Xiaohan
Zhu
,
Ruijie
Mi
,
Jie
Yin
,
Fen
Wang
* and
Xingwei
Li
*
School of Chemistry and Chemical Engineering, Shaanxi Normal University, Xi'an 710062, China. E-mail: fenwang@snnu.edu.cn; lixw@snnu.edu.cn
First published on 3rd July 2023
The atroposelective synthesis of axially chiral acyclic olefins remains a daunting challenge due to their relatively lower racemization barriers, especially for trisubstituted ones. In this work, atroposelective C–H olefination has been realized for synthesis of open-chain trisubstituted olefins via C–H activation of two classes of (hetero)arenes in the coupling with sterically hindered alkynes. The employment of phenyl N-methoxycarbamates as arene reagents afforded phenol-tethered olefins, with the carbamate being a traceless directing group. The olefination of N-methoxy-2-indolylcarboxamides afforded the corresponding chiral olefin by circumventing the redox-neutral [4 + 2] annulation. The reactions proceeded with excellent Z/E selectivity, chemoselectivity, regioselectivity, and enantioselectivity in both hydroarylation systems.
Given the high reactivity and abundance of unsaturated reagents,7–9 alkenes8 and alkynes9 have been widely applied as substrates in C–H bond asymmetric catalysis. In particular, metal-catalyzed C–H functionalization of arenes with unsaturated reagents has provided tremendous synthetic methods in atroposelective catalysis. Transformations of an existing olefin moiety constitutes a straightforward approach to access axially chiral olefins. In this context, Shi10 realized ortho C–H functionalization of the arene ring assisted by a directing group tethered to a tetrasubstituted olefin, resulting in size-increase along the C–C axis (Scheme 1b). Alternatively, the same group realized C(vinyl)–H functionalization of trisubstituted olefins, leading to tetrasubstituted product.11 Asymmetric functionalization of alkynes provides another straightforward avenue for construction of axially chiral olefins, where sterically hindered alkynes are typically employed to ensure atropo-stability of the olefin product. In 2021, our group reported 1,2-dicarbofunctionalization of sterically hindered alkynes via C–H activation of arenes bearing two classes of migratable directing groups.12 The Song group recently tackled the limitation of sterically hindered alkynes and disclosed copper- and palladium-catalyzed atroposelective arylboration of simple diarylacetylenes using sterically hindered aryl bromides.13 Metal-catalyzed hydrofunctionalization of alkynes offers a convenient approach to prepare trisubstituted olefins. Our group realized palladium-catalyzed atroposelective hydrophosphination of 1-indoylacetylenes, affording vinylphosphines that are potentially chiral ligands.14 Later, Wang reported related hydrophosphinylation of 1-alkynylnaphthol via copper catalysis.15 The Zhu group reported hydroarylation of 1-alkynylnaphthalene via Ni-catalyzed reductive coupling between alkyne and aryl halides.16 Alternatively, organocatalysis delivers powerful protocols to access axially chiral styrenes via addition of a relatively bulky nucleophile (such as sulfinate) to sterically hindered and electronically activated alkyne such as 1-alkynyl-2-naphthol and variants, which reacts via a vinylidene–quinone methide,17 as have been elegantly reported by Yan. In addition, Tan reported the first atroposelective addition of C-based nucleophiles to sterically hindered alkynals via an iminium ion intermediate.18
C–H bond activation has emerged as a powerful strategy in modern asymmetric synthesis owing to the abundance of hydrocarbons.19 While increasing reports on asymmetric synthesis of chiral olefins have been documented via this strategy, these systems are mostly restricted to tetrasubstituted ones starting from a pre-existing olefin unit.20 In contrast, trisubstituted olefins generally bear a much lower racemization barrier due to molecular fluxionality and distortion of the CC bond. Consequently, they have been largely underexplored due to the synthetic challenges: the reaction kinetic barrier must be lower than that of the subsequent racemization. Although hydroarylation of alkenes have been extensively studied,21 the atroposelective hydroarylation of alkynes is highly challenging because internal alkynes are necessary, and the alkene product may suffer from low atropo-stability with the introduced of a relatively small aryl group. We focused on hydroarylation of sterically hindered and electronically activated alkynes such as N-alkynylindoles via substrate activation. Meanwhile, (hetero)arenes bearing a functionalizable directing group are employed as suitable reagents, especially by a traceless directing group. We hereby report RhIII-catalyzed atroposelective C–H olefination of two classes of arenes with excellent Z/E selectivity and enantioselectivity under mild conditions (Scheme 1c).
R | H | Me | i Pr | t Bu | Ph | Boc | Bn |
---|---|---|---|---|---|---|---|
a Reaction conditions: 1 (0.12 mmol), 2a (0.1 mmol), (R)-Rh1 cat. (4 mol%) and NaOAc (0.12 mmol) in MeOH (2 mL) at 25 °C for 24 h. The e.r. was determined by HPLC analysis using a chiral stationary phase. | |||||||
Yield (%) | N.D. | 67 | 46 | N.D. | N.D. | N.D. | 23 |
e.r. | — | 88:12 | 85:15 | — | — | — | 70:30 |
Encouraged by the preliminary results, we next examined the effect of the chiral catalyst on the reaction of a paratBu-substituted carbamate (Table 2), and the reaction proceeded with high yield and enantioselectivity to afford product 3 under the same reaction conditions when catalyzed by the (R)-Rh1 catalyst, indicative of the substrate effect. Switching to other catalysts ((R)-Rh2–4) only led to inferior reactivity and enantioselectivity (entries 2–4). Further investigation of solvent effects revealed that MeOH outperformed other common solvents (entries 5–8). The screening of base additives showed that some common carboxylates of sodium, potassium, and cesium slightly increased the enantioselectivity (entries 9–11). Decreasing the temperature to 10 °C resulted in higher enantioselectivity (96.5:3.5 e.r., entry 12). Further prolonging the reaction time to 48 h led to excellent yield with no change of the enantioselectivity (entry 14, conditions A).
Entry | Cat. | Additive | Solvent | Yield (%) | e.r. |
---|---|---|---|---|---|
a Reaction conditions: 1 (0.12 mmol), 2a (0.1 mmol), (R)-Rh cat. (4 mol%) and NaOAc (0.12 mmol) in MeOH (2 mL), 25 °C, 24 h, isolated yield. The e.r. was determined by HPLC analysis using a chiral stationary phase. b 10 °C. c 0 °C. d 10 °C, 48 h. N.D. = not detectable. | |||||
1 | (R)-Rh1 | NaOAc | MeOH | 84 | 95:5 |
2 | (R)-Rh2 | NaOAc | MeOH | 52 | 90:10 |
3 | (R)-Rh3 | NaOAc | MeOH | 31 | 91:9 |
4 | (R)-Rh4 | NaOAc | MeOH | 36 | 14:86 |
5 | (R)-Rh1 | NaOAc | PhMe | 28 | 93.5:6.5 |
6 | (R)-Rh1 | NaOAc | DCE | 46 | 94:6 |
7 | (R)-Rh1 | NaOAc | TFE | 20 | 87.5:12.5 |
8 | (R)-Rh1 | NaOAc | EA | 19 | 91:9 |
9 | (R)-Rh1 | NaOPiv | MeOH | 80 | 95:5 |
10 | (R)-Rh1 | KOAc | MeOH | 83 | 95.5:4.5 |
11 | (R)-Rh1 | CsOAc | MeOH | 82 | 95.5:4.5 |
12b | (R)-Rh1 | NaOAc | MeOH | 81 | 96.5:3.5 |
13c | (R)-Rh1 | NaOAc | MeOH | 72 | 96.5:3.5 |
14d | (R)-Rh1 | NaOAc | MeOH | 92 | 96.5:3.5 |
With the optimal conditions in hand, we next explored the scope and limitation of this coupling system. Initially, the scope of the 1-indolylacetylenes was investigated (Scheme 2). The coupling of these alkynes with electron-donating (alkyl and alkoxyl), electron-withdrawing (ester), and halogen (Cl and Br) groups at different positions of the indole ring afforded the axially chiral olefins in excellent yields and enantioselectivities (4–12, 91.5:8.5 to 99:1 e.r.). Extension of the alkyne terminus to phenyl groups bearing a large rang of electron-donating, -withdrawing, and halogen substituents at the meta and para positions were fully compatible (13–23). Of note, alkynes bearing an ortho-substituted phenyl was also compatible despite the somewhat enhanced steric effect, giving the desired product in excellent enantioselectivity and efficiency (20 and 21, >96:4 e.r.). The alkyne terminus was also successfully expanded to a 2-thienyl, affording the corresponding product 23 in 86% yield and 95.5:4.5 e.r. The bulky 2-substituent in the indole was not restricted to a sulfonyl group, and N-alkynylindoles bearing a 2-diphenylphosphoryl group also reacted smoothly in high enantioselectivity under the standard reaction conditions (25, 91.5:8.5 e.r.). The absolute configuration of the product 17 was determined by X-ray crystallography (CCDC 2243043†) to be (R), and the rest were assigned by analogy. To obtain the conformational stability of this class of product, racemization studies have been conducted for 12, and a ΔG≠rac = 28.9 kcal mol−1 was obtained in PhMe at 80 °C, indicating the relatively low stability.
We next evaluated the generality of phenyl methoxy-carbamate in the coupling with alkyne 2a under the optimal conditions, where pronounced substituent effects were observed (Scheme 3). Simple phenyl methoxycarbamate was tolerated in this reaction and attenuated enantioselectivity was obtained (26, 90.5:9.5 e.r.). Carbamates bearing an electron-donating group at para position of the benzene ring tend to react with high yield and enantiomeric ratios (27–30). Similarly, carbamates with a halogen, vinyl, or ester group at the para position all tend to react in good yields and slightly lower atroposelectivities (31–34). The presence of meta Me, OMe, and Br groups was also well tolerated, and the corresponding product was obtained with good yield and enantioselective control (35–37). Overall, the reaction enantioselectivity is affected by both the steric and electronic effects of the substituent. Moreover, an estrone-derived carbamate also proved effective in this coupling system with high d.r., albeit with a moderate conversion and a low regioselectivity, and the product (38) could be potentially useful in pharmaceutical development. Unfortunately, no reaction occurred when an ortho-substituted O-phenyl methoxycarbamate was employed as a substrate even under harsh conditions.
To better explore the scope of arene reagents in atroposelective C–H olefination, typically reactive heteroarenes such as indoles have been investigated. We have identified N-methoxy-2-indolylcarboxamide (39) as a suitable candidate (Scheme 4). Extensive studies have been conducted to control both the enantioselectivity and the chemical selectivity because it readily underwent redox-neutral [4 + 2] annulation via cleavage of an internal oxidizing N–OMe bond. Optimizations of the catalyst, solvent, and additive revealed that the rhodium spirocyclic Cp catalyst (R)-Rh4 (ref. 22) outperformed the rest, and the choice of iPrOH solvent was essential to suppress the [4 + 2] side reaction (see ESI†). Both excellent enantioselectivity and reactivity were realized under these conditions. The scope of this atroposelective system was briefly explored by using various alkyne reagents. It was found that introduction of alkyl, alkoxyl, halogen, ester, and CF3 group to the different positions of the alkyne has been well-tolerated, and excellent enantioselectivities ranging from 94.5:5.5 to 99:1 e.r. have been realized (40–49). The absolution configuration of product 41 was determined to be (R) by X-ray crystallography (CCDC 2260294†). The conformational stability of product 41 was also investigated (90 °C, toluene), and a racemization barrier of 30.2 kcal mol−1 was obtained, which is only slightly higher than that of the product 12. In contrast to the high reactivity of the indolyl N–Me substrate, no reaction was observed when the corresponding protic NH reagent was used, likely due to the N–N chelation that inhibited subsequent C–H activation (substrate inhibition).
Synthetic applications of representative products were next demonstrated (Scheme 5). To explore the scalability of this protocol, the reaction of carbamate 1a was performed at a mmol scale under a reduced catalyst leading, affording (R)-3 in 90% yield with 96.5:3.5 e.r. Oxidative C–O cyclization has been realized when the (R)-3 was treated with I2 under mild conditions, affording a C–N axially chiral biaryl 50 in a moderate yield (47%) and excellent enantioselectivity (95.5:4.5 e.r.). Electrophilic bromination of 3 with NBS was accomplished in 68% yield with 92.5:7.5 e.r. of the product 51. O-Triflation (52) of the axially chiral indole–phenol with Tf2O followed by the Sonogashira coupling with phenylacetylene afforded alkyne 53 in good yield. In all cases, only slight erosion of enantiopurity was detected.
Experimental studies have been briefly conducted to gain some insight into the mechanism of this reaction (Scheme 6). A control experiment using simple phenol as a substrate has been conducted under the standard conditions, but no desired product was detected (Scheme 6a). This observation indicated necessity of the carbamate directing group, and the phenol moiety was derived from decay of this traceless directing group. Indeed, an OMe carbamate was detected as a co-product (see Scheme 5a). The reaction likely follows a C–H activation pathway based on previously related reports of phenols bearing a heterocyclic directing group, and kinetic isotope effect (KIE) was then determined from two parallel reactions using 1 and 1-d5. A rather large value of KIE = 3.6 was obtained based 1H NMR analysis of the coupled products, indicating that the cleavage of the C–H bond is involved in the rate-limiting step (Scheme 6b). A Hammett plot of log(kR/kH) for various para-substituted phenyl methoxycarbamates revealed a decent linear correlation with a slight negative slope (Scheme 6c), suggesting stabilization of positive charge in the transition state of the turnover-limiting step. This is consistent with a C–H activation process where a more electron-rich arene is more susceptible to C–H activation. To explore the mechanism of the indolylcarboxamide system, a stoichiometric C–H activation reaction between the amide 39 and the Rh-4 complex was conducted in the presence of a base followed by saturation with PPh3,9b affording the 18-electron complex 54 in high yield as a single stereoisomer (Scheme 6d) that has been characterized by X-ray crystallography (CCDC 2265864†). Complex 54 was indeed catalytically active when designated as a catalyst for the coupling of amide 39 and 2a, affording coupled product 40 in high yield with 98.5:1.5 e.r. In complex 54, the bulky amide directing group is pointed inward for minimized repulsions with the chiral ligand, which will then dictate the orientation of an incoming alkyne 2a in the catalytic cycle (replacement of the PPh3 by the alkyne). A stereochemical control model9b,12,23 is then proposed to account for the observed enantioselectivity of the product 40 when catalyzed by the (R)-Rh4 catalyst on the basis of our previous report (Scheme 6e). The 2-Ts group of the alkyne is aligned upward, as dictated by the chiral environment of the rhodacycle for minimized steric repulsions with the arene ring. Enantio-determining migratory insertion of the Rh–C bond is followed by protonolysis of the resulting rhodium vinyl bond, affording the observed (R) product.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 2243043, 2260294 and 2265864. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3sc02714g |
This journal is © The Royal Society of Chemistry 2023 |