Reductive Cross-Coupling of Alkynes with Aryl Bromides via Zr/Pd Synergistic Catalysis

Abstract

The cocatalytic reductive cross-coupling of unactivated alkynes with aryl bromides is reported, using 2 mol % Pd­(PPh₃)₄ and 10 mol % Cp₂ZrCl₂ in combination with a hydrosilane reductant. Hydroarylated products are furnished in upward of 91% yield across 25 elaborated substrates, including those containing unprotected alcohols, esters, amides, and terminal alkenes. Syn-selective hydrozirconation of alkyne substrates prior to transmetalation ensures exceptional (>20:1) (E)-selectivity across all alkene products. Further, this protocol was successfully implemented on gram scale without loss of efficiency or selectivity.

The reductive cross-coupling of alkynes represents a powerful synthetic disconnection, owing to the broad availability of alkynes as starting materials and the synthetic utility of resulting olefin products. Co-catalytic systems employing palladium in combination with a second transition metal, such as copper, and a terminal reductant have proven particularly effective for selective alkyne hydroarylation. , However, the pronounced reactivity of metal hydride or vinyl-metal intermediates in these systems often compromises chemoselectivity on complex substrates, especially those bearing alkene, carbonyl, or alcohol functionality.

While this transformation can also be achieved using less reactive vinylzirconocene coupling partners, formed via stepwise, stoichiometric protocols (Figure a), the use of zirconium in this context has been limited due to the cost and instability of Schwartz’s reagent, Cp₂ZrHCl. , Moreover, due to the attenuated reactivity and higher steric demand of vinyl zirconocene complexes, subsequent transmetalation to zinc is often preferred, thus necessitating stoichiometric amounts of two metals. Collectively, these challenges underscore the need for a more streamlined and economical method while still achieving high efficiency. Nevertheless, these strategies have enabled numerous syntheses toward natural product targets and bioactive molecules, such as splicing modulator FR901464 and the microsclerodermins, a family of natural products with antifungal and cytotoxic activity (Figure b).

1.

Strategies and applications for Pd-catalyzed Csp²–Csp² cross coupling via vinyl-zirconocene intermediates

Previously, we reported a Zr-catalyzed partial hydrogenation of alkynes utilizing hydrosilane reductants in combination with protic additives. We anticipated that the intermediate vinyl zirconocene species II, furnished upon hydrozirconation of the alkyne substrate 1, could instead be intercepted for transmetalation with palladium complex V followed by reductive elimination to afford the desired cross-coupled product 3 (Figure c). While the Pd catalytic cycle is well precedented, the turnover of resulting zirconocene halide species III back to the zirconocene hydride (ZrH) I within this context has only been achieved in a single study by Sakai and co-workers. This report marks a significant step forward, successfully demonstrating effective cocatalysis enabled through zirconocene fluoro complexes that react directly with hydrosilane. However, limitations of this strategy reveal opportunities for further development, including reliance on a noncommercial zirconocene catalyst prepared via reaction with Me₃SnF, which itself requires prior synthesis, as well as the use of superstoichiometric quantities of fluoride additive that promotes gradual catalyst decomposition through Cp deligation from zirconium.

Based on our prior studies, we anticipated that the zirconocene difluoride complex previously employed could be bypassed altogether by opting to promote turnover of the zirconocene halide using diethyl amine or carbonate salts instead ( Figure d). Initial efforts here sought to leverage diethyl amine as a means to facilitate the formation of a μ-oxo zirconocene dimer from the dihalide complex and water. However, at elevated temperatures it became apparent that water could be omitted from this process entirely, suggesting instead the possible generation of an amido zirconocene intermediate (see Supporting Information). Subsequent reaction with hydrosilane to regenerate the ZrH complex would then close the catalytic cycle. Alternatively, we envisioned that carbonate salts could promote a similar process, possibly through an intermediate zirconocene carbonate complex.

We commenced investigation of this cocatalytic reductive cross-coupling approach using alkyne 1a and aryl bromide 2a (Table ). Using 2 mol % of Pd­(PPh₃)₄ and 10 mol % of Cp₂ZrCl₂ in combination with dimethoxy­(methyl)­silane (DMMS) as the reductant and HNEt₂ as the turnover reagent, naphthalene 3a was furnished in 86% isolated yield (Table , entry 1). The product was formed with exclusive (E)-olefin selectivity, consistent with syn-hydrometalation preceding transmetalation. Evidence for Me­(MeO)₂SiNEt₂ formation by ¹H NMR analysis of the crude reaction mixture supported the possible involvement of a zirconocene-amido species in the catalytic cycle (see Supporting Information).

1. Examination of Reaction Conditions .

Entry	Change to Conditions	% Yield (3a)
1	None	90 (86)
2	1.5 equiv Na2CO3, 1.5 equiv 1a, 65 °C	92 (87)
3	No base	ND
4	No zirconium catalyst	ND
5	No palladium catalyst	ND
6	Cp2ZrHCl instead of Cp2ZrCl2	91
7	5 mol % Pd(PPh3)2Cl2	80
8	DEMS instead of DMMS	59
9	Solvent change to THF, toluene, or dioxane	82–87

^aUnless specified otherwise, reactions were carried out on a 0.5 mmol scale using 1.0 equiv of 2a and 1.2 equiv of 1a under N₂ atmosphere for 24–27 h. Yields were determined by ¹H NMR spectroscopy of the crude reaction mixture, using mesitylene as an internal standard. ND = not detected.

^bIsolated yield.

^cGlovebox-free setup.

Similar yields were obtained using Na₂CO₃ in lieu of HNEt₂ at 65 °C (entry 2). In the absence of base (entry 3), zirconium (entry 4), or palladium (entry 5), 3a was not produced. Instead, one or both starting materials were consumed through competing reduction or other side reactions (Table S1). Using 10 mol % of Schwartz’s reagent directly afforded 91% yield of alkene 3a (entry 6), demonstrating the comparable catalytic performance of zirconocene dichloride in this context. Alternatively, employing a glovebox-free protocol that utilizes 5 mol % of a bench-stable palladium­(II) precatalyst in place of Pd­(PPh₃)₄ provided the desired product in 80% yield (entry 7). This yield could be further improved with higher catalyst loading (see Supporting Information). As observed previously in Zr-catalyzed alkyne hydrogenation, replacing dimethoxymethylsilane (DMMS) with slightly bulkier diethoxymethylsilane (DEMS) led to diminished yields (entry 8). Finally, the hydroarylation proceeded to similar effect in other solvents, such as toluene, tetrahydrofuran, and dioxane (entry 9), whereas N,N-dimethylformamide proved incompatible with this protocol given its ability to undergo Zr-catalyzed reduction (Table S1). , While methyl pivalate is an uncommon solvent in this context, it is recognized as a sustainable and greener alternative to conventional volatile nonpolar and halogenated solvents and was therefore selected for subsequent scope studies.

Next, we evaluated the cocatalytic reductive coupling across various scales and substrates (Table ). The reaction could be scaled to 1 mmol (89% yield) and further to 5 mmol (91% yield) without loss of efficiency or selectivity, delivering upward of 1.25 g of alkene 3a. Aryl bromides bearing chloride and amine functionality performed well (3b, 3c). Notably, this reaction demonstrated high chemoselectivity for hydrozirconation of the alkyne over that of monosubstituted aliphatic and aryl alkenes (3d–3e), as well as ester, carbamate, nitrile, carbonate, and sterically hindered amide substituents (3f–3i). Products 3e, 3f, and 3i were obtained using the alternative Na₂CO₃ conditions to avoid competing amide formation and undesired reduction pathways.

2. Exploration of (Hetero)­Aryl Bromide and Alkyne Scope .

^aReactions were carried out on a 1.0 mmol scale under N₂ atmosphere at 85 °C for 20–46 h. Reported yields represent isolated yields.

^bReaction was carried out at 65 °C using 1.5 equiv of alkyne and Na₂CO₃ instead.

^cReaction carried out on a 0.5 mmol scale instead.

Decorated aryl bromides, including boronic ester and mesylate derivatives, proved compatible with the reaction conditions (3i–3l). Although sterically hindered aryl bromides, such as mesityl bromide, exhibited poor reactivity (see Supporting Information), ortho-substituted styrene 3m could be obtained in 51% yield. In addition, a range of heterocycle-containing reactants were also accommodated (3n–3t), including heteroaromatics susceptible to zirconium-mediated reduction. Remarkably, example 3t demonstrates this protocol’s ability to resist protodemetalation pathways when unprotected alcohols are present within one of the substrates, demonstrating expanded functional group tolerability compared to other cocatalytic hydroarylation strategies. In addition to aliphatic alkynes, the system also accommodated phenylacetylene, delivering alkene 3u under Na₂CO₃ conditions that avoid competitive Sonogashira coupling observed with Et₂NH.

To demonstrate the applicability of this cocatalytic alkyne hydroarylation for the generation of complex molecular architectures, several industrially relevant candidates were selected for further assessment (Table ). Dofenapyn, a previously marketed pesticide, was successfully cross coupled with an indole derivative to furnish 3v in 66% yield. Antifungal agent naftifine (3w) was prepared using this strategy, providing an alternative synthetic approach. Cholesterol-appended benzoxazole 3x was afforded only in moderate yield, with apparent decomposition likely due to the proximity of the ether linkage. Finally, the hydroarylation of an adapalene-derived alkyne proceeded in high yield to give 3y. Again, the presence of an ester necessitated a change in the additive used.

3. Application of Cocatalytic Hydroarylation to Agrochemical and Pharmaceutical Compounds .

^aReactions were carried out on a 1.0 mmol scale under N₂ atmosphere at 85 °C for 20–46 h. Reported yields represent isolated yields.

^bReaction was carried out at 65 °C using 1.5 equiv of alkyne and Na₂CO₃ instead.

^cReaction carried out on a 0.5 mmol scale instead.

In summary, we have developed a Zr/Pd cocatalytic one pot cross-coupling strategy for accessing aryl alkenes without the need for stoichiometric transmetalation additives, such as zinc salts. This method exhibits broad functional group tolerance and excellent chemoselectivity across a diverse array of substrates. Further, the cross coupling was successfully carried out on a 5 mmol scale without loss of efficiency. Taken together, these results highlight the improved chemoselectivity profile achieved when merging ZrH catalysis with Pd-catalyzed cross coupling.

The data underlying this study are available in the published article and its Supporting Information.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.orglett.6c01949.

• General procedural information, additional optimization and mechanistic studies, characterization data, and spectra (PDF)

‡.

M.E. and A.R. contributed equally.

The authors declare no competing financial interest.