The use of photochemical reactions and bio-based starting materials for the synthesis of Nitrogen and Oxygen heterocycles

Abstract

The work detailed in this thesis can be broadly divided into two parts. The first part was concerned with investigating the scope and limitation of a novel light-mediated C–N aromatic ring-forming reaction from methoxy-containing biaryl aromatic oximes. Previously in our laboratories at Wits University, in an attempted synthesis of phenanthroviridone, an advanced biaryl intermediate 3-methoxy-5-methyl-2-(1,4,5-trimethoxynaphthalen-2-yl)benzaldehyde O-phenyloxime when exposed to UV irradiation yielded 1,10,12-trimethoxy-8-methylbenzo[c]phenanthridine, together with 3-methoxy-5-methyl-2-(1,4,5-trimethoxynaphthalen-2-yl)benzonitrile, instead of the desired natural product phenanthroviridone. This fortuitous synthesis of a phenanthridine nucleus is thought to arise from homolysis of the weak N–O bond from the O-phenyloxime intermediate, generating iminyl radicals. The putative iminyl radical then undergoes intramolecular cyclization, forging a C–N bond with concomitant expulsion of the ortho-methoxy group, furnishing the phenanthridine nucleus. This serendipitous discovery allowed for the construction of the biologically important phenanthridine framework using UV irradiation. For the first part of this PhD we set out to investigate the substrate scope and limitations and to determine if this reaction could be applied generally. As a consequence, simple biaryl intermediates carrying an oxime ester group ortho to the biaryl axis were synthesised. In addition, the second aryl ring (the accepting ring) of the said intermediates had varying degrees of substitution and positioning of methoxy groups. However, in all cases, the substrates had to possess at least one methoxy group also ortho to the biaryl axis. The substrates were then irradiated under similar conditions as in the reaction described above and the products were analysed using NMR spectroscopy, FTIR and HRMS. It was demonstrated that the reaction prefers electron-rich accepting rings as irradiation of 2',4',5'-trimethoxy-[1,1'-biphenyl]-2-carbaldehyde O-acetyl oxime gave 2,3-dimethoxyphenanthridine in 74% yield. On the other hand, irradiation of 2'-methoxy-[1,1'-biphenyl]-2-carbaldehyde O-acetyl oxime yielded 2'-methoxy-[1,1'-biphenyl]-2-carbonitrileexclusively. The reaction is also sensitive to the positioning of the methoxys groups on the accepting ring. A case in point is the irradiation of the ortho 2',3'-dimethoxy-[1,1'-biphenyl]-2-carbaldehyde O-acetyl oxime which gave 4-methoxyphenanthridine, while irradiation of the meta 2',4'-dimethoxy-[1,1'-biphenyl]-2-carbaldehyde O-acetyl oxime furnished 2',4'-dimethoxy-[1,1'-biphenyl]-2-carbonitrile with no 3-methoxyphenanthridine produced. The corresponding nitrile side product resulting from the elimination of acetic acid was produced in all cases. This novel photo-mediated methodology was successfully extended to the synthesis of the naturally occurring phenanthridine –trisphaeridine, albeit, overall in moderate yield. The synthesis of trisphaeridine was achieved through the photo cyclization of 6-(2,4,5-trimethoxyphenyl)benzo[d][1,3]dioxole-5-carbaldehyde O-acetyl oxime which gave 2,3-dimethoxy-[1,3]dioxolo[4,5-j]phenanthridine. This step was followed by the removal of the methoxy groups via the [1,3]dioxolo[4,5-j]phenanthridine-2,3-dione that was reduced to the desired trisphaeridine by the action of zinc in acetic acid. In turn 6-(2,4,5-trimethoxyphenyl)benzo[d][1,3]dioxole-5-carbaldehyde O-acetyl oxime was synthesized from the Suzuki coupling reaction between 1-bromo-2,4,5-trimethoxybenzene and(6-formylbenzo[d][1,3]dioxol-5-yl)boronic acid furnishing 6-(2,4,5-trimethoxyphenyl)benzo[d][1,3]dioxole-5-carbaldehyde. Subsequent oxime formation of the resultant aldehyde with hydroxylamine hydrochloride and esterification with acetyl chloride formed 6-(2,4,5-trimethoxyphenyl)benzo[d][1,3]dioxole-5-carbaldehyde O-acetyl oxime. The second part of this PhD thesis details the synthesis of cardol-and cardanol-based oxa-heterocyclic compounds. The aim was to use well-established reactions to generate compounds that are known to absorb UV light, with the atoms coming exclusively from bio-renewable resources. As a consequence, coumarin derivatives were synthesized from the reaction between naturally-occurring malic acid and cardanol or cardol in the presence of catalytic amounts of sulfuric acid. For example, reaction between cardol and malic acid gave 7-hydroxy-5-pentadecyl-2H-chromen-2-one. Further elaboration of the 7-hydroxy-5-pentadecyl-2H-chromen-2-one gave tricyclic annellated coumarin derivatives, such as 5-pentadecylpyrano[2,3-f]chromene-2,8-dione, 4-pentadecyl-7H-furo[3,2-g]chromen-7-one and 5-pentadecyl-2H-furo[2,3-h]chromen-2-one. Finally, a concise synthesis of cardol-derived tetrahydrocannabinol (THC) derivative ((6aR,10aR)-6,6,9-trimethyl-3-pentadecyl-6a,7,8,10a-tetrahydro-6H-benzo[c]chromen-1-ol) is also described. This synthesis was achieved through Lewis-acid mediated union between 5-pentadecylbenzene-1,3-diol and (1S,5R,6R)-2,7,7-trimethylbicyclo[3.1.1]hept-2-en-6-ol in dichloromethane