, the transition metals-based electrocatalyst own their activity from an active layer formed on the surface; e.g. NiOOH in the case of nickel. This active layer can be synthesized by sweeping the surface using multiple CV cycles in KOH (in case of utilizing in alkaline medium [45][46][47] . The cyclic voltammetric behaviors (in 1 M KOH solution) of the introduced CoCu-decorated CNFs with different Cu contents are introduced in Fig. 4A-C. Polarization was started by a potential scanning at a scan rate of 100 mVs −1 from 800 mV to − 200 mV (vs. Ag/AgCl reference electrode) in the cathodic direction and then the scan was reversed in the anodic direction back to 800 mV.
As shown in Fig. 4A, the behavior of the Cu 5% Co 95% -CNFs nanofibers, two pairs of redox peaks can be observed. On the other hand, as shown in Fig. 4(B,C), increasing the copper content in the electrospun solutions led to almost disappearing of these redox peaks. Moreover, there is considerable difference in the corresponding current densities of the three formulations; Cu 5% Co 95% -CNFs sample reveals relatively higher values. Appearing of redox peaks is an important finding as it can be considered as an indicator for good electrocatalytic activity.
To properly evaluate the electrocatalytic activity of the activated Cu x Co y -decorated CNFs, methanol electrooxidation was carried out in 1M KOH solution. Figure 5(A) displays the activities in presence of 1 M methanol (in 1 M KOH). As shown, the copper content has very strong influence on the obtained current densities. Surprisingly, the nanofibers obtained from electrospun solution having 5 wt% CuAc (Cu 5% Co 95% -CNFs) reveal current density (190 mA/cm 2 ) 100 times more than the other formulations which comparatively have inconsiderable activities. This finding can be attributed to appearing of the clear redox peaks for this distinctly active formulation (Fig. 4A).
It is noteworthy mentioning that, among the transition metals, nickel is the most widely used due to its high activity especially in methanol electrooxidation. As shown in Fig. 5(C), the performance of the introduced material has been compared to Ni-decorated nanofibers. The results affirmed the activity of the introduced Co/Cu-decorated CNFs as the corresponding current density was higher as shown in the figure.
The optimum methanol concentration is a process parameter for every electrocatalyst. In other words, as water is a reactant in the methanol oxidation reaction, the optimum methanol concentration should be assigned for the introduced catalyst. Fig. 5(B) displays the influence of methanol concentration on the obtained current density of the best nanofibers observed in Fig. 5(C). As shown in the figure, the methanol concentration increase leads to the increase of the current densities which specifies methanol oxidation on the surface of the presented nanofibers. It can be observed in Fig. 5(B) the optimum methanol concentration of 2 M which is consistent with many reported electrocatalysts.
The onset potential is a significant indicator among the appealed parameters to prove the electrocatalytic activity. Generally, in alcohols electrooxidation, more negative onset potential shows high activity and less over potential. As shown in Fig. 5(D), the observed onset potential of the introduced nanofibers is ~310 mV (vs. Ag/AgCl), and the achieved assessment is relatively low compared to numerous reported materials 48 . Moreover, it is clear that the onset potential is independent on the methanol concentration as well as the alloy composition. This discovery further supports the electrocatalytic performance of the introduced nanofibers.
As shown in Fig. 5(E), the best decorated CNFs show cathodic peak at ~50 mV, this peak is almost at the same potential of the cathodic peak in the activation of this formulation (Fig. 4A) which is responsible about the catalyst regeneration 49 . Therefore, the observed high negative current density in the cathodic direction can be attributed to either oxidation of the intermediate compounds during methanol oxidation in the anodic direction and/or reactivation of the surface. Both assumptions can be considered as acceptable reasons for the observed high activities.
It is known that the alloy structure does have electron configuration differs than the individual constituents which provides special physicochemical characteristics. Moreover, this electron configuration depends mainly on the composition. As a heterogeneous catalyst, the performance of the electrocatalyst in methanol oxidation depends on the electronic structure of the catalyst. From TEM EDX results (Fig. 3E), the formed Co/Cu nanoparticles possess an alloy structure. Accordingly, the influence of the composition of the CoCu NPs decorating the CNFs on the electrocatalytic activity has been investigated by synthesizing more CoCu-CNFs having more different compositions for the metallic nanoparticles. Typically, the electrospun solutions were adjusted to produce final CNFs decorated by metallic nanoparticles with composition of Cu 0% Co 100% , Cu 5% Co 95% , Cu 35% Co 65% , Cu 65% Co 35% , and Cu 100% Co 0% . Figure 5(F) displays the electrocatalytic performances in presence of 1 M methanol. As shown in the figure, pristine Co-and Cu-(Cu 0% Co 100% and Cu 100% Co 0% , respectively) decorated CNFs almost do not have electrocatalytic activity toward methanol oxidation. Moreover, increasing the copper content in the formulations having nanoparticles with the composition of Cu 35% Co 65% and Cu 65% Co 35% has also distinct negative impact on the activity. Interestingly, the same formulation discovered before (Cu 5% Co 95% ) still prominent as it reveals the maximum and incomparable activity to the other formulations. The distinctly big difference in the activity observed in Fig. 5(F), might be understood from the activation behaviors on the investigated formulations. Figure 6 displays the sweep cyclic voltogramms in presence of 1 M KOH for 50 cycles at scan rate 100 mV/s. Compared the best formulation (Fig. 4A), one cannot find characteristic redox peaks in the investigated formulations. Figure 7(A), displays comparison between the investigated alloys compositions compared to the best formulation, as shown the redox peaks in the optimum composition are clearly seen while almost no peaks could be found in the other prepared modified CNFs.
Overall, according to the obtained results, the observed distinguished electrocatalytic activity of the best formulation can be attributed to formation of ZOOH/Z(OH) 2 on the surface of the Z alloy where Z refers to the optimum composition (Cu 5% Co 95% ). Analogy to the transition metals, formation of the ZOOH layers can be clarified by the following reactions 50 Where Z(OH) 2 was synthesized on the surface of the Z alloy NPs due to its electron configuration which enhances formation of stable hydroxide layer. Therefore, in Fig. 4(A), increasing the number of potential sweeps resulting in progressive increase of current density values of the cathodic peak since of the entry of OH − into the Z(OH) 2 surface layer, which leads to the progressive formation of a thicker ZOOH layer analogous to the Z(OH) 2 /ZOOH transition 50 which results in high catalytic activity. Accordingly, the mechanism of electrooxidation of methanol using the introduced modified nanofibers matches the introduced one on the preceding literature 44,48,51 . Briefly, in the first step of the methanol oxidation, the prepared catalyst (Z) adsorb methanol and partially release protons in the second and third step more protons released. Normally the Z 3 COH species decomposes to produce Z and protons in step 4 51 .
EIS Characterization. Electrochemical impedance spectroscopy (EIS) was performed to examine the methanol oxidation. Nyquist plots for the investigated electrode in different methanol concentrations (0, 0.5, 1, 2 and 3 M + 1.0 M KOH) are displayed in Fig. 7(B). Considering the Nyquist plot of the presented electrode, the acquired plot specifies that the current leads to capacitive charging of the double layers. In other words, it can be concluded that the cell has polarizable electrodes; this assumption is confirmed by the insignificant influence of increasing methanol concentration on the Nyquist plots as shown in the figure. In the Nyquist plot, the Faradic reaction (methanol oxidation) commonly showed by capacitive loop with a diameter almost corresponding to the charge transfer resistance (R CT ). As shown in the figure lower charge transfer resistance is obtained at 3 M concentration of methanol.  Table 1 summarizes the values of the R CT in Ω cm 2 for the investigated electrode at different methanol concentrations. As shown, the charge transfer resistance inversely proportions to the methanol concentration which indicates good activity.
Chronoamperometry. Beside the sought-for high electrocatalytic performance, stability was another target from preparing of the alloy structure. Figure 7(C), shows the chronoamperogram of electrode recorded for nearly 1500 s at fixed potential (E = 0.4 V) in 3 M methanol + 1 M KOH at room temperature. As shown in the figure, there is an initial current drop, followed by a very slow decay. It is noteworthy mentioning that, only one working electrode has been used in all the electrochemical analysis. Multiple use of the electrode did not affect the performance which supports the good stability of the introduced catalyst. This figure supports the good stability of introduced catalyst which can be attributed to the alloy structure.

Conclusion
Electrospinning technique can be utilized to produce smooth, beads free and good morphology nanofibers composed of poly(vinyl alcohol), copper acetate and cobalt acetate. Calcination of the resultant electrospun mats in nitrogen atmosphere leads to form carbon nanofibers decorated by CoCu bimetallic alloy. The electrocatalytic activity toward methanol electrooxidation mainly depends on the composition of the bimetallic CoCu alloy nanoparticles attaching the carbon nanofibers. Typically, the optimum  composition is related to the bimetallic nanoparticles having 5 wt% Cu. The suggested non-precious electrocatalyst can be further improved to be more effectual for fuel cell applications.

Method
Materials. Copper (II) Acetate monohydrate (CuAc, 98%) and Cobalt (II) acetate tetra hydrate (CoAc, 98%) were bought from Showa chemicals co Ltd Japan and Junsei chemicals co Ltd Japan, respectively. Poly (Vinyl alcohol) PVA with a molecular weight 65000 g/mol was obtained from Aldrich, USA, Distilled water was used as solvent.
Preparation of CoCu CNF's. CoAc and CuAc aqueous solutions were firstly prepared by dissolving different amounts CuAc and CoAc (total amount of CoAc and CuAc was 1 g) in 3 ml of distilled water with 2 h stirring at room temperature and then mixed with 15 g PVA aqueous solution (10 wt%). Finally the mixture was stirred at 50 °C for 6 h to get see-through, clear and consistent mixture. The achieved sol-gel was electrospun at high voltage of 22 kV using DC power supply at room temperature with 65% relative humidity. The distance between needle tip (positive electrode) and rotating cylinder (negative electrode) was kept constant at 22 cm. The ready nanofiber mats were normally dried at room temperature for 12 h and then under vacuum for 24 h at 60 °C, lastly the dried nanofibers were calcined at 900 °C for 7 h in nitrogen atmosphere with heating rate of 2.0 °C/min.

Sample characterization.
The phase and crystallinity of the composite were characterized by X-ray diffract meter (XRD, Rigaku, Japan) with Cu-Kα (λ = 1.54056 Å) radiation over a range of 2Ѳ angle from 10° to 80°. The morphology of the products was observed by field-emission scanning electron microscopy (FESEM, Htachi S-7400, Japan) whereas the distribution of elements was measured using energy dispersive X-ray spectroscopy (EDX) analysis. While high resolution TEM images and selected area electron diffraction patterns were observed by JEOL JEM-2200FS transmission electron microscope (TEM) operating at 200 kV equipped with EDX (JEOL, Japan).
Electrochemical measurements. Electrochemical measurements were conducted using VersaSTAT4 (USA) and conventional three electrode electrochemical cell at room temperature. An Ag/AgCl, Pt wire, and 3 mm glassy carbon were used as the reference, counter and working electrode, respectively. For the preparation of the catalyst electrode, a 2 mg of the synthesized alloy NFs were spread in a suspension of 400 μ l propanol and 20 μ l of a nafion solution under ultrasonic stirring for 1 h to ensure good dispersion. A 15 μ l aliquot of the slurry was spread on the top of the sophisticated working electrode surface and dried at 80 °C for 30 min. The active zone of the utilized glassy carbon electrode was a circle with a diameter of 3 mm. Therefore, the active area was estimated (0.0713 cm 2 ) and all the data was normalized to this area.