Photocatalytic degradation of methylene blue with hematite nanoparticles synthesized by thermal decomposition of fluoroquinolones oxalato–iron(III) complexes

Abstract

[Fe(C₂O₄)(FQ)(H₂O)₂] complexes (H-FQ = ciprofloxacin (1), lomefloxacin (2) and norfloxacin (3)) were synthesized and characterized using a variety of analytical and spectral techniques such as elemental analysis, infrared spectroscopy, thermogravimetric analysis, ultraviolet-visible spectroscopy, magnetic and conductance measurements. The experimental studies were complemented by quantum chemical calculations in terms of geometry optimization, natural bond orbital analysis and molecular electrostatic potential maps. Electronic structures were discussed by TD-DFT. Hematite (α-Fe₂O₃) nanoparticles, as promising materials for different catalytic applications, were prepared in air via the controlled thermal decomposition of 1–3. Powder X-ray diffraction was used to identify the polymorph of iron oxide. The morphology of nano-hematite was investigated by a field emission scanning electron microscope coupled to energy-dispersive X-ray spectroscopy for surface analysis. The catalytic degradation of methylene-blue (MB), as an industrial pollutant, exposed to UV radiation in the presence of nano-α-Fe₂O₃ as a catalyst and hydrogen peroxide as the oxidant was studied at room temperature in water.

---

Introduction

So far, the oxalato ligand has appeared as a fertile key for design complexes of intriguing structural research applications.¹ Normally, the oxalato can act as a chelating or bridging ligand.^2,3 The [M^III(C₂O₄)₃]³⁻ complexes of Al and transition metal ions have been extensively studied.^4–16 In 1939,⁵ Blair and Jones synthesized potassium tris(oxalato)ferrate(III) trihydrate (KOF) from FeC₂O₄. Recently, a modified synthesis method has been introduced by Hussain et al.⁶ through the reaction of a mixture of K₂C₂O₄·H₂O and H₂C₂O₄·2H₂O with Fe(OH)₃. Crystal structures of anhydrous⁶ and hydrated forms⁷ of KOF showed the existence of infinite anionic distorted octahedral [Fe(C₂O₄)₃]³⁻ units and the three K⁺ ions are coordinated by water or oxalato ligands. The importance of KOF is ascribed to being a chemical actinometer. This complex undergoes a color change when it is heated⁸ or if the aqueous solution is exposed to UV radiation⁹ suggesting the reduction of Fe³⁺ and formation of an ion-radical intermediate, C₂O₄⁻˙.^10,11 The photo-chemical significance of KOF has been extended to the field of polymer chemistry e.g. the photo-polymerization of acrylamide¹² and methyl methacrylate¹³ in the aqueous acid solution of KOF has been achieved. Another application of KOF has been gained from the preparation of nano-Fe₂O₃ via the solid state thermolysis process.¹⁴ Different polymorphs nanoparticles were obtained at the different temperatures.^8,15,16

Fluoroquinolones (FQ's) represent a group of synthetic antibiotics distinguished by a good bioavailability, safety profile, long half-life (with once-daily dosing) and a broad antibacterial spectrum against Gram(−) bacteria.^17,18 Ciprofloxacin (H-Cf) (Scheme 1) is a second-generation quinolone drug used for the treatment of urinary zone infections, bone and joint infections, infectious diarrhea, typhoid fever and acute sinusitis.¹⁹ Norfloxacin (H-Nf) is the first FQ drug introduced to the market. Its antibacterial activity involves multi-antibiotic resistant, Gram(−) rods and beta-lactamase producing organisms. Lomefloxacin (H-Lf) is a third-generation quinolone used in cure of respiratory and urinary tracts, gynecological, ophthalmological and soft tissue infections. It has the advantage of being effective against some anaerobic bacteria.²⁰

Scheme 1 (a) Structures of the investigated fluoroquinolones (FQ's) and (b) synthesis of [Fe(C₂O₄)(FQ)(H₂O)₂] complexes.

The formation of FQ's complexes with transition and non-transition metal ions was the question of a number of literature reports.²¹ Several studies^22–31 concerning the interactions between H-Cf, H-Lf and H-Nf and Fe³⁺ ions were reported. A simple rapid spectral method was developed for determination of H-Cf in its pharmaceutical formulations using Fe(NO₃)₃.²² Formation constants of Fe³⁺ complexes of H-Cf and H-Nf were determined.²³ The water-soluble Fe(Nf)₂⁺ (λ_max = 377 nm) formed in the acidic medium was used for determination of Fe³⁺.²⁴ The complexation of H-Lf with Fe³⁺ ions in terms of stability constants and stoichiometries was studied by Riley et al.²⁵ Complexes of the type [Fe(Nf)₂(H₂O)₂]Cl₃·6H₂O,²⁶ [Fe(Nf)₃]Cl₃·12H₂O,²⁷ [Fe(Cf)(nitrilotriacetate)]·3.5H₂O,²⁸ [Fe(Cf)₃]²⁹ and [Fe(Lf)(H₂O)₄]Cl₃·H₂O³⁰ were proposed, where FQ's interact with Fe³⁺ via the carboxylate and pyridone oxygen atoms. Hydrothermal reaction of H-Cf with Fe(OH)₃ in presence of oxalate yielded [Fe(H-Cf)(C₂O₄)₂]·H₂Cf·5H₂O.³¹

In the present study, synthesis, spectroscopic and thermal characterization of three mononuclear iron(III)–oxalato complexes containing fluoroquinolones (ciprofloxacin, lomefloxacin and norfloxacin) as a primary ligand (Scheme 1) are reported both experimentally and theoretically. Due to the wide variety of applications for iron oxide nanoparticle polymorphs in the fields of catalysis,³² sensors,³³ clinical diagnosis,³⁴ and magnetic storage,³⁵ a simple and nontoxic method for preparation of α-Fe₂O₃ nanoparticles based upon the thermal decomposition of [Fe(C₂O₄)(FQ)(H₂O)₂] complexes has been explored. The morphology has been characterized by different physical methods such as XRD, and FE-SEM. The surface analysis has been carried out by EDX. Finally, the degradation of MB by Fenton-like catalytic performance of α-Fe₂O₃ nanoparticles has been investigated with the aid of the ultraviolet radiation.

Results and discussion

Spectral and magnetic characterization

The reaction of KOF with one equivalent of FQ's (Scheme 1) affords complexes of the type [Fe(C₂O₄)(FQ)(H₂O)₂] (H-FQ = H-Cf (1), H-Lf (2) and H-Nf (3)). These complexes were characterized by elemental analysis, TGA, IR, UV-Vis, magnetic and molar conductance measurements. The active IR bands at 1707, 1624 cm⁻¹ (H-Cf·HCl), 1723, 1620 cm⁻¹ (H-Lf·HCl) and 1708, 1620 cm⁻¹ (H-Nf) (Fig. S1†) are assigned to ν(COOH)^carb and ν(C=O)^py (carb: carboxyl, py: pyridone). In complexes, the vanishing of ν(COOH)^carb mode might be a sign for the ionization of –COOH and consequently its involvement in chelation.³⁶ In general, the carboxylate ion gives rise to two bands: a strong asymmetric stretching band near 1650–1550 cm⁻¹ and a weaker symmetrical stretching one near 1400 cm⁻¹.³⁷ For example, complex 2 shows the ν^ass(C=O) and ν^ss(C=O) modes at 1554 and 1360 cm⁻¹³⁸ with Δν ≈ 200 cm⁻¹ characteristic of unidentate COO⁻ group,³⁹ while complex 3 displays only the asymmetric vibration at 1548 cm⁻¹. These modes are overlapped in 1. The pyridone C=O group is liberated from the intra-molecular H-bond interaction via the complex formation and interacts with metal ion giving a band at the higher wave numbers; 1628 (1), 1630 (2) and 1627 (3).⁴⁰ In contrast, the piperazine NH is not involved in chelation, since its stretching mode is located at higher wave numbers,⁴¹ 3436–3442 cm⁻¹ (1–3) compared with 3429–3436 cm⁻¹ observed in the free ligands. The weak/medium intensity bands at 1714, 1668 and 1476 cm⁻¹ (1), 1714, 1670 and 1459 cm⁻¹ (2), 1713, 1665 and 1477 cm⁻¹ (3) (Fig. S2†) allocated for ν(C=O)_ox, ν^ass(C=O)_ox and ν^ss(C=O)_ox (ox: oxalato) are taken as evidence for the bidentate nature of the oxalato ligand in the mononuclear complexes.^3,42

The electronic absorption spectra (Fig. S3†) of 1–3 in DMSO exhibit the intra-ligand transitions at 335, 320 and 285 nm (1), 340, 325 and 292 nm (2), 335, 320 and 285 nm (3).³⁸ An additional weak band at about 625 nm assigned to ⁶A_1g → ⁵T_1g transition in a distorted octahedral stereochemistry is observed in the studied complexes.²⁹ The effective magnetic moment values at 298 K, calculated by the spin-only terms, are 4.86, 6.19 and 6.14μ_B for complexes 1–3 that are in agreement with those expected for high spin arrangement Fe(III) complexes (S = 5/2; μ_eff = 5.0–6.1μ_B).⁴³

DFT/TD-DFT

To obtain an insight into the geometrical and electronic structures of the investigated complexes, cis-[Fe(C₂O₄)(FQ)(H₂O)₂] were optimized (Fig. 1) by DFT/B3LYP method combined with 6-31G(d). It was found that the cis-isomer is more stable than the trans-isomer by 18.543 kJ mol⁻¹. The complexes were characterized as local minima through harmonic frequency analysis. The Fe³⁺ ion displays a distorted FeO₆ octahedral geometry. Four oxygen atoms coming from the chelating oxalato [FeO₂ = 1.840 Å and FeO₇ ≈ 1.882 Å] and bidentate FQ drug [FeO₃ ≈ 1.893 Å and FeO₄ ≈ 1.877 Å] are coordinated to the metal ion. The octahedral geometry is completed by two water molecules [FeO₅ = 2.033 Å and FeO₆ = 2.026 Å] lying in a cis-position. The Fe³⁺–oxalato bonds are not equal, since the oxalato oxygen atoms are participated in two H-bonds of different strengths [O₂⋯H₅₄–O₆ (2.403 Å, <95.5°) and O₇⋯H₅₁–O₅ (2.094 Å, <107.3°)]. As shown in Fig. 1, another two H-bonds affecting the regularity of the geometry [O₄⋯H₅₃–O₆ (1.989 Å, <110.0°) and O₆⋯H₅₂–O₅ (2.543 Å, <95.5°)] are also observed. The strongest H-bond in the complexes is O₄⋯H₅₃–O₆, since the angle of this interaction is 110.0°.³⁷ As shown in Table 1, the optimized bond lengths and angles around the iron(III) ion are unaffected by the change in the type of substituent attached to the studied FQ's drugs.

Fig. 1 Local minimum structures of complexes (a) 1, (b) 2 and (c) 3 obtained at DFT/B3LYP/6-31G(d) level of theory.

Table 1 Selected bond lengths (Å) and angles (°) for complexes 1–3 calculated by DFT/B3LYP/6-31G(d) method

Bond lengths (Å)				Angles (°)
	1	2	3		1	2	3
FeO2	1.840	1.840	1.840	O2FeO3	93.6	93.4	93.6
FeO3	1.893	1.895	1.893	O2FeO4	97.1	97.1	97.1
FeO4	1.877	1.876	1.877	O2FeO5	165.9	166.1	165.9
FeO5	2.033	2.033	2.033	O2FeO6	87.5	87.7	87.5
FeO6	2.026	2.024	2.026	O2FeO7	85.8	85.8	85.8
FeO7	1.882	1.881	1.882	O3FeO4	96.1	96.1	96.1
				O3FeO6	175.7	175.8	175.7
				O4FeO7	174.5	174.5	174.5
				O4FeO6	79.6	79.7	79.6
				O6FeO7	95.9	95.8	95.9

---

The nature of the electronic transitions observed in the UV-Vis spectra of the complexes has been studied by time-dependent DFT.^44–46 The lowest 30 singlet-to-singlet spin-allowed excitation states were calculated using the same functional and basis set for the optimization. The effect of solvent (DMSO) was performed using the default polarizable continuum model. The calculated d–d excitation wave lengths, energies of other excitation transitions (f > 0.002) and their assignments are taken into consideration. The stimulated spectrum of complex 1 (Fig. S4†) shows three transitions at 445, 404, and 374 nm assigned to H(β) → L/L+1(β), H(β) → L(β) and H-4/H-5(β) → L(β) (H: HOMO; L: LUMO). The transitions at 445 nm have a ground state composed of Fe d character, whereas the excitation states are of Fe-d_z² (LUMO) and Fe-d_x²−y²/(FQ and oxalate)-π*orbitals (LUMO+1) forming d → d_z² and d → d_x²−y²/LMCT transitions (Fig. 2) characterized to the octahedral Fe³⁺ complexes. The band at 374 nm has a LMCT character, originating from the piperazine moiety (HOMO-4) and (FQ and oxalate)-π orbitals (HOMO-5) and going to Fe d_z² orbital. The TD-DFT spectrum of 2 is characterized by five transitions at 465, 420, 375, 360 and 325 nm with oscillator strengths of 0.0003, 0.0153, 0.0848, 0.0437 and 0.0101 corresponding to H(β) → L/L+1(β), H(β) → L(β), H(β) → L+1(β), H-4/H-2(β) → L(β) and H-1(β) → L+1(β). The descriptions of frontier MO's and the relocation of the electron density of complexes are nearly the same as those described in complex 1. This is in agreement with the experimental findings, since the three complexes show similar electronic spectrum (Fig. S4†). The calculated spectrum of 3 exhibits the same transitions as observed in 1. Comparison between the calculated and experimental spectra indicated the suitability of the applied calculation method for this size of compounds, where the experimental bands are deviated from the calculated ones by about 10 nm.

Fig. 2 Theoretical electronic absorption transitions of complex 1 in DMSO.

Natural bond orbital analysis and molecular electrostatic potential

Natural bond orbital (NBO) analysis47,48 and second order perturbation theory analysis of Fock matrix provide details about the type of hybridization, nature of bonding and strength of the interactions between Fe3+ ion and donor sites.49 The electronic arrangement of Fe in complexes 1–3 is [Ar]4s0.283d6.074p0.435s0.015p0.01 with 6.780 valence electrons. The natural charge is 1.200 and the occupancies of Fe 3d are: dxy1.900 dxz1.608 dyz1.352 dx2−y20.563 dz20.644 (1), dxy1.925 dxz1.539 dyz1.338 dx2−y20.558 dz20.707 (2) and dxy1.919 dxz1.609 dyz1.356 dx2−y20.539 dz20.644 (3). The 3d-electronic population is corresponding to the oxidation state Fe(I), not Fe(III) that is in agreement with ligand to dFe electron transfer. The strength of interactions between Fe3+ ion and donor sites can be estimated by the second order perturbation theory. The larger the E2 value, the more intensive is the interaction between the electron donors and electron acceptors. For complex 1, the E2 values are 1.07, 0.72, 1.27, 1.33, 0.89 and 1.32 kcal mol−1 for LP(3)O2 → RY*(3)Fe, LP(2)O3 → RY*(4)Fe, LP(3)O4 → RY*(2)Fe, LP(2)O5 → RY*(3)Fe, LP(2)O6 → RY*(4)Fe and LP(3)O7 → RY*(2)Fe respectively. Similar values are observed for the other investigated complexes as tabulated in Table S1.†

Molecular electrostatic potential (MEP) map is a very useful descriptor in understanding the sites for electrophilic and nucleophilic reactions as well as H-bonding interactions.^44,46 The MEP plots (Fig. 3) of 1–3 are characterized by a positive region (blue) coming from H-atoms of FQ's and the coordinated water molecules. The negative charge region is mainly located on the oxalato O atoms and the minor contribution is coming from the carboxylate group. A zero potential area (green) is covered the remaining parts of the investigated complexes.

Fig. 3 Molecular electrostatic potential map for complexes (a) 1, (b) 2 and (c) 3. The electron density isosurface is 0.004 a.u. Different values of the electrostatic potential are represented by different colors: red represents the regions of the most electro negative potential, blue represents regions of most positive electrostatic potential and green represents regions of zero potential. Potential increases in the following order: red < orange < yellow < green < blue.

Thermogravimetric analysis

The thermogram of 1 exhibits four decomposition steps at 90, 233, 325 and 477 °C. In the 1^st step, only one coordinated water molecule is eliminated (mass loss: found, 3.73%; calcd, 3.53%). The 2^nd stage is accompanied by a mass loss amounts to 20.74% (up to 310 °C), which finds a parallelism with the calculated value (20.78%) responsible for loss of another water molecule and 2CO₂. The 3^rd and 4^th steps bring the total mass loss up to 90.57%. Such mass loss is close to the calculated value (89.02%) expected for the formation of metallic iron as a residue. Similar, four degradation stages are observed in the TG curve of 2 at 143, 214, 331 and 405 °C. The 1^st step reveals loss of 1.5H₂O molecules (mass loss, found, 4.73%; calcd, 5.09%). The 2^nd step is assigned to removal of 0.5H₂O + 2CO₂ with a mass loss amounts to 18.25% (calcd 18.30%). The 3^rd and 4^th stages are attributed to degradation of one Lf molecule giving a metallic iron as a final residue (found, 10.33%; calcd 10.56%). The thermogram of 3 displays five decomposition steps at 102, 235, 315, 426 and 454 °C. The 1^st stage (up to 200 °C) is responsible for loss of 2H₂O (mass loss, found, 7.18%; calcd 7.22%). The 2^nd thermal event is accompanied by a mass loss amounts to 21.54%. This value goes parallel to the calculated one (21.48%) that is attributed to the elimination of 2CO₂ + 1/2F₂ molecules. The 3^rd and 4^th stages are corresponding to the pyrolysis of the rest of the organic part with overall mass loss amounts to 88.83% (calcd 88.75%) leaving metallic iron as a residue.

α-Fe₂O₃ nanoparticles were prepared by heating complexes 1–3 in an electric furnace (800 °C) under the ambient air condition before cooling to room temperature. Their XRD patterns (Fig. S5†) indicated that samples 1 and 2 are composed of single rhombohedral phase characteristic of α-Fe₂O₃ nanoparticles. All the diffraction peaks are in agreement with the reported JCPDS card no. 04-011-9585. Sample 3 is mainly α-Fe₂O₃ phase contaminated with α-FeOOH (goethite) (JCPDS card no. 00-034-1266). Fig. 4 shows FE-SEM images of the investigated α-Fe₂O₃ nanoparticles. Other FE-SEM images with a diameter of 2 μm are given in Fig. S6.† As expected, the pristine nanoparticles of iron oxides bend to aggregate into large clusters because of anisotropic dipolar attraction in nanosized form of Fe₂O₃. In other words, the starting of sintering led to many grown agglomerates. The influence of the type of the coordinated FQ's has been reflected in size of the nanoparticles. The diameter of the nanoparticles is found to be 50–120 nm (1), 28–54 nm (2) and 47–82 nm (3). Therefore, the thermal decomposition of lomefloxacin complex 3 gave rises to smallest iron oxide nanoparticles. EDX analysis of these particles indicates the presence of Fe and O composition in the pure as-grown iron oxide as no other peak related with any impurity has been detected (Fig. 5). The sizes of the current nanoparticles are comparable to the results of several previous reports prepared iron oxide nanoparticles from the thermal decomposition conditions (13–180 nm).^8,50

Fig. 4 FE-SEM images of α-Fe₂O₃ nanoparticles obtained from thermal decomposition of (a) 1, (b) 2 and (c) 3.

Fig. 5 EDX analysis of α-Fe₂O₃ nanoparticles obtained from thermal decomposition of (a) 1, (b) 2 and (c) 3.

Photo-degradation process

Methylene-blue (MB) is one of the generally industrial used dyes in the fields of printing, textiles, etc.⁵¹ Its industrial effluents are the major source of harmful to the living organisms through the decrease of the dissolved oxygen capacity of water and by blocking sunlight. Severe exposure to MB led to permanent burn to the eyes of organisms, nausea and vomiting.⁵² Therefore, the treatment of effluents containing MB is one of the challenging problems in the field of environmental chemistry. Moreover, the choice of MB in this work, as a test pollutant, to explore the catalytic performance of the synthesized nanoparticles is attributed to the high solubility in water with no possibility to volatility and gas-phase pyrolysis. Its blue solution is characterized by unique band at 665 nm, which allows easy spectrophotometrically monitoring of the degradation process.^53,54 Among the different techniques used to eliminate MB, the modified Fenton reaction with various types of the synthesized catalysts⁵⁵ was found to be effective through the creation of highly reactive free radicals HO˙ and safe by the formation of harmless chemicals such as CO₂ and H₂O. The percent of degradation depends upon the formation rate of these radicals. Heterogeneous Fenton catalyst has an advantage of working over a wider pH-range compared with the traditional one, which is limited to pH < 4. However, its disadvantage is the slower degradation rate of organic pollutants compared to the classical method. Recently, the coupling of UV irradiation with Fenton-like reagent has attracted the consideration of some working groups interested in the improvement of the degradation rate of organic pollutants such as phenol⁵⁶ and methylene-blue.⁵⁷

To evaluate the catalytic performance of the synthesized α-Fe₂O₃ nanoparticles, the absorption spectra of the illuminated samples at various time intervals were monitored. The degradation process of MB was initiated by addition of H₂O₂ solution to the MB solution and immediately turning on the ultraviolet irradiation. For comparison, the MB degradation experiment was also done without α-Fe₂O₃ nanoparticles, but in presence of H₂O₂ and/or UV light (Fig. S7†). Fig. 6 shows the time-dependent absorption electronic spectra of MB as a function of the illumination time at 365 nm in presence of H₂O₂/α-Fe₂O₃ nanoparticles (sample no. 2). Similar spectra reported for sample no. 1 are giving in Fig. S8.† It can see that intensity of the characteristic band of MB at 658 nm remains nearly unchanged in the simple systems, MB + H₂O₂ and MB + UV for about 200 min. In other words, control experiments indicated that the degradation of MB is hardly proceeded in absence of catalyst (Fig. 7). The degradation efficiency [R = 100 × ((A_t − A₀)/A₀)] of MB was determined, where A₀ and A_t are the initial absorbance and that obtained after the illumination time (t). Combined of H₂O₂ with the UV irradiation increases the percent of discolorization of MB up to about 59%. The addition of α-Fe₂O₃ nanoparticles to the MB + H₂O₂ + UV system degraded about 94% (1), 97% (2) and 98% (3) of MB exposed to the UV light within about 100 min. The obtained results are comparable to the previously published data used α-Fe₂O₃ nanostructured fibers as a catalyst.⁵⁸ The photo-catalytic degradation of MB was found to be first order reaction [ln(C_t/C₀) = kt]. Its ln(C_t/C₀) plot shows a linear relationship with the illumination time (Fig. 8), where slope is the rate constant (k). The calculated rate constant for the photo-catalytic reactions was 1.26 × 10⁻² (1), 1.89 × 10⁻² (2) and 2.44 × 10⁻² min⁻¹ (3), while it is only about 2.00 × 10⁻⁴ (MB + UV), 1.00 × 10⁻⁴ (MB + H₂O₂) and 26 × 10⁻⁴ min⁻¹ (MB + UV + H₂O₂) for blank samples. The t_1/2 value was found to be 55 (1), 37 (2) and 28 min (3) compared with 2.5 days (MB + UV), 5 days (MB + H₂O₂) and 4.5 hours (MB + UV + H₂O₂). It is worthy to mention that almost no degradation of MB was observed in dark, which reveals that the mechanism of the reaction is photo-degradation rather than adsorption. As shown in Fig. 7, the decrease in the particle size of the synthesized hematite nanoparticles as those coming from the thermal decomposition of complex 2, compared with compound 1, led to an improvement in the catalytic performance. The reactivity of iron oxide particles has been shown to greatly increase as their dimensions are reduced.⁵⁹ Decreasing the average particle size increases the specific surface area and thus increases the number of active surface sites. However, the presence of traces of iron oxyhydroxide in sample 3 may be the motivation for the high catalytic performance. The hydroxyl radical that comes from FeOOH may be the primary oxidant that initiate the degradation of the pollutant.

Fig. 6 Time-dependent UV-Vis absorption spectra for the photocatalytic degradation of MB with α-Fe₂O₃ (2) [10 mL dye (14 mg L⁻¹), 0.1 mL H₂O₂ (35%), 10 mg α-Fe₂O₃, λ = 365 nm].

Fig. 7 Ultraviolet light photo-degradation of MB under different conditions.

Fig. 8 Linear plots of ln(C_t/C₀) vs. time for the decolorization of MB under different conditions.

Conclusion

Reaction of ciprofloxacin, lomefloxacin and norfloxacin with K₃[Fe(C₂O₄)₃] afforded complexes of the type cis-[Fe(C₂O₄)(FQ)(H₂O)₂], which were characterized by several analytical and spectral tools. The experimental studies were complemented by quantum chemical calculations. Cis-isomer is found to be more stable than the trans-complex. Electronic spectra were assigned by the aid of TD-DFT calculations. Thermal decomposition of the synthesized complexes is a simple and nontoxic method for preparation of hematite (α-Fe₂O₃) nanoparticles. The purity of the hematite nanoparticles was verified by EDX analysis. The prepared hematite nanoparticles exhibited good photo-catalytic activity in the degradation process of an industrial pollutant methylene-blue by means of Fenton-like reagent.

Experimental

Instruments

TGA was performed in nitrogen atmosphere (20 mL min⁻¹) in a platinum crucible with a heating rate of 10 °C min⁻¹ using a Shimadzu DTG-60H simultaneous DTG/TG apparatus. FT IR spectra were recorded as KBr pellets using a Jasco FTIR 460. Elemental microanalysis was performed using Elementer Vario EL III. Magnetic measurement was carried out on a Sherwood scientific magnetic balance using Gouy method,³⁹ and Hg[Co(SCN)₄] was used as a calibrant. Electronic spectra were scanned on a Shimdazu Lambda 4B spectrophotometer. A digital Jenway 4310 conductivity with a cell constant of 1.02 was used for the molar conductance study. The iron oxide nanoparticle morphologies were investigated by field emission scanning electron microscope (Quanta FEG 250) attached with energy dispersive X-ray analyses for surface analysis. Hematite nanoparticles were characterized by X-ray powder diffraction (PANalytical, X'Pert PRO) operated with copper target (K_α, λ = 1.54056 Å) at 40 kV and 25 mA.

Synthesis

One mmol of ciprofloxacin·HCl (366 mg), lomefloxacin·HCl (386 mg) and norfloxacin (319 mg) was added to one mmol of hot aqueous solution of K₃[Fe(C₂O₄)₃]·3H₂O (25 mL, 491 mg) (supplied from Sigma chemical company) and heated to reflux (1–3 h), where yellow (1), orange (2) and red (3) complexes were precipitated, respectively. The pH of the reaction mixture was raised by addition of ammonia solution. Lower molar conductance values (in DMF) were reported for the studied complexes compared with the reported values⁴⁵ for 1 : 1 (65–90 Ω⁻¹ cm² mol⁻¹) and 1 : 2 (130–170 Ω⁻¹ cm² mol⁻¹) electrolytes in DMF, indicating their non-ionic nature. α-Fe₂O₃ nanoparticles were prepared by heating complexes 1–3 in an electric furnace under ambient air condition before cooling to room temperature.

• Complex 1: color: yellow. Elemental analysis (%): calcd C₁₉H₂₁FFeN₃O₉: C 44.73, H 4.15, N 8.24, found C 44.68, H 4.31, N 9.04. FT IR (cm⁻¹): 3436 ν(NH), 1714 ν(C=O)_ox, 1668 ν^ass(C=O)_ox, 1628 ν(C=O)_py, 1476 ν^ss(C=O)_ox, 1386, 1271, 1181, 1106, 1034. UV-Vis (DMF, 10⁻⁴, nm): 285, 320, 335, 625. Molar cond. (10⁻³ M, DMF, Ω⁻¹ cm² mol⁻¹): 18.14. μ_eff (μ_B, 298 K): 4.86.

• Complex 2: color: orange. Elemental analysis (%): calcd C₁₉H₂₂F₂FeN₃O₉: C 43.04, H 4.18, N 7.92, found C 43.45, H 4.22, N 8.39. FT IR (cm⁻¹): 3431 ν(NH), 1714 ν(C=O)_ox, 1670 ν^ass(C=O)_ox, 1630 ν(C=O)_py, 1554, ν^ass(C=O), 1523, 1459 ν^ss(C=O)_ox, 1402, 1360 ν^ss(C=O), 1323, 1273, 1164, 1123, 1093, 1051. UV-Vis (DMF, 10⁻⁴, nm): 292, 325, 340, 625. Molar cond. (10⁻³ M, DMF, Ω⁻¹ cm² mol⁻¹): 9.73. μ_eff (μ_B, 298 K): 6.14.

• Complex 3: color: red. Elemental analysis (%): calcd C₁₈H₂₁FFeN₃O₉: C 43.39, H 4.25, N 8.43, found C 43.92, H 4.19, N 8.64. FT IR (cm⁻¹): 3442 ν(NH), 1713 ν(C=O)_ox, 1665 ν^ass(C=O)_ox, 1627 ν(C=O)_py, 1548, ν^ass(C=O), 1519, 1477, 1386, 1272, 1188, 1139, 1098, 1036. UV-Vis (DMF, 10⁻⁴, nm): 285, 320, 335, 625. Molar cond. (10⁻³ M, DMF, Ω⁻¹ cm² mol⁻¹): 10.44. μ_eff (μ_B, 298 K): 6.15.

Quantum chemical calculations

Ground state geometry optimization, natural bond orbital analysis and molecular electrostatic potential maps of complexes 1–3 were obtained at DFT/B3LYP/6-31G* level of theory using Gaussian03.⁶⁰ The complexes were characterized as local minima through harmonic frequency analysis. Electronic transitions were calculated by TD-DFT.^44–46 The effect of solvent (DMSO) was performed using the default polarizable continuum model.⁶¹

Catalytic activity

The photo-catalytic degradation of MB was investigated by taking 10 mL of the dye solution (14 mg L⁻¹, pH = 5) with 0.1 mL hydrogen peroxide (35% v/v) in presence of 10 mg of nano-α-Fe₂O₃. The suspension was stirred to ensure that all the active sites of catalysts are in contact with the dye solution. Then the sample was illuminated with a UV hand lamp (365 nm) positioned perpendicular at a distance of 2 cm. The irradiation was interrupted in regular intervals to take electronic spectra on the UV-visible spectrophotometer until no more changes in the characteristic band of MB was observed.

References

1. K. S. Min and M. P. Suh, J. Solid State Chem., 2000, 152, 183; O. Castillo, I. Muga, A. Luque, J. M. Gutiérrez-Zorrilla, J. Sertucha and P. Román, Polyhedron, 1999, 18, 1235; Y. Akhriff, J. Server-Carrió, A. Sancho, J. García-Lozano, E. Escrivá, J. V. Folgado and L. Soto, Inorg. Chem., 1999, 38, 1174.
2. K. V. Krishnamurty and G. M. Harris, Chem. Rev., 1961, 61(3), 213; O. Schott, J. Ferrando-Soria, A. Bentama, S. Stiriba, C. Ruiz-Pérez, M. Andruh, F. Lioret and M. Julve, Inorg. Chim. Acta, 2011, 376, 358.
3. L. Androš, M. Jurić, P. Planinić, D. Žilić, B. Rakvin and K. Molčaov, Polyhedron, 2010, 29, 1291.
4. M. J. Schmelz, T. Miyazawa, S. Mizushima, T. J. Lane and J. V. Quagliano, Spectrochim. Acta, 1967, 9, 51.
5. J. C. Blair and E. M. Jones, in Inorganic Synthesis, ed. H. S. Booth, McGraw-Hill Book Co., New York, 1939, vol. 1, p. 35.
6. A. Saritha, B. Raju, M. Ramachary, P. Raghavaiah and K. A. Hussain, Phys. B. Condens. Matter., 2012, 407, 4208.
7. P. C. Junk, J. Coord. Chem., 2005, 58(4), 355.
8. A. Saritha, B. Raju, D. Narayana Rao, A. Roychowdhury, D. Das and K. A. Hussain, Adv. Powder Technol., 2015, 26(2), 349.
9. C. A. Parker and C. C. Hotchord, J. Phys. Chem., 1959, 63, 22; C. A. Parker, Trans. Faraday Soc., 1954, 50, 1213.
10. W. W. Wendlandt and E. L. Simmons, J. Inorg. Nucl. Chem., 1966, 28, 2420.
11. M. Izakovič, J. Šima and V. Brezová, J. Photochem. Photobiol., A, 2004, 167, 81.
12. K. Sahul, L. V. Natarajan and Q. Anwaruddin, J. Polym. Sci., Polym. Lett. Ed., 1977, 15, 605.
13. A. Abbas and I. Tajuddin, J. Photochem., 1986, 35, 87.
14. A. S. Brar and B. S. Randhawa, J. Solid State Chem., 1985, 58, 153; A. S. Brar and B. S. Randhawa, Radiochem. Radioanal. Lett, 1981, 47, 219; A. S. Brar and B. S. Randhawa, Radiochem. Radioanal. Lett, 1980, 44, 377; A. A. Medvinskii and V. I. Poskevnev, Zh. Strukt. Khim., 1974, 15, 623; G. M. Bancroft, K. G. Dharmawardena and A. G. Maddock, Inorg. Chem., 1970, 9, 223.
15. T. M. Muziol, G. Wrzeszcz and L. Chrzaszcz, Polyhedron, 2011, 30, 169; R. Prasad, Sulaxna and A. Kumar, J. Therm. Anal. Calorim., 2005, 81, 441.
16. P. K. Gallagher and C. R. Kurkjian, Inorg. Chem., 1965, 5, 214; D. Broadbent, D. Dollimore and J. Dollimore, J. Chem. Soc. A, 1967, 451; W. W. Wendlandt and T. D. George, J. Inorg. Nucl. Chem., 1961, 21, 69.
17. O. Paulsen, Drugs Today, 1987, 23, 269.
18. V. T. Andriole, Clin. Infect. Dis., 2005, 41, S113; Quinolone Antimicrobial Agents, ed. D. C. Hooper and E. Rubinstein, ASM Press, Washington, DC, 3rd edn, 2003; D. C. Hooper, J. S. Wolfson, E. Y. Ng and M. N. Swartz, Am. J. Med., 1987, 82(suppl. 4A), 12.
19. D. E. King, R. Malone and S. H. Lilley, Am. Fam. Physician, 2000, 61, 2741.
20. H. K. Mayer and J. A. Ellal, Am. J. Med., 1999, 92, 58S; G. S. Ogawa and R. Hyndiuk, Int. Ophthalmol. Clin., 1993, 33, 59.
21. I. Turel, Coord. Chem. Rev., 2002, 232(1–2), 27.
22. L. Fratini and E. E. S. Schapoval, Int. J. Pharm., 1996, 127, 279.
23. D. S. Lee, H. Han, K. Kim, W. B. Park, J. K. CHO and J. H. Kim, J. Pharm. Biomed. Anal., 1994, 12(2), 157; J. Al-Mustafa and B. Tashtoush, J. Coord. Chem., 2003, 56(2), 113.
24. P. B. Issopoulos, Analyst, 1989, 114, 627.
25. D. L. Ross and C. M. Riley, Int. J. Pharm., 1992, 87, 203.
26. F. Gao, P. Yang, J. Xie and H. Wang, J. Inorg. Biochem., 1995, 60, 61.
27. S. A. Sadeek, J. Mol. Struct., 2005, 753, 1.
28. S. C. Wallis, L. R. Gahan, B. G. Charles and T. W. Hambely, Polyhedron, 1995, 14(20–21), 2835.
29. G. Psomas, J. Inorg. Biochem., 2008, 102, 1798.
30. H. F. Abdel-Halim, G. G. Mohamed, M. M. I. El-Dessouky and W. H. Mahmoud, Spectrochim. Acta, 2011, 82, 8.
31. L. C. Yu, Z. L. Tang, P. G. Yi and S. L. Liu, J. Coord. Chem., 2009, 62(6), 894.
32. J. L. R. Costa, G. S. Marchetti and M. C. Rangel, Catal. Today, 2002, 77, 205; C. Huang, H. Zhang, Z. Sun, Y. Zhao, S. Chen, R. Tao and Z. Liu, J. Colloid Interface Sci., 2011, 364, 298.
33. J. Zhang, S. Rama, R. S. Srivastava and R. D. K. Misra, Acta Biomater., 2008, 4, 40; S. Ghoshal, A. A. M. Ansar, S. O. Raja, A. Jana, N. R. Bandyopadhyay, A. K. Dasgupta and M. Ray, Nanoscale Res. Lett., 2011, 6, 540.
34. Z. Medarova, W. Pharm, C. Farrar, V. Petkova and A. Moore, Nat. Med., 2006, 13(3), 372.
35. A. H. Lu, E. L. Salabas and F. Schüth, Angew. Chem., Int. Ed., 2007, 46, 1222.
36. F. Gao, P. Yang, J. Xie and H. Wang, J. Inorg. Biochem., 1995, 60, 61.
37. A. M. Mansour and O. R. Shehab, Spectrochim. Acta, Part A, 2014, 128, 263.
38. D. K. Saha, D. Deobagkar, U. Sandbhor, K. Shirisha, C. E. Anson, S. Padhye and A. K. Powell, Bioorg. Med. Chem. Lett., 2004, 14, 3027.
39. N. T. Abdel-Ghani, M. F. Abo El-Ghar and A. M. Mansour, Spectrochim. Acta, Part A, 2013, 104, 134.
40. I. Turel, I. Leban and N. Bukovec, J. Inorg. Biochem., 1994, 56, 273.
41. A. M. Mansour and R. R. Mohamed, RSC Adv., 2015, 5, 5415.
42. A. Bentama, O. Schott, J. Ferrando-Soria, S. Stiriba, C. Ruiz-Pérez, M. Julve and J. Pasán, Inorg. Chim. Acta, 2012, 389, 52; X. Chen, L. Liu, J. Ma, L. Yi, P. Cheng, D. Liao, S. Yan and Z. Jiang, J. Mol. Struct., 2005, 750, 94.
43. M. Massacesi and G. Ponticelli, Spectrochim. Acta, Part A, 1981, 37(12), 1035.
44. A. M. Mansour, Polyhedron, 2014, 78, 10.
45. A. M. Mansour, Dalton Trans., 2014, 15950.
46. O. R. Shehab and A. M. Mansour, J. Mol. Struct., 2013, 1047, 121.
47. A. E. Reed, L. A. Curtiuss and F. Weinhold, Chem. Rev., 1988, 88, 899.
48. N. T. Abdel-Ghani and A. M. Mansour, Inorg. Chim. Acta, 2011, 373, 249.
49. A. M. Mansour, Inorg. Chim. Acta, 2013, 408, 186.
50. P. Guardia, J. Pérez-Juste, A. Labarta, X. Batlle and L. M. Marza, Chem. Commun., 2010, 46, 6108.
51. B. Zargar, H. Parham and M. Rezazade, J. Chin. Chem. Soc., 2011, 58, 694.
52. M. Rafatullaha, O. Sulaimana and R. Hashim, J. Hazard. Mater., 2010, 177, 70.
53. M. Amini, B. Pourbadiei, T. P. A. Ruberu and L. K. Woo, New J. Chem., 2014, 38, 1250.
54. I. Ursachi, A. Stancu and A. Vasile, J. Colloid Interface Sci., 2012, 377, 184.
55. P. V. Nidheesh, RSC Adv., 2015, 5, 40552.
56. M. A. Andrade, R. J. Carmona, A. S. Mestre, J. Matos, A. P. Carvalho and C. O. Ania, Carbon, 2014, 76, 183; G. Thennarasu and A. Sivasamy, Powder Technol., 2013, 250, 1.
57. S. Chin, E. Park, M. Kim and J. Jurng, Powder Technol., 2010, 201, 171; E. M. Seftel, M. Niarchos, Ch. Mitropoulos, M. Mertens, E. F. Vansant and P. Cool, Catal. Today, 2015, 252, 120.
58. Y. Liu, Y. Li, H. Yu, Z. Lv, Y. Yu, S. Zhan and X. Yang, J. Sol-Gel Sci. Technol., 2011, 58, 716.
59. E. O. Kajander and N. Ciftcioglu, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 8274.
60. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle and J. A. Pople, GAUSSIAN 03 (Revision A.9), Gaussian, Inc., Pittsburgh, 2003.
61. N. T. Abdel-Ghani and A. M. Mansour, Spectrochim. Acta, Part A, 2012, 91, 272.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra12157d

---