Visible-Light-Triggered C-C and C-N Bond Formation by C-S Bond Cleavage of Benzylic Thioethers

Article abstract of DOI:10.1021/acs.orglett.8b02196

The cleavage of sulfidic C-S bonds under visible-light irradiation was harnessed to generate carbocations under neutral conditions and synthesize valuable di- and triarylalkanes as well as benzyl amines. To this end, photoredox catalysis and direct photoinduced C-S bond cleavage are used as complementary approaches and participate in the versatility of the general strategy. Extensive mechanistic studies have demonstrated the diversity of the reaction mechanism at work in these different reactions.

Full text of DOI:10.1021/acs.orglett.8b02196

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Visible-Light-Triggered C−C and C−N Bond Formation by C−S Bond
Cleavage of Benzylic Thioethers
†,∥
§
Matteo Lanzi,†,‡,∥ Jeremy Merad, Dina V. Boyarskaya,† Giovanni Maestri,‡ Clemence Allain,
́
́
́
,†
́
and Geraldine Masson*
†
́
́
Institut de Chimie des Substances Naturelles, CNRS UPR 2301, Universite Paris-Sud, Universite Paris-Saclay, 1, av. de la Terrasse,
91198 Gif-sur-Yvette Cedex, France
‡
̀
Universita di Parma, Dipartimento SCVSA, 17/A Parco Area delle Scienze, 43124 Parma, Italy
§
́
PPSM, ENS Cachan, CNRS, Universite Paris-Saclay, 94235 Cachan, France
S
* Supporting Information
ABSTRACT: The cleavage of sulfidic C−S bonds under
visible-light irradiation was harnessed to generate carbocations
under neutral conditions and synthesize valuable di- and
triarylalkanes as well as benzyl amines. To this end,
photoredox catalysis and direct photoinduced C−S bond
cleavage are used as complementary approaches and
participate in the versatility of the general strategy. Extensive mechanistic studies have demonstrated the diversity of the
reaction mechanism at work in these different reactions.
ulfides are ubiquitous and simple functional groups
prevalent in numerous pharmaceuticals and natural products.10
As a complementary approach to the existing methods,11 we
envisioned the photocatalyzed oxidative C−S bond cleavage of
benzyl thioethers12 as a potential soft route to access such
molecules (Scheme 1). The generation of radical cation 1 by
S
constitutive of many bioactive molecules1 and commer-
cially available compounds. Synthetic sequences involving the
cleavage of sulfidic C−S bonds followed by the new formation
of C−C or C−heteroatom bonds are then anticipated as highly
suitable to achieve last-stage transformations or increase the
molecular complexity. Moreover, the sulfur atom is hardly
electronegative and the C−S bond is poorly polarized, making
thioethers insensitive to soft Brønsted acids and bases. As a
consequence, C−S bond cleavages can constitute highly
chemoselective transformations. Great achievements have
been made to perform the catalytic activation of sulfidic C−
S bonds relying on transition metal complexes2 and Lewis
acids.3 However, most of these approaches display a narrow
scope due to the ability of the Lewis basic sulfur atom to
poison electron-deficient catalysts.4
Scheme 1. Work Hypothesis
In the meantime, the selective oxidation of thioethers is
efficiently ensured by photoredox catalysis since it does not
proceed by coordination of the sulfur atom to the catalyst.
Even if this approach has been extensively studied to
synthesize sulfoxides,5 it remains a surprisingly underexplored
strategy concerning the fragmentation of thioethers.6 Only in
2013, Bowers and co-workers made a proof of this concept
with the first example of photocatalyzed glycosylation of
thioglycosides induced by selective oxidation of a thioether.7
More recently, our group reported an efficient synthesis of α-
substituted amides by photoredox-catalyzed C−S bond
cleavages of α-amidosulfides.8 Despite these few examples,9
the synthetic potential of the visible light photoredox catalyzed
fragmentation of thioethers remains surprisingly underex-
plored.
single electron oxidation of the substrates should, in principle,
lead to the formation of benzylic carbocation 2. This
intermediate could then undergo the addition of various
nucleophiles under neutral conditions. We report herein the
achievement of a photoredox-catalyzed synthesis of di- and
triarylalkanes. Additionally, we set up a complementary
catalyst-free process in which the C−S bond cleavage was
induced by simple visible-light irradiation of the substrate.
On the basis of our previous work,8 we initially attempted to
functionalize 4-ethylthio(phenylmethyl)phenol 3a with 1,3,5-
trimethoxybenzene 4a in the presence of photocatalyst 5 in
acetonitrile/t-BuOH (8:2, v/v) under visible light irradiation
In the course of a research program, we became interested in
the synthesis of di- and triarylalkanes since these moieties are
Received: July 13, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b02196
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
(5 W blue LEDs) with molecular oxygen as an oxidant. To our
delight, we were able to isolate the desired triarylmethane
derivative 6a in 81% yield after 24 h (Table 1, entry 1). The
Scheme 2. Substrates Scope of the Desulfitative Arylation
Table 1. Optimization of the Desulfitative Arylation
a
Process
b
entry
3
atmosphere (1 atm)
x equiv
yield (%)
1
2
3
4
5
6
7
8
9
10
3a
3b
3c
3d
3d
3d
3d
3d
3d
3d
O2
O2
O2
O2
O2
O2
O2
air
10
10
10
10
0
0
0
0
0
84
71
56
99
94
c
94
0
d
83
14
11
argon
O2
e
0
a
General conditions: 3 (0.10 mmol, 1 equiv), 4a (0.15 mmol), 5
(0.025 equiv), t-BuOH (0 or 10 equiv) in CH3CN (0.5 mL)
b
irradiated with 5 W blue LEDs at rt for 24 h. Isolated yields.
c
d
Reaction performed on a 1 mmol scale (of 3d). Reaction performed
e
in the dark. Reaction performed without catalyst.
nature of the thioether moiety has demonstrated a crucial
influence on the reaction outcome. Indeed, when R is a tert-
butyl (3b) or a phenyl (3c) group, the substitution efficiency
was significantly decreased (entries 2−3). However, S-benzyl
derivative 3d afforded 6a in a nearly quantitative yield (entry
4). Unlike our previous study,8 the presence of t-BuOH does
not have a prominent effect when 3d and 4a constitute the
reaction partners (entry 5). Pleasingly, the reaction can be
performed on a 1 mmol scale without any change in rates or
yields (entry 6). Control experiments showed that the reaction
failed to proceed in the absence of light (entry 7). A slower
substitution was observed in an open-air flask (entry 8),
highlighting the positive effect of O2 on the reaction rate. This
observation was supported by a dramatic conversion drop
when the arylation was performed under an argon atmosphere
(entry 9). Then, we found that when the flask was irradiated
for 24 h in the absence of catalyst, 6a was isolated in 11% yield
(entry 10). This striking observation suggested a potential,
although poorly efficient, direct photoactivation of substrate 3d
under blue LED irradiation.13
With the optimized reaction conditions, the scope of the
photocatalyzed arylation of benzylthioethers with 4a was then
explored (Scheme 2a). Diarylmethanes bearing strong
donating (+M) substituents provided triarylmethanes 6a−6d
with good to excellent yields. In contrast, products 6e and 6f
were isolated in 12% and 1% yields, respectively. For this last
product the hydroxyl in the meta position does not participate
in the carbocation stabilization and only −I destabilizing effect
is operating. In the synthesis of 6e and 6f, sulfoxides arising
from the oxygenation of the radical cation intermediate 1 were
identified as the main reaction products (see Supporting
Information (SI)). A similar trend was brought to light for
anisole derivatives, since the photocatalytic arylation was found
a
b
Isolated yields. Temperature measured in a reaction tube at 3 cm of
c
d
1
the bulb. Yields determined by H NMR. Reaction performed over
48 h.
efficient for electron-donating aromatic substituents (6g−j).
On the other hand, the presence of electron-withdrawing
groups as m-F and p-CF3 resulted in the formation of products
6k and 6l in low yields. The reaction proceeded smoothly
when combining highly electron-rich phenols with electron-
poor arenes (6m). Heteroaromatic substituents such as the 3-
indolyl or 3-quinolinonyl group were also competent reaction
partners, delivering products (6n−r) that have potential
applications in medicinal chemistry. Extension of this novel
photocatalyzed protocol to diverse secondary benzylic
thioethers was also successful providing efficient access to
biologically relevant diarylalkanes.10 The desulfitative arylation
can be efficiently applied to propargylic thioethers as well as
cyclic and acyclic benzylic substrates 6s−w. Additionally, this
transformation provides an efficient route to anticancer
B
DOI: 10.1021/acs.orglett.8b02196
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
isoerianin analogue 6x.14 It has to be noticed that the presence
of t-BuOH can have a slight impact on the reaction efficiency
without us being able to establish a clear relationship with the
structure of the substrate.
Encouraged by these results, we turned our efforts toward
extending this protocol to other nucleophiles (Scheme 2b).
Diversely substituted arenes and heteroarenes readily partici-
pated in C−C bond formation to afford the expected
triarylmethanes 7a−7h in good yields. Moreover, the
construction of C−N bonds can also be achieved thanks to
this method. Various azole derivatives were successfully
incorporated giving rise to important drugs building blocks
7i and 7j.15 Nonaromatic N-nucleophiles such as carbamates,
anilines, and azides were also found to be viable reagents (7k−
7n).
SI). In a final step, the addition of the nucleophile onto the
carbocation could be assisted by the basic superoxide radical
10.
Correlations between the Hammett parameters (σ) of
aromatic substituents and the corresponding yields obtained
after 24 h (see SI) have highlighted a drop in yields in the
presence of substituents with σ values ≥0. This could be
explained by a dramatic increase in the oxidation potentials of
the related substrates. As a matter of fact, oxidation potentials
of 3e (precursor of 6e) (E1/2 = +1.18 V vs Fc/Fc+ in CH3CN)
and 3k (precursor of 6k) (E1/2 = +1.14 V vs Fc/Fc+ in
CH3CN) are elusive for Ru(III) suggesting another reaction
mechanism for this work (Scheme 3, green square). In this case,
the sulfur oxidation (step a) could rely on the formation of
1
singlet oxygen O2 by energy transfer between Ru(II)* and
O2.18 In order to shatter this main scope limitation,
photocatalysts with higher oxidation potentials were tested in
the arylation of 3e without significant improvement of the
yields (see SI). On the other hand, these disappointing results
could be explained by poor stabilization of the carbocation
intermediate 2, which slows C−S bond cleavage favoring the
oxygenation of the radical cation 1. Unfortunately, the use of
alternative oxidants to prevent the formation of sulfoxide 11
uniformly resulted in no conversion. Additionally, our attempts
to improve the nucleofugality of thiyl radicals 8 (Scheme 3b,
compounds 3g and 3h) remained unsuccessful (see SI).
With the desire to set up a complementary strategy and
based on our previous observation that the benzylthioethers
participate in arylative C−S bond cleavage under catalyst-free
conditions (Table 1, entry 14), we hypothesized that an
appropriate choice of conditions would allow the formation of
triarylalkanes while preventing the competing overoxidation.
Even if 3e remained unreactive under 5 W blue LED
irradiation an optimization stage (see SI) revealed that the
use of a more intense 25 W CFL bulb and higher dilution
supported the formation of desired product 6e in 39% yield
(Scheme 2). The synthetic potential of this new set of
conditions was then explored. Electron-rich substrates under-
went the smooth addition of 4a although in lower yields
compared to the Ru-catalyzed process. However, the catalyst-
free substitution furnishes enhanced efficiency in the case of
more electron-deficient substrates. Fluorinated products 6k
and 6l were then obtained in 65% and 76% yields, respectively.
This highlights the complementarity between the two
methodologies described herein. Highly deactivated thioether
was converted to triarylmethane 6f in 10% yield only.
However, this remains 10 times higher than with the
photoredox approach. The photoinduced desulfidation also
turned out to be more suitable to access diarylalkyne 6s.
Various arenes and heteroarenes can be used as the reaction
partners, as demonstrated by the formation of compounds 7a,
7o, 7p in rather good yields. Even if the mechanism of this
transformation is not fully understood, it is expected to involve
a visible-light-induced homolytic cleavage of the C−S bond
(see SI).19,20
To gain insight into the mechanism of the photoredox-
catalyzed thioether C−S bond cleavage, we first determined
the redox potential of several thioethers. Model substrate 3d
displayed an oxidation potential E1/2 = +0.85 V vs Fc/Fc+ in
CH3CN (see SI) that makes a direct SET thermodynamically
unlikely from the excited stated of 5a (Ru(II)*/Ru(I): E0 =
+0.37 V vs Fc/Fc+ in CH3CN).16 This was confirmed by
Stern−Volmer experiments in which no significant fluores-
cence quenching was observed in the presence of 3d or
trimethoxybenzene 4a (see SI). Based on these observations
and the requirement of O2 for the reaction to proceed
efficiently, we assumed that Ru(III) (Ru(III)/Ru(II): E0 =
+0.89 V vs Fc/Fc+ in CH3CN) could be the active oxidant of
electron-rich substrates (Scheme 3, red square).
Scheme 3. Suggested Reaction Mechanisms
The mesolytic cleavage of the C−S bond in radical cation 1
was supported by several experimental proofs. The involve-
ment of a carbocation intermediate (SN1-type process) was
unambiguously demonstrated by the complete loss of
enantiomeric excess through the arylation of optically pure
thioether 3d (see SI). Also, disulfide 9 resulting from the
dimerization of thiyl radical 8a was isolated in 83% yield when
3d was reacted with 4a (Scheme 3).17 This fast dimerization
prevents radical propagation as demonstrated by kinetic
experiments involving alternation of irradiation and dark (see
To conclude, we have described an unprecedented visible-
light photoredox catalyzed synthesis of diarylalkane and
triarylmethane derivatives under neutral conditions. The
reported strategy relied on the oxidative weakening of sulfidic
C−S bonds allowing the in situ generation of the reactive
carbocations. An appreciable range of substrates and
nucleophiles can be engaged in this transformation that uses
O2 as a terminal oxidant. Extensive mechanistic studies allowed
C
DOI: 10.1021/acs.orglett.8b02196
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
̈
E.; Sikorski, M.; Konig, B.; Cibulka, R. ChemCatChem 2012, 4, 620.
(e) Gu, W.; Chai, Y.; Yang, Q.; Li, P.; Yao, Y. Green Chem. 2013, 15,
357. (f) Chao, D.; Zhao, M. ChemSusChem 2017, 10, 3358. (g) Ye,
C.; Zhang, Y.; Ding, A.; Hu, Y.; Guo, H. Sci. Rep. 2018, 8, 2205.
(6) Lee, M.; Neukirchen, S.; Cabrele, C.; Reiser, O. J. Pept. Sci. 2017,
23, 556.
us to understand the initial limitations in term of substrate
electronic properties. This work finally revealed the possibility
of inducing homolytic C−S bonds cleavage under direct
visible-light irradiation. This catalyst-free approach efficiently
complemented the first methodology since it ensures the
formation of C−C and C−N bonds from more electron-
deficient substrates. We hope this work will highlight the
synthetic potential of the low oxidation state sulfur photo-
activation.
(7) Wever, W. J.; Cinelli, M. A.; Bowers, A. A. Org. Lett. 2013, 15,
30.
(8) (a) Jarrige, L.; Levitre, G.; Masson, G. J. Org. Chem. 2016, 81,
7230. (b) Lebee, C.; Languet, M.; Allain, C.; Masson, G. Org. Lett.
́
2016, 18, 1478.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b02196.
(9) An example of photocatalyzed C−S bond cleavage from: Gao,
X.-F.; Du, J.-J.; Liu, Z.; Guo, J. Org. Lett. 2016, 18, 1166.
(10) For relevant reviews on this topic, see: (a) Ameen, D.; Snape,
T. J. MedChemComm 2013, 4, 893. (b) Mondal, S.; Panda, G. RSC
Adv. 2014, 4, 28317.
■
S
(11) For recent examples, see: (a) Peng, L.; Li, Y.; Li, Y.; Wang, W.;
Pang, H.; Yin, G. ACS Catal. 2018, 8, 310. (b) Moon, P. J.; Fahandej-
Sadi, A.; Qian, W.; Lundgren, R. J. Angew. Chem., Int. Ed. 2018, 57,
4612. (c) Poremba, K. E.; Kadunce, N. T.; Suzuki, N.; Cherney, A.
H.; Reisman, S. E. J. Am. Chem. Soc. 2017, 139, 5684.
(d) Vasilopoulos, A.; Zultanski, S. L.; Stahl, S. S. J. Am. Chem. Soc.
2017, 139, 7705. (e) Zhou, Q.; Cobb, K. M.; Tan, T.; Watson, M. P.
J. Am. Chem. Soc. 2016, 138, 12057. (f) Friis, S. D.; Pirnot, M. T.;
Buchwald, S. L. J. Am. Chem. Soc. 2016, 138, 8372. For selected
recent examples on preparation of triarylalkanes, see: (g) Kim, J. H.;
Greßies, S.; Boultadakis-Arapinis, M.; Daniliuc, C.; Glorius, F. ACS
Catal. 2016, 6, 7652. (h) Nambo, M.; Crudden, C. M. ACS Catal.
2015, 5, 4734. (i) Gomes, R. F. A.; Coelho, J. A. S.; Frade, R. F. M.;
Trindade, A. F.; Afonso, C. A. M. J. Org. Chem. 2015, 80, 10404.
(j) Matthew, S. C.; Glasspoole, B. W.; Eisenberger, P.; Crudden, C.
M. J. Am. Chem. Soc. 2014, 136, 5828.
Experimental procedures, isolated products character-
ization data, detailed mechanistic investigations (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: geraldine.masson@cnrs.fr.
ORCID
Giovanni Maestri: 0000-0002-0244-8605
Clemence Allain: 0000-0002-2908-264X
Geraldine Masson: 0000-0003-2333-7047
Author Contributions
∥M.L. and J.M. contributed equally.
Notes
́
́
(12) For simple syntheses of benzylic sulfides, see: Parnes, R.;
Pappo, D. Org. Lett. 2015, 17, 2924 and references therein.
(13) For a recent review on this topic, see: Liu, W.; Li, C.-J. Synlett
2017, 28, 2714.
The authors declare no competing financial interest.
́
(14) Messaoudi, S.; Hamze, A.; Provot, O.; Treguier, B.; Rodrigo De
ACKNOWLEDGMENTS
■
Losada, J.; Bignon, J.; Liu, J.-M.; Wdzieczak-Bakala, J.; Thoret, S.;
Dubois, J.; Brion, J.-D.; Alami, M. ChemMedChem 2011, 6, 488.
(15) (a) Song, C.; Dong, X.; Yi, H.; Chiang, C.-W.; Lei, A. ACS
Catal. 2018, 8, 2195−2199. (b) Yang, Y.; Lan, J.; You, J. Chem. Rev.
2017, 117, 8787. (c) Correa, A.; Cornella, J.; Martin, R. Angew. Chem.,
Int. Ed. 2013, 52, 1878. (d) Yan, W.; Wang, Q.; Chen, Y.; Petersen, J.
L.; Shi, X. Org. Lett. 2010, 12, 3308.
We thank ICSN for funding. M.L. and G.M. are thankful for
support from SIR Grant RBSI14NKFL, J.M. thanks the CNRS
for postdoctoral fellowships, and D.B. thanks Labex Charm3at
for financial support of Master.
REFERENCES
■
(16) Zhu, Q.; Gentry, E. C.; Knowles, R. R. Angew. Chem., Int. Ed.
(1) (a) Ilardi, E. A.; Vitaku, E.; Njardarson, J. T. J. Med. Chem. 2014,
57, 2832. (b) Feng, M.; Tang, B.; Liang, S. H.; Jiang, X. Curr. Top.
Med. Chem. 2016, 16, 1200.
2016, 55, 9969.
́
̀
(17) Denes, F.; Pichowicz, M.; Povie, G.; Renaud, P. Chem. Rev.
2014, 114, 2587.
(2) For recent reviews about transition-metal-catalyzed C−S bond
activation, see: (a) Modha, S. G.; Mehta, V. P.; Van der Eycken, E. V.
Chem. Soc. Rev. 2013, 42, 5042. (b) Pan, F.; Shi, Z.-J. ACS Catal.
2014, 4, 280. For recent examples, see: (c) Shibata, T.; Mitake, A.;
Akiyama, Y.; Kanyiva, K. S. Chem. Commun. 2017, 53, 9016.
(d) Uetake, Y.; Niwa, T.; Hosoya, T. Org. Lett. 2016, 18, 2758.
(e) Bhanuchandra, M.; Baralle, A.; Otsuka, S.; Nogi, K.; Yorimitsu,
H.; Osuka, A. Org. Lett. 2016, 18, 2966. (f) Gao, K.; Yorimitsu, H.;
Osuka, A. Angew. Chem., Int. Ed. 2016, 55, 4573.
(3) For a unique example using Lewis acid catalysis, see: (a) George,
N.; Bekkaye, M.; Alix, A.; Zhu, J.; Masson, G. Chem. - Eur. J. 2014, 20,
3621. For a unique example using Brønsted acid catalysis, see:
(b) George, N.; Bekkaye, M.; Masson, G.; Zhu, J. Eur. J. Org. Chem.
2011, 2011, 3695.
(18) For examples of SET arising from singlet oxygen, see:
(a) Foote, C. S.; Peters, J. W. J. Am. Chem. Soc. 1971, 93, 3795.
(b) Matsuura, T.; Yoshimura, N.; Nishinaga, A.; Saito, I. Tetrahedron
1972, 28, 4933. (c) . (d) Saito, I.; Matsuura, T.; Inoue, K. J. Am.
Chem. Soc. 1981, 103, 188.
(19) For hemolytic, see: (a) Bonesi, S. M.; Crespi, S.; Merli, D.;
Manet, I.; Albini, A. J. Org. Chem. 2017, 82, 9054. (b) D’Ischia, M.;
Testa, G.; Mascagna, D.; Napolitano, A. Gazz. Chim. Ital. 1995, 125,
315. (c) Sucholeiki, I. Tetrahedron Lett. 1994, 35, 7307. (d) Fleming,
S. A.; Jensen, A. W. J. Org. Chem. 1996, 61, 7040. (e) Fleming, S. A.;
Rawlins, D. B.; Samano, V.; Robins, M. J. J. Org. Chem. 1992, 57,
5968. (f) Fleming, S. A.; Rawlins, D. B.; Robins, M. J. Tetrahedron
Lett. 1990, 31, 4995.
(20) Photoinduced S−S bond cleavage: Deng, Y.; Wei, X.-J.; Wang,
(4) For selected examples, see: (a) Gao, N.; Zheng, S.; Yang, W.;
Zhao, X. Org. Lett. 2011, 13, 1514. (b) Zheng, S.; Huang, W.; Gao,
N.; Cui, R.; Zhang, M.; Zhao, X. Chem. Commun. 2011, 47, 6969.
(c) Lipshutz, B. H.; Ghorai, S. Org. Lett. 2009, 11, 705.
̈
H.; Sun, Y.; Noel, Y.; Wang, X. Angew. Chem., Int. Ed. 2017, 56, 832.
(5) For selected examples, see: (a) Baciocchi, E.; Crescenzi, C.;
Lanzalunga, O. Tetrahedron 1997, 53, 4469. (b) Fujita, S.; Kakegawa,
N.; Yamagishi, A. J. J. Phys. Chem. B 2006, 110, 2533. (c) Li, W.; Xie,
́
́
Z.; Jing, X. Catal. Commun. 2011, 16, 94. (d) Dad’ova, J.; Svobodova,
D
DOI: 10.1021/acs.orglett.8b02196
Org. Lett. XXXX, XXX, XXX−XXX