Augmenting the catalytic activity of CoFe₂O₄ by substituting rare earth cations into the spinel structure

Abstract

The current research work evaluates enhancement in photo-Fenton activity of pristine CoFe₂O₄ by inserting a very small quantity of rare earth (RE) cations into its spinel structure. For this purpose, RE metal (Ce³⁺ and La³⁺) doped CoFe₂O₄ (having compositions CoLa_xFe_2−xO₄; x = 0.1, 0.2, 0.3, 0.4, 0.5 and CoCe_xFe_2−xO₄; x = 0.1, 0.2, 0.3) was synthesized by employing a sol–gel methodology. The samples were well characterized using diverse techniques (FTIR, powder XRD, HR-TEM, VSM, UV-DRS, BET surface area analyzer). The photo-Fenton reactions were performed using five probe molecules i.e. textile dyes (reactive black 5 and remazol brilliant yellow) and phenolic analogues (o-nitrophenol, m-nitrophenol and p-nitrophenol) using two inorganic oxidizing agents (hydrogen peroxide and potassium peroxymonosulphate). Interestingly, significant enhancement in photo-Fenton activity was noticed for RE metal (La/Ce) substituted samples in comparison with pristine parent CoFe₂O₄ irrespective of the nature of the inorganic oxidant used. The maximum photo-Fenton activity was observed for compositions CoCe_0.1Fe_1.9O₄ and CoLa_0.4Fe_1.6O₄. The enhancement in activity of x = 0.1 and x = 0.4 for Ce and La doped CoFe₂O₄ respectively was correlated to the octahedral site preference of RE metal ions, reduced band gap and enhanced surface area. This work announces the prominent role of RE metal ions in photo-Fenton activity enhancement. In addition, the magnetic nature of CoFe₂O₄ means that the RE metal (Ce³⁺ and La³⁺) doped CoFe₂O₄ could be efficiently recovered from reaction mixtures.

---

1. Introduction

Magnetic crystalline cobalt ferrite (CoFe₂O₄) nanoparticles are a subject of immense interest and have grabbed the attention of researchers worldwide. CoFe₂O₄ is a promising and versatile magnetic material as it exhibits fascinating and exotic properties like moderate saturation magnetization, high coercivity, strong magneto-crystalline anisotropy, chemical stability, mechanical hardness, high electrical resistivity, large magnetostrictive coefficient and high Curie temperature (793 K).^1–3 The above mentioned features along with stability of CoFe₂O₄ with time make it an appropriate candidate for electronic devices, magnetic recording applications, magnetic resonance imaging (MRI), drug delivery, medical science, microwave devices, sensors, catalysis etc.^4–11 CoFe₂O₄ is emerging as a focus of research interest in the field of catalysis due to its inherent magnetic features, which enables easy recovery of the catalyst after the reaction.

A group of 15 elements of the lanthanide series along with chemically similar scandium (Sc) and yttrium (Y) collectively constitute rare earth (RE) metals.¹² The most convenient and uncomplicated route to use RE metals in different applications is in the form of oxides. RE metal oxides manifest different qualities which make them striking contender for catalytic applications.¹³ In the recent times different rare earth metal oxide like europium oxide (Eu₂O₃),¹³ cerium oxide (CeO₂),¹⁴ lanthanum oxide (La₂O₃),¹⁵ praseodymium oxide (Pr₆O₁₁),¹⁶ gadolinium oxide (Gd₂O₃),¹⁷ terbium oxide (Tb₄O₇),¹⁸ holmium oxide (Ho₂O₃),¹⁹ samarium oxide (Sm₂O₃),²⁰ neodymium oxide (Nd₂O₃)²¹ etc. have been used by researchers in diverse catalytic reactions. But the issue regarding efficient extraction of catalyst after the reaction needs to be resolved from the green chemistry prospect as it holds back it application on commercial scale. A relatively easy and rational solution to this problem is to make the catalyst magnetic in character.

A literature survey suggests that different properties of ferrites are controlled by spin coupling of 3d electrons arising from Fe–Fe interactions. When partial substitution of Fe³⁺ in spinel ferrites is done by RE metal ions belonging to 4f series, RE–Fe interactions come into picture. This leads to 3d–4f coupling which subsequently modifies properties of ferrite.²² Therefore, in order to utilize excellent properties of rare earth metals and cobalt ferrite simultaneously; small quantity of rare earth (RE) metal ion was introduced in the spinel structure of cobalt ferrite. Substitution of RE metal in CoFe₂O₄ leads to interesting results such as reduction of Curie temperature and coercivity,²³ distortion in the spinel structure of ferrites²⁴ etc. But our main concern is to enhance catalytic activity of CoFe₂O₄ with RE metal doping.

Amid different catalytic reactions, the current aim is focused on improving the Fenton activity of CoFe₂O₄ by doping with two RE metal ions viz. cerium (Ce³⁺) and lanthanum (La³⁺). By doing so, not only the catalyst maintains its magnetic character but also we can take advantage of superior traits of RE metals. Although researchers have tuned different properties of CoFe₂O₄ with RE metal doping,^24–29 but limited literature is available on tailoring Fenton activity of CoFe₂O₄ with RE metal substitution. Hence, two series of composition CoCe_xFe_2−xO₄ (x = 0.0, 0.1, 0.2 and 0.3) and CoLa_xFe_2−xO₄ (x = 0.0, 0.1, 0.2, 0.3, 0.4 and 0.5) were prepared by sol–gel methodology where RE was replaced with Fe³⁺ in CoFe₂O₄ in steps of 0.1 and characterized using various techniques. The effect of RE metal doping on photo-Fenton degradation of five different organic pollutants (dyes and phenolic derivatives) using two different inorganic oxidants i.e. hydrogen peroxide (H₂O₂) and potassium peroxymonosulphate (KHSO₅) was investigated. Besides this, stability and recyclability of magnetic catalyst was also examined for promoting its application in eradicating hazardous pollutants.

2. Experimental

2.1 Chemicals

All the reagents used in the preparation of ferrite and photocatalytic experiments were of analytical grade. Cobalt nitrate (Co(NO₃)₂·6H₂O, 99%), ferric nitrate (Fe(NO₃)₃·9H₂O, 98%), citric acid (99.57%) and hydrogen peroxide (30% w/v) were purchased from Fischer Scientific. Ethylene glycol (99%) supplied by Merck was used as received. Oxone (triple salt having composition 2KHSO₅·KHSO₄·K₂SO₄; source of potassium peroxymonosulphate), o-, m-, and p-nitrophenol were obtained from Avra synthesis. The two model dyes i.e. reactive black 5 (C₂₆H₂₁N₅Na₄O₁₉S₆) and remazol brilliant yellow (C₂₁H₁₈O₁₃S₄N₄NaK) were procured from textile industry. All the solutions were prepared in deionized water.

2.2 Synthesis of ferrite

Nanocrystalline RE metal (La³⁺ and Ce³⁺) doped cobalt ferrite possessing composition CoRE_xFe_2−xO₄ (RE = La; x = 0.1, 0.2, 0.3, 0.4, 0.5 and RE = Ce; x = 0.1, 0.2, 0.3) were synthesized by employing sol–gel methodology.^30,31 During the reaction respective metal nitrates were used as precursors, citric acid as fuel and ethylene glycol as gelating agent. In a typical reaction corresponding AR grade metal nitrates were dissolved as per their stoichiometric amounts in minimum amount of distilled water at 80 °C. Citric acid solution was prepared by separately weighing it in 1 : 1 molar ratio to metal salts followed by thorough mixing in pre-dissolved nitrate precursor. The mixture was subjected to heating and stirring on a hot plate at temperature of 80 °C. After few minutes ethylene glycol was added to induce gel formation. Subsequently water evaporated from beaker and resulted in viscous liquid (gel) formation. The gels were dried on hot plate and crushed to attain powder. Finally the obtained sample was calcined in muffle furnace at 400 °C for 2 hours to get desired ferrite.

2.3 Photocatalytic activity

The effect of RE-doping on photo-Fenton activity of CoFe₂O₄ was evaluated by degradation of five model organic pollutants i.e. remazol black 5 (RB5), remazol brilliant yellow (RBY), o-nitrophenol (ONP), m-nitrophenol (MNP) and p-nitrophenol (PNP). All the experiments were conducted under visible irradiation with 150 W Xe lamp as light source at room temperature. In a typical photo-Fenton experimental run, desired photocatalyst was added to appropriate concentration of dye/phenolic solution. The mixture was stirred in dark for 30 minutes until adsorption desorption equilibrium was reached. Thereafter oxidant was added and the solution was irradiated under visible light. During the degradation process aliquots were withdrawn from the reaction mixture at selected time interval, centrifuged to remove catalyst and analyzed using UV-vis spectrophotometer.

2.4 Physical characterization

The formation of M–O bond in ferrites was characterized using Fourier transform infrared analysis using Perkin Elmer Spectrophotometer in the range 400–800 cm⁻¹. The particle size of prepared nano-ferrites was obtained using FEI Tecnai (G2 F20) high resolution transmission electron microscope (HRTEM) operating at 200 keV. The elemental composition of all the materials was confirmed using Wavelength Dispersive X-ray Fluorescence (WD-XRF) spectrometer (Bruker S8-Tiger). The crystalline structure and phase identification of pure and Ce-doped CoFe₂O₄ was identified by powder X-ray diffraction technique (XRD, Panalytical's X'Pert Pro spectrophotometer) by employing CuK_α radiation (λ = 1.54 Å, 2θ = 10–80°). Magnetic measurements of prepared samples were made using vibrating sample magnetometer (VSM) (155, PAR) at room temperature. The visible light activity of all the ferrites was examined using UV-vis diffuse reflectance spectra (DRS) (JASCO, V-750) in the wavelength range of 400–900 cm⁻¹. The Brunauer–Emmett–Teller (BET) surface area of all the materials was estimated by physical adsorption of N₂ at −196 °C using BET surface area analyzer ((11-2370) Gemini, Micromeritics, USA). The concentration of pollutant at any time was measured using UV-vis spectrophotometer (JASCO, V-750).

3. Results and discussion

3.1 Characterization of RE metal doped CoFe₂O₄

3.1.1 Fourier transform infrared spectroscopy (FTIR). The bands obtained in FTIR analysis provide information regarding position acquired by ions in the crystal, reaction completion and distribution of cations in spinel structure.³² Literature reveals that due to vibration of oxygen ions with highest valency metal cations, in ferrites two characteristic absorption bands emerge below 600 cm⁻¹.³³ The appearance of two bands in all the samples confirmed formation of spinel structure of ferrites. A strong band at higher frequency (ν₁) was observed at ∼590 cm⁻¹ which was due to intrinsic stretching vibration of entire tetrahedral unit (M_tet–O). However, lower band (ν₂) comparably much weaker in intensity was observed at ∼410 cm⁻¹ corresponding to octahedral metal stretching (M_oct–O).³⁴ The infrared spectra of all the Ce an La doped CoFe₂O₄ (CoCe_xFe_2−xO₄; x = 0.1, 0.2, 0.3; CoLa_xFe_2−xO₄; x = 0.1, 0.2, 0.3, 0.4, 0.5) was recorded and is presented in Fig. S1 (ESI†). The existence of these two bands in spinel ferrites has already been reported by different researchers.^32–34

3.1.2 Structural, elemental and particle size analysis. In order to get further insight into the shape, size and composition of synthesized RE metal doped CoFe₂O₄ nanoparticles, HRTEM coupled with EDAX (energy dispersive X-ray spectroscopy) analysis was performed for all the samples. The typical TEM micrographs of CoCe_0.1Fe_1.9O₄ and CoLa_0.4Fe_1.6O₄ at low and high resolution are shown in Fig. 1(a) and (b), 2(a) and (b) respectively. The images display high agglomeration in sample due to magnetic feature of ferrites. Further the average particle size as observed from the images is ∼20 nm and ∼10 nm for Ce and La doped samples respectively.

Fig. 1 Low resolution TEM image (a), high resolution TEM image (b), lattice fringes (c), SAED pattern (d) and EDAX pattern (e) of CoCe_0.1Fe_1.9O₄.

Fig. 2 Low resolution TEM image (a), high resolution TEM image (b), lattice fringes (c), SAED pattern (d) and EDAX pattern (e) of CoLa_0.4Fe_1.6O₄.

Crystallinity of prepared RE metal doped samples is approved by HRTEM investigation. The high magnification images at 10 nm (Fig. 1(c)-Ce doped; Fig. 2(c)-La doped) represented spacing of ∼0.25 nm corresponding to most intense (311) plane of spinel structure. In addition to this, lattice spacing of 0.29 nm was also noticed which matched well with 220 (plane) in both the doped materials. The labeled lattice spacing between lattice fringes are in good agreement with the planes of spinel CoFe₂O₄. Further the selected area electron diffraction (SAED) pattern in Fig. 1(d)-Ce doped and Fig. 2(d)-La doped illustrated diffraction rings corresponding to (311), (400) and (440) planes of spinel structures of ferrite, thereby confirming crystalline nature of prepared sample.

In the end EDAX analysis was performed to determine chemical composition of the samples. The typical spectra of CoCe_0.1Fe_1.9O₄ and CoLa_0.4Fe_1.9O₄ presented in Fig. 1(e) and 2(e) respectively demonstrated that the sample consisted of all the elements thus confirming purity of synthesized material. In addition to EDAX analysis, the presence of all the elements in stoichiometric proportions in the prepared ferrites was confirmed using wavelength dispersive X-ray fluorescence spectrometer (WD-XRF). The theoretical (theo.) and observed (obs.) weight (%) percentage all the materials are presented in Table S1 (ESI†). The data in table clearly illustrates that observed weight (%) of all the elements are closely related to theoretical values. Hence this study provides vivid evidence for the presence of all the elements i.e. Co, Fe, O, Ce and La in all the materials.

3.1.3 Identifying the crystal structure and phase purity of RE metal doped CoFe₂O₄. Phase analysis and structural interpretation of RE-doped CoFe₂O₄ was performed by powder X-ray diffraction (XRD) analysis. Fig. 3 and 4 displays acquired XRD patterns of pristine and doped samples together with standard powder diffraction data of CoFe₂O₄ (JCPDS card no. 00-001-1121). The figures undoubtedly exhibit essential peaks at 2θ values of 31°, 36°, 43°, 57° and 62° corresponding to (hkl) planes of (220), (311), (400), (511) and (440) respectively. A close examination of XRD patterns revealed no additional reflection peaks corresponding to any impurity subsequently validating formation of cubic spinel structure with Fd3̄m space group.

Fig. 3 Powder X-ray diffractographs of pure and Ce-doped CoFe₂O₄.

Fig. 4 Powder X-ray diffractographs of La-doped CoFe₂O₄.

However the doping of RE metal into the parent lattice resulted in broadening of peaks which was attributed to reduction of crystallite size. This observation was confirmed by estimating the average crystallite size using standard Debye Scherer equation.³⁵ The calculations were performed by measuring peak widths of most intense (311) plane and the computed values are listed in Table 1. The values signify an obvious decrease in crystallite size with RE-doping with respect to CoFe₂O₄. Such decline with RE metal doping in ferrites has been earlier reported by different researchers.^36,37

Table 1 Different parameters of pristine and RE-doped CoFe₂O₄ (RE = Ce, La)

Ferrite	DXRD	Eg (eV)	Ms (emu g^−1)	Hc (Oe)
CoFe2O4	25.8	1.26	65	1250
CoCe0.1Fe1.9O4	19.6	1.15	29	1225
CoCe0.2Fe1.8O4	9.8	1.20	27	325
CoCe0.3Fe1.7O4	6.4	1.25	26	300
CoLa0.1Fe1.9O4	9.6	1.21	32	675
CoLa0.2Fe1.8O4	9.0	1.17	27	180
CoLa0.3Fe1.7O4	5.1	1.12	18	200
CoLa0.4Fe1.6O4	3.8	1.07	13	140
CoLa0.5Fe1.5O4	2.8	1.10	10	110

---

3.1.4 Optical properties of photo-catalyst. Optical property estimated using UV-vis diffuse reflectance spectra (DRS) is an effective tool for estimating band gap of both crystalline and amorphous semiconducting materials. The band structure of spinel ferrites suggests energy is required to excite electron from O-2p orbital (valence band) to Fe-3d orbital (conduction band). Therefore, the energy required for electronic transition is directly proportional to the band gap of spinel ferrites.³⁸ The UV-vis absorption plots of pure and RE-doped CoFe₂O₄ shown in Fig. 5 (Ce-doped) and Fig. 6 (La-doped) specify absorption edges of all the samples in visible region. In order to evaluate band gap of all the compositions following Tauc relation is used:³⁹

αhν = A(hν − E_g)ⁿ (1)

where hν, h, α, E_g, A are photon energy, Plank's constant, absorption coefficient, optical band gap and constant respectively. Here n is the index that characterizes optical absorption process and its value depends on nature of electronic transition in semiconductor. Its theoretical value is 2 for indirect allowed, 1/2 for direct allowed, 3 for indirect forbidden and direct forbidden electronic transition. The absorption coefficient α of ferrites is determined form DRS spectra using following principle relations:⁴⁰

I = I⁰ e^αt (2)

A = log(I₀/I) (3)

α = 2.303(A/t) (4)

where A and t are absorption and thickness of ferrite respectively.

Fig. 5 UV-visible absorption spectra of CoCe_xFe_2−xO₄ (x = 0.0, 0.1, 0.2, 0.3).

Fig. 6 UV-visible absorption spectra of CoLa_xFe_2−xO₄; x = 0.0, 0.1, 0.2, 0.3, 0.4, 0.5.

Spinel ferrites are direct band gap materials⁴¹ hence the value of n in present case in 1/2. A graph between (αhν)² and hν is generally referred to as Tauc plot from which the unique value of band gap is estimated by extrapolating straight line to (αhν)² = 0 axis. The Tauc plots of CoCe_xFe_2−xO₄; x = 0.0, 0.1, 0.2, 0.3 and CoLa_xFe_2−xO₄; x = 0.0, 0.1, 0.2, 0.3, 0.4, 0.5 composition are given in Fig. S2 and S3 (ESI†) respectively and the acquired band gap values are listed in Table 1. The values in table suggest significant decrease in band gap value up to x = 0.1 (Ce-doped) and x = 0.4 (La-doped) and thereafter the value increases. However, the results indicate that the entire values are quite narrow hence all the materials are suggested to have exceptional photo-response under visible light irradiation.

3.1.5 Magnetic properties. The magnetic properties of ferrites are highly influenced by dopant ion and resulting cation distribution.⁴² Magnetic feature of pure and RE-doped CoFe₂O₄ was determined by recording room temperature hysteresis (M–H) loops using vibrating sample magnetometer (VSM). The dependence of magnetization (M) on applied magnetic field (H) of pristine and doped CoFe₂O₄ nanoparticles are presented in Fig. 7 and 8 respectively. The deduced values of hysteresis elements i.e. M_s and H_c from M–H curves are listed in Table 1. The estimated M_s of CoFe₂O₄ at room temperature is 65 emu g⁻¹ which is much lower than its bulk counterpart (80 emu g⁻¹).⁴³ This decline in the M_s was correlated with reduction of particle size as compared to bulk CoFe₂O₄. In addition large surface to volume ratio of nanoparticles and surface spin canting also contribute to this effect.

Fig. 7 Hysteresis loops of pristine and Ce-doped CoFe₂O₄.

Fig. 8 Hysteresis loops of pristine and La-doped CoFe₂O₄.

It is interesting to notice from Table 1 that the value of M_s declines with RE metal doping in CoFe₂O₄. The net magnetization according to Neel's two sub-lattice model of ferrimagnetism is expressed as difference between magnetization at B and A site as presented below:⁴⁴

n_B = M_B(χ) − M_A(χ) (5)

where n_B is the net magnetic moment, M_B(χ) and M_A(χ) are the magnetic moment at B and A site respectively. The model suggests that three kinds of interactions exist between A and B sites viz. A–A, B–B and A–B. Amid these, A–B interaction is the strongest. Owing to the large ionic radius of Ce³⁺ (1.020 Å) and La³⁺ (1.17 Å), the metals tend to occupy octahedral site. When RE³⁺ substitutes Fe³⁺ at B site, it is equivalent to replacing ion with high magnetic moment (Fe³⁺-5 μ_B) with low magnetic moment ion (Ce³⁺-2.6 μ_B and La³⁺-0 μ_B) thus causing decrease in the value of M_s.⁴⁵ Additionally, the hardness of CoFe₂O₄ declined and the material progressed towards softness with RE³⁺ doping. This was validated by decrease in coercivity values as displayed in Table 1. In spite of the fact that M_s decreased with RE³⁺ substitution, the material remains sufficiently magnetic so that it can be easily recovered from reaction mixture.

3.1.6 Surface area calculation using BET. The activity of catalysts eminently relies on its surface area. This is attributed to the fact that greater surface area of material provides more active sites on the surface for the reaction to take place. In order to authenticate this, surface area of pure and RE-doped CoFe₂O₄ (CoCe_xFe_2−xO₄; x = 0.0, 0.1, 0.2, 0.3 and CoLa_xFe_2−xO₄; x = 0.0, 0.1, 0.2, 0.3, 0.4, 0.5) compositions was recorded using BET surface analyzer at 77 K. Initially sample preparation was accomplished with flow of N₂ gas at 150 °C for 2 hours. The as obtained BET plots of pristine and doped (Ce and La) CoFe₂O₄ given in Fig. S4 and S5† respectively undoubtedly depict linear nature of curves. Different parameters estimated from plots like slope, intercept, Q_m, C and specific surface area (S_total) of pristine and doped CoFe₂O₄ are tabulated in Table 2. The data of S_BET indicates sharp increase in surface area with RE doping. However, the difference in values is minimal in doped samples with maximum surface area of x = 0.1 (Ce-doped) and x = 0.4 (La-doped) compositions.

Table 2 Surface area parameters of all the pure and RE-doped CoFe₂O₄

Ferrite	Slope (g cm^−3)	Intercept (g cm^−3)	Qm (cm^3 g^−1)	C	Stotal
CoFe2O4	0.081386	0.000481	12.2149	170.076638	∼53
CoCe0.1Fe1.9O4	0.052251	0.000544	18.9412	97.043888	∼82
CoCe0.2Fe1.8O4	0.052875	0.000651	18.6825	82.263406	∼81
CoCe0.3Fe1.7O4	0.056007	0.000725	17.6267	78.280415	∼77
CoLa0.1Fe1.9O4	0.069364	0.000829	14.2466	84.719109	∼62
CoLa0.2Fe1.8O4	0.063732	0.000621	15.5395	103.696957	∼67
CoLa0.3Fe1.7O4	0.062705	0.000761	15.7565	83.379580	∼69
CoLa0.4Fe1.6O4	0.060391	0.000507	16.4264	120.156142	∼72
CoLa0.5Fe1.5O4	0.064443	0.000994	15.2820	65.852580	∼66

---

3.2 Catalytic studies

After the successful synthesis and characterization of RE-doped CoFe₂O₄ (CoCe_xFe_2−xO₄; x = 0.0, 0.1, 0.2, 0.3 and CoLa_xFe_2−xO₄; x = 0.0, 0.1, 0.2, 0.3, 0.4, 0.5), the next step was to evaluate the photo-Fenton activity. The main objective of current work was to improve the activity of CoFe₂O₄ by incorporating RE³⁺ metal ion in its structure. Therefore, a series of heterogeneous photo-Fenton reactions were conducted using five model pollutants; remazol black 5 (RB5), remazol brilliant yellow (RBY), o-nitrophenol (ONP) m-nitrophenol (MNP) and p-nitrophenol (PNP) using two inorganic oxidants i.e. hydrogen peroxide (HP) and potassium peroxymonosulphate (PMS). Different operational parameters (catalyst dosage, initial pH of solution and oxidant concentration) of photo-Fenton degradation process using all the target molecules, CoFe₂O₄ as catalyst and both the inorganic oxidants was optimized and the results are presented below:

3.2.1 Factors influencing photo-catalytic degradation.
(a) Catalyst loading. One of the important parameter that has to be optimized in photo degradation reactions is catalyst loading. Fig. S6(A) and (B) (ESI†) shows the dependence of dosage of CoFe₂O₄ on RB5, RBY, ONP, MNP and PNP degradation using HP and PMS as oxidants respectively. It was observed that increase in dosage up to 0.5 g L⁻¹ lead to increase in rate of reaction, but beyond this concentration the rate decreased. The initial increase in catalyst concentration lead to production of more active species but after 0.5 g L⁻¹ the solution became turbid which lead to decrease in penetration of light. This observation of decrease in rate of reaction after specific catalyst concentration has been reported by Cai et al.⁴⁶ Hence 0.5 g L⁻¹ was optimized catalyst dosage and reactions were performed with this concentration.
(b) Variation of pH. The pH of the medium is an important parameter which affects the photo degradation performance. The pH effect in the range 2–10 was estimated for the entire pollutant degradation using both the oxidants and the results are presented in Fig. S7 (ESI†). It was observed that rate of degradation increased with increasing acidity of solution and the maximum efficiency observed at pH 2.5 when HP was used as oxidant. This result was attributed to formation of heterogeneous photo-Fenton reagent at low pH. However, when PMS was used as oxidant interesting results were observed as it was active at wide pH range with maximum degradation rate at neutral pH. Hence, optimum pH for the photodegradation was 2.5 and neutral when HP and PMS were used as oxidants respectively.
(c) Oxidant dosage. The oxidant dosage is a critical feature in photo-Fenton degradation reaction as it plays important role in determining the treatment cost. The rate of reaction at different H₂O₂ and PMS concentrations in presence of CoFe₂O₄ and all the pollutants is given in Fig. S8 (ESI†). The graph clearly indicated increase in rate of degradation with increase in concentration from of HP from 2.2 to 8.8 mM. This result was attributed to formation of ˙OH in high concentration. However, by increasing the concentration beyond 8.8 mM slight decrease in reaction rate was observed which might be due to scavenging of ˙OH by H₂O₂ (eqn (6) and (7)).⁴⁷ Therefore, all the reactions were performed at 8.8 mM H₂O₂.

OH˙ + H₂O₂ → H₂O + HOO˙ (6)

HOO˙ + OH˙ → H₂O + O₂ (7)

Contrary to this when PMS was used as inorganic oxidant maximum degradation rate was observed at 2.2 mM which was low four times less as compared to HP. Beyond 2.2 mM slight decrease in rate was noticed hence the reaction were performed at 2.2 mM PMS concentration.

3.2.2 Effect of RE (Ce and La) metal substitution on photo-Fenton activity of CoFe₂O₄. A thorough and deep analysis of different operational parameters suggested that the optimized reaction conditions during the degradation of all the pollutants were: [catalyst]-0.5 g L⁻¹, pH-2.5, [HP] = 8.8 mM and [catalyst]-0.5 g L⁻¹, pH-natural, [PMS]-2.2 mM while using HP and PMS respectively as oxidants. In the present case the effect of RE³⁺ substitution on photo-Fenton activity of CoFe₂O₄ was studied under the above mentioned optimized reaction conditions.

In order to evaluate influence of RE metal substitution on photo-Fenton activity of CoFe₂O₄, the % degradation versus time graphs recorded for the deterioration of model pollutants in the presence of pristine and RE-doped CoFe₂O₄. The respective graphs of Ce and La doping using HP and PMS as oxidants are given in Fig. 9–12. All the reactions followed pseudo first order kinetics and the corresponding rate constant values using HP and PMS as oxidants are given in Tables 3 and 4 respectively. Clearly effect of RE-doping on photo-Fenton activity of CoFe₂O₄ was observed from the tables and figures. Although CoFe₂O₄ exhibited fair activity in the presence of both the oxidants but significant enhancement with Ce and La doping was noticed.

Fig. 9 % degradation versus time graphs of all the pollutants in the presence of pure and Ce-doped CoFe₂O₄ using HP as oxidant.

Fig. 10 % degradation versus time graphs of all the pollutants in the presence of pure and Ce-doped CoFe₂O₄ using PMS as oxidant.

Fig. 11 % degradation versus time graphs of all the pollutants in the presence of pure and La-doped CoFe₂O₄ using HP as oxidant.

Fig. 12 % degradation versus time graphs of all the pollutants in the presence of pure and La-doped CoFe₂O₄ using PMS as oxidant.

Table 3 Rate constant k (min⁻¹) of pure and RE-doped CoFe₂O₄ in the presence of all the probe molecules using HP as oxidant

Ferrite	RB5	RBY	PNP	MNP	ONP
CoFe2O4	0.099	0.030	0.057	0.048	0.036
CoCe0.1Fe1.9O4	0.141	0.092	0.090	0.060	0.072
CoCe0.2Fe1.8O4	0.105	0.077	0.072	0.055	0.044
CoCe0.3Fe1.7O4	0.102	0.075	0.071	0.052	0.040
CoLa0.1Fe1.9O4	0.128	0.056	0.104	0.084	0.117
CoLa0.2Fe1.8O4	0.162	0.096	0.141	0.128	0.165
CoLa0.3Fe1.7O4	0.196	0.098	0.155	0.132	0.175
CoLa0.4Fe1.6O4	0.249	0.134	0.194	0.142	0.186
CoCe0.5Fe1.5O4	0.209	0.104	0.178	0.102	0.094

---

Table 4 Rate constant k (min⁻¹) of pure and RE-doped CoFe₂O₄ in the presence of all the probe molecules using PMS as oxidant

Ferrite	RB5	RBY	PNP	MNP	ONP
CoFe2O4	0.211	0.198	0.076	0.028	0.032
CoCe0.1Fe1.9O4	0.584	0.447	0.104	0.050	0.067
CoCe0.2Fe1.8O4	0.306	0.266	0.091	0.040	0.054
CoCe0.3Fe1.7O4	0.250	0.220	0.085	0.036	0.047
CoLa0.1Fe1.9O4	0.370	0.213	0.053	0.047	0.064
CoLa0.2Fe1.8O4	0.506	0.379	0.055	0.045	0.077
CoLa0.3Fe1.7O4	0.608	0.488	0.059	0.051	0.087
CoLa0.4Fe1.6O4	0.733	0.506	0.079	0.07	0.096
CoCe0.5Fe1.5O4	0.597	0.489	0.067	0.064	0.084

---

3.2.3 Mechanism of photo-Fenton degradation process. The mechanism of photo-Fenton degradation process using spinel ferrite as heterogeneous catalyst has been elaborate in our previous report.³¹ The cations (M = Co, Fe) present in CoFe₂O₄ while using HP as oxidant undergo Fenton reaction according to following equations:

Mⁿ⁺ + H₂O₂ → M⁽ⁿ⁺¹⁾⁺ + OH˙ + OH⁻ (8)

M⁽ⁿ⁺¹⁾⁺ + H₂O₂ → M⁺ + HOO˙ + H⁺ (9)

Similarly when PMS is used as oxidizing agent the generalized equation can be represented as:

M⁽ⁿ⁺¹⁾⁺ + HSO₅⁻ → Mⁿ⁺ + SO₅˙⁻ + H⁺ (10)

Mⁿ⁺ + HSO₅⁻ → M⁽ⁿ⁺¹⁾⁺ + SO₄˙⁻ + OH⁻ (11)

The equations clearly illustrate that the main active species involved during the degradation process are HO˙ (HP as oxidant) and HO˙, SO₄˙⁻ (PMS as oxidant). Due to Fe(II,III) and Co(II,III) cycling, the stability of ferrite system is maintained during the degradation process and the active species are generated continuously.

(a) Reason for augmentation in activity with Ce-doping. The enhancement in activity with Ce-doping was attributed to photo-Fenton like behavior of Ce³⁺ analogous to Fe³⁺ according to following equations:⁴⁸

Ce³⁺ + H₂O₂ → Ce⁴⁺ + OH˙ + OH⁻ (12)

Ce⁴⁺ + H₂O₂ → Ce³⁺ + HOO˙ + H⁺ (13)

Hence the presence of additional Ce³⁺/Ce⁴⁺ redox pair resulted in improvement in activity of Ce-doped samples. Moreover, Ce³⁺ tend to occupy octahedral sites which are exposed on the surface thereby enhancing the photo-Fenton activity.

Nevertheless, the maximum activity of x = 0.1 composition was detected. This observation was related to band gap of ferrites. Table 1 indicates lowest band gap for x = 0.1 composition hence can absorb a large fraction of visible light. Furthermore, the highest surface area (Table 2) of x = 0.1 composition also contributed to this effect. An increase in amount of Ce beyond this concentration displayed a negative effect on photo-Fenton activity. Hence profitable results in photo-Fenton activity were obtained by introducing very small amount of Ce³⁺ (x = 0.1) in CoFe₂O₄.

(b) Reason for augmentation in activity with La-doping. A steep rise in activity was observed with La-doping up to x = 0.4, thereafter slight decrease was noticed. It is evident from literature that La³⁺ does not itself undergo any photo-Fenton like degradation. But doping of La³⁺ in montmorillonite clay significantly improved its activity.⁴⁹ In ferrites the improvement in activity may be related to its inherent octahedral occupancy owing to large ionic radius. Moreover, activity of x = 0.4 composition was best which was attributed to reduced surface area (Table 1) and enhanced surface area (Table 2) in comparison to pristine CoFe₂O₄.

3.2.4 Recyclability. Recyclability of material is an essential criterion that has to be taken into account for long term practical use in catalysis. In order to test this, the nanocatalyst was magnetically separated, washed with distilled water and dried in oven at 100 °C. Subsequently the catalyst was subjected to next run with fresh dye/phenolic solution under identical conditions and results. The typical results of pristine and Ce-doped CoFe₂O₄ using RBY as probe molecule are summarized in Fig. 13. The outcome suggested that catalyst did not undergo any substantial loss in activity even upto four consecutive cycles. Similar observations were noticed for La-doped materials. Hence the study clarifies that pristine and doped cobalt ferrite can act as heterogeneous Fenton catalyst for long term use at industrial scale.

Fig. 13 (A) Recyclability results of all the Ce-doped materials obtained during degradation of RBY. (B) Recyclability results of all the La-doped materials obtained during degradation of RBY.

4. Conclusion

To conclude, the doping of rare earth metals into spinel structure of ferrites has a positive effect on photo-Fenton degradation process. To illustrate this, Ce and La doped CoFe₂O₄ having composition CoCe_xFe_2−xO₄; x = 0.1, 0.2, 0.3 and CoLa_xFe_2−xO₄; x = 0.1, 0.2, 0.3, 0.4, 0.5 were prepared using sol–gel technique. Initially the influence of RE³⁺ doping on structural, magnetic and optical properties was examined in detail. FTIR analysis confirmed presence of two characteristic bands corresponding to tetrahedral and octahedral stretching frequency of spinel ferrite family in both the compositions. Powder XRD not only confirmed phase purity and crystallinity of all the samples (Ce and La doped CoFe₂O₄) but also approved formation of cubic spinel structure with Fd3̄m space group. In addition to this reduction in crystallite size was prominently observed with RE metal doping. The UV-DRS spectra clearly illustrated reduction in band gap with RE metal substitution while the M–H hysteresis loops of Ce/La doped samples demonstrated characteristic magnetic feature of prepared ferrites. In addition, decline in saturation magnetization was noticed with RE metal substitution. After successful synthesis and characterization of RE-doped CoFe₂O₄ samples, the nanospinels were exploited as photo-Fenton catalyst for degradation of textile dyes and phenolic analogues by employing hydrogen HP and PMS as oxidizing agents. Results suggested significant enhancement in photo-Fenton degradation process with RE-doping with best activity at x = 0.1 (Ce-doped) and x = 0.4 (La-doped), irrespective of inorganic oxidant used. The magnetic catalyst was easily recoverable using an external magnet hence the recyclability was also studied up to 4 successive cycles without any substantial loss in activity.

Acknowledgements

This work was financially supported by Council of Scientific and Industrial Research (CSIR). (01(02833)/15/EMR-II) and PURSE-II. The author (R. Sharma) also expresses deep gratitude to CSIR for providing the fellowship. The authors are also highly grateful to Kunash Instruments Pvt. Ltd. Thane (W), for doing the surface area analysis.

References

1. S. K. Gore, R. S. Mane, M. Naushad, S. S. Jadhav, M. K. Zate, Z. A. Alothman and B. K. N. Hui, Dalton Trans., 2015, 44, 6384–6390.
2. S. Mosivand and I. Kazeminezhad, RSC Adv., 2015, 5, 14796–14803.
3. C. Murugesan and G. Chandrasekaran, RSC Adv., 2015, 5, 73714–73725.
4. M. Houshiar, F. Zebhi, Z. J. Razi, A. Alidoust and Z. Askari, J. Magn. Magn. Mater., 2014, 371, 43–48.
5. H. Wu, G. Liu, X. Wang, J. Zhang, Y. Chen, J. Shi, H. Yang, H. Hu and S. Yang, Acta Biomater., 2011, 7, 3496–3504.
6. S. Amiri and H. Shokrollahi, Mater. Sci. Eng., C, 2013, 33, 1–8.
7. A. Hannour, D. Vincent, F. Kahlouche, A. Tchangoulian, S. Neveu and V. Dupuis, J. Magn. Magn. Mater., 2014, 353, 29–33.
8. E. R. Kumar, P. S. P. Reddy, G. S. Devi and S. Sathiyaraj, J. Magn. Magn. Mater., 2016, 398, 281–288.
9. A. M. Kulkarni, K. S. Pandit, P. V. Chavan, U. V. Desai and P. P. Wadgaonkar, RSC Adv., 2015, 5, 70586–70594.
10. R. Dumitru, F. Papa, I. Balint, D. C. Culita, C. Munteanu, N. Stanica, A. Ianculescu, L. Diamandescu and O. Carp, Appl. Catal., A, 2013, 467, 178–186.
11. Y. Hammiche-Bellal, A. Benadda, L. Meddour-Boukhobza, S. Barama, A. Djadoun and A. Barama, Catal. Commun., 2013, 42, 62–67.
12. Y. Zhua, Y. Zheng and A. Wang, Int. J. Biol. Macromol., 2015, 72, 410–420.
13. L. Feng, S. Yao, X. Zhao, L. Yan, C. Liu and W. Xing, J. Power Sources, 2012, 197, 38–43.
14. C. M. Lousada, M. Yang, K. Nilsson and M. Jonsson, J. Mol. Catal. A: Chem., 2013, 379, 178–184.
15. X. Chen, S. A. C. Carabineiro, S. S. T. Bastos, P. B. Tavares, J. J. M. Orfao, M. F. R. Pereira and J. L. Figueiredo, J. Environ. Chem. Eng., 2013, 1, 795–804.
16. R. Wang, G. Lu, W. Qiao, Z. Sun, H. Zhuang and J. Yu, Carbon, 2015, 95, 940–948.
17. M. Ledwaba, N. Masilela, T. Nyokong and E. Antunes, J. Mol. Catal. A: Chem., 2015, 403, 64–76.
18. Y. Zhang, J. Deng, H. Zhang, Y. Liu and H. Dai, Catal. Today, 2015, 245, 28–36.
19. G. A. M. Hussein, J. Anal. Appl. Pyrolysis, 1996, 37, 111–147.
20. J. Gao, Y. Zhao, W. Yang, J. Tian, F. Guan, Y. Ma, J. Hou, J. Kang and Y. Wang, Mater. Chem. Phys., 2002, 77, 65–69.
21. M. V. Boscoa, G. G. Fougaa and A. E. Bohéa, Thermochim. Acta, 2012, 540, 98–106.
22. M. A. Ahmed, S. T. Bishay, R. M. Khafagy and N. M. Saleh, J. Magn. Magn. Mater., 2014, 350, 73–80.
23. R. C. Kambale, K. M. Song, C. J. Won, K. D. Lee and N. Hur, J. Cryst. Growth, 2012, 340, 171–174.
24. M. A. Khana, M. J. Rehman, K. Mahmood, I. Ali, M. N. Akhtar, G. Murtazae, I. Shakirf and M. F. Warsi, Ceram. Int., 2015, 41, 2286–2293.
25. Z. Karimi, Y. Mohammadifar, H. Shokrollahi, S. K. Asl, G. Yousefi and L. Karimi, J. Magn. Magn. Mater., 2014, 361, 150–156.
26. X. Wu, H. Yun, H. Dongn and L. Geng, Ceram. Int., 2014, 40, 5905–5911.
27. V. S. Puli, S. Adireddy and C. V. Ramana, J. Alloys Compd., 2015, 644, 470–475.
28. K. S. Lohar, A. M. Pachpinde, M. M. Langade, R. H. Kadam and S. E. Shirsath, J. Alloys Compd., 2014, 604, 204–210.
29. A. K. Nikumbh, R. A. Pawar, D. V. Nighot, G. S. Gugale, M. D. Sangale, M. B. Khanvilkar and A. V. Nagawade, J. Magn. Magn. Mater., 2014, 355, 201–209.
30. M. Sajjia, M. Oubaha, M. Hasanuzzaman and A. G. Olabi, Ceram. Int., 2014, 40, 1147–1153.
31. R. Sharma, S. Bansal and S. Singhal, RSC Adv., 2015, 5, 6006–6018.
32. A. Hassan, M. A. Khan, M. Shahid, M. Asghar, I. Shakir, S. Naseem, S. Riaz and M. F. Warsi, J. Magn. Magn. Mater., 2015, 393, 56–61.
33. W. B. Cross, L. Affleck, M. V. Kuznetsov, I. P. Parkin and Q. A. Pankhurst, J. Mater. Chem., 1999, 9, 2545–2552.
34. G. D. Tang, Z. F. Shang, X. Y. Zhang, J. Xu, Z. Z. Lin, C. M. Zhen, W. H. Qi and L. L. Lang, Phys. B, 2015, 463, 26–29.
35. S. Singhal, R. Sharma, T. Namgyal, S. Jauhar, S. Bhukal and J. Kaur, Ceram. Int., 2012, 38, 2773–2778.
36. M. A. Rehman, I. Yusoff and Y. Alias, J. Hazard. Mater., 2015, 299, 316–324.
37. X. Zhao, W. Wang, Y. Zhang, S. Wu, F. Li and J. P. Liu, Chem. Eng. J., 2014, 250, 164–174.
38. S. Chaturvedi, R. Das, P. Poddar and S. Kulkarni, RSC Adv., 2015, 5, 23563–23568.
39. R. Dom, R. Subasri, N. Y. Hebalkar, A. S. Chary and P. H. Borse, RSC Adv., 2012, 2, 12782–12791.
40. K. N. Harish, H. S. Bhojya Naik, P. N. Prashanthkumar and R. Viswanath, Catal. Sci. Technol., 2012, 2, 1033–1039.
41. K. K. Bharathi, M. Noor-A-Alam, R. S. Vemuri and C. V. Ramana, RSC Adv., 2012, 2, 941–948.
42. Y. Xu, J. Sherwood, Y. Qin, R. A. Holler and Y. Bao, Nanoscale, 2015, 7, 12641–12649.
43. H. Kumar, R. C. Srivastava, J. Pal Singh, P. Negi, H. M. Agrawal, D. Das and K. H. Chae, J. Mater. Chem., 2016, 401, 16–21.
44. R. Sharma and S. Singhal, Phys. B, 2013, 414, 83–90.
45. G. Dixit, J. P. Singh, R. C. Srivastava and H. M. Agrawal, J. Magn. Magn. Mater., 2013, 345, 65–71.
46. C. Cai, Z. Zhang, J. Liu, N. Shan, H. Zhang and D. D. Dionysiou, Appl. Catal., B, 2016, 182, 456–468.
47. R. Sharma and S. Singhal, J. Chem. Technol. Biotechnol., 2015, 90, 955–962.
48. A. D. Bokare and W. Choi, J. Hazard. Mater., 2014, 275, 121–135.
49. Z. Huang, P. Wu, H. Li, W. Li, Y. Zhua and N. Zhu, RSC Adv., 2014, 4, 6500–6507.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra14325c

---