A photoreducible copper(ii)-tren complex of practical value: Generation of a highly reactive click catalyst

Article abstract of DOI:10.1002/chem.201302032

A detailed study on the photoreduction of the copper(II) precatalyst 1 to generate a highly reactive cuprous species for the copper(I)-catalyzed alkyne-azide cycloaddition (CuAAC) click reaction is presented. For the photoactive catalyst described herein, the activation is driven by a photoinduced electron transfer (PET) process harnessing a benzophenone-like ketoprofenate chromophore as a photosensitizer, which is equally the counterion. The solvent is shown to play a major role in the Cu^II to Cu ^I reduction process as the final electron source, and the influence of the solvent nature on the photoreduction efficiency has been studied. Particular attention was paid to the use of water as a potential solvent, aqueous media being particularly appealing for CuAAC processes. The ability to solubilize the copper-tren complexes in water through the formation of inclusion complexes with β-CDs is demonstrated. Data is also provided on the fate of the copper(I)-tren catalytic species when reacting with O₂, O ₂ being used to switch off the catalysis. These data show that partial oxidation of the secondary benzylamine groups of the ligand to benzylimines occurs. Preliminary results show that when prolonged irradiation times are employed a Cu^I to Cu⁰ over-reduction process takes place, leading to the formation of copper nanoparticles (NPs). Finally, the main objective of this work being the development of photoactivable catalysts of practical value for the CuAAC, the catalytic, photolatent, and recycling properties of 1 in water and organic solvents are reported. Copyright

Full text of DOI:10.1002/chem.201302032

FULL PAPER
DOI: 10.1002/chem.201302032
A Photoreducible Copper(II)-Tren Complex of Practical Value: Generation
of a Highly Reactive Click Catalyst
Lydie Harmand, Romain Lambert, Luca Scarpantonio, Nathan D. McClenaghan,
Dominique Lastꢀcouꢁres, and Jean-Marc Vincent*[a]
Abstract: A detailed study on the pho-
toreduction of the copper(II) precata-
lyst 1 to generate a highly reactive
cuprous species for the copper(I)-cata-
has been studied. Particular attention
was paid to the use of water as a poten-
tial solvent, aqueous media being par-
ticularly appealing for CuAAC pro-
cesses. The ability to solubilize the
copper-tren complexes in water
through the formation of inclusion
complexes with b-CDs is demonstrated.
Data is also provided on the fate of the
copper(I)-tren catalytic species when
reacting with O2, O2 being used to
switch off the catalysis. These data
show that partial oxidation of the sec-
ondary benzylamine groups of the
ligand to benzylimines occurs. Prelimi-
nary results show that when prolonged
irradiation times are employed a CuI to
Cu0 over-reduction process takes place,
leading to the formation of copper
nanoparticles (NPs). Finally, the main
objective of this work being the devel-
opment of photoactivable catalysts of
practical value for the CuAAC, the cat-
alytic, photolatent, and recycling prop-
erties of 1 in water and organic sol-
vents are reported.
lyzed
alkyne-azide
cycloaddition
(CuAAC) click reaction is presented.
For the photoactive catalyst described
herein, the activation is driven by
a
photoinduced electron transfer
(PET) process harnessing a benzophe-
none-like ketoprofenate chromophore
as a photosensitizer, which is equally
the counterion. The solvent is shown to
play a major role in the CuII to CuI re-
duction process as the final electron
source, and the influence of the solvent
nature on the photoreduction efficiency
Keywords: click chemistry · copper ·
homogeneous catalysis · photoin-
duced electron transfer · photo-
chemistry · photolatent catalyst
Introduction
Most of the transition-metal catalysts exhibit an oxida-
tion-state dependent activity. Notable examples include pal-
ladium-catalyzed coupling reactions for which the active
species is the reduced Pd0, whereas the oxidized PdII is inac-
tive. Typically in these reactions a PdII precatalyst is em-
ployed, the active Pd0 species being generated in situ, typi-
cally through phosphine ligand oxidation or reductive elimi-
nation processes.[4] Electrochemical reduction has also
emerged as a powerful activation method.[4] Rather surpris-
ingly, photoinduced electron transfer (PET) to a metal
cation as a way to activate a catalyst has been largely unex-
plored. We recently reported the synthesis, preliminary sen-
sitized photoreduction studies and catalytic activity of
The design of chemical processes that can be controlled by
light is a research area with strong potential for develop-
ment.[1] Catalysts whose activity can be switched on/off upon
light irradiation are of particular interest.[2] Considering
transition-metal catalysts, the generation of a highly active
catalytic species on demand by the light irradiation of an in-
active precatalyst presents several key benefits: 1) handling
metal species that are stable under ambient conditions;
2) switching on catalysis at a given time and place; 3) tuning
the onset of the generated active catalyst by controlling the
photoactivation process, which would be valuable for highly
exothermic reactions conducted on a large scale. To date,
most examples of photocontrolled transition-metal catalysts
are the so-called photocaged catalysts, which rely on the
photodissociation or photodecomposition of ligands in coor-
dinatively saturated metal complexes to generate vacant
active sites.[2,3]
[Cu(tBuBz3tren)ketoprofenate]ketoprofenate
complex
1 (Figure 1), a prototype photolatent catalyst whose activa-
tion is driven by a PET process.[5] The copper(I)-catalyzed
alkyne–azide cycloaddition (CuAAC) is an example of a re-
action for which the catalyst is inactive in its oxidized cupric
state but active in its reduced cuprous form.[6] This click re-
action is now widely employed in all areas of the chemical
sciences.[7,8] Within the last ten years a range of catalytic sys-
tems have been developed, most of them being based on
either copper(II) precatalysts to which an external sacrificial
reductant, such as sodium ascorbate, is added or directly re-
duced copper(I) catalysts.[9] In recent years, light-induced
click reactions have received special attention,[10] and a few
examples of photodriven CuAAc processes have been re-
[a] Dr. L. Harmand, R. Lambert, Dr. L. Scarpantonio,
Dr. N. D. McClenaghan, Dr. D. Lastꢀcouꢁres, Dr. J.-M. Vincent
Universitꢀ de Bordeaux, Institut des Sciences Molꢀculaires
UMR-CNRS 5255, 351, cours de la libꢀration
33405 Talence Cedex (France)
E-mail: jm.vincent@ism.u-bordeaux1.fr
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201302032.
Chem. Eur. J. 2013, 19, 16231 – 16239
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
16231

detail, with the solvent playing a major role in the process
because it represents the terminal electron source. As water
is a particularly appealing reaction medium for the CuAAC,
special attention was paid to the solubilization of the cop-
per(II) precatalysts in water, exploiting the formation of in-
clusion complexes with b-cyclodextrins (b-CDs). Preliminary
results showing the formation of copper nanoparticles (NPs)
under prolonged irradiation times are described. Data is
provided on the fate of the catalyst when the highly reactive
photogenerated copper(I) catalyst reacts with O2, O2 being
used to switch off the catalysis. Finally, the main objective of
this work being the development of photoactivable catalysts
of practical value for the CuAAC, the catalytic, photolatent,
and recycling properties of 1 in aqueous and organic sol-
vents are reported.
Figure 1. Structural formulas of complexes 1 and 2 (left) and molecular
structure of the [Cu(tBuBz3tren)acetate]+ cation (right).
ported.[11] Of particular interest is the use of free-radical
photoinitiators to mediate the copper(II) to copper(I) re-
duction, which were applied to the preparation of poly-
mers.[11b,c]
Results and Discussion
In our preliminary communication we have shown that al-
though 1 is inactive as a catalyst at room temperature in
methanol, it is quantitatively and rapidly reduced to its cu-
Design, synthesis, and characterization of complexes 1–4:
With the aim of developing a photoactivable catalyst of
practical value, precatalyst 1 was designed to have a simple
preparation, yet present three key structural/functional fea-
tures: 1) A tetradentate tren-derived ligand, which should
favor the stabilization of the CuI reduced species, thus limit-
ing the formation of colloidal Cu0, which may occur during
the photoreduction process. Moreover we,[12] and others,[13]
have previously shown that the copper(I)-tren complexes
are excellent click catalysts; 2) Three available tert-butylben-
zyl (tBuBz) substituents are incorporated on the tren unit,
which are excellent guests for b-CDs. This should allow the
solubilization of 1 in both organic solvents and water, in
a strategy inspired by the seminal work of Monflier and co-
workers who exploited such supramolecular interactions not
only to solubilize organometallic complexes in water, but
also modulate the catalyst selectivities;[14] 3) A ketoprofen-
ate ligating counterion as photosensitizing group. The keto-
profen 2-(3-benzoylphenyl)propionic acid is a well-known
commercially available non-steroidal anti-inflammatory
drug containing the benzophenone chromophore and a car-
boxylic acid group. This choice was prompted by the work
of Chow and Buono-Core who reported that certain aromat-
ic ketones, in particular benzophenones, could sensitize the
photoreduction of transition-metal complexes ML2 such as
ACHTUNGTRENNUNGprous state upon light irradiation (TLC lamp, l=365 nm) to
provide a highly reactive click catalyst (Figure 2).[5] The pho-
Figure 2. Reaction profile showing the reversible photolatent properties
of 1 for the CuAAC.
togenerated cuprous [Cu(tBuBz3tren)]+ species proved to
be extremely sensitive towards O2. Consequently, by simply
introducing air into the reaction medium the catalysis could
be switched off, and then subsequently switched on by bub-
bling Ar followed by irradiation (Figure 2). Preliminary
flash photolysis studies reveal that the photoreduction pro-
cess involves the triplet benzophenone chromophore, which
is efficiently quenched (kq =1.0ꢃ108 sꢀ1), most likely by
a charge-transfer process from the copper(II) ion.
In the present article, we present the scope and limita-
tions of the first generation of copper-tren-ketoprofenate
photolatent click catalysts. The influence of the solvent
nature on the photoreduction efficiency has been studied in
[CuACHTNGUTRENNUG CAHTUNGTRENNNUG
(acac)2][15] and [Ni(acac)2] (acac=acetylacetonate),[16] or
[Cu(Bp)2] (Bp=dihydrobis(1-pyrazolyl)borate) in hydrogen-
donating solvents.[17]
In our preliminary account we reported the synthesis and
characterization of the copper(II) complex [Cu(t-
BuBz3tren)ketoprofenate]ketoprofenate 1 and its non-pho-
toactivable
structural
analogue
[Cu(t-
BuBz3tren)acetate]acetate 2.[5] The precatalyst (compound
1) was obtained in 90% yield by treating the tBuBz3tren
ligand in methanol with the copper(II)-carboxylate complex
[Cu2(ketoprofenate)
prepared by treating [Cu
sulfonate) with two equivalents of the sodium salt of keto-
(H2O)2] (3) (Figure 3). Complex 3 is
AHCTUNGTRENNUNG
16232
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16231 – 16239

A Photoreducible Copper(II)-Tren Complex
FULL PAPER
Figure 3. Synthesis and proposed structure of the copper(II)-ketoprofen-
ate complex 3.
profen in water. The precipitate that immediately forms is
filtered and dried affording 3 as a green powder in 97%
yield. Complex 3 was characterized by elemental analysis,
UV/Vis, FTIR, and EPR spectroscopies. In particular, the
X-band EPR spectrum of a solid, powdered sample of 3
measured at room temperature (Figure S1, the Supporting
Information) is characteristic of a magnetic copper(II)
dimer possessing the typical “paddle-wheel” structure with
four carboxylates symmetrically bridging the copper(II)
ions.[18] The complex [Cu(tBuBz3tren)acetate]acetate 2 was
obtained in a similar manner as its structural analogue 1 by
Figure 4. Photographs of aqueous solutions of 1 before (left) and after
addition (right) of 10 equivalents of RAME-b-CD (final concentration
5.2 mm), and graph showing the percentage of solubilized 1 upon addition
of 1 to 10 equivalents of RAME-b-CD. The inset shows the evolution of
the visible absorption spectrum of the solution after each addition of one
equivalent of CD.
assembly forms in which each tBuBz substituent is included
in a RAME-b-CD. The molecular models presented in
Figure 5 show that 3 RAME-b-CDs could easily fit around
the [Cu(tBuBz3tren)]2+ cation, and that steric hindrance
should not hamper the formation of such an assembly. For
1, the 5:1 ratio between the CD and the complex, necessary
to achieve almost complete solubilization, suggests that not
only the tert-butylphenyl groups, but also the benzophenone
moiety of the ketoprofenate anions, are encapsulated by the
RAME-b-CD. Confirmation was sought through MS studies
(see below).
treating the tBuBz3tren ligand with [Cu2ACHTNUTRGENNUG(OAc)4ACHTUNGTREN(NUNG H2O)2]. The
X-ray analysis of single crystals of 2 (Figure 1) revealed
a monomeric structure with the copper(II) ion in a distorted
trigonal-bipyramidal geometry and an acetate occupying the
axial coordination site in a syn monodentate mode. By anal-
ogy, and in agreement with characterization, we proposed
a
similar structure for 1. Finally, the complex [Cu(t-
BuBz3tren)Cl]·Cl 4, was prepared in 91% yield from
[CuCl2]·5H2O and tBuBz3tren.
Supramolecular complexes between 1–2, 4 and b-CDs in
water: Major applications of the CuAAC reactions involve
biomolecule substrates (notably polypeptides, oligosacchar-
ides, and oligonucleotides), which are intrinsically water-
soluble.[19] The search for highly reactive click catalysts that
are compatible with water-soluble reactants is thus of special
interest.[20] Exploiting supramolecular host–guest interac-
tions between the b-CDs and the tBuBz substituents of the
tren ligand should provide a photolatent catalyst useful for
both organic and aqueous reaction conditions. The photola-
tent precatalyst 1 is a lipophilic complex, which is soluble in
most organic solvents except alkanes. Complex 1 is com-
pletely insoluble in water (as verified by UV/Vis spectrosco-
py, see Figure 4), whereas [Cu(tBuBz3tren)Cl]·Cl (4) is
slightly soluble (<1 mm), and 2 is moderately soluble (up to
10 mm). Consequently, the solubilization studies were con-
ducted with complexes 1 and 4, employing the highly water-
soluble randomly-methylated b-cyclodextrin (RAME-b-CD)
as the host. Rapid solubilization occurs when the RAME-b-
CD is added to aqueous suspensions of 1 or 4. Titration ex-
periments revealed that the addition of 5 and 3 equivalents
of the CD was enough to solubilize 90–95% of complexes
1 (Figure 4) and 4 (Figure S2, the Supporting Information),
respectively. These results suggest that for complex 4,
Figure 5. Side and top views of CPK models of the minimized (MM2,
CaChe 3.1) supramolecular [Cu(tBuBz3tren)ꢁ3b-CD]2+ cation. A b-CD
ꢀ
with permethylated primary CH2OH functions has been considered for
this study to reflect the steric hindrance of the smaller rim found in the
RAME-b-CD.
Nuclear Overhauser spectroscopy (2D NOESY) and dif-
fusion-ordered spectroscopy (DOSY) NMR experiments did
not afford direct evidence of such an interaction in solution.
This is ascribed to the paramagnetic nature of the copper(II)
1
ꢀ
ꢀ
ꢀ
ion, affording H NMR spectra in which the Ph CH2 N, N
ꢀ
ꢀ
ꢀ
CH2CH2 N, and CHCH3 CO2 resonances are not ob-
served, whereas all the others are very broad (see the Sup-
porting Information, Figures S3 and S4 for the 600 MHz
a
supramolecular
[Cu(tBuBz3tren)ꢁ3RAME-b-CD]2+
Chem. Eur. J. 2013, 19, 16231 – 16239
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16233

J.-M. Vincent et al.
1H NMR spectra of 1 in MeOD, 1+10 equiv RAME-b-CD
in D2O, and RAME-b-CD in D2O). Interestingly, the
1D NMR spectra show that the anomeric protons of the
RAME-b-CD are shifted upfield by about d=0.1 ppm when
1 is present, indicating that an interaction with a guest effec-
tively takes place (Figure S4, the Supporting Information).
The 2D NOESY and DOSY experiments conducted on the
diamagnetic tBuBz3tren ligand confirmed the formation of
inclusion complexes. Addition of three equivalents of
RAME-b-CD to a suspension of the tBuBz3tren ligand in
D2O led to fast and complete solubilization (the 600 MHz
1H NMR spectrum of the resulting solution is given in Fig-
ure S3, the Supporting Information). The 2D NOESY spec-
trum (Figure 6) reveals intense cross-peaks between the pro-
tons of the tert-butylphenyl groups of the ligand and the
fusing at a much slower rate (D=3.54ꢃ10ꢀ11 m2sꢀ1) than the
free RAME-b-CD (D=1.77ꢃ10ꢀ10 m2sꢀ1), corresponding to
a hydrodynamic radius about five times higher.
Mass spectrometry studies have been conducted on aque-
ous solutions of 1 and 4 in the presence of b-CD. For the
MS studies, b-CD instead of RAME-b-CD was employed,
the latter being a mixture of randomly-methylated CDs,
with the major product bearing 14 methyl groups. The b-CD
was found to be less efficient in solubilizing 1 and 4 in
water. Only approximately 40% of 4 was solubilized with
three equivalents of b-CD, compared to approximately 95%
when employing RAME-b-CD. Nonetheless, the ESI-MS of
an aqueous solution of 4 and b-CD shows a peak at m/z=
1819.7 ascribed to the [Cu(tBuBz3tren)Clꢁb-CD]+ supra-
molecular cation (Figure S5, the Supporting Information).
With complex 1, the peak could not be observed in the posi-
tive mode as the [Cu(tBuBz3tren)ketoprofenateꢁb-CD]+
supramolecular cation would afford a peak at m/z=2034.9,
which is above the upper detection limit for ESI, that is,
2000 Da. However, in the negative mode an intense peak at
m/z=1387.5 Da assigned to the anionic inclusion complex
[ketꢁCD]ꢀ is observed (Figure S5, the Supporting Informa-
tion).
ꢀ
ꢀ ꢀ
ꢀ
ꢀ
CH , CH2 and OCH3 protons of the CD. The DOSY
experiments (Figure 6) confirmed the clean formation of
a supramolecular [tBuBz3trenꢁ3RAME-b-CD] complex dif-
Photoreduction studies of complexes 1 and 2: The main ob-
jective of this study was to assess the solvent scope and thus
determine the best conditions for the click reactions.
Indeed, the solvent plays a major role in the photoreduction
process as the potential terminal electron donor. Chow,
Buono-Core and co-workers, extensively studied the impact
of the solvent in the benzophenone photosensitized photore-
duction of transition-metal complexes ML2 such as [Cu-
ACHUTNGRENNUG CAHTUNGTRENNUGN
(acac)2],[15] [Ni(acac)2],[16] or [Cu(Bp)2].[17] The photoreduc-
tion proved to be particularly efficient in hydrogen-donating
solvents such as alcohols (MeOH, EtOH, iPrOH). Mecha-
nistic studies showed that photoreduction involves quench-
ing of the triplet aromatic ketone [Eq. (1)], most likely by
a charge-transfer complex intermediate [Eq. (2)], perform-
ing a monoelectronic oxidation of the ligand (L) to generate
a ligand-centered radical [Eq. (3)], which abstracts a hydro-
gen atom [Eq. (4)] from the solvent (RCH2OH). Overall,
the process generates the reduced complex ML, the proton-
ꢀ
C
ated ligand L H and an a-hydroxy radical R CHOH.
hv
*
Ph CO
Ph CO
ð1Þ
ð2Þ
ƒ!
2
2
Ph2CO þ ½M ðacacÞ2ꢂ ! Ph2CO þ ½MIIIðacacÞ2ꢂ
II
*
Cꢀ
III
II
Cꢀ
Cꢀ
Cþ
Ph2CO þ ½M ðacacÞ2ꢂ ! Ph2CO þ ½ðacacÞ M ðacacÞꢂ
ð3Þ
ð4Þ
Cꢀ
Cþ
Ph2CO þ ½ðacacÞ MIIðacacÞꢂ þ RCH2OH !
I
C
Ph2CO þ ½M ðacacÞꢂ þ Hacac þ R CHOH
Figure 6. a) 2D NOESY NMR spectrum (400 MHz) of a D2O solution of
tBuBz3tren (20 mm)+3 equiv RAME-b-CD (60 mm); b) DOSY spectrum
of the solution (black) superimposed with the DOSY spectrum (green)
of the RAME-b-CD alone (60 mm).
Photoreduction followed by UV/Vis spectroscopy: In our pre-
liminary work,[5] the photoreduction and catalytic studies
16234
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16231 – 16239

A Photoreducible Copper(II)-Tren Complex
FULL PAPER
Figure 7. Photoreduction of 1 in various solvents (1.85 mm, l=365 nm, deoxygenated by freeze–pump–thaw cycles, sealed quartz cuvettes) followed by
UV/Vis absorption spectroscopy: a) in THF (irradiation times=0, 4, 8, 12, 16 min from top spectrum to bottom spectrum, respectively); b) in toluene (ir-
radiation times=0, 8, 16, 24, 30 min from top spectrum to bottom spectrum, respectively); c) in CH3CN (irradiation times=0, 10, 30, 52, 70 min from top
spectrum to bottom spectrum, respectively).
were conducted in MeOH, a good H-donating and a green
solvent,[21] which are two favorable characteristics for the
title study. Irradiations were conducted at l=365 nm in
strictly anaerobic conditions (solutions degassed by freeze–
pump–thaw cycles, sealed cell) to eliminate O2 which, if
present, would react very rapidly with the copper(I) species
ultimately leading to ligand oxidation (see below). The
copper photoreduction process in MeOH proved to be very
efficient, the quantum (Fred) yield being estimated at 0.17.
Typical reductions of MeOH solutions of 1 (3 mL, 1.85 mm)
in 1 cm quartz UV cuvettes proceed in approximately 30
min, affording colorless solutions in agreement with the for-
mation of a [Cu(tBuBz3tren)]+ cationic species. Importantly,
the photoinduced reduction did not occur in 2 (in which ace-
tate replaces the ketoprofenate), which showed the funda-
mental photostability of the inorganic moiety when irradiat-
ing at 365 nm. In CH2Cl2, a poor H-donating/oxidizable sol-
vent, the photoreduction was sluggish, approximately 75%
of 1 being reduced after 3 h of irradiation. We report here
that in THF the reduction is very efficient, a colorless solu-
tion being obtained after irradiating for less than 20 min
(Figure 7). In toluene the complex 1 is rapidly photoreduced
in about 30 min, whereas in CH3CN, a poorer H-donating
solvent, the reduction was completed in about 1 h
(Figure 7).
treating a copper(I) species with acetylenes.[15a] Importantly,
when complex 2 is irradiated in water in the presence of
RAME-b-CD and excess propargyl alcohol, no reduction
occurs (Figure S7, the Supporting information), confirming
that with 1 the reduction is driven by the benzophenone
sensitizer.
1
Photoreduction followed by H NMR spectroscopy: Similarly
to observations in MeOD, sharp resonances, in particular
ꢀ
ꢀ
ꢀ
ꢀ
those of the benzylic CH2 and NCH2CH2N groups,
which are not observable with 1, emerge in the 1H NMR
spectra upon irradiation of a [D8]toluene solution, in agree-
ment with the formation of a diamagnetic (d10) cuprous
[Cu(tBuBz3tren)]+ cationic species (Figure 8). Concomitant-
ly, the broad resonances of 1, such as those of the tert-butyl
groups, disappear while the corresponding upfield-shifted
sharp resonances appear.
The photoreduction in water was tested by using a solution
(3 mL) containing 1 (1.84 mm) and 10 equiv of RAME-b-
CD. As shown by UV/Vis spectroscopy (Figure S6, the Sup-
porting Information) the reduction does not occur in water,
as expected for a poorly oxidizable/H-donating solvent.
Nonetheless, we hypothesized that in CuAAC reaction con-
ditions, the alkyne could potentially serve as the sacrificial
reductant and hence favor the reduction. When the photore-
duction is conducted in the presence of 200 equivalents
(conditions of catalysis experiments), the solution turns col-
orless within minutes with the concomitant formation of an
amorphous yellow precipitate (Figure S7, the Supporting In-
formation). The yellow precipitate, which is stable to air and
insoluble in any common solvent, is characteristic of poly-
1
Figure 8. Evolution of the H NMR spectrum of a [D8]toluene solution of
1 (5 mm, deoxygenated by Ar bubbling) upon irradiation of the NMR
tube at 365 nm.
Behavior of the cuprous [Cu(tBuBz3tren)]+ species in the
presence of O2: We previously showed that the photogener-
ated cuprous species was extremely sensitive towards O2,
and that such reactivity could be exploited to switch off the
reaction. After having been stopped, the reaction could be
switched on after deoxygenation of the reaction mixture fol-
lowed by irradiation (Figure 2). The reaction of copper(I)-
meric alkynyl-copper(I) complexes, possessing an infinite
[22]
ꢀ
ꢃ
Cu Cu ladder structure with bridging RC C groups. In-
terestingly, a range of such alkynyl-copper(I) complexes
have been previously prepared by Chow and Buono-Core
through the benzophenone-sensitized photoreduction of
[CuACHTUNGTRENNUNG(acac)2] in the presence of various acetylenes or, by
Chem. Eur. J. 2013, 19, 16231 – 16239
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16235

J.-M. Vincent et al.
polyamino complexes with O2 frequently leads to hydroxyl-
ation reactions at oxidizable positions of the ligand, imply-
ing copper(II)-superoxo, -hydroperoxo, or -peroxo inter-
mediates.[23] It was thus considered important to determine
whether these oxidative processes were taking place when
reduced 1 was treated with O2, and to see if major alteration
of the catalyst structure occurred. When a colorless toluene
solution of reduced 1 is exposed to air, the solution instanta-
neously turned green, indicating that the generated cuprous
species rapidly reacts with O2. Analysis of the ESI-MS spec-
trum of the resulting solution (Figure 9) shows peaks at
gassed by multiple freeze–pump–thaw cycles. Accordingly,
when less-effective degassing is conducted (Ar bubbling),
a more intense absorption at 410 nm is obtained (Figure S10,
the Supporting Information). This new absorption is charac-
teristic of a metal-to-ligand charge-transfer (MLCT) band
involving the copper(I) and a coordinating imine group that
originates from the oxidation of the benzylic amines of the
ligand by the oxidizing species produced when O2 reacts
with CuI. The attribution of this band was confirmed by re-
cording the UV/Vis spectrum of the cuprous [Cu(trisimine-
tren)Br] complex (Figure S11, the Supporting Information),
prepared in situ by treating CuBr and the trisimine ligand in
toluene under strictly anaerobic conditions. The resulting
spectrum shows an intense MLCT band at 410 nm (e=
4200mꢀ1 cmꢀ1). From this e value, it was estimated that ap-
proximately 10% of the amine groups in 1 were oxidized
into imines during the photoreduction process conducted
from the solution degassed by Ar bubbling and closed with
a rubber stopper. For the reduction conducted in a sealed
cell degassed by freeze–pump–thaw cycles, the amount of
imine that forms due to traces of O2 is estimated to be ap-
proximately 2.5%.
The amine-to-imine oxidation is thus favored when re-
duced 1 reacts with O2. Such reactivity is most probably re-
lated to that occurring in the copper(I)-catalyzed aerobic
oxidation of alcohols to aldehydes.[24] Based on our observa-
tions and previous extensive studies on related copper(I)/O2
systems, a tentative mechanism to account for the imine for-
mation is proposed in Scheme 1. Further studies, which go
Figure 9. ESI-MS spectrum of a toluene solution of 1 (1 mm) that was
photoreduced (deoxygenated by Ar bubbling) and subsequently reoxi-
dized with air. A comparative spectrum of the native complex 1 is given
in Figure S8 (the Supporting Information).
m/z=898.5 and 896.5, that is, 2 and 4 mass units less than
expected for the [Cu(tBuBz3tren)ketoprofenate]+ ion, but
also peaks at m/z=641.4 (minor), 643.4 (100%) and 645.4
(60%), that is, at 2, 4 and 6 mass units lower than the [Cu(t-
BuBz3tren)]+ cation. This agrees with the partial oxidation
of the benzylic amines into imines, resulting from the reac-
tion of the cuprous [Cu(tBuBz3tren)]+ cation with O2. Hy-
droxylation products that should afford peaks at 662.4 and
916.5 (+16 mass units) are not observed. The occurrence of
oxidation processes was further confirmed by reacting com-
plex 1 with 45 equivalents of H2O2 in MeOH for 3 h. The
ESI-MS spectrum of the resulting solution (Figure S9, the
Supporting Information) showed an intense peak at m/z=
Scheme 1. Proposed mechanism for the aerobic amine-to-imine ligand
oxidation.
641.4 attributed to the tris
cation.
The appearance of imine functions can be observed when
following the photoreduction process by UV/Vis spectrosco-
py. For instance, in the spectra recorded during the reduc-
tion in toluene (Figure 7), while the absorption band corre-
G
beyond the scope of this work, should be conducted to char-
acterize the reaction intermediates and confirm this mecha-
nism. Overall these results show that when adventitious
traces of O2 are present in the reaction medium, or when air
is introduced to stop the reaction, a rather efficient oxida-
tive process occurs leading to the formation of substantial
amounts of the oxidized imine ligand.
ꢀ
sponding to the d d copper(II) transition is disappearing
a new band appears at approximately 410 nm. The appear-
ance of this absorption band is correlated to the residual O2
present in the toluene solution of the sealed cuvette de-
Formation of copper nanoparticles (NPs) by prolonged irra-
diation: Chow and Buono-Core previously reported that
16236
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16231 – 16239

A Photoreducible Copper(II)-Tren Complex
FULL PAPER
prolonged irradiation in the benzophenone-sensitized reduc-
NPs (Figure 10). It can thus be assumed that the CuI to Cu0
over-reduction is mostly due to a photoreduction process
rather than disproportionation. It should also be underlined
that under catalytic conditions, the photogenerated cop-
per(I) species will be rapidly “trapped” by the alkyne and
azide present in large excess and then engaged in the cata-
lytic cycle, thus probably limiting the formation of
copper(0).
tion of [CuACHTUNGTRENNUNG
(acac)2],[15] and [Cu(Bp)2],[16] gave rise to the for-
mation of metallic copper particles due to disproportiona-
tion and/or photolysis of the cuprous species. They showed
that in the presence of strongly coordinating ligands (L)
such as phenanthroline, bipyridine, or triphenylphosphine,
copper(I) could be trapped to afford the corresponding [Cu-
ACHTUNGTRENNUNG(acac)Lx] complexes, which proved to be much less prone to
decomposition into metallic copper. The presence of the tet-
radentate tren ligand in 1 might limit the over-reduction
process, by lowering the reduction potential of the CuII/CuI
Photolatent and catalytic properties of 1: In our preliminary
report, reactions were conducted in methanol under an inert
atmosphere (Ar bubbling). The precatalyst was activated by
irradiating the reaction mixture at 365 nm (TLC lamp) for
15–30 min. The cuprous species generated was very active,
the reaction between the unprotected sugar b-d-galactopyra-
nosyl azide 5 and propargyl alcohol being completed in
about 3 hours with a catalyst loading of 0.5 mol%. From the
azide 5, a range of triazoles were isolated in good to excel-
lent yields. Interestingly, the reaction could be stopped im-
mediately at any time by simply introducing air in the reac-
tion medium. Degassing followed by irradiation switched on
the reaction (Figure 1).
I
0
1
and Cu /Cu , compared with copper salts. The E = of the
2
CuII/CuI couple for 1 measured in acetonitrile was found at
a rather low value of ꢀ0.34 V vs. SCE.[5] Nonetheless, we
observed that prolonged irradiation of 1 leads to CuI to Cu0
reduction and the formation of copper NPs. This is clearly il-
lustrated when following the photoreductions in THF
(Figure 10) or toluene (Figure S12, the Supporting Informa-
The latent behavior and click reactivity of 1 was further
probed in various organic solvents, namely the reaction be-
tween the soluble peracetylated b-D-galactopyranosyl azide
7 and propargyl alcohol substrates (Table 1). Click reactions
Table 1. Reactivity of 1 in various solvents.
Figure 10. UV/Visible spectra of a THF solution of 1 (1.85 mm, deoxygen-
ated by freeze–pump–thaw cycles, sealed cell) recorded after irradiation
times=30, 45, 60, 75, and 90 min (l=365 nm) in order of increasing in-
tensity at 550 nm, and photograph of the resulting homogeneous solu-
tion.
Solvents
Reaction times
to 100% conversion[a]
[D8]THF
CD2Cl2
[D8]toluene
CD3CN
<10 min
<10 min
<60 min
ꢄ40 h
tion). After typically 30–45 and 60 min of irradiation in
THF and toluene, respectively, the solutions turn progres-
sively to a henna color, with Plasmon resonances appearing
at 563 and 569 nm that are characteristic of dispersed Cu
NPs (particle diameter>4 nm).[25] To try to determine
whether the Cu0 was produced by disproportionation or
photoreduction of the CuI species, a yellowish toluene solu-
tion of reduced complex 1 (degassed by freeze–pump–thaw
cycles, sealed cuvette), obtained after 26 min of irradiation,
was left in the dark. The solution slowly recovers its initial
blue color, whereas fine particles of metallic copper appear
concomitantly. The UV/Vis spectrum of the supernatant
after seven days reveals that approximately 55% of the orig-
inal absorption at 826 nm has been recovered (Figure S13,
the Supporting Information), that is, close to the 50% theo-
retical value expected for a disproportionation process.
These results show that disproportionation of the cuprous
[Cu(tBuBz3tren)]+ occurs, but that this is a rather slow pro-
cess compared with the kinetics of appearance of the Cu
[a] Followed by using 1H NMR spectroscopy (reaction conditions given
in the Supporting Information).
were run in NMR tubes at room temperature (208C) in de-
oxygenated deuterated solvents (Ar bubbling, 20 min) with
1 (0.5 mol%). It is noteworthy that under ambient condi-
tions 1 exhibits no residual background activity, a crucial pa-
rameter when considering effective photoswitchable cata-
1
lysts (the H NMR spectra recorded at various times for the
reaction conducted in THF are provided in Figure S14, the
Supporting Information). Catalyst activation was carried out
by irradiating the NMR tube for 10 min at 365 nm by using
a TLC lamp. The reactions conducted in THF and CH2Cl2
proved to be extremely fast; the triazole 16 being formed
quantitatively at the end of the irradiation time, that is,
within 10 min. Interestingly, this shows that the reaction
Chem. Eur. J. 2013, 19, 16231 – 16239
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16237

J.-M. Vincent et al.
medium is not limited to good H-donating solvents. Indeed,
the photoreduction of 1 in CH2Cl2 in the absence of the re-
actant is sluggish, as approximately 75% of 1 is reduced
after 3 h irradiation.[5] However, as described above for the
photoreduction in water, under the reaction conditions, the
alkyne may serve as a sacrificial reductant. In toluene the
reaction proceeded very effectively (less than 60 min),
whereas in acetonitrile the reaction was very slow, with com-
plete conversion being reached after approximately 40 h.
Click reactions implying a range of alkynes and azides
were conducted in round-bottom flasks at low catalyst load-
ing (0.5 mol%, except for 13 which required 1 mol%) and
irradiation times of 30 min. Toluene was chosen as the sol-
vent, in which the triazoles that form are usually poorly
soluble. The triazoles (8–16) were obtained in good to excel-
lent yields, the products being conveniently isolated by fil-
tration after precipitation at ꢀ188C (Figure 11). A catalyst
tions, approximately 90% conversion was observed in 10 h,
the triazole 6 (Figure 2) being isolated in 83% yield after
chromatography.
Conclusion
PET processes could represent a powerful strategy for the
photoactivation of copper catalysts that are active in their
low-valent cuprous state. A benzophenone-like chomophore
proved to be a particularly useful photosensitizer to effi-
ciently generate a highly reactive click copper(I) species on-
demand from the copper(II)-tren precatalyst 1 upon light ir-
radiation at 365 nm. The convenient introduction of the ben-
zophenone chromophore in 1 through the ketoprofenate
counterion represents a rather general strategy, which could
be extended to other cationic metal complexes. We have
shown that the reaction medium is not limited to good H-
donating solvents, such as MeOH, THF, or toluene. Indeed,
under the reaction conditions, the alkyne may serve as a sac-
rificial reductant, opening the way to photolatent click catal-
ysis in aqueous media. The preliminary results reported
herein, exploiting the supramolecular precatalyst formed be-
tween 1 and the b-RAME-CD have shown that photolatent
click catalysis based on PET can indeed be conducted in
water. This study has shown that the photogenerated cu-
AHCTUNGTREGpNUNN rous-tren species is highly sensitive to O2. Although this re-
activity can be exploited to efficiently switch off the catalyt-
ic activity, by simply introducing air into the reaction
medium, it also induces side-oxidative reactions leading to
ligand oxidation with the formation of imines, thus raising
concerns on the chemical integrity of the catalyst. We also
noticed that over-reduction processes, ascribed to a photo-
sensitized process rather that disproportionation, may occur
leading to the formation of copper(0) and, ultimately, to
copper NPs. Although these are possible limitations, the
photoactivable precatalyst 1, which can be stored indefinite-
ly under ambient conditions, has strong potential for appli-
cations in click chemistry. These results may pave the way to
the development of a new generation of photoactivable cat-
alysts, not only for copper chemistry, but for other transi-
tion-metal-catalyzed reactions. Studies along these lines are
currently under investigation in our group.
Figure 11. Structures and reaction yields for triazoles 8–16.
recycling experiment was attempted in the following
manner. The filtrate recovered after isolation of the triazole
9 (1st run) was loaded with the reactants, degassed (20 min
Ar bubbling) and irradiated for 30 min. After 24 h, the solu-
tion was placed at ꢀ188C and the triazole isolated by filtra-
tion (2nd run). A third and fourth run were conducted in
the same manner. Yields of the isolated products of 89, 81,
78, and 71% were obtained for the 1st, 2nd, 3rd, and 4th
run, respectively, showing a satisfactory recyclability for 1.
As azide 5 and propargyl alcohol are both water-soluble,
the reaction was finally tested in “in water” conditions
(D2O, reaction profile followed by 1H NMR spectroscopy
given in the Figure S14) using 0.5 mol% 1 in the presence of
10 equivalents of RAME-b-CD. Both an excess of propargyl
alcohol (2 equiv) and continuous irradiation were necessary
for the reaction to proceed effectively. Under these condi-
Acknowledgements
N. Pinaud is warmly acknowledged for the recording of the 2D NOESY
and DOSY NMR spectra. The University of Bordeaux 1, the CNRS, the
Rꢀgion Aquitaine, European Research Council (FP7/2008-2013) ERC
grant agreement no. 208702, are gratefully acknowledged for their finan-
cial support.
[1] For a recent review on “photocontrolled tools”, see: C. Brieke, F.
Rohrbach, A. Gottschalk, G. Mayer, A. Heckel, Angew. Chem.
2012, 124, 8572–8604; Angew. Chem. Int. Ed. 2012, 51, 8446–8476.
16238
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16231 – 16239

A Photoreducible Copper(II)-Tren Complex
FULL PAPER
[2] For a recent review on photocontrolled catalysts, see: R. S. Stoll, S.
Hecht, Angew. Chem. 2010, 122, 5176–5200; Angew. Chem. Int. Ed.
2010, 49, 5054–5075.
M. Canipelle, S. Menuel, E. Monflier, Organometallics 2010, 29,
6668–6674; d) N. Six, A. Guerriero, D. Landy, M. Peruzzini, L. Gon-
salvi, F. Hapiot, E. Monflier, Catal. Sci. Technol. 2011, 1, 1347–1353.
[15] a) Y. L. Chow, G. E. Buono-Core, Can. J. Chem. 1983, 61, 795–800;
b) Y. L. Chow, G. E. Buono-Core, Can. J. Chem. 1983, 61, 801–808;
c) Y. L. Chow, G. E. Buono-Core, J. Am. Chem. Soc. 1982, 104,
3770–3771; d) Y. L. Chow, G. E. Buono-Core, J. Am. Chem. Soc.
1986, 108, 1234–1239.
[16] a) Y. L. Chow, G. E. Buono-Core, C. W. B. Lee, J. C. Scaiano, J. Am.
Chem. Soc. 1986, 108, 7620–7627; b) Y. L. Chow, G. E. Buono-Core,
J. Chem. Soc. Chem. Commun. 1985, 592–594.
[17] a) G. E. Buono-Core, A. H. Klahn, F. Aros, V. Astorga, Polyhedron
1996, 15, 363–366; b) G. E. Buono-Core, A. H. Klahn, C. Baha-
mondes, F. Aros, M. Tejos, V. Astorga, Inorg. Chim. Acta 1997, 257,
241–245.
[3] Selected examples: a) B. K. Keitz, R. H. Grubbs, J. Am. Chem. Soc.
2009, 131, 2038–2039; b) D. Wang, K. Wurst, W. Knolle, U. Decker,
L. Prager, S. Naumov, M. R. Buchmeiser, Angew. Chem. 2008, 120,
3311–3314; Angew. Chem. Int. Ed. 2008, 47, 3267–3270; c) L. De-
laude, M. Szypa, A. Demonceau, A. F. Noels, Adv. Synth. Catal.
2002, 344, 749–756; d) A. Fꢄrstner, L. Ackermann, Chem.
Commun. 1999, 95–96; e) M. Picquet, C. Bruneau, P. H. Dixneuf,
Chem. Commun. 1998, 2249–2250; f) A. Hafner, A. Mꢄlhebach,
P. A. van der Schaaf, Angew. Chem. 1997, 109, 2213–2216; Angew.
Chem. Int. Ed. 1997, 36, 2121–2124; g) A. L. Prignano, W. C. Tro-
gler, J. Am. Chem. Soc. 1987, 109, 3586–3595; h) J. C. Mitchener,
M. S. Wrighton, J. Am. Chem. Soc. 1981, 103, 975–977.
[4] A. Jutand, Chem. Rev. 2008, 108, 2300–2347.
[18] a) M. Melnꢅk, Coord. Chem. Rev. 1982, 42, 259–293; b) M. Contel,
P. R. Villuendas, J. Fernꢆndez-Gallardo, P. J. Alonso, J.-M. Vincent,
R. H. Fish, Inorg. Chem. 2005, 44, 9771–9778; c) A. Motreff, R. Cor-
rea da Costa, H. Allouchi, M. Duttine, C. Mathoniꢁre, C. Duboc, J.-
M. Vincent, Inorg. Chem. 2009, 48, 5623–5625.
[5] L. Harmand, S. Cadet, B. Kauffmann, L. Scarpantonio, P. Batat, G.
Jonusauskas, N. D. McClenaghan, D. Lastꢀcouꢁres, J.-M. Vincent,
Angew. Chem. 2012, 124, 7249–7253; Angew. Chem. Int. Ed. 2012,
51, 7137–7141.
[6] a) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless,
Angew. Chem. 2002, 114, 2708–2711; Angew. Chem. Int. Ed. 2002,
41, 2596–2599; b) C. W. Tornøe, C. Christensen, M. Meldal, J. Org.
Chem. 2002, 67, 3057–3064.
[7] See the themed issues: a) “Applications of click chemistry”, guest
Eds.: M. G. Finn, V. Fokin, Chem. Soc. Rev. 2010, 39, 1231–1405;
b) “Click chemistry in polymer science”, guest Ed. W. H. Binder,
Macromol. Rapid Commun. 2008, 29, 943–1185.
[8] Selected reviews: a) M. Meldal, C. W. Tornøe, Chem. Rev. 2008, 108,
2952–3015; b) V. D. Bock, H. Hiemstra, J. H. van Maarseveen, Eur.
J. Org. Chem. 2006, 51–68; c) L. Liang, D. Astruc, Coord. Chem.
Rev. 2011, 255, 2933–2945.
[9] Review focusing on the copper(I) click catalysts: S. Dꢅez-Gonzꢆlez,
Catal. Sci. Technol. 2011, 1, 166–178.
[19] Selected examples: a) V. Hong, S. I. Presolski, C. Ma, M. G. Finn,
Angew. Chem. 2009, 121, 10063–10067; Angew. Chem. Int. Ed. 2009,
48, 9879–9883; b) A. Kumar, K. Lib, C. Cai, Chem. Commun. 2011,
47, 3186–3188; c) S. I. Presolski, V. Hong. S.-H. Cho, M. G. Finn, J.
Am. Chem. Soc. 2010, 132, 14570–14576.
[20] For a recent example of a water-soluble click catalyst working in “in
water” conditions, see: C. Gaulier, A. Hospital, B. Legeret, A. F.
Delmas, V. Aucagne, F. Cisnetti, A. Gautier, Chem. Commun. 2012,
48, 4005–4007.
[21] C. Capello, U. Fischer, K. Hungerbꢄhler, Green Chem. 2007, 9, 927–
934.
[22] a) B. R. Buckley, S. A. Dann, D. P. Harris, H. Heaney, E. C. Stubbs,
Chem. Commun. 2010, 46, 2274–2276; b) S. S. Y. Chui, M. F. Y. Ng,
C.-M. Che, Chem. Eur. J. 2005, 11, 1739–1749.
[10] M. A. Tasdelen, Y. Yagci, Angew. Chem. 2013, 125, 6044–6053;
Angew. Chem. Int. Ed. 2013, 52, 5930–5938.
[23] Selected examples: a) M. Rolff, J. N. Hamann, F. Tuczek, Angew.
Chem. 2011, 123, 7057–7061; Angew. Chem. Int. Ed. 2011, 50, 6924–
6927; b) K. D. Karlin, J. C. Hayes, Y. Gultneh, R. W. Cruse, J. W.
McKown, J. P. Hutchinson, J. Zubieta, J. Am. Chem. Soc. 1984, 106,
2121–2128; c) P. L. Holland, K. R. Rodgers, W. B. Tolman, Angew.
Chem. 1999, 111, 1210–1213; Angew. Chem. Int. Ed. 1999, 38, 1139–
1142; d) O. Sander, A. HenB, C. Nꢇther, C. Wꢄrtele, M. C. Holthau-
sen, S. Schindler, F. Tuczek, Chem. Eur. J. 2008, 14, 9714–9729;
e) D. Maiti, D.-H. Lee, K. Gaoutchenova, C. Wꢄrtele, M. C. Holth-
ausen, A. A. N. Sarjeant, J. Sundermeyer, S. Schindler, K. D. Karlin,
Angew. Chem. 2008, 120, 88–91; Angew. Chem. Int. Ed. 2008, 47,
82–85.
[11] a) M. A. Tasdelen, Y. Yagci, Tetrahedron Lett. 2010, 51, 6945–6947;
b) M. A. Tasdelen, G. Yilmaz, B. Iskin, Y. Yagci, Macromolecules
2012, 45, 56–61; c) B. J. Adzima, Y. Tao, C. J. Kloxin, C. A. DeFor-
est, K. S. Anseth, C. N. Bowman, Nat. Chem. 2011, 3, 256–259; d) B.
Sandmann, B. Happ, J. Vitz, M. D. Hager, P. Burtscher, N. Moszner,
U. S. Schubert, Polym. Chem. 2013, 14, 3938–3942.
[12] a) N. Candelon, D. Lastꢀcouꢁres, A. Khadri Diallo, J. Ruiz Aran-
zaes, D. Astruc, J.-M. Vincent, Chem. Commun. 2008, 7, 41–43;
b) L. Harmand, M.-H. Lescure, N. Candelon, M. Duttine, D. Lastꢀ-
couꢁres, J.-M. Vincent, Tetrahedron Lett. 2012, 53, 1417–1420.
[13] a) P. L. Golas, N. V. Tsarevsky, B. S. Sumerlin, K. Matyjaszewski,
Macromolecules 2006, 39, 6451–6457; b) L. Liang, J. Ruiz, D.
Astruc, Adv. Synth. Catal. 2011, 353, 3434–3450.
[24] O. Das, T. K. Paine, Dalton Trans. 2012, 41, 11476–11481.
[25] a) I. Lisiecki, M. P. Pileni, J. Am. Chem. Soc. 1993, 115, 3887–3896;
b) J. C. Scaiano, K. G. Stamplecoskie, G. L. Hallett-Tapley, Chem.
Commun. 2012, 48, 4798–4808.
[14] Selected examples: a) C. Machut, J. Patrigeon, S. Tilloy, H. Bricout,
F. Hapiot, E. Montflier, Angew. Chem. 2007, 119, 3100–3102;
Angew. Chem. Int. Ed. 2007, 46, 3040–3042; b) L. Caron, H. Bric-
out, S. Tilloy, A. Ponchel, D. Landy, S. Fourmentin, E. Monflier,
Adv. Synth. Catal. 2004, 346, 1449–1456; c) J. Patrigeon, F. Hapiot,
Received: May 27, 2013
Revised: August 9, 2013
Published online: October 14, 2013
Chem. Eur. J. 2013, 19, 16231 – 16239
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16239