THE ACTION OP PYRIDINE ON DULCITOL HEXANITRATE by GEORGE GORDON MCKEQWN A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE In the Department of CHEMISTRY We accept this thesis as conforming to the standard required from candidates for the degree of MASTER OF SCIENCE. Members of the Department of Chemistry THE UNIVERSITY OP BRITISH COLUMBIA September, 1952. ABSTRACT Dulcitol hexanitrate was prepared In 90% yield by nitrating dulcitol with nitric and sulphuric acid. The optically inactive hexanitrate crystallized from ethanol and water to give fine, colorless needles melting at 98-99°C. When the pure dulcitol hexanitrate was dissolved in anhydrous pyridine, an exothermic reaction occurred with the evolution of a gas. On dilution of the solution with water, a crystalline dulcitol pentanitrate separated in 65% yield. The pentanitrate was re- crystallized from ethanol and water as fine, color• less needles, which melted at 85-86°C; it was optically inactive and did not reduce Fehling's solution. The pentanitrate was methylated with silver oxide and methyl iodide, and then reduced with hydrogen over palladized charcoal to give an optically inactive monomethyl dulcitol. This product crystallized from ethanol as heavy, colorless crystals which melted at 149-150°C. Periodate oxidation of the monomethyl dulcitol indicated that it was a racemic mixture of 3- and 4-methyl D-dulcitol (D-galactitol)• Hence the action of pyridine selectively removed a nitric acid ester group from the 3-(or chemically equivalent 4-) position in the D-dulcitol hexanitrate molecule leaving a hydroxyl group. ACKNOWLEDGEMENTS The writer wishes to express his sincere thanks to Dr. L.D. Hayward for his ready encouragement and sound advice in the direction of this investigation. Grateful acknowledgements are also made to the Standard Oil Company of B.C. for the award of a fellowship, to the National Research Council of Canada for a summer grant, and to the Department of Veterans Affairs. TABLE OP CONTENTS _____ GENERAL INTRODUCTION 1 HISTORICAL INTRODUCTION 2 DISCUSSION OP RESULTS 9 A. Dulcitol Hexanitrate 9 B. Dulcitol Pentanitrate 9 C. Monomethyl Dulcitol Pentanitrate 12 D. Monomethyl Dulcitol 13 EXPERIMENTAL 19 Special Precautions 19 A. Materials 19 Nitric Acid 19 Pyridine 19 Dulcitol 19 Palladized Charcoal Catalyst 19 Sodium Metaperiodate 20 Diphenyiamine Reagent 20 Silver Oxide 20 B. Analytical Methods 21 Nitrogen 21 Methoxyl 21 Oxidations with Periodate 21 C. Dulcitol Hexanitrate 23 Preparation of Dulcitol Hexanitrate 23 Hydrogenolysis of Dulcitol Hexanitrate 24 TABLE OF CONTENTS (continued) Page D. Dulcitol Pentanitrate 24 Denitration of Dulcitol Hexanitrate 24 Hydrogenolysis of Dulcitol Pentanitrate 26 Methylation of Dulcitol Pentanitrate 26 E. Monomethyl Dulcitol 27 Hydrogenolysis of Methyl Dulcitol Pentanitrate 27 Nitration of Monomethyl Dulcitol 28 Periodate Oxidation of Monomethyl Dulcitol 29 BIBLIOGRAPHY 31 GENERAL INTRODUCTION It has long been known that D-mannitol hexanitrate is unstable to pyridine and other bases. Prom the action of pyridine on D-mannitol hexanitrate, a product had been isolated in good yield which analysis had indicated to be a D-mannitol pentanitrate. Dulcitol hexanitrate had also been treated with pyridine to yield a product believed to be a dulcitol pentanitrate. In 1950, L.D. Hayward undertook to determine the position of the unstable nitrate group in the D-mannitol compound. He showed conclusively that a nitrate group was selectively removed from the 4-position (or 3- which is Identical chemically and optically) producing 1,2,3,5,6, D-mannitol pentanitrate. The present work is a similar study on the op• tically inactive dulcitol compound (dulcitol has a meso structure and may be designated as D- or L- dulcitol or galactitol). One might anticipate that the chemically equivalent 3- and 4-positions would be attacked. However, if such was the case, the two products would not be identical, but would be optical isomers. That is, the product would be a racemic mixture of 1,2,3,5,6, D-dulcitol pentanitrate and 1,2,3,5,6, L-dulcitol pentanitrate. HISTORICAL INTRODUCTION In recent years, considerable study has been made orf the selective removal of nitric acid ester groups from carbohydrate molecules. Apart from the intrinsic interest, selective denitration could be of great value in synthesis work requiring specific substitution. Nitrate ester groups have many qualities of a good blocking agent - introduction Is relatively easy, usually by direct nitration, complete removal is quantitative by a Kuhn hydrogenation1 and no wandering of the groups has 2 ever been reported. Denitration first appeared as the undesirable slow decomposition of guncotton. Guncotton, when stored for extended periods, became very sensitive to shock and, Indeed, detonated spontaneously. Technically, the problem was solved by Abel's process of distintegrating the nitrated fibre on Hollander machines and washing thoroughly with warm water. In" 1911, Walter4 investigated the action of weak organic bases on guncotton. Dimethylaniline, phenylhy- drazine, o- and p-toluidine and naphthylamine were employed. 5 ' Becker and Hunold reported degradation of guncotton by 6 7 diphenylamine. Angeli and later Giannini studied the decomposition of guncotton by pyridine at room temperature. Analysis of the carbohydrate products showed degradation and oxidation but only a small degree of denitration. 8 In 1944, Gladding and Purves found that pure, dry pyridine caused a vigorous decomposition of dissolved, stabilized gunco.tton at steam bath temperature. Nitrogen dioxide was evolved as a volatile pyridine complex that readily crystallized above the solution on cooling. Segall,9 in 1946, was interested in the relative reactivity of the nitrate groups in cellulose trinitrate and studied its behavior when treated with pyridine in the presence of hydroxylamine, methoxyamine, and their corresponding hydrochlorides. Pyridine with excess hydroxylamine attacked cellulose trinitrate at room temperature liberating one mole of nitrogen per glucose residue. The product obtained in 98$ yield, was approximately a dinitrate. The nitrate groups attacked were shown to be secondary in nature. This cellulose dinitrate was the first cellulose nitrate reported as being stable in pyridine and Segall suggested that the instability of cellulose trinitrate was due to a specific nitrate group .in a definite position in the glucose residue. Attempts to locate the position of this group were un• successful. Methoxyamine in pyridine yielded the same dinitrate but there was no noticeable yield of nitrogen. Segall explained this on the basis of the weaker reducing action of methoxyamine. Segall also treated cellulose trinitrate with excess hydroxylamine hydrochloride in pyridine. The gas _ 4 - liberated contained 85$ nitrous oxide, the remainder being nitrogen. The product obtained in 85$ yield approx• imated a ketoxime dinitrate. In 1949, L.D. Hayward10 investigated the action of hydroxylamine in pyridine on methyl -^-D-glucopyranoside tetranitrate to see if any light might be shed on Segall's results with cellulose trinitrate. Hayward found that a similar denitration proceeded vigorously, with the evolution of 1.26 moles of pure nitrogen. The reaction was halted after several hours by pouring the mixture into water. Ether extraction yielded a sirup in 80% yield which was shown to contain methyl -fl -D-glucopyranoside -2,3,6 - trinitrate (53$), methyl -D-glucopyranoside - 3,6 - dinitrate (33$) and unidentified methyl -4-D-glucoside trinitrate (14$). Since 70$ of the nitrate groups were removed from the 4-position, this research did not directly explain the reactions of cellulose trinitrate in which the 4-position is not available. C.S. Rooney11 followed up the work of Hayward by treating methylyS-D-glucopyranoside tetranitrate with hydroxylamine hydrochloride in pyridine. A slow reaction occurred with evolution of a gas composed of 70$ nitrous oxide and 30$ nitrogen. The carbohydrate products consisted of a complex mixture of partially nitrated methylglucosides and completely denitrated polyoxime products. Methyly$-D- glucopyranoside -2,3,6-trinitrate and a substance believed - 5 - to be methyl T^-D-glucopyranoside-2,6-dinitrate were isolated, 12 E.P. Swan has shown that methyl-#-D-glucopy- ranoside tetranitrate is acted upon by quinoline at 50°C with the evolution of a gas. Angeli studied the denitration of methyl and ethyl nitrate with hydroxylamine. The corresponding alcohols were formed. 14 Ryan and Casey studied the action of primary, secondary and tertiary amines on various organic nitrate esters. No analyses of the carbohydrate products were attempted. Dimethylaniline reacted with mannitol hexanitrate,at an elevated temperature to evolve a gas consisting of 70$ nitrous oxide and 30$ nitrogen. The study of partial denitration of the nitric acid esters of sugar alcohols began in 1863, when 15 Tichanowltsch Passed dry ammonia gas through an ethereal solution of D-mannitol hexanitrate(II). A dark, viscous layer separated and nitrogen was evolved. The supernatant liquid yielded a solid which proved to be a D-mannitol pentanitrate, together with a sirupy substance which appeared to be an anhydromannitol tetranitrate. The dark, viscous layer yielded a product which appeared to be an anhydromannitol tetraamine. In 1903, Wigner16 treated mannitol hexanitrate with alcoholic pyridine, to obtain a mannitol pentanitrate in 80-90$ yield similar to that obtained by