Redox control of a dendritic ferrocenyl-based homogeneous catalyst

Article abstract of DOI:10.1002/anie.201408314

The application of a dendrimer in a redox-switchable catalytic process is reported. A monomeric and the corresponding dendritic ferrocenylphosphane ligand were used to develop well-defined controllable catalysts with distinct redox states. The corresponding ruthenium(II) complexes catalyze the isomerization of the allylic alcohol 1-octen-3-ol. By adding a chemical oxidant or reductant, it was possible to reversibly switch the catalytic activity of the complexes. On oxidation, the ferrocenium moiety withdraws electron density from the phosphane, thereby lowering its basicity. The resulting electron-poor ruthenium center shows much lower activity for the redox isomerization and the reaction rate is markedly reduced.

Full text of DOI:10.1002/anie.201408314

Angewandte
Chemie
DOI: 10.1002/anie.201408314
Redox-Switchable Catalysis
Redox Control of a Dendritic Ferrocenyl-Based Homogeneous
Catalyst**
Paul Neumann, Hanna Dib, Anne-Marie Caminade, and Evamarie Hey-Hawkins*
Abstract: The application of a dendrimer in a redox-switch-
able catalytic process is reported. A monomeric and the
corresponding dendritic ferrocenylphosphane ligand were
used to develop well-defined controllable catalysts with distinct
redox states. The corresponding ruthenium(II) complexes
catalyze the isomerization of the allylic alcohol 1-octen-3-ol.
By adding a chemical oxidant or reductant, it was possible to
reversibly switch the catalytic activity of the complexes. On
oxidation, the ferrocenium moiety withdraws electron density
from the phosphane, thereby lowering its basicity. The resulting
electron-poor ruthenium center shows much lower activity for
the redox isomerization and the reaction rate is markedly
reduced.
such as ring-opening polymerization,[3] ring-closing metathe-
sis,[4] and the Mitsunobu reaction.[5] Thus catalysts have been
switched to change the solubility of the catalyst (for catalyst
recycling)[4b] or to modulate the activity of the transition
metal (electronic communication between the redox-active
group and the catalytic center). However, so far no reports
have involved dendritic phosphane-containing ligands
(Scheme 1).
Owing to the well-defined and hyperbranched molecular
architecture of dendrimers, the concentration and location of
the immobilized catalyst can be precisely controlled.[6]
Catalytic sites grafted to the surface of dendrimers are in
close proximity to each other, and the resulting high local
catalyst concentration may cause a “positive dendritic effect”,
that is, an increase in activity with increasing generations.[7]
Additionally, dendritic catalysts are nano-objects that can be
easily separated by precipitation or nanofiltration. Herein, we
report the first example of redox-switchable catalysis with
a dendritic catalyst.
Ferrocene is a widely used redox-active group because it
can be easily functionalized and displays a high degree of
reversibility. For our study, we employed the unsymmetrically
disubstituted 1,1’-ferrocenylphosphane 1, in which one cyclo-
pentadienyl ring is substituted with an anchoring group for
linkage to the dendrimers (Scheme 2). According to Allgeier
and Mirkin, 1 belongs to the class of substitutionally inert
redox-active ligands.[1] Monomeric ferrocenylphosphane
ligand (1-ML) was synthesized for comparison with the
dendritic ligand in catalytic testing.
The ferrocenylphosphane ligand 1 was synthesized from
1,1’-dibromoferrocene[8] through two successive Negishi
cross-coupling reactions (Scheme 2a,b), followed by depro-
tection of the phenol linker (Scheme 2c) and reduction of the
phosphane sulfide (Scheme 2d). The corresponding mono-
meric ligand 1-ML contains an anisole group instead of the
phenol substituent. The grafting experiments were performed
with a slight excess of 1 with respect to the terminal P(S)Cl2
groups of the dendrimer (Scheme 2e). The progress of the
reaction was monitored by 31P NMR spectroscopy. The
dendritic ligand was obtained by precipitation in high yield
and purity. The corresponding heterobimetallic complexes
were obtained by reaction of the monomeric (1-ML) and
dendritic (1-G1) phosphane ligands with [{Ru(p-cym)Cl2}2]
(Scheme 2i,f).
R
edox-switchable catalysis (RSC) is a field of growing
importance, in which redox-active functionality is incorpo-
rated within a ligand framework to allow the catalytic activity
of the coordinated metal centers to be influenced in situ.[1]
Oxidation and reduction influence the electron-donating
ability of the ligand and thus result in altered activity or
selectivity of the catalyst, which may facilitate a new trans-
formation altogether. The ultimate goal is to design a catalyst
that displays orthogonal activity for different substrates on
changing its electronic nature.
Wrighton et al. were the first to describe the concept of
RSC for a rhodium(I) bisphosphino cobaltocene complex.[2]
While the catalytic hydrogenation of cyclohexene is approx-
imately 16 times faster with the reduced complex, the
(chemically) oxidized complex is the faster and more durable
hydrosilylation catalyst. This original outcome was explained
by a more electron-rich rhodium center in the reduced form
promoting oxidative addition of H2.
Following the pioneering work of Wrighton et al., RSC
has been applied to several areas of homogeneous catalysis,
[*] M. Sc. P. Neumann, Prof. Dr. E. Hey-Hawkins
Institute of Inorganic Chemistry, Universitꢀt Leipzig
Johannisallee 29, 04103 Leipzig (Germany)
E-mail: hey@uni-leipzig.de
Homepage: http://www.uni-leipzig.de/chemie//hh
Dr. H. Dib, Prof. Dr. A.-M. Caminade
Laboratoire de Chimie de Coordination du CNRS
205 route de Narbonne, BP 44099, 31077 Toulouse Cedex 4 (France)
[**] This work was supported by the Europꢀischer Sozialfonds im
Freistaat Sachsen. Support from the Deutsche Forschungsgemein-
schaft (HE 1376/34-1, joint DFG-ANR project “DENDSWITCH”),
COST Action CM1302 SIPs and the Graduate School BuildMoNa is
gratefully acknowledged. We thank Chemetall GmbH and Umicore
AG & Co. KG for generous donations of chemicals.
To find a suitable chemical oxidant and reductant for the
redox switch, the complexes were investigated electrochemi-
cally. The FeII/FeIII redox potential does not change signifi-
cantly on complexation, and both FeII and RuII undergo well
separated, fully reversible one-electron redox processes
(Table 1). Both dendritic and monomeric complexes show
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201408314.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Scheme 2. Synthetic pathway to monomer ML and dendrimer G1
terminated by 1 or [1-Ru(p-cym)Cl2]: a) nBuLi, ZnCl2, [Pd(PPh3)4], 4-Br-
C6H4-O(TBDMS); b) nBuLi, ZnCl2, [Pd(PPh3)4], 4-Br-C6H4-P(S)Ph2;
c) TBAF·3H2O, THF, 3 h, RT; d) Raney nickel, toluene, MeCN, 12 h,
RT; e) 14 equiv 1, G1, Cs2CO3, THF, 12 h, RT; f) 1-G1, 6 equiv [{Ru(p-
cym)Cl2}2], CH2Cl2, 2 h, RT; g) nBuLi, ZnCl2, [Pd(PPh3)4], 4-Br-C6H4-
OMe; h) Raney nickel, toluene, MeCN, 72 h, 808C; i) 1-ML, 0.5 equiv
[{Ru(p-cym)Cl2}2], CH2Cl2, 2 h, RT; TBDMS=tert.-butyldimethylsilyl,
TBAF=tetra-n-butylammonium fluoride, ML=monomeric ligand,
G1 =1st generation dendrimer, p-cym=p-cymene.
Table 1: Electrochemical data for 1-ML, 2-ML, 1-G1, and 2-G1.[a]
Compound
E1/2 (FeII/FeIII) [V]
E1/2 (RuII/RuIII) [V]
1-ML
2-ML
1-G1
0.59
0.57
0.59
0.58
–
1.19
–
2-G1
1.27
[a] Data obtained from CV and square-wave voltammetry in THF with
0.1m [nBu4N][PF6] electrolyte and referenced against SCE.
oxidize all of the ferrocene units but does not interfere with
the RuII redox process.
The transition metal catalyzed isomerization of allylic
alcohols, which is a useful and atom-economical synthetic
process en route to saturated carbonyl compounds,[10] was
chosen to illustrate the use of metallodendritic catalysts in
RSC. As a result of the well-known ability of transition metals
to facilitate the migration of carbon–carbon double bonds, the
allylic alcohol is turned into an enol, which then readily
isomerizes into the corresponding carbonyl compound.[11]
Besides rhodium(I), ruthenium(II) is predominantly used
for this transformation.[12] This reaction has already been
applied in numerous multistep syntheses,[13] as well as with
metallodendritic catalysts.[14] Complexes such as [(h6-are-
ne)RuCl2(L)], in which L is a monodentate phosphane ligand
(PPh3 or analogues), show particularly high activity.[12,15]
Scheme 1. Redox reaction of the dendritic catalyst 2-G1 (green=active
catalyst; red=poorly active ligand).
similar redox behaviors, with a potential window between 0.6
and 1.1 V for a prospective chemical oxidant. It is also
noteworthy that the dendritic complexes only show a single
FeII redox wave. Therefore, the structurally flexible dendri-
mer undergoes fast rotational changes in solution that ensure
that all 12 ferrocene moieties are oxidized on the electro-
chemical timescale. Of the chemical oxidants studied, [Fe{h5-
C5H4C(O)Me}Cp][BF4][9] was found to be the most promising
candidate for RSC (E1/2 = 0.78 V vs. SCE in THF) since it can
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Since a poorly soluble base is used for activation,[16] an
induction period is observed that corresponds to formation of
the active species.[12] To minimize this effect and obtain more
easily comparable results, the catalyst was preheated with
Cs2CO3 in THF for 5 min at 608C. Both 2-ML and 2-G1 show
high catalytic activity in the conversion of 1-octen-3-ol to 3-
octanone, but the dendritic catalyst 2-G1 shows higher activity
than 2-ML (60 min, TOF 400) and leads to full conversion
after 40 min (TOF 600). This outcome can be explained by the
increased stability of the dendritic catalyst or by cooperative
effects of neighboring catalytic entities on substrate migration
or the stabilization of transition states.
One equivalent of [Fe{h5-C5H4C(O)Me}Cp][BF4] was added,
and subsequently the activity of the dendrimer dropped
significantly (Figure 1). In contrast to 2-ML, the oxidized
species [2-G1][BF4]12 partly precipitates from the reaction
mixture, as was anticipated given its high positive charge.
However, [2-G1][BF4]12 is still slightly soluble in THF,
showing 8% of the original activity of 2-G1. On reduction
with [FeCp*2], 2-G1 was reformed and a significant increase in
catalytic activity was observed. In the case of the dendritic
redox event, a combination of electronic communication and
an additional solubility effect is assumed. Figure 2 illustrates
After examining the catalytic properties of 2-ML and 2-
G1, the monomeric complex 2-ML was used in RSC
(Figure 1). On oxidation with 1 equiv of [Fe{h5-
Figure 2. Comparison of the experimental rate constants kobs derived
from linear regression of the RSC sectors A, B, and C: redox-switched
isomerization of 1-octen-3-ol catalyzed by 2-ML (left) and 2-G1 (right).
the difference in reaction kinetics between the monomeric
and dendritic redox events. The plots of conversion versus
time were classified into three RSC sectors: A (before
oxidation), B (after oxidation), and C (after reduction). The
experimental rate constants kobs could be observed by linear
regression of the sectors (see the Supporting Information). In
sector A, the rate constant of this pseudo-first-order reaction
is increased six-fold by using the dendritic catalyst. In both
cases, the decrease in the reaction rate upon oxidation is
remarkable considering that only one electron is removed
from the conjugated p system. The similarity of the rate
constants in sectors C and A proves the reversibility of both
redox events. These outcomes are interesting firstly for the
investigation of orthogonal activity by using different sub-
strates in the same reaction and secondly for facile catalyst
recovery (by simple filtration).[4b]
In conclusion, the first application of a dendritic transition
metal complex in redox-switchable catalysis and the first
example of a redox-switchable isomerization of an allylic
alcohol were achieved by employing a phosphane ligand that
contains a redox-active ferrocene moiety closely connected to
the catalytic center through a conjugated p system. As
expected, the dendritic catalyst 2-G1 partly precipitates on
oxidation, and this observation suggests that loss of catalytic
activity could be partly due to reduced solubility. However,
the oxidized monomeric catalyst [2-ML][BF4] and some of the
oxidized dendrimer [2-G1][BF4]12 remain in solution. Thus,
the reversible nature of the redox event is clearly derived
Figure 1. Plot of conversion versus time: redox-switched isomerization
of 1-octen-3-ol catalyzed by 2-ML (black squares) and 2-G1 (blue
triangles); monitored by GC–MS; conditions: 0.25 mol% ruthenium,
0.625 mol% Cs2CO3 in THF at 608C (preheating); Ox: 1 equiv [Fe{h5-
C5H4C(O)Me}Cp][BF4]; Red: 1 equiv [FeCp*2] (Cp*=C5Me5).
C5H4C(O)Me}Cp][BF4] the reaction rate was markedly
reduced. After reduction with [FeCp*2] (Cp* = C5Me5),[9]
full conversion was observed with a degree of activity similar
(96%) to that of the original catalyst before oxidation. Since
the oxidized catalyst does not precipitate from the reaction
mixture, these findings, in accord with Wrighton’s previous
work,[2] clearly demonstrate an electronic effect of the redox-
active ligand. Once the ferrocene is oxidized, FeIII withdraws
electron density from the phosphane and thus decreases its
basicity. Consequently, the decrease in electron density in the
À
Ru P bond creates an electron-poor transition metal and thus
a far less active catalyst for the isomerization of allylic
alcohols. In this manner, it is possible to reversibly tune the
electronic properties of the ruthenium catalyst in situ.
To demonstrate that RSC can also be applied to den-
drimers, it is necessary to investigate the reversibility of the
dendritic redox event during the reaction. Therefore, the
experiment was repeated with the dendritic complex 2-G1.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[8] T.-Y. Dong, L.-L. Lai, J. Organomet. Chem. 1996, 509, 131.
[9] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877.
[10] For reviews: a) R. C. van der Drift, E. Bouwman, E. Drent, J.
Organomet. Chem. 2002, 650, 1; b) R. Uma, C. Crꢁvisy, R. Grꢁe,
Chem. Rev. 2003, 103, 27; c) V. Cadierno, P. Crochet, J. Gimeno,
Synlett 2008, 1105; d) L. Mantilli, C. Mazet, Chem. Lett. 2011, 40,
341; e) N. Ahlsten, A. Bartoszewicz, B. Martꢂn-Matute, Dalton
Trans. 2012, 41, 1660; f) P. Lorenzo-Luis, A. Romerosa, M.
Serrano-Ruiz, ACS Catal. 2012, 2, 1079; g) J. Garcꢂa-ꢃlvarez,
S. E. Garcꢂa-Garrido, P. Crochet, V. Cadierno, Curr. Top. Catal.
2012, 10, 35.
from electronic communication of the redox-active phos-
phane ligand and the catalytically active transition metal,
ruthenium(II). These findings extend the field of stimuli-
responsive catalysts to redox-switchable phosphane ligands.
Redox control not only enables a better understanding of
catalytic mechanisms but also facilitates the development of
catalysts with in situ orthogonal activity for different sub-
strates. Moreover, dendritic catalysts can bridge the gap
between homogeneous and heterogeneous catalysis.
[11] Mechanistic studies: a) B. M. Trost, R. J. Kulawiec, J. Am. Chem.
Soc. 1993, 115, 2027; b) D. V. McGrath, R. H. Grubbs, Organo-
metallics 1994, 13, 224; c) V. Branchadell, C. Crꢁvisy, R. Grꢁe,
Chem. Eur. J. 2003, 9, 2062; d) V. Cadierno, S. E. Garcꢂa-
Garrido, J. Gimeno, A. Varela-ꢃlvarez, J. A. Sordo, J. Am.
Chem. Soc. 2006, 128, 1360; e) N. Ahlsten, B. Martꢂn-Matute,
Adv. Synth. Catal. 2009, 351, 2657; f) M. Batuecas, M. A.
Esteruelas, C. Garcꢂa-Yebra, E. Onate, Organometallics 2010,
29, 2166.
Received: August 17, 2014
Published online: && &&, &&&&
Keywords: dendrimers · ferrocene · homogeneous catalysis ·
.
phosphanes · redox-switchable catalysis
[1] A. M. Allgeier, C. A. Mirkin, Angew. Chem. Int. Ed. 1998, 37,
894; Angew. Chem. 1998, 110, 936.
[12] P. C. L. Menꢁndez-Rodrꢂguez, V. Cadierno, J. Mol. Catal. A
2013, 366, 390.
[2] I. M. Lorkovic, R. R. Duff, M. S. Wrighton, J. Am. Chem. Soc.
1995, 117, 3617.
[13] a) M. Ito, S. Kitahara, T. Ikariya, J. Am. Chem. Soc. 2005, 127,
6172; b) S. Bovo, A. Scrivanti, M. Bertoldini, V. Beghetto, U.
Matteoli, Synthesis 2008, 2547; c) N. Tanaka, T. Suzuki, T.
Matsumura, Y. Hosoya, M. Nakada, Angew. Chem. Int. Ed. 2009,
48, 2580; Angew. Chem. 2009, 121, 2618; d) A. Bouziane, T.
Regnier, F. Carreaux, B. Carboni, C. Bruneau, J.-L. Renaud,
Synlett 2010, 207; e) G. Sabitha, S. Nayak, M. Bhikshapathi, J. S.
Yadav, Org. Lett. 2011, 13, 382.
[3] a) C. K. A. Gregson, V. C. Gibson, N. J. Long, E. L. Marshall,
P. J. Oxford, A. J. P. White, J. Am. Chem. Soc. 2006, 128, 7410;
b) E. M. Broderick, N. Guo, T. Wu, C. S. Vogel, C. Xu, J. Sutter,
J. T. Miller, K. Meyer, T. Cantat, P. L. Diaconescu, Chem.
Commun. 2011, 47, 9897; c) E. M. Broderick, N. Guo, C. S.
Vogel, C. Xu, J. Sutter, J. T. Miller, K. Meyer, P. Mehrkhoda-
vandi, P. L. Diaconescu, J. Am. Chem. Soc. 2011, 133, 9278.
[4] a) K. Arumugam, C. D. Varnado, Jr., S. Sproules, V. M. Lynch,
C. W. Bielawski, Chem. Eur. J. 2013, 19, 10866; b) M. Sꢀßner, H.
Plenio, Angew. Chem. Int. Ed. 2005, 44, 6885; Angew. Chem.
2005, 117, 7045.
[14] P. Servin, R. Laurent, L. Gonsalvi, M. Tristany, M. Peruzzini, J.-P.
Majoral, A.-M. Caminade, Dalton Trans. 2009, 4432.
ˇ
ˇ
ˇ
ˇ
[15] J. Schulz, I. Cꢂsarovꢄ, P. Stepnicka, Eur. J. Inorg. Chem. 2012,
5000.
[16] a) J. E. Bꢅckvall, U. Andreasson, Tetrahedron Lett. 1993, 34,
5459; b) R. Uma, M. K. Davies, C. Crevisy, R. Grꢁe, Eur. J. Org.
Chem. 2001, 3141; c) V. Cadierno, S. E. Garcia-Garrido, J.
Gimeno, Chem. Commun. 2004, 232.
[5] C. A. Fleckenstein, H. Plenio, Adv. Synth. Catal. 2006, 348, 1058.
[6] A.-M. Caminade, C.-O. Turrin, R. Laurent, A. Ouali, B.
Delavaux-Nicot, Dendrimers—Towards Catalytic, Material and
Biomedical Uses, Wiley, Chichester, UK, 2011.
[7] a) B. Helms, J. M. J. Frꢁchet, Adv. Synth. Catal. 2006, 348, 1125;
b) A. M. Caminade, A. Ouali, R. Laurent, C. O. Turrin, J. P.
Majoral, Chem. Soc. Rev., DOI: 10.1039/C4CS00261J.
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Redox-Switchable Catalysis
Under control: Both monomeric and
dendritic (see picture) ferrocenylphos-
phane ruthenium(II) catalysts were
reversibly switched off and on by chem-
ical oxidation and reduction during iso-
merization of the allylic alcohol 1-octen-3-
ol. This outcome is mainly due to elec-
tronic communication between the redox-
active unit and the catalytic center. Such
redox control could facilitate the devel-
opment of catalysts with orthogonal
activity for different substrates.
P. Neumann, H. Dib, A.-M. Caminade,
E. Hey-Hawkins*
&&&& — &&&&
Redox Control of a Dendritic Ferrocenyl-
Based Homogeneous Catalyst
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!