Palladium-catalyzed regioselective and enantioselective acylation of aryl C-H bonds and nickel-catalyzed cross-coupling of 2,2-difluorovinyl benzoates with arylboronic acids for modular synthesis of gem-difluoroalkenes

Abstract

Transition metal-catalyzed C-C bond formation is a promising approach for organic synthesis, especially for the syntheses of multi-functionalized molecules. In recent decades, many successful examples in Pd-catalyzed direct C-H bond functionalization have been reported. The first part of this thesis explored the Pd-catalyzed direct ortho- C-H acylation of aromatic ketones. This is the first successful attempt of direct C-H acylation of aromatic ketones. To begin, aromatic ketones was treated with benzaldehyde in the presence of the cyclopalladated complex of aromatic ketone and TBHP. However, no acylation products were obtained. We then turned to employ α-oxocarboxylic acids as aroyl surrogates and (NH4)2S2O8 as oxidant. We were gratified that 44% of the acylation product was isolated with 10 mol % of the cyclopalladated complex as catalyst. After optimization, this protocol was able to generate ortho-acylated aromatic ketones (1,2-diacylbenzenes) in up to 90% yield and with 19 examples. The acylation reaction was inhibited by the addition of TEMPO as radical scavenger, with the adduct of benzoyl radical and TEMPO being isolated. It is proposed that the reaction occurred with the intermediary of cyclopalladated complexes and acyl radicals. We later turned to the Pd-catalyzed enantioselective acylation reaction for the synthesis of enantioenriched diarylmethylamines derivatives. Treating the N-protected diarylmethylamines (triflimides) with α-diketones, TBHP, TBAB, Pd(OAc)2, Na2CO3, CsOAc and mono-protected amino acids in THF at 80 °C for 16 h, the corresponding acylated triflimides were produced in up to 71% yield with up to 96% ee. After optimization, the triflimides can successfully coupled with both benzils and alkyl α-diketones, such as diacetyl and 3,4-hexadione. KIE experiments showed a kH/kD value of 1.8, which suggests that the reaction occurred with a rate-determining C-H activation step. A series of dosage dependent experiments has been performed with TEMPO as radical scavenger. When the dosage of TEMPO increased, the yield of the acylation product dropped accordingly. Geminal difluorovinyl moiety is known to be a surrogate of ketone groups, which have found important application in drug design. The final part of this thesis demonstrates a Ni-catalyzed modular synthesis of gem-difluoroalkenes. In this work, a series of 2,2-difluorovinyl benzoates (BzO-DFs) has been synthesized and their reactivities have been studied in the Ni-catalyzed cross-coupling reactions with various nucleophiles, including arylboronic acids, Grignard reagents, organozinc reagents and trifluoroborates. In the presence of Ni(cod)2 (10 mol %) as catalyst and dppf (12 mol %) as ligand, BzO-DFs coupled with arylboronic acids to afford gem-difluorovinyl ethers in up to 89% yield with 44 examples. The reaction also demonstrated good functional group tolerance, BzO-DFs bearing functionalities such as tertiary amines, trimethylsilyl groups, ethereals, alkenes and phthalimides were converted successfully. Some natural product derivatives have also been modified accordingly to further demonstrate the applicability of this reaction. Besides, this protocol was found to be effective for the synthesis of gem-difluoroalkenes, for example, 4,4'-(2,2-difluoroethene-1,1-diyl)bis(methoxybenzene) has been synthesized in 65% with its structure confirmed by X-ray crystallography. BzO-DFs can also couple with other nucleophiles under modified conditions. For example, in the presence of Ni(cod)2 (10 mol %) and PPh3 (20 mol %), BzO-DF coupled with 4-methoxyphenylmagnesium bromide to afford the corresponding gem-difluorovinyl ether in 45%. Under Ni-catalyzed conditions, C(sp2)-C(sp3) bond formation has also been achieved in the cross-coupling of BzO-DFs and benzylzinc(II) bromide and the gem-difluorovinyl ether was obtained in 38%. With a photocatalytic protocol, which employed NiCl2(dtbpy) (10 mol %) as catalyst and Ir[dF(CF3)ppy]2(bpy)PF6 (3 mol %) as photocatalyst, the coupling of BzO-DF with potassium benzyltrifluoroborate afforded the desired gem-difluorovinyl ether in 54% after irradiating with blue LED (26 W) for 18 h. The Ni-catalyzed reaction was found to be unaffected by the presence of BHT. Also, when the reaction was performed in the presence of α-cyclopropylstyrene as a radical probe, the yield of the coupling product was not affected and the radical probe was recovered completely. These results suggested that a radical mechanism was untenable.