Photochemical arylation of Bronsted acids with 2-azidobenzimidazole

Article abstract of DOI:10.1002/ejoc.201101711

Irradiation of N-benzylated 2-azidobenzimidazoles at 300 nm in the presence of an excess amount of carboxylic acids results in a novel regioselective synthesis of 2-amino-6-oxybenzimidazoles in isolated yields of 60-70 %. It is also possible to regioselectively introduce 6-bromo, 6-chloro, and 6-triflyloxy groups. Irradiation in dichloromethane in the absence of external nucleophiles revealed that after loss of nitrogen, the benzimidazolylnitrene probably undergoes coarctate ring opening to an N-cyano diaza-o-xylylene intermediate. Photolysis of 2-azidobenzimidazoles in the presence of carboxylic or sulfonic acidsleads to regioselective oxygenation of the 6-position in good yield. Copyright

Full text of DOI:10.1002/ejoc.201101711

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201101711
Photochemical Arylation of Brønsted Acids with 2-Azidobenzimidazole
Alex Sudakow,[a] Peter G. Jones,[b] and Thomas Lindel*[a]
Keywords: Azides / Nitrogen heterocycles / Oxygenation / Photolysis / Nitrenes
Irradiation of N-benzylated 2-azidobenzimidazoles at 300 nm
triflyloxy groups. Irradiation in dichloromethane in the ab-
in the presence of an excess amount of carboxylic acids re- sence of external nucleophiles revealed that after loss of ni-
sults in a novel regioselective synthesis of 2-amino-6-oxy-
benzimidazoles in isolated yields of 60–70%. It is also pos-
sible to regioselectively introduce 6-bromo, 6-chloro, and 6-
trogen, the benzimidazolylnitrene probably undergoes co-
arctate ring opening to an N-cyano diaza-o-xylylene inter-
mediate.
of 2-lithiated imidazole derivatives, as applied to the synthe-
sis of several pyrrole-imidazole alkaloids.[6] As initial candi-
dates, we irradiated 2-azido-1-methylimidazole (1) and 2-
azido-1-benzylimidazole (2) in EtOH or water in a Rayonet
photoreactor (RPR 3000/3500 lamps with emission maxima
at 300 and 350 nm, respectively, quartz reaction vessel).
However, after evolution of nitrogen, intractable product
mixtures were obtained.
Introduction
Photochemical conversions are of general interest for
chemical biology, because they are orthogonal to most bio-
chemical processes. In this communication we report on a
new possibility to arylate carboxylic and other Brønsted ac-
ids under photochemical conditions starting from 2-azido-
benzimidazole. The arylation is regioselective and proceeds
in good yields.
1
The H NMR spectra of the crude product mixtures ob-
It is known that five-membered heterocycles with an
azido group in the ortho-position to one of the heteroatoms
undergo ring opening under thermal or photochemical con-
ditions.[1] The reaction commences with loss of dinitrogen
and formation of the nitrene, followed by cleavage of the
bond between the ring heteroatom and the nitrene-substi-
tuted atom. In a seminal publication, such opening reac-
tions have been classified by Herges as proceeding via co-
arctate transition states.[2] The resulting formal ene-ene-
nitriles have cyclized further to interesting heterocycles, for
example, in the case of the high-yielding conversion of N-
amino-2-azido-1,3,4-triazoles into 3-amino-1,2,4,5-tetraz-
ines,[3] the reaction of 1-amino-2-azidoimidazoles into a
mixture of 1,2,3-triazoles and 3-amino-1,2,4-triazines,[4] or
the irradiation of 3-azidopyrazole-4-carbaldehydes into a
mixture of 4-cyanopyrazoles and 4-cyano-3-azofurans.[5]
tained from 1 indicated loss of the olefinic imidazole CH
signals. This prompted us to protect the olefinic double
bond of imidazole by annulation of a benzene ring and to
investigate the corresponding benzimidazoles. When 2-
azido-1-methylbenzimidazole was irradiated in MeOH or
DCM (λmax = 300 nm) the orange-colored 1,2-bis(1-methyl-
benzimidazol-2-yl)diazene was isolated in very low yield
(3%). The picture did not change in 10-fold higher dilution
(0.015 m). A little more promising was the behavior of 2-
azido-1-benzylbenzimidazole (3) upon irradiation (λmax
=
Results and Discussion
We decided to investigate 2-azidoimidazoles, because
they can be obtained in a facile manner through azidation
[a] Institute of Organic Chemistry, TU Braunschweig
Hagenring 30, 38106 Braunschweig, Germany
Fax: +49-531-391-7744
E-mail: th.lindel@tu-bs.de
[b] Institute of Inorganic and Analytical Chemistry,
TU Braunschweig
Hagenring 30, 38106 Braunschweig, Germany
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201101711.
Scheme 1. 2-Azidoimidazoles employed in initial irradiation experi-
ments.
Eur. J. Org. Chem. 2012, 681–684
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
681

A. Sudakow, P. G. Jones, T. Lindel
SHORT COMMUNICATION
350 nm) in DCM (0.01 m, Scheme 1). The major product
(13% isolated yield) was 2-phenylbenzimidazol-1-carbo-
nitrile (4), as proven by X-ray analysis.
To enable facile judgment by 19F NMR spectroscopy be-
fore workup, the benzyl group was replaced by a p-fluoro-
benzyl group. The synthesis of fluorinated azidobenzimid-
azole 6 required N-benzylation of benzimidazole (5), fol-
lowed by 2-azidation after lithiation and quenching with
tosylazide (53%).[7] We optimized the azidation by chang-
ing the solvent from THF to Et2O and by workup em-
ploying Na4P2O7 (Scheme 2).[8] Azide 6 can be stored in
the dark at room temperature for several months without
decomposition.
Scheme 3. 6-Hydroxylation of 2-azidobenzimidazole with simulta-
neous protection of the terminal amino group. 1H–15N HMBC cor-
relations are shown.
Sulfonic acids were also suitable. The synthesis of 2-
amino-6-sulfonyloxybenzimidazole 11 was possible in Me-
SO3H/DCM (1:8) in high yield (78%, Scheme 4). Of par-
ticular interest with respect to subsequent C–C coupling re-
actions will be the incorporation of CF3SO3H, which pro-
vided aryl triflate 12 (65%). It was also possible to obtain
6-chlorinated and 6-brominated benzimidazoles 13 and 14,
respectively, in acceptable yields (50–60%) by employing the
corresponding mineral acids (Scheme 4). Chlorinated com-
pound 13 was also formed in the presence of NH4Cl, which
shows that ammonia is not incorporated.
Scheme 2. Regioselective 6-oxygenation of fluorinated 2-azidobenz-
imidazole 6 by irradiation in the presence of carboxylic acids.
Irradiations (λmax = 300 nm) of 6 were carried out by
employing potential reaction partners (e.g., acetic acid,
benzaldehyde, furan, diethylamine) as cosolvents in DCM.
We replaced the quartz by a borosilicate vessel, which cut
the short-wavelength section of the emission spectrum off
below 295 nm and diminished the number of byproducts in
some cases. The 19F NMR spectra of the crude product
mixtures revealed that very selective reactions occurred with
carboxylic acids (Scheme 2), whereas less acidic reaction
partners led to complex mixtures. Irradiation of azide 6 in
AcOH/DCM (1:8) gave product 8 (confirmed by X-ray
analysis) in 60% yield and in pure AcOH the yield was Scheme 4. Regioselective sulfonylations, halogenations, and meth-
oxylations of 2-azidobenzimidazole 6.
92%. In the case of trifluoroacetic acid, transfer of the tri-
fluoroacetyl group from the oxygen to the amino nitrogen
took place during aqueous workup, resulting in overall 6-
Differing from carboxylic and sulfonic acids, alcohols
hydroxylation (Scheme 3). In a quartz tube, the yield of were not incorporated into the 6-position when used as the
63% (for 10) was diminished to 28%. The structure of prod- only cosolvent of DCM. However, 6-methoxylation was
1
uct 10 was secured by H–15N HSQC and HMBC experi- achieved by applying a mixture of MeOH and phosphoric
ments with the decisive correlation observed between 5-H acid (Scheme 4), indicating the participation of protonated
(δ = 6.77 ppm) and protonated N-3 (δ = –247.6 ppm, [D6]- intermediates. Interestingly, the benzylic methylene group
DMSO).
682
was methoxylated in 52% yield on irradiation of 6 in Na-
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 681–684

Photochemical Arylation of Brønsted Acids with 2-Azidobenzimidazole
OMe/MeOH. This makes it likely that the benzylic carbon
In the presence of methoxide, imine 20 is intercepted by
becomes part of an imine moiety under deprotonating con- attack at the most electropositive imine carbon, followed by
ditions.
Scheme 5 outlines possible reaction mechanisms. On irra-
cyclization to methoxylated product 16.
diation, azide 6 loses dinitrogen to afford the corresponding
nitrene. In the presence of Brønsted acids, the protonated
form of that nitrene predominates and cleavage of N1–C2
via a coarctate transition state affords protonated N-cyano-
diazaxylylene 17. McClelland and co-workers proposed a
similar ring opening for the nitrene derived from 2-azido-1-
methylimidazole in aqueous solution.[9] Pozharskii and co-
workers have shown that oxidation of 1,2-diaminobenz-
imidazole leads to the corresponding coarctate ring opening
after oxidation, affording 3-amino-1,2,4-triazines through
re-cyclization.[10] Proton shift at 17 to the more basic nitro-
gen affords intermediate 18. Nucleophilic attack occurs at
C6, which is the β-position of an α,β-unsaturated iminium
moiety. Ring closure to the benzimidazole would be fol-
lowed by proton transfer from C6 to the exocyclic nitrogen
to afford, for instance, aryl triflate 12.
Conclusions
In summary, we have discovered a new efficient method
for the synthesis of 6-oxygenated and -halogenated amino-
benzimidazoles, which may become of preparative use in
medicinal chemistry. Currently, 2-aminobenzimidazoles are
assembled from suitably substituted o-phenylendiamines
with the amino group introduced by either nucleophilic at-
tack at 2-chlorobenzimidazole[12] or condensation with urea
derivatives.[13] An important application of our reaction
may become the use of 2-azidobenzimidazoles for photola-
beling of carboxylic acids, as they occur in proteins as side
chains of aspartic and glutamic acid. We are currently in-
vestigating that option.
Experimental Section
General Procedure for the Irradiation of 6 in the Presence of Carb-
oxylic or Sulfonic Acids: A solution of 6 (50.0 mg, 0.187 mmol) and
a carboxylic acid or sulfonic acid (2 mL) in DCM (16 mL) was
irradiated for 2 h at λmax = 300 nm (borosilicate glass, Rayonet
RPR-200 reactor, equipped with eight RPR-3000 Å lamps, 35 °C).
Saturated aqueous NaHCO3 was added, until pH 8–9 was reached.
The organic phase was separated, and the aqueous phase was ex-
tracted with EtOAc (3ϫ). The combined organic phases were dried
with MgSO4, filtered, and concentrated. The crude residue was
purified by column chromatography on silica. For further experi-
mental procedures and data, see the Supporting Information.
CCDC-858062 (for 4) and -858063 (for 8) contain the supplemen-
tary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures, spectroscopic and analytic data, and
copies of the NMR spectra for all new compounds.
Acknowledgments
Scheme 5. Possible reaction mechanisms under protonating and
non-protonating conditions via coarctate ring opening (Ar = p-
fluorophenyl).
Financial support of this research by the Deutsche Forschungsge-
meinschaft (DFG, Li 597/5–1) is gratefully acknowledged. We also
thank Merck KGaA (Darmstadt, Germany) for the generous gift
of chromatography materials. BASF Group (Ludwigshafen, Ger-
many) and Honeywell Specialty Chemicals Seelze GmbH (Seelze,
Germany) are thanked for the donation of solvents.
Ring cleavage of an intermediate nitrene is also in agree-
ment with the products obtained in the absence of Brønsted
acids. Although isolated in a yield of only 13%, structure 4
proves that cleavage of the imidazole section can occur. In
the absence of a nucleophile, intramolecular attack by the
cyanamide nitrogen becomes the main pathway (Scheme 5).
After [1,5] hydrogen shift to aromatic intermediate 20, cycli-
zation would afford imidazoline 21, which has to undergo
oxidation to product 4. Compound 4 has been obtained by
Wentrup and co-workers by a different pathway as the main
product starting from 5-phenyltetrazolo[1,5-c]quinazoline
or 4-phenyltetrazolo[1,5-a]quinoxaline (59 or 88%).[11]
[1] For a recent review, see L. D. Shirtcliff, S. P. McClintock,
M. M. Haley, Chem. Soc. Rev. 2008, 37, 343–364.
[2] R. Herges, Angew. Chem. 1994, 106, 261–281; Angew. Chem.
Int. Ed. Engl. 1994, 33, 255–276.
[3] H. H. Takimoto, G. C. Denault, Tetrahedron Lett. 1966, 7,
5369–5373.
[4] a) M. Nakajima, R. Hisada, J. P. Anselme, J. Org. Chem. 1978,
43, 2693–2696; b) R. Hisada, M. Nakajima, J. P. Anselme, Tet-
rahedron Lett. 1976, 17, 903–904.
[5] N. Svenstrup, K. B. Simonsen, N. Thorup, J. Brodersen, W.
Dehaen, J. Becher, J. Org. Chem. 1999, 64, 2814–2820.
Eur. J. Org. Chem. 2012, 681–684
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
683

A. Sudakow, P. G. Jones, T. Lindel
SHORT COMMUNICATION
[6] For the use of 2-azidoimidazoles in the synthesis of pyrrole-
imidazole alkaloids, see: a) T. Lindel, M. Hochgürtel, Tetrahe-
dron Lett. 1998, 39, 2541–2544; b) T. Lindel, M. Hochgürtel,
J. Org. Chem. 2000, 65, 2806–2809; c) I. Kawasaki, N. Sakagu-
chi, N. Fukushima, N. Fujioka, F. Nikaido, M. Yamashita, S.
Ohta, Tetrahedron Lett. 2002, 43, 4377–4380; d) G. Breckle, K.
Polborn, T. Lindel, Z. Naturforsch. B 2003, 58, 451–456; e)
M. L. Meketa, S. M. Weinreb, Org. Lett. 2006, 8, 1443–1446;
f) I. Kawasaki, N. Sakaguchi, A. Khadeer, M. Yamashita, S.
Ohta, Tetrahedron 2006, 62, 10182–10192; g) M. L. Meketa,
S. M. Weinreb, Org. Lett. 2007, 9, 853–855; h) M. L. Meketa,
S. M. Weinreb, N. Yoichi, N. Fusetani, J. Org. Chem. 2007, 72,
4892–4899; i) Y.-G. Wang, B. I. Morinaka, J. C. P. Reyes, J. J.
Wolff, D. Romo, T. F. Molinski, J. Nat. Prod. 2010, 73, 428–
434; j) N. Jacobi, T. Lindel, Eur. J. Org. Chem. 2010, 5415–
5425.
[7] C. J. Lovely, H. Du, R. Sivappa, M. R. Bhandari, Y. He, R.
Dias, J. Org. Chem. 2007, 72, 3741–3749.
[8] P. Zanirato, S. Cerini, Org. Biomol. Chem. 2005, 3, 1508–1513.
[9] T. A. Gadosy, R. A. McClelland, J. Am. Chem. Soc. 1999, 121,
1459–1465.
[10] A. F. Pozharskii, I. M. Nanavyan, V. V. KuzЈmenko, Mendeleev
Commun. 1992, 2, 33–35.
[11] C. Wentrup, C. Thetaz, E. Tagliaferri, H. J. Lindner, B.
Kitschke, H.-W. Winter, H. P. Reisenauer, Angew. Chem. 1980,
92, 556; Angew. Chem. Int. Ed. Engl. 1980, 19, 566–567.
[12]
For recent examples, see a) Y. Ogino, N. Ohtake, Y. Nagae, K.
Matsuda, M. Moriya, T. Suga, M. Ishikawa, M. Kanesaka, Y.
Mitobe, J. Ito, T. Kanno, A. Ishihara, H. Iwaasa, T. Ohe, A.
Kanatani, T. Fukami, Bioorg. Med. Chem. Lett. 2008, 18,
5010–5014; b) D. A. Pizzi, C. P. Leslie, A. Mazzali, C. Seri, M.
Biagetti, J. Bentley, T. Genski, R. Di Fabio, S. Contini, F. M.
Sabbatini, L. Zonzini, L. Caberlotto, Bioorg. Med. Chem. Lett.
2010, 20, 7120–7123; c) M. D. Rosen, H. Venkatesan, H. M.
Peltier, S. D. Bembenek, K. C. Kanelakis, L. X. Zhao, B. E. Le-
onard, F. M. Hocutt, X. Wu, H. L. Palomino, T. I. Brond-
stetter, P. V. Haugh, L. Cagnon, W. Yan, L. A. Liotta, A.
Young, T. Mirzadegan, N. P. Shankley, T. D. Barrett, M. H.
Rabinowitz, ACS Med. Chem. Lett. 2010, 1, 526–529.
[13] For recent examples, see a) S. Ramurthy, S. Subramanian, M.
Aikawa, P. Amiri, A. Costales, J. Dove, S. Fong, J. M. Jansen,
B. Levine, S. Ma, C. M. McBride, J. Michaelian, T. Pick, D. J.
Poon, S. Girish, C. M. Shafer, D. Stuart, L. Sung, P. A.
Renhowe, J. Med. Chem. 2008, 51, 7049–7052; b) F. Palomares-
Alonso, H. Jung-Cook, J. Pérez-Villanueva, J. C. Piliado, S.
Rodríguez-Morales, G. Palencia-Hernández, N. López-Bal-
biaux, A. Hernández-Campos, R. Castillo, F. Hernández-Luis,
Eur. J. Med. Chem. 2009, 44, 1794–1800.
Received: November 29, 2011
Published Online: December 28, 2011
684
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 681–684