Nabyendu Pramanik,
Satinath Sarkar,
Dipanwita Roy,
Sudipto Debnath,
Sukla Ghosh,
Saikat Khamarui and
Dilip K. Maiti*
Department of Chemistry, University of Calcutta, 92 A. P. C. Road, Kolkata-700009, India. E-mail: dkmchem@caluniv.ac.in; Fax: +91-33-2351-9755; Tel: +91-33-2350-9937
First published on 17th November 2015
The first synthesis and diverse oxidative cyclization catalysis properties of high-valent MoVI–triazole are demonstrated towards highly selective construction of benzimidazoles, benzothiazoles, isoxazolines, isoxazoles and their chiral analogues.
Benzimidazoles are versatile compounds used in a wide range of scientific and industrial applications such as important building blocks for many organic syntheses, catalysis, fluorescence, chemosensing, crystal engineering, corrosion science, materials, organic electronics, medicinal research, pharmaceuticals, agrochemicals, textiles and cosmetics.5 The widely used oxidative cyclocondensation of o-phenylenediamines and aldehydes to benzimidazoles has a serious problem of forming two products6d namely 2-substituted and 1,2-disubstituted benzimidazoles (Table 1). In addition to the huge number of reports on the synthesis of valuable benzimidazoles,6 the development of an efficient, benign and highly selective strategy is still in demand, especially for affording 2-substituted benzimidazoles and their chiral analogues7 bearing free-NH for a wide range of applications in the nanoscience and pharmaceutical industry. The scope of the benign oxidative cyclization catalysis can be extended towards the synthesis of the therapeutically useful benzothiazoles utilising labile 2-aminothiophenols and aldehydes.8,9 Isoxazolines10 and isoxazoles11 are essential building blocks of numerous natural products, useful synthons, chiral macrocycles and bioactive compounds for a wide range of medicinal applications.12 In fact these are the most frequently synthesised five member heterocycles through 1,3-dipolar nitrile oxide cycloadditions.13,14
| Entry | Catalysta | Reaction conditionsb,c | Conversion (%) | Yieldd (%) |
|---|---|---|---|---|
| a Catalyst loading: 10 mol%.b Oxygen from air.c rt.d Isolated product after purification.e 8 mol%.f 7 mol%.g 10 mol%.h Not isolated. | ||||
| 1 | MoO2(HL) | DCM, MgSO4, 7 h | 98 | 68 |
| 2 | MoO2(HL) | PhMe, MgSO4, 24 h | 60 | 45 |
| 3 | MoO2(HL) | MeOH, MgSO4, 12 h | 40 | 25 |
| 4 | MoO2(HL) | MeCN, MgSO4, 16 h | 80 | 56 |
| 5 | MoO2(HL) | THF, MgSO4, 6 h | 100 | 86 |
| 6 | MoO2(HL)e | THF, MgSO4, 7 h | 100 | 86 |
| 7 | MoO2(HL)f | THF, MgSO4, 20 h | 90 | 79 |
| 8 | MoO2(HL), CeCl3g | THF, urea–H2O2, MgSO4, 6 h | 100 | 89 |
| 9 | MoO2(HL) | PhI(OAc)2, THF, MgSO4, 12 h | 100 | 55 |
| 10 | MoO2(HL) | THF, MgSO4, Ar, 9 h | 10 | 5h |
| 11 | MoO2(HL) | THF, MgSO4, O2, 6 h | 100 | 88 |
| 12 | MoO2(acac)2 | THF, MgSO4, 24 h | 50 | 35 |
The organometallic chemistry of 3,5-bis(2-hydroxyphenyl)-1H-1,2,4-triazoles (H3L, Scheme 1) was first studied by Ryabukhin in 1983.15 To the best of our knowledge, an MoVI-complex with an “ONO” ligand is unknown in the literature. We have devised a simple procedure to access MoVI–1,2,4-triazole complexes from MoO2(acac)2 under benign reaction conditions (Scheme 1). The structure of the new high-valent compound was established by single crystal X-ray crystallography (panel A, Fig. 1),16 NMR, FTIR, SQUID (panel B) and EPR (panel C) analyses. Interestingly the XRD structure of triclinic symmetry with a P
space group reveals a higher binding site of MoVI, which is associated with water and DMF [MoO2(HL)(H2O)DMF]. The loosely bound MoVI–N (HL), MoVI–O (H2O) and MoVI–O (DMF) may be useful for binding substrates during oxidative catalysis processes. The MoVI-complex is expected to be diamagnetic with a d0 electronic configuration,17 and the higher oxidation state species revealed an unexpectedly high magnetic moment, measured using SQUID (0.54 BM, panel B), with a hyperfine X-band EPR spectrum (panel C). The free-N–H of the complex may play a vital role during the catalytic process through transforming coordinated N–MoVI σ-bonds as per the requirement of the push–pull mechanism of the catalytic cycles.
To explore the oxidative catalytic activity of the new high-valent metal complex we initially executed a cyclocondensation cum oxidation reaction using 10 mol% of MoO2(HL)(H2O)(DMF), o-phenylenediamine (1a, 1 mmol), benzaldehyde (2a, 1.1 mmol), desiccant MgSO4 and solvent dichloromethane under aerobic conditions at ambient temperature (entry 1, Table 1). Gratifyingly the desired product 2-(phenyl)-1H-benzimidazole (3a, Fig. 1) was obtained with excellent selectivity. The possible byproduct 4a (Scheme 2) was not found in the post reaction mixture, which is the major concern of most of the reported methods. The moderate yield (68%) of the desired product 3a (entry 1, Table 1) led us to optimise the reaction using nonpolar, polar and protic solvents (entries 2–5), and the yield of the benzimidazole 3a was significantly improved (86%, entry 5) in THF medium, probably due to better stabilization of the intermediates in the transition state through coordination of THF as ligand. However we found a small amount of corresponding aldodimine, which was not converted into 3a. The catalyst loading was optimised to 8 mol% (entries 6–8). There was a little improvement in the yield using co-catalyst CeCl3 (10 mol%)9d,18 and urea–H2O2, but not with PhI(OAc)2 (ref. 19) (entries 8 and 9). The role of molecular oxygen as a stoichiometric oxidant in the oxidative cyclocondensation process (entries 10 and 11) was essential and it was verified by conducting two separate experiments, in the absence (entry 10) and presence of oxygen (entry 11). Our attempt to use simple MoO2(acac)2 as a catalyst (entry 12) was unsuccessful, which indicates that the presence of HL around the MoVI is essential for empowering it as a work-horse in the catalysis.
With this promising result in hand (entry 6, Table 1), the versatility of the benign synthetic approach involving the catalyst MoO2(HL)(H2O)DMF was successfully examined (Scheme 2) by the cyclocondensation cum oxidation of various functionalised aromatic aldehydes (2b–f) with different o-phenylenediamines (1a–c) at ambient temperature under similar reaction conditions to afford the corresponding benzimidazoles (3b–m) with a fast reaction rate (6–7 h) and high yield (74–86%). Electron-deficient, electron-rich and heteroaromatic substituents were tolerated in this catalytic process. Our aim was to develop a benign strategy for synthesising sensitive and labile molecules. This benign strategy was utilized successfully towards the direct synthesis of valuable glyceral- and glycal-based chiral benzimidazoles (3n–p).
Next, we explored the possibility of synthesising another ubiquitous framework, benzothiazole, using the powerful high-valent catalyst (Scheme 3). Gratifyingly the optimised reaction conditions (entry 6, Table 1) afforded the desired product 2-phenylbenzothiazole (6a, Scheme 3) from 2-aminothiophenol (5a) and benzaldehyde (2a). However, the yield (45%) and reaction time (24 h) were not encouraging due to oxidative dimerization of precursor 2-aminothiophenol (5a) through the formation of S–S bonds in the presence of molecular oxygen. After several attempts using various potential oxidants [PhI(OAc)2, PhIO, NMPO etc.], CeCl3 (10 mol%) was found to be an efficient co-catalyst for the oxidative cyclization process using non-aqueous urea–hydrogen peroxide as a stoichiometric oxidant. Herein the role of CeCl3 (ref. 20) is to supply oxygen from urea–hydrogen peroxide for the oxidative cyclization process. The heterocycle was obtained in high yield (83%) after 4 h at room temperature under an argon atmosphere. We have executed the reaction in the absence of the MoO2(HL)(H2O)DMF catalyst and the reaction was unsuccessful. Further, the dehydrative cyclisation cum oxidation process was carried out between 5a bearing an oxidation prone –SH group, and functionalised aromatic aldehydes (2b–i), which rapidly (4–5 h) furnished the respective benzothiazoles (6b–g) in excellent yields (82–89%). The sugar-based chiral benzothiazole 6h was successfully synthesised with high yield (82%). Most of the reported catalytic methods are incapable of direct access to the sugar-based chiral benzothiazole. The in situ generated urea was recovered and recycled through the formation of urea–hydrogen peroxide using hydrogen peroxide.
After investigating the applicability of the new high-valent MoO2(HL)(H2O)DMF catalyst, it was further examined for the synthesis of another frequently synthesised five member N-heterocycle isoxazoline generated via intermolecular 1,3-dipolar cycloaddition of nitrile oxides and functionalised olefins. The benzaldehyde aldoxime (7a) and ethyl acrylate (8a) were treated with the combo catalyst MoO2(HL)(H2O)DMF (8 mol%)–CeCl3 (10 mol%) at ambient temperature. To our delight the desired isoxazoline 9a (Scheme 4) was generated selectively without the formation of the corresponding other regioisomer 10a. However, the cycloaddition reaction was arrested in the absence of the MoO2(HL)(H2O)DMF catalyst. The versatility of the benign synthetic approach is verified with several precursors bearing electron-rich and electron-deficient aromatic substituents, which were tolerated in this benign approach to achieve isoxazolines 9b–l with excellent regioselectivity. The respective regioisomer (10) was not found using this powerful catalysis process, which is the major concern of the existing methods. The heterocycle based isoxazoline (9i) and sugar-based chiral isoxazolines (9l–o) were also achieved in good yields.
The versatility of this benign catalysis of high-valent MoVIO2(HL)(H2O)DMF was also verified in the asymmetric intramolecular cycloaddition reaction of nitrile oxide with alkynes for the synthesis of chiral fused-isoxazoles of allyl-substituted glycalaldoximes and pentose sugar analogues (11, Scheme 5). This synthetic tool with sugar-based nitrile oxides provides an excellent opportunity for constructing nature-like and unnatural organic molecules. Under the catalytic conditions, the chiral oximes (11a–d) underwent smoothly the reaction in a highly stereoselective fashion to produce sugar-based chiral isoxazoles (12a–d) in good yields (72–76%). The CeCl3 co-catalyst has the crucial role as an oxygen carrier from urea–H2O2 to the MoVI-catalyst for executing the oxidative intramolecular dipolar cycloaddition reactions.
Possible mechanisms of the diverse cyclocondensation cum oxidation and oxidative cycloaddition reactions are shown in Scheme 6. The first step for the oxidative cyclocondensation cum oxidation is expected to proceed through coordination of the high-valent catalyst with the substrates (I, cycle A, Scheme 6) and subsequent selective cyclisation to form intermediate II. It smoothly releases the desired products, benzimidazoles (3) and benzothiazoles (5), through reductive elimination of the MoV-complex (III), which is eventually regenerated to the MoVI-complex for the next catalytic cycle through oxidation with oxygen or CeCl3–urea·H2O2. The eliminating C–H of intermediate II for benzimidazole is relatively more acidic than that of benzothiazole due to the presence of two more electron withdrawing C–N bonds. Thus the molecular oxygen–MoVI-complex easily transforms II into benzimidazole (3). The highly regioselective synthesis of isoxazolines (9) and isoxazoles (12) also followed a similar catalytic cycle (cycle B, Scheme 6). Herein the role of CeCl3 is as an oxygen carrier to MoV–OH for regenerating the active MoVI-catalyst using H2O2–urea as a stoichiometric oxidant.20 However involvement of co-catalyst CeCl3 (IV → V) in the cyclisation step can’t be avoided. The high regioselectivity in all of the oxidative cyclisation processes can be explained in terms of the strong binding of the substrates and successive hetero- and homonuclear coupling through the large binding site of the powerful high-valent metal-complex during the formation of I → II (A) and IV → V (B).
In conclusion, we have discovered a high-valent MoO2(HL)(H2O)(DMF) complex with the rarely used H3L ligand and found it to be a powerful catalyst for oxidative cyclocondensation using oxygen as a stoichiometric oxidant to furnish 2-substituted benzimidazoles selectively. In contrast to the common epoxidation catalysis by MoVI-complexes, herein the MoO2(HL)(H2O)(DMF) complex has shown the diverse highly selective synthesis of ubiquitous 2-substituted benzimidazoles, 2-substituted benzothiazoles, isoxazolines, isoxazoles and their chiral analogues. Herein we demonstrated the MoVI(HL) as an attractive catalyst for developing benign, simple and general oxidative cyclization processes with excellent selectivity. We anticipate that the discovery of a new high-valent catalyst, its novel properties and robust oxidative cyclization catalysis will find considerable application in chemical science towards devising new strategies for functionalising molecules and discovering prospective new high-valent catalysts.
Footnote |
| † Electronic supplementary information (ESI) available: Experimental procedures, characterization data and NMR spectra. CCDC 1400488. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5ra21825j |
| This journal is © The Royal Society of Chemistry 2015 |