3,4,6-Tri-0-acetyl-D-glucal (1) reacted with carbon monoxide and hydrogen in the presence of dicobalt octacarbonyl to yield a mixture of two epimeric anhydrodeoxyheptitols, namely, 4,5,7-tri-0-acetyl-2,6-anhydro-3-deoxy-D-manno-heptitol (2) and 4,5,7-tri-0-acetyl-2,6-anhydro-3-deoxy-D-gluco-heptitol (3). De-0-acetylation of the mixture, followed by chromatographic separation, yielded 2,6-anhydro-3-deoxy-D-manno-heptitol (4) and 2,6-anhydro-3-deoxy-D-gluco-heptitol (5). Compounds (4) and (5) were oxidised with periodate to yield dialdehydes which on reduction with sodium borohydride afforded enantiomeric tetrol ethers. Reaction of 3,4,6-tri-0-acetyl-D-glucal (1) with carbon monoxide and deuterium, followed by de-0-acetylation and chromatographic separation gave 2,6-anhydro-5-deoxy-D-manno-heptitol-1,1,3-²H3(cis)(6) and 2,6-anhydro-3-deoxy-D-gluco-heptitol-1, 1,3-²H3(cis) (7) . Examination of the proton magnetic resonance spectra of the normal (4,5) and deuterated anhydrodeoxy heptitols (6,7) revealed their structures and showed that cis-addition of carbon monoxide and hydrogen to the double bond of (1) had taken place. Reaction of the mixture of partially acetylated heptitols (2) and (3) with p-toluenesulphonyl chloride followed by fractional crystallisation of the products gave pure 4,5,7-tri-0-acetyl-2,6-anhydro-3-deoxy-1-0-(p-toluenesul-phonyl)-D-gluco-heptitol (8). Similarly,, the mixture of (2) and (3) reacted with p-bromobenzenesulphonyl bromide to give 4,5, 7-tri-0-acetyl-2,6-anhydro-1-0-(p-bromobenzenesulphonyl)-3-deoxy-D-gluco-heptitol (11), the structure of which was confirmed by X-ray structure analysis by A. Camerman and J. Trotter. Therefore, the absolute structures of compounds (4) and (5) were ascertained. Compounds (8) and (11) were converted to (5) by a series of reactions. Comparison of the exchange reaction of sodium iodide with 4,5,7-tri- 0-acetyl-2,6-anhydro-3-deoxy-1-0-(p-toluenesulphonyl)-D-gluco-heptitol (8) and with 4,5,7-tri-0-acetyl-2,6-anhydro-3-deoxy-1-0-(p-toluenesulphonyl)-D-manno-heptitol (14) revealed that the equatorial primary p-toluenesulphonoxy group of (8) was exchanged more readily than that of (14). The hydroformylation of (1) yielded two enantiomeric aldehydes (16a, 16b) which were separated chromatographically via their 2,4-dinitrophenyl-hydrazones (16b) and (17b). Both (16b) and (17b) were degraded to (4) and (5), respectively. 3,4-Di-0-acetyl-D-arabinal (18) reacted with carbon monoxide and hydrogen in the presence of dicobalt octacarbonyl to yield, upon de-0-acetylation and chromatographic separation, a mixture of two epimeric anhydro-deoxyhexitols, namely, 1,5-anhydro-4-deoxy-L-ribo-hexitol (21) and 1,5-anhydro-4-deoxy-D-lyxo-hexitol (22). Compounds (21) and (22) were converted into enantiomeric 2-deoxy-3-0-(2-hydroxyethyl)-L-glycero-tetritol (23) and 2-deoxy-3-0-(2-hydroxyethyl)-D-glycero-tetritol (24). Compound (23) was identical to an authentic sample of 2-deoxy-3-0-(2-hydroxyethyl)-L-glycero-tetritol. 1,2,4,6-Tetra-0-acetyl-3-deoxy-α-D-erythro-hex-2-enopyranose (29) reacted with carbon monoxide and hydrogen in the presence of dicobalt octacarbonyl to yield 1,2,3¹,4,6-penta-0-acetyl-3-deoxy-3-C-(hydroxymethyl)-α-D-gluco-pyranose (31) besides hydrogenolysed hydrohydroxymethylated products, a similar reaction of (29) with deuterium instead of hydrogen gave 1,2,3¹,4,6-penta-0-acetyl-3-deoxy-3-C- (hydroxymethyl) -α-D-gluco-pyranose-2,3¹,3¹ -²H3 (cis) (33). The structures of (31) and (33) were established by p.m.r. spectroscopy. 2,3,4,6-Tetra-0-acetyl-α-D-glucosyl bromide (26) reacted with sodium tetracarbonylcobaltate under compressed carbon monoxide followed by treatment with triphenylphosphine to afford 2,3,4,6-tetra-0-acetyl-β-D-glucosyl tri-carbonyl triphenylphosphine cobaltate (39) and 3,4,5,7-tetra-0-acetyl-2,6- anhydro-D-glycero-D-gulo-heptosoyl tricarbonyl triphenylphosphine cobaltate (41). Reduction of. (39) and (41) with sodium borohydride followed by acetylation gave 2,3,4,6-tetra-0-acetyl-1,5-anhydro-D-glucitol (40) and 1,3,4,5,7-penta-0-acetyl-2,6-anhydro-D-glycero-D-gulo-heptitol (42). Both (40) and (42) were compared with authentic samples and shown to be the same. / Science, Faculty of / Chemistry, Department of / Graduate

Methyl 2,3-anhydro-4,6-0-benzylidene- α-D-mannopyranoside was allowed to react with carbon monoxide and methanol in the presence of dicobalt octacarbonyl at elevated temperatures and pressures. The resulting product mixture contained unreacted epoxide, methyl 3-0-methyl-α -D-altropyranoside and methyl 4,6-0-benzylidene- α-D-altropyranoside. Methyl 2,3-anhydro-4,6-0-benzylidene- α-D- mannopyranoside was allowed to react with carbon monoxide (990 p.s.i.) and hydrogen (910 p.s.i.) in the presence of dicobalt octacarbonyl at 140°. The resulting product mixture contained two products. The n.m.r. indicated that the epoxide ring had not been opened. In one product, the benzylidene ring was completely removed and in the other component the benzylidene ring had been partially altered. When Brigl's anhydride was allowed to react with carbon monoxide (910 p.s.i.) and hydrogen (990 p.s.i.) in the presence of dicobalt octacarbonyl at 106° for 30 minutes, the resulting product mixture contained one major product and two minor products. The minor product was presumed to be a sugar and the major product (75-80%) had been characterized as 2,6-anhydro-D-glycero-D-gulo-heptitol. / Science, Faculty of / Chemistry, Department of / Graduate

Hydroformylation of 5,6-anhydro-1,2-0-isopropylidene-α-D-glucofuranose (XXIV) with carbon monoxide (70 atm.) and hydrogen (70 atm.) at a temperature of 100-105° C. gave 6-deoxy-1,2-0-i sopropylidene-α-D-gluco-heptodialdo-1,4-furanose-3,7-pyranose (XXV) in 78% yield. Minor quantities of the rearrangement product, 6-deoxy-5-keto-1 ,2-0- i sopropyl i dene-α-D-gl ucofuranose (XXVI), and the hydro-hydroxymethyl at i on product, 6-deoxy-1 ,2-0- isopropyl idene-α-D-gl uco-hepto-1 , 4-furanose (XXVI I), were isolated in 9% and 4% yields, respectively. Acetylation of crude (XXV) afforded two anomeric derivatives (XXIX, XXVII I). Under identical experimental conditions, 5 ,6-dideoxy-1 , 2-0-isopropyl idene-α-D-xylo-hex-5-enofuranose (XXXIII) gave 5,6-dideoxy-α-D-xylo-heptodialdo-1 ,4-furanose-3,7-pyranose (XXXIV) in 51% yield. A minor amount of 5,6-dideoxy-α-D-xy1ohepto-1,4-furanose (XXXV) in about 5% yield was also detected by thin layer chromatography (T.L.C.) Fusion of 5 ,7-di-0-acetyl-6-deoxy-1,2-0-isopropylidene-α-D-gluco-heptodialdo-1,4-furanose-3,7-pyranose (XXVIII) with 5,6-dimethyl-benzimidazole using chloroacetic acid as a catalyst at 170-175°C. gave two anomeric nucleosides 1-(3'-0-acetyl-2'-deoxy-6’,7'-0-isopropylidene heptodialdo-4’,7'-pyrariose-α(and β-)-L-gulopyranosyl) -5,6-dimethyl -benzimidazole (XXXVII, XXXVIII) in 42% yield. These nucleosides were separated by multiple ascending development on preparative T.L.C. plates of silica gel G. impregnated with 2% G.E.Phosphor. Assignment of structures of the nucleosides was based on an analysis of their N.M.R. spectra. / Science, Faculty of / Chemistry, Department of / Graduate

published_or_final_version / Chemistry / Doctoral / Doctor of Philosophy

Thesis (Ph. D.)--University of Hong Kong, 1998. / Includes bibliographical references.

The purpose of this research was to directly compare the behavior of zeolites containing rhodium with that of homogeneous rhodium species as catalysts for liquid phase hydroformylation of 1-hexene in order to study the effects of zeolite immobilization. NaX zeolite was cation exchanged with several rhodium salts and used as hydroformylation catalysts at 50°C and 125°C in the presence of: triphenylphosphine (PPh₃), dimethylphenylphosphine (PMe₂Ph), and the poison for zeolite surface and solution rhodium: triphenylmethylmercaptan (Ph₃CSH). The results of these experiments were compared with those of several homogeneous catalysts under similar conditions. It was found that previously reported results of intrazeolitic activity with RhNaX at 50°C were probably incorrect, since, the addition of PMe₂Ph, Ph₃CSH, or both, virtually halted all reactivity of RhNax. The catalytic results at 125°C did not conclusively indicate the location of the active rhodium. Thus, intrazeolitic activity at 125°C may or may not have been observed, and needs further investigation. Reaction profiles were obtained for several of the catalyst systems, using an automatic sampling system. From these profiles, it was found that the addition of excess PMe₂Ph halted isomerization of 1-hexene to 2-hexenes for the zeolite-supported rhodium, and hindered, but did not stop isomerization for the homogeneous catalysts. Also, as expected, it was observed that the homogeneous catalysts reacted to completion faster than the heterogeneous catalyst. In addition, the effects of such treatments as preheating in air and precarbonylation of the heterogeneous catalysts were studied. Pretreatments had no effect upon the catalysis. Also, no activity was observed from the heterogeneous catalysts at 125°C unless phosphines were present. Finally, the hydrogenation of 1-hexene was studied. Heterogeneous and homogeneous rhodium catalysts showed hydrogenation activity which was accompanied by isomerization at 60°C and 125°C. / Master of Science