Angewandte
Chemie
[
8]
Abstract: This works introduces hypervalent bis-catecholato
silicon compounds as versatile sources of alkyl radicals upon
visible-light photocatalysis. Using Ir[(dF(CF )ppy) (bpy)]-
first synthesized by Frye, as substrates amenable to the
oxidation process. In that context, the reports by Nishigaishi
[9a]
et al. on the photoallylation of benzil-type derivatives and
3
2
[
9b]
(
PF6)
(dF(CF )ppy = 2-(2,4-difluorophenyl)-5-trifluorome-
dicyanoarenes using an allylbiscatecholato silicate reagent
under photoinduced electron transfer (PET) conditions drew
our attention. Despite narrow scope and modest yields, these
seminal studies drove us to design an unprecedented photo-
catalyzed oxidation of catechol-based silicates, under low-
energy visible-light activation and using readily available
3
thylpyridine, bpy = bipyridine) as catalytic photooxidant,
a series of alkyl radicals, including highly reactive primary
ones can be generated and engaged in various intermolecular
homolytic reactions. Based on cyclic voltammetry, Stern–
Volmer studies, and supported by calculations, a mechanism
involving a single-electron transfer from the silicate to the
photoactivated iridium complex has been proposed. This
oxidative photocatalyzed process can be efficiently merged
[10,11]
photoactive catalysts,
to generate functionalized alkyl
radicals and notably primary ones. The radical intermediates
were engaged in a series of homolytic transformations,
including dual catalysis involving nickel-catalyzed cross-
coupling (Scheme 1).
with nickel-catalyzed Csp
2
ÀCsp
3
cross-coupling reactions.
O
ver the last decades, radical chemistry has entered
mainstream organic synthesis so that homolytic steps are
nowadays quite frequently encountered in multistep synthe-
[
1]
ses of complex molecules. Therefore, the need to access
radical intermediates from an increasing number of method-
ologies and associated precursors remains high, this need
being even more vivid for highly reactive entities such as
primary radicals. Redox pathways display high versatility as
C-centered radicals have been easily generated by single-
electron transfer (SET) reduction of substrates bearing
electron-withdrawing functional groups. Conversely, SET
oxidative processes of electron-rich substrates, for instance
organometallics, are well known. Nevertheless, finding or-
ganometallic precursors which exhibit sufficient stability and
Scheme 1. Hypervalent silicon compounds as a source of C-radicals.
Our first task was to devise the most reliable access to bis-
catecholato silicates. We focused on potassium silicates
obtained from the reaction of commercially available or
readily prepared trialkoxysilanes with catechol in the pres-
[
1,2]
function compatibility is not that obvious.
Soft carbanions
[12]
such as ate complexes or hypervalent derivatives are intrigu-
ing alternatives with better handling properties. For instance,
organoborates and notably alkyl trifluoroborates can be
ence of potassium methoxide in methanol (method A). We
found that the presence of 18-crown-6 (18-C-6) ether was
highly beneficial for the stability of the hypervalent deriva-
[3]
[13,14]
oxidized and have recently been used in various efficient
tives.
Various bis-catecholato silicates such as primary
[4]
reactions under photocatalytic conditions. However, the
main drawbacks reside in the formation of polluting boron
fluoride side products and so far the difficulty to generate
(1a–i, 1m–n), secondary (1j, 1p), and aryl (1k–l) silicates
could be prepared in large scale, isolated as solids, and stored
at room temperature without any degradation (Scheme 2).
For the more sterically demanding cyclopentyl and tert-butyl
silicates, we had to use the described preparation of tetra-
[
5]
primary radicals under photocatalytic conditions. Hyper-
valent silicon compounds are also very attractive targets in
[
6]
[15]
terms of accessibility and potential reactivity. A seminal
study by Kumada et al. reported the stoichiometric copper(II)
oxidation of organopentafluorosilicates to provide organic
ethylammonium silicates, using trichlorosilane precursors
(method B) and obtained the corresponding silicates 1p and
1o in decent yields. The structures of 1d, 1j, 1n, and 1p were
confirmed by X-ray crystallography (see the Supporting
Information, SI).
[
7]
radicals. However, their extremely low solubility in organic
or aqueous media severely limit their use. We considered the
more soluble bis-catecholato hypervalent silicon derivatives,
We next examined their ability to generate carbon-
centered radicals under photooxidative conditions. Benzylsi-
licate 1c was initially chosen as representative substrate
because it was expected to lend itself to easy oxidation thanks
to the stability of the resulting radical and because of its
[
*] Dr. V. CorcØ, L.-M. Chamoreau, Dr. E. Derat, Prof. Dr. J.-P. Goddard,
Dr. C. Ollivier, Prof. Dr. L. Fensterbank
Institut Parisien de Chimie MolØculaire, UMR CNRS 8232
Sorbonne UniversitØs UPMC Univ Paris 06
[3b,4a,16]
known fast trapping by TEMPO.
To probe the oxida-
4
Place Jussieu, CC 229, 75252 Paris Cedex 05 (France)
tion process, 2.2 equiv of TEMPO were added in the reaction
mixture to trap any formed benzyl radical and to act as
sacrificial electron acceptor in the catalytic cycle (Sche-
E-mail: cyril.ollivier@upmc.fr
Prof. Dr. J.-P. Goddard
Laboratoire de Chimie Organique et Bioorganique EA 4566
UniversitØ de Haute-Alsace
Ecole Nationale SupØrieure de Chimie de Mulhouse
3
[4a]
me 4). The screening of different photocatalysts Ru(bpy)3-
(
(
PF6)2 (bpy = bipyridine), Ir[(dF(CF )ppy) (bpy)]PF (dF-
3 2 6
CF )ppy = 2-(2,4-difluorophenyl)-5-trifluoromethylpyri-
3
dine), and Ir[(dF(CF )ppy) (dtbbpy)]PF (dtbbpy = 4,4’-di-
Bis rue Alfred Werner, 68093 Mulhouse Cedex (France)
3
2
6
E-mail: jean-philippe.goddard@uha.fr
tert-butyl-2,2’-biyridine) under irradiation with blue LEDs
477 nm) was achieved in acetone, MeCN, MeOH, and DMF
(SI). A drastic solvent effect was observed when using DMF
(
Angew. Chem. Int. Ed. 2015, 54, 11414 –11418
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim