Rh(III)-Catalyzed Regio- and Chemoselective [4 + 1]-Annulation of Azoxy Compounds with Diazoesters for the Synthesis of 2H-Indazoles: Roles of the Azoxy Oxygen Atom

Article abstract of DOI:10.1021/acs.orglett.7b00631

A Rh(III)-catalyzed tandem C-H alkylation/intramolecular decarboxylative cyclization of azoxy compounds with diazoesters for the synthesis of 3-acyl-2H-indazoles is disclosed. The azoxy instead of the azo group enables a distinct approach for cyclative capture, leading to a [4 + 1]-annulation rather than a classic [4 + 2] manner. The azoxy oxygen atom is traceless after annulation, and further removal from the product is not required. This reaction features a complete regioselectivity for unsymmetrical azoxybenzenes and a compatibility of monoaryldiazene oxides.

Full text of DOI:10.1021/acs.orglett.7b00631

Letter
pubs.acs.org/OrgLett
Rh(III)-Catalyzed Regio- and Chemoselective [4 + 1]-Annulation of
Azoxy Compounds with Diazoesters for the Synthesis of
2H‑Indazoles: Roles of the Azoxy Oxygen Atom
Zhen Long, Zhigang Wang, Danni Zhou, Danyang Wan, and Jingsong You*
Key Laboratory of Green Chemistry and Technology of Ministry of Education, College of Chemistry, Sichuan University, 29
Wangjiang Road, Chengdu 610064, P.R. China
S
* Supporting Information
ABSTRACT: A Rh(III)-catalyzed tandem C−H alkylation/intramolecular
decarboxylative cyclization of azoxy compounds with diazoesters for the synthesis
of 3-acyl-2H-indazoles is disclosed. The azoxy instead of the azo group enables a
distinct approach for cyclative capture, leading to a [4 + 1]-annulation rather than
a classic [4 + 2] manner. The azoxy oxygen atom is traceless after annulation, and
further removal from the product is not required. This reaction features a complete regioselectivity for unsymmetrical
azoxybenzenes and a compatibility of monoaryldiazene oxides.
he indazole backbone has been recognized as one of the
most privileged structures found in natural products,
Scheme 1. Annulations of Azo and Azoxy Compounds
T
pharmaceuticals, and bioactive compounds.1 Among these
heterocyclic structures, 2H-indazoles, especially 3-acyl-2H-
indazoles, are not only crucial pharmacophores (Figure 1)2 but
Figure 1. Selected biologically active 3-acyl-2H-indazoles.
also valuable synthetic intermediates that can be transformed
into useful molecules.3 Although numerous efficient methods
have been well developed for the synthesis of 1H-indazoles,1,4
only a limited number of reports have documented the
regioselective synthesis of 2H-indazoles.5,6 In particular, the
known routes toward 3-acyl-2H-indazoles often suffer from
multistep synthetic precursors, harsh reaction conditions, and/or
limited substrate scope. Thus, it is highly desirable to develop
concise and regioselective strategies for preparing 3-acyl-2H-
indazoles from readily accessible starting materials.
During the past two decades, transition-metal-catalyzed C−H
bond activation and subsequent functionalization have emerged
as powerful tools to construct heterocycles.7 Recently, the azo
group has attracted interest as an incorporated directing group to
build heteroaromatic backbones. Pd-, Rh-, Co-, or Re-catalyzed
C−H annulations of azobenzenes with aldehydes have been
developed for the synthesis of 2H-indazoles (Scheme 1, eq 1).8
However, unsymmetrical azobenzenes as the reaction substrate
may suffer from site-selectivity problems due to the competitive
coordination of two electronically similar nitrogen atoms to the
metal center. The incorporation of the oxygen atom into the azo
group to form the azoxy unit could avoid such a competitive
coordination. Although two kinds of potentially reactive C−H
bonds still exist, the electronically distinctly different nitrogen
and oxygen atoms are capable of selectively coordinating to the
metal center.9 Recent examples on the palladium-catalyzed C−H
functionalization of azoxybenzenes demonstrated that the
nitrogen atom in the azoxy group is a stronger coordinating
atom than the O atom.10 Thus, the azoxy group could be an ideal
incorporated directing group to implement both regio- and
chemoselective C−H annulations.
Diazo compounds as carbine precursors play a prominent role
in transition-metal-catalyzed C−H coupling or cyclization due to
Received: March 1, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00631
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
their easy preparation and high reactivity.11 Recently, Lee and
Kim independently disclosed the Rh(III)-catalyzed C−H bond
alkylation/cyclization of azobenzenes with cyclic diazotized
Meldrum’s acid, delivering six-membered cinnolin-3(2H)-ones
through a [4 + 2]-annulation (Scheme 1, eq 2).12 Acyclic α-
diazoesters could not fully accomplish the cyclization and tended
to remain in the alkylation step. Thus, we supposed that the
reaction of azoxybenzenes with α-diazo compounds could enable
five-membered 2H-indazoles for the following reasons. The
incorporated strongly electronegative oxygen atom significantly
decreases electron density on the reaction site, N atom, and thus
increases its electrophilic ability toward the α-C(sp3) of the
alkylated intermediate. In other words, the incorporation of the
oxygen atom enables to reverse the electrophilic/nucleophilic
properties of the reactive N atom that determine the [4 + 1]- or
[4 + 2]-annulation (Scheme 1, eq 3).
and PivOH (1.0 equiv) in a 1:1 mixture of DCE/dioxane (1:1, v/
v) at 130 °C for 24 h (Table S1, entry 18).
Subsequently, we explored the substrate scope. The reactivity
of diazoesters was investigated first (3aa−ar) (Scheme 2).
Scheme 2. Scope of Diazoesters and Symmetrical
a
Azoxybenzenes
Although the azoxy groups have recently been employed as the
directing group for the o-C−H bond functionalization of
azoxybenzenes (mainly the palladium-catalyzed acylation), the
investigated substrate scope is mostly limited to symmetrical
azoxybenzenes and the merged oxygen atom remained in the
products.10,13 Following our continuous interest in the rhodium-
catalyzed C−H functionalization of N-oxides,14 we herein report
the azoxy group as an incorporated directing group for a tandem
C−H alkylation/cyclization for the synthesis of 3-acyl-2H-
indazoles from easily available symmetrical and unsymmetrical
di- and monoaryldiazene oxides (see section II in the SI for
synthetic details).
Initially, we found that azoxybenzenes could smoothly react
with [Cp*RhCl2]2 in methanol in the presence of NaOAc at 80
°C for 24 h (eqs S4 and S5). The X-ray crystallographic analysis
of isolated complexes I and II (64% and 40% yields, respectively)
disclosed the five-membered rhodacycle complex (Figure 2). For
a
Reaction conditions: 1 (0.2 mmol) and 2 (1.5 equiv) in 1.0 mL of
DCE/dioxane (1:1, v/v) at 130 °C for 24 h under N2. For the R2
group details, see the SI. 1.0 mmol scale. Di-tert-butyl 2-
diazomalonate was used. 1.0 mL of DCE at 140 °C.
b
c
d
Gratifyingly, a wide range of diazoesters smoothly reacted with
3,3′-azoxytoluene 1a. First, various α-diazomalonates were
tested, all of which proceeded smoothly in good yields. For
example, diisopropyl 2-diazomalonate (2c) reacted with 1a to
afford 3ac in an 80% yield. Interestingly, when the R2 group is t-
Bu, the decarboxylated 3af was isolated as the sole product in a
34% yield, probably because of the easy hydrolysis of tert-butyl
ester under the acidic conditions.15 In addition to α-
diazomalonates, α-diazo-β-keto esters could also undergo the
cyclization. The carbon−carbon single bond cleavage of α-diazo-
β-keto esters involved a decarboxylation rather than a
deacylation, affording the corresponding 3-acyl-2H-indazoles
3ag−ar in satisfactory yields. To our delight, both α-diazo-β-alkyl
keto esters and α-diazo-β-(hetero)aryl keto esters underwent the
cyclization. N-Heterocycle-containing diazoesters such as methyl
2-diazo-3-oxo-3-(pyridin-2-yl)propanoate (2t), methyl 2-diazo-
2-(pyridin-2-yl)acetate (2u), and methyl 2-(benzo[d]thiazol-2-
yl)-2-diazoacetate (2v) failed to undergo this annulation. Other
diazo reagents such as EtCO2CHN2 (2w), EtCO2C(CH3)N2
(2x), EtCO2CPhN2 (2y), and EtCO2C(CF3)N2 (2z) did not
deliver the desired products.
Figure 2. ORTEP diagrams of complexes I and II. Thermal ellipsoids
are shown at the 50% probability level.
both symmetrical azoxybenzene 1f and unsymmetrical azox-
ybenzene 1p, the rhodium(III) center coordinates to the
nitrogen atom rather than the oxygen atom, which would lead
to a 2H-indazole backbone rather than a benzo[c]isoxazole.
Subsequently, we performed the reaction of 3,3′-azoxytoluene 1a
with dimethyl 2-diazomalonate 2a for the reaction condition
optimization (Table S1). Gratifyingly, the desired 2H-indazole
3aa was obtained in a 11% yield in the presence of [Cp*RhCl2]2
in combination with AgSbF6 and KOAc in 1,2-dichloroethane
(DCE) at 120 °C for 24 h (Table S1, entry 1). When PivOH and
AcOH were used as the additive instead of KOAc, the yield
increased to 71% and 67%, respectively (Table S1, entries 2 and
3). However, only a trace amount of 3aa was obtained in the
absence of PivOH, AcOH, and KOAc (Table S1, entry 6). These
results indicated that the acidic nature of additives had a more
significant effect than the type of carboxylates (pivalate and
acetate). Dioxane proved to be superior to DCE, methanol, THF,
and toluene (Table S1, entries 2 and 9−13). The best result was
obtained using [Cp*RhCl2]2 (2.5 mol %)/AgSbF6 (10 mol %)
We next investigated the scope of symmetrical azoxybenzenes
(3ba−oa) (Scheme 2). Ortho-, meta-, and para-substituted
azoxybenzenes were all compatible with the optimal conditions,
but ortho-substituted azoxybenzenes gave a relatively low yield
B
DOI: 10.1021/acs.orglett.7b00631
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
possibly due to steric congestion. Meta-substituted azoxyben-
zenes with either electron-donating or electron-withdrawing
groups all gave an exclusively regioselective product at the less
sterically hindered position.
Scheme 4. Investigation of Mechanism
Most importantly, the current protocol was suitable for
unsymmetrical azoxybenzenes and monoaryldiazene oxides.
These azoxy compounds were easily prepared from the reaction
of 1-aryl-2-(tosyloxy)diazene 1-oxide with Grignard reagent (for
details, see section II in the SI), thus enabling a diversified library
of 3-acyl-2H-indazoles (Scheme 3). Notably, in contrast to
Scheme 3. Scope of Unsymmetrical Azoxybenzenes and
a
Monoaryldiazene Oxides
annulated product 3fa in an almost quantitative yield. In contrast,
the isolated decarboxylative molecule 5 could not afford any
desired 3fa (Scheme 4, eq 2). Unlike the alkylated azoxybenzene
4, the alkylated azobenzene 6 could not enable a [4 + 1]-
annulation but led to a [4 + 2] manner to form the six-membered
cinnolin-3(2H)-one 7 (Scheme 4, eq 3), indicating a pivotal
function of the azoxy oxygen atom in cyclative capture. These
results suggest that an alkylated azoxybenzene might be involved
in the intermediate process.
On the basis of the above experimental results and previous
reports,12 a plausible reaction mechanism is depicted in Scheme
5. First, the active [Cp*Rh(III)] species A is generated by the
Scheme 5. Proposed Pathway to 3-Acyl-2H-indazole
a
The reaction was run under the same conditions shown in Scheme 2.
1.0 mL of DCE at 140 °C.
b
unsymmetrical azobenzenes,12a unsymmetrical azoxybenzenes
bearing two electronically highly similar aromatic rings could
afford an exclusive regioselectivity (3qa, 3ra, and 3qg) (Scheme
3). A thiophene-yl-containing azoxy compound was also a
suitable substrate (3va). Furthermore, monoaryldiazene oxides
efficiently coupled with both α-diazomalonates and α-diazo-β-
keto esters to give 2H-indazoles (3wa, 3xa, 3wg, and 3xg)
(Scheme 3).
To gain insight into the mechanism, a series of experiments
were carried out. First, the hydrogen−deuterium exchange
experiments were performed. Treatment of azoxybenzene 1f
with PivOD led to significant deuterium scrambling (eq S1), thus
indicating that the cleavage of the ortho C−H bond is a reversible
process. Then kinetic isotope effect (KIE) experiments were run
with the reactions of 1f and/or [D10]-1f with ethyl
diazoacetoacetate 2g. The KIE values for the parallel and
intermolecular competitive reactions were detected to be 2.3 and
2.6, respectively (eqs S2 and S3).
We next investigated the stoichiometric and catalytic reactivity
of the rhodacycle complexes I and II (eqs S6−S8). Complexes I
and II could efficiently catalyze the cyclization of 1f and 1p to
furnish the expected products 3fg and 3pa in 55% and 64%
yields, respectively. A stoichiometric amount of complex II could
also react with 2a to afford 3pa in a 53% yield. These
observations suggest that the five-membered cyclometalated
complex is probably the intermediacy in the catalytic cycle.
Fortunately, the reaction of 1f with 2a at a lower temperature
(90 °C) led to the alkylated product 4 in a 53% yield (Scheme 4,
eq 1). Surprisingly, only in a mixture solvent of DCE/dioxane
(1:1, v/v) was 4 heated at 130 °C for 24 h to produce the
anion exchange of AgSbF6 with [Cp*RhCl2]2. The electrophilic
[Cp*Rh(III)] species coordinates to the azoxy group and then
undergoes a directed reversible C−H bond cleavage to form the
five-membered rhodacycle intermediate B, followed by coordi-
nation to a diazoester to form the diazonium intermediate C.
Subsequently, the reaction undergoes two possible pathways. In
pathway a, the metal−carbene intermediate D is formed with the
extrusion of nitrogen gas from C, followed by a migratory
insertion of the carbene into the rhodium−carbon bond to form
the six-membered rhodacycle species E. In pathway b, a direct
intramolecular 1,2-migratory insertion of the aryl group affords
E. The protonation of E next produces the alkylated intermediate
F and regenerates the active rhodium species A. Finally, F could
undergo intramolecular nucleophilic attack of the malonate ester
by oxygen of azoxy, followed by CO2 extrusion and intra-
molecular cyclization to deliver 3pa with the release alcohol
C
DOI: 10.1021/acs.orglett.7b00631
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(confirmed by GC−MS). In contrast, without the help of an
extra electron-drawing ester group, 5 difficultly undergoes the
intramolecular nucleophilic attack and subsequent decarbox-
ylation.
Finally, using 3fa as a representative example, we investigated
the synthetic potential of the reaction. As shown in Scheme 6,
REFERENCES
■
(1) (a) Elguero, J. Pyrazoles and their Benzo Derivatives. In
Comprehensive Heterocyclic Chemistry; Katritzky, A. R., Rees, C. W.,
Eds.; Pergamon: New York, 1984; Vol. 5, pp 167−303. (b) Cerecetto,
́ ́ ́
H.; Gerpe, A.; Gonzalez, M.; Aran, V. J.; de Ocariz, C. O. Mini-Rev. Med.
Chem. 2005, 5, 869. (c) Schmidt, A.; Beutler, A.; Snovydovych, B. Eur. J.
Org. Chem. 2008, 2008, 4073. (d) Haddadin, M. J.; Conrad, W. E.;
Kurth, M. J. Mini-Rev. Med. Chem. 2012, 12, 1293.
Scheme 6. Transformations of 3fa to Functional Molecules
(2) (a) Halim, R.; Harding, M.; Hufton, R.; Morton, C. J.; Jahangiri, S.;
Pool, B. R.; Jeynes, T. P.; Draffan, A. G.; Lilly, M. J.; Frey, B. WO 2012/
051659 A1, 2012. (b) Steffan, R. J.; Matelan, E. M. WO 2006/050006
A2, 2006. (c) Aman, W.; Lee, J.; Kim, M.; Yang, S.; Jung, H.; Hah, J.-M.
Bioorg. Med. Chem. Lett. 2016, 26, 1188.
(3) (a) Schmidt, A.; Merkel, L.; Eisfeld, W. Eur. J. Org. Chem. 2005,
2005, 2124. (b) Zhou, Y.; Liu, Q.; Lv, W.; Pang, Q.; Ben, R.; Qian, Y.;
Zhao, J. Organometallics 2013, 32, 3753. (c) Vidyacharan, S.; Sagar, A.;
Sharada, D. S. Org. Biomol. Chem. 2015, 13, 7614. (d) Cheng, Y.; Li, G.;
Liu, Y.; Shi, Y.; Gao, G.; Wu, D.; Lan, J.; You, J. J. Am. Chem. Soc. 2016,
138, 4730.
(4) (a) Inamoto, K.; Saito, T.; Katsuno, M.; Sakamoto, T.; Hiroya, K.
Org. Lett. 2007, 9, 2931. (b) Yu, D.-G.; Suri, M.; Glorius, F. J. Am. Chem.
Soc. 2013, 135, 8802. (c) Chen, C.-Y.; Tang, G.; He, F.; Wang, Z.; Jing,
H.; Faessler, R. Org. Lett. 2016, 18, 1690. (d) Wang, Q.; Li, X. Org. Lett.
2016, 18, 2102.
treatment of 3fa with KOH/EtOH and subsequent decarbox-
ylation provided 2-phenyl-2H-indazole 8 in a 90% yield. Next, 8
reacted with dimethyl acetylenedicarboxylate (DMAD) and
furan-2-carbaldehyde to deliver the 4-aminoquinoline derivative
93c and the Indazo-Fluor fluorophore 10,3d respectively. The
quaternization of 3fa with dimethyl sulfate and subsequent
hydrolysis produced the indazolium-3-carboxylate 11, which can
in situ generate a N-heterocyclic carbene by thermal decarbox-
ylation.3a
In summary, we have developed a Rh(III)-catalyzed
regioselective tandem C−H alkylation/cyclization of azoxy
compounds with diazoesters to afford 3-acyl-2H-indazole
derivatives. The azoxy group instead of the azo group enables a
[4 + 1]-annulation rather than a conventional [4 + 2] manner.
This reaction shows a broad functional group tolerance, a
complete regioselectivity for unsymmetrical azoxybenzenes and a
compatibility of monoaryldiazene oxides. The azoxy group has
been disclosed as an incorporated directing group, which could
have pronounced implications in future work and eventually lead
to new, valuable synthetic reactions.
́
(5) For selected examples, see: (a) Halland, N.; Nazare, M.; R’kyek, O.;
Alonso, J.; Urmann, M.; Lindenschmidt, A. Angew. Chem., Int. Ed. 2009,
48, 6879. (b) Wu, C.; Fang, Y.; Larock, R. C.; Shi, F. Org. Lett. 2010, 12,
2234. (c) Kumar, M. R.; Park, A.; Park, N.; Lee, S. Org. Lett. 2011, 13,
3542. (d) Genung, N. E.; Wei, L.; Aspnes, G. E. Org. Lett. 2014, 16, 3114.
(6) For the synthesis of 3-acyl-2H-indazoles, see: (a) Luo, G.; Chen, L.;
Dubowchik, G. J. Org. Chem. 2006, 71, 5392. (b) Bunnell, A.; O’Yang,
C.; Petrica, A.; Soth, M. J. Synth. Commun. 2006, 36, 285. (c) Unsinn, A.;
Knochel, P. Chem. Commun. 2012, 48, 2680. (d) Wang, C.-D.; Liu, R.-S.
Org. Biomol. Chem. 2012, 10, 8948. (e) Yong, W.-S.; Park, S.; Yun, H.;
Lee, P. H. Adv. Synth. Catal. 2016, 358, 1958.
(7) (a) Yamaguchi, J.; Yamaguchi, A. D.; Itami, K. Angew. Chem., Int.
Ed. 2012, 51, 8960. (b) Ackermann, L. Acc. Chem. Res. 2014, 47, 281.
(c) Huang, H.; Ji, X.; Wu, W.; Jiang, H. Chem. Soc. Rev. 2015, 44, 1155.
(d) Song, G.; Li, X. Acc. Chem. Res. 2015, 48, 1007.
(8) (a) Li, H.; Li, P.; Wang, L. Org. Lett. 2013, 15, 620. (b) Lian, Y.;
Bergman, R. G.; Lavis, L. D.; Ellman, J. A. J. Am. Chem. Soc. 2013, 135,
7122. (c) Hummel, J. R.; Ellman, J. A. J. Am. Chem. Soc. 2015, 137, 490.
(d) Geng, X.; Wang, C. Org. Lett. 2015, 17, 2434. (e) Jeong, T.; Han, S.
H.; Han, S.; Sharma, S.; Park, J.; Lee, J. S.; Kwak, J. H.; Jung, Y. H.; Kim,
I. S. Org. Lett. 2016, 18, 232.
ASSOCIATED CONTENT
* Supporting Information
■
S
(9) Balch, A. L.; Petridis, D. Inorg. Chem. 1969, 8, 2247.
(10) (a) Li, H.; Li, P.; Zhao, Q.; Wang, L. Chem. Commun. 2013, 49,
9170. (b) Sun, M.; Hou, L.-K.; Chen, X.-X.; Yang, X.-J.; Sun, W.; Zang,
Y.-S. Adv. Synth. Catal. 2014, 356, 3789. (c) Zhang, D.; Cui, X.; Yang, F.;
Zhang, Q.; Zhu, Y.; Wu, Y. Org. Chem. Front. 2015, 2, 951.
(11) (a) Davies, H. M. L.; Manning, J. R. Nature 2008, 451, 417.
(b) Doyle, M. P.; Duffy, R.; Ratnikov, M.; Zhou, L. Chem. Rev. 2010,
110, 704. (c) Xiao, Q.; Zhang, Y.; Wang, J. Acc. Chem. Res. 2013, 46, 236.
(d) Hu, F.; Xia, Y.; Ma, C.; Zhang, Y.; Wang, J. Chem. Commun. 2015,
51, 7986.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.7b00631.
Detailed experimental procedures, characterization data,
1
and H and 13C NMR spectra of key intermediates and
final products (PDF)
X-ray crystallogrphic data for complex I (CIF)
X-ray crystallogrphic data for complex II (CIF)
(12) (a) Son, J.-Y.; Kim, S.; Jeon, W. H.; Lee, P. H. Org. Lett. 2015, 17,
2518. (b) Sharma, S.; Han, S. H.; Han, S.; Ji, W.; Oh, J.; Lee, S.-Y.; Oh, J.
S.; Jung, Y. H.; Kim, I. S. Org. Lett. 2015, 17, 2852.
(13) (a) Li, H.; Xie, X.; Wang, L. Chem. Commun. 2014, 50, 4218.
(b) Xu, N.; Li, P.; Xie, Z.; Wang, L. Chem. - Eur. J. 2016, 22, 2236.
(14) (a) Huang, X.; Huang, J.; Du, C.; Zhang, X.; Song, F.; You, J.
Angew. Chem., Int. Ed. 2013, 52, 12970. (b) Huang, X.; Liang, W.; Shi, Y.;
You, J. Chem. Commun. 2016, 52, 6253.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: jsyou@scu.edu.cn.
ORCID
Jingsong You: 0000-0002-0493-2388
Notes
(15) Xia, Y.; Liu, Z.; Feng, S.; Zhang, Y.; Wang, J. J. Org. Chem. 2015,
80, 223.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by grants from the National NSF of
China (No. 21432005).
■
D
DOI: 10.1021/acs.orglett.7b00631
Org. Lett. XXXX, XXX, XXX−XXX