C«H.O» Acetic anhydride

It was found inadvisable to use more than four molecules of form-amide [ (47) when R = H] per molecule of anthranilic acid and the condensation produces best results when the mixture is heated at 120 -130°C for 2 hr followed by further heating at 170°-180 C for 2 hr. Other variants of this reaction involve the use of ammonium o-acylaminobenzoates, anthranilic acid in the presence of nitriles and acetic anhydride, o-acetamidonitrile with acetic anhydride or hydrogen peroxide, anthranilic esters and aliphatic or aromatic amides or amidines, isatoic anhydride with amides or amidines, and anthranilic esters with aryl iminochlorides in acetoned The mechanism proposed by Bogert and Gotthelf has had experimental supporR and is represented in Scheme 12. 

N,N -Diacetylkasuganobiosamine. To a suspension of kasuganobiosamine (500 mg., 1.6 mmoles) in 10 ml. of methanol was added 2.5 ml. of acetic anhydride. The mixture became a colorless transparent solution after allowing to stand at room temperature for 5 hours. After adding 20 ml. of water, the reaction mixture was allowed to stand at room temperature for 2 hours. An oily material was obtained after removal of the solvent, and dissolved in 50 ml. of water to adjust to pH 5.0 with dilute sodium hydroxide. The solution was placed and passed on a column (1.5 x 20 cm.) of Amberlite CG-50 (H form). Development of the chromatogram with water afforded fractions positive to Lemieux reaction. A colorless powder (620 mg.) was obtained after removal of the solvent. The powder was dissolved in methanol (15 ml.) to remove insoluble material. To the filtrate was added ether until no precipitate was produced. This procedure was repeated twice, affording a colorless material, m.p. 143°-150°C., [ ]D20 +113° (c=l.l, H20) which was identified to be 2V,N -diacetyl derivative of 4. The yield was 313 mg. (0.79 mmole). Anal. Calcd. for C16H2809N2 C, 48.97 H, 7.19 O, 36.70 N, 7.14. Found C, 48.93 H, 7.46 O, 37.10 N, 6.92. 

Acetylation of Kasuganobiosamine (4). Acetic anhydride (2 ml.) and pyridine (4 ml.) were added to kasuganobiosamine (212 mg., 0.688 mmole) and the solution was allowed to stand at room temperature for 24 hours. The reaction mixture was treated with ice-water and extracted three times with chloroform. The organic layer was dried over sodium sulfate. After removal of the solvent, an oily material was treated with 30 ml. of ether to deposit a crystalline material. It was further recrystallized from ether-ethyl alcohol to afford a needle crystalline material (409 mg., 0.679 mmole), m.p. 242°-243°C., and identified to be heptaacetate of 4. Anal Calcd. for C26H38014N2 C, 51.58 H, 6.44 O, 36.69 N, 4.41. Found C, 51.27 H, 6.44 O, 36.42 N, 4.53. 

The first examples of a homogeneous reduction of this type were reported in 1971. Cobalt carbonyl was found to reduce anhydrides such as acetic anhydride, succinic anhydride and propionic anhydride to mixtures of aldehydes and acids. However, scant experimental details were recorded [94]. In 1975, Lyons reported that [Ru(PPh3)3Cl2] catalyzes the reduction of succinic and phthalic anhydrides to the lactones y-bulyrolaclone and phthalide, respectively [95], The proposed reaction sequence for phthalic anhydride is shown in Scheme 15.15. Conversion of phthalic anhydride was complete in 21 h at 90 °C, but yielded an equal mixture of the lactone, phthalide (TON = 100 TOF 5) and o-phthalic acid, which is presumably formed by hydrolysis of the anhydride by water during lactoniza-tion. Neither acid or lactone were further hydrogenated to any extent using this catalyst system, under these conditions. 

Sardesai and Sunthankar studied the cyclization of diethyl )V-(2-amino-phenyl)aminomethylenemalonate (162, R = H) (57MI2 59MI1). No cyclization occurred in refluxing xylene in the presence or absence of a catalyst (p-toluenesulfonic acid or sodium hydroxide), or in acetic anhydride, or in a mixture of acetic anhydride and concentrated sulfuric acid. Benzimidazole and benzimidazolone were obtained in 20% and 66% yields, respectively, when 162 (R = H) was distilled in vacuo. Benzimidazolone was the product when 162 (R = H) was heated in boiling diphenyl ether, o-Phenylenediamine was reacted with diethyl acetylmalonate at 140°C for 4 hr to give 2-methylbenzimidazole and diethyl malonate (85S555). 

When the 3-thiourea derivative (59) was heated in boiling ethanol for 3 h, and then the evaporated reaction mixture was treated with 10% NaOH solution at 100°C for 20 min, anhydro 2-methyl-3-mercapto-4-hydroxy-5,6,7,8-tetrahydro[l,2-6]pyridazinium hydroxide (61) was obtained (71CPB159). The mercapto group was alkylated with benzyl bromide and was treated with HgCla in boiling ethanol to yield the 3-chloromercurithio derivative. Anhydro 3,4-dihydroxy-2-methyl-5,6,7,8-tetrahydropyrido[l,2-f ]pyridazinium hydroxide (62) was O-acylated with acetic anhydride, but the structure of the product was not elucidated (71CPB159). 

A further synthesis of thiazolo[3,2-6][l,2,4]triazoles starts with 2,3-diaminothiazoles. Treatment of 2-aminothiazole with O-mesitylensulfonyl-hydroxylamine results in the formation of salt 171, which on heating with acetic anhydride (140-150°C) gives a mixture of 172 and 173. 2-Phenylthia-zolo[3,2-h]-5-triazole could be obtained from 171 and benzoylchloride ( 200°C) in 68% yield. When 2-acylamino-3-aminothiazolium salts 174 are heated 20-30°C above their melting point for 1 h the corresponding thiazolo[3,2-h]-5-triazoles were obtained. Much better results were achieved by using PPA (100-110°C, 1.5 h 97% yield) (73JHC947). 

There is an interesting difference in properties between fluoro compounds and the corresponding hydrogen compounds that can be explained by the assumption of the formation of C—H---X bonds. For example, trifluoroacetyl chloride, F CC0C1, has a boiling point below 0°C, whereas that of acetyl chloride is 51°C similarly, trifluoroacetic acid anhydride, (FjCCO O, boils at 20°C and acetic acid anhydride at 137°C. 

---