Enantioselective difunctionalization of alkenes by a palladium-catalyzed Heck/borylation sequence

A palladium catalyzed enantioselective Heck/borylation reaction of alkene-tethered aryl iodides was realized, delivering a variety of 2,3-dihydrobenzofuranyl boronic esters in high yield with excellent enantioselectivity. Asymmetric synthesis of chromane boronic ester, indane boronic ester and indoline boronic ester was also accomplished. The protocol offers an efficient access to the corresponding chiral benzocyclic boronic esters, which are notably important chemical motifs in synthetic transformations.


Introduction
Versatile transformations of carbon-boron (C-B) bonds have been recognized as an ideal platform for the preparation of value-added molecules. 1 Much effort has been devoted to developing efficient methods for the construction of C-B bonds, 2 in which, transition-metal-catalyzed asymmetric carboboration of alkenes represents one of the most step-and atom-economic tools for facile access to enantioenriched borylated compounds. 3 Over the past few years, asymmetric carboboration of alkenes has been elegantly achieved by using dual metal synergistic or single metal catalysis, such as Cu/Pd-, 4 Pd-, 5 Cu-6 or Ni 7 -catalysis (Scheme 1a). Despite the breakthroughs in this process, its application in the construction of chiral boron-containing benzocycles is underdeveloped.
Benzocyclic skeletons are important building blocks prevalently found in biologic and pharmaceutical molecules. 8 Owing to organoboron compounds' high propensity of functionalization, the introduction of boron into benzocyclic skeletons can provide a promising avenue to expediently access valuable functionalized molecules bearing benzocyclic skeletons by downstream synthetic manipulation. In 2018, an enantioselective intermolecular Cu-catalyzed borylacylation to access chiral borylated 3,3-disubstituted oxindoles was remarkably demonstrated by Lautens et al. (Scheme 1b). 9 As another appealing alternative, the palladium-catalyzed domino Heck/ borylation process which involves carbopalladation of a double bond and boron trapping of s-alkylpalladium species, has provided an elegant strategy to form borylated compounds bearing benzocyclic skeletons. 10 Despite major progress in the racemic Heck/borylation sequence, the exploration of its asymmetric variants to efficiently furnish chiral boroncontaining benzocyclic compounds has remained an unresolved issue. Recently, Hooper's group realized the synthesis of borylated indanes in moderate enantioselectivity through the palladium-catalysed carboborylation reaction. 11 Tong et al. developed a Pd-catalyzed asymmetric vinylborylation of (Z)-1iodo-dienes with B 2 Pin 2 for enantioselective construction of 3,3disubstituted tetrahydropyridines. 12 On the basis of these ndings and the studies toward asymmetric cascade Heck reactions in our group, 13 we tended to exploit a highly enantioselective Heck/borylation sequence for the construction of the C-B bond starting from alkene-tethered aryl iodides and boronic esters. However several issues make the protocol challenging: (1) a direct Miyaura-type borylation seems to have an edge; 14 (2) competitive side reactions such as the reductive Heck reaction and carboiodination should be avoided; 13a,b (3) the transmetalation step may take place prior to the alkene insertion, which could alter the steric environment and affect the enantioselectivity. Herein, we report a highly enantioselective Pd-catalyzed tandem Heck/borylation sequence, conveniently accessing various chiral benzocyclic boronic esters in good yields (Scheme 1c).
Further screening of other reaction factors with (S,R S )-N-Me-Xu3 as the chiral ligand showed a subtle effect of solvents. Inferior results were obtained when the reaction was conducted in THF (Table 1, entry 2). However, both the reactivity and enantioselectivity were obviously promoted using Et 2 O as the solvent (Table 1, entry 3). Both toluene and chlorinated solvents failed to provide better results (Table 1, entries 4-6). Bases turned out to exert a profound inuence on the reaction outcomes, and Cs 2 CO 3 was still the best choice (Table 1, entries 7-10). The choice of the palladium source seemed not pivotal to the system, since relatively similar results were obtained when other palladium salts such as Pd(OAc) 2 , [PdCl(ally)] 2 and Pd(dba) 2 were employed (Table 1, entries 11-13). Noteworthy, the ee value was further increased to 91% when H 2 O was added (Table 1, entry 14). Finally, the loading of the catalyst and the reaction temperature could be lowered down without compromising the yield and the enantioselectivity (Table 1, entry 15).
Having optimized the enantioselective tandem Heck/ borylation protocol, we evaluated the compatability and stereochemical delity. Excitingly, the optimized reaction conditions were generally compatible with various substituents having different electronic properties on the allyl moiety (Scheme 2). Besides the methyl substrate 1a, alkenes with other linear alkyl groups including Et, n Pr and n Bu all worked well to furnish the desired products 2b-2d. Substrates equipped with sterically hindered branched groups such as i Pr and t Bu substituents all worked smoothly under the standard conditions, offering products 2e and 2f with high yields and ee values, suggesting the insensitivity of the system to the steric effect.  Moreover, benzyl allyl ethers bearing diverse functional groups including methyl, halogens (F and Cl) and CF 3 substituents at the para-position of the phenyl ring underwent this tandem Heck/Borylation smoothly, delivering the corresponding products 2g-2k in good yields with 92-94% ees. Substituents at the meta-and ortho-positions of the phenyl ring were also tolerated, and the desired products 2l-2r were formed in good yields with 90-94% ees. Besides, boronic esters 2s and 2t with di-or tri-uorine-substituted phenyl rings were prepared in 71-88% yields with 92-94% ees. Moreover, naphthyl and dibenzothienyl ancillaries reacted smoothly, delivering the corresponding compounds 2u and 2v in good yields with excellent ee values. Benzofuranyl boronic esters with the TMS moiety could also be achieved uneventfully (2w). The chlorinated alkyl group was also accommodated without any side reaction (2x). Notably, the reactions of substrates derived from phthalimide (1y) and indole (1z) worked equally well. Besides, the 2-phenylallyl substrate (1aa) could also deliver the desired product in moderate yield and with good enantioselectivity. Furthermore, we investigated the substrate scope with respect to various 2-iodophenol derivatives. As shown in Scheme 2, halogens (F and Cl), methyl, phenyl and cumyl located at the C5 position of 2,3-dihydrobenzofuranyl boronic esters (4a-4e) were obtained in good yields with excellent enantioselectivities. The desired product 4f with t Bu substituted at the C6 position could be smoothly generated (91% yield and 90% ee). The 2-naphthol-and indol-5-ol-derived allyl ethers 3g and 3h also worked well, leading to 4g and 4h with 93% and 80% ee, respectively. Next, we tried to assemble a six-membered product employing o-iodophenol-derived homoallyl ether 3i. Gratifyingly, the desired chromane 4i was generated with good yield and enantioselectivity. Further attempt to produce the indane product was found to be completely successful. Specifically, the substrate with a carbon linker (3j) showed excellent reactivity, providing 4j with the satisfactory result. The generality of this system was also demonstrated by the tolerance of the substrate with AcN as a tether, and the indoline adduct 4k was produced in 92% yield with 53% ee.
A gram-scale reaction was conducted to verify the practicability of our methodology. Starting from 5.0 mmol of allyl ether 3d, dihydrobenzofuran 4d was obtained in 91% yield and with 95% ee with half-loading of the Pd-catalyst and (S,R S )-N-Me-Xu3 (Scheme 3). The structure of 4d was conrmed by single crystal X-ray diffraction, and its absolute conguration was unambiguously determined. The versatility of the borylated 3,3-disubstituted 2,3-dihydrobenzofuran scaffold was demonstrated by Scheme 2 Exploration of the substrate scope.
carbon-boron bond construction. Oxidation with hydrogen peroxide and sodium phosphate monobasic afforded alcohol 5 in 99% yield with 94% ee (Scheme 3a). Treatment with KHF 2 converted 4d into the corresponding organotriuoroborate 6 in 98% yield (Scheme 3b). Allyl and furyl groups were successfully introduced by stereospecic coupling of the boronic ester 4d with vinyl magnesium bromide and lithiated furan (Scheme 3c and d).

Conclusions
In summary, we have developed a palladium-catalyzed enantioselective 1,2-carboboration of alkenes by a tandem Heck/ borylation approach via trapping of the s-alkylpalladium intermediate with a boron reagent. The methodology performs well over a broad scope of substrates, providing facile access to a series of 2,3-dihydrobenzofuran, chromane, indane and indoline in high yields with excellent enantioselectivities. The synthetic potential of our protocol was illustrated by further functionalization via transformations of the boron moiety to generate a library of compounds that could serve as medicinally relevant building blocks.

Data availability
Full experimental and characterisation data are provided as part of the ESI. †

Author contributions
Y. W. carried out most of the experiments and wrote the initial manuscript dra. L. W. screened the initial reaction conditions and supported the synthesis of substrates. B. X. performed part of the experiments. Z. Z. conceived the project. Z. Z., Y. L. and J. Z. directed the project and nalized the manuscript. All the authors co-wrote the paper. All authors discussed the results and commented on the manuscript.

Conflicts of interest
The authors declare no competing nancial interests.