Influence of cyclopentadienyl ring-tilt on electron-transfer reactions: Redox-induced reactivity of strained [2] and [3] Ruthenocenophanes

Article abstract of DOI:10.1002/chem.201403512

In contrast to ruthenocene [Ru(h⁵-C₅H₅)2] and di-methylruthenocene [Ru(h ⁵-C5H₄Me)₂] (7), chemical oxidation of highly strained, ring-tilted [2]ruthenocenophane [Ru(η⁵-C₅H₄)₂(CH₂)₂] (5) and slightly strained [3]ruthenocenophane [Ru(h⁵-C₅H4)2(CH₂)3] (6) with cationic oxidants containing the non-coordinating [B(C₆F₅)₄]^- anion was found to afford stable and isolable metal-metal bonded dicationic dimer salts [Ru(h⁵-C₅H₄)₂(CH₂)₂]₂[B(C₆F5)₄]₂ (8) and [Ru(h⁵-C₅H₄)₂-(CH₂)3]2[B(C₆F₅)4]2 (17), respectively. Cyclic voltammetry and DFT studies indicated that the oxidation potential, propensity for dimerization, and strength of the resulting Ru-Ru bond is strongly dependent on the degree of tilt present in 5 and 6 and thereby degree of exposure of the Ru center. Cleavage of the Ru-Ru bond in 8 was achieved through reaction with the radical source [(CH₃)₂NC(S)S-SC(S)N(CH₃)₂] (thiram), affording unusual dimer [(CH₃)₂NCS₂Ru(η⁵-C5H₄)(η³-C₅H₄)C₂H₄]₂[B(C₆F₅)₄]₂ (9) through a haptotropic η⁵-η³ ring-slippage followed by an apparent [2 + 2] cyclodimerization of the cyclopentadienyl ligand. Analogs of possible intermediates in the reaction pathway [C₆H₅ERu(η⁵-C₅H₄)₂C₂H₄] [B(C₆F₅)₄] [E = S (15) or Se (16)] were synthesized through reaction of 8 with C₆H₅E-EC₆H₅ (E = S or Se).

Full text of DOI:10.1002/chem.201403512

DOI: 10.1002/chem.201403512
Full Paper
&
Organometallic Chemistry
Influence of Cyclopentadienyl Ring-Tilt on Electron-Transfer
Reactions: Redox-Induced Reactivity of Strained [2] and
[3]Ruthenocenophanes
Andrew D. Russell,[a] Joe B. Gilroy,[a, c] Kevin Lam,[b, d] Mairi F. Haddow,[a] Jeremy N. Harvey,*[a]
William E. Geiger,*[b] and Ian Manners*[a]
Abstract: In contrast to ruthenocene [Ru(h5-C5H5)2] and di-
methylruthenocene [Ru(h5-C5H4Me)2] (7), chemical oxidation
of highly strained, ring-tilted [2]ruthenocenophane [Ru(h5-
C5H4)2(CH2)2] (5) and slightly strained [3]ruthenocenophane
[Ru(h5-C5H4)2(CH2)3] (6) with cationic oxidants containing the
non-coordinating [B(C6F5)4]À anion was found to afford
stable and isolable metalÀmetal bonded dicationic dimer
salts [Ru(h5-C5H4)2(CH2)2]2[B(C6F5)4]2 (8) and [Ru(h5-C5H4)2-
(CH2)3]2[B(C6F5)4]2 (17), respectively. Cyclic voltammetry and
DFT studies indicated that the oxidation potential, propensi-
ty for dimerization, and strength of the resulting RuÀRu
bond is strongly dependent on the degree of tilt present in
5 and 6 and thereby degree of exposure of the Ru center.
Cleavage of the RuÀRu bond in 8 was achieved through re-
action with the radical source [(CH3)2NC(S)SÀSC(S)N(CH3)2]
(thiram), affording unusual dimer [(CH3)2NCS2Ru(h5-C5H4)(h3-
C5H4)C2H4]2[B(C6F5)4]2 (9) through a haptotropic h5–h3 ring-
slippage followed by an apparent [2+2] cyclodimerization
of the cyclopentadienyl ligand. Analogs of possible inter-
mediates in the reaction pathway [C6H5ERu(h5-C5H4)2C2H4]
[B(C6F5)4] [E=S (15) or Se (16)] were synthesized through re-
action of 8 with C6H5EÀEC6H5 (E=S or Se).
Introduction
(nꢀ2). For [n]metallocenophanes containing more than four d
electrons, significant tilt angles (a=14–338), are associated
with high-energy, distorted structures,[3] where the inherent
ring-strain can be employed as a thermodynamic driving force
for ring-opening reactions.[4]
Strained [n]metallocenophanes 1,[1] and their analogues con-
taining other p-hydrocarbon ligands,[2] are a class of organo-
metallic molecules that have received widespread attention
due to their interesting structures and reactivity. Distortion of
the normally parallel cyclopentadienyl (Cp) rings (or their p-co-
ordinated counterparts) to a tilted orientation (as defined by
a) (Figure 1) occurs when the ansa linkage is sufficiently short
[a] Dr. A. D. Russell, Prof. J. B. Gilroy, Dr. M. F. Haddow, Prof. J. N. Harvey,
Prof. I. Manners
School of Chemistry, University of Bristol, Bristol
BS81TS (UK)
Figure 1. Structural metrics of [n]metallocenophanes.
E-mail: Jeremy.Harvey@bristol.ac.uk
Ian.Manners@bristol.ac.uk
Significantly, the release of ring-strain in [n]metalloceno-
phanes 1 can be exploited in the preparation of polymetallo-
cenes 2 (Figure 2), incorporating metallocene units in the
main-chain, via ring-opening polymerization (ROP) reactions.[5]
The most extensively studied system involves strained sila[1]-
ferrocenophanes (1, M=Fe, EnRx =SiRR’), which afford polyfer-
rocenylsilanes (2, M=Fe, EnRx =SiRR’) upon ROP. The interest-
ing properties of these metallopolymers, which complement
those of their all-organic counterparts, has resulted in their use
in a diverse range of applications, including as catalyst and
magnetic ceramic precursors,[6] etch resists,[7] the redox-active
component of photonic crystal displays,[8] redox-active films,
microcapsules and microspheres,[9] and self-assembled, nano-
structured materials.[10]
[b] Prof. K. Lam, Prof. W. E. Geiger
Department of Chemistry, University of Vermont
Burlington VT 05405-0125 (USA)
E-mail: William.Geiger@uvm.edu
[c] Prof. J. B. Gilroy
Present Address: Department of Chemistry
The University of Western Ontario, 1151 Richmond Street
London, ON (Canada) N6A5B7
[d] Prof. K. Lam
Present Address: Nazarbayev University
School of Science and Technology
53 Kabanbay Batyr Ave, Block 7, Office 7107
0100000, Astana (Republic of Kazakhstan)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201403512. Experimental, crystallographic,
and computational details; additional figures.
Chem. Eur. J. 2014, 20, 1 – 13
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
congested, electron-rich permethylated [3]ruthenocenophane
[Ru(h5-C5Me4)2(CH2)3]. This species was found to undergo a re-
versible one-electron oxidation in CH2Cl2, but afforded the RuIV
solvent adduct via a two-electron oxidation, when cyclic vol-
tammetry was conducted in MeCN (Scheme 1).[20a]
Figure 2. Polymetallocene 2, ring-slipped [n]metallocenophane 3 and [1]fer-
rocenophanium ion 4+.
The relative ease of synthesis and propensity for ring-open-
ing reactions of sila[1]ferrocenophanes has made them the
focus for the majority of [n]metallocenophane research. How-
ever, through variation of the incorporated metal atom,[11] and
bridging moiety (EnRx), a wide range of other [n]metalloceno-
phanes has also been prepared.[1a–i,2d] The known reactivity
modes of [n]metallocenophanes involve a variety of bond-
cleavage pathways whose nature and location is thus dictated
by the identity of the metal and bridging moiety. For example,
cleavage of the polarized CipsoÀE bond in sila[1]ferroceno-
phanes, with initiation of ROP, can be induced via the addition
of anionic reagents such as nBuLi,[12] or transition-metal cata-
lysts.[13] Oxidative-insertion of metal fragments into this bond
has also been demonstrated.[14] Cleavage of the EÀE bridge
bond in [2]metallocenophanes has been implicated in ring-
opening reactions, including homolytic bridge cleavage[15] and
insertion of main group elements or metal fragments.[16]
Scheme 1. Redox-induced reaction of octamethyl dicarba[2]ruthenoceno-
phane with MeCN.
In this paper we report full detailed studies of the redox
chemistry of highly strained [2]ruthenocenophane 5, slightly
strained [3]ruthenocenophane 6, and the unstrained analog di-
methylruthenocene 7 (Figure 3) and describe the formation of
novel dimerization products with RuÀRu bonds for the former
two species.[25] We also include electrochemical and computa-
tional investigations into the effect of tilt on both the oxida-
tion potential and propensity for dimerization. Furthermore,
our studies reveal an additional unprecedented reactivity
mode for [n]metallocenophanes, namely h5–h3 haptotropic
ring-slippage and dimerization of the Cp ring with CÀC bond
formation.
In contrast, UV irradiation of [n]ferrocenophanes results in
a selective weakening of the MÀCp bond and a selective cleav-
age in the presence of suitable initiators (such as Na[C5H5]) en-
ables ROP under mild conditions.[17] Haptotropic h5–h1 ring-
slippage of the Cp rings has been reported for several [1]- and
[2]metallocenophanes through the addition of phosphine
donors (usually under UV irradiation) to afford species such as
3 (Figure 2).[18] Furthermore, recent investigations illustrated
how changes in the redox state of [n]metallocenophanes can
alter their ansa structure and electron distribution, as observed
through the synthesis and reactivity of a [1]ferrocenophanium
ion 4+ (Figure 2).[19] In addition, several other reactivity modes
for strained [n]metallocenophanes that involve reactivity at the
metal centre have been identified.[11d,20]
Figure 3. Dicarba[2]ruthenocenophane 5, tricarba[3]ruthenocenophane 6
and dimethylruthenocene 7.
Results and Discussion
Electrochemical Studies of [2]Ruthenocenophane 5
Synthesis and characterization of dicarba[2]ruthenocenopha-
nium dimer 8
The well-known redox properties of ferrocenes, with readily
accessible and stable redox states (FeII/FeIII), has resulted in
their employment as a standard reference for electrochemical
studies,[21] as well as many other applications.[8,9c,22] The redox
properties of strained [n]ferrocenophanes[17c,23] have also been
studied and, except in the case of highly strained examples,
are similar to those observed for their unstrained ana-
logues.[12,17b] Investigations into the redox chemistry of ruthe-
nocenes, in the presence of electrolytes containing weakly-co-
ordinating anions, have revealed behavior that is more com-
plex than that of simple ferrocene derivatives, with evidence of
further reactivity occurring upon oxidation.[24] To date the elec-
trochemical properties of strained ruthenocenophanes are un-
explored, apart from the case of the slightly strained, sterically-
The redox properties of dicarba[2]ruthenocenophane 5 [a=
29.6(5)8, CÀC bridge=1.549(9) ꢁ], prepared utilizing a previous-
ly published “fly-trap” procedure involving the reaction of Li2-
[(C5H4)2(C2H4)] with RuCl2(DMSO)4,[11a] were investigated using
cyclic voltammetry (CV) in the presence of the weakly-coordi-
nating supporting electrolyte [nBu4N][B(C6F5)4].[26] Under identi-
cal conditions, CV studies of unstrained analogue ruthenocene,
[Ru(h5-C5H5)2] (RuCp2), displayed a reversible oxidation wave
(E1/2=0.41 V vs FcH, where FcH=FeCp20/+) corresponding to
the formation of the ruthenocenium ion, [RuCp2]+.[22b] In situ
voltammetry and NMR spectroscopy demonstrated that this
cation exists in equilibrium with the metalÀmetal bonded
dimer dication, [Ru2Cp4]2+ [24d]
In contrast, CV scans of 5 [scan
.
&
&
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
trum [lmax (CH3NO2) (e/MÀ1 cmÀ1)=462 (1172) and 642
(612) nm],[28] in good agreement with the electronic spectra re-
VI
ported for other diruthenium complexes containing Ru2
cores.[29] In contrast, the spectrum obtained for dicarba[2]ru-
thenocenophane 5 (Figure S2) displays no observable maxima
over the same energy range.
The structure of 8 was confirmed by single crystal X-ray dif-
fraction (Figure 5). Dimer
8 demonstrates unprecedented
metalÀmetal bonding between [n]metallocenophane units and
the structure contains one of the few known examples of an
unbridged RuIIIÀRuIII bond,[30] although a crystallographically
characterized osmocenium dimer dication has been previously
reported.[31] The RuÀRu bond length for 8 is 2.969(7) ꢁ, a signifi-
cant increase from the average RuÀRu bond lengths reported
for both diruthenium paddlewheel complexes containing sup-
porting co-ligands and involving a number of oxidation states
(2.235–2.599 ꢁ),[32] and also in ligand-unsupported RuIÀRuI
complexes (2.677–2.904 ꢁ).[33] However, the bond is compara-
ble in length to a ligand-supported [1.1]ruthenocenophane di-
cation [2.953(1) ꢁ].[34] Interestingly, in the latter case, the close
spatial proximity at which the metals were held together in
the neutral, bimetallic precursor appears to facilitate RuIIIÀRuIII
bond formation. This type of effect cannot be operative in the
formation of dimer 8, which supports the hypothesis that the
tilted structure of dicarba[2]ruthenocenophane 5, and more
exposed Ru center, significantly favours dimer formation. The
a angles between the two sets of Cp rings are 48.18(18) and
41.83(10)8 for each metallocenophane unit respectively.[35] The
significant increase in this value from the dicarba[2]ruthenoce-
nophane precursor 5 [a=29.6(5)8],[11a] presumably reflects the
increased tilt required to accommodate RuÀRu bond forma-
tion.
Figure 4. Cyclic voltamogram of a CH2Cl2 solution of 5 (ca. 1 mm analyte,
0.05 M [nBu4N][B(C6F5)4], scan rate=0.2 VsÀ1).
rate (n) = 0.2 VsÀ1] show that it undergoes an irreversible one-
electron oxidation (Figure 4, Epa =0.05 V vs FcH) giving a single
electroactive product having an irreversible reduction at Epc =
À0.33 V vs FcH (Figure 4). Bulk anodic electrolysis (Eappl =0.4 V)
confirmed the one-electron oxidation of 5, its exclusive conver-
sion to a single product, and the subsequent quantitative re-
duction of that product (Eappl =À0.7 V) back to 5. These obser-
vations, in concert with the literature precedent for the dimeri-
zation of ruthenocene, suggested that the metalÀmetal
bonded dicarba[2]ruthenocenophanium dimer dication 8 is the
product afforded upon oxidation of 5.
Although the irreversibility of the oxidation precludes a pre-
cise determination of the half-cell potential (E1/2) for the 50/+
couple, a value of approximately 0 V vs FcH is a reasonable es-
timate. In contrast to the highly concentration and tempera-
ture-dependent voltammetry of ruthenocene,[24b,d] the redox
properties of 5 vary little with changes in the medium. Chemi-
cal reversibility of the 50/+ couple was not observed even at
relatively low analyte concentrations (e.g., 10À4 m), suggesting
formation of a dimer dication that is thermodynamically more
highly favoured than in the case of the ruthenocenium ion.
Chemical oxidation of dicarba[2]ruthenocenophane 5 was
conducted to further characterize the product identified
through the electrochemical studies. To this end, reaction of
a stoichiometric quantity of acetylferrocenium tetra(pentafluor-
ophenyl)borate [FcCOCH3][B(C6F5)4],[21] with dicarba[2]rutheno-
cenophane 5 (in CH2Cl2) resulted in the immediate generation
of a dark green precipitate (Scheme 2). Following filtration and
removal of residual solvent in vacuo, dication 8 was isolated as
a dark, microcrystalline, green
Further Reactivity of Dicarba[2]ruthenocenophanium Dimer
(8)
Reaction with thiram, Me2NC(S)SÀSC(S)NMe2
We postulated that the RuÀRu bond in dimer 8 (RuÀRu=
2.969 ꢁ) might undergo facile homolytic cleavage, potentially
through addition of a stable radical species, and thus facilitate
novel reactivity modes at the Ru metal center. Furthermore,
the literature precedent for reactivity of disulfides with Ru
complexes makes this class of compounds ideal candidates for
reactions with dimer 8.[37] We therefore investigated the reac-
tivity of 8 with dimethylcarbamothioylsulfanyl N,N-dimethylcar-
bamodithioate (thiram). A solution of thiram in CD3NO2 was
solid (74% yield).[27]
Dimer 8 was analysed by 1H
NMR and 13C spectroscopy
which observed resonances as-
signed to both the h5-C5H4 rings
and (CH2)2 bridge (see Support-
ing Information for details). Visi-
ble spectroscopic studies of 8
(Figure S1) displayed two ab-
sorption maxima in the spec-
Scheme 2. Synthesis of dimer 8 via chemical oxidation of dicarba[2]ruthenocenophane 5.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
3
&
&
These are not the final page numbers! ÞÞ

Full Paper
1.95 were assigned to the protons of the C2H4 bridge.
13C{1H} NMR spectroscopic and elemental analyses
were conducted and also found to be consistent
with the structural assignment. Definitive characteri-
zation of 9 was obtained through X-ray crystallo-
graphic analysis of a single yellow crystal, grown
from a slowly cooled solution of CH3NO2/diethyl
ether (Figure 6).
Species 9 represents one of the few examples of
a structurally characterized organometallic complex
containing an h3-coordinated cyclopentadienyl
ligand.[38] The shift from the h5 to the h3-allyl coordi-
nation mode results in folding of the internal Cp ring,
a feature consistent with other organometallic com-
plexes of this type.[39] The degree of ring folding of
the Cp ring in h3 bound organometallic complexes is
defined by w, the angle between the planes formed
by the allylic and olefinic ring fragments (see Fig-
ure S3 for details),[40] and is considerably greater in 9
Figure 5. a) Solid-state structure of dication of 8. b) Different orientation of the solid-
state structure of 8 to illustrate 908 offset of metallocene units.[36] Thermal ellipsoids
shown at 50% probability level. Hydrogen atoms, solvent molecules, and counter ions
have been omitted for clarity. Selected bond lengths [ꢁ] and angles [8]: Ru1ÀRu2
2.969(7), C1ÀC11 1.487(7), C6ÀC12 1.553(7), C11ÀC12 1.499(8): C2ÀC1ÀC11 125.2(3), C5À
C1ÀC11 126.5(4), C1ÀC11ÀC12 108.5(4), C11ÀC12ÀC6 112.7(4), C7ÀC6ÀC12 125.8(4), C10À
C6ÀC12 126.8(3).
added to a solution of 8 (also in CD3NO2) and stirred
for 16 h affording a red-brown solution. 1H NMR
spectroscopic analysis in situ indicated complete con-
sumption of the starting material. Upon work up,
a yellow crystalline solid was obtained in poor isolat-
ed yield (7%) which, when analyzed, was identified
as the unusual and unexpected dimeric species 9
(Scheme 3). Cleavage of the RuÀRu bond in 8, with
formation of a cyclobutane ring between Cp rings in
an h3 coordination mode, is involved in the formation
of 9. Although the product is
formed in a very poor isolated
yield, 1H NMR spectroscopic
analysis of the reaction solution
indicated 33% conversion to 9,
suggestive that difficulty in the
crystallisation of this species dra-
matically reduces the amount of
isolable product.
Scheme 3. Reaction of dimer 9 with thiram.
Species 9 was characterized
1
by H NMR spectroscopic analy-
sis (in CD3NO2) and displayed
a doublet of doublets of dou-
blets at d=6.13, assigned to the
central allylic proton and two
multiplets at d=5.90–5.89 and
5.77–5.75 assigned to the other
two allyl protons. Multiplets at
d=5.26–5.24 and 4.47 were as-
signed to the a and b-protons
of the h5-C5H4 rings. A doublet
of doublets at d=4.99 was as-
signed to the cyclobutane
Figure 6. Solid-state structure of the dication of 9. Thermal elipsoids at 50% probability level. Hydrogen atoms,
solvent molecules and counter ions omitted for clarity. Selected bond lengths [ꢁ] and angles [8]: Ru1ÀS1
2.3892(7), Ru1ÀS2 2.3930(7), C6ÀC12 1.498(3), C1ÀC11 1.534(3), C11ÀC12 1.535(4), C1ÀC2 1.569(3), C1ÀC2 1.593(3),
C2ÀC3 1.511(3), C3ÀC4 1.423(3), C4ÀC5 1.418(3): C1ÀC2ÀC1 90.07(16), C2ÀC1ÀC2 89.93(16), C6ÀC12ÀC11 114.4(2),
C1ÀC11ÀC12 112.73(19), C3ÀC4ÀC5 104.6(2), C8ÀC9ÀC10 109.3(2).
proton, and two overlapping singlets at d=3.17 assigned to
the CH3 groups. Finally, two doublets of doublets of doublets
at d=2.71 and 2.22 and two multiplets at 2.14–2.08 and 2.01–
(w=30.448), than in other analogous complexes (17.4–
20.08),[38a,41] potentially reflecting the steric demands involved
in formation of the cyclobutane ring.
&
&
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
The significant value for w in
9
is reflected in the bond
lengths between the Ru and the
folded Cp ring carbons, in which
the RuÀallyl bonds are consider-
ably shorter (RuÀC(3)/(4)/(5)=
2.117–2.240 ꢁ) than the RuÀC(1)
and RuÀC(2) bond lengths
(3.053–3.062 ꢁ). The bond angles
around the cyclobutane ring are
as expected for such species
(89.9–90.18). It is interesting to
note the relative regiochemistry
of the dicarba spacers, which lie
in a 1,3- arrangement about the
cyclobutane ring.
Scheme 4. Postulated mechanism of formation of dimer 9 via intermediates 10 and 11.
Due to the increased strength
of MÀCp bonds relative to those
for other p-conjugated ligands containing a C5 ring (e.g., Ind= mer, presumably to minimise steric interactions. However,
indenyl, Flu=Fluorenyl),[42] haptotropic h5 to h3 shifts of the Cp given the moderate (33%) conversion to 9, the potential for-
ligand are extremely rare. Despite being proposed as a mecha- mation of the different regioisomers and diastereoisomers of 9
nistic step in a range of organometallic reactions,[43] this haptic- cannot be excluded.[48]
ity transformation has only actually been detected in solu-
Cycloaddition reactions involving metal-coordinated p-sys-
tion,[44] or implied from X-ray crystallographic analysis of the re- tems are relatively unexplored, with the majority of examples
sulting h3-Cp bound product, on a handful of occasions.[41] In employing metal-coordinated allyl groups in cycloaddition re-
contrast, examples in the literature of h5–h3 haptotropic shifts actions to form larger p-ring systems.[49] Most significant to our
in organometallic complexes containing Cp ligands fused to observations, diruthenium species 12 undergoes a reaction
larger p-conjugated ring systems (e.g., Ind, Flu) are much more with anthracene following bromide abstraction with AgOTf. In
common, forming the basis of the “indenyl effect”, whereby addition to the expected coordination product 13, tetramer 14
the kinetics of associative ligand substitution reactions are no- is also afforded through dimerization based on the internal C6-
ticeably increased for complexes with these ligands, relative to ring (Scheme 5).[50] The authors also postulated that species 13
their Cp analogues.[43,45]
is, in fact, an intermediate in the formation of 14.
We postulate a mechanism of formation for 9 (Scheme 4) in
which homolytic cleavage of both RuÀRu and SÀS bonds facili-
tate the coordination of the dithiocarbamate with consequent
Reactivity of dimer 8 with PhEÀEPh (E=S or Se)
oxidation of the RuIII metal center affording intermediate 10. To provide some support for the proposed mechanism of for-
Intramolecular coordination of the thiocarbamate group (C=S) mation of unusual dimer 9 we attempted to synthesize ana-
might then induce the h5–h3 haptotropic shift of the Cp logues of the postulated reaction intermediate 10 (see
ring,[46] to maintain the 18-electron count of the metal to Scheme 3) by using a source of organothio radicals that were
thereby afford transient intermediate 11. Formation of 9 could incapable of chelation to the Ru centre. To this end, to a solu-
then be facilitated through a 2+2 cycloaddition of the alkene tion of dimer 8 (in CD3NO2) was added a solution of phenyl di-
moeity in the folded h3-Cp ring of 11.
sulfide (also in CD3NO2). Over the 4-day reaction the dark
The structure of 9, with the dicarba bridges coordinated in green solution gradually turned dark brown, and in situ 1H
a 1,3-arrangement about the cyclobutane ring, is consistent NMR spectroscopic analysis indicated complete conversion to
with previously reported 2+2
cycloadditions, where the head-
to-tail regioisomer is favoured
when employing an electron
rich arene.[47] Thus, the electron
donating dicarba bridge in inter-
mediate 11 may stabilize
a formal positive charge in the
alternative resonance form (Fig-
ure S4), from which the 1,3-
product is afforded. In addition,
stereoselective dimerization to
form 9 affords the exo diastereo- Scheme 5. Dimerization of anthracene upon coordination to dimer 12.[50]
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
induce h5–h3 haptotropic trans-
formations in organometallics
complexes containing coordinat-
ed p-C5 ring systems,[38a,41,44] re-
sulted in no observable reaction
with 15. However, reaction of 15
with excess PMe3 afforded
a single new phosphorus-con-
taining product, as observed by
in situ 31P NMR spectroscopy
Scheme 6. Synthesis of 15 and 16.
1
a new product. Upon work up and recrystallization, species 15
was obtained in good yield (76%) (Scheme 6).
[d=24.0 ppm]. Although, H NMR spectroscopic analysis indi-
cated the formation of a new product(s), it also suggested that
a h5–h3 haptotropic transformation had not taken place.[52]
The reaction of 15 with N-heterocyclic carbene 1,3-bis(2,6-
diisopropylphenyl)imidazol-2-ylidene (IPr) resulted in further re-
1
Species 15 was characterized by H NMR, which displayed
broad multiplets at d=7.43–7.41 and 7.17–7.14, assigned to
the phenyl protons, two multiplets at 6.69–6.68 and 5.58 as-
signed to the Cp protons and a singlet at 3.31 ppm assigned
to the CH2 backbone. 13C NMR spectroscopic and elemental
analyses were also conducted and were found to be consistent
with the assigned structure of 15. Further confirmation of the
structure of 15 was obtained through X-ray crystallographic
analysis of a single crystal formed through slow cooling a satu-
rated solution of 15 in CH2Cl2/hexane (Figure 7).
The a angle for species 15 was found to be 38.778, an in-
crease from the angle in precursor 5 [a=29.6(5)8], but a de-
crease compared to that observed in dimer 8 [a = 48.18(18)8,
48.83(10)8], suggesting that the steric strain associated with
the formation of the RuÀRu bond in 8 increases the
tilt angle. The RuÀS bond length (2.4107 ꢁ) in 15 is
similar to that in related species,[51] and the phenyl
ring is positioned about 908 with respect to the di-
carba-bridge CÀC bond, presumably to minimize
steric repulsion with the Cp rings.
1
activity as demonstrated by in situ H NMR spectroscopic anal-
ysis. The spectrum, similar to that obtained for 9, displayed
multiple signals assigned to Cp protons, and is consistent with
a species of lower symmetry, formed after a potential hapticity
shift and dimerization of the Cp ligand (see S9 for details). Un-
fortunately, attempts to grow single crystals for X-ray crystallo-
graphic analysis from the products of both this reaction, and
the reaction of 15 with PMe3, were unsuccessful.
The isolation of 15 and 16, which are analogues to the pro-
posed reaction intermediate 10, provides evidence to support
the first step in the postulated mechanism for the formation
A similar reactivity pathway to that affording 15
was also observed upon treatment of 8 with phenyl
diselenide. The analogous species 16 (Scheme 6) was
isolated as a red/brown crystalline solid in a moderate
1
yield (53%). The product was analyzed by H and 13C
Figure 7. a) Solid-state structure of cation of 15. b) Different orientation of the solid-state
structure of 15. Thermal elipsoids at 50% probability level. Hydrogen atoms and counter
anions omitted for clarity. Selected bond lengths [ꢁ] and angles [8]: Ru1ÀS1 2.4107(12),
C1ÀC11 1.507(7), C6ÀC12 1.517(7), C11ÀC12 1.529(8), S1ÀC13 1.763(5): Ru1ÀS1ÀC13
109.04(16), C1ÀC11ÀC12 108.9(5), C6ÀC12ÀC11 109.6(4).
NMR spectroscopic and elemental analyses, which
were consistent with the assigned structure. Crystals
suitable for X-ray diffraction analysis were obtained
through slow cooling a solution of CH2Cl2/hexane
(Figure 8). The alpha angle in 16 was shown to be
40.18, a slight increase compared to that in 15 and
the RuÀSe bond length of 2.5751(5) ꢁ was found to
be similar to other species containing a RuÀSe
bond.[37c]
Next, our attention turned to whether the unusual
h5 to h3 slippage of the Cp ligand could be induced
in intermediate 15, a process we propose to be ach-
ieved by chelation in the case of intermediate 10
(Scheme 4). Typically, ring-slippages of this nature are
induced through the addition of a two electron
donor, or by reduction of the metal, in both cases
forcing the Cp ligand into the lower coordination
mode to preserve the 18-electron count at the metal
center.[38b] Several attempts to add two electron
Figure 8. a) Solid-state structure of cation of 16. b) Different orientation of the solid-state
structure of 16. Thermal elipsoids at 50 % probability level. Hydrogen atoms and counter
anions omitted for clarity. Selected bond lengths [ꢁ] and angles [8]: Ru1ÀSe1 2.5751(5),
C1ÀC11 1.501(6), C6ÀC12 1.505(6), C11ÀC12 1.516(7), Se1ÀC13 1.915(4): Ru1ÀSe1ÀC13
donors, including CO and MeCN which are known to 105.07(11), C1ÀC11ÀC12 108.8(3), C6ÀC12ÀC11 109.7(4).
&
&
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
for 9 (Scheme 4). However, the inability to isolate a product
containing an h3-coordinated Cp ring from the addition of a 2-
electron donor to 15 means that no definitive evidence to sup-
port the presence of intermediate 11 in the proposed reaction
pathway could be obtained.
temperatures (À208C) as, in contrast to 5+, more complex
electrochemical behaviour is observed when the cyclic voltam-
metric analysis was conducted at ambient temperatures. Thus,
further reactivity of monocation 6+ and the formation of un-
identified byproducts in addition to 17 occurs.[55] These obser-
vations, in concert with the reported electrochemically reversi-
ble behavior of cation 6+ at fast scan rates, are consistent with
a rate of dimerization for this species that is slower than for 6+
, and a sufficiently long lifetime in solution to facilitate reactivi-
ty pathways other than dimerization.
Electrochemical Properties of Tricarba[3]ruthenocenophane
6 and Dimethylruthenocene 7
The contrasting redox behaviour observed for dicarba[2]ruthe-
nocenophane 5 relative to that reported for unstrained ruthe-
nocene, prompted further investigations into the effect of tilt
on the redox properties of ruthenocene-containing species. To
this end, electrochemical studies of less tilted tricarba[3]ruthe-
nocenophane 6 (a=14.88) and unstrained dimethylrutheno-
cene 7 (a=08) were also conducted (Figure 3). The latter spe-
cies, bearing similar Cp ring substituents to 5, can be regarded
as an untilted, acyclic analogue to 5 and 6, allowing the effects
of tilt alone on the redox properties to be assessed.
Bulk chemical oxidation of 6 was conducted in order to iso-
late dimer 17. The contrasting oxidation potentials for the two
species was considered when selecting a suitable oxidant.
Thus, for species 6 (E1/2 =0.28 V vs FcH, Epa =0.34 V vs FcH),
which oxidizes at a higher potential than 5 (Epa =0.05 V vs
FcH), acetylferrocenium tetra(pentafluorophenyl)borate (0.27 V
vs FcH), employed in the formation of 8, is not sufficiently oxi-
dizing. We therefore turned our attention to tris(4-bromophe-
nyl)aminium
tetra(pentafluorophenyl)borate
[N(C6H4Br)3]
[B(C6F5)4], (0.7 V vs FcH).[21] Upon addition of a stoichiometric
amount of [N(C6H4Br)3][B(C6F5)4] to a solution of 6 (CH2Cl2,
À208C), a brown precipitate was afforded (yield=39%).[56] 1H
NMR spectroscopic analysis revealed downfield signals for the
protons environments in species 17 relative to neutral precur-
sor 6, consistent with the observations made for dimer 8. Thus,
two triplets at d=5.35 and 5.15 were assigned to the a- and
b-Cp protons, a multiplet at 2.13–2.12 assigned to the protons
on carbons 1 and 3 of the tricarba bridge, and a multiplet at
1.97–1.92 ppm assigned to the protons of the central bridging
carbon (Figure S5). 13C NMR spectroscopic and elemental anal-
yses were also conducted and found to be consistent with the
assigned structure (see Supporting Information for details). UV/
Vis spectroscopic analysis of 17 (Figure S6) displayed two ab-
sorptions [lmax (CH3NO2) (e/mÀ1 cmÀ1)=400 (7480) and 608
(1500) nm] similar to those for both dimer 8 and the other di-
Electrochemical studies of tricarba[3]ruthenocenophane 6 and
the synthesis and characterization of tricarba[3]ruthenoceno-
phanium dimer 17
Species 6, which has previously been synthesized through re-
duction of a [3]ruthecenophan-1-one precursor,[53] and crystal-
lographically characterized,[54] was prepared in this instance
through a similar procedure to that employed for 5.[11a] Thus,
RuCl2(DMSO)4 and Li2[(h5-C5H4)2(CH2)3] were combined in THF
at À788C. Upon work up, species 6 was afforded as a pale
yellow crystalline solid in low yield (22%). 1H NMR spectroscop-
ic analysis was conducted and found to be consistent with the
data reported previously for this species.[53]
With compound 6 in hand, we investigated its electrochemi-
cal properties. Similar to 5, CV scans of 6 (n=0.2 VsÀ1, CH2Cl2)
showed that it undergoes a one electron oxidation to afford
a single electroactive product (Figure 9) although at a higher
potential (by ca. 280 mV) than
VI
ruthenium species containing Ru2 cores.[29] Finally, the obser-
vation that bulk chemical reduction of 17 (at À208C) reaffords
for [2]ruthenocenophane 5. The
oxidation was found to be an
electrochemically irreversible pro-
cess under standard electro-
chemical conditions (1 mm ana-
lyte, n=0.2 VsÀ1), and dimeriza-
tion of the monocation and for-
mation of the dicationic dimer
17 occurs rapidly. However, in
contrast to 5, which displays
temperature and concentration
independent
electrochemical
properties, an electrochemically
reversible oxidation of species 6
(E1/2 =0.28 V vs FcH) to monocat-
ion 6+ was observed at higher
scan rates (n=5 VsÀ1).
Interestingly, cation 6+ ap-
Figure 9. Cyclic voltammogram of a CH2Cl2 solution of 6 (ca. 1 mm analyte, 0.05 M [nBu4N][B(C6F5)4], scan
pears only stable at reduced rate=0.2 VsÀ1, À208C).
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
species 6 (see Supporting Information for details), consistent
with the observations for 8, strongly supports the conclusion
that dimer 17 is formed upon oxidation. Unfortunately at-
tempts to grow crystals of 17 suitable for X-ray crystallograph-
ic analysis were unsuccessful.[57]
Scheme 7. Oxidation of 7 with [N(C6H4Br)3][B(C6F5)4].
Electrochemical studies of dimethylruthenocene 7
Next our attention turned to the investigation of the electro-
chemical properties of dimethylruthenocene 7 as a direct, un-
strained analog of 5. Cyclic voltammetric studies (n=0.2 VsÀ1
)
of this species displayed contrasting electrochemical behaviour
to that observed for ruthenocenophanes 5 and 6 (Figure 10).
Similar to the case for ruthenocene,[24d] the cyclic voltammo-
gram obtained for 7 had a quasi-Nerstian shape (E1/2 =0.44 V
vs FcH) and a new reduction peak appeared at a lower poten-
tial than expected for the reduction of cation 7+. However, the
position of the peak is significantly more negative (Epc =
À1.37 V vs FcH) than expected for the reduction of a dicationic
dimer. Furthermore, the peak is more prominent at slower
scan rates, which is inconsistent with both the formation of
a dimer and the electrochemical properties of ruthenocene. In
Figure 11. Solid-state structure of cation of 18. Thermal elipsoids at 50%
probability level. Hydrogen atoms and counter anion omitted for clarity. Se-
lected bond lengths [ꢁ] and angles [8]: Ru1ÀBr1 2.5469(6), C1ÀC6 1.472(5),
Ru1ÀC5 2.198(3), Ru1ÀC2 2.200(3), Ru1ÀC3 2.254(3), Ru1ÀC4 2.221(3), Ru1À
C1 2.244(3): Ru1ÀC1ÀC6 127.0(2).
1
situ H NMR spectroscopic analysis did not allow for identifica-
tion of the product.[58]
To investigate further, chemical oxidation of 7 was conduct-
ed utilizing the same aminium oxidant as employed in the for-
mation of dimer 17. To this end, a stoichiometric amount of
[N(C6H4Br)3][B(C6F5)4] was added to a solution of 7 (CH2Cl2,
258C). No precipitate was formed and the dark blue solution
gradually turned black over the course of the 16 h reaction.
Upon work up, a black crystalline product was afforded in near
quantitative yield (96%) and was identified, upon analysis, as
[Ru(h5-C5H4Me)2Br][B(C6F5)4] (18) (Scheme 7).
The product was fully characterized by NMR spectroscopy,
which displayed downfield resonances relative to those for 7.
Thus, two triplets at d=5.73 and 5.63 were assigned to the Cp
protons and a singlet at 2.35 ppm was assigned to the methyl
groups. 13C NMR spectroscopic analysis was conducted and
found to be consistent with the assigned structure. Full struc-
tural confirmation of the product was obtained through X-ray
crystallographic analysis of a single crystal of 18, grown
through slow cooling a CH2Cl2/hexane solution (Figure 11).
The two Cp rings in 18 are tilted away from the RuÀBr bond
and the methyl substituents are positioned in a pseudo trans
configuration, presumably to reduce steric interactions. The Cp
rings adopt an almost eclipsed arrangement. We postulate
that upon oxidation, strongly electrophilic cation 7+ is prone
to undergo reactions that allow the reattainment of an 18 elec-
tron configuration, and that bromide abstraction from the by-
product [N(C6H4Br)3], is apparently the preferred reaction path-
way, rather than dimerization. Further attempts to oxidize 7
with other oxidants such as Ag[B(C6F5)4] also failed to result in
the successful isolation of either the dimer dication or the new
oxidation product formed from the electrochemical studies.[59]
Computational Investigations into the Effect of Ring-Tilt on
Oxidation Potential and Dimerization Propensity
To provide further insight into the effect of Cp ring tilt on the
propensity for dimerization of oxidized ruthenocene
containing species, computational studies were con-
ducted using the B3LYP density functional, including
dispersion corrections (see Supporting Information
for details). The geometry optimized structure of 8
was in good agreement with the structure deter-
mined from X-ray crystallography (see S21). Reported
energies are based on single-point calculations with
a larger basis set and a continuum solvent model.
The singly occupied molecular orbitals (SOMOs) for
the hypothetical monocarba[1]ruthenocenophanium
cation 19+, dicarba[2]ruthenocenophanium 5+, the
tricarba[2]ruthenocenophanium cation 6+ and the
parent ruthenocenium cation (RuCp2+) differ sub-
stantially (Figure 12). Crucially, the SOMO orbitals for
Figure 10. Cyclic voltammogram of a CH2Cl2 solution of 7 (ca. 1 mm analyte, 0.05 M
[nBu4N][B(C6F5)4], scan rate=0.5 VsÀ1, À208C).
&
&
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
19+, 5+ and 6+, which have significant d character, protrude
from the exposed Ru metal center, facilitating RuÀRu bond for-
mation. Calculations also suggest that the standard free
energy (at 298 K) for dissociation of the dimeric structure in-
creases with higher tilt angle a (Figure 12).[60] Experimentally,
the equilibrium constant for the reversible dimerization of
compounds FeCp2 and oxidized FeCp2+, the energy increased
markedly upon tilting. However the energy difference between
the two species did not change greatly upon increasing
a angle (Figure S8)[62] until very large values (a>258). This is
consistent with electrochemical studies of ferrocenophanes,
which show that their redox potential is similar to that of ferro-
cene.[23,17c] RuCp2 likewise has a marked preference for the
RuCp2 in CH2Cl2 was estimated to be 9ꢂ104 mÀ1 at 243 K,[24d]
+
+
corresponding to a dissociation free energy of 5.5 kcalmolÀ1 at
that temperature. Given the different solvent, and the many re-
maining sources of errors in the computational method, this is
in fair agreement with the calculated value of 7.3 kcalmolÀ1 at
298 K, and the greater calculated free energy of dissociation
for the dimer of 5+ (8) is consistent with its irreversible forma-
tion.
non-bent structure with a=08. In contrast, RuCp2 barely in-
creases in energy upon tilting, until a>208, so that the cation
lies lower in energy relative to the neutral species as the angle
increases (Figure S9). These results are consistent with the ob-
served reduction in oxidation potential for ruthenocene con-
taining species as a function of tilt.
Analysis of the variation in orbital energies with a for FeCp2
(Figure S10) and RuCp2 (Figure S11) provided further insight
into the experimental observations. For FeCp2 our calculations,
which are in good agreement with previously reported studies
into bent metallocenes,[3] describe the mixing of one of the oc-
cupied e2’ orbitals with the a1’ orbital, affording two orbitals of
the same symmetry. The upper of these two orbitals (4a1 =
HOMO) is raised slightly in energy upon tilting (DE=ca.
39.4 kJmolÀ1 at a=308). Qualitatively similar results are report-
ed for RuCp2, however in this case, the 4a1 orbital is raised sig-
nificantly in energy upon distortion of the metallocene to even
modest tilt angles (DE=ca. 99.8 kJmolÀ1 at a=308). These dif-
ferences can be used as a rationale for the observed electro-
chemical properties, and thus facile removal of an electron
from the 4a1 HOMO in tilted ruthenocenophanes is achieved
relative to their Fe analogues.
Calculations correctly predict that the one-electron oxidation
of 6 (0.50 V vs FcH), 5 (0.25 V vs FcH), and hypothetical 19
(À0.09 V vs FcH) will be more facile than that of RuCp2 (0.79 V
vs FcH). In addition, the difference in oxidation potentials be-
tween the Ru and Fe congeners decreases with increasing tilt
angle a, and thus are larger in the metallocenes (DE(calcd)
=
0.79 V) than for the tricarba (DE(calcd) =0.59 V), dicarba
(DE(calcd) =0.18 V) and hypothetical monocarba (DE(calcd)
À0.6 V) metallocenophanes (See Table S2).[61]
=
The E1/2 value for 5, which cannot be measured directly due
to the electrochemical irreversibility of its oxidation, can be es-
timated (ca. E1/2 =0.00 V vs FcH) from the experimental anionic
peak potential obtained (Epa =0.05 V vs FcH). The differences
between calculated (calcd) and experimental (exptl) E1/2 values
for both species
6
(E1/2(calcd)ÀE1/2(exp) =0.22 V) and RuCp2
(E1/2(calcd)ÀE1/2(exp) =0.25 V), are consistent with the electrochemi-
cal value estimated for 5 (E1/2(calcd)ÀE1/2(estimated) =0.25 V) (see
Figure 12).
Conclusion
The effect of tilt upon the energies of neutral and oxidized
metallocenes was also investigated. In the case of the iron
In summary, the redox properties of ruthenocenophanes have
been studied, representing a rare investigation into the elec-
Figure 12. SOMO representations, structural and electrochemical data for 19+, 5+, 6+ and ruthenocenium. Isodensity value=0.03 (see Supporting Informa-
tion for computational details). a) Angle taken from calculated optimized structure. B) From electrochemical measurements. c) From calculations. d) E1/2 esti-
mated from experimental Epa value.
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
9
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Mꢃller, J. Am. Chem. Soc. 2012, 134, 7924; h) R. A. Musgrave, A. D. Rus-
sell, I. Manners, Organometallics 2013, 32, 5654; i) S. Sadeh, G. Schatte,
J. Mꢃller, Chem. Eur. J. 2013, 19, 13408.
tron-transfer chemistry of non-iron metallocenophanes. Unlike
the analogous metallocenophanes of Fe and Co, which under-
go reversible one-electron oxidations, removal of an electron
from dicarba[2]ruthenocenophane 5 afforded dicationic dimer
8 through metalÀmetal bond formation. To the best of our
knowledge, this is the first example of metalÀmetal bond for-
mation between strained [n]metallocenophanes, resulting in
one of the few examples of a ligand-unsupported RuIIIÀRuIII
bond. The reactivity of dimer 8 has been explored and,
through the reaction with thiram, unusual dimer 9 has been
isolated. The formation of dimer 9 from 8 provides a rare ex-
ample of a crystallographically characterized product from
a h5–h3 haptotropic ring slip of a cyclopentadienyl ligand, with
an apparent 2+2 dimerization of the resulting alkene. Species
15 and 16, analogues of intermediate 10 in the postulated
mechanism of formation of 9, were synthesized from the reac-
tion of 8 with dichalcogenides PhEÀEPh (E=S, Se).
[2] a) A. Berenbaum, I. Manners, Dalton Trans. 2004, 2057; b) C. Elschen-
broich, F. Paganelli, M. Nowotny, B. Neumꢃller, O. Burghaus, Z. Anorg.
Allg. Chem. 2004, 630, 1599; c) M. Tamm, A. Kunst, T. Bannenberg, E.
Herdtweck, P. Sirsch, C. J. Elsevier, J. M. Ernsting, Angew. Chem. 2004,
116, 5646; Angew. Chem. Int. Ed. 2004, 43, 5530; d) P. Chadha, J. L.
Dutton, M. J. Sgro, P. J. Ragogna, Organometallics 2007, 26, 6063;
e) C. L. Lund, J. A. Schachner, J. W. Quail, J. Mꢃller, J. Am. Chem. Soc.
2007, 129, 9313; f) H. Braunschweig, M. Kaupp, C. J. Adams, T. Kupfer, K.
Radacki, S. Schinzel, J. Am. Chem. Soc. 2008, 130, 11376; g) M. Tamm,
Chem. Commun. 2008, 3089; h) H. Braunschweig, T. Kupfer, Acc. Chem.
Res. 2010, 43, 455.
[3] J. C. Green, Chem. Soc. Rev. 1998, 27, 263.
[4] a) W. Finckh, B. Z. Tang, D. A. Foucher, D. B. Zamble, R. Ziembinski, A. J.
Lough, I. Manners, Organometallics 1993, 12, 823; b) H. K. Sharma, F.
Cervantes-Lee, J. S. Mahmoud, K. H. Pannell, Organometallics 1999, 18,
399.
[5] a) V. Bellas, M. Rehahn, Angew. Chem. 2007, 119, 5174; Angew. Chem. Int.
Ed. 2007, 46, 5082; b) G. R. Whittell, M. D. Hager, U. S. Schubert, I. Man-
ners, Nat. Mater. 2011, 10, 176.
[6] a) M. Ginzburg, M. J. MacLachlan, S. M. Yang, N. Coombs, T. W. Coyle,
N. P. Raju, J. E. Greedan, R. H. Herber, G. A. Ozin, I. Manners, J. Am.
Chem. Soc. 2002, 124, 2625; b) J. Q. Lu, T. E. Kopley, N. Moll, D. Roitman,
D. Chamberlin, Q. Fu, J. Liu, T. P. Russell, D. A. Rider, I. Manners, M. A.
Winnik, Chem. Mater. 2005, 17, 2227.
[7] a) J. Y. Cheng, C. A. Ross, V. Z.-H. Chan, E. L. Thomas, R. G. H. Lammer-
tink, G. J. Vancso, Adv. Mater. 2001, 13, 1174; b) I. Korczagin, R. G. H.
Lammertink, M. A. Hempenius, S. Golze, G. J. Vancso, Adv. Polym. Sci.
2006, 200, 91; c) V. P. Chuang, J. Gwyther, R. A. Mickiewicz, I. Manners,
C. A. Ross, Nano. Lett. 2009, 9, 4364.
The redox properties of less strained tricarba[3]ruthenoceno-
phane 6 were also investigated and found to show some simi-
larities to those of 5, with RuÀRu bond formation affording
dimer 17 upon oxidation. However, evidence supporting
a lower propensity for dimerization of less tilted cation 6+ was
also detected, including electrochemical reversibility under cer-
tain conditions and the further reactivity modes of 6+ reported
at ambient temperatures.
In line with these observations, computational studies re-
vealed that the propensity for dimerization of ruthenoceno-
phanes increases with increasing tilt angle and also provided
an explanation for the lower oxidation potential for [2]rutheno-
cenophane 5 relative to [3]ruthenocenophane 6, and other
less tilted analogues.
[8] D. P. Puzzo, A. C. Arsenault, I. Manners, G. A. Ozin, Angew. Chem. 2009,
121, 961; Angew. Chem. Int. Ed. 2009, 48, 943.
[9] a) L. Espada, K. H. Pannell, V. Papkov, L. Leites, S. Bukalov, I. Suzdalev, M.
Tanaka, T. Hayashi, Organometallics 2002, 21, 3758; b) K. Kulbaba, A.
Cheng, A. Bartole, S. Greenberg, R. Resendes, N. Coombs, A. Safa-Sefat,
J. E. Greedan, H. D. H. Stçver, G. A. Ozin, I. Manners, J. Am. Chem. Soc.
2002, 124, 12522; c) Y. Ma, W.-F. Dong, M. A. Hempenius, H. Mçhwald,
G. J. Vancso, Nat. Mater. 2006, 5, 724.
[10] a) V. S. Papkov, M. V. Gerasimov, I. I. Dubovik, S. Sharma, V. V. Dementiev,
K. H. Pannell, Macromolecules 2000, 33, 7107; b) Y. Ma, W.-F. Dong, M. A.
Hempenius, H. Mçhwald, G. Vancso, Angew. Chem. 2007, 119, 1732;
Angew. Chem. Int. Ed. 2007, 46, 1702; c) D. A. Rider, K. Liu, J.-C. Eloi, L.
Vanderark, L. Yang, J. Y. Wang, D. Grozea, Z. H. Lu, T. P. Russell, I. Man-
ners, ACS Nano 2008, 2, 263; d) P. A. Rupar, L. Chabanne, M. A. Winnik, I.
Manners, Science 2012, 337, 559.
Future work in this area will investigate the influence of
ring-tilt on the redox chemistry of other metallocenophanes
and related species, in addition to investigations of the unusu-
al h5–h3 haptotropic ring slip and other reactivity modes at the
Ru metal center.
Acknowledgment
[11] a) J. M. Nelson, A. J. Lough, I. Manners, Angew. Chem. 1994, 106, 1019;
Angew. Chem. Int. Ed. Engl. 1994, 33, 989; b) H. Braunschweig, M. Gross,
K. Radacki, Organometallics 2007, 26, 6688; c) H. Braunschweig, F.
Breher, M. Kaupp, M. Gross, T. Kupfer, D. Nied, K. Radacki, S. Schinzel, Or-
ganometallics 2008, 27, 6427; d) H. Braunschweig, M. Gross, K. Radacki,
C. Rothgaengel, Angew. Chem. 2008, 120, 10127; Angew. Chem. Int. Ed.
2008, 47, 9979; e) U. F. J. Mayer, J. B. Gilroy, D. O’Hare, I. Manners, J. Am.
Chem. Soc. 2009, 131, 10382.
We thank the EPSRC for financial support (A.D.R.), the Natural
Sciences and Engineering Research Council of Canada and the
European Union Marie Curie Actions Program for postdoctoral
fellowships (J.B.G.). Work carried out at the University of Ver-
mont was generously supported by the National Science Foun-
dation, CHE-1212339.
[12] R. Rulkens, A. J. Lough, I. Manners, S. R. Lovelace, C. Grant, W. E. Geiger,
J. Am. Chem. Soc. 1996, 118, 12683.
[13] a) K. Temple, F. Jꢄkle, J. B. Sheridan, I. Manners, J. Am. Chem. Soc. 2001,
123, 1355; b) M. Kumar, A. J. Metta-Magana, K. H. Pannell, Organometal-
lics 2008, 27, 6457.
[14] K. Temple, F. Jꢄkle, A. J. Lough, I. Manners, J. Chem. Soc. Dalton Trans
1997, 711.
Keywords: electron-transfer reactions
chemistry · ruthenium
·
organometallic
[1] a) K. H. Pannell, V. V. Dementiev, L. Hong, F. Cervantes-Lee, M. T. Nguyen,
A. F. Diaz, Organometallics 1994, 13, 3644; b) D. E. Herbert, U. F. J. Mayer,
I. Manners, Angew. Chem. 2007, 119, 5152; Angew. Chem. Int. Ed. 2007,
46, 5060; c) D. M. Khramov, E. L. Rosen, V. M. Lynch, C. W. Bielawski,
Angew. Chem. 2008, 120, 2299; Angew. Chem. Int. Ed. 2008, 47, 2267;
d) C. Moser, F. Belaj, R. Pietschnig, Chem. Eur. J. 2009, 15, 12589; e) T.
Arnold, H. Braunschweig, M. Gross, M. Kaupp, R. Mꢃller, K. Radacki,
Chem. Eur. J. 2010, 16, 3014; f) B. Bagh, J. B. Gilroy, A. Staubitz, J. Mꢃller,
J. Am. Chem. Soc. 2010, 132, 1794; g) B. Bagh, G. Schatte, J. C. Green, J.
[15] a) D. E. Herbert, J. B. Gilroy, A. Staubitz, M. F. Haddow, J. N. Harvey, I.
Manners, J. Am. Chem. Soc. 2010, 132, 1988; b) J. B. Gilroy, A. D. Russell,
A. J. Stonor, L. Chabanne, S. Baljak, M. F. Haddow, I. Manners, Chem. Sci.
2012, 3, 830.
[16] a) W. Finckh, B. Z. Tang, A. J. Lough, I. Manners, Organometallics 1992,
11, 2904; b) M. Herberhold, U. Steffl, W. Milius, B. Wrackmeyer, Angew.
Chem. 1997, 109, 1545; Angew. Chem. Int. Ed. Engl. 1997, 36, 1508; c) H.
Bera, H. Braunschweig, A. Oechsner, F. Seeler, R. Sigritz, J. Organomet.
Chem. 2010, 695, 2609.
&
&
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
10
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[17] a) M. Tanabe, G. W. M. Vandermeulen, W. Y. Chan, P. W. Cyr, L. Vanderark,
D. A. Rider, I. Manners, Nat. Mater. 2006, 5, 467; b) D. E. Herbert, J. B.
Gilroy, W. Y. Chan, L. Chabanne, A. Staubitz, A. J. Lough, I. Manners, J.
Am. Chem. Soc. 2009, 131, 14958; c) D. E. Herbert, U. F. J. Mayer, J. B.
Gilroy, M. J. Lꢅpez-Gꢅmez, A. J. Lough, J. P. H. Charmant, I. Manners,
Chem. Eur. J. 2009, 15, 12234.
[18] a) T. Mizuta, Y. Imamura, K. Miyoshi, J. Am. Chem. Soc. 2003, 125, 2068;
b) Y. Imamura, K. Kubo, T. Mizuta, K. Miyoshi, Organometallics 2006, 25,
2301; c) D. E. Herbert, M. Tanabe, S. C. Bourke, A. J. Lough, I. Manners, J.
Am. Chem. Soc. 2008, 130, 4166.
[19] G. Masson, D. E. Herbert, G. R. Whittell, J. P. Holland, A. J. Lough, J. C.
Green, I. Manners, Angew. Chem. 2009, 121, 5061; Angew. Chem. Int. Ed.
2009, 48, 4961.
[20] a) K. Hashidzume, H. Tobita, H. Ogino, Organometallics 1995, 14, 1187;
b) F. Schaper, O. Wrobel, R. Schwçrer, H. H. Brintzinger, Organometallics
2004, 23, 3552.
[38] a) G. Huttner, H. H. Brintzinger, L. G. Bell, P. Friedrich, V. Bejenke, D. J.
Neugebauer, J. Organomet. Chem. 1978, 145, 329; b) J. M. O’Connor,
C. P. Casey, Chem. Rev. 1987, 87, 307.
[39] L. F. Veiros, Organometallics 2000, 19, 5549.
[40] For species 9, which does not contain an olefinic ring fragment, the
angle w is between the planes formed from the C(3)-C(4)-C(5) and C(3)-
C(2)-C(1)-C(5) ring carbons.
[41] V. Raaij, L. Zsolnai, G. Huttner, H. H. Z. Brintzinger, Z. Anorg. Allg. Chem.
Anorg. Allg. Chem. 1989, 577, 217.
[42] M. J. Calhorda, C. A. Gamelas, C. C. Rom¼o, L. F. Veiros, Eur. J. Inorg.
Chem. 2000, 2000, 331.
[43] E. Rerek, F. Basolo, Organometallics 1983, 2, 372.
[44] W. Simanko, W. Tesch, V. N. Sapunov, K. Mereiter, R. Schmid, K. Kirchner,
Organometallics 1998, 17, 5674.
[45] L. F. Veiros, M. J. Calhorda, Comments Inorg. Chem. 2001, 22, 375.
[46] Trapping of intermediate 11 was attempted through the reaction of an
equimolar amounts of dimer 8 and thiram, in the presence of five
equivalents of PhC(O)CC(O)CPh, at room temperature with stirring for
[21] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877.
[22] D. A. Durkee, H. B. Eitouni, E. D. Gomez, M. W. Ellsworth, A. T. Bell, N. P.
Balsara, Adv. Mater. 2005, 17, 2003.
1
4 d. H NMR spectroscopic analysis of the crude indicated conversion to
dimer 9. Unreacted PhC(O)CC(O)CPh was recovered upon attempted
isolation of 9 through crystallization.
[47] a) E. J. Corey, J. D. Bass, R. LeMahieu, R. B. Mitra, J. Am. Chem. Soc. 1964,
86, 5570; b) R. Huisgen, L. A. Feiler, G. Binsch, Chem. Ber. 1969, 102,
3460.
[23] a) J. K. Pudelski, D. A. Foucher, C. H. Honeyman, A. J. Lough, I. Manners,
S. Barlow, D. O’Hare, Organometallics 1995, 14, 2470; b) R. Rulkens, D. P.
Gates, D. Balaishis, J. K. Pudelski, D. F. McIntosh, A. J. Lough, I. Manners,
J. Am. Chem. Soc. 1997, 119, 10976.
[24] a) M. G. Hill, W. M. Lamanna, K. R. Mann, Inorg. Chem. 1991, 30, 4687;
b) S. Trupia, A. Nafady, W. E. Geiger, Inorg. Chem. 2003, 42, 5480; c) C. E.
Banks, K. L. Robinson, H.-P. Laing, A. W. Meredith, N. S. Lawrence, Elec-
troanal. 2007, 19, 555; d) J. C. Swarts, A. Nafady, J. H. Roudebush, S.
Trupia, W. E. Geiger, Inorg. Chem. 2009, 48, 2156; e) E. I. Rogers, N. S.
Lawrence, R. G. Compton, J. Electroanal. Chem. 2011, 657, 144.
[25] Communication: A. D. Russell, J. B. Gilroy, K. Lam, M. F. Haddow, J. N.
Harvey, W. E. Geiger, I. Manners, Chem. Eur. J. 2012, 18, 8000.
[26] W. E. Geiger, F. Barriꢆre, Acc. Chem. Res. 2010, 43, 1030.
[27] To establish the chemical reversibility of the synthesis of dimer 8, re-
duction with an excess of decamethylcolbaltocene in CD3NO2 was stud-
ied. 1H NMR spectroscopic analysis revealed complete conversion of
dimer 8 to the precursor dicarba[2]ruthenocenophane 5.
[28] The UV/Vis solvent cut-off for MeNO2 was taken to be 400 nm. For de-
tails see: G. W. C. Kaye, T. H. Laby, Tables of Physical and Chemical Con-
stants, 16edth edthe National Physics Laboratory, Middlesex, 1995.
[29] a) K. M. Kadish, L.-L. Wang, A. Thuriere, E. Van Caemelbecke, J. L. Bear,
Inorg. Chem. 2003, 42, 834; b) K. M. Kadish, R. Garcia, T. Phan, J. Wellh-
off, E. Van Caemelbecke, J. L. Bear, Inorg. Chem. 2008, 47, 11423.
[30] a) L. F. Warren, V. L. Goedken, J. Chem. Soc. Chem. Commun. 1978, 909;
b) M. B. Hursthouse, R. A. Jones, K. M. Abdul Malik, G. Wilkinson, J. Am.
Chem. Soc. 1979, 101, 4128.
[31] M. W. Droege, W. D. Harman, H. Taube, Inorg. Chem. 1987, 26, 1309.
[32] a) F. A. Cotton, T. Ren, J. L. Eglin, J. Am. Chem. Soc. 1990, 112, 3439; b) J.-
L. Zuo, E. Herdtweck, F. E. Kꢃhn, J. Chem. Soc. Dalton Trans. 2002, 1244;
c) P. Angardis, In Multiple Bonds Between Metal Atoms (Eds.: F. A. Cotton,
C. A. Murillo, R. A. Walton), 3edrd edSpringer, New York, 2005, p. 377;
d) S. K. Patra, N. Sadhukhan, J. K. Bera, Inorg. Chem. 2006, 45, 4007.
[33] a) M. O. Albers, D. C. Liles, E. Singleton, J. E. Stead, M. M. de V. Steyn, Or-
ganometallics 1986, 5, 1262; b) K.-B. Shiu, C.-H. Li, T.-J. Chan, S.-M. Peng,
M.-C. Cheng, S.-L. Wang, F.-L. Liao, M. Y. Chiang, Organometallics 1995,
14, 524; c) T. Funaioli, F. Marchetti, G. Fachinetti, Chem. Commun. 1999,
2043.
[48] Reaction of 8 with thiram to afford 9 was observed to proceed in the
absence of light, suggesting that the reaction does not proceed via
a photochemically induced 2+2 cycloaddition mechanism.
[49] a) S. F. Pedersen, R. R. Schrock, M. R. Churchill, H. J. Wasserman, J. Am.
Chem. Soc. 1982, 104, 6808; b) Z. L. Lutsenko, G. G. Aleksandrov, P. V.
Petrovskii, E. S. Shubina, V. G. Andrianov, Y. T. Struchov, A. Z. Rubezhov,
J. Organomet. Chem. 1985, 281, 349; c) T. L. Dzwiniel, N. Ektin, J. M.
Stryker, J. Am. Chem. Soc. 1999, 121, 10640; d) T. L. Dzwiniel, J. M. Stryk-
er, J. Am. Chem. Soc. 2004, 126, 9184.
[50] H. Salembier, J. Mauldin, T. Hammond, R. Wallace, E. Alqassab, M. B.
Hall, L. M. Pꢇrez, Y.-J. A. Chen, K. E. Turner, E. Bockoven, W. Brennessel,
R. M. Chin, Organometallics 2012, 31, 4838.
[51] Q.-F. Zhang, F. K. M. Cheung, W.-Y. Wong, I. D. Williams, W.-H. Leung, Or-
ganometallics 2001, 20, 3777.
[52] The new environments that were observed by 1H NMR for the Cp pro-
tons displayed characteristic splittings observed for [n]metalloceno-
phanes such as 5 and 6, with triplets assignable to both the a and
b protons. This observation suggests that the h5–h3 haptotropic trans-
formation of the Cp ring did not occur in this instance.
[53] S. Kamiyama, T. Suzuki, T. Kimura, A. Kasahara, Bull. Chem. Soc. Jpn.
1978, 51, 909.
[54] S. Ohba, Y. Saito, S. Kamiyama, A. Kasahara, Acta Crystallogr. Sect. C
1984, 40, 53.
[55] After aniodic bulk electrolysis of 6 at 258C (Eappl =0.4 V), the cyclic vol-
tammagram displayed peaks associated with the reduction of dimer 17
in addition to other unidentified oxidation products. Following bulk re-
duction (Eappl =À0.7 V), the peak corresponding to the oxidation of 6 is
reobserved, in addition to peaks for unidentified byproducts. Contrast-
ingly, when the bulk oxidation of 6 is conducted at À208C, the cyclic
voltammogram is broader (potentially resulting from a slower electron-
transfer rate), however, upon bulk reduction, electrochemical behaviour
is identical to 6, suggesting no additional reactivity of cation 6+ occurs
at reduced temperatures.
[56] Stability of monocation 6+ was found to be temperature dependant.
When the reaction was conducted at room temperature, a complex
mixture of products resulted from which the dimer could not be isolat-
ed. Conducting the reaction at À208C afforded the dimer 17 in moder-
ate yield.
[57] Single crystals of 17 were produced through slow cooling a CH3NO2/
Et2O solution of 17 to À408C. Analysis by X-ray crystallography deter-
mined a monoclinic unit cell with the parameters of a=17.7253, b=
23.4512, c=16.9187 ꢁ, b=101.6538, V=6887.80 ꢁ3. A structural solu-
tion was obtained in the monoclinic space group C2/c and consists of
1 molecule of [B(C6F5)4]À crystallizing with 1/2 molecule of 17 per asym-
metric unit with solvent occupying the voids in the lattice. Both the sol-
vent and 17 displayed serious molecular disorder making satisfactory
[34] U. T. Mueller-Westerhoff, A. L. Rheingold, G. F. Swiegers, Angew. Chem.
1992, 104, 1398; Angew. Chem. Int. Ed. Engl. 1992, 31, 1352.
[35] The differences in the a angles between the two dicarba[2]ruthenoce-
nophane units in dimer 8 appear to arise due to crystal packing effects.
[36] The torsion angles between the four Cp centroid positions and the two
Ru metal centres are as follows: 87.2(1), 88.9(1), 89.1(1), 94.8(1)8. The
variation in these values reflects slight distortions in each dicarba[2]ru-
thenocenophane.
[37] a) W. R. Cullen, A. Talaba, S. J. Rettig, Organometallics 1992, 11, 3152;
b) Y. Nishibayashi, H. Imajima, G. Onodera, Y. Inada, M. Hidai, S. Uemura,
Organometallics 2004, 23, 5100; c) E. P. L. Tay, S. L. Kuan, W. K. Leong,
L. Y. Goh, Inorg. Chem. 2007, 46, 1440; d) Y. Miyake, S. Endo, M. Yuki, Y.
Tanabe, Y. Nishibayashi, Organometallics 2008, 27, 6039.
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
11
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
refinement of 17 impossible and, for this reason, the full structure and
analysis have not been included in the manuscript.
[58] The complexity of the spectrum obtained from in situ 1H NMR spectro-
scopic analysis of the oxidation of 7, indicated that multiple products
had been formed, presumably through decomposition of highly elec-
trophilic 7+.
[59] Treatment of 7 with Ag[B(C6H3(CF3)2)4] resulted in Ag metal precipita-
tion from solution. However, in situ 1H NMR spectroscopic analysis of
the reaction did not show signals that could be assigned to any of the
proton environments in 7 (or product(s) thereof), potentially due to the
formation of a paramagnetic byproduct(s). Despite several attempts, no
oxidation product could be crystallized.
[61] Calculations were performed employing the coordinates obtained for
the geometry optimized structures of untilted FeCp2 and RuCp2, respec-
tively (a=08). These structures were then tilted by the required
a angle, the CpcentroidÀMÀCpcentroid bond angle was constrained, and the
calculations were performed. In addition, the same geometry optimized
structures were employed as starting coordinates for the calculations
+
+
for oxidized species FeCp2 and RuCp2 .
+
[62] Two total energies were calculated for the oxidized species FeCp2 and
+
RuCp2 as a function of a, corresponding to removal of an electron
from either the 4a1 or 2b1 orbitals (derived from e2’ upon tilting) (see
Figures S9 and S10). As oxidation will likely remove an electron from
the higher energy 4a1 orbital, these results are used for comparison.
[60] Three potential structures for dimer 17 were found. The structures vary
through the position of the central carbons of the tricarba bridges. Cal-
culations for dimer 17b afforded the greatest RuÀRu bond enthalpy,
and therefore this data is used as representative for dimer 17. See Fig-
ure S7 in the Supporting Information for more details.
Received: June 6, 2014
Published online on && &&, 0000
&
&
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
12
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
&
Organometallic Chemistry
A. D. Russell, J. B. Gilroy, K. Lam,
M. F. Haddow, J. N. Harvey,* W. E. Geiger,*
I. Manners*
&& – &&
Strained ruthenocenophanes undergo
dimerization following electron loss to
form stable and isolable dicationic
dimer complexes. These species under-
go homolytic cleavage of the RuÀRu
bond upon addition of radical sources,
facilitating new reactivity modes includ-
ing reactivity at the metal centre and an
unusual h5–h3 haptrotropic Cp ring slip.
Influence of Cyclopentadienyl Ring-Tilt
on Electron-Transfer Reactions: Redox-
Induced Reactivity of Strained [2] and
[3]Ruthenocenophanes
Chem. Eur. J. 2014, 20, 1 – 13
www.chemeurj.org
13
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ