Transition metal catalysts,such as palladium,ruthenium,and rhodium have shown powerful catalytic ability in the formation of C(sp² )-C(sp²) bonds ^[1]. However,their relatively high price and toxicity have limited their applications. Therefore,chemists are increasingly turning their attention to cheaper transition metals,such as copper and iron ^[2]. Iron,being one of the most abundant metals on earth,offers several advantages. It is relatively inexpensive,readily available,and environmentally friendly ^[3]. To date,chemists have reported various iron-catalyzed organic transformations ^[4],including nucleophilic addition,substitution,reduction,oxidation,hydrogenation,cycloaddition,isomerization,rearrangement,as well as direct C-H transformations. Ironcatalyzed direct C-H activation has attracted continuous interests because of its accessibility,activity,and selectivity ^[5].

Direct acylation of aromatic rings is an important transformation in organic synthesis. The classic syntheticmethod for acylation is the well-known Friedel-Crafts reaction of carboxylic acid derivatives. Another method is the transition-metal-catalyzed C-H activation of aldehydes,carbonyl acids,anda-dicarbonyl compounds ^[6]. Inrecent years,the area of cheap transition metal-catalyzed direct acylation of aldehyde involving the sp²-hybridized C has been actively explored. Wang et al. reported a copper-catalyzed ortho-acylation of phenols with aryl aldehydes; a one-step reaction between 2-substituted aryl aldehydes and phenols produced xanthones in high yields [7a] (Scheme 1). Meanwhile,Lei et al. described a copper-catalyzed oxidative coupling of alkenes with aldehydes [7b],and Studer et al. reported a FeCp²-catalyzed preparation of fluorenones and xanthones from ortho-formyl biphenyls and ortho-formyl biphenylethers. In addition,the synthesis of 6-aroylated phenanthridines from readily available 2-isocyano-biphenyls and aromatic aldehydes has also been investigated ^[8].

Our group is actively engaged in the development of cascade synthesis of heterocycles and has developed a number of tandem reactions to synthesize different heterocycles ^[9]. We have previously reported a copper-catalyzed cascade synthesis of azoquinazolinones from substituted 2-halobenzamides and different N-heterocycles [9e]. However,attempts to synthesize polycyclic- fused azo[l,2-a]indol-ones from 2-bromobenzaldehyde and 1H-benzo[d]imidazole using a similar strategy were unsuccessful. Even though Rosevear and Wilshire have previously reported the reaction between azoles and o-fluorobenzaldehydes without metal catalysts,more substituted products were obtained than cyclic ketones ^[10]. As iron salts could catalyze both C-H activation/direct acylation of aldehyde and C-N bond formation,we seek to apply iron-based catalysis in the synthesis of azo[l,2-a]indol-ones. Herein,we present a highly efficient approach for the synthesis of polycyclic-fused azo[1,2-a]indolones via an iron-catalyzed tandem reaction of azoles with 2-fluoroaldehydes (Scheme 1) involving an intermolecular S_NAr and an intramolecular direct acylation reaction.

Scheme 1

	Download: JPG larger image
Scheme 1.One-pot synthesis of xanthone and 11H-indolo[l,2-a]benzimidazol-11- one.

All reagents and solvents were pure analytical grade materials purchased from commercial sources and used without further purification,unless stated otherwise. All melting points are uncorrected. The NMR spectra were recorded in CDCl₃ on a 500 MHz instrument with TMS as the internal standard. TLC was performed using 0.2 mm thick silica gel plates (GF254). Visualization of the TLC plates was accomplished under UV light. The columns were hand-packed with silica gel 60 (200-300). All reactions were carried out in an oven-dried Schlenk tube equipped with a magnetic stir bar under a N₂ atmosphere. All products were confirmed by ¹ H NMR and ¹³C NMR. Unknown compounds were confirmed by high-resolution mass spectroscopy (HR-MS) analysis. HR-MS spectra were recorded using an electrospray quadrupole time-of-flight (ESI-Q-TOF) mass spectrometer.

FeCl₃ (8 mg,0.05 mmol,10 mol%),K₃PO₄ (212 mg,1 mmol),Nheterocycles 1 (0.50 mmol),and o-fluorobenzaldehydes 2 (0.50 mmol) were added to an oven-dried Schlenk tube (with a magnetic stirrer bar within) equipped with a Teflon valve. The tube was repeatedly evacuated and back-filled with N₂ (3 times). Under a counter flow of N₂,DMF (2.0 mL) was added by a syringe,and the mixture was stirred for about 16 h at 130 ℃. When the reaction was complete,the mixture was extracted with ethyl acetate (3 × 15 mL),and the organic layer was washed with brine (3 × 10 mL) and dried over anhydrous Na₂SO₄. Subsequently,the solvent was removed and the product was purified by column chromatography on silica gel (PE:EA = 5:1-1:1,v/v) to give the desired product 3.

During the preliminary experiments,the reaction between benzimidazole 1a and o-fluorobenzaldehyde 2a was chosen as a model reaction to optimize the reaction conditions,including the screening of catalyst,base,and solvent. The results are summarized in Table 1. Firstly,the reactants with 10 mol% FeCl₃ and anhydrous K₂CO₃ in DMSO (at 130 ℃ under N₂) afforded 11Hindolo[l,2-a]benzimidazole-11-one 3a in 61% yield (Table 1,entry 1). When various solvents were tested (Table 1,entries 1-5),DMF gave an improved yield of 73% whereas only 40% of 3a was obtained in toluene (Table 1,entries 2 and 4). Other solvents,such as NMP and 1,4-dioxane,were incompatible with the reaction (Table 1,entries 3 and 5). Stronger bases (NaOEt,Cs₂CO₃,and t-BuOK) did not work well in the reaction,giving ＜50% yields of 3a (Table 1,entries 6-8); and the organic base DBU was also ineffective,giving only trace amount of the product (Table 1,entry 10). To our delight,the yield of 3a increased to 94% when K₃PO₄ was used as a base (Table 1,entry 9). The effect of different iron sources on product formation was also investigated. FeCl₃-6H₂O gave no product,suggesting that water is detrimental to the reaction (Table 1,entry 11). In contrast,with 10 mol% Fe2(SO₄)₃ as the catalyst,3a was obtained in 84% yield (Table 1,entry 12). Neither Fe nor FeCl₂ was suitable for the reaction,as they afforded only 37% and 34% isolated yields,respectively (Table 1,entries 13 and 15). In the absence of an iron source,the product was obtained in 30% yield (Table 1,entry 14),which was close to Rosevear’s report ^[10]. Therefore,FeCl₃ was shown to have good catalytic activity in facilitating the synthesis of 11H-indolo[l,2- a]benzimidazole-11-one. Four different Lewis acids,including CuCl₂,ZnCl₂,InCl₃,and La(OTf)₃,were also studied for their catalytic potential (Table 1,entries 17-20). However,these Lewis acids were less efficient compared to FeCl₃. When the reaction was carried out under ambient air,a low product yield was obtained (Table 1,entry 21).

Table 1

Table 1 Optimization of the reaction conditions for the synthesis of 11H-indolo[l,2- a]benzimidazole-ll-one.a

Table 1 Optimization of the reaction conditions for the synthesis of 11H-indolo[l,2- a]benzimidazole-ll-one.a

We then investigated the scope of the iron-catalyzed tandem reaction of substituted o-fluorobenzaldehyde (2a-2e) and different N-heterocyclic compounds (1a-1e) for the synthesis of azo[l,2- a]indol-ones under the optimized conditions aforementioned (10 mol% FeCl₃,anhydrous K₃PO₄ in DMF; under N₂ at 130 ℃).

As shown in Table 2,most 11H-indolo[l,2-a]benzimidazol-11-ones were obtained in higher yields compared to 9H-imidazo[l,2- a]indol-9-ones and 11H-indolo[l,2-b]indazol-11-ones. In general,the electron-donating groups (e.g. -OCH₃) ino-fluorobenzaldehydes negatively affected the reaction,resulting in lower product yields,when compared to compounds bearing the electron-withdrawing groups (e.g. -Cl and -Br) (Table 2,entries 2-5,9-12,and 16-17). However,the yield of CH₃O-substituted azo[l,2-a]indolones increased when K₂CO₃ was used as a base (Table 2,entry 12). The position of the substituents also affected product yield. o-Fluorobenzaldehyde bearing the -Cl group at the para position showed higher activity compared to 5-chloro-2-fluorobenzaldehyde (Table 2,entries 3-4 and 10-11). The use of 5,6-dimethylbenzimidazole and 4-methyl-1H-imidazole as substrates gave lower product yields (Table 2,entries 6-7 and 13-14). Other N-heterocyclic compounds,such as 1,2,4-triazole,indazole,and pyrazole,were also investigated. The indazole 1e reacted with o-fluorobenzaldehydes to afford 11H-indolo[l,2-b]indazol-11-ones in moderate yields (Table 2,entries 15-16). However,other Nheterocyclic compounds did not give their corresponding products.

Table 2

Table 2 FeCl3-catalyzed tandem reaction of substituted o-fluorobenzaldehydes and N-heterocycles.a

Table 2 FeCl3-catalyzed tandem reaction of substituted o-fluorobenzaldehydes and N-heterocycles.a

The proposed pathway for the formation of 11H-indolo[l,2- a]benzimidazol-11-ones invovles two steps (Scheme 2). The reaction might first proceed through an intermolecular S_NAr reaction between benzimidazole 1 and o-fluorobenzaldehyde 2 (step a),consequently forming the intermediate 2-(benzimidazol- 10-y1)benzaldehyde 4,which has been isolated and confirmed by spectral data. Subsequently,the conversion of the intermediate 4 to product 3a was studied (Scheme 3). While the desired compound 3a was readily obtained under the optimum conditions,it was found that the product yield was positively correlated to the reaction time. However,3a was only obtained in 27% yield without the addition of FeCl₃. Hence,it was concluded that FeCl₃ facilitated the intramolecular cyclization process to give the target product 3a in step b (Scheme 2). According to the literature ^{[10, 11]},a possible mechanism for the cyclization of intermediate 4 to give 11Hindolo[l,2-a]benzimidazol-11-ones is shown in Scheme 4. The mechanism proposed a base- and iron-initiated C-2 (benzimidazole) nucleophilic addition of the aldehyde followed by the formation of the hydroxyl intermediate (Ⅰ). In the presence of a base,intermediate (Ⅰ) can be isomerized to a carbonyl intermediate (Ⅱ). Then it is aromatized by trace air at 130 ℃ to generate the target product. Although the intramolecular cyclization process could proceed without FeCl₃,the product was obtained in lower yield. Therefore,FeCl₃ could potentially function as a Lewis acid to increase the electrophilicity of the aldehyde,leading to a higher yield of the cyclic products.

Scheme 2

	Download: JPG larger image
Scheme 2.The proposed reaction pathway for the one-pot synthesis of 11Hindolo[l,2-a]benzimidazol-11-one.

Scheme 3

	Download: JPG larger image
Scheme 3.The reaction of the intermediate 4 under different conditions.

Scheme 4

	Download: JPG larger image
Scheme 4.A possible mechanism for the cyclization of 11H-indolo[l,2- a]benzimidazol-11-one.

In summary,we have developed an iron-catalyzed cascade reaction between azoles and 2-fluoroaldehydes to synthesize azo[l,2-a]indol-ones via an S_NAr/acylation. This method circumvents the use of expensive catalytic agents,giving access to the desired polycyclic-fused azo[l,2-a]indolone compounds in good yields,and could be a valuable tool in the synthesis of heterocyclic molecules with biological and medicinal activities.

This work was financially supported by the Natural Science Foundation of China (Nos. 21302135 and 21272169),and Taizhou Science & Technology program (No. 111ZD02).