Palladium-Catalyzed Reductive Heck Hydroarylation of Unactivated Alkenes Using Hydrosilane at Room Temperature

Abstract

The reductive Heck hydroarylation of unactivated alkenes has emerged as an essential reaction for regioselective hydroarylation. Herein, we report a palladium-catalyzed reductive Heck hydroarylation of unactivated alkenes under mild conditions with enhanced functional group tolerance using hydrosilane as the reducing reagent. Under the optimal conditions, the alkylarene yields increased, resulting in minimal undesired products. Mechanistic studies using deuterated reagents indicated the involvement of two competing catalytic cycles.

The Mizoroki–Heck coupling¹⁻⁴ of aryl halides and alkenes is an effective protocol for C(sp²)–C(sp²) bond formation, which enables the synthesis of functionalized alkenes⁵⁻⁷ used for producing pharmaceuticals, agrochemicals, and natural products. Reductive Heck hydroarylation of alkenes,⁸⁻¹⁰ a variant of Mizoroki–Heck coupling, is an attractive approach for the single-step construction of alkylarenes via C(sp²)–C(sp³) bond formation using hydride reagents. Aiming for a protecting group-free total synthesis,¹¹ the reductive Heck hydroarylation has recently been applied to construct complex carbon skeletons of natural products.

The first palladium-catalyzed reductive Heck hydroarylation was reported by Cacchi¹² in 1983 using α,β-unsaturated ketones and aryl iodides. Since then, inter- and intramolecular reductive Heck hydroarylation reactions have been developed using various substrates that have been classified into four types: activated (conjugated),¹²⁻¹⁴ unactivated (nonconjugated),¹⁵⁻²⁰ strained²¹ (lack β-hydrogens), and tethered alkenes^22,23 (aryl halide groups in the same molecules). However, unactivated alkenes have not been widely used in this reaction, except for intramolecular coupling, because substrates with no electronic differences between two olefinic sites cannot undergo regioselective hydroarylation. Therefore, inter- and intramolecular reductive Heck hydroarylations have been conducted using electronically biased alkenes, such as activated alkenes.

In 2018, Wu and Loh¹⁶ reported the first regioselective reductive Heck hydroarylation of unactivated alkenes with a directing group using proton sponge as the hydride source (Figure 1a). The directing group, Dauglis’s 8-aminoquinoline²⁴ (AQ), controlled the regioselectivity of the migratory insertion to achieve the regioselective hydroarylation of unactivated alkenes.²⁵⁻²⁸ Subsequently, Loh and Wu^18,19 and Gong and Song²⁰ independently developed the reductive Heck hydroarylations of unactivated alkenes with the same directing group using alcohols as the hydride sources (Figures 1b and c). These pioneering studies revealed that AQ is an effective directing group for regioselectivity and that α-hydrogen-bearing alkylamines or alcohols can be used as hydride sources. Fatty acid alkylarene derivatives are important skeletons frequently found in different pharmaceuticals (Figure 1e).²⁹⁻³¹ However, despite affording valuable alkylarenes of fatty acids, certain drawbacks are associated with these reactions: (I) high temperatures (>100 °C) are required; (II) polar functional groups, such as amino groups, are unsuitable because they undergo side reactions such as N-alkylation^32,33 with aldehydes or imine generation when alcohols or amines are used as hydride sources; and (III) easily reducible substituents, such as bromo and benzyl-protecting hydroxy groups, are not tolerated by palladium–hydride species. Overcoming these drawbacks will lead to significant progress in the reductive Heck hydroarylation of unactivated alkenes. Therefore, using a new reducing reagent, we investigated the palladium-catalyzed reductive Heck hydroarylation of unactivated alkenes at room temperature under mild conditions using hydrosilane as the hydride source (Figure 1d).

Figure 1.

Intermolecular reductive Heck hydroarylation of unactivated alkenes and pharmaceuticals. (a) Amine and (b and c) alcohol as the hydride sources. (d) This study: hydrosilane as the hydride source. (e) FDA-approved pharmaceuticals.

For the initial screening, different reported hydride sources^{8−10,15−20} were investigated using the 4-pentenoic acid derivative 1a with an AQ directing group and p-iodoanisole 2a (3.0 equiv) in the presence of Pd(OAc)₂ (10 mol %) and 6,6′-dimethyl-2,2′-bipyridyl (10 mol %) (Table 1a, entries 1–4). These hydride sources afforded either no reaction or a low desired product (3aa) yield. Therefore, we focused on triethylsilane, which is a reducing reagent used in the dehalogenation^34,35 of aryl halides and the palladium-catalyzed cycloreductions of haloene-ynes,³⁶ particularly the room-temperature dehalogenation of aryl halides using triethylsilane. Based on previous studies, we used triethylsilane as the hydride source; however, no reaction occurred (Table 1a, entry 5). Subsequently, we examined the effects of adding cesium fluoride. Fluoride anions coordinate with organosilane reagents to form highly reactive pentacoordinated silicates,^37,38 which can undergo transmetalation^39,40 with the Pd^II ligand. Consequently, 3aa was obtained in 16% yield in our developed reaction (Table 1a, entry 6). Combining hydrosilane and cesium fluoride was effective at room temperature. Thus, we screened various hydrosilanes (Table 1a, Entries 6–9, See Supporting Information (SI), p. S5), among which polymethylhydrosiloxane (PMHS) gave the highest yield of the desired product. Next, we examined the reaction with and without bipyridyl ligands and found that the reaction without bipyridyl ligands afforded the highest yield among these trials (Table 1a, entries 9–12, see SI, p S6). To further improve the yield, we analyzed the crude mixture after the reaction reached completion (Table 1a, entry 12). The crude mixture contained the starting material 1a (38%), desired compound 3aa (21%), undesired olefin migration product 1b (19%), and undesired reduced product 1c (16%). Based on this result, we hypothesized different pathways for the formation of the desired and undesired products (Table 1b) depending on the coupling partner that underwent oxidative addition to the palladium catalyst. In the desired pathway, the reaction proceeds via the Ar–Pd^II–I species generated via oxidative addition of Ar–I to Pd⁰, whereas in the undesired pathway side reactions such as olefin migration (chain walking reaction^41,42) and C=C double bond reduction⁴³ proceed via the Si–Pd^II–H⁴⁴ species generated from the Pd⁰ catalyst and Si–H. We hypothesized that decreasing the yield of the undesired products, such as 1b and 1c, would improve the yield of 3aa. To ensure that the reaction proceeded via the desired pathway, we attempted to create a condition under which Ar–Pd^II–I was produced in larger excess than Si–Pd^II–H. As the formation of the Si–Pd^II–H species could be avoided with a limited amount of hydrosilane, we proceeded with slow addition of Si–H over 6 h using a syringe pump (see SI, p S7); the yield of 3aa increased (82%), whereas those of the byproducts decreased (0% 1b and 11% 1c) (Table 1c, entries 12 and 13). Screening of the reaction solvents indicated that the highest yield of 3aa (87%) was obtained using ethyl acetate (Table 1a, entries 13–15).

Table 1.

^a

^a(a) Reaction optimization. (b) Overview: desired and undesired paths from 1a. (c) Effect of slow addition (entries 12 and 13).

^dNMR yield was determined using dibenzylether as an internal standard.

^eSlow addition of Si–H over 6 h.

Under the optimized reaction conditions, the substrate scope of aliphatic alkenes for the room-temperature reductive Heck hydroarylation (Figure 2a) was examined. As expected, using butenoic and pentenoic acid derivatives gave the desired γ- and δ-arylated products, respectively, in high yields (3aa, 3da, 3ea, and 3ha). Although alkenes bearing branched alkyl chains were expected to afford high yields owing to the Thorpe–Ingold effect,^45,46 the yields were lower than those achieved using the derivatives with nonsubstituted alkyl chains (3fa, 3ga, and 3ha). By contrast, internal alkenes gave higher yields than terminal alkenes, although the migratory insertion of internal alkenes was less likely than that of terminal alkenes (3da and 3ea). Further, we evaluated aryl iodides containing electron-withdrawing groups (EWGs), electron-donating groups (EDGs), and polar functional groups (Figure 2b). The coupling reagent had the following reactivity characteristics: (I) The ortho-, meta-, and para-substituents did not significantly affect the reactivity; however, the ortho-substituted reagents had relatively low reactivity (3aa, 3ab, and 3ac). (II) The reaction with the coupling reagents and EWGs proceeded slower than that with the EDGs (selected examples: 57% 3an and 73% 3aq). (III) Functional groups that could be reduced by hydrosilanes, such as aldehydes,^47,48 methyl ketones,^47,48 and nitriles,^49,50 were also applicable. Additionally, easily reducible substituents, such as bromo, benzyl-protecting hydroxy, and nitro groups, were applicable under these reductive conditions (3af, 3am, and 3av). (IV) Polar functional groups, such as amino or hydroxy groups, were applicable (3ad and 3aj). (V) The yields were comparable to or higher than those previously reported for the reductive Heck hydroarylation,^{8−10,15−20} which required high reaction temperatures.

Figure 2.

(a) Scope of aliphatic alkenes. (b) Scope of aryl iodides. ^aReaction conditions: 1 (0.25 mmol), p-iodoanisole (0.75 mmol), Pd(OAc)₂ (0.025 mmol), CsF (0.5 mmol), PMHS (0.5 mmol, slow addition over 6 h), and AcOEt (1.1 mL). Reaction time of 24 h. ^b3.1 equiv of PMHS (0.75 mmol, slow addition over 9 h) was used. ^c4.1 equiv of PMHS (1.0 mmol, slow addition over 12 h) was used. Reaction time of 48 h. ^d4.1 equiv of PMHS (1.0 mmol, slow addition over 12 h) and 2.4 mL of AcOEt were used. Reaction time of 48 h. ^eYield determined after purification.

To gain mechanistic insights into the reductive Heck hydroarylation using hydrosilanes, we examined the incorporation of deuterium using deuterated triethylsilane (Et₃SiD). The results indicated that 3aa-d1 was deuterated (0.8 D) only at the γ-position of the alkyl chain (Figure 3a), suggesting that the alkyl–palladium species generated by migratory insertion underwent transmetalation by triethylsilane, thereby affording the deuterated products, which is consistent with the reported mechanism.⁸⁻¹⁰ The partial incorporation of deuterium (0.8 D) at the γ-position suggests competition between protodepalladation^51,52 by an amide proton and transmetalation by Si–D. Therefore, we examined reductive Heck hydroarylation using a deuterated amide (1a, 0.9D) (Figure 3b). Incorporation of deuterium at the γ-(0.8 D) and δ-positions (0.2 D) was observed, and the occurrence of protodepalladation by amide deuterons was supported.

Figure 3.

Mechanistic experiments. (a) Deuteration experiment using Si–D. (b) Using deuterated amide 1a (0.9D). ^aSi–D added over 6 h using a syringe pump. ^bYield determined after column chromatography.

The results reported herein and previously⁸⁻¹⁰ suggest two plausible catalytic cycles for the reductive Heck hydroarylation: main and side cycles (Figure 4). In the main cycle, Pd^III is first reduced to Pd⁰II by hydrosilane,⁴³ followed by the oxidative addition of Ar–I to Pd⁰. The resulting Ar–Pd^II–I species III is coordinated by the AQ directing group of the substrate and undergoes regioselective migratory insertion⁵³ to form the key intermediate (V) of the five- or six-membered palladacycle. After transmetalation⁴⁰ with the pentacoordinated silicate generated from Si–H and CsF, the alkyl–Pd^II–H intermediate (VI) is obtained. Finally, reductive elimination completes the catalytic cycle, thus producing the γ- or δ-arylated desired product (X). However, the main cycle does not fully rationalize the results of incorporating deuterium, in which the δ- (0.2 D) and γ-positions (0.8 D) are deuterated (Figure 3b). This recovered deuterated product (3aa-d2) suggested the formation of a benzyl palladium intermediate as the product precursor. Thus, we propose a side cycle where, following the regioselective β-H elimination of palladacycle V_n=1 (see SI, p S9), the Pd–H species undergoes rotation and migratory insertion, thereby affording the benzyl palladium intermediate (IX). This intermediate is protonated by an amide proton, thus producing X. Finally, Pd^III, released by the protodepalladation of product precursor IX, is reduced again by Si–H to regenerate Pd⁰II, which becomes available for the next catalytic cycle.

Figure 4.

Proposed catalytic/main and side cycles (n = 0 or 1).

In summary, we developed the room-temperature reductive Heck hydroarylation of unactivated alkenes. The key to this process was the slow addition of PMHS as the hydride source. These findings indicate that the reaction proceeds at room temperature with high functional group tolerance, thereby affording the products in moderate-to-high yields. This reaction system does not require specific ligands, and each reagent is relatively inexpensive. Therefore, this reaction is useful for laboratory-scale and industrial synthesis. Further studies to elucidate this reaction mechanism are ongoing in our laboratory.

Data Availability Statement

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

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.joc.3c02488.

• Investigation of optimal conditions, control experiments, preparation of substrates, typical procedure and synthesis for alkylarenes, characterization data, and NMR spectra (PDF)

Author Contributions

T.S. and Y.M. performed the experiments and analyzed the data. T.S., Y.M., R.N., and T.K. conceived and designed the experiments and prepared the manuscript.

The authors declare no competing financial interest.