Solvent/oxidant-switchable synthesis of multisubstituted quinazolines and benzimidazoles via metal-free selective oxidative annulation of arylamidines

Article abstract of DOI:10.1021/ol500864r

A fast and simple divergent synthesis of multisubstituted quinazolines and benzimidazoles was developed from readily available amidines, via iodine(III)-promoted oxidative C(sp³)-C(sp²) and C(sp ²)-N bond formation in nonpolar and polar solvents, respectively. Further selective synthesis of quinazolines in polar solvent was realized by TEMPO-catalyzed sp³C-H/sp²C-H direct coupling of the amidine with K₂S₂O₈ as the oxidant. No metal, base, or other additives were needed.

Full text of DOI:10.1021/ol500864r

Letter
pubs.acs.org/OrgLett
Solvent/Oxidant-Switchable Synthesis of Multisubstituted
Quinazolines and Benzimidazoles via Metal-Free Selective Oxidative
Annulation of Arylamidines
Jian-Ping Lin,† Feng-Hua Zhang,† and Ya-Qiu Long*
CAS Key Laboratory of Receptor Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555 Zuchongzhi
Road, Shanghai 201203, China
S
* Supporting Information
ABSTRACT: A fast and simple divergent synthesis of
multisubstituted quinazolines and benzimidazoles was devel-
oped from readily available amidines, via iodine(III)-promoted
oxidative C(sp3)−C(sp2) and C(sp2)−N bond formation in
nonpolar and polar solvents, respectively. Further selective
synthesis of quinazolines in polar solvent was realized by
TEMPO-catalyzed sp3C−H/sp2C−H direct coupling of the
amidine with K2S2O8 as the oxidant. No metal, base, or other additives were needed.
Scheme 1. Approaches To Access Quinazolines and
Benzimidazoles via Metal-Free Oxidative C−H Activation
uinazoline and benzimidazole are among the most
Q common motifs present in drugs and bioactive
compounds with a broad spectrum of pharmcological
activities.1 Typically, quinazolines constitute an important
category of marketed targeted cancer therapies (e.g., Iressa,
Tarceva and Tykerb). Although there are various well-
established methods to construct the two privileged structures,
most of them require special starting materials, multistep
procedures, and harsh reaction conditions,2 as well as the use of
transition metals,3 impeding the biological studies and
therapeutic applications of these drug-like scaffolds.
In the past decade, oxidative C−H functionalization to form
new C−C or C−N bonds has emerged as an elegant and robust
synthetic method for interesting products, featuring atom
economy and straightforwardness.4 However, many of these
methods are based on transition-metal catalysis, bringing cost
and toxicity concerns. As an alternative, metal-free versions
have become very important for avoiding more expenditure and
environmental issues.5 The direct construction of the nitrogen
containing heterocycles via metal-free oxidative couplings is yet
limited;5 only select excellent examples were applied to
quinazoline and benzimidazole synthesis (Scheme 1),6 with a
limited substrate scope and prolonged reaction time.
Distinct from the literature-reported one-way strategies,
herein we report an unprecedented divergent synthesis of
quinazolines and benzimidazoles directly from the same starting
material amidine 3, in a solvent- or oxidant-switchable manner
under metal-free and base-free conditions (Scheme 1). By using
iodobenzene diacetate as the only reagent, a broad range of
multisubstituted quinazolines and benzimidazoles were fur-
nished via an intramolecular oxidative C(sp3)−C(sp2) and
C(sp2)−N bond formation in nonpolar and polar solvents,
respectively. Further selective synthesis of quinazolines in polar
solvents was achieved by TEMPO-catalyzed oxidative sp3C−H
arylation.
Structurally diverse amidines 3 were easily prepared from
readily available carboxylic acid, aniline, and alkyl/benzylamine,
via two different condensation sequences (Scheme 2), based on
the imine−enamine tautomerism,7 ensuring a broad substrate
availability scope.
Scheme 2. Two Synthetic Routes to the Designed Amidines
Received: March 23, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ol500864r | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
We designed the N-benzyl-N′-phenyl benzimidamide sub-
strate 3a to build quinazoline via a novel intramolecular
C(sp3)−H arylation strategy. Our initial effort commenced with
hypervalent iodine(III), i.e. iodobenzene diacetate, as an
oxidation reagent in toluene. Gratifyingly, the direct C(sp3)−
H activation/C(sp3)−C(sp2) bond formation occurred and the
desired quinazoline product 4a was obtained in 63% yield in
the absence of a transition metal catalyst. From systematic
screening of a range of solvents, oxidants and its ratio, substrate
concentration, reaction temperature, and time (see Supporting
Information (SI) for the details), the best result was achieved
when 4 equiv of PhI(OAc)2 were used as the oxidant with
nonpolar solvent toluene at 120 °C for 1 h (Scheme 3, 95% for
butyl substituted quinazoline was generated in a comparable
yield (4i, 76%).
Then, the effect of the substituent attached on the aniline
ring (i.e., R2) was examined. The bromo, chloro, or fluoro
group had little influence on the reaction (Scheme 3, 4j−4l and
4r), thus offering the possibility of introducing further
substituents by additional coupling reactions. However, the
strong electron-withdrawing group substituted at the para
position of the aniline ring remarkably retarded the direct
annulation reaction (4m, 5%). The electron-donating group at
the para position of the aniline ring also adversely affected the
conversion (4n and 4o). But the introduction of an electron-
donating group to the meta position marginally reduced the
reaction yield (4p and 4q). When the aniline ring was
substituted at the meta position, there were two positions for
the oxidative coupling to occur, favoring the less sterically
hindered C−H bond, and thus the regioselectivity increased as
the substituent size increased. For example, the meta-methyl
substituted substrate produced a mixture of two regioisomers in
a 5:1 ratio (4p), while tert-butyl substitution led to nearly a
single product (4q).
Scheme 3. Synthesis of Quinazolines via Metal-Free
a
−c
Oxidative sp3C−H/sp2C−H Cyclization of Amidines
Finally, we surveyed the scope of the R3 moiety. A range of
substrates bearing electron-donating or weak electron-with-
drawing group substituted aromatic rings all underwent the
oxidative cyclization smoothly and gave the desired quinazoline
products in excellent yields (Scheme 3, 4s, 4t, and 4w). Even a
strong electron-withdrawing group on the aromatic ring only
caused a slight decrease in the yield (4u and 4v). However, this
conversion was incompatible with the aliphatic substituent
(4x).
Interestingly, from screening the reaction conditions for
quinazoline synthesis, we found that benzimidazole was formed
as a major product when polar solvents were used, such as
EtOH, MeCN. Therefore, we were intrigued to further
investigate the annulation of N-benzyl-N′-phenyl benzimida-
mide 3a to generate the benzimidazole 5a by a direct sp2 C−H/
N−H coupling (Table 2). Through reaction conditions
optimization (see SI), we found that the best result was
obtained by using 3 equiv of PhI(OAc)2 in acetonitrile (with
0.1 M optimal concentration) at 120 °C for 1 h, furnishing 1,2-
substituted 5a in 92% yield.
To further evaluate the scope of the oxidative synthesis of
benzimidazoles in the polar solvent, a survey of N-phenyl-
amidine substrates 3 was conducted (Scheme 4). In general, the
substrate scope was very similar to that of the quinazoline
synthesis in toluene, with a diverse array of R1, R2, and R3
groups being compatible with the reaction conditions in good
to excellent yields (5a−5i for R1, 5j−5r for R2, 5x−5z for R3).
Notably, the alkyl group at R3 was well tolerated for this
transformation. The exception included the following: alky R1
substituted substrate 3i performed the sp2 C−H amination to
give 5i in only 40% yield; strong electron-withdrawing group R2
disfavored this conversion (5m). Compared to the recently
reported synthesis of benzimidazole via an intramolecular
oxidative coupling reaction,6e−g the major advantage of this new
approach is the broad substrate scope with more structurally
diverse substituents at position 1 and 2 thus enabling a rapid
derivatization of the drug-like scaffold.
a
Reaction conditions A: 3 (0.4 mmol), PhI(OAc)2 (1.6 mmol),
toluene (40 mL), 120 °C, 1 h. bReaction conditions B: 3 (0.2 mmol),
K2S2O8 (0.6 mmol), TEMPO (0.04 mmol), MeCN (4 mL), 120 °C, 3
c
h. Yield of isolated pure product under conditions A (isolated yield
under conditions B). n.d. = not detected. t = 4 h. t = 12 h, 35%
starting material recovered.
d
e
f
4a). Notably, the substrate concentration played an important
role in this oxidative cyclization reaction, and a lower reaction
concentration proved beneficial with 0.01 M being the best.
Equipped with a set of optimized conditions, we explored the
substrate scope of the oxidative C(sp3)−C(sp2) coupling
reaction delivering quinazolines, as depicted in Scheme 3.
First, we investigated the scope of the R1 substituent. Various
aromatic groups with electron-withdrawing and -donating
substituents at various positions were well tolerated, affording
the corresponding quinazolines in good to excellent yields
(4a−4g). However, the bulky naphthyl group disfavored the
formation of quinazoline (4h, 46%). Notably, this method is
applicable to the synthesis of 2-alkylquinazolines; e.g., 2-tert-
To gain insight into the reaction mechanism, we carried out
several control experiments (Scheme 5). First, the N-
methylated amidine 6 was subjected to the optimized
conditions for the quinazoline synthesis, but no reaction
occurred, indicating that the N−H was essential for the
B
dx.doi.org/10.1021/ol500864r | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. Synthesis of Benzimidazoles via PhI(OAc)2-
Promoted Oxidative Annulation in Acetonitrile
Scheme 6. Proposed Mechanism for the PhI(OAc)2-
Mediated Oxidative Annulation
a
,b
cation species C, which could undergo an intramolecular
Friedel−Craft reaction to afford intermediate E, which is
further oxidized to give the final product quinazoline 4. On the
other hand, when the reaction is performed in a polar solvent,
due to the solvation stabilization effect, intermediate A is apt to
be cleaved into cationic intermediate F, which is tautomerized
to nitrenium ion G. Then, the nitrogen bearing lone pair
electrons attack the benzene cation to generate intermediate H,
which delivers the heteroaromatic product 5 by proton
elimination. Meanwhile, cationic intermediate F can undergo
an intermolecular nucleophilic substitution reaction to afford
intermediate I. Intermediate I can be further rearranged to
compound J, which can serve as the substrate to undergo the
oxidative cyclization under the reaction conditions.
a
Reaction conditions: 3 (0.4 mmol), PhI(OAc)2 (1.2 mmol), MeCN
b
(4 mL), 120 °C, 1 h. Yields of isolated pure products.
Scheme 5. Control Experiments
Serendipitously, when exploring the mechanism of the
PhI(OAc)2-mediated benzimidazole synthesis by adding 0.2
equiv of TEMPO, we were surprised to harvest the quinazoline
as the major product in CH3CN, whereas the original dominant
product 5a was obtained in a substantially lowered yield (see
SI). The unusual result evoked our intense interest to develop a
new synthesis for quniazolines in polar solvents via a metal-free
TEMPO-catalyzed C(sp3)−H/C(sp2)−H coupling reaction. By
carefully examining the oxidation system with respect to the
oxidant species, the equivalence, the combination, the reaction
temperature, and time (see SI), the best result was achieved by
using 3 equiv of K2S2O8 as the oxidant and 0.2 equiv of
TEMPO as the additive in CH3CN at 120 °C for 3 h, affording
the quinazoline 4a in 94% yield.
Then, structurally diverse substrates were subjected to the
optimized conditions to evaluate the scope and generality of the
new protocol (Scheme 3; the yield is in parentheses). In
general, the TEMPO-catalyzed synthesis of quinazoline in
acetonitrile displayed superior reaction efficiency compared to
the PhI(OAc)2-mediated synthesis of quinazoline in toluene,
with an improved yield for the same product. The structural
scope and regioselectivity were very similar to those under the
PhI(OAc)2−toluene system. Several exceptions occurred:
furanyl as the R1 substituent on the amidine moiety was not
compatible with this procedure (4g); an obvious electronic
effect was observed with the R2 substituent on the aniline ring
(Scheme 3, products 4j−4r); an electron-donating group was
beneficial for the sp3C−H arylation (4n−4q), while the
coupling. Second, in the presence of 2 equiv of radical trapping
agent TEMPO, the quinazoline product 4a was still formed in
95% yield, ruling out a free radical mechanism. Then a chiral
substrate 8 yielded the corresponding benzimidazole 10 in
polar solvent with optical activity retention, while, in the
nonpolar solvent, the generated corresponding hydrogenated
quinazoline 9 almost lost all optical activity (see SI). Fianlly,
when using EtOH as the solvent, para-ethoxy substituted side-
product 5a′ was isolated in 6% yield. These results imply that
an aromatic cationic intermediate is involved.
Based on the observations above and previous reports on
hypervalent iodine and amidine involved reactions,8 we
proposed a plausible mechanism, as shown in Scheme 6.
Initially, treatment of the substrate 3 with PhI(OAc)2 could
furnish an N-(phenylacetoxyiodo)imidamido species A.8b−d
Then, in a nonpolar solvent, the iodoimidamido species A is
prone to conversion to the neutral N-benzylidene-N′-phenyl-
carboxamidine B by an intramolecular attack of the acetate ion
on the benzylic C(sp3)−H of A. Assisted by the proton acid
formed in situ, intermediate B is transformed to a benzylic
C
dx.doi.org/10.1021/ol500864r | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
electron-withdrawing group compromised the transformation
(4m, 15% yield), suggesting a cationic intermediate was
involved. Interestingly, the pyridino and aliphatic R3 substituted
amidines were competent in giving the desired products in
good yields (4v, 4x′), complementary to the hypervalent
iodine/nonpolar solvent protocol.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: yqlong@simm.ac.cn.
The exact mechanism of the TEMPO-catalyzed K2S2O8-
promoted oxidative Csp3−Csp2 coupling reaction was not clear
at this stage. But the reaction condition screening experiments
excluded the free radical mechanism (SI). Furthermore, the N-
methlylated substrate 6 failed to produce 3,4-dihydroquinazo-
line 7, and the electron-withdrawing p-CF3-phenyl substituted
amidine delivered compound 11 as a major product when
extending the reaction time to 5 h (Scheme 7).
Author Contributions
†J.-P.L. and F.-H.Z. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the National Natural Science
Foundation of China (81123004, 81325020 and
81361120410).
Scheme 7. Electron-Withdrawing Substituent Effect
REFERENCES
■
(1) (a) Bansal, Y.; Silakari, O. Bioorg. Med. Chem. 2012, 20, 6208.
(b) Michael, J. P. Nat. Prod. Rep. 2008, 25, 166. (c) Marzaro, G.;
Guiotto, A.; Chilin, A. Expert Opin. Ther. Pat. 2012, 22, 223.
(2) (a) Jin, Y.; Li, H. Y.; Lin, L. P.; Tan, J.; Ding, J.; Luo, X.; Long, Y.
Q. Bioorg. Med. Chem. 2005, 13, 5613. (b) Connolly, D. J.; Cusack, D.;
O’Sullivan, T. P.; Guiry, P. J. Tetrahedron 2005, 61, 10153. (c) Witt,
A.; Bergman, J. Curr. Org. Chem. 2003, 7, 659. (d) Lin, S. Y.; Isome, Y.;
Stewart, E.; Liu, J. F.; Yohannes, D.; Yu, L. B. Tetrahedron Lett. 2006,
47, 2883. (e) McGowan, M. A.; McAvoy, C. Z.; Buchwald, S. L. Org.
Lett. 2012, 14, 3800. (f) Baars, H.; Beyer, A.; Kohlhepp, S. V.; Bolm,
C. Org. Lett. 2014, 16, 536.
Taken together, a possible mechanism was proposed in
Scheme 8, in which the stoichiometric secondary oxidant
Scheme 8. A Possible Mechanism of the TEMPO-Catalyzed
Annulation Using K2S2O8 as the Oxidant
(3) (a) Peng, J. S.; Ye, M.; Zong, C. J.; Hu, F. Y.; Feng, L. T.; Wang,
X. Y.; Wang, Y. F.; Chen, C. X. J. Org. Chem. 2011, 76, 716. (b) Zou,
B. L.; Yuan, Q. L.; Ma, D. W. Angew. Chem., Int. Ed. 2007, 46, 2598.
(c) Zheng, N.; Anderson, K. W.; Huang, X. H.; Nguyen, H. N.;
Buchwald, S. L. Angew. Chem., Int. Ed. 2007, 46, 7509. (d) Han, B.;
Yang, X. L.; Wang, C.; Bai, Y. W.; Pan, T. C.; Chen, X.; Yu, W. J. Org.
Chem. 2012, 77, 1136. (e) Yuan, H.; Yoo, W. J.; Miyamura, H.;
Kobayashi, S. Adv. Synth. Catal. 2012, 354, 2899. (f) Brasche, G.;
Buchwald, S. L. Angew. Chem., Int. Ed. 2008, 47, 1932. (g) Saha, P.;
Ramana, T.; Purkait, N.; Ali, M. A.; Paul, R.; Punniyamurthy, T. J. Org.
Chem. 2009, 74, 8719.
(4) (a) Kumar, R.; Van der Eycken, E. V. Chem. Soc. Rev. 2013, 42,
1121. (b) Mousseau, J. J.; Charette, A. B. Acc. Chem. Res. 2013, 46,
412. (c) Zhang, C.; Tang, C. H.; Jiao, N. Chem. Soc. Rev. 2012, 41,
3464.
(5) Samanta, R.; Matcha, K.; Antonchick, A. P. Eur. J. Org. Chem.
2013, 2013, 5769.
K2S2O8 transformed TEMPO into an oxoammonium salt that
operated as the primary oxidant and hydrogen acceptor.9 In this
oxidation system, substrate 3 was oxidized to iminium
intermediate A, which then underwent an intramolecular
Friedel−Craft reaction to give intermediate B. Further
oxidation afforded the hydroquinone C, which was finally
aromatized to give the stable product quinazoline 4.
In summary, we have developed an unusual solvent- or
oxidant-switchable direct synthesis of quinazolines and
benzimidazoles from the common N-alkyl-N′-arylamidine
under metal- and base-free conditions. This is the first example
to achieve the chemoselectivity between oxidative sp3C−H/
sp2C−H and N−H/sp2C−H coupling simply by the choice of
solvent or oxidant, affording a broad range of multisubstituted
quinazolines and benzimidazoles in excellent yields.
(6) (a) Yan, Y. Z.; Zhang, Y. H.; Feng, C. T.; Zha, Z. G.; Wang, Z. Y.
Angew. Chem., Int. Ed. 2012, 51, 8077. (b) Zhang, J. T.; Zhu, D. P.; Yu,
C. M.; Wan, C. F.; Wang, Z. Y. Org. Lett. 2010, 12, 2841. (c) Yan, Y.
Z.; Wang, Z. Y. Chem. Commun. 2011, 47, 9513. (d) Han, B.; Wang,
C.; Han, R. F.; Yu, W.; Duan, X. Y.; Fang, R.; Yang, X. L. Chem.
Commun. 2011, 47, 7818. (e) Huang, J. B.; He, Y. M.; Wang, Y.; Zhu,
Q. Chem.Eur. J. 2012, 18, 13964. (f) Zhang, C.; Zhang, L. R.; Jiao,
N. Green Chem. 2012, 14, 3273. (g) Alla, S. K.; Kumar, R. K.; Sadhu,
P.; Punniyamurthy, T. Org. Lett. 2013, 15, 1334.
(7) Lin, J. P.; Long, Y. Q. Chem. Commun. 2013, 49, 5313.
(8) (a) Uneyama, K.; Kobayashi, M. J. Org. Chem. 1994, 59, 3003.
(b) Kim, H. J.; Kim, J.; Cho, S. H.; Chang, S. J. Am. Chem. Soc. 2011,
133, 16382. (c) Kikugawa, Y.; Nagashima, A.; Sakamoto, T.;
Miyazawa, E.; Shiiya, M. J. Org. Chem. 2003, 68, 6739. (d) Zhdankin,
V. V.; Koposov, A. Y.; Yashin, N. V. Tetrahedron Lett. 2002, 43, 5735.
(9) Mancheno, O. G.; Stopka, T. Synthesis-Stuttgart 2013, 45, 1602.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, reaction conditions screening,
compound characterization data, and copies of NMR spectra.
D
dx.doi.org/10.1021/ol500864r | Org. Lett. XXXX, XXX, XXX−XXX