Rhodium(I)-Catalyzed Decarbonylative Aerobic Oxidation of Cyclic α-Diketones: A Regioselective Single Carbon Extrusion Strategy

Article abstract of DOI:10.1021/acs.orglett.7b03837

A rhodium-catalyzed decarbonylative aerobic oxidation of cyclic α-diketones has been developed for the first time, where the regioselective formations of α-pyrones and isocoumarins have been achieved. The current decarbonylative aerobic oxidation pathway proceeds via the C-C bond cleavage followed by a C-O bond formation, representing a biomimetic oxidation approach to unsaturated six-membered cyclic lactones. The unique ability of rhodium catalysts to induce the decarbonylative aerobic oxidation opens up a new synthetic toolbox that utilizes the "regioselective single carbon" extrusion strategy.

Full text of DOI:10.1021/acs.orglett.7b03837

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Rhodium(I)-Catalyzed Decarbonylative Aerobic Oxidation of Cyclic
α‑Diketones: A Regioselective Single Carbon Extrusion Strategy
Gangadhararao Golime, Hun Young Kim, and Kyungsoo Oh*
Center for Metareceptome Research, College of Pharmacy, Chung-Ang University, 84 Heukseok-ro, Dongjak, Seoul 06974, Republic
of Korea
S
* Supporting Information
ABSTRACT: A rhodium-catalyzed decarbonylative aerobic
oxidation of cyclic α-diketones has been developed for the
first time, where the regioselective formations of α-pyrones
and isocoumarins have been achieved. The current decarbon-
ylative aerobic oxidation pathway proceeds via the C−C bond
cleavage followed by a C−O bond formation, representing a
biomimetic oxidation approach to unsaturated six-membered
cyclic lactones. The unique ability of rhodium catalysts to
induce the decarbonylative aerobic oxidation opens up a new synthetic toolbox that utilizes the “regioselective single carbon”
extrusion strategy.
elective activation of carbon−carbon bonds remains a
decarbonylative cycloaddition of cyclobutendiones with a
pendant alkene.11 Recently, the Dong group also explored the
Rh(I)-catalyzed decarbonylative coupling of alkynes12 and the
decarbonylative coupling of isatins with alkynes.13 While the
ketone decarbonylation strategy holds great synthetic value in
new aryl−aryl bond-forming reactions and provides ample
opportunity for redesigning the molecular structures, these
methods typically require strict anaerobic reaction conditions
due to the sensitive nature of organometallic species. To advance
the selective activation strategy of C−C bonds, we envisioned a
biomimetic oxidation approach inspired by mechanistic studies
of nonenzymatic C−C bond cleavage by nonheme metal
dioxygenase models.14 To the best of our knowledge there has
been no research effort toward effecting C−C bond cleavage/C−
O bond formation: the decarbonylative aerobic oxidation strategy
(Scheme 1C).
S
formidable challenge in organic synthesis,1 as they typically
rely on the oxidative addition of transition metal complexes to
highly strained molecules such as cyclopropanes,2 cyclobutanes,3
cyclobutanones,4 and directing-group-containing substrates5
(Scheme 1A). The recent development of Rh-catalyzed6 and
Scheme 1. Activation of C−C Bonds via Rh(I) Catalysis
In the course of development of a new synthetic strategy to
biologically relevant isocoumarin derivatives,15 we envisaged the
decarbonylative aerobic oxidation of ortho-naphthoquinones to
isocoumarins (Table 1). Thus, we subjected ortho-naphthoqui-
none 1a,16 prepared in two steps using Friedel−Crafts acylation
followed by Cu(I)-catalyzed aerobic oxidation, under refluxing
chlorobenzene conditions in the presence of 5 mol %
dirhodium(I) catalyst in open air (entry 1). While the reaction
temperature was as high as 130 °C, 1a did not undergo any
reaction, demonstrating the stability of 1a under the aerobic
conditions with the dirhodium catalyst. Next, we screened a
variety of phosphine ligands, otherwise under the same reaction
conditions (entries 2−5). Gratifyingly, we observed the
formation of two new products, isocoumarin 2a and coumarin
3a, in a 6:1 ratio. Among the screened phosphine ligands, the use
Ni-mediated7 decarbonylative C−C bond forming reactions
significantly expanded the scope of the C−C bond cleavage/C−
C bond formation in catalytic reactions beyond the traditional-
metal-mediated decabonylation reaction of aldehydes.8 Thus,
after the pioneering work of Tehanishi in 1974 on the
decabonylative C−C bond formations of α- and β-diketones
where RhCl(PPh3)3 was transformed to RhCl(CO)(PPh3)3,9 the
Murakami group in 1994 reported the Rh(I)-mediated
decarbonylation of cyclic ketones (Scheme 1B).10 The
Yamamoto group in 2006 reported the Rh(I)-catalyzed
Received: December 8, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b03837
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Optimization of the Rh(I)-Catalyzed
Decarbonylative Aerobic Oxidation
Scheme 2. Electronic Effect of ortho-Naphthoquinones
a
b
entry
Rh (mol %)
ligand (%)
solvent
yield (%)
NR
1
2
3
4
5
6
7
8
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (5)
Rh(PPh)3Cl (10)
−
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
DMI
o-Xyl
PhEt
PhMe
PhCl
PhCl
PhCl
PhCl
dppe (10)
dppb (10)
dppbz (10)
Xant (10)
dppb (10)
dppb (10)
dppb (10)
dppb (10)
dppb (10)
dppb (10)
dppb (10)
dppb (10)
dppb (5)
dppb (10)
dppb (10)
62/8
70/12
62/17
58/11
15/6
12/6
NR
Rh(PPh3)3(CO)H (10)
RhCl3 (10)
c
9
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (2.5)
[Rh(COD)Cl]2 (5)
[Rh(COD)Cl]2 (5)
15/2
66/11
61/10
28/9
10
11
d
12
e
13
82/14(68/9)
44/8
a
10 mol % catalyst used.
e
14
ef
,
15
16
73/13
8:1 ratios (2a−2e). The isolated yields of isocoumarins were in
the range 62−73%, demonstrating the synthetic utility of the
current C−C bond cleavage followed by a decarbonylative
aerobic oxidation. We also found that the electronic feature of
ortho-naphthoquinones could impact the regioselective for-
mation of isocoumarins. Thus, the increase of the electronic
density of the arenes improved the selective formation of
isocoumarins (2f−h, 2j), raising the overall reaction efficiency up
to an 81% yield (2g).
Upon examination, the alkyl-substituted ortho-naphthoqui-
nones compared to those that are aryl-substituted react faster due
to the strong electron-donating effect of an alkyl group at the C-4
position; however, the reaction was also accompanied by rapid
decomposition of the starting materials. Previously, we observed
the significant reactivity enhancement of 4-alkyl substituted 2-
naphthol derivatives during the oxidation16 as well as the
oxidative coupling reaction.17 In order to overcome the high
reactivity of these substrates, we used a high catalyst loading up to
10 mol % in a short reaction time at 100 °C or a low reaction
temperature of 50 °C with a catalyst loading of 15 mol %
(Scheme 3). We found that the regioselectivity pattern was also
relevant to the electronic feature of the arenes; thus, higher
selectivities were observed for 1m−1n compared to 1k−1l.
Likewise, the electron-rich ortho-naphthoquinones 1p and 1r, in
a direct comparison to 1o and 1q, significantly influenced the
overall efficiency of this novel oxidation reaction. The primary
alkyl chloride containing ortho-naphthoquinone 1s provided the
desired isocoumarin 2s in 37% yield.
eg
,
<1/0
a
Reaction conditions: 1a (0.1 mmol), Rh catalyst (mol %), ligand
b
1
(mol %) in solvent (0.1 M) at 130 °C. Yield was determined by H
c
NMR for 2a/3a. Isolated yields in parentheses. Reaction at 150 °C.
d
e
f
Reaction at 110 °C. Reaction for 60 h. Reaction using oxygen
g
balloon. Reaction under argon.
of 1,4-bis(diphenylphosphino)butane (dppb) provided 2a and
3a in 70% and 12% yield, respectively (entry 3). A subsequent
screening of rhodium salts revealed that the desired decarbon-
ylative aerobic oxidation pathway of 1a was only unique to the
[Rh(COD)Cl]2 catalyst (entries 6−8). The use of other solvents
suggested that the reaction temperature was a critical parameter
rather than the nature of solvents (entries 9−12). Thus, while the
use of 1,3-dimethyl-2-imidazolidinone (DMI) at 150 °C
primarily led to a low yields of products (entry 9), the reaction
at 130 °C using o-xylene (o-Xyl) and ethylbenzene (PhEt)
provided the comparable selectivities and yields (entries 10−11).
Also, the reaction at 110 °C in toluene was significantly slower
(entry 12). The reaction time could be extended to 60 h where
the yields of 2a and 3a were 84% and 14%, respectively (entry
13). The catalyst loading significantly helped the rate of reaction,
and the use of a 2.5 mol % catalyst loading did not lead to full
conversion, only providing a 52% yield (entry 14). Our control
experiments revealed that the molecular oxygen in the air was
sufficient to induce the decabonylative oxidation reaction
without an oxygen balloon (entry 15). However, the reaction
only proceeded in the presence of molecular oxygen to initiate
the desired synthetic process (entry 16).
The elaborated ortho-naphthoquinones 1t−1v readily pro-
vided the isocoumarin derivatives in excellent yields of 67−80%
(Scheme 4). In addition, while the current methodology could be
extended to unsubstituted ortho-naphthoquinone 1w, the yield of
the product remained low, possibly due to the decomposition of
the ortho-naphthoquinone 1w by the phosphine ligands.
Interestingly, we observed the preferential formation of
coumarins 3x−3y, when 4-methoxy or aryl substituted 1x−1y
were used. Although we do not have a plausible explanation for
With the optimized reaction conditions in hand, the reaction
scope of the Rh(I)-catalyzed decarbonylative aerobic oxidation
was studied using ortho-naphthoquinones with different
electronic and steric characters (Scheme 2). First, we
investigated the electronic influence of 4-aryl substituents of 1.
The electronic effect of aryl substituents was minimal, providing
the selective formation of isocoumarins over coumarins in 5:1−
B
DOI: 10.1021/acs.orglett.7b03837
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Alkyl-Substituent Effect of ortho-Naphthoquinones
Scheme 5. (A) Structural Features of α-Diketones; (B)
Proposed Reaction Mechanism
a
b
10 mol % catalyst loading at 100 °C. 15 mol % catalyst loading at 50
°C.
Scheme 4. Further Substrate Scope
preferential decarbonylation of α-diketones where the carbonyl
group with a more electron-donating group, not steric, was
removed. We also observed the formation of anhydride 4 in
about 1% yield. The possibility of 4 being the intermediate to the
lactone 2aa was ruled out because the anhydride 4 failed to
provide the lactone 2aa in a control experiment. In addition, we
investigated the conformational importance of α-diketones.
Thus, benzil 1ab did not undergo the desired decarbonylation19
and other acyclic α-diketones that adopt the conjugated s-cis
conformation such as 1ac and 1ad only underwent the
decarbonylation reaction to give the α,β-unsaturated carbonyl
compounds 5a−5b in 20−65% yields,20 clearly demonstrating
the importance of the cyclic nature of α-diketones for the desired
aerobic oxidation.
Mechanistically, the aerial oxidation of phosphine ligand
(dppb) is possible; however, our reaction monitoring revealed
that the phosphine ligand was not readily oxidized until the ortho-
naphthoquinones were completely consumed. Also, our control
experiments with the oxidized phosphine ligands revealed that
while the use of dppb(O)2 did not provide the isocoumarin
derivatives, dppb(O) was a still active ligand for the
decarbonylative oxidation (see the Supporting Information).
Based on the redox-active ortho-naphthoquinones that prevent
the oxidation of phosphine ligands,21 we propose a plausible
reaction mechanism (Scheme 5B) that manifests (1) the ligand/
substrate-accelerated oxidative addition to provide the rhoda-
cycle 1a and (2) the disproportionation of molecular oxygen to
generate carbon dioxide22 and phosphine oxides from phosphine
ligand oxidation.
a
b
10 mol % catalyst loading in PhCl at 130 °C. 10 mol % catalyst
c
loading in PhCH3 at 100 °C. 5 mol % catalyst loading in o-Xyl at 146
°C. 5 mol % catalyst loading in PhCl at 130 °C. 5 mol % catalyst
loading in DMI at 150 °C.
d
e
the selective formation of coumarins over isocoumarins, these
results strongly imply the possibility of accessing coumarin
derivatives with suitable substrates or reaction-parameter-
control. The use of 9,10-phenanthrenequinone 1z provided the
desired 6H-benzo[c]chromen-6-one 2z, albeit in a low yield of
35% due to the formation of byproduct 2z′ from the solvent
insertion.18
To understand the structural feature of α-diketones for the
decabonylative aerobic oxidation, we tested other α-diketones
(Scheme 5A). Upon subjecting 1aa to our optimized conditions,
the preferential formation of lactone 2aa was observed in 71%
yield. The regioselective formation of 2aa strongly implied the
C
DOI: 10.1021/acs.orglett.7b03837
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(5) For selected examples, see: (a) Chatani, N.; Ie, Y.; Kakiuchi, F.;
Murai, S. J. Am. Chem. Soc. 1999, 121, 8645. (b) Dreis, A. M.; Douglas,
C. J. J. Am. Chem. Soc. 2009, 131, 412. (c) Lei, Z.-Q.; Li, H.; Li, Y.; Zhang,
X.-S.; Chen, K.; Wang, X.; Sun, J.; Shi, Z.-J. Angew. Chem., Int. Ed. 2012,
51, 2690.
(6) For pioneering work, see: (a) Daugulis, O.; Brookhart, M.
Organometallics 2004, 23, 527. For the five-membered rings, see:
(b) Xia, Y.; Lu, G.; Liu, P.; Dong, G. Nature 2016, 539, 546. (c) Zeng, R.;
Chen, P.-H.; Dong, G. ACS Catal. 2016, 6, 969. (d) Xia, Y.; Wang, J.;
Dong, G. Angew. Chem., Int. Ed. 2017, 56, 2376.
In summary, we have discovered a novel decarbonylative
aerobic oxidation reaction of cyclic α-diketones to give the
regioselective formation of isocoumarins. Without relying on the
highly strained and directing-group-containing substrates, the
current decarbonylative aerobic oxidation of α-diketones readily
proceed in good to excellent regioselectivities and yields thanks
to the inherent metal coordination capability of the adjacent two
carbonyl groups in α-diketones. Another key feature of the
current method is the use of molecular oxygen to effect the
oxidation of rhodacycle intermediates, eliminating the precarious
nature of rhodium catalysts which typically requires a strict inert
reaction atmosphere. The extension of the decarbonylative
aerobic oxidation reactions is currently underway, and our results
will be reported in due course.
(7) (a) Ruhland, K.; Obenhuber, A.; Hoffmann, S. D. Organometallics
2008, 27, 3482. (b) Morioka, T.; Nishizawa, A.; Furukawa, T.; Tobisu,
M.; Chatani, N. J. Am. Chem. Soc. 2017, 139, 1416.
(8) For selected examples, see: (a) Morimoto, T.; Fuji, K.; Tsutsumi,
K.; Kakiuchi, K. J. Am. Chem. Soc. 2002, 124, 3806. (b) Shibata, T.;
Toshida, N.; Takagi, K. Org. Lett. 2002, 4, 1619. (c) Fessard, T. C.;
Andrews, S. P.; Motoyoshi, H.; Carreira, E. M. Angew. Chem., Int. Ed.
2007, 46, 9331. (d) Fristrup, P.; Kreis, M.; Palmelund, A.; Norrby, P. O.;
Madsen, R. J. Am. Chem. Soc. 2008, 130, 5206. (e) Akanksha; Maiti, D.
Green Chem. 2012, 14, 2314. (f) Modak, A.; Deb, A.; Patra, T.; Rana, S.;
Maity, S.; Maiti, D. Chem. Commun. 2012, 48, 4253.
(9) Kaneda, K.; Azuma, H.; Wayaku, M.; Tehanishi, S. Chem. Lett.
1974, 3, 215.
(10) Murakami, M.; Amii, H.; Ito, Y. Nature 1994, 370, 540.
(11) Yamamoto, Y.; Kuwabara, S.; Hayashi, H.; Nishiyama, H. Adv.
Synth. Catal. 2006, 348, 2493.
(12) (a) Dermenci, A.; Whittaker, R. E.; Dong, G. Org. Lett. 2013, 15,
2242. (b) Whittaker, R. E.; Dong, G. Org. Lett. 2015, 17, 5504.
(c) Dermenci, A.; Whittaker, R. E.; Gao, Y.; Cruz, F. A.; Yu, Z.-X.; Dong,
G. Chem. Sci. 2015, 6, 3201.
(13) Zeng, R.; Dong, G. J. Am. Chem. Soc. 2015, 137, 1408.
(14) For a review, see: Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L.,
Jr. Chem. Rev. 2004, 104, 939.
(15) For recent reviews, see: (a) Pal, S.; Chatare, V.; Pal, M. Curr. Org.
Chem. 2011, 15, 782. (b) Saeed, A. Eur. J. Med. Chem. 2016, 116, 290.
(16) (a) Kim, H. Y.; Oh, K. Org. Lett. 2014, 16, 5934. (b) Kim, H. Y.;
Takizawa, S.; Oh, K. Org. Biomol. Chem. 2016, 14, 7191.
(17) Kim, H. Y.; Takizawa, S.; Sasai, H.; Oh, K. Org. Lett. 2017, 19,
3867.
(18) Our control experiments confirmed that the Rh(I)-dppb catalyst
was responsible for the formation of byproduct 2z′, eliminating the
possibility of a simple solvent insertion to ortho-naphthoquinone 1z.
(19) The central C−C bond of benzil is 1.54 Å, illustrating the clear
single bond character due to the lack of conjugation between the two
carbonyl groups. Thus, the two benzoyl groups are twisted with each
other to have a dihedral angle of 117°; see: Shen, Q.; Hagen, K. J. Phys.
Chem. 1987, 91, 1357.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.7b03837.
Experimental procedures and characterization data for all
new compounds (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: kyungsoooh@cau.ac.kr.
ORCID
Kyungsoo Oh: 0000-0002-4566-6573
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by the National Research
Foundation of Korea (NRF) grants funded by the Korean
government (MSIP) (NRF-2015R1A5A1008958 and NRF-
2015R1C1A2A01053504).
REFERENCES
■
(1) For recent reviews, see: (a) Murakami, M.; Matsuda, T. Chem.
Commun. 2011, 47, 1100. (b) C−C Bond Activation. In Topics in
Current Chemistry; Dong, G., Ed.; Springer-Verlag: Berlin, 2014; Vol.
346. (c) Marek, I.; Masarwa, A.; Delaye, P.-O.; Leibeling, M. Angew.
Chem., Int. Ed. 2015, 54, 414. (d) Somerville, R. J.; Martin, R. Angew.
Chem., Int. Ed. 2017, 56, 6708. (e) Chen, P.-H.; Billett, B.; Tsukamoto,
T.; Dong, G. ACS Catal. 2017, 7, 1340.
(20) The reaction of 1ad to 5b was accompanied with the formation of
dimerization products, which accounts for the remaining 60% yields.
(21) For the ortho-naphthoquinones as redox active aerobic catalysts,
see: (a) Goriya, Y.; Kim, H. Y.; Oh, K. Org. Lett. 2016, 18, 5174. For
recent reviews on redox-active ligands, see: (b) Lyaskovskyy, V.; de
Bruin, B. ACS Catal. 2012, 2, 270. (c) Praneeth, V. K. K.; Ringenberg, M.
R.; Ward, T. R. Angew. Chem., Int. Ed. 2012, 51, 10228. (d) Luca, O. R.;
Crabtree, R. H. Chem. Soc. Rev. 2013, 42, 1440. (e) Chirik, P. J. Acc.
Chem. Res. 2015, 48, 1687. (f) Broere, D. L. J.; Plessius, R.; van der Vlugt,
J. I. Chem. Soc. Rev. 2015, 44, 6886.
(2) Matsuda, T.; Tsuboi, T.; Murakami, M. J. Am. Chem. Soc. 2007,
129, 12596.
(3) For selected examples, see: (a) Satoh, T.; Jones, W. D.
Organometallics 2001, 20, 2916. (b) Matsumura, S.; Maeda, Y.;
Nishimura, T.; Uemura, S. J. Am. Chem. Soc. 2003, 125, 8862.
(c) Masselot, D.; Charmant, J. P. H.; Gallagher, T. J. Am. Chem. Soc.
2006, 128, 694. (d) Shibata, T.; Nishizawa, G.; Endo, K. Synlett 2008,
2008, 765. (e) Seiser, T.; Roth, O. A.; Cramer, N. Angew. Chem., Int. Ed.
2009, 48, 6320. (f) Shigeno, M.; Yamamoto, T.; Murakami, M. Chem. -
(22) For the generation of carbon dioxide from the reaction of Rh(I)−
BINAP complex and dioxygen, see: Bunten, K. A.; Farrar, D. H.; Poe, A.
̈
J.; Lough, A. Organometallics 2002, 21, 3344.
Eur. J. 2009, 15, 12929. (g) Crep
Chem., Int. Ed. 2010, 49, 620.
́
in, D.; Dawick, J.; Aïssa, C. Angew.
(4) For recent examples, see: (a) Lu, G.; Fang, C.; Xu, T.; Dong, G.;
Liu, P. J. Am. Chem. Soc. 2015, 137, 8274. (b) Chen, P.-H.; Sieber, J.;
Senanayake, C. H.; Dong, G. Chem. Sci. 2015, 6, 5440. (c) Zhou, X.;
Dong, G. J. Am. Chem. Soc. 2015, 137, 13715. (d) Deng, L.; Xu, T.; Li,
H.; Dong, G. J. Am. Chem. Soc. 2016, 138, 369. (e) Zhou, X.; Ko, H. M.;
Dong, G. Angew. Chem., Int. Ed. 2016, 55, 13867. (f) Zhou, X.; Dong, G.
Angew. Chem., Int. Ed. 2016, 55, 15091.
D
DOI: 10.1021/acs.orglett.7b03837
Org. Lett. XXXX, XXX, XXX−XXX