An efficient synthesis of 2-quinoxalinecarboxylic acid

Article abstract of DOI:10.1021/op049951d

Development of a cost efficient and scaleable process for 2-quinoxalinecarboxylic acid is described. The primarily goals of the development work were to improve the overall yield of the process, to minimize the use of environmentally unacceptable material

Full text of DOI:10.1021/op049951d

Organic Process Research & Development 2004, 8, 666−669
An Efficient Synthesis of 2-Quinoxalinecarboxylic Acid
Arthur E. Harms*
Science and Technology Department, Albany Molecular Research, Inc., 21 Corporate Circle,
Albany, New York 12203, U.S.A.
Abstract:
tions, the hydrazone/osazone7 of a sugar is utilized. For
Development of a cost efficient and scaleable process for
2-quinoxalinecarboxylic acid is described. The primarily goals
of the development work were to improve the overall yield of
the process, to minimize the use of environmentally unaccept-
able materials, and to obtain a material with a high level of
purity. A variety of approaches were examined, and the most
efficient method was a condensation of o-phenylenediamine with
a monosaccharide followed by a mild peroxide oxidation.
example, fructose was converted to the fructose osazone/
hydrazone and condensed with o-phenylenediamine to
provide a moderate yield of the quinoxaline skeleton. While
some of these methods worked rather well, they required
materials that would be difficult to handle on large scale,
namely the hydrazine starting materials. In addition to the
handling of the toxic starting materials, large quantities of
the hydrazone would be present in the waste stream.
Results
In the development of an efficient route to 2-quinoxa-
linecarboxylic acid, we examined a number of different
approaches. One of the first routes investigated the straight-
forward approach of alkylating o-phenylenediamine with
ethyl 2,3-dibromopropanoate in the presence of triethylamine
(TEA, eq 1). Unfortunately, the reaction resulted in an
unacceptable yield of material that required a difficult
purification.
Introduction
The 2-substituted quinoxaline skeleton is a relatively
unexploited heterocycle in drug discovery. Only a few
examples of its use are found in the literature.1-3 The
synthesis of 2-quinoxalinecarboxylic acid has been reported
many times in the chemical literature, as early as 1935.4 Other
synthetic methods for the preparation of quinoxaline-
carboxylic acid include the oxidation of a nitroalkyl4 or
alkenyl5 substituent, usually with potassium permanga-
nate- or silver(I)-mediated hydrolysis of a tribromomethyl
side chain.7 Synthesis utilizing biocatalytic oxidation of
2-methylquinoxaline has also been described.8 A latent form
of carboxylic functionality is the 2-furyl group, which can
easily and quantitatively be cleaved by potassium perman-
ganate to yield the corresponding quinoxalinecarboxylic
acid.4 Another approach, which has had the most success,
involves the reaction of o-phenylenediamine with monosac-
charides9a-d followed by the oxidation of the resulting
condensed product.10,11 However, the yield of the condensa-
tion step has been poor (∼15-30%). In some other publica-
Deprotonation of quinoxaline with a variety of bases
(lithium diisopropylamide (LDA), potassium and lithium bis-
(trimethylsilyl)amide (KHMDS and LHMDS, respectively))
with subsequent quench using carbon dioxide failed to
provide any desired quinoxalinecarboxylic acid (eq 2). In a
similar approach, 2-chloroquinoxaline was submitted to
conditions of lithium-halogen exchange (eq 3). Under these
conditions, the starting material was consumed with no
detection of the desired quinoxalinecarboxylic acid.
* To whom correspondence should be addressed. E-mail: arthur.harms@
albmolecular.com.
(1) Seitz, L. E.; Suling, W. J.; Reynolds, R. C. J. Med. Chem. 2002, 45, 5604.
(2) Komatsu, M.; Sato, Hideaki, T.; Shinichi, M.; Masahiro, Magata, K.;
Yoshida, H.; Ueyama, A.; Nishi, T. Preparation of Quinoxaline-2-
Carboxamides as Antidiabetics. PCT Appl. WO 9509159 A1, April 6, 1995.
(3) Toshima, K.; Ohta, K.; Kano, T.; Nakamura, T.; Nakata, M.; Matsumura,
S. J. Chem. Soc., Chem. Commun. 1994, 19, 2295.
(4) Maurer, K.; Schiedt, B.; Schroeter, H. Chem. Ber. 1935, 68, 1716.
(5) Fanta P. E.; Stein, R. A.; Rickett, R. M. W. J. Am. Chem. Soc. 1958, 80,
4577.
(6) Keller-Scheirlien, W.; Prelog, V. HelV. Chim. Acta. 1957, 24, 205.
(7) Brown, B. R. J. Chem. Soc. 1949, 2577.
(8) Wong, J. N.; Watson, H., Jr.; Bouressa, J. F.; Burns, M. P.; Crawley, J. J.;
Doro, A. E.; Guzek, D. B.; Hintz, M. A.; McCormick, E. L.; Scully, D.
A.; Siderewicz, J. M.; Taylor, W. J.; Truesdale, S. J.; Wax, R. G. Org.
Process Res. DeV. 2002, 4, 477.
(9) (a) Griess, P.; Harrow, G. Chem. Ber. 1887, 20, 2205. (b) Hinsberg, O.
Chem. Ber. 1893, 26, 3092. (c) Ohle, H. Chem. Ber. 1934, 67, 155. (d)
Maurer, K.; Schiedt, B.; Schroeter, H.; Chem. Ber. 1935, 68, 1716.
(10) (a) Toldy et al. Acta Chim. Acad. Sci. Hung. 1954, 4, 303. (b) Muller, A.;
Varga, I. Chem. Ber. 1939, 72, 1993.
Another approach employing 2-chloroquinoxaline was to
displace the chloride with a nitrile. A neat mixture of
(11) Maurer, K.; Boettger, B. Chem. Ber. 1938, 71, 1183.
666
•
Vol. 8, No. 4, 2004 / Organic Process Research & Development
10.1021/op049951d CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/16/2004

potassium cyanide and 2-chloroquinoxaline was heated to
250 °C with only starting material being observed.
hydrazone of a hexose. Fructose was converted to the
fructose osazone and condensed with o-phenylenediamine
to provide a moderate yield of the quinoxaline skeleton
(Scheme 3). While some of these methods worked rather
Scheme 3
Direct chemical oxidation of 2-methylquinoxaline gener-
ally provides pyrazine dicarboxylic acid; however, an earlier
report described a procedure that converted a styryl group
to a carboxylic acid (Scheme 1). Condensation of 2-meth-
Scheme 1
well, the additional steps and potential hazardous reagents
made this approach less attractive. With these issues in mind,
a decision was made to optimize the route that employed an
unmodified sugar and o-phenylenediamine.
To this end, the synthesis of the quinoxaline ring system
was approached from the condensation of fructose, glucose,
or mannose with o-phenylenediamine (Scheme 4). A variety
ylquinoxaline with benzaldehyde provided the styryl deriva-
tive.6 At this stage, potassium permanganate was used to
cleave oxidatively the olefin to provide the desired product.
Unfortunately, oxidation only provided a 68% recovery of
material that required further purification (50-70% purity).
A final attempt to functionalize quinoxaline was inves-
tigated. In this approach, the literature cites methods where
acyl radicals add to quinoxaline to provide the desired
quinoxalinecarboxylic acid derivative in good yields.13 In
our hands, the yield rarely exceeded 30% (Scheme 2). The
Scheme 4
of conditions were investigated to optimize this transforma-
tion (Table 1). Conditions found in Table 1 were conducted
on 3-6 g of the appropriate sugar.
Scheme 2
Initial investigations duplicated the reported low yields
of the desired polyhydroxylated quinoxaline. The reaction
provided even lower yields of product when conducted at
>80 °C (entries 1, 2). Stronger acids such as trifluoroacetic
acid (entry 5) and hydrochloric acid failed to yield any
desired product. Variations on the equivalents of sugar and
phenylenediamine, as well as the rate of addition, did not
significantly improve the yields. The optimum temperature
for this condensation was briefly examined. After 1.5 h at
60 °C, no desired product was observed (entry 4). Only after
heating to 80 °C was the product formed, and heating at 80
°C for an extended period of time led to the best yield for a
single addition of fructose (entry 7).The next step of the
Scheme 5
resulting amide was easily hydrolyzed to provide the desired
acid (eq 5).
The approach that had the most success was an improved
route using the procedures9-11 that used o-phenylenediamine
and a monosaccharide. The earliest reports found that when
o-phenylenediamine was mixed with glucose,9a fructose,9b,c
or mannose,9d a crystalline product was formed. The ele-
mental analysis matched the expected condensation product.
However, the yields have been poor (∼15-30%). In later
publications,12 some preparations utilized the osazone or
synthesis of 2-quinoxalinecarboxylic acid was the oxidation
of the polyhydroxylated side chain. Initial attempts utilizing
basic hydrogen peroxide provided only a 30% yield of the
desired acid. After additional experiments, the yields were
increased to 70% (based on 1.5 g of the desired polyhy-
droxylated quinoxaline; a 60% yield was obtained on an 8.5
g reaction). One of the reasons for the yield improvement
can be attributed to the purity of the starting material used
in the peroxide oxidation. Subsequent oxidations from
(12) (a) Henseke, G.; Winter, M. Chem. Ber. 1956, 89, 956. (b) Henseke, G.;
Ba¨hner, K. J. Chem. Ber. 1958, 91, 1605.
(13) Minisci, F.; Gardini, G. P. Tetrahedron Lett. 1970, 11, 15.
Vol. 8, No. 4, 2004 / Organic Process Research & Development
•
667

Table 1. Conditions examined for the preparation of 2-tetrahydrobutylquinoxaline
phenylenediamine
entry
(equiv)
sugar (equiv)
solvent
temp (°C)
time (h)
yield (%)
1
2
3
4
5
6
7
1
1
1
1
2
1
1
mannose (1)
glucose (1)
glucose (1)
fructose (1)
fructose (1)
fructose (2)
fructose (1)
H2O/50% AcOH (10:1)
10% AcOH
H2O/50% AcOH (10:1)
H2O/50% AcOH (10:1)
H2O/50% TFA (4:1)
H2O/ 50% AcOH (10:1)
H2O/ 50% AcOH (10:1)
reflux
reflux
80
60
80
80
80
2
2
2
1.5
2
2
8
11
31
NR
decomp
25
18
46
material obtained from the fructose osazone path failed to
provide a reliable yield. However, material arising from the
fructose approach appears to be of higher quality, thereby
resulting in better yields in the following oxidation. It should
be also mentioned that the original side-chain oxidation
reported the addition of sodium hydroxide to a solution of
2-tetrahydroxybutylquinoxaline in 6% hydrogen peroxide at
70 °C.11 However, this order of addition resulted in an
exothermic reaction with severe foaming that was difficult
to control. Alternatively, if 30% hydrogen peroxide was
slowly added to a solution of the quinoxaline in aqueous
sodium hydroxide, minimal foaming was observed, but the
reaction was still quite exothermic. It was envisaged that
the excessive foaming formation resulted from the excess
hydrogen peroxide (14 volumes based on 2-tetrahydroxy-
butylquinoxaline). To examine this possibility, the amount
of hydrogen peroxide was halved to 7 volumes to provide
crude 2-quinoxalinecarboxylic acid (57% yield, based on 5
g of 2-tetrahydroxybutylquinoxaline). Under these conditions,
the mixture did not foam excessively, but the reaction was
still slightly exothermic. The amount of 30% hydrogen
peroxide was halved again to 3.5 volumes to afford a 57%
yield of the crude acid. No significant exotherm event or
foaming problems were encountered under these conditions,
and the product was formed in comparable purity and yield.
Experimental Section
Reagents and solvents were obtained from commercial
sources and used as received. Proton nuclear magnetic
resonance spectra were obtained on a Bruker AC-300
spectrometer at 300 MHz using dimethyl sulfoxide as the
solvent. Carbon nuclear magnetic resonance spectra were
obtained on a Bruker AMX-300 at 75 MHz using dimethyl
sulfoxide as the solvent. The melting points were obtained
on a Thomas-Hoover capillary melting point apparatus.
Preparation of 2-Tetrahydroxybutylquinoxaline (1).
1,2-Phenylenediamine (450 g, 4.16 mol) was placed in a 5-L
three-neck round-bottom flask equipped with a mechanical
stirrer and thermocouple. Aqueous acetic acid (2.0 L, 10%
solution) was added to the reactor, and the mixture was
stirred for 30 min. D-Fructose (750 g, 4.16 mol) was added
portionwise over 20 min. The mixture was heated to 80 °C
and stirred for 18 h at 80 °C. The reaction mixture was cooled
to 10 °C for 5 h. The solids were filtered through Shark
skin filter paper using a Bu¨chner funnel. The solids were
washed with water (2 × 500 mL). A rubber dam was placed
over the funnel, and the funnel was kept under vacuum for
3 h. The light-brown solids were placed in a vacuum oven
for 18 h (oven temp ) 60 °C, 1 mmHg). The desired product
1
was obtained in a 40% yield (400 g); mp 181-184 °C; H
NMR (300 MHz, DMSO-d6) δ 9.19 (s, 1H), 8.05-8.11 (m,
2H), 5.63 (d, 2H), 5.17 (d, 1H), 4.74 (d, 1H), 4.64 (d, 1H),
4.42 (t, 1H), 3.65-3.72 (m, 3H); 13C NMR (75 MHz,
DMSO-d6) δ 159.4, 145.2, 141.0, 140.8, 129.9, 129.2, 128.8,
128.6, 74.3, 72.5, 71.2, 63.5.
Conclusions
A cost-efficient and scaleable synthesis of 2-quinoxa-
linecarboxylic acid utilizes a two-step process. The first step
of the process involves a condensation reaction of D-fructose
with o-phenylenediamine in the presence of 50% acetic acid
to construct the quinoxaline skeleton. After a detailed
investigation on this process, the best yield was obtained
when the reaction temperature was kept at 80 °C. The second
step of the process involves an oxidation of the side chain
of 1 using 30% basic hydrogen peroxide as an oxidant. A
modification in the order of addition of the reagents and their
concentration proved to be critical to minimize the amount
of foaming and the exothermic event. The exothermic event
and foaming problems were completely eliminated when a
minimum amount of hydrogen peroxide was used in the
reaction. In summary, a large quantity of quinoxalinecar-
boxylic acid was synthesized via a two-step process in a 22%
overall yield starting from reasonably priced starting materi-
als. It is worth noting that no further purification process
was performed on the final product since this procedure
provides the product with an acceptable level of purity
(>95%).
Preparation of 2-Quinoxalinecarboxylic acid (2). In a
5-L four-neck round-bottom flask equipped with a mechan-
ical stirrer, thermocouple, reflux condenser, and liquid
addition funnel, sodium hydroxide (155.4 g, 3.88 mol,
pellets) was dissolved in water (1575 mL). Once the internal
temperature reached 40 °C, 2-tetrahydroxybutylquinoxaline
(70.0 g, 280 mmol) was added, and the suspension was
stirred for 45 min. A 30% solution of hydrogen peroxide
(130 mL) was introduced into the reaction mixture over a
15 min period (no change in internal temperature was
observed). The reaction mixture was slowly heated to 60
°C over a period of 3 h. Slowly, the remainder of the 30%
hydrogen peroxide solution (115 mL) was added over a
period of 45 min (maximum temperature reached was 91
°C). The reaction mixture was stirred for an additional 30
min. The reaction mixture was then refluxed for 15 min.
The reaction was cooled to 80 °C, and the liquid was
decanted away from the tar-like material. The tar-like
668
•
Vol. 8, No. 4, 2004 / Organic Process Research & Development

material was discarded. The reaction was allowed to cool to
room temperature. The pH of the solution was adjusted to 2
with concentrated hydrochloric acid. The mixture was
allowed to stand overnight at room temperature. The off-
white solids were filtered through Shark skin filter paper and
washed with water (2 × 100 mL). A rubber dam was placed
over the funnel, and the funnel was kept under vacuum for
3 h. The off-white solids were placed in a vacuum oven for
18 h (oven temp ) 60 °C, 1 mmHg). The desired product
2H), 7.91-8.08 (m, 2H); 13C NMR (75 MHz, DMSO-d6) δ
165.3, 145.2, 143.6, 142.8, 140.8, 132.4, 131.3, 103.1, 129.0,
128.2.
Acknowledgment
We thank Dr. Farahnaz Mohammadi from Albany Mo-
lecular Research, Inc., for her assistance with the preparation
of this manuscript.
Received for review March 3, 2004.
OP049951D
1
was obtained in a 53% yield (25.7 g); mp 209-210 °C; H
NMR (300 MHz, DMSO-d6) δ 9.50 (s, 1H), 8.20-8.31 (m,
Vol. 8, No. 4, 2004 / Organic Process Research & Development
•
669