TiO₂-CoFe₂O₄ and TiO₂-CuFe₂O₄ Composite Films: A New Approach to Synthesis, Characterization, and Optical and Photocatalytic Properties

Abstract

Here, we report a new simple and fast method of cobalt and copper ferrites film fabrication on the Ti/TiO₂ surface. The approach is based on the deposition of gelatin gel containing copper and cobalt nitrates on the surface of porous oxide-silicon coatings on titanium obtained by plasma electrolytic oxidation (PEO) followed by two-stage annealing at 300 °C and 800 °C to yield ferrite films with good adhesion to PEO layer. The presence of Co/Cu ferrite phases was confirmed by EDX analysis, XRD, and Mössbauer spectroscopy. TiO₂-CoFe₂O₄ and TiO₂-CuFe₂O₄ composite films have excellent performance in the photocatalytic degradation of indigo carmine as a model dye at pH 3 under UV and visible irradiation. The suggested approach to obtain ferrite/TiO₂ composite films is promising for the development of magnetic materials, sensors, catalysts, and photocatalysts for various applications.

1. Introduction

Materials based on transition metal ferrites are increasingly being used in various fields of science and technology due to their excellent magnetic, catalytic, and photo- and electrocatalytic properties, among others [1,2,3]. Application of nanoparticles of transition metal ferrites for magnetic resonance imaging, cancer diagnosis at the primary stage, magnetothermal, anticancer and antimicrobial therapies, hyperthermia, and drug delivery have been recently reviewed [4,5].

Recent advances in the synthesis of cobalt ferrite nanoparticles for environmental applications, including treatment of water/wastewater, environmental remediation, and photocatalytic degradation of dyes, have been reported in [5].

Copper and cobalt ferrite nanowires and nanoparticles have been successfully used for the decomposition of ciprofloxacin [6,7]. These particles have demonstrated high efficiency when used repeatedly, highlighting their potential in the fabrication of materials to combat antibiotic pollution.

Many researchers have reported the application of ferrite nanoparticles as efficient catalysts and photocatalysts for dye degradation. The dyes degradation depends on many factors, such as unique charge separation, morphology, and crystalline nature of the materials. The influence of the annealing temperature during the synthesis of cobalt ferrite on the efficiency of methylene blue degradation was shown in [8]. The authors found that at the optimal CoFe₂O₄ annealing temperature of 400 °C efficiency of photocatalytic performance reached 74%.

The authors of [9] reported the efficient use of copper ferrite nanoparticles for the photocatalytic degradation of malachite green. It has been demonstrated that the photocatalytic activity depends on the degree of incorporation of copper ions into the magnetite host structure. The sample with the lowest degree of incorporation of copper ions showed the highest photocatalytic activity.

There are a number of problems with the application of catalysts as individual particles in the powdered form. Namely, the difficulty of nanoparticle separation at the end of the technological cycle makes it impossible to scale up the process of wastewater decontamination from organic pollutants and reuse catalysts. In order to solve this problem, in recent years, many research efforts have focused on developing composites by combining powders with large surface areas with different solid substrates, such as nanoparticle carriers [10,11,12]. Additionally, the combination of semiconductors with different electronic structures can lead to a decrease in the rate of electron-hole recombination and thereby enhance the interaction of electrons and holes with organic pollutants to achieve high photocatalytic activity.

As a substrate for the deposition of active materials nanoparticles, in particular transition metal ferrites, it is promising to use oxide layers with a rough developed surface and a porous structure, which can be generated by plasma electrolytic oxidation (PEO) on a valve metal in aqueous electrolytes containing alkali metal metasilicates, for example, Na₂SiO₃ [13,14,15,16]. The PEO method is based on the electrochemical formation of oxide layers on valve metals under the action of spark or microarc electric discharges at the metal/electrolyte interface.

The formation of iron-containing oxide PEO coatings on metals can yield magnetic materials, such as “metal/magnetic film” systems with ferromagnetic properties. These materials can be utilized for various purposes, such as constructing electromagnetic screens or structures that absorb electromagnetic radiation [17].

For example, Fe₂O₃ nanoparticles have successfully immobilized on a titanium substrate using the PEO process [18]. Enhancement of photocatalytic performance can be achieved through the immobilization of compounds with strong visible light absorption. This approach has been proven effective for the acceleration of methylene blue degradation rates when exposed to UV radiation, as reported by the authors [18].

Authors of [19] successfully fabricated TiO₂ film containing Fe₂O₃ on titanium subjected to the PEO process and demonstrated the high efficiency of the composite for the decomposition of methylene blue in visible light.

Here, we suggest a new approach to fabricate composite photocatalysts via the deposition of cobalt and copper ferrite stabilized by gelatin on Ti/TiO₂ PEO-coatings and demonstrate the application of these composites for the degradation of indigo carmine dye. In comparison with earlier reported methods of transition metals ferrites films fabrication using such eco-unfriendly reagents as 2-methoxyethanol [20,21], ethanolamine [20], ethylene glycol monomethyl ether [22], we have used gelatin as a green stabilizer in aqueous solutions without any other additives.

2. Materials and Methods

2.1. Materials

All reagents were of analytical grade.

2.2. Synthesis of the Ferrites

Cobalt and copper ferrites powders were synthesized using Co(NO₃)₂·6H₂O, Cu(NO₃)₂·3H₂O, Fe(NO₃)₃·9H₂O, and pure edible gelatin powder as stabilizing agent. First, 0.4 g of gelatin was dispersed in 10 mL of distilled water at room temperature. Then, 0.4947 g (1.7 mmol) of cobalt nitrate or 0.4107 g (1.7 mmol) of copper nitrate and 1.3736 g (3.4 mmol) of iron nitrate were added to the gelatin solution [23]. The resulting gelatin-based gels to fabricate powders and films of cobalt and copper ferrites were obtained by keeping these dispersions under stirring at 100 °C for 1 h. In order to obtain powders, the resulting gels were annealed in a SNOL 7.2/1100 muffle furnace (Lithuania) with a ceramic chamber in the air in 2 stages: (1) annealing at 300 °C for 2 h, (2) crushing materials obtained at stage 1, and further annealing at 800 °C for 30 min.

2.3. Fabrication of the PEO-Coatings

PEO coatings were formed on VT1-0 titanium plates (wt. %: 0.7 Al, 0.25 Fe, 0.10 Si, 0.07 C, 0.04 N, 0.2 O, 0.01 N; other impurities—0.30 and Ti-the rest) with a size of 20 × 20 × 0.5 mm³. Before PEO-treatment, the surfaces of the samples were subjected to mechanical grinding and chemical polishing at 60–80 °C in a mixture of concentrated HF:HNO₃ acids (at a volume ratio of 1:3) for 3 s, then washed with distilled water and dried in the air [24].

An electrolyte based on distilled water and Na₂SiO₃·5H₂O was used to form PEO coatings on titanium. The silicate–alkaline electrolyte was chosen because it was earlier demonstrated as efficient media to form coatings with a developed surface and high moisture absorption [25]

The PEO processing was carried out in a 1000 mL polypropylene beaker. The electrolyte was magnetically stirred and cooled by cold water, which was pumped through a spiral-shaped nickel coil, keeping the electrolyte temperature below 35 °C. A computer-controlled TER4-100/460N (ZAO Converter, Saransk, Russia) unipolar thyristor unit was used as a power source. The nickel coil and the processed samples were connected to the negative and positive poles of the power source, respectively.

The oxide films were formed for t = 10 min in DC mode at current density i = 0.1 A/cm². After PEO treatment, the samples were washed with distilled water and dried in the air at ~70 °C.

2.4. Ferrite Films Fabrication on the PEO-Coatings

Cobalt and copper ferrites containing composites were fabricated by distributing 200 µL of the gelatin-based gels on 4 cm² of the surface area of the PEO coatings. Then, the samples were dried at 80 °C for 1 h and annealed in a muffle furnace to form ferrites phases on the PEO coatings. Annealing in a muffle furnace was carried out under the same conditions as described in Section 2.2 for the synthesis of powder cobalt and copper ferrites.

2.5. Characterization of the Powders and Coatings

The thickness of the coatings was measured on a thickness gauge VT-201 (Control. Measurement. Diagnostics. LLC, Moscow, Russia).

Morphology and elemental analysis were performed using Hitachi TM3000 (Hitachi High-Technologies Corporation, Tokyo, Japan) scanning electron microscope (SEM) equipped with Bruker Quantax 70 (Bruker AXS GmbH, Karlsruhe, Germany) energy dispersive X-ray (EDX) spectrometer at an accelerating voltage of 15 kV.

X-ray powder diffraction analysis (XRD) of ferrites powders and films on the PEO-coatings was carried out using a D8 ADVANCE X-ray Bruker diffractometer (Bruker AXS GmbH, Karlsruhe, Germany) in CuK_α-radiation. The identification of the compounds in their composition was carried out using the EVA search program with a PDF-2 database.

The Mössbauer spectra were obtained at room temperature using a Wissel (Wissenschaftliche Elektronik GmbH, Ortenberg, Germany) spectrometer in transmission geometry with a 512-multichannel analyzer. ⁵⁷Co in a rhodium matrix with an activity of 35 mCi was used as a radiation source. The Mössbauer spectra were fitted using the WinNormos program in order to obtain the values of isomer shift (δ), quadrupole splitting (Δ), hyperfine field (H), linewidth (Γ), and relative sub-spectrum area. The velocity scale was calibrated using the spectrum of metallic iron (α-Fe). The value of isomer shifts was determined relative to the center of gravity of the α-Fe spectrum.

The acid–base properties of the composite surface were studied using a SevenCompact pH meter (Mettler-Toledo GmbH, Greifensee, Switzerland). Therefore, 40 mL of distilled water was introduced into the potentiometer. After stabilization of the potential of the glass electrode, the sample was placed in water, and the dependence of pH on time was recorded for 15 min.

Diffuse reflectance spectra were recorded in the wavelength range from 300 to 800 nm on an SF-2000 spectrophotometer (OKB Spectr LLC, St. Petersburg, Russia) with a diffuse reflectance attachment with a spectral resolution of 1 nm (or Shimadzu). Halogen and deuterium lamps were used as radiation sources, and the standard was BaSO₄. The optical band gap (BG) was determined from the position of the fundamental absorption edge according to the Tauc Equation (1):

$${\left(\mathsf{\alpha }·\mathrm{h}\mathrm{v}\right)}^{\frac{1}{\mathrm{n}}}=\mathrm{A}\left(\mathrm{h}\mathrm{v}-{\mathrm{E}}_{\mathrm{g}}\right),$$

(1)

where E_g is the energy of the optical band gap, h is Planck’s constant, ν is the oscillation frequency of electromagnetic waves, α is optical absorption, A is a constant, and n depends on the type of interband electronic transition (n = 1/2 for a direct allowed transition). The band gap was determined by approximating the linear part of the decay of the Tauc curve along the abscissa axis, along which the incident photon energy hν was plotted.

2.6. Photocatalytic Tests

The photocatalytic properties of TiO₂-Co(Cu)Fe₂O₄ film composites were studied in the reaction of indigo carmine (IC) degradation under UV irradiation. The photocatalytic tests were carried out in a quartz beaker filled with 50 mL of IC solution (10 mg/L, pH 4.6 and 2.9). The initial pH of the IC dye aqueous solution was 4.6, while 2.9 was the pH of IC solution acidified with sulfuric acid (0.5 mol/L).

When the photocatalytic tests were performed under UV irradiation, the composites were placed vertically near the wall of a quartz beaker filled with IC solution. When the tests were carried out under visible light irradiation, the sample was placed horizontally on a special plastic perforated holder 1 cm below the solution surface. A 100 W SB-100P irradiator (maximum radiation with a wavelength of 365 nm) and a 35 W Xenon lamp (wavelength range of 510–680 nm) were used as sources of UV and visible light, respectively. The distance between the light source and the sample surface was precisely fixed at 5 cm in both cases.

Before irradiation, the IC solution with the sample was kept in the darkness under stirring for 30 min to ensure the adsorption-desorption equilibrium. Then the initial absorbance of the IC solution was measured as the reference, and the composite was exposed to UV or visible light for 3 h under constant stirring. The photocatalytic activity of the samples was evaluated by monitoring the decrease in the absorbance of IC dye at a wavelength of 610 nm using a UV-2600PC scanning UV–Vis spectrophotometer (Shimadzu, Kyoto, Japan) in a quartz cuvette with an optical length of 1 cm. The dependence of absorbance on concentration obeyed the Bouguer–Lambert–Beer law that allowed the determination of the IC degradation as follows:

$$\mathsf{\chi }=\frac{{\mathrm{A}}_0-\mathrm{A}}{{\mathrm{A}}_0}·100\mathrm{\%},$$

(2)

where A₀ and A are the absorbances of the IC solution before and after reaction.

3. Results and Discussion

3.1. Films Fabrication Method

Figure 1 shows the sequence of steps to obtain Ti/TiO₂-CoFe₂O₄ and Ti/TiO₂-CuFe₂O₄ film composites.

Gelatin-based solutions containing cobalt/copper and iron nitrates, which were used for the film’s fabrication, were homogeneous and transparent, without signs of opalescence. After the distribution of the resulting gelatin-based gels over the PEO-coating and drying at 80 °C, the thick gel layer was formed on Ti/TiO₂ surface. Subsequent two-stage annealing resulted in the formation of a ferrite phase on the PEO-coating, which displays high adherence to the titanium substrate. Almost the same values of the thicknesses of ferrite coatings were measured in different areas for the CoFe₂O₄-PEO and CuFe₂O₄-PEO composites; they were equal to 18 ± 2 and 23 ± 2 μm, respectively, indicating the uniformity of the deposited coatings.

3.2. Morphology

Figure 2 shows the morphology and EDX maps of the cobalt and copper ferrites powders and films on the Ti/TiO₂ PEO coatings.

Although SEM images (Figure 2a,b) do not give information on the size of primary ferrite particles due to their agglomeration to the large flakes after combustion of the gelatin matrix, EDX analysis confirmed uniform distribution of Co/Cu and Fe over the surface of the sample. Figure 2c,d demonstrate the difference between the morphology of powdered ferrites and ferrite films deposited on Ti/TiO₂ PEO coatings formed in a silicate electrolyte. After annealing the gel-like layer, the composite surface repeated the morphology of the initial PEO-coating, which has a coral-like structure [25].

The atomic ratios between Fe and Co/Cu in the samples (

Table 1. Elemental analysis of Co- and Cu- ferrites as powders and films on the PEO-coatings.

Sample	Atomic Ratios in Samples Fe:Cu/Co	Cu/Co, at %	Fe, at %	O, at %	Ti, at %	Si, at %
CoFe2O4	2:1.15	17.64	30.75	51.60
CuFe2O4	2:1.07	16.16	30.09	53.74
CoFe2O4-PEO	2:1.01	2.98	5.93	64.40	6.00	20.69
CuFe2O4-PEO	2:1.04	2.43	4.66	66.31	6.67	19.93

) were close to the theoretical value of 2:1, hence indicating the successful synthesis of cobalt and copper ferrite phases in both powder and film forms. To elucidate the structure of Co- and Cu- ferrites powders and films in detail, the obtained materials were investigated using X-ray diffraction and Mössbauer spectroscopy. The optoelectronic properties were also analyzed for ferrites films on the PEO coatings.

3.3. X-ray Diffraction

Figure 3 shows the X-ray diffraction patterns of the cobalt and copper ferrites powders and films on the Ti/TiO₂ PEO-coatings, which are in accordance with the literature data for crystalline cobalt and copper ferrites [26,27,28,29], which confirms the phase composition of the obtained samples.

XRD patterns of ferrite films on the Ti/TiO₂ PEO-coatings contain, in addition to ferrite phases, phases of Ti and rutile modification of TiO₂ and SiO₂, which are in agreement with XRD data in [30,31,32].

In addition to XRD analysis, Mössbauer spectra were obtained to confirm the phase composition of the materials.

3.4. Mössbauer Spectroscopy

Mössbauer spectra of the cobalt and copper ferrites powders were typical for ferrite compounds (Figure 4 and

Table 2. Mossbauer parameters for Co- and Cu- ferrites powders, T = 298 K *.

Sample	δ, mm/s	Δ, mm/s	H, kOe	Γ, mm/s	Relative area, %	Assignment
CoFe2O4	0.23	−0.05	498.8	0.45	67.73	A site: Th Fe(III)
	0.52	0.17	505.1	0.51	32.27	B site: Oh Fe(III)
CuFe2O4	0.26	−0.02	481.7	0.47	61.58	A site: Th Fe(III)
	0.36	−0.33	507.6	0.37	38.42	B site: Oh Fe(III)

* Isomer shift (δ), quadrupole splitting (Δ), hyperfine field (H), linewidth (Γ). Values of δ are reported relative to α-Fe metal. Fitting error in the values of δ, Δ, and Γ remained below 0.01 mm/s.

). In agreement with the literature data [23,33,34], the spectrum of cobalt ferrite powder was fitted with two components: the first sextet had an isomer shift of 0.23 mm/s, quadrupole splitting of −0.05 mm/s, and hyperfine field of 498.8 kOe, corresponding to the high-spin Fe(III) in tetrahedral (Th) A site of spinel structure; the second sextet had an isomeric shift 0.52 mm/s, quadrupole splitting of 0.17 mm/s, and hyperfine field of 505.1 kOe, corresponding to the high-spin Fe(III) in octahedral (Oh) B site of spinel structure.

The spectrum of copper ferrite powder was fitted with two components: the first sextet had an isomer shift of 0.26 mm/s, quadrupole splitting of −0.02 mm/s, and hyperfine field of 481.7 kOe, corresponding to high-spin Fe(III) in tetrahedral A site of the spinel structure; the second sextet had an isomeric shift 0.36 mm/s, quadrupole splitting of −0.33 mm/s and hyperfine field of 507.6 kOe, corresponding to high-spin Fe(III) in octahedral B site of spinel structure.

For spinel ferrites, the general formula is (M²⁺_1−xFe³⁺_x)[M²⁺_xFe³⁺_2−x]O₄ (cations in parentheses correspond to tetrahedral A sites, in square brackets to octahedral B sites), and x is the inversion parameter quantifying the distribution of M²⁺ and Fe³⁺ cations among these sites [35]. The relative area values of each component of the Mössbauer spectra general formula was calculated as (Co_0.32Fe_0.68)[Co_0.68Fe_1.32]O₄ and (Cu_0.38Fe_0.62)[Cu_0.62Fe_1.38]O₄ for cobalt and copper ferrites, respectively.

3.5. Indigo Carmine Degradation

Preliminary tests have shown that irradiation of indigo carmine (IC) solutions with natural pH = 4.6 does not lead to a noticeable dye decolorization in both UV and visible light. However, the decrease in IC solution pH to 2.9 facilitated dye decolorization in the presence of composite films. The absorption spectra of IC at pH 2.9 obtained before and after UV and visible irradiation are shown in Figure 5a. In the absence of catalyst (blank experiment), the IC degradation degree was insignificant and did not exceed 2, 5, and 13% after 3 h in darkness and exposure to visible and UV irradiation, respectively.

When the CuFe₂O₄-containing composites were used as photocatalysts, the IC degradation degree reached 74% and 45% under UV and visible irradiation, respectively. Composites containing CoFe₂O₄ showed a lower efficiency of IC degradation, which was ~59% under UV and 44% under visible light, respectively (Figure 5b). Thus, we concluded that the efficiency of IC photodegradation depended on both the pH of the solution and the type of ferrite.

A decrease in the IC solution pH affects both the surface charge of the coatings and the acid-base equilibrium of the IC in the aqueous solution. IC is an anionic dye, which exacts in the anionic form in the pH range of 3–11. The point of zero charge of CuFe₂O₄ and CoFe₂O₄ is 5.4 and 7.2, accordingly [36,37], while the surface of TiO₂ is negatively charged at pH > 3 [38].

The acid-base properties of the surface coatings were studied by recording pH versus time curves in the absence and presence of samples after stabilization of the pH of distilled water. It can be seen from Figure 6 that the pH of water in the absence of samples does not change over time. In the presence of both samples, a gradual increase in water pH over time is observed. One can determine the pH value corresponding to the isoionic state of the surface from the equilibrium pH value. pH of the isoionic state was 5.58 and 5.6 for the CoFe₂O₄-PEO and CuFe₂O₄-PEO composites, respectively. An increase in water pH in the presence of composites over time indicates the predominance of basic Lewis sites on the surface, which form Bronsted acid sites after hydration. This process is accompanied by the release of the hydroxy groups into the water and, thus, an increase in pH. It can be assumed that the predominance of basic Lewis centers on the surface of the ferrite films prevents the attraction of negatively charged IC anions. Obviously, when the pH of the solution decreases, protonization of the basic Lewis centers increases, which promotes the interaction of the film surface with IC anions.

Therefore, at the pH of the solution below, three composite surfaces are positively charged due to protonation, indicating favorable conditions for interactions with anionic dye.

Since indigo carmine has pKa = 12.2, its amino groups are expected to be in the protonated form in acid media and the unprotonated form in highly alkaline media [39]. The negative charge of IC in an acidic medium comes from the ${\mathrm{S}\mathrm{O}}_3^{-}$ group. At pH < 3, sufficiently strong electrostatic interactions arise between the positively charged surface of TiO₂-ferrite composites and IC anions, which leads to their strong adsorption on the surface. A noticeable decolorization of IC at a low pH results from the direct interaction of the dye molecule with the photocatalyst, which is necessary for the photocatalytic reaction to occur. A similar effect of the pH on the IC degradation was reported in [40,41,42].

The first-order kinetic equation was used o describe the degradation kinetics of IC under the action of UV and visible radiation for 180 min. The kinetic equation is expressed as:

$$Ln\left(\frac{C_0}{C}\right)=-\mathrm{k}\mathrm{t},$$

(3)

where C₀ and C (mg/L) are the initial and final concentrations of IC in the solution, respectively; k is the IC degradation rate constant; and t is the reaction time (min). The Bouguer Lambert–Beer law establishes a relationship between the concentration of a solution and its optical density, expressed in the equation A = ε × b × C, where A, ε, b, and C are the absorbance of the solution, molar absorptivity, path length, and solution concentration, respectively. Therefore, the rate constant (k) can be determined graphically from the equation:

$$Ln\left(\frac{A_0}{A}\right)=\mathrm{k}\mathrm{t}.$$

(4)

It is clearly seen that in all cases, ln(A₀/A) linearly depends on time with the correlation coefficient (R²) > 0.99 (Figure 7 and

Table 3. Degradation rate constants (k) and correlation coefficients (R²) for IC degradation under UV and VIS irradiation.

Sample	UV		VIS
	k, min−1	R2	k, min−1	R2
Blank IC	0.57 × 10−3	0.99	0.41 × 10−3	0.99
CoFe2O4-PEO	5.00 × 10−3	0.99	3.02 × 10−3	0.99
CuFe2O4-PEO	7.97 × 10−3	0.99	3.60 × 10−3	0.99

). The calculated IC degradation rate constants are given in Table 3. In the absence of catalysts, the degradation rate of IC in both visible and UV light is much lower than in their presence. It can be seen from Table 3 that the CuFe₂O₄-PEO sample provides a higher rate of IC degradation than the CoFe₂O₄-PEO sample under the action of UV and visible radiation.

In order to give more insight into the photocatalytic activity of the samples, their optoelectronic properties were studied.

3.6. Optoelectronic Properties

The absorption spectrum fitting method with the Tauc model was utilized to estimate the optical band gap in PEO coatings containing cobalt and copper ferrites. To determine the optoelectronic properties of the synthesized samples, diffuse reflectance spectra (DRS) were taken in the range of 300–800 nm (Figure 8a). It is noticeable that the Ti/TiO₂ (PEO) sample exhibits a unique spectrum for titanium dioxide and does not absorb within the λ = 400–800 nm region, which distinguishes it from the modified surface samples [43]. CuFe₂O₄-PEO and CoFe₂O₄-PEO samples exhibit a broad absorption peak within the λ = 400–800 nm range, indicating superior light absorption in the UV to visible range compared to Ti/TiO₂. Therefore, these materials show high potential as photocatalysts with a wide range of applications.

Extrapolation of the linear part of the (αhν)² curve to the (hν) axis showed that the energy of the direct allowed transition for the Ti/TiO₂, CoFe₂O₄-PEO, and CuFe₂O₄-PEO sample is 3.11, 1.68, and 1.89 eV, respectively (Figure 8b). The value obtained for Ti/TiO₂ corresponds to the literature data for titanium dioxide (~3.0 eV) [44]. According to [45,46,47], the value of the band gap for CoFe₂O₄ is ~1.4–1.76 eV, which correlates well with our results. For CuFe₂O₄, according to the literature data [48,49,50], the values range between 1.2 and 2.1 eV. Some discrepancies between the data obtained and the results of other authors may be associated with the complex compositional structure of the samples, leading to a shift in the edges of the valence and conduction bands.

3.7. Charge Carrier Mechanism

For a better understanding of the mechanism of charge carrier separation in titanium dioxide-based CoFe₂O₄ or CuFe₂O₄ composites, the potentials of the conduction (CB) and valence (VB) bands at the point of zero charge were calculated using Equations (5) and (6) [51]:

$${\mathrm{E}}_{\mathrm{V}\mathrm{B}}=\mathrm{X}-{\mathrm{E}}^{\mathrm{e}}+\frac{1}{2}{\mathrm{E}}_{\mathrm{g}},$$

(5)

$${\mathrm{E}}_{\mathrm{C}\mathrm{B}}={\mathrm{E}}_{\mathrm{V}\mathrm{B}}-{\mathrm{E}}_{\mathrm{g}},$$

(6)

where E_g is the band gap; E_CB is the conduction band potential; E_VB is the valence band potential; X is the absolute electronegativity of the semiconductor, calculated as the geometric mean of the absolute electronegativity of the constituent atoms; and E^e is the energy of free electrons on the hydrogen scale (~4.5 eV) [52].

To calculate the parameters and plot the energy diagrams, the values of E_g for pure phases were taken from [53,54,55]. The calculated X, E_g, E_CB, and E_VB values for TiO₂ and CoFe₂O₄/CuFe₂O₄ are presented in

Table 4. Values of Mulliken semiconductor electronegativity X, band gap energy E_g, conduction band bottom E_CB, and valence band top E_VB for individual components of the formed heterostructures.

Sample	X, eV	Eg, eV (Ref)	ECB, eV	EVB, eV
TiO2	5.81	3.11 [53]	−0.25	2.87
CoFe2O4	5.81	1.72 [54]	−0.45	2.17
CuFe2O4	5.85	2.04 [55]	−0.33	2.37

.

For a better understanding of the charge transfer process, an energy level diagram was plotted for individual semiconductors and formed heterostructures.

Figure 9a presents an energy diagram for a CoFe₂O₄-PEO sample. Before coupling, the Fermi level (E_f) of the TiO₂ (n-type) semiconductor is closer to the edge of the conduction band (CB), and for CoFe₂O₄ (p-type) is closer to the valence band potential (VB) [56]. The CB potential of CoFe₂O₄ (−0.45 eV) is less negative than that of TiO₂ (−0.25 eV), while the VB potential of TiO₂ (+2.87 eV) is more positive than that of CoFe₂O₄ (+2.17 eV). After joining two semiconductors, E_f CoFe₂O₄ moves up, while E_f TiO₂ moves down until E_f TiO₂ and CoFe₂O₄ are aligned. Thus, a p-n heterojunction is formed at the interface between CoFe₂O₄ and TiO₂. Because of the formation of the p-n heterojunction, the region close to the p–n interface is charged, creating an internal electric field. In this process, the TiO₂ and CoFe₂O₄ bands bend, the overall energy level of CoFe₂O₄ increases, and the TiO₂ level decreases [57]. For the CuFe₂O₄-PEO composite, the energy diagram has a similar form. Figure 9b shows that the VB CuFe₂O₄ edge (+2.37 eV) is much more positive than the VB TiO₂ (+2.87 eV), while the CB CuFe₂O₄ potential (+0.33 eV) is positive and the CB TiO₂ is negative (−0.25 eV). Under UV light irradiation, both TiO₂ and CoFe₂O₄/CuFe₂O₄ can absorb photons giving rise to electrons and holes, whereas, under the action of visible light, the formation of photogenerated charges occurs only in CoFe₂O₄/CuFe₂O₄. However, we assume that the IC degradation mechanism was similar in both cases.

Because the position of CB CoFe₂O₄/CuFe₂O₄ is higher than that of TiO₂, electrons can easily move from CB CoFe₂O₄/CuFe₂O₄ to CB TiO₂. In turn, photogenerated holes can easily move from the higher VB TiO₂ (+2.87 eV) to the lower VB CoFe₂O₄ (+2.17 eV). Therefore, the photogenerated holes and electrons can be separated effectively, which significantly enhances photocatalytic activity. Because the CB of both ferrites is more negative than O₂/•O₂⁻ potential (−0.33 eV vs. NHE), photogenerated electrons can react with O₂ to generate •O₂⁻ radicals, which are active species with strong oxidizing ability and can react with IC. Before the connection of two semiconductors, the valence band positions of CoFe₂O₄ (2.17 eV) and CuFe₂O₄ (2.37 eV) are more positive than the potential OH•/OH^-(1.99 eV), indicating that •OH can be generated via the oxidation reaction of h+. Moreover, only E_CB (CuFe₂O₄) is more positive than OH•/H₂O (2.27 eV). This suggests that the formation of hydroxyl radicals in the composites with copper ferrite occurs more efficiently than in the composites based on cobalt ferrite. Therefore, IC can be degraded via the /•O²⁻, h+, or OH• oxidation pathway.

Therefore, based on the data given in the diagrams, it is impossible to give an unambiguous answer to the question of why copper ferrite is more active than cobalt ferrite. However, according to the literature, which describes other methods to determine the potentials of the valence band and the conduction band, CuFe₂O₄ has a more negative character of the conduction band (−1.1 eV), while cobalt ferrite is more positive (0.65 eV). According to [58], copper ferrite has a strong reduction ability, and cobalt ferrite has a moderate redox ability, which determines the higher photocatalytic activity of the first compared to the second. In [59], the authors also explained the higher photocatalytic activity of copper ferrite compared to nickel ferrite by the more negative nature of its conduction band.

4. Conclusions

In this paper, we reported a new method for the fabrication of cobalt and copper ferrite films on PEO coatings. The Fe:Cu/Co atomic ratios obtained by EDX analysis resulted in being close to 2:1. XRD patterns indicate the presence of cobalt and copper ferrite phases both in crystalline powders and in films on PEO-coatings. The Mössbauer spectra confirmed the structure of the obtained copper and cobalt ferrites: the presence of two sextet subspectra related to the tetrahedral Fe(III) A site and octahedral Fe(III) B site was determined. The photocatalytic properties of the obtained materials were demonstrated by the decomposition of indigo carmine (IC) as a model dye. It was shown that in the presence of CuFe₂O₄ -containing composite, the IC degradation degree reached 74% and 45% under UV and visible irradiation, respectively. While CoFe₂O₄ -containing composite provided IC degradation degrees of 59% and 44% under the same conditions, respectively. The high photocatalytic activity of the formed film composites is explained by the formation of p-n heterojunctions between TiO₂ and CoFe₂O₄ (CuFe₂O₄), and pH dependence of IC degradation is determined by acid-base properties of the coatings promoting IC attraction in acidic medium.

In view of the simplicity, rapidity, and feasibility of scaling up, the proposed method of ferrite film composite fabrication is promising for the development of magnetic materials, sensors, catalysts, and photocatalysts.