Quantitative conformational study of redox-active [2]rotaxanes, part 2: Switching in flexible and rigid bistable [2]rotaxanes

Article abstract of DOI:10.1002/chem.200701384

Translational movement of the macrocycle in two structurally similar bistable [2]rotaxanes, which is induced by a four-step electrochemical process in solution, has been investigated by using a methodology developed in the preceding article (Chem. Eur. J.

Full text of DOI:10.1002/chem.200701384

FULL PAPER
DOI: 10.1002/chem.200701384
Quantitative Conformational Studyof Redox-Active [2]Rotaxanes, Part 2:
Switching in Flexible and Rigid Bistable [2]Rotaxanes
Kirill Nikitin,* Elena Lestini, Jacek K. Stolarczyk,* Helge Müller-Bunz, and
Donald Fitzmaurice[a]
Abstract: Translational movement of
the macrocycle in two structurally simi-
lar bistable [2]rotaxanes, which is in-
duced by a four-step electrochemical
process in solution, has been investigat-
ed by using a methodology developed
in the preceding article (Chem. Eur. J.
2008, 14, 1107–1116). Both [2]rotaxanes
contain a crown ether that can be ac-
commodated by either of two intercon-
nected viologen recognition sites.
These sites are substantially different
in terms of their affinity towards the
crown ether and they possess consider-
ably different electrochemical reduc-
tion potentials. The two [2]rotaxanes
differ in the length and the rigidity of a
bridge that links these sites. A combi-
nation of molecular mechanics model-
ling and NOE spectroscopy data pro-
vides information about the conforma-
tions of both [2]rotaxanes in the parent
oxidation state when the crown ether
exclusively populates the strong recog-
nition site. To determine the popula-
tion of the recognition sites at subse-
quent stages of reduction, a paramag-
netic NMR technique and cyclic vol-
tammetry were used. The key finding is
that the flexibility of the connecting
bridge element between the recogni-
tion sites interferes with shuttling of
the crown ether in [2]rotaxanes. It can
be demonstrated that the more flexible
trimethylene bridge is folded, thus lim-
iting the propensity of the crown ether
to shuttle. Consequently, the crown
ether populates the original site even
in the second reduced state of the flexi-
ble [2]rotaxane. On the contrary, in the
[2]rotaxane in which two viologen sites
are connected by a larger and more
rigid p-terphenylene bridge, the pre-
dominant location of the crown ether
at the weak recognition site is achieved
after just one single electron reduction.
Keywords: conformation analysis ·
cyclic voltammetry
·
molecular
modeling · rotaxanes · supramolec-
ular chemistry
Introduction
Functional supermolecules, macromolecules and closely
related interlocked molecules are attractive nanoscale com-
ponents and have been extensively studied in solution and
at surfaces.[4–7] These nanoscale components have been suc-
cessfully attached to various nanoparticles and nanostruc-
tured electrodes.[8] The properties of such systems can be
tuned by controlling the size and composition of the nano-
particles or contact surfaces. However, the functionality de-
pends in particular on the ability to control the orientation
and relative positions of the components of these nanoscale
systems. For axle-based systems, such as switchable [2]ro-
taxanes, it is therefore important to ensure that the axle re-
mains normal to the contact surface, the recognition sites lo-
cated on the axle do not interfere with each other and the
recognition sites are restricted from approaching the contact
surface. Similar arguments can be made for molecular
switch tunnel junctions,[2e] in which the rotaxane molecules
should remain stretched between the surfaces of the electro-
des. These conditions effectively require the axle of the
Self-assembled nanoscale architectures emerged over
a
decade ago and continue to inspire and motivate researchers
in many fields.[1] Potential applications of nanoscale archi-
tectures are particularly promising in the field of microelec-
tronics.[2] It is envisaged that the self-assembly of such archi-
tectures in solution and at surfaces can be effectively carried
out from nanoscale molecular and condensed-phase compo-
nents.[1,3]
[a] Dr. K. Nikitin, E. Lestini, Dr. J. K. Stolarczyk, Dr. H. Müller-Bunz,
Prof. D. Fitzmaurice
School of Chemistry and Chemical Biology
University College Dublin
Belfield, Dublin 4 (Ireland)
Fax: (+353)1-716-2127
E-mail: kirill.nikitin@ucd.ie
jacek.stolarczyk@ucd.ie
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2008, 14, 1117 – 1128
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1117

[2]rotaxane to be rigid, in particular, in the bridging section
between the recognition sites.
cally address and switch the above [2]rotaxane at the surface
of a TiO2 nanoparticle or at the surface of a nanostructured
titania electrode, but the process was accompanied by large
conformational changes to the molecule.[12]
The orientation of the axle and the stability of attachment
has been addressed by using tripodal linkers to adsorb the
nanoscale components at the surface of semiconductors or
metals.[9] To this end, we have investigated tripodal [2]ro-
taxanes (Figure 1) that were attached to the surface of TiO2
nanoparticles and nanostructured electrodes by a tridentate
phosphonic acid linker L.[10–12] It was possible to electroni-
In our preceding paper (Part 1)[13a] we presented a novel
methodology to quantitatively study the conformations and
co-conformations[13b] of [2]rotaxanes in solution and in dif-
ferent oxidation states. We combined molecular modelling
with NOESY, paramagnetic line-broadening and suppres-
sion spectroscopy (PASSY),[13c] electrochemical and refer-
ence crystallographic data to analyse a model, single-station
[2]rotaxane. In this paper we apply the same methodology
to study two closely related switchable tripodal [2]rotaxanes
(1 and 3, Figure 1) in solution. The study is directed towards
the detailed characterisation of the switching process and its
dependence on the overall conformation of the [2]rotaxanes.
A general representation of a switchable tripodal [2]ro-
taxane is given in Figure 1a. Two different viologen recogni-
tion sites (V1 and V2) are interconnected by a bridge (B)
and thread macrocyclic crown ether C. The integrity of the
system is maintained by the bulkiness of tripodal linker L
and stopper group S, which is connected to V2 by a flexible
polyether moiety PE. Although C is free, in principle, to
travel (shuttle) along the axle, it is predominantly localised
at V1 or V2.
It has previously been shown that the affinity of C to V1
and V2 can be altered electro- or photochemically by reduc-
ing V1 and V2.[14,15] A single- or double-electron reduction
leads to a redistribution of C between the viologen sites.
Studies of shuttling in the above and related [2]rotaxane sys-
tems, however, appear to suggest that electrochemical shut-
tling in [2]rotaxanes may be accompanied by intramolecular
stacking of V1 and V2 and folding of B.[14,15]
The two [2]rotaxanes studied herein differ in the structure
[12]
of B that links V1 and V2. In the [2]rotaxane 1·4PF6
(Figure 1, bridge B1), the viologens are interconnected by
three methylene groups. It is referred to throughout the
paper as a flexible rotaxane because the distance between
the nitrogen atoms of V1 and V2 is relatively short (2.7–
5.3 Š) and can, in principle, freely change owing to free ro-
tation. In fact the angle formed by the N···N’ axes of V1 and
V2 is variable in the range of at least 0 to 1808, thus leaving
the possibility for V1 and V2 to stack in space. It should also
be noted that 1·4PF6 contains two flexible units, B1 and PE.
To distinguish between the independent movements of these
units, the conformational changes are referred to as B-fold-
ing and PE-folding, respectively.
The other [2]rotaxane (3·4PF6) incorporates a terpheny-
lene bridge (B2) between the two viologen stations
(Figure 1, bridge B2). In 3·4PF6 the distance between the ni-
trogen atoms of V1 and V2 is relatively long (15.2–16 Š, see
the Supporting Information) and does not change signifi-
cantly as a result of free rotation. For the same reason, the
angle formed by the N···N’ axes of V1 and V2 is variable in
the range of only 0 to 1428, thus excluding the possibility for
V1 and V2 to stack. Bridge B2 also effectively plays the role
of a spacer between viologens V1 and V2.
Figure 1. a) General representation of switchable [2]rotaxane in which L:
linker, C: crown ether, V1 and V2: viologen sites, B1 and B2: bridges, PE:
flexible chain and S: stopper. b) Structures of [2]Rotaxanes 1 and 3,
which show the numbering used in Tables 1, to 4, and axles, 2 and 4.
1118
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1117 – 1128

Flexible and Rigid Bistable [2]Rotaxanes
FULL PAPER
The aim of this study is to understand the significance of
possible intramolecular folding processes and their role in
electrochemically controlling the shuttling of C. Our find-
ings highlight both the similarities and the differences in be-
haviour of the flexible and rigid [2]rotaxanes in solution. In
particular, they address the question as to whether inserting
a rigid bridge into the bistable [2]rotaxane enables more
precise control over the position and orientation of self-as-
sembled nanoscale systems. The significance of these find-
ings should be considered in the context of the related stud-
ies of molecular shuttles in solution and at the surface.[12–16]
Upon completion of the present study it came to our atten-
tion that an important and very recent paper is devoted to
shuttling in functionally rigid [2]rotaxanes.[16a] It was demon-
strated that the rigid spacer possesses a very low barrier for
the movement of a macrocycle component in a [2]rotaxane.
This paper describes the design of 1·4PF6 and 3·4PF6 and
the elucidation of their conformations in solution in the
parent tetracationic state, 14+ and 34+, from NOE spectros-
different. Specifically, affinity of the crown ether to more
electron-deficient V1 is higher than to less electron-deficient
V2. In fact, it has been shown for a closely related inter-
locked system that the crown ether is located on V1.[14a,b]
Polyether chain PE links V2 to rigid aromatic stopper S in
1·4PF6 and 3·4PF6. Linker PE, which includes adjacent aro-
matic groups, is 18 Š long in its stretched conformation and
is flexible. In Part 1 we demonstrated that owing to its flexi-
bility the PE chain adopts a folded conformation and en-
sures that 7 is localised at the viologen site.[13a] It has been
suggested that the folded conformation of the PE chain is
possibly stabilised by CH–p interactions.
In summary, [2]rotaxanes 1·4PF6 and 3·4PF6 were de-
signed to contain rigid and flexible moieties in the sequence
L-V1-B-V2-PE-S. Whereas both 1·4PF6 and 3·4PF6 contain
an identical flexible PE moiety, the flexibility and size of
the central bridging element (B) that connects the two func-
tional viologens is significantly different. It is expected that
the size of B and the B-folding significantly influence the
conformation and shuttling in the above interlocked mole-
cules.
Synthesis: [2]Rotaxanes 1·4PF6 and 3·4PF6 were prepared
by a threading method that uses high-concentration condi-
tions[10–12] rather than from axles 2·4PF6 and 4·4PF6, respec-
tively, with slipping of 7.[17–19] Owing to the large size of the
bulky tetraarylmethane stoppers, the slipping approach
would not be successful. The synthesis of 1·4PF6 has been
described elsewhere.[12] Similarly, the preparation of 3·4PF6
and 4·4PF6 was carried out as shown in Scheme 1 starting
from 8·PF6. In the final step, key precursor 12·3PF6 was
treated with the stopper component, 13 or 13a, in a mini-
mum volume of benzonitrile.[10] In the presence of 7, [2]ro-
taxane 3·4PF6 was obtained in good yields, whereas in the
absence of 7 only 4·4PF6 was formed.
To unambiguously assign the reduction potentials of V1
and V2 in compounds 1 to 4, a direct comparison was made
by using structurally analogous model compounds 5·2PF6
and 6·2PF6. Model compound 5·2PF6, which contains V1 in
the absence of V2, was used to assign the reduction poten-
tials of V1 in compounds 1 to 4.[11,12] Model compound
6·2PF6, which contains only V2, was used to assign the re-
duction potentials of V2. Compound 6·2PF6 was prepared by
alkylation of 3,3’-dimethyl-4,4’-bipyridine with 13 (see the
Experimental Section). Reference compounds 5 and 6 ena-
bled the comparative structural and dynamic characterisa-
tion of 1·4PF6 and 3·4PF6 in solution.
copy data. The conformations of the radical cations (13+
C
and 33+ ) derived from the [2]rotaxanes after single-elec-
tron reduction in solution are also discussed. Cyclic voltam-
metry data are used to elucidate possible conformational
changes in 12+, 32+ and 1+, 3+ reduction states of the [2]ro-
taxanes.
C
Results and Discussion
Design of tripodal [2]rotaxanes: [2]Rotaxanes 1·4PF6 and
3·4PF6 and their corresponding axle components, 2·4PF6
and 4·4PF6, used in this work are shown in Figure 1.
Both 1·4PF6 and 3·4PF6 contain tripodal linker L. Linker
L is a large rigid aromatic structure that prevents dethread-
ing of C. To provide substantial solubility in most common
solvents, the phosphonate groups of L are in the form of
ethyl esters. L is directly and rigidly connected to V1.
Viologens V1 and V2, which are present in 1·4PF6 and
3·4PF6, are both electroactive and are able to accept up to
two electrons each when reduced chemically or electro-
chemically.[11,12,14] With regards to V1 and V2, it should be
noted that relatively electron-poor V1 is fully reduced at sig-
nificantly more positive potentials than relatively electron-
rich V2.[12] This is possibly because the radical cation of V1
can potentially form a conjugated planar structure that delo-
calises the additional electron across both pyridine rings.
Further delocalisation is possible because V1 is conjugated
with the p-phenylene moiety of L. In comparison, as a result
of electron density donation from the methyl substituents at
the 3 and 3’ positions, viologen moiety V2 is relatively less
electron poor. Additionally, the radical cation of V2 cannot
adopt a planar conformation owing to steric repulsion of the
two methyl substituents at the 3 and 3’ positions and cannot
delocalise the donated electron.[14f] As a consequence, V2 is
reduced at more negative potentials than V1 (see below).
Owing to the same electronic and steric factors, the affinity
of crown ether 7 towards the V1 and V2 sites is significantly
Conformations of [2]rotaxanes in the parent state: One pos-
sible way to characterise the conformations of [2]rotaxanes
in their parent states is by X-ray crystallography. This tech-
nique has been successfully completed for a number of
[2]pseudorotaxanes that contain viologens and crown ethers,
but not for [2]rotaxanes that contain bulky, oily stopper
groups, which impede the formation of an ordered crystal-
line phase.[19–21] Until recently, little was known about solid-
state and solution structures of viologen-based [2]rotaxanes
that incorporate 7, and in particular, the co-conformation of
Chem. Eur. J. 2008, 14, 1117 – 1128
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1119

K. Nikitin, J. K. Stolarczyk et al.
for this purpose.[22,23] To this end, an almost complete assign-
1
ment of the observed H NMR spectroscopy resonances was
carried out on the basis of COSY correlations and related
data from structural analogues.[13–15] All of the protons in
1·4PF6 and 3·2PF6 were numbered (Figure 1) so that analo-
gous protons would have the same numbers in the two com-
pounds. For convenience a full list of the chemical shifts, ob-
served NOE intensities and respective calculated distances
is given in Table S1 in the Supporting Information. In the ar-
omatic region (d=7.5 to 8.0 ppm), some loss of information
was unavoidable as a result of overlap of the adjacent peaks.
In Part 1 we demonstrated that a combination of quantita-
tive NOE and crystallographic reference data can form the
basis of a detailed analysis of the structure of a model [2]ro-
taxane in solution.[13a] In this paper we followed the same
methodology to characterise two tripodal [2]rotaxanes,
1
1·4PF6 and 3·4PF6, by using H NMR NOE spectroscopy at
298 K in acetonitrile. In separate experiments eight equiva-
lent aromatic protons of 7 (denoted H13) and six methyl pro-
tons of the V2 moiety (H23/H27) were irradiated for both ro-
taxanes. The results were expressed, as previously, in terms
of average effective interatomic distances (d) in 1·4PF6 and
3·4PF6 in solution by using the solid-state crystallographic
data from analogous [2]pseudorotaxane and viologen refer-
ence compounds.[13a] In this manner two sets of distances
were obtained for each rotaxane, distances dax between aro-
matic protons H13 and other groups of equivalent Hx pro-
tons, and distances dmx between methyl protons H23/H27 and
protons Hx. These results, however, do not allow the confor-
mations and co-conformations of 1·4PF6 and 3·4PF6 to be
fully characterised and visualised. Therefore, we used molec-
ular mechanics (MM) minimisation (Macromodel 8.5 with
AMBER force field) to find the structures that best
matched the experimentally determined distances dax and
dmx. The least-squares method was used for fitting. Owing to
the fact that dax and dmx represent average distances between
multiple equivalent irradiated protons H13 or H23/H27 and
multiple protons Hx, it was impossible to directly use them
as constraints in MM minimisation. Accordingly, a multiple-
step Euler-type optimisation procedure was employed with
distance constraints refined during each step to achieve the
best fit with dax and dmx. The resulting optimised structures
of 1·4PF6 and 3·4PF6 are shown in Figure 2.
Scheme 1. i) 1) benzonitrile, 368C, 2 d, 2) KPF6 (aq), CH3NO2, total 83%;
ii) 1) benzonitrile, ambient temperature, 6 d, 2) KPF6 (aq), CH3NO2, total
83%; iii) 1) benzonitrile, ambient temperature, 10 d, 2) KPF6 (aq),
CH3NO2, total 69%; iv) 1) benzonitrile, ambient temperature, 10 d,
2) KPF6 (aq), CH3NO2, total 70%.
For the calculations of average distances dax (i.e., when
protons H13 were irradiated), the corresponding average dis-
tance between protons H13 and H14 in the solid-state struc-
ture of an analogous [2]pseudorotaxane was used as a refer-
ence (Table 1).[13a] Clearly, crown ether 7 is localised at V1
adjacent to L in both 1·4PF6 and 3·4PF6 in solution. A com-
parison of 1·4PF6 and 3·4PF6 reveals that in each case there
are substantial NOEs for the protons in V1 (H9–H12 in
Figure 1). The calculated distances da9 to da12 (from protons
H13 to protons H9–H12) are in the range of 3.7 to 4.1 Š,
which are similar to the distances in the crystals of analo-
gous [2]pseudorotaxanes.[13a]
the [2]rotaxanes, that is, which viologen site is predominant-
ly occupied.[12–15]
1H NMR spectroscopy techniques, based on NOE, are
commonly used for structural characterisation in solution. In
particular, NOE spectroscopy of [2]rotaxanes has been used
It is also clear that 7 is not localised at V2 adjacent to S in
both 1·4PF6 and 3·4PF6 in solution. Both 1·4PF6 and 3·4PF6
1120
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1117 – 1128

Flexible and Rigid Bistable [2]Rotaxanes
FULL PAPER
[24]
than 5.86 Š
(16.12 Š in the
model of 3·4PF6). The crown
ether, therefore, is also not ac-
commodated by B2. These find-
ings are in full agreement with
the expected position of the
crown ether moiety in the
parent oxidation state.[12,14a,b]
In 1·4PF6, significant NOEs
were only measured for the
methyl substituents (protons
H23/H27), which are, therefore,
located relatively close to the
crown
ether
(d23 =6.23 Š),
which indicates B-folding of the
molecule (Figure 2a). By com-
parison, it is apparent that the
V1-B2-V2 section of 3·4PF6 is
not folded because no measura-
ble NOE was detected for the
methyl substituents (protons
H23) of the V2 moiety (Fig-
ure 2b).
Figure 2. Model conformations of the [2]rotaxanes a) 1·4PF6 b) 3·4PF6.
Table 1. Effective interatomic distances dax [Š] in 1·4PF6 and 3·4PF6.
It can be concluded that flexi-
ble trimethylene bridge B1 in
1·4PF6 is folded so that the
methyl groups of V2 can ap-
Position
Notation
d in 1[a]
experimental
d in 3[a]
experimental
model
model
L
V1
H8
H9
H10
H11
H12
H14
H18
H42
H43
H44
H45
H19
H23/H27
H20/H24
H21/H25
H22/H26
H39
3.73ꢀ0.11
4.02ꢀ0.16
3.82ꢀ0.12
3.80ꢀ0.12
4.07ꢀ0.17
3.20ꢀ0.07[b]
4.40ꢀ0.19
4.61ꢀ0.25
–
4.07
4.41
4.30
4.37
4.91
3.39
5.93
3.83
–
3.96ꢀ0.15
4.04ꢀ0.16
3.71ꢀ0.11
3.75ꢀ0.11
4.04ꢀ0.16
3.20ꢀ0.07[b]
4.04
3.69
3.90
4.19
4.83
3.28
5.07
–
4.22
4.41
6.22
16.12
20.31
18.75
18.41
18.92
10.59
proach
7
from the outside,
whereas V2 as a recognition site
does not accommodate 7. On
the contrary, rigid terphenylene
bridge B2 in 3·4PF6 does not
permit B-folding and no NOE
is detected.
With regards to bulky stopper
S, no measurable NOE was ob-
served for any of the protons
(Table S1 in the Supporting In-
formation). These findings are
consistent with the conclusion
that 7 populates V1 and 7 is not
approached closer than about 6
to 7 Š by S. However, no infor-
mation can be gained with re-
gards to the conformation and
possible folding of the PE
chain.
7
[c]
B1, B2
B1
B2
–
–
4.04ꢀ0.16
–
–
–
–
>5.86
5.44ꢀ0.06
>5.86
>6.33
>5.86
>5.86
B1, B2
V2
>5.86
7.13
6.10
7.88
6.94
7.96
15.03
6.23ꢀ1.04
>5.86
>5.86
>5.86
>7.60
>5.86
>7.75
S
[a] If no corresponding signal has been observed in the NOE spectrum, the minimum distances (shown in ital-
ics) have been estimated on the basis of the noise level in the spectrum and the reference distance dref. [b] Ref-
erence distance dref was obtained from a crystal structure of a related [2]pseudorotaxane.[13a] [c] The value
cannot be accurately determined as a result of overlap with the irradiation peak.
show no measurable NOE of the protons of V2 (closely
overlapping H20, H21, H24, H25 and closely overlapping H22,
H26). Furthermore, with regards to B, which links V1 to V2,
it is noted that 7 is not localised at the bridge, or in other
words, the bridge does not provide a recognition site for 7.
In 1·4PF6, the short distances for bridge methylene protons
H18 (da18 =4.40 Š) and H42 (da42 =4.61 Š), but not for the
third pair of methylene protons H19 (da19 >5.86 Š) confirm
that 7 is localised at V1. In 3·4PF6, relatively short distances
associated with B2 were calculated for H43 (d43 =4.04 Š) and
H45 (d45 =5.44 Š), however, the distance to H19 is far greater
Further insights into the structure of 1·4PF6 and 3·4PF6 in
solution were obtained when the methyl protons of the V2
moiety (H23 and H27) were irradiated at their respective res-
onance frequency in subsequent NOE experiments. In this
case the required effective reference distance (deref) was cal-
culated between the six protons of the two 3,3’-methyl
groups (H23/H27 in 1·4PF6 and 3·4PF6) and the two protons
at the remaining 3,3’ positions (H22/H26 in 1·4PF6 and
3·4PF6). The required structural data were sourced from the
X-ray crystal structure of 3,3’-dimethyl-1,1’-diethyl-4,4’-bi-
Chem. Eur. J. 2008, 14, 1117 – 1128
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1121

K. Nikitin, J. K. Stolarczyk et al.
pyridinium hexafluorophospane, 14·2PF6. The resulting dis-
tance (3.79 Š) was used as a reference. The ensuing data are
presented in Table 2.
macrocycle and estimation of the conformation of the axle
in the reduced state of viologen-based [2]rotaxanes.[13a]
The reduction of V1 can be achieved by using an appro-
priate reducing reagent or by applying an appropriate nega-
tive potential. Typically it is carried out by using Zn metal
as a reducing agent. The reduction can be readily achieved
for 1·4PF6 and 3·4PF6, the model rotaxanes that only con-
tain the V1 moiety, and for the model compound 5·2PF6. To
determine if this agent also reduces V2, model compound
6·2PF6, which only contains the V2 unit, was tested. No evi-
dence for a reduction process with V2 was found on addition
of Zn metal to a solution of 6·2PF6 in either acetonitrile or
methanol under air-free conditions. This result clearly indi-
cated that whereas V1 is reduced, V2 is not reduced by Zn
metal and this reagent can be used to obtain the [2]ro-
taxanes in singly reduced radical cation states 1·3PF6 and
3·3PF6.
Table 2. Effective interatomic distances dmx [Š] in 1·4PF6 and 3·4PF6.
Position
Notation
d in 1[a]
experimental
d in 3[a]
experimental
model
model
V1
V2
S
H12
H22/H26
H39
H28
H29
>8.35
3.79ꢀ0.26[b]
6.71ꢀ0.84
7.27ꢀ0.80
>8.35
6.87
3.75
6.83
6.19
6.36
6.10
>8.35
3.79ꢀ0.26[b]
7.42ꢀ1.15
>8.35
18.84
3.75
7.30
5.91
6.09
20.31
>8.35
>9.44
7
H13
7.38ꢀ1.08
[a] If no corresponding signal has been observed in the NOE spectrum,
the minimum distances (shown in italics) have been estimated on the
basis of the noise level in the spectrum and reference distance dref
.
[b] The reference distance was calculated from the crystal structure of
14·2PF6.
The 1H NMR spectra of 1·4PF6 and 3·4PF6 and their
C
C
Clearly, in both 1·4PF6 and 3·4PF6 the PE chain is folded
so that protons H39 of the stopper approach the methyl
groups of V2 as close as 6.7 to 7.4 Š. The data also confirm
an earlier observation that in the case of 1·4PF6, the methyl
groups of V2 approach 7 at 7.3 Š owing to the B-folding of
the trimethylene bridge. No evidence of similar B-folding
was found for 3·4PF6 in which the viologens are mechanical-
ly separated by non-flexible terphenylene bridge B2.
cation radicals 1·3PF6 and 3·3PF6 were recorded in de-
gassed acetonitrile at 258C and the relaxation times T1 were
measured by the saturation–inversion method.[13a,c] Calculat-
ed PRE values are presented in Tables 3 and 4 (for full T1
data see the Supporting Information). The effective average
Table 3. Paramagnetic relaxation enhancement and estimated effective
[a]
distances [Š] in 1·4PF6
The model conformations of 1·4PF6 and 3·4PF6 are pre-
sented in Figure 2 in which V1 and V2 are highlighted in
blue and green, respectively. The crown ether, which forms
a sandwich structure with V1, is shown in red. Whereas the
PE chain is clearly folded in both 1·4PF6 and 3·4PF6, violo-
gens V1 and V2 only form an acute angle in 1·4PF6 as a
result of folding of the bridge. In this way, V2 can come into
close proximity with the crown ether. A key finding is that 7
is localised at and interacting with V1 in both cases.
Position
V1
Notation
PRE [sꢁ1
]
dxe
H9
H10
H11
PSZ
PSZ
PSZ
<5.1[b]
<5.1[b]
<5.1[b]
H12
PSZ
<5.1[b]
V2
B1
H20/H24
H21/H25
H23/H27
H18
H42
H19
H39
H5
H13
H14–H17
1.2ꢀ0.3
1.0ꢀ0.3
7ꢀ2
12.0ꢀ1.8
12.4ꢀ1.9
9.0ꢀ1.4
<5.1[b]
PSZ
PSZ
<5.1[b]
122ꢀ11
1.3ꢀ0.2
2.5ꢀ0.5
83ꢀ5
58ꢀ3
5.6ꢀ0.6
11.8ꢀ1.5
10.6ꢀ1.5
5.9ꢀ0.7
6.3ꢀ0.7
S
L
7
Conformations of the [2]rotaxanes in the first reduced state:
After determining that V1 was predominantly populated in
the parent state and the axles adopted a folded conforma-
tion, a further objective of this study was to understand the
conformations and co-conformation adopted by 1 and 3
when V1 was reduced and the affinity of the crown ether to-
wards this recognition site significantly diminished.[25]
Owing to sufficient differences in the reduction potentials
between V1 and V2 it was possible to selectively reduce V1
adjacent to L. This allowed the co-conformation in the
singly reduced radical cation states of 1·3PF6 and 3·3PF6 to
be determined by using the paramagnetic relaxation en-
hancement (PRE) technique outlined below.
Recently we described a novel NMR methodology that
allows the structural and conformational changes of the
[2]rotaxane molecule in different oxidation states to be
monitored.[13a] It was shown that in radical cations of re-
duced viologens the unpaired electron density causes meas-
urable PRE. The PRE values depend drastically on the dis-
tance from the unpaired electron. This provides a means of
reasonably accurate characterisation of the location of the
[a] ꢁ1.00”10ꢁ2 moldmꢁ3 solution in [D3]acetonitrile, 258C. [b] Based on
T1 below 2.5”10ꢁ3 s.
Table 4. Paramagnetic relaxation enhancement and estimated effective
distances in 3·4PF6.[a]
Position
V1
Notation
PRE [sꢁ1
]
dxe
H9
H10
H11
H12
H23 H27
H18
H45
H19
H39
H4
PSZ
PSZ
PSZ
PSZ
<5.1[b]
<5.1[b]
<5.1[b]
<5.1[b]
>14
V2
B2
0.00ꢀ0.09
PSZ
<5.1[b]
8.1ꢀ1.6
>14
10.6ꢀ3.3
11.2ꢀ2.5
4.7ꢀ0.8
5.7ꢀ1.0
1.0ꢀ0.3
0.0ꢀ0.3
0.2ꢀ0.2
0.14ꢀ0.07
26ꢀ3
S
L
7
H13
H14–H16
8.4ꢀ0.7
[a] ꢁ10ꢁ2 moldmꢁ3 solution in [D3]acetonitrile, 258C. [b] Based on T1
below 2.5”10ꢁ2 s.
1122
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1117 – 1128

Flexible and Rigid Bistable [2]Rotaxanes
FULL PAPER
distances from the centre of the paramagnetic suppression
zone (PSZ) to each of the proton groups Hx were also calcu-
lated.[13a] The distances from the centre of the PSZ to pro-
tons H5 (in the model conformation of 1·4PF6) and from
protons H4 (in the model conformation of 3·4PF6) were
used as references because these distances were expected to
remain unchanged upon reduction of the rotaxanes.
The analysis of proton PRE data (Table 3) shows that in
1·3PF6, as expected, the strongest paramagnetic effect is ex-
perienced by V1 (protons H9–H12) and all adjacent protons
(PSZ). The PRE is too high to be measured in the PSZ and
no quantitative information can be obtained. For the rest of
the molecule the proton PRE values were obtained and
were used to determine the conformation of the mole-
cule.[13a]
of the noticeable PRE (1.2 sꢁ1) for stopper terminal protons
H39 that, owing to the folding of B1 and the PE chain, stop-
per S is in the vicinity of the PSZ of V1. These results prove
that B- and PE-folding persist in 1·3PF6. This is in excellent
agreement with the above NOE findings on the conforma-
tion of the parent oxidation state 1·4PF6.
To conclude, a single-electron reduction of 1·4PF6 in solu-
tion does not seem to induce dramatic changes in the popu-
lation of the recognition sites by 7. Whereas the interactions
of 7 with V1 are significantly weakened as a result of addi-
tional electron density on this site, the crown ether still pre-
dominantly populates the V1 site. This can be explained by
considering the folding of the flexible trimethylene bridge,
which possibly represents an additional steric barrier for 7
to overcome.
To facilitate the analysis of the co-conformational changes
in the [2]rotaxanes upon reduction, the fractions of the en-
semble (population) of the rotaxane molecules in which 7 is
located at V1 and V2 are denoted a1 and a2, respectively.
Owing to the affinity of the crown ether towards the recog-
nition sites, the sum of a1 and a2 should be equal to 1, that
is, only the two above co-conformations are considered to
be relevant. The values of a1 and a2 could also be under-
stood to be the probabilities of finding 7 at either V1 or V2,
respectively.
Probabilities a1 and a2 in the first single-electron reduced
state, 1·3PF6, were determined as follows: According to
Table 3 the PRE value for aromatic protons H13 of 7 in
1·3PF6 is 83 sꢁ1. If 7 remained at V1 (a1 =1) its PRE would
be significantly higher, specifically 125 sꢁ1 as was previously
shown for a structurally similar [2]rotaxane.[13a] However,
for 7 located at V2 the PRE would have been significantly
lower, specifically about 7 sꢁ1 (Table 3, PRE value for V2).
This may suggest that the population is split between the
sites. Assuming that the observed PRE value is a linear
combination of the relaxation rates at both sites, it can be
calculated that a1 =0.65ꢀ0.10 in the reduced state 1·3PF6,
and hence, a2 =0.35ꢀ0.10.[26] The details of the calculation
are given in the Supporting Information. This result suggests
that 7 is predominantly, but not completely localised at V1
in the reduced state, 1·3PF6.
The analysis of the proton PRE data in 3·4PF6 (Table 4)
shows that in this case, as expected, the strongest PRE is ex-
perienced by V1 itself (H9–H12) and the surrounding protons
(PSZ). Importantly, it is again possible to elucidate the pop-
ulation of the viologen sites by 7 in the first single-electron
reduced state 3·3PF6. In this case the PRE value for aromat-
ic protons H13 of 7 in 3·3PF6 is 26 sꢁ1. Following the above
argument and by using data from Table 3 for the V2 site of
1·3PF6, it can be calculated that the probability value of a1
is 0.21ꢀ0.10 in the reduced state, 3·3PF6, and the probabili-
ty value of a2 is 0.79ꢀ0.10.[26] The results cannot be ex-
plained in terms of shift of the crown ether, as for 3·3PF6,
because it would require a highly unlikely localisation of the
crown at bridge B2. It can be concluded, therefore, that 7
does predominantly populate V2 in the reduced state,
3·3PF6.
The PRE values for V2 are negligible. Therefore, as ex-
pected, viologen station V2 in 3·3PF6 is not approaching the
PSZ of V1. It can also be concluded on the basis of negligi-
ble PRE values for the t-butyl groups of the stopper that the
stopper does not approach the PSZ of V1. This is in excel-
lent agreement with the above NOE findings for the parent
state of 3·4PF6.
Apparently, single-electron reduction of rigid 3·3PF6
causes significant changes in the population of the recogni-
tion sites by 7 as shown in Figure 3. In this case 7 mainly
populates the second, originally weakly binding viologen
recognition site V2.
The decrease in the PRE value for protons H13 can also
be interpreted differently. It can be attributed to a shift of
the average position of these crown ether protons with re-
spect to V1, which leads to an increase of the effective dis-
By comparing the findings for both rigid and flexible tri-
podal [2]rotaxanes it is concluded, therefore, that as a result
of the folded conformation of the axle in 1, the distribution
of the population of the viologen recognition sites by the
crown ether is sensitive to, but is not effectively controlled
by, single-electron reduction in solution. Upon single-elec-
tron reduction the crown ether continues to predominantly
populate the originally strongly binding site V1. On the con-
trary, the rigid design of 3 provides more efficient conditions
for electrochemical control of distribution of the crown
ether between recognition sites V1 and V2. In this case a
single-electron reduction causes inversion of the population,
a process that may be termed shuttling or Brownian switch-
ing.[27]
1
tances by a factor of (125= ) . This corresponds to a shift of
=
6
83
1
only 0.3 Š towards the trimethylene bridge, that is, = th of
20
the length of the bipirydyl unit. It means that the crown
ether is still structurally localised at V1. This finding is in
qualitative agreement with our previous findings on the sur-
face of titanium oxide.[12]
Note that the PRE values for H27 of the methyl substitu-
ents in V2 (7 sꢁ1) are higher than the PRE values for the in-
plane aromatic protons H20 of V2 (1 sꢁ1). Consequently, it
C
can be concluded that V2 in 1·3PF6 is folded back towards
V1 so that the methyl substituents of V2 approach the PSZ
of V1 relatively closely. It can also be concluded on the basis
Chem. Eur. J. 2008, 14, 1117 – 1128
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1123

K. Nikitin, J. K. Stolarczyk et al.
the second and third potentials.
For this reason the maxima of
combined second and third re-
duction peaks are quoted for
systems 1 to 4. It is noted that
this behaviour was expected on
the basis of a number of CV ex-
periments with individual violo-
gen stations V1 and V2 and
their various mixtures under a
range of conditions.
Firstly, let us consider the re-
duction of bis-viologen system
2·4PF6, which involves four
one-electron steps. The assign-
ment of the reduction poten-
tials of 2·4PF6 can be made by
comparing the reduction poten-
tials of tripodal model com-
pound 5·2PF6, the V1 model ro-
taxane[13a] and model compound
6·2PF6. Owing to the presence
of an additional positively
Figure 3. Shuttling in the [2]rotaxanes. Left: flexible 1·4PF6, right: rigid 3·4PF6.
Conformations of the second and further reduced states of 1
and 3: It has previously been shown that cyclic voltammetry
(CV) can provide insights into intramolecular shuttling pro-
cesses in switchable [2]rotaxanes.[11,12,13c,14,15] These insights
are possible because the donor–acceptor interaction of 7
with the viologen moieties shifts the reduction potential of
the viologens to more negative values. The reduction of viol-
ogens is known to be associated with their dimerisation,
which leads to positive reduction potential shifts,[28] and with
conformational changes, which lead to incomplete reduc-
tion.[29] Recently it has been shown that the reduction re-
quires flattening of the structure of the viologens, and conse-
quently, the reduction potentials of viologens depend criti-
cally on the dihedral angle between the two pyridinium
rings.[14f] Consequently, the reduction requires more negative
potentials if the ability of a viologen to adopt a flat confor-
mation is restricted as it is in all 3,3’-dimethylviologens, V2.
The CVs of [2]rotaxanes 1·4PF6 and 3·4PF6 and the corre-
sponding axles 2·4PF6 and 4·4PF6 were recorded (see the
Supporting Information). Tripodal model compound 5·2PF6
and model compound 6·2PF6 were also studied to assign any
reduction peaks observed in 2·4PF6 and 4·4PF6. The CV
measurements were carried out at a scan rate of 10 mVsꢁ1
by using a solution of tetraethylammonium hexafluorophos-
phate (0.05 moldmꢁ3) as an electrolyte and non-aqueous
Ag/AgNO3 as the reference electrode. The list of the respec-
tive peak potentials (Ep) is given in Table 5. For comparison,
the reduction potentials of the single-station model [2]ro-
taxane described in detail Part 1 are also given.[13a]
Table 5. Cyclic voltammetry data.[a]
Compound
Ep [V][b]
2+
+
+
0
2+
+
+
0
V1 !V1
V1 !V1
V2 !V2
V2 !V2
1·4PF6
2·4PF6
3·4PF6
4·4PF6
5·2PF6
6·PF6
ꢁ0.75
ꢁ0.52
ꢁ0.78
ꢁ0.69
ꢁ0.63
–
ꢁ0.98
ꢁ1.06
ꢁ1.08
ꢁ1.06
ꢁ0.96
–
ꢁ1.46
ꢁ1.40
ꢁ1.58
ꢁ1.48
–
–
ꢁ1.07
ꢁ1.56
V1 model rotaxane[c] ꢁ0.77
ꢁ1.12
V1 model axle[c]
ꢁ0.67
ꢁ1.00
[a] Acetonitrile, 258C, 0.05 moldmꢁ3 tetraethylammonium hexafluoro-
phosphate, 5.00”10ꢁ2 moldmꢁ3 in 1–6. [b] Error bar ꢀ0.01 V. [c] Refer-
ence [13a].
charged moiety (V2) the reduction potential of V1 in 2·4PF6
is shifted to more positive values compared with tripodal
model compound 5·2PF6 (ꢁ0.63 V) and model rotaxane
(ꢁ0.67 V).[14a,b,29] As a result, the first reduction potential of
2·4PF6 is ꢁ0.52 V. It can also be concluded qualitatively that
the second reduction potential of 2·4PF6 is within ꢁ0.96 to
ꢁ1.0 V, as the second reduction potential of 5·4PF6
(ꢁ0.96 V) and the model single-station rotaxane (ꢁ1.00 V).
The third and the fourth reduction potentials of 2·4PF6
(ꢁ1.06 and ꢁ1.40 V) are associated with the reduction of
V2. This can be determined on the basis of the reduction po-
tentials of V2 in 6·2PF6 (ꢁ1.07 and ꢁ1.55 V).[29] It is con-
cluded, therefore, that the electrochemical reduction of
2·4PF6 involves two single-electron reductions of V1 fol-
lowed by two single-electron reductions of V2. Consequent-
ly, the formation of an interim reduced species of 2·2PF6
Owing to the strong overlaps of the reduction peaks of
flexible 1·4PF6 and rigid 3·4PF6 in the region at approxi-
mately ꢁ1 V, it is difficult to make precise assignments of
1124
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1117 – 1128

Flexible and Rigid Bistable [2]Rotaxanes
FULL PAPER
C
C
with two co-existing viologen radicals, V1 and V2 , is not ex-
pected. This species, if formed, would undergo intramolecu-
lar viologen–viologen association and lead to pronounced
positive shifts in the second and third potentials.
It can be expected that unlike the first reduced state of
flexible 1·3PF6, in rigid 3·3PF6 the second reduction poten-
tial of V1 will only be marginally shifted to negative values
as a result of a weak interaction with 7 after the first reduc-
tion of V1. On the other hand, the first reduction potential
of V2 will be considerably negatively shifted, thus maintain-
ing the electrochemical gap between V1 and V2. This would
exclude the formation of two co-existing radical cation viol-
It was possible to analyse the interaction of the same viol-
ogen moieties of 1·4PF6 with 7 and the position of 7 after
the reduction potentials of the viologen moieties in 2·4PF6
were assigned. The value of the first reduction peak of
1·4PF6 (ꢁ0.75 V) is shifted towards more negative values
compared with 2·4PF6 (ꢁ0.52 V) as a result of the interac-
tion of V1 with 7. This is in full agreement with the reduc-
tion potentials of the model single-station rotaxane
(ꢁ0.77 V) and the above NMR spectroscopy findings that
the crown ether mainly populates the V1 site in 1·4PF6, as
shown in Figure 3. It also is consistent with previous findings
on the surface of titanium dioxide[12] and in solution.[14a,b]
It is expected that in 1·3PF6 the second reduction poten-
tial of the V1 site will be shifted to negative values, possibly
to ꢁ1.1 V as in the model rotaxane, owing to the interaction
with 7 after the first reduction of V1. This requires, however,
the V2 site to be reduced first at approximately ꢁ1.07 V
with the formation of two co-existing radical cation violo-
gens (Figure 3). As these viologen radicals are conforma-
tionally permitted to interact and dimerise, owing to B-fold-
ing, the resulting reduction potential is considerably shifted
towards more positive values, ꢁ0.98 V. This unexpected shift
indicates the early onset of p–p-type stacking of V1 and V2
and the crown ether in 1 in comparison with 2. Finally, the
second reduction potential of V2 in 1·PF6 (ꢁ1.46 V) is dis-
placed towards more negative values compared with 2
(ꢁ1.40 V), which indicates that 7 populates V2 in the re-
duced state, 1·PF6.[11,12] Proposed co-conformational and
conformational changes in 1 in solution are illustrated in
Figure 3. It can be concluded that the population of the viol-
ogen sites by 7 changes over the course of the four consecu-
tive reduction steps. In first three oxidation states, 1·4PF6,
1·3PF6 and 1·2PF6, crown ether 7 predominantly populates
V1. Predominant population of V2 takes place in 1·PF6 (after
triple reduction).
+
+
C
C
ogens, V1 and V2 . Additionally, their possible association
is precluded by long and rigid bridge B2. Although, owing to
the overlap of the second and the third reduction peaks, the
observed electrochemical shift cannot be accurately mea-
sured, it is clear that the second reduction peak is slightly
shifted towards more negative values and there is no unusu-
al positive shift associated with dimerisation. Finally, the
second reduction potential of V2 in 3·PF6 (ꢁ1.58 V) is sub-
stantially displaced towards more negative values compared
with 4·PF6 (ꢁ1.48 V), which indicates that 7 populates V2 in
the reduced state, 3·PF6.
Proposed co-conformational and conformational changes
in 3 in solution are shown in Figure 3. It can be concluded
that the population of the viologen sites by 7 changes over
the course of four consecutive reduction steps. In the first
oxidation state (3·4PF6), crown ether 7 predominantly popu-
lates V1. Predominant population of V2 takes place in
3·3PF6, 3·2PF6 (after double reduction) and in 3·PF6 (after
triple reduction).
Comparing the two rotaxane systems it is noted that,
unlike in 1, the inversion of the site population (switching)
happens in 3 after the first single-electron reduction. A
likely explanation is that in 1·3PF6 flexible bridge B1 folds
back, which allows additional stabilisation of 7 at V1 by pos-
sible simultaneous interaction with both V1, and to a lesser
extent, V2. On the contrary, in 3·3PF6 rigid bridge B2 ex-
cludes the possibility of such additional stabilisation.
Conclusion
Now let us consider the reduction of the rigid bis-viologen
system, 4·4PF6. The relatively long and rigid bridge (B2) be-
tween V1 and V2 separates the two stations and prevents B-
folding. Accordingly, the reduction potentials of V1 and V2
in 4·4PF6 are not shifted to more positive values when com-
pared with V1 and V2 in model compounds 5·2PF6 and
6·2PF6, respectively. The first two reduction potentials of
4·4PF6 (ꢁ0.69 and ꢁ1.06 V) can be assigned, therefore, to
V1. The third and fourth reductions of axle compound
4·4PF6 (ꢁ1.06 and ꢁ1.48 V) are the reduction potentials of
the viologen moiety V2. The interaction of the viologen moi-
eties in 3·4PF6 with 7 and the position of 7 can be analysed
on the basis of the electrochemical shifts. The value for the
first reduction of 3·4PF6 (ꢁ0.78 V) is displaced, owing to the
interaction of V1 with 7, towards more negative values com-
pared with 4·4PF6 (ꢁ0.69 V). This is in full agreement with
above findings by NOE spectroscopy that the crown ether
populates the V1 site in 3·4PF6.
The observation of intramolecular shuttling processes in so-
lution and at the surface is a challenging task for modern
structural techniques because the changes in the spectral
characteristics of switching molecular systems are subtle.
1
A H NMR spectroscopy study of switchable 1·4PF6 and
3·4PF6 in solution demonstrates that the crown ether is lo-
cated at the V1 recognition site in the parent state. The PRE
technique can be used to determine the population of the
recognition sites and the conformation of the first reduced
states of 1·3PF6 and 3·3PF6. It can be demonstrated that in
1·4PF6 the folding of the bridge interconnecting V1 and V2
results in additional stabilisation of the crown ether at V1.
Consequently, the crown ether populates this site in the first
reduced state, 1·3PF6. The electrochemical behaviour of
1·4PF6 further demonstrates that the crown ether populates
V1 even in the second reduced state (1·2PF6).
In the case of 3·4PF6, the possibility of bridge folding is
precluded in all reduction states. Consequently, switching
Chem. Eur. J. 2008, 14, 1117 – 1128
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1125

K. Nikitin, J. K. Stolarczyk et al.
sodium hydrogen carbonate, dried over MgSO4 and evaporated under re-
duced pressure to give 9 as a white solid (0.088 g, 100%). M.p. 1728C;
1H NMR (300 MHz, CDCl3, 258C): d=7.45 (d, J=8.0 Hz, 4H), 7.34 (d,
J=8.0 Hz, 4H), 7.13 (s, 2H), 4.57 (s, 4H), 2.27 ppm (s, 6H); elemental
analysis calcd (%) for C22H20Br2: C 59.49, H 4.54, Br 35.98; found: C
59.58, H 4.28, Br 35.87.
(when the population of V2 is higher than V1) happens after
the first reduction of 3. However, even in this case the repo-
sitioning of the crown ether from V1 to V2 is not complete
after the first reduction. This incomplete switching is possi-
bly owing to low affinity of the crown ether for V2.
A comparison of the electrodynamics of switchable tripo-
dal [2]rotaxanes 1·4PF6 and 3·4PF6 in solution strongly sug-
gests that the design of the bridge between V1 and V2 in
these [2]rotaxanes determines the degree of pre-organisa-
tion of the [2]rotaxane and its ability to switch. This implica-
tion is valuable for the development and design of new,
more effective switchable molecular systems and is in good
agreement with the most recent findings published by Stod-
dart and co-workers.[16]
Compound 10·2PF6: Monocation 8·PF6 (0.050 g, 0.04 mmol) and
9
(0.045 g, 0.1 mmol) were dissolved in benzonitrile (0.2 mL) and the mix-
ture was kept at 368C for 2 d. The mixture was purified by column chro-
matography (silica gel; MeOH/nitromethane/saturated aqueous KPF6
80:19:1 v/v) to give 10·2PF6 as a yellow solid (0.059 g, 83%). Decomp
220–2308C; 1H NMR (300 MHz, CD3OD, 258C): d=9.21 (d, J=7.0 Hz,
2H), 9.18 (d, J=7.0 Hz, 2H), 8.64 (d, J=7.0 Hz, 2H), 8.57 (d, J=7.0 Hz,
2H), 7.87–7.95 (m, 6H), 7.70–7.75 (m, 10H), 7.55–7.65 (m, 8H), 7.42–
7.58 (m, 10H), 7.31 (d, J=7.0 Hz, 2H), 7.10–7.14 (m, 2H), 5.93 (s, 2H),
4.58 (s, 2H), 4.00–4.18 (m, 12H), 2.24 (s, 6H), 1.35 ppm (t, J=7.0 Hz,
18H); 31P NMR: d=20.81 (s), ꢁ140 (septet, J=711 Hz); MS (ES): m/z
(%): 1475 (19)
A
C87H86BrF12N2O9P5: C 59.16, H 4.91, N 1.59; found: C 58.90, H 4.76, N
1.61.
Experimental Section
Compound 12·3PF6: Dication 10·2PF6 (0.058 g, 0.033 mmol) and 3,3’-di-
methylbypiridyl (11, 0.016 g, 0.09 mmol) were dissolved in benzonitrile
(0.16 mL) and the mixture was kept at room temperature for 6 d. Chro-
matography (silica gel; MeOH/nitromethane/saturated aqueous KPF6
70:20:10 v/v) gave 12·3PF6 as a beige solid (0.055 g, 83%). Decomp
2308C; 1H NMR (300 MHz, CD3OD, 258C): d=9.18 (d, J=6.2 Hz, 2H),
9.12 (d J=6.2 Hz, 2H,), 8.92 (s, 1H), 8.76 (d, J=6.2 Hz, 1H), 8.56–8.65
(m, 6H), 7.78–7.85 (m, 7H), 7.66–7.73 (m, 10H), 7.58–7.63 (m, 8H),
7.40–7.44 (m, 13H), 7.04 (s, 2H), 5.90 (s, 2H), 5.80 (s, 2H), 4–4.2 (m,
12H), 2.30 (s, 3H), 2.17 (s, 6H), 2.12 (s, 3H), 1.35 (t, J=7.1 Hz, 18H);
31P NMR: d=23.8 (s), ꢁ140 ppm (septet, J=711 Hz); MS (ES): m/z (%):
1869 (30) [12·2PF6+]; elemental analysis calcd (%) for C99H98F18N4O9P6:
C 58.99, H 4.90, N 2.78; found: C 58.89, H 4.66, N 2.70.
General methods: Reagents were purchased from Sigma-Aldrich. All re-
actions were conducted in a nitrogen atmosphere. Melting points were
estimated by using a Gallenkamp melting point device and were not cor-
rected. NMR spectra were recorded by using Varian Inova 300 and 500
spectrometers in the solvent indicated at 258C. Proton NMR spectra
were recorded at 299.89 and 499.82 MHz and phosphorus NMR spectra
at 121.39 MHz. Mass spectra were recorded by using a Micromass LCT
mass spectrometer. Crystal data were collected by using
a Bruker
SMART APEX CCD area detector diffractometer. Paramagnetic sup-
pression NMR spectroscopy was carried out as described elsewhere.[13c]
Tripodal [2]rotaxane 1·4PF6 and axle 2·4PF6,[12] tripods 5·2PF6 and
8·PF6,[11] crown ether 7,[30a] stopper bromide 13,[10] and viologen
[2]Rotaxane 3·4PF6: Trication 12·3PF6 (0.053 g, 0.026 mmol), crown
ether 7 (0.038 g, 0.071 mmol) and stopper 13 (0.038 g, 0.05 mmol) were
dissolved in benzonitrile (0.15 mL) and kept at ambient temperatures for
10 d. Chromatography (silica gel; MeOH/nitromethane/saturated aque-
ous KPF6 70:20:10 then MeOH/nitromethane/1m aqueous NH4PF6
70:20:10 v/v) followed by extraction with nitromethane gave 3·4PF6 as a
red solid (0.060 g, 69%). M.p. 1608C; 1H NMR (500 MHz, CD3CN,
258C): d=9.18 (d, J=7.0 Hz, 2H), 8.98 (d, J=7.0 Hz, 2H), 8.88 (s, 1H),
8.79 (s, 1H), 8.76 (d, J=6.2 Hz, 1H), 8.69 (d, J=6.2 Hz, 1H), 8.04 (d, J=
7.0 Hz, 2H), 7.92 (d, J=7.0 Hz, 2H), 7.75–7.90 (m, 18H), 7.74 (d, J=
8.0 Hz, 6H), 7.50–7.64 (m, 10H), 7.40–7.46 (m, 4H), 7.30–7.34 (m, 6H),
7.16–7.22 (m, 10H), 7.03 (d, J=9.0 Hz, 2H), 6.80 (d, J=9.0 Hz, 2H) 6.16
(s, 8H), 5.99 (s, 2H), 5.80 (s, 2H), 5.67 (s, 2H), 4.15–4.20 (m, 4H), 4.08–
4.15 (m, 12H), 3.82–3.90 (m, 4H), 3.55–3.70 (m, 24H), 3.45–3.50 (m,
8H), 2.25 (s, 6H), 2.22 (s, 3H), 2.20 (s, 3H), 1.29 (t, J=7.1 Hz, 18H),
1.27 (s, 27H); 31P NMR: d=23.70 (s), ꢁ140 ppm (septet, J=712 Hz); MS
(ES): m/z (%): 1543.5 (100) [3·2PF62+]; elemental analysis calcd (%) for
[30b,c]
14·2PF6
were prepared as described elsewhere.
Compound 6·2PF6: Bromide 13 (0.076 g, 0.1 mmol) and 3,3’-dimethyl-
4,4’-bipyridyl (0.009 g, 0.05 mmol) were added to a mixture of CH3CN
(2 mL) and CH2Cl2 (2 mL) and the mixture was left at room temperature
for 6 d. The mixture was separated by column chromatography (silica
gel; acetone/MeOH/nitromethane/saturated aqueous KPF6 50:35:10:5 v/
v), the retained fraction was evaporated under reduced pressure, the resi-
due was extracted with CH3Cl, the organic layer was washed with water
and evaporated under reduced pressure to give 6 as a white solid
1
(0.070 g, 70%). M.p. 2268C; H NMR (300 MHz, CD3CN, 258C): d=8.77
(s, 2H), 8.67 (d, J=6.0 Hz, 2H), 7.74 (d, J=6.0 Hz, 2H), 7.44 (d, J=
9.0 Hz, 4H), 7.30 (d, J=9.0 Hz, 12H), 7.14–7.20 (m, 16H), 7.03 (d, J=
9.0 Hz, 4H), 6.80 (d, J=9.0 Hz, 4H), 5.66 (s, 4H), 4.14–4.18 (m, 4H),
4.08–4.12 (m, 4H), 3.83–3.88 (m, 8H), 2.087 (s, 6H), 1.27 ppm (s, 54H);
MS (ES): m/z (%): 865.5 (100) [6+ꢁstopper], 846.5 (45) [6 PF62+]; ele-
C
mental analysis calcd (%) for C108H126F12N2O6P2: C 70.57, H 6.91, N 1.52;
found: C 70.80, H 6.83, N 1.74.
C
175H195F24N4O22P7: C 62.20; H 5.82; N 1.66; found: C 62.26; H 5.54; N
1,4-Bis(4-hydroxymethylphenyl)-2,5-dimethylbenzene: 1,4-Dibromo-2,5-
dimethylbenzene (0.264 g, 1 mmol), 4-hydroxymethylboronic acid
(0.320 g, 2.1 mmol), toluene (2 mL), ethanol (6 mL), potassium hydrogen
phosphate (0.700 g, 5 mmol) and dichloro-1,1’-bis-(diphenylphosphinofer-
rocene)palladium dicholomethane (0.016 g, 0.02 mmol) were heated with
stirring at 708C for 36 h. The mixture was extracted with EtOAc and pu-
rified by column chromatography (silica gel; EtOAc) to give the product
as a white solid (0.24 g, 75%). M.p. 1678C; 1H NMR (300 MHz, MeOH,
258C): d=7.40 (d, J=8.0 Hz, 4H), 7.32 (d, J=8.0 Hz, 4H), 7.10 (s, 2H),
4.67 (s, 4H), 2.23 ppm (s, 6H); elemental analysis calcd (%) for
C22H22O2: C 82.99, H 6.96; found: C 82.66, H 6.86. The product was crys-
tallised from CH3Cl/MeOH and monoclinic needles that were suitable
for X-ray diffractometry analysis were obtained.
1.51.
Axle 4·4PF6: Trication 12·3PF6 (0.040 g , 0.02 mmol) and 13 (0.021 g,
0.028 mmol) were dissolved in benzonitrile (0.1 mL) and kept at room
temperature for 10 d. Chromatography (silica gel; MeOH/nitromethane/
saturated aqueous KPF6 70:20:10; then MeOH/nitromethane/1m aqueous
NH4PF6 70:20:10 v/v) followed by extraction with nitromethane gave
4·4PF6 as a beige solid (40 mg, 70%). Decomp 200–2108C; 1H NMR
(300 MHz, CD3COCD3, 258C): d=9.60 (d, J=7.1 Hz, 2H), 9.58 (d, J=
7.1 Hz, 2H), 9.42 (s, 1H), 9.32 (s, 1H), 9.27 (d, J=7.1 Hz, 1H), 9.18 (d,
J=7.1 Hz, 1H), 8.96 (d, J=7.1 Hz, 2H), 8.91 (d, J=7.1 Hz, 2H), 8.20–
8.24 (m, 2H), 8.04 (d, J=9.0 Hz, 2H), 7.70–7.92 (m, 22H), 7. 63 (d, J=
9.0 Hz, 2H), 7.45–7.54 (m, 12H), 7.3 (d, J=6.2 Hz, 6H), 7.10–7.14 (m,
12H), 6.82 (d, J=9.0 Hz, 2H), 6.26 (s, 2H), 6.13 (s, 2H), 5.98 (s, 2H),
4.05–4.20 (m, 16H), 3.85–3.92 (m, 4H), 2.42 (s, 3H), 2.40 (s, 3H), 2.26 (s,
6H), 1.29 (t, J=7.2 Hz, 18H), 1.28 (s, 27H); 31P NMR: d=23.70 (s),
ꢁ140 ppm (septet, J=712 Hz); MS (ES): m/z (%): 1275.8 (100)
Compound
9:
1,4-Bis(4-hydroxymethylphenyl)-2,5-dimethylbenzene
(0.064 g, 0.2 mmol) was suspended in THF(2 mL) and phosphorus tribro-
mide (0.110 g, 0.4 mmol) was added. After stirring at room temperature
for 3 h the mixture was diluted with diethyl ether, washed with aqueous
1126
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1117 – 1128

Flexible and Rigid Bistable [2]Rotaxanes
FULL PAPER
[4·2PF62+]; elemental analysis calcd (%) for C147H155F24N4O12P7: C 62.11;
H 5.50; N 1.97; found: C 61.96; H 5.50; N 1.81.
7811; c) L. Wei, K. Padmaja, W. J. Youngblood, A. B. Lysenko, J. S.
Lindsey, D. F. Bocian, J. Org. Chem. 2004, 69, 1461–1469.
[10] K. Nikitin, B. Long, D. Fitzmaurice, Chem. Commun. 2003, 282–
283.
[11] B. Long, K. Nikitin, D. Fitzmaurice, J. Am. Chem. Soc. 2003, 125,
5152–5160.
Cyclic voltammetry: All cyclic voltammograms were recorded in solution
under the following conditions: The working electrode was an isolated
platinum wire. The counter electrode was an isolated platinum foil. The
reference electrode was a non-aqueous Ag/Ag+ electrode filled with
10 mm AgNO3 in the electrolyte solution. The electrolyte solution was
tetraethylammonium hexafluorophosphate (0.05 moldmꢁ3) in dry CH3CN
(distilled over calcium hydride). All solutions were degassed by three
freeze–thaw cycles under vacuum prior to measurement. All cyclic vol-
tammograms were recorded by using a Solartron SI 1287 potentiostat
controlled by a LabView program running on a Macintosh Power PC at a
scan rate of 10 mVsꢁ1 unless otherwise stated.
[12] B. Long, K. Nikitin, D. Fitzmaurice, J. Am. Chem. Soc. 2003, 125,
15490–15498.
[13] a) S. Altobello, K. Nikitin, J.K. Stolarczyk, E. Lestini, D. Fitzmaur-
ice, Chem. Eur. J. 2008, 14, 1107–1116 (preceding article); b) the
term conformation is used herein to indicate any rotational confor-
mation of the components of the [2]rotaxane and co-conformation is
used to indicate the relative position of the crown ether in a [2]ro-
taxane. c) K. Nikitin, D. Fitzmaurice, J. Am. Chem. Soc. 2005, 127,
8067–8076.
[14] a) P. R. Ashton, R. Ballardini, V. Balzani, A. Credi, K. R. Dress, E.
Ishow, C. J. Kleverlaan, O. Kocian, J.A. Preece, N. Spencer, J. F.
Stoddart, M. Venturi, S. Wenger, Chem. Eur. J. 2000, 6, 3558–3574;
b) V. Balzani, M. Clemente-León, A. Credi, B. Ferrer, M. Venturi,
A. H. Flood, J. F. Stoddart, Proc. Natl. Acad. Sci. USA 2006, 103,
Acknowledgement
This research was funded by Science Foundation Ireland and EU Marie
Curie Host Development grants.
´
1178–1183; c) C. P. Collier, E. W. Wong, M. Belohradsky,
F . M.
Raymo, J. F. Stoddart, P. J. Kuekes, R. S. Williams, J.R Heath, Sci-
ence 1999, 285, 391–394; d) C. P. Collier, G. Mattersteig, E. W.
Wong, Y. Luo, K. Beverly, J. Sampaio, F. M. Raymo, J. F. Stoddart,
J. R. Heath, Science 2000, 289, 1172–1175; e) S. J. Loeb, J. Tiburcio,
S. J. Vella, Chem. Commun. 2006, 1598–1600; f) A. C. Benniston, A.
Harriman, P. Li, J. P. Rostron, R. W. Harrington, W. Clegg, Chem.
Eur. J. 2007, 13, 7838–7851.
[1] a) P. W. K. Rothemund, Nature 2006, 440, 297–302; b) D. Philp, J. F.
Stoddart, Angew. Chem. Int. Ed. Engl 1996, 35, 1154–1196; Angew.
Chem. 1996, 108, 1242–1286; c) F. Mancin, E. Rampazzo, P. Tecilla,
U. Tonellato, Chem. Eur. J. 2006, 12, 1844–1854; d) P. M. Mendes, S.
Jacke, K. Critchley, J. Plaza, Y. Chen, K. Nikitin, R. E. Palmer, J.A.
Preece, S. D. Evans, D. Fitzmaurice, Langmuir 2004, 20, 3766–3768.
[2] a) C. V. Cojocaru, F. Ratto, C. Harnagea, A. Pignolet, F. Rosei, Mi-
croelectron. Eng. 2005, 80, 448–456; b) D. Bishop, Bell Lab. Techni-
cal Journal 2005, 23–28; c) W. Bows, Investors Chronicle 2002, 139,
30; d) J. E. Green, J. W. Choi, A. Boukai, Y. Bunimovich, E. John-
ston-Halperin, E. DeIonno, Y. Luo, B. A. Sheriff, K. Xu, Y. S. Shin,
H.-R. Tseng, J. F. Stoddart, J. R. Heath, Nature 2007, 445, 414–417;
e) Y. Luo, C. P. Collier, J. O. Jeppesen, K.A. Nielsen, E. DeIonno,
G. Ho, J. Perkins, H.-R. Tseng, T. Yamamoto, J. F. Stoddart, J. R.
Heath, ChemPhysChem 2002, 3, 519–525.
[3] a) C. T. Lin, M. T. Kao, K. Kurabayashi, E. Meyhofer, Small 2006, 2,
281–287; b) D. K. Smith, Chem. Commun. 2006, 34–44; c) P.M
Mendes, A. H. Flood, J. F. Stoddart, Appl. Phys. A: Mater. Sci. Pro-
cess. 2005, 80, 1197–1209; d) G. M. Whitesides, J. P. Mathias, C. T.
Seto, Science 1991, 254, 1312–1319.
[4] a) J.-M. Lehn, Supramolecular Chemistry, Concepts and Perspectives,
VCH, Weinheim, 1995; b) R. J Forster, T. E. Keyes, J. G. Vos, Inter-
facial Supramolecular Assemblies, Wiley, Chichester, 2003.
[15] a) Y. H. Jang, W. A. Goddard, J. Phys. Chem. B 2006, 110, 7660–
7665; b) T. Yamamoto, H.-R. Tseng, J. F. Stoddart, V. Balzani, A.
Credi, F. Marchioni, M. Venturi, Collect. Czech. Chem. Commun.
2003, 68, 1488–1514.
[16] a) S. Nygaard, K. C.-F. Leung, I. Aprahamian, T. Ikeda, S. Saha,
B. W. Laursen, S.-Y. Kim, S. W. Hansen, P. C. Stein, A. H. Flood,
J. F. Stoddart, J. O. Jeppesen, J. Am. Chem. Soc. 2007, 129, 960–970;
b) J. W. Choi, A. H. Flood, D. W. Steuerman, S. Nygaard, A. B.
Braunschweig, N. N. P. Moonen, B. W. Laursen, Y. Luo, E. DeIonno,
A. J. Peters, J. O. Jeppesen, K. Xu, J. F. Stoddart, J. R. Heath, Chem.
Eur. J. 2006, 12, 261–279.
ˇ
[17] M. Asakawa, P. R. Ashton, R. Ballardini, V. Balzani, M. Belohrad-
´
sky, M. T. Gandolfi, O. Kocian, L. Prodi, F. M. Raymo, J. F. Stoddart,
M. Venturi, J. Am. Chem. Soc. 1997, 119, 302–310.
[18] F. M. Raymo, K. N. Houk, J. F. Stoddart, J. Am. Chem. Soc.; 1998,
120, 9318–9322.
[19] a) M. C. T. Fyfe, J. F. Stoddart, Acc. Chem. Res. 1997, 30, 393–401;
b) P. L. Anelli, P. R. Ashton, R. Ballardini, V. Balzani, M. Delgado,
M. T. Gandolfi, T. T. Goodnow, A. E. Kaifer, D. Philp, M. Pietrasz-
kiewicz, L. Prodi, M. V. Reddington, A. M. Z. Slawin, N. Spencer,
J. F. Stoddart, C. Vincent, D. J. Williams, J. Am. Chem. Soc. 1992,
114, 193–218.
[20] P. R. Ashton, R. Ballardini, V. Balzani, M. Belohradsky, M. T. Gan-
dolfi, D. Philp, L. Prodi, F. M. Raymo, M. V. Reddington, N. Spenc-
er, J. F. Stoddart, M. Venturi, D. J. Williams, J. Am. Chem. Soc. 1996,
118, 4931–4951.
[21] F. Huang, L. N. Zakharov, A. L. Rheingold, M. Ashraf-Khorassani,
H. W. Gibson, J. Org. Chem. 2005, 70, 809–813.
[5] a) R. F. Service, Science 1997, 277, 1036–1037; b) L, Cusack, S. N.
Rao, D. Fitzmaurice, Chem. Eur. J. 1997, 3, 202–207.
[6] a) C. Niemeyer, Angew. Chem. 2001, 113, 4254–4287; Angew. Chem.
Int. Ed. 2001, 40, 4128–4158; b) J. J. Davis, Chem. Commun. 2005,
3509–3513; c) R. J. Alvarado, J. Mukherjee, E. J. Pacsial, D.
Alexander, F. M. Raymo, J. Phys. Chem. B 2005, 109, 6164–6173.
[7] a) J. Berna, D. A. Leigh, M. Lubomska, S. M. Mendoza, E. M.
Perez, P. Rudolf, G. Teobaldi, F. Zerbetto, Nat. Mater. 2005, 4, 704–
710; b) A. Shipway, I. Willner, Acc. Chem. Res. 2001, 34, 421–432;
c) F. Cecchet, P. Rudolf, S. Rapino, M. Margotti, F. Paolucci, J. Bag-
german, A. M. Brouwer, E. R. Kay, J.K. Y. Wong , D. A. Leigh, J.
Phys. Chem. B 2004, 108, 15192–15199; d) W. R. Browne, B. L. Fer-
inga, Nat. Nanotechnol. 2006, 1, 25–35; e) E. R. Kay, D. A. Leigh, F.
Zerbetto, Angew. Chem. 2007, 119, 72–196; Angew. Chem. Int. Ed.
2007, 46, 72–191.
[22] a) C. A. Hunter, J. Am. Chem. Soc. 1992, 114, 5303–5311; b) A. C.
Benniston, A. Harriman, V. M. Lynch, J. Am. Chem. Soc. 1995, 117,
5275–5291.
[23] a) P. R. Ashton, R. Ballardini, V. Balzani, I. Baxter, A. Credi,
M. C. T. Fyfe, M. T. Gandolfi, M. Gómez-López, M.-V. Martꢁnez-
Dꢁaz, A. Piersanti, N. Spencer, J. F. Stoddart, M. Venturi, A. J. P.
White, D. J. Williams, J. Am. Chem. Soc. 1998, 120, 11932–11942;
b) J. F. Stoddart, S. A. Vignon, Collect. Czech. Chem. Commun.
2005, 70, 1493–1576.
[8] a) K. G. Thomas, P. V. Kamat, Acc. Chem. Res. 2003, 36, 888–898;
b) U. Drechsler, B. Erdogan, V. M. Rotello, Chem. Eur. J. 2004, 10,
5570–5579; c) S. Mann, W. Shenton, M. Li, S. Connolly, D. Fitz-
maurice, Adv. Mater. 2000, 12, 147–150; d) A. P. Alivisatos, Science
1996, 271, 933–937; e) A. Hagfeldt, M. Grätzel, Chem. Rev. 1995,
95, 49–68.
[24] The signal in the NOE spectrum that corresponds to protons H18
(d=6.0) overlaps with the intense excitation peak of H13 (d=6.2) in
this case.
[9] a) E. Galoppini, W. Guo, P. Qu, G. J. Meyer, J. Am. Chem. Soc.,
2001, 123, 4342–4343; b) E. Galoppini, W. Guo, W. Zhang, P. G.
Hoertz, P. Qu, G. J. Meyer J. Am. Chem. Soc., 2002, 124, 7801–
Chem. Eur. J. 2008, 14, 1117 – 1128
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1127

K. Nikitin, J. K. Stolarczyk et al.
[25] It is noted that although in some cases the radical cations formed
when a viologen undergoes a single-electron reduction may dimerise
in solution, it has been shown that relatively bulky [2]rotaxanes are
present as monomers in solution, see reference [13c] and J. A.
Alden, J. A. Cooper, F. Hutchinson, F. Prieto, R. G. Compton, J.
Electroanal. Chem. 1997, 432, 63–70.
[26] We assume that shuttling is fast on the NMR spectroscopy timescale
and that bridge B represents a sufficiently high barrier and is not oc-
cupied itself.; see also: D. A. Leigh, A. Troisi, F. Zerbetto, Angew.
Chem. 2000, 112, 358–361; Angew. Chem. Int. Ed. 2000, 39, 350–
353.
chem. Soc. 2006, 153, E179–E183; c) X. Tang, T. W. Schneider, J. W.
Walker, D. A. Buttry, Langmuir 1996, 12, 5921–5933.
[29] The relatively low reduction potential of V2 in 6·2PF6 (ꢁ1.55 V) is
possibly associated with structural folding and encapsulation of the
viologen moiety; a process that was demonstrated earlier in: F.
Marchioni, M. Venturi, P. Ceroni, V. Balzani, M. Belohradsky, A. M.
Elizarov, H.-R. Tseng, J. F. Stoddart, Chem. Eur. J. 2004, 10, 6361–
6368.
[30] a) R. C. Helgeson, T. L. Tarnowski, J. M. Timko, D. J. Cram, J. Am.
Chem. Soc. 1977, 99, 6411–6418; b) C. Stoehr, M. Wagner, J. Prakt.
Chem. 1893, 48, 1; c) E. Lestini, K. Nikitin, H. Müller-Bunz, D. Fitz-
maurice, Chem. Eur. J. 2008, 14, 1095–1106.
[27] M. N. Chatterjee, E. R. Kay, D. A Leigh, J. Am. Chem. Soc. 2006,
128, 4058–4073.
[28] a) J.A. Alden, J.A. Cooper, F. Hutchinson, F. Prieto, R. G. Comp-
ˇ
ton, J. Electroanal. Chem. 1997, 432, 63–70; b) L. Pospꢁsil, J. Fiedler,
Received: September 3, 2007
ˇ
M. Hromadovꢂ, M. Gꢂl, M. Valꢂsek, J. Pecka, J. Michl, J. Electro-
Published online: November 28, 2007
1128
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 1117 – 1128