Organocatalytic Asymmetric Synthesis of Si-Stereogenic Silyl Ethers.

Functionalized enantiopure organosilanes are important building blocks with applications in various fields of chemistry; nevertheless, asymmetric synthetic methods for their preparation are rare. Here we report the first organocatalytic enantioselective synthesis of tertiary silyl ethers possessing "central chirality" on silicon. The reaction proceeds via a desymmetrizing carbon-carbon bond forming silicon-hydrogen exchange reaction of symmetrical bis(methallyl)silanes with phenols using newly developed imidodiphosphorimidate (IDPi) catalysts. A variety of enantiopure silyl ethers was obtained in high yields with good chemo- and enantioselectivities and could be readily derivatized to several useful chiral silicon compounds, leveraging the olefin functionality and the leaving group nature of the phenoxy substituent.

Chiral molecules bearing a carbon stereogenic center are ubiquitous in nature and have been the focus of fundamental and applied studies during the past decades.[1] In contrast, the asymmetric synthesis of their heavier congeners bearing a stereogenic silicon atom has been far less investigated. As enantiopure organosilanes are currently gaining substantial importance in material science, in polymer synthesis, as chiral ligands, and in scent and medicinal chemistry, an expansion of the synthetic toolbox to enable their preparation is highly desirable.[2−8] A literature survey reveals that Si-stereogenic silanes have been created either via desymmetrization and diastereoselective synthesis from achiral silicon compounds or via the kinetic resolution of racemic silanes by means of transition metal and enzyme catalysis.[9−13] Indeed, the catalytic construction of Si-stereogenic centers appears to be a topic of high current relevance in asymmetric catalysis.[14−24] However, to the best of our knowledge, organocatalytic asymmetric approaches to enantiopure silanes constitute an unmet challenge. Inspired by previous work on asymmetric catalysis of the silicon–hydrogen exchange reaction,[25,26] we considered a catalytic desymmetrizing silyl ether formation from bis(methallyl)silanes 1 with alcohols 2, in which olefin protonation occurs from an acid catalyst HX* (3) to form a β-silyl-stabilized cationic intermediate, followed by nucleophilic attack of the alcohol to generate the chiral silyl ether 5 (eq 1).

We report here that instead of the intermolecular addition of the alcohol, an intramolecular cation−π cyclization with the second olefin occurs, ultimately leading to a silylium ion equivalent, which reacts with the alcohol to form an enantioenriched silyl ether 4, featuring a new C–C σ-bond (eq 2). We have discovered a highly enantioselective, desymmetrizing, carbon–carbon bond forming reaction of symmetrical bis(methallyl)silanes with phenols. The perfectly atom-economic reaction is catalyzed by newly developed imidodiphosphorimidates (IDPi) and delivers several structurally distinct silyl ethers. We also describe the utilization of the obtained products in the synthesis of various Si-stereogenic silanes and suggest a mechanism of this unusual transformation.

As the starting point, bis(methallyl)silane 1a and 2,6-dimethyl-phenol 2a were reacted with IDPi catalyst 3a, featuring a 1-naphthyl substituent at the 3,3′-position of the BINOL backbone and a trifluoromethyl sulfonyl group in the inner core. Product 4a was obtained as the main product with only small amounts of ether 5a (Table ). Notably, we obtained the new chiral organosilane 4a in 73% yield and a promising enantioselectivity of 64:36 e.r. (Table , entry 1). A subsequent investigation of the reaction conditions revealed toluene as the optimal solvent, affording the desired product in 86% yield and 86:14 e.r. (entries 2–4). Lowering the temperature to −20 °C suppressed the formation of side product 5a and led to an increased yield without significantly affecting the enantioselectivity (entry 5). Spirocyclic-fluorenyl-substituted IDPi catalysts, which have been preferred motifs in our previous Si-ACDC studies, were subsequently investigated (entries 6–9).[27−36] We eventually identified IDPi 3e as the optimal catalyst, which enabled quantitative formation of product 4a with 97:3 e.r. (entry 9).

Table 1

Reaction Developmenta

entry	catalyst	solvent	T (°C)	yield (%)b	e.r.c
1	3a	diethyl ether	25	73	64:36
2	3a	dichloromethane	25	76	86:14
3	3a	cyclohexane	25	90	79:21
4	3a	toluene	25	86	86:14
5	3a	toluene	–20	94	86.5:13.5
6	3b	toluene	–20	>95	85:15
7	3c	toluene	–20	>95	89.5:10.5
8	3d	toluene	–20	>95	95:5
9	3e	toluene	–20	>95	97:3

Performed with 2,6-dimethylphenol 2a (0.025 mmol), 1a (1.5 equiv), and IDPi catalysts 3a–3e (2.5 mol %) in solvent (0.25 mL, 0.1 M).

Yields determined by 1H NMR spectroscopy using 1,3,5-trimethoxybenzene as internal standard.

Enantiomeric ratios (e.r.) determined by HPLC.

Having identified these optimized conditions, we investigated other substrates for this new catalytic asymmetric approach to Si-stereogenic organosilanes. As illustrated in Table , benzyl-substituted chiral silane 4a was isolated in 92% yield with 97:3 e.r. on a 0.2 mmol scale. Related silanes 4b–4d were prepared in similarly good yields and enantioselectivities (87–96% yield and 96:4 to 98:2 e.r., respectively) from the corresponding benzyl silanes with electron-donating or -withdrawing groups at the para-position of the phenyl ring. In addition, substituents with diverse electronic properties in meta- and ortho-positions of the phenyl group were well tolerated, affording chiral organosilanes 4e–4i with comparable outcome. 3,4-Dimethyl-substituted product 4j was obtained in 93% yield and an e.r. of 93.5:6.5. The use of substrate 1k bearing a bulkier 2-naphthylmethyl group on the silicon atom delivered 4k with good yield and 95:5 e.r. While the reactions of benzyl-substituted silanes (1a–1k) occurred with good chemoselectivity, aryl-substituted substrates (1l–1n) initially led to the formation of a side product and required further optimization of the reaction conditions (see the Supporting Information, Table S3). Ultimately, this gave products 4l–4n in 78–84% yields with 95:5 e.r. The silicon-stereogenic silane 4o, bearing a thiophenyl moiety, could also be obtained in satisfactory yield and with high enantioselectivity. Product 4p was generated from silane 1p, bearing three methallyl groups, with 76:24 e.r., and no further substitution was detected. Submitting silane 1q to the reaction conditions revealed the key role of the methyl substituent on the silicon atom; indeed, its replacement with an ethyl group delivered product 4q in quantitative yield but with a significantly decreased enantioselectivity (70.5:29.5 e.r.).

Table 2

Substrate Scopea

Performed with 2,6-dimethylphenol 2a (0.2 mmol), 1 (1.5 equiv), and IDPi catalyst 3a or 3e (2.5 mol %) in toluene (2.0 mL, 0.1 M) at −20 °C for 24 h. Isolated yields with enantiomeric ratio (e.r.) determined by HPLC analysis.

With IDPi 3e.

With IDPi 3a.

Despite extensive attempts, we were unable to obtain suitable single crystals of our new organosilicon compounds for the determination of their absolute configuration by single-crystal X-ray diffraction. Therefore, the “crystalline sponge method” was applied to elucidate the absolute structure by X-ray analysis (CS-XRD).[37−40]

In this method, single crystals of a flexible MOF, the crystalline sponge (CS), are used as a scaffold to arrange the analyte molecules into the pores and apply X-ray crystallographic analysis to determine the structure. With this method, we could unequivocally determine the absolute structure of the analyte molecules. The space group symmetry of the original CS changed from centrosymmetric (C2/c) to non-centrosymmetric (C2), to accommodate enantiopure molecules.

As depicted in Figure a, the absolute configuration of product 4b was assigned to be R by CS-XRD. Several independent additional experiments using the same and the opposite enantiomer confirmed the correctness of this assignment (see the Supporting Information, Figures S13–S30 and Tables S4–S6).

Figure 1

Absolute configuration determination of (R)-4b. (a) The crystalline sponge method was employed. (b) Calculated (blue curve) and experimental (red curve) CD spectra of 4b.

The absolute configuration of products 4b and 4d was furthermore confirmed to be R by computational and experimental CD spectroscopy (Figure b and Figure S33 in the Supporting Information).[41]

To illustrate the practical utility of our methodology and the synthetic value of the newly obtained enantiopure organosilicon products, a preparative-scale reaction was performed and readily delivered 1.06 g of product 4a in 96% yield and 95:5 e.r. (Figure ). Hydroboration/oxidation of olefin 4a gave compound 6, with moderate diastereoselectivity and without erosion of enantiopurity.[42] Upon cyclopropanation or epoxidation, silane products 7 and 8 could be obtained in 40% and 55% yield, respectively, with identical e.r.[43,44] Hydrogenation of 4a led to product 9 in 88% yield. Moreover, a sequential strategy was employed for the conversion of the Si–O bond into a Si–C bond. Accordingly, in situ treatment of product 9 with diisobutylaluminum hydride gave hydrosilane 10.[45] A subsequent Pt-catalyzed hydrosilylation with 1-octene provided quaternary silane 11.[18] While the absolute stereochemistry of this product remains to be confirmed, it has previously been shown that the reduction of alkoxysilanes with diisobutylaluminum hydride and also the hydrosilylation of 1-octene, respectively, proceed via retention.[46,47] The direct construction of a Si–C bond by treatment of silyl ether 4a with n-BuLi occurred with substantial loss of enantioselectivity and gave quaternary silane 12, mostly with retention of configuration at the silicon stereocenter.[48] A Pt-catalyzed intramolecular hydrosilylation of unsaturated hydrosilane 13, which is readily obtained via reduction of 4a, gave product 14 with moderate d.r. and retention of configuration.[46] The Si–H bond in silane 13 could be converted into a Si–OH group via dehydrogenative coupling with water in the presence of Pd/C.[49] This transformation has been shown to proceed with inversion of configuration, furnishing chiral silanol 15.[50] Finally, boronate 16 was prepared in 90% yield as a 1:1 mixture of diastereomers via hydroboration.[51]

Figure 2

Derivatizations. Isolated yields with e.r. determined by HPLC and d.r. measured by 1H NMR spectroscopy or HPLC. (a) BH3SMe2 (1.0 equiv), 0 °C, THF, 1 h, then H2O2 (7.8 equiv), rt, EtOH, 3 h. (b) Et2Zn (2.0 equiv), CH2I2 (4.0 equiv), 0 °C–rt, DCE, 18 h. (c) m-CPBA (2.0 equiv), 0 °C–rt, DCM, 7 h. (d) Pd/C (0.1 equiv), H2 (balloon), rt, MeOH, 12 h. (e) Pd/C (0.1 equiv), H2 (balloon), rt, MeOH, 12 h, then DIBAL-H (2.0 equiv), 0 °C–rt, hexanes, 18 h. (f) Pt(dvds) (0.05 equiv), 1-octene (2.0 equiv), 50 °C, hexanes, 48 h. (g) n-BuLi (3.0 equiv), 0–35 °C, Et2O, 48 h. (h) DIBAL-H (2.0 equiv), 0 °C–rt, hexanes, 18 h. (i) Pt(dvds) (0.05 equiv), 50 °C, hexanes, 48 h. (j) Pd/C (0.1 equiv), H2O (3.0 equiv), 0 °C, ethyl acetate, 24 h. (k) n-BuLi (0.1 equiv), HBPin (3.0 equiv), 130 °C, toluene, 18 h.

To elucidate the reaction mechanism, we conducted several additional experiments. Using phenyl-substituted silane 1l as starting material, we could isolate side product 5b, which enabled us to rule out its possible role as an intermediate, undergoing an intermolecular hydroallylation. As expected, mixing compounds 5b and 1l with 1.1 equiv of phenol 2b and IDPi 3a as catalyst in toluene at room temperature for 24 h did not lead to any detectable amounts of product 4l (Figure , eq 1). Further, when the reaction between substrates 1a and 2a was conducted for only 12 h at a reduced catalyst loading (1 mol %), cyclic Si-stereogenic silane 17a was isolated in 20% yield. Reaction progress kinetic studies suggested silane 17a to be a (“parasitic”) intermediate of the reaction (see the Supporting Information, Figure S5). Interestingly, the e.r. of this six-membered silane product was determined to be 53:47 (eq 2). That cyclization is indeed not the enantio-determining step was then confirmed when we reacted racemic silane 17a with phenol 2a, which cleanly furnished product 4a in 75% yield and 96:4 e.r. (Figure , eq 3). As such, the enantio-determining step is supposed to be the Si–O bond formation, and steric effects on enantiocontrol in the corresponding transition states were elucidated by DFT studies (see the Supporting Information, Figures S31 and S32 and Table S7). Furthermore, in light of the formation of product 4p from the corresponding tris-methallyl silane 1p, with significant enantioselectivity, olefin protonation as the enantio-determining step is also unlikely.

Figure 3

Control experiments. Reaction of eq 1 was performed with 5b (0.1 mmol), silane 1l (1.5 equiv), phenol 2b (1.1 equiv), and IDPi 3a (2.5 mol %) in toluene (1.0 mL, 0.1 M) at rt for 24 h. Reaction of eq 2 was performed with 2a (0.2 mmol), silane 1a (1.5 equiv), and IDPi 3e (1.0 mol %) in toluene (2.0 mL, 0.1 M) at −20 °C for 12 h and terminated by the addition of Et3N. Reaction of eq 3 was performed with rac-17a (0.2 mmol), 2a (1.1 equiv), and IDPi 3e (2.5 mol %) in toluene (2.0 mL, 0.1 M) at −20 °C for 24 h.

Based on these results, a plausible reaction mechanism can be proposed (Figure ). Accordingly, the catalytic cycle commences with the protonation of symmetrical silane 1 by IDPi 3 to provide ion pair I, the carbocation of which is stabilized by silicon hyperconjugation. Subsequent cation−π cyclization takes place to afford the ion pair II. Deprotonation of its cyclic cation gives compound 17, an isolable intermediate that can reversibly be protonated to regenerate ion pair II. Alternatively, Si–C bond cleavage would lead to silylium-based ion pair III. Finally, reaction of this intermediate with phenol 2a furnishes product 4 and regenerates catalyst 3.

Figure 4

Proposed mechanism.

In conclusion, we have realized an organocatalytic asymmetric synthesis of Si-stereogenic silyl ethers that proceeds via a C–C bond forming desymmetrization and is enabled by our IDPi catalysts. Various non-natural, enantioenriched silane products could be generated and were utilized in the synthesis of valuable silane derivatives with potential application in material and medicinal chemistry. Our approach features scalability, broad substrate scope, operational simplicity, and mechanistic novelty. Particularly, we observed and characterized an unprecedented six-membered cyclic chiral silane, the protonated form of which may act as an intermediate in the catalytic cycle. Our newly developed strategy provides a practical and efficient access to Si-stereogenic compounds, which may find utilization in the synthesis of silicon-containing materials, pharmaceuticals, and chiral ligands for transition-metal-catalyzed reactions.