Visible-Light-Induced Synthesis of Carbazoles by in Situ Formation of Photosensitizing Intermediate

Article abstract of DOI:10.1021/acs.orglett.7b00681

A visible-light-induced synthesis of N-H carbazoles from easily accessible 2,2′-diaminobiaryls in the absence of any external photosensitizer is reported. The process only requires ^tBuONO and natural resources, visible light, and molecular oxygen for the synthesis of N-H carbazoles. Experimental and computational studies support that the in situ formation of a visible-light-absorbing photosensitizing intermediate, benzocinnoline N-imide, is responsible for the activation of triplet molecular oxygen to singlet oxygen that, in turn, promotes the synthesis of carbazole.

Full text of DOI:10.1021/acs.orglett.7b00681

Letter
pubs.acs.org/OrgLett
Visible-Light-Induced Synthesis of Carbazoles by in Situ Formation of
Photosensitizing Intermediate
Tanmay Chatterjee,† Geum-bee Roh,† Mahbubul Alam Shoaib,‡ Chang-Heon Suhl,§ Jun Soo Kim,*,‡
Cheon-Gyu Cho,*,§ and Eun Jin Cho*,†
†Department of Chemistry, Chung-Ang University, Seoul 06974, Republic of Korea
‡Department of Chemistry & Nanoscience, Ewha Womans University, Seoul 03760, Republic of Korea
§Department of Chemistry, Hanyang University, Seoul 04763, Republic of Korea
S
* Supporting Information
ABSTRACT: A visible-light-induced synthesis of N-H carbazoles from
easily accessible 2,2′-diaminobiaryls in the absence of any external
t
photosensitizer is reported. The process only requires BuONO and
natural resources, visible light, and molecular oxygen for the synthesis of N-
H carbazoles. Experimental and computational studies support that the in
situ formation of a visible-light-absorbing photosensitizing intermediate,
benzocinnoline N-imide, is responsible for the activation of triplet
molecular oxygen to singlet oxygen that, in turn, promotes the synthesis
of carbazole.
Scheme 1. Design of a New Synthetic Strategy for N-H
Carbazoles from 2,2′-Diamino-1,1′-biaryls
ver the past decade, various synthetic methods have been
Odeveloped that utilize visible light to drive organic
reactions.1 However, because of the inability of many organic
molecules to absorb visible light, the success of the development
ofvisible-light-induced synthesishaslargelyreliedontheuseofan
external photosensitizer (PS) that absorbs the visible-light energy
and utilizes it for the transformation of non-photoabsorbing
organic reactants into products, either through electron transfer
or energy transfer.1,2 In recent years, several external PS-free,
visible-light-induced synthetic methods have been developed by
utilizing the photochemical activity of various electron donor−
acceptor (EDA) complexes that can generate radical species for
further synthetic transformations by single-electron-transfer
(SET) processes.3 The Melchiorre group showed that in situ
formation of chiral enamines can directly act as a PS without
relying on EDAchemistry.4 TheGlorius5a and the Hong5b groups
demonstrated that the substrates and/or products of a chemical
reaction can also act as a PS in the visible-light induced processes.
In this study, we propose a visible-light-induced synthesis of
carbazole2a,6 utilizingeasilyaccessible2,2′-diaminobiaryls7,8 anda
nitrite source that occurs autonomously by in situ formation of a
photosensitizing intermediate without requiring any external PS.
Carbazoles are valuable heterocycles in many applications
because of their ubiquity in numerous bioactive natural products,
pharmaceutical agents, and diverse novel functional materials.9,10
The in situ diazotization of 2,2′-diaminobiphenyl (1a)
generates intermediate I, which undergoes intramolecular
cyclization to produce highly conjugated nitrogen-rich tricyclic
intermediates, N-H dibenzo[d,f ][1,2,3]triazepine (II), and
benzocinnoline N-imide (III), which remain in equilibrium by
rearrangement(Scheme1). Inthe1970s, theReesgroupreported
the formation of II and III from the reactions of 2,2′-
diaminobiphenyl (1a), and some other products were also
obtained under their conditions.11 We hypothesized that II and/
or III might work as a visible-light-absorbing photosensitizer to
sensitize triplet molecular oxygen to its higher energy singlet state
(1O2), which in turn could convert II to N-H carbazole (2a). The
design of such a protocol originated from the well-known fact that
highly conjugated and heteroatom (N,O,S)-containing organic
molecules such as porphyrins, rose bengal, eosin, and methylene
blue are very effective PSs for the activation of molecular oxygen
that, in turn, promotes chemical reactions.12
To our delight, the reaction of 2,2′-diaminobiphenyl 1a with
1.5 equiv of tBuONO in oxygen atmosphere under visible-light
[24 W compact fluorescent lamp (CFL)] irradiation produced N-
Received: March 7, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00681
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
H carbazole 2a without external PS. Various solvents, including
acetonitrile (MeCN), dimethylformamide (DMF), dimethyl
sulfoxide (DMSO), MeOH, 2,2,2-trifluoroethanol (TFE),
dichloromethane (DCM), and dichloroethane (DCE), were
screened; DCE showed the best result (Scheme 2). When the
formed intermediate obtained from 1a can possibly work as PS in
the presence of visible light and molecular oxygen.
To determine the PS of the reaction, first the time-dependent
density functional theory (TD-DFT) calculations of II and III
were performed.13 A functional B3LYP level of theory was used in
the calculations with a basis set of 6-311++g(d,p). The calculation
results indicate that III has a significant absorption in the visible
region with an absorption maximum at 376 nm whereas II did not
absorb light in the visible region (Figure 2a). Fortunately, the PS
Scheme 2. Preliminary Results of Carbazole Synthesis Using
ab
,
1a under Visible-Light Irradiation
a
b
Reaction scale: 0.1 mmol. Yields were determined by 1H NMR
using bromoform as the internal standard.
Figure 2. (a) Calculated absorption spectra of II and III using TD-DFT
with B3LYP/6-311++G(d,p) and (b) UV−vis spectra of III in DCE
(10−4 M).
reaction mixture was completely deoxygenated by repeated
vacuum−freeze−thaw cycles, no carbazole 2a was produced,
confirming the critical role of molecular oxygen in the reaction.
When the reaction was carried out in the dark, a trace amount of
2a was produced, confirming that the process is promoted by
visible light. The use of blue LEDs as the visible light source or the
use of CFL along with a 400 nm UV cutoff filter still worked but
decreased the yield slightly, indicating that the PS would absorb
light in the near-UV−vis region. Further optimization revealed
thatuseof 1.3equiv oftBuONOin0.1MDCEisoptimum forthis
reaction. Notably, the use of an external PS like fac-Ir(ppy)3, fac-
Ir(dFppy)3,[Ru(bpy)3]Cl2,andEosinYdidnotimprovetheyield
of 2a (for optimization details, see Table S1).
candidate III could be isolated under dark conditions and
1
characterized by H and 13C NMR spectroscopy and high-
resolution mass spectroscopy (HRMS). The UV−vis spectrum of
III (Figure 2b) showed a significant absorption in the visible
region with a maximum at 383 nm, similar to that of the reaction
mixture of 1a (Figure 1) and also consistent with the calculation
results.
In addition, several experiments further confirmed that III is
the intermediate of the reaction and works as a PS for the process.
First, the isolated III was converted to N-H carbazole (2a) under
oxygen atmosphere and visible-light irradiation, supporting that
IIIisindeedoneoftheintermediates(Scheme3a).14 Next, IIIwas
To prove that the reaction is promoted by the in situ generated
photosensitizing intermediate, the UV−vis spectra of the reaction
mixture using 1a as the substrate were recorded at different times
(Figure 1). A new broad peak appeared in the visible region above
Scheme 3. Additional Experiments To Prove that III Is the
Photosensitizing Intermediate of the Process
subjected to the reaction of benzyl alcohol 3. It is known that
singlet oxygen generated by a PS under visible-light irradiation
can lead to the oxidation of 3 to benzaldehyde 4.15 In the reaction
of 3 with 5 mol % of III and molecular oxygen under visible-light
irradiation (Scheme 3b), 4 was formed despite a low yield,
indicating that III can act as a PS. The reaction of 3 did not
proceed in the absence of either III or O2.
Figure 1. UV−vis spectra of 1a, 2a, and the reaction mixture of 1a were
recorded at different time intervals (20 min, 4, 8, and 20 h) in DCE (10−4
M). Inset: plot of time-dependent change in absorption at 400 and 420
nm.
To determine whether the reaction of 1a to 2a proceeds
through a free-radical pathway, additional experiments were
performed in the presence of radical scavengers including
TEMPO and 1,4-dinitrobenzene. The reactions were found to
be unaffected in terms of yield, thus supporting a concerted
mechanismforthesynthesisofcarbazolethroughN−NandC−N
bond cleavages in II, followed by the new C−N bond formation.
Based on these results, we propose a reaction mechanism as
outlined in Scheme 4. Among the proposed intermediates II and
III, III was found to be the PS from the combined experimental
400 nm, with a maximum at 360 nm after 20 min of reaction.
Then, the intensity of the peak decreased, whereas the intensity of
the corresponding N-H carbazole 2a peak increased with reaction
time, indicating the transient formation of a visible-light-
absorbing intermediate during the reaction and its conversion
to 2a. The formation of photoabsorbing intermediate is clearly
shown in the inset of Figure 1 from the absorbance values at 400
and420nminthevisiblerange. Theresultsindicatethattheinsitu
B
DOI: 10.1021/acs.orglett.7b00681
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. Proposed Reaction Mechanism
Scheme 5. Substrate Scope of the Synthesis of N-H
ab
,
Carbazoles
and computational studies. Compound III is excited to its
1
corresponding high-energy singlet state [III] by absorbing
photons from visible light; subsequent intersystem crossing
(ISC) produces a relatively lower energy and long-lived triplet
state 3[III], which returns to its ground state by transferring its
energy to the triplet oxygen (3O2), generating high-energy singlet
oxygen (1O2). According to the literature reports,12b,16 1O2 can
react with II by two possible pathways, ene reaction (path A) and
[2 + 2]-cycloaddition (path B). The ene reaction produces a
reactive intermediate II-A which immediately decomposes to N-
hydroxycarbazole II-B17 in a concerted step through the
homolytic cleavage of O−O, N−N, and C−N bonds followed
by the formation of new C−N bond and hydrogen atom
abstraction from any H atom sources present in the reaction
medium. The concomitant liberation of molecular nitrogen
would be the driving force for the process. Finally, II-B
decomposes to more stable N-H carbazole 2a through homolytic
N−O bond cleavage followed by H atom abstraction. On the
a
b
c
d
Reaction scale: 0.3 mmol. Isolated yield. 1.0 mmol scale. Reaction
was performed at 80 °C.
substrate with a methyl substituent (para to the NH2 group), 5-
methyl-[1,1′-biphenyl]-2,2′-diamine (1j), produced 3-methyl-
9H-carbazole (2j) in 65% yield, the substrate with a methoxy
(metatotheNH2 group)substituent, 4-methoxy-[1,1′-biphenyl]-
2,2′-diamine (1k), produced 2-methoxy-9H-carbazole (2k) in
lowyield(26%). Thelowerreactivityof1kcanbeattributedtothe
stabilization of the corresponding diazonium salt by the
resonance effect of the methoxy group. Increasing the reaction
temperature to 80°C did not significantly improve theyield of 2k.
Similarly, the reaction of 4,4′-dimethoxy[1,1′-biphenyl]-2,2′-
diamine (1m) produced 2,7-dimethoxy-9H-carbazole (2m), an
alkaloid (clausine V), in a low yield (30%). A substrate containing
a heteroaromatic moiety, i.e., 2-(2-aminophenyl)pyridin-3-amine
(1n), also participated in the transformation to furnish δ-
carboline (2n), the core moiety in anti-infective agents and blue-
phosphorescent organic light-emitting diodes.19
This method was also used to synthesize N-H benzo[c]-
carbazoles (Scheme 6). Regardless of the electron density of the
substituents in the starting substrates (5a−d), the corresponding
N-H benzo[c]carbazoles (6a−d) were synthesized in moderate-
to-good yields (55−70%). On the other hand, the reaction of
[1,1′-binaphthalene]-2,2′-diamine (7) furnished 7H-dibenzo-
[c,g]carbazole (8) in a low yield (25%), probably because of the
1
other hand, in path B, [2 + 2]-cycloaddition of II with O2
produces highly reactive intermediate II-C that immediately
decomposes to 2a through the homolytic cleavage of O−O, N−
N, and C−N bonds followed by the formation of new C−N bond
along with the generation of nitric oxide (•NO) or dinitrogen
dioxide (N2O2) which subsequently oxidize to nitrogen dioxide.
Despitethefeasibilityofbothpathways, pathAismorelikelytobe
operative over path B because intermediate II-A is expected to
have lower energy than II-C according to the theoretical study of
1
the reaction of formal hydrazone with O2 done by the Greer
group.16
Next, the substrate scope of the synthesis of free N-H
carbazoles was investigated using various substituted 2,2′-
diamino-1,1′-biaryls (Scheme 5). Notably, this method using
free amino groups avoids the need for additional N-protection
and deprotection steps to synthesize free N-H carbazole
derivatives.2a,9c In general, 10−25% of the corresponding
benzo[c]cinnoline derivatives were also obtained as the side
product. Halogen (−Br, −Cl, −F)-substituted 2,2′-diamino-1,1′-
biaryls produced the corresponding carbazoles (2b−d). In
particular, 1-bromocarbazole 2b is a good precursor for the
synthesis of various host materials for phosphorescent organic
light-emitting diodes by further functionalization at the 1-
position.18 Substrates with various electron-withdrawing groups
(such as −Br, −Cl, −F, −OCF3, −COMe, −CO2Me, −CN, and
−NO2) smoothly underwent cyclization to furnish the desired
carbazoles (2b−i). Regarding the reactivity of substrates with
electron-donating substituents, although the reaction of a
Scheme 6. Substrate Scope of the Synthesis of N-H
Benzo[c]carbazoles
a b
,
a
b
Reaction scale: 0.3 mmol. Isolated yield.
C
DOI: 10.1021/acs.orglett.7b00681
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Escudero-Adan
́
, E. C.; Melchiorre, P. Angew. Chem., Int. Ed. 2015, 54,
steric effect when forming the key intermediate (see II in Scheme
1).
1485. (d)Sun, X.;Wang, W.;Li,Y.;Ma, J.;Yu, S. Org.Lett. 2016,18, 4638.
(e) Beatty, J. W.; Douglas, J. J.; Miller, R.; McAtee, R. C.; Cole, K. P.;
Stephenson, C. R. J. Chem. 2016, 1, 456. (f) For a recent review, see:
Lima, C. G. S.; Lima, T. M.; Duarte, M.; Jurberg, I. D.; Paixao, M. W. ACS
Catal. 2016, 6, 1389. (g)Deng, Y.; Wei, X.-J.;Wang, H.; Sun, Y.;Noel, T.;
Wang, X. Angew. Chem., Int. Ed. 2017, 56, 832.
(4) Silvi, M.; Arceo, E.; Jurberg, I. D.; Cassani, C.; Melchiorre, P. J. Am.
Chem. Soc. 2015, 137, 6120.
In conclusion, we developed an autonomous visible-light-
induced method for the synthesis of N-H carbazoles from easily
available 2,2′-diaminobiaryls through the in situ formation of a
visible-light-absorbing photosensitizing intermediate, benzocin-
noline N-imide. Combined experimental and theoretical studies
confirmed that in the presence of visible light benzocinnoline N-
imide activates triplet molecular oxygen to generate singlet
oxygen that plays akey role incarbazole synthesis by reacting with
the dibenzo[d,f ][1,2,3]triazepine intermediate. This simple
method has several advantages: (1) external PS-free protocol
utilizing natural resources such as visible light and molecular
oxygen, (2) direct synthesis of free N-H carbazoles without N-
protection/deprotection steps, (3) broad substrate scope with a
high functional group tolerance under mild reaction conditions,
and (4) no regioselectivity problem.
̃
̈
(5) (a) Sahoo, B.; Hopkinson, M. N.; Glorius, F. Angew. Chem., Int. Ed.
2015, 54, 15545. (b) Kim, I.; Min, M.; Kang, D.; Kim, K.; Hong, S. Org.
Lett. 2017, 19, 1394.
(6) For visible-light-induced synthesis of carbazoles, see: (a) Hernan-
dez-Perez, A. C.; Collins, S. K. Angew. Chem., Int. Ed. 2013, 52, 12696.
(b) Yuan, Z.-G.; Wang, Q.; Zheng, A.; Zhang, K.; Lu, L.-Q.; Tang, Z.;
Xiao, W.-J. Chem. Commun. 2016, 52, 5128. Also see ref 2a.
(7) For examples of synthetic accessibility of 2,2′-diamino-1,1′-biaryls,
see: (a) Lim, Y. − K.; Jung, J. − W.; Lee, H.; Cho, C. − G. J. Org. Chem.
2004, 69, 5778. (b) Schulz, L.; Enders, M.; Elsler, B.; Schollmeyer, D.;
Dyballa, K. M.; Franke, R.; Waldvogel, S. R. Angew. Chem., Int. Ed. 2017,
DOI: 10.1002/anie.20161261.
(8) Previously, the Yamato group utilized 2,2′-diamino-1,1′-biaryls for
the synthesis of carbazoles in the presence of perfluorinated sulfonic acid
resin (50 wt % Nafion-H) at high temperature (200 °C): Yamato, T.;
Hideshima, C.; Suehiro, K.; Tashiro, M.; Prakash, G. K. S.; Olah, G. A. J.
Org. Chem. 1991, 56, 6248.
(9) For examples of books and reviews on carbazoles, see:
(a) Chakraborty, D. P.; Roy, S. Progress in the Chemistry of Organic
Natural Products; Herz, W., Kirby, G. W., Steglich, W., Tamm, Ch., Eds.;
Springer-Verlag: New York, 1991; Vol. 57, p 72. (b) Schmidt, A. W.;
Reddy, K. R.; Knolker, H.-J. Chem. Rev. 2012, 112, 3193. (c) Roy, J.; Jana,
A. K.; Mal, D. Tetrahedron 2012, 68, 6099.
ASSOCIATED CONTENT
* Supporting Information
■
S
TheSupportingInformationisavailablefreeofchargeontheACS
Publications website at DOI: 10.1021/acs.orglett.7b00681.
DFT calculation details, additional experimental results,
experimental details, analytical data of the synthesized
compounds, and NMR spectra (PDF)
AUTHOR INFORMATION
Corresponding Authors
̈
■
(10) For examples to show application of carbazole derivatives in novel
functional materials, see: (a) Grazulevicius, J. V.; Strohriegl, P.;
Pielichowski, J.; Pielichowski, K. Prog. Polym. Sci. 2003, 28, 1297.
*E-mail: jkim@ewha.ac.kr.
*E-mail: ccho@hanyang.ac.kr.
*E-mail: ejcho@cau.ac.kr.
(b) Brunner, K.; van Dijken, A.; Borner, H.; Bastiaansen, J. J. A. M.;
̈
ORCID
Kiggen, N. M. M.; Langeveld, B. M. W. J. Am. Chem. Soc. 2004, 126, 6035.
(c) Liu, Z.; Guan, M.; Bian, Z.; Nie, D.; Gong, Z.; Li, Z.; Huang, C. Adv.
Funct. Mater. 2006, 16, 1441.
Jun Soo Kim: 0000-0001-8113-0605
Eun Jin Cho: 0000-0002-6607-1851
Notes
(11) (a) Gait, S. F.; Rees, C. W.; Storr, R. C. J. Chem. Soc. D 1971, 1545.
(b) Gait, S. F.; Peek, M. E.; Rees, C. W.; Storr, R. C. J. Chem. Soc., Chem.
Commun. 1972, 982.(c)Gait, S.F.;Peek, M. E.;Rees, C. W.;Storr, R.C. J.
Chem. Soc., Perkin Trans. 1 1974, 1248. (d) Gait, S. F.; Peek, M. E.; Rees,
C. W.; Storr, R. S. J. Chem. Soc., Perkin Trans. 1 1975, 19.
(12) For examples of reviews on singlet oxygen, see: (a) DeRosa, M. C.;
Crutchley, R. J. Coord. Chem. Rev. 2002, 233−234, 351. (b) Ghogare, A.
A.; Greer, A. Chem. Rev. 2016, 116, 9994.
(13) For the details of DFT calculations, see Figures S1−S4 and Tables
S2−S4 in the SI. The DFT calculation was also performed for
intermediate I, but the spectrum of I showed weak absorption in the
visible region, which is negligible; see Figure S2.
(14) Along with N-H carbazole, 24% of benzo[c]cinnoline was also
formed from III through N−N bond cleavage by photolysis.
(15) (a) Lang, X.; Chen, X.; Zhao, J. Chem. Soc. Rev. 2014, 43, 473.
(b) Mehrabi-Kalajahi, S. S.; Hajimohammadi, M.; Safari, N. J. Iran. Chem.
Soc. 2016, 13, 1069.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Thisworkwas supported bytheNational Research Foundation of
Korea (NRF-2014R1A1A1A05003274, 2014R1A5A1011165,
and 2012M3A7B4049657).
REFERENCES
■
(1) For some recent reviews on visible-light-induced photocatalysis,
see: (a) Yoon, T. P.; Ischay, M. A.; Du, J. Nat. Chem. 2010, 2, 527.
(b) Narayanam, J. M. R.; Stephenson, C. R. J. Chem. Soc. Rev. 2011, 40,
102. (c) Prier, C. K.;Rankic, D. A.;MacMillan, D. W. C. Chem. Rev. 2013,
113,5322. (d)Koike, T.;Akita, M. Top.Catal. 2014,57,967.(e)Romero,
N. A.; Nicewicz, D. A. Chem. Rev. 2016, 116, 10075. (f) Cambie,
́
D.;
́
(16) Rudshteyn, B.; Castillo, A.; Ghogare, A. A.; Liebman, J. F.; Greer,
A. Photochem. Photobiol. 2014, 90, 431.
Bottecchia, C.; Straathof, N. J. W.; Hessel, V.; Noel, T. Chem. Rev. 2016,
̈
116, 10276.
(17) II-B cannot absorb light in the visible region, which excludes its
roleasaPSforthisreaction;see:Buxton, P.C.;Heaney, H.;Mason,K. G.;
Sketchley, J. M. J. Chem. Soc., Perkin Trans. 1 1974, 2695.
(18) Kim, S. M.; Byeon, S. Y.; Hwang, S.-H.; Lee, J. Y. Chem. Commun.
2015, 51, 10672.
(19) (a) Mazu, T. K.; Etukala, J. R.; Jacob, M. R.; Khan, S. I.; Walker, L.
A.;Ablordeppey, S.Y. Eur.J. Med.Chem. 2011,46,2378. (b)Im, Y.;Lee, J.
Y. Synth. Met. 2015, 209, 24.
(2) Forexamples ofourrecent workson visiblelightphotocatalysis, see:
(a) Choi, S.; Chatterjee, T.; Choi, W. J.; You, Y.; Cho, E. J. ACS Catal.
2015,5, 4796. (b)Chatterjee, T.;Choi, M.G.;Kim, J.;Chang, S.-K.;Cho,
E. J. Chem. Commun. 2016, 52, 4203. (c) Choi, Y.; Yu, C.; Kim, J. S.; Cho,
E. J. Org. Lett. 2016, 18, 3246. (d) Chatterjee, T.; Iqbal, N.; You, Y.; Cho,
E. J. Acc. Chem. Res. 2016, 49, 2284.
(3) For examples of visible-light-induced EDA chemistry and their
application in synthetic organic chemistry, see: (a) Rosokha, S. V.; Kochi,
J. K. Acc. Chem. Res. 2008, 41, 641. (b) Davies, J.; Booth, S. G.; Essafi, S.;
Dryfe, R. A. W.; Leonori, D. Angew. Chem., Int. Ed. 2015, 54, 14017.
(c) Kandukuri, S. R.; Bahamonde, A.; Chatterjee, I.; Jurberg, I. D.;
D
DOI: 10.1021/acs.orglett.7b00681
Org. Lett. XXXX, XXX, XXX−XXX