Practical amination of nitropyridones by silylation

Article abstract of DOI:10.1021/op800201u

A practical method for coupling nitropyridones (1) with primary amines by treatment with hexamethyldisilazane has been developed, avoiding the use of hazardous reagents such as POCl₃. The activation of the pyridone by a nitro group is necessary for efficient coupling, leading to aminonitropyridines (3) in good yields. Regioisomers other than 3-nitro-4-pyridone (1) were found to be substantially less reactive but would undergo coupling with primary amines.

Full text of DOI:10.1021/op800201u

Organic Process Research & Development 2008, 12, 1261–1264
Technical Notes
Practical Amination of Nitropyridones by Silylation
Robert A. Singer* and Michae¨l Dore´
Chemical Research and DeVelopment, Pfizer Global Research and DeVelopment, Groton Laboratories, Eastern Point Road,
Groton, Connecticut 06340, U.S.A.
Scheme 1. General route to azabenzimidazoles (5) via
4-amino-3-nitropyridine (3)
Abstract:
A practical method for coupling nitropyridones (1) with primary
amines by treatment with hexamethyldisilazane has been devel-
oped, avoiding the use of hazardous reagents such as POCl3. The
activation of the pyridone by a nitro group is necessary for efficient
coupling, leading to aminonitropyridines (3) in good yields.
Regioisomers other than 3-nitro-4-pyridone (1) were found to be
substantially less reactive but would undergo coupling with
primary amines.
and found that primary amines coupled readily with 1b in
isopropanol under ambient conditions to furnish 3a-d in good
isolated yields (Table 1). Upon reaction completion, the products
crystallized following addition of water to the reaction mixture.
While this was a viable process alternative to using the
corresponding chloronitropyridine, the desired methoxynitro-
pyridines were not as readily available nor as inexpensive as
the corresponding hydroxy-substituted derivatives.7
Introduction
During our process research and development work on
potential drug candidates in early development, we encountered
the need to prepare various aminonitropyridines (3) as precursors
for azabenzimidazoles (5). Originally we used a conventional
approach to prepare this family of compounds by displacement
of a chloronitropyridine (2) with the appropriate primary
alkylamine (Scheme 1).1 The chloronitropyridine was prepared
from the corresponding nitropyridone (1) with POCl3.1,2 Since
2 and some chloronitropyridine derivatives are irritants and
potential sensitizers,3 we elected to utilize more benign starting
materials that would be sufficiently activated to undergo
substitution. Herein we describe the process research and
development of a safer activation of nitropyridones to enable
coupling with primary amines.
8
In order to avoid the use of POCl3 and a separate
activation step of the nitropyridones, we explored the use
of hexamethyldisilazane (HMDS). We speculated that
HMDS could be sufficient to carry out in situ activation
(4) (a) Mugnaini, C.; Petricci, E.; Botta, M.; Corelli, F.; Mastromarino,
P.; Giorgi, G. Eur. J. Med. Chem. 2007, 42, 256–262. (b) Cain, G. A.;
Beck, J. P. Heterocycles 2001, 55, 439–446. (c) Zhao, H.; Serby,
M. D.; Xin, Z.; Szczepankiewicz, B. G.; Liu, M.; Kosogof, C.; Liu,
B.; Nelson, L. T. J.; Johnson, E. F.; Wang, S.; Pederson, T.; Gum,
R. J.; Clampit, J. E.; Haasch, D. L.; Abad-Zapatero, C.; Fry, E. H.;
Rondinone, C.; Trevillyan, J. M.; Sham, H. L.; Liu, G. J. Med. Chem.
2006, 49, 4455–4458.
Results and Discussion
Aside from the conversion of 1 to a halide such as 2, other
conventional approaches to the activation of nitropyridone (1)
include the formation of the sulfonate4 or phosphonium salts.5
While these options were likely to be viable, they still utilized
hazardous or toxic reagents. As another option, we were
intrigued by reports in the literature2,6 that methoxynitropyridines
had demonstrated the ability to undergo substitution with
primary amines. We explored the viability of this substrate class
(5) Phosphonium salts have often been formed with reagents such as BOP,
see: (a) Kang, F.-A.; Kodah, J.; Guan, Q.; Li, X.; Murray, W. V. J.
Org. Chem. 2005, 70, 1957–1960. (b) Wan, Z.-K.; Binnum, E.; Wilson,
D. P.; Lee, J. Org. Lett. 2005, 7, 5877–5880. (c) Wan, Z.-K.;
Wacharasindhu, S.; Binnum, E.; Mansour, T. Org. Lett. 2006, 8, 2425–
2428.
(6) (a) Bamford, M. J.; Alberti, M. J.; Bailey, N.; Davies, S.; Dean, D. K.;
Gaiba, A.; Garland, S.; Harling, J. D.; Jung, D. K.; Panchal, T. A.;
Parr, C. A.; Steadman, J. G.; Takle, A. K.; Townsend, J. T.; Wilson,
D. M.; Witherington, J. Bioorg. Med. Chem. Lett. 2005, 15, 3402–
3406. (b) Lanza, T. J.; Durette, P. L.; Rollins, T.; Siciliano, S.;
Cianciarulo, D. N.; Kobayashi, S. V.; Caldwell, C. G.; Springer, M. S.;
Hagmann, W. K. J. Med. Chem. 1992, 35, 252–258.
* Author to whom correspondence may be sent. E-mail:
robert.a.singer@pfizer.com.
(1) Crey-Desiolles, C.; Kotera, M. Bioorg. Med. Chem. 2006, 14, 1935–
1941.
(7) The price of 4-methoxy-3-nitropyridine is about $5/g to $20/g in cost
(depending on the choice of supplier) while the corresponding
hydroxypyridine is about $1/g to $4/g. In addition we had issues with
the quality of 4-methoxy-3-nitropyridine; one supplier provided
samples contaminated with up to 50% of N-methyl-3-nitro-4-pyridone.
(8) POCl3 can be hazardous for both laboratory use and the environment,
see: (a) Kapias, T.; Griffiths, R. F. J. Hazard. Mater. 2001, 81, 223–
249.
(2) Provencal, D. P.; Gesenberg, K. D.; Wang, H.; Escobar, C.; Wong,
H.; Brown, M. A.; Staab, A. J.; Pendri, Y. R. Org. Process Res. DeV.
2004, 8, 903–908.
(3) Material Safety Data Sheets on 2-chloro-3-nitropyridine and 4-chloro-
3-nitropyridine indicate that these materials are severe irritants and
that the 2-chloro isomer is a skin sensitizer.
10.1021/op800201u CCC: $40.75
Published on Web 11/07/2008
2008 American Chemical Society
Vol. 12, No. 6, 2008 / Organic Process Research & Development
•
1261

Table 1. Addition of amines to 1ba
Table 2. Addition of amines to 3-nitro-4-pyridone (1)a
a Reaction conditions: Used 1 equiv of 1, 1.5 equiv of HMDS, 2.0 equiv of
amine in CH3CN (4 mL/g 1) at 60 °C. b Isolated yield. c Reaction carried out in
DMAC. d Reaction carried out at 75 °C.
a Reaction conditions: Used
1 equiv of 1b, 2.0 equiv of amine, ambient
conditions in isopropanol (1.5 M relative to 1b). b Isolated yield.
of the hydroxypyridine tautomer (1a) by silylation to
enable substitution by primary amines. A benefit of using
HMDS was that the silylating reagent would not react with
the amine to be coupled nor with the nitrogen of the
pyridine ring,9 and could also serve to remove water from
the system through formation of siloxane byproduct.
Others have reported the use of HMDS for activating
heterocycles, such as pyridines, toward amination reac-
tions, but high temperatures and Lewis acids were
typically required since the systems studied were not
activated with electron-withdrawing groups.10 We were
pleased to find that we could adapt this in situ activation
to a practical protocol that could be readily implemented
on a kilogram scale. It was unnecessary to execute the
process in a stepwise manner; rather, the amine (2.0 equiv)
and HMDS (1.5 equiv) were combined with 3-nitro-4-
pyridone (1) in acetonitrile and heated to 60 °C for 8-24
h. Upon reaction completion, the reaction mixture was
concentrated to remove solvent and siloxane byproduct.
The products were typically crystallized from the crude
reaction mixture. The polarity of the reaction solvent
seemed to play a critical role since no reaction was
observed in THF, dichloromethane, or acetone. While
DMF or DMAC could be used, the reactions were
slower,11 and dimethylamine adducts formed in DMF as
a result of decomposition to dimethylamine during heat-
ing.12 In addition, acetonitrile possessed the advantage of
enabling solvent removal by distillation to facilitate
workup and isolation. The order of addition of HMDS
and the amine had no impact on the rate of the reactions,
demonstrating that HMDS does not interact with the
amine. We also observed no difference in relative rates
whether using 1 equiv or 3 equiv of HMDS. This suggests
that the main role of HMDS is silylation of the hydroxyl
group and additional activation by silylation of the nitro
group is unlikely.13 For examining the reaction scope,
3-nitro-4-pyridone (1) was coupled with various primary
amines in a tandem process (Table 2).
The preparation of benzimidazole 5a was demonstrated
on a kilogram scale.14 We decreased the charge of
propylamine to 1.5 equiv and lowered the reaction
temperature due to the volatility of propylamine. We also
noticed with optimization of the reaction that fewer polar
side products formed with a lower loading of propylamine.
The addition was complete within 48 h by using 1.5 equiv
of HMDS in acetonitrile with 1.5 equiv of propylamine
at 45 °C. After crystallization from 10:1 water/acetonitrile,
the product was isolated in 82% yield. The product was
then reduced under hydrogenation conditions in THF (8
mL/g) and crystallized from toluene to afford diamine 4a
as white crystals in 78% yield. To complete the preparation
of benzimidazole 5a, diamine 4a was dissolved in iso-
propylacetate (20 mL/g of 4a) and treated with 4 equiv
of chloroacetic anhydride under ambient conditions (Scheme
2). The acylation proceeded rapidly, but the ring closure
required stirring for several hours. The excess anhydride
(9) Due to the high affinity of silicon to oxygen, silylation always occurs
on oxygen rather than nitrogen, see: (a) Vorbruggen, H.; Krolikiewicz,
K. Liebigs Ann. Chem. 1976, 745–761. (b) Vorbruggen, H.; Krolik-
iewicz, K.; Niedballa, U. Liebigs Ann. Chem. 1975, 988–1002. (c)
Cook, M. J.; Katritzky, A. R.; Linda, P. AdV. Heterocycl. Chem. 1974,
17, 255–356.
(10) HMDS has been used in the past to activate hydroxy pyridines toward
substitution by amines, see: (a) Vorbruggen, H.; Krolikiewicz, K.
Chem. Ber. 1984, 117, 1523–1541. (b) Vorbruggen, H. Angew. Chem.,
Int. Ed. Engl. 1972, 11, 305–306.
(11) Polar solvents generally are favored for aromatic substitution reactions
which most likely progress through a charged intermediate such as a
Meisenheimer salt, see: (a) Meisenheimer, J. Liebigs Ann. Chem. 1902,
323, 205–246. (b) Illuminati, G.; Stegel, F. AdV. Heterocycl. Chem.
1983, 34, 305–444. (c) Brewis, D. M.; Chapman, N. B.; Paine, J. S.;
Shorter, J.; Wright, D. J. J. Chem. Soc., Perkin Trans. 2 1974, 15,
1787–801.
(12) The decomposition of DMF to dimethylamine and CO is well-known,
see: (a) Neumeyer, J. L.; Cannon, J. G. J. Org. Chem. 1961, 26, 4681–
4682.
1262
•
Vol. 12, No. 6, 2008 / Organic Process Research & Development

Scheme 2. Processing route to benzimidazole 5a
Conclusion
In summary, we have developed a practical one-pot proce-
dure for silylating nitropyridones and coupling with primary
amines which was demonstrated on a kilogram scale. From our
survey in this area there are some clear limitations on the class
of substrates viable for utilizing this methodology; however,
the highly activated nitropyridones appear to readily undergo
amination. Regioisomers other than 3-nitro-4-pyridone were
found to be substantially less reactive. This technology is well
suited for the preparation of azabenzimidazoles which typically
utilize primary amine additions to activated nitropyridines.
Finally, this methodology also possesses the advantage of
avoiding hazardous reagents such as POCl3 which are often
used for the activation of heterocycles to undergo additions by
amines.
initially allowed formation of a diacylated intermediate
which then would undergo ring closure. The acylation and
ring closure tolerated heating up to 40 °C, but higher
temperatures (60-70 °C) resulted in increased levels of
side products generally arising from alkylation at the
R-chloride. Maintaining a pH in the range of 3-6 was
essential for efficient ring closure following the acylation.
For this reason the anhydride was optimal for targeting
the desired pH range; however, the corresponding acid
chloride, chloroacetyl chloride, was ineffective in the
cyclization unless the reaction was buffered with pyridine
or other mild bases. Once complete, the reaction was
worked up with base, and the product (5a) was crystallized
as the HCl salt from isopropanol in 67% yield.
Experimental Section
General Procedure for Amination of Nitropyridones. To
a suspension of the nitropyridone (5.00 g, 35.7 mmol) in
acetonitrile (20 mL), was added HMDS (11.3 mL, 53.5 mmol)
and the mixture was stirred for 15 min. The amine (71.4 mmol)
was added and the resulting mixture was heated at 60 °C for
16 h. Acetonitrile was removed under reduced pressure, and
isopropanol (10 mL) was added and then concentrated. The
crude material was purified by crystallization. Isopropanol (10
mL) was added, and the mixture was cooled at 0 °C and stirred
for 5 min. Water (75 mL) or isopropyl ether (75 mL) was added
dropwise, and the crystals formed were stirred at 0 °C for 3 h.
The crystals were filtered, washed with either water (15 mL)
or isopropyl ether (15 mL), and dried affording the pure
aminonitropyridine as crystals.
Other regioisomers of nitropyridone, 5-nitro-2-pyridone (6),
3-nitro-2-pyridone (7), and 3-hydroxy-2-nitropyridine (8), were
evaluated and were found to be much less reactive than 1 as
substrates (Table 3). These results illustrate the sensitivity of
the electrophile to electronic effects.
Preparation of 3e on Kilogram Scale. To a suspension of
the nitropyridone (5.5 kg, 39.3 mol) in acetonitrile (22 L), was
added HMDS (12.4 L, 59.0 mol), and the mixture was stirred
for 15 min. Then propylamine (3.50 kg, 59.2 mol) was added,
and the resulting mixture was heated at 45 °C for 48 h.
Isopropanol (10 L) was charged, and the reaction mixture was
concentrated in vacuo to remove excess HMDS. This was
solvent exchanged for acetonitrile, and the volume was adjusted
to ∼12 L. Water (75 L) was added to crystallize the product.
After 3 h of stirring, the crystals were filtered and washed with
water to remove residual acetonitrile. The product was dried at
45 °C to yield yellow-green crystals (5.85 kg, 82%). Mp )
Table 3. Addition of amines to 5-nitro-2-pyridone (6),
3-nitro-2-pyridone (7), or 3-hydroxy-2-nitropyridine (8)a
1
65-66 °C; H NMR (400 MHz, DMSO-d6) δ 8.97 (s, 1H),
8.36 (s, 1H), 8.20 (d, J ) 6.0 Hz, 1H), 6.95 (d, J ) 6.0 Hz,
1H), 3.32 (dt, J ) 13.5, 6.0 Hz, 2H), 1.52-1.61 (m, 2H), 0.88
(t, J ) 7.5 Hz, 3H); 13C NMR (100 MHz, DMSO-d6) δ 153.3,
149.0, 148.8, 129.8, 109.2, 44.2, 22.1, 11.7; Anal. Calcd for
C8H11N3O2: C, 53.03; H, 6.12; N, 23.19. Found: C, 53.10; H,
6.15; N, 23.22.
Preparation of Diamine 4a by Hydrogenation. To a
solution of 3e (5.4 kg, 29.8 mol, 1.0 equiv) in THF (43 L) was
added 10% palladium on carbon (180 g, 50% wet by weight).
The Hastelloy vessel was sealed, charged with nitrogen, and
pressurized to 50 psi of hydrogen. After 2 h of stirring, the
reaction was filtered through a pad of Celite (pad washed with
THF). The filtrate was concentrated in vacuo and displaced with
toluene. The product diamine was recrystallized by heating in
a Reaction conditions: Used 1.0 equiv of nitropyridone, 2.0 of equiv amine, 75
°C, CH3CN (4 mL/g nitropyridone). b Isolated yield. c Product not isolated.
(13) Silylation of nitro groups to facilitate nucleophilic aromatic substitution
has been observed, see: Makosza, M.; Surowiec, M. Tetrahedron 2003,
59, 6261–6266.
(14) We had tested compound 5a as a potential mutagen and found that it
was negative in the Sprial Ames test as well as in the in Vitro
Micronucleus test.
Vol. 12, No. 6, 2008 / Organic Process Research & Development
•
1263

toluene (24 L) to afford white crystals (needles). These were
conditions. The crystals were filtered, washed with iso-
filtered and dried under vacuum at 50 °C to obtain 4a (3.5 kg,
78%). H NMR (400 MHz, CDCl3) δ 7.94 (d, J ) 5.3 Hz,
propanol, and dried at 50 °C to produce white crystals of
5a (3.6 kg, 67%) as an HCl salt. H NMR (400 MHz,
1
1
1H), 7.85 (s, 1H), 6.45 (d, J ) 5.3 Hz, 1H), 4.22 (s, 1H), 3.11
(dd, J ) 6.0 Hz, 2H), 2.43 (s, 2H), 1.62-1.71 (m, 2H), 1.00 (t,
J ) 7.5 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 145.4, 144.1,
138.0, 128.6, 105.0, 45.0, 22.6, 11.8.
CD3OD) δ 9.34 (s, 1H), 8.61 (d, J ) 6.6 Hz, 1H), 8.32
(d, J ) 6.6 Hz, 1H), 5.13 (s, 2H), 4.51 (t, J ) 7.5 Hz,
2H), 2.00 (sxt, J ) 7.5 Hz, 2H), 1.05 (t, J ) 7.5 Hz, 3H);
13C NMR (100 MHz, CD3OD) δ 158.4, 145.9, 139.2,
135.1, 133.9, 109.7, 46.9, 35.3, 22.9, 10.2.
Preparation of Benzimidazole 5a. To a solution of
chloroacetic anhydride (15 kg, 4.0 equiv) in isopropyl
acetate (197 L) was added 4a (3.3 kg, 21.8 mol, 1.0 equiv)
under ambient conditions. After stirring for 18 h at 20-25
°C, the reaction mixture was washed once with a 5.0 M
aqueous solution of NaOH (50 L), then once with a 1.0
M aqueous solution of NaOH (33 L), and once with a
saturated aqueous solution of NaCl (33 L). The organic
layer was concentrated to about half of the batch volume.
In a separate vessel acetyl chloride (2.22 kg, 1.3 equiv)
was added to isopropanol (50 L) to make anhydrous HCl.
The solution of crude product in isopropyl acetate was
slowly added to the solution of anhydrous HCl in
isopropanol to precipitate the product. The product formed
a white precipitate and was stirred for 3 h under ambient
Acknowledgment
We thank Brian Li and Rich Buzon for the demonstration
of this methodology in the Pfizer Groton Kilo Laboratory.
Supporting Information Available
Experimental procedures, characterization data and 1H NMR
and 13C NMR spectra for products 3a-d, 6a, and 7a are
provided. This material is available free of charge via the
Internet at http://pubs.acs.org.
Received for review August 19, 2008.
OP800201U
1264
•
Vol. 12, No. 6, 2008 / Organic Process Research & Development