8410
X. Liu et al. / Tetrahedron Letters 48 (2007) 8409–8412
Especially in the case of p-bromoacetophenone, even
though the acetyl group7 is vulnerable to nucleophilic
attack and potential enolization, the reaction still
proceeded to afford the desired product in 61% yield
(entry 9). The success of this reaction might be due to
the possibilities that the ketone arylation was slower
than the amination, or the hydrazine reacted with the
ketone to form a hydrazone thus protecting it from
arylation at the a methyl group. However, the failure
to obtain product from an arylbromide carrying a
formyl group suggested that the protection mechanism
is less likely. The reaction was also found to be effective
in the presence of an acidic N–H group (entries 12 and
13): the 4-Boc-amino phenyl bromide was aminated
with 1-aminopiperidine in 74% yield.
catalyst/ligand
base
solvent
temperature, time
N
NH2
N
Br
+
NH
3
1
2
Scheme 1.
N,N-dimethylhydrazine, using Pd[P(o-tolyl)3]2Cl2 and
BINAP or PPh3 and Cs2CO3. The reactions were carried
out at 150 °C without solvent and the scope of this
methodology is not clear. Herein, we wish to report
our effort in the synthesis of N,N-dialkylarylhydrazines
and their conversion into anilines.
In addition to 1-aminopiperidine, other N,N-dialkyl-
hydrazines have also been employed in the amination
studies (entries 2, 4, 10, 11 and 13). 1,1-Dimethylhydr-
azine reacted with aryl bromides to give yields similar to
the ones obtained with 1-aminopiperidine (entries 1–4).
The sterically more hindered SAMP [(S)-1-amino-2-
methoxylpyrrolidine] was also arylated to provide the
desired product in 75% yield (entry 13). Since dialkyl
hydrazines were readily available through a simple
transformation from dialkylamines, the amination
chemistry shown here could be used with great conve-
nience in the isosteric replacement of benzyl amines with
arylhydrazines (Fig. 1).
We began our studies with the model reaction as shown
in Scheme 1. The commercially available 1-aminopiperi-
dine, 2, was coupled with phenyl bromide, 1, under vari-
ous conditions to give hydrazine 3. The yields were
obtained through the analysis of the GC chromato-
grams, with trimethoxybenzene as an internal standard.
As expected, the reactions worked best with bulky, elec-
tron-rich phosphines such as P(t-Bu)3 as ligands;2,3
PPh3, PCy3, BINAP and DPPF failed to produce signi-
ficant amounts of product with Cs2CO3 as the base, in
toluene. Using Pd2(dba)3/P(t-Bu)3 as the catalyst sys-
tem, we also tested different bases in this reaction.
Although strong bases such as KO–t-Bu were effective,
anhydrous solvents were needed and functional groups
were less tolerated. Both K2CO3 and K3PO4 were suit-
able bases, however, Cs2CO3 provided the best yield.
The reaction proceeded in DME giving 3 in a yield simi-
lar to the one run in toluene, whereas the use of DMF as
a solvent gave inferior results. It is necessary to point
out that the reactions gave comparable yields regardless
of the moisture content in the solvent, toluene.
One of the well-known applications of hydrazines in
organic synthesis was demonstrated by the Enders
group8 in using SAMP/RAMP [(R)-1-amino-2-
methoxylpyrrolidine] as a chiral auxiliary to prepare
secondary amines enantioselectively. The success of this
methodology prompted us to study the possibility of
combining our amination chemistry with an easy
cleavage method to provide a two-step sequence for
preparing aniline derivatives. Many ammonia equiva-
lents9 were developed for the synthesis of aniline deriva-
tives from aryl bromides, each with its own advantage
and limitation. One of the advantages of using the
N,N-dialkylhydrazines as ammonia surrogates is that
the arylation chemistry tolerates a variety of functional
groups, and methods for cleavage of N–N bonds are
known in the literature.8 Furthermore, after cleavage
of the N–N bonds, the byproduct would be low mole-
cular weight sec-amines, which could be easily washed
away in the workup. With this in mind, we briefly tested
our hypothesis. We first heated hydrazine 4 with excess
borane8 in refluxing THF for 4 h, and then released ani-
line 5 from the borane complex via methanolysis. To our
dismay, the aniline was isolated only in 30% yield
(Scheme 2). We then turned our attention to palla-
dium-catalyzed transfer hydrogenation to cleave the
N–N bond. The Boc-protected amine 6 was heated at
reflux with ammonium formate and 10 mol % Pd/C in
MeOH for 12 h. To our delight, the desired aniline 7
was obtained in 75% yield. As predicted, purification
of the product was easy via a simple aqueous workup
to wash away the byproduct piperidine. This result
demonstrated the potential use of these N,N-di-
alkylhydrazines as ammonia surrogates.
With these results in hand, we then set out to explore the
scope of this reaction. One equivalent of ArBr was cou-
pled with 1.5 equiv of hydrazines in the presence of
Pd2(dba)3 (4 mol % Pd) as a catalyst, 4 mol % P(t-
5
Bu)3ÆHBF4 as a ligand and 2.5 equiv of Cs2CO3 as a
base in toluene at 100 °C [Although the reactions were
generally carried out using 4 mol % Pd as a catalyst,
phenyl bromide was successfully coupled with 1-amino-
piperidine using 1 mol % Pd as a catalyst to provide
3 in 71% isolated yield (Table 1, entry 1).]6 Electron-
neutral ArBr and sterically hindered ArBr were suitable
substrates as the o-tolyl bromide and o,o-dimethylphe-
nyl bromide were aminated in excellent yields (entries
3 and 5). Aryl bromides bearing electron withdrawing
groups such as m-trifluoromethyl also reacted with 1-
aminopiperidine to give the desired product in 61% yield
(entry 6). ArBr bearing strong electron donating groups,
such as 4-N,N-dimethylamino-phenylbromide, were also
tested, however, the results were much poorer (data not
shown).
A variety of functional groups such as cyano, ester,
ketone (entries 8–11) were tolerated in the reaction.
Liu, Xiaoxiang
