Aerobic transition-metal-free visible-light photoredox indole C-3 formylation reaction

Article abstract of DOI:10.1021/cs5005129

An aerobic visible-light-promoted indole C-3 formylation reaction catalyzed by Rose Bengal has been developed. This transition-metal-free process employs molecular oxygen as the terminal oxidant and uses TMEDA as the one-carbon source through C-N bond cleavage. The reaction is compatible with a variety of functional groups.

Full text of DOI:10.1021/cs5005129

Letter
pubs.acs.org/acscatalysis
Aerobic Transition-Metal-Free Visible-Light Photoredox Indole C‑3
Formylation Reaction
Xiang Li, Xiangyong Gu, Yongjuan Li, and Pixu Li*
Key Laboratory of Organic Synthesis of Jiangsu Province, College of Chemistry, Chemical Engineering, and Materials Science,
Soochow University, 199 RenAi Road, Suzhou, Jiangsu 215123, China
*
S Supporting Information
ABSTRACT: An aerobic visible-light-promoted indole C-3 for-
mylation reaction catalyzed by Rose Bengal has been developed. This
transition-metal-free process employs molecular oxygen as the
terminal oxidant and uses TMEDA as the one-carbon source through
C−N bond cleavage. The reaction is compatible with a variety of
functional groups.
KEYWORDS: aerobic oxidation, visible light photoredox, indole formylation, Rose Bengal, transition-metal-free
he chemistry of indole derivatives has attracted and
continues to attract interest from synthetic organic
yield of 3a was indeed increased to 35% (Table 1, entry 2). We
T
then screened a variety of photosensitizers, including Ru/Ir
polybipyridyl complexes and several organic dyes. It turned out
that Rose Bengal was the best catalyst among all the sensitizers
tested (entries 3−8). A survey of solvents, such as DMF,
chemists because the indole moiety exists in countless
biologically active molecules. Indole C-3 formylation is one of
the important methods for direct C-3 functionalization of
indole. The traditional methods for the synthesis of 3-
DMSO, DMA, and CH
better yield than MeCN (entries 9−12). It was found that
MeCN/H O (5:1) was the optimal mixed solvent system. The
Cl , mixed with water did not provide a
2
2
1
2
formylindoles were Vilsmeier−Haack and Reimer−Tiemann
reactions. These protocols suffered from harsh conditions, low
selectivity, and lack of functionality tolerance. In 2011, Su and
co-workers reported a Ru-catalyzed indole C-3 formylation
2
desired product was obtained in 48% (entry 13). To obtain a
synthetically useful protocol, we further optimized the reaction
3
using N-methylaniline as the carbonyl source. Cu-catalyzed C-
conditions using an O balloon (53%, entry 14), raising the
2
4
5
3
indole formylation reactions using TMEDA or DMSO as
reaction temperature to 60 °C (60%, entry 15), and adding 4
equiv of KI as the additive (75%, entry 16). At this stage, a set
of control reactions were carried out. No product 3a was
observed in the absence of catalyst, without visible-light
the one-carbon units were developed soon after. Recently,
metal-free indole C-3 formylation reactions were reported using
6
similar carbonyl sources. In the above-mentioned reports, the
key strategy was the C−N or C−S bond cleavage of the
nontraditional formyl sources through the formation of
iminium ion or sulfur ylide intermediates.
irradiation, or under N atmosphere (entries 14−16). The
2
results indicated that the reaction was a visible-light photoredox
reaction and molecular oxygen was required. Although our
reaction conditions are analogous to the reported Cu-catalyzed
7
Organic dyes, such as Rose Bengal and eosin Y, and Ru/Ir
4b,c
diindolemethane from indoles and TMEDA,
no diindole-
polybipyridyl complexes are known to oxidize amine to the8
corresponding radical cation under visible light irradiation.
Upon losing a hydrogen atom, the amine radical cation forms
an iminium ion. Many new visible-light-promoted reaction9s
have been developed utilizing the iminium ion intermediate.
Iminium ion generated by visible-light photoredox catalysis
methane side-product was observed in our reaction mixture. A
reaction performed in a more diluted solution did not improve
the yield, either.
With the optimized conditions in hand, we then turned our
focus to the scope and functional group compatibility of the
reaction. First, the substitution groups on the indole nitrogen
were studied. Primary and secondary alkyl groups, methyl,
ethyl, and isopropyl groups afforded 70%, 60%, and 71% yields,
respectively (Table 2, 3a−c). Allyl, cyclopropylmethyl, and
benzyl groups were tolerated (55%, 74%, and 47% respectively
for 3d−f). 2-(1H-indol-1-yl)Ethanol with a free hydroxyl group
afforded 45% yield of the corresponding C-3 formylated
product (3g), and its Boc-protected derivative gave the desired
10
could undergo electrophilic substitution with indole, resulting
11
in new indole C-3 functionalization reactions. In continuation
of our interest in visible-light-promoted aerobic oxidation
1
2
reactions, we started our investigation of the visible-light-
promoted indole C-3 functionalization reaction utilizing
molecular oxygen as the terminal oxidants.
To our delight, upon irradiation of visible light, the reaction
13
of N-methylindole (1a) with TMEDA in the presence of 2
mol % Rose Bengal under air afforded 3-formyl-N-methylindole
(
3a) in 12% yield in acetonitrile (Table 1, entry 1). Water was
Received: February 2, 2014
Revised: April 29, 2014
added into the reaction because it was required to hydrolyze
the iminium ion intermediate to afford the final product. The
©
XXXX American Chemical Society
1897
dx.doi.org/10.1021/cs5005129 | ACS Catal. 2014, 4, 1897−1900

ACS Catalysis
Letter
a
Table 1. Optimization of Reaction Conditions
Table 2. Visible-Light-Promoted Aerobic C-3 Formylation of
Indoles
,
a b
b
entry
catalyst
solvent
yield
c
1
2
3
4
5
6
7
8
9
Rose Bengal
Rose Bengal
MeCN
12%
35%
25%
28%
MeCN−H O (100:1)
MeCN−H O (100:1)
MeCN−H O (100:1)
MeCN−H O (100:1)
MeCN−H O (100:1)
MeCN−H O (100:1)
MeCN−H O (100:1)
DMSO−H O (100:1)
2
c
2
Ru(bpy) (PF )
3
6
2
c
2
Ru(bpy) Cl
3
2
c
Ir(ppy)3
trace
trace
trace
<5%
trace
5%
2
Rhodamine B
Alzarin Red S
2
2
Eosin Yellowish
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
2
2
1
1
1
1
1
1
1
0
1
2
3
4
5
6
DMF−H O (100:1)
2
CH Cl −H O (100:1)
22%
trace
48%
53%
2
2
2
DMA−H O (100:1)
MeCN−H O (5:1)
MeCN−H O (5:1), O ballon
MeCN−H O (5:1), O balloon, 60 °C 60%
2
2
2
2
2
2
d
MeCN−H O (5:1), 4 equiv KI, O
75%
2
2
2
2
2
balloon, 60 °C
d
1
1
1
7
8
9
MeCN−H O (5:1), 4 equiv KI, O
N.R.
N.R.
N.R.
2
balloon, 60 °C
d
d
Rose Bengal
Rose Bengal
MeCN−H O (5:1), 4 equiv KI, O
2
balloon, 60 °C, no light
MeCN−H O (5:1), 4 equiv KI, N
2
balloon, 60 °C
a
Reaction conditions: N-methylindole (0.5 mmol), TMEDA (2
equiv), photocatalyst (5 mol %), solvent (3 mL), room temp, air
balloon, using a 14 W fluorescent lamp as the visible light source, 72 h.
b
c
d
GC yield. 2 mol % catalyst was used. 48 h.
product, too (42%, 3h). It was notable that free indole worked
under the standard conditions and gave 3-formylindole in 52%
isolated yield (3i). Not surprisingly, when indole nitrogen was
protected with strong electron-withdrawing groups, such as N-
Boc and N-Ts, very little products were observed (3j and 3k).
Presumably, the electron-withdrawing groups lowered the
electron density of the indole ring and retarded the
electrophilic attack of iminium ion.
The substitution groups on the carbon skeleton of N-
methylindole ring were investigated, as well. For methyl
substitution at different positions, little impact on the reaction
outcome was observed. Substitution at the C-6 position
afforded a little better yield (71%, 3o) than the others (61%,
5
8%, 62%, and 56% for 2-, 4-, 5-, and 7-substitution,
respectively, 3l−n, 3p). Larger substitution at the C-2 position
did not affect the reaction (2-Ph for 62%, 3q). Electron-
donating groups did not facilitate the reaction (58% for 5-OMe,
a
Reaction conditions: indoles (1 mmol), TMEDA (2 equiv), Rose
Bengal (5 mol %), KI (4 equiv), MeCN−H O (5:1, 6 mL), under O
2
2
6
2% for 6-OMe, and 60% for 4-OBn, 3r−t). Halogen
balloon at 60 °C, using a 14 W fluorescent lamp as the visible light
functionalities gave slightly lower yields. Substituents on the
C-6 position generally afforded lower yields than that on the C-
b
source. Isolated yield.
5
position (3u vs 3v, 3w vs 3x, and 3y vs 3z). Again, strong
electron-withdrawing groups, 7-aza- or 4-CN, shut down the
reaction. Other electron-rich aromatic compounds, such as
furan and pyrrole, were subjected to the standard conditions;
however, no desired formylated products were observed.
under visible light irradiation sensitized by Rose Bengal. Several
reactions were carried out to investigate the reaction pathway
(Scheme 1). Tetraphenylporphine (TPP), a porphyrin dye
14
sensitizer known to generate singlet oxygen, was used as the
catalyst instead of Rose Bengal. No C-3 formylated product was
observed in either MeCN−H O or CH Cl (eq a). Although
It is known that molecular oxygen could form two reactive
1
speciessinglet oxygen (O ) through energy transfer or
2
2
2
2
•
−
superoxide radical anion (O ) via single electron transfer
the addition of DABCO, a singlet oxygen quencher, did not
2
1
898
dx.doi.org/10.1021/cs5005129 | ACS Catal. 2014, 4, 1897−1900

ACS Catalysis
Letter
Scheme 1. Investigations of Reaction Pathway
progressed. Only 7% of C-3 formylated product was detected
(
eq c). Therefore, it was believed that the combination of KI/
H O or I did not promote the reaction. The real function of
2
2
2
KI in this reaction needs further investigation.
On the basis of the above mechanistic studies and literature
reports, a possible mechanism is proposed (Figure 1).
change the reaction outcome, the addition of TEMPO, a radical
scavenger, significantly reduced the yield of the reaction to only
8
% (eq b). Deuterated solvent increases the lifetime of singlet
oxygen, so a reaction was performed in CD CN and D O (eq
3
2
Figure 1. Proposed mechanism.
c). The reaction progress was monitored by GC. No rate
acceleration or yield increase was observed compared with the
reaction in nondeuterated solvent. The results from the above
reactions strongly suggested that the reaction underwent the
superoxide radical anion pathway, not the singlet oxygen
pathway.
Oxidative quenching of the visible-light-excited Rose Bengal
(
RB*) by TMEDA resulted in the formation of Rose Bengal
•
−
radical anion (RB ) and radical cation 4. Upon transfer of an
The role of KI in the reaction was studied, too (Scheme 2).
Different equivalents of KI were tested in the reaction. It turned
electron to O to form the superoxide radical anion, Rose
2
Bengal was regenerated and completed the photoredox cycle.
On the other hand, the radical cation 4 gave up a hydrogen
atom, presumably to the superoxide radical anion to afford
hydrogen peroxide anion and iminium ion 5. Alternatively, the
radical cation 4 could give up a proton and form an α-amino
carbon radical, which could go through a second oxidation to
afford iminium ion 5. Electrophilic addition of iminium ion 5 to
N-methylindole (1a) gave indole iminium ion 6. Upon
rearomatization by giving up a proton, C-3-substituted N-
methylindole intermediate 7 was formed. Because it was more
facile for intermediate 7 to form iminium ion than TMEDA, a
second visible-light photoredox cycle occurred and generated
iminium ion 8. Hydrolysis of 8 cleaved the C−N bond and
afforded 3-formyl-N-methylindole (3a). To support the
postulation, the reaction of N-methylindole with TMEDA
under standard conditions was analyzed by LC−MS. In the
reaction mixture, the mass of intermediate 7 and N,N,N′-
trimethylethylenediamine (9) were observed. In addition, N-[2-
Scheme 2. Investigations of the Role of KI
(
dimethylamino)ethyl]-N-methylformamide, a byproduct pre-
sumably from the hydrolysis of iminium ion intermediate 5, was
detected by mass spectrometry (see Supporting Information).
These observations were good evidence for the proposed
reaction pathway.
In summary, a mild transition-metal-free visible-light photo-
redox indole C-3 formylation reaction using TMEDA as the
one-carbon source was developed. The reaction uses molecular
oxygen as the terminal oxidant. Inexpensive and readily
available organic dye, Rose Bengal, acts as the photosensitizer
in the reaction.
out that the yields increased with an increase in the KI loading
(
eq a). I generated in situ through the oxidation of KI by trace
2
amount of H O might act as the catalyst. To rule out this
2
2
possibility, stoichiometric H O was added to the mixture of N-
2
2
methylindole, TMEDA, and KI. No 3-formyl-N-methylindole
was formed (eq b). When a stoichiometric amount of I was
2
added to the mixture of N-methylindole and TMEDA, a fast
formation of 3-iodo-N-methylindole was observed, and it was
converted to unknown messy products as the reaction time
1
899
dx.doi.org/10.1021/cs5005129 | ACS Catal. 2014, 4, 1897−1900

ACS Catalysis
ASSOCIATED CONTENT
Letter
Stephenson, C. R. J. J. Org. Chem. 2012, 77, 4425−4431. (c) Freeman,
D. B.; Furst, L.; Condie, A. G.; Stephenson, C. R. J. Org. Lett. 2012, 14,
■
* Supporting Information
S
9
4−97. (d) Wang, Z. Q.; Hu, M.; Huang, X. C.; Gong, L. B.; Xie, Y.
1
13
Experimental procedures, characterization data, and H and C
NMR spectra for all products. This material is available free of
charge via the Internet at http://pubs.acs.org.
X.; Li, J. H. J. Org. Chem. 2012, 77, 8705−8711. (e) Zhu, S.; Rueping,
M. Chem. Commun. 2012, 48, 11960−11962.
(12) (a) Gu, X.; Li, X.; Chai, Y.; Yang, Q.; Li, P.; Yao, Y. Green Chem.
2
013, 15, 357−361. (b) Liu, D.; Zhou, H.; Gu, X.; Shen, X.; Li, P.
Chin. J. Chem. 2014, 32, 117−122.
13) Cai, S.; Zhao, X.; Wang, X.; Liu, Q.; Li, Z.; Wang, D. Z. Angew.
Chem., Int. Ed. 2012, 51, 8050−8053.
14) Jiang, G.; Chen, J.; Huang, J.-S.; Che, C.-M. Org. Lett. 2009, 11,
4568−4571.
AUTHOR INFORMATION
Corresponding Author
Email: lipixu@suda.edu.cn.
■
(
*
(
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support is provided by NSFC (21102097), the
Scientific Research Foundation for the Returned Overseas
Chinese Scholars, State Education Ministry, and Beijing
National Laboratory for Molecular Sciences (BNLMS).
REFERENCES
■
(
1
1) (a) Vilsmeier, A.; Haack, A. Ber. Dtsch. Chem. Ges. 1927, 60, 119−
22. (b) Seshadri, S. J. Sci. Ind. Res. 1973, 32, 128−149. (c) Tasneem.
Synlett 2003, 138−139.
(
2) (a) Reimer, K. Ber. Dtsch. Chem. Ges. 1876, 9, 423−424.
(
(
1
b) Reimer, K.; Tiemann, F. Ber. Dtsch. Chem. Ges. 1876, 9, 824−828.
c) Reimer, K.; Tiemann, F. Ber. Dtsch. Chem. Ges. 1876, 9, 1268−
278. (d) Wynberg, H. Chem. Rev. 1960, 60, 169−184. (e) Wynberg,
H.; Meijer, E. W. The Reimer−Tiemann Reaction. Organic Reactions;
John Wiley and Sons: New York, 1982; Vol. 28; pp 1−36
(
(
3) Wu, W.; Su, W. J. Am. Chem. Soc. 2011, 133, 11924−11927.
4) (a) Chen, J.; Liu, B.; Liu, D.; Liu, S.; Cheng, J. Adv. Synth. Catal.
2
012, 354, 2438−2442. (b) Zhang, L.; Peng, C.; Zhao, D.; Wang, Y.;
Fu, H. J.; Shen, Q.; Li, J. X. Chem. Commun. 2012, 48, 5928−5930.
c) Zhao, D.; Wang, Y.; Zhu, M. X.; Shen, Q.; Zhang, L.; Du, Y.; Li, J.
X. RSC Adv. 2013, 3, 10272−10276.
5) Chiba, S.; Wang, Y.-F.; Zhang, F.-L. Synthesis 2012, 44, 1526−
534.
6) (a) Li, L.-T.; Huang, J.; Li, H.-Y.; Wen, L.-J.; Wang, P.; Wang, B.
(
(
1
(
Chem. Commun. 2012, 48, 5187−5189. (b) Li, L.-T.; Li, H.-Y.; Xing,
L.-J.; Wen, L.-J.; Wang, P.; Wang, B. Org. Biomol. Chem. 2012, 10,
9
519−9522. (c) Fei, H.; Yu, J.; Jiang, Y.; Guo, H.; Cheng, J. Org.
Biomol. Chem. 2013, 11, 7092−7095.
7) For selected reports on organic-dye-promoted photocatalytic
(
reactions, see: (a) Liu, H.; Feng, W.; Kee, C. W.; Zhao, Y.; Leow, D.;
Pan, Y.; Tan, C.-H. Green Chem. 2010, 12, 953−956. (b) Hari, D. P.;
Ko
S.; Ko
d) Pan, Y. H.; Kee, C. W.; Chen, L.; Tan, C. H. Green Chem. 2011,
3, 2682−2685. (e) Hari, D. P.; Schroll, P.; Koenig, B. J. Am. Chem.
Soc. 2012, 134, 2958−2961.
8) (a) Maestri, M.; Gratzel, M. Ber. Bunsen-Ges. Phys. Chem. 1977,
1, 504−507. (b) Bock, C. R.; Connor, J. A.; Gutierrez, A. R.; Meyer,
T. J.; Whitten, D. G.; Sullivan, B. P.; Nagle, J. K. J. Am. Chem. Soc.
979, 101, 4815−4824.
9) (a) Yoon, T. P.; Ischay, M. A.; Du, J. N. Nat. Chem. 2010, 2,
27−532. (b) Narayanam, J. M. R.; Stephenson, C. R. J. Chem. Soc.
Rev. 2011, 40, 102−113. (c) Teply, F. Collect. Czech. Chem. Commun.
̈
̈
nig, B. Org. Lett. 2011, 13, 3852−3855. (c) Neumann, M.; Fuldner,
̈
nig, B.; Zeitler, K. Angew. Chem., Int. Ed. 2011, 50, 951−954.
(
1
(
8
1
(
5
́
2
4
7
6
(
1
011, 76, 859−917. (d) Ravelli, D.; Fagnoni, M. ChemCatChem 2012,
, 169−171. (e) Shi, L.; Xia, W. J. Chem. Soc. Rev. 2012, 41, 7687−
697. (f) Xuan, J.; Xiao, W.-J. Angew. Chem., Int. Ed. 2012, 51, 6828−
838. (g) Fukuzumi, S.; Ohkubo, K. Chem. Sci. 2013, 4, 561−574.
h) Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Chem. Rev. 2013,
13, 5322−5363. (i) Ravelli, D.; Fagnoni, M.; Albini, A. Chem. Soc.
Rev. 2013, 42, 97−113.
(
(
10) Li, Z. P.; Li, C. J. J. Am. Chem. Soc. 2005, 127, 6968−6969.
11) (a) Chen, M.; Huang, Z. T.; Zheng, Q. Y. Chem. Commun. 2012,
48, 11686−11688. (b) Dai, C.; Meschini, F.; Narayanam, J. M. R.;
1
900
dx.doi.org/10.1021/cs5005129 | ACS Catal. 2014, 4, 1897−1900