Protonmotive force: Development of electrostatic drivers for synthetic molecular motors

Article abstract of DOI:10.1002/chem.200500519

Ferrocene has been investigated as a platform for developing protonmotive electrostatic drivers for molecular motors. When two 3-pyridine groups are substituted to the (rapidly rotating) cyclopentadienyl (Cp) rings of ferrocene, one on each Cp, it is shown that the (Cp) eclipsed, π-stacked rotameric conformation is preferred both in solution and in the solid state. Upon quaternization of both of the pyridines substituents, either by protonation or by alkylation, it is shown that the preferred rotameric conformation is one where the pyridinium groups are rotated away from the fully π-stacked conformation. Electrostatic calculations indicate that the rotation is caused by the electrostatic repulsion between the charges. Consistently, when the π-stacking energy is increased π-stacked population increases, and conversely when the electrostatic repulsion is increased π-stacked population is decreased. This work serves to provide an approximate estimate of the amount of torque that the electrostatically driven ferrocene platform can generate when incorporated into a molecular motor. The overall conclusion is that the electrostatic interaction energy between dicationic ferrocene dipyridyl systems is similar to the π-stacking interaction energy and, consequently, at least tricationic systems are required to fully uncouple the π-stacked pyridine substituents.

Full text of DOI:10.1002/chem.200500519

FULL PAPER
DOI: 10.1002/chem.200500519
Protonmotive Force: Development of Electrostatic Drivers for Synthetic
Molecular Motors
[
a]
James D. Crowley, Ian M. Steele, and Brice Bosnich*
Abstract: Ferrocene has been investi-
gated as a platform for developing pro-
tonmotive electrostatic drivers for mo-
lecular motors. When two 3-pyridine
groups are substituted to the (rapidly
rotating) cyclopentadienyl (Cp) rings
of ferrocene, one on each Cp, it is
shown that the (Cp) eclipsed, p-stacked
rotameric conformation is preferred
both in solution and in the solid state.
Upon quaternization of both of the
pyridines substituents, either by proto-
nation or by alkylation, it is shown that
the preferred rotameric conformation
is one where the pyridinium groups are
rotated away from the fully p-stacked
conformation. Electrostatic calculations
indicate that the rotation is caused by
the electrostatic repulsion between the
charges. Consistently, when the p-
stacking energy is increased p-stacked
population increases, and conversely
when the electrostatic repulsion is in-
creased p-stacked population is de-
creased. This work serves to provide an
approximate estimate of the amount of
torque that the electrostatically driven
ferrocene platform can generate when
incorporated into a molecular motor.
The overall conclusion is that the elec-
trostatic interaction energy between di-
cationic ferrocene dipyridyl systems is
similar to the p-stacking interaction
energy and, consequently, at least trica-
tionic systems are required to fully un-
couple the p-stacked pyridine substitu-
ents.
Keywords: electrostatic
interac-
tions · molecular motors · molecular
switches · protonmotive rotation ·
supramolecular chemistry
[
5]
[6]
Introduction
that are driven by protonmotive, electronmotive, and
[7]
photonmotive forces. The types of systems employed are
[8]
Biological molecular motors are crucial for the functioning
and organization of living systems. These motors can trans-
port cellular material along microtubules or actin fila-
wide ranging. Herein we describe a systematic study of
what might be regarded as drivers (engines) for potential
molecular motors and machines. For this purpose we have
elaborated the ferrocene molecule and have exploited the
rotation of the cyclopentadienyl (Cp) rings for induced
rotary motion.
[1]
ments, can catalyze the production of ATP by proton-in-
[2]
duced rotation, and are responsible for locomotion in
[3]
living things. In most cases the motors are driven by the
hydrolysis of ATP. Unlike macroscopic motors, molecular
[4]
motors operate by two mechanisms, the “power stroke”
where movement is induced by a potential gradient and the
Results and Discussion
“Brownian ratchet” where stochastic Brownian motion is
rectified to cause net coherent motion. Biological motors
are complex molecular systems and attempts to develop syn-
thetic analogues present challenges of considerable com-
plexity. Even so, progress has been made in devising systems
The [Fe(Cp) ]-based driver: Ferrocene, 1, is a stable mole-
2
[9,10]
cule, it prefers an eclipsed conformation of its Cp rings,
[11]
2, the distance between the Cp rings is 3.298 Š, and the
ꢀ
1 [9,12]
barrier to rotation of the Cp rings is 0.9(3) kcalmol .
As
[
a] Dr. J. D. Crowley, Dr. I. M. Steele, Prof. B. Bosnich
Department of Chemistry
The University of Chicago
5
735 South Ellis Avenue, Chicago, IL 60637 (USA)
Fax : (+1)773-702-0805
E-mail: bos5@uchicago.edu
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2006, 12, 8935 – 8951
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8935

ꢀ
1
a consequence of the low rotational barrier, the Cp rings are
in rapid rotation even at very low temperatures. Inspection
of about 1 kcalmol that refer
to the (five degenerate)
[13]
of crystal structures of 1,1’-diarylsubstituted ferrocenes re-
veals that in nearly all of the cases the Cp rings are eclipsed
and the aryl groups on the Cp rings are p-stacked. This p-
eclipsed conformations. For the
molecule 7, it is assumed that
the two pyridine groups will p-
stack (in the eclipsed Cp con-
formation). Calculations
[
14,15]
stacking also obtains in solution.
Aromatic p–p stacking
[
16]
[19]
occurs at distances of 3.30 to 3.50 Š. This distance span is
slightly greater than the interplanar Cp–Cp separation
and
[20]
experiments indicate that a pair of symmetrically disposed
(
3.298 Š), nonetheless, p–p stacking of 1,1’-diaryl ferrocenes
face-to-face p-stacked benzene molecules is stabilized by
about 2kcalmol . Thus, assuming a similar value for two p-
ꢀ
1
is almost universally observed.
To convert ferrocenes into electrostatic drivers, we chose
,1’-substitution by basic aromatic groups, which are neutral
stacked pyridine groups, the total well depth for the Cp-
eclipsed p-stacked rotamer of 7, is about 3 kcalmol . Al-
ꢀ
1
1
but which can be made positively charged by quarterniza-
tion or by protonation. The basic idea is outlined by 3 and
though it is probably correct to assume a sinusoidal varia-
[12]
tion for the pure Cp rotation,
it is not clear what the
4
. The molecule 3 is expected to prefer the Cp rings in an
shape of the energy well for the p–p stacking is when the
two pyridine groups are rotated (parallel) with respect to
each other. Although a complex function probably obtains
for this process, we have used a sine function whose depth is
ꢀ
1
2
kcalmol
at the eclipsed p-stacked rotamer and whose
half wave encompasses plus 1088 to minus 1088 rotation of
the Cp rings from the energy minimum (see Supporting In-
formation). The energy well was extended to ꢁ1088 of Cp
rotation because at smaller rotations, the two pyridine sub-
stituents overlap with each other to varying degrees. Addi-
tion of the two rotational functions, one for Cp rotation and
the other for p–p interaction, gives the energy profile for
Cp rotation of 7. This is shown as the lower curve in
Figure 1.
eclipsed conformation and the pyridine groups p-stacked as
shown. Upon addition of acid, the pyridine groups will
become protonated. As a consequence, electrostatic repul-
sion between the two positively charged pyridine groups will
cause the Cp rings to rotate and adopt an orientation that
minimizes the Coulombic interaction. We, however, have
found for both ferrocene- and ruthenocene-substituted sys-
tems that if a basic ring nitrogen atom is positioned such
that the Cp ring can conjugate to form a fulvene structure
when the nitrogen is protonated, the complexes become un-
stable in acid solution. An example of this conjugation is
Whereas the neutral compound 7 and its analogues are
soluble in solvents of low dielectric constant such as CH Cl ,
2
2
upon quaternization they become insoluble in these solvents
ꢀ
ꢀ
with the SbF6 counter ion. The SbF6 ion was used to avoid
hydrogen bonding between the counter ion and the proto-
nated pyridine groups. Acetone and acetonitrile are good
solvents for both the neutral and charged species. To obtain
an estimate of the electrostatic repulsion in these two sol-
vents, the Coulombic energy was calculated by using the
[17]
shown in 6. (Protonation of the [FeCp ] iron atom does
2
(
(
bulk) dielectric constant of acetone (20.2) and acetonitrile
36.2) for diprotonated 7. When the pyridinium groups are
stacked, the effective dielectric constant is likely to be lower
than that of the bulk solvent because the charges are partly
screened by the aromatic rings. As rotation away from the
stacked rotamer increases, the solvent dielectric constant is
likely to become more realistic. Point charges located on the
nitrogen atoms were used and the separations between the
charges were calculated by rotating the pyridinium groups
while retaining their coplanar relationship. As we show by
NOE experiments later, the pyridine (and pyridinium) sub-
stituents rotate about the single bond that connects them to
the Cp rings. It was possible to include such rotation in the
calculation but such detail did not seem to provide anything
new associated with these electrostatic drivers.
not occur in acetonitrile solutions even with 100-fold excess
1
[18]
of trifluoroacetic acid as evidenced by H NMR spectra.
)
Thus the electrostatic drivers discussed here are ferrocene
systems bearing aromatic substituents which cannot engage
in the proton-induced conjugative shift that leads to decom-
position.
Electrostatic interactions: The essential elements of the
electrostatic drivers can be illustrated by the assumed be-
havior of 1,1’-di-3-pyridylferrocene, 7, in the (drawn) meso
conformation. Rotation of ferrocene itself will produce a si-
nusoidal energy profile where there are five wells of a depth
Figure 1 shows the Coulombic repulsion that is calculated
for Cp ring rotation in acetonitrile and acetone solutions for
the diprotonated meso rotamer of 7. Addition of these Cou-
lombic plots to the (lower) Cp-rotation-p–p-stacking plot
gives the two curves shown in Figure 1. Rotation occurs in
8936
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8935 – 8951

Electrostatic Drivers for Synthetic Molecular Motors
FULL PAPER
Figure 1. The calculated electrostatic repulsion curves for the meso ro-
tamer in acetonitrile (c) and acetone (c) solutions. The calculated
energy profile for the ferrocene Cp rotation (c). The sum of the Cp
rotation and the electrostatic repulsion in acetonitrile (c), and in ace-
tone (c). The nitrogen–nitrogen distance at 08 is 3.30 Š. The maximum
Figure 2. The calculated electrostatic repulsion curves for a racemic ro-
tamer in acetonitrile (c) and acetone (c) solutions. The calculated
energy profile for the ferrocene Cp rotation (c). The sum of the Cp
rotation and the electrostatic repulsion in acetonitrile (c), and in ace-
tone (c). The nitrogen–nitrogen distance at 08 is 4.10 Š. The maximum
ꢀ1
ꢀ1
electrostatic repulsion energy is 2.77 kcalmol
in acetonitrile and
electrostatic repulsion energy is 2.77 kcalmol
in acetonitrile and
ꢀ
1
ꢀ1
4
.96 kcalmol in acetone. The space-filling models show the rotation of
4.96 kcalmol in acetone. The space-filling models show the rotation of
the pyridinium groups for the angles shown in the lowest ordinate.
the two protonated 3-pyridine groups for the angles shown in the lowest
ordinate.
lations. A similar situation as found for the acetone solution
obtains for the interaction of a 2+ with 1+ interaction in
acetonitrile solution.
either direction with the same energy as expected for the
(
achiral) meso rotamer. Inspection of the resultant energy
profile in acetonitrile solution shows that there is little or no
preference for any rotamer. Shallow energy wells of about
the same depth occur at the eclipsed Cp conformations. The
analogous energy profile for acetone solution indicates that
rotation of the pyridinium groups will occur either to the
A consideration of the energy curves in Figure 1 and
Figure 2suggests that generally there exist energy wells in
the racemic rotamer that are more stable than those of the
meso pyridinium rotamer. It should, however, be empha-
sized that the variation in energy for any composite curve is
small and detailed discussion is circumscribed by the ap-
proximate nature of the calculations.
ꢁ
728 or ꢁ1448 wells. At the present level of calculation, the
results suggest that in acetone or solutions with lower die-
lectric constants, electrostatic repulsion is likely to cause ro-
tation of the pyridinium groups. For the meso rotamer, the
present approximate calculations do not predict that the pyr-
idinium groups will be significantly rotated in acetonitrile
solutions when each of the pyridinium groups carries one
Synthesis: The synthesis of the symmetrical and unsymmet-
rical 1,1’-diaryl ferrocenes followed the excellent procedures
[21,22]
reported by Braga.
The details are given in the experi-
mental section and are summarized in Equations (1) to (7)
in Scheme 1. The Suzuki couplings proceed in moderate to
good yields. It was found that mono-substitution of the di-
boronic acids proceeds best by short microwave induction.
Addition of the second substituent to the monoboronic acid
was found to give the best yields under standard Suzuki con-
(
positive) charge each. For the interaction of a 2+ group
with a 1+ group, these calculations would predict rotation.
As noted, however, the solvent bulk dielectric constant may
not be an appropriate constant when the groups are p-
stacked, and as experiments here show, some rotation of the
pyridinium groups obtains in acetonitrile solutions.
ditions. Using acetone as a solvent, AgSbF removal of the
6
Figure 2shows the analogous curves for one enantiomer
of the racemic rotamer of 7. There are two features of inter-
est shown by these energy relationships. First, the profile is
unsymmetrical. As a consequence, rotation of the pyridini-
um groups is biased in one direction; biased clockwise for
one enantiomer and biased counterclockwise for the other.
Second, in acetonitrile solution the pyridinium groups will
rotate and prefer to lie oriented at an angle of less than 728
according to these calculations. In acetone solution, the pyri-
dinium groups will rotate further than in acetonitrile and
will lie either at ꢂ708 or at ꢁ1448 according to these calcu-
chloro ligands from the palladium complex in Equation (5)
in the presence of pyridine, yields two products; one is the
dipyridine complex (75% yield) and the other is the monop-
yridine complex, which has the other coordination position
occupied by a s-bonded carbon atom of the acetone anion
(15% yield). This latter complex is presumably formed after
acetone is deprotonated by pyridine. The two products in
Equation (5) are separated by crystallization.
The preparation of 3-bromo-8-nitroquinoline [Eq. (6)] has
[23]
been described
but the two nitro isomers were misas-
signed; the minor component is the 8-nitro isomer. This is
Chem. Eur. J. 2006, 12, 8935 – 8951
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8937

B. Bosnich et al.
Scheme 1. a) 3-bromopyridine, Na
2
CO
, dioxane, NaOH (aq), DME, [Pd
], 1008C, 20 h; e) [Pd(NCCH Cl
CO , dioxane, NaOH (aq), DME, [Pd
reflux 36 h. dppf=1,1’-diphenylphosphinoferrocene, DME=dimethoxyethane.
3
, dioxane, NaOH (aq), DME, [Pd
(dppf)Cl
], acetone, RT, 1 h; f) i) 2equiv AgSbF
(dppf)Cl ], 1008C, 24 h; h) NH
A
H
R
U
G
2
], 1008C, 24 h; b) i) CH
3
I, CH
2
Cl
2
, RT, 16 h; ii) AgSbF
6
, acetone, RT,
CO , dioxane,
1
h; c) 3-bromopyridine, Na
2
CO
3
A
H
R
U
G
2
2
3
NaOH (aq), DME, [Pd(dppf)Cl
A
C
H
T
R
E
U
N
G
2
A
H
R
U
G
3
)
2
2
6
acetone, RT, 1 h; g) 3-bromo-8-nitroquinoline, Na
2
3
A
H
R
U
G
2
2
2
3
8938
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8935 – 8951

Electrostatic Drivers for Synthetic Molecular Motors
FULL PAPER
confirmed by X-ray diffraction (see later). Reduction of the
nitro group to the amine is efficiently carried out by the
graphite/hydrazine ethanol procedure. All of these com-
pounds are conveniently stable both as solids and in solu-
tion.
rience an upfield chemical shift compared to those of 8, in-
dicating that the pyridine substituents of 7 are p-stacked at
least to some extent. A similar comparison of 9 with 10 indi-
cates that the pyridinium groups of 9 are not extensively p-
stacked because of the small differences in chemical shift
[24]
(
Dd) for the two compounds. That the neutral compound 7
The 1,1’pyridylferrocene systems: The four systems investi-
gated were 7, 8, 9, and 10 (Table 1). The monosubstituted
is p-stacked to a greater extent than the charged analogue 9
is consistent with the electrostatic arguments presented ear-
lier.
For these ferrocene systems to act as protonmotive driv-
ers, compound 7 should rotate from its eclipsed p-stacked
rotameric configuration upon protonation of the pyridine
Table 1. Chemical shifts of the protons of the pyridine substituents of
various ferrocene compounds in [D
3
]acetonitrile solutions.
Chemical shift (d)
groups. Two acids were used: CF CO H, which is not disso-
Compound
H
a
H
b
H
c
H
d
3
2
[25]
ciated in dry acetonitrile and HSbF ·6H O which, in ace-
6
2
tonitrile solutions, was found to be highly conducting, sug-
ꢀ
gesting dissociation of the proton from SbF .
6
8
.73
.46
8.38
7.24
7.86
The protonation of 7 and 8 with CF CO H in dry acetoni-
trile was studied by examination of H NMR spectral shifts
of the pyridine protons. Five equivalents of CF CO H were
required to fully protonate 8 as indicated by the chemical
shifts becoming insensitive to further addition of acid. Con-
ductivities of 10 m solutions of fully protonated 8 were
65 W cm mol , which is that expected for a 1:1 electro-
lyte in acetonitrile solution (120–160 W cm mol ).
Hence the protonated form of 8 appears to be fully dissoci-
ated in acetonitrile solution with the CF CO counterion.
3
2
1
3
2
8
8.28
7.01
7.48
ꢀ3
ꢀ
1
2
ꢀ1
1
ꢀ
1
2
ꢀ1 [26]
difference (Dd)
ꢀ0.27
ꢀ0.10
ꢀ0.23
ꢀ0.38
ꢀ
3
2
A similar titration with CF CO H and the dipyridine
8.63
8.45
7.83
8.34
3
2
system, 7, required 10 equivalents of CF CO H for full pro-
3
2
tonation (see Supporting Information). The conductivity in
ꢀ
3
acetonitrile for a 10 m solution of the protonated form of 7
ꢀ
1
2
ꢀ1
was 208 W cm mol , which is slightly lower than expected
for 2:1 electrolyte in dry acetonitrile (220–
00 W cm mol ). The much stronger acid HSbF ·6H O
a
ꢀ
1
2
ꢀ1 [26]
3
6
2
8.55
8.38
7.76
8.24
fully protonated 8 and 7 upon the addition of 1 and 2equiv-
ꢀ
3
alents of HSbF ·6H O to the 10 m acetonitrile solutions, re-
6
2
spectively.
difference (Dd)
ꢀ0.08
ꢀ0.07
ꢀ0.07
ꢀ0.10
Table 2lists the chemical shifts for the two protonated
systems at 258C. The two acids cause almost the same
proton chemical shifts for the two compounds. The chemical
shift differences (Dd) are also similar for the two acids, with
Dd for the CF CO H protonation being larger in magnitude.
compounds 9 and 10 were used to determine the chemical
shifts of the protons on the pyridine rings when no p-stack-
ing is present. In the disubstituted cases 7 and 8 p-stacking
3
2
This may suggest that HSbF ·6H O is most effective in un-
6
2
coupling the p-stacking.
[14,15]
is expected to induce (upfield) chemical shifts
of the
The solid-state structures of five compounds bearing only
3-pyridine or pyridinium substituents were determined by
X-ray crystallography to try to obtain a more detailed un-
derstanding of the structures. We were unable to obtain a
structure of the neutral di-3-pyridine compound 7, because
the crystals were always twinned, even after crystallization
pyridine protons. Although these compounds are highly
fluxional, the chemical shifts for the disubstituted species 7
and 8 are related to the concentration weighted sum of the
chemical shifts of all of the possible rotamers. Thus, for ex-
ample the greater the average concentration of the p-
stacked rotamers, the greater the upfield shift of the signal
from several solvent combinations. Based on the work of
1
[22]
for any proton in the H NMR spectrum. The absence of p-
Braga et al.
and the present studies, there seems little
stacking would be indicated by chemical shifts that are the
same as for the corresponding monomers.
The chemical shifts of the pyridine protons of 8 and 7 in
doubt that the two pyridine groups of 7 are p-stacked.
Figure 3 and Figure 4 show two perspectives of the struc-
ꢀ
tures of 10 and 9 as their SbF6 salts. Table 3 lists the appro-
[
D ]acetonitrile solutions and the corresponding differences
priate crystallographic data. The bond lengths and angles
are unexceptional. The information germane to the present
study is summarized in Table 4. The present structures and
3
in chemical shifts are listed in Table 1. It will be noted that
in acetonitrile solutions all of the pyridine protons in 7 expe-
Chem. Eur. J. 2006, 12, 8935 – 8951
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8939

B. Bosnich et al.
Table 2. Chemical shifts of the protons of the pyridinium substituent of the mono- and disubstituted ferro-
the CF CO H and CF SO H
3
2
3
3
cenes in [D
3
]acetonitrile solutions when protonated by CF
3
2 6 2
CO H and HSbF ·6H O.
derived salts one pyridinium
group is protonated but hydro-
gen-bonded to either the
Chemical shift (d)
CF
3
CO
2
H
6 2
HSbF ·6H O
ꢀ
ꢀ
or CF SO
3
Compound
H
a
H
b
H
c
H
d
H
a
H
b
H
c
H
d
CF CO
group,
3
2
3
whereas the other pyridinium
group is also protonated but is
not hydrogen-bonded. Thus,
because of the hydrogen-bond-
ing in the crystal, one pyridini-
um group is essentially neutral
while the other is positively
charged. That the pyridinium
groups are p-stacked in these
cases, is no surprise.
8
8
.66
.46
8.58
8.44
8.24
7.86
8.66
8.60
8.46
8.427.88
8.40
7.66
8.50
8.32
7.76
To avoid the effects of hy-
drogen-bonding on the rotam-
er configuration in the solid,
difference (Dd)
ꢀ0.20
ꢀ0.18
ꢀ0.20
ꢀ0.20
ꢀ0.16
ꢀ0.14
ꢀ0.10
ꢀ0.12
ꢀ
we have employed the SbF6
salts, which are not expected to
hydrogen-bond. The di-quaternary methyl compound, 9, was
also investigated in order to see if similar structures were
6ꢀ
Figure 3. Two perspectives of the solid-state structure of 10 as the SbF
salt. Thermal ellipsoids are at 50% probability. The hydrogen atoms
ꢀ
shown are calculated, they were not located. The SbF
clarity.
6
ion is omitted for
those reported bearing one aromatic ring on each Cp ring
have the aromatic rings roughly coplanar with the Cp ring
to which it is attached.
The diprotonated salts of 7 formed from CF COOH and
3
6ꢀ
Figure 4. Two perspectives of the solid-state structure of 9 as the SbF
CF SO H, form crystals that contain molecules that have
3
3
salt. Thermal ellipsoids are at 50% probability. The hydrogen atoms
the pyridinium substituents eclipsed (see Figure 12and
Figure 13 in the Supporting Information). For the cases of
ꢀ
shown are calculated, they were not located. The SbF
for clarity.
6
ions are omitted
8940
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8935 – 8951

Electrostatic Drivers for Synthetic Molecular Motors
FULL PAPER
Table 3. Crystallographic data for 7·2
A
H
R
U
G
6
), 9, and 10.
not dominant in deciding the
rotamer configurations in these
cases, the small Cp–Cp rota-
tion angle suggests that the
Compound
7·2
A
H
R
U
G
6
)
9
10
formula
formula weight
space group
a [Š]
b [Š]
c [Š]
C
840.78
P1
7.341(8)
10.439(11)
18.683(19)
89.914(16)
80.141(17)
71.143(15)
1333(2)
20
H
18FeN
2
+2
2
H
6
O
C
830.68
C2/c
18.055(3)
11.654(2)
13.251(3)
90.0 (3)
107.755(3)
90.0 (3)
2655.3(8)
22
H
22FeN
2
+2SbF
6
20 2 6
C H18FeN + SbF
513.90
C2/c
20.865(5)
6.3412(14)
27.133(6)
90.0
110.312(4)
90.0
¯
Coulombic
repulsion
and
eclipsed p-stacking forces are
similar in energy. The solution
1
a [8]
b [8]
H NMR spectral data and the
calculations shown in Figure 1
and Figure 2are also consistent
with the conclusion that the re-
pulsive and stabilizing forces
are of a similar magnitude. The
p–p-stacking distances py–py
g [8]
V [Š ]
3
3366.7(13)
Z
2
4
8
cryst. size
color, habit
0.20”0.10”0.03 mm
red, needle
2.095
2.647
221(2)
0.20”0.20”0.15 mm
red, fragment
2.078
2.658
100(5)
0.17”0.15”0.04 mm
red-orange, plate
2.028
2.527
100(5)
ꢀ
3
1
calcd, g [cm ]
ꢀ1
m [mm
T [K]
]
(
Table 4) are within the range
expected for such interactions.
If the stabilizing forces in
the eclipsed p-stacked rotamer
are of similar energy to the
electrostatic forces for the di-
pyridinium systems, then it fol-
wavelength [Š]
0.71073 (MoKa
11.56
26.73
)
0.71073 (MoKa
3.67
10.17
)
0.71073 (MoKa)
5.09
8.64
[
a]
R(F) [%]
2
[a]
R
A
C
H
T
R
E
U
N
G
(wF ) [%]
Table 4. Important angles and distances for several pyridine or pyridini-
um substituted ferrocenes.
lows that if the p–p attractive forces were increased to a
modest degree, the electrostatic repulsion may not be suffi-
cient to overcome the stabilizing forces. This was investigat-
ed for the methylpyridinium and 8-aminoquinolyl combina-
tion, 11.
Angles
Distance
Py–Py
Compound
Cp–Cp
8]
Cp–Py
[8]
1
Cp–Py
2
[8]
[
a]
[b]
[b]
[c]
[
[Š]
1
0.85
14.7
–
–
3
2
7.524.10
4.10
3.34
9.62
5.8
6.7
3.28
[
a] Cp–Cp rotation angle; eclipsed angle is 08. [b] Interplanar angle be-
tween Cp and Py rings. [c] Average interplanar separation between the
pyridine rings.
obtained in the solid state for two different quaternized
ꢀ
SbF6 salts (Figure 4 and Figure 5). Both solid state struc-
tures have the substituents rotated away from the eclipsed
p-stacked rotamer. Both structures are in the racemic con-
+
Figure 5. Two perspectives of the solid-state structure of 7(H )
2
as the
formation. The Cp–Cp rotations are similar, 29.628 and
ꢀ
SbF
6
salt. Thermal ellipsoids are at 50% probability. The hydrogen
+
3
7.528 for 7(H ) and 9, respectively. In both cases some p–
ꢀ
2
atoms shown are calculated, they were not located. The SbF6 ions are
omitted for clarity.
p overlap occurs. Assuming that crystal packing forces are
Chem. Eur. J. 2006, 12, 8935 – 8951
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8941

B. Bosnich et al.
Table 5 lists the pyridine
proton chemical shifts and
chemical shift differences (Dd)
for free monoprotonated 11 in
acetonitrile solution as well as
those for the reference mole-
cule 10. It was found that
The N-methylpyridinium group can be regarded as elec-
tron-poor, whereas the 8-aminoquinoline group is electron-
rich; this, together with the greater p-surface of the latter
should lead to greater p–p interaction than when the quino-
line group is replaced by a pyridine moiety. As can be seen
in Table 5, the chemical shift differences (Dd) between 10
and 11 are much larger than observed for difference be-
tween the mono- 8, and dipyridine 7 analogues. This sug-
gests a greater p-stacked population for 11 than for 7, con-
sistent with the assumption that 11 will have a greater p–p
interaction energy.
6
0 equivalents of CF CO H
3 2
were required to fully monop-
rotonate 11 and that 10 equivalents of HSbF ·6H O were
6
2
necessary to achieve full monoprotonation of 11 (see Sup-
porting Information). The large number of acid equivalents
+
The two monoprotonated complexes, 11 (H ), one de-
rived from CF CO H and the other from HSbF ·6H O pro-
3
2
6
2
Table 5. Chemical shifts of the protons of the N-methylpyridinium sub-
tonation, have essentially the same Dd values, suggesting
that both acids fully protonate 11 without significant ion-
pairing. The magnitudes of the shifts clearly indicate that,
on average, the two groups are not fully separated and that
there exist significant populations of p-stacked species.
The crystal structures of 11 and the monoprotonated form
stituents in [D
3
]acetonitrile solutions for the compounds 10 and 11 and
for CF CO H and HSbF
3
2
6
·6H
2
O monoprotonation of 11.
Chemical shift (d)
Compound
H
a
H
b
H
c
H
d
+
ꢀ
1
1(H ) both as SbF salts were determined. The crystallo-
6
graphic data are listed in Table 6 and the structures are
shown in Figure 6 and Figure 7 in two orientations. The sali-
ent parameters are listed in Table 7.
8
7
.63
.89
8.45
7.71
7.83
6.97
8.34
7.45
6
Table 6. Crystallographic data for 11 and the 11·HSbF .
Compound
11
11·HSbF
6
formula
formula weight
space group
a [Š]
b [Š]
c [Š]
C
656.06
P2
7.478(2)
15.871(5)
19.594(6)
90.0
90.0
90.0
2325.3(12)
4
25
H
22FeN
3
+SbF
6
C
889.81
P1
24
H
23FeN
3
+ H
2
O +2SbF
6
¯
1
2
1
2
1
10.348(20)
10.69(2)
14.03(3)
70.44(3)
81.60(3)
69.69(3)
1347(4)
2
difference (Dd)
difference (Dd)
difference (Dd)
ꢀ0.74
ꢀ0.74
7.99
ꢀ0.86
7.35
ꢀ0.89
7.93
a [8]
b [8]
[
a]
[a]
[a]
[a]
g [8]
V [Š ]
8
.37
3
Z
cryst. size
color, habit
0.25”0.15”0.03 mm
red, plate
1.874
1.854
100(5)
0.30”0.20”0.10 mm
black, plate
2.155
2.580
100(5)
ꢀ0.26
ꢀ0.46
7.98
ꢀ0.48
7.34
ꢀ0.41
7.94
ꢀ
3
1
calcd [gcm ]
ꢀ
1
m [mm
T [K]
]
wavelength [Š]
0.71073 (MoKa
)
0.71073 (MoKa
10.30
)
[b]
[b]
[b]
[b]
8.37
[
a]
R(F) [%]
3.31
7.42
2
2
[a]
R
A
H
R
U
G
27.28
1
2
o
2
2
2
2
=
2
[
a] Quantity minimized=R
A
H
R
U
G
;
R=ꢀD/
c
o
2
2
o
2
2
c
ꢀ
(F
o
), D=j(F
o
ꢀF
c
)j w=1/[s (F ) + (aP) + bP], P=[2 + Max(F
o
,
ꢀ0.26
O.
ꢀ0.47
ꢀ0.49
ꢀ0.40
0
)]/3.
[
a] CF
3
CO
2
H. [b] HSbF
6
·6H
2
necessary to monoprotonate 11 is, at least in part, connected
with the repulsion of the neighboring positive charge on the
N-methylpyridinium group. That only monoprotonation
occurs under these conditions was confirmed by the appear-
ance of isosbestic points for the electronic absorption spec-
tra recorded for various concentrations of acid (see Support-
ing Information). Further, for the fully protonated complex
with CF CO H, the conductivity was found to be
The results illustrated in Figure 6 and Figure 7 and tabu-
lated in Table 7 show that the solid-state structures for 11
+
and its monoprotonated form 11(H ) are essentially the
same. These observations most probably indicate greater p–
p stacking interactions for these compounds as compared to
the previous ones. The structure in the solid state, however,
is also governed by the exigencies of crystal packing and the
coincidental structures may be a reflection of these forces
rather than the interplay of electrostatic repulsion forces
3
2
ꢀ
1
2
ꢀ1
209 W cm mol which, as noted, is close to that expected
1
for a 2:1 electrolyte in acetonitrile.
with those associated with p–p interactions. The H NMR
8942
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8935 – 8951

Electrostatic Drivers for Synthetic Molecular Motors
FULL PAPER
Table 7. Important angles and distances for several pyridine or pyridini-
um substituted ferrocenes.
Angles
Cp–Py
[8]
Distance
Py–Q
[Š]
Compound
Cp–Cp
Cp–Q
[8]
[
a]
[b]
[b]
[c]
[
8]
0
.5213.10
8.90
8
3.2
1
.00
7.40
2.10
3.35
[
a] Cp–Cp rotation angle; eclipsed angle is 08. [b] Interplanar angle be-
tween Cp and pyridine (Py) or quinoline (Q) rings. [c] Average interpla-
nar separation between the pyridine and quinoline rings.
necessarily reflect the preferred rotamer especially when
salts are involved.
6ꢀ
Figure 6. Two perspectives of the solid-state structure of 11 as the SbF
Overall, the preceding data indicate that the interaction
between two 1+ charged ferrocene substituents is capable
of causing rotation of the p-stacked entities in acetonitrile
solution. The electrostatic repulsion and the forces that
induce an eclipsed, p-stacked configuration, however, are of
similar energies. If this is the case, the electrostatic repulsion
between 2+ and 1+ charges at similar distances with analo-
gous groups in acetonitrile solution is expected to provide a
higher population of the un-p-stacked rotamers. This issue is
investigated in the next section.
salt. Thermal ellipsoids are at 50% probability. The hydrogen atoms
ꢀ
shown are calculated, they were not located. The SbF
clarity.
6
ion is omitted for
Rotamers formed by stronger repulsive electrostatic forces:
To increase the electrostatic repulsion in these ferrocene
systems, we have prepared the bipyridyl (bipy) compound
1
2 [Eq. (3)]. Attempts to incorporate Pd
A
H
R
U
G
ligand using [Pd(CH CN) Cl ] failed. The product was a
AHCTREUNG
3
2
2
dimer where two PdCl units were bound to a bipy ligand of
one ferrocene and the 3-pyridine of the other ferrocene (see
1
Supporting Information for the crystal structure). The H
NOE cross-peaks are shown in CD CN at 208C for 12. Most
3
of the cross-peaks are unsurprising but the cross-peaks that
are seen between ortho-protons of the Cp units and the at-
tached pyridine groups (indicated by *) indicate that the
+
Figure 7. Two perspectives of the solid-state structure of 11(H )
2
as the
ꢀ
SbF
6
salt. Thermal ellipsoids are at 50% probability. The hydrogen
ꢀ
atoms shown are calculated, they were not located. The SbF
omitted for clarity.
6
ions are
+
data for 11(H ), however, clearly indicate that the equilibri-
um is not completely tilted to the p-stacked rotamer in solu-
tion. Thus, crystal structures, although informative, need not
Chem. Eur. J. 2006, 12, 8935 – 8951
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8943

B. Bosnich et al.
pyridine groups are freely rotating about the Cp units. In
principle, cross-peaks are expected between the pyridine on
one Cp unit and the corresponding pyridine on the other.
Despite considerable effort, none was observed between the
pyridine groups on different Cp units.
As shown in Equation (4), the 3-pyridine group of 12 can
be methylated selectively and this product can form the
PdCl complex to the bipy group. This product can then be
2
converted to the Pd(Py)2 and Pd
A
H
R
U
G
[Eq. (5)]. In Figures 8–12are shown the X-ray diffraction
crystal structures of systems carrying a variety of charges.
ꢀ
All charged crystals are formed as SbF6 salts. Table 8 lists
6ꢀ
Figure 10. Two perspectives of the solid-state structure of 14 as the SbF
ꢀ
salt. Thermal ellipsoids are at 50% probability. The SbF
drogen atoms are omitted for clarity.
6
ion and the hy-
the crystallographic data for each of the crystals. Important
data related to p-stacking of these complexes are given
Table 9.
As can be seen from Figure 8, Figure 9 and Figure 10 and
the data in Table 9, the pyridine or pyridinium group is p-
stacked on the Cp-bound pyridine of bipy ligand. The two
substituents are essentially eclipsed (Cp–Cp angles are
small). In all of the cases the nitrogen atom of the pyridine
or pyridinium group is trans to the corresponding nitrogen
of the Cp-bound pyridine group of bipy. The dicationic
system 15, shows small Cp–Cp rotation (17.348) but consid-
erable p–p overlap remains (see Figure 13 in the Supporting
Information). The distance between N(5) and Pd(1) is
Figure 8. Two perspectives of the solid-state structure of 12. Thermal el-
lipsoids are at 50% probability. The hydrogen atoms shown are calculat-
ed, they were not located.
6
.22 Š. This distance can be compared to that found for the
ꢀ
pyridinium N(1) and N(2) distances of the SbF salts of
7
6
+
(H ) and 9; these are 5.66 Š and 6.16 Š, respectively.
2
These similar distances indicate that Coulombic repulsion
may be the determining factor in the rotation of the Cp
rings, as assumed. It will be noted that for the palladium
complex 15, a distance of 6.22 Š between the metal ion and
the pyridinium nitrogen atom will result in more p-overlap
between the corresponding pyridine rings than in the cases
+
of 7(H ) and 9. For the tricationic system, 16, however, the
2
pyridinium and the dicationic [bipyPd(py) ] units are rotated
2
to almost maximum mutual displacement (Figure 12). The
N-methylpyridinium groups are disordered in the crystal,
with almost equal population of the two shown positions.
The Cp carbon atoms attached to the N-methylpyridinium
group are not resolved for the two positions. It is clearly dis-
ordered (Figure 12). The distances between the two posi-
tions of the quaternary nitrogen atom and the palladium
6ꢀ
Figure 9. Two perspectives of the solid-state structure of 13 as the SbF
salt. Thermal ellipsoids are at 50% probability. The hydrogen atoms
ꢀ
shown are calculated, they were not located. The SbF
clarity.
6
ion is omitted for
8944
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8935 – 8951

Electrostatic Drivers for Synthetic Molecular Motors
FULL PAPER
6ꢀ
Figure 12. Two perspectives of the solid-state structure of 16 as the SbF
ꢀ
salt. Thermal ellipsoids are at 50% probability. The SbF
hydrogen atoms are omitted for clarity.
6
ions and the
atom are 10.85 and 11.10 Š. This is the first case in which
the two groups are fully displaced from each other in the
solid state. As noted previously, the crystal structure de-
pends not only on the intramolecular forces but also on the
crystal packing interactions. Even so, the structures in the
solid state are broadly consistent with chemical shift data in
CD CN solutions at 278C (Table 10).
3
The data in Table 10 indicate that the compounds 12–14
exist in mainly the eclipsed, p-stacked rotameric conforma-
tion. The cationic compound 13 appears to have a higher
6ꢀ
Figure 11. Two perspectives of the solid-state structure of 15 as the SbF
ꢀ
salt. Thermal ellipsoids are at 50% probability. The SbF
hydrogen atoms are omitted for clarity.
6
ions and the
Table 8. Crystallographic data for 12, 13, 14, 15, and 16.
Compound
12
13
14
16
15
formula
formula
weight
space group
a [Š]
b [Š]
c [Š]
a [8]
b [8]
C
834.56
25
H
19FeN
3
C
26
H
22FeN
3
+SbF
6
C
26
H
22Cl
2
FeN
3
Pd+C
2
H
3
N+SbF
6
C
34
H
31FeN
4
OPd+C
3
H
6
O+2SbF
6
C
36
H
32FeN
5
Pd+3SbF
6
66 08 3. 0. 47 6
886.4212 1397.11
¯
¯
P
P2
1
P1
7.580(2)
P1
1
P2 /c
9.643(2)
11525(3)
17.768(4)
79.162(4)
80.400(4)
70.768 (4)
1819.2(7)
4
8.117(3)
10.183(4)
14.664(6)
90.0
101.720(6)
90.0
8.496(4)
15.244(7)
16.211(8)
8 2. 548(8)
82.125(8)
78.219(8)
2024.7(17)
2
18.949(7)
11.741(4)
19.635(7)
90.0
101.035(6)
90.0
10.167(3)
19.454(6)
90.515(5)
100.723(5)
98.632(5)
1455.3(7)
2
g [8]
V [Š ]
3
1186.8(8)
2
4288(3)
4
Z
cryst. size
color, habit
0.20”0.15”0.04 mm 0.20”0.05”0.02 mm 0.20”0.10”0.03 mm
0.40”0.20”0.03 mm
red-brown, plate
1.974
2.203
100(5)
0.25”0.20”0.03 mm
purple, plate
2.164
2.721
100(5)
red, plate
1.524
red, ribbon
1.869
red, plate
2.023
ꢀ
3
1
calcd [gcm
]
ꢀ1
m [mm
T [K]
]
0.846
1.818
2.275
100(5)
100(5)
100(5)
wavelength
Š]
R(F) [%]
0.71073 (MoKa
)
0.71073 (MoKa
)
0.71073 (MoKa
)
0.71073 (MoKa
)
0.71073 (MoKa)
[
[
a]
4.69
10.80
4.84
7.23
5.11
10.31
5.31
11.12
7.75
19.30
2
[a]
R(wF ) [%]
1
2
2
o
2
c
2
2
o
2
=
2
2
2
o
2
2
[
a] Quantity minimized=R(wF )=ꢀ[w(F ꢀF ) ]/ꢀ[(wF ) ] ; R=ꢀD/ꢀ(F
o
), D=j(F
o
ꢀF
c
)j w=1/[s (F ) + (aP) + bP], P=[2Fc + Max(F
o
, 0)]/3.
Chem. Eur. J. 2006, 12, 8935 – 8951
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8945

B. Bosnich et al.
Table 9. Important angles and distances for the crystal structures illus-
trated in Figures 8–12.
Table 10. Chemical shifts of the protons of the pyridine or N-methylpyri-
dinium substituent for the compounds 12 to 16 in [D ]acetonitrile solu-
3
tion.
Angles
Cp–Py
[8]
Distance
Py–Py
[Š]
Compound
Cp–Cp
Cp–Bipy
[8]
Chemical Shift (d)
[
a]
[b]
[b]
[c]
[
8]
Compound
H
a
H
b
H
c
H
d
8
8
.73
.48
8.38
7.24
7.86
2
1
.71
.44
5.10
9.00
3.39
3.26
8.08
ꢀ0.30
8.45
6.90
ꢀ0.34
7.83
7.60
ꢀ0.26
8.34
17.30
24.50
difference (Dd)
ꢀ0.25
7
.43
8.30
11.10
3.29
3.24
–
8.63
8
.13
8.06
7.36
7.87
1
7.34
11.90
8.80
difference (Dd)
ꢀ0.50
ꢀ0.39
ꢀ0.48
ꢀ0.47
1
59.64
160.39
5.90
5.90
8
.55
8.39
7.46
7.83
16.60
ꢀ
difference (Dd)
ꢀ0.08
ꢀ0.06
ꢀ0.37
ꢀ0.51
[
a] Cp-Cp rotation angle; eclipsed angle is 08. [b] Interplanar angle be-
tween Cp and pyridine (or pyridinium), py, and the Cp attached pyridine
of bipy, Cp-bipy. [c] Average interplanar separation between the pyri-
dines on the different Cp units.
8
.54
8.37
ꢀ0.08
8.36
7.66
ꢀ0.17
7.81
8.08
ꢀ0.26
8.24
population of the eclipsed, p-stacked form than the neutral
compound 12, an observation that supports the assertion
that the p–p stacking is more effective for charged–neutral
than for neutral–neutral aromatic pairs of molecules. The
proton chemical shift differences of 14 are different from
those observed for 12 and 13, in that 14 has only a small up-
field shift for the H and H protons. The H and H protons
difference (Dd)
ꢀ0.08
8
.55
a
b
c
d
of 14, on the other hand, show substantial upfield shifts.
These observations suggest that the favored p-stacked con-
formation for 12 and 13 has all four protons (H to H )
difference (Dd)
ꢀ0.08
ꢀ0.09
ꢀ0.02
ꢀ0.10
a
d
roughly equally disposed over the requisite pyridine group
of bipy, whereas 14 has the pyridinium group displaced such
that H and H lie over the requisite pyridine of the bi-
c
d
pyPdCl2 and that the corresponding Ha and Hb are not
aligned over the p-cloud of the same pyridine. The dication-
ic complex 15 shows chemical shifts of its pyridinium pro-
tons which suggest that the major rotameric conformation is
a structure that has the two groups rotated but some overlap
remains for the Hc and Hd protons. For the tricationic
8946
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8935 – 8951

Electrostatic Drivers for Synthetic Molecular Motors
FULL PAPER
1
system, H NMR chemical shifts indicate the least amount
system (Figure 11) when a pair of 1+ charges is involved
the most populated rotamer is one where considerable p–p
interaction remains. It is perhaps surprising that electrostatic
repulsions between ferrocene substituents are of a similar
magnitude to the attractive interactions between p-substitu-
ents. Probably the high dielectric constant of acetonitrile at-
tenuates the Coulombic repulsion as the two charges sepa-
rate.
of p-stacking in solution, as expected from the electrostatic
considerations and from the results of X-ray diffraction.
+
Reversibility: The diprotonated compound 7(H ) and the
monoprotonated system 11(H ), revert to their original ro-
2
+
tameric conformations when the protons are neutralized by
addition of triethylamine to acetonitrile solutions. Similarly,
the dipyridine palladium complex 16 is converted cleanly to
the dichloro analogue 14, upon addition of two equivalents
of tetra-n-butylammonium chloride in acetonitrile solution.
Thus in both cases these electrostatic drivers are cleanly re-
versible, a condition necessary for the development of mo-
lecular motors.
During the reviewing time of this article, two reports on
the use of ferrocenes as rotatory motors have appeared.
One of these involved a gas-phase study of the preferred ro-
tameric conformation of the ferrocene carboxylates [Fe-
ꢀ
2ꢀ
ꢀ
ꢀ
A
H
R
U
(C H CO ) ] and [Fe
A
H
R
U
G
A
H
R
U
G
5
4
2
2
5
4
2
[27]
shown
that the conformation of the dicarboxylate was
where the two carboxylate groups were rotated away from
[28]
each other. The other report, concerned the use of ferro-
cene rotation for inducing concerted rotation about a distant
single bond when the ferrocene rings were caused to rotate
by photochemical isomerization of diazo groups. These two
examples demonstrate that the ferrocene group is likely to
find increasing use in the design of rotatory motors.
The data presented here serve to indicate the approxi-
mate amount of protonmotive torque that can be generated
by these electrostatic drivers and are a prerequisite for de-
veloping the ferrocene platform for the construction of mo-
lecular motors.
Discussion
This study of the potential of using substituted ferrocenes as
protonmotive drivers for developing molecular motors and
machines has shown that it may be possible to use them for
this purpose. It is clear, however, that in solvents of high di-
electric constant, such as acetonitrile, the amount of torque
generated by electrostatic repulsion by the dicationic sys-
tems is insufficient to drive the two substituents away from
populations where some p-overlap occurs. For the tricationic
driver 16, however, sufficient torque is present to essentially
uncouple the two substituents.
This report has not investigated in detail one other com-
plicating factor in developing these electrostatic drivers. As
the dielectric constant of the medium is decreased, the elec-
trostatic interaction will be increased but the tendency for
ion-pairing will also increase. These two phenomena are
likely to act in opposition to each other and lead to diminu-
tion of the electrostatic repulsion that is gained in solvents
of low dielectric constant.
Experimental Section
General procedures: All reagents were obtained from commercial suppli-
ers and were used without further purification. All reactions were per-
formed under an atmosphere of argon, unless otherwise specified. H and
C NMR spectra were recorded using a Bruker DRX500 or a Bruker
DMX500 Fourier transform spectrometer. Proton and carbon chemical
shifts, d, are reported in ppm, and referenced to tetramethylsilane
TMS). Coupling constants, J, are reported in hertz. Electronic absorp-
1
1
3
(
Overall, the results observed in solution are consistent
with those observed in the solid state, despite the possible
effects of crystal packing on the solid-state structures. As
noted earlier and as is shown in Figure 1 and Figure 2, the
approximate electrostatic calculations indicate that the Cou-
lombic interactions for a pair of 1+ substituents are similar
to the stabilizing interactions that lead to the eclipsed con-
formation of the Cp rings and of the two substituents. The
results presented here are consistent with these conclusions.
Inspection of Table 1, Table 2, and Table 4 and the struc-
tures shown in Figure 4 and Figure 5 indicate the most popu-
lated rotamers of the di-1+ systems are those where some
overlap between the two pyridine substituents remains. The
data in Table 5 and Table 7 and the structures shown in
Figure 6 and Figure 7 indicate that the greater p–p interac-
tion (than for the dipyridine systems) between the two sub-
stituents is nearly sufficient to overcome the electrostatic re-
pulsion. Consistently, as illustrated by the structures in Fig-
tion spectra were obtained by using a Perkin Elmer Lambda 6 UV/Vis
spectrophotometer. For studying protonation of substrates at a given
wavelength and acid concentration, the absolute change in absorbance, (j
DAj), was calculated by subtracting the acid dependent absorbance from
the absorbance in the absence of acid, (jDAj
o
). The absolute change in
chemical shift, (jDdj) is calculated similarly. Elemental analyses were
performed by Desert Analytics, Tucson, Arizona. Conductance measure-
ꢀ3
ments were performed at 238C with 1”10 m samples using an YSI Sci-
entific model 35 conductance meter. Acetonitrile was dried over CaH
2
,
tetrahydrofuran (THF) was dried over potassium/benzophenone ketyl,
diethyl ether was dried over sodium/benzophenone ketyl, and methylene
chloride was dried over CaH . Dry trifluoroacetic acid (TFA) was ob-
2
tained by distillation from a solution containing 5% trifluoroacetic anhy-
dride. Thin-layer chromatography was carried out using precoated silica
gel (Whatman PE SIL G/UV) or precoated aluminum oxide (J. T. Baker,
aluminum oxide IB-F). Silica gel 60 Š (Merck, 230–400 mesh) and alumi-
num oxide 58 Š (either activated, basic, Brockman I or activated, neu-
tral, Brockman I) were used for chromatography as indicated. Celite is
[
21]
J. T. Baker Celite 503. Ferrocene-1,1’-diboronic acid, 5-bromo-2,2’-bi-
[
29]
pyridine
were prepared by literature methods. Ferroceneboronic
graphite, and hexafluoroantimonic acid hexahydrate
·6H O) were obtained from the Aldrich Chemical Co. and used
as received.
3-Pyridylferrocene] (8): A 100-mL flask was charged with 3-bromopyri-
dine (1.00 g, 6.32mmol) dissolved in a mixture of dioxane (10 mL) and
[
30]
acid,
HSbF
(
6
2
1
ures 10–12and by the H NMR spectral data in Table 10 the
two substituents can be completely separated when a 2+
with 1+ interaction obtains (Figure 12) but for a similar
[
Chem. Eur. J. 2006, 12, 8935 – 8951
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8947

B. Bosnich et al.
1
m aqueous Na
2
CO
3
(10 mL). The catalyst PdCl
2
A
H
E
N
[1,1’-bis(diphenylphos-
tion and was washed with cold CH
2
Cl
2
, Et
2
O, and pentane. The red iodo
phino)ferrocene] (0.250 g, 0.34 mmol) was added to the reaction mixture
followed by a mixture of ferroceneboronic acid (0.85 g, 3.69 mmol) and
6
salt (0.170 g) was dissolved in acetone (3 mL) and AgSbF (0.188 g,
0.55 mmol) was added as a solid to the red solution. AgI precipitated im-
mediately and the mixture was stirred at room temperature for 30 min.
The resulting red suspension was filtered through celite and the solvent
was removed under reduced pressure. The solid red residue was dissolved
3
m aqueous NaOH (3 mL) in DME (10 mL). The reaction mixture was
refluxed for 24 h. At the end of this period the reaction mixture was stir-
red until it had cooled to room temperature and was then poured into
ice-water (200 mL). The resulting mixture was extracted with ethyl ace-
into CH
2 2
Cl and was vapor-diffused with hexanes. Red blocks formed
tate (3”100 mL). The organic layer was washed with NH
100 mL), water (100 mL), and brine (100 mL). The resulting orange sol-
ution was dried with Na SO and concentrated to an orange-red solid.
The solid residue was dissolved in CH Cl and chromatographed on basic
alumina with CH Cl /hexanes (50:50). The eluted solvent was removed
under reduced pressure to yield an orange powder, which was recrystal-
lized by vapor diffusion of CH Cl solution with hexanes. The orange
needles (0.63 g, 55%) were collected by filtration and were washed with
4
Cl solution
and were filtered and were washed with cold hexanes and vacuum-dried
1
(
(0.185 g, 75%). H NMR (CD
3
CN, 278C, 500 MHz): d=4.25 (s, 6H), 4.56
1
(sb, 4H), 4.90 (sb, 4H), 7.76 (m, 2H), 8.24 (d, J=8.0 Hz, 2H), 8.38 (d,
2
4
1
J=5.6 Hz, 1H), 8.55 ppm (s, 1H); conductivity:
L
M
A
H
R
U
G
3
CN)=
2
2
ꢀ
1
2
ꢀ1
259 W cm mol ; ESI-MS (CH
3
CN): m/z 606 [MꢀSbF
6
2
2
ysis calcd (%) for C22
31.34, H 2.55, N 3.03.
H
22
F
12FeN
2
Sb
2
2
3
3
-Bromo-8-nitroquinoline and 3-bromo-5-nitroquinoline: To a solution of
-bromoquinoline (10.0 g, 48.0 mmol) in concentrated H SO (20 mL)
1
3
cold hexanes and vacuum-dried. H NMR (CD CN, 278C, 500 MHz): d=
2
4
1
1
4
1
1
.04 (s, 5H), 4.37 (t, J=1.8 Hz, 2H), 4.81 (t, J=1.8 Hz, 2H), 7.26 (m,
cooled in a water/ice bath was added dropwise 16 mL of a concentrated
SO /concentrated HNO mixture (6:2). The reaction mixture was
1
H), 7.88 (m, 1H), 8.41 (dd, J=4.7 Hz, J
H); ESI-MS (MeOH): m/z: 263 [M], 264 [M+1]; conductivity: L
2
=1.5 Hz, 1H), 8.80 ppm (s,
H
2
4
3
M
-
maintained at low temperature, stirred rapidly, and monitored by thin-
layer chromatography until all of the 3-bromoquinoline had been con-
sumed (ca. 3 h). The mixture was diluted with water (50 mL) and NaOH
added until the solution reached pH 10–11. The resulting solution was ex-
tracted with diethyl ether (2”50 mL), the organic phase dried over anhy-
drous magnesium sulfate, and the solvent was removed under reduced
pressure to give an ꢂ8:2mixture of 3-bromo-5-nitroquinoline and 3-
bromo-8-nitroquinoline, respectively. Crystallization of this mixture from
ꢀ
1
2
ꢀ1
A
C
H
T
R
E
U
N
G
(CH
N-Methyl-3-pyridylferrocene]
with 8 (0.130 g, 0.49 mmol) dissolved in CH
.30 mL, 4.94 mmol) was added to the orange solution and the reaction
3
CN)=0.84 W cm mol .
[
A
H
R
U
G
6
) (10): A 10-mL flask was charged
Cl (2mL). CH I (0.69 g,
2
2
3
0
mixture was stirred at room temperature for 16 h, during which time a
red precipitate formed. The precipitate was isolated by filtration and was
washed with cold CH
was dissolved in acetone (3 mL) and AgSbF
2
Cl
2
, Et
2
O and pentane. The red iodo salt (0.200 g)
(0.170 g, 0.49 mmol) was
1
ethyl acetate afforded 3-bromo-5-nitroquinoline (8.8 g, 72%). H NMR
6
1
2
1
added as a solid to the red solution. A precipitate formed immediately
and the mixture was stirred at room temperature for 30 min. The orange
suspension was filtered through celite and the solvent was removed
under reduced pressure. The solid orange residue was dissolved into
(CDCl
3
): d=9.26 (dd, J=2.1 Hz, J=0.7 Hz, 1H), 9.04 (d, J=6.4 Hz,
1
1H), 8.44 (m, 2H), 7.86 ppm (t, J=8.1 Hz, 1H); ESI-MS (MeOH): m/z:
9 5 2 2
253 [M], 254 [M+1]; elemental analysis calcd (%) for C H BrN O : C
42.72, H 1.99, N 11.07; found: C 42.59, H 2.15, N 11.28.
CH
2
Cl
2
and vapor diffused with hexanes. Orange needles were obtained.
The solvent was removed from the mother liquor to yield an off-white
residue which contains both 3-bromo-8-nitroquinoline and 3-bromo-5-ni-
troquinoline. This residue was recrystallized twice from boiling cyclohex-
anes to yield the desired 3-bromo-8-nitroquinoline (2.05 g, 17%).
They were collected and were washed with cold hexanes and vacuum-
dried (0.220 g, 95%). H NMR (CD CN, 278C, 500 MHz): d=4.15 (s,
3
1
1
1
5
H), 4.25 (s, 3H), 4.60 (t, J=1.8 Hz, 2H), 4.90 (t, J=1.9 Hz, 2H), 7.82
1
1
1
1
1
(
m, 1H), 8.34 (d, J=5.9 Hz, 1H), 8.45 (d, J=8.2Hz, 1H), 8.63 ppm (s,
3
H NMR (CDCl ): d=9.05 (d, J=2.1 Hz, 1H), 8.43 (d, J=2.1 Hz, 1H),
1
1
1
1
1
3
H); ESI-MS (CH
24 W cm mol ; elemental analysis calcd (%) for C16
7.39, H 3.14, N 2.73; found: C 37.71, H 2.98, N 2.35.
3
CN): m/z 278 [MꢀSbF
6
]; conductivity: L
M
A
H
R
U
G
3
CN)=
8.05 (d, J=7.4 Hz, 1H), 7.96 (d, J=8.1 Hz, 1H), 7.66 ppm (t, J=
7.8 Hz, 1H); ESI-MS (MeOH): m/z: 253 [M], 254 [M+1]; elemental
analysis calcd (%) for C H BrN O : C 42.72, H 1.99, N 11.07; found: C
ꢀ
1
2
ꢀ1
16 6
H F
9
5
2
2
A
C
H
T
R
E
U
N
G
[1,1’-(3-Pyridyl)ferrocene] (7): In 250-mL flask 3-bromopyridine (3.00 g,
8.98 mmol) was dissolved in a mixture of dioxane (50 mL) and 1m aque-
ous Na CO (40 mL). The catalyst [PdCl (1,1’-bis(diphenylphosphino)fer-
42.76, H 2.01, N 10.98.
1
[1-(3-Pyridyl)-1’-(boronic acid)ferrocene] (18): Into a 100-mL flask was
added 3-bromopyridine (1.16 g, 7.31 mmol) dissolved in a mixture of di-
oxane (10 mL) and 1m aqueous Na CO (10 mL). The catalyst [PdCl2-
2
3
2
ACHTREUNG
rocene)] (0.60 g, 0.82mmol) was added to the reaction mixture followed
by a mixture of ferrocene-1,1’-diboronic acid (2.00 g, 7.31 mmol) and 3m
aqueous NaOH (5 mL) in DME (35 mL). The reaction mixture was re-
fluxed for 24 h. At the end of this time, the reaction mixture was stirred
until it had cooled to room temperature and was then poured into ice-
water (250 mL). The resulting mixture was extracted with ethyl acetate
2
3
A
H
R
U
G
the reaction mixture followed by a mixture of ferrocene-1,1’-diboronic
acid (2.00 g, 7.31 mmol) and 3m aqueous NaOH (3 mL) in DME
(10 mL). The reaction mixture was refluxed for 12min (5 min ramp,
7 min hold) in a CEM microwave reactor (power 150 W). At the end of
this period the reaction mixture was stirred until it had cooled to room
temperature and was then poured into ice-water (200 mL). The resulting
mixture was extracted with ethyl acetate (3”100 mL). The organic layer
(
3”100 mL). The organic layer was washed with NH
100 mL), water (100 mL), and Brine (100 mL). The resulting orange sol-
SO and was concentrated to an orange-red
solid. The solid residue was dissolved into CH Cl and was chromato-
graphed on silica gel with CH Cl /MeOH (98:2). The eluted solvent was
removed under reduced pressure yielding an orange powder, which was
recrystallized by vapor diffusion of CH Cl solution with hexanes. The
4
Cl solution
(
ution was dried over Na
2
4
2
2
was washed with NH Cl solution (100 mL), water (100 mL), and brine
4
2
2
(100 mL). The resulting orange solution was dried with Na SO and was
2
4
concentrated to an orange-red solid. The solid residue was dissolved into
acetone and chromatographed on silica gel with acetone as the eluant.
The eluted solvent was removed under reduced pressure yielding an
orange powder, which was slurried in hexanes, sonicated, then filtered.
The orange powder (1.36 g, 62%) was washed with cold hexanes and
2
2
orange needles (1.68 g, 68%) were collected by filtration and were
washed with cold hexanes and were vacuum-dried. H NMR
(
1
1
1
1
1
6
[D ]acetone, 278C, 500 MHz): d=4.34 (t, J=1.86 Hz, 4H), 4.75 (t, J=
1
2
1
1
.88 Hz, 4H), 7.06 (m, 2H), 7.58 (m, 2H), 8.31 (dd, J=4.7 Hz, J=
.5 Hz, 2H), 8.54 ppm (d, J=1.8 Hz, 2H); ESI-MS (MeOH): m/z: 340
6
vacuum-dried. H NMR ([D ]acetone, 278C, 500 MHz): d=4.20 (t, J=
1
1
1
1.81 Hz, 2H), 4.34 (t, J=1.90 Hz, 2H), 4.36 (t, J=1.82Hz, 2H), 4.74 (t,
ꢀ
1
2
ꢀ1
1
1
[
M], 341 [M+1]; conductivity: L
M
A
H
R
U
G
3
CN)=0.59 W cm mol ; elemen-
J=1.91 Hz, 2H), 6.71 (sb, 2H), 7.24 (m, 1H), 7.90 (m, 1H), 8.36 (d, J=
tal analysis calcd (%) for C20
H
16FeN
2
4.64 Hz, 1H), 8.73 ppm (s, 1H); ESI-MS (MeOH): m/z 306 [M], 307
7
0.72, H 5.01, N 8.18.
[1,1’-(N-Methyl-3-pyridyl)ferrocene]
charged with 7 (0.100 g, 0.29 mmol) dissolved in CH
0.417 g, 0.18 mL, 2.95 mmol) was added to the orange solution and the
[M+1], 336 [M+2MeOHꢀ2H O]; elemental analysis calcd (%) for
2
15 2
C H14BFeNO : C 58.70, H 4.60, N 4.56; found: C 60.25, H 4.70, N 4.45.
A
C
H
T
R
E
U
N
G
A
H
R
N
(SbF
6
)
2
(9):
A
10-mL flask was
Cl (2mL). CH
2
2
3
I
[1-(3-Pyridyl)-1’-(3-(8-nitroquinoline)ferrocene] (19): Into a 50-mL flask
was added 3-bromo-8-nitroquinoline (1.00 g, 3.95 mmol) dissolved in a
(
reaction mixture was stirred at room temperature for 16 h, during which
time a deep red precipitate formed. The precipitate was isolated by filtra-
mixture of dioxane (10 mL) and 1m aqueous Na
2 3
CO (10 mL). The cata-
lyst [PdCl (1,1’-bis(diphenylphosphino)ferrocene)] (0.40 g, 0.54 mmol)
2
ACHTREUNG
8948
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8935 – 8951

Electrostatic Drivers for Synthetic Molecular Motors
FULL PAPER
was added to the reaction mixture followed by a mixture of 18 (1.00 g,
ture was stirred until it had cooled to room temperature and was then
poured into ice-water (100 mL). The resulting mixture was extracted with
3
.26 mmol) and 3m aqueous NaOH (3 mL) in DME (10 mL). The reac-
tion mixture was refluxed for 24 h. At the end of this period the reaction
mixture was stirred until it had cooled to room temperature and was
then poured into ice-water (100 mL). The resulting mixture was extracted
with ethyl acetate (3”100 mL). The organic layer was washed with
ethyl acetate (3”100 mL). The organic layer was washed with NH
ution (100 mL), water (100 mL), and brine (100 mL). The resulting
orange solution was dried over Na SO and was concentrated to an
orange-red solid. The solid residue was dissolved into CH Cl and was
chromatographed on silica gel with CH Cl /acetone (90:10). The eluted
4
Cl sol-
2
4
2
2
NH
sulting orange solution was dried with Na
orange-red solid. The solid residue was dissolved into CH
matographed on silica gel with CH Cl /THF (90:10). The eluted solvent
4
Cl solution (100 mL), water (100 mL), and brine (100 mL). The re-
SO and concentrated to an
Cl and chro-
2
2
2
4
solvent was removed under reduced pressure yielding orange powder,
which was recrystallized by vapor-diffusion of an acetone solution with
hexanes. The orange needles (0.75 g, 70%) were collected by filtration
2
2
2
2
1
was removed under reduced pressure yielding a red-orange powder,
which was recrystallized by vapor diffusion of an acetone solution with
hexanes. The red-orange needles (0.75 g, 53%) were collected by filtra-
tion and were washed with cold hexanes and were vacuum dried.
and were washed with cold hexanes and were vacuum dried. H NMR
1
1
([D
6
]acetone, 278C, 500 MHz): d=4.38 (t, J=1.8 Hz, 2H), 4.40 (t, J=
1
1
1.8 Hz, 2H), 4.80 (t, J=1.8 Hz, 2H), 4.84 (t, J=1.8 Hz, 2H), 6.96 (m,
1H), 7.39 (m, 1H), 7.58 (m, 1H), 7.70 (dd, J=8.2Hz, J=2.3 Hz, 1H),
7.91 (m, 1H), 8.18 (m, 2H), 8.44 (d, J=7.9 Hz, 1H), 8.55 (d, J=1.6 Hz,
1H), 8.59 (d, J=2.1 Hz, 1H), 8.67 ppm (m, 1H); ESI-MS (MeOH): m/z:
1
2
1
1
1
1
H NMR ([D
6
]acetone, 278C, 500 MHz): d=4.43 (t, J=1.9 Hz, 2H), 4.51
1
1
1
1
(
(
1
8
t, J=1.9 Hz, 2H), 4.86 (t, J=1.9 Hz, 2H), 5.02 (t, J=1.9 Hz, 2H), 6.64
1
1
2
m, 1H), 7.35 (m, 1H), 7.69 (t, J=7.4 Hz, 1H), 7.84 (dd, J=4.8 Hz, J=
.6 Hz, 1H), 7.94 (dd, J=8.2Hz, J=1.3 Hz, 1H), 8.01–8.03 (m, 1H),
.05 (d, J=2.3 Hz, 2H), 8.41 (m, 1H), 8.91 ppm (d, J=2.1 Hz, 1H);
3
417 [M], 418 [M+1]; elemental analysis calcd (%) for C25H19FeN : C
71.96, H 4.59, N 10.07; found: C 72.08, H 4.69, N 9.98.
1
2
1
1
[
1-(N-Methyl-3-pyridyl)-1’-(5-2,2’-dipyridyl)ferrocene]
mL flask was charged with 12 (0.150 g, 0.35 mmol) dissolved in CH
2mL). CH I (0.507 g, 3.59 mmol) was added to the orange solution and
A
H
R
U
G
6
) (13): A 10-
ESI-MS (MeOH): m/z 435 [M], 436 [M+1]; elemental analysis calcd (%)
for C24 : C 66.23, H 3.94, N 9.65; found: C 66.50, H 4.31, N
.46.
1-(N-Methyl-3-pyridyl)-1’-(3-(8-nitroquinoline)ferrocene]
0-mL flask was charged with 19 (0.150 g, 0.34 mmol) dissolved in
CH Cl (3 mL). CH I (0.520 g, 3.68 mmol) was added to the red solution
2
Cl
2
H
3 2
17FeN O
(
3
9
the reaction mixture was stirred at room temperature for 16 h, during
which time a deep red precipitate formed. The precipitate was isolated
[
A
H
R
U
G
6
(SbF ) (20): A
1
by filtration and was washed with cold CH
2 2 2
Cl , Et O, and pentane. The
2
2
3
deep red iodo salt (0.200 g) was dissolved in acetone (5 mL) and AgSbF
6
and the reaction mixture was stirred at room temperature for 16 h,
during which time a deep red precipitate formed. The precipitate was iso-
(0.122 g, 0.35 mmol) was added as a solid to the red solution. AgI precipi-
tated immediately and the mixture was stirred at room temperature for
30 min. The reaction mixture was filtered through celite and the solvent
was removed under reduced pressure. The solid red residue was dissolved
lated by filtration and was washed with cold CH
The deep red iodo salt (0.191 g) was dissolved in acetone (5 mL) and
AgSbF (0.114 g, 0.33 mmol) was added as a solid to the red solution.
2 2 2
Cl , Et O, and pentane.
6
into CH
2 2
Cl and was vapor-diffused with hexanes. The red needles so ob-
AgI precipitated immediately and the mixture was stirred a room tem-
perature for 30 min. The reaction mixture was filtered through celite and
the solvent removed under reduced pressure. The solid red residue was
dissolved into acetone and was vapor-diffused with hexanes. The red
tained were filtered and were washed with cold hexanes and vacuum-
dried (0.220 g, 90%). H NMR (CD CN, 278C, 500 MHz): d=4.01 (s,
3
1
1
1
1
3H), 4.51 (t, J=1.8 Hz, 2H), 4.63 (t, J=1.8 Hz, 2H), 4.87 (t, J=1.8 Hz,
1
2H), 4.92 (t, J=1.8 Hz, 2H), 7.35 (m, 1H), 7.45 (m, 1H), 7.55 (m, 1H),
1
1
blocks so obtained were collected by filtration and then were washed
7.87 (d, J=5.9 Hz, 1H), 7.95 (m, 1H), 7.97 (d, J=8.2Hz, 1H), 8.05 (d,
J=8.2Hz, 1H), 8.13 (s, 1H), 8.55 (m, 2H), 8.70 ppm (m, 1H); conduc-
1
1
with cold hexanes and vacuum-dried (0.185 g, 78%). H NMR (CD
3
CN,
ꢀ
1
2
ꢀ1
2
78C, 500 MHz): d=3.92 (s, 3H), 4.59 (sb, 2H), 4.65 (sb, 2H), 4.88 (sb,
tivity:
[MꢀSbF
3.32, N 6.29; found: C 46.29, H 3.50, N 5.84.
1-(N-Methyl-3-pyridyl)-1’-(5-2,2’-dipyridyl)ferrocenePdCl
mL flask was charged with 13 (0.130 g, 0.19 mmol) dissolved in acetone
8 mL). [Pd(NCCH Cl ] (0.051 g, 0.19 mmol) was added as a solid to
L
M
A
H
R
U
G
3
CN)=137 W cm mol
;
ESI-MS (CH
3
CN): m/z: 432
Sb: C 46.74, H
1
2
H), 5.04 (sb, 2H), 7.00 (m, 1H), 7.70 (d, J=7.8 Hz, 1H), 7.75 (m, 1H),
6
H
22
F
6
FeN
3
1
1
1
7
.77 (d, J=8.1 Hz, 1H), 7.98 (d, J=8.1 Hz, 1H), 8.02(d, J=7.4 Hz,
1
1
1
H), 8.05 (d, J=2.0 Hz, 1H), 8.14 (s, 1H), 8.65 ppm (d, J=2.0 Hz, 1H);
[
2
] (14): A 25-
ꢀ
1
2
ꢀ1
conductivity: L
M
A
C
H
T
R
E
U
N
G
(CH
3
CN)=137 W cm mol ; ESI-MS (CH
3
CN): m/z:
: C
4
50 [MꢀSbF
6
]; elemental analysis calcd (%) for C25
H
20
F
6
FeN O
3 2
(
A
H
R
U
G
3
)
2
2
4
3.77, H 2.94, N 6.13; found: C 44.00, H 2.98, N 6.21.
the red solution. The reaction mixture was stirred at room temperature
for 1 h, during which time a purple solid formed. The solid was isolated
[
1-(N-Methyl-3-pyridyl)-1’-(3-(8-aminoquinoline)ferrocene]
A 10-mL flask was charged with 20 (0.100 g, 0.14 mmol) dissolved in
CH CN (3 mL) and EtOH (3 mL). ·H (0.020 g. 0.020 mL,
.39 mmol) and graphite (0.500 g) were added to the red solution and the
A
H
R
U
G
6
)
(11):
by filtration and was washed with Et
was dissolved into CH CN and was vapor-diffused with Et
needles so obtained were collected by filtration and were washed with
2
O and pentane. The purple residue
3
N
2
H
4
2
O
3
2
O. The purple
0
1
reaction mixture was heated at reflux for 36 h. The reaction mixture was
filtered through celite and the solvent was removed under reduced pres-
sure. The deep red residue was slurried in hexanes and was filtered. The
cold hexanes and were vacuum-dried (0.160 g, 97%). H NMR (CD
3
CN,
278C, 500 MHz): d=4.46 (s, 3H), 4.62(m, 4H), 4.90 (m, 4H), 7.45 (m,
1
1
1H), 7.68 (m, 1H), 7.81 (d, J=8.2Hz, 1H), 8.01 (d, J=8.4 Hz, 1H),
1
1
solid red residue was dissolved into CH
2
Cl
2
and was vapor-diffused with
8.06 (m, 1H), 8.17 (d, J=8.1 Hz, 1H), 8.22 (m, 1H), 8.39 (d, J=5.9 Hz,
1H), 8.55 (s, 1H), 8.70 (d, J=2.2 Hz, 1H), 9.22 ppm (m, 1H); conductiv-
1
hexanes. The red blocks so obtained were collected by filtration and
were washed with cold hexanes and were vacuum-dried (0.080 g, 82%).
ꢀ1
2
ꢀ1
ity:
[MꢀSbF
36.94, H 2.62, N 4.97; found: C 36.61, H 2.71, N 4.65.
1-(N-Methyl-3-pyridyl)-1’-(5–2,2’-dipyridyl)ferrocenePd(Py)
and 16): A 25-mL flask was charged with 14 (0.100 g, 0.11 mmol) dis-
solved in acetone (8 mL). AgSbF (0.085 g, 0.24 mmol) was added as a
L
M
A
H
R
U
G
3
CN)=126 W cm mol
;
ESI-MS (CH
3
CN): m/z: 610
1
H NMR (CD
3
CN, 278C, 500 MHz): d=3.77 (s, 3H), 4.48 (t, J
1
=1.7 Hz,
6
H
22Cl
2
F
6
FeN PdSb: C
3
1
1
1
2
2
7
1
1
H), 4.59 (t, J=1.7 Hz, 2H), 4.82 (t, J=1.7 Hz, 2H), 4.96 (t, J=1.7 Hz,
H), 5.27 (sb, 2H), 6.88–6.90 (m, 2H), 6.95–6.97 (m, 1H), 7.33 (t, J=
.8 Hz, 1H), 7.44 (d, J=5.7 Hz, 1H), 7.62(d, J=2.1 Hz, 1H), 7.89 (sb,
1
[
2 6 3
](SbF ) (15
A
H
R
U
G
1
1
H), 8.40 ppm (d,
J
1
=2.3, 1H); conductivity:
22 W cm mol ; ESI-MS (CH
CN): m/z: 42 0 M[ ꢀSbF
FeN
L
M
+
A
H
R
U
G
3
CN)=
6
ꢀ
1
2
ꢀ1
3
6
] ; elemental
solid to the purple solution. The reaction mixture was stirred at room
temperature for 1 h, during which time AgCl precipitated. The AgCl was
removed by Schlenk filtration. Pyridine (0.22 g, 0.26 mmol) was added to
the solution and the reaction mixture was stirred at room temperature
for 1 h. The solvent volume was reduced to half of the initial amount and
analysis calcd (%) for C25
5.65, H 3.54, N 6.69.
1-(3-Pyridyl)-1’-(5-2,2’-dipyridyl)ferrocene] (12):
charged with 5-bromo-2,2’-bipyridine (1.25 g, 5.31 mmol) dissolved in a
mixture of dioxane (10 mL) and 1m aqueous Na CO (10 mL). The cata-
lyst [PdCl (1,1’-bis(diphenylphosphino)ferrocene)] (0.40 g, 0.54 mmol)
was added to the reaction mixture followed by a solution of 18 (0.80 g,
.60 mmol) in 3m aqueous NaOH (3 mL) in DME (10 mL). The reaction
mixture was refluxed for 24 h. At the end of this period the reaction mix-
H
22
F
6
3
Sb: C 45.77, H 3.38, N 6.40; found: C
4
[
A 50-mL flask was
2
3
the purple acetone solution was vapor-diffused with Et
purple and red/brown crystals was obtained during the crystallization
from acetone/Et O. These crystals were separated by hand under a mi-
2
O. A mixture of
2
ACHTREUNG
2
2
croscope. Yield: The dark purple crystals of 16 (0.126 g, 75%). Red-
brown crystals of 15 (0.020 g, 15%). Compound 16 alone can be generat-
Chem. Eur. J. 2006, 12, 8935 – 8951
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8949

B. Bosnich et al.
ꢀ
1
ed without the formation of 15, in an analogous procedure to above by
using acetontrile (8 mL) as the solvent.
repulsion energy was 2.77 kcalmol in acetonitrile solution and 4.96 kcal
mol in acetone solution.
ꢀ
1
1
1
4
6
H NMR (CD
3
CN, 278C, 500 MHz): d=4.27 (s, 3H), 4.37 (sb, 2H),
General procedure for crystallographic structural determinations of the
ferrocene rotors
1
.43 (sb, 2H), 4.51 (sb, 2H), 4.83 (sb, 2H), 7.00 (s, 1H), 7.35 (d, J=
.3 Hz, 1H), 7.57 (t, J=4.8 Hz, 1H), 7.75 (t, J=6.3 Hz, 2H), 7.81 (t,
J=6.2Hz, 3H), 8.14–8.18 (m, 3H), 8. 24 (d, J=8.1 Hz, 2H), 8.33–8.38
m, 3H), 8.55 (sb, 1H), 9.00 (d, J=5.2Hz, 2H), 9.03 ppm (d, J=5.2Hz,
1
1
5
Crystallization conditions: Crystals of 7·2
diffusion of diethyl ether into an acetonitrile solution of the ferrocene
rotor 7 in the presence of 10 equivalents of TFA. Crystals of 7·2(TfOH)
A
H
R
U
G
1
1
1
1
(
2
ACHTREUNG
2
+
H); ESI-MS (CH
3
CN): m/z: 467 [Mꢀ2SbF
6
]
,
1090 [M+
were obtained by vapor diffusion of diethyl ether into an acetonitrile sol-
ution of the ferrocene rotor 7 in the presence of two equivalents of
+
ꢀ1
2
ꢀ1
ClꢀPyꢀ2SbF
6
] ; conductivity: L
M
A
H
R
U
G
3
CN)=333 W cm mol ; elemen-
tal analysis calcd (%) for C36
32 5 3
H F18FeN PdSb : C 30.79, H 2.30, N 4.99;
TfOH. Crystals of 7·2
tion containing 10 equivalents of HSbF
A
H
R
U
G
6
) were obtained by layering an ethanol solu-
·6H O on top of a dichlorome-
found: C 30.38, H 2.37, N 4.80.
6
2
1
1
(
5 H NMR (CD
3
1
CN, 278C, 500 MHz): d=2.32 (s, 3H), 2.81 (s, 2H), 4.19
thane solution of 7. Crystals of 9 were obtained by vapor diffusion of
hexanes into an acetone solution of 9. Crystals of 10 were obtained by
vapor diffusion of hexanes into a dichloromethane solution of 10. Crys-
tals of 11 were obtained by vapor diffusion of hexanes into a dichlorome-
1
1
s, 3H), 4.37 (t, J=1.8 Hz, 2H), 4.40 (t, J=1.8 Hz, 2H), 4.53 (t, J=
.8 Hz, 2H), 4.88 (t, J=1.8 Hz, 2H), 7.35 (d, J=5.3 Hz, 1H), 7.59 (t,
J=4.8 Hz, 1H), 7.66 (t, J=6.3 Hz, 1H), 7.81 (t, J=6.2Hz, 2H), 8.14–
.18 (m, 3H), 8.30–8.36 (m, 3H), 8.37 (d, J=8.1 Hz, 2H), 8.54 (sb, 1H),
.08 ppm (d, J=5.5 Hz, 2H); ESI-MS (CH
conductivity: L
1
1
1
1
1
1
1
8
9
thane solution of 11. Crystals of 11·HSbF
ethanol solution containing 10 equivalents of HSbF
6
were obtained by layering an
·6H O on top of a di-
1
3
ꢀ1
CN): m/z: 910.6 [MꢀSbF
CN)=267 W cm mol ; elemental analysis calcd
OPdSb : C 29.55, H 2.33, N 4.05; found: C 29.66,
6
];
6
2
ꢀ
1
2
M
A
C
H
T
R
E
U
N
G
(CH
3
chloromethane solution of 11. Crystals of 12 were obtained by vapor dif-
fusion of hexanes into a dichloromethane solution of 12. Crystals of 13
were obtained by vapor diffusion of hexanes into a dichloromethane sol-
ution of 13. Crystals of 14 were obtained by vapour diffusion of diethyl
ether into an acetonitrile solution of the 14. Crystals of 16 were obtained
by vapor diffusion of diethyl ether into an acetone solution of the both
(
%) for C34
H 2.45, N 4.28.
Titration of 7 with TFA: A series of 0.494-mm solutions of 7 in dry
CD CN containing varying amounts of dry TFA, ranging from 0.325 mm
to 38.9 mm, were prepared and were examined by absorption spectrome-
try (258C). The ferrocene rotor 7 in CD CN solution is orange, and TFA
H
32
F
18FeN
4
3
3
1
5 and 16, the red-brown crystals of 15 were separated from the purple
3
crystals of 16 under a microscope. Crystals of 16 were obtained by vapor
diffusion of diethyl ether into an acetonitrile solution of 16. Crystals of
is colorless. The acid–base mixtures vary in color from orange to red. The
change in absorbance (jDAj) was plotted against the number of equiva-
lents of TFA. This plot indicated that 10 equivalents of TFA are required
to fully protonate 7 (see Supporting Information). The protonation of 7
was also studied using H NMR spectroscopy (278C). A series of 2.35 mm
3
solutions of 7 in dry CD CN containing varying amounts of dry TFA,
ranging from 1.59 mm to 156 mm, were prepared. The data is consistent
with the results obtained by absorption spectra (see Supporting Informa-
tion).
1
9 were obtained by vapor diffusion of hexanes into a dichloromethane
solution of 19 (see Supporting Information).
1
Data collection: A suitable crystal was selected under a stereo-micro-
scope while immersed in Fluorolube oil to avoid possible reaction with
air. The crystal was removed from the oil using a tapered glass fiber that
also served to hold the crystal for data collection. The crystal was mount-
ed and centered on a Bruker SMART APEX system at 100 K. Still
images showed the diffractions to be sharp. Frames separated in recipro-
cal space were obtained and provided an orientation matrix and initial
cell parameters. Final cell parameters were obtained from the full data
set.
Titration of 7 with HSbF
dry CD CN containing varying amounts of HSbF
.356 mm to 8.54 mm, were prepared and were examined by absorption
spectrometry (258C). The protonation of 7 with HSbF ·6H O was also
studied by H NMR spectroscopy (278C), and for this purpose a series of
.94-mm solutions of 7 in dry CD CN containing varying amounts of
HSbF ·6H O, ranging from 0.755 mm to 72.5 mm, were prepared. The
change in absorbance or chemical shift versus the equivalents of
HSbF ·6H O indicated that two equivalents of HSbF ·6H O are required
to protonate 7 (see Supporting Information).
Titration of 11 with TFA and with HSbF ·6H
TFA and HSbF ·6H O were carried out in an analogous manner to those
above. The results for the titration of 11 with TFA indicated that approxi-
mately 60 equivalents of TFA are required to fully protonate 11 (see Sup-
porting Information). For the titration of 11 with HSbF
changes in chemical shift were plotted against the equivalents of
HSbF ·6H O; these plots indicated that 10 equivalents of HSbF ·6H
are required to fully protonate 11 (see Supporting Information).
6
·6H
2
O: A series of 0.894-mm solutions of 7 in
3
6
·6H O, ranging from
2
0
6
2
1
A “full sphere” data set was obtained which samples approximately all of
reciprocal space to a resolution of 0.75 Š using 0.38 steps in w using 10 s
integration times for each frame. Data collection was made at 100 K. In-
tegration of intensities and refinement of cell parameters were done
2
3
6
2
[
32]
[32]
6
2
6
2
using SAINT. Absorption corrections were applied using SADABS
based on redundant diffractions.
6
2
O: The titrations of 11 with
Structure solution and refinement: The space group of the crystal was de-
termined based on systematic absences and intensity statistics. Direct
methods were used to locate the Fe and the other heavy atoms (such as
Pd and Sb), and most C atoms from the E-map. Repeated difference
Fourier maps allowed recognition of all expected atoms. The N assign-
ments were based on the abnormal small thermal parameter on the origi-
nally assigned C atoms. Following anisotropic refinement of all non-H
atoms, ideal H atom positions were calculated. Final refinement was ani-
6
2
6 2
·6H O the
6
2
6
2
O
sotropic for all non-H atoms isotropic-riding for H atoms. Except for
Electrostatic repulsion calculation: The Coulombic repulsion energy for
the ferrocene rotor 7 was calculated as the sum of all the cationic(+) in-
teractions using Equation (8), in which U=electrostatic repulsion energy,
ꢀ
slight disorder in some of the SbF
6
groups, no anomalous bond lengths
or thermal parameters were noted for the molecules. All ORTEP dia-
grams have been drawn with 50% probability ellipsoids.
ꢀ19
e (elementary charge)=1.602177”10 C, N
A
(Avogadro constant)=
2
3
ꢀ1
ꢀ12 ꢀ1
2
ꢀ1
6
.023”10 mol
pe
,
e
o
(vacuum permittivity)=8.85419”10
J
C m
,
CCDC-271205 (7·2
CCDC-268425 (11), CCDC-270740 (11·HSbF
A
H
R
U
G
6
)), CCDC-268421 (9), CCDC-268429 (10),
), CCDC-268415 (12),
ꢀ
10 ꢀ1
2
ꢀ1
4
o
(vacuum permittivity)=1.11265”10
CN, 298 K), 20.2 (acetone, 298 K), z (charge)=z
+. The charges are assumed to be point charges located at the centers
of the nitrogen atoms. The cation–cation distances were obtained from
J
C m , e (dielectric con-
6
stant)=36.2(CH
3
a
=z
b
=
CCDC-268428 (13), CCDC-268423 (14), CCDC-268427 (16), CCDC-
268424 (15) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
1
[
31]
molecular modeling.
2
2
N
A
z e
U ¼ 4
ð8Þ
pee r
0 ab
Calculated minimum cation–cation distances for the meso rotamer rab
=
3
.30 Š, and 4.10 Š for the racemic rotamer. The maximum electrostatic
8950
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8935 – 8951

Electrostatic Drivers for Synthetic Molecular Motors
FULL PAPER
Acknowledgement
Leigh, Top. Curr. Chem. 2005, 262, 133–177; g) G. S. Kottas, L. I.
Clarke, D. Horinek, J. Michl, Chem. Rev. 2005, 105, 1281–1376;
h) S. P. Fletcher, F. Dumur, M. M. Pollard, B. L. Feringa, Science
This work was supported by grants from the Basic Sciences Division of
the U.S. Department of Energy.
2
005, 310, 80–82.
[
9] A. Haaland, Acc. Chem. Res. 1979, 12, 415–422.
[
[
10] S. Carter, J. N. Murrell, J. Organomet. Chem. 1980, 192, 399–408.
11] a) P. Seiler, J. D. Dunitz, Acta Crystallogr. Sect. B 1979, 35, 2 02 0–
[
1] a) A. Houdusse, H. L. Sweeney, Curr. Opin. Struct. Biol. 2001, 11,
2
1
1
1
032; b) P. Seiler, J. D. Dunitz, Acta Crystallogr. Sect. B 1979, 35,
068–1074; c) J. D. Dunitz, L. E. Orgel, A. Rich, Acta Crystallogr.
956, 9, 373–375; d) J. D. Dunitz, L. E. Orgel, Nature 1953, 171,
21–122.
182–194; b) A. F. Huxley, Philos. Trans. R. Soc. London Ser. B 2000,
355, 433–440; c) F. J. Kull, S. A. Endow, J. Cell Sci. 2002, 115, 15–
23; d) D. Stock, A. G. Leslie, J. E. Walker, Science 1999, 286, 1700–
1705; e) A. Yildiz, M. Tomishige, R. D. Vale, P. R. Selvin, Science
2004, 303, 676–678; f) C. L. Asbury, A. N. Fehr, S. M. Block, Science
2003, 302, 2130–2134; g) W. Hua, J. Chung, J. Gelles, Science 2002,
295, 844–848; h) R. A. Cross, Curr. Biol. 2004, 14, R158–R159.
[
[
12] T. N. Doman, C. R. Landis, B. Bosnich, J. Am. Chem. Soc. 1992, 114,
264–7272.
7
13] ConQuest, 1.4; software for searching the Cambridge Crystallo-
graphic Data Centreꢁs database of crystal structures, http://
www.ccdc.cam.ac.uk; The Cambridge Crystallographic Data Centre:
UK, 2004.
14] a) F. Gelin, R. P. Thummel, J. Org. Chem. 1992, 57, 3780–3783;
b) A. Kasahara, T. Izumi, Y. Yoshida, I. Shimizu, Bull. Chem. Soc.
Jpn. 1982, 55, 1901–1906; c) I. Shimizu, H. Umezawa, T. Kanno, T.
Izumi, A. Kasahara, Bull. Chem. Soc. Jpn. 1983, 56, 2023–2028;
d) I. Shimizu, Y. Kamei, T. Tezuka, T. Izumi, A. Kasahara, Bull.
Chem. Soc. Jpn. 1983, 56, 192–198.
[
2] a) M. Yoshida, E. Muneyuki, T. Hisabori, Nat. Rev. Mol. Cell Biol.
2001, 2, 669–677; b) G. Oster, H. Wang, Biochim. Biophys. Acta
2000, 1458, 482–510; c) P. D. Boyer, Annu. Rev. Biochem. 1997, 66,
717–749; d) P. Dimroth, Biochim. Biophys. Acta 2000, 1458, 374–
386; e) T. C. Elston, G. Oster, Biophys. J. 1997, 73, 703–721; f) H. C.
[
Berg, Philos. Trans. R. Soc. London Ser. B 2000, 355, 491–501;
g) D. J. Derosier, Cell 1998, 93, 17–20.
[
[
3] a) Molecular Motors (Ed.: M. Schliwa), Wiley-VCH, Weinheim,
2
003; b) M. Schliwa, G. Woehlke, Nature 2003, 422, 759–765; c) K.
[
15] M. Inouye, Y. Hyodo, H. Nakazumi, J. Org. Chem. 1999, 64, 2704–
Kinbara, T. Aida, Chem. Rev. 2005, 105, 1377–1400.
2
710.
4] a) G. Oster, H. Wang, Molecular Motors (Eds.: M. Schliwa), Wiley-
VCH, Weinheim, 2003, chap. 8, p. 207; b) G. Oster, H. Wang, Trends
Cell Biol. 2001, 11, 196–202; c) G. Oster, H. Wang, Appl. Phys. A
[
16] a) C. Janiak, J. Chem. Soc. Dalton Trans. 2000, 3885–3896; b) C. A.
Hunter, K. R. Lawson, J. Perkins, C. J Urch, J. Chem. Soc. Perkin
Trans. 2 2001, 651–669.
17] a) S. Barlow, A. Cowley, J. C. Green, T. J. Brunker, T. Hascall, Orga-
nometallics 2001, 20, 5351–5359; b) A. A. Koridze, N. M. Astakho-
va, P. V. Petrovskii, J. Organomet. Chem. 1983, 254, 345–360; c) A.
Ceccon, G. Giacometti, A. Venzo, D. Paolucci, D. Benozzi, J. Orga-
nomet. Chem. 1980, 185, 231–239; d) J. J. Dannenberg, M. K. Leven-
berg, J. H. Richards, Tetrahedron 1973, 29, 1575–1584.
2
002, 75, 315–323; d) P. Reimann, Phys. Rep. 2002, 361, 57–265.
5] a) R. A. Bissel, E. Córdova, A. E. Kaifer, J. F. Stoddart, Nature
994, 369, 133–137; b) P. R. Ashton, R. Ballardini, V. Balzani, I.
[
[
[
1
Baxter, A. Credi, M. C. T. Fyfe, M. T. Gandolfi, M. Gomez-Lopez,
M.-V. Martinez-Diaz, A. Piersanti, N. Spencer, J. F. Stoddart, M.
Venturi, A. J. P. White, D. J. Williams, J. Am. Chem. Soc. 1998, 120,
1
1932–11942; c) J. D. Badjic, V. Balzani, A. Credi, S. Silvi, J. F.
18] a) M. Rosenblum, J. O. Santer, J. Am. Chem. Soc. 1959, 81, 5517–
Stoddart, Science, 2004, 303,1845–1849; d) F. Ibukuro, T. Kusukawa,
M. Fujita, J. Am. Chem. Soc. 1998, 120, 8561–8562; e) J. D. Crowley,
A. J. Goshe, I. M. Steele, B. Bosnich, Chem. Eur. J. 2004, 10, 1944–
5
518; b) U. T. Mueller-Westerhoff, T. J. Haas, G. F. Swiegers, T. K.
Leipert, J. Organomet. Chem. 1994, 472, 229–246; c) G. Cerichelli,
G. Illuminati, G. Ortaggi, A. M. Giuliani, J. Organomet. Chem. 1977,
1
955.
6] a) W-Y. Sun, T. Kusukawa, M. Fujita, J. Am. Chem. Soc. 2002, 124,
1570–11571; b) M. Asakawa, P. R. Ashton, V. Balzani, A. Credi,
1
27, 357–370.
19] F. Tran, J. Weber, T. A. Wesolowski, Helv. Chim. Acta 2001, 84,
489–1503.
20] J. R. Grover, E. A. Walters, E. T. Hui, J. Phys. Chem. 1987, 91,
233–3237.
[
[
[
1
1
C. Hamers, G. Mattersteig, M. Montalti, A. N. Shipway, N. Spencer,
J. F. Stoddart, M. S. Tolley, M. Venturi, A. J. P. White, D. J. Williams,
Angew. Chem. 1998, 110, 357–361; Angew. Chem. Int. Ed. 1998, 37,
3
[
[
21] R. Knapp, M. Rehahn, J. Organomet. Chem. 1993, 452, 235–240.
22] a) D. Braga, M. Polito, M. Bracaccini, D. DꢁAddario, E. Tagliavini,
L. Sturba, F. Grepioni, Organometallics 2003, 22, 2142–2150; b) D.
Braga, M. Polito, M. Bracaccini, D. DꢁAddario, E. Tagliavini, D. M.
Proserpio, F. Grepioni, Chem. Commun. 2002, 1080–1081.
3
33–337; c) V. Balzani, J. Becher, A. Credi, M. B. Nielsen, F. M.
Raymo, J. F. Stoddart, A. M. Talarico, M. Venturi, J. Org. Chem.
000, 65, 1947–1956.
7] a) D. A. Leigh, J. K. Y. Wong, F. Dehez, F. Zerbetto, Nature 2003,
24, 174–179; b) A. M. Brouwer, C. Frochot, F. G. Gatti, D. A.
Leigh, L. Mottier, F. Paolucci, S. Roffia, G. W. Wurpel, Science 2001,
91, 2124–2128; c) N. Koumura, R. W. J. Zijistra, R. A. Van Delden,
2
[
4
[
23] S. Doherty, E. G. Robins, I. Pal, C. R. Newman, C. Hardacre, D.
Rooney, D. A. Mooney, Tetrahedron: Asymmetry 2003, 14, 1517–
2
1
527.
24] B. H. Han, D. H. Shin, S. Y. Cho, Tetrahedron Lett. 1985, 26, 6233–
234.
N. Harada, B. L. Feringa, Nature 1999, 401, 152–155; d) T. Muraoka,
K. Kinbara, Y. Kobayashi, T. Aida, J. Am. Chem. Soc. 2003, 125,
[
[
6
5
612–5613; e) F. M. Hawthorne, J. I. Zink, J. M. Skelton, M. J.
25] K. Izutsu, Acid-Base Dissociation Constants in Dipolar Aprotic Sol-
vents, Blackwell, London, 1990.
26] W. J. Geary, Coord. Chem. Rev. 1971, 7, 81–122.
27] X.-B. Wang, D. Bing, H.-K. Woo, L.-S. Wang, Angew. Chem. 2005,
Bayer, C. Liu, E. Livshits, R. Baer, D. Neuhauser, Science 2004, 303,
1
3
849–1851; f) J. V. Hernandez, E. R. Kay, D. A. Leigh, Science 2004,
06, 1532–1537; g) Y. Liu, A. H. Flood, P. A. Bonvallet, S. A.
[
[
Vignon, H.-R. Tseng, T. J. Huang, B. Brough, M. Baller, S. Magonov,
S. Solares, W. A. Goddard, C.-M. Ho, J. F. Stoddart, J. Am. Chem.
Soc. 2005, 127, 9745–9759.
1
17, 6176–6178; Angew. Chem. Int. Ed. 2005, 44, 6022–6024.
[
[
[
[
[
28] T. Muraoka, K. Kinbara, T. Aida, Nature 2006, 440, 512–515.
29] Y-Q. Fang, G. S. Hanan, Synlett 2003, 852–854.
30] F. S. Kamounah, J. B. Christensen, J. Chem. Res. Synop. 1997, 150.
31] PC Spartan Pro, Version 1.0.5, 1999, Wavefunction, Irvine, CA.
32] G. Sheldrick, SHELXTL (version 6.1) program library; G. Bruker
Analytical X-ray Systems, Madison, WI, 2000.
[
8] a) V. Balzani, M. Venturi, A. Credi, Molecular Devices and Ma-
chines, Wiley-VCH, Weinheim, 2003; b) V. Balzani, A. Credi, F. M.
Raymo, J. F. Stoddart, Angew. Chem. 2000, 112, 3486–3531; Angew.
Chem. Int. Ed. 2000, 39, 3349–3391; c) M. C. Jimenez-Molero, C.
Dietrich-Buchecker, J.-P. Sauvage, Chem. Commun. 2003, 14, 1613–
1
616; d) Molecular machines special issue (Ed.: J. F. Stoddart), Acc.
Chem. Res. 2001, 34, 409–522; e) E. R. Kay, D. A. Leigh in Func-
tional Artificial Receptors (Eds.: T. Schrader, A. D. Hamilton),
Wiley-VCH, Weinheim, 2005, pp. 333–406; f) E. R. Kay, D. A.
Received: May 10, 2005
Revised: March 28, 2006
Published online: July 5, 2006
Chem. Eur. J. 2006, 12, 8935 – 8951
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8951