Excited state relaxation dynamics of model green fluorescent protein chromophore analogs: Evidence for cis-trans isomerism

Article abstract of DOI:10.1021/jp206815t

Two green fluorescent protein (GFP) chromophore analogs (4Z)-4-(N,N-dimethylaminobenzylidene)-1-methyl-2-phenyl-1,4-dihydro-5H- imidazolin-5-one (DMPI) and (4Z)-4-(N,N-diphenylaminobenzylidene)-1-methyl-2- phenyl-1,4-dihydro-5H-imidazolin-5-one (DPMPI) were investigated using femtosecond fluorescence up-conversion spectroscopy and quantum chemical calculations with the results being substantiated by HPLC and NMR measurements. The femtosecond fluorescence transients are found to be biexponential in nature and the time constants exhibit a significant dependence on solvent viscosity and polarity. A multicoordinate relaxation mechanism is proposed for the excited state relaxation behavior of the model GFP analogs. The first time component (τ₁) was assigned to the formation of twisted intramolecular charge transfer (TICT) state along the rotational coordinate of N-substituted amine group. Time resolved intensity normalized and area normalized emission spectra (TRES and TRANES) were constructed to authenticate the occurrence of TICT state in subpicosecond time scale. Another picosecond time component (τ₂) was attributed to internal conversion via large amplitude motion along the exomethylenic double bond which has been enunciated by quantum chemical calculations. Quantum chemical calculation also forbids the involvement of hula-twist because of high activation barrier of twisting. HPLC profiles and proton-NMR measurements of the irradiated analogs confirm the presence of Z and E isomers, whose possibility of formation can be accomplished only by the rotation along the exomethylenic double bond. The present observations can be extended to p-HBDI in order to understand the role of protein scaffold in reducing the nonradiative pathways, leading to highly luminescent nature of GFP.

Full text of DOI:10.1021/jp206815t

ARTICLE
pubs.acs.org/JPCA
Excited State Relaxation Dynamics of Model Green Fluorescent
Protein Chromophore Analogs: Evidence for CisꢀTrans Isomerism
Shahnawaz Rafiq,† Basanta Kumar Rajbongshi,† Nisanth N. Nair, Pratik Sen,* and Gurunath Ramanathan*
Department of Chemistry, Indian Institute of Technology Kanpur, Kanpur, 208 016, UP, India
S Supporting Information
b
ABSTRACT: Two green fluorescent protein (GFP) chromo-
phore analogs (4Z)-4-(N,N-dimethylaminobenzylidene)-1-meth-
yl-2-phenyl-1,4-dihydro-5H-imidazolin-5-one (DMPI) and (4Z)-
4-(N,N-diphenylaminobenzylidene)-1-methyl-2-phenyl-1,4-di-
hydro-5H-imidazolin-5-one (DPMPI) were investigated using
femtosecond fluorescence up-conversion spectroscopy and
quantum chemical calculations with the results being substan-
tiated by HPLC and NMR measurements. The femtosecond
fluorescence transients are found to be biexponential in nature
and the time constants exhibit a significant dependence on
solvent viscosity and polarity. A multicoordinate relaxation mechanism is proposed for the excited state relaxation behavior of the
model GFP analogs. The first time component (τ1) was assigned to the formation of twisted intramolecular charge transfer (TICT)
state along the rotational coordinate of N-substituted amine group. Time resolved intensity normalized and area normalized
emission spectra (TRES and TRANES) were constructed to authenticate the occurrence of TICT state in subpicosecond time scale.
Another picosecond time component (τ2) was attributed to internal conversion via large amplitude motion along the exomethylenic
double bond which has been enunciated by quantum chemical calculations. Quantum chemical calculation also forbids the
involvement of hula-twist because of high activation barrier of twisting. HPLC profiles and proton-NMR measurements of the
irradiated analogs confirm the presence of Z and E isomers, whose possibility of formation can be accomplished only by the rotation
along the exomethylenic double bond. The present observations can be extended to p-HBDI in order to understand the role of
protein scaffold in reducing the nonradiative pathways, leading to highly luminescent nature of GFP.
an excited state lifetime of 3.3 ns,13,14 its isolated chromophore
1. INTRODUCTION
p-HBDI has a quantum yield that is almost 4000-fold less and a
The green fluorescent protein (GFP) of the jellyfish Aequorea
lifetime around 3300-fold less than the wtGFP.15 Further, the
victoria is very widely used as a genetically encoded noninvasive
denatured protein, isolated fragments of protein containing the
fluorescence marker in bioimaging.1 Some years after its discovery,
chromophore, and most of the synthetic model compounds of
it was shown that GFP could be cloned and expressed in other
the GFP chromophore, are all weakly fluorescent (ϕ ≈ 10ꢀ3) in
organisms.1,2 The utilityof GFP asa fluorescence markergained all
bulk solutions at standard conditions.15,16 These observations
around acceptance especially after the beautiful fluorescence
have led to an extensive investigation of the photophysical proper-
images obtained by Chalfie et al.1 Advanced developments pre-
ties of GFP and its mutants,17 the synthetic derivatives of the GFP
dominantly through mutagenesis to increase the stability and
chromophore and its related model systems, by theory and
diversify the spectral range enhanced the vast potential of GFP
experiment.14ꢀ16,18ꢀ26 The majority of the literature suggest that
and other fluorescent proteins.3ꢀ5 This protein is particularly
useful due to its stability and the fact that its chromophore is
generated in situ by an autocatalytic post-translational cyclization
HBDI and its analogs undergo excited state twisting that triggers
internal conversion (IC) making them weakly fluorescent.18ꢀ25 Such
weak fluorescence is essentially a characteristic feature of flexible
that does not require a cofactor.6 GFP’s meteoric rise as a
chromophores like triphenylmethane (TPM) dyes, for which a
molecular imaging tool took place after Roger Tsien’s group
developed GFP variants with much improved biological and
photophysical properties, as well as additional color variants with
large amplitude motion of phenyl rings is the main route of
internal conversion.23,27 The mechanism of IC in p-HBDI has
long been of the subject of debate. While some suggest a single
distinctive emissions ranging from blue and cyan to yellow.7ꢀ10
bond rotation,18 others suggest double bond rotation15,16,19 or
The chromophore responsible for the fluorescence of this protein
is p-hydroxybenzylideneimidazolinone (p-HBDI). The fluores-
cence arises from both its neutral and anionic forms, which is
a simultaneous and concerted rotation of both single and
double bonds known as the hula-twist.20 Weber et al.21 suggest
generated upon ultrafast excited state intramolecular proton
transfer in the neutral form.11,12 One intriguing feature about
the wild type green fluorescent protein (wtGFP) is that while the
chromophore in its intact protein has a quantum yield of ∼0.8 and
Received: July 17, 2011
Revised: October 13, 2011
Published: October 13, 2011
r
2011 American Chemical Society
13733
dx.doi.org/10.1021/jp206815t J. Phys. Chem. A 2011, 115, 13733–13742
|

The Journal of Physical Chemistry A
ARTICLE
Scheme 1. Two GFP Chromophore Analogs DMPI and DPMPI along with the Main GFP Chromophore p-HBDIa
a The p-HBDI analog studied extensively by others is the analog in the middle where the wiggly lines representing protein chains are now only methyl
groups. Same atom labels are used for all analogs.
a possibility of double bond rotation and a hula-twist in the
neutral chromophore, whereas Baldridge et al.22f suggested IC by
twisting of the exomethylene double bond in their alkyl sub-
stituted analogs in octaacid cavitand, while in solution phase, the
IC was by a combination of one bond flip and twisting of the
bridging single bond.24 Usman et al.25 performed ultrafast
polarization sensitive infrared (IR) spectroscopic studies on p-
HBDI. They suggested that both hula-twist and rotation around
the exomethylene double bond can lead to IC. While the double
bond rotation or hula-twist is accepted as the twisting motion
that triggers IC, mechanistic explanations of these energetically
uphill processes are not yet available. Since cisꢀtrans isomeriza-
tion is an obvious outcome of double bond rotation and hula-
twist, both isomers should be detectable at the end of any
photoirradiation.
sample was up-converted in another nonlinear crystal (0.5 mm β
barium borate, θ = 38°, ϕ = 90°) by using the fundamental beam
as a gate pulse. The up-converted light was dispersed in a
monochromator and detected by photon counting electronics.
A cross-correlation function obtained with the use of Raman
scattering from water displayed a full width at half-maximum
(fwhm) of 350 fs. Femtosecond up-conversion measurements
were performed for both the molecules in various glycerol-
methanol mixtures of different viscosities and also in a number
of isoviscous solvents of different polarities. The decays were
deconvoluted using a Gaussian shape for the exciting pulse using
commercial software (IGOR Pro, WaveMetrics). The up-conver-
sion decay transients were measured at their emission maxima,
unless otherwise mentioned. All measurements reported here
were made at 20 °C.
To furnish a deeper understanding of the mechanism of IC, we
designed and synthesized two new analogs of HBDI that differ in
the bulkiness of the electron donor substituent with considerable
design inputs from studies by earlier workers on dimethylamino
group containing fluorophores such as the dansyl group.28 The
other analog was designed to increase the bulkiness and thus act
as a reporter for large amplitude motion. Fluorescence up-
conversion studies and quantum chemical calculations were
performed to understand the excited state relaxation mechanism
of the analogs under consideration. The present study was
undertaken to reveal the nonradiative pathways operational in
GFP chromophore analogs.
2.4. Irradiation Experiment and HPLC Analysis. A solution
of DMPI in acetonitrile was irradiated at 467 nm (absorption at
467 nm ≈ 1.54) using a xenon arc lamp (450 W, 18 V, 25 A,
Ushio code UXL-450SꢀO) in a spectrofluorimeter (Fluorolog
3-21, Horiba Jobin-Yvon) with an excitation slit of 10.0 nm for 11
h. This solution was then injected in an analytical HPLC
(Amersham Biosciences, P-900, C-18 reverse phase column,
150 ꢁ 46 mm, kromasil 700C-18, particle size 5 μm) and
compared with the chromatograms obtained before irradiation.
An isocratic solvent of 65% methanol in water was used as the
eluent and the flow rate was kept at 1.0 mL/min. In a separate
experiment, DMPI (8.0 mg) was dissolved in deutero-chloro-
1
form and its H NMR was recorded in a 400 MHz NMR
spectrometer. The solution in the NMR tube was irradiated
similarly using an excitation slit 5.0 nm for 16 h. Its H NMR
2. EXPERIMENTAL SECTION
1
was recorded in a 400 (400 MHz) JEOL NMR spectrometer and
13C NMR was recorded in a 500 (125 MHz) JEOL NMR
spectrometer.
2.1. Synthesis of GFP Chromophore Analogs. The mol-
ecules (Scheme 1) were synthesized using previously published
procedures29ꢀ33 (see Supporting Information). The two com-
1
pounds were fully characterized by H NMR, 13C NMR and
Mass (available in Supporting Information, Figures S1ꢀS4).
2.2. Steady State Measurements. Steady state absorption
spectra and fluorescence spectra were measured using a com-
mercial UVꢀvis spectrophotometer (JASCO V-550) and fluori-
meter (Fluorolog 3-21, Horiba Jobin-Yvon), respectively. All
measurements have been done at 22 °C.
2.3. Femtosecond Fluorescence Up-conversion Measure-
ments. The fluorescence transients were measured in a femtose-
cond fluorescence up-conversion setup (FOG-100, CDP Corp.).
The details of this setup are discussed elsewhere.34 Briefly, the
sample was excited at 435 nm using the second harmonic of a
mode-locked Ti-sapphire laser (Tsunami, Spectra Physics),
pumped by a 5 W Millennia (Spectra Physics). To generate the
second harmonic, we used a nonlinear crystal (1 mm β barium
borate, θ = 25°, ϕ = 90°). The fluorescence emitted from the
3. THEORETICAL CALCULATIONS
Computation of the first singlet excited state (S1) potential
energy surface of DMPI molecule was done by time-dependent
density functional theory (TDDFT)35 as implemented in the
Gaussian03 software package.36 For all TDDFT calculations 20
excited states were included and were done employing 6-31G**
basis set and B3LYP37 density functional. The potential energy
surface (PES) corresponding to the electronic ground state (S0)
was computed using the density functional theory (DFT) at the
same level. The three-dimensional potential energy surface was
constructed along the dihedral angles γ (C4ꢀC7ꢀC8ꢀC13)
and β(N3ꢀC4ꢀC7ꢀC8) for S0 and S1 states. PES of the ground
state and the first excited state were computed by varying
dihedral angle γ(C4ꢀC7ꢀC8ꢀC13) between ꢀ2° to 98° in
13734
dx.doi.org/10.1021/jp206815t |J. Phys. Chem. A 2011, 115, 13733–13742

The Journal of Physical Chemistry A
ARTICLE
Figure 1. Steady state absorption of the two synthetic GFP chromophore analogs; (a) DMPI and (b) DPMPI in three different solvents:
(1) Cyclohexane (blue line); (2) Methanol (black line); and (3) 50% gylcerolꢀmethanol mixture (red line). Steady state fluorescence spectra of
(c) DMPIand(d) DPMPIinthreedifferent solvents: (1) Cyclohexane (blue line); (2) Methanol (black line); and (3) 50% gylcerolꢀmethanol mixture (red line).
20° increments, and β(N3ꢀC4ꢀC7ꢀC8) from 0° to 180° by
20° increments, including 90°, keeping all of the other degrees of
freedom frozen at the X-ray structure. Thus, a total of 66 ground
state and excited state calculations were performed. Calculations
have been performed in vacuum as well as in acetonitrile medium
(using polarizable continuum model, PCM). Since the excited
state of the molecule has charge transfer character, the potential
energy barriers computed using TDDFT at the S1 excited state
may have quantitative errors.38 Thus, results of TDDFT will only
be used here for obtaining a qualitative picture of the excited state
dynamics. Much accurate coupled cluster methods are highly
computationally demanding for the computation of the entire
potential energy surface.39
methanol is at 533 nm with fluorescence quantum yield of 0.002.
In 50% glycerolꢀmethanol mixture, the emission maximum of
DMPI is red-shifted to 549 nm with a 13-fold increase in
fluorescence quantum yield (Φ = 0.026). For DPMPI, there is
almost no shift in the emission maximum from methanol
(622 nm) to 50% glycerolꢀmethanol (623 nm) because the
extended conjugation causes an increase in electron density in
the σ bond between the N,N-diphenylamine moiety and the
phenyl ring, which renders the former moiety rigid. Both DMPI
and DPMPI also show substantial polarity dependence in their
emission spectra. In DMPI, the emission shows moderate
bathochromic shift from cyclohexane (482 nm) to methanol
(533 nm) and 50% glycerolꢀmethanol (549 nm), while in
DPMPI, there is large bathochromic shift from cyclohexane
(493 nm) to methanol (622 nm). The observed solvatochro-
mism in our analogs is different from that exhibited by
p-HBDI.22e In cyclohexane, a prominent vibrational structure is
observed for both molecules. As the solvent polarity increases,
the vibrational structure vanishes, and the emission maximum
shifts to longer wavelengths. In methanol, DMPI has an emission
maximum at 533 nm and DPMPI is characterized by the
maximum at 622 nm.
4.2. Femtosecond Fluorescence Up-Conversion Study. 4.2.1.
Effect of Viscosity on Excited State Relaxation Dynamics. The
fluorescence transients of both synthetic analogs in various glycerolꢀ
methanol mixtures were measured at their respective emission
maximum and are shown in Figure 2. The logarithmic plots are
shown in Figure S5a of Supporting Information. The decay traces
were fitted by a sum of two exponentials and the resulting time
constants are shown in Table1. For DMPI, in pure methanol, the
two time constants are τ1 = 590 fs and τ2 = 4.3 ps. As we increase the
proportion of glycerol from 0% to 50% glycerol in methanol, τ1
increases to 1.1 ps and τ2 to 10 ps. For DPMPI, τ1 increases from 7.9
to 17.3 ps and τ2 from 37.7 to 81 ps with increase in viscosity from
pure methanol to 50% glycerolꢀmethanol mixture. The viscosity
dependence of the nonradiative decay pathway is often described in
terms of following the power law empirical relationship.27
4. RESULTS AND DISCUSSION
4.1. Steady State Absorption and Emission. The UVꢀvis
absorption spectra of both synthetic GFP chromophore analogs
(DMPI and DPMPI) are characterized by the presence of a single
S1 r S0 absorption band having a maximum located between 430
and 480 nm in a range of solvents as shown in Figure 1 (left panel).
All the absorption spectra have a similar shape, and the position of
the maximum depends on the nature of the substituent groups,
polarity of the solvent and on the extent of hydrogen bonding. The
absorption maximum of DMPI is found to be 460 and 470 nm in
0% to 50% glycerol in methanol, respectively. This bathochromic
shift is assigned to the increase in polarity with increasing propor-
tion of glycerol. Similarly for DPMPI, the shift in absorption
maximum occurs from 460 to 469 nm with increase in glycerol
proportion from 0% to 50%. The absorption maximum of DMPI
shifts from 435 nm in cyclohexane to 460 nm in methanol.
Similarly, DPMPI exhibits a shift in absorption maximum from
449 nm (in cyclohexane) to 460 (in methanol). Further, there is a
significant bathochromic shift as we go from DMPI to DPMPI
probably due to enhanced conjugation.
The fluorescence spectra shown in Figure 1 (right panel) of
both molecules are characterized by a significant Stokes shift
relative to the absorption spectra. The fluorescence intensity of
both molecules shows a strong dependence on solvent viscosity
and increases monotonically with an increase in the solvent
viscosity. Such behavior in these analogs is quite different
compared to p-HBDI.22e The emission maximum of DMPI in
τnr ¼ ηα
ð1Þ
Where η is viscosity of the solvent and α is a factor that gives us
the degree of dependence and whose value varies from 0.1 to
1.0.27 The time constants of both DMPI and DPMPI were
13735
dx.doi.org/10.1021/jp206815t |J. Phys. Chem. A 2011, 115, 13733–13742

The Journal of Physical Chemistry A
ARTICLE
Figure 2. Femtosecond fluorescence up-conversion transients of (a) DMPI and (b) DPMPI in different glycerolꢀmethanol mixtures with varying
viscosity. As the viscosity of the medium increases, the decay becomes slower in both cases.
Table 1. Time Constants τ1 and τ2 of Two Synthetic Analogs,
DMPI and DPMPI Tabulated As a Function of Viscosity of
GlycerolꢀMethanol Mixtures
acetonitrile, and methanol are, respectively, 0.006, 0.005, 0.003,
and 0.002), observed increase in quantum yield with increase in
viscosity, and viscosity dependent lifetime, we assign the long
time component (τ2) to the decay of S1 to S0 state by internal
conversion via a large amplitude motion.16ꢀ21 Weber et al.21
showed the major and dominant relaxation pathway in zwitter-
ionic p-HBDI is the large amplitude motion of benzene ring. If
the excited state of DMPI involves a charge transfer coordinate,
then the molecule will be zwitterionic in nature and hence the
involvement of the large amplitude motion of the substituted
phenyl ring may have a part to play.
4.2.2. Iso-Viscosity Analysis. Fluorescence transients of DMPI
and DPMPI were measured in four polar aprotic solvents of
similar viscosity but different polarities at their respective emis-
sion maximum. The decay transients are shown in Figure 3
(logarithmic plots are shown in Figure S5b of the Supporting
Information) and the fitting parameters are tabulated in Table 2.
In all of the solvents, the immediately formed FranckꢀCondon
(FC) state relaxes within a time scale of a few hundred
femtoseconds to an intermediate state followed by relaxation
to the ground state through internal conversion on a time scale of
glycerol in methanol
DMPI
DPMPI
percentage
viscosity (cP)
τ1 (ps)
τ2 (ps)
τ1 (ps)
τ2 (ps)
0
0.543
1.145
0.59
0.65
0.67
0.86
0.86
1.10
4.3
4.8
7.9
8.2
37.7
40.2
46.5
52.7
65.5
81.0
10
20
30
40
50
2.415
5.4
9.6
5.093
6.7
11.1
14.1
17.3
10.741
22.651
7.7
10.0
plotted against the viscosity of the solvent (See Supporting
Information, Figure S6). In both molecules, the τ1 and τ2
increase with an increase in solvent viscosity. For DMPI, τ1
and τ2 follow power law viscosity dependence with α values of
0.51 and 0.46, respectively. While for DPMPI, the α value for the
two time constants is 0.51. The dependence of τ1 and τ2 time
constants on the microviscosity of glycerolꢀmethanol mixtures
suggests the involvement of a large amplitude motion in
their excited state relaxation dynamics. The relative comparison
of the time constants of these molecules furnishes information
about the effect of substitution on the excited state dynamics of
these molecules. In pure methanol, as we go from DMPI to
DPMPI, wherein the two methyl groups on nitrogen have been
substituted by two bulky phenyl groups, there is a substantial
increase in magnitude of time constants. While the faster time
component τ1 increases from 590 fs to 7.9 ps, the slower time
component τ2 increases from 4.3 to 37 ps. The increase in
magnitude of time constants can be ascribed mainly to the
increased conjugation and also the frictional resistance offered
by solvent molecules to the torsional motion of the N,N-
diphenylamine moiety.
Previously it has been proposed by many authors, that “hula-twist”,
a volume conserving torsional motion of the rings about the
bridging bond is responsible for the excited state relaxation of
GFP chromophore and its analogs.10,11,23 The experimental
observation of the analogs considered in this work does not
support the involvement of hula-twist as a primary relaxation
coordinate, because of the evidence of large amplitude motion.
On the basis of the low quantum yield of DMPI in nonviscous
solvents (quantum yields of DMPI in cyclohexane, ethyl acetate,
sub-hundred picoseconds. For DMPI, in cyclohexane (λem
=
482 nm), time constant τ1 is 850 fs and as we change the
dielectric media from cyclohexane (ε = 2) to acetonitrile (ε = 35),
τ1 decreases to 650 fs, while the dependence of τ2 is not
monotonous. The dependence of time constant τ1 on the
polarity of the solvent implies the decay of the FC state to a
relatively relaxed state, whose stability is a function of solvent
polarity. As the solvent polarity increases, the solvent molecules
stabilize that state in such a way that the rate of depletion of the
FC state increases, leading to a faster time constant. This is
possible only if that state has a charge transfer character and will
be more stable in more polar solvents. It is worth mentioning that
this time constant also depends on solvent viscosity and on the
extent of substitution (ca. DMPI vs DPMPI).
These observations indicate the involvement of amplitude
motion in the system, during the transfer of charge between
donor and acceptor moiety. In order to justify the viscosity
dependence of both the time constants, we propose a multi-
coordinate relaxation mechanism of the excited state. Upon
photoexcitation, the locally excited (LE) state depletes to the
charge transfer (CT) state via the rotational motion of N-sub-
stituted amine moiety attached to the phenyl ring. This provides
rigidity to the σ bond between the amine group and the phenyl
ring and hence the charge transfer time constant is prone to the
13736
dx.doi.org/10.1021/jp206815t |J. Phys. Chem. A 2011, 115, 13733–13742

The Journal of Physical Chemistry A
ARTICLE
Figure 3. Femtosecond fluorescence up-conversion transients of (a) DMPI and (b) DPMPI in four isoviscous, aprotic solvents having different values
of dielectric constant; cyclohexane (blue O), chloroform (green (), acetone (black +), and acetonitrile (red 3).
Table 2. Polarity Dependence of Time Constants τ1 and τ2 of DMPI and DPMPI in Four Isoviscous, Aprotic Solvents
(Cyclohexane, Chloroform, Acetone, and Acetonitrile) of Different Dielectric Constantsa
solvent details
viscosity (cP)
DMPI
DPMPI
solvent
dielectric constant
τ1 (fs)
τ2 (ps)
τ1 (ps)
τ2 (ps)
cyclohexane
chloroform
acetone
0.89
2.00
4.72
850 (0.53)
810 (0.51)
790 (0.48)
650 (0.50)
590 (0.50)
730 (0.55)
800 (0.55)
8.8 (0.47)
5.5 (0.49)
8.7 (0.52)
7.2 (0.50)
4.3 (0.50)
6.3 (0.45)
6.4 (0.45)
4.6 (0.52)
3.9 (0.68)
5.1 (0.68)
2.0 (0.69)
7.9 (0.70)
9.1 (0.55)
10.7 (0.4)
96 (0.48)
39 (0.32)
61 (0.32)
75 (0.31)
37 (0.30)
57 (0.45)
85 (0.60)
0.54
0.30
20.56
35.95
32.63
24.35
17.43
acetonitrile
methanol
ethanol
0.341
0.543
1.087
2.61
n-butanol
a The other three solvents (methanol, ethanol, and n-butanol) depict the dependence of time constants on the hydrogen bond donating capability of
solvents. The parameters in parentheses represent the amplitude of the respective time constants.
Figure 4. (a)Intensity normalized time-resolved emission spectra (TRES) of DMPI in acetonitrile, constructed from the decay transients obtained at
ten different wavelengths. The time-resolved emission spectra were plotted for 0 ps (red b), 0.1 ps (black, 9), 0.5 ps (green, 2), 1 ps (blue,1), 3 ps
(pink, () and 5 ps (orange, +). (b) Time-resolved area normalized emission spectra (TRANES) of DMPI in acetonitrile. The symbols have their usual
meaning as mentioned for TRES. The TRANES reports the occurrence of an isoemissive point and hence implies the existence of two states in the
excited state deactivation process.
viscosity of the medium. The involvement of the torsional
motion of N-substituted amine moiety also explains the large
magnitude of the CT time component in DPMPI than DMPI.
Subsequently, the CT state relaxes via conical intersection to the
ground state along the torsional motion of N-substituted
phenylamine moiety about the exomethylenic double bond,
paving the path for the cisꢀtrans isomerization.
4.2.3. TRES and TRANES Measurements. Fluorescence up-
conversion transients of DMPI in acetonitrile were recorded
at eleven different wavelengths throughout the emission
13737
dx.doi.org/10.1021/jp206815t |J. Phys. Chem. A 2011, 115, 13733–13742

The Journal of Physical Chemistry A
ARTICLE
spectrum with the relaxation dynamics exhibiting substantial
emission wavelength dependence. As already mentioned, the
excited state relaxation of DMPI involves two time constants.
The faster time constant can be the solvent relaxation or any
other excited state reaction like intramolecular charge transfer,
while the longer time constant corresponds to the internal
conversion via large amplitude motion from S1 to S0 potential
energy surface. The time-resolved intensity normalized and area
normalized emission spectra (TRES and TRANES) were con-
structed using the parameters of best fit to the fluorescence
decays and the steady state emission spectra.40,41 Figure 4a shows
the intensity normalized time-resolved emission spectra (TRES)
of DMPI constructed at six times viz., 0, 0.1, 0.5, 1, 3, and 5 ps.
There is a continuous shift in the wavenumber corresponding to
maximum intensity up to 5 ps, which marks the limit of any
depletion of LE state and formation of CT state, with an overall
Stokes shift of 1000 cmꢀ1. Constructed TRANES (Figure 4b) is
characterized by the presence of an iso-emissive point represent-
ing the equilibrium between the two states viz., locally excited FC
state on the higher wavenumber side and charge transfer CT state
on the lower wavenumber state. As the time evolves from 0 to 5
ps, the intensity of the LE state decreases and that of the CT state
increases, after which there is practically no change in the
intensity of either of the states. The occurrence of a single iso-
emissive point in the TRANES of DMPI suggests a two state
decay kinetics representing the dielectric dependent decay of LE
state to CT state within 3ꢀ5 ps, followed by the depletion of
TICT state to the ground state through internal conversion via
large amplitude motion.
4.2.4. Effect of Intermolecular Hydrogen Bonding on Relaxa-
tion Behavior. Intermolecular hydrogen bonding is one of the
significant nonradiative decay pathways of the S1 state in a large
number of chromophores especially containing a hydrogen
bond donor or acceptor group. We measured the fluorescence
transients of both the synthetic analogs in alcoholic solutions of
different hydrogen bond donating capabilities, but with almost
similar viscosities and polarities. In methanol, for DMPI, the CT
time constant τ1 = 590 fs and IC time constant τ2 = 4.3 ps.
Going down to n-butanol in the series, τ1 increased to 807 fs and
τ2 incremented to 6.4 ps (Table 2). A similar type of increase is
observed for DPMPI. The more viscous nature of higher order
alcohols can also induce an increase in the magnitude of the
time constants. However, after subtracting the viscosity effect
from the experimentally obtained parameters, the time con-
stants are found to depend on the H-bonding ability of the
solvents as well. The reason is ascribed to the presence of
intermolecular hydrogen bonding between solvent molecules
and the chromophore.
Similar hydrogen bonding dependent dynamics have also been
proposed by Vauthey and co-workers,23 wherein the charge
transfer results in an increase of electron density on the carbonyl
group for other GFP chromophore analogs. This enhances the
acidity of amino group and the basicity of carbonyl group, which
in turn leads to an enhancement of hydrogen bonding ability in
the excited state.42ꢀ44 The stretching vibrational modes of the
hydrogen bonds act as a sink for the S1 f S0 nonradiative decay
and thus in this way the electronic energy gets dissipated among
the vibrational modes of the hydrogen bonds.23 The increase of
hydrogen bonding ability while going up from n-butanol to
methanol induces stronger intermolecular hydrogen bonding
interactions, which leads to an acceleration of the S1 state decay
and consequently a decrease of time constants. A similar
Figure 5. (a) Calculated three dimensional potential energy surface of
the S1 state of DMPI as a function of dihedral angles β and γ. The S0
surface is dropped for clarity of the picture. (b) Energy profiles of the S0
and S1 states as a function of β (rotation about C4ꢀC7 double bond),
keeping γ fixed at ꢀ2°.
conclusion is reached in the case of methanol and acetonitrile.
Both solvents have almost similar viscosities and dielectric
properties, but differ in their hydrogen bond donating abilities
(methanol is a more efficient hydrogen bond donor than
acetonitrile). The S1 relaxation time constant (τ2) of DMPI
decreases from 7.2 ps in acetonitrile to 4.3 ps in methanol, which
is a direct consequence of intermolecular hydrogen bonding. The
similar gradation is observed for DPMPI as well.
The results of steady state fluorescence and femtosecond up-
conversion measurements comprehend the viscosity-depen-
dent decay of the excited state. The strong viscosity dependence
suggests a large amplitude motion of some bulky groups playing
a role in the internal conversion process. For some GFP
chromophore analogs, it has already been proposed that tor-
sional motion along the exomethylene double bond15,16,19 is the
primary relaxation coordinate for internal conversion, which moti-
vated us to look for the nonradiative decay channel responsible for
the depletion of the excited state by theoretical calculations.
4.3. Potential Energy Surface (PES) Calculations. The
potential energy surfaces of the S0 and S1 states of DMPI were
calculated in vacuum and are shown in Figure 5. These surfaces
are observed to form a “conical intersection”16,19c,45 when β is
90°, and the coordinates representing the point are (γ,β) =
(ꢀ2°,90°) labeled by AC in Figure 5a. At the dihedral angles of
ground state optimal values [(γ,β) = (ꢀ2°,0°)] is a local
minimum (FC) on the S1 surface. The dihedral angles (γ,β) =
(90°,0°) and (90°,180°) represent two other minimum energy
13738
dx.doi.org/10.1021/jp206815t |J. Phys. Chem. A 2011, 115, 13733–13742

The Journal of Physical Chemistry A
ARTICLE
Figure 6. HPLC chromatograms of DMPI (a) before and (b) after irradiation at 467 nm for 11 h in acetonitrile. An isocratic 65% methanol in water was
used as the eluent on a C-18 reverse phase analytical column with a flow rate 1 mL/min.
structures on the S1 surface (labeled M0 and M00, respectively),
and are nearly degenerate with FC with an energy difference of
only 0.25 kcal/mol. The potential energy barrier along the
minimum energy pathway (MEP) for FC f AC transition, as
indicated by the dotted lines in Figure 5a, is about 10 kcal/mol
(see also Figure 5b). Most interestingly, the MEP has no change
along the γ coordinate, aligned parallel to the β coordinate axis.
The position of the transition state along the MEP is at β = 60°.
The forward and reverse barrier separating FC and M0 is
estimated to be less than a kT at room temperature. However
the gap between the S0 and S1 states is too large (57.7 kcal/mol)
to provide coupling between these two states (Figure S7a,
Supporting Information), for which it appears that rotation along
the C7ꢀC8 single bond cannot lead to internal conversion. The
minimum M00 is kinetically inaccessible due to large potential
energy barriers separating FC/M0 and M00. Since the S1 potential
energy approaches the S0 surface and forms a conical intersection
at β = 90°, one may expect Z-E isomerization about the
exomethylene double bond (C4ꢀC7) of DMPI. The S0 surface
reveals that the E-isomer is 11 kcal/mol less stable than the
Z-isomer (Figure 5b).
The excited state charge transfer in the molecule introduces a
partial double bond character to the C7ꢀC8 bond, thus the
barrier going from M0 to M00 is expected to be much larger than
that estimated by TDDFT calculations (see Supporting Informa-
tion Figure S7b). We believe that this is the result of the artifact of
TDDFT in describing the charge-transfer at the excited states
while employing the ground state density functionals. It is known
that the excitation energy is underestimated by TDDFT in such
cases.38 The results of the theoretical calculations alone seem to
contradict the internal conversion by the single bond rotation
involving the phenyl ring in the model systems of HBDI, as
suggested by many authors18 including us. Overall, this strongly
suggests IC by rotation along the exomethylene double bond
(C4ꢀC7) as the main relaxation coordinate. This also clearly
rules out a “hula twist” mechanism for internal conversion in
these molecules. Voityuk and co-workers19c found a 3.2 kcal/mol
energy barrier for the double-bond rotation in the cation and a 15
kcal/mol energy gap for the double-bond rotation in the anion of
a model system of HBDI, leading them to propose that IC is
indeed possible in both cases. We also have calculated the S0 and
S1 potential energy surfaces using polarizable continuum model
(PCM) with acetonitrile as a solvent. Incorporation of the
implicit solvent effect does not alter the main conclusions derived
in vacuo (see Figure S8, Supporting Information).
4.4. HPLC and NMR Analysis. HPLC chromatograms of
Z-DMPI at 270 and 470 nm obtained from a solution in
acetonitrile show one absorption peak at 10.1 min (Figure 6a).
This solution was irradiated at 467 nm by a xenon short arc lamp
(450 W) for 11 h. HPLC chromatograms at 270 and 470 nm of
this irradiated solution show two close but separated peaks at 10.3
and 12.7 min (Figure 6b) with approximate area of 60% and 40%,
respectively. This observation suggests isomerization about the
exomethylene double bond and suggests the presence of Z-DMPI
(60%) and E-DMPI (40%) in the solution. Further evidence of
isomerization about the exomethylene double bond was obtained
from analysis of 1H and 13C NMR of the irradiated sample.
1
Before irradiation, H NMR of a solution of Z-DMPI in
deutero-chloroform shows all of the protons (Figure 7a)—two
singlets from the methylamine and dimethylamine protons, one
aromatic doublet from the Ha protons, one olefinic singlet from
the Hc proton, two aromatic multiplets from the Hd, He, and Hf
protons and one aromatic doublet from the Hb protons. On the
basis of the integration of protons and coupling constants in 1H
NMR, 13C NMR chemical shifts, (Figure S3 and S9, Supporting
Information), and crystal structure,29 DMPI has been authenti-
cated as a single compound, with Z-configuration. After irradia-
tion, each of the above signals is appearing as doublet except the
N-methyl group singlet (Figure 7b). The appearance of the dual
signals is better understood from the integrations of the signals
(Figure S10, Supporting Information). Although the N-methyl
group is giving only one singlet, its integration is equivalent to
five protons (Figure S10, Supporting Information), indicating
overlap of two N-methyl groups of two isomeric Z-DMPI and E-
DMPI. A new aromatic doublet from the Hb0 protons is shifted
further downfield because of the possibility of formation of an
intramolecular hydrogen bond with the carbonyl oxygen
atom.29 A new olefinic singlet from the Hc0 proton is also
observed downfield. The ratio of the integrations of the Hb0
protons to Hb protons or He0 protons to He protons or Hc0
protons to Hc protons shows isomerization of 43% Z-DMPI to
E-DMPI. The 13C NMR shows the presence of all of the
equivalent carbon atoms (Figure S11, Supporting Information)
of two isomeric DMPI.
The experimental observation of the existence of Z- and E-
isomers on irradiation and the existence of conical intersection
13739
dx.doi.org/10.1021/jp206815t |J. Phys. Chem. A 2011, 115, 13733–13742

The Journal of Physical Chemistry A
ARTICLE
1
Figure 7. (a) H NMR (400 MHz, CDCl3) of Z-DMPI before irradiation. The signals are assigned to protons as labeled in the structures of
Z-DMPI and E-DMPI shown in (c). (b) 1H NMR (400 MHz, CDCl3) of Z-DMPI and E-DMPI after irradiation of Z-DMPI in CDCl3 at 467 nm for 16 h.
The signals are assigned to protons as labeled in structures of Z-DMPI and E-DMPI.
to internal conversion via rotation about the latter bond. Further-
more, the proton at the para-position of the benzene ring attached
to the 2-position of the imidazolin-5-one ring is shifted upfield as
compared to the protons at the ortho- and the meta-positions,
suggesting buildup of electron density at the para-position, which
may be attributed to the charge transfer from the N,N-substituted
amine.
Scheme 2. In DMPI, Intramolecular Charge Transfer Occurs
First for Which the Exo-Methylene Double Bond (C4ꢀC7)
Becomes Predominantly a Single Bond (Right Side)a
The excited state dynamics of DMPI in acetonitrile can thus be
summarized as shown in Scheme 3. Excitation of the Z-isomer
forms the locally excited (LE) state which undergoes an ultrafast
intramolecular charge transfer along the rotational coordinate of
N-substituted amine on a time scale of ∼650 fs to form the
twisted intramolecular charge transfer (TICT) state. The TICT
state subsequently undergoes twisting with a 7 ps time compo-
nent about the exomethylenic double bond (C4ꢀC7) having a
reduced double-bond character and reaches the region of conical
intersection with approximately perpendicular geometry. The
torsionally relaxed excited state crosses to the “hot ground state”
of either of the Z- or E-isomer. The vibrationally excited ground
a After charge transfer, DMPI can now undergo rotation about the
resulting C4ꢀC7 single bond (dihedral angle β).
between the excited and the ground state substantiate the con-
sideration ofinvolvementof torsional motion of the exomethylenic
double bond as the dominant excited state relaxation coordinate
for the present analogs. The charge transfer provide rigidity to the
σ bond between amine and phenyl moiety and also reduces double
bond character of the exomethylenic bond (Scheme 2) leading
13740
dx.doi.org/10.1021/jp206815t |J. Phys. Chem. A 2011, 115, 13733–13742

The Journal of Physical Chemistry A
ARTICLE
Scheme 3. Schematic Excited State Dynamics of DMPI in
Acetonitrile
’ ACKNOWLEDGMENT
We would like to dedicate this work to our colleague Prof. N.
Sathyamurthy on his 60th birthday. R.G. thanks DST and ISRO
for partially supporting this research. S.R. and B.K.R. thank CSIR,
India for a junior research fellowship (JRF) and senior research
fellowship (SRF), respectively. We also gratefully acknowledge
DST-FIST for the computational facility. We thank Professor
KankanBhattacharyya for providingthe femtosecondfluorescence
up-conversion facility required for the present study under De-
partment of Science and Technology (DST), India. This work is
financially supported by the Board of Research in Nuclear Sciences
(BRNS), Department of Atomic Energy, Government of India.
’ REFERENCES
(1) Chalfie, M.; Tu, Y.; Euskirchen, G.; Ward, W. W.; Prasher, D. C.
Science 1994, 263, 802–805.
(2) Prasher, D.; McCann, R. O.; Cormier, M. J. Biochem. Biophys. Res.
Commun. 1985, 126, 1259–1268.
(3) Tsien, R. Y. Annu. Rev. Biochem. 1998, 67, 509–544.
(4) Shaner, N. C.; Lin., M. Z.; McKeown, M. R.; Steinbach, P. A.;
Hazelwood, K. L.; Davidson, M. W.; Tsien, R. Y. Nat. Methods 2008,
5, 545–551.
state finally relaxes to the minimum of ground state potential
energy surface by thermal equilibration.
(5) Shaner, N. C.; Steinbach, P. A.; Tsein, R. Y. Nat. Methods 2005,
2, 905–909.
5. CONCLUSIONS
(6) Prendergast, F. G. Methods Cell Biol. 1998, 58, 1–18.
(7) Heim, R.; Cubitt, A. B.; Tsien, R. Y. Nature 1995, 373, 663–664.
(8) Heim, R.; Tsien, R. Y. Curr. Biol. 1996, 6, 178–182.
(9) Heim, R.; Prasher, D. C.; Tsien, R. Y. Proc. Natl. Acad. Sci. U.S.A.
1994, 91, 12501–12504.
(10) Orm€o, M.; Cubitt, A. B.; Kallio, K.; Gross, L. A.; Tsein, R. Y.;
Remington, S. J. Science 1996, 273, 1392–1395.
(11) Chattoraj, M.; King, B. A.; Bublitz, G. U.; Boxer, S. G. Proc. Natl.
The observed solvent dependent (viz., polarity, viscosity, etc.)
emission characteristics of both the analogs lead us to propose a
multicoordinate excited state relaxation mechanism. The first time
component is assigned to twisted intramolecular charge transfer
(TICT) corresponding to the depletion of LE to CT state via the
rotational motion of N-substituted amine group. In case of
DPMPI, due to extended conjugation, the excited state charge
transfer leads to more rigid σ bond between N-substituted amine
and the phenyl group and hence increases the magnitude of the
time constant. The TICT state then decays to the point of conical
intersection via the torsional motion of N-substituted phenyla-
minemoiety about theexomethylenic doublebondaspredictedby
theoretical calculations. The twist about the exomethylenic double
bond leads to cisꢀtrans isomerization as evident from HPLC and
1H NMR studies. The torsional motion along the single bond
cannot lead to IC because of very high S0ꢀS1 energy gaps, which
does not allow any coupling between the two states. The TICT
and IC deactivation channels make DMPI and DPMPI very
weakly fluorescent. From this study, an analogy may be established
between the reported analogs and p-HBDI to understand the role
of the protein scaffold in suppressing the nonradiative pathways
leading to highly fluorescent nature of wild-type GFP.
Acad. Sci. U.S.A. 1996, 93, 8362–8367.
(12) Brejc, K.; Sixma, T. K.; Kitts, P. A.; Kain, S. R.; Tsien, R. Y.;
Orm€o, M.; Remington, S. J. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 2306–2311.
(13) Meech, S. R. Chem. Soc. Rev. 2009, 38, 2922–2934.
(14) Vengris, M.; van Stokkum, I. H. M.; He, X.; Bell, A. F.; Tonge,
P. J.; van Grondelle, R.; Larsen, D. S. J. Phys. Chem. A. 2004, 108,
4587–4598.
(15) Niwa, H.; Inouye, S.; Hirano, T.; Matsuno, T.; Kojima, S.;
Kubota, M.; Ohashi, M.; Tsuji, F. I. Proc. Natl. Acad. Sci. U. S. A. 1996,
93, 13617–13622.
(16) Follenius-Wund, A.; Bourotte, M.; Schmitt, M.; Iyice, F.; Lami,
H.; Bourguignon, J.-J.; Haiech, J.; Pigault, C. Biophys. J. 2003, 85,
1839–1850.
(17) (a) Leiderman, P.; Genosar, L.; Huppert, D.; Shu, X.; Remington,
S. J.; Solntsev, K. M.; Tolbert, L. M. Biochemistry 2007, 46, 12026–12036.
(b) Kummer, A. D.; Kompa, C.; Lossau, H.; P€ollinger-Dammer, F.; Michel-
Beyerle, M. E.; Silva, C. M.; Bylina, E. J.; Coleman, W. J.; Yang, M. M.;
Youvan, D. C. Chem. Phys. 1998, 237, 183–193. (c) Kummer, A. D.;
Wiehler, J.; Rehaber, H.; Kompa, C.; Steipe, B.; Michel-Beyerle, M. E.
J. Phys. Chem. B 2000, 104, 4791–4798. (d) Kummer, A. D.; Wiehler, J.;
Sch€uttrigkeit, T. A.; Berger, B. W.; Steipe, B.; Michel-Beyerle, M. E.
Chembiochem 2002, 3, 659–663. (e) Kummer, A. D.; Kompa, C.; Niwa,
H.; Hirano, T.; Kojima, S.; Michel-Beyerle, M. E. J. Phys. Chem. B 2002,
106, 7554–7559. (f) Chen, M. C.; Lambert, C. R.; Urgitis, J. D.; Zimmer, M.
Chem. Phys. 2001, 270, 157–164. (g) Mauring, K.; Krasnenko, V.; Miller, S.
J. Lumin. 2007, 122ꢀ123, 291–293.
’ ASSOCIATED CONTENT
S
Supporting Information. Detailed synthetic procedures,
b
NMR spectral characterization of compounds and calculated
potential energy surfaces, as mentioned in the text. This material
is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
(18) (a) Martin, M. E.; Negri, F.; Olivucci, M. J. Am. Chem. Soc.
2004, 126, 5452–5464. (b) Rajbongshi, B. K.; Sen, P.; Ramanathan, G.
Chem. Phys. Lett. 2010, 494, 295–300. (c) Stavrov, S. S.; Solntsev, K. M.;
Tolbert, L. M.; Huppert, D. J. Am. Chem. Soc. 2006, 128, 1540–1546.
(19) (a) Gepshtein, R.; Huppert, D.; Agmon, N. J. Phys. Chem. B
2006, 110, 4434–4442. (b) Altoe, P.; Bernardi, F.; Garavelli, M.;
Orlandi, G.; Negri, F. J. Am. Chem. Soc. 2005, 127, 3952–3963.
Corresponding Author
*Fax: +91-512-2597436; E-mail: gurunath@iitk.ac.in (G.R.);
psen@iitk.ac.in (P.S.).
Author Contributions
†These authors contributed equally to this work.
13741
dx.doi.org/10.1021/jp206815t |J. Phys. Chem. A 2011, 115, 13733–13742

The Journal of Physical Chemistry A
ARTICLE
(c) Voityuk, A. A.; Michel-Beyerle, M.-E.; R€osch, N. Chem. Phys. Lett.
1998, 296, 269–276. (d) Yang, J.-S.; Huang, G.-J.; Liu, Y.-H.; Peng, S.-M.
Chem. Commun. 2008, 1344–1346. (e) Solntsev, K. M.; Poizat, O.;
Dong, J.; Rehault, J.; Lou, Y.; Burda, C.; Tolbert, L. M. J. Phys. Chem. B
2008, 112, 2700–2711.
Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A. Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.;
Millam, J. M.; Klene,M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;
Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich,
S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.;
Fox, D. J. Gaussian 03, rev C.02; Gaussian, Inc.: Wallingford, CT, 2004.
(37) (a) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988,
37, 785–789. (b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(38) (a) Autschbach, J. Chem. Phys. Chem. 2009, 10, 1757–1760.
(b) Niehaus, T. A.; March, N. H. Theor. Chem. Acc. 2010, 125, 427–432.
(c) List, N. H.; Olsen J. M.; Rocha-Rinza, T.; Christiansen, O.; Kongsted,
J. Int. J. Quantum Chem. 2011, published online (DOI: 10.1002/
qua.23059).
(20) (a) Webber, N. M.; Litvinenko, K. L.; Meech, S. R. J. Phys.
Chem. B 2001, 105, 8036–8039. (b) Megley, C. M.; Dickson, L. A.;
Maddalo, S. L.; Chandler, G. J.; Zimmer, M. J. Phys. Chem. B 2009,
113, 302–308. (c) Baffour-Awuah, N. Y. A.; Zimmer, M. Chem. Phys.
2004, 303, 7–11. (d) Litvinenko, K. L.; Webber, N. M.; Meech, S. R.
Chem. Phys. Lett. 2001, 346, 47–53. (e) Mandal, D.; Tahara, T.; Webber,
N. M.; Meech, S. R. Chem. Phys. Lett. 2002, 358, 495–501. (f) Mandal,
D.; Tahara, T.; Meech, S. R. J. Phys. Chem. B 2004, 108, 1102–1108.
(g) Litvinenko, K. L.; Webber, N. M.; Meech, S. R. J. Phys. Chem. A 2003,
107, 2616–2623. (h) Maddalo, S. L.; Zimmer, M. Photochem. Photobiol.
2006, 82, 367–372. (i) Litvinenko, K. L.; Webber, N. M.; Meech, S. R.
Bull. Chem. Soc. Jpn. 2002, 75, 1065–1070. (j) Toniolo, A.; Olsen, S.;
Manohar, L; Martínez, T. J. Faraday Discuss. 2004, 127, 149–163.
(21) Weber, W.; Helms, V.; McCammon, J. A.; Langhoff, P. W. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 6177–6182.
(22) (a) Branchini, B. R.; Lusins, J. O.; Zimmer, M. J. Biomol. Struct.
Dyn. 1997, 14, 441–448. (b) Voliani, V.; Bizzarri, R.; Nifosì, R.;
Abbruzzetti, S.; Grandi, E.; Viappiani, C.; Beltram, F. J. Phys. Chem. B
2008, 112, 10714–10722. (c) Chen, K.-Y.; Cheng, Y.-M.; Lai, C.-H.;
Hsu, C.-C.; Ho, M.-L.; Lee, G.-H.; Chou, P.-T. J. Am. Chem. Soc. 2007,
129, 4534–4535. (d) Dong, J.; Solntsev, K. M.; Poizat, O.; Tolbert, L. M.
J. Am. Chem. Soc. 2007, 129, 10084–10085. (e) Dong, J.; Solntsev, K. M.;
Tolbert, L. M. J. Am. Chem. Soc. 2006, 128, 12038–12039. (f) Baldridge,
A.; Samanta, S. R.; Jayaraj, N.; Ramamurthy, V.; Tolbert, L. M. J. Am.
Chem. Soc. 2011, 133, 712–715. (g) Olsen, S.; Smith, S. C. J. Am. Chem.
Soc. 2008, 130, 8677–8689.
(39) Kurashige, Y.; Nakajima, T.; Kurashige, S.; Hirao, K. J. Phys.
Chem. A 2007, 111, 5544–5548.
(40) (a) Maroncelli, M.; Fleming, G. R. J. Chem. Phys. 1987, 86,
6221. (b) Rafiq, S.; Yadav, R.; Sen, P. J. Phys. Chem A 2011, 115, 8335–
8343.
(41) (a) Koti, A. S. R.; Krishna, M. M. G.; Periasamy, N. J. Phys.
Chem. A 2001, 105, 1767–1771. (b) Koti, A. S. R.; Periasamy, N. Proc.
Ind. Acad. Sci. 2001, 113, 157–163. (c) Koti, A. S. R.; Periasamy, N.
J. Chem. Phys. 2001, 115, 7094–7099.
(42) Sherin, P. S.; Grilj, J.; Tsentalovich, Y. P.; Vauthey, E. J. Phys.
Chem. B 2009, 113, 4953–4963.
(23) Petkova, I.; Dobrikov, G.; Banerji, N.; Duvanel, G.; Perez, R.;
Dimitrov, V.; Nikolov, P.; Vauthey, E. J. Phys. Chem. A 2010, 114, 10–20.
(24) Baldridge, A.; Samanta, S. R.; Jayaraj, N.; Ramamurthy, V.;
Tolbert, L. M. J. Am. Chem. Soc. 2010, 132, 1498–1499.
(43) F€urstenberg, A.; Vauthey, E. Photochem. Photobiol. Sci. 2005,
4, 260–267.
(44) Nishiya, T.; Yamauchi, S.; Hirota, N.; Baba, M.; Hanazaki, I.
J. Phys. Chem. 1986, 90, 5730–5735.
(25) Usman, A.; Mohammed, O. F.; Nibbering, E. T. J.; Dong, J.;
Solntsev, K. M.; Tolbert, L. M. J. Am. Chem. Soc. 2005, 127, 11214–11215.
(26) (a) Tolbert, L. M.; Baldridge, A.; Kowalik, J.; Solntsev, K. M.
Acc. Chem. Res. DOI: 10.1021/ar2000925. (b) Bravaya, K. B.; Grigorenko,
B. L.; Nemukhin, A. V.; Krylov, A. A. I. Acc. Chem. Res. DOI: 10.1021/
ar2001556.
(45) Levine, B. G.; Martínez, T. J. Annu. Rev. Phys. Chem. 2007,
58, 613–634.
(27) (a) Velsko, S. P.; Fleming, G. R. Chem. Phys. 1982, 65, 59–70.
(b) Duxbury, D. F. Chem. Rev. 1993, 93, 381–433. (c) Ben-Amotz, D.;
Harris, C. B. J. Chem. Phys. 1987, 86, 4856–4870. (d) Sundstr€om, V.;
Gillbro, T.; Bergst€orm, H. Chem. Phys. 1982, 73, 439–458. (e) Rafiq, S.;
Yadav, R.; Sen, P. J. Phys. Chem. B 2010, 114, 13988–13994.
(28) Diwu, Z.; Lu, Y.; Zhang, C.; Klaubert, D. H.; Haugland, R. P.
Photochem. Photobiol. 1997, 66, 424–431.
(29) Rajbongshi, B. K.; Ramanathan, G. J. Chem. Sci. 2009,
121, 973–982.
(30) Badr, M. Z.; El-Sherief, H. A. H.; Tadros, M. E. Bull. Chem. Soc.
Jpn. 1982, 55, 2267–2270.
(31) Bhattacharjya, G.; Agasti, S. S.; Ramanathan, G. ARKIVOC
2006, 10, 152–161.
(32) Hoshina, H.; Tsuru, H.; Kubo, K.; Igarashi, T.; Sakurai, T.
Heterocycles 2000, 53, 2261–2274.
(33) Mukerjee, A. K.; Kumar, P. Heterocycles 1981, 16, 1995–2034.
(34) Sen, P.; Roy, D.; Mondal, S. K.;Sahu, K.; Ghosh, S.; Bhattacharyya,
K. J. Phys. Chem. A 2005, 109, 9716–9722.
(35) (a) Jamorski, C.; Casida, M. E.; Salahub, D. R. J. Chem. Phys.
1996, 104, 5134–5147. (b) Petersilka, M.; Gossmann, U. J.; Gross,
E. K. U. Phys. Rev. Lett. 1996, 76, 1212–1215. (c) Bauernschmitt, R.;
Ahlrichs, R.; Hennrich, F. H.; Kappes, M. M. J. Am. Chem. Soc. 1998,
120, 5052–5059. (d) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub,
D. R. J. Chem. Phys. 1998, 108, 4439–4449. (e) Stratmann, R. E.;
Scuseria, G. E.; Frisch, M. J. J. Chem. Phys. 1998, 109, 8218–8224.
(36) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;
13742
dx.doi.org/10.1021/jp206815t |J. Phys. Chem. A 2011, 115, 13733–13742