α‐Amidoaldehydes as Substrates in Rhodium‐Catalyzed Intermolecular Alkyne Hydroacylation: The Synthesis of α‐Amidoketones

Abstract We show that readily available α‐amidoaldehydes are effective substrates for intermolecular Rh‐catalyzed alkyne hydroacylation reactions. The catalyst [Rh(dppe)(C6H5F)][BArF 4] provides good reactivity, and allows a broad range of aldehydes and alkynes to be used as substrates, delivering α‐amidoketone products. High yields and high levels of regioselectivity are achieved. The use of α‐amidoaldehydes as substrates establishes that 1,4‐dicarbonyl motifs can be used as controlling groups in Rh‐catalyzed hydroacylation reactions.

Transition-metal-catalyzed hydroacylation hase mergeda sa powerfula nd robustm ethod to preparek etones. [1a] Hydroacylation involves the addition of an aldehyde across the carboncarbon p-bond of an alkene or alkyne, to form ak etoneo ra n enone, respectively (Scheme 1a). [1] Rhodium(I) complexes are the mostc ommonly used catalysts, and the success of these reactions is often related to the stabilityo fk ey metal-acyl hydride intermediates, which are formed after oxidative addition of the aldehyde to the rhodium complex. [2] In addition to productive hydroacylation, these metal-acyl hydride intermediates can undergo reductived ecarbonylation, which delivers deactivated rhodium complexes and reduced substrates. Ac ommon approacht ol imit decarbonylation is to use aldehydes containing additional coordinating groups that facilitate the formation of as table chelated intermediate, whichi su sually a5 -membered metallocycle. Chelating motifs featuring phosphines, [3] alkenes, [4] amines, [5] hydroxyl groups, [6] ands ulfides [7] have been previously reported( Scheme 1b). Although this strategy has proven effective, the installation and removal,o rf urther transformation, of the chelating group is an inherent limitation. Recently our laboratory has demonstrated that simple carbonyl groups,s uch as ketones, esters and amides, when positioned b to the aldehyde group, can be used to mediate rhodium-catalyzed hydroacylation reactions( Scheme 1c). [8] The ability to exploit carbonylg roups-arguably the most versatile functional groupi ns ynthetic chemistry-in this way,p rompted us to investigate new applications of this concept, and in particular to targett he preparation of synthetically valuable a-amidoketones.
a-Amidoketones are present in many biologically relevant molecules, [9] as well as being widely used as buildingb locks in organic synthesis (Scheme 2). [10] This is exemplified by their use as precursorst oo xazoles, [11] thiazoles [12] and imidazoles, [13] which are among the most commonh eterocycles in aw ide range of biological and medicinala pplications. [14] Unsurprisingly,c onsiderable effort hasb een devoted to their synthesis;a selection of approaches is shown in Scheme 3. [15] Many of these methods rely on the use of complex substrates, that are themselves synthetically challenging. In this Communication we report am ild, robust and atom economic method to pre-pare a-amidoenones exploitingt he Rh I -catalyzedh ydroacylation of alkynes. Importantly,t he reaction uses readily available a-amidoaldehydes as substrates, and demonstrates that, in the contexto fR h-catalyzedh ydroacylation, the presumed 6-membered metallocycles accessed from 1,4-dicarbonyl compounds, are effective stabilizingm otifs.
We began our study by investigating the addition of 1octyne to aldehyde 1a to give enone 2a using av ariety of Rhcatalysts (Table 1). Evaluation of av ariety of commercial bisphosphines wasu ndertaken,w ith af ocus on ligands that had previously been shown to be effective in hydroacylation chemistry.I nitial results revealed that the smallest bite-angle ligand,d cpm, gave moderate conversiona fter 18 hours at 80 8Cu sing 1,2-dichloroethane (DCE) as solvent (entry 1). [16] Using the larger bite-angle ligand dcpe resulted in very low conversion (entry 2);h owever,e xchanging the cyclohexyl groups for phenylg roups significantly improved reactivity, with dppe delivering a9 3% conversion (entry 3). Further increasingt he bite angle (dppp)r esulted in al ower conversion, but with increased regioselectivity (entry 4). Alternative wide bite angle ligands, such as BINAP,D PEphosa nd Xantphos were not effective (entries 5-7). Havinge stablished dppe as the most effective ligand for the reaction, we next evaluated ap reformed catalyst [Rh(dppe)(C 6 H 5 F)][BAr F 4 ], [17] with the aim of achieving good reactivity with ad ecreased catalyst loading. Pleasingly, full conversion (80 %i solated yield) and increased regioselectivity (7:1 to 10:1) wasa chieved using 5mol %c atalyst (entry 8), and importantly,the reaction could be performed at 40 8Ci nC H 2 Cl 2 ,c ompared to 80 8Cn eeded in the originalr eaction. We attributet he higher conversionst oa ne ffective higher catalyst concentration when using the preformed complex, and the modest increase in regioselectivity is likely a functiono ft he lower reaction temperature.A lthough we have not undertaken detailed mechanistic studies, by analogy to the 1,3-dicarbonyl substrates we propose an irreversible alkyne insertionstep that is product determining. [8b] With the optimized reaction conditions in hand, the scope of this process was explored with respectt ob oth the alkynes Scheme2.Examples of medicinally relevant compoundsc ontainingt he aamidoketone or oxazole motif.
Having established the robustness of the hydroacylationr eaction towards the alkyne component, we next evaluated the scope of the aldehyde partner,i nc ombination with phenylacetylene.V ariation of the N-substituent was wellt olerated( 2s-u), and pleasingly an aldehyde featuring an NÀHs ubstituent also workedw ell, delivering adduct 2v in 80 %y ield and > 20:1 linear selectivity.V ariation of the a-substituent was then investigated, with benzyl, methyl and isobutyl substituted aldehydes reacting successfully to provide enones with excellent selectivities (2w-y). However,w hen sterically larger substituents werep resent in the a-position the yield of the product enone decreased, as seen in a-isopropyl enone 2z.V ariation of the acyl moiety of the amido group gave mixed results;abenzylamide gave the desired product in high yield and excellent regioselectivity (2aa), however,a na cetylamide resulted in deminished yields (2ab). Use of aC bz-carbamate, in place of an amide, delivered the correspondingk etone (2ac)i no nly 26 %y ield, and likely reflects the less Lewis basic nature of the group. The sterically demanding cyclohexyl amide was also a poor substrate, providing enone 2ad in 28 %y ield when 10 mol %o fc atalyst was used. Alleviating potentiala llylic strain by replacingt he N-Bn substituent with aN ÀHg roup led to improved yields, with cyclohexyl amide-substituted enone 2ae now produced in 44 %y ield.
To illustratet he synthetic utility of the products, we synthesised oxazole 3 from hydroacylationp roduct 2w (Scheme 5). Cyclization in presence of POCl 3 in tolueneu nder reflux conditions gave oxazole 3 in excellent (92 %) yield. [18] In conclusion, we have developed the Rh I -catalyzed hydroacylationo fa lkynes as am ethod to synthesize a-amidoketones. The use of a-amidoaldehydes as substrates establishes that 6membered rhodacyclic intermediates provide suitable chelating stability to enablep roductive hydroacylation. This robust methodt olerates aw ider ange of functionalg roups to form enones in good to excellent yields with generally high levels of regiocontrol. The use of preformed catalyst [Rh(dppe)(C 6 H 5 F)] [BAr F 4 ]w as found to be crucial to achieveh igh reactivity and regioselectivity.