Gated electron sharing within dynamic naphthalene diimide-based oligorotaxanes

Article abstract of DOI:10.1002/anie.201309680

The controlled self-assembly of well-defined and spatially ordered π-systems has attracted considerable interest because of their potential applications in organic electronics. An important contemporary pursuit relates to the investigation of charge transport across noncovalently coupled components in a stepwise fashion. Dynamic oligorotaxanes, prepared by template-directed methods, provide a scaffold for directing the construction of monodisperse one-dimensional assemblies in which the functional units communicate electronically through-space by way of π-orbital interactions. Reported herein is a series of oligorotaxanes containing one, two, three and four naphthalene diimide (NDI) redox-active units, which have been shown by cyclic voltammetry, and by EPR and ENDOR spectroscopies, to share electrons across the NDI stacks. Thermally driven motions between the neighboring NDI units in the oligorotaxanes influence the passage of electrons through the NDI stacks in a manner reminiscent of the conformationally gated charge transfer observed in DNA. Birds of a feather flock together: Mutually π-stacked small-molecule naphthalene diimide (NDI) [24]crown-8 derivatives self-assemble dynamically on a repeating -(CH₂CH₂NH₂⁺CH ₂)-template to afford the stepwise evolution of discrete-length oligorotaxanes, which are shown to communicate electronically through their co-facially interacting π-surfaces, sharing electrons across linear arrays containing one, two, three, and four NDI units as determined by cyclic voltammetry and EPR/ENDOR spectroscopies.

Full text of DOI:10.1002/anie.201309680

.
Angewandte
Communications
DOI: 10.1002/anie.201309680
Self-Assembly
Gated Electron Sharing Within Dynamic Naphthalene Diimide-Based
Oligorotaxanes**
Alyssa-Jennifer Avestro, Daniel M. Gardner, Nicolaas A. Vermeulen, Eleanor A. Wilson,
Severin T. Schneebeli, Adam C. Whalley, Matthew E. Belowich, Raanan Carmieli,
Michael R. Wasielewski,* and J. Fraser Stoddart*
Abstract: The controlled self-assembly of well-defined and
spatially ordered p-systems has attracted considerable interest
because of their potential applications in organic electronics.
An important contemporary pursuit relates to the investigation
of charge transport across noncovalently coupled components
in a stepwise fashion. Dynamic oligorotaxanes, prepared by
template-directed methods, provide a scaffold for directing the
construction of monodisperse one-dimensional assemblies in
which the functional units communicate electronically
through-space by way of p-orbital interactions. Reported
herein is a series of oligorotaxanes containing one, two, three
and four naphthalene diimide (NDI) redox-active units, which
have been shown by cyclic voltammetry, and by EPR and
ENDOR spectroscopies, to share electrons across the NDI
stacks. Thermally driven motions between the neighboring
NDI units in the oligorotaxanes influence the passage of
electrons through the NDI stacks in a manner reminiscent of
the conformationally gated charge transfer observed in DNA.
Electron or hole transport within covalent D-B-A molecules
is most often through-bond,[6] whereas transport in self-
assembled noncovalent D-B-A systems is generally through-
space.[7]
Many approaches[8–12] to the fabrication of through-space
“molecular wire” systems with discrete architectures often
require elegant, multi-step syntheses. Alternative strategies
make use of less complex covalent building blocks that self-
assemble into oligomers having relatively low polydisper-
sities[9] as well as template-directed methods[10] to control the
self-assembly of planar aromatics into 1D architectures.[11] For
example, Wu et al.[12] have used G-quadruplexes to construct
discrete assemblies of electron and hole charge carriers, while
Matile et al.[13] have established a general bottom-up method
for effecting the template-directed, self-organizing surface
initiated polymerization (SOSIP) of functional monomers
into vertical multicomponent photosystems. By relying on
a combination of hydrogen bonding, p–p stacking and
reversible ring-opening polymerization, oriented long-range
photoactive layers can be produced on indium tin oxide (ITO)
surfaces with low polydispersities. Meanwhile, Fujita et al.[14a]
have combined the host-guest chemistry of columnar Pd-
coordination cages with the direct binding nanogap technique
developed by Kiguchi and co-workers[15] in order to obtain
conductance measurements on molecular junctions composed
of discretely assembled p-stacks containing four, five, and six
aromatic units.[14b] In this case, the number of aromatic guests
in the stack is pre-determined by the length and size of the
host, the initial preparation of which is limited by the
availability of appropriately long and/or rigid ligands.
Recently, we capitalized upon a highly efficient template-
E
lectron transport has been shown to play a crucial role in
the functioning of biological systems,[1] light-harvesting cas-
cades,[2] molecular actuators and switches,[3] and polymeric
materials.[4] Understanding how electrons move from one
molecular component to the next over long discrete distances
remains an important goal in the development of nano-
materials for organic molecular electronics. Covalent donor-
bridge-acceptor (D-B-A) systems have been used exten-
sively[5] to 1) demonstrate charge transport in one-dimen-
sional (1D) systems having well-defined D–A distances and
orientations, and 2) expand our knowledge of how molecular
structure and dynamics impact charge-transfer processes.
[*] A.-J. Avestro, D. M. Gardner, Dr. N. A. Vermeulen, E. A. Wilson,
Dr. S. T. Schneebeli, Dr. A. C. Whalley, Dr. M. E. Belowich,
Prof. R. Carmieli, Prof. M. R. Wasielewski, Prof. J. F. Stoddart
Center for the Chemistry of Integrated Systems (CCIS) and
Argonne-Northwestern Solar Energy Research (ANSER) Center
Department of Chemistry, Northwestern University
2145 Sheridan Road, Evanston, IL 60208 (USA)
E-mail: m-wasielewski@northwestern.edu
would like to thank both KACSTand NU for their continued support
of this research. M.R.W. is supported by the Chemical Sciences,
Geosciences, and Biosciences Division, Office of Basic Energy,
DOE, under grant no. DE-FG02-99ER-14999. A.-J.A. gratefully
acknowledges the National Science Foundation (NSF) for the award
of a Graduate Research Fellowship (GRF) under Grant No. DGE-
0824162. D.M.G. was supported by the Department of Defense
through the National Defense Science & Engineering Graduate
Fellowship (NDSEG) Program. S.T.S. thanks the International
Institute for Nanotechnology (IIN) at NU for an IIN Postdoctoral
Fellowship and the QUEST high-performance computing center at
NU for a research allocation of computer time. R.C. is supported by
the ANSER Center, an Energy Frontier Research Center funded by
the U.S. Department of Energy, Office of Science, Office of Basic
Energy Sciences under Award Number DE-SC0001059.
stoddart@northwestern.edu
Homepage: http://chemgroups.northwestern.edu/wasielewski
http://stoddart.northwestern.edu
[**] We would like to thank Dr. Saman Shafaie for collecting high-
resolution mass spectrometric data, Drs. Charlotte Stern and
Amy A. Sargeant for single-crystal structure determination, and Drs.
James M. Holcroft and Marco Frasconi for helpful discussions. This
research is part (Project 34-947) of the Joint Center of Excellence in
Integrated Nano-Systems (JCIN) at King Abdul-Aziz City for Science
and Technology (KACST) and Northwestern University (NU). We
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201309680.
4442
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 4442 –4449

Angewandte
Chemie
directed clipping strategy[16] utilizing the concept of dynamic
covalent chemistry[17] (DCC) to prepare a series of rigid
oligorotaxanes[18] featuring a precise number of [24]crown-8
rings which encircle an oligoammonium dumbbell containing
(NDI)—which encircle oligoammonium dumbbells Dx where
x also represents the number of –(CH2CH2NH2 CH2)–
+
recognition sites for the rings. NDI is an attractive functional
unit because of 1) its large p-surface, 2) its propensity to self-
assemble into co-facial stacks, 3) its excellent photochemical/
thermal stability, 4) its predictable n-type redox behavior and
5) its electronic tenability via naphthalene core substitution
methods, making it a suitable candidate for fundamental
investigations of charge and energy transfer processes.[19]
Soluble NDI derivatives are also easily prepared and
modified by attaching substituents to the diimide nitrogens
without affecting their electronic properties.[20] The oligo-
rotaxanes [n]NDIxR were prepared by mixing x equivalents
each of the acyclic precursor 1-NDI and tetraethylene glycol
bis(2-aminophenyl)ether (2) in the presence of 1.0 equiv of
the dumbbell Dx using the clipping procedure.[16,18] For more
details, see the Experimental Section and Scheme S6 in the
Supporting Information. A control NDI derivative 3 was also
prepared for comparison. All of the compounds were fully
+
x –(CH2CH2NH2 CH2)– recognition sites spaced 3.4 ꢀ apart,
an ideal distance for sustaining p–p stacking interactions.
These oligorotaxanes can be isolated in near quantitative
yields (up to 96%) in < 10 min at room temperature.
Employing this same approach, we envisioned the self-
assembly of redox-active units into ordered 1D arrays for
the systematic investigation of through-space electron trans-
port over well-defined distances. Here we report 1) the
application of DCC in the preparation of n-type single-
molecular arrays, featuring electron-deficient naphthalene
diimides and 2) the utility of mechanically interlocked
architectures in the broader development of organic elec-
tronic materials, while at the same time assessing 3) the
influence of the mechanical bond, that is, the dynamic nature
of the rotaxane components, on electron transfer.
1
Herein, we describe the synthesis and intramolecular
electron sharing properties of a new series (Scheme 1) of
oligorotaxanes [n]NDIxR featuring one, two, three, and four
rings (i.e., x = 1–4)—functionalized with naphthalene diimide
characterized by H and 13C NMR spectroscopies, as well as
by high-resolution electron-spray ionization (HR-ESI) mass
spectrometry (Supporting Information, Sections 2 and 3). The
1
crude H NMR spectra (Figure 1) of the [2]-, [3]-, [4]-, and
Scheme 1. Synthesis of the [n]NDIxR series starting from 1.0 equiv of an oligoammonium dumbbell Dx (where x=1, 2, 3, or 4) and x equiv of
acyclic clipping precursors 1-NDI and 2 to afford the thermodynamically controlled self-assembly of [2]-, [3]-, [4]-, and [5]rotaxanes containing one,
two, three, and four NDI units, respectively. The constitutionally unsymmetrical NDI derivative 3 was prepared to in order to serve as a control
compound.
Angew. Chem. Int. Ed. 2014, 53, 4442 –4449
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4443

.
Angewandte
Communications
[5]rotaxanes reveal that, in all cases, the oligorotaxanes are
formed quantitatively in < 10 min, demonstrating the power
of the clipping methodology for the production of assemblies
with precise control in an efficient manner. The resonances
for the NDI protons, Ha and Hb, in [2]NDI1R appear
(Figure 1a) as
a “tight” AB system centered around
8.80 ppm despite the asymmetry of the NDI unit, whereas
for [3]NDI2R, the signals for the same protons separate
(Figure 1b) into
a well-defined AX system (d = 8.40,
8.20 ppm) indicative of p–p stacking interactions between
the NDI units. This separation is exaggerated even more so in
the spectra (Figure 1c,d) of [4]NDI3R and [5]NDI4R. In
agreement with the literature[18] the resonances for the imine
and aromatic protons shift towards higher and higher fields as
the number of rings increase.[21]
Since growing single crystals of [n]NDIxR suitable for X-
ray diffraction analysis has so far evaded us,[22] we focused our
Figure 2. a) Structural formula and single-crystal X-ray structure of the
related [3]rotaxane [3]Ph2R. Hydrogen atoms have been omitted for the
sake of clarity. The solid-state structure of [3]Ph2R shows a co-facial
arrangement of the phenyl groups yet negligible p–p interactions as
a consequence of the slipped-stack relationship between the rings
which creates a 5.4 ꢀ centroid-to-centroid distance between them.
b) Structural formula and graphical representation of the DFT-calcu-
lated (B3LYP-D3/6-31G**) geometry-optimized structure of [3]NDI2R,
which was constructed from the solid-state structure obtained for
[3]Ph2R. Inset: A plan view of the geometry-optimized structure of
[3]NDI2R superimposed upon the corresponding space-filling represen-
tation depicting 1) a 598 twist between the NDI units and phenylene
rings and 2) the two co-facial NDI units at a 3.4 ꢀ plane-to-plane
distance apart.
attention on quantum mechanical modeling, starting from the
crystal structure (Figure 2a; see also the Supporting Infor-
mation, Section 4) of the related [3]rotaxane[23] [3]Ph2R
which is devoid of NDI units para-substituted on the Ph
rings. Density functional calculations[24,25] of [3]NDI2R
afforded (Figure 2b) the two parallel-aligned NDI units
with an average plane-to-plane separation of 3.4 ꢀ and an
NDI-C6H4 dihedral angle of 598. Populating the molecule with
an additional single electron to form the mono-reduced
radical anion ([3]NDI2R)Cꢀ generates a SOMO (Figure 3)
spanning both NDI units which suggests that the unpaired
electron is shared within the [3]rotaxane. On the other hand,
variable-temperature (VT) 1H NMR spectra (Supporting
Information, Section 3.2) for the [3]-, [4]-, and [5]rotaxanes
show slight broadening of the resonances for the NDI protons
(Ha and Hb) as the temperature is decreased, whereas the
signals for these same protons in the [2]rotaxane [2]NDI1R
remain sharp. This observation implies interconversion
between various co-conformations of the NDI-appended
rings on the 1H NMR timescale at 298 K, as well as an
increase in the population of the co-facial co-conformations at
lower temperatures. Although in solution the rotaxane
components are free to rotate and can still adopt many
different co-conformations, it is not unreasonable to expect
Figure 1. Partial 1H NMR spectra (500 MHz/CD3CN/298 K) of the
[n]NDIxR series. The resonances for the NDI protons Ha and Hb
separate and shift significantly upfield as the number of rings is
increased from one to two in a) [2]NDI1R and b) [3]NDI2R, respectively,
an indication of the existence of p–p interactions between the NDI
units. Two sets of resonances occurring in c) 2:1 and d) 1:1 ratios are
consistent with the constitutionally heterotopic nature of the outer
(blue signals) and inner (red signals) rings of the [4]- and [5]rotaxanes
[4]NDI3R and [5]NDI4R, respectively.
4444 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 4442 –4449

Angewandte
Chemie
Electron paramagnetic resonance (EPR) spectroscopy at
X-band (9.5 GHz) was used to explore (Figure 5) electron
sharing within the singly reduced ([n]NDIxR)Cꢀ series at
270 K. Specifically, if an electron is being shared across
identical NDI monomers on the timescale of the NDICꢀ
monomer electron-nuclear hyperfine couplings (ca. 107 sꢀ1),
then a narrowing of the spectral width proportional to the
square root of the number of monomers should be
observed.[28] Although this spectral narrowing was not
observed for ([3]NDI2R)Cꢀ when compared with
([2]NDI1R)Cꢀ and the control 3Cꢀ, partial narrowing is
apparent in the spectra of ([4]NDI3R)Cꢀ and ([5]NDI4R)Cꢀ.
The EPR spectra (Figure 5) of ([n]NDIxR)Cꢀ at 270 K in
Figure 3. Graphical representation of the DFT-calculated (B3LYP-D3/6-
31G** level of theory) singly occupied molecular orbital (SOMO) of
([3]NDI2R)Cꢀ. Orbital isosurfaces are illustrated at 0.015 electrons
Bohrꢀ1
.
the proximate NDI units to favor co-facial p-interactions
based on the available data.[23,24,26]
In order to examine the interactions between the NDI
units in the [n]NDIxR series, we directed our attention
towards investigating their redox properties by cyclic voltam-
metry (CV) in solution. We also performed CV experiments
(Supporting Information, Figure S28) on 1-NDI and the
control compound 3 to probe the effects of the mechanical
bond on the reduction potential of the NDI units. The
[2]rotaxane (Figure S29) [2]NDI1R exhibits two one-electron
reductions at ꢀ0.73 and ꢀ1.20 V corresponding[19,20] to the
generation of NDICꢀ and NDI2ꢀ, respectively. On scanning at
even more negative potentials, a third one-electron process is
observed at ꢀ1.44 V. This redox process is also evident in the
CV of 1-NDI, but not in the control 3, leading us to associate
this feature with the reduction of the pyridine ring.[27]
Figure 4, which illustrates the first-wave reductions of the
[n]NDIxR series of oligorotaxanes, reveals a separation of the
first one-electron process observed (Figure 4a) in the case of
[2]NDI1R into x discrete one-electron steps as a result of the
electronic communication between the NDI units in the
arrays as the number of NDI units is increased. The first
reduction (ꢀ0.67 V) of [3]NDI2R is easier to perform (Fig-
ure 4b) than that (ꢀ0.73 V) of [2]NDI1R owing to the
favorable interactions between the two electron-deficient
NDI units. On the other hand, the fact that the second NDI
reduction (ꢀ0.88 V) of [3]NDI2R to is more difficult to
achieve on account of the destabilizing interactions present in
the electron-rich system. Similar trends are observed (Fig-
ure 4c) for [4]NDI3R where three reductions (ꢀ0.59, ꢀ0.73,
and ꢀ0.85 V) are observed for the generation one, two, and
three NDICꢀ in the [4]rotaxane, respectively. Indeed, the NDI
geometries and rotaxane co-conformations adopted by
[5]NDI4R become increasingly more complex and less well-
defined (Figure 4d) as it is reduced at ꢀ0.57 V down to
ꢀ0.97 V from the mono-reduced species to the tetra-anion.
Nevertheless, in each of these examples, the emergence of
x discrete redox processes suggests that, even when up to four
electrons are introduced into an array like [5]NDI4R,
communication between the NDI units is preserved despite
the potential for unfavorable charge repulsion to occur.
Figure 4. Cyclic voltammograms (CVs) of the [n]NDIxR series (scan
rate 50 mVsꢀ1) depicting successive first-electron reductions of one,
two, three, and four NDI units in a) [2]-, b) [3]-, c) [4]-, and d) [5]rotax-
anes, respectively. All the experiments were performed at 298 K in Ar-
purged CH2Cl2 solutions (1.0 mm) with 0.1m TBAPF6 as the support-
ing electrolyte. The splitting of first-wave reductions into x one-electron
steps is a clear indication of electronic coupling between NDI units in
the arrays.
Angew. Chem. Int. Ed. 2014, 53, 4442 –4449
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4445

.
Angewandte
Communications
this timescale, perfect sharing is likely to be hindered by the
distribution of different co-conformations.
Continuous-wave electron-nuclear double resonance
(ENDOR) spectroscopy (Figure 6) allows for the detection
of unresolved hyperfine couplings in the EPR spectra of
radicals,[29] thus providing higher resolution data on electron
sharing. On account of the flexible nature of the NDIs in the
oligorotaxanes, we hypothesized that a larger extent of
electron sharing might be observable with ENDOR spectros-
copy. Isotropic hyperfine coupling constants (hfccs) for the
singly reduced ([n]NDIxR)Cꢀ series were measured in CH2Cl2
at 270 K (Table 1) and follow the resonance condition
vꢁENDOR ¼ jvn ꢁ ðaH=2Þj, where vEꢁNDOR is the ENDOR tran-
sition frequency and nn is the proton Larmor frequency. The
ENDOR spectrum (Figure 6b) of ([2]NDI1R)Cꢀ shows the
expected ENDOR signal consistent with previously assigned
spectra of NDICꢀ derivatives with an unsubstituted core.[19,30]
The ENDOR spectrum (Figure 6c) of ([3]NDI2R)Cꢀ shows
a near-perfect one-half reduction of the hfccs established by
([2]NDI1R)Cꢀ compared to those calculated for the asym-
metrically substituted control 3Cꢀ. This observation provides
conclusive evidence that the unpaired electron is being fully
shared across two NDI units on a timescale ꢂ 107 sꢀ1.
The ENDOR spectra (Figure 6d,e) of ([4]NDI3R)Cꢀ and
([5]NDI4R)Cꢀ appear to exhibit fractional sharing in which the
broader linewidths provide evidence of several populations
that we suspect arise from dynamic motions taking place
between the NDI units and the partial disruption of p-orbital
overlap. The spectra, however, are narrowed still further as
a result of the increased probability that any two NDI units
Figure 5. EPR Spectra of a) 3Cꢀ and b–e) the ([n]NDIxR)Cꢀ series after
reduction with 1.0 equiv of CoCp2. All spectra were recorded in Ar-
purged CH2Cl2 solutions (0.75 mm) at 270 K at X-band (9.5 GHz) with
0.020 mT modulation amplitude. The hyperfine splitting of
b) ([2]NDI1R)Cꢀ and c) ([3]NDI2R)Cꢀ remain partially resolved, whereas
the spectra for d) ([4]NDI3R)Cꢀ and e) ([5]NDI4R)Cꢀ reveal a featureless
signal. Inhomogenous broadening of the EPR spectra relative to (a)
monomeric NDI 3 is attributed to the slower molecular tumbling rate
of the larger oligorotaxanes.
Table 1: Hyperfine coupling constants (hfccs) obtained from electron-
nuclear double resonance (ENDOR) spectra.
Compound
Hyperfine coupling constants[a] (hfccs) [MHz]
3Cꢀ (theor.)[b]
3Cꢀ (expt.)
5.27
5.34
5.85
3.05
3.82
2.39
4.84
–
4.76
2.49
3.11
–
0.75
0.43
0.56
0.28
0.90
1.02
([2]NDI1R)Cꢀ[c,d]
([3]NDI2R)Cꢀ[c,e]
([4]NDI3R)Cꢀ[f]
([5]NDI4R)Cꢀ[f]
CH2Cl2 are inhomogenously broadened relative (Figure 5a)
to the control 3Cꢀ because of the slower molecular tumbling
rates for the larger compounds, as well as a loss in resolution
of the hyperfine splitting resulting from electron sharing,
except for the spectra (Figure 5b,c) of ([2]NDI1R)Cꢀ and
([3]NDI2R)Cꢀ in which the hyperfine splittings are still
[a] hfcc values reflect coupling of the unpaired electron to either the
asymmetric protons on the NDI core or the sp3 hybridized methylene
protons nearest the imide nitrogen atom. Calculations show that the
naphthalene protons closest to the phenyl substituent are more strongly
coupled, that is, exhibit higher hfccs, than those protons near the hexyl
chain. [b] Theoretical hfccs for the control 3Cꢀ were calculated at the
UB3LYP/EPR-II level of theory. [c] The differences in hfccs are resolved for
molecules ([2]NDI1R)Cꢀ and ([3]NDI2R)Cꢀ. The weaker hyperfine coupling
constant observed below 1.0 MHz is a result of the interactions between
the unpaired electron and the protons of the methylene nearest the imide
nitrogen atom. [d] The experimental hfccs determined for ([2]NDI1R)Cꢀ
are in agreement with the theoretical and experimental hfccs recorded for
the control 3Cꢀ. [e] Reduction of all hfccs for ([3]NDI2R)Cꢀ relative to
([2]NDI1R)Cꢀ by a factor of two demonstrates sharing of the electron
between two identical NDI units in the [3]rotaxane. [f] Although
substantial reduction of the hfccs is also apparent for ([4]NDI3R)Cꢀ and
([5]NDI4R)Cꢀ, the extent of electron sharing (or efficiency of p-orbital
overlap) between the NDI units is ultimately impacted by the various
NDI geometries and rotaxane co-conformations that can be adopted,
which makes it difficult to assign the 0.90 and 1.02 MHz couplings.
partially
resolved.
The
spectra
(Figure 5d,e)
of
([4]NDI3R)Cꢀ and ([5]NDI4R)Cꢀ exhibit a single unresolved
line. The lack of significant spectral narrowing in the case of
([3]NDI2R)Cꢀ led us to hypothesize that there are a number of
internal nuclear reorganization events—such as intramolec-
ular rotation of the ring around the dumbbell inducing NDI
motion—which create a high barrier to electron sharing on
the timescale of the EPR experiment. We attempted to limit
intramolecular motions by lowering the temperature of
([3]NDI2R)Cꢀ to 200 K; however, no change in EPR linewidth
was observed (Figure S30). For the remaining oligorotaxanes,
linewidth analysis using Equation (1) in Ref. [28] gives values
for NS of 2.21 and 1.83 for ([4]NDI3R)Cꢀ and ([5]NDI4R)Cꢀ,
respectively, providing conclusive evidence that, although the
electron can be shared across these higher order rotaxanes on
4446 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 4442 –4449

Angewandte
Chemie
nature of NDI units on the ring, leading to imperfect stacking.
Furthermore, the flexible nature of the rotaxane components
resulting from the mechanical bond enables other modes of
disorder.
In summary, we have demonstrated through-space elec-
tron sharing in a series of mechanically interlocked molecules
featuring single-molecular 1D arrays of NDIs, capable of
forming co-facial stacks. In contrast with previous approaches
to preparing through-space molecular wires where non-trivial
syntheses and uncontrolled assembly of molecular compo-
nents can limit the size of a system and produce ambiguous
results, we have been able to control sequentially 1) the
number of NDI units and 2) their relative dispositions
employing a template-assisted strategy and mechanical bond-
ing. As a result, we have been able to show in a systematic
fashion the through-space delocalization of an electron
through co-facial stacks of NDIs with only a moderate loss
in transport as the length of the array is increased. The present
data suggests that the electron sharing process is conforma-
tionally gated by the relative motions of NDI units within the
stack as a consequence of the mechanical bond, which allows
individual rotaxane components to move in and out of active
electron sharing domains. Strategies to attenuate and control
these degrees of freedom in order to elucidate and improve
electron transport in these materials are currently in progress
in our laboratories. These experimental results demonstrate
the untapped utility of mechanically interlocked molecules in
constructing functional organic motifs, which may serve to
enhance our understanding of organic electronics in the
future.
Received: November 6, 2013
Published online: March 12, 2014
Figure 6. ENDOR Spectroscopy of a) 3Cꢀ and b–e) the ([n]NDIxR)Cꢀ
series after reduction with 1.0 equiv of CoCp2 provides conclusive
evidence for electron sharing between the NDI units in the linear
oligorotaxane arrays. All spectra were recorded in Ar-purged CH2Cl2
solutions (0.75 mm) at 270 K. The proton Larmor frequency is indi-
cated by n0. The near-perfect one-half reduction of the proton-hyperfine
interactions in c) ([3]NDI2R)Cꢀ confirms that the electron is being
shared fully across the two NDI units on the ENDOR timescale. The
ENDOR spectra of d) ([4]NDI3R)Cꢀ and e) ([5]NDI4R)Cꢀ reveal fractional
sharing of the unpaired electrons and the existence of different
populations, most likely as a consequence of the dynamic p–p
interactions between three and four NDI units in the arrays, respec-
tively.
Keywords: electron transport ·
.
mechanically interlocked molecules · radical anions ·
self-assembly · template-directed synthesis
[1] a) D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 1993, 26,
198 – 205; b) C. J. Murphy, M. R. Arkin, Y. Jenkins, N. D.
Ghatlia, S. H. Bossmann, N. J. Turro, J. K. Barton, Science
1993, 262, 1025 – 1029; c) J. C. Genereux, J. K. Barton, Chem.
Rev. 2010, 110, 1642 – 1662.
[2] a) M. R. Wasielewski, Acc. Chem. Res. 2009, 42, 1910 – 1921;
ˇ
b) R. Bhosale, J. Misek, N. Sakai, S. Matile, Chem. Soc. Rev.
are within a favorable geometry for electron sharing to occur.
In short, the ENDOR data suggests that, while each and every
NDI unit in the array is involved in mediating an electron
through the array, their p-orbitals may not be perfectly
aligned into active electron sharing domains at all times. This
idea is consistent with previous reports in the literature[31] on
non-mechanically interlocked assemblies where electron
sharing is only observed to occur through a fraction of the
system and can be attributed to the concept of conformational
gating.[32] In this case, electron sharing is mediated by the
thermally driven intramolecular motions which enable the
alignment of NDI p-orbitals within the oligorotaxane. Indeed,
the degree of p-orbital overlap between co-facial NDI units
could be limited certainly by the slipped-stacked and twisted
2010, 39, 138 – 149; c) F. Garo, R. Hꢁner, Angew. Chem. 2012,
124, 940 – 943; Angew. Chem. Int. Ed. 2012, 51, 916 – 919; d) P.
Frischmann, K. Mahata, F. Wꢂrthner, Chem. Soc. Rev. 2013, 42,
1847 – 1870; e) G. Sforazzini, E. Orentas, A. Bolag, N. Sakai, S.
Matile, J. Am. Chem. Soc. 2013, 135, 12082 – 12090.
[3] a) A. Takai, T. Yasuda, T. Ishizuka, T. Kojima, M. Takeuchi,
Angew. Chem. 2013, 125, 9337 – 9341; Angew. Chem. Int. Ed.
2013, 52, 9167 – 9171; b) A. C. Fahrenbach, S. C. Warren, J. T.
Incorvati, A.-J. Avestro, J. C. Barnes, J. F. Stoddart, B. A.
Grzybowski, Adv. Mater. 2013, 25, 331 – 348.
[4] a) V. Percec, M. Glodde, T. K. Bera, Y. Miura, I. Shiyanovskaya,
K. D. Singer, V. S. K. Balagurusamy, P. A. Heiney, I. Schnell, A.
Rapp, H.-W. Spiess, S. D. Hudson, H. Duan, Nature 2002, 417,
384 – 387; b) T. Aida, E. W. Meijer, S. I. Stupp, Science 2012, 335,
813 – 817; c) H. Usta, M. D. Yilmaz, A.-J. Avestro, D. Boudinet,
Angew. Chem. Int. Ed. 2014, 53, 4442 –4449
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4447

.
Angewandte
Communications
M. Denti, J. F. Stoddart, A. Facchetti, Adv. Mater. 2013, 25,
4327 – 4334.
[16] a) P. T. Glink, A. I. Oliva, J. F. Stoddart, A. J. P. White, D. J.
Williams, Angew. Chem. 2001, 113, 1922 – 1927; Angew. Chem.
Int. Ed. 2001, 40, 1870 – 1875; b) M. Horn, J. Ihringer, P. T. Glink,
J. F. Stoddart, Chem. Eur. J. 2003, 9, 4046 – 4054.
[17] a) S. J. Rowan, S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders,
J. F. Stoddart, Angew. Chem. 2002, 114, 938 – 993; Angew. Chem.
Int. Ed. 2002, 41, 898 – 952; b) J. M. Lehn, Chem. Soc. Rev. 2007,
36, 151 – 160; c) M. E. Belowich, J. F. Stoddart, Chem. Soc. Rev.
2012, 41, 2003 – 2024; d) Y. Jin, C. Yu, R. J. Denman, W. Zhang,
Chem. Soc. Rev. 2013, 42, 6634 – 6654.
[5] a) W. B. Davis, W. A. Svec, M. A. Ratner, M. R. Wasielewski,
Nature 1998, 396, 60 – 63; b) C. A. Hunter, S. Tomas, J. Am.
Chem. Soc. 2006, 128, 8975 – 8979; c) P. M. Beaujuge, J. M. J.
Frꢃchet, J. Am. Chem. Soc. 2011, 133, 20009 – 20029; d) A.
Coskun, J. M. Spruell, G. Barin, W. R. Dichtel, A. H. Flood,
Y. Y. Botros, J. F. Stoddart, Chem. Soc. Rev. 2012, 41, 4827 –
4859.
[6] a) H. Oevering, M. N. Paddon-Row, M. Heppener, A. M. Oliver,
E. Cotsaris, J. W. Verhoeven, N. S. Hush, J. Am. Chem. Soc. 1987,
109, 3258 – 3269; b) E. A. Weiss, M. J. Ahrens, L. E. Sinks, A. V.
Gusev, M. A. Ratner, M. R. Wasielewski, J. Am. Chem. Soc.
2004, 126, 5577 – 5584; c) M. Scott, T. Miura, A. Butler Ricks,
Z. E. X. Dance, E. M. Giacobbe, M. T. Colvin, M. R. Wasielew-
ski, J. Am. Chem. Soc. 2009, 131, 17655 – 17666.
[7] a) F.-R. F. Fan, J. Yang, L. Cai, D. W. Price, Jr., S. M. Dirk, D. V.
Kosynkin, Y. Yao, A. M. Rawlett, J. M. Tour, A. J. Bard, J. Am.
Chem. Soc. 2002, 124, 5550 – 5560; b) S. A. Serron, S. A.
Aldridge III, C. N. Fleming, R. M. Danell, M.-H. Bailk, M.
Sykora, D. M. Dettelbaum, T. J. Meyer, J. Am. Chem. Soc. 2004,
126, 14506 – 14514; c) T. M. Wilson, T. A. Zeidan, M. Hariharan,
F. D. Lewis, M. R. Wasielewski, Angew. Chem. 2010, 122, 2435 –
2438; Angew. Chem. Int. Ed. 2010, 49, 2385 – 2388; d) S.
Sengupta, D. Ebeling, S. Patwardhan, X. Zhang, H. von Ber-
lepsch, C. Bçttcher, V. Stepanenko, S. Uemura, C. Hentschel, H.
Fuchs, F. C. Grozema, L. D. A. Siebbeles, A. R. Holzwarth, L.
Chi, F. Wꢂrthner, Angew. Chem. 2012, 124, 6484 – 6488; Angew.
Chem. Int. Ed. 2012, 51, 6378 – 6382.
[18] a) M. E. Belowich, C. Valente, R. A. Smaldone, D. C. Friedman,
J. Thiel, L. Cronin, J. F. Stoddart, J. Am. Chem. Soc. 2012, 134,
5243 – 5261; b) A.-J. Avestro, M. E. Belowich, J. F. Stoddart,
Chem. Soc. Rev. 2012, 41, 5869 – 6216.
[19] a) S. C. Bhosale, S. V. Bhosale, S. K. Bhargava, Org. Biomol.
Chem. 2012, 10, 6455 – 6468; b) Z. Zhu, C. J. Cardin, Y. Gan,
H. M. Colquhoun, Nat. Chem. 2010, 2, 653 – 660; c) A. Das, S.
Ghosh, Macromolecules 2013, 46, 3939 – 3949; d) M. B. Avinash,
T. Govindaraju, Adv. Mater. 2012, 24, 3905 – 3922; e) M. B.
Avinash, E. Verheggen, C. Schmuck, T. Govindaraju, Angew.
Chem. 2012, 124, 10470 – 10474; Angew. Chem. Int. Ed. 2012, 51,
10324 – 10328; f) R. S. Lokey, Y. Kwok, V. Guelev, C. J. Pursell,
L. H. Hurley, B. L. Iverson, J. Am. Chem. Soc. 1997, 119, 7202 –
7210; g) A. R. Smith, B. L. Iverson, J. Am. Chem. Soc. 2013,
1035, 12783 – 12789; h) S. V. Bhosale, C. H. Jani, S. J. Langford,
Chem. Soc. Rev. 2008, 37, 331 – 342; i) N. Sakai, J. Mareda, E.
Vauthey, S. Matile, Chem. Commun. 2010, 46, 4225 – 4237; j) S.
Tu, S. H. Kim, J. Joseph, D. A. Modarelli, J. R. Parquette,
ChemPhysChem 2013, 14, 1609 – 1617; k) H. Shao, T. Nguyen,
N. C. Romano, D. A. Modarelli, J. R. Parquette, J. Am. Chem.
Soc. 2009, 131, 16374 – 16376; l) S. Guha, F. S. Goodson, S. Roy,
L. J. Corson, C. A. Gravenmier, S. Saha, J. Am. Chem. Soc. 2011,
133, 15256 – 15259; m) N. Ponnuswamy, F. B. L. Cougnon, J. M.
Clough, G. D. Pantos¸, J. K. M. Sanders, Science 2012, 338, 783 –
785; n) S. P. Black, A. R. Stefankiewicz, M. M. J. Smulders, D.
Sattler, C. A. Schalley, J. R. Nitschke, J. K. M. Sanders, Angew.
Chem. 2013, 125, 5861 – 5864; Angew. Chem. Int. Ed. 2013, 52,
5749 – 5752; o) C. Shao, M. Stolte, F. Wꢂrthner, Angew. Chem.
2013, 125, 7630 – 7634; Angew. Chem. Int. Ed. 2013, 52, 7482 –
7486; p) S. Sengupta, F. Wꢂrthner, Acc. Chem. Res. 2013, 46,
2498 – 2512; q) Y. S. Chong, W. R. Carroll, W. G. Burns, M. D.
Smith, K. D. Shimizu, Chem. Eur. J. 2009, 15, 9117 – 9126; r) X.
Guo, D. Zhang, D. Zhu, Adv. Mater. 2004, 16, 125 – 130.
[20] X. Zhan, A. Facchetti, S. Barlow, T. J. Marks, M. A. Ratner,
M. R. Wasielewski, S. R. Marder, Adv. Mater. 2011, 23, 268 – 284.
[21] The peak integrations are consistent with the ratios of the
constitutionally heterotopic outer and inner rings (blue and red
in the 1H NMR spectra) present in the [n]NDIxR series—that is,
2:1 and 1:1 for x = 3 and 4, respectively.
[22] All efforts resulted in single crystals that were either twinned or
unable to diffract with high enough resolution. The ability of the
rings to rotate or twist around the dumbbells in order to
accommodate the sterically bulky groups could result conceiv-
ably in multiple favorable co-conformations in the solid state.
Surprisingly, out of the entire [n]NDIxR series, the most
successful crystallizations were achieved from solutions
(0.9 mgmLꢀ1) of [5]NDI4R in CH2Cl2/n-heptane grown at 08C,
yet these crystals were only able to diffract up to 2.8 ꢀ using a Cu
beam source. We attribute this observation to the more rigid
structure afforded by multiple interactions when x = 4.
[8] a) Y. Morisaki, S. Ueno, A. Saeki, A. Asano, S. Seki, Y. Chujo,
Chem. Eur. J. 2012, 18, 4216 – 4224; b) M. Wielopolski, A.
Molina-Ontoria, C. Schubert, J. T. Margraf, E. Krokos, J.
Kirschner, A. Gouloumis, T. Clark, D. M. Guldi, N. Martin, J.
Am. Chem. Soc. 2013, 135, 10372 – 10381.
[9] J. E. Bullock, R. Carmieli, S. M. Mickley, J. Vura-Weis, M. R.
Wasielewski, J. Am. Chem. Soc. 2009, 131, 11919 – 11929.
[10] a) S. Anderson, H. L. Anderson, J. K. M. Sanders, Acc. Chem.
Res. 1993, 26, 469 – 475; b) R. Cacciapaglia, L. Manodolini,
Chem. Soc. Rev. 1993, 22, 221 – 231; c) R. Hoss, F. Vçgtle, Angew.
Chem. 1994, 106, 389 – 398; Angew. Chem. Int. Ed. Engl. 1994,
33, 375 – 384; d) G. A. Breault, C. A. Hunter, P. C. Mayers,
Tetrahedron 1999, 55, 5265 – 5293; e) T. J. Hubin, D. H. Busch,
Coord. Chem. Rev. 2000, 200, 5 – 52; f) M.-J. Blanco, J.-C.
Chambron, M. C. Jimꢃnez, J.-P. Sauvage, Top. Stereochem.
2002, 23, 125 – 173; g) D. H. Busch, Top. Curr. Chem. 2005, 249,
1 – 65; h) C. D. Meyer, C. S. Joiner, J. F. Stoddart, Chem. Soc.
Rev. 2007, 36, 1705 – 1723.
[11] a) T. R. Kelly, R. L. Xie, C. K. Weinreb, T. Bregant, Tetrahedron
Lett. 1998, 39, 3675 – 3678; b) Y. Xia, P. Yang, Y. Sun, Y. Wu, B.
Mayers, B. Gates, Y. Yin, F. Kim, H. Yan, Adv. Mater. 2003, 15,
353 – 389; c) L. Palmer, S. I. Stupp, Acc. Chem. Res. 2008, 41,
1674 – 1684.
[12] Y.-L. Wu, K. E. Brown, M. R. Wasielewski, J. Am. Chem. Soc.
2013, 135, 13322 – 13325.
[13] a) N. Sakai, M. Lista, O. Kel, S. Sakurai, D. Emery, J. Mareda, E.
Vauthey, S. Matile, J. Am. Chem. Soc. 2011, 133, 15224 – 15227;
b) M. Lista, J. Areephong, N. Sakai, S. Matile, J. Am. Chem. Soc.
2011, 133, 15228 – 15230; c) N. Sakai, S. Matile, J. Am. Chem.
Soc. 2011, 133, 18542 – 18545.
[14] a) Y. Yamauchi, M. Yoshizawa, M. Akita, M. Fujita, J. Am.
Chem. Soc. 2010, 132, 960 – 966; b) M. Kiguchi, T. Takahashi, Y.
Takahashi, Y. Yamauchi, T. Murase, M. Fujita, T. Tada, S.
Watanabe, Angew. Chem. 2011, 123, 5826 – 5829; Angew. Chem.
Int. Ed. 2011, 50, 5708 – 5711.
[23] Crystal data for the [3]rotaxane [3]Ph2R: C95H109F12N9O14P2,
MW= 1890.891, monoclinic, space group P1n1, a =
24.5464(18), b = 15.2507(10), c = 25.4313(19) ꢀ, b = 95.403(5)8,
V= 9477.9(12) ꢀ3, T= 100(2) K, Z = 4, 1calc = 1.325 gcmꢀ3, m-
(Cu-Ka) = 1.188, F(000) = 3968. Independent measured reflec-
tions 16367. R1 = 0.1117, wR2 = 0.3415 for 9167 independent
observed reflections [2q ꢃ 57.628, I > 2s(I)]. CCDC 963939
[15] M. Kiguchi, O. Tal, S. Wohlthat, F. Pauly, M. Krieger, D. Djukic,
J. C. Cueves, J. M. van Ruitenbeek, Phys. Rev. Lett. 2008, 101,
046801.
4448 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 4442 –4449

Angewandte
Chemie
([3]Ph2R) contains 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.
For more details, see: A. Cisak, P. J. Elving, Electrochim. Acta
1965, 10, 935 – 946.
[28] Sharing of an unpaired spin between identical monomeric units
is characterized by narrowing of the EPR spectral linewidth by
a factor proportional to the square root of the number of
[24] The computed structure of [3]NDI2R was obtained by modeling
the NDI units directly onto the X-ray crystal structure obtained
for the related [3]rotaxane [3]Ph2R (see Ref. [23]), which is
devoid of the NDI functionality, and performing an uncon-
strained geometry optimization at the B3LYP-D3/6-31G** level
of theory (Ref. [25]). Density functional theory (DFT) measure-
ments were only carried out on the [2]rotaxane [3]NDI2R since
calculations on the higher order [n]rotaxanes failed to converge
in a reasonable amount of time.
[25] a) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785 – 789;
b) A. D. Becke, J. Chem. Phys. 1993, 98, 1372 – 1377; c) S.
Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010,
132, 154104.
[26] UV/Vis spectroscopy (Figure S27) of the [n]NDIxR series
(5.0 mm in CH2Cl2) reveals the existence of three absorption
bands between 330 and 400 nm characteristic of other NDI
derivatives in the literature (Ref. [19]). An intensity reversal of
the absorptions centered at 360 and 380 nm, relative to one
another as the number of NDI units is increased, may indicate
weak exciton coupling between the NDI units in the [n]rotax-
anes in CH2Cl2 solution. For more details, see: a) M. Kasha,
H. R. Rawles, M. L. El-Bayoumi, Pure Appl. Chem. 1965, 11,
371 – 392; b) A. D. Q. Li, W. Wang, L. Q. Wang, Chem. Eur. J.
2003, 9, 4594 – 4601; c) J. M. Giaimo, J. V. Lockard, L. E. Sinks,
A. M. Scott, T. M. Wilson, M. R. Wasielewski, J. Phys. Chem. A
2008, 112, 2322 – 2330.
monomers
involved,
NS DHM
as
given
by
Equation (1):
ꢁ ꢀ
ꢂ
pffiffiffiffiffiffi
DHN
¼
1
which relates the linewidth of the monomer (DHM) to the
observed linewidth (DHN) when an unpaired spin is shared over
NS molecular sites. For more details, see: J. R. Norris, R. A.
Uphaus, H. L. Crespi, J. J. Katz, Proc. Natl. Acad. Sci. USA 1971,
68, 625 – 628.
[29] H. Kurreck, B. Kirste, W. Lubitz, Electron Nuclear Double
Resonance Spectroscopy of Radicals in Solution, VCH, New
York, 1988.
[30] a) J.-F. Penneau, B. J. Stallman, P. H. Kasai, L. L. Miller, Chem.
Mater. 1991, 3, 791 – 796; b) L. L. Miller, R. G. Duan, Y. Hong, I.
Tabakovic, Chem. Mater. 1995, 7, 1552 – 1557; c) L. L. Miller,
K. R. Mann, Acc. Chem. Res. 1996, 29, 417 – 423.
[31] a) S. G. Chen, H. M. Branz, S. S. Eaton, P. C. Taylor, R. A.
Cormier, B. A. Gregg, J. Phys. Chem. B 2004, 108, 17329 – 17336;
b) Y. K. Che, A. Data, X. M. Yang, T. Naddo, J. C. Zhao, L.
Zang, J. Am. Chem. Soc. 2007, 129, 6354 – 6355; c) T. M. Wilson,
M. J. Tauber, M. R. Wasielewski, J. Am. Chem. Soc. 2009, 131,
8952 – 8957.
[32] The term “conformational gating” has been used to describe the
thermally driven motion of base pairs to mediate hole transfer
within double-stranded DNA. For more examples, see: a) G. I.
Likhtenshtein, J. Photochem. Photobiol. A 1996, 96, 79 – 92; b) S.
Delaney, J. K. Barton, J. Org. Chem. 2003, 68, 6475 – 6483;
c) M. A. OꢄNeill, J. K. Barton, J. Am. Chem. Soc. 2004, 126,
11471 – 11483; d) E. A. Weiss, M. J. Tauber, R. F. Kelley, M. J.
Ahrens, M. A. Ratner, M. R. Wasielewski, J. Am. Chem. Soc.
2005, 127, 11842 – 11850.
[27] Since subjecting the [n]rotaxane samples (1.0 mm in Ar-purged
CH2Cl2) to multiple redox cycles beyond ꢀ1.44 V resulted in
decomposition of the molecules, presumbly as a consequence of
the irreversible reduction of the pyridine ring, applying high
negative potentials were avoided when recording CV spectra.
Angew. Chem. Int. Ed. 2014, 53, 4442 –4449
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4449