Green Chemistry
Communication
Table 1 Screening of bidentate ligandsa
powerful ligand. The excellent yield was maintained with a
catalyst loading of 1 mol%, provided that the copper to ligand
ratio was maintained at 1 : 2 (entry 14, cf. entries 13 and 15).
The reaction required more time to complete; however,
elevating the temperature to 130 °C allowed completion in
3 hours with 1 mol% catalyst and within 24 hours with
0.1 mol% catalyst (entries 16 and 17). Coupling with bromo-
benzene also proceeded with excellent yields down to 0.1 mol%
catalyst loading, but required a somewhat higher temperature
and longer reaction time (entries 18–20). Chlorobenzene only
resulted in trace amounts of the product.
Aryl halides with both electron withdrawing and electron
donating groups coupled to imidazole gave good to excellent
yields (Table 3). In contrast to the tendency normally observed,
the highest yields were observed with electron donating
substituents.
To explore its scope, the protocol was investigated with
diverse nitrogen coupling partners (Table 4). Coupling of
4-nitroaniline with bromobenzene or iodobenzene under
standard conditions gave 76% and 88% yield, respectively
(entries 1 and 2). Unsubstituted indole was coupled with iodo-
and bromobenzene in excellent yield (entries 3 and 4).
Selective N-arylation of the indole of melatonin in the presence
of the acetamide proceeded smoothly (entry 5). Furthermore,
5-methoxytryptamine was selectively N-arylated at the indole in
quantitative yield with no trace of coupling to the aliphatic
amine (entry 6).9 Several primary and secondary aliphatic
amines were investigated, including 3-amino-1-propanol,
piperidine and N,N-dimethylpropylenediamine, but with no
reaction taking place. This prompted us to also investigate
unprotected histamine as a substrate. Again, we observed
clean N-arylation at the aromatic ring, with a 4.3 : 1 ratio
between arylation of the less hindered Nτ-position and the
more hindered Nπ-position in 91% and 72% combined yield
with iodobenzene and bromobenzene, respectively (entries 7
and 8). Coupling of histamine with 3-iodotrifluoromethyl-
benzene also proceeded smoothly (entry 9). To our surprise,
N-arylated histamines are virtually unknown and neither
Nπ- nor Nτ-phenylhistamine has been reported previously.
We therefore also performed the same reaction using an
established procedure in toluene,9 resulting in only 21% of the
Nτ-coupled product together with 4% of the Nα,Nτ-diarylated
product (see the ESI†). This confirms that histamine indeed is
a difficult substrate. Thus, the present protocol can facilitate
easy access to previously unexplored histamine analogues.
Interestingly, N-arylation of 2-aminobenzimidazole pro-
ceeded with complete selectivity for the aromatic position with
both iodobenzene and bromobenzene (entries 10 and 11).
In the extension of these results, purines were explored as sub-
strates for our protocol. N-Arylpurines have previously been
synthesised by coupling with arylboronic acids. Apart from
nucleophilic aromatic substitutions, only a small handful of
examples of direct coupling between aryl halides and purines
have been reported in low to moderate yields, and none with
xanthines or adenine. Thus, these compounds also represent a
difficult substrate class. Satisfactory yields and high selectivity
Entry
Ligand
Time (h)
Yieldb (%)
1
2
3
4
5
6
7
8
9
None
A
B
C
21
21
21
21
21
21
48
48
48
20
38
58
35
48
93
46
38
75
D
E (DPPhen)
F
G
H
a Reaction conditions: iodobenzene (0.5 mmol), methyl indole-3-
carboxylate (0.6 mmol), KOH (1.0 mmol), CuBr (5 mol%), ligand
(10 mol%), PEG-400 (0.1 mmol), water (1 mL), 100 °C, N2 atmosphere.
b Isolated yields.
The reaction conditions with E were investigated further to
determine the robustness of the reaction (Table 2). The coup-
ling of imidazole with iodobenzene has frequently been
studied,3b,c,e,i,6a,b,10 and we adopted this reaction to facilitate
comparison with these reports. Coupling under the same
conditions as in Table 1 produced quantitative yield (Table 2,
entry 1). Lowering the temperature to 60 °C required a sig-
nificantly extended reaction time but also resulted in excellent
yield (entry 2). When the temperature was elevated to 130 °C
in a microwave reactor the reaction was complete in 3 hours
(entry 3). Changing the copper source produced good to
excellent yields for Cu(I)-salts (entries 4 and 5) but poor yield
with CuCl2 (entry 6). Bases such as Cs2CO3 or K3PO4 are often
used for similar reactions,6 but did not perform well under
the conditions used here, most likely a result of their much
lower basicities in water than in aprotic solvents (entries 7 and
8). Decreasing the amount of PEG-400 below 20 mol% also
resulted in lower yields (entries 9 and 10). PEG-400 contributes
to solubilise the reactants and produce homogenous reaction
mixtures, and the required amount can vary somewhat with
the substrate (see Table 4). Changing the ratio between copper
and ligand to 1 : 1 gave a moderate drop in yield (entry 11),
whereas a 1 : 3 ratio gave a more pronounced drop (entry 12),
probably due to sequestering of copper by an excess of the
This journal is © The Royal Society of Chemistry 2013
Green Chem., 2013, 15, 336–340 | 337
Engel-Andreasen, Jens
