B
A. R. Hajipour et al.
LETTER
(not aryl amines) were formed as single products. The temperature for this transformation. The cross-coupling
high stability of the catalyst, even under acidic conditions, reaction did not perform well at lower temperatures (60–
compared to other copper(Ι) catalysts, such as copper(I) 80 °C), and a lower amount of 4 was obtained between
oxide, and the preference for aryl azides make it a unique 90–95 °C in comparison with the one with a temperature
catalyst for copper-catalyzed reactions. Here, the catalyst of 100 °C (Table 1, entries 4–6). After that, various sol-
was efficiently prepared from Cu2+ (not Cu+) salt as re- vents were used for the optimization of the reaction.
ported by Hashimoto and co-workers.29 Butyraldehyde Among different solvents, DMF–H2O (9:1) gave better
acted as a weak reducing agent to reduce Cu2+ to Cu+ yields than other solvents (Table 1, entries 7–12).
which reacted with Fe3+ to produce CuFeO2 in the pres-
When the reaction temperature was increased to 115 °C,
ence of base. It is notable that the copper and iron ions are
the conversion rose to 55% (Table 1, entry 13). Because
in +1 and +3 oxidation states, respectively. The stability
of the instability of aryl azides at higher temperatures, in-
of CuFeO2 and copper(I) iodide was not comparable.
creasing the temperature to 125 °C led to a decreasing
CuFeO2 is stable even at 220 °C, but copper(I) iodide is
amount of product (Table 1, entry 14), and this fact had
oxidized readily at lower temperatures. CuFeO2 is also
been confirmed by Thatcher and co-workers.26 Apparent-
stable under acidic conditions, unlike copper(I) iodide,
ly, the reaction is sensitive to temperature changes, and
which may be readily oxidized in common solvents like
the optimized temperature was selected as 115 °C. When
the amount of CuFeO2 was increased to 15 mol%, the
DMF.
Therefore, in continuation of our research to introduce the yield of product was not affected much, but further in-
first heterogeneous catalyst for coupling of NaN3 with creasing the amount of catalyst led to lower yields (Table
aryl halides, we developed an efficient, convenient, and 1, entries 2 and 15). It should be noted that decreasing the
mild procedure for the cross-coupling of aryl halides and amount of catalyst from 10 mol% to 5 mol% led to a lower
sodium azide using L-proline as ligand. This methodology yield of product (Table 1, entry 16). According to these re-
gives access to a wide variety of aryl azides.
sults, the reaction conditions with 10 mol% of CuFeO2 are
the most appropriate. The increased catalytic activity of
CuFeO2, the so-called delafossites with isotypic crystal
structures, and its selectivity to form aryl azides is proba-
bly due to the interesting layered oxide structure of this
compound,. Finally, in an effort to develop better reaction
conditions, different ligands and bases were tested for the
preparation of aryl azides in the presence of 10 mol% of
CuFeO2, and the results are summarized in Table 1 (en-
tries 17–23). The results show that among the different
tested ligands, L-proline and (S)-alanine showed better
yields. Therefore, it can be concluded that this copper-cat-
alyzed azidation of aryl halides is promoted by amino ac-
ids.
To perform the reaction, we undertook an intensive
screening of reaction variables using sodium azide (1) and
4-bromoanisole (2) as a representative substrate (Scheme
1). The selection of 4-bromoanisole as a model substrate
was based on the lower amount of 4-methoxyaniline (3)
formed (46%) under Helquist’s conditions.20 Our goal
was to apply heterogeneous CuFeO2 as a catalyst and to
increase the yield of 4-methoxyaniline compared to the re-
port of Helquist and co-workers. In these screenings, a
mixture of 1 and 2 (with a mole ratio of 2:1) in EtOH–H2O
(7:3) was stirred at 100 °C in the presence of 30 mol% L-
proline and 30 mol% NaOH, and the progress of the reac-
tion was monitored by thin-layer chromatography (TLC).
To explore the generality and scope of this method, vari-
ous starting materials were used. We studied the reactivity
of various aryl bromides with sodium azide (1). Under the
optimized conditions (Table 1, entry 12), only aryl azides
were obtained as products and no aryl amines were ob-
served. All reactions showed fair yields of aryl azide de-
rivatives in the presence of air and were complete in ten
hours.
OMe
OMe
OMe
CuFeO2, ligand
base, solvent
NaN3
+
1
NH2
Br
N3
2
Some of the aryl bromides reacted well with 1, affording
the aryl azides in fair to excellent yields (Table 2, entries
2, 3, 9, 13–15), except in the cases with aryl bromides pos-
Scheme 1 Reaction of sodium azide with 4-bromoanisole
To evaluate the role of the catalyst, the reaction was per- sessing strong electron-withdrawing groups in the para
formed in the absence of catalysts, and no product was ob- position (Table 2, entries 7 and 16) or strong electron-do-
tained (Table 1, entry 1). Using a stoichiometric amount nating groups (Table 2, entry 11). The yield of product in
of catalyst, we surprisingly obtained a low amount of aryl the case of entry 16 (Table 2) was very low (almost 8%),
azide instead of aryl amine (Table 1, entry 2).
and the isolation was too difficult; also entries 7 and 11
(Table 2) represent no product. The same optimized con-
ditions were used for the coupling of different aryl iodides
with compound 1, but with a slightly different reaction
temperature. Finally, the optimized temperature was se-
lected as 95 °C for these compounds. However, some of
the aryl iodides (Table 2, entries 1, 4, 8, 18) gave good
In the next step, we carried out the same procedure with a
catalytic amount of CuFeO2 (10 mol%) and, surprisingly,
4-methoxyphenyl azide 4 (35%) was produced as a single
product instead of 3 (Table 1, entry 3). We then selected
lower and higher temperatures in order to optimize the
Synlett 2014, 25, A–E
© Georg Thieme Verlag Stuttgart · New York