Communications
DOI: 10.1002/anie.200803312
Homogeneous Catalysis
A General and Selective Copper-Catalyzed Cross-Coupling of Tertiary
Grignard Reagents with Azacyclic Electrophiles**
Lukas Hintermann,* Li Xiao, and Aurꢀlie Labonne
Catalytic cross-coupling reactions continue to revolutionize
in the absence of a catalyst, substitution products were not
formed.[21] However, the addition of a catalytic amount (3–
5 mol%) of copper(I) iodide to the same reactant combina-
tion led to a selective catalytic cross-coupling; namely, 2a was
converted in high yield into 1a with only a trace (< 0.5%) of
the accompanying disubstituted product (Table 1). A catalytic
quantity of copper is all that is needed to achieve the highly
selective monosubstitution of dichloropyridine (2a).[22] This
simple, yet highly efficient protocol was additionally explored
for a range of chloroazacycles (Table 1).[23] Selective mono-
substitution was also observed with quinazolines 2d and 2e,
and quinoxaline 2 f (Table 1, entries 3–5). Trichloroazacycles
such as pyrimidine 2b and cyanuric chloride (2c) were readily
converted into disubstituted products (Table 1, entries 7, 8,
and 11). Limiting the amount of the Grignard reagent and
lowering of the reaction temperature led to selective mono-
alkylations (Table 1, entries 6, 9, and 10). The latter con-
ditions may be compared to those of a noncatalyzed
alkylation (2c!3c: 1128C, 3.5 h, 24% yield)[24] to illustrate
the impressive catalytic acceleration by copper(I). Trisubsti-
tution was achieved in the case of cyanuric chloride (2c) by
performing consecutive alkylations in one pot (Table 1,
entry 12). The results in Table 1 imply that the catalytic
substitution is sensitive to electronic effects, because the ease
and rate of alkylation increase with each additional chlorine
or ring-nitrogen atom in the substrate. The selectivity profile
includes a peculiar specificity for tertiary Grignard reagents;
analogous reactions of 2a with either secondary alkyl
(isopropyl, cyclohexyl) or aryl nucleophiles gave mixtures
containing monosubstituted, disubstituted, dehalogenated, or
reductively coupled products, in addition to unreacted start-
ing material.
The reaction was also extended to higher tertiary
Grignard reagents (Table 2). Dimethylalkyl (Table 2,
entries 1–7), cycloaliphatic (Table 2, entry 8), and cage-type
Grignard reagents (Table 2, entries 12 and 13) underwent the
cross-coupling reaction. Limitations became apparent only
for combinations of the most hindered Grignard reagent
reacting with the least active substrate (2a; Table 2, entry 9).
The products in Table 2 are new compounds which would be
difficult to obtain by other methods.[25] The immediate
practical value of the new cross-coupling reaction is exempli-
fied by the straightforward syntheses of compounds that have
been previously obtained by involved procedures (Scheme 1).
Phosphane 14[17] having a shielded pyridine unit is a
powerful ligand for bifunctional catalysis,[14] including the
ruthenium-catalyzed anti-Markovnikov hydration of terminal
alkynes.[15,26] The previous synthesis (6 steps, 3.3% overall
yield)[17] is now replaced by the two-step protocol shown in
Scheme 1 (62% overall yield). The tert-amylpyridylphos-
ꢀ
the way of constructing C C bonds, particularly for olefinic,
aromatic, and heteroaromatic compounds.[1] The cross-cou-
pling reactions of C(sp3)-centered substrates bearing b-
hydrogen atoms remains challenging because of the compet-
ing b-hydride elimination and isomerization[2] of the catalyst-
bound alkyl groups.[3] The catalytic cross-coupling of sterically
hindered tertiary alkyl nucleophiles is considered to be
particularly difficult,[4] and such reactions have not been
investigated systematically.[5–9] We present herein a general
and selective copper-catalyzed cross-coupling of tertiary
Grignard reagents with chloro-azacyclic electrophiles which
gives access to a range of heterocyclic building blocks for use
in supramolecular chemistry,[10] molecular opto-electronics,[11]
and pharmaceutical chemistry,[8e,12,13] and in the preparation
of valuable bifunctional ligands for transition-metal cataly-
sis.[14]
In our work on bifunctional catalysis with bulky aza-aryl
phosphane ligands,[15] we needed heterocyclic building blocks
(1a–c) containing tertiary alkyl groups and electrophilic sites
for additional synthetic elaboration. These heterocycles tend
to be synthesized in lengthy and, at times, low-yielding
condensation routes.[16,17] Alternative syntheses involving
metal-catalyzed tertiary alkylation address the above-men-
tioned problems and solve the issue of regioselective mono-
versus di- or trisubstitution in doubly or triply chlorinated
substrates.[18] Fꢀrstner and co-workers have addressed the
latter problem for the iron-catalyzed alkylations of dichlor-
oazacycles[19] with Grignard reagents, but an extension of
their protocol to tertiary alkyl nucleophiles was unsuccess-
ful.[20]
When we combined tBuMgCl with either 2,6-dichloropyr-
idine (2a) or cyanuric chloride (2c) at ambient temperature,
[*] Dr. L. Hintermann, Dr. L. Xiao, Dr. A. Labonne
Institut fꢀr Organische Chemie, RWTH Aachen University
Landoltweg 1, 52074 Aachen (Germany)
E-mail: lukas.hintermann@oc.rwth-aachen.de
[**] This work was supported by the DFG (Emmy Noether Programme).
We thank Prof. Carsten Bolm for continued support.
Supporting information for this article is available on the WWW
8246
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 8246 –8250
Hintermann, Lukas
