[2,2']paracyclophane-based π -conjugated molecular wires reveal molecular-junction behavior

Article abstract of DOI:10.1021/ja109745a

The electronic coupling as well as the attenuation factor (β), which depends primarily on the nature of the molecular bridge and is used as a benchmark to test the molecular wire behavior, have been determined in a systematic study carried out on a series

Full text of DOI:10.1021/ja109745a

COMMUNICATION
pubs.acs.org/JACS
[2,20]Paracyclophane-Based π-Conjugated Molecular Wires Reveal
Molecular-Junction Behavior
Agustín Molina-Ontoria,† Mateusz Wielopolski,‡ Julian Gebhardt,‡ Andreas Gouloumis,† Timothy Clark,§
||
Dirk M. Guldi,*,‡ and Nazario Martín*,†,
†Departamento de Química Orgꢀanica, Facultad de Química, Universidad Complutense, E-28040 Madrid, Spain
‡Department of Chemistry and Pharmacy and Interdisciplinary Center for Molecular Materials (ICMM), Friedrich-Alexander-
Universit€at Erlangen-N€urnberg, Egerlandstrasse 3, 91058 Erlangen, Germany
§Computer-Chemie-Centrum, Cluster of Excellence Engineering of Advanced Materials, University of Erlangen, N€agelsbachstrasse 25,
91052 Erlangen, Germany
Facultad de Ciencias, IMDEA-Nanociencia, Campus Cantoblanco, E-28049 Madrid, Spain
S Supporting Information
b
the rate constant becomes linear because the weaker length
ABSTRACT: The electronic coupling as well as the
attenuation factor (β), which depends primarily on the
nature of the molecular bridge and is used as a benchmark
to test the molecular wire behavior, have been determined in
a systematic study carried out on a series of ZnP/C60
conjugates connected through a [2,20]paracyclophane-
oligophenylenevinylene (pCp-oPPV). The convergent
synthesis involves a series of Horner-Emmons olefination
reactions or double palladium-catalized Heck-type reac-
tions. ZnP-pCp-C60 conjugates were finally obtained by
the 1,3-dipolar cycloaddition reaction of the in situ-gener-
ated azomethyne ylide containing the ZnP-pCp moiety to
the [60]fullerene using Prato conditions. Experimental
(UV-vis, fluorescence, transient absorption spectroscopy,
and solution electrochemistry) and theoretical studies
revealed that the pCps act as molecular junctions. If hole
transfer is assumed to be the dominant charge transfer (CT)
mechanism, CT is facilitated in one direction (from C60 to
ZnP via pCp) but disfavored in the other direction (from
ZnP to C60 via pCp).
dependence associated with the hopping regime facilitates the
transport of charges over sizable distances. β quantifies the
charge transfer (CT) capability of the bridge and in turn is a
bridge-specific parameter. As such, it depends both on the
magnitude of the electronic coupling between the electron donor
and electron acceptor and on the energy of the localized CT
states (i.e., electron or hole).
Recent work on π-conjugated electron donor-acceptor
conjugates has provided a wealth of information.4 However,
relatively little is known about the physicochemical features
of π systems that contain subunits with strong through-space
interactions, such as [2,20]paracyclophane (pCp). In pCp,
the close proximity of the arene moieties results in strong
electronic and structural interactions between the two rings. As
a consequence, styryl[2,20]paracyclophane derivatives exhibit
highly conductive behavior due to efficient through-space π-π
coupling.5
In the current investigation, on the basis of our previous work
on charge transport, that is, charge separation and charge
recombination through molecular wires,6 we have bridged a
series of ZnP-C60 conjugates by [2,20]paracyclophane-
oligophenylenevinylene (pCp-oPPV) π systems having different
lengths. Neither the magnitude of electronic coupling nor the
attenuation factor between electron donors and electron accep-
tors linked by pCps has previously been determined.
Compounds 1-4 (Chart 1; for details, see the Supporting
Information) were prepared using 4,12-diformylparacyclo-
phane7 (5) in the case of 1 or pCps bearing substituted bis-
benzaldehydes (10) for 2-4. In the first step, these were coupled
to a zinc porphyrin (ZnP) using a procedure similar to that
previously reported for related endohedral fullerenes.8 In the
second step, the correspondingly formed monoaldehydes were
attached to C60 via a 1,3-dipolar cycloaddition reaction.9 The
synthesis of precursor monoaldehyde 14 for the preparation of 3
was carried out by consecutive Wittig and Heck reactions (see
the Supporting Information).
π-Conjugated organic molecules have been widely used for
studying charge-transport processes in solution and/or the solid
state.1 In this context, electronic coupling/communication with a
length dependence that is directly related to the electronic
properties of the individual constituents (electron donor, bridge,
and electron acceptor) is decisive.1 Efficient charge transport is
in fact realized only when the synergy between different parts
leads to “molecular wire” behavior, which describes π-conjugated
molecules capable of transporting charges efficiently over length
scales far beyond those of chemical bonds.2
For short molecular bridges, resonant-tunneling electron
transfer leads to an exponential dependence on the length3
described by k = k0e-βR, where R is the distance between the
electroactive constituents and β is the attenuation factor, which
depends primarily on the nature of the molecular bridge. For
long molecular bridges, the relationship between the length and
Received: November 12, 2010
Published: February 7, 2011
r
2011 American Chemical Society
2370
dx.doi.org/10.1021/ja109745a J. Am. Chem. Soc. 2011, 133, 2370–2373
|

Journal of the American Chemical Society
COMMUNICATION
Chart 1. Different ZnP-pCp-C60 Triads (1-4) Prepared
in This Work
Figure 1. Differential absorption spectra (visible and near-IR) obtained
upon femtosecond flash photolysis (420 nm, 90 nJ) of 3 in argon-
saturated THF with several time delays between 0 and 3000 ps at room
temperature. Inset: corresponding time profiles at 455, 530, and 1000
nm that monitor the charge-separation dynamics.
quantum yields are 4.0 ꢀ 10-3 for 1, 1.6 ꢀ 10-2 for 2, 1.9 ꢀ
10-2 for 3, and 2.6 ꢀ 10-2 for 4. This suggests that there
are sizable interactions between the strongly fluorescent ZnP
and electron-accepting C60 units in the excited state. Increasing
the solvent polarity by using benzonitrile strengthens the
fluorescence quenching. In contrast, increasing the donor-accep-
tor separation weakens it. In complementary time-resolved
measurements, we determined ZnP fluorescence lifetimes of
200, 930, 1270, and 1700 ps for 1, 2, 3, and 4, respectively. The
latter are a quantitative match of the fluorescence quantum yields.
Complementary transient absorption measurements were
used to investigate the nature of the deactivation processes
evolving from photoexcitation. The characteristic features due
to singlet excited states of the references 11 and 19 (see the
Supporting Information) are, for example, formed instanta-
neously. In the case of pCp-oPPV, singlet-singlet absorptions
develop that include a transient bleach between 430 and 500 nm
and a broad transient maximum at 600 nm (Figure S4).
Characteristics of the ZnP singlet excited state are a bleach
around 560 nm and transients at 480 nm and between 580 and
750 nm. Fast intersystem-crossing processes (1011 s-1) govern
the fate of the singlet excited states in 11 and 19 and lead to the
triplet manifolds. The singlet excited state lifetimes of 11 and 19
were determined to 21 and 56 ps, respectively. The correspond-
ing triplet-triplet absorptions were found as broad transients
between 500 and 900 nm (Figure S4).
Moving on to the donor-acceptor systems 1-4, we focused
on the generation of the singlet excited states of ZnP. In
accordance with previous publications, however, we may safely
assume that starting with the C60 singlet excited state would lead
to similar reaction patterns.10 At the very beginning of the
experiments, the transient absorption spectra of 1-4 were those
found for the instantaneous formation of the ZnP singlet excited
state. However, instead of the fast intersystem crossing dynamics
and growth of the triplet, we saw that the singlet-singlet absorp-
tion decayed quickly in the presence of C60 (Figure 1). The
singlet excited state lifetimes ranged from 200 ps for 1 to 830 ps
for 2, 1430 ps for 3, and 1890 ps for 4, resembling the corres-
ponding fluorescence lifetimes. At the end of the singlet
decay, new transient features emerged with maxima in the
The solution electrochemistry of 1-4 was studied using cyclic
voltammetry at room temperature and compared with that of
ZnP and C60-pCp, as shown in Figure S1 and Tables S1 and S2
in the Supporting Information. They are all electrochemically
active in both the anodic and cathodic scan directions. In the
cathodic scans, 1-4 exhibit four reductions at -0.84, -1.42,
-1.79, and -2.04 V. The first, second, and fourth reversible one-
electron reductions agree reasonably well with the C60 reduction
in C60-pCp. The third quasi-reversible one-electron reduction,
on the other hand, corresponds to the ZnP reduction and is in
excellent agreement with that seen for ZnP itself. In the anodic
scan, only the quasi-reversible one-electron oxidation of ZnP
emerges at 0.53 V. In short, the redox features of 1-4 are simply
the superimposition of those seen for the references. This leads
us to conclude that no significant electronic interactions between
the different redox- and photoactive constituents exist in the
ground state.
The UV-vis absorption spectra of 1-4 are dominated by the
Soret and Q bands of ZnP at 430 and 550/604 nm, respectively
(Figure S2). Characteristic absorptions of C60 and pCp-oPPV are
found in the UV region and show a significant overlap. In
particular, C60 absorbs between 250 and 320 nm, whereas the
broad pCp-oPPV absorption maximum in THF gradually shifts
from 304 to 329 nm. In comparison with the absorption of
pristine oPPVs of equal length, the absorption maxima of pCp-
oPPV are blue-shifted by ∼20 nm. This suggests weaker π
conjugation across the pCp core. All three component moieties
(ZnP, pCp-oPPV, and C60) emit singlet excited state energy in
different regions of the solar spectrum. While the fluorescence of
ZnP is typically observed in the region between 600 and 700 nm,
pCp-oPPVs emit between 430 and 550 nm. The C60 emission
with its maximum at 715 nm is, however, masked by the stronger
ZnP emission; quantum yields for these processes are 6.0 ꢀ 10-4
and 4.0 ꢀ 10-2, respectively. Interestingly, in THF, the ZnP-
centered fluorescence is appreciably quenched (Figure S3) in all
of the ZnP-pCp-oPPV-C60 conjugates, regardless of the
system being excited into the Soret- or Q-band region; the
2371
dx.doi.org/10.1021/ja109745a |J. Am. Chem. Soc. 2011, 133, 2370–2373

Journal of the American Chemical Society
COMMUNICATION
Figure 3. Visualization of the electrostatic potentials of the BS states of
2, 3, and 4.
Figure 2. Dependence of the charge-separation (kCS) and charge-
recombination (kCR) rate constants on the center-to-center separation
(RCC) in nitrogen-saturated THF at room temperature for the directly
connected 1-4. The slopes represent the β values.
Further insight into the electronic conduction across the pCp
linkage was provided by molecular orbital (MO) calculations.
Figure S7 shows that the HOMO and LUMO are localized on
ZnP and C60, respectively. Increased localization of the electron
density on the pCp-oPPV bridge is found for the lower-energy
orbitals (HOMO-1 to HOMO-3). In 1, the space between the
donor and acceptor is fully conjugated because of the short
donor-acceptor distance, as indicated by the orbital coefficients.
The HOMO-2 orbital extends over the bridge. Increasing the
pCp-oPPV length in 2-4 changes the situation. While the
HOMO/HOMO-1 and LUMO remain localized on ZnP and
C60, respectively, the HOMO-2 and HOMO-3 are localized
exclusively on the pCp-oPPVs. Importantly, no electron density
can be found across the pCps, demonstrating the interruption of
through-bond π-conjugation.
Configuration interaction (CI) calculations in a simulated
solvent environment (diethyl ether) indicated that the presence
of pCps leads to localized bridge CT states (BS states), which are
stabilized within the π-electron framework of the oPPVs and are
nearly degenerate (i.e., within ∼2 kcal mol-1) with the CT states
(Table S3). In these bridge states, the positive charge actually
resides on the pCps, and the stabilization within the ZnP-pCp-
oPPV-C60 systems is apparently stronger on C60, as shown in
Figure 3.
As Table S3 shows, these states contribute to the overall CT
mechanisms because of their relatively low energies. Further, we
note that in our CI calculations, excitations from the HOMO-2
and HOMO-3 orbitals made the greatest contributions to the CT
states (see Table S4). Together with the fact that these orbitals lie
within 0.1 eV of each other, we may confirm that the through-
space conjugation in the pCps is sufficient to guarantee efficient
coupling between the two sides of the pCp cores (Table S4).
In view of the fact that hole transport dominates the charge
transport in π-conjugated molecular wires,12 these states may act
as intermediate steps in the CT process since they are stabilized
by smaller changes of the dipole moment. On the basis of these
results, we postulate that the pCps act as molecular junctions: CT
in one direction (from C60 via pCp to ZnP) is facilitated, whereas
it is disfavored in the opposite direction (from ZnP via pCp to C60).
Moreover, the local ionization potentials of 2-4 (Figure S8)
show low values (green) around the bonds of the bridge, whereas
high values (orange) are centered on the phenyl rings of the
bridge. This suggests that hole transfer is favored along a
homogeneous pathway of easily oxidizable sites.
visible (580-710 nm) and near-IR (800-1200 nm) regions.
The visible maxima at 480, 580, and 620 nm correspond to the
one-electron-oxidized π-radical cation of ZnP (ZnP•þ), while
the near-IR maximum at 1000 nm resembles the signature of
the one-electron-reduced form of C60 (C60•-). In other words,
the fast deactivation of the ZnP singlet excited state is linked to
the formation of the ZnP•þ-pCp-oPPV-C60•- radical ion pair
state. Although we cannot rule out the presence of the triplet
markers for C60 at 700 nm (see Figure S5) and ZnP at 840 nm
(see Figure S4), their formation is unlikely to occur because of
their high energies of ∼1.5 eV. In other words, the radical ion pair
states, with energies of 1.35 eV for 1, 1.40 eV for 2, 1.40 eV for 3,
and 1.42 eV for 4, are sufficiently below the two triplets.11
The ZnP•þ-pCp-oPPV-C60•- radical ion pair state lifetimes
exceeded the time scale of our femtosecond experiments (3 ns)
and were determined in nanosecond experiments. The spectral
•-
fingerprints of the ZnP•þ-pCp-oPPV-C60 radical ion pair
state, as seen immediately after the nanosecond laser pulse, were
useful probes in this context (Figure S6). The kinetics of the
ZnP•þ-pCp-oPPV-C60•- radical ion pair state decays obeyed
clean unimolecular rate laws. In strictly oxygen-free THF, the
lifetimes are 466 ns for 1, 974 ns for 2, 1373 ns for 3, and 1422 ns
for 4.
The contrast between the fast charge-separation processes
[i.e., 5.2 ꢀ 109 s-1 (1), 3.4 ꢀ 109 s-1 (2), 3.1 ꢀ 109 s-1 (3), and
2.1 ꢀ 109 s-1 (4)] and the slow charge-recombination processes
[i.e., 2.2 ꢀ 106 s-1 (1), 1.0 ꢀ 106 s-1 (2), 7.3 ꢀ 105 s-1 (3), and
7.0 ꢀ 105 s-1 (4)] leads us to conclude that the two processes are
located in the normal and inverted regions of the Marcus
parabola, respectively. In this context, the small reorganization
energy of C60 in electron-transfer reactions is certainly decisive.
However, the facts that pCps interrupt the homogeneous
through-bond π conjugation in oPPVs on the one hand and
institute strong through-space π-π coupling interactions on the
other hand are equally important.
Thus, we postulate that low-lying, through-space delocalized
states facilitate CT in one direction (i.e., charge separation) but
retard it in the opposite one (i.e., charge recombination).
Consequently, we analyzed the dependence of the CT rate con-
stants on the donor-acceptor separation. Figure 2 shows the
resulting plots. Their slopes give β values of 0.039 ( 0.001 Å-1
for charge separation and 0.045 ( 0.001 Å-1 for charge
recombination.
Finally, a comparison of the CT behavior of the pCp-oPPV
linkers to fully conjugated oPPV systems6 clearly visualizes
2372
dx.doi.org/10.1021/ja109745a |J. Am. Chem. Soc. 2011, 133, 2370–2373

Journal of the American Chemical Society
COMMUNICATION
the effects of the pCp units. The 20 nm blue shift of the pCp-
oPPV absorption maxima relative to the oPPV absorptions
(discussed above; see Figure S2) hints at weaker π-conjugation
of the pCp-oPPVs due to through-space interactions across pCp.
Similarly, visualization the HOMO-2 and HOMO-3 (Figure S7)
confirm the fact that at larger bridge lengths, the pCps break the
through-bond conjugation. In the end, these effects are reflected
in a β value that is ∼4 times greater than that for fully conjugated
oPPV bridges (i.e., 0.04 in pCp-oPPV vs 0.01 in oPPV).
Additionally, the inhomogenous coupling due to the pCp units
leads to different β values for charge separation and charge
recombination. In the oPPVs, on the other hand, the extremely
homogeneous distribution of π electrons along the bridge results
in equal transport properties for both directions (i.e., charge
separation and charge recombination).
Fernꢀandez, G.; Wielopolski, M.; Atienza, C.; Sꢀanchez, L.; Gouloumis, A.;
Clark, T.; Martín, N.; Guldi, D. M. J. Am. Chem. Soc. 2009, 131, 12218.
(7) Morisaki, Y.; Lin, L.; Chujo, Y. Chem. Lett. 2009, 38, 734.
(8) Wolfrum, S.; Pinzꢀon, J. R.; Molina-Ontoria, A.; Gouloumis, A.;
Martín, N.; Echegoyen, L.; Guldi, D. M. Chem. Commun., in press.
(9) Prato, M.; Maggini, M. Acc. Chem. Res. 1998, 31, 519.
(10) Giacalone, F.; Segura, J. L.; Martín, N.; Ramey, J.; Guldi, D. M.
Chem.—Eur. J. 2005, 11, 4819.
(11) Energies are based on the Born dielectric continuum model of
the solvent as estimated using the expression by Weller (see: Weller, A.
Z. Phys. Chem. (Munich) 1982, 133, 93).
(12) (a) Xue, Y.; Datta, S.; Ratner, M. A. J. Chem. Phys. 2001, 115,
4292. (b) Derosa, P. A.; Seminario, J. M. J. Phys. Chem. B 2001, 105, 471.
(c) Taylor, J.; Brandbyge, M.; Stokbro, K. Phys. Rev. Lett. 2002, 89, No.
138301.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures and
b
complete spectroscopic and structural analysis, including Figures
S1-S8. This material is available free of charge via the Internet at
http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
dirk.guldi@chemie.uni-erlangen.de; nazmar@quim.ucm.es
’ ACKNOWLEDGMENT
Financial support by the MEC of Spain (Projects CTQ2008-
0795/BQU and Consolider-Ingenio 2010C-07-25200), the
CAM (MADRISOLAR-2 Project S2009/PPQ-1533), the EU
(FUNMOL FP7-212942-1), and the Deutsche Forschungsge-
meinschaft (SFB 583 and Clusters of Excellence Engineering of
Advanced Materials) is acknowledged. A.M.-O. and A.G., thank
the CAM for a research grant and the MEC of Spain for a Ramꢀon y
Cajal Contract, respectively.
’ REFERENCES
(1) (a) Nitzan, A.; Ratner, M. A. Science 2003, 300, 1384. (b) Davis,
W. B.; Svec, W. A.; Ratner, M. A.; Wasielewski, M. R. Nature 1998, 396,
60. (c) Choi, S. H.; Kim, B.-S.; Frisbie, D. C. Science 2008, 320, 1482.
(2) (a) Roberston, N.; McCowan, C. A. Chem. Soc. Rev. 2003, 32, 96.
(b) Sikes, H. D.; Smalley, J. F.; Dudek, S. P.; Cook, A. R.; Newton, M. D.;
Chidsey, C. E. D.; Feldberg, S. W. Science 2001, 291, 1519.
(3) (a) Conwell, E. M. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 8795.
(b) Giese, B. Acc. Chem. Res. 2000, 33, 631.
(4) (a) James, D. K.; Tour, J. M. Top. Curr. Chem. 2005, 257, 33. (b)
Weiss, E. A.; Wasielewski, M. R.; Ratner, M. A. Top. Curr. Chem. 2005,
257, 103. (c) Guldi, D. M.; Illescas, B. M.; Atienza, C. M.; Wielopolski,
M.; Martín, N. Chem. Soc. Rev. 2009, 38, 1587.
(5) (a) Seferos, D. S.; Trammell, S. A.; Bazan, G. C.; Kushmerick,
J. G. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 8821. (b) Bartholomew,
G. P.; Bazan, G. C. Acc. Chem. Res. 2001, 34, 30. (c) Kahnt, A.; Guldi,
D. M.; de la Escosura, A.; Martínez-Díaz, M. V.; Torres, T. J. Mater.
Chem. 2008, 18, 77.
(6) (a) Giacalone, F.; Segura, J. L.; Martín, N.; Guldi, D. M. J. Am.
Chem. Soc. 2004, 126, 5340. (b) de la Torre, G.; Giacalone, F.; Segura,
J. L.; Martín, N.; Guldi, D. M. Chem.—Eur. J. 2005, 11, 1267. (c)
Atienza, C.; Martin, N.; Wielopolski, M.; Haworth, N.; Clark, T.; Guldi,
D. M. Chem. Commun. 2006, 3202. (d) Atienza-Castellanos, C.;
Wielopolski, M.; Guldi, D. M.; van der Pol, C.; Bryce, M. R.; Filippone,
S.; Martin, N. Chem. Commun. 2007, 5164. (e) Molina-Ontoria, A.;
2373
dx.doi.org/10.1021/ja109745a |J. Am. Chem. Soc. 2011, 133, 2370–2373