Ultrafast forward and backward electron transfer dynamics of coumarin 337 in hydrogen-bonded anilines as studied with femtosecond UV-pump/IR-probe spectroscopy

Article abstract of DOI:10.1021/jp108090b

Femtosecond infrared spectroscopy is used to study both forward and backward electron transfer (ET) dynamics between coumarin 337 (C337) and the aromatic amine solvents aniline (AN), N-methylaniline (MAN), and N,N-dimethylaniline (DMAN), where all the aniline solvents can donate an electron but only AN and MAN can form hydrogen bonds with C337. The formation of a hydrogen bond with AN and MAN is confirmed with steady state FT-IR spectroscopy, where the C= O stretching vibration is a direct marker mode for hydrogen bond formation. Transient IR absorption measurements in all solvents show an absorption band at 2166 cm^-1, which has been attributed to the C≡N stretching vibration of the C337 radical anion formed after ET. Forward electron transfer dynamics is found to be biexponential with time constants τ_ET¹ = 500 fs, τ_ET ² = 7 ps in all solvents. Despite the presence of hydrogen bonds of C337 with the solvents AN and MAN, no effect has been found on the forward electron transfer step. Because of the absence of an H/D isotope effect on the forward electron transfer reaction of C337 in AN, hydrogen bonds are understood to play a minor role in mediating electron transfer. In contrast, direct π-orbital overlap between C337 and the aromatic amine solvents causes ultrafast forward electron transfer dynamics. Backward electron transfer dynamics, in contrast, is dependent on the solvent used. Standard Marcus theory explains the observed backward electron transfer rates.

Full text of DOI:10.1021/jp108090b

ARTICLE
pubs.acs.org/JPCA
Ultrafast Forward and Backward Electron Transfer Dynamics of
Coumarin 337 in Hydrogen-Bonded Anilines As Studied with
Femtosecond UV-Pump/IR-Probe Spectroscopy
Hirendra N. Ghosh,*,†,‡ Sandeep Verma,† and Erik T. J. Nibbering*,‡
†Radiation & Photochemistry Division, Bhabha Atomic Research Centre, Trombay Mumbai -400085, India
‡Max Born Institute for Nonlinear Optics and Short Pulse Spectroscopy, Max Born Strasse 2A, D-12489, Berlin, Germany
S Supporting Information
b
ABSTRACT: Femtosecond infrared spectroscopy is used to
study both forward and backward electron transfer (ET)
dynamics between coumarin 337 (C337) and the aromatic
amine solvents aniline (AN), N-methylaniline (MAN), and
N,N-dimethylaniline (DMAN), where all the aniline solvents
can donate an electron but only AN and MAN can form
hydrogen bonds with C337. The formation of a hydrogen
bond with AN and MAN is confirmed with steady state FT-IR
spectroscopy, where the CdO stretching vibration is a direct
marker mode for hydrogen bond formation. Transient IR
absorption measurements in all solvents show an absorption
band at 2166 cm-1, which has been attributed to the CꢀN
stretching vibration of the C337 radical anion formed after
ET. Forward electron transfer dynamics is found to be biex-
ponential with time constants τET1 = 500 fs, τET2 = 7 ps in all
solvents. Despite the presence of hydrogen bonds of C337
with the solvents AN and MAN, no effect has been found on the forward electron transfer step. Because of the absence of an H/D
isotope effect on the forward electron transfer reaction of C337 in AN, hydrogen bonds are understood to play a minor role in
mediating electron transfer. In contrast, direct π-orbital overlap between C337 and the aromatic amine solvents causes ultrafast
forward electron transfer dynamics. Backward electron transfer dynamics, in contrast, is dependent on the solvent used. Standard
Marcus theory explains the observed backward electron transfer rates.
those biological systems.6 Moreover, several studies have demon-
1. INTRODUCTION
strated that hydrogen bonds can mediate electron transfer
Considerable progress in both theory and experiment has
efficiently.7 As such it has generally been recognized that the
been witnessed in the last few decades in the field of electron
transfer (ET) dynamics.1 Photoinduced electron transfer (ET)
function of hydrogen bonds in biological systems most likely
extends beyond simply providing the structural scaffolding for
reactions between donor and acceptor moieties are ubiquitous in
light harvesting and solar energy conversion2 and biological
the donors and acceptors which participate in the redox process.
systems.3 The dynamical time scales of ET and the associated
However, whether hydrogen bonds play the most dominating
role in ET remains a topic for continuous study, as every ET
efficiencies are governed by electronic coupling between donor
and acceptor. Underlying mechanisms for such ET reactions
include through space couplings, and through bond couplings,
that can either be of covalent nature or consist of hydrogen bonds
process between donor and acceptor units may also be controlled
by solvent reorganization or other types of donor-acceptor
interactions such as π-π orbital overlap.
Photoinduced bimolecular ET between donor and acceptor
or involve π-π orbital overlap linking donor and acceptor
molecules in liquid solution has been a widely studied topic8-10
units.4 Whereas through space couplings usually involve a large
Time-resolved studies have, until now, mostly focused on the
range of spatial degrees of freedom, couplings through chemical
forward ET transfer after photoexcitation of either donor or
bonds possess a high degree of directionality. Hydrogen bonds
control the separation distance and relative orientation between
donor and acceptor.5 In the case of ET in proteins and DNA,
Received: August 26, 2010
electronic coupling is thus strongly affected by the hydrogen
Revised:
November 2, 2010
bond networks making up the three-dimensional structure of
Published: December 30, 2010
r
2010 American Chemical Society
664
dx.doi.org/10.1021/jp108090b J. Phys. Chem. A 2011, 115, 664–670
|

The Journal of Physical Chemistry A
ARTICLE
acceptor molecules. Many of these studies have relied on time-
resolved fluorescence techniques, revealing the electronic excited
state decay of the chromophore after photoinitiation, caused by
oxidative ET with an acceptor molecule8 or by reductive ET from
a donor molecule.9,10 Time-resolved UV/vis pump-probe spec-
troscopy of electronic transitions of the donor and acceptor
molecules and of the radical cations and anions formed upon ET,
on the other hand, may in principle provide insight in both
forward and backward ET dynamics. This approach is, however,
strongly hampered by the extreme broadening typical for elec-
tronic transitions, and as such structural information on the ET
process has remained limited.11 In contrast, transient vibrational
spectroscopy may provide more in-depth information on the
geometry of donor-acceptor reaction pairs, as has recently been
obtained on the photoinduced oxidative ET between π-π
stacked perylene-tetracyanoethene molecules.10 Until now,
however, such an approach has not been applied to photoin-
duced ET between donor and acceptor species where hydrogen
bond dynamics could be playing a major role.
We present here femtosecond infrared spectroscopic results
on the photoinduced reductive ET of a coumarin dye and
aromatic amine solvents. Yoshihara and co-workers reported
on ultrafast electron transfer dynamics in photoexcited acceptor
coumarin dyes dissolved in donating aromatic amines. Here a
collection of coumarin dyes were used, to explore ET driving
force dependencies, and the aromatic family of anilines with
different substituents on the amino group, including the solvents
aniline (AN), N-methylaniline (NMAN), and N,N-dimethylani-
line (DMAN), to investigate the effects of hydrogen bonding
dynamics. Hydrogen bonding is expected to occur between the
hydrogen bond donating groups of AN and NMAN and the
CdO moiety of the coumarin dyes, whereas this is not possible
with DMAN. Because in most of these studies the fluorescence
upconversion technique was used, information has been ob-
tained only on the forward ET process. In the forward ET
reaction between these coumarin dyes and the aromatic amine
solvents, only a minor H/D isotope effect has been found, sugg-
esting that ET is not strongly mediated by any possible site-
specific solute-solvent hydrogen bonding. Instead, with ET
reaction rates for the investigated systems often faster than
solvation dynamics, electronic couplings appear to play a more
important role dictating the ET dynamics.
Ultrafast IR spectroscopy provides key insight into the dy-
namics of site-specific hydrogen bond interactions between
solute and solvent affected by electronic excitation, as has been
demonstrated by Chudoba et al.12 on coumarin 102 (C102)
dissolved in chloroform or hydrogen bonded with methanol in
nonpolar solvent. Palit et al.13 used this approach to investigate
the hydrogen bond dynamics of C102 in the electronic excited
state in AN solvent. C102-AN, however, does not exhibit an
ultrafast electron transfer reaction. Recently, using time-depen-
dent density functional theory, Liu et al. calculated that the
intermolecular C102-AN hydrogen bond, connecting the CdO
group of C102 and the NH of the amino group of AN, is
significantly strengthened upon electronic excitation.14 Wang
et al.11 reported on the ultrafast forward electron transfer
dynamics between coumarin 337 (C337) and DMAN solvent
using a combined ultrafast UV/IR and UV/vis spectroscopic
approach. Whereas a forward ET rate with a 4 ps time constant
was reported, no details were given on the backward ET process,
and in addition no hydrogen bonding effects using other aniline
solvents were explored.
Scheme 1. Molecular Structures of C337 (Acceptor) and
N-Methylated Anilines (Donors)
We now present results on C337 dissolved in AN, NMAN,
and DMAN (Scheme 1). We have carried out femtosecond UV-
pump/IR-probe measurements of C337 dissolved in these aro-
matic amine solvents by exciting C337 to the S1-state using pump
pulses tuned at 400 nm and probing absorbance changes of
vibrational marker bands. Using this approach we can determine
both forward and backward electron transfer rates. Whereas
effects of hydrogen bonds turn out to be of minor magnitude, we
conclude that other electronic coupling mechanisms, in particu-
lar stacking of the π-orbitals of C337 and the aromatic solvents,
play a more important role.
2. EXPERIMENTAL SECTION
To follow the temporal evolution of vibrational marker modes,
we used a femtosecond infrared spectroscopic laser system
consisting of a 1 kHz amplified Ti:sapphire regenerative and
booster amplifier system (Spitfire Pro, Spectra Physics) and
frequency conversion stages. The second harmonic output of
the amplified laser system (wavelength 400 nm, pulse duration
55 fs, energy 3-7 μJ, spot diameter 300 λm) was used to excite
C337. Tunable mid-IR probe pulses (100-150 fs duration, 10 nJ
energy) were generated by difference frequency mixing of signal
and idler pulses from a near-infrared optical parametric amp-
lifier.15 Probe and reference pulses were derived using reflections
from a BaF2 wedge and focused onto the sample with off-axis
parabolic mirrors (focal diameter 100 mm). After spectral
dispersion with a polychromator, the probe pulses were detected
by a multichannel mid-IR detector array (IR associates) with a
spectral resolution of 4-8 cm-1, depending on the spectral
region probed. The whole pump-probe setup was purged
with nitrogen gas to avoid spectral and temporal reshaping of
the mid-IR pulses by the absorption of water vapor and CO2 in
air. For reference purposes the sample was replaced by a
polished ZnSe window to determine the zero delay point
(including the chirp of the IR pulse) and to control the time
resolution of the cross-correlation (fwhm 120-140 fs). The
relative polarization of the pump and probe pulses were set
under magic angle to detect ET dynamics not affected by
rotational diffusion effects.
Sample solutions were circulated through a flow cell consisting
of 1 mm thick BaF2 windows separated by a 100 μm thick Teflon
spacer, to guarantee that for every laser shot a new fraction of the
solution was excited. We used 10 mM solutions of C337
(Aldrich) in the femtosecond pump-probe experiments. Aniline
(AN), N-methylaniline (NMAN), and dimethylaniline (DMAN)
were obtained from Fluka. Deuterated aniline-d2, C6H5ND2,
was obtained by H/D exchange using D2O (Deutero GmbH),
followed by extraction of the organic phase and further drying.
All measurements were performed at room temperature, T =
25 ( 2 °C.
665
dx.doi.org/10.1021/jp108090b |J. Phys. Chem. A 2011, 115, 664–670

The Journal of Physical Chemistry A
ARTICLE
Figure 2. Transient IR absorption spectra of C337 in acetonitrile at
different time delays after excitation at 400 nm. Panel A shows the CꢀN
stretching region. Panel B depicts the upper part of the fingerprint
region.
find that the CꢀN band is located at 2222, 2220, and 2218 cm-1
for DMAN, NMAN, and AN, respectively. Such small shifts may
be caused by dielectric constant changes, or by the indirect effect
of hydrogen bonds when an aromatic amine solvent molecule
complexes with the CdO group, or when π-π stacking occurs
between C337 and the aromatic amine solvent molecules. The
spectral width of the CꢀN transition also increases slightly when
going from DMAN, via NMAN, to AN as solvent. One can
conclude from these steady FT-IR spectra that C337 has a
stronger interaction with the solvent molecules when going from
DMAN, via MAN, to AN, likely caused by a direct coupling
between the electrical dipoles of C337 and the solvent and/or
π-π stacking between the aromatic π-orbitals, accompanied by
an increasing hydrogen bond strength with an additional solvent
molecule at the CdO group.
3.2. Transient IR Absorption Spectroscopy of C337 in
Acetonitrile. We first present transient IR results of C337 in
the non-hydrogen bonding and nonreacting solvent acetonitrile
(ACN). Panel B of Figure 2 shows the transient IR absorption
spectra of C337 in the spectral region from 1580 to 1800 cm-1
recorded at several pulse delays after exciting with the 400 nm
pump pulse. The transient spectra show the time-resolution
limited generation of bleach signals at 1597, 1626, and 1737
cm-1, directly correlated with the peak positions of CdC and
CdO stretching signals of C337 in the electronic ground state. In
addition to these bleach signals, a positive absorption band at
1664 cm-1 appears within temporal resolution, which we assign
as the CdO stretching vibration of C337 in the first electronic
excited state. This band decays on a time scale >1 ns (beyond the
maximum scanning range of our delay stage), in accordance with
bleach recovery with a similar time constant, suggesting an S1-
state lifetime larger than 1 ns. We have determined the electronic
excited state lifetime using time-correlated single photon count-
ing of the fluorescence emission to be 3.7 ns (see Supporting
Information, Figure S3). We have also carried out transient
absorption studies of the CꢀN stretching frequency region from
2160 to 2240 cm-1 (panel A, Figure 2). The transient spectra
show a clear bleach signal at 2230 cm-1 indicative of the CꢀN
stretching mode of C337 in the S0-state. We have identified a
positive signal located at 2180 cm-1 and attribute this to the
CꢀN stretching mode of C337 in the S1-state. We note that
Figure 1. Panel A: Ground state infrared spectrum of C337 in
acetonitrile (ACN). The infrared absorption at 1500-1650 cm-1 can
be attributed to the stretching mode CdC ring vibration, the band in the
1690-1740 cm-1 region to the stretching mode of the CdO group, and
the band in the 2220-2250 cm-1 region to the stretching mode of the
CꢀN vibration. Panel B: Steady state IR spectra of C337 in AN, MAN,
and DMAN from 1680 to 2240 cm-1 showing only the CdO and CꢀN
stretching modes.
3. EXPERIMENTAL RESULTS
3.1. Steady-State IR Spectra of C337. Panel A of Figure 1
shows the FT-IR spectra of C337 in acetonitrile. The bands in the
1500-1650 cm-1 region can be attributed to CdC ring vibra-
tional modes. The sharp band located at 1722 cm-1 is the CdO
stretching vibrational transition of C337. The CꢀN marker
mode has its transitional frequency at 2223 cm-1. This provides
us with at least two distinct marker modes, the CdO and CꢀN
vibrations, with which we can probe the photoinduced ET
reaction. Hydrogen bonding with protic solvents is expected to
occur at the CdO group of C337 (panel B, Figure 1). Indeed,
when dissolving C337 in AN, MAN, and DMAN, the frequency
position of the CdO marker mode clearly exhibits a red-shift in
AN and to a lesser extent in MAN, whereas in DMAN the CdO
stretching band is located at the same position as in other non-
hydrogen bonding solvents (panel A, Figure 1). The two-peaked
structure of the CdO stretching mode in MAN (1718 and 1709
cm-1) and in AN (1717 and 1706 cm-1) could be assigned to a
non-hydrogen-bonded C337 and a C337-solvent hydrogen-
bonded complex, respectively, or to a Fermi resonance splitting
(often observed for coumarin dyes) of the CdO stretching band
of C337 hydrogen bonded to the solvent. Indeed, a small solvent
shift due to changes in dielectric constant of the solvent when
going from DMAN to MAN or AN may occur without having a
hydrogen bond at the CdO moiety. Another possibility is the
interaction between the π-orbitals of C337 and nearby aromatic
amine solvent molecules that may be larger for AN and less so for
DMAN due to steric hindrance of the double-alkylated amino
group. Minor frequency shifts also happen in the case of the
CꢀN stretching band, where hydrogen bonds with the solvent
are not expected to directly occur at this functional group. We
666
dx.doi.org/10.1021/jp108090b |J. Phys. Chem. A 2011, 115, 664–670

The Journal of Physical Chemistry A
ARTICLE
Figure 3. Transient IR absorption spectra of C337 in different amine
solvents as function of time delay after excitation at 400 nm. Panel A: in
DMAN. Panel B: in MAN. Panel C: in AN.
Figure 4. Transient population kinetics at 2166 cm-1 of C-337 in AN
(red filled dots), MAN (green filled triangles), and DMAN (open blue
squares). Panel A: Early time population kinetics, showing the forward
ET dynamics. Panel B: Long time population kinetics, showing the
solvent dependent back ET dynamics. Panel C: Comparison of the
population kinetics in AN-h2 and AN-d2, showing the absence of H/D
isotope effects on the forward and backward ET dynamics in aniline.
upon electronic excitation of C337 this CꢀN stretching vibra-
tion has, besides its frequency downshift of 50 cm-1, a significant
decrease in oscillator strength compared to its value in the S0-
state, in strong contrast to the CdO stretching vibration which
shows merely a frequency downshift of 73 cm-1 without much
change in oscillator strength. The fact that both bands downshift
in frequency provides insight into the charge redistribution upon
excitation to the S1-state.
cm-1 band, and only a small difference on the decay of the band
(Figure 4C).
3.3. Transient IR Absorption of C337 in Aromatic Amine
Solvents. The aromatic amine solvents have a pronounced
steady state absorption in the spectral range of the CdO
stretching vibration, preventing probing this important marker
mode when exciting C337 in these solvents. We thus have to rely
on the transient response of the CꢀN marker band to learn
about the ET dynamics. Figure 3 shows the transient IR spectra
of C337 in DMAN, MAN, and AN, recorded in the CꢀN
stretching region. Apart from the bleach signal located at the
CꢀN stretching transition frequency of C337 in the S0-state, we
observe a positive transient signal at 2166 cm-1. Such a
frequency shift is significantly larger than the frequency shift
we have recorded of the CꢀN stretching vibration of C337 when
going from the S0-state to the S1-state. Moreover, now this
positive signal has about similar magnitude as the bleach signal at
2230 cm-1. Whereas the bleach appears within temporal resolu-
tion, and recovers with a time constant of 110 ps in DMAN, the
positive band at 2166 cm-1 grows in a nonexponential fashion on
picosecond time scales and decays with the same 110 ps time
constant as the bleach recovers (Figure 4A, B). The decay of the
positive signal and the bleach recovery are for all solvents
identical, however with a varying solvent dependent time con-
stant: we measure it to be 40 ps in MAN and 66 ps in AN. The
nonexponential rise can be fitted with a biexponential function,
using in contrast the same time constants for all solvents: τ1= 500
fs and τ2=7 ps, where now the relative magnitudes weakly depend
on the electron donating solvent used: A1 = 33% and A2 = 67%
for DMAN, A1 =27% and A2 =73% for MAN, and A1 =23% and
A2=77% for AN. We have in addition performed an experiment
using AN-d2 and found no isotope effect on the rise of the 2166
4. DISCUSSION
Based on the fact that in the aromatic amine solvents the S1-
state of C337 is short-lived, converting via an intermediate state
with a CꢀN stretching vibration located at a different frequency
position back to the electronic ground state, we conclude that
upon photoexcitation of C337 reductive forward electron trans-
fer from the electron donating amine solvents occurs, generating
a C337•- radical anion and the aromatic amine radical cation.
This ion pair then follows the backward electron transfer reaction
pathway recovering C337 in the electron ground state and
neutral solvent molecule. The fact that photoexcitation of
C337 in aromatic amine solvents initiates electron transfer is in
full accordance with the large body of work obtained using other
coumarin dyes.8,16 Whereas most of these studies involve time-
resolved emission studies, providing information only on the
forward electron transfer step, Wang et al. performed UV/vis
electronic pump-probe spectroscopy of C337 in DMAN, and
by monitoring the transient absorption at 500 nm, where the
DMAN•þ is understood to absorb, they derived a 4 ps time
constant for the forward electron transfer reaction.11 Wang et al.
did not discuss the back electron reaction of the charge-separated
species. We also performed femtosecond UV/vis pump-probe
spectroscopy of C337 in DMAN. A straightforward interpreta-
tion appears to be impossible due to strong overlap of the
absorption of DMAN•þ with the stimulated emission of C337
at early pulse delays (see Supporting Information, Figures S1 and
S2), which may be affected by solvation dynamics as well.
Measurement of the transient IR response of the CꢀN stretching
vibration of C337•- appears to lead to results that are easier to
interpret. Whereas Wang et al. do not report on the ET dynamics
667
dx.doi.org/10.1021/jp108090b |J. Phys. Chem. A 2011, 115, 664–670

The Journal of Physical Chemistry A
ARTICLE
that can be derived from the time-dependent absorption of the
CꢀN stretching marker mode, we have derived both forward and
backward ET reaction time scales in all solvents. We note that the
small additional 14 cm-1 red-shift of the CꢀN stretching
vibration, when C337 in the S1-state converts to C337•-,
indicates that the additional electron density in C337•- is located
in an orbital with antibonding character along the CꢀN bond,
but that a larger effect on the CꢀN bond is obtained when
promoting an electron in C337 from the HOMO to the LUMO
by photoexcitation.
We do not observe a large solvent dependence in forward ET
transfer rates, when comparing the experimental kinetics traces
obtained in DMAN, MAN, and AN. As a result, there is no direct
connection between a pronounced red-shift in CdO stretching
vibration, i.e., the formation of a hydrogen bond between the
CdO group of C337 and a solvent molecule, and the forward ET
process. In addition to that, the absence of a pronounced isotope
effect in the forward and backward ET steps in AN strongly
suggests that hydrogen bonds of the solvent to the solute, or
between solvent molecules, are not playing a key role in mediat-
ing ET between C337 and the electron donating aromatic amine
solvent molecules. Instead, other factors, such as solvation
dynamics of the polar solvent or differences in electronic
coupling coefficients, are likely more important in dictating ET
rates. Similar conclusions were drawn in the reports by Yoshihara
and co-workers.8
ET in the three aromatic amine solvents. This suggests that the
coupling between the solvation shells and the ET reaction
coordinate remains moderate.
Di Labio and Johnson19 have shown that π-π interactions
play a major role in proton-coupled electron transfer reactions.
Recently Takai et al.20 demonstrated enhanced electron transfer
properties of cofacial porphyrin dimers through π-π interac-
tions. In the present investigation also we suggest that electronic
interactions between the π-orbitals of C337 and the aromatic
amine solvent molecules play a key role. With this we can
interpret the nonexponential behavior of the forward ET reaction
as being indicative of a distribution of configurations with
different spatial orientations of C337 and the solvent molecules.
The fastest component of 0.5 ps may be caused by those
configurations where the π-orbital overlap between C337 and
the aromatic amine solvent is optimal, whereas for the slower 7 ps
component couplings may be smaller due to less favorable
relative orientations. We can assume that once ET has occurred
the C337•- radical anion and the DMAN•þ/MAN•þ/AN•þ
radical cation remain as an ion pair in solution, because ion pair
separation does not occur in the aromatic amine solvents that
have only moderate values for the dielectric constant. As a result
the fate of the ion pair is to convert back to the ground state by
backward ET.
In contrast we do observe a solvent dependence in backward
ET rates, with, however, a less obvious order, as the fastest
backward ET occurs in MAN (40 ps), followed by AN (66 ps),
and the slowest backward ET is found for DMAN (110 ps). With
backward ET rates much longer than solvation dynamics, we may
assume that the standard Marcus theory for ET applies. Now the
feasibility of backward ET reaction from reduced coumarin
(C337•-) to oxidized DMAN•þ/MAN•þ/AN•þ is dictated by
the standard free energy change (-ΔGo) for the ET reaction
following Marcus theory. The free energy change for the back-
ward ET reaction can be expressed as17
We now calculate the free energy changes (ΔG0) for the
forward ET reactions of photoexcited C-337 and anilines. The
free energy changes (ΔG0) for the ET reactions between the
ground state anilines (donors) and the excited state C-337
(acceptor) in neat aniline solvent can be expressed as17
e2
o
ΔGCSET ¼ EðD=DþÞ - EðA=A - Þ - E00
-
ð1Þ
εrDA
where E00 is the 0-0 transition energy (between the S0 and S1
states) of the C337, E(A/A-) is the reduction potential of C337,
E(D/Dþ) is the oxidation potential of neat anilines, and e2/εrDA
is the stabilization energy of the product ion pair state in anilines
(where e is the charge of an electron, ε is the dielectric constant of
the solvent, and rDA is the distance between donor-acceptor
pair). Using eq 1 we have determined ΔGCSETo to be -0.55 eV
for AN, -0.67 eV for MAN, and -0.72 eV for DMAN.
Considering Marcus ET theory,18 we expect faster ET reaction
rates when going from AN, via MAN, to DMAN. Based on the
fact that the influence of hydrogen bonds seems to be absent, and
the fact that the ET driving force does not govern the ET rates, all
the donor-acceptor pairs have special-and similar-interactions
dominating the ET dynamics, which allows them to react with
similar reaction rates.
From Shirota et al. we learn that the early time component of
solvation dynamics in DMAN is about two times slower than in
MAN and about three times slower than in AN, whereas the long
time components are similar.8 For DMAN the solvation correla-
tion function can be fitted with biexponential decay with time
constants of τsolv,1=3.8 ps (46%) and τsolv,2=15.0 ps (54%). For
MAN these constants are τsolv,1 = 2.0 ps (25%) and τsolv,2 =16.2
ps (75%) and for AN τsolv,1 = 1.2 ps (28%) and τsolv,2 = 17.8 ps
(72%). One may anticipate a direct interplay between the
rearranging solvation shells and the forward ET coordinate, as
solvation dynamics and forward ET occur on similar time scales.
However, we do not observe any change in dynamics for forward
e2
ΔGCRET ¼ - EðD=DþÞ þ EðA=A - Þ þ
ð2Þ
o
εrDA
Following eq 2 the calculated value of free energy change for
DMAN is ΔGCRET = -1.97 eV, for MAN it is ΔGCRET
o
o
=
-2.02 eV, and for AN it is ΔGCRETo = -2.14 eV. Considering the
fact that a similar type of interaction between donor-acceptor
pairs occurs in all solvents studied here, we can assume that the
coupling matrix elements for the backward ET reaction are
similar for all three cases. Following Marcus ET theory, this
estimation of the driving force for ET suggests that the back ET
reaction for all the donor-acceptor pairs falls in the inverted
regime of ET reactions. We can clearly observe that back ET in
MAN is faster (40 ps) than in AN (66 ps), which follows Marcus
ET theory. On the other hand, we expect the back ET reaction
rate in DMAN to be the fastest of all three solvent cases. Instead,
we observe the slowest reaction time (110 ps) in C337/DMAN.
We suggest that the magnitudes of the π-π interactions, which
dictate the electronic couplings between donor and acceptor,
may be affected by steric hindrance factors, where the geometric
arrangement of the two methyl groups in DMAN might be the
reason for slowing down the BET reaction in DMAN.
5. CONCLUSION
We have investigated forward and backward electron transfer
(ET) dynamics between coumarin 337 (C337) (acceptor) and
668
dx.doi.org/10.1021/jp108090b |J. Phys. Chem. A 2011, 115, 664–670

The Journal of Physical Chemistry A
ARTICLE
the aromatic amine solvents (donor) aniline (AN), N-methylani-
line (MAN), and N,N-dimethylaniline (DMAN) with the help of
femtosecond infrared spectroscopy. Monitoring the CdO
stretching vibration of C337 in the S0-state reveals that a
hydrogen bond between C337 and a solvent molecule occurs
in AN and MAN. Electron transfer from amine solvents to
photoexcited C337 can be deduced from the appearance of the
CꢀN stretching vibration band at 2166 cm-1 of C337 radical
anion radical in the transient IR spectrum. Forward ET dynamics
is found to be biexponential with time constants τET1 = 500 fs,
Collin, J.-P.; Gavi~na, P.; Heitz, V.; Sauvage, J.-P. Eur. J. Inorg. Chem.
1998, 1–14. (i) Sauvage, J.-P.; Collin, J.-P.; Chambron, J.-C.; Guillerez,
S.; Coudret, C.; Balzani, V.; Barigelleti, F.; De Cola, L.; Flamigni, L.
Chem. Rev. 1994, 94, 993–1019. (j) Imahori, H.; Norieda, H.; Yamada,
H.; Nishimura, Y.; Yamazaki, I.; Sakata, Y.; Fukuzumi, S. J. Am. Chem.
Soc. 2001, 123, 100–110. (k) Hagfeldt, A.; Gr€atzel, M. Acc. Chem. Res.
2000, 33, 269–277. (l) Bignozzi, C. A.; Argazzi, R.; Indelli, M. T.;
Scandola, F. Sol. Energy Mater. Sol.Cells 1994, 32, 229–244. (m)
O'Regan, B.; Gr€atzel, M. Nature 1991, 353, 737–740.
(3) (a) Krauss, N. Curr. Opin. Chem. Biol. 2003, 7, 540–550. Willner,
I.; Willner, B. Coord. Chem. Rev. 2003, 245, 139–151. (b) Lubitz, W.;
Lendzian, F.; Bittl, R. Acc. Chem. Res. 2002, 35, 313–320.
2
τET = 7 ps in all solvents. This shows that hydrogen bonds
(4) (a) Piotrowiak, P. Chem. Soc. Rev. 1999, 28, 143. (b) Ward, M. D.
Chem. Soc. Rev. 1997, 26, 365.(c) Sessler, J. L.; Wang, B.; Springs, S. L.;
Brown, C. T. In Comprehensive Supramolecular Chemistry; Atwood, J. L.,
Davies, J. E. D., MacNicol, D. D., Vogtle, F., Murakami, Y., Eds.;
Pergamon: Oxford, 1996; Vol. 4, p 311.(d) Deng, Y.; Roberts, J. A.;
Peng, S.-M.; Chang, C. K.; Nocera, D. G. Angew. Chem., Int. Ed. Engl.
1997, 36, 2124. (e) Kirby, J. P.; Roberts, J. A.; Nocera, D. G. J. Am. Chem.
Soc. 1997, 119, 9230. (f) Hayashi, T.; Miyahara, T.; Kumazaki, S.;
Ogoshi, H.; Yoshihara, K. Angew. Chem., Int. Ed. Engl. 1996, 35, 1964. (g)
Damrauer, N. H.; Hodgkiss, J. M.; Rosenthal, J.; Nocera, D. G. J. Phys.
Chem. B 2004, 108, 6315. (h) Hodgkiss, J. M.; Damrauer, N. H.; Pressꢀe,
S.; Rosenthal, J.; Nocera, D. G. J. Phys. Chem. B 2006, 110, 18853. (i)
Reece, S. Y.; Nocera, D. G Annu. Rev. Biochem. 2009, 78, 673.
(5) Cooley, L. F.; Headford, C. F. L.; Elliott, C. M.; Kelly, D. F. J. Am.
Chem. Soc. 1988, 110, 6673.
(6) (a) Bowler, B. E.; Meade, T. J.; Mayo, S. L.; Richards, J. H.; Gray,
H. B. J. Am. Chem. Soc. 1989, 111, 8757. (b) Therien, M. J.; Selman, M.;
Gray, H. B.; Chang, I.-J.; Winkler, J. R. J. Am. Chem. Soc. 1990, 112, 2420.
(c) Beratan, D. N.; Betts, J. N.; Onuchic, J. N. Science 1991, 252, 1285.
(d) Philip, D.; Stoddart, J. F. Angew. Chem., Int. Ed. Engl. 1996, 35,
1154.
(7) (a) de Rege, P. J. F.; Williams, S. A.; Therien, M. J. Science 1995,
269, 1409–1413. (b) Turro, C.; Chang, C. K.; Leroi, G. E.; Cukier, R. I.;
Nocera, D. G. J. Am. Chem. Soc. 1992, 114, 4013–4015. (c) Kirby, J. P.;
Roberts, J. A.; Nocera, D. G. J. Am. Chem. Soc. 1997, 119, 9230–9236.
(d) Roberts, J. A.; Kirby, J. P.; Wall, S. T.; Nocera, D. G. Inorg. Chim. Acta
1997, 263, 395–405. (e) Williamson, D. A.; Bowler, B. E. J. Am. Chem.
Soc. 1998, 120, 10902–10911.
between C337 and the amine solvents do not play a major role in
the forward ET dynamics, which is in addition supported by the
absence of an H/D isotope effect on this ET reaction step in AN.
Due to the fact that driving forces for the forward ET reaction are
significantly different in AN, MAN, and DMAN, we suggest that
electronic interactions between the π-orbitals of C337 and the
aromatic amine solvent molecules play a key role in this process.
We have also followed the back ET dynamics by monitoring the
transient decay of anion radical marker mode (2166 cm-1) of
C337 and found this reaction step to be solvent dependent.
Standard Marcus theory again fails to predict the right correlation
between driving force and reaction rate in the backward ET step.
We suggest that steric hindrance of the dimethylamino group in
the C337/DMAN may be a reason for this.
’ ASSOCIATED CONTENT
S
Supporting Information. Transient UV/vis absorption
b
spectra of C337 in ACN and in DMAN measured in the visible
region at different time delays after exciting C337 at 400 nm
(Figure S1) and associated kinetic traces (Figure S2). Time-
resolved emission kinetics of C337 in acetonitrile after excitation
at 400 nm and detection at 490 nm (Figure S3). This material is
available free of charge via the Internet at http://pubs.acs.org.
(8) (a) Nagasawa, Y.; Arkadiy, T.; Yartsev, P.; Tominaga, K.; Bisht,
P. B.; Johnson, A. E.; Yoshihara, K. J. Phys. Chem. 1995, 99, 653–662. (b)
Nagasawa, Y.; Arkadiy, T.; Yartsev, P.; Tominaga, K.; Johnson, A. E.;
Yoshihara J. Am. Chem. Soc. 1993, 115, 7922–7923. (c) Pal, H.;
Nagasawa, Y.; Tominaga, K.; Yoshihara, K. J. Phys. Chem. 1996, 100,
11964–11974. (d) Shirota, H.; Pal, H.; Tominaga, K.; Yoshihara, K.
J. Phys. Chem. A 1998, 102, 3089–3102. (e) Rubtsov, I. V.; Shirota, H.;
Yoshihara, K. J. Phys. Chem. A 1999, 103, 1801–1808. (f) Shirota, H.; Pal,
H.; Tominaga, K.; Yoshihara, K. J. Phys. Chem. A 1998, 102, 3089–3102.
(9) (a) Singh, A. K.; Mondal, J. A.; Ramakrishna, G.; Ghosh, H. N.;
Bandyopadhyay, T.; Palit, D. K. J. Phys. Chem. A 2005, 109, 4014–4023.
(b) Glusac, K.; Goun, A.; Fayer, M. D. J. Chem, Phys 2006, 125, 054712.
(10) (a) Mohammed, O. F.; Banerji, N.; Lang, B.; Nibbering, E. T. J.;
Vauthey, E. J. Phys. Chem. A 2006, 110, 13676–13680. (b) Mohammed,
O. F.; Adamczyk, K.; Banerji, N.; Dreyer, J.; Lang, B.; Nibbering, E. T. J.;
Vauthey, E. Angew. Chem., Int. Ed. 2008, 47, 9044–9048.
’ AUTHOR INFORMATION
Corresponding Author
*H.N.G.: e-mail hnghosh@barc.gov.in; fax 00-91-22-25505151.
E.T.J.N.: e-mail nibberin@mbi-berlin.de..
’ REFERENCES
(1) (a) Marcus, R. A. J. Chem. Phys. 1956, 24, 966. (b) Marcus, R. A.
J. Chem. Phys. 1956, 24, 979. (c) Marcus, R. A. Annu. Rev. Phys. Chem.
1964, 15, 155. (d) Barbara, P. F.; Jarzeba, W. Adv. Photochem. 1990, 15,
1. (e) Weaver, M. J.; McManis, G. E., III. Acc. Chem. Res. 1990, 23, 294.
(f) Electron Transfer in Inorganic, Organic and Biological Systems; Bolton,
J. R., Mataga, N., McLendon, G. L., Eds.; American Chemical Society:
Washington, DC, 1991.(g) Newton, M. D.; Sutin, N. Annu. Rev. Phys.
Chem. 1984, 35, 437. (h) Yoshihara, K.; Tominaga, K.; Nagasawa, Y.
Bull. Chem. Soc. Jpn. 1995, 68, 696. (i) Zusman, L. D. Chem. Phys. 1980,
49, 295. (j) Zusman, L. D. Chem. Phys. 1988, 119, 51. (k) Rips, I.;
Jortner, J. J. Chem. Phys. 1987, 87, 2090. (l) Jortner, J.; Bixon, M. J. Chem.
Phys. 1988, 88, 167. (m) Gao, Y. Q.; Marcus, R. A. J. Chem. Phys. 2000,
113, 6351. (n) Marcus, R. A. J. Electroanal. Chem. 1997, 438, 251.
(2) (a) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2001, 34,
40–48. (b) Barbara, P. F.; Meyer, T. J.; Ratner, M. A. J. Phys. Chem. 1996,
100, 13148–13168. (c) Hagfeldt, A.; Gr€atzel, M. Chem. Rev. 1995, 95,
49–68. (d) Bard, A. J.; Fox, M. A. Acc. Chem. Res. 1995, 28, 141–145. (e)
Paddon-Row, M. N. Acc. Chem. Res. 1994, 27, 18–25. (f) Wasielewsky,
M. R. Chem. Rev. 1992, 92, 435–461. (g) Baranoff, E.; Collin, J.-P.;
Flamigni, L.; Sauvage, J.-P. Chem. Soc. Rev. 2004, 33, 147–155. (h)
(11) Wang, C.; Akhremitchev, B.; Walker, G. C. J. Phys. Chem. A
1997, 101, 2735.
(12) (a) Chudoba, C.; Nibbering, E. T. J.; Elsaesser, T. Phys. Rev.
Lett. 1998, 81, 3010. Chudoba, C.; Nibbering, E. T. J.; Elsaesser, T.
J. Phys. Chem. A 1999, 103, 5625. (b) Nibbering, E. T. J.; Tschirschwitz,
F.; Chudoba, C.; Elsaesser, T. J. Phys. Chem. A 2000, 104, 4236.
(13) Palit, D. K.; Zhang, T.; Kumazaki, S.; Yoshihara, K. J. Phys.
Chem. A 2003, 107, 10798–10804.
(14) Liu, Y.; Ding, J.; Shi, D.; Sun, J. J. Phys. Chem. A 2008, 112,
6244–6248.
(15) Kaindl, R. A.; Wurm, M.; Reimann, K.; Hamm, P.; Weiner,
A. M.; Woerner, M. J. Opt. Soc. Am. 2000, B 17, 2086.
669
dx.doi.org/10.1021/jp108090b |J. Phys. Chem. A 2011, 115, 664–670

The Journal of Physical Chemistry A
ARTICLE
(16) (a) Shannon, C. F.; Eads, D. D. J. Chem. Phys. 1995, 103, 5201.
(b) Castner, E. W.; Kennedy, D., Jr.; Cave, R. J. J. Phys. Chem. A 2000,
104, 2869–2885.
(17) Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 259.
(18) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265.
(19) Di Labio, G. A.; Johnson, E. R. J. Am. Chem. Soc. 2007, 129,
6199–6203.
(20) Takai, A.; Gros, C.; Barbe, J.-M.; Guilard, R.; Fukuzumi, S.
Chem.;Eur. J. 2009, 15, 3110–3122.
670
dx.doi.org/10.1021/jp108090b |J. Phys. Chem. A 2011, 115, 664–670