2.1. XRD Patterns of α-NixFe1-xOOH NPs
Figure 1 shows the XRD patterns of Ni0, Ni5, Ni10, Ni15, Ni20, Ni30, Ni40, and Ni50 samples. The XRD peaks of goethite (ICDD card No.00-029-0713) with space group (Pbnm) were observed in all XRD patterns with increasing Ni content. Moreover, a new peak appeared at 2
θ = 11.56° corresponding to the (0 0 3) plane of α-Ni(OH)
2 crystalline phase (ICDD card No.01-076-6904) with space group (R3m: H), and the intensity of this peak increased with increasing x content from 0.30 to 0.50 as shown in the yellow rectangle in
Figure 1. The percentage of the α-Ni(OH)
2 crystalline phase increased from 7.63% to 9.35% while ‘x’ increased from 0.30 to 0.50 as shown in
Table 1. Additionally, the elevation of background intensity observed in the XRD patterns depicted in
Figure 1, as the Ni content increased from 0.30 to 0.50, suggests the appearance of an amorphous Ni hydroxide phase. The diffraction lines of α-FeOOH were slightly shifted towards lower 2
θ values with increasing Ni content, implying a minor expansion in the unit cell due to Ni/Fe substitution. The unit cell parameters of goethite in Ni0, Ni5, Ni10, Ni15, Ni20, Ni30, Ni40, and Ni50 samples obtained by Rietveld refinement of XRD patterns are listed in
Table 1. The slight variation in ionic radius between Fe
3+ (0.645 Å) and Ni
2+ (0.69 Å) [
16] ions caused a small expansion of the unit cell of α-FeOOH. In line with previous studies [
17,
18], it was observed that the expansion along the b-axis direction was most remarkable. The non-uniform expansion is likely to be attributed to different distortions in the octahedral sites occupied by Ni
2+ and Fe
3+ ions, stemming from their disparate electron configurations (3d
8 and 3d
5, respectively). The incorporation of Ni into α-FeOOH can be attributed partially to its incongruent release from the initial Fe-Ni hydroxide (Ni-ferrihydrite) [
19]. The calculated average size of α-Ni
xFe
1-xOOH NPs by the Scherrer equation from the FWHM of the (110) diffraction line is summarized in
Table 1.
2.2. Thermogravimetric and Differential Thermal Analysis (TG-DTA) of α-NixFe1-xOOH
Thermogravimetric (TG) and differential thermal analysis (DTA) curves for Ni
xFe
1-xOOH NPs with ‘x’ 0, 0.10, 0.20, 0.30, 0.40, and 0.50 are shown in
Figure 2a and
Figure 2b, respectively.The weight loss at temperatures up to approximately 140 °C corresponds to the release of H
2O molecules adsorbed on the surface of Ni
xFe
1-xOOH NPs crystals or trapped within the interstitial spaces between them [
19,
20]. This low-temperature weight loss increased in the Ni
xFe
1-xOOH NPs samples with increased Ni content, which can be explained by more adsorbed water in the NiFeOOH NPs samples with a large surface area [
20]. Also, different molar fractions of incorporated Ni
2+ ions and different crystal facets in Ni-doped Ni
xFe
1-xOOH NPs crystals can also affect the amount of adsorbed water in Ni
xFe
1-xOOH NPs samples. The weight loss at temperatures between about 150 and 290 °C can be attributed to the dehydroxylation of surface hydroxyl groups [
18,
19,
20]. This weight loss also increases in the Ni
xFe
1-xOOH NPs with a higher Ni content due to the larger surface area. The total weight loss percentage for Ni
xFe
1-xOOH NPs in the temperature range from RT to 1000 °C increased with Ni content increased, from 14.24% in Ni0 to 26.63% in Ni50, as shown in
Figure 2 (a).
Figure 2 (b) shows the DTA curves of the Ni
xFe
1-xOOH NPs with ‘x’ 0, 0.10, 0.20, 0.30, 0.40, and 0.50. DTA curve of Ni
xFe
1-xOOH NPs showed a single endothermic peak in the temperature range from 230- 300°C due to the phase transformation from α-FeOOH to α-Fe
2O
3 [
18,
19]. As the amount of Ni increased from 0 to 0.2 mol.%, the endothermic peak became smaller and somewhat shifted to higher temperatures from 251°C in Ni0 to 262°C in Ni20, while in Ni30 shifted in the opposite direction and observed at 241 °C. The DTA curves of Ni40 and Ni50 showed new endothermic reactions at lower temperatures caused by the lower degree of crystallinity or the increased amorphous phase [
19]. These results agree with XRD data (see
Figure 1 and
Table 1).
2.3. FTIR Spectra of α-NixFe1-xOOH NPs
FTIR spectra of Ni
xFe
1-xOOH NPs with ‘x’ 0, 0.10, 0.20, 0.30, 0.40, and 0.50, and Fe
2O
3 and NiO as standard reference materials, carried out in the wavenumber range of 400–4000 cm
−1, are shown in
Figure 3. The transmission band showed approximately 3454 cm
-1 ascribed to -OH [
21] in Ni0 and shifted to a lower wavenumber, and the peak intensity increased with increasing Ni content. The small transmission peak observed at 3130 cm
-1 corresponds to the O-H stretching vibration of the hydroxyl group [
18,
22]. The band center of the hydroxyl group gradually shifts to a higher wave number, and the peak intensity decreases with increasing Ni content, as shown in
Figure 3. The intense two transmission peaks were observed at 797 cm
-1 and 895 cm
-1 related to Fe-O-H bending vibrations out of plane and in plane [
17,
18,
19,
22], respectively. The intensity of both transmission peaks gradually decreased with Ni content. These two bands slightly shifted towards higher wave numbers up to 0.30 mol% Ni content, and with high Ni content (0.40 and 0.50), the center of these bands changed to lower wavenumbers. It is known that bending bands' positions depend on crystallinity degree and the transition metal substituted with Fe [
17,
22]. In this study, Ni substituted with Fe caused a shift toward lower wavenumbers with increasing Ni content from Ni0 to Ni30, while in Ni40 and Ni50, an increase in particle size caused a change to a higher wavenumber. These results agree with XRD and DTA (see
Figure 1 and
Figure 2). The transmission band was observed at 645 cm
-1 in pure-FeOOH due to Fe-O symmetric stretching [
22], and the band center gradually shifted to a lower wavenumber with increasing Ni content from 0 to 0.5. A new transmission peak appeared in higher Ni concentration samples (Ni40 and Ni50) at 472 and 476 cm
-1, respectively. It was noted that the recent peak compared with that shown in the FTIR spectrum of NiO, can be attributed to Ni-O [
23]. The small transmission bands located at 1639 cm
-1 in Fe
2O
3 are due to the stretching vibration of -OH from absorbed water [
21,
23,
24]. Also, this peak showed in Ni
xFe
1-xOOH NPs at 1639 cm
-1 in Ni0 and slightly shifted to 1643 cm
-1, and the peak intensity gradually increased with increasing Ni content.
2.4. XAFS Spectra of α-NixFe1-xOOH NPs
The XAFS and EXAFS measurements are powerful tools for investigating the local atomic and electronic properties of materials, providing a comprehensive understanding of their structure and reactivity [
25,
26].
Figure 4 shows the XANES spectra and Fourier transforms of
Fe-/
Ni-
K-edges of α-Ni
xFe
1-xOOH NPs with ‘x’ 0, 0.05, 0.10, 0.15, 0.20, 0.30, 0.40, and 0.50 and Fe foil, Fe
3O
4, and α-Fe
2O
3 as standard materials in the
Fe K-edge and with ‘x’ from 0.10, 0.15, 0.20,0.30, 0.40, and 0.50 and Ni-foil and NiO as references materials at
Ni K-edge.
Figure 4 (a) displays the normalized
Fe K-edge XAFS spectra of α-Ni
xFe
1-xOOH NPs. The
Fe K absorption edge was observed at normalized intensity= 0.5 for Fe-foil, Fe
3O
4, and Fe
2O
3 being 7117.26±0.02, 7120.67±0.02, and 7121.79±0.02 eV, respectively. The oxidation state in Fe-foil, Fe
3O
4, and α-Fe
2O
3 is 0, 2.66, and 3, respectively. The absorption edge of α-Ni
xFe
1-xOOH NPs was observed at 7122.65±0.01, 7122.49±0.01, 7122.27±0.01,7122.15±0.01, 7122.23±0.01, 7121.91±0.01, 7122.08±0.01, and 7122.28±0.01 with ‘x’ 0, 0.05, 0.10, 0.15, 0.20, 0.30, 0.40, and 0.50, respectively as shown in insert figure in
Figure 4 (a). Increasing oxidation state of Fe results in a shift of the absorption edge to higher energy [
27]; according to the above, values of the absorption edge of α-Ni
xFe
1-xOOH NPs are higher than the values of Fe
2O
3 (Fe
3+), this implies that none of the α-Ni
xFe
1-xOOH NPs samples contained Fe
2+, and they contained only Fe
3+. The pre-edge peak was found at 7113 eV for α-Ni
xFe
1-xOOH NPs, at 7114 eV for both Fe-foil and Fe
2O
3, being designated as the absorption edge. The pre-edge features in the XANES spectrum reveal electronic states and symmetry of unoccupied orbitals in the absorbing atom [
27,
28]. Analyzing the intensity, energy position, and shape of pre-edge peaks helps understand the electronic structure, oxidation state, and local environment of Fe atoms in the material [
29]. The intensity of the pre-edge peak is lower when the octahedral site has high symmetry. α-FeOOH with the octahedral structure of FeO
6 shows an increase in the pre-edge peak intensity and therefore a decrease in symmetry, with increasing Ni content.
Figure 4 (c) displays the normalized
Ni K-edge XANES spectra of Ni
xFe
1-xOOH NPs with ‘x’ 0.10, 0.15, 0.20, 0.30, 0.40, and 0.50 and Ni-foil and NiO as reference materials. The pre-edge peak of Ni foils, NiO, and Ni
xFe
1-xOOH NPs was shown at 8321.40, 8319.68, and 8319.48 eV, respectively [
30]. The height of the pre-edge peak intensity of NiO and Ni
xFe
1-xOOH NPs has approximately the same value, and the change in the pre-edge is minimal. Therefore, we cannot find any significant difference in the oxidation state of Ni and the symmetry of the structure [
31].
In contrast, the
Ni K normalized absorption peak of Ni
xFe
1-xOOH NPs and reference materials are shown in the inserted figure in
Figure 4 (c). The absorption edge of Ni-foil, Ni
xFe
1-xOOH NPs, and NiO is at 8327.40, (8329.21-8329.48), and 8328.76 eV, respectively. According to the absorption edge's higher value compared to NiO, the oxidation state of Ni in all Ni
xFe
1-xOOH NPs samples is higher than +2. Both Ni15 and Ni20 are samples containing less Ni
3+, while Ni10 has the highest level of Ni
3+.
Figure 4 (b) shows the Fourier transform of
Fe K-edge EXAFS (FT-EXAFS) for α-Ni
xFe
1-xOOH NPs. The FT-EXAFS of α-Ni
xFe
1-xOOH NPs, in the first coordination, was observed at 1.57 Å, and the peak intensity decreased with increasing ‘x’ from 0.10 to 0.50, and this peak could be attributed to Fe-O [
29,
30]. Moreover, the second peak of Fe-foil, Fe
3O
4, and Fe
2O
3 was shown at 2.21 Å, 2.91 Å, and 2.68 Å, respectively. The second peak of Ni
xFe
1-xOOH NPs was also shown between the Fe
2O
3 and Fe
3O
4. Also, this peak was shifted to a higher Fe-Fe distance, and the peak intensity decreased by reducing the Fe content from 1 to 0.50. All observed peaks in this range (2.21 - 2.91) Å are attributed to Fe-Fe (or Fe-Ni) [
32,
33].
Figure 4 (d) shows the
Ni-K FT-EXAFS of Ni
xFe
1-xOOH NPs, and it was clear that two peaks at 1.63 Å and 2.77 Å, can be attributed to Ni-O and Ni- Me (Me = Ni or Fe), respectively [
33,
34,
35]. The first peak was moved to a higher distance between Ni and O 1.75 Å, but the second peak shifted to a lower Ni- Me distance of 2.62 Å for Ni
xFe
1-xOOH NPs [
35]. The intensity of the first peak decreased, and the peak position was moved to a higher radius with increasing Ni content. These results obtained from
Ni K and
Fe K FT-EXAFS and XANES of the Ni
xFe
1-xOOH NPs agree with XRD results.
2.5. 57Fe- Mössbauer Spectra of α-NixFe1-xOOH NPs
Figure 5 shows
57Fe-Mössbauer spectra of α-Ni
xFe
1-xOOH NPs ‘x’ from 0.00 to 0.50 measured at RT in the velocity range from 12 mm s
-1 to -12 mm s
-1 and from 3 mm s
-1 to -3 mm s
-1. The Mössbauer parameters in these two velocity ranges are listed in
Table 2 and
Table 3.
Figure 5 (a) shows the
57Fe- Mössbauer spectra of Ni0, Ni10, Ni20, Ni30, Ni40, and Ni50 measured at RT in the wide velocity range from 12 mm s
-1 to -12 mm s
-1. The Mössbauer spectra exhibits a sextet with Mössbauer parameters characteristic of α-Ni
xFe
1-xOOH NPs ‘x’ from 0.00 to 0.30 as the predominant component, while the sextet disappeared after increasing the Ni content to 0.30. The α-FeOOH NPs and α-Ni
xFe
1-xOOH NPs spectra were analyzed by incorporating the hyperfine magnetic distribution (
Bhf); their average values are presented in
Table 2. The slight changes in isomer shift, except for the Ni0 sample, indicate that the site is distorted goethite, as shown in
Table 2. Moreover, the presence of a quadrupole doublet, which increases as the Ni content increases, can be ascribed to the presence of Fe
3+ ions in a low-crystalline or amorphous phase. This phase could be attributed to compounds such as Ni-Fe hydroxide, low crystalline Ni-ferrite, or Ni-goethite. The higher proportion of the low-crystalline or amorphous phase aligns with the increased background observed in the XRD patterns, as depicted in
Figure 1. Also, the line width value increased from 0.196 to 0.53 mm s
-1, this may be confirmed that the result obtained from XRD, wherein the amorphous phases increased with increasing Ni content. In addition, the particle size of the goethite drops to a nano-scale, and the superparamagnetic component increases as a function of increasing Ni content [
11,
14,
15]. The presence of the doublet suggests nanosized goethite particles in Ni40 and Ni50. To analyze the doublet component of FTMS spectra of Ni20, Ni30, Ni40, and Ni50 measured at RT in the wide velocity range for more detail, a narrower velocity range from 3 mm s
-1 to -3 mm s
-1 was applied and the spectra are presented in
Figure 5 (b). One doublet showed in Ni20, Ni30, Ni40, and Ni50 spectra in the wide velocity range from 12 mm s
-1 to -12 mm s
-1 was analyzed as a doublet component. The two doublets were considered to be derived from the superparamagnetic and amorphous (or ferrihydrite) components [
15].
On the other hand, the FTMS spectra of Ni0, Ni5, Ni10, and Ni20 were measured in the range of velocity from 12 mm s
-1 to -12 mm s
-1 at 86 K, as shown in
Figure 6 (a). Goethite was present in all spectra, and a doublet derived from amorphous (or ferrihydrite) are also present in Ni10 and Ni20. Additionally, the FTMS spectra of Ni20 and Ni50 were measured in a narrower range, as shown in
Figure 6 (b).
A sextet of FeOOH and a doublet of amorphous (ferrihydrite) were present in Ni20, and Ni50 had two doublets derived from ferrihydrite and a superparamagnetic phase. These results are in good agreement with XRD and XAFS measurements, as shown in
Figure 1 and
Figure 4. From
Table S2, the quadrupole splitting of the doublet derived from amorphous or ferrihydrite increased from 0.70 mm s
-1 for Ni20 to 0.89 mm s
-1 for Ni50. Therefore, it is considered that amorphous aging is progressing, and ferrihydrite is being formed [
17,
19]. This is consistent with the fact that Ni50 has a specific surface area lower than Ni20 as shown by BET analysis [
35]. In addition, Sample Ni20 contains goethite and an amorphous phase, while sample Ni50 also contains crystalline α-Ni(OH)
2, which is a possible reason for the lower surface area. The FTMS spectra of Ni10 and Ni20 were measured at low temperatures from 20K to 300K, as shown in
Figure S1, and the Mössbauer parameters of Ni10 and Ni20 are summarized in
Table S3 and
Table S4.
As shown in
Table 2 the Mössbauer parameters measured at room temperature indicated in a sextet and a doublet related to goethite and a superparamagnetic phase. Also, a doublet shown at RT is divided into two doublets related to superparamagnetic and amorphous phases. Additionally, with decreasing temperature from 300 K to 20 K, the doublets disappeared in both samples Ni10 and Ni20.
3.6. BET Analysis of α-NixFe1-xOOH NPs
The experiment involved varying the relative pressure (
P/
P0), where
P represents the adsorption equilibrium pressure, and
P0 represents the saturation vapor pressure) within the range of 0 to 1. The number of gas molecules adsorbed was quantified and graphed against the relative pressure, resulting in an isotherm.
Figure 7 displays the isotherms of α-Ni
xFe
1-xOOH NPs. There are many factors that influence the shape of the curve of an isotherm, such as the existence and dimensions of pores and adsorption energy. In 1985, the International Union of Pure and Applied Chemistry (IUPAC) published a classification system comprising six types of adsorption/desorption isotherms. Over the past three decades, two additional types have been introduced. The isotherm shape observed for Ni0 and Ni10 is classified as type II according to this classification. Type II indicates the absence of pores or the potential presence of macropores (pores with sizes exceeding 50 nm) [
36]. The IV and V types exhibit a unique phenomenon known as hysteresis, wherein the adsorption and desorption processes do not align as typically [
37].
Hysteresis, closely associated with capillary condensation, primarily occurs in the mesopore range [
38]. The hysteresis pattern observed can offer insights into the shape and structure of the pore. However, due to the diverse nature of hysteresis patterns, establishing a direct relationship between them and pore shape and structure is challenging.
On the other hand, the adsorption isotherms for Ni20, Ni30, Ni40, and Ni50 fall into the H4-type hysteresis classification. Type H4 is indicative of the presence of slit-type pores. This type of hysteresis pattern can sometimes be observed when micropores (pores with sizes of 2 nm or less) are present, as seen in type I isotherms [
39]. Based on the isotherm shape depicted in
Figure 7, it can be inferred that Ni0 to Ni10 possesses macropores, while Ni20 to Ni50 exhibits micropores. The presence of micropores can also be deduced from the rapid increase in adsorption at low relative pressures and the characteristics of the hysteresis loop. The BET method can extract information regarding the adsorption process, specifically from monolayer adsorption to multilayer adsorption. The sample's surface area can be determined by accurately calculating the degree of monolayer adsorption.
This can be achieved by multiplying the quantity of single-molecule adsorption by the cross-sectional area occupied by a single gas molecule as follows [
37].
P, P0, Vm, and C are parameters related to adsorption equilibrium pressure, saturation vapor pressure line, monolayer adsorption volume (e.g., adsorption volume when the gas molecules form a monolayer on a solid surface), and adsorption heat, respectively. The established relationship between P/P0 values in the range of 0.05 to 0.35 reveals significant insights.
The surface area, determined by applying the BET method, is presented in
Table 4. Based on these calculations, the surface area of Ni0 was found to be 45.1 m
2g
-1. With an increase in the Ni content from Ni0 to Ni20, the surface area progressively rises, reaching the highest value of 174.0 m
2g
-1 for Ni20. However, as the Ni content is further increased, the surface area decreases, reaching 96.3 m
2g
-1 for Ni40. Based on this result, it can be expected that the electrochemical properties of Ni20 will be the best due to the largest surface area compared to all prepared samples.
3.7. Bandgap Energy Derived from DRS of α-NixFe1-xOOH NPs
The band gap energy (
Eg) of α-Ni
xFe
1-xOOH NPs was determined using diffuse reflection UV-Vis spectroscopy. In this method, a powder sample is irradiated by incident light, and the incident light undergoes multiple transmissions, absorptions, and reflections within the powder. Consequently, absorption patterns are obtained from the reflections in different directions. The spectra obtained were analyzed using the Kubelka-Munk K-M function [
40].
Eg, which represents the energy difference between the valence band's top and the conduction band's bottom level [
41], is a crucial parameter in assessing photocatalytic activity.
Eg can be determined by constructing a Tauc plot [
42,
43]:
(2),
where
h is Planck's constant,
ν is the frequency,
α is the absorption coefficient,
A is the proportionality constant,
Eg is the band gap energy, and
n is a parameter depending on the type of transition in the material. For direct allowed transitions, it is taken as
n = 1/2; for indirect allowed transitions
n = 2 [
44].
Figure 8 displays the plot of (
hνα)
² versus (
hν) obtained from diffuse reflectance spectroscopy (DRS) measurements of Ni
xFe
1-xOOH NPs, enabling estimation of the
Eg values. The
Eg value for pure goethite (Ni0) was determined to be 2.71±0.01 eV.
Table 4 shows that the
Eg values for all Ni-doped samples were smaller than that of Ni0. Notably, the smallest
Eg value of 2.06±0.01 eV was observed for the Ni30 sample, determined as the x-axis intercept of the fitted straight lines depicted in
Figure 8.
This comparison highlights the greater effectiveness of Ni addition in reducing the band gap. Photogenerated carriers are produced when a photocatalyst absorbs photons with energy larger than its
Eg. Consequently, in this study, the Ni
xFe
1-xOOH NPs exhibit high excitability when exposed to a metal-halide lamp emitting light with wavelengths ranging from 250 to 750 nm. This excitation greatly enhances the photocatalytic performance [
45]. In conclusion, the above observations validate that the Ni
xFe
1-xOOH NPs photocatalysts investigated in this study possess an appropriate band gap and effectively utilize light.
3.8. Photo-Fenton Catalytic Ability of α-NixFe1-xOOH NPs
The reaction rate constant
k of methylene blue decomposition is calculated using the following equation (3) [
14]:
c0 is the concentration before the reaction of MB, and ct means the concentration of MB at time t.
Figure 9 illustrates the relationship between ln(
ct/
c0) and t for α-Ni
xFe
1-xOOH NPs. The values of
k, representing the rate constant for MB decomposition, were determined and are presented in
Table 4. For Ni0, the value of
k is 6.4±0.1×10
-3, while the maximum value of
k, 15.8±0.6×10
-3 min
-1, is observed for Ni10. As Ni is added to goethite, the
k values show an increasing trend from Ni5 to Ni10. However, for Ni15 and subsequent samples, the
k values are smaller than that for Ni0.
In the case of Ni15 and later samples, a decrease in goethite content is observed, accompanied by the generation of a significant amount of low-crystalline substances. This change in composition is believed to be the underlying cause of the lower k values observed. Notably, Ni10 exhibits the highest k value among the samples. This can be attributed to the findings from DRS and FeMS data, which suggest a reduction in the band gap and non-oxidation processes occurring in the Ni10 sample.
3.9. Electrical Properties - Impedance Spectra and DC Conductivity
Experimental data obtained through Solid-State Impedance Spectroscopy (SS-IS) are presented in a complex impedance plane, known as the Nyquist diagram, see
Figure 10 (a-c). Analysis of these plots involves employing electrical equivalent circuit (EEC) modeling, which utilizes a complex nonlinear least-square (CNLLSQ) fitting procedure. It is apparent that the impedance spectra observed in all examined samples exhibit a distinct semicircle, representing the bulk electrical process within the investigated α-Ni
xFe
1-xOOH NPs. This behavior can be effectively described by an equivalent R-CPE circuit. The parameters for each circuit element (
R,
A, and
α) were directly derived from the measured impedance data using the CNLLSQ method.
In the present study, we took a step further and examined the conductivity spectra of all the samples. The conductivity spectra of samples with Ni10 and Ni20 at different temperatures are shown in
Figure 10 (a-b), respectively. Although the conductivity isotherms have a similar shape, overall spectral features can be observed as follows. Firstly, there exists a frequency-independent conductivity plateau at low frequencies. This particular feature is associated with the long-range transportation of charge carriers and represents the overall resistance observed in the impedance spectra or DC conductivity. In addition, frequency-dependent conductivity, commonly referred to as conductivity dispersion, manifests itself with increasing frequency in the form of a power-law. This behavior arises from localized movements of charge carriers occurring over short distances. We used values of the fitting parameter
R obtained from modelling along with sample geometry to determine the total DC conductivity, as shown in
Table 5. The DC conductivity obtained is in good correlation with observed DC plateaus in conductivity spectra; see
Figure 11 (a-b).
The DC conductivity in our samples demonstrates a temperature-dependent behavior that follows an Arrhenius relationship, indicating semiconducting characteristics, see
Figure 11 (c). Consequently, the activation energy for the DC conductivity, E
DC, was determined for individual samples from the slope of log(σ
DCT) vs. 1000/T using the equation:
where
σDC is the DC conductivity,
σ0*is the pre-exponent,
kB is the Boltzmann constant, and
T is the temperature (K). The activation energy,
EDC, and DC conductivity,
σDC, at 90 °C for all investigated α-Ni
xFe
1-xOOH NPs samples are presented in
Table 5.
Continuing our analysis, we turn the focus to the comparison of the conductivity spectra at 110 °C, shown in
Figure 12 (a). Firstly, as previously mentioned, the shape of the conductivity spectra does not change with the composition. This consistency indicates that the mechanism of electrical transport remains unaffected. However, it is evident that the modification of α-Ni
xFe
1-xOOH NPs and the increase in Ni content has a negative effect, resulting in a decrease in DC conductivity, see
Figure 12 (b) and
Table 5. As the Ni/(Fe+Ni) ratio increases, the DC conductivity exhibits a nearly linear decline, from 5.52×10
-10 (Ω cm)
-1 for Ni10 to 5.30×10
-12(Ω cm)
-1 for the Ni50 sample. Conversely, the activation energy for DC conductivity,
EDC, follows the opposite trend, with values increasing in the 73.0–82.5 kJmol
-1 range.
The goethite structure can be described as parallel double chains of edge-sharing octahedra. These chains consist of Fe
III bonded to three oxide ions and three hydroxides extending along the [001] direction and are linked to neighboring double chains by corner sharing.
Guskos et al. [
46] conducted an electrical study on goethite and proposed that charge transport occurs through thermally activated three-dimensional hopping of electrons via oxygen vacancies, based on DC electrical measurements conducted over a wide temperature range.
However, it is important to note that our understanding of the electronic structure of iron oxides and oxyhydroxides, including goethite, remains incomplete. In our study for Ni
xFe
1-xOOH NPs, the obtained values for activation energy (0.75–0.85 eV), see
Table 5, are almost 3 times lower in comparison to goethites studied by
Guskos et al [
46]. Our findings align more closely with the activation energy values reported for magnetite and hematite. Similar values are observed in the literature for various materials with disordered or partially disordered structures and dominant electron transport [
47,
48,
49,
50].
Furthermore, Vitaly et al. [
51], showed that RT charge transport in goethite is primarily governed by thermally activated hopping of small polarons, with the associated mobility being higher compared to other iron oxyhydroxide (FeOOH) polymorphs. Small polarons are formed when electrons self-trap onto an iron center, resulting in conduction through phonon-mediated hops between centers [
52]. Different inequivalent paths for electron hopping characterized by different Fe–Fe bond distances and species bridging two neighboring Fe atoms are identified [
51]. The pathway involving migration along the double chain ([001] direction) through shared octahedral edges, with electron transport mediated by O and OH species, is characterized by Fe atoms with parallel spins and with the shortest Fe
2+-Fe
3+ distance of approximately 3 Å.
Su et al. [
53], studied the electrical conduction mechanism of goethite under pressures up to 17.1 GPa using Impedance spectroscopy. The results indicate a pressure-induced conduction mechanism transition around 5 GPa from mixed protonic-electronic conduction to pure electronic conduction which is associated with the pressure-induced magnetic state transition.
In our study, we used IS method in an inert atmosphere (LN2), so the protonic contribution is expected to be inhibited and does not contribute to total conductivity. Thus, the obtained trend in our study could lead to the conclusion that as the Ni content is increasing, the structure and bonding are affected in the studied NixFe1-xOOH NPs samples, which leads to the increase of polaron hopping activation energies, thus decreasing the ground state carrier mobility compared to pure goethite. Also, at the same time, with Ni doping and a decrease in Fe content, the charge carrier concentration decreases. Additionally, the presence of mixed metal centers (Fe, Ni) does not appear to have a favorable effect on polaron transport. Overall, the effects above collectively contribute to a decrease in the DC conductivity as the Ni content increases in the studied NixFe1-xOOH nanoparticles.
3.10. The Electrochemical Properties of α-NixFe1-xOOH NPs/Li- and Na-Ion Batteries
Table 6 shows the discharge capacity and capacity retention of Ni0, Ni10, Ni20, and Ni30 samples as a cathode in LIB under a low current rate at 5 mA g
-1; the initial discharge capacity of Ni0 was the highest, reaching 1069 mAhg
-1. The discharge capacity decreased with the increase in Ni content to 250 mAhg
-1 for Ni30. Moreover, the capacity retention of all samples suddenly dropped after 10 cycles. The capacity retention of Ni0, Ni20, and Ni30 was less than 1 %, while Ni10 was 7.4 %.
On the other hand, after a tenfold increase of the current rate to 50 mA g-1, the initial discharge capacity of Ni0, Ni10, and Ni20 was 333, 350, and 494 mAhg-1, respectively, and the discharge capacity increased with the Ni content.
However, capacity retention decreased when the Ni content was increased. We can conclude that after increasing the current rate from 5 to 50 mA g
-1, the initial capacity of all samples decreased, while the capacity retention enhanced, as shown in
Table 6 and
Figure 13 (a).
In contrast, the discharge capacities of Ni0, Ni10, Ni20, Ni30, and Ni50 were 110, 116, 223, 189, and 202 mAhg
-1, respectively, which was proportional to the size of the surface area as shown in
Table 4 [
54,
55].
Additionally, the capacity retention after 30 cycles of all samples after adding Ni is lower than that for Ni0, as shown in
Table 7 and
Figure 13 (b). According to the aforementioned results concerning α-Ni
xFe
1-xOOH NPs as cathode material in both LIB and SIB, with a specific focus on capacity retention after 30 cycles, one can deduce that the performance of Ni
xFe
1-xOOH NPs as active cathode materials is more comparable in SIB compared to LIB.