Radical Hydrosilylation of Alkynes Catalyzed by Eosin y and Thiol under Visible Light Irradiation

Article abstract of DOI:10.1021/acs.orglett.8b00909

A visible light-promoted hydrosilylation of alkynes has been explored and achieved using 1 mol % organic dye Eosin Y as the photocatalyst and a catalytic amount of thiol as the radical quencher. The corresponding alkenylsilanes were provided with high reg

Full text of DOI:10.1021/acs.orglett.8b00909

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Radical Hydrosilylation of Alkynes Catalyzed by Eosin Y and Thiol
under Visible Light Irradiation
Jing Zhu,† Wei-Chen Cui,† Shaozhong Wang,* and Zhu-Jun Yao*
State Key Laboratory of Coordination Chemistry, Jiangsu Key Laboratory of Advanced Organic Materials, School of Chemistry and
Chemical Engineering, Nanjing University, 163 Xianlin Avenue, Nanjing, Jiangsu 210023, China
S
* Supporting Information
ABSTRACT: A visible light-promoted hydrosilylation of
alkynes has been explored and achieved using 1 mol %
organic dye Eosin Y as the photocatalyst and a catalytic
amount of thiol as the radical quencher. The corresponding
alkenylsilanes were provided with high regio- and stereo-
selectivites in the reactions of various terminal and internal
alkynes. The experimental evidence shows that the reaction is
preferentially initiated by a single electron transfer process, and a photoredox pathway is suggested.
lkenylsilanes are widely applied as versatile silicon-
containing building blocks in organic synthesis and
connection with our previous investigations on organic dye
photocatalysis,8 we herein wish to report a visible light-
promoted transition-metal-free hydrosilylation of alkynes using
Eosin Y as the photocatalyst and a catalytic amount of thiol as
the radical quencher.
A
organosilicon polymer synthesis.1 Consequently, a considerable
number of protocols have been developed to access this type of
silicon molecule. Among them, hydrosilylation of alkynes is the
most favorable and straightforward one with perfect atom
economy.2 In recent decades, intensive efforts have been
devoted to the regio- and stereoselective hydrosilylation of
mono- and disubstituted alkynes by tuning transition-metal
complexes and coordinating ligands.3 In contrast, the radical
hydrosilylation of alkynes has received less attention.4
A survey of bond-dissociation energies and oxidation
potentials of commercially available hydrosilanes confirmed
that the combined use of tris(trimethylsilyl)silane (TTMS)
(E1/2Ox = 0.73 V vs SCE) and Eosin Y (E1/2Red = 0.83 V vs SCE
for excited state of Eosin Y) may enable the oxidative scission
of a Si−H bond and controllable generation of silyl radicals
under proper visible-light activation conditions.9−12 First, the
hydrosilylation of but-3-ynyl acetate was examined under
irradiation of 10 W white LEDs with 1 mol % of Eosin Y.
To our delight, the reaction proceeded in a highly regio- and
stereoselective manner in a mixed solvent of 1,4-dioxane and
water (100:1, v/v), affording anti-Markovnikov Z-alkenylsilane
1 as a major product (Table 1, entry 1). The yield was further
improved by introducing a catalytic amount of basic additives
(Table 1, entries 2−4). Thiol reagents were attempted as the
hydrogen donors (Table 1, entries 5−8). The combination of
triisopropylsilanethiol and potassium carbonate resulted in
almost quantitative yield and excellent regio- and stereo-
selectivity. Our experiments also revealed that (1) no
hydrosilylation occurred in the absence of light; (2) poor
conversion was observed when silanes (triphenylsilane and
dimethylphenylsialne) with high oxidation potentials were used,
instead of tris(trimethylsilyl)silane; (3) replacement of Eosin Y
with other organic photocatalysts afforded inferior results; and
(4) photocatalyst, thiol, and base additives are all essential for
this hydrosilylation (for details, see Table S1 in the Supporting
Information (SI)).
Hydrogen atom transfer (HAT) was usually employed as a
common strategy in silyl radical additions of carbon−carbon
multiple bonds. An additional carbon- or heteroatom-centered
radical is essential to ensure the formation of silyl radical(s).4−6
Due to the hydridic property of the Si−H bond, electrophilic
t
radicals such as BuO· or RS· were preferred over nucleophilic
ones. However, these reactions suffered from use of peroxidants
and azo-type compounds as radical initiators. Very recently, a
variety of radicals could be generated through visible light
photocatalysis in a controllable catalytic manner, involving
either a single electron transfer (SET) or an energy transfer
process.7 Hence, it may be likely to access a silyl radical through
an oxidative pathway based on a SET between an oxidizable
hydrosilane and an excited state of photocatalyst. This may also
enable a unified radical hydrosilylation for alkynes with different
electronic properties and substitution patterns (Scheme 1). In
Scheme 1. Proposal on Radical Hydrosilylation of Alkynes
Received: March 20, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b00909
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
a
Table 1. Optimization of Reaction Conditions
Scheme 2. 1,2-Disubstituted Alkenylsilanes
additive
b
c
entry
base
thiol
yield (%)
Z/E ratio
1
2
3
4
5
6
7
8
−
−
−
−
−
I
72
78
74
86
63
78
79
99
18/1
14/1
14/1
16/1
45/1
7/1
DIPEA
Na2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
II
III
IV
14/1
56/1
a
Reaction conditions: alkyne (0.3 mmol), silane (0.36 mmol), EY (1
mol %), dioxane/H2O (100/1 v/v, 3.0 mL), 10 W white LEDs. The
b
amount of mentioned thiol additive or base is 5 mol %. NMR yield
c
using 1,3,5-trimethoxybenzene as an internal standard. The Z/E ratios
1
were determined by H NMR analysis of the crude products.
The scope of this transition-metal-free hydrosilylation was
then explored under the optimized conditions, including
electron-neutral, -deficient, and -rich terminal alkynes (Scheme
2). All these reactions worked well, delivering 1,2-disubstituted
alkenylsilanes 1−22 in good to excellent yields with variable Z/
E selectivities. A variety of functional groups, including acetate,
alkyl chloride, alcohol, N-Boc protected amine, benzaldehyde,
fluorobenzene, pyridine, thiophene, acetal, carboxylate, phe-
noxy, phenylthio, and diphenylphosphino were tolerated. The
substitution patterns on propargylic carbon were found to affect
the distribution of two stereoisomers. Terminal alkynes with
primary alkyl substituents gave Z-alkenylsilanes 1−4 predom-
inantly. However, in the case of secondary and tertiary alkyl-
substituted alkynes, the proportion of Z-adduct was reduced to
a certain extent. A hydroxyl group at the propargylic position
may cause a decreased stereoselectivity, providing alkenylsilanes
5 and 8 in a Z/E ratio lower than 15/1. It is noteworthy that Z-
alkenylsilanes 7−8 with a tertiary allylic carbon would isomerize
to the corresponding thermodynamically more stable E-
stereoisomers during silica gel column purification, affording
the products with an inverse Z/E ratio. Similar inversions or
partial erosions of Z/E ratio were also observed in aromatic and
heteroaromatic alkenylsilanes 9−16. It seems likely that
electron-withdrawing groups (acetal, ester) on the alkyne
terminus make Z-configuration isomers stable, and alkenylsi-
lanes 17−19 could be isolated without any decrease of the Z/E
ratio. When a diyne substrate was attempted, an electron-
deficient alkyne is preferred over the electron-neutral one.
Furthermore, the heteroatoms attached on the alkyne terminus
affected the reactivity and stereoselectivity of the hydro-
silylation, in which a longer reaction time was needed for Z-
alkenylsilane 20 and the formation of Z-alkenylsilanes 21−22
became unfavorable due to the increasing steric repulsion
a
The hydrosilylation was performed as entry 8 of Table 1. The yields
and Z/E ratios out of parentheses correspond to the isolated products,
while the Z/E ratios of crude products are shown inside parentheses.
b
c
0.3 mmol TTMSS was used. The reaction time is 24 h.
between a heteroatom-containing group and a bulky tris-
(trimethylsilyl)silyl radical.
Next, the regio- and stereoselectivity of this transformation
was further examined upon a range of unsymmetrical internal
alkynes (Scheme 3). All the reactions of aromatic internal
alkynes provided excellent regioselectivities and good yields,
affording only one type of alkenylsilane regioisomers 23−26
with a tris(trimethylsilyl)silyl substituent locating at the β-
position toward the aromatic group. However, the stereo-
selectivities of the products derived from aromatic internal
alkynes were poor, according to their Z/E ratios. In contrast, an
α-addition of a silyl radical to electron-deficient alkynes was
observed, delivering alkenylsilanes 27−31 preferentially in a
stereoselective manner.13 The α/β selectivity varies, depending
on the substituents (alkyl, aryl, trimethylsilyl) on the terminus
of electron-deficient alkynes. Poor regioselectivity was observed
in the presence of a methyl substituent. When primary,
secondary alkyl substituents and a phenyl group are appended,
the α/β selectivity increases. Perfect α/β selectivity was
achieved in the case of a trimethylsilyl substituted electron-
deficient alkyne. In addition, the terminal phenoxy group led to
B
DOI: 10.1021/acs.orglett.8b00909
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
bears a cyclopropyl group, its reaction with tris(trimethylsilyl)-
silane under the standard conditions afforded a mixture of the
expected product alkenylsilane 33 and allenylsilane 34 in a 1:1
ratio (Scheme 4, I). Obviously, the emergence of an
allenylsilane product evidenced that a vinyl radical adjacent to
the cyclopropane ring was involved in this free-radical reaction,
resulting in homolytic cleavage of a carbon−carbon bond in the
absence of any alkynophilic Lewis acids. To gain insight into
the role of the thiol and quenching pathway of the transient
vinyl radical, a variety of isotope labeling experiments were
performed. When oct-1-yne was treated with deuterated
tris(trimethylsilyl)silane under the standard conditions, an
alkenylsilane was detected with 34% incorporation of
deuterium at the C2 position (Scheme 4, IIa). A similar
installation of deuterium at the same position of the product
was observed when H2O was replaced with D2O (Scheme 4,
IIb). However, no deuterium was installed in the product when
1,4-dioxane-D8 was applied as the solvent (Scheme 4, IIc).
These results indicate that the new hydrogen atom in the
alkenylsilane adduct originates from not only hydrosilane but
also water, and it is irrelevant with the solvent 1,4-dioxane.
Furthermore, in the absence of a thiol additive, 84% deuterium
incorporation was observed by using deuterated silane
((TMS)3SiD, Scheme 4, IId), while less than 4% deuterium
was installed when D2O was employed (Scheme 4, IIe). It
means that two radical-quenching pathways may be involved, in
which both hydrosilane and H2O can serve as the hydrogen
atom source. Most importantly, only in the presence of thiol
can the hydrogen atom transfer of water be implemented.
Finally, our parallel experiments revealed that the progress of
the reaction was frozen after removing the light source during
the reaction, while continuous irradiation gave the product in
99% yield (Figure 1). The quantum yield of the reaction was
Scheme 3. Trisubstituted Alkenylsilanes
a
50 W white LEDs were used, and the reaction time is 24 h. The yields
and Z/E ratios outside of parentheses correspond to the isolated
products, while the Z/E ratios of crude products are shown inside
b
parentheses. The reaction time is 60 h.
a high regio- and stereoselective formation of 32, even though
the yield is moderate after extending the reaction time.
To explain the mechanism of this type of alkyne hydro-
silylation, we designed several experiments to capture some
crucial evidence (Scheme 4). When one terminal of acetylene
Scheme 4. Mechanistic Studies
Figure 1. Hydrosilylation of oct-1-yne with/without light.
determined to be 0.257 (for details, see the SI), which suggests
that a photoredox pathway may be involved rather than a
radical chain propagation.
A mechanism was thus proposed based on the above
experimental evidence (Scheme 5). Under the irradiation of
white LEDs, a single electron transfer (SET) occurs between
Red
the excited state EY* (E1/O2x = 0.83 V vs SCE) and
tris(trimethylsilyl)silane (E1/2 = 0.73 V vs SCE), affording
silyl radical cation A and reduced EY. The radical cation was
converted to silyl radical B through deprotonation promoted by
a base or water. A subsequent intermolecular addition of a silyl
radical across a triple bond takes place, resulting in the
C
DOI: 10.1021/acs.orglett.8b00909
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
different mechanisms, which involve either a single electron
transfer or a hydrogen atom transfer in the radical-initiation
step and radical-quenching step. At this stage, the thiol in this
reaction is preferred to serve as a hydrogen donor in the
radical-quenching step rather than a hydrogen abstractor of
hydrosilane in the radical-initiation step.
Scheme 5. A Proposed Mechanism for the Hydrosilylation
In conclusion, visible-light-promoted metal-free hydrosilyla-
tion of alkynes has been developed by using catalytic amounts
of Eosin Y as a photocatalyst, thiol as a radical quencher, and
potassium carbonate as a base additive. A variety of terminal
and internal alkynes with different electronic properties were
successfully transformed to the corresponding di- and
trisubstituted alkenylsilanes with good functional group
tolerance. Control experiments indicate that the hydrosilylation
is initiated preferentially by a SET between hydrosilane and
excited Eosin Y, and a photoredox mechanism is preferred.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b00909.
formation of interconvertible vinyl radicals C and D.14 After a
hydrogen atom abstraction from thiol, the Z- and E-
alkenylsilane adducts will be yielded accompanied by thiol
radical E. A single electron transfer between thiyl radical E and
reduced EY will complete the catalytic cycle, affording the
photocatalyst and a thiol anion F, the latter of which is prone to
recombine with a proton to regenerate thiol. The following
should be noted: (1) The relative reactivity of the
intermolecular HAT and thermodynamic stability of silylated
vinyl radicals C and D are pivotal factors controlling the final
distribution of regioisomers and stereoisomers. For terminal
alkynes (R1 = H), anti-Markovnikov Z-alkenylsilanes derived
from radical C were isolated as a major product, even though
radical D is thermodynamically more stable than C. The
quenching of radical D should be slower than that of C, oweing
to the bulky steric hindrance of the silyl group, which makes the
abstraction of hydrogen atom disadvantageous. The excellent
Z-slectivity was also observed in trisubstituted alkenylsilane 32.
However, in the case of trisubstituted aromatic alkenylsilanes
(R2 = Ar), poor stereoselectivities were encountered, which
indicates that the quenching of radical D is competitive.
Furthermore, the introduction of an electron-deficient group
(R1 = CO2Me) on the internal alkyne might facilitate the
quenching of radical D, providing the E-alkenylsilane as a major
product. (2) From the point of view of polar effect, the thiol
might be preferred as a major hydrogen contributor of a
nucleophilic vinyl radical rather than the silane, even though a
hydrogen atom transfer from the silane to vinyl radical is
practical, especially in the absence of a thiol additive, as
evidenced by our deuterium experiments. (3) An alternative
chain-process mechanism, in which the thiol behaves as a
polarity-reversal catalyst in both the silyl-radical-initiation step
and vinyl-radical-quenching step,6c,15 is not supported by the
Experimental procedures, characterizations, and copies of
1H and 13C NMR spectra of products (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: wangsz@nju.edu.cn (S.W.).
*E-mail: yaoz@nju.edu.cn (Z.J.Y.).
ORCID
Shaozhong Wang: 0000-0002-7766-4433
Zhu-Jun Yao: 0000-0002-6716-4232
Author Contributions
†J.Z. and W.-C.C. contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The work was financially supported by the National Natural
Science Foundation of China (21572098, 21532002,
21472087). We thank Prof. Li-Zhu Wu and Dr. Tao Lei of
the Technical Institute of Physics and Chemistry, Chinese
Academy of Sciences for the determination of quantum yields.
REFERENCES
(1) (a) Fleming, I.; Dunogues
Wiley: New York, 2004; pp 57−575. (b) Muzafarov, A. M. Silicon
Polymers; Springer: Heidelberg, 2011.
■
̀
, J.; Smithers, R. Organic Reactions;
(2) (a) Lim, D. S. W.; Anderson, E. A. Synthesis 2012, 44, 983−1010.
(b) Cui, H.; Ma, B.; Cui, C. Organometallics 2012, 31, 7339−7342.
(3) (a) Trost, B. M.; Ball, Z. T. Synthesis 2005, 2005, 853−887.
́
(b) Marciniec, B.; Maciejewski, H.; Pietraszuk, C.; Pawluc, P. In
Hydrosilylation: A Comprehensive Review on Recent Advances, Vol. 1;
Marciniec, B., Ed.; Springer: Berlin, 2009; chapters 2 and 3. (c) Sun, J.;
Deng, L. ACS Catal. 2016, 6, 290−300.
quantum-yield calculations (see SI for more details). Poor
Ox
hydrosilylation was observed when triphenyl silane (E1/2
=
Ox
0.81 V vs SCE) or dimethylphenylsilane (E1/2 = 0.91 V vs
SCE) with higher oxidation potential was used, though a HAT
between this kind of silane and the thiyl radical is feasible.6g
Based on our current experimental evidence, the distinguishing
reactivity between the corresponding silylated vinyl radical and
the silylated alkyl radical may be a determinant leading to
(4) (a) Kopping, B.; Chatgilialoglu, C.; Zehnder, M.; Giese, B. J. Org.
Chem. 1992, 57, 3994−4000. (b) Amrein, S.; Timmermann, A.;
Studer, A. Org. Lett. 2001, 3, 2357−2360. (c) Yang, X.; Wang, C.
Angew. Chem. 2018, 130, 935−940.
(5) (a) Chatgilialoglu, C. Acc. Chem. Res. 1992, 25, 188−194.
(b) Chatgilialoglu, C. Chem. Rev. 1995, 95, 1229−1251. (c) Chatgi-
D
DOI: 10.1021/acs.orglett.8b00909
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
lialoglu, C. Organosilanes in Radical Chemistry; John Wiley & Sons,
Ltd.: Chichester, 2004.
(6) (a) Selin, T. G.; West, R. J. Am. Chem. Soc. 1962, 84, 1860−1863.
(b) Ballestri, M.; Chatgilialoglu, C.; Clark, K. B.; Griller, D.; Giese, B.;
Kopping, B. J. Org. Chem. 1991, 56, 678−683. (c) Dang, H.-S.;
Roberts, B. P. Tetrahedron Lett. 1995, 36, 2875−2878. (d) Romain, E.;
Fopp, C.; Chemla, F.; Ferreira, F.; Jackowski, O.; Oestreich, M.; Perez-
Luna, A. Angew. Chem., Int. Ed. 2014, 53, 11333−11337. (e) Zhang, L.;
Hang, Z.; Liu, Z.-Q. Angew. Chem., Int. Ed. 2016, 55, 236−239.
(f) Yang, Y.; Song, R.-J.; Ouyang, X.-H.; Wang, C.-Y.; Li, J.-H.; Luo, S.
Angew. Chem., Int. Ed. 2017, 56, 7916−7919. (g) Zhou, R.; Goh, Y.;
Liu, H.; Tao, H.; Li, L.; Wu, J. Angew. Chem., Int. Ed. 2017, 56, 16621−
16625.
(7) For selected reviews, see: (a) Prier, C. K.; Rankic, D. A.;
MacMillan, D. W. C. Chem. Rev. 2013, 113, 5322−5363. (b) Schultz,
D. M.; Yoon, T. P. Science 2014, 343, 1239176−1−1239176−8.
(c) Shaw, M. H.; Twilton, J.; MacMillan, D. W. C. J. Org. Chem. 2016,
81, 6898−6926. (d) Skubi, K. L.; Blum, T. R.; Yoon, T. P. Chem. Rev.
2016, 116, 10035−10074. (e) Capaldo, L.; Ravelli, D. Eur. J. Org.
Chem. 2017, 2017, 2056−2071.
(8) Zhu, J.; Yuan, Y.; Wang, S.; Yao, Z.-J. ACS Omega 2017, 2, 4665−
4677.
(9) For BDE values, see: (a) Luo, Y. R. Handbook of Bond
Dissociation Energies in Organic Compounds; CRC Press, LLC: Boca
Raton, FL, 2003. (b) Wang, H.-J.; Fu, Y. J. Mol. Struct.: THEOCHEM
2009, 893, 67−72.
(10) For electrochemical measurements of silanes, see SI.
(11) For reviews on applications of tris(trimethylsilyl)silane, see:
(a) Chatgilialoglu, C. Chem. - Eur. J. 2008, 14, 2310−2320.
(b) Chatgilialoglu, C.; Lalevee, J. Molecules 2012, 17, 527−555.
́
(12) For reviews on photocatalyst EY, see: (a) Romero, N. A.;
Nicewicz, D. A. Chem. Rev. 2016, 116, 10075−10166. (b) Hari, D. P.;
Konig, B. Chem. Commun. 2014, 50, 6688−6699. (c) Ravelli, D.;
̈
Fagnoni, M.; Albini, A. Chem. Soc. Rev. 2013, 42, 97−113.
(13) For a platinum-catalyzed-hydrosilylation of electron-with-
drawing internal alkynes: Rooke, D. A.; Ferreira, E. M. Angew.
Chem., Int. Ed. 2012, 51, 3225−3230.
(14) Wille, U. Chem. Rev. 2013, 113, 813−853.
́ ̀
(15) (a) Roberts, B. P. Chem. Soc. Rev. 1999, 28, 25−35. (b) Denes,
F.; Pichowicz, M.; Povie, G.; Renaud, P. Chem. Rev. 2014, 114, 2587−
2693.
E
DOI: 10.1021/acs.orglett.8b00909
Org. Lett. XXXX, XXX, XXX−XXX