Exploiting conformational dynamics to facilitate formation and trapping of electron-transfer photoproducts in metal complexes

Article abstract of DOI:10.1021/ja1055559

Three new photoinduced electron donor-acceptor (D-A) systems are reported which juxtapose a Ru(II) excited-state donor with a bipyridinium acceptor via a conformationally active asymmetric aryl-substituted bipyridine ligand participating in the bridge between D and A. Across the series of complexes 1-3, steric bulk is sequentially added to tune the inter-ring dihedral angle θ between the bipyridine and the aryl substituent. Driving forces for photoinduced electron transfer (ΔG_ET) and back electron transfer (ΔG_BET) are reported based on electrochemical measurements of 1-3 as well as Franck-Condon analysis of emission spectra collected for three new donor model complexes 1′-3′. These preserve the substitution patterns on the aryl substituent in their respective D-A complexes but remove the bipyridinium acceptor. Both ΔG_ET and ΔG_BET are invariant to within 0.02 eV across the series. Upon visible photoexcitation of each of the D-A systems with ～100 fs laser pulses at 500 ± 10 nm, an electron-transfer (ET) photoproduct is observed to form with a time constant of τ_ET = 29 ps (1), 37 ps (2), and 57 ps (3). That ET remains relatively rapid throughout this series, even as steric bulk significantly increases the inter-ring dihedral angle θ, is attributed to the effects of ligand-based torsional dynamics driven by intraligand electron delocalization in the D-A excited state manifold prior to ET. The lifetimes of the charge-separated states (τ_BET) are also reported with τ_BET = 98 ps (1), 217 ps (2), and 789 ps (3), representing a more than 8-fold increase across the series. This is attributed to reverse conformational dynamics in D⁺-A^- driven by steric repulsions, which serves to minimize electronic coupling to the ground state. Steric control of ligand geometry and the range over which θ changes during conformational dynamics provides a new strategy to facilitate the formation and storage of charge-separated excited states.

Full text of DOI:10.1021/ja1055559

Published on Web 08/04/2010
Exploiting Conformational Dynamics To Facilitate Formation and Trapping of
Electron-Transfer Photoproducts in Metal Complexes
Heather A. Meylemans, Joshua T. Hewitt, Mirvat Abdelhaq, Paul J. Vallett, and Niels H. Damrauer*
Department of Chemistry and Biochemistry, UniVersity of Colorado at Boulder, Boulder, Colorado 80309
Received June 24, 2010; E-mail: niels.damrauer@colorado.edu
transfer, and conduction in molecular systems4-12 (see ref 3 for a
more extensive bibliography).
Abstract: Three new photoinduced electron donor-acceptor
(D-A) systems are reported which juxtapose a Ru(II) excited-
state donor with a bipyridinium acceptor via a conformationally
active asymmetric aryl-substituted bipyridine ligand participating
in the bridge between D and A. Across the series of complexes
1-3, steric bulk is sequentially added to tune the inter-ring
dihedral angle θ between the bipyridine and the aryl substituent.
Driving forces for photoinduced electron transfer (∆GET) and back
electron transfer (∆GBET) are reported based on electrochemical
measurements of 1-3 as well as Franck-Condon analysis of
emission spectra collected for three new donor model complexes
1′-3′. These preserve the substitution patterns on the aryl
Figure 1. (Left) D-A and respective D complexes with varying steric
substituent in their respective D–A complexes but remove the
bipyridinium acceptor. Both ∆GET and ∆GBET are invariant to within
bulk on the conformationally active aryl substituent (θ indicated in red).
(Right) Representative transient absorption kinetics collected for 1 (green),
0.02 eV across the series. Upon visible photoexcitation of each
of the D-A systems with ∼100 fs laser pulses at 500 ( 10 nm,
an electron-transfer (ET) photoproduct is observed to form with
a time constant of τET ) 29 ps (1), 37 ps (2), and 57 ps (3). That
ET remains relatively rapid throughout this series, even as steric
bulk significantly increases the inter-ring dihedral angle θ, is
attributed to the effects of ligand-based torsional dynamics driven
by intraligand electron delocalization in the D*-A excited state
manifold prior to ET. The lifetimes of the charge-separated states
(τBET) are also reported with τBET ) 98 ps (1), 217 ps (2), and
789 ps (3), representing a more than 8-fold increase across the
series. This is attributed to reverse conformational dynamics in
D+-A- driven by steric repulsions, which serves to minimize
electronic coupling to the ground state. Steric control of ligand
geometry and the range over which θ changes during conforma-
tional dynamics provides a new strategy to facilitate the formation
and storage of charge-separated excited states.
2 (purple), and 3 (black) in acetonitrile at 294 K at λprobe ) 607 nm following
∼100 fs pulsed excitation at 500 nm. Solid lines are fits of the data to a
biexponential model (rise and decay). (Right inset) Transient and fit for 3
expanded to 5 ns.
Of recent interest in our group are systems (see Figure 1 for
species explored herein) that juxtapose a Ru(II) excited-state donor
with a bipyridinium acceptor via a conformationally active asym-
metric aryl-substituted bipyridine ligand serving as the bridge.
Diquaternary ammonium acceptors are commonly used for both
bimolecular and unimolecular ET studies with d6 transition metal
donor complexes (e.g., refs 1, 13, and 14). In our systems metal-
to-ligand charge-transfer (MLCT) states populating the π* system
of the asymmetric ligand are expected to initiate ligand-based
molecular geometry changes along torsional nuclear coordinates,
i.e., planarization of θ (we consider the absolute value of this
quantity θ between 0° and 90°) driven by excited-state intraligand
electron delocalization.2,15 Such motions can facilitate ET if they
(a) lower the reorganization energy in the D*-A f D+-A-
reaction, as we have previously argued to be important,2 and/or
(b) change the electronic coupling between these states, both by
varying the distance of ET and by tuning superexchange through
orbital overlap changes.3 Opportunities to hinder BET arise if
reverse motions, driven by steric effects, decrease electronic
coupling between D+-A- and D-A states.
Recent computational work from our group explored structure
and excited-state conformational activity of these systems.3 In DFT
models of 1, 2, and 3, θ ) 35°, 50°, and 86°, respectively, for
ground-state and, by inference, the Franck-Condon state geom-
etries. The vibrationally relaxed 3MLCT states were also modeled,
leading to the finding that each species has a more planarized aryl
substituent relative to the bipyridine in the excited state prior to
ET by 13° (1), 15° (2), and 31° (3). While we expect kET to decrease
across the series as electronic communication between D and A
through the π* system is diminished (as θ increases), the planarizing
torsional motion can limit the effects and ideally permit facile ET
in each of the systems.
Electron transfer (ET) research exploiting structural manipula-
tions of donor-acceptor (D-A) systems has expanded our funda-
mental understanding of reactivity between electronic states and
has guided strategies for solar energy conversion.1 We have been
interested in developing D-A design principles wherein excited-
state conformational dynamics on low-frequency nuclear modes that
are nominally orthogonal to high-frequency ET reaction coordinates
are exploited to both facilitate charge separation (ET) and hinder
energy wasting recombination reactions (back electron transfer,
BET).2,3 Ideal systems would avoid relying on driving force and
ET distance to create and store charge-separated states, thus
circumventing a key source of inefficiency common to both
synthetic and natural solar energy conversion strategies. Our
approach exploits the general concept that thermally accessible
nuclear motions can impact intercomponent electronic couplings
important for determining rates of ET, exchange-mediated energy
9
11464 J. AM. CHEM. SOC. 2010, 132, 11464–11466
10.1021/ja1055559  2010 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Measured ET Properties in Acetonitrile
A second important conclusion from computational models is
that complete or nearly complete reVerse switching of θ is expected
following the formation of the ET photoproduct D+-A-. An
optimized geometry where θ ) 33° was previously reported for
the lowest energy triplet state (namely, D+-A-) of 1.3 Similar
calculations (B3LYP functional with PCM acetonitrile continuum;
basis set identical to ref 3)16 were carried out here for 2 and 3 with
comparable results: for 2, θ ) 47° whereas for 3, θ ) 88°. Thus
θ is expected to be approximately the same for both the ground
state and the ET product for 1-3. A simplistic model for tuning
electronic coupling between reactant (D+-A-) and product (D-A)
states for the BET reaction suggests important consequences of these
reverse switching motions. If coupling is proportional to cos(θ),
as is expected from superexchange treatments of biaryl ET
bridges,3,7,10,17 then kBET would be strongly affected when reverse
switching motions reach larger values of θ such as that predicted
in 3.
1
2
3
∆GET /eV
-0.58
-0.58
-0.59
-1.55
1.75 ( 0.1
57 ( 3
0.127 ( 0.002
789 ( 9
∆GBET /eV
-1.53
-1.55
a
kET /1010 s-1
3.4 ( 0.2
29 ( 2
1.02 ( 0.03
98 ( 3
2.7 ( 0.1
37 ( 2
0.46 ( 0.01
217 ( 4
τET /ps
a
kBET /1010 s-1
τBET /ps
a Rate constants from data collected at 294 K. Error bars represent 2σ
determined from fitting three separate measurements of kinetics
collected at λprobe ) 607 nm.
a rise in the transient absorption intensity with a time constant of
29 ps (τrise), followed by a decay of 98 ps (τdecay). These kinetics
are understood in the context of observations and analysis by our
group on related bis-heteroleptic D-A species where the ancillary
2
ligand (L) is modified to tune ∆GET and ∆GBET
.
For example,
changing L from 2,2′-bipyridine (bpy) to 4,4′,5,5′-tetramethyl-2,2′-
bipyridine (tmb) is accompanied by a modification in ∆GBET from
-1.7 to -1.5 eV and a decrease in τdecay (λprobe ) 607 nm kinetics)
from 160 to 73 ps. This is the expected result for Marcus inverted
region kinetics which allows us to assign τdecay to τBET and, by
inference, τrise to τET. Kinetics collected for 1-3 at 400 nm (also
containing MV+ absorption) and 460 nm (MLCT bleach) behave
as expected.
To measure the effects of geometry and torsion-conformational
dynamics on kET and kBET, we have synthesized compounds 1-3.
The asymmetric electroactive ligand needed for 1 has been reported
by us.2 Those needed for 2 and 3 were synthesized according to
Scheme 1. Donor complexes 1′-3′ were also synthesized which
preserve the methyl substitution pattern on the aryl group but
remove the bipyridinium acceptor. These are used to estimate the
stored 3MLCT excited-state energy prior to ET. As shown in Figures
S1-S3 (Supporting Information), there is excellent agreement
between the absorption spectra of each donor model and the
respective D-A complex, thus justifying their use. 1H NMR
chemical shifts and accurate mass spectrum analyses for 1-3 and
The systematic introduction of steric bulk at the aryl substituent
of the asymmetric ligand has the effect of increasing τET by about
a factor of 2 across the series 1-3 (from 29 ps (1) to 57 ps (3)).
The structural modifications appear responsible, which in and of
itself is an important finding. It is emphasized, however, that the
increase in τET across the series is modest in light of the significant
ground-state structural differences of the bridging ligand with
respect to θ (vide supra). Here we believe that intraligand electron
delocalization2,15 in D*-A prior to ET, the effect which drives
-
1′-3′ as PF6 salts are reported in the Supporting Information.
Scheme 1. Synthetic Strategy for Electroactive Ligands in 2 and 3
3
the predicted decrease in θ in the MLCT states, plays a critical
role, facilitating ET even in the most sterically encumbered system
3. This finding is consistent with calculations of the 3MLCT states
of 1′-3′ showing spin population on the aryl substituent in each
of these systems.3
Importantly, whereas τET for the sterically encumbered 3
increases by less than a factor of 2 relative to 1, τBET increases by
more than a factor of 8 (from 98 ps (1) to 789 ps (3)). This leads
to the dramatic lengthening of the transient absorption decay for 3
versus 1 shown in Figure 1. The ratio between the lifetime of charge
separation (τBET) and the time it takes to achieve it (τET) may be
considered as a figure of merit. Here the structural and dynamical
modifications put in place across the series allow for substantial
tuning of τBET/τET from 3.4 in 1 to 5.9 in 2 to 14 in 3. The very
favorable properties of 3 in terms of this ratio are interpreted to be
a consequence of the range over which excited-state torsional
motions occur. The initial planarization of the dihedral angle by
31°, driven by intraligand electron delocalization, is sufficient to
achieve electronic coupling that permits relatively efficient forward
ET. By contrast, the energy-wasting BET is inefficient. The reverse
switching driven by steric repulsions accesses a dihedral angle (θ
) 88°) where the reduced acceptor (MV+) is significantly decoupled
from the Ru(III) center through the π* system of the bridging ligand.
Notably, with electronic coupling proportional to cos(θ), the range
over which θ changes in 3 permits a significantly stronger contrast
between τBET and τET than in either 1 or 2 (see also discussion in
ref 3). The electronic coupling is of course nonzero, as evidenced
by the finite lifetime τBET ) 789 ps. Through-solvent coupling may
play a role.20 Further, at 294 K a range of dihedral angles will be
thermally accessible, leading to larger electronic coupling. Finally,
The driving forces for ET and BET in 1-3 have been determined
through electrochemical measurements (see Table S1, Supporting
Information) and through emission spectroscopy of the donor
models 1′-3′ (see Figures S1-S3). Franck-Condon analysis18
yields ∆GMLCT as a measure of the stored energy in the long-lived
3MLCT of the donor models and therefore the stored excited-state
energy in 1-3 prior to ET (Table S1). The first oxidation and
reduction potentials measured for 1-3 allow for the determination
of the free energy of formation of the ion-pair state D+-A-
(∆GIP).19 The quantities ∆GIP and ∆GMLCT are used to determine
∆GET () -∆GMLCT + ∆GIP) and ∆GBET () -∆GIP) listed in Table
1. As can be seen, there is little variation in either ∆GET or ∆GBET
for 1-3. This is an ideal situation in efforts to unravel specific
effects of the bridging-ligand geometry and torsional dynamics on
kET and kBET in a structural series of D-A complexes.
Transient absorption spectroscopy with ∼100 fs time resolution
has been utilized to measure ET rate constants for 1-3 in
acetonitrile at 294 K. Here we focus on single-wavelength kinetics
(Figure 1) at λprobe ) 607 nm following short-pulse excitation into
the MLCT band at λpump ) 500 nm. The probe wavelength was
chosen in order to interrogate the well-known visible absorption
of the radical formed when MV2+ is reduced by one electron.2
Reductive spectroelectrochemistry on 1-3 confirms this expectation
(see Figures S4-S6, Supporting Information). Compound 1 shows
9
J. AM. CHEM. SOC. VOL. 132, NO. 33, 2010 11465

C O M M U N I C A T I O N S
ET pathways where orbital symmetry dictates that θ ) 90° is not
the electronic coupling minimum may participate, as has been
observed in bis-porphyrin systems.6 Additional computational
efforts are needed to explore these points in greater detail.
details; complete ref 16. This material is available free of charge via
the Internet at http://pubs.acs.org.
References
(1) Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH: Weinheim,
2001.
(2) Meylemans, H. A.; Lei, C. F.; Damrauer, N. H. Inorg. Chem. 2008, 47,
4060.
(3) Meylemans, H. A.; Damrauer, N. H. Inorg. Chem. 2009, 48, 11161.
(4) Ratner, M. A.; Madhukar, A. Chem. Phys. 1978, 30, 201.
(5) Hoffman, B. M.; Ratner, M. A. J. Am. Chem. Soc. 1987, 109, 6237.
(6) Helms, A.; Heiler, D.; McLendon, G. J. Am. Chem. Soc. 1991, 113, 4325.
(7) Davis, W. B.; Ratner, M. A.; Wasielewski, M. R. J. Am. Chem. Soc. 2001,
123, 7877.
(8) Weiss, E. A.; Tauber, M. J.; Kelley, R. F.; Ahrens, M. J.; Ratner, M. A.;
Wasielewski, M. R. J. Am. Chem. Soc. 2005, 127, 11842.
(9) Laine, P. P.; Bedioui, F.; Loiseau, F.; Chiorboli, C.; Campagna, S. J. Am.
Chem. Soc. 2006, 128, 7510.
(10) Benniston, A. C.; Harriman, A. Chem. Soc. ReV. 2006, 35, 169.
(11) Eng, M. P.; Martensson, J.; Albinsson, B. Chem. Eur. J. 2008, 14, 2819.
(12) Huynh, M. H. V.; Meyer, T. J. Chem. ReV. 2007, 107, 5004.
(13) Yonemoto, E. H.; Saupe, G. B.; Schmehl, R. H.; Hubig, S. M.; Riley, R. L.;
Iverson, B. L.; Mallouk, T. E. J. Am. Chem. Soc. 1994, 116, 4786.
(14) Weber, J. M.; Rawls, M. T.; MacKenzie, V. J.; Limoges, B. R.; Elliott,
C. M. J. Am. Chem. Soc. 2007, 129, 313.
(15) Damrauer, N. H.; Boussie, T. R.; Devenney, M.; McCusker, J. K. J. Am.
Chem. Soc. 1997, 119, 8253.
(16) Frisch, M. J.; et al. Gaussian 09, Revision A.1; Gaussian, Inc.: Wallingford,
CT, 2009.
These results demonstrate that synthetic design principles, guided
by measurement and prediction of excited-state dynamics on low-
frequency vibrational coordinates, can be used to increase the
lifetime of ET photoproducts without greatly reducing their rate of
formation. Researchers interested in charge separation at interfaces
while avoiding recombination reactions s a topic critical to the
development of solar energy conversion strategies s may want to
consider use of what may appear to be an unlikely choice: namely,
π-systems for which conjugation appears restricted due to the
influence of steric repulsions on inter-ring dihedral angles. In our
laboratory, temperature-dependent experiments are underway to
establish values of electronic coupling for the ET and BET reactions
in this series.
Acknowledgment. This work was supported by the National
Science Foundation under CHE-0847216. Computational resources
were due to the NSF through their NCSA TeraGrid.
(17) McConnell, H. J. Chem. Phys. 1961, 35, 508.
(18) Kober, E. M.; Caspar, J. V.; Lumpkin, R. S.; Meyer, T. J. J. Phys. Chem.
1986, 90, 3722.
(19) Weller, A. Z. Phys. Chem. 1982, 133, 93.
(20) Kumar, K.; Lin, Z.; Waldeck, D. H.; Zimmt, M. B. J. Am. Chem. Soc.
1996, 118, 243.
Supporting Information Available: Synthetic, experimental, and
data fitting procedures; 1H NMR shifts, accurate mass spectrum
analyses, and absorption and emission spectra for new compounds;
spectroelectrochemical data, XYZ coordinates, and computational
JA1055559
9
11466 J. AM. CHEM. SOC. VOL. 132, NO. 33, 2010