Palladium-Catalyzed Stereoselective Construction of 1,3-Stereocenters Displaying Axial and Central Chirality via Asymmetric Alkylations

The concurrent construction of 1,3-stereocenters remains a challenge. Herein, we report the development of stereoselective union of a point chiral center with allenyl axial chirality in 1,3-position by Pd-catalyzed asymmetric allenylic alkylation between racemic allenyl carbonates and indanone-derived β-ketoesters. Various target products bearing a broad range of functional groups were afforded in high yield (up to 99%) with excellent enantioselectivities (up to 98% ee) and good diastereoselectivities (up to 13:1 dr).


Introduction
Over the past decades, extensive demands for chiral non-racemic compounds from various fields have significantly motivated the development of asymmetric catalysis [1][2][3][4][5]. Hence numerous catalytic asymmetric methodologies have been developed for enantioselective construction of chiral structures [6][7][8][9]. Whereas many methods are available for the synthesis of chiral molecules containing one single stereocenter or even two adjacent stereocenters (Scheme 1a), much less efforts have been paid to the concurrent creation of nonadjacent stereocenters in an enantio-and diastereoselective manner, due in part to the difficulties in high levels of simultaneous stereocontrol posed by increased distance between the two chiral centers [10][11][12][13][14]. Of note, the limited reports on enantio-and diastereoselective construction of 1,3-stereocenters focused mostly on molecules bearing two point chiral centers, while those containing different types of chirality, for example, central and axial chiral motifs, have been rarely explored.
As a prominent example of axial chiral molecules, chiral allenes constitute an important structural motif widely present in a variety of organic molecules including natural products, pharmaceutical agents, chiral catalysts, and ligands for coordination chemistry etc., as represented by the selected compounds in Scheme 1b [15][16][17][18][19]. Interestingly, Enprostil [20], a prostaglandin analogue used for the treatment of acute duodenal ulcer disease, shows a distinct 1,3-stereocenters consisting of a point chiral center and allenyl axial chirality.

Optimization of the Reaction Conditions
In our preliminary investigation, allenylic carbonate 2a was selected as the model substrate for the asymmetric alkylation with β-ketoester 1a in the presence of Cs2CO3 Scheme 1. Representative allenes with different functions, different methods for the synthesis of allenes and construction of 1,3-stereocenters bearing allenyl axial and central chirality.

Optimization of the Reaction Conditions
In our preliminary investigation, allenylic carbonate 2a was selected as the model substrate for the asymmetric alkylation with β-ketoester 1a in the presence of Cs 2 CO 3 using a palladium catalyst (Table 1, entries 1-26). Initially, several classic biphosphine ligands were screened to explore the effect of the chiral ligands on the Pd-catalyzed allenylic alkylation reaction. With L1 or L3 as the ligand, product 3a could be afforded in high yields (84% and 89%) with good enantioselectivities (−71% ee and −67% ee) (entries 1 and 3). In the reaction with Trost ligand L2, no product was observed (entry 2), while L4 afforded product 3a in 55% yield with −73% ee and 15:1 dr (entry 4). Further screening led to the observation that ligands L5-L7 were sluggish for this allenylation, whereas ligand L6 afforded the product 3a with high diastereoselectivity (11:1) but low yield and enantioselectivity (20%, −7% ee) (entry 6). Then, the ligand (R)-and (S)-SegPhos were applied in this reaction. Interestingly, using ligand (S)-SegPhos (L9) the product 3a was obtained in 90% yield with 4:1 dr and 91% ee (entry 9). Subsequently, increasing the amount of 2a further increased the yield of 3a to 97%, and the enantioselectivity of 3a to 93%, respectively (entry 10). With L9 as the ligand, subsequent investigation on the solvent effect showed that THF was the best one, affording 97% yield 3a with 6:1 dr and 92% ee (entry 15). Next, we investigated the effect of base, and observed that the product 3a was formed in 98% yield, with 7:1 dr and 93% ee (entry 20), when the base of the reaction was NaHCO 3 . Further optimization was performed and found that the reaction at 0.05 M could give product 3a in 98% yield with 8:1 dr and 96% ee at −10 • C.

Substrate Scope for the Asymmetric Alkylations of Allenylic Carbonates 2
Next, the substrate scope with respect to allenylic carbonates 2 were investigated (Scheme 3). The electronic nature and position of the substituents on the arylallenyl carbonates backbone of the substrates were also examined. When a chloro group was present at the C2 position in the aryl, high yield and enantioselectivity were observed for the formation of the alkylation product 3n (80% yield, 97% ee). The products (3o and 3p) bearing electron-donating substituents were obtained in 99% yield with 93-97% ee. Meanwhile, halogenation in the aryl C4 position of the substrates (2f and 2g) were well-tolerated, giving the corresponding products (3r and 3s) with 91-97% ee and a slight decrease in yield (84-87%). Moreover, 2-naphthyl-substituted allenylic carbonate also worked smoothly (73% yield, 97% ee). Compared to model substrate 2a, various substituted arylallenyl carbonates (2b-2h) could afford the target products (3n-3t) with relatively higher diastereoselectivities as expected. In addition, substrates 2i with an n-propyl group and 2j with isopropyl afforded products 3u and 3v with slightly decreased enantioselectivities and diastereoselectivities. The asymmetric alkylation of allenylic carbonate with symmetri- cal cyclohexyl substituent delivered the product 3w in moderate enantioselectivity and diastereoselectivity, albeit with a lower yield. alkylation reaction well and afforded products (3e, 3f, and 3h-3l) in 96-99% yields, with high enantioselectivities (95-97% ee) and diastereoselectivities (7:1-12:1 dr). Notably, the six-membered cyclic β-ketoester (1m) underwent the alkylation reaction smoothly to afford product 3m in high yield and excellent enantioselectivity (99%, 97% ee), though with a moderate dr (2:1).

Scheme 2.
Substrate scope for the reactions of β-ketoesters 1 with arylallenyl carbonate 2a. The reaction was carried out on a 0.2 mmol scale with Pd2dba3 (2.5 mol%) and L9 (5 mol%) in 4.0 mL THF with NaHCO3 (0.2 mmol); the ratio of 1/2a is 1.0/1.2; isolated yields are given; the dr was determined by 1 H NMR of crude product; the ee was determined by chiral HPLC.
The reaction was carried out on a 0.2 mmol scale with Pd 2 dba 3 (2.5 mol%) and L9 (5 mol%) in 4.0 mL THF with NaHCO 3 (0.2 mmol); the ratio of 1/2a is 1.0/1.2; isolated yields are given; the dr was determined by 1 H NMR of crude product; the ee was determined by chiral HPLC.

Substrate Scope for the Asymmetric Alkylations of Allenylic Carbonates 2
Next, the substrate scope with respect to allenylic carbonates 2 were investigated (Scheme 3). The electronic nature and position of the substituents on the arylallenyl carbonates backbone of the substrates were also examined. When a chloro group was present at the C2 position in the aryl, high yield and enantioselectivity were observed for the formation of the alkylation product 3n (80% yield, 97% ee). The products (3o and 3p) bearing electron-donating substituents were obtained in 99% yield with 93-97% ee. Meanwhile, halogenation in the aryl C4 position of the substrates (2f and 2g) were well-tolerated, giving the corresponding products (3r and 3s) with 91-97% ee and a slight decrease in yield (84-87%). Moreover, 2-naphthyl-substituted allenylic carbonate also worked smoothly (73% yield, 97% ee). Compared to model substrate 2a, various substituted arylallenyl carbonates (2b-2h) could afford the target products (3n-3t) with relatively higher diastereoselectivities as expected. In addition, substrates 2i with an n-propyl group and 2j with isopropyl afforded products 3u and 3v with slightly decreased enantioselectivities and diastereoselectivities. The asymmetric alkylation of allenylic carbonate with symmetrical cyclohexyl substituent delivered the product 3w in moderate enantioselectivity and diastereoselectivity, albeit with a lower yield. Scheme 3. Substrate scope for the reactions of β-ketoester 1a with allenylic carbonates 2. The reaction was carried out on a 0.2 mmol scale with Pd2dba3 (2.5 mol%) and L9 (5 mol%) in 4.0 mL THF with NaHCO3 (0.2 mmol); the ratio of 1a/2 is 1.0/1.2; isolated yields are given; the dr was determined by 1 H NMR of crude product; the ee was determined by chiral HPLC.

Gram-Scale Reaction and Product Derivatization
To demonstrate the practicality of the transformation, a gram-scale reaction was conducted and found that the reaction of β-ketoester 1j (2.5 mmol) with allenylic carbonate 2a (3.0 mmol) gave product 3j (1.01 g) in 98% yield, with 97% ee and 12:1 dr Scheme 3. Substrate scope for the reactions of β-ketoester 1a with allenylic carbonates 2. The reaction was carried out on a 0.2 mmol scale with Pd 2 dba 3 (2.5 mol%) and L9 (5 mol%) in 4.0 mL THF with NaHCO 3 (0.2 mmol); the ratio of 1a/2 is 1.0/1.2; isolated yields are given; the dr was determined by 1 H NMR of crude product; the ee was determined by chiral HPLC.

Gram-Scale Reaction and Product Derivatization
To demonstrate the practicality of the transformation, a gram-scale reaction was conducted and found that the reaction of β-ketoester 1j (2.5 mmol) with allenylic carbonate 2a (3.0 mmol) gave product 3j (1.01 g) in 98% yield, with 97% ee and 12:1 dr (Scheme 4).

FOR PEER REVIEW 6 of 15
Scheme 4. Gram-scale reaction and product derivatization.

Plausible Mechanism of the Palladium-Catalyzed Alkylation of β-Ketoester 1
A plausible mechanism for the palladium-catalyzed allenylic alkylation of indanone β-ketoester is depicted in Scheme 5 [43,44]. Oxidative addition of 2a with Pd(0) would immediately undergo delocalization to yield intermediate A.
In the presence of NaHCO3, the nucleophile 1a attacks the terminal carbon atom of intermediate A to afford the target product 3a and regenerate Pd(0).

General Information
Unless otherwise noted, materials were purchased from commercial suppliers and used without further purification. Column chromatography was performed on silica gel (200~300 mesh). Enantiomeric excesses (ee) were determined by HPLC (Agilent, Palo Alto, CA, USA) using corresponding commercial chiral columns as stated at 30

Plausible Mechanism of the Palladium-Catalyzed Alkylation of β-Ketoester 1
A plausible mechanism for the palladium-catalyzed allenylic alkylatio β-ketoester is depicted in Scheme 5 [43,44]. Oxidative addition of 2a wit immediately undergo delocalization to yield intermediate A. In the presen the nucleophile 1a attacks the terminal carbon atom of intermediate A to af product 3a and regenerate Pd(0).

General Information
Unless otherwise noted, materials were purchased from commercial used without further purification. Column chromatography was performe (200~300 mesh). Enantiomeric excesses (ee) were determined by HPLC Alto, CA, USA) using corresponding commercial chiral columns as stated UV detector at 254 nm. Optical rotations (JiaHang Instruments, Shangha reported as follows: [α] T D (c g/100 mL, solvent). All 1 H NMR and 19 F NMR Scheme 5. Plausible mechanism of the palladium-catalyzed alkylation of β-ketoester 1.

General Information
Unless otherwise noted, materials were purchased from commercial suppliers and used without further purification. Column chromatography was performed on silica gel (200~300 mesh). Enantiomeric excesses (ee) were determined by HPLC (Agilent, Palo Alto, CA, USA) using corresponding commercial chiral columns as stated at 30 • C with UV detector at 254 nm. Optical rotations (JiaHang Instruments, Shanghai, China) were reported as follows: [α] T D (c g/100 mL, solvent). All 1 H NMR and 19 F NMR spectra were recorded on a Bruker Avance II 400 MHz (Bruker, Karlsruhe, Germany) and Bruker Avance III 600 MHz (Bruker, Karlsruhe, Germany), respectively, 13 C NMR spectra were recorded on a Bruker Avance II 101 MHz or Bruker Avance III 151 MHz with chemical shifts reported as ppm (in CDCl 3 , TMS as an internal standard). Data for 1 H NMR are recorded as follows: chemical shift (δ, ppm), multiplicity (s = singlet, d = doublet, t = triplet, m = multiplet, br = broad singlet, dd = double doublet, coupling constants in Hz, integration). HRMS (ESI) was obtained with a HRMS/MS instrument (LTQ Orbitrap XL TM, Agilent, Palo Alto, CA, USA). The characterization data is available in Supplementary Material.