1). Synthesis of Dibenzcycloheptadienone. 2). Stereochemistry of Two Metabolic Diols

Part I. An attempt has been made to synthesise derivatives of dibenzoyoloheptadiene by appropriate transformation of the two carboxylic acid groups of diphenic acid (I; R=CO2H) By an adaptation of the McFadyen-Stevens process(14) the aldehydo acid (I; R = CHO) has been prepared and hence via the derived acrylic (I; R = CH = CH.CO2H) and propionic (I; R = CH2.CH2.CO2H) acids cyclisation to dibenzcyclo-heptadienone (II) has been demonstrated. The yields, however, are not good and attempted improvement by desul-phftrisation. of the p-tolyl thiol ester (I; R = CO.S.C6H4.CH3 (p)) afforded only the alcohol (I; R = CH2OH). An alternative method, involving the synthesis of (III; R = CO2H) and the projected expansion of the carboxylio to the propionic acid side chain, failed by reason of the unreactive nature of the ester (III; R = CO2Me) and the facility with which (III; R = CO2H) yielded the diphenyl-methylolid (IV) Part II. As a contribution to knowledge of the stereochemical configuration of the 1:2-dihydroxy-1:2-dihydro-naphthalene and - anthracene, metabolites of the respective aromatic hydrocarbons isolated by Young(8) and by Boyland(1) from rat and rabbit urine, the cis- and trans- forms of the diols (V) and (VI) have been synthesised. In each case the configuration has been assigned on the basis of the general criteria established by Criegee. In the case of (V) optical resolution of the transcompound into its d and 1l- forms shows that the former is identical with the dihydride obtained by Young through catalytic hydrogenation of the naphthalene metabolite. In the case of (VI) the physical properties of the synthetic dl-trans- compound indicate its identity with the dihydride of the optimally inactive anthracene metabolite described by Boyland. In conjunction therefore with the fact that the 9:10-dihydroxy-9:10-dihydrophenanthrene, the corresponding metabolite of phenanthrene from rats, is said to be optically active(34) and must accordingly be the trans isomer, it appears that the metabolic process of hydroxylation leads to the trans- and not to the cis- configuration