A multistep flow process for the synthesis of highly functionalized benzoxazoles

Article abstract of DOI:10.1021/ol402706a

An efficient and scalable transformation of 3-halo-N-acyl anilines to the corresponding benzoxazoles within a continuous flow reactor is reported. This transformation proceeds via base-mediated deprotonation, ortho-lithiation, and intramolecular cyclization to provide unstable lithiated benzoxazole moieties. The subsequent in-line electrophilic quench results in the formation of substituted benzoxazoles in high yield and quality. Continuous flow technology allowed for accurate temperature control and immediate in-line quench while minimizing the hold-up time for the unstable lithiated intermediates thereby minimizing associated byproduct formation.

Full text of DOI:10.1021/ol402706a

ORGANIC
LETTERS
2013
Vol. 15, No. 21
5546–5549
A Multistep Flow Process for the
Synthesis of Highly Functionalized
Benzoxazoles
†
Jorg Sedelmeier,* Fabio Lima, Alain Litzler, Benjamin Martin, and
€
Francesco Venturoni*
Novartis Pharma AG, Fabrikstrasse 14, 4002 Basel, Switzerland
joerg.sedelmeier@novartis.com; francesco.venturoni@novartis.com
Received September 20, 2013
ABSTRACT
An efficient and scalable transformation of 3-halo-N-acyl anilines to the corresponding benzoxazoles within a continuous flow reactor is reported.
This transformation proceeds via base-mediated deprotonation, ortho-lithiation, and intramolecular cyclization to provide unstable lithiated
benzoxazole moieties. The subsequent in-line electrophilic quench results in the formation of substituted benzoxazoles in high yield and quality.
Continuous flow technology allowed for accurate temperature control and immediate in-line quench while minimizing the hold-up time for the
unstable lithiated intermediates thereby minimizing associated byproduct formation.
The benzoxazole moiety and the related R-hydroxy
anilines represent an important pharmacophore in many
biologically active compounds. Substituted benzoxazoles
have been shown to exhibit antitumor, hypoglycaemic,
antiallergic, antihistaminic, antiparasitic, herbicidal, anti-
tubercular, antihelmintic, COX-2 inhibitory, antifungal,
antibacterial, anticancer, anticonvulsant, HIV reverse trans-
criptase inhibitor, and insecticidal activities, among others.1
As such, a variety of synthetic methods for the formation
of benzoxazoles have been developed,2 and we adopted
such an approach2f in the synthesis of benzoxazole 10a
(Scheme 2), a key intermediate of a recent research investi-
gation. We focused our attention toward continuous
flow techniques since major drawbacks in traditional batch
chemistry had already been observed, including quality-
critical deviations such as formation of dimeric benzoxa-
zoles, as well as complete loss of batches on multikilogram
scale due to the lack of robustness. The continuous manu-
facturing system used for our investigation is of in-house
design and integrates precooling loops for all reagent
streams, two tubular reactors, three Syrdos2 continuous
syringe pumps,3 pressure sensors for each reagent stream,
and a Coriolis mass flow controller unit4 to monitor the
overall flow rate (Figure 1). The entire setup is controlled
by HiTecZang software5 allowing for online monitoring
of flow rates, mass flow, temperatures, and system pres-
sures. The integrated software control features enable
automatic safety shut-down or emergency actions based
upon user predefined parameters thereby increasing the
overall safety of the continuous flow process. To effi-
ciently remove the reaction heat generated in the mixing
zones, narrow-bore metal T-pieces (Swagelok)6 have
been used. Perfluoroalkoxy (PFA) tubing with an inner
diameter (id) = 0.78 mm served as residence volumes and
have been precisely cooled to the desired temperatures
using cryostat baths.
Although outside of the scope of this manuscript, the
benzoxazoles can subsequently be hydrolyzed to the
corresponding ortho-hydroxy anilines (Figure 1).7
The initial benzoxazole synthesis was optimized using
the pivaloyl protected aniline 2a8 as a substrate, with the
† Current address: Department of Chemistry, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139, United States.
r
10.1021/ol402706a
Published on Web 10/22/2013
2013 American Chemical Society

systematically investigated. On combination with 2 equiv
of n-butyllithium the protected aniline 2a underwent base-
mediated elimination of the 3-chloro substituent to afford
the aryne intermediate 4a, followed by intramolecular
cyclization. Trapping of moiety 5a with a THF solution
of acetic acid as the model electrophile provided the
benzoxazole 6a (Scheme 1).9
The optimal preparation of 6a was achieved using a
0.25 M concentration of aniline derivative 2a (1.0 equiv) in
tetrahydrofuran, a 1.25 M solution of n-butyllithium (2.4
equiv) in hexane, and a 0.75 M solution of the correspond-
ing electrophile (3.0 equiv) in tetrahydrofuran. The lithia-
tion chemistry occurred at À10 °C with a residence time of
300 s, whereas the electrophilic quench required À5 °C and
a 60 s reaction time. Maintaining a temperature as close to
À10°C aspossibleforthe lithiation chemistry wasfound to
be a critical process parameter, since any warmer tempera-
tures led to the formation of undesired byproducts10 over
a prolonged reaction time. Calorimetric investigations
showed an instantaneous reaction and the exothermic
nature (510 kJ/mol) of the base-mediated transformation;
it resulted in an adiabatic temperature rise of about 55 °C
(considering a 0.17 M solution) in the first mixing zone.11
To efficiently remove the reaction heat generated, an
appropriate mixing device had to be identified. In order
to avoid hot spots it was decided that a nonoptimal mixing
regime would serve to dissipate the reaction energy over
the length of the tubular reactor, avoiding thermal accu-
mulation in a single mixing point (adiabatic conditions).
This reasoning led us to choose a metal Swagelok T-piece
as a mixing device.12 To further facilitate and control the
heat dissipation all reagent streams were prechilled before
mixing. Adhering to the above protocol ensured a high-
quality product 6a that required only a simple extractive
workup followed by evaporation of the solvent.
Figure 1. General reactor setup.
solvent, residence time, mixing efficiency, reaction tem-
perature, reagent concentration, and stoichiometries being
(1) (a) Lokwani, P.; Nagori, B. P.; Batra, N.; Goyal, A.; Gupta, S.;
Singh, N. J. Chem. Pharm. Res. 2011, 3, 302–311. (b) Hausner, S. H.;
Alagille, D.; Koren, A. O.; Amici, L.; Staley, J. K.; Cosgrove, K. P.;
Baldwin, R. M.; Tamagnan, G. D. Bioorg. Med. Chem. Lett. 2009, 19,
543–545. (c) Smith, C. J.; Ali, A.; Chen, L.; Hammond, M. L.; Anderson;
Chen, Y.; Eveland, S. S.; Guo, Q.; Hyland, S. A.; Milot, D. P.; Sparrow,
C. P.; Wright, S. D.; Sinclair, P. J. Bioorg. Med. Chem. Lett. 2010, 20,
346–349. (d) Yang, Z.; Fairfax, D. J.; Maeng, J.-H.; Masih, L.;
Usyatinsky, A.; Hassler, C.; Isaacson, S.; Fitzpatrick, K.; DeOrazio,
R. J.; Chen, J.; Harding, J. P; Isherwood, M.; Dobritsa, S.; Christensen,
K. L.; Wierschke, J. D.; Bliss, B. I.; Peterson, L. H.; Beer, C. M.; Cioffi,
C.; Lynch, M.; Rennells, W. M.; Richards, J. J; Rust, T.; Khmelnitsky,
Y. L.; Cohen, M. L.; Manning, D. D. Bioorg. Med. Chem. Lett. 2010, 20,
6538–6541. (e) Jin, C.; Fix, S. E.; Kepler, J. A.; Cook, C. E. Bioorg. Med.
Chem. Lett. 2012, 22, 1705–1708. (f) Wang, X.; Cui, M.; Yu, P.; Li, Z.;
Yang, Y.; Jia, H.; Liu, B. Bioorg. Med. Chem. Lett. 2012, 22, 4327–4331.
(g) Sessions, E. H.; Yin, Y.; Bannister, T. D.; Weiser, A.; Griffin, E.;
Pocas, J.; Cameron, M. D.; Ruiz, C.; Lin, L.; Schurer, S. C.; Schroter, T.;
LoGrasso, P.; Feng, Y. Bioorg. Med. Chem. Lett. 2008, 18, 6390–6393.
(h) Ouyang, L.; Huang, Y.; Zhao, Y.; He, G.; Xie, Y.; Liu, J.; He, J.; Liu,
B.; Wei, Y. Bioorg. Med. Chem. Lett. 2012, 22, 3044–3049. (i) Deng, G.;
Meng, Q.; Liu, Q.; Xu, X.; Xu, Q.; Ren, F.; Guo, T. B.; Lu, H.; Xiang,
J. N.; Elliott, J. D.; Lin, X. Bioorg. Med. Chem. Lett. 2012, 22, 3973–
3977. (j) Keely, N. O. M.; Meegan, J. Curr. Trends Med. Chem. 2009, 6,
1–17. (k) Sweis, R. F.; Hunt, J. A.; Kallashi, F.; Hammond, M. L.; Chen,
Y. Bioorg. Med. Chem. Lett. 2011, 21, 1890–1895. (l) Hou, J.; Li, Z.;
Fang, Q.; Feng, C.; Zhang, H.; Guo, W.; Wang, H.; Gu, G.; Tian, Y.;
Liu, P.; Liu, R.; Lin, J.; Shi, Y. K.; Yin, Z.; Shen, J.; Wang, P. G. J. Med.
Chem. 2012, 55, 3066–3075. (m) Certal, V.; Halley, F.; Virone-Oddos,
A.; Delorme, C.; Karlsson, A.; Rak, A.; Thompson, F.; Filoche-
In order to demonstrate the substrate scope and syn-
thetic value of the continuous flow process, conditions
for the direct conversion of 2a to various benzoxazoles
€
€
(3) HiTec Zang SyrDos Precision Syringe Doser. http://www.hitec-
zang.de/en/laboratory-devices/syringe-doser.html (accessed Oct 30,
2013).
(4) Mass flow controller Bronkhorst Schweiz. http://www.bron-
khorst.ch/en/ (accessed Oct 30, 2013).
(5) HiTec Zang Automation and Evaluation Software. http://www.
hitec-zang.de/en/laboratory-automation/software.html (accessed Oct
30, 2013).
(6) Swagelok. http://www.swagelok.com/ (accessed Oct 30, 2013).
(7) (a) Ono, M.; Yamakawa, K.; Kobayashi, H.; Itoh, I. Heterocycles
1988, 27, 881–884. (b) Bruyneel, F.; Enaud, E.; Billottet, L.; Vanhulle, S.;
Marchand-Brynaert, J. Eur. J. Org. Chem. 2008, 72–79. (c) Lagadic, E.;
Garcia, Y.; Marchand-Brynaert, J. Synthesis 2012, 44, 93–98.
(8) Busch-Petersen, J.; Leonard, T. B.; Palovich, M. R.; Sarau,
H. M.; Lazaar, A. L.; Widdowson, K. L. US 20100256157, 2010.
(9) (a) Reavill, D. R.; Richardson, S. K. Synth. Commun. 1990, 20,
1423–1436. (b) Clark, R. D.; Caroon, J. M. J. Org. Chem. 1982, 47,
2804–2806.
ꢀ
Romme, B.; El-Ahmad, Y.; Carry, J. C.; Abecassis, P. Y.; Lejeune, P.;
Vincent, L.; Bonnevaux, H.; Nicolas, J. P.; Bertrand, T.; Marquette,
J. P.; Michot, N.; Benard, T.; Below, P.; Vade, I.; Chatreaux, F.;
Lebourg, G.; Pilorge, F.; Angouillant-Boniface, O.; Louboutin, A.;
Lengauer, C.; Schio, L. J. Med. Chem. 2012, 55, 4788–4805. (n) Expert
Opin. Ther. Pat. 2003, 13, 717À720.
(2) (a) Thomas, B.; George, J.; Sugunan, S. Ind. Eng. Chem. Res.
2009, 48, 660–670. (b) Ren, P.; Salihu, I.; Scopelliti, R.; Hu, X. Org. Lett.
2012, 14, 1748–1751. (c) Wu, M.; Hu, X.; Liu, J.; Liao, Y.; Deng, G.-J.
ꢀ
Org. Lett. 2012, 14, 2722–2725. (d) Bastug, G.; Eviolitte, Ch.; Marko,
(10) As major byproducts the corresponding dimers (for structure see
Supporting Information, Figure 1) and the protodehalogenated benzox-
azole (when using an electrophile other than acid) have been observed
during the batch optimization process.
(11) Experiment was performed in a METTLER RC1 reaction
calorimeter. Software: WinRC NT V.7. Glass reactor 0.9 L equipped
with 4-bladed glass propeller stirrer, glass cover, and glass inserts.
I. E. Org. Lett. 2012, 14, 3502–3505. (e) Cui, W.; Kargbo, R. B.; Sajjadi-
Hashemi, Z.; Ahmed, F.; Gauuan, J. F. Synlett 2012, 23, 247–250. (f)
Yoshida, Y.; Hamada, Y.; Umezu, K.; Tabuchi, F. Org. Process Res.
Dev. 2004, 8, 958–961. (g) Bonnamour, J.; Bolm, C. Org. Lett. 2008, 10,
2665–2667. (h) Reavill, D. A.; Richardson, S. K. Synth. Commun. 1990,
20, 1423–1436. (i) Kalkhambkar, R. G.; Laali, K. K. Tetrahedron Lett.
2012, 53, 4212–4215. (j) Turker, L.; Sener, E.; Yalcin, I.; Akbulut, U.;
€
€
(12) Schwolow, S.; Hollmann, J.; Schenkel, B.; Roder, T. Org.
Process Res. Dev. 2012, 16, 1513–1522.
Kayalidere, I. Sci. Pharm. 1990, 58, 107–113. (k) Schnurch, M.;
€
Hammerle, J.; Stanetty, P. Sci. Synth. 2010, 11.13, 1–56.
Org. Lett., Vol. 15, No. 21, 2013
5547

Scheme 1. Model Reaction and Reaction Mechanism
Scheme 3. Expansion of Benzoxazole Diversitya
Scheme 2. Transformation of Aniline 2a to Benzoxazole
Derivatives Using Various Electrophilesa
a 2bÀd in dry THF (0.25 M, 0.750 mL/min), n-butyllithium in hexane
(1.25 M, 0.360 mL/min), electrophile in THF (0.75 M, 0.750 mL/min).
b Residence time in the first reactor = 500 s, and residence time in the
second reactor = 60 s. c Residence time in the first reactor = 300 s, and
residence time in the second reactor = 60 s.
chemical behavior. However, in contrast, we encountered
a significantly slower and less exothermic reaction for the
monochloroaniline derivatives (2b, 2c). A plausible expla-
nation is that the proton in the 2-position is less acidic than
those in 2a and 2d; therefore, the reactor volume (residence
time) was adjusted accordingly, keeping the overall flow
rates (mixing efficiency) unchanged.
With a protocol available to prepare a broad variety
of different benzoxazoles (6À14), we turned our attention
toward scale-up of the continuous manufacturing process.
We considered this to be an important proof-of-concept
effort due to the exothermic nature of the two consecutive
reaction steps, the potential clogging of the reactor
system with lithium salts, and the generation of un-
desired byproduct formation in case of insufficient
temperature control. Our aim was to demonstrate that
the flow setup could be operated under steady state
conditions for several hours, thereby generating multi-
gram quantities of benzoxazole product. The integrated
control and safety features were managed by the soft-
ware, which constantly monitors the system temperature,
flow rates, and pressures, which encouraged us to directly
scale the two-step sequence to a 100 g scale.
a 2a in dry THF (0.25 M, 0.750 mL/min), n-butyllithium in hexane
(1.25 M, 0.360 mL/min), electrophile in THF (0.75 M, 0.750 mL/min);
residence time in first reactor = 300 s, and residence time in second
reactor = 60 s. b Yield based on 1H NMR.
by quenching with different electrophiles were devised
(Scheme 2).
Following the described flow protocol, functionalities
such as iodine, alkyl, esters, aldehydes, amides, boronic
esters,13 and alcohols could be introduced into the 7-posi-
tion of the benzoxazole moiety14 enabling further
modifications.
In an effort to expand on the product diversity of these
flow reactions, additional modifications to the aniline
moiety were investigated (Scheme 3). Regarding the
substitution pattern on the aromatic ring, the electron-
deficient 3,4-dichloroaniline 2a (Scheme 2) and the
3-chloro-4-fluoroaniline moiety 2d showed comparable
With this aim in mind, we maximized the throughput
of the transformation by adapting the concentration of
the feedstock solutions, flow rates, and the reactor volumes
(Scheme 4), without observation of clogging or quality
issues related to hot spots.
Under the adjusted reaction conditions a 0.5 M solution
of 2a (100 g, 406 mmol, 4 mL/min) was reacted with a
2.5 M solution of n-butyllithium (1.92 mL/min) at a total
flow rate of 5.92 mL/min using a reactor volume of 30 mL.
The reaction stream was quenched in-line with a 1.5 M
solution of phenyl isocyanate in THF (4 mL/min) prior
to a 10 mL residence coil. The final reaction mixture
(13) Boronic ester 12a was revealed to be instable at ambient condi-
tions and would require immediate consumption. The concept of
production and immediate usage of boronic intermediates has been
demonstrated recently by Buchwald, S., Prof.; Shu, W.; Pellegatti, L.;
Oberli, M. A.; Buchwald, S. Angew. Chem., Int. Ed. 2011, 50, 10665–
10669.
(14) For an overview of electrophiles being used, see Table 1 in the
Supporting Information.
5548
Org. Lett., Vol. 15, No. 21, 2013

and isolation of the desired product occurred in comparable
yield and purity as for the small-scale flow reaction.
In summary, we have developed an efficient, rapid, and
reproducible process for the formation of benzoxazoles
under continuous flow conditions. We demonstrated the
reliable and robust scale-up by running the continuous
manufacturing process in steady state mode processing
400 mmol of starting material. The value of this process lies
in the use of readily available and inexpensive aniline
precursors delivering highly functionalized benzoxazole
products in good yields and purities.
Scheme 4. Scale-up of Key Intermediate 10aa
Acknowledgment. We thank Dr. Flavien Susanne, Dr.
Jutta Hollmann, and Dr. Berthold Schenkel (Novartis
Pharma, Basel) for chemical engineering, Dr. Isabelle
Sylvie Gallou (Novartis Pharma, Basel) for useful discus-
sions and support, and Stephane Schmitt (Novartis Pharma,
Basel) for analytical support.
a 2a in dry THF (0.5 M, 4.0 mL/min), n-butyllithium in hexane
(2.5 M, 1.92 mL/min), PhNCO in THF (1.5 M, 4.0 mL/min); residence
time in the first reactor = 300 s, and residence time in the second
reactor = 60 s. Throughput of product (space-time-yield) has been
normalized to production time [1 h] and reactor volume [1 L].
b
was collected into saturated aqueous ammonium chloride
solution over a period of 200 min. After extraction, phase
separation, removal of solvent, and crystallization from
heptane/EtOAc, the corresponding benzoxazole 10a was
isolated in 86% yield (114 g, 348 mmol). A throughput of
0.85 kg (10a) per hour per liter of reactor volume was thus
achieved. During the extended processing period, no opera-
tional issues were encountered due to fouling of the reactor,
Supporting Information Available. Experimental pro-
cedures and full characterization (1H and 13C NMR data
and spectra, UV and HRMS) for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 21, 2013
5549