Photoinduced electron transfer in covalent ruthenium-anthraquinone dyads: Relative importance of driving-force, solvent polarity, and donor-bridge energy gap

Article abstract of DOI:10.1039/c2cp23240e

Four rigid rod-like molecules comprised of a Ru(bpy)₃ ²⁺ (bpy = 2,2′-bipyridine) photosensitizer, a 9,10-anthraquinone electron acceptor, and a molecular bridge connecting the two redox partners were synthesized and investigated by optical spectroscopic and electrochemical means. An attempt was made to assess the relative importance of driving-force, solvent polarity, and bridge variation on the rates of photoinduced electron transfer in these molecules. Expectedly, introduction of tert-butyl substituents in the bipyridine ligands of the ruthenium complex and a change in solvent from dichloromethane to acetonitrile lead to a significant acceleration of charge transfer rates. In dichloromethane, photoinduced electron transfer is not competitive with the inherent excited-state deactivation processes of the photosensitizer. In acetonitrile, an increase in driving-force by 0.2 eV through attachment of tert-butyl substituents to the bpy ancillary ligands causes an increase in electron transfer rates by an order of magnitude. Replacement of a p-xylene bridge by a p-dimethoxybenzene spacer entails an acceleration of charge transfer rates by a factor of 3.5. In the dyads from this study, the relative order of importance of individual influences on electron transfer rates is therefore as follows: solvent polarity ≥ driving-force > donor-bridge energy gap. the Owner Societies 2012.

Full text of DOI:10.1039/c2cp23240e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
PCCP
Cite this: Phys. Chem. Chem. Phys., 2012, 14, 2685–2692
www.rsc.org/pccp
PAPER
Photoinduced electron transfer in covalent ruthenium–anthraquinone
dyads: relative importance of driving-force, solvent polarity, and
donor–bridge energy gapw
Jihane Hankache and Oliver S. Wenger*
Received 14th October 2011, Accepted 13th December 2011
DOI: 10.1039/c2cp23240e
2+
Four rigid rod-like molecules comprised of a Ru(bpy)3 (bpy = 2,20-bipyridine) photosensitizer,
a 9,10-anthraquinone electron acceptor, and a molecular bridge connecting the two redox
partners were synthesized and investigated by optical spectroscopic and electrochemical means.
An attempt was made to assess the relative importance of driving-force, solvent polarity, and
bridge variation on the rates of photoinduced electron transfer in these molecules. Expectedly,
introduction of tert-butyl substituents in the bipyridine ligands of the ruthenium complex and a
change in solvent from dichloromethane to acetonitrile lead to a significant acceleration of charge
transfer rates. In dichloromethane, photoinduced electron transfer is not competitive with the
inherent excited-state deactivation processes of the photosensitizer. In acetonitrile, an increase in
driving-force by 0.2 eV through attachment of tert-butyl substituents to the bpy ancillary ligands
causes an increase in electron transfer rates by an order of magnitude. Replacement of a p-xylene
bridge by a p-dimethoxybenzene spacer entails an acceleration of charge transfer rates by a factor
of 3.5. In the dyads from this study, the relative order of importance of individual influences on
electron transfer rates is therefore as follows: solvent polarity Z driving-force > donor–bridge
energy gap.
In this work we have sought to assess experimentally the
Introduction
relative importance of variations in driving-force, solvent
2+
The Ru(bpy)3
complex represents an extremely popular
polarity, and donor–bridge energy gaps in a given system.
We reasoned that a donor–acceptor dyad with an inherently
low driving force for electron transfer might be particularly well
suited for this purpose because small variations in each of the
choice for investigations of photoinduced electron transfer,1–3
but in combination with 9,10-anthraquinone as an electron
acceptor it has received rather limited attention,4–13 at least when
compared to more potent redox partners such as viologens or
tertiary amines. 9,10-Anthraquinone (AQ) is a comparatively
three abovementioned parameters might have a particularly
2+
strong effect. Our choice therefore fell upon the Ru(bpy)3
/
weak oxidant,14 hence a small driving-force is usually associated
anthraquinone couple, which was incorporated into rigid
rod-like donor–bridge–acceptor molecules as shown in
Scheme 1. As a benchmark system, a dyad comprised of a
2+
with electron transfer between photoexcited Ru(bpy)3
and
AQ.15–17 The influence of driving-force and solvent variation
on the rates of electron transfer in donor–bridge–acceptor
molecules is nowadays fairly well understood, and the validity
of theoretical models has been thoroughly tested and confirmed
by numerous experimental investigations.1–3,18–21 The role of the
molecular bridge between the donor and the acceptor has been
explored in many biological and artificial systems,22–30 and one
has obtained a fairly complete picture of what factors contribute
to the overall charge transfer efficiency.
Georg-August-Universitat Gottingen, Institut fur Anorganische
¨
Chemie, Tammannstrasse 4, D-37077 Gottingen, Germany.
¨
¨
¨
E-mail: oliver.wenger@chemie.uni-goettingen.de;
Fax: +49 (0)551 39 3373; Tel: +49 (0)551 39 19424
w Electronic supplementary information (ESI) available: Detailed
synthetic protocols and product characterization data. See DOI:
10.1039/c2cp23240e
Scheme 1 Chemical formulas of the six molecules synthesized and
investigated in this work.
c
This journal is the Owner Societies 2012
Phys. Chem. Chem. Phys., 2012, 14, 2685–2692 2685

View Article Online
2+
Ru(bpy)3
donor (Ru), a p-xylene (xy) spacer, and an AQ
laser equipped with an OPO from Opotek. Attempts to measure
transient absorption were made using the same LP920-KS
instrument. Before all luminescence lifetime measurements,
samples were deoxygenated thoroughly by bubbling nitrogen
gas through the solutions.
acceptor (Ru-xy-AQ) was used. Driving-force variation was
achieved through attachment of tert-butyl groups to the
ancillary bpy ligands (Ru(tBu)-xy-AQ), and donor–bridge
energy gaps were altered through replacement of the p-xylene
spacer by a p-dimethoxybenzene (dmb) unit (Ru-dmb-AQ
and Ru(tBu)-dmb-AQ). The influence of solvent polarity was
explored by studying intramolecular electron transfer in these
dyads in dichloromethane and acetonitrile. As reference
molecules, two simple ruthenium complexes (Ru and Ru(tBu))
were employed (Scheme 1, left).
Results and discussion
Synthesis
Synthesis of the donor–bridge–acceptor molecules shown in
Scheme 1 was carried out following the synthetic strategy outlined
in Scheme 2. The key building blocks are 2,5-dimethyl-4-
trimethylsilylphenylboronic acid (2a) or 2,5-dimethoxy-4-
trimethylsilylphenylboronic acid (2b), 2-bromo-9,10-anthra-
quinone (3), and 5-(tri-n-butyltin)-2,20-bipyridine (1). While
molecule 3 is a commercially available chemical, the preparation
of molecules 1, 2a, and 2b has been previously described by us
and others.47–55 Coupling of the anthraquinone acceptor to the
p-xylene or p-dimethoxybenzene unit occurs conveniently
under Suzuki-type conditions, and subsequent trimethylsilyl
deprotection with iodine monochloride yields a compound
(5a/5b) which can readily be coupled to the bipyridine ligand
(1). The final step is the reaction with Ru(bpy)2Cl2,56 giving
the four dyads shown in Scheme 1 in overall yields ranging
from 26% to 41% with respect to the 2-bromo-9,10-anthra-
quinone (3) starting material. Detailed synthetic protocols and
product characterization data are given in the ESI.w
While it is a priori clear how each of the individual variations of
our systems (solvent exchange, tert-butyl group attachment to
bpy ancillary ligands of the electron donor,31 change of the
bridging spacer between the donor and acceptor) will qualitatively
affect photoinduced ruthenium-to-anthraquinone electron transfer,
the relative importance of these factors is more difficult to predict.
The importance of donor–bridge energy matching and so-called
tunneling energy gaps has long been known,32–34 and there have
appeared numerous studies on this subject in recent years,35–41
including the demonstration of accelerated charge transfer in a
p-dimethoxybenzene bridged system compared to molecules
with other phenylene spacers.42–46 The open question is
therefore not whether such acceleration will also be observed
in the Ru-dmb-AQ and Ru(tBu)-dmb-AQ dyads relative to the
Ru-xy-AQ and Ru(tBu)-xy-AQ molecules, but how important
this effect will be compared to other experimental parameters
that are known to affect electron transfer rates. The goal of the
present study was therefore to assess and quantify the relative
importance of variations in driving-force, solvent polarity, and
donor–bridge energy gap for a given donor–acceptor system.
Optical absorption spectroscopy
Fig. 1 shows the optical absorption spectra of the four dyads
and the two reference complexes shown in Scheme 1. The left
panels (a, c) exhibit data measured in dichloromethane, the right
panels (b, d) show data obtained from acetonitrile solutions.
Absorption spectra of Ru, Ru-xy-AQ, and Ru-dmb-AQ are
shown in the upper half (panels a, b) while the respective data
of Ru(tBu), Ru(tBu)-xy-AQ, and Ru(tBu)-dmb-AQ are given in
Experimental section
Synthetic procedures and product characterization data are
1
given in the ESI.w H NMR spectroscopy was performed on
Bruker Avance DRX 300 and Bruker B-ACS-120 spectrometers.
Electron ionization mass spectrometry (EI-MS) was performed
using a Finnigan MAT8200 instrument, elemental analysis was
performed on a Vario EL III CHNS analyzer from Elementar.
For cyclic voltammetry, a Versastat3-100 potentiostat from
Princeton Applied Research equipped with a Pt disk working
electrode and a platinum counter electrode was employed. A
silver wire served as a quasi-reference electrode. Ferrocene (Fc)
was used as an internal reference. The supporting electrolyte was
a 0.1 M solution of tetrabutylammonium hexafluorophosphate,
and prior to voltage sweeps at rates of 100 mV sꢀ1, nitrogen gas
was bubbled through the solutions. Optical absorption spectra
were measured on a Cary 300 spectrometer from Varian. Steady-
state luminescence spectra were recorded on a Fluorolog-3
instrument (FL322) from Horiba Jobin-Yvon, equipped
with a TBC-07C detector from Hamamatsu. Time-resolved
luminescence experiments were conducted on the same Fluorolog-3
instrument using the FL-1061PC Fluorohub for detection in
TCSPC mode and a NanoLed-407 as a pulsed excitation source,
or alternatively, using an LP920-KS instrument from Edinburgh
Instruments. The detection system of the LP920-KS spectrometer
consisted of an R928 photomultiplier and an iCCCD camera
from Andor. The excitation source was a Quantel Brilliant b
Scheme 2 Synthetic strategy for synthesis of the bipyridine–bridge–
anthraquinone ligands used in this work. (a) Pd(PPh3)4, toluene,
water, Na2CO3; (b) ICl, CH2Cl2; (c) Pd(PPh3)4, m-xylene.
c
2686 Phys. Chem. Chem. Phys., 2012, 14, 2685–2692
This journal is the Owner Societies 2012

View Article Online
Fig. 1 Optical absorption spectra of the molecules shown in
Fig. 2 Steady-state luminescence spectra of the molecules shown in
Scheme 1 in dichloromethane (a, c) and acetonitrile (b, d) obtained
after excitation at 450 nm.
Scheme 1 in dichloromethane (a, c) and acetonitrile (b, d).
the lower half (panels c, d). The key absorption features are the
same in all compounds in both solvents, namely the metal-to-
ligand charge transfer (MLCT) band of the ruthenium complex
centered at around 450 nm, and a bpy-localized p–p* transition
at around 290 nm.57 All dyads exhibit a shoulder to the lower
energy side of the 290 nm absorption band which may be
attributed to anthraquinone-localized transitions. An interesting
observation is the additional absorption between 370 and 450 nm
in the two dyads with the p-dimethoxybenzene spacer (dotted
traces): in Ru-dmb-AQ as well as in Ru(tBu)-dmb-AQ a broad
absorption with a maximum at around 400 nm is easily
discernable between the MLCT band at 450 nm and the
p–p* absorption at 290 nm. This additional band is tentatively
assigned to an electronic transition from the p-dimethoxybenzene
unit to the ruthenium complex. This observation may indicate
possible electronic delocalization between the ruthenium(II)
chromophore and the p-dimethoxybenzene spacer that could
be of importance for long-range electron transfer between the
ruthenium and anthraquinone redox partners.
molecules with tert-butyl substituted ancillary ligands, the relative
luminescence intensities follow the same order given above,
namely, Ru(tBu) > Ru(tBu)-xy-AQ > Ru(tBu)-dmb-AQ (Fig. 2c
and d). Thus, in the Ru(tBu)-xy-AQ and Ru(tBu)-dmb-AQ dyads
there appears to be more efficient nonradiative excited-state
deactivation than in the Ru(tBu) reference complex not only in
CH3CN, but also in the less polar CH2Cl2 solvent.
In dichloromethane, the dyad emission band maxima are
somewhat red-shifted with respect to the reference complexes.
This may reflect the fact that the bpy ligand bearing the
AQ-bridge moiety has its p* LUMO at somewhat lower
energy than the bpy and tert-butyl bpy ancillary ligands.
3
In principle, the emissive MLCT excited states of d6 metal
diimine complexes can be depopulated efficiently by triplet–
triplet energy transfer,58–66 provided the reaction energetics
are favorable. However, for Ru and Ru(tBu) the 3MLCT
energies are 2.12 eV and 2.16 eV,57 respectively, while the
lowest triplet excited state of anthraquinone is at 2.69 eV.67
Consequently, ruthenium-to-anthraquinone triplet–triplet energy
transfer is endergonic by more than 0.5 eV and a highly improbable
process. Electron transfer from photoexcited ruthenium to
anthraquinone is an alternative possibility, and hence we
examine the electrochemical properties of the molecules shown
in Scheme 1 next.
Steady-state luminescence spectroscopy
All six molecules shown in Scheme 1 are emissive in dichloro-
methane and acetonitrile solution after excitation into the MLCT
band at 450 nm. The resulting luminescence spectra obtained in
the two solvents are shown in Fig. 2, which is built up in the same
way as Fig. 1: emission data obtained from dichloromethane
solutions are on the left (a, c), spectra from acetonitrile solutions
on the right (b, d).
Electrochemistry
In each panel, the luminescence spectra are normalized to
the emission intensity of the Ru or Ru(tBu) reference complexes.
Fig. 2a shows that in dichloromethane solution Ru and Ru-xy-AQ
emit with nearly equal intensities, while the luminescence of the
Ru-dmb-AQ dyad is even somewhat stronger. From the acetonitrile
data of the same compounds in Fig. 2b a different picture emerges:
under these conditions the luminescence intensities decrease along
the series: Ru > Ru-xy-AQ > Ru-dmb-AQ, and there is a factor of
4 difference between the highest (Ru) and lowest (Ru-dmb-AQ)
emission intensities. This observation suggests that the change
in solvent from CH2Cl2 to CH3CN activates an additional
nonradiative excited-state depopulation mechanism. For the
Fig. 3 shows cyclic voltammetry data for the six molecules
shown in Scheme 1 in dichloromethane (left) and acetonitrile
(right) solution, measured in the presence of 0.1 M tetrabutyl-
ammonium hexafluorophosphate as a supporting electrolyte.
In all voltammograms there is a wave at 0.0 V caused by the
ferrocenium/ferrocene (Fc+/Fc) couple, which was used as an
internal reference in all experiments.
Fig. 3a shows that oxidation of the Ru reference complex
occurs at a potential of 0.97 V vs. Fc+/Fc in CH2Cl2 (solid
trace), while Ru(tBu) is oxidized already at 0.84 V vs. Fc+/Fc
under the same conditions (dashed trace). The electron-donating
tert-butyl substituents therefore facilitate metal oxidation by
c
This journal is the Owner Societies 2012
Phys. Chem. Chem. Phys., 2012, 14, 2685–2692 2687

View Article Online
(Ru(tBu)) reveals that reduction processes associated with the
ruthenium complexes occur at a similar potential, indeed, these are
the well known ligand-based reductions of Ru(a-diimine)32+-type
complexes.57 Thus, in the voltammograms of the dyads
(Fig. 3b, c, e and f) at potentials below ꢀ1.2 V vs. Fc+/Fc
there are overlapping waves due to bpy-based and AQ-localized
reductions. However, the shapes of the most prominent
reduction waves at ꢀ1.3 V vs. Fc+/Fc, particularly in CH3CN,
are reminiscent of the voltammogram reported for isolated 9,10-
anthraquinone and are therefore attributed to an AQ-localized
reduction process.14 All E(AQ0/ꢀ)-values are summarized in the
fourth column of Table 1. Nearly the same reduction potential is
found for all four dyads shown in Scheme 1, irrespective of
whether CH2Cl2 or CH3CN is used as a solvent.
Thermodynamics of photoinduced electron transfer
Fig. 3 Cyclic voltammograms for the six molecules shown in
Scheme 1 in dichloromethane (left) and acetonitrile (right): (a, d) Ru
(solid trace) and Ru(tBu) (dotted trace); (b, e) Ru-xy-AQ (solid) and
Ru(tBu)-xy-AQ (dotted); (c, f) Ru-dmb-AQ (solid) and Ru(tBu)-dmb-AQ
(dotted).
Based on the electrochemical data from the preceding paragraph,
we are now in a position to estimate the driving force for electron
transfer (DGET) between photoexcited ruthenium and anthra-
quinone in the four dyads shown in Scheme 1. Eqn (1) is
commonly used for the purpose of estimating driving forces
for photoinduced electron transfer.1,37,68
130 mV, an effect which is well known and which can also be
observed in the p-xylene (Fig. 3b) and p-dimethoxybenzene
(Fig. 3c) bridged dyads.57 Likewise, the same effect is observed
in acetonitrile for all molecules, albeit in somewhat attenuated
form. For convenience, all electrochemical potentials for
ruthenium oxidation (E(RuII/III)) extracted from Fig. 3 have
been summarized in the third column of Table 1. The data
show that the Ru unit with unsubstituted ancillary bpy ligands
is easier to oxidize by B100 mV in acetonitrile compared to
dichloromethane, while in the Ru(tBu)-based systems the
difference between solvents only amounts to roughly 50 mV.
For estimation of the driving force for photoinduced electron
transfer from ruthenium to anthraquinone, the electrochemical
potential for one-electron reduction of the anthraquinone
moiety (E(AQ0/ꢀ)) is particularly relevant. From literature data,
we expect this reduction process to occur near ꢀ1.3 V vs. Fc+/Fc,
and we indeed observe prominent reduction waves at this
potential in Fig. 3b, c, e and f. Inspection of the voltammograms
of the pure reference complexes in Fig. 3a (Ru) and Fig. 3d
DGET = eꢁ(Eox ꢀEred) ꢀ E00 ꢀ e2/(4pe0esRDA
)
(1)
In eqn (1), Eox and Ered are the electrochemical potentials for
ruthenium oxidation (E(RuII/III)) and anthraquinone
reduction (E(AQ0/ꢀ)), respectively, while E00 is the energy of
3
the photoactive MLCT state of the ruthenium complex. e0 is
the vacuum permittivity, and es stands for the dielectric
constant of the solvent (dichloromethane: 8.93, acetonitrile:
35.94).69 r represents the average Born radius of the two redox
partners (assumed to be 4.5 A), and RDA is the center-to-center
donor–acceptor distance (13.3 A). Using the redox potentials
from Table 1 and a 3MLCT energy of 2.12 eV for the
unsubstituted Ru complex, we find that for Ru-xy-AQ and
Ru-dmb-AQ photoinduced electron transfer is endergonic by
0.09 eV/0.10 eV in dichloromethane, while in acetonitrile
DGET = 0.04 eV/0.02 eV for the same two dyads (last column
of Table 1). As expected, photoinduced electron transfer is less
endergonic in the more polar CH3CN solvent, partly due to
the fact that ruthenium oxidation occurs more readily under
these conditions (see the third column of Table 1 and the
preceding paragraph).
Table 1 Electrochemical potentials (E) for oxidation of Ru(II) to
Ru(III) and for reduction of AQ0 to AQꢀ in volts versus Fc+/Fc; free
energy (DGET) associated with photoinduced electron transfer from
ruthenium to anthraquinone (calculated with eqn (1) and the parameters
given in the text)
The differences in DGET between solvents are likely responsible
for the discrepancies in luminescence behavior observed for the
Ru-xy-AQ/Ru-dmb-AQ dyads in dichloromethane (Fig. 2a) and
acetonitrile (Fig. 2b). In CH2Cl2, DGET E 0.1 eV and hence
excited-state deactivation by electron transfer is inefficient, and
the Ru, Ru-xy-AQ, and Ru-dmb-AQ luminescence intensities are
all about equal (Fig. 2a). In CH3CN, DGET o 0.1 eV, leading to
noticeable luminescence quenching in the dyads with respect to
the Ru reference complex as a consequence of intramolecular
photoinduced electron transfer. It appears plausible that the
normalized luminescence intensities in this case (Fig. 2b) reflect
the relative electron transfer efficiencies, but this aspect will be
further examined below.
Compound
Solvent
E(RuII/III
)
E(AQ0/ꢀ
)
DGET/eV
Ru
Ru
CH2Cl2
CH3CN
CH2Cl2
CH3CN
CH2Cl2
CH3CN
CH2Cl2
CH3CN
CH2Cl2
CH3CN
CH2Cl2
CH3CN
0.97
0.88
0.98
0.90
1.00
0.90
0.84
0.80
0.83
0.79
0.83
0.79
Ru-xy-AQ
Ru-xy-AQ
Ru-dmb-AQ
Ru-dmb-AQ
Ru(tBu)
ꢀ1.35
ꢀ1.29
ꢀ1.34
ꢀ1.27
0.09
0.04
0.10
0.02
Ru(tBu)
Ru(tBu)-xy-AQ
Ru(tBu)-xy-AQ
Ru(tBu)-dmb-AQ
Ru(tBu)-dmb-AQ
ꢀ1.33
ꢀ1.29
ꢀ1.33
ꢀ1.29
ꢀ0.12
ꢀ0.11
ꢀ0.12
ꢀ0.11
As far as the dyads with the Ru(tBu) photosensitizer are
concerned, the change in solvent from CH2Cl2 to CH3CN has
c
2688 Phys. Chem. Chem. Phys., 2012, 14, 2685–2692
This journal is the Owner Societies 2012

View Article Online
qualitatively the same influence as on the dyads with the
ordinary Ru complex without tert-butyl groups: photoinduced
electron transfer in Ru(tBu)-xy-AQ/Ru(tBu)-dmb-AQ becomes
more favorable in acetonitrile (Table 1, fifth column). Excited-state
quenching by electron transfer is efficient already in CH2Cl2, hence
the emission of Ru(tBu)-xy-AQ and Ru(tBu)-dmb-AQ is weaker
than that of Ru(tBu) already in dichloromethane (Fig. 2c) and not
only in acetonitrile (Fig. 2d).31
Kinetics for photoinduced electron transfer
With the thermodynamics for photoinduced electron transfer in
all four dyads at hand, we may now turn to the kinetics of this
excited-state deactivation process. Time-resolved luminescence
and transient absorption spectroscopy are among the two most
widely used experimental techniques for exploration of the
dynamics of electron transfers occurring from excited states.
In the dyads shown in Scheme 1, photoinduced electron transfer
is expected to produce an oxidized ruthenium complex and a
reduced anthraquinone moiety, both of which have clearly
identifiable absorptions in the visible spectral range. However,
all our attempts to detect oxidized ruthenium (e.g., by a bleach
in transient absorption at around 450 nm) or reduced anthra-
quinone (e.g., by searching positive transient absorption signals
at 390 nm or 550 nm)17,70 by nanosecond transient absorption
spectroscopy have failed, presumably due to rapid disappearance
of the respective photoproducts by thermal electron transfer in
the opposite sense. Indeed, this is not an uncommon problem
in investigations of photoinduced electron transfer, including
prior studies of ruthenium–benzoquinone dyads.71 Thus, our
kinetic investigations are limited to time-resolved luminescence
experiments. We recall that excited-state quenching by triplet–triplet
energy transfer could be ruled out based on energetic grounds
(see above).
Fig. 4 Luminescence decays of the molecules shown in Scheme 1 in
CH2Cl2 (a, c) and CH3CN (b, d) measured after pulsed excitation
at 407 nm or 450 nm. Labels are as follows: 1: Ru, 2: Ru-xy-AQ, 3:
Ru-dmb-AQ, 4: Ru(tBu), 5: Ru(tBu)-xy-AQ, 6: Ru(tBu)-dmb-AQ.
Table 2 Luminescence lifetimes (t) and excited-state quenching rate
constants (kQ) in deoxygenated solution. Excitation occurred at
407 nm or 450 nm, detection was at 600 nm. kQ-values were calculated
using eqn (2) and reflect rate constants for photoinduced ruthenium-
to-anthraquinone electron transfer. 1: Ru, 2: Ru-xy-AQ, 3: Ru-dmb-AQ,
4: Ru(tBu), 5: Ru(tBu)-xy-AQ, 6: Ru(tBu)-dmb-AQ
t/ns t/ns t/ns
kQ/sꢀ1
2
kQ/sꢀ1
3
Solvent
1
2
3
CH2Cl2 535 696 1865 o1.4 ꢃ 106
o 5.4 ꢃ 105
CH3CN 866 300 124
(2.2 ꢂ 0.2) ꢃ 106 (6.9 ꢂ 0.5) ꢃ 106
t/ns t/ns
t/ns
kQ/sꢀ1
kQ/sꢀ1
Solvent
4
5
6
5
6
CH2Cl2 1231 1394
CH3CN 947 47
1510
14
o7.2 ꢃ 105
o 6.6 ꢃ 105
(2.0 ꢂ 0.1) ꢃ 107 (7.0 ꢂ 0.4) ꢃ 107
Fig. 4 shows the luminescence decays of the six molecules
shown in Scheme 1 detected at 600 nm after pulsed excitation
at 407 nm or at 450 nm. The left panels (a, c) show data
measured in dichloromethane, while the right panels (b, d)
exhibit lifetimes determined in acetonitrile. With the exception
of Ru(tBu)-xy-AQ and Ru(tBu)-dmb-AQ in acetonitrile
(Fig. 4d), all emission decays are single exponential, and fits
to the experimental data in Fig. 4 yield the lifetime values (t)
given in Table 2. Inspection of panels (a) and (c) in Fig. 4
reveals that all six molecules shown in Scheme 1 exhibit
luminescence lifetimes longer than 500 ns in deoxygenated
dichloromethane solution. For the molecules based on the Ru
photosensitizer, the relative magnitude of the emission life-
times reflects the relative luminescence intensities in Fig. 2a:
Ru has the shortest lifetime (535 ns) and the weakest intensity,
while Ru-dmb-AQ has the longest lifetime (1865 ns) and
strongest intensity; Ru-xy-AQ is between the two, both
regarding the lifetime (696 ns) and emission intensity. In
acetonitrile, the strongest emitter (Fig. 2b) is the isolated
Ru complex exhibiting a luminescence lifetime of 866 ns,
Ru-dmb-AQ shows the weakest emission and shortest lifetime
(124 ns), while Ru-xy-AQ is in between both regarding the
intensity and lifetime (300 ns). The changes in emission
intensity and luminescence lifetime upon solvent replacement
from CH2Cl2 to CH3CN are both consistent with an increased
(50%) (67%)
950 950
(50%) (33%)
efficiency of photoinduced ruthenium-to-anthraquinone electron
transfer as a result of increased driving-force (DGET, Table 1).
In the absence of transient absorption data which could
provide direct kinetic information regarding the build-up of
electron transfer photoproducts, it is common to use eqn (2) to
determine excited-state quenching rate constants (kQ).1,66
ꢀ1
ꢀ1
kQ = tdyad ꢀ tref
(2)
In eqn (2), tdyad is the lifetime of the ruthenium–bridge–
anthraquinone dyad, and tref is the luminescence lifetime of
a suitable reference complex. It is customary to use complexes
such as Ru and Ru(tBu) as reference compounds. For the
Ru-xy-AQ and Ru-dmb-AQ dyads this procedure results in
kQ-values of (2.2 ꢂ 0.2) ꢃ 106 sꢀ1 and (6.9 ꢂ 0.5) ꢃ 106 sꢀ1 in
acetonitrile. For dichloromethane solution, this procedure is
not applicable because there is no significant luminescence
quenching or lifetime shortening in these two particular dyads
with respect to the reference complex. Therefore, instead of
applying eqn (2) to the dichloromethane data, we simply
c
This journal is the Owner Societies 2012
Phys. Chem. Chem. Phys., 2012, 14, 2685–2692 2689

View Article Online
estimate an upper limit for kQ of Ru-xy-AQ and Ru-dmb-AQ
in CH2Cl2 in the following way: in the (unrealistic) limiting
case in which excited-state deactivation would occur entirely
by Ru-to-AQ electron transfer, kQ = tdyadꢀ1. In the presence of
other excited-state deactivation pathways (e.g., luminescence,
multiphonon relaxation), kQ o tdyadꢀ1, hence by using the
approximation kQ E tdyadꢀ1 one can estimate the upper limit of
kQ for Ru-xy-AQ and Ru-dmb-AQ in CH2Cl2 even without
having a suitable reference point at disposition.
the luminescence lifetime of the Ru reference complex under
identical conditions. This observation, together with the
finding of (slightly) increasing emission intensities along the
series: Ru o Ru-xy-AQ o Ru-dmb-AQ in CH2Cl2 (Fig. 2a),
indicates that nonradiative relaxation becomes less efficient
along this series of compounds. The origin of this effect is not
understood.
Summary and conclusions
In the dyads with the Ru(tBu) photosensitizer, the time-
resolved luminescence data in Fig. 4c provide no evidence for
significant excited-state quenching in dichloromethane, and
one is forced to conclude that the electron transfer rate
constant is below 7.2 ꢃ 105 sꢀ1 for Ru(tBu)-xy-AQ and below
6.6 ꢃ 105 sꢀ1 in Ru(tBu)-dmb-AQ in CH2Cl2. In deoxygenated
acetonitrile (Fig. 4d), t = 947 ns for Ru(tBu), while the
Ru(tBu)-xy-AQ and Ru(tBu)-dmb-AQ dyads both exhibit
biexponential decays, with a fast decay component yielding
lifetimes of 47 ns (xy system) and 14 ns (dmb system),
respectively, and a second decay component which is markedly
slower and which can be fitted to a lifetime of B950 ns in both
cases. This slower component is attributed to emission of a Ru(tBu)
impurity present in the Ru(tBu)-xy-AQ and Ru(tBu)-dmb-AQ
samples. By standard analytical methods (NMR, elemental
analysis) this impurity is not detectable (see the ESIw), but
given the fast luminescence decay of Ru(tBu)-xy-AQ (47 ns)
and Ru(tBu)-dmb-AQ (14 ns) in acetonitrile (Table 2), even
minute amounts of comparatively strongly emissive impurities
now become noticeable. The same effect may explain the
observation of nearly identical emission intensities for the
Ru(tBu)-based dyads in CH2Cl2 (Fig. 2c) and CH3CN
(Fig. 2d). Given the much shorter lifetimes in acetonitrile than
in dichloromethane, the inherent dyad emission intensities
are expected to be significantly lower in CH3CN, and it is
likely that a Ru(tBu) impurity contributes substantially to the
dyad emissions in Fig. 2d. Be that as it may, given the long
Ru(tBu)-xy-AQ/Ru(tBu)-dmb-AQ lifetimes in CH2Cl2 compared
to the Ru(tBu) reference complex, application of eqn (2) makes
sense only for acetonitrile solvent in which substantial lifetime
shortening in the dyads is observed. In this case, rate constants
of (2.0 ꢂ 0.1) ꢃ 107 sꢀ1 and (7.0 ꢂ 0.4) ꢃ 107 sꢀ1 are obtained for
the xylene- and dimethoxybenzene-bridged systems, respectively
(Table 2).
The key findings from this work are summarized in Scheme 3.
From the kQ-values in Table 2 we learn that the change in
solvent from CH2Cl2 (Scheme 3a) to CH3CN (Scheme 3b)
increases charge transfer rates to the extent that electron
transfer becomes a competitive excited-state deactivation pathway.
Introduction of electron-donating tert-butyl substituents in the
bpy ancillary ligands and the associated increase in ꢀDGET by
B0.2 eV (Table 1) entail an acceleration by an order of magnitude
(e.g., from 2.2 ꢃ 106 sꢀ1 for Ru-xy-AQ to 2.0 ꢃ 107 sꢀ1 for
Ru(tBu)-xy-AQ, Scheme 3c). Finally, replacement of the p-xylene
spacer by a p-dimethoxybenzene unit leads to an increase in
electron transfer rates by a factor of 3.5 (e.g., from 2.0 ꢃ 107 sꢀ1
in Ru(tBu)-xy-AQ to 7.0 ꢃ 107 sꢀ1 in Ru(tBu)-dmb-AQ,
Scheme 3d). Thus, the relative importance of the three
effects decreases along the series: solvent polarity Z tert-butyl
substitution > bridge modulation.
The influence of the molecular bridge on electron transfer
rates is relatively small in the present case. The p-dimethoxy-
benzene molecule is oxidized much more easily than p-xylene, and
therefore it is prone to mediating charge transfer predominantly
via a hole transfer mechanism, as previously demonstrated for
3+
hole transfers originating from photogenerated Ru(bpy)3
complexes.72,73 In the present case, we are dealing with
excited-state electron transfer from ruthenium(II) to anthra-
quinone, or equivalently, hole transfer from anthraquinone to
photoexcited ruthenium(II). Based on the electrochemical
potential for one-electron oxidation of p-dimethoxybenzene
(1.1 V vs. Fc+/Fc)74,75 and the reduction potential of anthra-
quinone (ꢀ1.3 V vs. Fc+/Fc, Table 2), hole injection from
anthraquinone to p-dimethoxybenzene is energetically uphill
by 2.4 eV. In the case of the p-xylene dyads, the respective
injection barrier amounts to more than 3.0 eV, based on an
oxidation potential > 1.7 V vs. Fc+/Fc for p-xylene.74 Given the
fact that the donor–bridge energy gap decreases by only about
20% between the xylene- and dimethoxybenzene-bridged dyads,
Curiously, the emission lifetimes of the Ru-xy-AQ and
Ru-dmb-AQ dyads in CH2Cl2 are significantly longer than
Scheme 3 Energetics and kinetics of photoinduced electron transfer as a function of solvent, driving-force, and bridge modulation.
c
2690 Phys. Chem. Chem. Phys., 2012, 14, 2685–2692
This journal is the Owner Societies 2012

View Article Online
the experimentally observed acceleration of charge transfer
rates by a factor of 3.5 is remarkable.
26 B. Albinsson, M. P. Eng, K. Pettersson and M. U. Winters, Phys.
Chem. Chem. Phys., 2007, 9, 5847.
27 B. Albinsson and J. Martensson, J. Photochem. Photobiol., C,
2008, 9, 138.
28 W. R. Browne, N. M. O’Boyle, J. J. McGarvey and J. G. Vos,
Chem. Soc. Rev., 2005, 34, 641.
29 F. Scandola, M. T. Indelli, C. Chiorboli and C. A. Bignozzi, Top.
Curr. Chem., 1990, 158, 73.
30 A. C. Benniston and A. Harriman, Chem. Soc. Rev., 2006, 35,
169.
31 H. A. Meylemans, C. F. Lei and N. H. Damrauer, Inorg. Chem.,
2008, 47, 4060.
32 H. M. McConnell, J. Chem. Phys., 1961, 35, 508.
33 M. N. Paddon-Row, Acc. Chem. Res., 1994, 27, 18.
34 M. N. Paddon-Row, A. M. Oliver, J. M. Warman, K. J. Smit,
M. P. Dehaas, H. Oevering and J. W. Verhoeven, J. Phys. Chem.,
1988, 92, 6958.
The bottom line is that in our dyads a 0.2 eV increase in
driving-force has a stronger influence on charge transfer rates
than the 0.6 eV decrease in the donor–bridge energy gap
associated with the replacement of a p-xylene spacer by a
p-dimethoxybenzene unit. However, this is because the
absolute magnitude of the barrier for hole injection into the
bridge is large for the particular systems investigated here. It is
conceivable that for other donor–bridge–acceptor combina-
tions with inherently smaller injection barriers, variations in
donor–bridge energy gaps outweigh the effect of small driving-
force changes—even in the tunneling regime, before hopping
processes become energetically viable.
35 M. P. Eng and B. Albinsson, Angew. Chem., Int. Ed., 2006,
45, 5626.
36 M. P. Eng and B. Albinsson, Chem. Phys., 2009, 357, 132.
37 K. Kilsa, J. Kajanus, A. N. Macpherson, J. Martensson and
B. Albinsson, J. Am. Chem. Soc., 2001, 123, 3069.
38 J. Wiberg, L. J. Guo, K. Pettersson, D. Nilsson, T. Ljungdahl,
J. Martensson and B. Albinsson, J. Am. Chem. Soc., 2007,
129, 155.
39 K. Pettersson, J. Wiberg, T. Ljungdahl, J. Martensson and
B. Albinsson, J. Phys. Chem. A, 2006, 110, 319.
40 O. S. Wenger, Acc. Chem. Res., 2011, 44, 25.
41 O. S. Wenger, Chem. Soc. Rev., 2011, 40, 3538.
42 S. E. Miller, A. S. Lukas, E. Marsh, P. Bushard and
M. R. Wasielewski, J. Am. Chem. Soc., 2000, 122, 7802.
Acknowledgements
This work was supported by the Swiss National Science
Foundation through grant number 200021-117578 and the
Deutsche Forschungsgemeinschaft through grant number
INST 186/872-1.
Notes and references
1 V. Balzani, Electron transfer in chemistry, VCH Wiley, Weinheim,
2001.
43 C. Lambert, G. Noll and J. Schelter, Nat. Mater., 2002, 1, 69.
¨
44 D. Hanss, M. E. Walther and O. S. Wenger, Chem. Commun.,
2010, 46, 7034.
45 M. E. Walther and O. S. Wenger, ChemPhysChem, 2009, 10,
1203.
46 J. Hankache and O. S. Wenger, Chem. Rev., 2011, 111, 5138.
47 N. Belfrekh, C. Dietrich-Buchecker and J.-P. Sauvage, Tetrahedron
Lett., 2001, 42, 2779.
2 A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and
A. Von Zelewsky, Coord. Chem. Rev., 1988, 84, 85.
3 O. S. Wenger, Coord. Chem. Rev., 2009, 253, 1439.
4 C. J. da Cunha, E. S. Dodsworth, M. A. Monteiro and A. B.
P. Lever, Inorg. Chem., 1999, 38, 5399.
5 R. Frank, G. Greiner and H. Rau, Phys. Chem. Chem. Phys., 1999,
1, 3481.
48 V. Hensel and A. D. Schluter, Liebigs Ann., 1997, 303.
¨
49 V. Hensel, K. Lutzow, J. Jacob, K. Gessler, W. Saenger and
6 R. Frank and H. Rau, Helv. Chim. Acta, 2001, 84, 3837.
7 D. G. McCafferty, D. A. Friesen, E. Danielson, C. G. Wall,
M. J. Saderholm, B. W. Erickson and T. J. Meyer, Proc. Natl.
Acad. Sci. U. S. A., 1996, 93, 8200.
8 D. R. Striplin, S. Y. Reece, D. G. McCafferty, C. G. Wall,
D. A. Friesen, B. W. Erickson and T. J. Meyer, J. Am. Chem.
Soc., 2004, 126, 5282.
A. D. Schluter, Angew. Chem., Int. Ed., 1997, 36, 2654.
¨
50 M. E. Walther and O. S. Wenger, Inorg. Chem., 2011, 50,
10901.
51 D. Hanss, J. C. Freys, G. Bernardinelli and O. S. Wenger, Eur. J.
Inorg. Chem., 2009, 4850.
52 A. Bouillon, A. S. Voisin, A. Robic, J. C. Lancelot, V. Collot and
S. Rault, J. Org. Chem., 2003, 68, 10178.
53 M. S. Jensen, R. S. Hoerrner, W. J. Li, D. P. Nelson, G. J. Javadi,
P. G. Dormer, D. W. Cai and R. D. Larsen, J. Org. Chem., 2005,
70, 6034.
9 F. Vogtle, M. Plevoets, M. Nieger, G. C. Azzellini, A. Credi,
¨
L. De Cola, V. De Marchis, M. Venturi and V. Balzani, J. Am.
Chem. Soc., 1999, 121, 6290.
10 K. Sumi, M. Furue and S. Nozakura, Photochem. Photobiol., 1985,
42, 485.
54 D. Hanss and O. S. Wenger, Inorg. Chem., 2008, 47, 9081.
55 D. Hanss and O. S. Wenger, Inorg. Chem., 2009, 48, 671.
56 B. P. Sullivan, D. J. Salmon and T. J. Meyer, Inorg. Chem., 1978,
17, 3334.
57 D. M. Roundhill, Photochemistry and Photophysics of Metal
Complexes, Plenum Press, New York, 1994.
58 K. S. Schanze, D. B. MacQueen, T. A. Perkins and L. A. Cabana,
Coord. Chem. Rev., 1993, 122, 63.
59 K. S. Schanze, L. A. Lucia, M. Cooper, K. A. Walters, H. F. Ji and
O. Sabina, J. Phys. Chem. A, 1998, 102, 5577.
11 J. Otsuki, H. Ogawa, N. Okuda, K. Araki and M. Seno, Bull.
Chem. Soc. Jpn., 1997, 70, 2077.
12 L. Mishra, C. S. Choi and K. Araki, Chem. Lett., 1997, 447.
13 Y. Pellegrin, R. J. Forster and T. E. Keyes, Inorg. Chim. Acta,
2009, 362, 1715.
14 A. Babaei, P. A. Connor, A. J. McQuillan and S. Umapathy,
J. Chem. Educ., 1997, 74, 1200.
15 S. L. Mecklenburg, D. G. McCafferty, J. R. Schoonover,
B. M. Peek, B. W. Erickson and T. J. Meyer, Inorg. Chem.,
1994, 33, 2974.
60 M. E. Walther and O. S. Wenger, Dalton Trans., 2008, 6311.
61 J. C. Freys and O. S. Wenger, Eur. J. Inorg. Chem., 2010, 5509.
62 S. Boyde, G. F. Strouse, W. E. Jones and T. J. Meyer, J. Am.
Chem. Soc., 1989, 111, 7448.
63 G. J. Wilson, A. Launikonis, W. H. F. Sasse and A. W. H. Mau,
J. Phys. Chem. A, 1997, 101, 4860.
64 M. D. Ward, Coord. Chem. Rev., 2006, 250, 3128.
65 R. T. F. Jukes, V. Adamo, F. Hartl, P. Belser and L. De Cola,
Coord. Chem. Rev., 2005, 249, 1327.
16 K. A. Opperman, S. L. Mecklenburg and T. J. Meyer, Inorg.
Chem., 1994, 33, 5295.
17 J. Hankache and O. S. Wenger, Chem. Commun., 2011, 47, 10145.
18 M. R. Wasielewski, Chem. Rev., 1992, 92, 435.
19 P. Y. Chen and T. J. Meyer, Chem. Rev., 1998, 98, 1439.
20 P. F. Barbara, T. J. Meyer and M. A. Ratner, J. Phys. Chem., 1996,
100, 13148.
21 H. B. Gray and J. R. Winkler, Proc. Natl. Acad. Sci. U. S. A., 2005,
102, 3534.
22 H. B. Gray and J. R. Winkler, Annu. Rev. Biochem., 1996, 65,
537.
66 B. Schlicke, P. Belser, L. De Cola, E. Sabbioni and V. Balzani,
J. Am. Chem. Soc., 1999, 121, 4207.
23 J. R. Winkler and H. B. Gray, Chem. Rev., 1992, 92, 369.
24 B. Giese, Annu. Rev. Biochem., 2002, 71, 51.
25 M. Cordes and B. Giese, Chem. Soc. Rev., 2009, 38, 892.
67 N. J. Turro, Molecular Photochemistry, New York, Amsterdam,
1967.
68 A. Weller, Z. Phys. Chem. (Leipzig), 1982, 133, 93.
c
This journal is the Owner Societies 2012
Phys. Chem. Chem. Phys., 2012, 14, 2685–2692 2691

View Article Online
69 G. G. Gurzadyan and S. Steenken, Chem.–Eur. J., 2001, 7,
1808.
70 F. D. Lewis, A. K. Thazhathveetil, T. A. Zeidan, J. Vura-Weis and
M. R. Wasielewski, J. Am. Chem. Soc., 2010, 132, 444.
72 O. S. Wenger, Inorg. Chim. Acta, 2011, 374, 3.
73 D. Hanss, M. E. Walther and O. S. Wenger, Coord. Chem. Rev.,
2010, 254, 2584.
74 O. Nicolet and E. Vauthey, J. Phys. Chem. A, 2002, 106,
5553–5562.
71 M. Borgstrom, O. Johansson, R. Lomoth, H. B. Baudin, S. Wallin,
¨
¨
L. C. Sun, B. Akermark and L. Hammarstrom, Inorg. Chem., 2003,
42, 5173.
75 M. A. Haga, E. S. Dodsworth, G. Eryavec, P. Seymour and
A. B. P. Lever, Inorg. Chem., 1985, 24, 1901.
c
2692 Phys. Chem. Chem. Phys., 2012, 14, 2685–2692
This journal is the Owner Societies 2012