(N,N,N′,N′-tetramethylethylenediamine), and TMHD (2,2,6,6-
tetramethyl-3,5-heptanedione) resulted in low yields (Table
1, entries 6-8). Next, we probed the solvent effect and found
that DMSO was considerably superior to dioxane, toluene,
or DMF (Table 1, entries 9-11). Finally, we found that the
nature of the bases had a pronounced impact on the process.
Cs2CO3, K3PO4, DBU, and KOH were all ineffective (Table
1, entries 12-15). We compared the conventional preheated
oil bath reaction with microwave reactions, and we found
that the yield decreased in the latter case (Table 1, entries 5
and 16). In summary, the optimum results were obtained
when amine (1.0 equiv) and aryl halide (1.5 equiv) were
allowed to react with Fe2O3 (0.1 equiv), L-proline (0.2 equiv),
and NaOtBu (2.0 equiv) stirred in DMSO at 135 °C for 24 h.
Table 1. Optimization of the Catalysis Conditions
yield
(%)
entry
catalyst
ligand
base
solvent
1
2
3
-
-
FeCl3
-
NaOtBu DMSO
53
52
77
67
85
52
66
75
trace
0
L-proline NaOtBu DMSO
L-proline NaOtBu DMSO
4
Fe(ClO4)2·6H2O L-proline NaOtBu DMSO
5
6
7
8
Fe2O3
Fe2O3
Fe2O3
Fe2O3
Fe2O3
Fe2O3
Fe2O3
Fe2O3
Fe2O3
Fe2O3
Fe2O3
Fe2O3
L-proline NaOtBu DMSO
DMEDA NaOtBu DMSO
TMEDA NaOtBu DMSO
Having determined the optimized conditions, we examined
the scope of the process with respect to amine substrates.
We applied the new process to a variety of nitrogen-
containing compounds including aliphatic primary amines,
aliphatic secondary amines, benzylamine, anilines, and
nitrogen heterocycles with iodobenzene. The desired ami-
nation products were obtained in moderate to good yields
(Table 2). We noticed that all primary amines worked well,
affording the corresponding coupling product in good yields
(Table 2, entries 1-4). The steric hindrance of the secondary
amines often makes them more sluggish than primary amines
toward coupling with aryl halides. Remarkably, we found
that aliphatic secondary amines also worked well (Table 2,
entries 5-7). Although satisfactory results were obtained in
the case of primary amines, poor coupling yields were
obtained for benzylamine (Table 2, entry 8). Moderate yields
were observed for nitrogen heterocycles such as pyrazole,
indole, and benzoimidazole (Table 2, entries 9-11). As
shown, anilines containing an electron-withdrawing group
typically afforded lower yields than anilines and those
containing an electron-donating group (Table 2, entries
12-14).
We then investigated the scope of the process with respect
to aryl halides (Table 3). We tested a variety of substituted
aryl halides under the optimized reaction conditions with
morpholine as the model amine. As expected, the corre-
sponding N-arylation products were obtained in moderate
to excellent yields. Aryl iodides and aryl bromides were more
reactive than aryl chlorides and gave the corresponding
N-arylated products in higher yields. The coupling reactions
of dihalogenated aryl halides with morpholine were also
tested, and the chlorides showed lower reactivity as compared
with the bromides and iodides. The results indicate the
reactivity order of aryl halides: iodides > bromides .
chlorides (Table 3; entries 1, 10, and 11). In general, no
significant electronic effects were observed for the electron-
rich and electron-poor substituted aryl halides (Table 3;
entries 2, 5, 8, and 10). However, the steric effect was
significant. The reactions of para- or meta-substituted aryl
halides afforded higher yields (Table 3; entries 2, 3, 5, 6,
and 9), while more steric ortho-substituted aryl halides were
less reactive. Only a trace of the product was observed when
2-iodobenzotrifluoride was coupled with morpholine (Table
3; entries 4 and 7). Furthermore, we investigated the coupling
TMHD
NaOtBu DMSO
9
L-proline NaOtBu DMF
L-proline NaOtBu toluene
L-proline NaOtBu dioxane
10
11
12
13
14
15
16a
0
0
L-proline Cs2CO3
L-proline K3PO4
L-proline DBU
L-proline KOH
DMSO
DMSO
DMSO
DMSO
trace
0
0
L-proline NaOtBu DMSO
51
a Performed under microwave, 160 °C, 30 min.
are limited in terms of the substrates for which they can be
used; in particular, they are ineffective for aniline and
alkylamine derivatives. Therefore, iron-catalyzed C-N cross-
coupling reactions need to be modified to expand the scope
of these methods and to employ more universal ligands. In
this study, we wish to report efficient Fe2O3-catalyzed
N-arylation of aryl halides with various amines using
L-proline as the ligand. In comparison to currently existing
methods of the C-N bond formation, our proposed approach
has several distinguishing features that are worth mentioning:
(i) the approach employs an environment-friendly and
economically competitive catalytic system that is a combina-
tion of readily available iron salts and an universal ligand;
(ii) the approach offers experimental simplicity and can be
performed without the need for protection from air or
moisture; (iii) in this approach, a broader scope of substrates,
both aliphatic and aromatic amines and various substituted
aryl halides, can be applied.
Initially, we carried out a set of experiments using
iodobenzene (1.5 equiv) and morpholine (1.0 equiv) as model
substrates for optimizing the reaction conditions, and the
results are summarized in Table 1. Only a moderate amount
of the expected product was formed without the iron source
(Table 1, entries 1 and 2). However, the product was obtained
in 77% yield when FeCl3 and L-proline were used as the
catalyst and ligand in DMSO, respectively, at 135 °C. Next,
we screened the iron sources, and preliminary results showed
that iron salts such as Fe2O3 and Fe(ClO4)2·6H2O were also
effective when L-proline was used as the ligand (Table 1,
entries 4 and 5). The most efficient and air-stable Fe2O3 was
selected to examine the effect of the ligands. Attempts
to use DMEDA (N,N′-dimethylethylenediamine), TMEDA
4514
Org. Lett., Vol. 10, No. 20, 2008