A solvent free approach for Knoevenagel condensation: facile synthesis of 3-cyano and 3-carbethoxycoumarins

1 Introduction

Coumarin constitutes an important class of oxygen heterocyclic compounds and occupies a special place in the field of natural and synthetic organic chemistry, owing to their potent and diverse biological activities [1]. A large number of natural products and synthetic analogues bearing the coumarin structural unit, display diverse and important biological activities [2–4]. 3-Cyano and 3-carbethoxycoumarins are the well-known starting materials for the synthesis of coumarin-3-carboxylic acid and coumarin-3-carboxamide, which have been used to synthesise modified cephalosporins [5], penicillins [6], and oxygen bridged tetrahydropyridones [7]. Cyano and carbethoxy functional groups are also useful for introducing units of biological interest such as triazole, thiadiazine and thiadiazole to coumarin nucleus for increasing biological activity of the compounds [8]. Beside synthetic applications, 3-cyanocoumarins have also been evaluated for their biological activities such as antimicrobial properties [9].

3-Cyanocoumarins are generally obtained by the reaction of 2-hydroxybenzaldehydes with malononitrile or ethylcyanoacetate. This reaction, first involves the condensation of the previously mentioned two reactants to give benzylidene derivatives, which undergo further cyclisation to give 3-cyanocoumarin. Various reagents have been used for this reaction, which includes aqueous alkali [10], aqueous K₂ CO₃ using phase transfer catalyst [11], pyridine [12], piperidine [13], etc. The toxic nature of piperidine or pyridine, longer reaction time, poor yields and use of hazardous organic solvents during the reaction, are the important limitations of this method. Moreover in some cases, a mixture of products is formed, as cyano groups normally undergo hydrolysis if an aqueous or alcoholic medium is used, thus giving a mixture of products which includes 3-carboxycoumarins.

Carbon-carbon bond formation is an important requirement for many organic syntheses [14] and Knoevenagel condensation is a key route for affecting this [15, 16]. This reaction is reported to be catalysed by various organic bases such as piperidine [17], pyridine [18], inorganic Lewis acids and bases such as ZnCl₂ [19], CaO [20] and ZnO [21], using either conventional heating or microwave irradiations [22]. The use of ultrasound [23] and ionic liquid [24] has also been made for this reaction.

Because of the public concern over environment degradation, many efforts are being made to carry out the reaction, under eco-friendly conditions by eliminating the use of hazardous chemicals, particularly solvents [25]. Nowadays the grinding technique has evolved as an important and cheap method to carry out the reactions under solvent free conditions with greater selectivity and higher yields [26–29].

In continuation of our work to develop solvent free routes for organic synthesis [30–32], using grinding technique, herein we report a facile synthesis of 3-cyano and 3-carbethoxycoumarins by making use of solvent free Knoevenagel condensation (Scheme 1).

Scheme 1

Synthesis of 3-cyano and 3-carbethoxycoumarin via solvent free Knoevenagel condensation.

2 Materials and methods

Grinding in all the reactions was carried out in a porcelain mortar with a pestle. Basic alumina of Brockmann activity-I and 150 mesh size from Aldrich was used. Melting points were taken in an open capillary, on an electrically heated metal block and are uncorrected. All the compounds were characterised from their spectral data (IR and ¹H-NMR) and compared with an authentic sample (CO-TLC and CO-IR). The ¹H NMR spectra were recorded on a Bruker Avance II 400 spectrometer at 400 MHz in CDCl₃ using TMS as an internal standard. Chemical shifts (δ) are reported in ppm and coupling constant in Hz. The IR spectra were recorded using a Perkin Elmer spectrometer.

3 Results and discussion

In an attempt to devise a solvent free approach for Knoevenagel condensation, a mixture of benzaldehyde and malononitrile was ground well for 5 min, in a porcelain mortar with a pestle using basic alumina as the support and catalyst. Reactants were found to have reacted completely after 5 min of grinding as confirmed by TLC. The yellow solid mass obtained after grinding was worked up to give a compound which was identified as benzylidenemalononitrile, based on its melting point (compared to a reference compound) and spectral data. Using this procedure, differently substituted benzylidenemalononitriles (3a–e) were obtained with 88 to 92% yield (Scheme 1). Encouraged by the results of this reaction, the reaction of 2-hydroxybenzaldehyde and malononitrile was taken up next, with a view to obtaining 3-cyanocoumarin. The reaction was performed in the presence of basic alumina as described earlier and the product obtained was identified as 3-cyano-2-imino-2H-1-benzopyran (4a). So instead of 3-cyanocoumarin, we obtained 3-cyano-2-imino-2H-1-benzopyran. The imino compound obtained was then hydrolysed with methanolic HCl to give 3-cyanocoumarin. In the next experiment, an attempt was made to prepare 3-cyanocoumarin from 2-hydroxybenzaldehyde and malononitrile in a single pot. Thus after grinding the mixture of 2-hydroxybenzaldehyde and malononitrile over basic alumina for 5 min, p-toluenesulphonicacid was added to the reaction mixture and was again ground for 5 min, and left at room temperature for another 10 min, when the initially formed 3-cyano-2-imino-2H-1-benzopyran was found to have disappeared from the reaction mixture as was confirmed by TLC. The reaction mixture finally gave 3-cyanocoumarin. It appears that the water molecule of crystallisation, present in the monohydrate of PTSA or the moisture absorbed by PTSA during the reaction was sufficient for the hydrolysis of 3-cyano-2-imino-2H-1-benzopyran into 3-cyanocoumarin. A reaction between 2-hydroxybenzaldehyde and ethylcyanoacetate, using basic alumina as support, under similar conditions, produced 3-carbethoxycoumarin. It is clear that it is the –CN functional group not –COOC₂ H₅ functional group of ethylcyanoacetate which participates in cyclisation. So it may be the case of trans-selectivity of the Knoevenagel reaction which leads to (E)-ethylbenzylidenecyanoacetate as the major product. In the case of ethylcyanoacetate, if after grinding the reaction mixture with PTSA, it was kept at room temperature for 1 h, the 3-carboethoxycoumarin began converting to 3-carboxycoumarin. However, complete conversion could not take place even after keeping the reaction mixture for 1 day. Reaction conditions were then optimised with respect to the amount of basic alumina used in step one and PTSA in step two, by taking the model reaction of 2-hydroxybenzaldehyde with malononitrile (Table 1). 3-Cyano-2-imino-2H-1-benzopyran was obtained with a maximum yield in step one, when 5 g of basic alumina was used and, in step two, the use of 1 g–1.5 g of PTSA gave a maximum yield of 3-cyanocoumarin. Thus by making use of these optimised conditions, using this procedure, various substituted 3-cyano and 3-carbethoxycoumarins (5a–j) were successfully synthesised (82–90% yield) under solvent free conditions, directly reacting 2-hydroxybenzaldehyde with malononitrile or ethylcyanoacetate in one pot without isolating the imino compounds (Table 2).

4 Conclusion

In conclusion it can be stated that the present approach for Knoevenagel condensation provides a very simple, solvent free and convenient synthesis of benzylidenemalononitriles in high yields by making use of grinding technique at room temperature. This approach can also be used for the synthesis of 3-cyano and 3-carbethoxycoumarins in a single pot under solid phase conditions.