A Palladium-Catalysed Allylic Alkylation Cascade: Towards the Total Synthesis of Thromboxanes A₂ and B₂
The design and development of new chemical reactions is crucial to the ongoing success of organic synthesis research. In this work the scope and utility of a recently discovered regioselective palladium-catalysed allylic alkylation (Pd-AA) cascade was explored through increasing the range of non-symmetric pyran-based biselectrophiles and β-dicarbonyl bis-nucleophiles that can be used in this reaction. Four differentially protected tri-substituted dihydropyrans based on glucose were synthesised, including 2,3-unsaturated silyl glycosides and α,β-unsaturated lactones. These substrates were assessed as bis-electrophiles in the Pd-AA cascade. One silyl glycoside bis-electrophile, possessing a carbonate leaving group, was shown to be an excellent substrate for reaction with a number of cyclic bis-nucleophiles. Furthermore, a series of regioisomeric methylated 4-hydroxycoumarins were synthesised, tested and found to be equally effective as bis-nucleophiles in the Pd-AA cascade with both acyclic and cyclic bis-electrophiles. Advances made during this research include a novel Ferrier reaction with silanol nucleophiles, which was found to produce silyl glycosides, albeit in low yields. Additionally, several Perlin aldehydes were generated by the Ferrier-type hydrolysis of 3,4,6-tri-O-acetyl-D-glucal and led to the discovery of discrepant structural assignments in the literature. Furthermore, a ¹³C NMR shielding template was generated as a tool for the stereochemical assignment of tri-substituted dihydropyrans. An extended variant of the Pd-AA cascade was achieved by employment of the bisnucleophile Meldrum’s acid with the optimal tri-substituted bis-electrophile in the presence of H₂O. The reaction afforded a γ-butyrolactone that could serve as a potential intermediate en route to the synthesis of the biologically interesting compounds thromboxanes A₂ and B₂. This extended Pd-AA cascade, although currently unoptimised, is capable of performing five synthetic transformations in one-pot and holds the potential to improve on the current syntheses of the thromboxanes.