Organic Letters
Letter
a b
,
Scheme 4. Evaluation of the Substrate Scope by Variation of Activated Alkenes
a
Unless otherwise specified, reactions were performed on a 0.2 mmol scale of alkenes 3b−y using 1.5 equiv of tert-butyl bromide (1a), 1.5 equiv of
the anhydride 2a, 20 mol % of NiBr2·diglyme, 30 mol % of ligand L1, 3 equiv of Zn, and 1 equiv of MgCl2 in 0.5 mL of DMAc at room temperature
b
c
d
1
for 24 h. Yields of the isolated products after column chromatography. Reaction was performed on a 6 mmol scale. Determined by H NMR
e
spectroscopy. Reactions were performed with 2 equiv of 2a, 2e, 2p, or 2r, 4 equiv of Zn, 2 equiv of MgCl2, and 4 Å molecular sieves.
delivering the coupling products 4aai−aam, 4apm, and 4aen
bearing a quaternary center in moderate to good yields. This
method was also applicable to the α,β-unsaturated amide,
ketone, and sulfone (4aao−aaq). Moreover, styrenes 3r−x with
electron-withdrawing substitution proved to be competent
precursors in this Ni-catalyzed reaction, furnishing the
corresponding products 4aar, 4apr, 4arr, and 4aas−aax in
moderate to high yields. Additionally, the reaction employing 2-
vinylpyridine (3y) also formed the alkylacylation product 4aay
in a moderate yield. Notably, this reaction could be simply scaled
up to 6 mmol, giving compound 4aac in 75% yield.
We conducted a set of control experiments to shed light on
the mechanism of this Ni-catalyzed reaction (Scheme 5). First, a
radical clock reaction using the α-cyclopropyl styrene 3z as a
precursor was carried out. In this case, the ring opening product
4aaz′ was afforded in 62% yield, whereas the formation of the
direct coupling product 4aaz was not observed (Scheme 5A).
This result indicates an addition of tert-butyl radical to the
olefinic unit, leading to the generation of a benzyl radical 5,
which could induce the cyclopropane ring opening. Next, we
performed a stoichiometric reaction using Ni(COD)2, which
delivered none of compounds 4aaz, 4aaz′, or hydroalkylation
product 4aaz″ (Scheme 5B). However, we noticed that benzoic
anhydride was completely consumed, suggesting that Ni(0)
favors an irreversible oxidative addition with benzoic anhydride
over tert-butyl bromide, and the reaction ceases at this stage.
This observation is consistent with Gong’s report.14a Sub-
sequently, Zn was replaced by tetrakis(dimethylamino)ethylene
(TDAE), and no reaction occurred in this case (Scheme 5C). To
study the effect of Zn in this reaction, we conducted the reaction
in the absence of Ni. Although the carboacylation product was
not formed, it turned out that Zn was capable of promoting the
hydroalkylation reaction, producing compound 4aaz″ in 38%
yield. In the absence of MgCl2, the yield of 4aaz″ decreased to
5%, revealing that MgCl2 could assist the Zn-mediated
hydroalkylation (Scheme 5D). When the acrylates (3a, 3i, or
3k) were used as the precursors instead of the styrene derivative
3z, the reactions without nickel could still deliver the
alkylacylation products 4aaa, 4aai, and 4aak in low yields
(Scheme 5E). These results suggest that Zn is able to mediate
the addition of alkyl bromides to the alkenes in a radical
pathway, whereas nickel plays a crucial role in the step for
installation of an acyl moiety. Furthermore, Ni(0) proved to be
able to mediate the hydroalkylation process, as well, albeit in an
efficiency lower than that with Zn (Scheme 5F). In addition, no
hydroalkylation reaction occurred in the absence of Ni(COD)2.
On the basis of the preliminary results of the mechanistic
investigations, we proposed a plausible catalytic cycle for the
major reaction pathway in Scheme 6. Initially, a Ni(0) species I
was formed under the reductive conditions, which undergoes
subsequent oxidative addition with the anhydrides 2 to afford a
Ni(II) intermediate II. Meanwhile, the alkyl bromides 1 are
reduced by Zn or a low-valent Ni species,15 and the resultant
alkyl radical III performs a Giese addition16 to the electron-
C
Org. Lett. XXXX, XXX, XXX−XXX