Manganese porphyrins as efficient catalysts in solvent-free cyclohexane oxidation

2.1 Reagents

Analytical grade reagents CH₃OH, CH₂Cl₂, CHCl₃ were purchased from VETEC and previously distilled. PhIO was prepared following a protocol in the literature (Saltzman and Sharefkin, 1963), stored at −20 °C and periodically analyzed by iodometric titration. All other reagents and solvents were used without previous purification.

2.2 Equipment

UV–Vis absorption spectroscopy was performed in an HP 8453A diode-array spectrophotometer at a wavelength range of 190–1100 nm with a 10 mm optical path quartz cuvette. Solvent elimination was performed in a Büchi rotary evaporator model R-114B-480. Porphyrin solubilization was assisted with a Unique® MaxiClean 1400, 40 kHz ultrasound device. Thin-layer chromatographic analysis and metalation monitoring through the loss of fluorescence from the porphyrin ring were performed with a 365-nm wavelength ultraviolet lamp adapted to a Spectroline dark cabinet, Model CX-20. A Sartorius analytical balance, model BP210D, with maximum load of 210 g and accuracy of 0.01 mg was also used. ¹H NMR spectra were recorded with a Bruker spectrometer, Model Avance DRX-400 (400 MHz). Infrared spectra were obtained with a Perkin Elmer spectrometer; model BXFTIR; the samples were prepared in KBr pellets previously oven dried for 3 h at 150 °C. Mass spectra were obtained with a Bruker MicroTOF QII electronspray ionization mass spectrometer (ESI-MS) operating in positive and negative modes, using methanol as a solvent. A Corning hot plate magnetic stirrer was used in catalytic reactions. The cyclohexane oxidation products were analyzed in a Shimadzu GC-17A gas chromatograph fitted with a flame ionization detector (FID) and an Altech Econo-Cap capillary column 30.00 m long with 0.32 mm diameter and 0.25 film thickness using a Carbowax® stationary phase. The flame gas was a synthetic air-hydrogen mixture, hydrogen was used as carrier. The oven temperature was programmed to rise from 80 °C to 150 °C at 5 °C min⁻¹, and then kept for 1 min. The injector and detector temperatures were set to 220 °C.

2.3 Synthesis of the metalloporphyrin catalysts

2.3.1

2.3.1 Synthesis of second-generation metalloporphyrins MnP1 and cis/trans-MnP2

H₂P1 and a mixture of cis/trans-H₂P2 isomers, precursors in the synthesis of second- and third-generation catalysts, were prepared according to da Silva et al. (2014) and characterized by ¹H NMR (Fig. S1, Supplementary Material) and UV–Vis absorption spectroscopy (Fig. 2 and Fig. S2, Supplementary Material).

Close

Fig. 2 UV–Vis absorption spectra of cis/trans-H₂P2 (4.32 × 10⁻⁶ mol L⁻¹), cis/trans-MnP2 (7.48 × 10⁻⁶ mol L⁻¹) and cis/trans-MnBr₈P2 (1.49 × 10⁻⁵ mol L⁻¹) in CH₂Cl₂.

Export to PPT

Manganese porphyrin complexes were prepared following (Borovkov et al., 1999) with modifications. MnP1 was obtained from 98.5 mg (0.149 mmol) of H₂P1 and cis/trans-MnP2 from 146.0 mg (0.207 mmol) of cis/trans-H₂P2 isomer mixture dissolved in 15 mL and 20 mL of CH₂Cl₂, respectively. 200 µL of collidine (2,4,6-trimethyl pyridine) were subsequently added to each reaction. A methanol solution (5 mL) of manganese(II) acetate tetrahydrate (Mn(OAc)₂·4H₂O) was added 20-fold in molar excess to the porphyrin solution. The systems were refluxed and magnetically stirred for 12 h, then the solvent was removed on the rotary evaporator. Metalloporphyrins purification was done by column chromatography (column height = 15 cm; column diameter = 2 cm) with SiO₂ (Sigma-Aldrich, 60 Å, 70–230 mesh) as stationary phase. Elution was initially done with CH₂Cl₂ (for separation of the free base porphyrin) and then with a CH₂Cl₂:CH₃OH (10:1) mixture. The resulting solids were eluted with CH₃OH on an ion exchange resin column Dowex 2X-8 (Sigma-Aldrich, 100–200 mesh, chloride form).

MnP1: Yield 76% (85.0 mg; 0.11 mmol). UV–Vis (CHCl₃) λ_max (nm) (log ε): 375 (4.39); 478 (4.70); 581 (3.64); 618 (3.66). FTIR (Fig. S3a, Supplementary Material) in KBr (cm⁻¹): (1596, 1488) ν C⚌C; (1518) ν_as NO₂; (1342) ν_sNO₂; (1284) δ of the porphyrin skeleton; (1010) δ Mn-N. ESI-TOF (Fig. S4a, Supplementary Material) [Mn^IIIMNPTPP]⁺ m/z 712.14 (100%). Anal. Calcd for C₄₄H₂₇ClMnN₅O₂.1CH₃OH: C, 69.28; H, 4.00; N, 8.98. Found: C, 69.22; H, 3.62; N, 8.78.

Cis/trans-MnP2: Yield 80% (131.87 mg; 0.16 mmol). UV–Vis (CHCl₃), λ_max (nm) (log ε): 377 (4.79); 478 (5.04); 581 (4.09); 618 (4.11). FTIR (Fig. S3b, Supplementary Material) in KBr (cm⁻¹): (1596, 1488) ν C⚌C; (1518) ν_as NO₂; (1344) ν_sNO₂; (1282) δ of the porphyrin skeleton; (1010) δ Mn-N. ESI-TOF (Fig. S4b, Supplementary Material) [Mn^IIIDNPDPP]⁺ m/z 757.13 (100%). Anal. Calcd for C₄₄H₂₆ClMnN₆O₄.1CH₃OH: C, 65.50; H, 3.66; N, 10.18. Found: C, 65.41; H, 3.59; N, 10.03.

2.4 Oxidative catalysis of cyclohexane

Solutions of second- and third-generation manganese porphyrins (1.0 × 10⁻³ mol L⁻¹) in dichloromethane were prepared for the oxidation reactions of cyclohexane with solvent in a volumetric flask. The solutions were transferred to 2.0-mL flasks with threaded stoppers and Teflon/silicon septa and maintained in a water bath at approximately 20 °C. The mass of the oxygen donor, PhIO or PhI(OAc)₂, (2.0 × 10⁻⁶ mol), was measured in 2.0-mL glass flasks on an analytical balance to a oxidant/MnP molar ratio of 10. The flasks were sealed with threaded stoppers and Teflon/silicon septa. 200 μL of metalloporphyrin in dichloromethane solution and 100 μL of substrate (cyclohexane) were added to the reaction flasks with Hamilton® microsyringes.

For the oxidation reactions of cyclohexane without solvent, the mass of metalloporphyrin (2.0 × 10⁻⁷ mol) was also measured using an analytical balance in the same 2.0-mL glass flask containing the oxygen donor, to which 300 μL of cyclohexane were added with Hamilton® microsyringes. The reaction mixture was manually stirred for 1 min at room temperature before the reactions with and without the solvent. The flasks were placed in a water bath at 25 °C immediately afterwards and magnetically stirred for 90 min. After the stirring period, the reaction was inhibited by adding a saturated solution of sodium sulfite and sodium tetraborate in dichloromethane (50.0 μL). Next, 50.0 μL of bromobenzene internal standard solution (4.8 × 10⁻² mol L⁻¹) in dichloromethane were added.

In addition to the reactions above, some control reactions were also performed. First, tests with the absence of manganese porphyrin were performed to check oxidation products formation in a non-catalytic medium, as well as tests in the absence of the oxidant. Next, manganese porphyrin was replaced with manganese acetate in a procedure similar to the oxidation without solvent in order to analyze the importance of choosing metalloporphyrin as a catalyst. Finally, cyclohexanol was oxidized in a procedure similar to the oxidation with the solvent to analyze the mechanism of formation of cyclohexanone.

An aliquot of 0.5 μL of each reaction mixture was injected into chromatograph CG 17A. All reactions were performed in quadruplicate and the products were analyzed by gas chromatography using the internal standard method (Ribani et al., 2004). The yields were calculated based on the oxygen donor, iodosylbenzene or iodobenzene diacetate.

3.1 Synthesis of the metalloporphyrin catalysts

Separation of the isomers, both in the free-base cis/trans-H₂P2 and in the metalloporphyrins was attempted during the synthesis of cis/trans-MnP2 and cis/trans-MnBr₈P2. To our knowledge, there is no record in the literature of this procedure. Therefore, we attempted to separate the metalloporphyrin isomer mixture by thin-layer chromatography with silica and alumina plates using mixtures of solvents in various proportions. As it wasn’t successful, we decided to use the isomer mixture in the catalytic oxidation reactions.

Second-generation metalloporphyrins were obtained in reactions with manganese(II) acetate tetrahydrate, which are favored by the basic nature of the acetate anion, as it promotes the deprotonation of the pyrrole hydrogen atoms of the porphyrin ring, thus enabling the metalation reaction (Wijesekera and Dolphin, 1994). The complexation reaction was favored by adding 2,4,6-collidine (C₈H₁₁N), which has the capacity to capture the hydrogen ion released during the reaction and favors the metallic ion coordination (Borovkov et al., 1999).

The metalation reactions were monitored by thin-layer chromatography using CH₂Cl₂ as an eluent. H₂P1 and cis/trans-H₂P2 moved along the chromatographic silica plate (Rf ∼0.7), while MnP1 and cis/trans-MnP2 were retained close to the application spot (Rf ∼0.06).

The loss of fluorescence under ultraviolet light of the metalloporphyrins could also be observed. This phenomenon can be attributed to the presence of heavy atoms, in this case, manganese ion, that increase the radioactive decay rate for an intersystem crossing to the excited triplet state due to spin-orbit coupling, which results in phosphorescence, and in a decrease in the fluorescence quantum yield (Harriman, 1981).

The metalloporphyrins were characterized by UV–Vis absorption spectroscopy and the results were consistent with the spectra of other manganese porphyrins in the literature (do Nascimento et al., 2005; Silva et al., 2007; Horvath et al., 2012; da Silva et al., 2014). The H₂P1, MnP1 and MnBr₈P1 spectra (Fig. S2, Supplementary Material) did not differ from the cis/trans-H₂P2, cis/trans-MnP2 and cis/trans-MnBr₈P2 spectra (Fig. 2), respectively, because they are all meso-substituted porphyrin compounds with substitutions on the para-positions of the phenyl group. Thus, the influence of the π electronic density of the phenyl group on the π density of the porphyrin ring is very small, since they are located in different planes due to steric hindrance (Kim et al., 1972).

Analysis of the spectra in Fig. 2 shows that the Soret band of the third-generation metalloporphyrins presented a bathochromic shift in comparison to the second-generation metalloporphyrins. This happened because the introduction of electron-withdrawing bromine atoms in the β-pyrrolic positions decreases the electronic density in the porphyrin macrocycle and changes the planar conformation into a saddle conformation (Valicsek et al., 2011). Thus, the HOMO orbitals are destabilized in brominated metalloporphyrins due to a reduction in the conjugation of the macrocycle π orbitals. Furthermore, the HOMO and LUMO orbitals are stabilized due to the inductive effect. These combined factors provoke a reduction in the energy between the HOMO-LUMO orbitals (Autret et al., 1996; do Nascimento et al., 2005; da Silva et al., 2008).

After bromination of the metalloporphyrins, two oxidation states of the metallic center were observed, Mn(II) and Mn(III) (Fig. S8, Supplementary Material). β-octabrominated metalloporphyrins tend to have a metallic center in oxidation state 2+ because the introduction of bromine atoms into β-pyrrolic positions of the macrocycle favors a shift of the Mn(III)/Mn(II) redox potential to more positive values (anodic shift) in relation to the non-brominated ones. This increase in potential is sufficient to stabilize the central metallic ion in a lower oxidation state, even in the presence of air. This is attributed to the electron withdrawing inductive effect of bromine in direct conjugation with the porphyrin π system, which decreased the electronic density in both metallic center and conjugated π system of the macrocycle, resulting in the favoring of the complex with the Mn(II) ion (Bhyrappa and Krishnan, 1991; Hariprasad et al., 1996; Ghosh et al., 2001). The presence of the Mn(II) complex together with Mn(III) complex provokes an alteration in the metalloporphyrin UV–Vis absorption spectrum, leading to the appearance of a new band at 465 nm. Analysis of the spectrum of the Mn(II) complex isolated during the metalloporphyrin purification by alumina column revealed a Soret band at 472 nm characteristic of β-halogenated Mn(II) metalloporphyrins (Boucher, 1972). This fact affected the yield of the synthesis of third-generation catalysts, as part of the product obtained contained Mn(II). The Mn(II) fraction was solubilized in chloroform and stirred for one week in an attempt to force oxidation, without success. Thus, this fraction was not studied in the catalytic processes because it is less stable to demetalation than that with Mn(III) (Reboucas et al., 2008).

The absence of N—H (pyrrole) stretching band and the presence of Mn—N (pyrrole) stretching band (−1010 cm⁻¹ for MnP1, 1024 cm⁻¹ for MnBr₈P1, 1010 cm⁻¹ for cis/trans-MnP2 and 1026 cm⁻¹ for cis/trans-MnBr₈P2) (Boucher and Katz, 1967; Boucher, 1968) in vibrational spectroscopy (infrared region) characterization indicates the formation of the complexes (see Fig. S3, Supplementary Material for more information).

Mass spectrometry analysis (Figs. S4 and S5, Supplementary Material) corroborates the structure proposed for all metalloporphyrins, with the complete β-octabromination of the macrocycles of third generation metalloporphyrins as they presented peaks at m/z 712.14 (100%) for MnP1, 757.13 (100%) for cis/trans-MnP2, 1343.42 (100%) for MnBr₈P1 and 1388.41 (100%) for cis/trans-MnBr₈P2, associated with the loss of chloride ion. The experimental results were very close to the expected theoretical mass. ¹H NMR spectroscopy of H₂Br₈P1 and cis/trans-H₂Br₈P2 after demetalation of MnBr₈P1 and cis/trans-MnBr₈P2 (Fig. S7, Supplementary Material) revealed the absence of hydrogen in β-pyrrolic positions (8.5–9 ppm), an indication of octabromination of the macrocycle.

3.2 Catalytic study of the oxidation of cyclohexane

Cyclohexane was used as a substrate in the catalysis reactions. Because it is considered a standard substrate in oxidation reactions, it allows evaluating the efficiency of metalloporphyrin catalysts.  The main products obtained in the reactions catalyzed by metalloporphyrins are cyclohexanol (Cy-ol) and cyclohexanone (Cy-one) (Meunier, 1992). Two different oxygen donors were used in the systems investigated, PhIO and PhI(OAc)₂, both in presence and in absence of CH₂Cl₂, the reaction solvent (Table 1).

Control reactions were also performed in either the absence of catalysts or of oxidants. In addition, cyclohexanol was oxidized to check whether cyclohexanone resulted from the oxidation of cyclohexanol during the reaction.

3.3 Reactions with solvent

Cyclohexane was oxidized in the presence of CH₂Cl₂ for comparison of the catalytic activity of second-generation metalloporphyrins (MnP1 and cis/trans-MnP2) and the novel third-generation metalloporphyrins, MnBr₈P1 and cis/trans-MnBr₈P2, with that of other catalysts found in the literature. Furthermore, these results served as a comparison parameter for the reactions without solvent.

The cyclohexane oxidation control reactions using manganese(II) acetate tetrahydrate (Mn(OAc)₂·4H₂O) in the presence of PhIO or PhI(OAc)₂ did not afford the products, which gives evidence of the need for interaction of metallic ion with the porphyrin macrocycle. The reactions using only metalloporphyrins catalysts in absence of oxidants or only in presence of oxidants did not yield oxidation products either.

In relation to the classical first-generation metalloporphyrin [Mn^IIITPPCl] using PhIO as an oxidant (Balcells et al., 2008; da Silva et al., 2015a, 2015b), MnP1 and cis/trans-MnP2 afforded an increase in the total yield and selectivity for alcohol, from 25 and 56% for the first generation, to approximately 30 and 72% for the second generation, respectively (entries 1 and 5, Table 1). The insertion of the nitro group (—NO₂) in the —para position of the phenyl group (at the macrocycle’s meso position) of the second-generation metalloporphyrins led to noticeably better catalytic results. The reason is that the nitro group, an electronic density withdrawer, increases high-valence active species reactivity, Mn^V(O)P, facilitating the transfer of the oxygen atom to the substrate (Latifi et al., 2011). Additionally, the inductive effect provoked by the nitro groups may turn the oxygen of high-valence active species more electrophilic, which hinders the escape of the radical and leads to the collapse of the cage or of the ionic pair, thus favoring the formation of Cy-ol (Hill and Schardt, 1980; Smegal and Hill, 1983; Smegal et al., 1983; Latifi et al., 2011).

Comparison of second-generation metalloporphyrins MnP1 and cis/trans-MnP2 showed that they have a similar catalytic behavior (entries 1 and 2 versus 5 and 6, respectively, Table 1) for the same oxidant. No negative influence was observed for the insertion of another nitro group in cis/trans-MnP2. However, for the third-generation metalloporphyrins (entries 3 and 4 versus 7 and 8, Table 1), catalyst cis/trans-MnBr₈P2 was significantly more efficient than MnBr₈P1, giving better yields, greater selectivity for Cy-ol and a smaller degree of destruction of the catalyst with both oxidants.

Third-generation metalloporphyrins MnBr₈P1 and cis/trans-MnBr₈P2 gave better catalytic results than the respective second-generation metalloporphyrins, MnP1 and cis/trans-MnP2, with both oxidants used, achieving greater product yield and smaller degree of destruction of the catalyst, in agreement with the literature (Hoffmann et al., 1990; da Silva et al., 2017). This fact is a consequence of the introduction of bromine atoms into β-pyrrolic positions that withdraw electronic density from the metallic center, destabilizing the high-valence active species Mn^V(O)P and facilitating the transfer of the oxygen atom to the substrate (Latifi et al., 2011). Furthermore, they lead to a macrocycle conformation change from planar to saddle, reducing the possibility of occurrence of oxidative destruction processes (Valicsek et al., 2011). When comparing the degree of destruction of the catalysts, it was found that the third generation catalysts had a lower degree of destruction than the second generation catalysts, which in turn had a lower degree of destruction than the first generation catalyst (Mn^IIITPPCl) (da Silva et al., 2014). This result evidences the importance of the substituents (bromine in β-pyrrolic positions and —NO₂ at aryl group) on the periphery of porphyrin macrocycle. This is because the presence of these functional groups diminishes the process of oxidative destruction, that is, a metalloporphyrin, causes oxidation of another metalloporphyrin. The lower degree of destruction is even more pronounced in the third-generation metalloporphyrins as a function of the distortion caused by the bromine atoms at the β-pyrrole positions, which make difficult the approximation of the metalloporphyrins and, consequently, decrease the oxidative destruction process.

A parallel can be drawn between non-symmetric metalloporphyrins with the nitro (—NO₂) or the amine group (—NH₂) as substituents. Second-generation [Mn^IIIAPTPPCl], which presents only an amine group in a para-position of the meso-phenyl group (da Silva et al., 2014), had a total yield of 32% in the presence of PhIO and 60% selectivity for alcohol. Thus, in comparison to MnP1 (entry 1, Table 1), which has only one nitro group in a para- position of the meso-phenyl group, the yield was the same. The reason is that although the amine group is an electronic density donor group, it might behave as a Lewis base for the metallic ion of another metalloporphyrin, increasing the reactivity of the high-valence active species Mn^V(O)P, compensating the electronic density donor effect caused by the amine group and being equivalent to the electronic density withdrawal effect of the nitro group (da Silva et al., 2014). However, greater selectivity was observed for MnP1 (70%) in relation to [Mn^IIIAPTPPCl] (60%), possibly due to the electronic density withdrawal effect of the nitro group. Comparatively to the catalytic activity of [Mn^IIIBr₉APTPPCl], a third generation metalloporphyrin with total yield of 52% and selectivity of 71% for alcohol (da Silva et al., 2014), MnBr₈P1 (entry 3, Table 1) performed better, since the systems with this metalloporphyrin had a total yield of 80% with similar selectivity, 70%. This may be attributed to steric hindrance caused by the insertion of bromine atoms into the macrocycle, which makes the coordination of the amine group with the metallic center more difficult. Thus, [Mn^IIIBr₉APTPPCl] has opposite electronic effects: the mesomeric effect, donor of electrons to the macrocycle, provoked by the amine group, and the inductive effect, electron withdrawer, caused by the bromine atoms in β-pyrrolic positions. In contrast, besides MnBr₈P1 having the same eight bromine atoms that withdraw electronic density in β-pyrrolic positions, it also has a nitro group that provokes the same effect (inductive) and thus contributes to the greater reactivity of the high-valence active species.

Second-generation metalloporphyrin cis-[Mn^IIIDAPDPPCl], with an amine group in the -para position of two meso-phenyl groups, had a total product yield of 32% when PhIO was used, similar to that obtained with cis/trans-MnP2 (entry 5, Table 1). This fact may also be attributed to the similar reactivity of the high-valence active species Mn^V(O)P, despite the opposite electronic effects of the catalysts, as cis-[Mn^IIIDAPDPPCl] is influenced by intermolecular coordination from the amine group, while cis/trans-MnP2 is susceptible to the inductive effect from the nitro group. Similarly to the brominated metalloporphyrins with one nitro group in the macrocycle, cis-[Mn^IIIBr₁₂DAPDPPCl], a third-generation metalloporphyrin (da Silva et al., 2015a), had a total yield of 50%, smaller than that of 89% obtained for cis/trans-MnBr₈P2 (entry 7, Table 1).

MnP1 and cis/trans-MnP2 are also more efficient catalysts in comparison to second-generation metalloporphyrin Mn^III(TCMPP)Cl (da Silva et al., 2008), which has carbomethoxy groups (—COOCH₃) in –para positions of the meso-phenyl, as it led to a yield of 16% in the presence of PhIO and 21% in the presence of PhI(OAc)₂ for Cy-ol, with selectivity for alcohol at around 75% in both systems. In contrast, MnP1 and cis/trans-MnP2 (entries 1, 2, 5 and 6, Table 1) had higher yields for Cy-ol in all systems (approximately 22% in PhIO and 33% in PhI(OAc)₂) and similar selectivity in PhIO and slightly smaller in PhI(OAc)₂. Third-generation metalloporphyrins, MnBr₈P1 and cis/trans-MnBr₈P2 (entries 3, 4, 7 and 8, Table 1), had better catalytic results (56 and 70% yields with PhIO and 32 and 38% yields with PhI(OAc)₂, respectively, for Cy-ol), in comparison to Mn^III(Br₈TCMPP)Cl(da Silva et al., 2008), which had a yield of 21% in PhIO and 29% in PhI(OAc)₂ for Cy-ol, with selectivity for alcohol of approximately 78% in both systems. These data allow the inference that the electronic density withdrawing inductive effect provoked by the nitro group is more effective than that provoked by the carbomethoxy group (—COOCH₃).

In da Silva et al. (2017), metalloporphyrin [Mn^IIIT4MPPCl] had a total yield of 23% and 78% selectivity for alcohol in the presence of PhIO. Comparing MnP1 and cis/trans-MnP2 (entries 1 and 5, Table 1), the metalloporphyrin with the nitro group had greater yields (approximately 30% in relation to the total yield) and similar selectivity values. This result can be justified by the presence of the nitro group in MnP1 and cis/trans-MnP2 that withdraws electronic density from the macrocycle and the metallic ion, while the methoxy group donates electronic density to the porphyrin macrocycle. [Mn^IIIT3,5DMPPCl] (da Silva et al., 2015c), which has methoxy groups in –meta positions in the phenyl groups, had a total yield of 21% with PhIO. Therefore, the insertion of another four methoxy groups in the porphyrin ring did not affect the results significantly. MnP1 and cis/trans-MnP2 had a better catalytic performance than [Mn^IIIT3,5DMPPCl]. MnBr₈P1 and cis/trans-MnBr₈P2 were compared to third-generation metalloporphyrins [Mn^IIIBr₁₂T4MPPCl] (da Silva et al., 2017) (46% total yield and 80% selectivity for Cy-ol) and [Mn^IIIBr₁₂T3,5DMPPCl] (da Silva et al., 2015c) (50% total yield and 96% selectivity for Cy-ol) in PhIO and the same pattern of behavior of second-generation metalloporphyrins was observed: an increase in the total yield of metalloporphyrins with nitro groups (80% for MnBr₈P1 and 89% for cis/trans-MnBr₈P2) in relation to the metalloporphyrins with methoxy groups.

The use of PhI(OAc)₂ with second-generation catalysts MnP1 and cis/trans-MnP2 resulted in an increase in product yield in relation to PhIO. For MnP1, the relative percent increase in total yield (Cy-ol + Cy-one) was 74% and for cis/trans-MnP2, 57%; however, both systems had a slight decrease in selectivity for Cy-ol. The use of PhI(OAc)₂ led to an increase in the degree of destruction of MnP1 and cis/trans-MnP2, because high-valence active species distinct from that present in the reaction medium with PhIO must have been formed, as proposed in a kinetic study conducted by Collman et al. (2000). In contrast, third-generation metalloporphyrins MnBr₈P1 and cis/trans-MnBr₈P2 had better results with PhIO. In this case, there was a relative percent decrease in the total yield of 41% for MnBr₈P1 and 36% for cis/trans-MnBr₈P2 and a decrease in the selectivity for Cy-ol. These differences may be attributed to the different high-valence active species formed with each oxidant (Collman et al., 2000). Additionally, the degree of destruction of the catalysts increased with PhI(OAc)₂ in relation to the PhIO system. This result may be associated with the distorted saddle conformation of brominated metalloporphyrins, which results in a smaller overlapping of π orbitals. As a result, the compound is more susceptible to destruction in the presence of various oxidant species in the reaction medium, and, as different oxidant species are formed in systems with PhI(OAc)₂ (Silva and Lopes, 2005), the degree of destruction is greater when this oxidant is used. In face of the different behaviors of the PhIO and PhI(OAc)₂ systems, mechanistic analysis associated with a theoretical study is necessary to better understand systems with PhI(OAc)₂ and how it reacts with the macrocycle of second- and third-generation metalloporphyrins. The study of the mechanism of oxidation of cyclohexane is equally important as it is directly related to structural modifications made to the metalloporphyrins used as catalysts in the oxidation of C—H bonds. Various oxidation reactions of Cy-ol catalyzed with cis/trans-MnBr₈P2 were conducted to this end, as Cy-one, one of the oxidation products of cyclohexane, may be a product of Cy-ol oxidation. These systems presented a yield of 56% for Cy-one with PhIO, and 80% with PhI(OAc)₂ (entries 9 and 10, Table S1, Supplementary Material). The degree of destruction of the metalloporphyrin catalysts was approximately the same in both systems. These results corroborate the proposed mechanism of formation of Cy-one from Cy-ol, which directly affects the selectivity of the catalysis reactions with and without solvent. A mechanistic proposal for this reaction based on studies by Smegal and Hill (1983) is presented in Fig. 3. Initially the MnP reacts with the oxidant to generate the high-valence active specie Mn^V(O)P (1); then the oxygen atom of that specie interacts with the hydrogen atom of the cyclohexanol OH group and, simultaneously, with the hydrogen atom of the carbon bonded to the OH group, forming a five-membered ring (2); then, in a concerted way, the electron pair of C—H bond is transferred to the C—O bond and Cy-one and a metalloporphyrin bound to a water molecule are formed (3); next, the water molecule is lost and the catalyst is regenerated (4).

Close

Fig. 3 Proposed mechanism of oxidation of cyclohexanol into cyclohexanone.

Export to PPT

3.4 Reactions without solvent

Catalysis reactions without solvent have a great impact in green chemistry because they reduce the production of residues and the use of toxic solvents. In the literature, these reactions are performed mainly in the presence of molecular oxygen at high temperatures and pressures, these systems offer wide application for industry (Liu and Guo, 2012). In traditional industrial processes, the oxidation of cyclohexane has a conversion of approximately 5%, and studies have been conducted to improve it. Metalloporphyrin catalysts stand out as they afford a conversion of approximately 10% (Guo et al., 2003a, 2003b; Liu and Guo, 2012).

Seeking to reduce the consumption of energy in reactions without solvent, the use of metalloporphyrin catalysts has been investigated in the oxidation of cyclohexane in systems under mild reaction conditions, that is, at room temperature and pressure. The systems without solvent, given in Table 2, like the systems with solvent in Table 1, presented a better product yield and a smaller degree of destruction of third-generation metalloporphyrins MnBr₈P1 and cis/trans-MnBr₈P2 (entries 13, 14, 17 and 18, Table 2) when compared to second-generation catalysts MnP1 and cis/trans-MnP2 (entries 11, 12, 15 and 16, Table 2) for both oxidants. The reason is that third-generation metalloporphyrins have bromine atoms in β-pyrrolic positions that withdraw the electronic density and thus increase the reactivity of the high-valence active species (Latifi et al., 2011).

The systems with second-generation metalloporphyrins MnP1 and cis/trans-MnP2 with oxidant PhIO (entries 11 and 15, Table 2) had a similar Cy-ol yield. These results are similar to those obtained with solvent (entries 1 and 5, respectively, Table 1); however, the catalytic systems are 100% selective for Cy-ol in the absence of solvent, which is desirable as it avoids additional product separation steps.

In systems with second-generation metalloporphyrins with PhI(OAc)₂ as an oxidant (entries 12 and 16, Table 2), only Cy-ol was formed, but in very low yields. These results differ from those obtained with solvent (entries 2 and 6, Table 1), possibly because PhI(OAc)₂ is not soluble in the system in reactions without solvent, and, therefore, may hinder the formation of PhIO in situ, which agrees with the results in In et al. (2003).

In systems with third-generation catalysts MnBr₈P1 and cis/trans-MnBr₈P2, the total yield decreased (Cy-ol + C-one) in reactions without solvent and with PhIO and PhI(OAc)₂ (entries 13 and 17, 14 and 18, respectively, Table 2) in relation to the systems with CH₂Cl₂ (entries 3 and 7, 4 and 8, respectively, Table 1). However, the systems without solvent were much more selective for Cy-ol with both oxidants, reaching 96–100% in the presence of PhIO. Cis/trans-MnBr₈P2 was a much more efficient catalyst in the formation of products than MnBr₈P1, also in reactions with solvent. Furthermore, third-generation metalloporphyrins were also more efficient than their second-generation analogues.

The reduction in Cy-ol yield in reactions without solvent in relation to reactions with solvent may have been due to a smaller number of active sites available for the formation of the high-valence active species. Consequently, the Cy-one yield was smaller, possibly due to the oxidation of Cy-ol. This oxidation is also unfavored, since the amount of cyclohexane in the medium is much greater than that used in the reactions with solvent.

The degree of destruction for MnP1 and MnBr₈P1 was smaller in the systems without solvent when compared to the systems with CH₂Cl₂, since the oxidative destruction process is smaller in the reaction medium without a complete soluble metalloporphyrin. Surprisingly, cis/trans-MnP2 and cis/trans-MnBr₈P2 had opposite tendencies, with a greater degree of destruction in systems without solvent. In view of this, we believe that the higher degree of destruction of the porphyrin macrocycle for cis/trans-MnP2 and cis/trans-MnBr₈P2 in the oxidation reactions of cyclohexane without solvent is due to the higher reactivity of these metalloporphyrins with two nitro groups, in comparison with those with only one nitro group (MnP1 and MnBr₈P1). We believe that this happens because these groups in the para-positions of the mesoaryl groups changes the electronic density of the macrocycle, making it more susceptible to oxidative destruction, as can be evidenced in the work described by Zakavi et al. (2011). This result corroborates with the work developed by Song et al. (2005) that any modification in the systems involving metalloporphyrins in the biomimetic catalysis is sufficient to completely change their behavior, these authors classified this behavior as “chameleon oxidant” because changes in the reaction environment affect the reactivity and selectivity of the active species. Further studies are necessary to understand the mechanisms involved in the process of destruction of these metalloporphyrins.