New methods are described for the preparation of 5,6‐dihydro‐11‐morphanthridone, 5‐substituted 5,6‐dihydro‐11‐morphanthridones and nuclear halogenated derivatives. Basic Grignard reagents convert these ketones to carbinols, which through dehydration affords a convenient route to straight chain and cyclic 11‐aminoalkylidene derivatives.

Because of our interest in the preparation of basic substituted morphanthridines (1,2 ) , we have investigated the Bischler-Napieralski cyclization of 2-acylaminobenzophenones in an effort to obtain the useful intermediates, 11morphanthridones. This cyclization method was applied quite successfully for the preparation o€ 6substituted morphanthridines from 2acylaminodiphenylmethanes (3,4), a s well a s for the synthesis of related oxazepines and thiazepines (4, 5).However, when we treated 2formylamino-5chlorobenzophenone (la) with hot polyphosphoric acid, only 2 -amino -5chlorobenzophenone was isolated. This is in agreement with the reported recovery of 2-aminodiphenylmethane from the attempted polyphosphoric acid cyclization of its 2fonnylamino analog, a s reported by Jilek, et al. (3). Heating the 2acetylaminoketone (Ib) with polyphosphoric acid resulted in the formation in high yield of a bright yellow crystalline solid, Neither the analytical nor the spectral data of the product were in agreement with those for the expected 11-morphanthridone (11). The elemental assay indicated an empirical formula C28H&12N20, while its infrared spectrum was notably lacking in extinctions in the 3.3 to 3.6 p region, indicating the absence of methyl-and methylene groups. This was confirmed by its nuclear magnetic resonance spectrum.Furthermore, the compound showed a carbonyl absorption a t 6.12 p , which is in the same position as a carbonyl absorption of its precursor Ib, but which had only approximately half its intensity. Comparison of the ultraviolet spectra of the compound in alcohol and in alcohol containing 0.1 N hydrochloric acid, showed a shift to shorter wave-length in the latter case (Table I). All this evidence suggested that the cyclization product was the benzoylanilinoquinoline 111. A possible reaction mechanism involves acid catalyzed aldol cyclization concomitant with anil formation with the simultaneously formed benzoylaniline (6).Structure proof of 111 a s well a s added support for the proposed mechanism was obtained by an alternate synthesis. Base catalyzed cyclization of the acetylaminobenzophenone Ib according to the method of Hauser (7) afforded 6-chloro-4-phenylcarbostyril (IV) (a), which when treated with phosphorus oxychloride (9) yielded 2,6-dichloro-Pphenylquinoline (V). This was then made to react with 2-amino -5chlorobenzophenone in order to afford III. Comparison of the infrared spectrum of m with that of the cyclization product, as well a s lack of depression of their mixture melting points, proved the compounds to be identical.A s expected, similar cyclization of Z-propionylamino-5-chlorobenzophenone afforded the 3-methyl analog of III. TABLE I Ultraviolet Spectra, x max. IU/.I (c x I$) Ia (EtOH) III (EtOH) III (EtOH + 0.1 N HC1) 205-6 (24,5) 222-3 (47.3) 251 (30,6) 238 (24,36) 257-8 (34,8) 368 (9.5) 329 (1.97) 285 (30.6) 343 (5,12) 3 9 2 4 (9,O) M. I a. R = H 0. R = M e I 111 r I b -CI -a CI qc'-III Ph m I V V

No abstract

Butyl lithium converts 5‐substituted‐5,6‐dihydromorphanthridines to the 11‐position anion. Treatment of this lithium anion with aminoalkyl halides affords the 11‐aminoalkylated derivatives. Catalytic hydrogenation causes reductive cleavage of benzylamino bridge of the morphanthridine ring, a reaction which is used to structurally relate these compounds to the known 11‐aminoalkylidene‐5, 6‐dihydromorphanthridines

No abstract