Si-Yi Liu†
a,
Zi-Qi Zhu†*a,
Fei-Xiao Chenb,
Ni-Na Fenga,
Shao-Fei Ni*bc and
Feng Shi*ad
aSchool of Petrochemical Engineering, Institute of Functional Heterocycles, School of Materials Science & Engineering, Changzhou University, Changzhou, 213164, China. E-mail: fshi@jsnu.edu.cn; zzq@cczu.edu.cn
bCollege of Chemistry & Chemical Engineering, Shantou University, Shantou, 515063, China. E-mail: sfni@stu.edu.cn
cChemistry and Chemical Engineering Guangdong Laboratory, Shantou, 515063, P. R. China
dSchool of Chemistry and Materials Science, Jiangsu Normal University, Xuzhou, 221116, China
First published on 7th October 2025
The first diastereo- and enantioselective synthesis of indole-based atropisomers bearing Si-stereogenic centers has been established via the strategy of catalytic asymmetric reductive silylation. By this strategy, a wide range of indole-based atropisomers bearing Si-stereogenic centers were synthesized in moderate to good yields with overall excellent enantioselectivities and good diastereoselectivities (up to 87% yield, 94% ee, 7.3:
1 dr). This work not only provides a new strategy for catalytic asymmetric synthesis of atropisomers bearing Si-stereogenic centers, but also has realized the first catalytic asymmetric synthesis of indole-based atropisomers with a Si-stereogenic center, which will add an important member to the family of indole-based atropisomers bearing multiple chiral elements. Moreover, this work also presents the first catalytic asymmetric reductive silylation of indole scaffolds, which adds new content to enantioselective C–Si bond formation.
![]() | ||
Fig. 1 Current status of catalytic asymmetric synthesis of atropisomers bearing a silicon-stereogenic center, the remaining challenges and our design. |
As shown in Fig. 1c, there are only very limited reports on catalytic asymmetric synthesis of atropisomers bearing Si-stereogenic centers. In 2021, Gu's group applied the desymmetrization strategy in catalytic asymmetric ring-opening/acylation of torsionally strained silafluorenes under the catalysis of an Rh/chiral ligand,10a synthesizing biaryl atropisomers bearing Si-stereogenic centers via the cleavage of Si–C bonds (eqn (1)). In the same year, He's group reported another strategy of catalytic asymmetric dehydrogenative coupling under the catalysis of an Rh/chiral ligand, constructing bridged biaryl atropisomers bearing Si-stereogenic centers (eqn (2)).10b In spite of these elegant approaches, the established strategies rely on using presilylated biaryl precursors as substrates, and the activation of C–Si/C–H bonds necessitates noble metal catalysis. Moreover, the constructed scaffolds are confined to biaryl atropisomers bearing Si-stereogenic centers. Therefore, the remaining challenges in this field mainly include (1) developing new strategies for catalytic asymmetric synthesis of atropisomers bearing Si-stereogenic centers; (2) designing a new class of atropisomers bearing Si-stereogenic centers and their effective synthetic methods; and (3) controlling the enantioselectivity and diastereoselectivity of atropisomers bearing Si-stereogenic centers.
To tackle these challenges, based on our understanding of chiral indole chemistry,15 we conceive designing a new class of indole-based atropisomers bearing Si-stereogenic centers (Fig. 1d). In recent years, catalytic asymmetric synthesis of indole atropisomers has become an emerging area due to their wide applications in developing chiral catalysts or ligands and discovering bioactive molecules.15b,16,17 Although this research area developed rapidly, the catalytic asymmetric synthesis of indole atropisomers bearing heteroatom-stereogenic centers is still unknown, which might be ascribed to the formidable challenges in such transformations such as overcoming the relatively lower rotational barriers and less stable configuration of indole atropisomers, simultaneously generating indole axial chirality and heteroatom central chirality with high stereocontrol. So, to disclose this unknown chemistry, we envision merging the scaffold of axially chiral indole with Si-central chirality, thus upgrading to novel indole atropisomers bearing Si-stereogenic centers.
To realize the catalytic asymmetric synthesis of this new class of indole atropisomers bearing Si-stereogenic centers, we design a nickel/chiral ligand (L*)-catalyzed reductive silylation of 2-bromo-3-arylindoles as a new strategy toward this goal, which involves enantioselective C–Si bond formation (Fig. 1d). It should be noted that enantioselective C–Si bond formation through the strategy of catalytic asymmetric reductive silylation is rarely reported.18 Nevertheless, we consider that the power of asymmetric Ni-catalysis19 should provide a promising opportunity for realizing enantioselective C–Si bond formation via reductive silylation.
In our hypothesis (Fig. 1d), the Ni-catalyst will undergo oxidative addition with 2-bromo-3-arylindoles to generate NiII species, which will be reduced to NiI intermediates I. Then, the NiI intermediates will undergo another oxidative addition with silanes to give intermediates (R)-II and (S)-II, which further undergo reductive elimination to yield the designed indole atropisomers bearing Si-stereogenic centers. Notably, in the structures of NiIII intermediates (R)-II and (S)-II, the low rotational barrier and weak coordination between nickel and the alkoxyl (OR) group will facilitate tautomerization between the two stereoisomers, thus realizing the dynamic kinetic asymmetric transformation (DyKAT)20 in the reaction process and controlling the stereoselectivity of the indole atropisomers bearing Si-stereogenic centers.
Therefore, the significance of this work lies in that it will not only provide a new strategy for catalytic asymmetric synthesis of atropisomers bearing Si-stereogenic centers, but also accomplish the first catalytic asymmetric synthesis of indole-based atropisomers with a Si-stereogenic center. Moreover, this work will present the first catalytic asymmetric reductive silylation of indole scaffolds with excellent stereocontrol.
Entry | Ligand | Solvent | Yieldb (%) | eec (%) | drd |
---|---|---|---|---|---|
a Reaction conditions: 1a (0.1 mmol), 2a (0.3 mmol), NiBr2(dme) (0.01 mmol), ligand (0.01 mmol), NaI (0.3 mmol), Mn (0.3 mmol), and solvent (1 mL), T °C, 12 h.b Isolated yields.c The ee value was determined by HPLC.d The dr value was determined by HPLC and 1H NMR.e NiCl2(dme) was used as a catalyst.f Ni(cod)2 was used as a catalyst. | |||||
1 | L1 | THF | 12 | −10 | 1.1![]() ![]() |
2 | L2 | THF | Trace | — | — |
3 | L3 | THF | 39 | 5 | 1.2![]() ![]() |
4 | L4 | THF | 33 | 66 | 2.3![]() ![]() |
5 | L5 | THF | 9 | −71 | 2.3![]() ![]() |
6 | L6 | THF | Trace | — | — |
7 | L7 | THF | 41 | −14 | 1.1![]() ![]() |
8 | L8 | THF | 7 | −2 | 1.1![]() ![]() |
9e | L4 | THF | 29 | 69 | 2.4![]() ![]() |
10f | L4 | THF | 10 | 65 | 2.5![]() ![]() |
11 | L4 | Dioxane | Trace | — | — |
12 | L4 | 2-MeTHF | 44 | 73 | 2.4![]() ![]() |
13 | L4 | MTBE | Trace | — | — |
14 | L4 | Toluene | Trace | — | — |
15 | L4 | MeCN | Trace | — | — |
In further study, other reaction parameters such as additives, bases and temperature were optimized (Table 2). The evaluation of different additives (entries 1–5) disclosed that using NH4Cl as an additive instead of NaI could generate 3aa in a higher yield with better diastereo- and enantioselectivity (entry 3 vs. Table 1, entry 12). Surprisingly, the addition of a stoichiometric base benefited the enantiocontrol of the chiral Ni-catalyst possibly by the neutralization of the generated HBr (entries 6–10). For example, the use of K2CO3 as a base afforded atropisomer 3aa in 89% ee and 4:
1 dr (entry 7). The subsequent variation of the reaction temperature disclosed a delicate effect on the catalytic asymmetric silylation (entries 11–13). Namely, elevating the temperature to 60 °C led to diminished stereoselectivity of product 3aa (entry 13), whereas lowering the temperature to 40 °C boosted the diastereomeric ratio of 3aa to 6.7
:
1 albeit with a reduced yield of 39% (entry 12). At last, to increase the yield, the reaction time was extended to 96 h, resulting in a greatly improved yield of 61% with a high enantioselectivity of 93% ee and a good diastereoselectivity of 5.6
:
1 dr (entry 14). So, the conditions in entry 14 were established as the optimized conditions for the synthesis of indole atropisomers bearing Si-stereogenic centers via catalytic asymmetric reductive silylation.
Entry | Additives | Base | Yieldb (%) | eec (%) | drd |
---|---|---|---|---|---|
a Reaction conditions: 1a (0.1 mmol), 2a (0.3 mmol), NiBr2(dme) (0.01 mmol), L4 (0.01 mmol), additives (0.3 mmol), Mn (0.3 mmol), bases (0.2 mmol) and 2-methyltetrahydrofuran (1 mL), T °C, 12 h.b Isolated yields.c The ee value was determined by HPLC.d The dr value was determined by HPLC and 1H NMR.e The reaction temperature is 30 °C.f The reaction temperature is 40 °C.g The reaction temperature is 60 °C.h The reaction time was 96 h. | |||||
1 | LiCl | — | 61 | 78 | 3.1![]() ![]() |
2 | KF | — | 49 | 68 | 2.6![]() ![]() |
3 | NH4Cl | — | 63 | 79 | 3.3![]() ![]() |
4 | NaF | — | 48 | 66 | 2.7![]() ![]() |
5 | NaCl | — | 60 | 69 | 2.6![]() ![]() |
6 | NH4Cl | Cs2CO3 | 13 | 84 | 2.9![]() ![]() |
7 | NH4Cl | K2CO3 | 45 | 89 | 4![]() ![]() |
8 | NH4Cl | Na2CO3 | 42 | 88 | 4.4![]() ![]() |
9 | NH4Cl | K3PO4 | 44 | 89 | 5.1![]() ![]() |
10 | NH4Cl | Et3N | 17 | 85 | 4.2![]() ![]() |
11e | NH4Cl | K2CO3 | 12 | 82 | 6.1![]() ![]() |
12f | NH4Cl | K2CO3 | 39 | 88 | 6.7![]() ![]() |
13g | NH4Cl | K2CO3 | 31 | 84 | 3.8![]() ![]() |
14f,h | NH4Cl | K2CO3 | 61 | 93 | 5.6![]() ![]() |
Following the establishment of optimized reaction conditions, we evaluated the substrate compatibility of the catalytic asymmetric silylation for the synthesis of indole atropisomers bearing Si-stereogenic centers (Fig. 2). At first, the substrate scope of 2-bromo-3-arylindoles 1 was examined, and it was found that a wide range of substrates 1 were amenable to the strategy of catalytic asymmetric reductive silylation, leading to the generation of numerous indole-based atropisomers 3 bearing Si-stereogenic centers in acceptable yields (35–77%) with overall high enantioselectivities (66–94% ee) and moderate to good diastereoselectivities (1.6:
1–7.3
:
1 dr). In detail, 3-arylindoles 1 with R1 substituents in the C4–C5 position of the indole ring could smoothly take part in the silylation reaction to give atropisomers 3ba–3ga in generally high enantioselectivities. Due to the C4-steric effect of the indole ring, C4-fluoro-substituted substrate 1b exhibited much lower reactivity than others and required longer reaction time to generate product 3ba. Moreover, C6-substituted 3-arylindoles 1h–1k bearing alkyl, aryl, and halogen substituents served as competent substrates for the catalytic asymmetric silylation, which afforded products 3ha–3ka bearing both indole axial chirality and Si-central chirality in generally high enantioselectivities (89–91% ee) and good diastereoselectivities (3.2
:
1–5.9
:
1 dr). Notably, electron-withdrawing substituents such as the trifluoromethyl group obviously decreased the reactivity of 3-arylindole 1l, which participated in the reaction in a very sluggish way to generate the corresponding product 3la. Additionally, C7-substituted 3-arylindoles 1m–1o were also suitable coupling partners for methyl(phenyl)silane 2a, and these reactions proceeded smoothly to give products 3ma–3oa in a highly enantioselective manner (90–94% ee) with good diastereoselectivities (4
:
1–7.1
:
1 dr). In addition to monosubstituted 3-arylindoles, C6,C7-disubstituted 3-arylindoles were capable of undergoing the Ni-catalyzed asymmetric reductive silylation, affording atropisomers 3pa–3sa bearing Si-stereogenic centers in moderate yields with good stereoselectivities. Furthermore, the R2 group in the moiety of naphthalene could be changed from methoxyl to ethoxyl, and this substrate 1s displayed similar reactivity and stereoselectivity to its analogue 1a to give product 3sa in 64% yield with 93% ee and 4.8
:
1 dr. Notably, 2,6-disubstituted phenyl-type substrates 1t and 1u smoothly participated in the catalytic asymmetric reductive silylation under standard conditions to give the corresponding products 3tf and 3ua in acceptable yields with moderate enantioselectivities.
![]() | ||
Fig. 2 Substrate scope for the synthesis of indole atropisomers bearing Si-stereogenic centers.a Reaction conditions: 1 (0.1 mmol), 2a (0.3 mmol), NiBr2(dme) (0.01 mmol), L4 (0.01 mmol), NH4Cl (0.3 mmol), Mn (0.3 mmol), K2CO3 (0.2 mmol) and 2-methyltetrahydrofuran (1 mL), 40 °C, 96 h, isolated yields, the ee value was determined by HPLC, the d.r. value were determined by 1H NMR. The absolute configuration of the major diastereomer of product 3pa was determined to be (Ra,S) by the X-ray diffraction analysis of its single crystal.21 b The reaction was carried out for 168 h. c The reaction was carried out for 240 h. d The reaction was carried out for 144 h. |
Subsequently, the substrate scope of silanes 2 in the catalytic asymmetric reductive silylation reaction was explored (Fig. 2). Generally, a variety of silanes 2 bearing different Ar/R3 substituents could serve as reliable silylating reagents toward the synthesis of indole atropisomers 3 bearing Si-stereogenic centers in a highly enantioselective manner (82–94% ee) with moderate to good diastereoselectivities (3.1:
1–6.4
:
1 dr). Specifically, a series of Ar groups bearing either electron-donating or electron-withdrawing substituents at ortho-, meta- and para-positions of the phenyl ring could be utilized, and these substrates 2b–2h smoothly underwent catalytic asymmetric reductive silylation with 2-bromo-3-arylindole 1a to yield the desired products 3ab–3ah in moderate to good yields with high enantiocontrol. It seemed that the existence of an ortho-substituent was helpful to improve the diastereoselectivity, as exemplified by substrate 2b bearing an ortho-methoxy group, which afforded product 3ab in a high diastereoselectivity of 5.0
:
1 dr. Moreover, naphthyl-substituted silanes 2i–2j demonstrated outstanding suitability in the reaction, giving products 3ai and 3aj in good yields with high diastereo- and enantioselectivities. Moreover, the R3 substituent could be changed from the methyl group to other alkyl groups such as ethyl, n-propyl and t-butyl, and these silanes 2k–2m successfully underwent reductive silylation with 2-bromo-3-arylindole 1a to produce indole atropisomers with multiple chiral elements in moderate yields (52–61%) with excellent stereoselectivities (90–94% ee, 4.5
:
1–6.4
:
1 dr).
To demonstrate the practicality of the catalytic asymmetric reductive silylation in the synthesis of indole-based atropisomers with a Si-stereogenic center, we performed one-mmol-scale synthesis and some transformations of product 3aa (Scheme 1). The one-mmol-scale reaction between 1a and 2a smoothly afforded indole-based atropisomer 3aa in a high yield of 90% with a good stereoselectivity of 85% ee and 3.5:
1 dr (Scheme 1a). Furthermore, several functionalizations of the Si–H group in product 3aa were performed through transition metal-catalyzed reactions (Scheme 1b–d). For example, Pt-catalyzed olefination of 3aa gave rise to axially chiral indole-based silane 4 bearing an all-carbon quaternary Si-stereogenic center (Scheme 1b). Moreover, compound 3aa could undergo Rh-catalyzed stereoselective Si–O coupling with acetone or 4-methoxyphenol, giving products 5 and 6 via the formation of a Si–O bond in a nearly retained stereoselectivity (Scheme 1c and d).
Based on previous reports,18 we proposed a mechanism for the catalytic asymmetric reductive silylation (Scheme 2a). As exemplified by the formation of product 3aa, the transformation started from the coordination and reduction of the NiII-catalyst to generate a chiral Ni0-catalyst (Ni0-L4). Then, the oxidative addition of 2-bromo-3-arylindole 1a to the Ni0-catalyst afforded NiII intermediate A, which was further reduced by Mn powder to give NiI intermediates B. Then, the oxidative addition of silane 2a to intermediate B gave NiIII intermediate C bearing a Si-stereogenic center, which further underwent reductive elimination to afford the final indole atropisomer 3aa with a Si-stereogenic center and release the NiI complex. The generated NiI complex was further reduced by Mn powder to regenerate the Ni0-catalyst and accomplish the catalytic cycle.
To further understand the reaction pathway, we calculated the rotational barriers of substrate 1a, intermediates A–C and product 3aa by a theoretical method (Scheme 2b, see the SI for details). It was discovered that the rotational barriers of substrate 1a and product 3aa are much higher than the 24.0 kcal mol−1 required for isolating two atropisomers at room temperature, while those of intermediates A–C are lower than 24.0 kcal mol−1. These results indicated that this reaction was not a kinetic resolution (KR) process, but a dynamic kinetic asymmetric transformation (DyKAT). Namely, due to the low rotational barrier of intermediates A–C, the tautomerization between the atropisomers of A–C and A′–C′ could readily occur, thus facilitating the DyKAT process, and the final step of reductive elimination determined the absolute configuration of the axial chirality in product 3aa.
To verify the proposed reaction pathway, we performed some control experiments (Scheme 3). Firstly, we subjected 2-bromo-2′-methoxy-1,1′-binaphthalene 7 (a substrate with the indole ring replaced with a naphthalene moiety) to the reaction conditions. As shown in Scheme 3a, substrate 7 failed to participate in the reductive silylation, and no reaction occurred under standard conditions. This result indicated that the indole ring is very important for the reactivity of the substrates in the catalytic asymmetric reductive silylation. To further understand the reactivity difference between substrates 7 and 1, theoretical calculations on the rotational barriers in the reaction system of substrate 7 were conducted, including supposed intermediates D–F and product 8. Unlike 2-bromo-3-arylindole 1, the rotation in this binaphthalene system is more restricted. The DFT calculations revealed significantly high rotational barriers for substrate 7 (36.3 kcal mol−1), the corresponding nickelacycle intermediates D–F (44.8, 36.7, and 47.1 kcal mol−1, respectively), and the supposed product 8 (47.9 kcal mol−1). Notably, the high rotational barriers of the key intermediates D–F indicate that the interconversion between the atropisomers would be exceedingly slow under the catalysis of Ni/L4. This high configurational stability of the binaphthalene system contrasts sharply with our designed naphthyl-indole system, wherein a much lower rotational barrier is essential for the rapid equilibration between atropisomers. These computational results strongly support our observation that the unique structural features of the naphthyl-indole scaffold facilitate the DyKAT pathway.
Secondly, to demonstrate the role of the methoxy group in stabilizing or directing nickelacycle formation, the methoxy group in substrate 1 was replaced by a methyl group. As shown in Scheme 3b, under standard conditions, methyl-substituted substrate 1v underwent Ni-catalyzed reductive silylation to give product 3va in a low yield of 17% (eqn (3)), which supported the proposed coordination-based mechanism, i.e., the coordination of the oxygen atom in the methoxy group to the nickel catalyst is crucial for the reductive silylation reaction. Moreover, we employed N-methyl-substituted substrate 1w in the catalytic asymmetric reductive silylation reaction. As shown in eqn (4), this substrate 1w could react with 2a to give the corresponding product 3wa in 29% yield with 63% ee and 3.1:
1 dr. This result primarily ruled out the possibility that the observed racemization arises from tautomeric interconversion of the indole ring in one of the intermediates, involving temporary formation of an sp3 carbon at the stereogenic axis.
Finally, to clarify the nature of the observed diastereomers, we synthesized products bearing either axial or silicon chirality and studied their configurational stability (Scheme 4a). Specifically, we utilized symmetric silane (diphenylsilane) 2n as a substrate to synthesize product 3an bearing only axial chirality (60% ee, eqn (5)) and employed substrate 1x in the reductive silylation reaction to generate product 3xa bearing only silicon chirality (55% ee, eqn (6)). Then, we conducted epimerization experiments of products 3an and 3xa in iPrOH at 70 oC for 12 hours. It was found that indole atropisomer 3an underwent epimerization with a decreased ee value (32% ee), while the ee value of chiral silane 3xa was well retained (55% ee).
Notably, in the presence of chiral ligand L9, substrate 1x could undergo nickel-catalyzed asymmetric reductive silylation with 2a to give product 3xa bearing a silicon stereocenter in a moderate enantioselectivity of 55% ee albeit with an unsatisfactory yield of 26% (eqn (6)). By changing the additive from NH4Cl to NH4Br, the yield of 3xa could be improved to 45% with a nearly retained enantioselectivity of 53% ee. So, this result indicated that this nickel-catalyzed asymmetric reductive silylation has moderate selectivity for forming the silicon stereocenter in simpler systems, and the atropisomer formation indeed assists the selectivity of the silicon stereocenter.
Furthermore, we isolated both diastereomers of 3aa and tested whether they converge into the same diastereomeric mixture upon heating. As shown in Scheme 4b, by column chromatography (petroleum ether:
acetone = 500
:
1) on silica gel, we isolated the major diastereomer of 3aa with 11.5
:
1 dr and the minor diastereomer of 3aa with <1
:
19 dr. Then, we conducted epimerization experiments by heating both diastereomers of 3aa in iPrOH at 70 °C. It was found that the dr value of the major diastereomer dropped to 1.08
:
1 after 38 hours (eqn (7)), while the dr value of the minor diastereomer dropped to 1
:
1.06 after 78 hours (eqn (8)). Evidently, both diastereomers of 3aa converged into a similar diastereomeric mixture upon heating. So, these results demonstrated that the diastereomers are epimers at the stereogenic axis.
CCDC 2448426 contains the supplementary crystallographic data for this paper.
Footnote |
† Equally contributed to this work. |
This journal is © The Royal Society of Chemistry 2025 |