A continuous kilogram-scale process for the manufacture of 7-ethyltryptophol

Article abstract of DOI:10.1021/op1003083

An expeditious and multikilogram-scale process for the synthesis of 7-ethyltryptophol via a continuous flow reactor from 2-ethylphenylhydrazine and 4-hydroxybutyraldehyde in higher and high yield was described. The main steps in this synthesis involved not only the generation of the hydrazone intermediate in situ but also the catalysis of the subsequent [3 + 3] sigmatropic rearrangement in the tandem loop reactor. Decomposition of the intermediate hydrazone was found to be a key factor resulting in low yield.

Full text of DOI:10.1021/op1003083

TECHNOLOGY REPORT
pubs.acs.org/OPRD
A Continuous Kilogram-Scale Process for the Manufacture of
-Ethyltryptophol
7
†
,‡
†
,†
Yanwen Lv, Zhiqun Yu, and Weike Su*
†
Key Laboratory of Pharmaceutical Engineering of Ministry of Education, College of Pharmaceutical Sciences, Zhejiang University of
Technology, Hangzhou 310014, P.R. China.
‡
College of Chemical and material Engineering, Quzhou University 324000, P.R. China.
ABSTRACT: An expeditious and multikilogram-scale process for the synthesis of 7-ethyltryptophol via a continuous flow reactor
from 2-ethylphenylhydrazine and 4-hydroxybutyraldehyde in higher and high yield was described. The main steps in this synthesis
involved not only the generation of the hydrazone intermediate in situ but also the catalysis of the subsequent [3 þ 3] sigmatropic
rearrangement in the tandem loop reactor. Decomposition of the intermediate hydrazone was found to be a key factor resulting in
low yield.
6
’
INTRODUCTION
of trifluoroacetic acid; (3) one-pot synthesis of the indole nucleus
7
using K-10 montmorillonite as catalyst; (4) one-pot synthesis with
Lewis acidic ionic liquids such as 1-butylpyridinium chloride 3AlCl
or choline chloride 2ZnCl as catalysts and solvents. However,
several kinds of reported byproducts might form through further
reactions of the intermediate hydrazone 3 or 1 with excess 2,3-
dihydrofuran (DHF) (Figure 1) during the one-pot synthesis me-
thods.
Our previous work on the synthesis of 1 has achieved a
certain progress in a batch reactor by taking the following
methods including the control of the pH during the formation
of the hydrazone, the removal of excess 4-hydroxybutyraldehyde
Application of continuous flow synthesis is growing rapidly in
3
3
both academia and the pharmaceutical industry. Continuous flow
reactors in synthesis offer the following unique advantages over
traditional batch vessels: (a) the decrease of reactor size and
increase in surface-to-volume ratio significantly reinforce mass
and heat transfer, (b) the feasibility and device flexibility of
continuous flow synthesis offer fewer transport limitations which
minimizes chemical hazards and moderately harsh conditions,
8
3
2
9
10
(c) precise control of reaction variables such as temperature,
pressure, residence time, and stoichiometry enhances the control
of the overall reaction rates and improves yield and selectivity,
with NaHSO , the adjustment of the addition rate of H SO , the
(
d) multistep synthesis in a single continuous flow process can be
set up by connecting several reactors or processes in series, and
e) continuous flow synthesis can readily be scaled-up in
3
2
4
dilution of the reaction system, and extraction of the product
formed in situ in the Fischer cyclization step. However, the
product yield was still not satisfying, and the process was a little
time consuming. Accordingly, we focused on a scalable and
efficient continuous process for the Fischer indole synthesis of 1.
(
1
numbering up and is ideal for automation.
Widespread occurrence of indoles in natural products and
biologically active compounds has led to a continued interest in
2
the practical synthesis of the indole nucleus. Of prevailing
pharmaceutical importance is 7-ethyltryptophol which is a key
intermediate for the clinically effective analgesic and antiinflam-
matory drug etodolac. Etodolac has been shown to possess an
exceptional safety profile with respect to the gastrointestinal and
renal tract, and it has also been proved to have the potential to
retard the progression of skeletal changes in rheumatoid arth-
’
RESULTS AND DISCUSSIONS
Process A. Our original process discussed in this document
was the direct conversion of the commonly used one-pot batch
methodology to a homogeneous continuous flow system.
Scheme 2 shows the continuous flow reactor experimental setup.
Solution A of commercially available 2 hydrochloride in ethylene
glycol/water (2:5) with a concentration of 0.8 M together with
solution B of DHF (1 equiv) in ethylene glycol/water (2:5) with
a concentration of 0.8 M were pumped into the residence loop
reactor (Hastelloy tube, 7 mm o.d., 6 mm i.d.) via a T-joint (7
mm i.d.) by two peristaltic pumps (P1, P2, Boading Longer,
China), respectively. The mixture flowed through residence loop
I which was preheated and kept at 100 °C by hot oil at a total flow
rate of 200 mL/min. The reactor allowed for a mixing period in
which the hydrazone could be generated immediately in situ and
3
ritis.
Despite the diverse and creative approaches that have been
developed so far, the classical Fischer indole synthetic methodol-
ogy which involves hydrazone formation and subsequent [3 þ 3]
sigmatropic rearrangement remains the benchmark method
4
(
Scheme 1). The development of one-pot approaches to the
assembly of the indole skeleton which obviates the isolation of
the unstable arylhydrazones is attracting considerable attention
due to both its economical and its ecological importance. Some
reported examples within this context are (1) one-pot synthesis
of 5-substituted N,N-dimethyltryptamine with 4% H SO as
2
4
5
catalyst; (2) one-pot synthesis of 1,2,3,4-tetrahydrocarbazole
Received: November 15, 2010
Published: March 18, 2011
with trifluoroacetic anhydride as catalyst, necessitating the recycling
r 2011 American Chemical Society
471
dx.doi.org/10.1021/op1003083
|Org. Process Res. Dev. 2011, 15, 471–475

Organic Process Research & Development
TECHNOLOGY REPORT
then catalyzed the subsequent [3 þ 3] sigmatropic rearrange-
ment in residence loop I; reaction was terminated in residence
loop II by cooling the mixture by cold water. After a residence
time of 30 min in loop I, streams were typically analyzed by
Process B. According to the problems of the first method, an
improved process was developed by generating the hydrazone 3
under neutral condition in residence loop I, and then catalyzing the
cyclization with 8% aqueous sulfuric acid in the tandem loop reactor
II (Scheme 4). Experimental parameters (flow rate, residence time,
and temperature) of synthesizing hydrazone were initially optimized
separately. The temperature of 60 °C with a residence time of about
5 min led to a conversion of 100% by HPLC. Isolation of this
intermediate was unnecessary; the hydrazone was used directly in the
cyclization step in a continuous flow manner. The [3 þ 3]
sigmatropic rearrangement catalyzed by sulfuric acid was undertaken
at 100 °C in the tandem residence loop II. With the aim at quenching
the reaction at reaction completion, the mixture was cooled to nearly
20 °C in loop III, and the residual acid was neutralized by 30%
aqueous NaOH. Full conversion of 3 was completed in 16 min;
however, side reactions were still detected by HPLC including 4
1
1
HPLC at about 65% of 1, less than 5% of 3, and byproducts
totally about 30%). Further analysis of byproducts indicated
(
only a trace of triol 4; however, a quantity of 7a and a novel
byproduct 8 (Scheme 3). The results demonstrated the con-
secutive reaction between 1 and DHF has dropped substantially
in the continuous flow process; however, hydrazone 3 still could
react with excess DHF while cyclizing to indole, and could also
cyclize to cinnoline.
Scheme 1. Synthesis of 7-ethyltryptophol 1
(
0.8%), 7a (13%), and 8 (8%). Yield of 1 was only about 55%
corrected for purity. After extensive efforts, we found that reducing
the concentration from 0.4 to 0.1 M reduced the rate formation of 8;
however, the impurty 7a remained. We were confused as to where
the DHF or 4-hydroxybutyraldehyde came from? The only possible
source was the decomposition of hydrazone 3.
Decomposition study. An extra decomposition test of 3 was
systematically investigated by varying temperature and acidity in
the same solvents and concentration with the above process
(
Figure 2). It was found that the decomposition rate was not so
fast within the first few minutes, which makes possible a higher
yield process by shortening the reaction time.
Scheme 3. Novel Side Reaction
Figure 1. Byproducts in the preparation of 1.
a
Scheme 2. Schematic of Process A
a
Solution A is 0.8 M 2 hydrochloride in ethylene glycol/water (2:5) with a flow rate of 100 mL/min, solution B is 0.8 M DHF in ethylene glycol/water
2:5) with a flow rate of 100 mL/min.
(
4
72
dx.doi.org/10.1021/op1003083 |Org. Process Res. Dev. 2011, 15, 471–475

Organic Process Research & Development
TECHNOLOGY REPORT
a
Scheme 4. Schematic of Process B
a
Solution A is 0.2-0.8 M 2 hydrochloride and NaOH (1 equiv) in ethylene glycol/water (2:5) with a flow rate of 100 mL/min, solution B is 0.2-0.8 M
DHF in ethylene glycol/water (2:5) together with 0.5-1 mL hydrochloric acid stirred for 10 minutes in advance with a flow rate of 100 mL/min;
solution C is 8% sulfuric acid with a flow rate of 50 mL/min.
Figure 2. With the increase of temperature and the decrease of pH values, the hydrazone decomposed much more rapidly. These procedures generated
only a small amount of product 1: pH = 7 (none), pH = 6 (3%), pH = 5 (10%).
Process C. On the basis of the results of decomposition study,
a new optimized process was then undertaken, and optimal
conditions were achieved by flowing a solution of 2 in ethylene
glycol/water (5:2) together with a stream of 4-hydroxybutyr-
aldehyde (1 equiv, generated from DHF beforehand) in ethylene
glycol/water (5:2) at 115 °C for 20 s, followed by the introduc-
tion of 115 °C 50% aqueous sulfuric acid into the loop reactor (7
mm o.d., 6 mm i.d.) for about 4 min. The mixture was cooled
when it flowed through the cooling loop, and the reaction was
terminated by neutralizing the residual acid with 30% aqueous
NaOH (Scheme 5). Ethylene glycol was removed by extraction
into water, while 1 was extracted by MTBE and subsequently
analyzed by HPLC, indicating at least 98% of 1, and only trace of
found that such conditions gave only moderate conversion
(Table 1) and selectivity (Figure 3). The precise control of
temperature distribution and the efficient heating and mixing
associated with flow technique minimized the formation of
byproducts and ensured that this chemical transformation pro-
ceeded at faster rates by increasing the amount of acid compared
to that from the batch system.
’
CONCLUSION
In summary, an expeditious and high-yielding process for the
synthesis of 1 from 2 and 4-hydroxybutyraldehyde via a con-
tinuous flow reactor has been set up. The process is amenable
both for the preparation of analogous compounds and for
scaling-up by operating several reactors with high throughput
in parallel. Due to the rapid decomposition of 3, it remains a
challenging task to obtain a more satisfying yield.
7
a and 8. The yield of 1 was 73-75% corrected for purity.
For comparative purposes, the reaction was run both as a one-
pot batch process on a 1-kg scale, using the same temperature
and concentration of reagents as optimized in Process C. It was
4
73
dx.doi.org/10.1021/op1003083 |Org. Process Res. Dev. 2011, 15, 471–475

Organic Process Research & Development
TECHNOLOGY REPORT
a
Scheme 5. Schematic of Process C
a
Solution A is 0.8 M 2 hydrochloride and NaOH (1 equiv) in ethylene glycol/water (5:2) with a flow rate of 230 mL/min; solution B is 0.8 M DHF in
ethylene glycol/water (5:2) together with 0.5-1 mL hydrochloric acid stirred for 10 minutes in advance with a flow rate of 230 mL/min; solution C is
50% sulfuric acid with a flow rate of 40 mL/min.
Table 1. Comparison of results
flowing through loop III while the indole cyclization was
terminated by neutralizing the residual acid with 30% aqu-
eous NaOH. The solution was mixed with NaCl (1.6 kg) and
extracted with MTBE (3 ꢀ 21 L), and the combined organic
a
b
operate manner
conversion (%)
purity (%)
yield (%)
batch process
93
60.2
82.6
45
75
extracts were washed with water (3.5 L), aqueous NaHSO
3
continuous process
100
a
b
(5%, 3.5 L), aqueous NaHCO (saturated, 3.5 L), and water
Conversion of hydrazone was determined by HPLC at 275 nm. Yield
3
(3.5 L). After solvent evaporation in vacuo, the initially
obtained oil solidified to afford 0.99 kg of reddish solid in
of 1 was corrected for purity.
8
1
2.6% purity and 75% yield corrected for purity of 1. On a
0-kg scale, this convenient procedure afforded a 73.6% yield
of 1 at 83.1% purity. Isolation of 1 was unnecessary, for it was
used directly in the further synthesis of etodolac in the
commerical process.
1
Characterization data of compound 1: H NMR (500
MHz, CDCl ) δ/ppm: 8.05 (br s, 1H), 7.47 (d, 1H, J = 8.0
3
Hz), 7.11-7.04 (m, 3H), 3.90 (t, 2H, J = 6.5 Hz), 3.03 (t, 2H,
J = 6.0 Hz), 2.85 (q, 2H, J = 8.0, 15.5, 23.0 Hz), 1.36 (t, 3H, J =
1
3
7
2
.8 Hz). C NMR (100 MHz, CDCl ) δ/ppm: 13.9, 24.1,
3
8.9, 62.6, 112.5, 116.4, 119.6, 120.6, 122.0, 126.5, 127.0,
1
35.1.
þ
Characterization data of compound 7a: MS: (M þ H )
1
2
75. H NMR (400 MHz, CDCl ) δ/ppm: 7.40 (s, 1H), 7.34-
.32(m, 2H), 7.27-7.25(m, 2H), 5.30(s, 1H), 3.80(t, 2H, J = 5.6
3
7
Hz), 3.70 (t, 2H, J = 5.8 Hz), 2.80 (t, 2H, J = 6.9 Hz), 2.75 (t, 2H, J
=
6.5 Hz), 2.57(q, 2H, J = 7.6, 15.1, 22.6Hz,), 2.00-1.93(m, 2H),
13
1
1
1
.10 (t, 3H, J = 7.6 Hz). C NMR (100 MHz, CDCl ) δ/ppm:
4.8, 23.0, 24.4, 27.0, 31.2, 62.1, 62.9, 115.9, 126.4, 126.5, 128.5,
29.6, 130.3, 139.3, 139.8, 151.7. IR (KBr) ν/cm : 3416, 2935,
3
-1
Figure 3. Comparison of byproducts.
2868, 1684, 1500, 1456, 1377, 1045, 764. HRMS (ESI) m/z:
þ
Calcd for C H N O , (M þ H ): 274.3572. Found: 274.3581.
1
6 22 2 2
þ
’
EXPERIMENTAL SECTION
Characterization data of compound 8: MS: (M þ H )
1
2
7
7
7
3
1
1
03. H NMR (400 MHz, CDCl ) δ/ppm: 7.68 (s, 1H),
3
All chemicals were purchased from commercial sources and
.65-7.58 (m, 3H), 4.22 (t, 2H, J = 5.6 Hz), 3.51 (q, 2H, J =
were used without further purification. Solution A [2 hydro-
chloride (1 kg, 5.79 mol) and sodium hydroxide (0.23 kg, 5.79
mol) in ethylene glycol (20 L) and water (8 L) under nitrogen]
and solution B [DHF (0.41 kg, 5.79 mol) with 0.5-1 mL
hydrochloric acid in ethylene glycol (20 L) and water (8 L)
which was stirred for 10 min in advance] were respectively
pumped into the residence loop I via a T-joint by P1 and P2 at a
flow rate of 230 mL/min, after a residence time of 20 s at 115 °C.
Solution C of 50% aqueous sulfuric acid (preheated to 115 °C)
was introduced into the tandem loop II by P3 at a flow rate of
.5, 15.0, 22.5 Hz), 3.41 (t, 2H, J = 5.6 Hz), 1.46 (t, 3H, J =
1
3
.5 Hz). C NMR (100 MHz, CDCl ) δ/ppm: 15.3, 24.4,
3
8.1, 61.9, 122.4, 124.1, 126.6, 128.3, 131.4, 143.8, 148.0,
-
1
55.3. IR (KBr) ν/cm : 3420, 2964, 2929, 2876, 1654,
556, 1503, 1453, 1057, 779. HRMS (ESI) m/z: Calcd for
þ
C H N O, (M þ H ): 202.2518. Found: 202.2521.
1
2
14
2
’
ASSOCIATED CONTENT
4
0 mL/min. Reagents were maintained in loop II for about 4 min,
b
S
Supporting Information. This material is available free
and then the mixture was cooled to nearly 20 °C by cold water
of charge via the Internet at http://pubs.acs.org.
4
74
dx.doi.org/10.1021/op1003083 |Org. Process Res. Dev. 2011, 15, 471–475

Organic Process Research & Development
AUTHOR INFORMATION
TECHNOLOGY REPORT
’
Corresponding Author
Fax: (þ86)57188320752. E-mail: pharmlab@zjut.edu.cn.
’
ACKNOWLEDGMENT
We are grateful for National Key Technology R&D Program
2007BAI34B00) and National Natural Science Foundation of
China (20876147) for financial support.
(
’
REFERENCES
(1) (a) Riva, E.; Gagliardi, S.; Mazzoni, C.; Passarella, D.; Rencurosi,
A.; Vigo, D.; Martinelli, M. J. Org. Chem. 2009, 74, 3540–3543. (b)
Watts, P.; Haswell, S. J. Drug Discovery Today 2003, 8, 586–593. (c)
Petersen, T. P.; Ritz ꢀe n, A.; Ulven, T. Org. Lett. 2009, 11, 5134–5137. (d)
Riva, E.; Gagliardi, S.; Mazzoni, C.; Passarella, D.; Rencurosi, A.; Vigo,
D.; Rencurosi, A. Tetrahedron 2010, 66, 3242–3247. (e) Kulkarni, A. A.;
Kalyani, V. S.; Joshi, R. A.; Joshi, R. R. Org. Process Res. Dev. 2009,
1
3, 999–1002. (f) Wahab, B.; Ellames, G.; Passey, S.; Watts, P.
Tetrahedron 2010, 66, 3861–3865.
2) (a) Sundberg, R. J. The Chemistry of Indoles; Academic Press:
New York, 1970. (b) Sundberg, R. J. Indoles; Academic Press: London,
(
1996. (c) Humphrey, G. R.; Kuethe, J. T. Chem. Rev. 2006,
106, 2875–2911.
(
3) (a) Hughes, P.; DeVirgilio, J.; Humber, L. G.; Chau, T.; Weich-
man, B.; Neuman, G. J. Med. Chem. 1989, 32, 2134–2137. (b) Soll, R. M.;
Guinosso, C.; Asselin, A. J. Org. Chem. 1988, 53, 2844–2847. (c)
Humber, L. G.; Christopher, A.; Demerson, C. A.; Swaminathan, P.;
Birdt, P. H. J. Med. Chem. 1986, 29, 871–874.
(4) (a) Robinson, B. The Fischer Indole Synthesis; Wiley-Interscience:
New York, 1982. (b) Hughes, D. L. Org. Prep. Proced. Int. 1993,
25, 607–632. (c) Gribble, G. W. J. Chem. Soc., Perkin Trans. 1
2
000, 1045–1075. (d) Fleck, T. J.; Chen, J. J.; Lu, C. V. Org. Process
Res. Dev. 2006, 10, 334–338. (e) Slade, J.; Parker, D.; Girgis, M.; Wu, R.;
Joseph, S.; Repi ꢁc , O. Org. Process Res. Dev. 2007, 11, 721–725.
(
5) Chen, C. Y.; Senanayake, C. H.; Bill, T. J.; Larsen, R. D.;
Verhoeven, T. R.; Reider, P. J. J. Org. Chem. 1994, 59, 3738–3741.
6) Hino, T.; Hasegawa, A.; Liu, J. J.; Nakagawa, M. Chem. Pharm.
Bull. 1990, 38, 59–64.
7) (a) Singh, P. R.; Surpur, M. P.; Patil, S. B. Tetrahedron Lett. 2008,
9, 3335–3340. (b) Dhakshinamoorthy, A.; Pitchumani, K. Appl. Catal.,
A 2005, 292, 305–311.
8) (a) Rebeiro, G. L.; Khadilkar, B. M. Synthesis 2001, 3, 370–372.
b) Morales, R. C.; Tambyrajah, V.; Jenkins, P. R.; Davies, D. L.; Abbott,
A. P. Chem. Commun 2004, 158–159.
9) (a) Kevin, R. C.; Jacqueline, C. S.; Sandra, L.; Richard, D. A. Org.
(
(
4
(
(
(
Lett. 2004, 6, 79–82. (b) Brodfuehrer, P. R.; Chen, B. C.; Sattelberg,
T. R.; Smith, P. R.; Reddy, J. P.; Stark, D. R.; Quinlan, S. L.; Thottathil, J.
G. R. K.; Wang, S. P. J. Org. Chem. 1997, 62, 9192–9202. (c) Anderson,
N. G.; Ary, T. D.; Berg, J. L.; Bernot, P. J.; Chan, Y. Y.; Chen, C. K.;
Davies, M. L.; DiMarco, J. D.; Dennis, R. D.; Deshpande, R. P.; Do,
H. D.; Droghini, R.; Early, W. A.; Gougoutas, J. Z.; Grosso, J. A.; Harris,
J. C.; Haas, O. W.; Jass, P. A.; Kim, D. H.; Kodersha, G. A.; Kotnis, A. S.;
LaJeunesse, J.; Lust, D. A.; Madding, G. D.; Modi, S. P.; Moniot, J. L.;
Nguyen, A.; Palaniswamy, V.; Phillipson, D. W.; Simpson, J. H.;
Thoraval, D.; Thurston, D. A.; Tse, K.; Polomski, R. E.; Wedding,
D. L.; Winter, W. J. Org. Process Res. Dev. 1997, 1, 300–310.
(
10) Lu, Y. W.; Lu, Z. X.; Su, W. K. Gaoxiao Huaxue Gongcheng
Xuebao 2010, 24, 127–131.
11) HPLC conditions: RP-18 column, 250 mm ꢀ 4.0 mm, eluted at
.0 mL/min with 50:25:25 0.03 M NaH PO aqueous/CH CN/
CH OH, 275 nm.
(
1
2
4
3
3
4
75
dx.doi.org/10.1021/op1003083 |Org. Process Res. Dev. 2011, 15, 471–475