Cyclisation and Debenzylatim of m-BenzyloxyphenylpopionicAcid as Competitive Reactions.By M. DONBROW.DURING an investigation into ring-closures yielding 5-alkyloxyindan-1-ones attempts were made to cyclise m-benzyloxyphenylpropionic acid by Friedel-Crafts methods. Aluminium chloride invariably caused debenzylation but significant amounts (1O-15%) of 5-benzyloxyindan-1-one survived when stannic chloride dissolved in benzene was used and the main product, 5-hydroxyindan-l-one, was relatively pure and in good yield (40-50%).The 5-benzyloxyindan-l-one was identical with a sample prepared by direct benzylation of 5-hydroxyindan-1-one. The partial retention of the benzyloxy-group and the clean debenzylation were considered to warrant the use of this system to study cyclisation and debenzylation as competitive reactions.The reactions were followed by quantitative analysis of the products, the experiments being designed to test variables such as concentration of reactants, excess of reagents, time, temperature, and solvent composition. Separation of the product into acidic, phenolic ketonic, phenolic non-ketonic, neutral ketonic, neutral non-ketonic, and ether-insoluble polymeric fractions revealed considerable complexity. The neutral fraction contained, in addition to the required ketone, traces of higher-melting ketones, diphenylmethane, and occasionally benzyl chloride. The phenolic ketone, mainly 5-hydroxyindan-l-one, contained other ketones, one of which was probably a benzyl-substituted hydroxyindan-1one C,,H,,02.Traces of 7-hydroxyindan-1-one derivatives were also detected. A violet complex always separated during the reaction and since a similar complex was precipitated during the cyclisation of the 5-methoxy-analogue (which was not however demethylated) , the violet compound was presumed to be a ketone-catalyst complex and not an ethercatalyst compound. Retention of the benzyl group was associated with the use of a large excess of stannic chloride and the rapid separation of the violet complex, though a connection between the two could not be proved. The use of less than 1-5 mol. of stannic chloride resulted in a substantial recovery of acid, even at high temperatures. The course of the
