Process development of 5-methoxy-1H-indole-2-carboxylic acid from ethyl 2-methylmalonate

Article abstract of DOI:10.1021/op980006x

Development is described of a new process for the preparation from malonates of 5-methoxy-1H-indole-2-carboxylic acid esters, useful intermediates in the synthesis of pharmaceutical compounds. The process uses readily available starting materials, produces little waste, can be operated safely on at least 1 molar scale, and gives high yields. The main areas of optimization included the azo coupling of a diazonium salt with malonate derivatives, the Japp-Klingemann rearrangement, and the Fischer indole synthesis.

Full text of DOI:10.1021/op980006x

Organic Process Research & Development 1998, 2, 214−220
Process Development of 5-Methoxy-1H-indole-2-carboxylic Acid from Ethyl
2-Methylmalonate
Yves Bessard†
Department of Process Research, LONZA Ltd., P.O. Box CH-3930 Visp, Switzerland
Scheme 1. Atevirdine mesylate (U-87201E, 1) and its
precursor (10)
Abstract:
Development is described of a new process for the preparation
from malonates of 5-methoxy-1H-indole-2-carboxylic acid es-
ters, useful intermediates in the synthesis of pharmaceutical
compounds. The process uses readily available starting materi-
als, produces little waste, can be operated safely on at least 1
molar scale, and gives high yields. The main areas of optimiza-
tion included the azo coupling of a diazonium salt with malonate
derivatives, the Japp-Klingemann rearrangement, and the
Fischer indole synthesis.
Strategy Leading to the Indole Ring. The choice of
synthetic pathway is very dependent on the complexity of
the indole target and on the availability of the starting
material. For the preparation of 5-methoxy-1H-indole-2-
carboxylic acid (10), the Fischer synthesis,4 which is certainly
the most widely used method, appears to be the most adapted
for our purpose. Two classical pathways can be envi-
sioned: reaction of a diazonium salt with a â-ketoester and
cyclization of the resulting hydrazonopyruvate [Pathway
(a)5-9] and condensation of a hydrazine with a pyruvate and
cyclization of the resulting hydrazonopyruvate [Pathway
(b)10-12] (Scheme 2). For economical reasons (availability
and costs of the hydrazine) it was decided to follow the
diazonium Pathway (a).
Introduction
5-Methoxy-1H-indole-2-carboxylic acid (10) has been
shown to be a useful intermediate for the preparation of
pharmaceutical products.1 Moreover, 10 has been employed
for the preparation of antiviral agents such as U-87201E2
(1, atevirdine mesylate; (Scheme 1), a non-nucleoside,
reverse transcriptase inhibitor of HIV. The increased chemi-
cal and medicinal applications of 10 and its derivatives have
led to renewed interest in the synthesis and chemistry of this
versatile heterocycle. As part of our program of development
of synthetic methods for the construction of heterocycles,
we report here a safe, economical, and reliable synthesis of
5-methoxy-1H-indole-2-carboxylic acid (10) from readily
available starting materials.3
The reaction of a diazonium salt with a â-ketoester is well
documented in the literature. Specifically, the 5-methoxy-
1H-indole-2-carboxylic acid (10) has been prepared from the
ethyl ester of 2-methylacetoacetic acid (13) and the di-
azonium salt 3 of p-anisidine (2; Scheme 3, process A). In
the most detailed procedure, published by T. Kralt,6 the
5-methoxy-1H-indole-2-carboxylic acid (10) was obtained
in a 58% overall yield from p-anisidine. We were able to
improve their procedure, and 10 was obtained in 75-80%
overall yield from p-anisidine. Nevertheless, this process
† Phone: (41) 27 948 5800. Fax: (41) 27 948 6180. E-mail:
yves.bessard@lonza.ch.
(1) (a) Le´on, P.; Garbay-Jaureguiberry, C.; Barsi, M. C.; Le Pecq, J. B.; Roques,
B. P. J. Med. Chem. 1987, 30, 2074. (b) Caubere, P.; Jamart-Gregoire, B.;
Caubere, C.; Bizot-Espiard, J. G.; Renard, P.; Adam, G. Eur. Pat. Appl. EP
624,575 (Chem. Abstr. 1995, 122, 81177f). (c) Morales-Rios, M. S.; Joseph-
Nathan, P. ReV. Soc. Quim. Me´x. 1989, 33, 331.
(2) (a) Romero, D. L.; Morge, R. A.; Biles, C.; Berrios-Pe`na, N.; May, P. D.;
Palmer, J. R.; Johnson, P. D.; Smith, H. W.; Busso, M.; Tan, C. K.;
Voorman, R. L.; Reusser, F.; Althaus, I. W.; Downey, K. M.; So, A. G.;
Resnick, L.; Tarpley, W. G.; Aristoff, P. A. J. Med. Chem. 1994, 37, 999.
(b) Livermore; D. G. H.; Bethell, R. C.; Cammack, N.; Hancock, A. P.;
Hann, M. M.; Green, D. V. S.; Lamont, R. B.; Noble, S. A.; Orr, D. C.;
Payne, J. J.; Ramsay, M. V. J.; Shingler, A. H.; Smith, C.; Storer, R.;
Williamson, C.; Willson, T. J. Med. Chem. 1993, 36, 3784. (c) Romero, D.
L.; Morge, R. A.; Genin, M. J.; Biles, C.; Busso, M.; Reusser, F.; Althaus,
I. W.; Resnick, L.; Tarpley, W. G.; Thomas, R. C. J. Med. Chem. 1993, 36,
1505. (d) Williams, T. M.; Ciccarone, T. M.; MacTough, S. C.; Rooney,
C. S.; Balani, S. K.; Condra, J. H.; Emini, E. A.; Goldman, M. E.; Greenlee,
W. J.; Kauffman, L. R.; O’Brien, J. A.; Sardana, V. V.; Schleif, W. A.;
Theoharides, A. D.; Anderson, P. S. J. Med. Chem. 1993, 36, 1291. (e)
Romero, D. L.; Thomas, R. C.; May, P. D.; Poel, T. J. (Upjohn Company,
USA) PCT Int. Appl. WO 96 18,628 (Chem. Abstr. 1996, 125, 142777g).
(f) Perrault, W. R.; Shephard, K. P.; LaPean, L. A.; Krook, M. A.;
Dobrowolski, P. J.; Lyster, M. A.; McMillan, M. W.; Knoechel, D. J.;
Evenson, G. N.; Watt, W.; Pearlman, B. A. Org. Proc. Res. DeV. 1997, 1,
106. (g) See also additional information in: Drugs of the Future 1992, 17,
891; 1993, 18, 57; 1994, 19, 9; 1996, 21, 72; and in Scrip Magazine October
1993, 39.
(3) Bessard, Y.; Imwinkelried, R. (LONZA Ltd.) Swiss Patent No. 687,327
(Chem. Abstr. 1997, 126, 144112f).
(4) Robinson, B. The Fischer Indole Synthesis; John Wiley & Sons: New York,
1982.
(5) Hughes, G. K.; Lions, F. J. Pr. R. Soc. N. S. Wales 1938, 71, 475.
(6) Kralt, T.; Asma, W. J.; Haeck, H. H.; Moed, H. D. Rec. TraV. Chim. Pays-
Bas 1961, 80, 313.
(7) Heath-Brown, B.; Philpott, P. G. J. Chem. Soc. 1965, 7185.
(8) Monge Vega, A.; Fernandez Alvarez, E. An. Quim. 1972, 68, 1153.
(9) Murakami, Y.; Yokoyama, Y.; Miura, T.; Hirasawa, H.; Kamimura, Y.;
Izaki, M. Heterocycles 1984, 22, 1211.
(10) Amorosa, M. Gazz. Chim. Ital. 1955, 85, 1445.
(11) Salituro, F. G.; Harrison, B. L.; Baron, B. M.; Nyce, P. L.; Stewart, K. T.;
Kehne, J. H.; White, H. S.; McDonald, I. A. J. Med. Chem. 1992, 35, 1791.
(12) Morales-Rios, M. S.; Joseph-Nathan, P. ReV. Soc. Quim. Me´x. 1989, 33,
331.
214
•
Vol. 2, No. 4, 1998 / Organic Process Research & Development
S1083-6160(98)00006-1 CCC: $15.00 © 1998 American Chemical Society and Royal Society of Chemistry
Published on Web 04/30/1998

Scheme 2. Pathway (a) from a diazonium derivative and Pathway (b) from a hydrazine derivative
Scheme 3. General reaction scheme: Process A from a â-ketoester and Process B from a malonatea
a Process A. (a) NaNO2/HCl, (b) CH3COONa (10 equiv) in C2H5OH/H2O, (c) HCl in C2H5OH, (g) KOH in H2O; overall yield, 80% (lit.6 58%). Process B. (a)
NaNO2/HCl, (d) Na2CO3 (1 equiv)/Et3N (1 equiv) in CH3OH/H2O, (e) C2H5ONa (0.2 equiv) in C2H5OH, (f) HCl (gas), (g) KOH in H2O; overall yield, 64%.
suffers two major drawbacks: ethyl 2-methylacetoacetate
(13) is not available in large quantities and the ring formation
is extremely exothermic, limiting the possibility of safe scale-
up. To overcome these drawbacks, we considered the use
of the readily available diethyl 2-methylmalonate (16) for
the coupling reaction with the diazonium salt 3. Surprisingly,
very little is known in the literature of the reaction of diethyl
2-methylmalonate and p-anisidine (Scheme 3, Process B).
Vol. 2, No. 4, 1998 / Organic Process Research & Development
•
215

Scheme 4. Coupling reaction: Process A from a â-ketoester, Process B from a malonate, and process C from
methyl-Meldrum’s acid
Table 1. Comparison of reaction conditions for the coupling reaction
entry no.
reactants
base/solvent systems
reaction time
conversiona to
1
2
3
4
5
6
7
8
9
ethyl 2-methylacetoacetate (13)
diethyl 2-methylmalonate (16)
diethyl 2-methylmalonate (16)
diethyl 2-methylmalonate (16)
diethyl 2-methylmalonate (16)
diethyl 2-methylmalonate (16)
diethyl 2-methylmalonate (16)
dimethyl 2-methylmalonate (15)
dimethyl 2-methylmalonate (15)
2-methylmalonic acid (14)
CH3COONa (10 equiv)/H2O/EtOH
CH3COONa (10 equiv)/H2O/EtOH
pyridine/H2O
Na2CO3(1.5 equiv)/pyridine/H2O
Na2CO3 (2.5 equiv)/pyridine/H2O/DMF
Na2CO3 (2 equiv)/H2O/MeOH
Na2CO3 (1 equiv)/Et3N (1 equiv)/H2O/MeOH
Na2CO3 (1 equiv)/Et3N (1 equiv)/H2O/MeOH
Na2CO3 (2 equiv)/H2O/MeOH
CH3COONa (10 equiv)/H2O/EtOH
CH3COONa (10 equiv)/H2O/EtOH
2 h
>90% (4)
10-15% (5)
60-65% (5)
85-90% (5)
90-95% (5)
85-90% (5)
85-90% (5)
>90% (11)
>90% (11)
b
22 h
24 h
24 h
20 h
20 h
3 h
1 h
4 h
4 h
<2 h
10
11
methyl-Meldrum’s acid (17)
>90% (6)
a The conversion of the reactant to the azo-intermediate was measured by 1H NMR. b Decomposition of 2-methylmalonic acid (14).
To our knowledge, 5-methoxy-1H-indole-2-carboxylic acid
(10) has never been prepared starting from a malonic acid
derivative, though some malonic acid derivatives have been
used for the preparation of azo- and hydrazono-intermediates
via coupling reactions with diazonium salts.13
1. Formation of the Diazonium Salt 3. The diazonium
salt was prepared under classical conditions by the reaction
of p-anisidine (2) with sodium nitrite in aqueous acid (0 °C/1
h). The reaction mixture was then used directly for the next
step (coupling reaction).
Even though the reaction of the diazonium salt of
p-anisidine14,15 with diethyl 2-methylmalonate and the reac-
tion between the diazonium salt of benzidine and toluidine
with the ethyl ester of 2-methyl(and 2-ethyl)malonic acid16
were reported more than 60 years ago, this concept has never
been utilized for indole formation. The low yields obtained
for these reactions can explain why the use of malonic
derivatives has never been attractive in indole chemistry.
2. Coupling Reaction (Scheme 4). The coupling
reaction was first studied following the â-ketoester route
(Process A). This reaction is almost always carried out under
basic conditions. Although potassium hydroxide5 was
sometimes used, sodium acetate6,9 was preferred, due to the
milder conditions, providing a cleaner reaction. Using the
reaction conditions described in the literature6 (Table 1, entry
1), ethyl 2-methylacetoacetate (13) was rapidly converted
to the corresponding azo-intermediate17 4 (>90% conver-
sion/2 h/0 °C). Nevertheless, such a good conversion could
only be obtained by using a very large excess of sodium
acetate (up to 10 equiv), which would drastically increase
the costs of the waste treatment. As mentioned, another
major drawback was that 2-methyl derivatives of â-ketoesters
are not available in bulk quantities.18,19
Results and Discussion
The two processes (Process A and Process B) were studied
in order to identify their scope and limitations for large scale
manufacture.
(13) (a) Heckendorn, R. HelV. Chim. Acta 1987, 70, 2118. (b) Heckendorn, R.
HelV. Chim. Acta 1990, 73, 1700. (c) Heckendorn, R. Bull. Soc. Chim. Belg.
1986, 95, 921. (d) Szantay, Cs.; Szabo, I.; Kalaus, Gy. Synthesis 1974, 354.
(14) Favrel, G. Bull. Soc. Chim. Fr. 1930, 47, 1290.
(15) Favrel, G. C. R. Hebd. Seances Acad. Sci. 1901, 132, 1336.
(16) Favrel, G. Bull. Soc. Chim. Fr. 1902, 27, 324.
(17) For simplificity, the intermediate formed during the coupling reaction, is
called the azo-intermediate (using the Heath-Brown and Philpott7 terminol-
ogy).
216
•
Vol. 2, No. 4, 1998 / Organic Process Research & Development

Scheme 5. Japp-Klingemann rearrangement
When we tried to couple diethyl 2-methylmalonate (16)
(Process B) with the diazonium salt under the same condi-
tions (entry 2), very disappointing results were obtained. The
conversion to the azo intermediate 5 was only 10-15%
complete after 2 h. Due to the relatively weak acidity of
the activated methylene group of the malonate 16 compared
to that of the â-ketoester 13, more strongly basic conditions
were required. However, the use of strong bases (sodium
and potassium hydroxide) led to an increasing formation of
side-products. Since the pH and also the polarity of the
reaction mixture were very important, a number of combina-
tions of solvent systems (water, methanol, ethanol, pyridine
dimethylformamide) and bases (sodium acetate and carbon-
ate, potassium acetate and carbonate, sodium and potassium
hydrogen carbonate, sodium and potassium hydroxide, tri-
methylamine, triethylamine, triisopropylamine) were then
studied. A selection of the results are reported in Table 1.
The optimum pH was found to be in the region of 9.5 (
0.5. The best result was obtained when the reaction was
carried out in a methanol/water mixture with 1 equiv of
sodium carbonate and 1 equiv of triethylamine. Under these
conditions 85-90% conversion was achieved within 3 h at
0 °C (entry 7). Under the same conditions, the reaction with
the corresponding dimethyl ester 15 is notably faster (entry
8). Even under relatively mild conditions (sodium acetate),
the free acid 14 decomposes before undergoing any reactions
(entry 10).
corresponding azo-intermediate 6 (entry 11). However, this
alternative route was, unfortunately, economically not com-
petitive.
3. Japp-Klingemann Rearrangement (Scheme 5).
Under the acidic conditions in Process A, the azo-intermedi-
ate 4 rearranges with cleavage of the acetyl group to give
the hydrazono-intermediate20 7 which immediately undergoes
Fischer cyclization to the indole ester 9. This “one-step”
reaction sequence is extremely fast and exothermic. In the
original preparation,6 a solution of HCl gas in ethanol was
added in one portion at 0 °C to 4 and the temperature rose
to reflux within a few seconds. On the other hand, no
reaction occurred if the temperature was maintained under
40 °C but as soon as the temperature reached 50 °C it was
almost impossible to keep the reaction under control. A slow
adddition of HCl was not helpful because there was an
induction period with almost no reaction, and then the
temperature suddenly rose to the boiling point. The same
problems are encountered with the reverse addition. The
best way was to introduce HCl to a refluxing solution of 4
in ethanol, but even under these conditions, the reaction was
hardly controllable, even at 100 mmol scale. Scale-up was
thus almost impossible.
In Process B, the azo-intermediate 5 did not rearrange
under acidic conditions to give the hydrazono-intermediate
7. The rearrangement, with cleavage of an ester group,
occurred only under basic conditions. At this stage, the key
point was to obtain the hydrazono-intermediate 7 selectively.
With mild bases (NaHCO3, Na2CO3, K2CO3) no reaction
occurred, and under relatively strong basic conditions
(NaOH/water, NaOH/ethanol, or NaOH/water/ethanol), the
As expected, the methyl-Meldrum’s acid (17) (Process
C) reacts very quickly with the diazonium salt to give the
(18) This was the situation when the project was initiated.
(19) Ethyl 2-methylacetoacetate (13) is most often prepared by alkylation of ethyl
acetoacetate. The use of cheap alkylating agents (such as dimethyl sulfate)
often requires a large excess of reagent, leading to a greater amount of
undesired bis-alkylated by-product and an expensive separation. On the other
hand, selective akylation methods use expensive alkylating agents.
(20) For simplificity, the intermediate formed during the Japp-Klingemann
reaction, is called the hydrazono-intermediate (using the Heath-Brown and
Philpott7 terminology).
Vol. 2, No. 4, 1998 / Organic Process Research & Development
•
217

Scheme 6. Fischer indole synthesis
hydrazono acid 8 was already present with the expected
hydrazono ester 7 before all of the azo ester 5 had reacted,
the ratio between the three compounds depending on the
excess of the base. Eventually, selective and fast reaction
conditions were found by using alcoholates (t-BuOK, t-
BuONa, CH3OK, CH3ONa, EtONa). During optimization,
the amount of base was progressively reduced from 2 to 0.2
equiv. Thus, the NaOEt (0.2 equiv)/ethanol system gave
selectively the desired hydrazono-intermediate 7 in 15 min
at room temperature. In this process, the hydrazono-
intermediate 7 did not have to be isolated, and the reaction
mixture was used directly for the next step (the Fischer
cyclization). However, for analytical purposes, a reaction
was worked up and the hydrazono-intermediate 7 was
isolated in 90% yield from p-anisidine.
The azo-intermediate 6 of the methyl-Meldrum’s acid
(17), Process C, behaved similarly compared to the azo-
intermediate 5. Thus, under conditions developed for the
Process B, the hydrazono-intermediate 7 was also obtained
selectively.
4. Fischer Indole Synthesis (Scheme 6). For the Fischer
cyclization, generally an ethanolic solution of HCl gas21 has
been used.5-7 The use of p-toluenesulfonic acid9 as well as
sulfuric acid10 has also been reported (many other conditions
such us Lewis acids in acetic acid have also been used, but
for differently substituted indoles). As already mentioned,
the azo-intermediate 4 (Process A) can be directly converted,
under acidic conditions, to the indole ester 9 without isolation
of the hydrazono-intermediate 7.
acids such as HClO4) were unsuccessful (recovery of the
starting material or unidentified material). Therefore, the
azo-intermediate 5 was transformed to the hydrazono-
intermediate 7 under basic conditions (NaOEt (0.2 equiv)/
ethanol) and, without any work-up, the reaction mixture
heated to reflux while adding gaseous HCl. The excess of
HCl was progressively reduced from 76 to 3 equiv. Before
we were able to develop a selective method for the
transformation of the azo-intermediate 5 to the hydrazono-
intermediate 7, the only clean reaction achievable was the
transformation of the azo-intermediate 5 to the hydrazono
acid 8. Unfortunately, this could be converted to the indole
acid 10 in only moderate yield (∼50%).
5. Hydrolysis of the Indole Ester (Scheme 6). The
literature procedure6 was modified in order to increase the
throughput of this step. A suspension of the indole ester 9
in water was heated to reflux in the presence of potassium
hydroxide (∼1.2 equiv). In less than 1 h, a clear solution
was obtained. After acidification, the indole acid 10
precipitated and could then be isolated by filtration. The
yield of this step was practically quantitative (>98%).
6. “Methyl Ester” Pathway (Scheme 6). The azo
intermediate 11 obtained from the dimethyl 2-methyl-
malonate (15) was also transformed (after the Japp-
Klingemann rearrangement to 18) to the indole ester 12 and
then to the indole acid 10 with 65% overall yield from
p-anisidine, following Process B.
Conclusions
Even with the improvements made (especially the increase
in yield from 58 to 80%), the original procedure (Process
A) presents too many drawbacks: the excessive amount of
salts generated by the process (needed for good conversion
in the coupling reaction); the highly exothermic reaction
leading to the indole ester 9 (reaction technically not
During the development of the Process B, considerable
effort was expended to transform the azo-intermediate 5 to
the indole ester 9 in one step. However, all of the conditions
tested (longer reaction time, up to 16 h, or use of stronger
(21) In presence of water, the reaction is inhibited.
218
•
Vol. 2, No. 4, 1998 / Organic Process Research & Development

controllable in large scale); and the availablity of the
expensive ethyl 2-methylacetoacetate in bulk quantities. With
Process B, the indole acid 10 is isolated in 65% overall yield
from p-anisidine, with an average purity normally >99%.
Compared to Process A, there is a significant reduction of
the amount of salts; the exothermicity of the cyclization step
can be safely controlled, and readily available starting
materials can be used. Even with a lower overall yield, this
process is economically more competitive than the Process
A. Process B was successfully scaled up from 100 mmol
(routine laboratory experiments) to 1000 mmol without
encountering any difficulties, yield drop, or loss of product
purity. The alternative route (Process C) using the methyl-
Meldrum’s acid for the coupling reaction is chemically
interesting but, owing to the cost of Meldrum’s acid, too
expensive.
stirring and external cooling and heating. To the crude azo
diester (5, 325 g) in ethanol (400 mL) a 21% solution of
sodium ethanolate (65.0 g, 200 mmol) is added dropwise
(in 20 min) at 25-30 °C. After 15 min, the conversion is
checked by TLC (2:1 hexane/ethyl acetate). A sample was
taken and worked up for analytical purposes. Beige solid,
mp 96-98 °C (lit.7 mp 94 °C; lit.23 mp 97 °C). 1H NMR
(CDCl3): δ 7.6 (br s, 1H), 7.15 (d fine structure, 2H), 6.87
(d fine structure, 2H), 4.32 (q, 2H), 3.80 (s, 3H), 2.10 (s,
3H), 1.38 (t, 3H).
Ethyl 5-Methoxy-1H-indole-2-carboxylate (9). The reac-
tion mixture of 7 is heated to reflux, and, simultaneously,
gaseous HCl (120 g, 3290 mmol) is added over 2h. Reflux
is maintained for 15 min after the end of addition of HCl.
The reaction mixture is cooled to room temperature, and
water (100 mL) is added. The reaction mixture is then
cooled to 0 °C for 2.5 h. The suspension is filtered and
washed first with precooled (0 °C) ethanol (4 × 100 mL)
and then with water (2 × 250 mL). The cake is pressed as
dry as possible, leaving the crude indole ester as a yellow
powder (9, ∼146 g). A sample was taken for analytical
purposes. Yellow solid, mp 159-161 °C (lit.7 mp 154-
157 °C; lit.9 mp 158-160 °C). 1H NMR (CDCl3): δ 9.1
(br s, 1H), 7.31 (d, 1H, J ) 8.9 Hz), 7.15 (d, 1H, J ) 0.9
Hz), 7.08 (d, 1H, J ) 2.4 Hz), 7.00 (dd, 1H, J ) 6.5, 2.5
Hz), 4.41 (q, 2H, J ) 7.2 Hz), 3.84 (s, 3H), 1.41 (t, 3H, J )
7.1 Hz). 13C NMR (DMSO-d6): δ 162.1, 154.8, 132.4,
128.0, 127.9, 117.0, 112.8, 108.2, 102.7, 61.0, 55.7, 14.4.
5-Methoxy-1H-indole-2-carboxylic Acid (10). The reac-
tion is conducted in a 2.5 L reactor with mechanical stirring
and external cooling and heating. To a suspension of the
crude indole ester (9, ∼146 g) in water (1300 mL) is added
potassium hydroxide (56 g, 850 mmol). The suspension is
heated to reflux for 1 h. The resulting clear solution is then
cooled to 0 °C, and concentrated aqueous HCl (150 mL) is
added (during the addition of HCl a heavy white precipitate
is formed). After 1.5 h at 0 °C, the precipitate is filtered,
washed with water (2 × 500 mL), and dried (vacuum, 20
°C, 16 h) to give 123.6 g (64%; overall yield from
p-anisidine) of white solid (99.2% assay by titration), mp
197.0-198.3 °C (lit.6 mp 196 °C; lit.10 mp 192-193 °C/
196 °C recrystallized from water). 1H NMR (DMSO-d6):
δ 12.8 (br s, 1H), 11.6 (br s, 1H), 7.35 (d, 1H, J ) 8.9 Hz),
7.10 (s, 1H), 7.02 (d, 1H, J ) 0.5 Hz), 6.91 (dd, 1H, J )
6.6, 2.3 Hz), 3.71 (s, 3H). 13C NMR (DMSO-d6): δ 162.7,
153.8, 132.6, 128.6, 127.1, 115.7, 113.3, 106.9, 102.0, 55.2.
Process A. From Ethyl 2-Methylacetoacetate22 (13).
Optimization is based on the literature procedure.6
Experimental Section
General Procedures. Reagents and solvents were re-
agent grade and used as received. All reactions were
conducted under nitrogen. Melting points were determined
on a Bu¨chi 535 apparatus and have not been corrected. 1H
NMR (400 MHz) and 13C NMR (400 MHz) spectra were
recorded on a Varian spectrometer. Chemical shifts are
reported as parts per million. Tetramethylsilane was used
as internal standard. Coupling constants J are given in hertz.
Process B. From 2-Methylmalonic Acid Diethyl Es-
ter22 (16). Diazonium Salt Solution 3. The reaction is
conducted in a 2.5 L reactor with mechanical stirring and
external cooling. A solution of sodium nitrite (73.2 g, 1060
mmol) in water (200 mL) is slowly added to a solution of
p-anisidine (125.7 g, 1000 mmol) in concentrated HCl (220
mL, 2100 mmol) and ice (1000 g). The temperature is kept
at 0 °C during the course of the addition (40 min) and then
for an additional 30 min.
Diethyl 2-(4-Methoxyphenylazo)-2-methylmalonate (5).
The reaction is conducted in a 10 L reactor with mechanical
stirring and external cooling. The diazonium salt solution
3 is added in portions (in 40 min) to a 0 °C stirred suspension
of diethyl 2-methylmalonate (16) (174.9 g, 1000 mmol),
sodium carbonate (106.5 g, 1000 mmol), and triethylamine
(100.2 g, 985 mmol) in water (500 mL) and methanol (1000
mL). After 2.5 h at 0 °C, water (2000 mL) is added to the
brown suspension (pH 10.1). The resulting solution is
concentrated (750 mL distillate) under vacuum (30 °C/50
mmHg). The reaction mixture is then extracted with toluene
(3 × 1000 mL), and the solvent evaporated under vacuum
(30 °C/50 mmHg), leaving the crude azo diester 5 as a red
oil (∼325 g). A sample was purified by chromatography
(silica gel, 2:1 ethyl acetate/hexane) for analytical purposes.
The red oil decomposes upon heating. 1H NMR (CDCl3):
δ 7.75 (d fine structure, 2H), 6.95 (d fine structure, 2H),
4.30 (q, 4H), 3.85 (s, 3H), 1.78 (s, 3H), 1.30 (t, 6H).
Ethyl 2-[(4-Methoxyphenyl)hydrazono]propionate (7).
The reaction is conducted in a 2.5 L reactor with mechanical
Diazonium Salt Solution 3. The reaction is conducted in
a 0.5 L reactor with mechanical stirring and external cooling.
A solution of sodium nitrite (7.5 g, 108 mmol) in water (20
mL) is added slowly to a solution of p-anisidine (12.6 g,
100 mmol) in 15% HCl (65 mL, 260 mmol) and ice (100
g). The resulting solution is kept at 0 °C during the course
of the addition (40 min) and then for an additional 30 min
before being used in the next step.
(22) The process can be followed by TLC (ethyl acetate/hexane); the order of
increasing polarity is as follows: indole acid > hydrazono ester > indole
ester > azo diester.
(23) Rydon, H. N.; Siddappa, S. J. Chem. Soc. 1951, 2462.
Vol. 2, No. 4, 1998 / Organic Process Research & Development
•
219

Ethyl 2-(4-Methoxyphenylazo)-2-methyl-3-oxobutanoate24
(4). The reaction is conducted in a 1 L reactor with
mechanical stirring and external cooling. The diazonium salt
solution 3 is added in portions (in 10 min) to a 0 °C stirred
suspension of ethyl 2-methylacetoacetate (13) (17.6 g [91%
assay], 111 mmol), sodium acetate (82 g, 1000 mmol) in
water (100 mL), and ethanol (100 mL). After 2 h at 0 °C,
the reaction mixture is slowly warmed to room temperature
and then extracted with toluene (5 × 80 mL). The toluene
extracts are dried with anhydrous sodium sulfate, and the
solvent is evaporated under vacuum (30 °C/50 mmHg),
leaving the crude azo diester (7) as a red oil (∼30 g). A
sample was taken for analytical purposes. The red oil
decomposes upon heating. 1H NMR (CDCl3): δ 7.76 (d,
2H), 6.98 (d, 2H), 4.27 (m, 2H), 3.88 (s, 3H), 2.34 (s, 3H),
1.64 (s, 3H), 1.28 (t, 3H).
Ethyl 5-Methoxy-1H-indole-2-carboxylate (9). The reac-
tion is conducted in a 1 L reactor with mechanical stirring
and external cooling. The crude product from the previous
step (7, ∼30 g) is dissolved in 0 °C ethanol (20 mL). To
this solution is added cold ethanol satured with dry hydrogen
chloride gas (50 mL, ∼500 mmol of HCl) in one portion,
while stirring. Upon the addition, the temperature rises
quickly to the boiling point and violent reflux continues for
3-5 min. The reaction mixture is stirred for another hour
at reflux. The reaction mixture is then cooled to 0 °C, and
water (100 mL) is added. The cold suspension is filtered
and washed with precooled ethanol (10 mL), then with water
(2 × 20 mL), and again with ethanol (10 mL).
addition (10 min) and then for an additional 30 min before
being used in the next step.
5-[(4-Methoxyphenyl)azo]-2,2,5-trimethyl-1,3-dioxane-
4,6-dione (6). The reaction is conducted in a 1 L reactor
with mechanical stirring and external cooling. The diazo-
nium salt solution 3 is added in portions (in 10 min) to a 0
°C stirred suspension of 2,2,5-trimethyl-1,3-dioxane-4,6-
dione (methyl-Meldrum’s acid, 17) (8.0 g, 50 mmol) and
sodium acetate (41 g, 500 mmol) in water (50 mL) and
ethanol (50 mL). After 2 h at 0 °C, water (200 mL) is added
to the reaction mixture and the yellow suspension is filtered,
washed with water (2 × 50 mL), and dried (vacuum, 20 °C,
16 h) to give 12.3 g (84%) of 6, mp 80-81 °C. 1H NMR
(CDCl3): δ 7.76 (d, 2H), 6.97 (d, 2H), 3.89 (s, 3H), 1.95 (s,
3H), 1.82 (s, 3H), 1.78 (s, 3H).
Ethyl 2-[(4-Methoxyphenyl)hydrazono]propionate (7).
The reaction is conducted in a 0.5 L reactor with mechanical
stirring and external cooling and heating. To a solution of
5-[(4-methoxyphenyl)azo]-2,2,5-trimethyl-1,3-dioxane-4,6-
dione (6) (1.50 g, 5.1 mmol) in ethanol (10 mL) is added
dropwise a 21% solution of sodium ethanolate (1.92 g, 6.0
mmol) in ethanol (during 5 min) at room temperature. The
solution is then stirred for 60 min. Water (30 mL) is added,
and the reaction mixture extracted with ethyl acetate (3 ×
15 mL). The combined organic phases are dried (magnesium
sulfate), and the solvent is evaporated to leave 1.15 g (95%)
of ethyl 2-[(4-methoxyphenyl)hydrazono]propionate (7).
Data of Other Isolated Intermediates. Dimethyl 2-(4-
Methoxyphenylazo)-2-methylmalonate (11). Prepared fol-
lowing Process B, from methyl 2-methylacetoacetate (15).
A sample was taken for analytical purposes. The red oil
decomposes upon heating. 1H NMR (CDCl3): δ 7.75 (d,
2H), 6.95 (d, 2H), 3.88 (s, 3H), 3.83 (s, 6H), 1.70 (s, 3H).
Methyl 5-Methoxy-1H-indole-2-carboxylate (12). Pre-
pared following Process B, from dimethyl 2-(4-methoxy-
phenylazo)-2-methylmalonate (11). Grey solid, mp 178-
180 °C. 1H NMR (CDCl3): δ 9.2-9.0 (br s, 1H), 7.32 (d,
1H), 7.15 (d, 1H), 7.08 (d, 1H), 7.00 (dd, 1H), 3.95 (s, 3H),
3.86 (s, 3H).
5-Methoxy-1H-indole-2-carboxylic Acid (10). The reac-
tion is conducted in a 1 L reactor with mechanical stirring
and external cooling and heating. To a suspension of the
crude indole ester (9, ∼20 g) in ethanol (120 mL) is added
potassium hydroxide (5.9 g, 90 mmol). The suspension is
heated to reflux for 0.5 h and then poured into ice-water
(250 mL) and acidified with concentrated HCl (25 mL) to
pH 1 (a heavy white precipitate is formed). The precipitate
is filtered, washed with water (2 × 25 mL) and dried
(vacuum, 20 °C, 16 h) to give 15.9 g (81% overall yield
from p-anisidine [58% overall yield lit.6]) of white solid
(97.4% assay by titration).
Process C. From Methyl-Meldrum’s Acid22 (17).
Diazonium Salt Solution 3. The reaction is conducted in a
0.5 L reactor with mechanical stirring and external cooling.
A solution of sodium nitrite (3.5 g, 50 mmol) in water (10
mL) is added slowly to a solution of p-anisidine (6.3 g, 50
mmol) in 15% HCl (22 mL, 130 mmol) and ice (50 g). The
resulting solution is kept at 0 °C during the course of the
2-[(4-Methoxyphenyl)hydrazono]propionic Acid (8). By-
product formed during Process B (Scheme 4). A sample
was isolated for analytical purposes. White solid, mp 130-
132 °C. 1H NMR (CDCl3): δ 7.8-7.7 (br s, 1H), 7.12 (d,
2H), 6.91 (d, 2H), 3.82 (s, 3H), 2.13 (s, 3H).
Acknowledgment
I thank my colleagues in the Chemical Research and
Development department for their support and especially
Andy Naughton and Walter Brieden for their valuable advice
during the redaction of this manuscript.
(24) An extended study has been carried out by Heath-Brown and Philpott7 and
confirms that the intermediate isolated after the coupling reaction is the
azo-intermediate 4 and not the hydrazono-intermediate 7. The latter is formed
in acidic conditions and is usually not isolated but is directly transformed
to the indole ester 9.
Received for review January 26, 1998.
OP980006X
220
•
Vol. 2, No. 4, 1998 / Organic Process Research & Development