Catalytic dehydrative etherification and chlorination of benzyl alcohols in ionic liquidsw

Dehydrative reactions can be used for a wide variety of transformations including etherifications, esterifications and thioetherifications. Generally, this route is more environmentally friendly than classical methods (Williamson and Ullmann) as the only byproduct is water, whereas the use of strongly basic alkoxides or phenoxides with alkyl or aryl halides in the Williamson method produces considerable amounts of salt byproducts. In etherification reactions, a wide variety of catalysts have been employed including alumina, phase transfer catalysts, Lewis acids, and several transition metal complexes. Among these catalysts, Pd(II) compounds have a special position, which has led to a recent review in this area. It should be noted that dehydrative etherification of benzylic alcohols has been investigated to a lesser extent than phenols, however, such catalytic reactions could be useful in synthetic chemistry, e.g. for protecting alcohols. In the area of catalytic dehydration reactions, we thought that the use of hydrophobic ionic liquids (ILs) as reaction media could provide an added benefit to homogeneous catalytic systems by aiding in water abstraction from the reaction mixture. Furthermore, we have employed microwave heating to reduce reaction times and energy consumption. In general, room temperature ILs have attracted great attention during the last decade, since they are non-volatile, non-flammable, potentially recyclable and benign solvents with the ability to dissolve a wide variety of compounds. In terms of ether synthesis, Ullmann andWilliamson methods have been performed using ILs as the reaction media, as well as the promoter. As far as we are aware, catalytic dehydrative etherification reactions have not been performed in ILs previously. In our preliminary studies, benzyl alcohol was selected as it is the simplest benzylic alcohol, Table 1. Conventional heating as well as microwave irradiation was used as the energy source. Scheme 1 presents the general reaction conditions and the benzyl alcohol derivatives used in further studies, Table 2. Two hydrophilic and three hydrophobic ILs were used along with several transition metal sources (Table 1, entries 1–15). 1-Butyl-3-methyl imidazolium hexafluorophosphate ([BMIm]PF6) gave the highest yields under microwave irradiation in combination with Pd(CH3CN)2Cl2 (Table 1, entry 1). It should be noted that other environmentally-friendly solvent systems—solvent-free (Table 1, entry 5) and aqueous (Table 1, entry 6)—were not suitable for this reaction. Additionally, reactions in conventional polar, aprotic organic solvents (Table 1, entries 7 and 8) only gave low yields of the desired product. The increased yields obtained in ILs may be due to the unique highly ionic environments afforded by these media that can support ionic intermediates that have been proposed for similar Pd-catalysed reactions in conventional reaction media. As many ether syntheses are pH dependent, the reactions were examined in the presence of several additives. Under basic conditions (Table 1, entries 16 and 17), no conversion was seen. Conducting the reaction in the presence of acetic acid led to the formation of benzyl acetate, 3 (Table 1, entry 18), while benzyl chloride, 2, was obtained in excellent yield (84%) in the presence of NH4Cl (Table 1, entry 19). NH4Cl is less hazardous to human health and the environment than other common chlorinating agents including thionyl chloride and concentrated hydrochloric acid. Further studies are ongoing to determine the applicability of this environmentally benign halogenation process. Control reactions have also been performed. In the absence of a transition metal, with and without additives, no conversion of the benzyl alcohol was seen. This indicates that although the hexafluorophosphate anion has a tendency to degrade and can potentially form HF in the reaction mixtures, the reactions are not acid catalysed and the metal, ideally palladium, is an essential component of the catalytic system. The degradation of [BMIm]PF6 was evident through the formation of a precipitate in many of the reactions especially upon prolonged heating. The precipitate contained significant amounts of Na and Si which could only have originated from the glass reaction vessels due to their sodium borosilicate content. Furthermore, mild etching of the vessels was seen after some reactions, presumably through in situ action of HF from PF6 degradation.z Based on our initial studies, reactions using other benzyl alcohol derivatives (Table 2) and ionic liquids were performed under similar conditions. When 4-methoxy benzyl alcohol was used as the substrate (Table 2, entry 4), the highest yield was obtained. We propose that this is due to the higher stability of the benzylic carbocation formed from this reagent. All other alcohols, with the exception of 4-nitro benzyl alcohol, were successfully coupled and yields of 50–60% obtained. Furthermore, it should be noted that no precipitate formed during the transformation of 4-methoxy benzyl alcohol. Whereas for Chemistry Department, Memorial University of Newfoundland, St. John’s, Newfoundland, Canada A1B 3X7. E-mail: [email protected]; Fax: +1 709 737-3702; Tel: +1 709 737-8089 w Electronic supplementary information (ESI) available: Instrumentation and chemicals, EDX analysis and SEM image of the precipitate from some reactions, GC-MS analysis, H, F and P-NMR data. See DOI: 10.1039/b909866f