Metal Complexes of a Redox-Active [1]Phosphaferrocenophane: Structures, Electrochemistry and Redox-Induced Catalysis

Article abstract of DOI:10.1002/chem.201700868

The synthesis and characterisation of several metal complexes of a redox-active, mesityl(Mes)-substituted [1]phosphaferrocenophane, FcPMes (1), are reported. Cyclic voltammetry studies on the bimetallic complexes [M(κ¹P-1)(cod)Cl] (M=Rh: 2; M=I

Full text of DOI:10.1002/chem.201700868

A Journal of
Accepted Article
Title: Metal Complexes of a Redox-Active [1]Phosphaferrocenophane:
Structures, Electrochemistry and Redox-Switchable Catalysis
Authors: Alexander Feyrer, Markus Armbruster, Karin Fink, and Frank
Breher
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201700868
Link to VoR: http://dx.doi.org/10.1002/chem.201700868
Supported by

10.1002/chem.201700868
Chemistry - A European Journal
COMMUNICATION
Metal Complexes of a Redox-Active [1]Phosphaferrocenophane:
Structures, Electrochemistry and Redox-Induced Catalysis
Alexander Feyrer,[a] Markus K. Armbruster,[b] Karin Fink,[b] and Frank Breher*[a]
Dedicated to Prof. Dieter Fenske on occasion of his 75th birthday.
Abstract: The synthesis and characterisation of several metal
complexes of redox-active, mesityl(Mes)-substituted [1]phos-
fully applied in redox-induced gold catalysis.[5] Our group has
reported on bimetallic complexes of the closely related, redox-
active C-[6] and Ge-based[7] metalloligands (VII, IX). The
electronic metal-to-metal communication, e.g. for IX, was found
to be efficient due to the close proximity of the metal atoms.
Applications in stimuli-responsive catalysis have, however, been
hampered so far due to the high moisture sensitivity of the
complexes. We have therefore turned our focus to phosphine-
based metalloligands while retaining the arrangement of the
metal atoms in the same region of space. Even though
[1]phosphaferrocenophanes[8] were studied in great detail in the
past, mainly for the synthesis of metallopolymers, investigations
of their transition metal complexes – especially the role of the
ferrocenophane unit[9] as a redox-active ligand – are rare. In
view of our interest to explore cooperative effects in multimetallic
complexes[10] we present herein our studies on their applicability
in redox-switchable/redox-induced catalysis.
a
phaferrocenophane, FcPMes (1), are reported. Cyclic voltammetry
studies on the bimetallic complexes [M(1P-1)(cod)Cl] (M = Rh: 2; M
= Ir: 4), [Rh(1P-1)2(CO)Cl] (3) and [AuCl(1P-1)] (5), in conjunction
with DFT calculations, provided indications for a good electronic
communication between the metal atoms. In order to confirm that the
ferrocenophane unit might be able to electrochemically influence the
reactivity of the coordinated transition metal, the rhodium complex 2
was employed as stimuli-responsive catalyst in the hydrosilylation of
terminal alkynes. All reactions were greatly accelerated with in situ
generated 2+ as a catalyst as compared to 2. Even more importantly,
a markedly different selectivity was observed. Both factors were
attributed to different mechanisms operating for 2 and 2+ (alternative
Chalk-Harrod and Chalk-Harrod mechanism, respectively). DFT cal-
culations revealed relatively large differences for the activation
barriers for 2 and 2+ in the reductive elimination step of the classical
Chalk-Harrod mechanism. Thus, the key to the understanding is a
cooperative “oxidatively induced reductive elimination” step, which
facilitates both a higher activity and a markedly different selectivity.
Stimuli-responsive homogeneous catalysts whose activity and/or
selectivity can be modulated by an external, non-invasive
stimulus have been the subject of many current research
activities.[1] Most efforts so far have been devoted to on–off
switching of catalytic activity. Selected examples containing a
redox-active metalloligand, which were employed in redox-
switchable/redox-induced catalysis, are depicted in Scheme 1.
Pioneering work in this area have been performed by Wrigthon
and co-workers by employing the oxidised and reduced form of I
in rhodium-catalysed hydrogenation and hydrosilylation
reactions.[2] In following years, the redox-controlled polymeri-
sation has received much attention by various groups (e.g., II –
IV, VIII).[3,4] More recently, some gold complexes featuring
redox-responsive carbene ligands (V, VI) have been success-
Scheme 1. Selected redox-switchable bimetallic complexes from the literature.
Most of them have been applied in redox-switchable/redox-induced catalysis.
[a]
Dr. A. Feyrer, Prof. Dr. F. Breher
Institute of Inorganic Chemistry
Division Molecular Chemistry
Karlsruhe Institute of Technology (KIT)
Engesserstr. 15
76131 Karlsruhe, Germany
E-mail: breher@kit.edu
Dr. M. K. Armbruster, PD Dr. K. Fink
Institute of Nanotechnology
Karlsruhe Institute of Technology (KIT)
Postfach 3630
The metalloligand 1[8g] (Scheme 2) was prepared according to
published procedures from dilithiated ferrocene TMEDA adduct
and MesPCl2 and obtained as dark-red needles after work-up.
The complexes 2, 4 and 5 were prepared by adding 1 to a
solution of the respective metal precursor at low temperatures
and obtained in good yields as orange to red needles (2 and 4)
or as dark-red crystals (5) after crystallisation (Scheme 2).
Reaction of 2 with CO in dichloromethane or toluene led to
complex 3 instead of the expected bis(carbonyl) complex (3’)
[b]
76021 Karlsruhe, Germany)
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201700868
Chemistry - A European Journal
COMMUNICATION
and was obtained as an orange solid after work-up (see also
footnote [11]). All complexes are slightly air-sensitive with 5
being additionally sensitive to light. Upon coordination to a
transition metal fragment, the 31P NMR chemical shifts are
detected at higher frequencies (δ31P = 25.4 – 37.5 ppm) as
compared to the free ligand (δ31P = 2.8 ppm; see Table S1-1,
Supporting Information) showing for the rhodium complexes
additional couplings of 1JRhP = 156.4 Hz (2) and 1JRhP = 122.7 Hz
(3).
transition metal fragment. The largest effect is observable when
pure ligand 1 (E0
= –0.01 V) is compared with the gold
= 0.44 V, Figure 2a). For the dinuclear
1/2
complex 5 (E0
1/2
complexes, anodic shifts of the first, iron-centred redox events
are observed in the series 1 < 2  4 < 5 (Table 1). This trend
nicely confirms the electronic communication between the metal
atoms in these scaffolds.
Table 1. Half-wave potentials, peak potential differences and corresponding
i
pc/ipa values of 1–5 in CH2Cl2 vs. Fc/Fc+ (internal standard); v = 100 mV/s.
E01/2 / V
-0.01
ipc/ipa
~1
Compound
Ep / mV
104
1
2
3
4
0.07 / 0.39
0.13 / 0.30
0.12a,b
68 / 106
101 / 106
96a
~0.7 / ~1
~1 / ~1
~1a
5
0.44
99
~1
a
b
v = 50 mV/s; A second redox event was observed at E0 = 0.48 V, which
1/2
presumably belongs to the Ir(I)/Ir(II) couple. However, the intensity was found
to be much lower (see Section S3).
For some complexes, e.g. 2 (Figure 2), a second redox event
was observed at E01/2(2) = 0.39 V, which was attributed to the
Rh(I)/Rh(II) couple, as previously discussed.[12] It has to be
noted, however, that we consequently had to use the electrolyte
[NBu4][Al{OC(CF3)3}4] [13] containing a weakly coordinating anion
for our studies. Even under these consitions, the first oxidation
of 2 at E01/2(1) = 0.07 V shows only quasi-reversible behaviour
with ipc/ipa  0.7 and Ep = 68 mV (see also section S3).
Scheme 2. Synthesis of 1–5.
Single crystals of all multimetallic complexes suitable for
structure determination by X-ray diffraction were obtained and
confirmed the identities of the products (Figure 1). The rhodium
and iridium atoms in 2, 3, and 4 are coordinated in the expected
square-planar geometry. The M–P distances are in accordance
with those of similar structures. The gold atom in 5 exhibits the
expected linear geometry with d(Au–P) = 222.3(4) pm in the
normal range. In all structures, the bonding motifs of the
ferrocenophane unit are not much influenced upon coordination
to
a transition metal (see section S2 of the Supporting
Information). The distances between the iron atom(s) and the
respective transition metal are very similar for 2 (460.0 pm) and
4 (458.7 pm), and are shorter in 3 (450.5 pm) and 5 (433.5 pm).
Figure 2. (a) Cyclic voltammograms of 1 (green), 5 (red) and 2 (blue) in
CH2Cl2 vs Fc/Fc+; v = 100 mV/s, Pt/[NBu4][Al{OC(CF3)3}4]/Ag. (b) Overlaid
calculated structures of q3’ (fogged background) and q3’+ (foreground) and q2
and q2+, both including calculated spin density of q3’+ and q2+ obtained at the
B3LYP/def2-TZVP level.
DFT computations were performed on the neutral and cationic
rhodium carbonyl complexes [Rh(1P-1)(CO)2Cl]0/+ (q3’ and q3’+,
Scheme 2). The structural changes upon going from q3 to q3’+
are relatively small. As expected, the spin density in q3’+ is
predominantly located on the Fe atom (Figure 2b). The
difference in the Tolman electronic parameters (TEP)[14] values
upon switching the redox-state of 1 in the complex [Rh(1P-
1)(CO)2Cl]0/+ was calculated to TEPcalc = 22 cm-1. Previous
calculations on the related metalloligand IX (Scheme 1) gave
similar values.[7] As noted before, these calculated TEP changes
for the gas-phase species overestimate the effect of changing
the redox-state of the ligand by a factor of ca. 2.[7] The
qualitative trend to decreasing the donor and/or increasing the
acceptor properties of 1 upon oxidation was further confirmed by
inspecting the neutral and oxidized form of the 1:2 complex
[Rh(1P-1)2(CO)Cl]0/+ (q3 and q3+, Scheme 2, Figure S5-2). Also
Figure 1. Molecular structures of the complexes 2–5. Selected distances (pm):
2: Fe…Rh 460.0(2), Rh–P 229.8(1); 3: Fe…Rh (av.) 450.5(1), Rh–P (av.)
231.2(1); 4: Ir…Fe 458.7(2), Ir–P 229.6(1); 5: Fe…Au 433.6(2), Au–P 222.3(4).
See also section S2 of the Supporting Information.
The electrochemical properties of all title compounds (1–5) were
probed with the aid of cyclic voltammetry. The E01/2 values for the
iron-centred oxidation (strongly) depend on the coordinated
This article is protected by copyright. All rights reserved.

10.1002/chem.201700868
Chemistry - A European Journal
COMMUNICATION
in this case, as for q2+ (Figure 2), the spin density in q3+ is
predominantly located on the Fe atoms. As expected, the
calculated CO stretching frequency considerably change from
in varying amounts (between 6 and 27%). Table 2 shows an
overview of the tested reactions alongside the collected activity
and selectivity data (cf. Scheme 3).

calc = 1970 cm-1 for q3 to calc= 1981 cm-1 for q3+.[15] The results
It has to be noted that in some cases the β-products underwent
isomerisation,[21] which led to a high yield of the respective β-(E)
alkene over time (Section S4, Supporting Information). However,
we will discuss selectivity issues on the basis of the preferred
kinetic products only. 4-methoxyphenyl acetylene was the only
alkyne from the tested ones, which was not consumed
completely (about 5% remained left) and needed an initiating
phase (see Figure S4-1). The obtained β-vinyl silanes seemingly
has a positive effect on the reaction rate. Generally, the rates
were found to be predominantly affected by the substituents on
the alkyne. A detailed analysis of the kinetics will be performed
in a separate study.
prompted us to apply the stimuli-responsive metal complexes in
redox-switchable/redox-induced catalysis.
Table 2. Hydrosilylation of terminal alkynes (cf. Scheme 3). Product
distributions are given for the time at which the substrates are consumed.
Subsequent isomerisation are not considered.
alkyne
(R’ =)
silane
HSiR3
(R =)
cat. mol
%
t
conv. β-
β-
α
poly
[min.] [%]
(Z) (E)
1
2
3
4
5
H
H
H
Me2Ph
Me2Ph
Me2Ph
Me2Ph
Me2Ph
2
1.0
1.0
0.25 130
1.0
0.25
80
5
>99
>99
>99
95
52 40b <1
7
2+
2+
2
0
0
48
32
40 14
45
28
6
32
0
0
0
25
0
27
0
65
46
35
6
0
0
4-MeO
4-MeO
90
15
65
10
2+
>99
In order to elucidate if the oxidation of the iron atom in 2 has an
effect on the reaction rates and/or the product distribution, one
equivalent (based on 2) of the acetyl ferrocenium salt
[FeCp(C5H4COMe)][Al{OC(CF3)3}4] was added to the catalyst
(mixture of 2 and the oxidation reagent are referred to as 2+).
This oxidation reagent was specifically chosen due to its redox
6
7
8
4-CF4
4-CF4
4-CF4
H
H
H
Me2Ph
Me2Ph
Me2Ph
Et3
2
1.0
1.0
0.25 105
1.0
0.5
1.0
>99c 48 44b
5
2+
2+
2
>99c
86c
98
0
12
42
0
59 41
74 14
32 <1
9
310
165
130
10
11
12
Et3
2+
2
92
>99
96
95
5
Me(OEt)2
29 43b <1
H
Me(OEt)2 2+
0.25 105
0
95
5
Determined by 1H NMR spectroscopy (no activity with pure ligand or pure
a
potential in CH2Cl2 (E0 = 0.27 V). This value lies between the
oxidising reagent; no metallopolymer formation during catalysis (31P NMR
spectroscopy); activity remains after adding a second batch of alkyne + silane).
1/2
determined half-wave potentials of 2 with E01/2(1) = 0.07 V and
E01/2(2) = 0.39 V (Figure 2) and should therefore only oxidise the
iron atom.
b
c(alkynes) = 0.02 – 0.05 mmol/ml. Isomerisation β-(Z)  β-(E) is observed
with time (Table S4-1). c The products precipitate during the reaction.
All reactions were greatly accelerated with 2+ as a catalyst as
compared to 2. The rate accelerating effect is illustrated in some
reaction profiles (Figure S4-1). For instance, the reaction of
phenyl acetylene with HSiMe2Ph was finished after only 5
minutes using 1 mol% of 2+ (80 minutes for 2, entries 1 and 2 in
Table 2). Similar conversion rates have been obtained for phenyl
acetylene and HSiMe2Ph using only 0.25 mol% of 2+ instead of 1
mol% of 2 (entries 3 and 1, Table 2; Figure S4-1). The rate
accelerating effect is slightly less pronounced for the substrate
containing the electron withdrawing CF3-substituent (entries 6-8).
The electron rich 4-MeO-C6H4CCH alkynes, however, is
converted even faster using 0.25mol% 2+ (15 min.) instead of 1
mol% 2 (90 min., entries 5 and 4). A similar trend is observed for
other silanes (entries 9–12).
The application of complexes featuring [n]metallocenophanes as
ligands in redox-switchable catalysis is – to the best of our
knowledge
– limited to one report on olefin metathesis
polymerisation catalysts based on the diaminocarbene-
[3]ferrocenophane VIII by the group of Bielawski.[16] In order to
investigate the applicability of the bimetallic complexes reported
herein in this area, we have studied the hydrosilylation[17] of
terminal alkynes and silanes using 2 and in situ generated 2+[18]
as catalysts. All reactions were performed in CD2Cl2 at room
temperature and were monitored by 1H NMR spectroscopy.
Control reactions on gram scale were also performed (section
S4).
Even more importantly, the selectivity of the catalysis
considerably changed and a very different product distribution
was obtained for 2 and 2+ (Table 2). While the reaction of
PhCCH and HSiMe2Ph catalysed by 2 led to mixture of the β-
(E)- and β-(Z)-alkene (40% and 52%), no β-(Z)-isomer is formed
with 2+ while increased polyphenyl acetylene formation is
observed. The selectivity is further increased at lower catalyst
loadings (0.25 mol%, 65% β-(E), entry 3). Although the
selectivity effect is much less pronounced for the substrate 4-
MeO-C6H4CCH (entries 4 and 5), the electron poor alkyne 4-
CF4-C6H4CCH again gave no β-(Z)-alkene with HSiMe2Ph and
1 mol% 2+ (only 12 % with 0.25 mol% 2+, entries 7 and 8). Only
for this substrate also the α product is formed in considerable
amounts (41% for 1 mol%; 14% for 0.25 mol%). The most
impressive results, however, were obtained with PhCCH and
the two silanes HSiEt3 and HSiMe(OEt)2 (entries 9–12). While 2
gave in each case a typical product distribution for Rh-catalysed
hydrosilylations, 2+ almost exclusively produced the β-(E)-alkene
(95% in each case, entries 10 and 12).
Scheme 3. Overview of the hydrosilylation reactions performed in this study
including a possible product distribution (those which have not been observed
are shown in grey).
In general, hydrosilylation reactions[19] of alkynes can lead to a
mixture of three isomers: the geminal product α and the β-(Z)
and β-(E) alkenes alongside other by-products (Scheme 3).
Neutral and cationic Rh(I) complexes are known to favour the β-
(Z) product, while the geminal product (α) is formed only in little
amounts or not at all (for a more detailed discussion of
mechanistic aspects, see below).[20] 2 was found to be an
efficient catalyst for the hydrosilylation of terminal alkynes with
periods after which 50% of the alkyne was consumed (t50)
between 17 and 58 minutes. The main products were the
respective β-(Z) and β-(E) alkenes. The competing reaction was
found to be the polymerisation of the alkyne, which was obtained
The widely accepted mechanism used to describe platinum-
catalysed hydrosilylation reactions is the Chalk-Harrod
mechanism (C.H., Scheme 4),[19,20] which involves key steps
such as the oxidative addition of silanes HSiR3, alkyne binding,
insertion of the latter into the Pt–H bond, and reductive
This article is protected by copyright. All rights reserved.

10.1002/chem.201700868
Chemistry - A European Journal
COMMUNICATION
elimination by Si–C bond formation and elimination of the β-(E)
vinylsilane. Since some occurrences, such as the formation of
alkynyl silanes (Scheme 4), cannot be explained by this
mechanism, the so-called modified Chalk-Harrod (m.C.H.)
mechanism was proposed later. In the m.C.H. mechanism, the
alkyne inserts into the M–Si bond to furnish a metal alkenyl
intermediate. The geminal α product or β-(Z) vinylsilane are
considered to form after isomerisation of the metal alkenyl
intermediate.
Overall, the observed effects in redox-induced catalysis[23] can
be explained as follows: the oxidation of the iron atom in the
metalloligand causes an additional positive charge on the
rhodium atom. Being in the formal oxidation state +3 before the
reductive elimination step, the additional charge lowers the
activation barrier to re-generate the +1 oxidation state of Rh.
This cooperative “oxidatively induced reductive elimination” step
obviously provides the key to a different catalytic cycle and thus
to both a higher catalytic activity and, in most cases, a markedly
different selectivity.
Experimental Section
Experimental Details: See the Supporting Information.
Acknowledgements
We acknowledge financial support from the DFG-funded trans-
regional collaborative research centre SFB/TRR 88 “Cooperative
effects in homo- and heterometallic complexes (3MET)”
(projects B1 and B4). We thank Wolfram Feuerstein for
calculating the g tensor of 2+.
Scheme 4. Chalk-Harrod (A) and modified Chalk-Harrod (B) mechanism for
the hydrosilylation of alkynes.
In contrast to platinum, rhodium-catalysed hydrosilylation
reactions occur preferentially along the m.C.H. mechanism and
therefore exhibits different selectivities. In a seminal paper,
Sakaki and co-workers investigated the elementary steps of both
mechanistic variants in the rhodium-catalysed hydrosilylation of
ethylene by quantum chemical calculations.[22] It was concluded
that the rate-determining step in the C.H. mechanism is the Si–C
reductive elimination (114.6 kJ mol-1 at DFT level). The ethylene
insertion into the Rh–H bond in the classical C.H. mechanism is
basically barrier-free (<1 kJ mol-1). Contrary to this, the rate-
determining step of the m.C.H. mechanism is either ethylene
insertion into the Rh–Si bond (56.5 kJ mol-1) or the oxidative
addition of the silane (65.7 kJ mol-1), depending on the level of
theory. Overall, the m.C.H. mechanism is therefore more
favourable in Rh-catalysed hydrosilylations.
Keywords: Redox-switchable catalysis, [1]ferrocenophane,
cooperative effects, bimetallic complexes, metalloligand
References
1
(a) A. M. Allgeier, C. A. Mirkin, Angew. Chem., Int. Ed. 1998, 37, 894; (b)
V. Lyaskovskyy, B. de Bruin, ACS Catal. 2012, 2, 270; (c) O. R. Luca, C.
H. Crabtree, Chem. Soc. Rev. 2013, 42, 1440; (d) V. Blanco, D. A. Leigh,
V. Marcos, Chem. Soc. Rev. 2015, 44, 5341; (e) V. K. K. Praneeth, M. R.
Ringenberg, T. R. Ward, Angew. Chem. Int. Ed. 2012, 51, 10228.
I. M. Lorkovic, R. R. Duff, M. S. Wrighton, J. Am. Chem. Soc. 1995, 117,
3617.
2
3
Reviews: (a) S. M. Guillaume, E. Kirillov, Y. Sarazin, J.-F. Carpentier,
Chem. Eur. J. 2015, 21, 7988; (b) A. J. Teator, D. N. Lastovickova, C. W.
Bielawski, Chem. Rev. 2016, 116, 1969.
4
5
Selected examples: (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) X. Wang, A. Thevenon, J. L. Brosmer, I. Yu, S. I. Khan, P.
Mehrkhodavandi, P. L. Diaconescu, J. Am. Chem. Soc. 2014, 136, 11264.
(a) L. Hettmanczyk, S. Manck, C. Hoyer, S. Hohloch, B. Sarkar, Chem.
Commun. 2015, 51, 10949; (b) S. Ibanez, M. Poyatos, L. N. Dawe, D.
́
̃
Gusev, E. Peris, Organometallics 2016, 35, 2747; (c) L. Hettmanczyk, L.
Suntrup, S. Klenk, C. Hoyer, B. Sarkar, Chem. Eur. J. 2017, 23, 576; for
closely related work, see (d) P. Neumann, H. Dib, A.-M. Caminade, E.
Hey-Hawkins, Angew. Chem. Int. Ed. 2015, 54, 311; (e) K. Arumugam,
C. D. Varnado, Jr., S. Sproules, V. M. Lynch, C.W. Bielawski, Chem.
Eur. J. 2013, 19, 10866; (f) M. Süßner, H. Plenio, Angew. Chem. Int.
Ed. 2005, 44, 6885.
Scheme 5. DFT-calculated (BP86/def-SV(P) and {B3LYP/def2-TZVPP})
reductive elimination using 2 or 2+ as catalyst (Section S5 of the Supporting
Information).
We speculated that in our case the introduction of an additional
positive charge on the metalloligand lowers the activation barrier
of the Si–C reductive elimination to such an extend that the
classical C.H. mechanism becomes favourable for the rhodium
catalyst 2+. Indeed, DFT calculations (BP86/def-SV(P) and
{B3LYP/def2-TZVPP}) provided clear evidence: the activation
barrier for the reductive elimination step was found to be
considerable smaller (64.4 kJ {66.7} mol-1) for 2+ as compared to
2 (105.5 {82.1} kJ mol-1, Scheme 5; see also Table S5-1). Our
calculations also predict that going form the starting to the
transition state both the positive charge and the spin density in
2+ is significantly shifted to the rhodium atom (q(Rh)Start = 0.13,
q(Rh)TS = 0.46; sd(Rh)Start = 0.00, sd(Rh)TS = 0.84; all for
B3LYP/def2-TZVPP).
6
(a) S. González-Gallardo, I. Kuzu, P. Ona-Burgos, T. Wolfer, C. Wang, K.
W. Klinkhammer, W. Klopper, S. Bräse, F. Breher, Organometallics 2014,
33, 941; (b) S. Styra, S. González-Gallardo, F. Armbruster, P. Ona-
Burgos, E. Moos, M. Vonderach, P. Weis, O. Hampe, A. Gren, Y.
Schmitt, M. Gerhards, F. Menges, M. Gaffga, G. Niedner-Schatteburg,
F. Breher, Chem. Eur. J. 2013, 19, 8436.
7
8
F. Walz, E. Moos, D. Garnier, R. Köppe, C. E. Anson, F. Breher, Chem.
Eur. J. 2017, 23, 1173.
Recent reviews on strained sandwich compounds: (a) D. E. Herbert, U. F.
J. Mayer, I. Manners, Angew. Chem. Int. Ed. 2007, 46, 5060; (b) M.
Tamm, Chem. Commun. 2008, 3089; (c) H. Braunschweig, T. Kupfer,
Acc. Chem. Res. 2010, 43, 455; (d) H. Bhattacharjee, J. Müller, Coord.
Chem. Rev. 2016, 314, 114. Selected examples on [1]phosphaferroceno-
phanes: (e) E.
K.
Sarbisheh, J.
C.
Green and J. Müller,
This article is protected by copyright. All rights reserved.

10.1002/chem.201700868
Chemistry - A European Journal
COMMUNICATION
Organometallics 2014, 33, 3508; (f) M. Herberhold, F. Hertel, W. Milius, B.
Wrackmeyer, J. Organomet. Chem. 1999, 582, 352; (g) T. Mizuta, Y.
Imamura, K. Miyoshi, J. Am. Chem. Soc. 2003, 125, 2068; (h) S. K. Patra,
G. R. Whittell, S. Nagiah, C.-L. Ho, W.-Y. Wong, I. Manners, Chem. Eur.
J. 2010, 16, 3240; (i) I. R. Butler, W. R. CuIlen, F. W. B. Einstein, S. J.
Rettig, A. J. Willis, Organometallics 1983, 2, 128; (j) H. Brunner, J.
Klankermayer, M. Zabel, J. Organomet. Chem. 2000, 601, 211; (k) D.
Seyferth, H. P. Withers, J. Organomet. Chem. 1980, 185, C1; (l) A. G.
Osborne, R. H. Whitely, J. Organomet. Chem. 1980, 193, 345.
9
Selected reviews on ferrocene-based ligands: (a) U. Siemeling, Eur. J.
Inorg. Chem. 2012, 3523; (b) D. Astruc, Eur. J. Inorg. Chem. 2017, 6.
10 (a) J. Chmela, M. E. Harding, D. Matioszek, C. E. Anson, F. Breher,
W. Klopper, ChemPhysChem 2016, 17, 37; (b) A. Feyrer, F. Breher,
Inorg. Chem. Front. 2017, DOI: 10.1039/c7qi00125h.
11 The reaction of 2 with CO was performed in a J. Young NMR tube. 1H and
31P NMR spectroscopic monitoring of the reaction revealed that 3 is
formed very fast; the spectra of the reaction mixture directly showed only
signals corresponding to 3 and free cod. Even though the other product
containing the missing equivalent of rhodium was not identified, it is
reasonable to assume that [Rh(CO)2Cl]2 was formed. Similar ligand
exchange processes have been observed before: See, e.g. (a) F.
Lorenzini, B. O. Patrick, B. R. James, Dalton Trans. 2007, 3224; see also
(b) S. Chen, E. Manoury, R. Poli, Eur. J. Inorg. Chem. 2014, 5820.
12 (a) D. M. Khramov, E. L. Rosen, V. M. Lynch, C. W. Bielawski, Angew.
Chem. Int. Ed. 2008, 47, 2267; see also (b) U. Siemeling, C. Färber, C.
Bruhn, Chem. Commun. 2009, 98.
13 (a) I. Raabe K. Wagner, K. Guttsche, M. Wang, M. Grätzel, G. Santiso-
Quiñonoes, I. Krossing, Chem. Eur. J. 2009, 15, 1966; (b) W. E. Geiger,
F. Barrière, Acc. Chem. Res. 2010, 43, 1030.
14 (a) C. A. Tolman, Chem. Rev. 1977, 77, 313; (b) D. J. Nelson, S. P.
Nolan, Chem. Soc. Rev. 2013, 42, 6723.
15 The computed vibrational frequencies were scaled by
F = 0.9956
(BP86/def2-SVP): M. K. Kesharwani, B. Brauer, J. M. L. Martin, J. Phys.
Chem. A 2015, 119, 1701.
16 C. D. Varnado, Jr., E. L. Rosen, M. S. Collins, V. M. Lynch, C. W.
Bielwaski, Dalton Trans. 2013, 42, 13251.
17 Another rare example on redox-switchable hydroelementation catalysis:
S. M. Shepard, P. L. Diaconescu, Organometallics 2016, 35, 2446.
18 All attempts to prepare the cationic complexes, e.g. 2+, on a preparative
scale have not yet been successful. Although EPR spectroscopic
investigations on in situ generated 2+ have been performed (cw X-band,
frozen solution, 100 K), no clear results have been obtained. While
CH2Cl2 solutions were found to be EPR silent, only a very weak EPR
signal was observed in THF, which was not reproducible. DFT
calculations on 2+ (see Section S5) indicated axial anisotropy with the
principle g values: g1 = 2.028, g2 = 2.030, g3 = 2.337. The much lower g
anisotropy of 2+ as compared to ferrocenium (Fc+, g = 1.30, g = 4.36,
see: R. Prins, F. J Reinders, J. Am. Chem. Soc. 1969, 91, 4929) can be
attributed to its lower symmetry (C2V or lower vs. D5d or D5h), which
causes a lifting of the orbital degeneracy. For a discussion on a closely
related compound, see e.g. H. Braunschweig, F. Breher, M. Kaupp, M.
Groß, T. Kupfer, D. Nied, K. Radacki, S. Schinzel, Organometallics 2008,
27, 6427.
19 (a) A. K. Roy, Adv. Organomet. Chem. 2008, 55, 1; (b) D. Troegel, J.
Stohrer, Coord. Chem. Rev. 2011, 255, 1440.
20 Selected examples: (a) M. V. Jiménez, J. J. Pérez-Torrente, M. I.
Bartolomé, V. Gierz, F. J. Lahoz, L. A. Oro, Organometallics 2008, 27,
224; (b) M. Iglesias, M. Aliaga-Lavrijsen, P. J. Sanz Miguel, F. J.
Fernández-Alvarez, J. J. Pérez-Torrente, L. A. Oro, Adv. Synth. Catal.
2015, 357, 350; (c) L. Busetto, M. C. Cassani, C. Femoni, M. Mancinelli,
A. Mazzanti, R. Mazzoni, G. Solinas, Organometallics 2011, 30, 5258.
21 M. F. Lappert, R. K. Maskell, J. Organomet. Chem. 1984, 264, 217.
22 S. Sakaki, M. Sumimoto, M. Fukuhara, M. Sugimoto, H. Fujimoto, S.
Matsuzaki, Organometallics, 2002, 21, 3788.
23 Although the term “redox-switchable catalysis” is a common description in
the literature, one referee of the manuscript argued that, in our case, the
term “redox-induced catalysis” might be more appropriate since the
reaction can be controlled in only one direction and was not/cannot be
brought back to the starting point. We highly appreciate this helpful
comment.
This article is protected by copyright. All rights reserved.

10.1002/chem.201700868
Chemistry - A European Journal
COMMUNICATION
Entry for the Table of Contents
COMMUNICATION
Alexander Feyrer, Markus Armbruster,
Karin Fink, and Frank Breher*
Page No. – Page No.
Metal Complexes of a Redox-Active
[1]Phosphaferrocenophane:
Structures, Electrochemistry and
Redox-Induced Catalysis
Redox-switchable catalysis: Bimetallic complexes of a [1]phosphaferrocenophane
are reported featuring a good electronic communication between the metal atoms.
A rhodium(I) complex (Figure) was employed as redox-active catalyst in the
hydrosilylation of terminal alkynes. Oxidatively induced reductive elimination
facilitates higher activity and markedly different selectivity for the oxidised complex
This article is protected by copyright. All rights reserved.