Practical routes toward the synthesis of 2-halo- and 2-alkylamino-4-pyridinecarboxaldehydes

Article abstract of DOI:10.1016/S0040-4039(01)01428-9

We recently required an efficient synthesis of 2-halo- and 2-alkylamino-4-pyridinecarboxaldehydes. Several routes to these compounds were investigated resulting in efficient and practical procedures from readily available and inexpensive starting materials.

Full text of DOI:10.1016/S0040-4039(01)01428-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6815–6818
Practical routes toward the synthesis of 2-halo- and
2-alkylamino-4-pyridinecarboxaldehydes
Lisa F. Frey,* Karen Marcantonio, Doug E. Frantz, Jerry A. Murry, Richard D. Tillyer,
Edward J. J. Grabowski and Paul J. Reider
Department of Process Research, Merck Research Laboratories, Merck & Co. Inc., PO Box 2000, Rahway, NJ 07065, USA
Received 25 July 2001; accepted 3 August 2001
Abstract—We recently required an efficient synthesis of 2-halo- and 2-alkylamino-4-pyridinecarboxaldehydes. Several routes to
these compounds were investigated resulting in efficient and practical procedures from readily available and inexpensive starting
materials. © 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
2. 2-Halopicolines as starting materials
We recently required an efficient synthesis of 2-halo-
and 2-alkylamino-4-pyridinecarboxaldehydes to sup-
port an ongoing drug development program. 2,4-Disub-
stituted pyridines are widely used substrates that appear
in compounds directed towards a range of therapeutic
areas.1 Despite this fact, general reviews that focus on
the synthesis of substituted pyridine rings do not cover
many examples of 2,4-disubstitution.2 An initial survey
of the literature indicated that 2-halo-4-pyridinecarbox-
aldehydes were typically prepared by a reduction–oxi-
dation sequence from 2-haloisonicotinic acid1a–c,1i or via
a POCl3 reaction with a 4-substituted pyridine N-oxide.
While these routes provide the desired materials in
reasonable yields, we were challenged with discovering
a more efficient procedure. In this communication, we
disclose efficient syntheses of 2-halo-4-pyridinecarbox-
aldehydes and three separate routes towards a 2-
alkylamino-4-pyridinecarboxaldeyde.
We initially investigated the conversion of the 2-halopi-
colines to the desired alkyl amino aldehyde 1. Amina-
tion of 2-bromopicoline 2a utilizing the Hartwig/
Buchwald protocol3 or 2-fluoropicoline 2b by simply
heating with (S)-a-methylbenzylamine provided the
desired 2-alkylaminopicoline 3 (Scheme 1). Unfortu-
nately, we were unable to functionalize the picolyl
methyl group while the amine side chain was in place.
As a result, we began looking at methods for first
oxidizing the 4-methyl group to the aldehyde oxidation
state, followed by amination at the 2-position. There
are numerous reports for oxidation of heteroaromatic
methyl groups to the aldehyde,4 acid5 and oxime.6 We
CN
CH3
N
O
CHO
N
N
X
X = Br, Cl, F
A search of commercially available 2,4-disubstituted
pyridines revealed that in addition to 2-chloro-isonico-
tinic acid, 2-halopicolines and 4-cyanopyridine N-oxide
were also commercially available. We set out to explore
the transformation of each of these materials into the
desired 2-substituted-4-pyridinecarboxaldehydes (Fig.
1).
X
X = Br, Cl,
H2N
Ph
COOH
Keywords: pyridine; a-methylbenzyl amine; amination.
* Corresponding author. Tel.: 732-594-3307; fax: 732-594-5170;
N
Cl
Figure 1.
e-mail: lisa frey@merck.com
–
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01428-9

6816
L. F. Frey et al. / Tetrahedron Letters 42 (2001) 6815–6818
In order to obtain the 2-alkylamino analogue, 5 was
CH3
N
CH3
protected as the dimethyl acetal and successfully ami-
nated with palladium catalysis. Acidic hydrolysis of the
dimethyl acetal afforded the title compound 1 (45%
yield from picoline 2a). Direct amination of the alde-
hyde 5, after in situ formation of the imine from
(S)-a-methylbenzylamine, using palladium catalysis
provided 1 in only 26% yield from 5 after acidic hydroly-
sis. As a result, we began investigating alternative ami-
nation protocols. Ullmann-type couplings have been
used for reactions of amines with aryl halides utilizing a
variety of copper sources.9 After exploring a wide range
of reaction conditions, we found that the imine from 5
and (S)-a-methylbenzylamine could be heated with sto-
ichiometric copper(I) acetate in the presence of cesium
trifluoroacetate and excess (S)-a-methylbenzylamine to
provide the desired product 1 after acidic hydrolysis
(Scheme 1). This three-step protocol provided a 48%
yield of the desired compound 1 from 2-bromopicoline
2a. While this method has some appealing qualities, it
requires the use of stoichiometric copper to effect the
amination. We investigated the 2-fluoro analogue in an
effort to overcome this limitation.
a
X
N
N
H
Ph
2a, X=Br
2b, X=F
3
b
NOH
CHO
CHO
X=Br
d or e
c
N
X
N
Br
N
N
H
Ph
4a, X=Br
4b, X=F
5
1
Scheme 1. Reagents and conditions: (a) (X=Br) Pd(OAc)2,
BINAP, (S)-a-methylbenzylamine, NaOtBu, toluene, 50°C or
(X=F) (S)-a-methylbenzylamine, 140°C; (b) tBuONO,
KOtBu, THF; (c) aq. HCl, CH2O; (d) (i) MeOH, H2SO4, (ii)
Pd(OAc)2, BINAP, (S)-a-methylbenzylamine, NaOtBu, tolu-
ene, 50°C, (iii) aq. HCl; (e) (i) (S)-a-methylbenzylamine,
toluene, (ii) CuOAc, CsOCOCF3, (S)-a-methylbenzylamine,
toluene, 110°C, (iii) aq. HCl.
Chloro and bromo pyridines require metal catalysis to
undergo substitution, while the corresponding fluorides
often undergo displacement under thermal conditions.
Amination of 2-fluoropicoline was successful in neat
(S)-a-methylbenzylamine at 140°C. However, amina-
tion of the 2-fluoro-4-oximino compound 4b was
accompanied by some dehydration of the oxime to the
nitrile. Heating 4b in the presence of (S)-a-methylben-
zylamine led to complex mixtures of 2-amino-4-oxime
6, 2-amino-4-nitrile 7, 2-fluoro-4-imine 8 and other
products (Scheme 2). Interestingly, reaction of the
oxime 4b and (S)-a-methylbenzylamine at 110°C with
Ti(OiPr)4 provided the dehydrated 2-amino-4-cyano
pyridine 7 in 66% yield. Reaction of this intermediate
with DIBAL provided the desired aldehyde 1 in an
overall yield of 41% from the 2-fluoropicoline 2b. Alter-
natively, oxime 4b can be dehydrated with thionyl
chloride, aminated by heating with (S)-a-methylbenzyl-
amine in DMAC, and the nitrile reduced using DIBAL
to give the desired aldehyde 1. This is a four-step
synthesis of 1 from 2-fluoro-picoline 2b with an overall
yield of 52%.
were pleased to find that subjecting 2-bromo or 2-fluoro
picoline to t-butyl nitrite in the presence of potassium
t-butoxide provided the desired oximes 4a and 4b in
high yield (Scheme 1).
Attempts to aminate the 2-bromo compound 4a under
standard palladium amination conditions3 resulted in
poor yields of the desired product. Presumably under
strongly basic conditions the oxime undergoes deproto-
nation rendering the aromatic ring less reactive.7
Instead, we chose to evaluate the amination on an
aldehyde substrate. Reaction of the 2-bromo oxime 4a
with formaldehyde under acidic conditions provided the
desired aldehyde 5 in 64% yield from the picoline 2a.8
This product is the first in the series of 2-halo-4-
pyridinecarboxaldehydes required by our laboratory.
NOH
CN
b or c
N
F
N
N
H
Ph
4b
7
3. Pyridine N-oxides as starting materials
a
Our success reducing the nitrile to the desired aldehyde
led us to investigate the 2-substituted-4-cyano pyridines
as potential intermediates in the synthesis of our desired
target. A search of the pertinent literature revealed two
observations: 4-pyridine carboxamide N-oxide reacts
with a mixture of POCl3 and PCl5 to give a 50% yield
of 2-chloro-4-cyano pyridine,10a while 4-cyanopyridine
N-oxide provides the 3-chloro-4-cyano pyridine under
similar conditions.10b The production of the 3-chloro
isomer was a curious result and we decided to reinvesti-
gate these reactions to determine whether we could
utilize them to our advantage. Interestingly, we found
N
NOH
CN
Ph
+
+
N
N
H
Ph
N
F
N
N
H
Ph
6
7
8
Scheme 2. Reagents and conditions: (a) (S)-a-methylbenzyl-
amine, 140°C; (b) Ti(OiPr)4, (S)-a-methylbenzylamine,
DMSO, 110°C; (c) (i) SOCl2, THF, (ii) (S)-a-methylbenzyl-
amine, 110°C, DMAC.

L. F. Frey et al. / Tetrahedron Letters 42 (2001) 6815–6818
6817
that heating 4-cyanopyridine N-oxide with POCl3,1i
either neat or in ACN, gave the 2-chloro-4-cyanopy-
ridine 9 as the predominant product (10:1 mixture of
isomers)11 (Scheme 3). The reaction required 24 h at
100°C to effect complete consumption of starting mate-
rial, and a pH and temperature controlled work-up to
safely quench the POCl3, while preventing degradation
of the product.12 In this way, 2-chloro-4-cyano pyridine
9 was obtained in 69% yield.
Isolation of the 2-alkylamino compound 1 was more
problematic, and as a result, the yields for this reaction
varied from 60 to 90%. At the completion of the
reaction it was necessary to decompose both aluminum
complexes, as well as oligomers that arose from the
imine intermediate. Workers at Lilly had demonstrated
that a similar naphthyridine substrate could be isolated
by quenching the reaction mixture into acetic acid and
then extracting the product into the organic layer.14 We
found this protocol to be extremely exothermic and
very difficult to control. In addition, we found that the
intermediates were not completely broken down unless
the pH was <4,15 while at pH <4, all of the product was
in the aqueous stream. Increasing the pH of the
aqueous substrate after complete decomposition of the
reaction intermediates resulted in a gel of aluminum
salts.16 As a result, we turned our attention towards
isolating the aldehyde from an acidic aqueous solution.
Direct crystallization of the desired compound was not
successful, so we were gratified to find that adding a
20% bisulfite solution to the acidic aqueous solution of
aldehyde resulted in crystallization of the bisulfite
adduct 10 in 80% yield (Scheme 3). To employ the
aldehyde in the subsequent steps, it was necessary to
break down the bisulfite adduct at high pH. This was
carried out most effectively by dissolving the solid in
saturated aqueous KHCO3 and extracting the resulting
solution with ethyl acetate.17 In this way the pure
aldehyde 1 (obtained in 44% overall yield from 4-
cyanopyridine N-oxide) could be isolated or used as a
solution.
Amination of 2-chloro-4-cyanopyridine 9 (Scheme 3)
utilizing palladium catalysis3 was initially problematic.
We found that some decomposition of the nitrile
occurred during the reaction, and we also observed
varying amounts of a dimer (bis-arylated a-methylben-
zylamine) depending on reaction conditions. These
issues were overcome by slow addition of 9 to a suspen-
sion of the other reagents. At the completion of the
reaction, an acetic acid wash was used to remove excess
(S)-a-methylbenzylamine and then the product was
extracted into aqueous HCl to allow for further purifi-
cation. 2-Alkylamino-4-cyanopyridine 7 was crystal-
lized from toluene–heptane in 82% yield from 9.13
The DIBAL reduction of both the 2-chloro and 2-alkyl-
amino-4-cyanopyridines proceeded very well using stan-
dard conditions (toluene, −10°C). Reduction of the
2-chloro-compound 9 gave 2-chloro-4-pyridine carbox-
aldehyde 11 in 80% yield. This represents an alternative
synthesis of 2-chloro-4-pyridinecarboxaldehyde 11 that
proceeds in 55% overall yield from 4-cyanopyridine
N-oxide.
4. Conclusion
CN
CN
CN
a
In summary, we have identified several routes to 2-sub-
stituted-4-pyridine carboxaldehydes from commercially
available starting materials. 2-Chloro-4-pyridine-car-
boxaldehyde 11 was prepared from 4-cyanopyridine
N-oxide in 55% yield, and 2-bromo-4-pyridinecarbox-
b
N
O
N
Cl
N
N
H
Ph
7
9
e
c
aldehyde
5
was prepared from 2-bromo-4-
methylpyridine in 64% yield. Both of these routes offer
attractive alternatives to the previously reported prepa-
rations of 2-halo-4-pyridinecarboxaldehydes. In addi-
tion, we identified three simple routes towards the
2-alkylamino-4-pyridinecarboxaldehyde 1 from 2-bro-
mopicoline (48% overall yield), 2-fluoropicoline18 (41%)
and 4-cyanopyridine N-oxide (44%). When the final
step is the Dibal reduction of the nitrile, isolation of 1
as the bisulfite adduct allows for easy purification and
improved long term stability and handling.
CHO
CHO
N
Cl
N
N
H
Ph
11
1
d
-
HO
SO3 Na+
Supporting information available. Includes detailed
descriptions of experimental details and new compound
characterization data.
N
H
N
H
Cl-
Ph
+
References
10
Scheme 3. Reagents and conditions: (a) POCl3, 100°C; (b)
Pd(OAc)2, BINAP, (S)-a-methylbenzylamine, NaOtBu, tolu-
ene, 50°C; (c) (i) Dibal, toluene, −15°C, (ii) aq. HCl; (d) aq.
NaHSO3; (e) (i) Dibal, toluene, −15°C, (ii) aq. HCl.
1. (a) Nobuhiko, H.; Noriaki Ishikawa-so, K.; Hajime
Green-copo Itoya, M.; Akihiko, H.; Yasuo, S.;
Yoshikazu, I.; Tesuaki, Y.; Akihiro S.; Masashi, N.
European Patent 0,282,077, 1988; (b) Adams, J. L.; Gal-

6818
L. F. Frey et al. / Tetrahedron Letters 42 (2001) 6815–6818
lagher, T. F.; Sisko, J.; Peng, Z. Q.; Osifo, I. K.; Boehm,
Chem. Soc. 1998, 120, 12459; (c) Kiyomori, A.; Marcoux,
J.-F.; Buchwalks, S. L. Tetrahedron Lett. 1999, 40, 2657;
(d) Paine, A. J. J. Am. Chem. Soc. 1987, 109, 1496.
10. (a) Libermann, D.; Rist, N.; Grumbach, F.; Cals, S.;
Moyeux, M.; Rouaix, A. Bull. Soc. Chim. Fr. 1958, 694;
(b) Rokash, J.; Girard, Y. J. Heterocycl. Chem. 1978, 15,
683.
11. In addition, reaction of 4-cyanopyridine N-oxide with
a-methylbenzyl amine in the presence of methylbenzyl
isocyanate did not yield the desired 2-amino-4-
cyanopyridine.
12. When 9 is aged in solutions of ACN–aq buffer, at pH 1
and at pH 11–12, the corresponding amide is slowly
formed. A similar pH controlled quench was reported:
Singh, B.; Lesher, G. Y.; Pennock, P. O. J. Heterocycl.
Chem. 1990, 27, 1841.
13. The ee of the a-methylbenzylamine was retained in the
product.
14. Anderson, B. A.; Becke, L. M.; Booher, R. N.; Flaugh,
M. E.; Harn, N. K.; Kress, T. J.; Varie, D. L.; Wepsiec,
J. P. J. Org. Chem. 1997, 62, 8634.
15. The use of acetic acid and tartrate salt appeared to
effectively break down the aluminum salts but not some
oligomeric species, which presumably arise from trimer-
ization of the imine intermediate. NMR studies of the
reaction mixtures after these types of quenches showed
complicated signals consistent with these types of inter-
mediates.
J. C. US Patent 5,869,660, 1999; (c) Rossi, T.; Andreotti,
D.; Tedesco, G.; Tarsi, L.; Ratti, E.; Feriani, A.; Pizzi,
D.; Gaviraghi, G.; Biondi, S.; Finizia, G. WO Patent
98/21210, 1998; (d) Huth, A.; Seidelmann, D.; Thierauch,
K.-H.; Bold, G.; Manley, P.; Furet, P.; Wood, J.; Mes-
tan, J.; Bruggen, J.; Ferrari, S.; Kruger, M.; Ottow, E.;
Menrad, A.; Schirner, M. WO Patent 00/27819, 2000; (e)
Klimesova, V.; Svoboda, M.; Waisser, K.; Pour, M.;
Kaustova, J. Il Farmaco 1999, 54, 666; (f) Lipinski, C. A.;
LaMattina, J. L.; Oates, P. J. J. Med. Chem. 1986, 29,
2154; (g) Pendalwar, S. L.; Chaudhari, D. T.; Patel, M.
R. Bull. Haffkine Inst. 1980, 8, 33; (h) Liverton, N. J.; et
al. J. Med. Chem. 1999, 42, 2180; (i) Watson, S. E.;
Markovich, A. Heterocycles 1998, 48, 2149.
2. (a) Mongin, F.; Queguiner, G. Tetrahedron 2001, 57,
4059; (b) Rewcastle, G. W.; Katritzky, A. R. Advances in
Heterocyclic Chemistry; Academic Press: New York,
1993; Vol. 56, p. 155.
3. (a) Wagaw, S.; Buchwald, S. L. J. Org. Chem. 1996, 61,
7240; (b) Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc.
1996, 118, 7217.
4. Vismara, E.; Fontana, F.; Minisci, F. Gazz. Chim. Ital.
1987, 117, 135.
5. (a) Reyes-Rivera, H. M.; Hutchins, R. O.; Dalton, D. R.
J. Het. Chem. 1995, 32, 665; (b) Telser, J.; Cruickshank,
K. A.; Schanze, K. S.; Netzel, T. L. J. Am. Chem. Soc.
1989, 111, 7221; (c) Garelli, N.; Vierling, P. J. Org. Chem.
1992, 57, 3046.
16. Quenching the DIBAL reaction into HCl–citric acid (to
break down aluminum complexes), followed by addition
of NaOAc (to allow for extraction of the product out of
the aqueous layer without gelling of aluminum salts) does
allow for isolation of the pyridine aldehyde from toluene–
heptane, but the work-up is tedious and involves several
steps.
17. The higher concentration of saturated aqueous KHCO3
over NaHCO3 allowed for lower losses to the aqueous
layer and a more efficient process.
6. (a) Tagawa, Y.; Hama, K.; Goto, Y. Heterocycles 1992,
34, 1605; (b) Tagawa, Y.; Arakawa, H.; Goto, Y. Hetero-
cycles 1989, 29, 1741; (c) Jaemoon Lee, personal
communication.
7. Alkylation of the oxime 4a with methyl iodide provided
the methyl oxime which, under palladium-catalyzed ami-
nation conditions, gave a mixture of the desired product
and nitrile 7.
8. (a) Cava, M. P.; Litle, R. L.; Napier, D. R. J. Am. Chem.
Soc. 1958, 80, 2257; (b) Barltrop, J. A.; Johnson, A. J.;
Meakins, G. D. J. Chem. Soc. 1951, 181.
9. (a) Loughhead, D. G. J. Org. Chem. 1990, 55, 2245; (b)
Ma, D.; Zhang, Y.; Yao, J.; Wu, S.; Tao, F. J. Am.
18. Alternatively, separation of the dehydration and amina-
tion steps provides 1 in four steps and 52% overall yield
from 2-fluoropicoline.