Characterizations
The FTIR analyzer was used to identify the functional groups on the Catalyst samples and SEP surface, which are shown in Fig. 1. The FTIR spectra showed strong absorption at 3421 cm− 1, which was due to stretching of the phenolic and aliphatic hydroxyl groups. Peaks at around 2921 cm− 1 that were related to the C-H functional group changed after the nickel-supported catalyst samples were prepared by SEP. The results showed that there were chemical interactions and ion changes between OH, C–H, C = O, and heavy metal ions in the bio-adsorption process of nickel (Foroutan F et al. 2019). The FT-IR spectrum of N-B-Ni2P/SEP, N-B-NiFeP/SEP and N-B-NiFeP/SEP-1 confirmed the existence of NO3− and OH− group in the Nickel iron load catalyst (Fig. 1). The bands at 1596, 1363 and 777 cm− 1 were characteristic vibration for H2O, -NO3−and Metal-O (M-O) (Lee S et al. 2019; H Yang et al. 2019), respectively, which proved again that the NiFe was formed on SEP. Compared with blank SEP, the change of absorption peak at 777 cm− 1 indicated that metal particles are attached to the surface of SEP. while the weak peaks at 1112 cm− 1 is characteristic of the C-N stretching mode. (Coates J. 2006) The absorption peaks of repeatedly used catalysts at 777 cm− 1 was not lowered, indicating that the catalytic process did not affect the transition metal particles on the surface of the carrier.
The morphology and microstructural information of the N-B-NiFeP/SEP and N, B-Ni2P/SEP were systematically studied using electron microscopy techniques. (Xiao C.L et al. 2016) The closely packed
NiFe coating deposited at room temperature did not change the fiber structure of steam-exploded poplar (Fig. 2a-b). It can be seen from nickel-plated iron or nickel scanning electron microscope (SEM) that there are a large number of voids in the bracket. SEM images (Fig. 2a-b) show the growth of NiFe layer Ni-P vertically arranged nano-thin sheets, with interconnected macroporous morphology, will not hinder the underlying macroporous structure. This interesting morphology is beneficial to electrocatalysis because it provides a large number of exposed catalytic active sites and enables electrons to travel rapidly along vertical nanoflakes. Energy dispersive X-ray (EDX) spectroscopy was used to further characterize the elemental composition and distribution of the Nmai-B-talk NiFeP-hand EDS samples by EDS surface scan (Fig. 2d-g). The results show that Ni, Fe, P, B and N are uniformly distributed in the sample, and the atomic ratio is 1.28 (Ni): 1.21 (Fe). It is further confirmed that B and N atoms have been successfully entered into SEP. The above results further prove that the NMagi B talk NiFeP hand SEP is successfully realized by introducing zero valent NMagi B atoms.
The detailed structural features of the obtained sample were firstly investigated by X-ray diffraction (XRD). All the diffraction peaks can be ascribed to hexagonal Ni2P (JCPDS card No. 03-065-1989) without any peaks for impurities, suggesting the N-B-Ni2P/SEP precursor was successfully converted into nickel phosphide/SEP. (Pinilla J.L et al. 2016; Sun Y.Q et al. 2020). The diffraction pattern for PC has a broad peak at 26, which were characteristic of the (002) plane of graphitic carbon (Fig. 3).
Compared with N-B-Ni2P/SEP, the four diffraction peaks of Ni2P in the XRD spectrum of Fe-doped catalyst (N-B-NiFeP/SEP) shift to a larger diffraction angle with the doping of Fe, indicating that Fe atoms enter the Ni lattice to form Fe-Ni alloy. The intensity of the diffraction peaks of 111, 201, 210 and 300 of Ni decreases with the doping of Fe, indicating that the doping of Fe will affect the crystallinity of the alloy particles.
The average crystallite size was determined to be about 10.78 nm for N-B-NiFeP/SEP, and 17.97 nm for N-B-Ni2P/SEP from the (111) reflection by utilizing Scherrer's equation relating the coherently scattering domains with Bragg peak widths: D = kλ/B cos(θ), in which k = 0.89 for spherical particles and B is the full angular width at half-maximum of the peak in radians. On the basis of these results, we come to the conclusion that the metal particles can be well dispersed on the surface of the fiber, and the doping of Fe will affect the crystallinity of the alloy particles. The above results showed that the addition of Ni could effectively promote the miniaturization of Fe grains (Mansouriieh N et al. 2016). The XRD pattern of NiFe/Ni-P electrode (Fig. 3) further confirms the amorphous nature of NiFe catalyst layer as no new peaks are observed besides those from the catalyst. In fact, it has been proposed that amorphous NiFe electrocatalysts are much more active than their crystalline counterparts due to their structural flexibility and a high density of co-ordinatively unsaturated sites that help in the adsorption of oxidized intermediates.
Table 1
Catalysts composition determined by ICP-MS and XPS
Catalysts
|
Fe loading
(wt%)
|
Ni loading
(wt%)
|
Fe:Ni
(molar ratio)
|
FeNi loading
(wt%)
|
|
N-B-NiFeP/SEP (1:1)
|
15.58
|
17.19
|
1:1.05
|
32.77
|
ICP-MS
|
N-B-Ni2P/SEP
|
/
|
27.71
|
/
|
27.71
|
N-B-NiFeP/SEP (1:1)
|
14.84
|
11.14
|
1:0.75
|
25.98
|
XPS
|
N-B-Ni2P/SEP
|
/
|
13.88
|
/
|
13.88
|
The XPS survey scan spectrum (Fig. 4a) clearly confirmed that all Ni, Fe, B, P, N, O, and C elements were in the samples. According to XPS analysis, the contents of Ni and Fe in N-B-NiFeP/SEP were measured to be 11.14 and 14.84 wt% (Table 1), respectively. The molar ratio and actual total loading content of Fe and Ni in the N-B-NiFeP/SEP and N-B-Ni2P/SEP catalysts were further determined by ICP-MS, as shown in Table 1. The results are in good agreement with the theoretical molar ratio, indicating that the Ni and Fe metal particles are uniformly dispersed on the steam explosion poplar carrier. It should be pointed out that Ni and Fe loading of N-B-NiFeP/SEP from ICP-OES (17.19 and 15.58 wt%) analysis is much higher than the outmost surface Ni and Fe content (11.14 and 14.84 wt%) measured by XPS. This result helps us to conclude that the tiny Ni and Fe are embedded in the carbon fiber instead of anchored on the surface. (Ding R et al. 2020) This phenomenon is more obvious in the nickel content of the N-B-Ni2P/SEP.
The high-resolution spectra of the Ni 2p region showed two peaks, 2p3/2 (856.82 eV) and 2p1/2 (874.47 eV) correspond to the Ni2+ derived from the oxidation of the Ni2P surface, respectively (with corresponding shakeup satellite peaks at 862.26 and 880.03 eV) (Ding Y et al. 2020). The Fe 2p spectrum (Fig. 4h) could be fitted into two separate peaks at 711.76 and 724.68 eV corresponding to the spin-orbit states of Fe 2p3/2and Fe 2p1/2, respectively. This also confirms that Fe predominantly exists in the Fe3+state. As shown in Fig. 4g, compared with N-B-Ni2P/SEP, the negative shift of Ni2p indicates a decrease in the number of electrons at Fe site and the accumulation of electrons around the Ni site (Jiao S.L et al. 2019). These changes in electron accumulation changed the distribution of electrons, thus changing the local electronic structure of the metal position.
The XPS spectrum (Fig. 4c) for O 1s of samples can be deconvoluted into two peaks at binding ener-gies of 531.08 and 532.08 eV, which were attributed to surface adsorbed water (-OH) and C-O species (oxygen vacancies), respectively. The oxygen vacancies indicate a defect site with low oxygen coordination, which decrease the barrier for the adsorption of OH− and promotes OER. In particular, N-B-NiFeP/SEP and N-B-Ni2P/SEP had a clear difference area of oxygen vacancies due to Fe metal ions (29.49% : 51.62%). (Xu W.J et al. 2018; Kim S.H et al. 2020)
As shown in Fig. 4b, the four components of C1s spectrum (284.77, 286.36, 288.49 and 291.54 eV) were attributed to sp2 C-C, sp3 C-C, C-O and carboxylic groups, respectively. In the high-resolution XPS spectra of P 2p exhibits three contributions, P 2p3/2 and P 2p1/2, located at respectively 129.49 and 130.46 eV (Fig. 4e), which can be assigned to Ni2P, and the peak at 133.72 eV can be caused by oxidized P species.
The B 1s spectrum (Fig. 4f) clearly evidences the presence of three chemical environments for phosphorus atoms (B-O, B-C, and B-Ni). The existence of B3+ in N-B-NiFeP/SEP and N-B-Ni2P/SEP catalyst is evidenced by the peak at 192.05 eV (Fig. 4f), which can be attributed to borate species surface oxidation. Compared with N-B-Ni2P/SEP, the peak intensity at 191.16 eV of N-B-NiFeP/SEP that corresponds to B-C bonds was higher, which indicated that some C atoms in carbon fiber are replaced by B atoms. Pleasantly, the peak at 187.66 eV can be attributed to B(0) in the Ni-B bonds, which matches well with the previous literature. This result suggests that there are abundant zero valent B atoms in the N-B-NiFeP/SEP and N-B-Ni2P/SEP after N2 treatment.
As for the high-resolution N 1s spectrum, in addition to the characteristics related to pyrrolic-N (402.07 eV) and pyridinic-N (400.00 eV), a characteristic peak with a 397.10 eV binding energy is observed in the N regions, that is ascribed to metal-nitrogen bonds, indicating zero valent N(0) atoms in the N-B-NiFeP/SEP and N, B-Ni2P/ SEP (Fig. 4d). The presence of N dopant in the sample will inherently improve the interaction ability with the reactants and produce a higher positive charge density on its adjacent carbon atoms, which may also contribute to the high activity of the sample. (Sun Y.Q et al. 2020) Therefore, the above results indicate that the zero valent N, B atoms were successfully doped into N, B-Ni2P/SEP and N-B-NiFeP/SEP.
The crystallization and graphitization degree of carbonized steam exploded poplar support on nickel-supported catalyst and Ni-Fe bimetallic catalyst were studied by Raman spectroscopy. In general, the ID/IG ratio less than one is ideal. As shown in Fig. 5, the carbon fiber carriers have high quality and crystallinity, the peak intensity ratio (ID/IG) is less than one, and the carbon samples show two distinct bands. The first band is the well-known D band, located at 1363cm− 1, attributed to disorder in the carbon structure, such as defects in the carbon structure or amorphous carbon. (M.S.Dresselhaus et al. 2002; Awadallah A.E et al. 2013) On the other hand, the vibration of sp2 carbon atoms in the graphitization region forms the G band located in 1589 cm− 1. (Buthainah Ali et al. 2017; Allaedini G et al. 2015) Generally, the ratio of D-band strength to G-band strength ID/IG is used to reflect the degree of graphitization. The ID/IG ratio of N-B-NiFeP/SEP is 0.89, which indicates that a large number of defects and irregular structures have been introduced into the carbon fiber carrier. The ID/IG ratio was further increased to 0.95 for N-B-Ni2P/SEP, representing the enhanced number of structural defects, increased localized sp3 defects in sp2 framework, high electrical conductivity.
The specific surface area and porosity of the obtained materials have been investigated by N2 adsorption-desorption experiments. In the curves of N-B-NiFeP/SEP and N-B-Ni2P/SEP (Fig. 6a), the type IV adsorption branches were corresponding to the mesoporous structure. According to IUPAC classification, the isotherms (Fig. 6) of the mixed oxides were classified as type IV with an H3 hysteresis loop, suggesting the existence of mesoporous materials with an incision-like pore geometry. The specific surface area of N-B-NiFeP/SEP and N-B-NiP/SEP were calculated to be 55.44 and 57.18 m2·g− 1, respectively. The pore size distributions are shown in Fig. 6b. The average pore size of N-B-NiFeP/SEP was about 11.86 nm, while those of N-B-Ni2P/SEP was around 8.42 nm. It was clear that N-B-NiFeP/SEP and N-B-Ni2P/SEP were mainly composed of micropores and mesopores around 10 nm. As shown in Fig. 6d, the average pore widths of three samples follow the order of N-B-NiFeP/SEP > N-B-Ni2P/SEP and the pore volumes of N-B-NiFeP/SEP and N-B-Ni2P/SEP were 0.19 and 0.11 cm3·g− 1, respectively. The pore structure of materials played an important and even decisive role in many properties of materials. Carbon materials as carriers, their porous properties were conducive to the diffusion of substrates and products, and can expose more active sites, thus improving the overall activity of the catalyst.
In addition, to investigate the kinetics of these catalysts, the Tafel slopes obtained from the LSV polarization curves were shown in Fig. 7c. The Tafel slope of N-B-NiFeP/SEP (101 mV dec− 1) was considerably smaller than those of N-B-Ni2P/SEP (151 mV dec− 1), confirming its faster OER kinetics. Ac-cording to our research results, it can be concluded that the synergetic effect of Ni-Fe bimetal load and carbon carrier played an important role in facilitating the kinetics of OER (Li Y.Y et al. 2020; Jiang J et al. 2018; Yue S et al. 2019).
Removal of 4-nitrophenol (4-NP) from wastewater is of significant importance in view of environment protection since 4-NP is a prevalent contaminant produced in industry and agriculture (Choi S et al. 2019; Ding R et al. 2020). It is known that 4-aminophenol (4-AP) is very useful and important in many applications including analgesic and antipyretic drugs, photographic developer, corrosion inhibitor and anticorrosion-lubricant. The reduction of 4-NP to 4-AP has been extensively used as a benchmark system to evaluate the catalytic activity of metal NPs (Guohui Chang et al. 2012; Yang Y et al. 2014).
Therefore, the reduction of 4-NP toward 4aminophenol (4-AP) in the presence of NaBH4 was selected as a model reaction to further confirm the generality of N-B-NiFeP/SEP and N-B-Ni2P/SEP. As shown in Fig. 8a, the adsorption peak of 4-NP was red-shifted from 317 to 400 nm immediately upon the addition of NaBH4 solution which corresponds to a color change from light yellow to yellow green, due to the formation of the 4-nitrophenolate ion under alkaline conditions. When catalyst was added, the intensity of characteristic peak at 400 nm quickly declined. The reduction of 4-NP was completed within 10 min over 3 mg N-B-Ni2P/SEP and N-B-NiFeP/SEP (Fig. 8c). Considering that the reductant concentration is much higher than that of 4-NP (CNaBH4/C4 − NP= 100) in the reaction mixture, the pseudo-first-order rate kinetics with respect to 4-NP concentration could be used to evaluate the catalytic rate. The reaction kinetics can be described as − ln(Ct/C0) = kt, where k is the rate constant at a given temperature and t is the reaction time. C0 and Ct are the 4-NP concentration at the beginning and at time t, respectively. As expected, a good linear correlation of ln(Ct/C0) vs. reaction time t was obtained (Fig. 8b), whereby the kinetic rate constant k was estimated to be 1.617 (R2 = 0.99) and 0.765 (R2 = 0.99) min− 1 for N-B-Ni2P/SEP and N-B-NiFeP/SEP. To compare different catalysts, we calculated the ratio of rate constant K over total weight of nickel catalyst, where K = k/m. Thus the activity factor K was calculated to be 539 and 255 min− 1·g− 1 for N-B-Ni2P/SEP and N-B-NiFeP/SEP, respectively. It is clear that N-B-Ni2P/SEP shows the largest activity factor.