Synergetic E ﬀ ect of Calcium Doping on Catalytic Activity of Manganese Ferrite: DFT Study and Oxidation of Hydrocarbon

: Manganese ferrite (MnFe 2 O 4 ) and calcium-doped manganese ferrite (Ca-MnFe 2 O 4 ) were synthesized, characterized, and tested for oxidation of hydrocarbons (CH) in a self-designed gas blow rotating (GBR) reactor. The uniformly sized and thermally stable MnFe 2 O 4 nanoparticles (molar ratio, 1 / 284.5) showed a reasonable catalytic activity (productivity: 366.17 mmolg − 1 h − 1 ) with 60% selectivity at 80 ◦ C, which was further enhanced by calcium doping (productivity: 379.38 mmolg − 1 h − 1 ). The suspicious behavior of Ca-MnFe 2 O 4 was disclosed experimentally and theoretically as well. ,


Experimental
The chemicals and reagents that were used in this study were of high purity grade (supplier: Sigma-Aldrich & Merck, Karachi, Sindh, Pakistan) and were used without any further purification. Gases like oxygen and nitrogen were supplied by BOC, Pakistan Ltd. For the removal of traces from the gases, specific filters (PerkinElmer: safe glass moisture trap, part number N930-1193 and PerkinElmer: gas line filter, part number N930-1178) were used.

Synthesis of the Catalysts
Iron oxide, MnFe 2 O 4 , and Ca-MnFe 2 O 4 nanoparticles were prepared by the precipitation/ co-precipitation method with modifications, as reported in the literature [42]. Briefly, solutions of FeCl 3 ·6H 2 O (0.2 M), MnCl 2 ·4H 2 O (0.1 M), and Ca (NO 3 ) 2 ·4H 2 O (0.05 M) were prepared in deionized water. The individual solution and/or mixture of solutions was titrated against 3 M solution of NaOH (pH 9), which was preheated to 95 • C. The mixture was aged for 2 h at constant stirring and then allowed to cool down at room temperature. The precipitates formed were separated by centrifugation, washed several times until they reached a neutral pH, and dried for ten hours in an oven at 80 • C. The resulting nanoparticles were crushed to a fine powder and stored in a desiccator.

Characterization of Catalysts
The catalysts were characterized by Transmission Electron Microscopy (TEM) and Scanning Electron Microscopy (SEM, JSM 5910, Jeol, Japan). The elemental analysis of samples was investigated by Energy Dispersive X-ray Spectroscopy (EDX) (JSM 5910, JEOL, Japan). The phase of the samples was determined by an X-ray Diffractometer (D/Max-II, Rigaku, Japan) with the radiation source CuKα with λ = 0.15418 nm and an operating voltage of 40 kV, in the 2θ range of 10 • -70 • at a step size of 0.05 • . Thermogravimetric analysis of the samples was investigated using a Thermal Gravimetric Analyzer (TGA, PerkinElmer, USA). The Fourier Transform Infrared Spectrometer (FT-IR) (Prestige 21, Shimadzu, Japan) was used to record the FT-IR spectra of the sample pellets (1.3 mm) with KBr (spectrum range 500-4000 cm −1 ).

DFT Study of the Catalyst
In this work density functional theory calculations were performed using the Full Potential Linearized Augmented Plane Wave method (FP-LAPW) implemented in Wien2k code [43]. To deal with exchange and correlation, energy-generalized gradient approximations [44] are used for electronic properties. In the FP-LAPW method, the wave function for potential and charge density expansion is composed of the combination of radial functions multiplied by spherical harmonics inside the atomic radii and plan waves in the interstitial region. The cut-off value for the planned wave RMT × K max = 8 is selected, while the charge density is Fourier-expanded up to G max = 14 (Ryd). For the integration of the Brillouinzone, we used 7 × 7 × 7 k-point. The self-consistent convergence is achieved up to 0.5 mRy/Bohr.

Oxidation Reaction
A total of 0.1849 mol of cyclooctane ( a CH) or cyclohexane ( b CH) or cyclohexene ( c CH) and the catalyst (6.5 × 10 −4 mol) were loaded into a self-designed gas blow rotating (GBR) reactor. The temperature of the reactor was maintained through heating tapes and a temperature controller (Digi-Sense, ColeParmer, USA). Furthermore, the reaction parameters were set up as desired from the temperature range: 60-80 • C, time: 25-240 min, catalyst: 0.01-0.25 g and oxygen partial pressure: 50-760 Torr. Iodometric titration with sodium thiosulfate was used for the determination of CHHP in the reaction mixture [45]. The CHHP was completely converted into CH by adding an excessive amount of triphenylphosphine (PPh 3 ). The amount of acid that was generated in the reaction was titrometrically (NaOH) determined [46]. The exact content (conc.) of alcohol (COH) was determined by deducting the content (conc.) of CHHP, as determined by iodometry from the amount of COH determined by gas chromatography (GC) analysis (Clarus 580, PerkinElmer, USA) equipped with flame ionization detector (FID), while using a capillary column (cross-linked methyl siloxane capillary column; length: 30 m, ID: 0.32 mm, and film thickness: 0.25 µm), while n-nonane was used as the internal standard. The % conversion was calculated from initial and final concentrations of CH, although the % selectivity to the product was determined as a ratio of COH/RCOR to the total conversion of the product, from standard calibration curves, as given in the Supporting Information (SI), CH ( Figure S1), COH ( Figure S2), and RCOR ( Figure S3).

Physical Characterization of the Catalysts
The characterization of Mn-ferrite was performed using various analytical techniques, as given in Figure 1. Figure 1a shows that the Mn-ferrite has uniform, ultra-small spherical structure as distinct entities with some aggregates. Similarly, Figure 1b, reveals that some of the nanostructures were interconnected with each other, or present in the form of bulbous clusters. The average grain intercept (AGI) method was used to determine the particle size of the nanoparticles (30-75 nm) which is also reflected from the TEM image. Figure 1c, presents the percent weight of each element as Mn (22%), The third observable region was in the range of 300−600 °C (weight loss of 2%) because of bulk oxygen and residual carbon contents. Beyond 600 °C, no more weight loss was observed [23,27,47]. Figure 2 shows the physical characterization of Ca-Mn-ferrite. The SEM image revealed that the particles are in aggregate form (Figure 2a), while the EDX spectra confirmed the presence of Ca doping (Figure 2b). Similarly, the XRD patterns show few extra peaks that are responsible for Ca at 2θ = 29.3°, 36.4° (Figure 2c). The surface areas for Mn-ferrite and Ca-Mn-ferrite were 48.9 and 41.5 m 2 /g, respectively. The decrease in surface area is most probably due to the blockage of pores and also the increase in particle size, as presented in Figure 2d. The adsorption/desorption isotherm of both catalysts shows that the pores have narrow bottleneck-like shape, which reflects the mesoporous texture [48].   [27]. The variation in the size calculated from XRD and SEM is due to the aggregation of the Mn-ferrite nanoparticles. The FT-IR spectrum of MnFe 2 O 4 is given in Figure 1e. Different bands for functional groups were observed; the band at 3441cm −1 is due to hydroxyl groups (OH) group residue, while the band at 651cm −1 is responsible for Fe-O vibration. Similarly, bands at 1459 and 1121cm −1 are assigned to the Mn-O-Fe bond stretching, while the peaks at 1721 and 2451cm −1 may be due to the presence of sp 2 and sp-hybridized carbon contents. Sahoo & coworkers [23] prepared Mn-ferrite for the catalytic degradation of dye pollutant. They predicted the FT-IR spectrum for Mn-ferrite, which is in close agreement with our data. Bellusci & co-workers [46] synthesized albumin-ferrite superparamagnetic nanoparticles by the reverse micelle method to attain high inorganic contents in polymeric materials. They also reported the FT-IR spectrum for Mn-ferrite nanoparticles, which was comparable to our investigation. The OH group may facilitate and enhance the surface activity of Mn-ferrite [46]. The TGA thermogram of Mn-ferrite that is presented in Figure 1f shows three different regions for weight loss. First weight loss (5%) in the range of 30-100 • C was due to surface water, while the second weight loss (3%) was observed in the range of 110-250 • C due to surface oxygen.
The third observable region was in the range of 300-600 • C (weight loss of 2%) because of bulk oxygen and residual carbon contents. Beyond 600 • C, no more weight loss was observed [23,27,47].  (Figure 2c). The surface areas for Mn-ferrite and Ca-Mn-ferrite were 48.9 and 41.5 m 2 /g, respectively. The decrease in surface area is most probably due to the blockage of pores and also the increase in particle size, as presented in Figure 2d. The adsorption/desorption isotherm of both catalysts shows that the pores have narrow bottleneck-like shape, which reflects the mesoporous texture [48].

Screening of the Catalyst
Reaction parameters were optimized for the efficient catalytic activity of magnetic nanoparticles for the green and selective oxidation of CH to COH/RCOR, as shown in Scheme 1.

Screening of the Catalyst
Reaction parameters were optimized for the efficient catalytic activity of magnetic nanoparticles for the green and selective oxidation of CH to COH/RCOR, as shown in Scheme 1.

Screening of the Catalyst
Reaction parameters were optimized for the efficient catalytic activity of magnetic nanoparticles for the green and selective oxidation of CH to COH/RCOR, as shown in Scheme 1. The initial substrate volume has an inverse relation with % production of COH/RCOR. The oxidation of CH conversion to COH/RCOR decreases with an increase in the initial volume of CH (mL), as shown in Figure 3. The observed trend may be attributed to the availability of CH in large amounts to the surface of Mn-ferrite. However, the reverse trend was observed in % selectivity, which linearly increases with the increase in the initial substrate volume. Most probably, more oxidation of CH occurred at a high concentration, instead of the deep oxidation of COH/RCOR because of the unavailability of active sites on the surface of the catalyst. The initial substrate volume has an inverse relation with % production of COH/RCOR. The oxidation of CH conversion to COH/RCOR decreases with an increase in the initial volume of CH (mL), as shown in Figure 3. The observed trend may be attributed to the availability of CH in large amounts to the surface of Mn-ferrite. However, the reverse trend was observed in % selectivity, which linearly increases with the increase in the initial substrate volume. Most probably, more oxidation of CH occurred at a high concentration, instead of the deep oxidation of COH/RCOR because of the unavailability of active sites on the surface of the catalyst.  Fang & co-workers [49] performed electrocatalytic oxidation of CH to CHOH/RCOR using a functional electrocatalytic membrane reactor (ECMR) assembled with an MnO nanostructure that was loaded on a porous titanium membrane as an anode in a ternary system of CH, water and acetic acid. They observed the same trend in selectivity and conversion per unit time. Table 1   Fang & co-workers [49] performed electrocatalytic oxidation of CH to CHOH/RCOR using a functional electrocatalytic membrane reactor (ECMR) assembled with an MnO nanostructure that was loaded on a porous titanium membrane as an anode in a ternary system of CH, water and acetic acid. They observed the same trend in selectivity and conversion per unit time. Table 1 reflects the higher catalytic activity of Mn-ferrite and Ca-doped Mn-ferrite used for the oxidation of CH in solvent-free conditions compared to all reported catalysts being used for the oxidation of COH/RCOR [29][30][31]50].
Scheme 2 (i-v) shows the oxidation of CH to COH/RCOR with molecular oxygen and Mn-ferrite. The reaction was initiated by the Fe of the Mn-ferrite in the presence of molecular oxygen, while Mn shows a synergistic effect. This observation was confirmed by carrying out experiments with simple iron oxide as a catalyst under the same set of reaction conditions; we obtained the low productivity value of 101.52 mmolg −1 h −1 , which is too low in comparison to the productivity value for Mn-ferrite (366.17 mmolg −1 h −1 ); thus, it is confirmed that Mn has a synergistic effect on the activity of Mn-ferrite, as shown in Table 2.  Similarly, the addition of calcium to Mn-ferrite further promotes the catalytic activity of MnFe2O4, as presented in Table 2, while reducing the ease of the magnetic filtration of the catalyst from the reaction mixture. Hou et al. [51] reported Ca-promoted Ni/α-Al2O3 for the dry reforming of CH4 and observed that the enhancing effect of Ca on the catalytic activity depends on its interaction with the support material.   Similarly, the addition of calcium to Mn-ferrite further promotes the catalytic activity of MnFe 2 O 4 , as presented in Table 2, while reducing the ease of the magnetic filtration of the catalyst from the reaction mixture. Hou et al. [51] reported Ca-promoted Ni/α-Al 2 O 3 for the dry reforming of CH 4 and observed that the enhancing effect of Ca on the catalytic activity depends on its interaction with the support material.
Theoretical Ca-MnFe 2 O 4 is presented in Figure 4b, which reveals that the Fe atom has a paramount impact on DOS structure for MnFe 2 O 4 ; similarly, Fe and Mn atoms have a great impact, while Ca contribution is too low in the DOS structure for Ca-MnFe 2 O 4 . The total DOS structure of MnFe 2 O 4 and Ca-MnFe 2 O 4 is given in Figure 5, which shows the similar band structure in spin-down states, while the band gaps of MnFe 2 O 4 (1.6 eV) and Ca-MnFe 2 O 4 (2.3 eV) are mainly determined by spin-up states. The total magnetic moment of the unit cell and total magnetic moment of the formula unit both decrease with the addition of Ca to MnFe 2 O 4 (110.29 µB to 93.63 µB) and (13.78 µB to 11.70 µB), respectively [11], as shown in Table S1, which shows close agreement with the experimental results. As represented in Scheme 2 (i−v), the Cat (II) was activated in the presence of molecular oxygen to produce Cat (III)-OO • , where (II) and (III) show the oxidation states. Hydrogen was abstracted from CH by Cat (III)-OO • and produced Cat (III)-OOH and cyclohexyl radical (CHy • ), which further gave cyclohexylhydroperoxide (CHHP) and CHy • in the presence of molecular oxygen and CH. The cyclohexyl radical (CHy • ) was recycled in the process. The cyclohexylperoxy radical was produced from the oxidation of (CHHP) by Cat (III)-OOH; a proton (H + ) was eliminated to produce Cat (II)-OOH.
Crystals 2020, 10, x FOR PEER REVIEW 10 of 17 up state of MnFe2O4 up to (0.7 eV), as shown in Figure 4a, while Ca doping up to 20% decreased the band gap energy by 0.22 eV. Ustundağ et al. [11] have also reported that the band gap energy of MnFe2O4 is reduced by the incorporation of calcium. The Density of States (DOS) structure of MnFe2O4 and Ca-MnFe2O4 is presented in Figure 4b, which reveals that the Fe atom has a paramount impact on DOS structure for MnFe2O4; similarly, Fe and Mn atoms have a great impact, while Ca contribution is too low in the DOS structure for Ca-MnFe2O4. The total DOS structure of MnFe2O4 and Ca-MnFe2O4 is given in Figure 5, which shows the similar band structure in spin-down states, while the band gaps of MnFe2O4 (1.6 eV) and Ca-MnFe2O4 (2.3 eV) are mainly determined by spin-up states. The total magnetic moment of the unit cell and total magnetic moment of the formula unit both decrease with the addition of Ca to MnFe2O4 (110.29 µB to 93.63 µB) and (13.78 µB to 11.70 µB), respectively [11], as shown in Table S1, which shows close agreement with the experimental results. As represented in Scheme 2 (i−v), the Cat (II) was activated in the presence of molecular oxygen to produce Cat (III)-OO˚, where (II) and (III) show the oxidation states. Hydrogen was abstracted from CH by Cat (III)-OO˚ and produced Cat (III)-OOH and cyclohexyl radical (CHy • ), which further gave cyclohexylhydroperoxide (CHHP) and CHy • in the presence of molecular oxygen and CH. The cyclohexyl radical (CHy • ) was recycled in the process. The cyclohexylperoxy radical was produced from the oxidation of (CHHP) by Cat (III)-OOH; a proton (H + ) was eliminated to produce Cat (II)-OOH. However, when the cyclohexylperoxy radical was recycled, a small amount of product was obtained, which may be due to the dissolved oxygen in the reactant or the bulk oxygen of oxide catalyst. Liu & co-workers [50] performed the oxidation of CH to KA oil, using C3N4/Au composites, and attained low conversion (10.54%) with high selectivity (100%) under visible light, without using any initiator. However, in the electrocatalytic oxidation of the CH with a Ti membrane decorated However, when the cyclohexylperoxy radical was recycled, a small amount of product was obtained, which may be due to the dissolved oxygen in the reactant or the bulk oxygen of oxide catalyst. Liu & co-workers [50] performed the oxidation of CH to KA oil, using C 3 N 4 /Au composites, and attained low conversion (10.54%) with high selectivity (100%) under visible light, without using any initiator. However, in the electrocatalytic oxidation of the CH with a Ti membrane decorated with nano-Mn oxide, 14.6% conversion, and high selectivity of 99.8% was obtained [49]. Similarly, Ti-containing zeolite catalysts were also efficiently used for oxidation of CH with 90% selectivity [31]. In the reported studies [31,49,50], the achieved % conversion was very poor and the method of preparation of the catalyst was also expensive; therefore, Mn-ferrite and Ca-Mn-ferrite are active, affordable, and efficient catalysts for the oxidation of CH and might be a good alternative for industrial use.
with nano-Mn oxide, 14.6% conversion, and high selectivity of 99.8% was obtained [49]. Similarly, Ticontaining zeolite catalysts were also efficiently used for oxidation of CH with 90% selectivity [31]. In the reported studies [31,49,50], the achieved % conversion was very poor and the method of preparation of the catalyst was also expensive; therefore, Mn-ferrite and Ca-Mn-ferrite are active, affordable, and efficient catalysts for the oxidation of CH and might be a good alternative for industrial use.  Figure 6 shows the kinetics of a model reaction where (Figure 6a) represents the effect of oxygen partial pressure on the conversion of CH. An increase in oxygen partial pressure is accompanied by an increase in the rate of conversion up to pO2 ≈ 570Torr. A further increase in pO2 did not show any enhancing effect but, instead, a small reduction in the rate of reaction was observed. To obtain the mechanism of surface reaction (between oxygen and CH), various reaction rate equations were carried out [53] based upon the Langmuir-Hinshelwood (LH) mechanism, the Eley-Rideal (ER) mechanism and the Mars-van Kreveleen mechanism.

Effect of Oxygen Partial Pressure
The data did not fit into equations based upon the ER and MK Mechanisms. The Langmuir-Hinshelwood mechanism is based upon the reaction of the reactant in an adsorbed state. However, as it is a solvent-free reaction, the CH concentration remains constant and the rate only depends on the amount of oxidant. Various adsorption isotherm equations for the representation of were used, including the Langmuir adsorption isotherm for non-dissociative and dissociative adsorption of oxygen, te Temkin isotherm and the Freundlich isotherm. However, none of these equations resulted in a better fit than the Langmuir equation for competitive adsorption.
The equation for competitive adsorption between oxygen (O2) and CH on a surface can be represented as: Surface covered by CH, = Therefore, rate of reaction = kr θCHθO2 = ( ) . ( For studying the effect of oxygen partial pressure at a constant CH concentration, Equation 3 is modified to be suitable for using the nonlinear least-square program (Curve Expert 1.4) in the following form (Equation 4):  Figure 6 shows the kinetics of a model reaction where (Figure 6a) represents the effect of oxygen partial pressure on the conversion of CH. An increase in oxygen partial pressure is accompanied by an increase in the rate of conversion up to pO 2 ≈ 570 Torr. A further increase in pO 2 did not show any enhancing effect but, instead, a small reduction in the rate of reaction was observed. To obtain the mechanism of surface reaction (between oxygen and CH), various reaction rate equations were carried out [53] based upon the Langmuir-Hinshelwood (LH) mechanism, the Eley-Rideal (ER) mechanism and the Mars-van Kreveleen mechanism.

Effect of Oxygen Partial Pressure
Crystals 2020, 10, x FOR PEER REVIEW 13 of 17 Figure 6. Kinetics of CH oxidation (a); effect of oxygen partial pressure on the rate of CH conversion using a MnFe2O4 catalyst, (b); comparison of experimental and calculated rates from non-linear leastsquare fit for oxygen partial pressure, (c); effect of change in CH volume on the rate of reaction, (d); correlation between experimental and calculated rates for CH volume.
Only rates that were obtained for VCH ≤ 50 show a similar plot, as shown in Figure 6d, with the constants obtained having similar values, as obtained in Figure 6b. The change in the rate of reaction with a change in the volume of CH could be due to the effect of volume change on the rate of transportation of oxygen to the catalyst surface. With the increase in volume, there is a reduction in the rate of diffusion of oxygen to the catalyst surface. This reduction in the rate of diffusion of oxygen The data did not fit into equations based upon the ER and MK Mechanisms. The Langmuir-Hinshelwood mechanism is based upon the reaction of the reactant in an adsorbed state. However, as it is a solvent-free reaction, the CH concentration remains constant and the rate only depends on the amount of oxidant. Various adsorption isotherm equations for the representation of θ O 2 were used, including the Langmuir adsorption isotherm for non-dissociative and dissociative adsorption of oxygen, te Temkin isotherm and the Freundlich isotherm. However, none of these equations resulted in a better fit than the Langmuir equation for competitive adsorption.
The equation for competitive adsorption between oxygen (O 2 ) and CH on a surface can be represented as: Surface covered by CH, θ CH = For studying the effect of oxygen partial pressure at a constant CH concentration, Equation (3) is modified to be suitable for using the nonlinear least-square program (Curve Expert 1.4) in the following form (Equation (4)): where a = k r , b = K CH C CH and c = K O2 . Using Equation (4), the best fit was obtained, as shown in Figure 6b, while the comparison of experimental data points and the calculated data points using the constants a, b, and c was obtained. The values of the constants were used for the rate calculation at various oxygen partial pressures. The correlation coefficient R 2 was 0.982, which is reasonably better than all other rate equations applied for oxygen partial pressure effects in this case.

Effect of Initial Substrate Volume
It is a solvent-free reaction and, therefore, the concentration of CH would remain constant and, so, normally, surface coverage for CH should remain constant, whatever the volume of CH may be. However, Figure 6c shows that, practically, the rate of reaction changes with a change in the volume of CH. Rate increases up to a maximum at 50 mL of CH, after which it rapidly decreases. In a competitive adsorption between two reactants at the surface, an increase in the concentration of one reactant increases the surface coverage and simultaneously decreases the surface coverage of another reactant; as a result, the rate of reaction increases and the maximum rate is achieved at a point where the surface coverage of the reactants becomes equal. A further increase in the concentration of either reactant would decrease the rate of reaction, as shown in Figure 6c, as the decline in the rate of reaction is much faster when the volume of CH is increased above 50 mL. This could be due to the fact that the volume of CH affects the surface concentration of O 2 (θ O 2 ). As a competitive adsorption, any reduction in the surface coverage of O 2 would increase the surface coverage of CH and, therefore, a change in the rate of reaction occurs. Thus, the equation for competitive adsorption was applied to the data where the concentration of CH (C CH ) was replaced by V CH , as shown in Equation (5).
This equation (Equation (5)) was applied for calculating the rate of reaction as a function of the volume of CH. All of the constants obtained from Figure 6b were used as such. However, the constant b (= K CH C CH ) was used as "K CH V CH ". Figure 6b has a constant volume of CH (20 mL); therefore, K CH was obtained by as b/V CH (=1.55 E −3 /mL). The experimental and calculated rates as a function of the volume of CH are shown in Figure 6d. There is an excellent correlation between experimental and calculated rates for CH volume ≤50 mL. For volume ≥60 mL, there is a very rapid reduction in the rate of reaction. Attempts to apply the equation of competitive adsorption to the whole data (10-90 mL of CH) using the non-linear least-square fit was unsuccessful.
Only rates that were obtained for V CH ≤ 50 show a similar plot, as shown in Figure 6d, with the constants obtained having similar values, as obtained in Figure 6b. The change in the rate of reaction with a change in the volume of CH could be due to the effect of volume change on the rate of transportation of oxygen to the catalyst surface. With the increase in volume, there is a reduction in the rate of diffusion of oxygen to the catalyst surface. This reduction in the rate of diffusion of oxygen to the catalyst surface is still faster than the rate of adsorption/reaction/desorption up to V CH = 50 mL. However, at V CH ≥ 60 mL, the rate of diffusion of oxygen to the catalyst surface becomes slower than the surface reactions, and the reaction becomes diffusion controlled. One reason could be that the kinetic equation is applicable only in the kinetically controlled region (≤50 mL). Activation energy (82.54 kJ/mol) was calculated from the Arrhenius plot shown in Figure 7. Bavykin et al. [54] have reported activation energy = 79 kJ/mol in a purely kinetic regime. An activation energy = 77.8 kJ/mol was also considered in a free-mass transfer zone [53]. Thus, the activation energy in the present case shows that, in these conditions, the reaction is free of mass transfer and is in good agreement with kinetic studies.

Separation and Recyclability of MnFe2O4
To confirm the heterogeneity of the catalysts, a series of experiments was carried out. In the first run, 20 mL reactant was loaded under the optimal set of parameters. The product was determined by iodometry and GC analysis. The catalyst was recycled five times with a stable catalytic activity in the oxidation of CH, as shown in Figure 8a. The magnetic nanoparticles were successfully separated from the reaction medium by using an external magnet, as shown in Figure 8b. The extended life span is an important feature for industrial catalysis. To confirm the heterogeneity of the catalysts, a series of experiments was carried out. In the first run, 20 mL reactant was loaded under the optimal set of parameters. The product was determined by iodometry and GC analysis. The catalyst was recycled five times with a stable catalytic activity in the oxidation of CH, as shown in Figure 8a. The magnetic nanoparticles were successfully separated from the reaction medium by using an external magnet, as shown in Figure 8b. The extended life span is an important feature for industrial catalysis.
To confirm the heterogeneity of the catalysts, a series of experiments was carried out. In the first run, 20 mL reactant was loaded under the optimal set of parameters. The product was determined by iodometry and GC analysis. The catalyst was recycled five times with a stable catalytic activity in the oxidation of CH, as shown in Figure 8a. The magnetic nanoparticles were successfully separated from the reaction medium by using an external magnet, as shown in Figure 8b. The extended life span is an important feature for industrial catalysis.

Conclusions
The catalysts (Mn-ferrite and Ca-doped Mn-ferrite) were efficiently used for the synthesis of COH/RCOR from the oxidation of CH in the presence of molecular oxygen under solvent-free conditions in a self-designed reactor (GBR reactor). The catalyst shows good productivity values of 366.17 and 379.38 mmolg −1 h −1 for Mn-ferrite and Ca-Mn-ferrite, respectively, at optimal conditions. The synergetic effect of Ca doping on the catalytic activity of MnFe2O4 was theoretically explored and corelated with experimental observations. The only discrepancy of Ca doping was reducing the magnetic properties of MnFe2O4. Furthermore, the reaction shows no mass transfer effect and it follows the Langmuir-Hinshelwood mechanism for the competitive adsorption of CH and oxygen. These catalysts are truly heterogenous with extended life spans and will be a green alternative to existing industrial processes.

Conclusions
The catalysts (Mn-ferrite and Ca-doped Mn-ferrite) were efficiently used for the synthesis of COH/RCOR from the oxidation of CH in the presence of molecular oxygen under solvent-free conditions in a self-designed reactor (GBR reactor). The catalyst shows good productivity values of 366.17 and 379.38 mmolg −1 h −1 for Mn-ferrite and Ca-Mn-ferrite, respectively, at optimal conditions. The synergetic effect of Ca doping on the catalytic activity of MnFe 2 O 4 was theoretically explored and corelated with experimental observations. The only discrepancy of Ca doping was reducing the magnetic properties of MnFe 2 O 4 . Furthermore, the reaction shows no mass transfer effect and it follows the Langmuir-Hinshelwood mechanism for the competitive adsorption of CH and oxygen. These catalysts are truly heterogenous with extended life spans and will be a green alternative to existing industrial processes.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4352/10/4/335/s1, Figure S1: Calibration curve of CH in a mixture solution of CHOH and RCOR, Figure S2: Calibration curve of COH in a mixture solution of CH and RCOR, Figure S3: Calibration curve of RCOR in a mixture solution of CH and COH, Figure S4: EDX spectrum of MnFe 2 O 4 , Table S1: Magnetic moment for Mn-ferrite and Ca-Mn-ferrite.