Synlett 2017; 28(19): 2604-2608
DOI: 10.1055/s-0036-1590962
cluster
© Georg Thieme Verlag Stuttgart · New York

Stereospecific Nickel-Catalyzed Borylation of Secondary Benzyl Pivalates

R. Martin-Montero
a   Institute of Chemical Research of Catalonia (ICIQ), Av. Paisos Catalans 16, 43007, Tarragona, Spain   Email: rmartinromo@iciq.es
,
T. Krolikowski
a   Institute of Chemical Research of Catalonia (ICIQ), Av. Paisos Catalans 16, 43007, Tarragona, Spain   Email: rmartinromo@iciq.es
,
C. Zarate
a   Institute of Chemical Research of Catalonia (ICIQ), Av. Paisos Catalans 16, 43007, Tarragona, Spain   Email: rmartinromo@iciq.es
,
a   Institute of Chemical Research of Catalonia (ICIQ), Av. Paisos Catalans 16, 43007, Tarragona, Spain   Email: rmartinromo@iciq.es
,
R. Martin*
a   Institute of Chemical Research of Catalonia (ICIQ), Av. Paisos Catalans 16, 43007, Tarragona, Spain   Email: rmartinromo@iciq.es
b   Catalan Institution for Research and Advanced Studies (ICREA), Passseig Lluis Companys, 23, 08010, Barcelona, Spain
› Author Affiliations

MINECO (CTQ2015-65496-R & Severo Ochoa Excellence Accreditation 2014-2018, SEV-2013-0319) and Cellex Foundation.
Further Information

Publication History

Received: 01 August 2017

Accepted after revision: 26 October 2017

Publication Date:
08 November 2017 (online)

 


Published as part of the Cluster C–O Activation

Abstract

A stereoselective nickel-catalyzed direct borylation of enantioenriched secondary benzyl pivalates is described. This methodology is characterized by an intriguing cooperativity of simple nickel and copper salts to promote the targeted C–B bond formation under mild reaction conditions. Unlike classical SN2-type processes, this protocol occurs with a neat retention of configuration, resulting in synthetically versatile benzyl boronic esters with excellent stereochemical fidelity.


#

In recent years, C–O electrophiles have received considerable attention as counterparts in a myriad of metal-catalyzed cross-coupling reactions.[1] Such interest primarily arises from the lack of toxicity and readily accessibility of alcohols, as well as the possibility of implementing orthogonal techniques in the presence of organic halide partners, thus becoming powerful synthetic alternatives for building up molecular complexity. Despite the wealth of literature data when forging C–C bonds,[1] the design of enantioselective cross-coupling reactions via functionalization of C–O electrophiles other than particularly activated organic sulfonates such as aryl triflates or tosylates is still in its infancy (Scheme [1], paths b vs a).[2]

Zoom Image
Scheme 1 Cross-coupling reactions of ester derivatives via C–O cleavage

Although the employment of chiral ligands has become routine in enantioselective C–C bond-forming reactions,[3] an attractive alternative in these endeavours consists of the implementation of stereospecific coupling reactions.[4] In these rather appealing scenarios, the reaction outcome is dictated by the stereochemistry of the starting precursor, thus avoiding the need for sophisticated, oftentimes expensive, ancillary chiral ligands. Illustrative examples are the elegant protocols described by Jarvo[5] or Watson[6] that make use of enantioenriched benzyl C–O electrophiles to promote a variety of stereospecific C–C bond-forming reactions with well-defined organometallic reagents, allowing to reliably transfer the stereochemical information in the starting precursors with high fidelity. Despite the advances realized, stereoselective C–O bond-cleavage reactions remain predominantly confined to C–C bond formations (Scheme [1], path b).[5] [6] In sharp contrast, the paucity of stereoselective C–heteroatom bond-forming reactions of C–O electrophiles other than particularly activated organic sulfonates is certainly striking,[7] a rather surprising observation if one takes into consideration the prevalence of C–heteroatom bonds in pharmaceuticals.[8] A remarkable step forward has recently been described by Tobisu and Chatani,[9] in which 2-pyridyl benzyl ethers can be used for such purposes via chelation control. However, the stereoselective C–heteroatom bond formation of benzyl esters still remains elusive. Prompted by our ongoing interest in C–O bond functionalization[10] and by the versatility of organoboranes as synthons in organic synthesis,[11] we wondered whether a stereospecific borylation of enantioenriched benzyl pivalates could be implemented (Scheme [2]). Herein, we present our investigations towards this goal, demonstrating the synergy of Ni and Cu catalysts for promoting a stereoretentive borylation event of benzyl pivalates under mild conditions, constituting the first example of a stereospecific C–heteroatom bond formation of readily accessible organic ester derivatives.

Zoom Image
Scheme 2 Ni/Cu-catalyzed stereospecific borylation of benzyl ester derivatives

Table 1 Screening of the Reaction Conditionsa

Entry

Deviation from standard conditions

2a (%)b

ees (%)c

1

none

95

95

2

IPr·HCl as the ligand

2

n.d.

3

PCy2Ph as the ligand

90

92

4

dcpe as the ligand

62

78

5

P i Pr3 as the ligand

0

6

THF as solvent

54

88

7

Dioxane as solvent

47

91

8

using PCy3 (20 mol%)

40

91

9

AgF in lieu of CsF

21

77

10

CuBr2 in lieu of CsF

0

11

in the absence of CuF2

84

93

12

in the absence of CsF

75

92

13

no CuF2, no CsF

56

93

14

in the absence of PCy3

0

15

no CsF, with CuF2 (1 equiv)

84

86

16

no CuF2, with CsF (1 equiv)

73

85

17

no Ni(COD)2

0

a Reaction conditions: 1a (0.20 mmol), B2nep2 (0.30 mmol), Ni(COD)2 (7.5 mol%), PCy3 (7.5 mol%), CuF2 (30 mol%), CsF (30 mol%) in PhMe, 50 °C.

b GC yields using decane as internal standard.

c Calculated by HPLC of the corresponding oxidized alcohol due to the inherent instability of 2a.

We began our study by evaluating the borylation reaction of 1a (99% ee) with B2nep2 (B2nep2 =5,5,5′,5′-tetramethyl-2,2′-bi-1,3,2-dioxaborinane). After a systematic optimization of the reaction conditions,[12] we found that a cocktail containing Ni(COD)2 (7.5 mol%), PCy3 (7.5 mol%), CuF2 (30 mol%), and CsF (30 mol%) provided the best results at 50 °C, affording 2a in 95% yield and 95% ees (Table [1], entry 1).[13] HPLC analysis revealed that the borylation event occurred with a neat retention of configuration. As expected, an intimate interplay of all reaction parameters was critical for success (Table [1], entries 2–7). For instance, while the presence of CsF or CuF2 was not absolutely required, their inclusion led to improved yields and enantioselectivities (Table [1], entries 11 and 12).[14] The latter is particularly noteworthy, indicating that nickel and copper salts might cooperatively promote the targeted C–O bond cleavage/C–B bond formation in high yields and excellent stereochemical fidelity under mild reaction conditions.[15] As expected, the nature of the ligand exerted a profound influence on the reaction outcome, with PCy3 providing the best results (Table [1], entries 2–5). Note, however, that a slight increase on the catalyst loading of PCy3 had a deleterious effect, obtaining 2a in lower yields (Table [1], entry 8). Intriguingly, no reaction took place in the absence of ligand (Table [1], entry 14) whereas lower reactivity was found when using more sterically encumbered B2pin2 as the boron source. Control experiments in the absence of Ni(COD)2 revealed that no borylation occurred, recovering quantitatively 1a in 99% ee (Table [1], entry 17).

Bolstered by these initial results, we next turned our attention to study the generality of our stereoselective borylation via C–O bond cleavage of benzyl ester derivatives (Scheme [3]). As shown, a host of differently substituted enantioenriched benzyl pivalates, easily within reach from the corresponding benzyl alcohols obtained via Corey–Bakshi–Shibata (CBS) reduction,[16] delivered the targeted compounds in good yields and enantioselectivities. The reaction worked equally well regardless of whether electron-rich (2b) or electron-poor substituents were included on the aryl backbone (2c). Unfortunately, the employment of quinoline (2e) or phenanthrene analogues (2f) resulted in a markedly loss of enantiomeric excess; while the former might suggest competitive binding of the nitrogen atom to the metal center, the loss of fidelity in the latter can tentatively be ascribed to racemization of the enantioenriched oxidative addition species by bimolecular mechanisms with exogeneous low-valent Ni(0)Ln.[5e] Substrates possessing aliphatic side chains other than methyl groups resulted in lower enantioselectivities (2hm). These observations suggest that transmetalation might be hampered by the presence of larger alkyl side chains, setting the basis for the low yields are tentatively attributed to parasitic β-hydride elimination events that might erode the chiral integrity by subsequent migratory insertion. In line with this notion, non-negligible amounts of homobenzyl borylation were observed in the crude reaction mixtures. Unfortunately, non-π-extended aryl pivalates were not suited as substrates for the targeted borylation event.[17]

Zoom Image
Scheme 3 Stereospecific nickel-catalyzed borylation of benzyl pivalates.a,b a As Table [1] (entry 1). b Isolated yields, with enantiomeric excesses of starting precursors in parentheses. c Ni(COD)2(10 mol%), PCy3 (20 mol%), CsF (1 equiv). d Ni(COD)2 (10 mol%), PCy3 (10 mol%), CuF2 (50 mol%), CsF (50 mol%), 55 °C. e Ni(COD)2 (5 mol%), PCy3 (5 mol%), No CuF2, CsF (50 mol%). f Ni(COD)2 (10 mol%), PCy3 (10 mol%).

Independently on the starting precursor utilized, the borylation of 1am invariably occurred with a neat retention of configuration. This observation was corroborated by direct comparison of the alcohols obtained by oxidation of 1am with authentic samples obtained by CBS reduction of the corresponding ketone derivatives. Taking into consideration that both transmetalation and reductive elimination should proceed with stereoretention,[18] our results suggest a scenario consisting of an initial oxidative addition occurring with retention of configuration. In line with recent literature reports,[6a] [5f] we propose that the pivalate motif initially binds the nickel catalyst, setting the stage for a directed SN2′ oxidative addition that generate a putative π-benzyl nickel(II) intermediate I (Scheme [4], top). Such observation tacitly explains the considerably higher reactivity of extended π-systems, as partial dearomatization is required en route to the corresponding π-benzyl species I. In line with this notion, little reactivity, if any, was found with regular arenes, a recurrent limitation found in a myriad of catalytic C–O bond-functionalization technologies.[19] [20] Consistent with a SN2′ oxidative addition by attack of the Ni(0)Ln at the ortho position, we found that benzyl pivalates containing ortho substituents (1n,o) failed to provide the targeted benzyl boronate intermediates. In these cases, a site selectivity switch en route to linear boronic esters 2n and 2o was obtained (Scheme [4], bottom). The formation of these products likely arises from a sequence consisting of a β-hydride elimination followed by a migratory insertion prior to a final C–B bond-forming reaction that takes place selectively at the periphery to avoid the clash with the proximal ortho substituent on the arene.

Zoom Image
Scheme 4 Mechanistic hypothesis

Prompted by the pivotal role of organoboron reagents as synthetic intermediates,[11] we next turned our attention to study the prospective impact of this protocol by exploring the reactivity of the corresponding benzyl boronic ester intermediates (Scheme [5]). Among the different alternatives, we found that in situ generated 2a could trigger a Suzuki–Miyaura reaction[21] with a catalytic protocol based on Pd2(dba)3/PPh3,[22] resulting in the formation of 3 in excellent overall yield from (S)-1a, thus constituting an alternative to existing methodologies for the preparation of enantioenriched diarylmethanes. Consistent with a well-precedented stereoretentive transmetalation,[18] the C–C bond formation occurred with retention of configuration. Inspired by the reaction conditions reported by Zweifel,[23] we next wondered whether α-vinyl arenes could be within reach from 2a. As shown in Scheme [4] (bottom, right), this turned out to be the case. Specifically, we found that a simple exposure of in situ generated 2a to vinylmagnesium bromide followed by addition of I2 and NaOMe/MeOH at –78 °C resulted in 4 in good overall yield and exquisite stereochemical fidelity.

Zoom Image
Scheme 5 Synthetic applicability

In summary, we have developed a stereospecific nickel-catalyzed borylation of enantioenriched benzyl pivalates with neat retention of configuration.[24] This method is characterized by the cooperativity of Cu and Ni salts to effect the targeted C–B bond-forming event with excellent stereochemical fidelity by a mechanism consistent with a stereoretentive oxidative addition. This work constitutes the first C–heteroatom bond formation from enantioenriched benzyl ester derivatives. Current investigations are focused on extending the scope of these reactions beyond C–B bond-forming reactions.


#

Acknowledgment

C. Zárate and R. Martin-Montero thank MINECO and La Caixa Foundation for predoctoral fellowships.

Supporting Information



Zoom Image
Scheme 1 Cross-coupling reactions of ester derivatives via C–O cleavage
Zoom Image
Scheme 2 Ni/Cu-catalyzed stereospecific borylation of benzyl ester derivatives
Zoom Image
Scheme 3 Stereospecific nickel-catalyzed borylation of benzyl pivalates.a,b a As Table [1] (entry 1). b Isolated yields, with enantiomeric excesses of starting precursors in parentheses. c Ni(COD)2(10 mol%), PCy3 (20 mol%), CsF (1 equiv). d Ni(COD)2 (10 mol%), PCy3 (10 mol%), CuF2 (50 mol%), CsF (50 mol%), 55 °C. e Ni(COD)2 (5 mol%), PCy3 (5 mol%), No CuF2, CsF (50 mol%). f Ni(COD)2 (10 mol%), PCy3 (10 mol%).
Zoom Image
Scheme 4 Mechanistic hypothesis
Zoom Image
Scheme 5 Synthetic applicability