Mono N-arylation of piperazine(III): Metal-catalyzed N-arylation and its application to the novel preparations of the antifungal posaconazole and its advanced intermediate

Article abstract of DOI:10.1016/S0040-4039(02)00556-7

A novel application of Pd⁰-catalyzed arylation to mono N-arylated piperazines, its mechanism, and its application towards the novel syntheses of the key differentially N,N′-diarylated piperazine antifungal intermediate N-(4-hydroxyphenyl)-N′-(4-aminophenyl)piperazine 5 as well as posaconazole 1 are described.

Full text of DOI:10.1016/S0040-4039(02)00556-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 3359–3363
Mono N-arylation of piperazine(III): metal-catalyzed
N-arylation and its application to the novel preparations of the
antifungal posaconazole and its advanced intermediate
Michael Hepperle, Jeffrey Eckert, Dinesh Gala,* Lan Shen, C. Anderson Evans and
Andrew Goodman
Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
Received 25 January 2002; revised 7 March 2002; accepted 12 March 2002
Abstract—A novel application of Pd0-catalyzed arylation to mono N-arylated piperazines, its mechanism, and its application
towards the novel syntheses of the key differentially N,N%-diarylated piperazine antifungal intermediate N-(4-hydroxyphenyl)-N%-
(4-aminophenyl)piperazine 5 as well as posaconazole 1 are described. © 2002 Elsevier Science Ltd. All rights reserved.
Mono and diarylated piperazines are key moieties of a
variety of biologically active compounds such as Neu-
ropeptide Y antagonists, G-Protein-coupled receptor
ligands, cholesterol ester transfer protein inhibitors,
5-HT1D receptor agonists GPIIb/IIIa antagonist, sero-
tonin-3 antagonists, and orally active broad-spectrum
triazole antifungals (1–4), the compounds of interest for
the current manuscript.1–7 These triazole antifungals
such as Sch 51048, Sch 56592, itraconazole, and
hydroxyitraconazole3 utilize an unsymmetrically N,N%-
diarylated piperazine moiety 5.
has been in seeking a mild, selective method for mono-
N-arylation of piperazine which could be utilized for
preparation of 5.
In previous publications1,2 we described a direct sequen-
tial N-arylation of piperazine nitrogens for the prepara-
tion of the common antifungal intermediate 5. There it
was shown that the reaction of equimolar amounts of
electron deficient arylhalides with piperazine, without
the use of catalyst, led to the mono N-arylated piperazi-
nes which were then converted to the differentially
N,N%-diaryl substituted piperazines. Electron rich aro-
matic halides do not react with piperazine under these
conditions. Thus, towards the synthesis of antifungal
intermediate 5 we reported that the reaction of piper-
azine 7 with p-chloro or bromonitrobenzene 6 led to
the mono arylated piperazine 8 (Scheme 1), which
reacted with boronic acid derivative 10 to generate 11.
N
N
NH2
HO
5: Common antifungal intermediate
Typical synthetic routes1–6 to these differentially N,N%-
diaryl substituted piperazines encompass the reaction of
anilines to the mono N-aryl piperazines (via carcino-
genic intermediates under high pressure and/or temper-
ature and specialized catalyst conditions) which are
then reacted with appropriate arylhalides. Alternately
these differentially substituted piperazines are made by
reacting electron withdrawing groups containing aryl-
halides either with a large excess of piperazine under
harsh reaction conditions or with mono-N protected
piperazine, usually as a carbamate, to avoid a mixture
of mono-, di-, and unsubstituted piperazines which
require chromatographic purifications.7,8 Our interest
Interestingly,
under
standard
Pd0
coupling
conditions9,10 8 reacted very slowly with p-bromoan-
isole 9 as after 3 days it led to 58% isolated yield of 11.
Compound 11 produced via either of the above reac-
tions were indistinguishable from the one obtained via
alternate synthesis and could be converted to the
desired compound 5 as shown in Scheme 1. For a large
scale preparation of 5, the use of 9 is more desirable
compared to the use of 10.11 The slow arylation of
piperazine 9 with 8 was intriguing. Very little informa-
tion is available in the literature pertaining to Pd0-cata-
lyzed N-arylation of piperazine.12 In view of novel and
useful results obtained in our laboratory on piperazine
* Corresponding author.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00556-7

3360
M. Hepperle et al. / Tetrahedron Letters 43 (2002) 3359–3363
O2N
Br
+
HN
NH
N
NH
O2N
8
6
7
(HO)2B
10
OMe
Br
OMe
9
Pdo, ref. 13
Cu(OAc)2/DMSO
H2/Pd/C
HBr
5
N
N
OMe
O2N
Ref. 1
11
Scheme 1. Reactivity of N-p-nitrophenyl piperazine.
N-arylation with electron deficient aromatic halides,1,2
we decided to investigate Pd0-catalyzed piperazine N-
arylation with non electron deficient aromatic halides in
general. This work has led us to the discovery that
almost equimolar amounts of non-electron deficient
arylhalides and piperazine react in the presence of a Pd0
catalyst to produce only mono N-arylated piperazines.
This finding has been applied to the preparation of the
common antifungal intermediate 5 as well as to
posaconazole, 1, and are described in this manuscript.
was imparted by insolubility or specific set of reagents.
Isolated 18 subjected to the reaction conditions with
fresh reagents and the catalyst essentially resulted in a
recovery of 18. While this mechanism remained under
investigation, we applied the above finding to the syn-
theses of 5 as detailed below.
As shown in Scheme 2, arylhalide 6 was reacted with 18
using conditions described previously for N-arylation
of piperazine with electron deficient aryl halides1 to
obtain compound 11 indistinguishable from the one
made via alternate routes shown in Scheme 1. Note that
the one-pot, two-step conversion, i.e. Pd0 coupling to
crude 18 followed by its reaction with 6, without acid
base work up, can be carried out in one pot to obtain
75% isolated yield of 11. The residual reagents and
byproducts form Pd0 coupling reaction did not interfere
with the conversion of 18 to 11 and the removal of one
acid–base work up minimized the loss of mass resulting
in an improved yield. Compound 11 was then con-
verted to the desired antifungal intermediate 5 via
demethylation followed by hydrogenation. All yields
reported in this manuscript are upoptimized. In this
preparation of 5, demethylation requires an excess of
refluxing aqueous HBr for a prolonged period (24+ h)
of time. The isolation of nitrophenol 22 from HBr
requires a pH control, it is labor intensive, and moder-
ate yielding (75–80%).
The results of piperazine mono N-arylation with Pd0
catalysis (experimental conditions derived from Ref. 9
are described in Ref. 13) are shown in Table 1. In
general the stronger the electron donating group on the
arylhalide, the higher the piperazine N-arylation yield.
A steric hindrance played a role (entries 4 and 5) in
lowering yields. Most importantly for our work, com-
pound 9 and 15 reacted with piperazine to produce
intermediates that were suitable for the preparation of
5.
Preliminary laboratory work undertaken to understand
factors that govern Pd0-catalyzed piperazine mono N-
arylation is detailed below. The use of racemic Binap in
place of chiral one during the formation of 18 did not
alter the outcome of the reaction ruling out any facial
selectivity imparted by the catalyst. Solubility of 9,
11–15 and mono N-aryl piperazines 16–21 in toluene,
the use of several lots of reagents, ligands, as well as
catalyst eliminated the possibility that the selectivity
Changing the reaction order, i.e. hydrogenation prior
to demethylation encountered difficulties as well.
Although insolubility of 11 in many solvents slowed
down the reduction, the main problems resided in the
fact that the resultant aniline intermediate formed a
HBr salt with aq. HBr. Thus, yet larger excess of HBr
and much longer (2+ days) refluxing time was needed
for the demethylation reaction. These difficulties were
overcome by the use of benzyloxy group in place of
methoxy group (Scheme 3). Benzyloxy derivative 23 is
more soluble compared to 11, and its deprotection does
not require harsh HBr treatment. The reaction of piper-
azine with 15 in the presence of Pd0 catalyst resulted in
mono N-arylated piperazine 21. The use of larger
amount of the catalyst allowed for improved yield of 21
with short reaction time. Compound 21 readily reacted
with the arylhalide 6 under the conditions described
previously1 to form 23. Compound 23 was then con-
Table 1. Pd0-catalyzed piperazine mono N-arylation
R
R
Pdo
HN
NH
N
NH
Br +
16-21
9, 11-15
7
c
R
Product/% yield
Reaction time (h)
1
2
3
4
5
6
H (11)
164/42
17/69
18/82
19/35
20/10
21/80
24
48
24
48
96
314
4-CH3 (12)
4-OCH3 (9)
2-CH3 (13)
2-OCH3 (14)
4-OCH2Ph (15)

M. Hepperle et al. / Tetrahedron Letters 43 (2002) 3359–3363
3361
Pdo, ref. 10
H2/Pd/C
HBr
6
N
N
NO2
HO
N
NH
5
MeO
7
9
+
11
acid, base
work up,
82%
K2CO3
[TBAI]
DMSO, ∆
77%
18
22
Scheme 2. Preparation of 5 via Pd0-catalyzed mono-N-arylation of piperazine.
Pd°
N
NH
BnO
+
Conditions:
7
15
acid-base
work up
80%
1. Pd/C dry, cyclohexadiene, EtOH, reflux, 24h, 81%.
2. Pd/C wet, NH4HCO2, MeOH, EtOAc, reflux, 12h, 80%.
3. Pd/C wet, THF, H2, 80 PSI, 70 °C, 40h, 94%.
4. Pd/C wet, THF, H2, 50 PSI, 50 °C, 20h, 99%.
21
K2CO3, [TBAI]
6
88%
DMSO, ∆
Condition 4
7
1. Pd°
N
N
NO2
N
N
NH2
BnO
BnO
15
2. K2CO3, [TBAI]
23
24
DMSO, ∆
6
Conditions 1-3
78%
5
Scheme 3. A short, convenient synthesis of antifungal intermediate 5.
verted to 5 via one of the several conditions depicted in
the scheme. Further work towards streamlining the
preparation of 5 showed that the isolation of 21 was
unnecessary towards the preparation of 23 as 21, pre-
pared in situ, readily reacted with 6 to form 23.
Improved yield for one-pot preparation reflects the loss
of 21 during the acid base workup. This two-pot syn-
thesis (15 to 23 to 5) via the transfer hydrogenation
(eliminating pressure reactor) is the most large scale
amenable preparation of 5 reported to date. Note that
the aniline 24 could be isolated, if needed, by using the
milder hydrogenation conditions. Compound 24, upon
further hydrogenation as shown in conditions 1–3 could
be converted to 5.
Initial work towards the preparation of antifungal 1
(Scheme 4) led to the following findings. The prepara-
tion of the arylhalide 27 from sulfonate 25 (available to
us6) progressed well. Next, the Pd0-catalyzed reaction
between arylhalide 27 and 7 under above described
conditions resulted in only the mono-N-aryl piperazine
28 albeit in small amount (ꢀ5% yield with 0.5 mol%
catalyst). On the other the reaction of triazolone 2915
with 7 under previously described reaction conditions1
led to degradation of 29 instead of the formation of 30.
O
F
N
N
N
O
N
R
N
Next we investigated the application of the above meth-
ods to the preparation of the azole antifungal
posaconazole 1. The known3–6 disconnections for the
synthesis of the above azole antifungals are shown in
Fig. 1. In the case of Sch 51048 and Sch 56592 the
chiral tetrahydrofuran moiety was appended to the
di-arylated piperazine containing the N,N%-substituted
urea entity. This was then cyclized to the hydroxy
protected triazolone, i.e. O-benzyl 1.4,6 For itracona-
zole, the original synthesis consisted of reacting the
preformed triazolone containing piperazine to the
dichlorophenyl dioxolane azole. This was then sub-
jected to triazolone-N-alkylation to the desired
product.3 Sepracor has recently used Pd0 catalysis to
append the preformed haloaryl triazolone to the mono-
N-substituted piperazine as shown in Fig. 1.5 We recog-
nized that the piperazine mono-N-arylation findings
offered an alternate, novel convergent approach (Fig. 2)
for the preparation of 1 similar to the one used above
for the preparation of differentially-N,N%-diarylated
piperazine 5. Furthermore, this chemistry may be appli-
cable to the preparation of a variety of bioactive classes
of N,N%-substituted piperazines and to important piper-
azines for combinatorial library work.1–7
O
F
1: Sch 56592, R = OH
2: Sch 51048, R = H
N N
N
Ref. 5
O
Cl
N
N
N
O
N
O
R'
N
O
Cl
3: Itraconazole, R' = H
4: Hydroxyitraconazole, R' = OH
N N
N
Figure 1. The antifungals and their synthetic disconnections.
O
F
N
N
N
N
O
N
N
OH
1
5
O
F
N N
HO
N
NH2
N
Figure 2. New disconnection for the synthesis of 1, and 5.

3362
M. Hepperle et al. / Tetrahedron Letters 43 (2002) 3359–3363
Scheme 4. Synthesis of antifungal 1 via piperazine mono-N-arylations methods.
Although additional studies are needed to fully under-
Thus, the outcomes of the initial efforts for both the
Pd0-catalyzed as the non-catalyzed piperazine mono
stand this mechanism, the above study generated suffi-
cient information for us to complete the novel synthesis
of 1.
N-arylation towards the preparation of
1 were
discouraging.
In view of the above mono-N-arylation of piperazine
with non-electron deficient aromatic halide, and in view
of Pd0-catalyzed amination of arylhalides in general,12
the low yield for the reaction between 7 and 27 was
intriguing. Good solubility of 7 and 27 in toluene, the
use of several lots of reagents, ligands, as well as the
catalyst eliminated the possibility that the low yield was
imparted by specific set of reagents. Increasing the
catalyst amount to 2 mol% led to ꢀ20% solution yield
of 9 suggesting a catalyst poisoning in this reaction.
Based on the above observations, the strategy for the
synthesis of 1 was revised (Scheme 6). First triazolone
29 was reacted with 7 under Pd0-catalyzed conditions to
produce compound 30 in 88% isolated yield.17 This may
be due to the possibility that product 30 of this reaction
forms faster than the catalyst deactivation. Coupling of
30, with 27 to form 31 was then accomplished also via
Pd0 catalysis. This reaction was slower compared to the
preparation of 30.18 Alternately, a one-pot coupling of
piperazine with 29 followed by the reaction of 30 with
27 can be achieved in a reasonable time (1 day) with a
fresh does of catalyst added to the reaction mixture
after the formation of 30. The product profile was less
clean in the latter case. Compound 31 was debenzylated
via the known transfer hydrogenation procedure6 to
obtain posaconazole 1, identical to the one obtained via
the previous route.6
This prompted us to further investigate Pd0-catalyzed
arylation of piperazine. Due to simplicity of the sub-
strates the reaction between 9 and 7 to form 18 (Scheme
5) was chosen as a model reaction to study the above
mechanism with NMR. In one set of experiments the
mixture of 18 and the catalyst (prepared as described
previously)9–13 was first subjected to the reaction condi-
tions (80°C, toluene). The use of this pretreated catalyst
for reaction between fresh 7 and 9 led to no additional
product 18. This suggested a catalyst deactivation dur-
ing pretreatment with 18. This deactivation can be
either due to chelation of the catalyst with 18 or due to
In summary, a Pd0-catalyzed method for the mono
N-arylation of piperazine from commercially available
inexpensive raw materials without the use of protecting
groups, high pressure/temperature equipment, or chro-
matographic purification has been discovered. This
method has been applied to a short, convenient prepa-
ration of the key differentially N,N%-diaryl substituted
piperazine antifungal intermediate. After a limited
mechanistic investigation, this mono-N-arylation
finding has been used to develop a novel synthesis of
the antifungal posaconazole. The Pd0 catalysis method
complements our recently published direct, mono N-
arylation of piperazine with arylhalides containing elec-
tron withdrawing groups.
1
catalyst poisoning by non-specific binding. H and 31P
NMR probing of this reaction showed that the forma-
tion of the catalyst progressed as reported previously.16
When this catalyst was treated with 18 and warmed, 31
P
NMR signals for the catalytic species significantly
decreased and 31P NMR signals for the BINAP
increased correspondingly. The catalyst by itself did not
generate BIINAP 31P NMR signals upon heating. Thus,
under the reaction conditions, mono N-arylpiperazine
18 appears to compete with BINAP for Pd2(dba)3 or
displace BINAP from the catalytic spices interfering
with the catalytic cycle.9,16 This NMR study suggests
that mono N-aryl piperazines displace BINAP from the
catalytic species there by preventing its diarylation.
Finally comparative rate/extent of catalyst deactivation
by product compared to the rate of piperazine N-aryla-
tion govern the extent of mono N-arylation reaction.
Pdo
+ HN
NH
N
NH
MeO
Br
MeO
9
7
18
Scheme 5. Model reaction for mechanism study via NMR.

M. Hepperle et al. / Tetrahedron Letters 43 (2002) 3359–3363
3363
Scheme 6. Novel synthesis of 1 via Pd0-catalyzed piperazine mono-N-arylation.
Acknowledgements
halides, please see: (a) Wolfe, J. P.; Wagaw, S.; Marcoux,
J.-F.; Buchwald, S. L. Acc. Chem. Res. 1998, 31, 805; (b)
Hartwig, J. F. Synlett 1997, 329; (c) Frost, C. G.; Men-
donc¸a, P. J. Chem. Soc., Perkin Trans. 1 1998, 2615.
13. Typical experimental procedure (unless noted otherwise):
To a solution of piperazine (5.5 mmol) in toluene was
added Pd2(dba)3 (0.025 mmol) and (+), (−), or ( )-
BINAP (0.075 mmol). Arylhalide (5 mmol) was added
dropwise, followed by NaOtBu (7.5 mmol) as a suspen-
sion in THF. The resulting mixture was heated to 80–
90°C for prescribed amount of time. The reaction was
cooled to ambient temperature and diluted with EtOAc
and water. After separation of the layers, the organic
phase was washed once with water. The combined
aqueous layers were back extracted with EtOAc. The
resulting combined organic layers were extracted with 1N
HCl. The combined acidic layers were treated with
NaOH to pH 12. The mixture was then extracted with
CH2Cl2, dried over Na2SO4, filtered, and concentrated
under reduced pressure to afford the product. CsCO3,
NaHMDS, or KHMDS as bases offered no advantage
over NaOtBu.
We thank Dr. George Wu for many discussions on the
mechanism aspect of this project.
References
1. Hepperle, M.; Eckert, J.; Gala, D. Tetrahedron Lett.
1999, 40, 5655.
2. Eckert, J.; Chan, T.-M.; Osterman, R. M.; Lambert, J.
B.; Gala, D. Tetrahedron Lett. 1999, 40, 5661.
3. Heeres, J.; Backx, L. J. J.; VanCutsem, J. J. Med. Chem.
1984, 27, 894.
4. Saksena, A. K.; Girijvallabhan, V. M.; Wang, H.; Liu, Y.
T.; Pike, R. E.; Ganguly, A. K. Tetrahedron Lett. 1996,
37, 5657.
5. Tanoury, G. J.; Senanayake, C. H.; Hett, R.; Kuhn, A.
M.; Kessler, D. W.; Wald, S. A. Tetrahedron Lett. 1998,
39, 6845.
6. Andrews, D. R.; Gala, D.; Gosteli, J.; Gunter, F.;
Mergelsberg, I.; Sudhakar, A. US Patent 5,625,064: April
29, 1997.
7. For recent literature on N-phenylpiperazine useful for (a)
Neuropeptide Y antagonist: Wood, P. L. Drug Discov.
Develop. 1998, 1, 34; (b) G-Protein-coupled receptor lig-
ands: Kubinyi, H. Drug Discov. Develop. 1998, 1, 4.
8. Coop, A.; Rice, K. C. Tetrahedron Lett. 1998, 39, 8925.
9. Wolfe, J. P.; Wagaw, S.; Buchwald, S. L. J. Am. Chem.
Soc. 1996, 118, 7215.
14. Twice the amount of the catalyst/ligands added to
shorten the reaction time.
15. A gift of this compound from Dr. Anantha Sudhakar is
appreciated.
16. Amatore, C.; Broeker, G.; Jutand, A.; Khalil, F. J. Am.
Chem. Soc. 1997, 119, 5176.
17. Laboratory work had indicated that under the Pd0 catal-
ysis condition, electron withdrawing group containing 29
first monoarylates piperazine to form 30. When 30 was
allowed to react with fresh catalyst and more 29, the bis
arylation set in. The bis arylation was slow (required ꢀ3
days for ꢀ80% yield of the symmetrically N,N%-bis-sub-
stituted piperazine).
10. Louie, J.; Driver, M. S.; Hamann, B. C.; Hartwig, J. F. J.
Org. Chem. 1997, 62, 1268.
11. Typically, the aromatic halides (Ar-X) are converted to
Grignard reagents (Ar-Mg-X) which up on reaction with
trimethylborate (B(OMe)3) followed by hydrolysis lead to
arylborates (Ar-B(OH)2).
18. Presumably due to the steric bulk associated with the
reacting species.
12. For reviews on Pd0-catalyzed amination of aromatic