Stereodivergent Olefination of Enantioenriched Boronic Esters

Abstract A stereodivergent coupling reaction between vinyl halides and boronic esters is described. This coupling process proceeds without a transition‐metal catalyst, instead proceeding by electrophilic selenation or iodination of a vinyl boronate complex followed by stereospecific syn or anti elimination. Chiral, nonracemic boronic esters could be coupled with complete enantiospecificity. The process enables the highly stereoselective synthesis of either the E or Z alkene from a single isomer of a vinyl coupling partner.

The stereodefined synthesis of multiply substituted alkenes continues to attract attention because of their importance in natural products,p harmaceuticals,a nd materials. [1] Of the many methods that exist, the Suzuki-Miyaura cross-coupling reaction is widely used as it enables the direct union between vinyl halides and boronates. [2] However,whilst sp 2 -hybridized and primary organoborons couple efficiently,the corresponding reactions with secondary or tertiary boronic esters do not, thus limiting its broader use. [3] An attractive feature of the Suzuki-Miyaura reaction is that it is stereospecific, [4] but if one geometrical isomer of ag iven coupling partner is much easier to make than the other, as is often the case,t his again limits its wider application. [5] Herein, we address both of these issues and describe as tereospecific method for coupling secondary boronic esters with as ingle geometrical isomer of avinyl halide,thus leading to either the E or Z isomer of the coupled product.
To develop as olution to these problems,w et urned our attention to the Zweifel olefination. [6] In this process,avinyl metal is combined with ab oronic ester,r esulting in the formation of ab oronate complex. Addition of iodine to the double bond gives an intermediate iodonium ion, which triggers a1 ,2-metallate rearrangement leading to a b-iodoboronic ester.U pon treatment with methoxide,t he biodoboronic ester undergoes anti elimination to produce as ingle isomer of the resulting alkene (Scheme 1c). [7] We reasoned that if the anti elimination could be diverted to a syn elimination instead, then we should be able to access the alternative geometric isomer from the same geometry of the vinyl metal. Such syn elimination processes are known for boron, but most are specific for trialkyl boranes,t he most notable example being Zweifelsu se of cyanogen bromide, involving the intermediacyofab-bromo cyanoborane (Scheme 1a). [8] In the realm of the more atom economic and readily available boronic esters, syn elimination processes have been reported for substrates with b-positioned N-oxide (Scheme 1b), [9] and carbamate moieties. [10] However, the lack of suitable electrophiles for introducing such functionality within our envisioned manifold, led us to consider syn elimination of a b-selenoxyboronic ester.W especulated that such an intermediate could be obtained through the electrophilic addition of ArSeCl to avinyl boronate followed by oxidation. If the selenoxide could attack ab oron atom instead of ah ydrogen atom, with formation of the strong BÀOb ond providing the driving force,t hen the desired syn elimination should result. However,asuccessful process would require a) chemoselective oxidation of as elenide in the presence of an easily oxidizable boronic ester [11] and b) selective elimination of the boronic ester in the presence of a b-hydrogen atom. [12] We commenced our study with E-vinyl bromide 1 (Table 1). Lithium-halogen exchange followed by addition of enantioenriched boronic ester 2 gave the desired vinyl boronate complex. Upon addition of sodium methoxide in methanol followed by iodine we isolated coupled product 3a in 80 %yield as asingle Z-isomer with complete enantiospecificity (entry 1). [13] Moreover,w ew ere pleased to find that when the same boronate complex was treated with PhSeCl, smooth conversion into the desired b-selenoboronic ester was Scheme 1. Previous work and strategy for stereodivergent olefination. pin = pinacol.
observed. This crude intermediate could be treated with sodium methoxide in methanol, thus triggering an anti elimination to afford the product in 80 %yield as asingle Z isomer without any loss of enantiomeric purity (entry 2). In certain cases (see later) this procedure could serve as au seful alternative to the Zweifel coupling.
We next turned our attention to the development of ap rotocol for syn elimination. Upon treatment of aT HF solution of the crude b-selenoboronic ester with aqueous hydrogen peroxide we obtained the coupled product in modest yield, but as a5 5:45 E/Z ratio (entry 3). When mCPBAw as employed as an oxidant we obtained the undesired Z isomer almost exclusively (entry 4). However, these reactions showed that chemoselective oxidation of the selenide did indeed occur in the presence of the boronic ester. Remarkably,w ef ound that filtration of the crude reaction solution of b-selenoboronic ester through ashort plug of silica gel, followed by addition of mCPBAi nd ichloromethane resulted in ac omplete switch in selectivity and 3b was obtained as as ingle E isomer in 71 %y ield (entry 5). [14] Changing the oxidation solvent to THF afforded the product in slightly improved yield, and still with complete stereo-and enantiospecificity (entry 6).
Having identified optimal reaction conditions for generating either the E or Z alkene,w ee xplored scope,i nitially focusing on variation of the boronic ester (Table 2). With nonbranched secondary boronic esters,b oth coupling processes proceeded efficiently to provide the corresponding alkenes in excellent yields and levels of selectivity,t ogether with complete enantiospecificity.N otably,b oronic ester 6, bearing an electron-rich trisubstituted alkene,a nd estercontaining boronic ester 4 underwent efficient coupling with no evidence of side reactions.B enzylic,n atural-productderived and menthol-derived boronic esters 8, 12,a nd 14,  coupled smoothly under our optimized reaction conditions for selenation-oxidation (Method B; > 98:2 E/Z)b ut with reduced selectivity in the Zweifel procedure (Method A). This issue is addressed later. Avinyl boronic ester could also be employed, thus enabling the stereodivergent synthesis of Z,E or E,E dienes 17 a and 17 b in 71 %a nd 79 %y ield, respectively.W ew ere also able to extend the process to the coupling of vinyl boronic esters and organolithiums,t hus accessing styrenes 18 a and 18 b with excellent yields and high stereoselectivity. Synthesis of methyl-substituted-alkenes by Zweifel coupling has previously been reported to be challenging because of the poor migratory aptitude of am ethyl group. [6g] In line with this observation, we found that coupling of 16 with methyl lithium led to 19 a in moderate yield, an issue which is addressed later. [15] We were pleased to find that our selenation-oxidation protocol enabled the synthesis of 19 b in good yield and excellent stereoselectivity.
We were interested in the trend where bulkier coupling partners resulted in lower Z/E selectivity in the Zweifel reaction. In this process,t he normally favored anti-elimination pathway brings the two substituents into close proximity, and the barrier to elimination will increase as the groups get larger.T his scenario may allow the less favored syn-elimination process to compete,t hus leading to am ixture of isomers (Scheme 2A). We rationalized that if we could disfavor the syn-elimination pathway further by using ap oorer leaving group,f or example as elenide in place of an iodide,high Z selectivity should be restored. Therefore,we turned to the reaction conditions we had previously developed for methoxide-promoted anti elimination of a b-selenoboronic ester (Table 1, entry 2).
Gratifyingly,w hen we carried out the cross coupling of benzylic boronic ester 8 under these conditions,t he coupled product 9a was obtained in 63 %y ield as as ingle Z-isomer (Scheme 2B). Moreover,when these conditions were applied to other Z-selective couplings that had previously given lower Z selectivity,the products 11 a, 13 a and 15 a were all obtained in good to excellent yields as as ingle alkene isomer. Additionally,u nder these conditions,t he coupling of 16 with MeLi proceeded smoothly,t hus affording isomerically pure 19 a in 61 %yield.
We next turned our attention to varying the vinyl halide coupling partner,f ocusing our attention on trisubstituted vinyl bromides,a ss tereospecific synthesis of trisubstituted alkenes is often more difficult (Table 3). [16] We were delighted to find that commercially available E-2-bromobut-2-ene (20) could be successfully coupled with enantioenriched boronic ester 2 to afford either isomer of the coupled product with excellent yields and stereoselectivity. [17] Thes ame coupled products could be obtained through stereodivergent coupling of isomeric Z-bromide 22.F or many trisubstituted vinyl halides only one geometrical isomer can be readily accessed. Forexample,vinyl bromide 23,prepared stereoselectively by hydrozirconation of the corresponding alkyne, [18] underwent coupling to afford either 24 a or 24 b in excellent yields and with near perfect stereocontrol. Similarly,v inyl bromide 25, prepared from 2-butyn-1-ol by hydroxyl-directed hydroalumination, [19] underwent stereodivergent coupling with 2 to afford the coupled products 26 a and 26 b in 61 %a nd 55 % yield, respectively.Finally,the methodology can be applied in settings relevant to complex molecule synthesis,asillustrated with boronic ester 27,w hich is readily prepared in high Zselectivity by cross-metathesis.Reaction with an alkyl lithium derived from the Roche ester and subsequent olefination gave either the E or Z trisubstituted alkene with high selectivity. Z-alkene 28 a represents the C9ÀC17 fragment of discodermolide and its ease of synthesis is especially notable.InNovartissformidable synthesis of discodermolide the synthesis of this trisubstituted alkene was one of the most challenging reactions they encountered. [20] Thep utative syn elimination of b-selenoxyboronic esters was investigated by DFT calculations using the B3LYP functional with ac c-PVDT(H,C)/cc-PVTZ(B,O)/RECP-DZ-(Se) basis set. Both diastereomers of the b-selenoxyboronic ester (diastereomeric at the selenium center) which would give (E)-but-2-ene (Scheme 3) show global minima involving astrong interaction between the selenoxide oxygen atom and the boron atom. These conformers were primed to undergo elimination with remarkably low barriers (0.4-2.2 kcal mol À1 ), Scheme 2. Highly Z-selective olefination. [a] Vinyl bromide (1.05 equiv), tBuLi (2.1 equiv), THF, À78 8 8C; then boronic ester (1.0 equiv), THF, À78 8 8C; then PhSeCl (1.2 equiv), THF, À78 8 8CtoRT; then NaOMe (5.0-20.0 equiv), THF/MeOH (1:1), 0 8 8CorRT.

Angewandte Chemie
Communications 788 www.angewandte.org which were more accessible than rotation about the Se-C-C-B dihedral. [21] Tr aditional selenoxide eliminations,involving the expulsion of phenylselenenic acid from other conformers, were also calculated for comparison, and showed significantly higher barriers.I nterestingly,e limination to give the vinyl boronic ester was calculated to be considerably more facile than elimination to give the allyl boronic ester (5.6-7.0 versus 10.0-11.0 kcal mol À1 ), thus suggesting that hydrogen atoms geminal to trivalent boron substituents are more activated. This observation is in agreement with related eliminations of b-sulfoxysilanes,w hich undergo faster eliminations (to give vinyl silanes) relative to nonsilyl derivatives. [12a,b] In conclusion, we have developed an efficient method for the stereodivergent coupling of vinyl halides with boronic esters.T his reaction proceeds with no requirement for at ransition metal and tolerates aliphatic,v inylic, and aromatic coupling partners.W here chiral, nonracemic boronic esters were employed, the coupling took place with complete enantiospecificity,a nd the process has been successfully applied to the stereodivergent synthesis of trisubstituted alkenes.D FT studies probing the mechanism of this interesting transformation suggest that syn elimination of a bselenoxyboronic ester is ar emarkably facile process.W e believe that this approach will find widespread application in the stereoselective synthesis of polysubstituted alkenes.