Angewandte
Communications
Chemie
Table 1: Optimization of the formation of 2a over 3 and 4.
tional groups. In order to address these issues, we furthered
1a:2a:3:4[a]
our investigation into the unique properties of these cross-
coupling reactions and herein we report the development of
a novel, solvent free, direct palladium catalyzed cross-
coupling of a wide range of distinct (hetero)arenes mediated
by organolithium reagents (Figure 1C). This simple and
straightforward one-pot procedure affords a wide variety of
biaryl products, including advanced intermediates, with
excellent yields in record coupling times and featuring low
E-factors. Moreover the need for inert conditions and the
separate formation of the organolithium partner were elim-
inated.
Entry
RLi (equiv)
[Pd] (x mol%)
1
2
3
4
5
6
7
8
s-Bu (2)
s-Bu (2)
s-Bu (2)
s-Bu (2)
t-Bu (2)
t-Bu (2)
t-Bu (2)
t-Bu (2)
t-Bu (2)
t-Bu (0.7)
t-Bu (0.7)
Pd/C (5)
57:0:24:0
57:14:13:15
57:6:3:33
54:17:4:25
66:0:34:0
36:47:18:0
35:51:14:0
32:55:13:0
5:76:19:0
8:81:11:0
0:92:8:0
Pd-PEPPSI-IPent (2)
Pd-PEPPSI-IPr (2)
Pd[P(t-Bu)3]2 (2)
XPhos/Pd2dba3 (1)
Pd[P(t-Bu)3]2 (2)
Pd-PEPPSI- IPent (2)
Pd-PEPPSI-IPr (2)
Pd-PEPPSI-IPr (2)
Pd-PEPPSI-IPr (0.1)
Pd-PEPPSI-IPr (1)
9[b]
10[b]
11[b]
In preliminary studies the Pd-catalyzed one pot homo-
coupling of 1-bromonaphthalene (1a) mediated by an alkyl
lithium to obtain binaphthalene was examined (2a,
Scheme 1). We reasoned it would be possible to bypass the
[a] Ratios determined by GC-MS. [b] Inert atmosphere was replaced by
dry air. See the Supporting Information for experimental details.
conversion which allowed to opt for the cheaper of the two.
The active form of the catalyst might be oxygen generated
nanoparticles as we recently demonstrated to be the case in
organolithium based transformations[6f] and that switching
from an inert atmosphere to dry air might promote the
formation of the active catalyst and enhance the conversion
further. We were pleased to see that under these conditions,
conversion to 2a enhanced to 76% (entry 9) with no propor-
tional increase in side products. The amount of catalyst could
be reduced to 1 mol% and of t-BuLi from two to 0.7 equiv (as
0.5 equiv are in theory needed, this actually represents a slight
excess of t-BuLi, entry 11). Overall this resulted in an 80%
isolated yield of 2a and an E-factor of 3.1 which is a significant
drop in comparison to previous, similar, methods (for
example 76[18]) towards the preparation of 2a.
These conditions proved general, providing homocoupled
products in high yields (Scheme 2). Sterically demanding
substrates such as TBDMS protected 2-bromonaphthol (1b)
and 2-bromoanisole (1c) similarly gave excellent yields
(91%, 90%). Electron rich 3,5- and 2,4-dimethoxybromo-
benzene (1d–e), who might suffer from a loss in selectivity
due to the added possibility of ortholithiation, or 4-bromo-
dimethylaniline (1 f) also gave good to excellent yields of the
homocoupled biphenyl products (2d–f, 77–96%). Aryl bro-
mides bearing electron withdrawing groups or heteroaryl
bromides also function well (2g–i, 89–96%). Employing aryl
iodides instead of aryl bromides (2a’, 2h’, 2j-m) showed no
significant difference in terms of conversion or selectivity, all
giving excellent yields of the corresponding homocoupled
biphenyl products (2a, 2h, 2j–o, 84–95%). Compound 2m
especially was highly interesting as the reaction proved
chemoselective for the iodine vs. the chloride. E-factors for
these reactions vary from 1.6 to 4.3 remaining significantly
lower than for reported methods. In the case of aryl chlorides
(2a’, 2p–r), however, lower conversions were obtained (46–
64%).
Scheme 1. Possible cross-coupling and lithium–halogen exchange reac-
tions.
need for the separate formation of the aryl lithium partner
due to the differences in kinetics between the various possible
coupling and lithium–halogen exchange reactions It was
anticipated that with the use of either t-Buli or s-BuLi the
cross-coupling between an aryl bromide (1a) and in situ aryl
lithium (1a’, leading to 2a) would occur before side reactions
that is, dehalogenation (leading to 3) or cross-coupling
between the aryl bromide and the alkyl lithium (leading to
4). This meant that upon addition of the alkyl lithium to the
aryl bromide, lithium halogen exchange would selectively
occur to form an aryl lithium intermediate which is immedi-
ately consumed in the coupling step avoiding any buildup of
the organolithium species (which would lead to 3 upon
quenching) bypassing many of the issues in previous reported
methods using organolithium reagents[6b] while retaining high
levels of selectivity. In contrast s-BuLi with various catalyst
(Table 1, entries 1–4), when t-BuLi[16] was added over a period
of 10 min onto neat 1-bromonaphthalene in the presence of
1 mol% Pd[P(t-Bu)3]2 or Pd-PEPPSI-IPent[17] about 50%
conversion to 2a occurred (entries 6 and 7). While no
formation of 4 could be observed under these conditions,
a significant amount of starting aryl bromide remained.
Switching to cheaper Pd-PEPPSI-IPr as catalyst (entry 8) saw
a slight increase in selectivity but no notable change in
Subsequently the possibility of heterocoupling of two
distinct aryl halides under the optimized conditions was
explored (Scheme 3). We envisioned that even a slight differ-
ence in the rate of lithium–halogen exchange between both
aryl halides would be sufficient to result in high selectivities
for the heterocoupled product vs. the homocoupled product.
Indeed when using 1-bromo-2,4-dimethoxybenzene (5a), in
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!