Progress in arylpiperazine synthesis by the catalytic amination reaction

Article abstract of DOI:10.1016/S0968-0896(02)00303-6

Careful base and solvent optimization for catalytic amination is described. A Pd-catalyzed amination between some arylbromide and unprotected piperazine (1 equiv) was efficiently carried out with Pd/BINAP catalyst in a toluene-DBU solvent system, which is useful for the one-pot preparation of unsymmetrical piperazine through amination and in-situ N-protection. Reaction with N-BOC-piperazine was also successful in toluene-DBU or more polar NMP with Cs₂CO₃ as a key base. No reports have previously reported such solvent and base optimization in arylpiperazine synthesis.

Full text of DOI:10.1016/S0968-0896(02)00303-6

Bioorganic & Medicinal Chemistry 10 (2002) 4023–4027
Progress in Arylpiperazine Synthesis by the Catalytic
Amination Reaction
Yasuhiro Torisawa,* Takao Nishi and Jun-ichi Minamikawa
Process Research Laboratory, Second Tokushima Factory, Otsuka Pharmaceutical Co. Ltd., Kawauchi-cho,
Tokushima 771-0182, Japan
Received 27 May 2002; accepted 3July 2002
Abstract—Careful base and solvent optimization for catalytic amination is described. A Pd-catalyzed amination between some
arylbromide and unprotected piperazine (1 equiv) was efficiently carried out with Pd/BINAP catalyst in a toluene–DBU solvent
system, which is useful for the one-pot preparation of unsymmetrical piperazine through amination and in-situ N-protection.
Reaction with N-BOC-piperazine was also successful in toluene–DBU or more polar NMP with Cs2CO3 as a key base. No reports
have previously reported such solvent and base optimization in arylpiperazine synthesis.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
Aripiprazole studies
Synthesis of arylamines and its congeners is obviously
an important task in modern medicinal chemistry.
Although a number of synthetic methods are available,
there still exists a demand for improved protocol which
allows an effective transformation in the presence of a
wide range of functional groups. We have thus been
interested in the application of some metal-catalyzed
reactions to the synthesis of Aripiprazole and other
drug substances.1,2
Aripiprazole is an important arylpiperazine derivative
developed by our company for the treatment of psy-
chotic disease (structure shown in Scheme 1).1 In rela-
tion to the study of its metabolite (8), Morita and
Uchida have revealed some beneficial aspect of the cat-
alytic amination over conventional routes.5 In this
study, a tetra-substituted arylpiperazine (7) was synthe-
sized in the yield of 94% as shown in Scheme 2. This
application clearly showed the superiority of Buchwald
amination over classical methods based on the conver-
sion from nitroaromatics. Furthermore, the precise
study (shown in Scheme 2) indicated a preferable
amount of the catalysts and selection of the amine
reagent for the reaction of 9.
We have already described a simple method for the
conversion of phenol into arylamine via the correspond-
ing triflate.3 In this study, coupling of the triflate (1)
with the aniline derivative (2) revealed the superiority of
the Pd catalyst over the Cu system. On the other hand,
some reactions were successfully carried out by the
Cu catalyst using the different substrates including the
activated halide (3).
Initial Studies
We then resumed our own process optimization for the
arylpiperazine (11) in the presence of a metal catalyst
under various base-solvent systems. A Cu-catalyzed
amination of the bromide (9) either with NH-piperazine
(10a) or N-Me piperazine (10b) afforded only less
than 15% of the desired aryl piperazine as shown in
Scheme 2. Albeit low yield, the utilization of the tol-
uene–DBU–Cs2CO3 system was noteworthy, because
almost no product was obtained from the reaction in
toluene–Cs2CO3.
Our arylpiperazine study has also disclosed a modified
Buchwald amination reaction with 4 and 5, in which the
addition of 18-crown-6 to Cs-promoted amination was
fruitful, as shown in Scheme 1.4
*Corresponding author. Tel.: +88-665-2126; fax: +88-637-1144;
e-mail: torisawa@fact.otsuka.co.jp
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00303-6

4024
Y. Torisawa et al. / Bioorg. Med. Chem. 10 (2002) 4023–4027
Contrary to the results with Cu catalyst, a survey with the
Pd–BINAP system (Buchwald amination) demonstrated
again the efficiency of the Pd catalyst. In the transfor-
mations shown above (9–11), 60% unoptimized yield
was attained in the initial survey using the toluene–
DBU–NaOtBu system with excess piperazines and Pd
catalyst (5 mol%).
Our study then focused on the Pd-catalyzed amination
with nearly 1equiv NH-piperazine with minimum catalyst
loading (1mol%) .
Amination with NH-Piperazine
Careful surveying started with employing less than
1 mol% Pd catalyst. Selected results are summarized in
Table 1. We were interested in the effect of added
organic base such as DBU, which can form a clear
solution and assist nuclephilic substitution by amines.
Compared with the previous results, this modification is
particularly useful in a one-pot operation to unsymme-
trical arylpiperazine. The notable features are as fol-
lows:
Scheme 1. Previous study.
ꢀ Addition of DBU formed a colorless clear solution
at first, which indicated DBU substrates (9, 10a)
interaction before reaction starts (see the structure
shown in Scheme 3).
ꢀ Addition of tBuONa started reaction in exothermic
t
way, but addition of BuONa before the piper-
azine led to a messy and sluggish reaction
(ꢁ45% yield with by-products).
ꢀ DBU could neither work as ligand nor as a key
t
base like BuONa. This indicated an interaction
between ligand–Pd–DBU(piperazine) complex
and tBuONa before reductive elimination as
delineated in Scheme 3.
We noticed here that the interaction of the amine and
Pd was an initial key factor for the successful conversion.
The reactions were strongly dependent on the amine
counterpart employed. Some unwanted complexation or
further interaction between amine and the catalyst
strongly affected the catalytic cycle. We thus tentatively
Table 1. Amination with 1 equiv NH-piperazine
9/10a
Catalysts
Base
11c
10 mM/12 mM
Pd(dba) (0.1 mM),
BINAP (0.3mM)
NaOBut
(added last)
75%
Pd(OAc)2 (0.2 mM)
BINAP (0.6 mM)
NaOBut
(added last)
70%
45%
Pd(OAc)2 (0.2 mM)
BINAP (0.6 mM)
NaOBut
(added before 10a)
Pd(OAc)2, BINAP (0.2/0.6)
Pd(OAc)2, BINAP (0.2/0.6)
Pd(OAc)2, P(o-tol)3 (0.2/0.6)
Cs2CO3
(added last)
<10%
25%
Cs2CO3
(added last)
NaOBut
(added last)
Trace
Scheme 2. Aripiprazole study.

Y. Torisawa et al. / Bioorg. Med. Chem. 10 (2002) 4023–4027
4025
speculate that DBU might assist such proper interaction
of the catalyst with one molar amine that can form a
crucial intermediate shown in Scheme 3.
indicated that a mixture of products was obtained with
tBuONa as a base.
To overcome the messy reaction caused by alkoxide
base (or NaH), we attempted the reaction of 9 and 10c
with Cs2CO3 in DMA and NMP. A recommended
reaction sequence was shown (Scheme 4), in which a
preformed cesium amide worked as a good precursor
for this amination in 80% yield (see Experimental). It
should be noted that crushed Cs salt was recommendable
and the reactions were almost complete within 4 h with
vigorous stirring under Ar atmosphere. Among the
cesium salts examined, Cs2CO3 was superior to CsF,
CsOH, CsHCO3 and CsOAc.
Whatever the true nature is, the reaction was a facile
process which can afford the N-protected arylpiper-
azines in one step from arylbromide and NH-piperazine
in ca. 75% overall yield through the sequential amination
and protection in the same flask. Our protocol thus
offers operational simplicity as well as efficient conver-
sion in a short reaction time, which is superior to the
existing protocols, usually requiring excess piperazine to
prevent the formation of bis-piprazine by-products, and
requiring long reaction time. None of the reported pro-
cedures attained such short time conversion with 1 equiv
of NH-piperazine without the assistance of DBU as a
co-solvent, which is summarized later for clarity.
Pd-Catalyzed Amination with DPPP or DPPB
It is also desirable to find a simple ligand that can
replace expensive BINAP. Search for an alternative
ligand is currently under active investigation, which
highlighted the bulky trialkylphosphines or others.6,7
Here, we did not focus on these special candidates,
which sometimes require special precautions in handling
or preparation. Instead, we have been interested in the
simplest example of the chelating phosphines DPPP and
DPPB.
Amination with N-BOC-Piperazine
We further attempted an improved catalytic amination
with mote bulky N-BOC piperazine (10c). The availability
of this derivative in bulk quantity now prompted us to
uncover its utility. As summarized in Table 2, the con-
version again indicated the superiority of DBU-based
protocol, especially in its fast and clean reaction in
comparison with the Cu catalyst. As indicated in the
last run, comparable reaction with Cu catalyst gave only
25% of the product with recovery of largely starting
material.
Results with such ligands as DPPP and DPPB in toluene–
DBU media are summarized in Table 3, which clearly
indicates that piperazine formation was successful only
in moderate yields.
Further reaction in more common polar solvents such
as DMF, DMA and NMP (N-methyl piperidinone)
In all cases in Table 3, o-dichlorobenezene was obtained
as a major product, which was also formed in the reaction
with PPh3 or P(o-Tol)3. These results clearly indicated
the bulkiness of phosphine is essential. In other words,
Scheme 4. Improved amination in NMP.
Table 3. Reaction with other ligands
Scheme 3.
Table 2. Amination with N-BOC piperazine
9/10a
Catalyst/solvents
Conditions
70–120ꢂ, 6 h
11c
10 mM/12 mM
Pd/dba (0.05 mM),
DPPP (0.12 mM)
DBU/tol
45%
Catalysts
10c
Reflux (h)
4
11c
Pd/dba (1 mol%)
BINAP (3mol%)
2 equiv
80%
Pd/dba (0.05 mM)
BIPPB (0.12 mM)
DBU/tol
70–120ꢂ; 6 h
70–120ꢂ; 46 h
3 0%
30%
Pd/dba (1 mol%)
BINAP (3mol%)
1.2 equiv
2 equiv
6
6
55%
25%
Pd(OAc)2 (0.24 mM)
BU3P (0.29 mM)
DPPP (0.24 mM) DBU/tol
Cul (20 mol%)
CuCO3 (120 mol%)

4026
Y. Torisawa et al. / Bioorg. Med. Chem. 10 (2002) 4023–4027
large aromatic ring (binaphtyl) is essential for key
reductive coupling reaction. We are now searching for a
readily available bulky phosphine, which has a tether
that can interact with DBU such as COOH group.
Thus, the order of the addition of reagents was an
important factor, which governed this catalytic amina-
tion. The key base BuONa should be added after all
reagents were mixed [ligand–Pd–DBU(amine) complex
formation] in the flask. Exothermic amination reaction
immediately started on warming around 70 ꢂC.
t
As reviewed in a recent article, reactions in some ionic
liquids or by the assistance of N-heterocyclic carbene
ligands will open another avenue of the aryl amination
and other important transformations.11
In the reaction with bulky and less reactive N-BOC-
piperazine, solvent optimization was critical for the effi-
cient conversion. We also found an optional system, in
which DMA and NMP was a good solvent in combi-
nation with a Cs bases. In this reaction of Cs base, pre-
formation of a Cs amide species was preferable. Thus,
Cs base was first mixed with amine and substrate,
before the addition of catalysts. In all reactions men-
tioned above, more solvated mixture was obtained than
those reported before.
Related studies
Besides our study, many other studies have dealt with
the same topics. Some of them should be summarized
below for comparison with our protocols. The synthesis
of arylpiperazines via Pd-catalyzed amination with
unprotected piperazine (10a) was first reported in 1996
by Zhao (Roche),8 in that the isolated yield of the
product has reached 50% with 4 equiv excess of piper-
azine by PdCl2[P(o-tol)3]2–NaOt-Bu combination. The
major problem was the bis-arylation reaction resulting
in the formation of undesired N,N-disubstituted
bisarylpiperazines.
Obviously, further investigation is necessary to find the
optimum ligand–solvent combination for the important
arylpiperazines, and one promising direction was indi-
cated in the aforementioned review article.11 Progress
along these lines will be reported in due course.
In a search for the effective ligand for catalytic amina-
tion, Nishiyama (Tosoh)9 discovered the new catalyst
system: Pd(II)/PtBu3 in o-xylene (i.e., at higher reaction
temperature), which shows high product (Ar–N/Ar–H)
selectivity and high TON: 6400 (mol product/mol Pd).
A wide variety of arylpiperazines including N-hetero-
arylpiperazines can be prepared using this system with
an excess amount of NH-piperazine (2–6 equiv).
Experimental
Materials and instrumentation
Chemicals were purchased from the commercial firms
indicated and used without further purification except
for brief crushing prior to use. TLC analysis was carried
1
out using Merck silica gel 60 F254 plate (Art 5715). H
and 13C NMR spectra were measured at 300 and
75 MHz, respectively, with tetramethylsilane (TMS) as
the internal standard.
For the work on 5-HT1A ligands, Thomas (Knoll Pharma-
ceuticals)10 reported a convenient procedure for the
bicyclic arylpiperazines, which involved the reaction of
N-BOC-piperazine (10c) with bicyclic bromoarenes
under Pd2dba3–P(o-tolyl)3 system, followed by depro-
tection of the BOC group by TFA. Yields were
strongly dependent upon the nature of the bromoarene,
ranging from 20 to 60%. This protocol has found wide
unitity as a standard procedure in medicinal chemistry
research.
Pd-catalyzed amination with NH-piperazine (Table 1)
A well-dried flask (200 mL) was first charged with the
starting bromide (9, 10 mM, TCI) and NH-piperazine
(10a, 12 mM, Wako), which was evacuated and back-
filled with argon through a balloon under gentle warming
(40 ꢂC). Solvent (dry toluene, 15 mL) was charged and
the mixture was bubbled with Ar for ca. 10 min, before
the two catalysts were delivered (BINAP and pd2dba3:
the amount indicated in Table 1; BINAP was purchased
from Kanto Chemical Co and pd2dba3 from Aldrich).
After the addition of DBU (TCI, dried by MS 4A and
flushed with Ar for 5 min; 1–2 mL) to the mixture, the
clear solution was warmed at 60–70 ꢂC while fine powder
of tBuONa (Wako, 15 mM initially, or other bases
shown in Table 1) was added in one portion to start
amination. After being heated for a while (1–4 h), the
dense colored mixture was monitored by tlc (AcOEt/
n-hexane) to reveal the formation of the polar products.
(Further amount of base was delivered if reaction was
incomplete.)
Concluding Remarks
In our present study, we have briefly investigated and
summarized some Cu and Pd catalyzed pathways to the
important arylpiperazine intermediate for Aripiprazole
and others. The simple and effective protocol was
developed through a careful solvent optimization
focusing on DBU, which can be easily removed in
extractive workup (see Experimental). Among the
amine additives examined, DBU was particularly
noteworthy because it is effective in both Cu- and
Pd-catalyzed reactions. DBU has previously been utilized
as a good solvent and base for such Pd-catalyzed reaction
as amidocarbonylation reaction, in which amine is a
nucleophile involved. We thus assume that DBU might
assist the ligand–Pd–DBU(amine) complex as depicted
above.
After cooling to rt, the mixture was diluted with ethylene
dichloride (dry reagent; 10–15 mL) and further treated
with Boc2O (TCI, 20–30 mM) to form a single less polar
spot on the tlc. After dilution with AcOEt–H2O, the

Y. Torisawa et al. / Bioorg. Med. Chem. 10 (2002) 4023–4027
4027
products were extracted with AcOEt and extracts were
washed well with H2O to remove polar material. The
crude products obtained after evaporation of the dried
extracts was further passed through a SiO2 short col-
umn to afford the very pure product as a faint yellow
caramel in the yields indicated in Table 1.
(BINAP: 30 mg and pd2dba3: 20 mg). Resulting solution
was warmed at 70–80 ꢂC for 1 h to iniꢂtiate the reaction.
After further heating for 3h at 110 C with vigorous
stirring, the dense colored mixture was monitored by tlc
(AcOEt/n-hexane) to reveal the formation of the single
polar product.
For the structural identification, the initial N-Boc-aryl
piperazine (11c) was further deprotected by the treat-
ment of excess HCl in MeOH (most conveniently from
AcCl+MeOH) to furnish the known hydrochloride, on
evaporation of the mixture, as a white solid, which was
identical with the commercially available sample from
Lancaster Co.
After cooling to rt, the mixture was diluted with
AcOEt–H2O, extracted with AcOEt and extracts were
washed well with H2O to remove polar material.
The crude products obtained after evaporation of the
extracts gave a nearly pure material, which was further
passed through a SiO2 column to afford the very pure
product (11c) as a yellow caramel in 80% yields.
Pd-catalyzed amination with N-BOC-piperazine (Table
2)
References and Notes
In a well-dried flask (200 mL) was first placed the starting
bromide (9, 10 mM) and N-BOC-piperazine (10b,
12 mM, Lancaster), which was evacuated well and
backfilled with argon through a balloon under gentle
warming to melt the crystals (40–50 ꢂC). Solvent (dry
toluene, 15 mL) was charged and the mixture was well
mixed and exchanged with Ar, before the two catalysts
were delivered (BINAP and pd2dba3: the amount indi-
cated in Table 2). After the addition of DBU (2 mL) to
the mixture, the resulting clear solution was warmed at
60 C while fine powder of BuOK (14 mM) was added
in one portion to start the reaction. After heating for
4 h, the mixture was monitored by tlc (AcOEt/n-hexane)
to reveal the formation of the single polar product (11c).
After cooling to rt, the mixture was diluted with
AcOEt–H2O, further extracted with AcOEt and extracts
were washed well with H2O to remove polar material.
The crude products obtained after evaporation of the
dried extracts was further passed through a SiO2 col-
umn to afford the very pure product (11c) as caramel in
the yields indicated in Table 2.
1. Oshiro, Y.; Sato, S.; Kurahashi, N.; Tanaka, T.; Kikuchi,
T.; Tottori, K.; Uwahodo, Y.; Nishi, T. J. Med. Chem. 1998,
41, 658.
2. Oshiro, Y.; Sakurai, Y.; Sato, S.; Kurahashi, N.; Tanaka,
T.; Kikuchi, T.; Tottori, K.; Uwahodo, Y.; Nishi, T. J. Med.
Chem. 2000, 43, 177.
3. Torisawa, Y.; Hashimoto, A.; Furuta, T.; Minamikawa, J.
Bioorg. Med. Chem. Lett. 1997, 7, 1821.
4. Torisawa, Y.; Nishi, T.; Minamikawa, J. Bioorg. Med.
Chem. Lett. 2000, 10, 2489.
5. Morita, S.; Kitano, K.; Matsubara, J.; Ohtani, T.; Kawano,
Y.; Otsubo, K.; Uchida, M. Tetrahedron 1998, 54, 4811.
6. (a) For recent reviews, see: Hartwig, J. F. Angew. Chem.,
Int. Ed. Engl. 1998, 37, 2046. (b) Wolfe, J. P.; Wagaw, S.;
Marcoux, J.-F.; Buchwald, S. L. Acc. Chem. Res. 1998, 31,
805. (c) Belfield, B. J.; Brown, R. G.; Foubister, A. J. Tetra-
hedron 1999, 55, 11399.
7. (a) For recent topics, see: Ali, M. H.; Buchwald, S. L. J.
Org. Chem. 2001, 66, 2560. (b) Parrish, C. A.; Buchwald, S. L.
J. Org. Chem. 2001, 66, 3820. (c) Huang, X.; Buchwald, S. L.
Org. Lett. 2001, 3, 3417. (d) Hundertmark, T.; Littke, A. F.;
Buchwald, S. L.; Fu, G. C. Org. Lett. 2000, 2, 1729. (e) Tomori,
H.; Fox, J. M.; Buchwald, S. L. J. Org. Chem. 2000, 65, 5334. (f)
Yin, J.; Buchwald, S. L. Org. Lett. 2000, 2, 1101. (g) Bolm, C.;
Hildebrand J. Org. Chem. 2000, 65, 169. (h) Zhang, C.; Trudell,
M. L. Tetrahedron Lett. 2000, 41, 595. (i) Watanabe, M.;
Nishiyama, M.; Yamamoto, T.; Koie, Y. Tetrahedron Lett. 2000,
41, 481. (j) Prashad, M.; Hu, B.; Lu, Y.; Draper, R.; Har, D.;
Repic, O.; Blacklock, T. J. J. Org. Chem. 2000, 65, 1158.
8. Zhao, S.-H.; Miller, A. K.; Berger, J.; Flippin, L. A.
Tetrahedron Lett. 1996, 37, 4463.
ꢂ
t
Pd-catalyzed amination with N-BOC-piperazine in NMP
In a well-dried flask (200 mL) was first placed the starting
bromide (9, 10 mM) and N-BOC-piperazine (10c,
12 mM), which was evacuated and backfilled with argon
through a balloon under gentle warming to melt the
solids (40 ꢂC). Solvent (dry NMP, 10 mL) was charged
and crushed cesium carbonate (Aldrich, 15 mM; or
other cesium salts such as CsF) was then added in one
portion. The mixture was warmed further to form a
purple solution, which was well mixed and exchanged
with Ar, before the two catalysts were delivered
9. Nishiyama, M.; Yamamoto, T.; Koike, Y. Tetrahedron
Lett. 1998, 39, 617.
10. Kerrigan, F.; Martin, C.; Thomas, G. H. Tetrahedron
Lett. 1998, 39, 2219.
11. Herrmann, W. A. Angew. Chem., Int. Ed. 2002, 41, 1290.