Tuning the excited-state dynamics of GFP-inspired imidazolone derivatives

Article abstract of DOI:10.1021/jp903900b

The excited-state dynamics of five derivatives of the GFP-chromophore, which differ by the position and nature of their substituents, has been investigated in solvents of various viscosity and polarity and in rigid media using femtosecond-resolved spectro

Full text of DOI:10.1021/jp903900b

10
J. Phys. Chem. A 2010, 114, 10–20
Tuning the Excited-State Dynamics of GFP-Inspired Imidazolone Derivatives
Irina Petkova,*,†,‡ Georgi Dobrikov,† Natalie Banerji,‡ Guillaume Duvanel,‡ Robert Perez,‡
Vladimir Dimitrov,† Peter Nikolov,† and Eric Vauthey*,‡
Institute of Organic Chemistry with Centre of Phytochemistry, Bulgarian Academy of Sciences, Akad. G.
BontcheV str. Bl. 9, Sofia 1113, Bulgaria, and Department of Physical Chemistry, UniVersity of GeneVa,
30 Quai Ernest-Ansermet, CH-1211 GeneVa 4, Switzerland
ReceiVed: April 28, 2009; ReVised Manuscript ReceiVed: September 22, 2009
The excited-state dynamics of five derivatives of the GFP-chromophore, which differ by the position and
nature of their substituents, has been investigated in solvents of various viscosity and polarity and in rigid
media using femtosecond-resolved spectroscopy. In polar solvents of low viscosity, like acetonitrile or methanol,
the fluorescence decays of all compounds are multiexponential, with average lifetimes of the order of a few
picoseconds, whereas in rigid matrices (polymer films and low temperature glasses), they are single exponential
with lifetimes of the order of a few nanoseconds and fluorescence quantum yields close to unity. Global
analysis of the fluorescence decays recorded at several wavelengths and of the transient absorption spectra
reveals the presence of several excited-state populations with slightly different fluorescence and absorption
spectra and with distinct lifetimes. These populations are attributed to the existence of multiple ground-state
conformers. From the analysis of the dependence of the excited-state dynamics on the solvent and on the
nature of the substituents, it follows that the nonradiative deactivation of all these excited chromophores
involves an intramolecular coordinate with large amplitude motion. However, depending on the solvent and
substituent, additional channels, namely, inter- and intramolecular hydrogen bond assisted nonradiative
deactivation, are operative. This allows tuning of the excited-state lifetime of the chromophore. Finally, an
ultrafast photoinduced intramolecular charge transfer is observed in polar solvents with one derivative bearing
a dimethylaminophenyl substituent.
1. Introduction
viscosity dependence in moderately viscous solvents.13,16 On the
basis of this and on quantum chemistry calculations, it has been
proposed that p-HBDI relaxes nonradiatively to the ground state
via a concerted, volume-preserving, twist motion around both
bridging bounds, the so-called “hula-twist”.15,16 On the other
hand, substantial viscosity dependence was reported in more
viscous solvents, such as polyalcohols.29 Additionally to being
ultrafast, the excited-state dynamics of p-HBDI was also found
to be biexponential, with time constants that are the same
throughout the emission spectrum but with wavelength-depend-
ent relative amplitudes.14-16 This effect has been attributed to
the existence of various conformations of the chromophore in
the ground state with a similar absorption spectrum.38,39 Thus,
photoexcitation leads to several S1 subpopulations with different
lifetimes.
To get a deeper insight into the relaxation mechanism of the
GFP chromophore, p-HBDI derivatives have been synthesized
and their photophysics has been investigated.24,25,27-32 For
example, Solntsev et al.32 have compared the ultrafast excited-
state dynamics of p-HBDI and m-HBDI in aqueous solutions
and have found that the deactivation by large amplitude motion
was slower with the m-HBDI, due possibly to a higher barrier
for Z/E isomerization.
The green fluorescent protein (GFP) from the jellyfish
Aequorea Victoria has found widespread applications in mo-
lecular biology, cell biology, biochemistry, biotechnology, and
molecular genetics.1-5 The chromophore responsible for the
emission of this protein is p-hydroxybenzylideneimidazolinone
(p-HBI). The fluorescence arises from its anionic form generated
upon ultrafast intramolecular excited-state proton transfer in the
neutral form.6-8 Although GFP was discovered more than 20
years ago, the photophysics of p-HBI and of its mutants is still
intensively investigated mainly because of the spectacular
decrease of its fluorescence quantum yield, Φfl, and lifetime, τ,
taking place when going from the protein (Φfl ) 0.79,9 τ ) 3.3
ns)8,10 to liquid solutions (Φfl < 10-3 11 τ < 1 ps).12-14 In this
respect, the photophysical properties of the synthetic dimethyl
derivative of the GFP chromophore (p-HBDI)13,15-26 and other
related model systems24,25,27-32 have been extensively investigated.
,
The absence of fluorescence of p-HBDI in solution has been
found to originate from an ultrafast internal conversion (IC)
from the S1 state.15,16,19,22,33,34 Such lack of emission is a
characteristic of flexible chromophores like for example triph-
enylmethane dyes,35-37 for which the relevant coordinate for
IC is the large amplitude motion of the phenyl rings. However,
contrary to these dyes and other chromophores, which also
undergo photoisomerization, the fluorescence lifetime and
quantum yield of p-HBDI were found to exhibit only weak
Other authors27-29 have also shown that when the methyl
groups of p-HBDI are replaced by phenyl or other bulky
substituents, the fluorescence quantum yield and lifetime exhibit
a much more pronounced viscosity dependence. In all these
cases, nonexponential excited-state decay, indicative of the
existence of several ground state conformers, was found as well.
* To whom correspondence should be addressed, Irina.Petkova@unige.ch
(I.P.) and Eric.Vauthey@unige.ch (E.V.).
† Institute of Organic Chemistry with Centre of Phytochemistry.
‡ Department of Physical Chemistry, University of Geneva.
We report here on our detailed investigation of the photo-
physics of 4-benzylidene-1,2-diphenyl-1H-imidazol-5(4H)-one-
10.1021/jp903900b  2010 American Chemical Society
Published on Web 12/11/2009

Dynamics of GFP-Inspired Imidazolone Derivatives
J. Phys. Chem. A, Vol. 114, No. 1, 2010 11
SCHEME 1: Synthesis and Structure of Compounds
1-5
intermediates. This limitation is caused by their hydrolytical
instability in water and in other protic solvents, leading to ring-
opening53 and loss of fluorescent properties. Alternative synthetic
methods have also been reported in literature,54-61 a microwave-
assisted synthesis of azalactones has been recently published.62
In a second step, the pure intermediates IIIa-d were heated
with the aromatic amines IVa,b and anhydrous KOAc in acetic
acid (AcOH) to afford the desired (Z)-2-phenyl-1-(aryl)-4-
(arylmethylene)-1H-imidazol-5(4H)-ones (1-5) (Scheme 1B).58-69
After column chromatography and/or crystallization, the prod-
ucts 1 and 3-5 were obtained as chemically and stereoisomeri-
cally pure Z-isomers in moderate to high yields. For compound
2 we observed mixtures of Z/E-isomers (by NMR), which could
not be isolated into the individual stereoisomers. Besides,
compound 2 is extremely insoluble in low polarity solvents and
only weakly soluble in polar solvents. For the imidazolones 1
and 5 there are previously published data.58,62,65,68-73
The solvents, acetonitrile (ACN), propionitrile (PrCN), bu-
tyronitrile (BuCN), methanol (MeOH), ethanol (EtOH), pentanol
(PeOH), diethyl ether (Et2O), n-hexane (HX), cyclohexane (CH),
decalin (DEC), and methylcyclohexane (MCH) were of the
highest commercially available purity and were used as received.
Poly(methyl methacrylate) (PMMA) films were prepared as
follows: 300 mg of PMMA was dissolved in 10-4 M dichlo-
romethane solutions of the investigated compounds. The solu-
tions were poured in Petri dishes, and the solvent was evaporated
for about 24 h. The thickness of the resulting films was uniform
(40-50 µm), and the concentration of the investigated com-
pounds was about 0.500-0.700% weight.
Steady-State Measurements. Steady-state absorption spectra
were recorded with a Perkin-Elmer Lambda 25 UV-vis
spectrophotometer set to 1 nm resolution. Corrected steady-
state fluorescence and excitation spectra were measured with a
Perkin-Elmer LS 55 spectrofluorometer, with 10 nm excitation
and emission slit widths. The fluorescence quantum yields were
determined at room temperature using solute concentrations
adjusted to achieve an optical density of 0.1 at the excitation
wavelength. The fluorescence standard for compounds 1-4 was
p-methoxy-benzylidenphtalide in ethanol (Φfl ) 0.14)74 and for
compound 5, coumarin in ethanol (Φfl ) 0.78).75
derivatives with various substituents in different media using
femtosecond-resolved emission and absorption spectroscopy.
These substituents differ by their electron donating or withdraw-
ing properties, their size, and their ability for intra- or
intermolecular hydrogen bonding (Scheme 1, compounds 1-5).
Our main goal was to establish all the different nonradiative
deactivation pathways of these fluorophores not only to get a
deeper insight into the origin of the weak fluorescence of the
GFP chromophore in solution but also to find out whether it is
possible to control them in order to develop new fluorophores
with tunable emission properties. Indeed, the widespread interest
on GFP and its mutants has been particularly focused on their
role in biology. As yet, it has not been fully realized that GFP-
like fluorophores could have a large potential in optoelectronics,
for example, as emitters in organic light-emitting diodes.40
The emission spectra in polymer films were measured in the
reflection mode.
The low-temperature luminescence measurements were per-
formed at 77 K with the standard phosphorescence accessory
of a Perkin-Elmer spectrofluorimeter MPF 44 using 4 mm
diameter quartz tubes, and the spectra were subsequently
corrected.
Time-Resolved Fluorescence Measurements. Fluorescence
dynamics on the subnanosecond time scale was measured by
up-conversion using a setup described in detail elsewhere.76,77
Excitation was achieved at 400 nm with the frequency-doubled
output of a Kerr lens mode-locked Ti:sapphire laser (Tsunami,
Spectra-Physics). The output pulses centered at 800 nm had
duration of 100 fs and a repetition rate of 82 MHz. The
polarization of the pump beam was at magic angle relative to
that of the gate pulses at 800 nm. Experiments were carried
out in a 1 mm rotating cell. The full width at half-maximum
(fwhm) of the instrument response function (IRF) was ca. 280
fs. No significant degradation of the samples was observed after
the measurements.
2. Experimental Section
Samples. The different substituted imidazolone derivatives
have been synthesized according to Scheme 1 (see also
Supporting Information). The first step includes the preparation
of (Z)-2-phenyl-4-(arylmethylene)oxazol-5(4H)-ones by imple-
mentation of the classic Erlenmeyer synthesis-condensation of
hippuric acid I with the aromatic aldehydes IIa-d in the
presence of anhydrous KOAc in acetic anhydride to obtain the
corresponding (Z)-2-phenyl-4-(arylmethylene)oxazol-5(4H)-ones
IIIa-d in moderate yields (Scheme 1A).41-46 As known in
literature,47-51 small amounts of (E)-2-phenyl-4-(arylmethyl-
ene)oxazol-5(4H)-ones were detected in the crude product by
TLC. The appropriate crystallization resulted to the pure target
Z-isomers IIIa-d. Although oxazolones are well-known for
their good fluorescent properties,52 they are mainly used as
Fluorescence decay measurements on a longer time scale were
performed using the time-correlated single photon counting
(TCSPC) technique. Excitation was carried out at a repetition
rate of 40 MHz with <90 ps pulses generated by laser diodes

12 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Petkova et al.
interaction being between the (dimethylamino)phenyl and the
carbonyl groups in 5.
A very weak solvatochromism was observed with 1-4:
compared to CH, the absorption maximum in ACN is slightly
blue-shifted, whereas in ethanol the maximum is red-shifted.
The number of solvents investigated is not large enough to allow
precise interpretation of this small solvent dependence. Never-
theless, the blue shift in ACN might be due to the lower
refractive index of this solvent, thus to weaker dispersion
interaction, while the red shift in EtOH might originate from
hydrogen bonding interaction. By contrast, the first absorption
band of 5 exhibits a substantial bathochromic shift (1300 cm-1)
by going from the nonpolar CH to the polar ACN and EtOH.
In this case, the shift from CH to ACN is attributed to
dipole-dipole interaction, whereas the additional shift in EtOH
is tentatively assigned to the effect of hydrogen bonding.
Figure 1. Absorption and fluorescence spectra of 1 and 5 in
cyclohexane at 300 K. The vertical lines denote the computed energies
of the ππ* transitions and the corresponding relative oscillator strengths
of the Z-conformer.
In PMMA films, the shape and the position of the longest
wavelength absorption band of 1-5 are very similar to those
measured in EtOH (Table 1).
either at 395 nm (PicoQuant model LDH-PC-400B) or at 469
nm (PicoQuant model LDH-PC-470). Fluorescence was col-
lected at 90° and was measured at magic angle with respect to
the polarization of the pump pulses. A 420 or 560 nm cutoff
filter placed in front of the photomultiplier tube (Hamamatsu,
H5783-P-01) ensured that no scattered excitation light could
reach the detector, whose output was connected to the input of
a TCSPC computer board module (Becker and Hickl, SPC-300-
12). The fwhm of the IRF was around 200 ps.
Transient Absorption (TA) Measurements. Most of the TA
measurements have been carried out with the setup described
in detail in ref 78. Additional measurements have been
performed with an upgraded version of this setup as discussed
in ref 79 that gives a better signal/noise ratio. Excitation was
performed at 400 nm using the frequency-doubled output of a
standard 1 kHz amplified Ti:sapphire system (Spectra-Physics).
The pump intensity on the sample was around 2 mJ/cm2. The
polarization of the probe pulses was at magic angle relative to
that of the pump pulses. All spectra were corrected for the chirp
of the white light probe pulses. The fwhm of the IRF was ca.
200 fs. The sample solutions were placed in a 1 mm path length
quartz cell where they were continuously stirred by N2-bubbling.
Their absorbance at the excitation wavelength was between 0.05
and 0.2, depending on the solvent.
The fluorescence spectra of 1-4 in all solvents investigated
are characterized by a Stokes shift amounting to about 4000
cm-1. Furthermore the fluorescence spectrum is independent of
the excitation wavelength and does not exhibit a strong
solvatochromism. The red shift introduced by the Ar1- and Ar2-
substituents can be as large as 990 cm-1 in both nonpolar and
polar media and is thus more pronounced than that found in
the absorption spectra.
On the other hand, the fluorescence spectrum of 5 is strongly
solvent dependent (Figure 2). In CH, a pronounced vibrational
structure is observed. As solvent polarity increases, this structure
vanishes and the fluorescence maximum shifts to longer
wavelengths (Figure 2). This effect suggests a substantial solvent
and/or intramolecular reorganization of the excited state in polar
solvents.
The fluorescence quantum yields, Φfl, of 1 and 3 in CH are
around 10-2, whereas those of the other compounds are about
10-3 (Table 1). The fluorescence quantum yields of 1-3 in
protic and aprotic polar solvents (EtOH, ACN) decrease
substantially, whereas those of 4 and 5 remain essentially
unchanged.
The fluorescence spectra of all compounds in frozen EtOH
matrix at 77 K are very similar to those in room temperature
liquids; however, the fluorescence quantum yields are much
larger and for 1, Φfl amounts to unity (Table 1). This is in
agreement with the observed increase of the fluorescent intensity
in PMMA films as well. Finally no phosphorescence could be
detected with all compounds investigated even at 77K.
Time-Resolved Fluorescence. Fluorescence up-conversion
measurements of 1-5 in various solvents were performed at
several wavelengths throughout the emission band. As shown
in Figure 3A for 3 in ACN, the fluorescence dynamics exhibits
substantial wavelength dependence, accelerating with increasing
wavelength. In order to get a deeper insight into the origin of
this effect, these time profiles were analyzed globally with a
sum of exponentials as trial function. Once the global lifetime
was determined, the amplitudes of the different decay compo-
nents were calculated by rescaling the integrated time profiles
according to the stationary fluorescence intensity, as explained
in ref 88. The data shown in Figure 3A could be satisfactory
reproduced using the convolution of the instrument response
function with the sum of three exponential functions with time
constants of 0.4, 1.2, and 11.9 ps. As illustrated in Figure 3B,
the shortest time constant is associated with a decay of the
fluorescence intensity at short wavelength and with a rise above
Computational Methods. The ground-state singlet gas-phase
geometries of 1-5 were fully optimized at the density functional
level of theory (DFT)80 using the B3-LYP functional,81,82 and a
def-TZVP83 basis set. Electronic vertical excitation energies and
oscillator strengths were then computed with time-dependent
density functional theory (TDDFT) using the same functional
and basis set. All calculations were performed using the
TURBOMOLE 5.8 package.84-87
3. Results
Steady-State Spectroscopy. The UV-vis absorption spectra
of 1-4 above 300 nm are very similar and consist of a broad
band with a maximum located between 375 and 400 nm
depending on the nature of the Ar1-substituent and with a
shoulder at shorter wavelength (Figure 1). The presence of
electron-withdrawing or electron-donating substituents in 2, 3,
and 4 leads to a bathochromic shift of the Frank-Condon
absorption transition up to 550 cm-1 (Table 1). Furthermore,
the strong electron-donating group in 5 results to a red shift of
the longest wavelength absorption band of more than 3600 cm-1
(Table 1, Figure 1).
These spectra point to a coupling between the Ar1-substituents
and the imidazolone moiety in all compounds, the strongest

Dynamics of GFP-Inspired Imidazolone Derivatives
J. Phys. Chem. A, Vol. 114, No. 1, 2010 13
TABLE 1: Absorption, λabs, and Fluorescence, λfl, Maxima in nm and Fluorescence Quantum Yields, Φfl, of 1-5
cyclohexane
300 K
λfl
acetonitrile
300 K
λfl
ethanol
PMMA
300 K
300 K
77 K
λabs
Φfl
λabs
Φfl
λabs
λfl
Φfl
λfl
Φfl
λabs
λfl
1
2
3
4
5
377
389
396
396
437
449
458
470
470
484
0.03
0.005
0.03
0.001
0.002
375
393
393
393
455
453
470
471
470
552
0.005
<10-3
0.002
0.003
0.002
377
403
397
396
467
447
475
470
471
540
0.004
<10-3
0.002
0.003
0.001
440
485
470
470
525
∼1
385
400
400
a
455
465
470
a
∼0.2
∼0.3
∼0.2
∼0.3
460
520
a Measurements not performed.
As vibrational relaxation results to spectral dynamics as well,
mostly as a narrowing of the emission band,90 it could also
contribute to the 0.4 ps component. On the other hand, the
amplitude factors of the two larger time constants are positive
at all wavelengths and their spectral dependences, especially
that associated with the longest one, resemble that of the
stationary fluorescence spectrum. Consequently, these two
components correspond to a decay of the excited-state popula-
tion. A biexponential population decay can be due to two distinct
emitting populations. However, it has also been shown that a
Gaussian distribution of decay times associated with a distribu-
tion of different populations can also be reproduced by a
biexponential function.91,92 Independently of whether the decay
is due to two or to a distribution of populations, the resemblance
of the associated amplitude spectra points to emitting species
with relatively similar fluorescence spectra.
Figure 2. Fluorescence spectra of 5 at 300 K in various solvents.
The best-fit parameters have been used to reconstruct time-
resolved emission spectra, those shown in Figure 3C having
been area-normalized to better distinguish the spectral changes
occurring during the first few picoseconds. The main feature is
a marked narrowing of the red side of the band. This change
can be explained by the slightly red-shifted emission spectrum
of the shorter-lived species compared to that of the long-lived
one.
Qualitatively similar fluorescence dynamics was observed in
the other nitrile solvents, the main difference being that all time
constants are larger than those in ACN (Table 2). In alcohols
on the other hand, the time profiles could not be correctly
reproduced with less than four exponential functions, one
accounting for the initial spectral dynamics and the other three
corresponding to population decays. Four exponentials were also
used in the nonpolar CH and DEC, but in this case the initial
spectral dynamics with a time constant τ1 should rather be
ascribed to the relaxation of intramolecular modes.
As the amplitude spectra of the population components are
all quite similar, an amplitude average population decay time
was calculated as
〈τ〉 )
a τ /
a
i
(1)
∑ ∑
i
i
i>1
i>1
where ai is the amplitude of the component i at a fixed
wavelength, in general at the stationary fluorescence maximum.
Table 2 shows that, if the average population decay times of 3
are sorted according to the class of solvents, namely, nitriles,
alcohols, and nonpolar solvents, 〈τ〉 increases with solvent
viscosity within each class. Moreover, at constant viscosity, 〈τ〉
is the shortest in alcohols and is clearly the largest in nonpolar
solvents.
The fluorescence dynamics of the other compounds was
measured in the same way in various solvents, and the data
were also analyzed globally, except for 1 for which the
fluorescence was measured at the maximum wavelength only.
Figure 3. (A) Fluorescence time profiles of 3 in acetonitrile. (B) Decay-
associated spectra obtained from the global analysis of the fluorescence
dynamics of 3 in acetonitrile. (C) Reconstructed area-normalized time-
resolved emission spectra and stationary spectrum (dotted).
about 470 nm. This corresponds essentially to a red shift of the
fluorescence band but not to a decay of the excited-state
population. Considering that the 0.4 ps time constant is close
to the solvation time of ACN,89 this process can be assigned to
a dynamic Stokes shift. However, excitation is performed at
400 nm, and about 0.2 eV is deposited in 3 as vibrational energy.

14 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Petkova et al.
TABLE 2: Time Constants Obtained from a Global
Analysis of the Fluorescent Time Profiles of 1-5 Measured
at Different Wavelengths
solvent
τ1 (ps)
τ2 (ps)
τ3 (ps)
τ4 (ps) 〈τ〉 (ps)
1a acetonitrile
ethanol
0.5 (0.35)b 1.1 (0.19) 22.1 (0.43)
0.5 (0.53) 3.7 (0.26) 23.6 (0.21)
0.6 (0.51) 6.3 (0.30) 56.0 (0.19)
0.4 (0.52) 1.4 (0.19) 115.0 (0.29)
pentanol
n-hexane
cyclohexane 0.4 (0.49) 1.9 (0.20) 108.0 (0.31)
2
3
acetonitrile
ethanol
cyclohexane <IRF
acetonitrile
<IRFc
<IRF
0.4
0.5
0.8
1.2
1.7
2.6
1.6
0.5
1.4
0.6
1.3
1.8
3.3
0.5
0.8
0.5
1.6
0.9
1.1
2.0
2.2
1.1
1.6
6.1
4.6
12.2
16.3
4.8
6.2
16.3
51.3
107.5
7.3
22.1
5.7
15.4
3.0
10.6
11.9
18.9
27.6
6.0
2.1
8.7
7.0
11.9
12.9
38.1
2.5
0.4
propionitrile 0.3
butyronitrile <IRF
methanol
ethanol
pentanol
cyclohexane <IRF
decalin
acetonitrile
butyronitrile 0.6
ethanol
pentanol
cyclohexane <IRF
n-hexane
acetonitrile
0.4
0.2
<IRF
24.9
13.2
45.1
128.3
212.2
<IRF
0.27
4
5
<IRF
<IRF
13.2
42.0
6.6
13.7
28.3
11.0
14.8
<IRF
0.2
11
butyronitrile <IRF
ethanola
<IRF
cyclohexanea 0.8 (0.48) 9.4 (0.45) 25.6 (0.07)
1.2 (0.52) 8.6 (0.48)
a Measured at the fluorescence maximum only. b Relative
amplitudes. c Instrument response function.
For this compound, a reliable assignment of the time constants
obtained from the multiexponential fit is therefore not possible.
The resulting time constants are listed in Table 2.
Although 1 was not investigated in detail, the solvent
dependence of its fluorescence dynamics does not differ much
from that of 3. The same is true for 2, except that the average
lifetime is faster than for 3 in all three types of solvents,
especially in CH.
Figure 4. (A) Fluorescence time profiles of 5 in acetonitrile. (B) Decay-
associated spectra obtained from the global analysis of the fluorescent
dynamics of 5 in acetonitrile. (C) Reconstructed area-normalized time-
resolved emission spectra and stationary spectrum (dotted).
On the other hand, 4 behaves similarly to 3 in nitriles and
alcohols but exhibits a much faster excited-state dynamics in
nonpolar solvents. In fact, the fastest population decay was
measured in CH.
TABLE 3: Fluorescence Lifetimes, τ, and Radiative Time
Constants, τr, of 1-5 in PMMA at 300 K and in Ethanol at
77 K
PMMA (300 K)
ethanol (77 K)
Finally, the largest wavelength dependence was found with
5 in polar solvents (Figure 4A). The amplitude spectrum
associated with the shortest time constant, τ1 ) 0.2 ps, points
to a substantial dynamic Stokes shift (Figure 4B). However,
the reconstructed, area-normalized, time-resolved emission
spectra reveal a conversion between two different emitting
species rather than a continuous spectral shift, as observed with
the other compounds (Figure 4C). The excited-state dynamics
of 5 seems therefore to differ considerably from that of the other
compounds. Because of the probable presence of different
species with more distinct emission spectra, an average popula-
tion decay time cannot really be determined with 5.
The fluorescence dynamics in solid matrices is much slower
than those in room temperature liquids and could thus be
measured by TCSPC. The fluorescence decays in both PMMA
films and EtOH glasses at 77 K could be well reproduced using
single exponential functions with lifetimes, τ, around 2.5 ns
(Figure S1 in Supporting Information). The latter together with
the radiative time constants, τr, calculated from Φfl and τ are
listed in Table 3. The lifetimes measured in PMMA are of the
same order of magnitude.
τ (ns)
τ (ns)
τr (ns)
1
2
3
4
5
2.9
1.7
1.9
a
3.4
2.5
2.3
2.8
2.9
3.4
12.5
7.7
14.0
9.7
1.8
a Measurement not performed.
nonradiative rate constant, calculated as knr ) τ-1 - τr-1 and
assuming τr independent of temperature, clearly departs from
an Arrhenius behavior, with a marked kink around 130 K
(Figure S3 in Supporting Information). Such an effect, already
reported for other GFP fluorophores15 as well as for Malachite
Green,93 reflects a change in the nature of the matrix occurring
above the glass transition temperature and can be considered
as a strong indication of the involvement of intramolecular
modes with large amplitude motion in the nonradiative deactiva-
tion of 1 and 5.
Transient Absorption Measurements. To have a deeper
insight into the excited-state dynamics of these compounds, TA
measurements have been performed with 1-5 in ACN, EtOH,
and CH. Figure 5A shows TA spectra recorded with 3 in ACN
The temperature dependence of the fluorescence lifetime of
1 and 5 down to 77 K was also investigated by TCSPC in MeOH
and MCH. The observed temperature dependence of the

Dynamics of GFP-Inspired Imidazolone Derivatives
J. Phys. Chem. A, Vol. 114, No. 1, 2010 15
TABLE 4: Time Constants Obtained from a
Multiexponential Global Analysis of the Transient
Absorption Spectra
solvent
τ1 (ps)
τ2 (ps)
τ3 (ps)
1
2
3
4
5
acetonitrile
ethanol
cyclohexane
acetonitrile
ethanol
cyclohexane
acetonitrile
ethanol
cyclohexane
acetonitrile
ethanol
0.4
0.2
0.6
0.4
0.7
<IRF
0.4
0.3
1.3
0.4
0.5
0.4
0.2
0.2
0.1
1.0
1.2
3.1
1.2
2.6
0.9
1.3
1.7
8.9
1.2
2.8
7.1
0.9
1.5
1.1
22.7
16.8
114.0
11.9
9.4
11.1
11.9
12.1
122
11.9
13.6
106.8
11.0
8.3
cyclohexane
acetonitrile
ethanol
cyclohexane
11.4
The TA spectra measured with 1 and 4 in all three solvents
are qualitatively very similar to those of 3. With 2 on the other
hand, the ESA band shape remains essentially unchanged over
the whole excited-state lifetime.
Figure 5. (A) TA spectra recorded with 3 in acetonitrile at different
time delays after excitation. (B) Decay-associated spectra obtained from
multiexponential global analysis.
A quite different behavior was observed with 5 in polar
solvents as illustrated in Figure 6A. The very early TA spectra
consist in two negative bands, one around 450 nm and the other
around 540 nm, and in two positive bands, one at 500 nm and
the other above 570 nm. The shortest wavelength band can be
safely ascribed to the bleach of the S0-S1 absorption, whereas
the other one arises from S1-S0 SE. Over a few hundreds of
femtoseconds, the SE band shifts to the red. As time proceeds,
the intensity of the long wavelength positive band decreases,
whereas the TA intensity around 500 nm increases to reach its
maximum at about 1.1 ps. Then the TA intensity over the whole
spectral window decreases, and after about 50 ps, the TA
spectrum remains unchanged until 1.8 ns, the maximum time
delay of the experiment.
As shown in Figure 6C, the time profile of the intensity of
the two positive TA bands suggests a precursor-successor
relationship of the populations responsible for these two bands.
To confirm the existence of such relationship, a global fit with
target analysis was performed assuming an A f B f C reaction
scheme, where both A f B and B f C steps follow a first-
order kinetics. The overall decay of the TA spectra was well
reproduced with this model, but strong deviation was observed
during the first 2 ps, as the initial rise could not be well
accounted for. However, much better agreement over the whole
time window was obtained upon addition of a third step to the
model, namely, assuming an A f B f C f D reaction scheme.
The time constants for the three consecutive steps amount to
0.3, 0.9, and 11.8 ps, and the spectra of the four species A, B,
C, and D are shown in Figure 6B. All A, B, and C spectra
exhibit a negative band below 490 nm due to the ground-state
depletion. A second negative band arising from SE is also
present in all spectra; however that of A peaks at 530 nm,
whereas those of B and C are centered at 550 nm. The species
D accounts for the very small residual TA spectrum measured
after about 50 ps. These findings agree very well with the
fluorescence up-conversion data that point to a conversion
between two emitting species with distinct spectra. The similar-
ity of the spectra associated with B and C suggests that these
two species are closely related, whereas that of A has a different
nature.
at different time delays after 400 nm excitation. The spectra
consist of a negative band located between 470 and about 530
nm, which can be ascribed to S1-S0 stimulated emission (SE)
and of a broad positive band above 530 nm that can be assigned
to S1 excited-state absorption (ESA). Comparison of the negative
SE band with the spontaneous fluorescence spectrum, which
extends from 470 to 650 nm, points to a substantial overlap of
SE and ESA spectra. As time proceeds, the intensity of both
bands exhibits a parallel decay to zero. A closer inspection of
the SE band allows some small changes of the shape with time
to be observed, especially a broadening of the red edge.
Similarly, the ESA band shape presents some temporal variation.
At early time, its maximum is located at about 600 nm with a
secondary maximum around 650 nm. About 2-3 ps after
excitation, the intensity at both wavelengths is identical and, at
later time, the main ESA maximum is at 650 nm. The decay of
the TA intensity over the whole spectral window can be well
reproduced with the sum of three exponential functions with
0.4, 1.3, and 11.9 ps time constants and with the decay-
associated spectra shown in Figure 5B. It appears that most of
the dynamics in the ESA region can be well accounted for with
the two slower components. The distinct decay-associated
spectra of these two components explain the temporal variation
of the ESA band shape. On the other hand, the 0.4 ps component
is required to reproduce the early spectral dynamics in the SE
band region. Given this very short time constant, the amplitude
of its associated spectrum might be overestimated relative to
that of the other two decay associated spectra. The three time
constants are in good agreement with those obtained from the
analysis of the fluorescence up-conversion data (Tables 2 and
4).
No significant difference can be observed in EtOH, apart from
the time constants, which are very close to those found in the
fluorescence measurements as well (Tables 2 and 4). In CH,
the TA spectra are also quite similar except for the ESA band,
which extends to longer wavelengths (Figure S3 in Supporting
Information). In this case again, a very good agreement of TA
and up-conversion time constants is obtained.
A qualitatively similar behavior was observed in EtOH.
However, the intensity of the 500 nm band is relatively smaller

16 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Petkova et al.
Figure 7. TA spectra recorded with 5 in cyclohexane at different time
delays after excitation.
Figure 8. Optimized ground-state geometries of Z and E isomers of
3 calculated at the B3LYP level of theory.
Figure 6. (A) TA spectra recorded with 5 in acetonitrile at different
time delays after excitation. (B) Species-associated spectra obtained
from target analysis assuming a A f B f C f D scheme. (C) TA
profiles at the maximum of the two positive bands and best biexpo-
nential global fit.
Figure 9. Frontier molecular orbitals of 5 computed at the B3LYP
level of theory. The orbital contours are drawn at the 0.02 au level.
than that in ACN. Performing the same analysis as in ACN
resulted in a good agreement with the data as well, with τAB
)
0.3 ps, τBC ) 1.5 ps and τCD ) 7.9 ps for the A f B, B f C,
and C f D steps, respectively. In EtOH, the τCD/τBC ratio
amounts to about 5, whereas it is as large as 13 in ACN. This
means that in EtOH, the population of species C does not
accumulate as much as that in ACN, explaining the smaller
relative intensity of the 500 nm TA band.
Finally, the TA spectra of 5 in CH differ substantially from
those in polar solvents (Figure 7) and resemble rather those
measured with 1-4. Apart from spectral changes occurring at
very early time and most probably associated with vibrational
relaxation, the shape of the ESA band remains constant over
the lifetime of the excited state.
The time constants obtained from the multiexponential global
analysis of the TA spectra of compounds 1 to 5 are summarized
in Table 4. It can be seen that in all cases, these time constants
match very well with those extracted from the fluorescence up-
conversion data. In fact, global analysis of both the fluorescence
and TA data could also be performed successfully and yielded
time constants between those obtained upon separate analysis
of the fluorescence and TA data. However, the components with
the time constants written in bold characters in Table 4 are not
present in the fluorescence data. These values, which always
correspond to the slowest components, can be assigned to
nonemitting species. For 2 and 4, their decay-associated spectra
are quite similar to that related to the 12 ps time constant shown
in Figure 5B. In the case of 5, this long-lived component is
associated with species D in Figure 6B. As it can be seen, the
amplitude of the associated spectrum is very small. Thus, species
D is most probably the ground state of 5 and the small associated
spectrum might originate from some photoproduct.
Quantum Chemistry Calculations. The ground-state cal-
culations predict the presence of two stable forms for all
compounds (Figure 8), except for 2 where no convergence could
be achieved for the E form. Comparison of the total energies
of the Z and E forms of 1, 3, and 5 shows that the Z isomer is
more stable by about 3 kcal/mol. For 4 on the other hand, the
E isomer is more stable than the Z form.
According to the TDDFT calculations, the S0-S1 absorption
band of 5 is essentially connected to a one-electron HOMO-
LUMO transition and has a ππ* character. The pronounced
bathochromic shift of both absorption and fluorescence bands
of 5 points to ππ* transition with some charge transfer (CT)
character as testified by the HOMO and LUMO shown in Figure
9. Thus optical excitation of 5 leads to a substantial charge
redistribution from the Ar1 substituent to one of the phenyl
groups on the imidazolone moiety.

Dynamics of GFP-Inspired Imidazolone Derivatives
J. Phys. Chem. A, Vol. 114, No. 1, 2010 17
TABLE 5: Viscosity, η, Static Dielectric Constant, εs, and
Kamlet-Taft Parameter, r, of the Solvents Used
4. Discussion
Multiexponential Excited-State Dynamics. The fluorescence
dynamics of all five compounds in liquid solution is clearly
multiexponential. Whereas the fastest dynamic components can
be safely ascribed to solvent and/or to vibrational relaxation,
the remaining components can be assigned to the decay of
excited-state populations with slightly different fluorescence
spectra.
solvent
ηa (cP)
εsa
Rd
n-hexane
cyclohexane
decaline
acetonitrile
proprionitrile
butyronitrile
methanol
0.298
0.90
1.88
2.02
0
0
0
0.15
0
2.42
2.15
0.325b
0.39b
0.515b
0.54
37.5c
27.2c
20.3c
32.7
0
This interpretation is corroborated by the TA measurements,
where a marked temporal variation of the ESA spectrum is
observed (Figure 5). The time scale of this spectral dynamics
is very similar to that found by fluorescence. Moreover, the
decay-associated spectra obtained from global analysis of the
TA spectra point to excited-state populations with slightly
different absorption spectra as well.
0.98
0.86
0.84
ethanol
pentanol
1.08
3.35
24.55
13.9
a From ref 94, if not specified at 25 °C. b At 30 °C. c At 20 °C.
d From ref 95.
dependence of IC via large amplitude motion is often discussed
in terms of the following empirical relationship:96
Multiexponential fluorescence decay has already been ob-
served with other derivatives of the GFP chromophore14-16 and
was attributed to the existence of several ground-state confor-
mations. The same interpretation can be proposed here for 1-5.
Considering the flexibility of these molecules, a distribution of
ground-state conformers with different torsion angles around
the single bonds between the imidazolone core and the two
phenyl substituents as well as around the ethylenic bridge can
be expected. Although the fluorescence and excited-state
absorption spectra of these conformers differ slightly, their
S0-S1 absorption band seems to be almost identical. Indeed,
the fluorescence excitation spectra of 1-5 are independent of
the emission wavelength. Furthermore, the fluorescence dynam-
ics of 3 recorded by up-conversion upon 370 and 420 nm
excitation are identical, within experimental error, to that
measured upon 400 nm excitation.
τnr ∝ ηR
(2)
where η is the solvent viscosity and R is a factor varying
between 0.1 and 1, R ) 1 corresponding to the high friction
limit in Kramers theory.97 This relationship has been rationalized
by Grote and Hynes in terms of frequency-dependent friction.98
Although eq 2 cannot be used very reliably with only three
solvents, doubly logarithmic plots of τ2 and τ3 vs η yield R
values of 0.6 and 1.5, respectively. Despite the large errors on
these two values, the faster decay component is clearly less
viscosity dependent than the slower one. This is in good
agreement with the assignment of these time constants to
different excited conformers. Thus, the motion responsible for
the nonradiative deactivation of the conformers does not
experience the same friction from the environment. Most
probably, the conformers with τ3 has to undergo larger amplitude
motion than the shorter lived ones, i.e., those with τ2.
The barrier height for photoisomerization and IC via large
amplitude motion may also depend on the solvent polarity, if
the electronic distribution of the transition state differs from
that of the initial state.99,100 For example, if the transition state
is more polar, the reaction rate increases with solvent polarity.
This is probably the case for 1 and 3, for which the fluorescence
time constants are clearly larger in nonpolar solvents, indepen-
dently of their viscosity. Consequently, the very large R values
most probably do not only reflect friction effects but also the
decrease of polarity that accompanies the increase of viscosity
when going from ACN to BuCN. A reliable determination of
R would require a series of solvents with the same polarity but
with different viscosity.
(ii) Intermolecular hydrogen bond formation has been shown
to substantially accelerate the excited-state decay of chro-
mophores with carbonyl groups.101-104 This interaction can be
viewed as an excited-state proton transfer leading to an unstable
product, which in turn decays to the ground state by back proton
transfer. To see whether such a process is also operative with
1-5, the excited-state dynamics of 1 and 2 should be compared
in protic and aprotic solvents of similar viscosity and dielectric
constant, εs (see Table 5). For example, the average excited-
state lifetime of 3 is markedly shorter in MeOH than in PrCN
and is also shorter in EtOH than in BuCN (Table 2). Similarly,
the excited-state lifetimes of the other compounds in EtOH are
at least as short as those in ACN, although EtOH is substantially
more viscous and less polar than ACN (Table 5). Therefore,
the shorter lifetime of 1-5 in protic solvents can be rather safely
ascribed to H-bond-assisted nonradiative deactivation occurring
additionally to IC by large amplitude motion. The shortening
On the other hand, the fluorescence decay of these compounds
in solid matrices, namely, PMMA films and EtOH glasses at
77 K, is single exponential. At 77 K, this could be explained
by a decrease of the torsional disorder at low temperature and
thus by the existence of a single conformer. However, this
interpretation cannot account for the single exponential fluo-
rescence decay in room temperature PMMA films. Therefore,
unless the polymeric environment imposes a well-defined
ground-state structure of 1-5, a distribution of conformers
should also be expected as in liquids. This implies that these
conformers have essentially the same fluorescence lifetime in
polymeric environments and consequently they have the same
radiative time constant, τr, and the same nonradiative dynamics.
Identical radiative time constants agree with the undistinguish-
able absorption spectra of these conformers. Thus, the multi-
exponential fluorescence decays in liquid solutions indicate that
the conformers have different nonradiative dynamics in a
nonrigid environment. The difference with solid matrices clearly
shows that the nonradiative pathways in liquids should be
associated with conformational dynamics. This will be discussed
in more detail below.
Nonradiative Decay Pathways. When considering the
structures of 1-5, several nonradiative deactivation pathways
of the S1 state can be envisaged: (i) IC by large amplitude motion
for all of them, (ii) intermolecular H-bonded assisted deactiva-
tion for all of them in protic solvents, (iii) intramolecular H-bond
assisted deactivation for 4, and (iv) intramolecular CT with 5.
(i) In principle, IC via large amplitude motion is the only
radiationless decay pathway that can be operative with 1 and 3
in nonprotic solvents. Table 2 shows that the fluorescence time
constants of 3 related to population dynamics, τ2 and τ3, increase
with viscosity in the three nitriles investigated. The viscosity

18 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Petkova et al.
of the excited-state lifetime observed by going from MeOH to
PeOH can be explained by both the decrease of the hydrogen-
bonding ability of the solvent, quantified by the Kamlet-Taft
parameter R (Table 5), and the increase of viscosity, that slows
down large amplitude motion.
The hydroxyl group in Ar1 in 2 could also be involved in
H-bonds with the solvent. Comparison of 2 with 1 and 3 also
points to a faster nonradiative deactivation of the excited-state
lifetime of 2. However, this difference is also present in CH,
which has no H-bond acceptance property, and thus the observed
acceleration of the nonradiative decay of 2 could be due to
another effect, like for example a decrease of the barrier height
for large amplitude motion around the ethylenic bond induced
by the presence of the electron-donating OH group. Indeed, the
position of the OH group in HBDI, para or meta, was found to
have a marked influence on the excited-state lifetime.32
to the charge recombination of the CT state. The excited-state
dynamics of 5 and of other imidazolone derivatives undergoing
similar intramolecular CT will be discussed in more detail
elsewhere.
Comparison with p-HBDI. p-HBDI is probably the most
studied model of the GFP chromophore. Its structure is
essentially the same as that of 2 except that Ar2 and the phenyl
substituent in position 7 are methyl groups. Therefore, the
excited-state dynamics of both p-HBDI and 2 can be expected
to be very similar. The fluorescence decay of p-HBDI in a series
of alcohols ranging from MeOH to decanol has been reported
to be biexponential with time constants around 200 fs and 1
ps, independently on the solvent viscosity.16 As shown in Table
2, the decay times found with 2 in ACN and MeOH do not
differ very much, confirming similar deactivation pathways of
p-HBDI and 2. Therefore by analogy with 2, it appears that,
depending on the solvent, the deactivation pathways of p-HBDI
should be IC by large amplitude motion and intermolecular
H-bond assisted deactivation. The number of solvents investi-
gated with 2 does not allow conclusions on the relative efficiency
of both channels to be drawn. However, the very weak viscosity
dependence reported for p-HBDI by Mandal et al.16 in alcohols
could be explained by the predominance of H-bond-assisted
deactivation. This is in full agreement with the recent study of
Yang et al., who found that the photoisomerisation yield of
p-HBDI is much smaller in protic than in aprotic solvents.25
Our study clearly shows that H-bonding to the imidazolinone
group is involved in the deactivation. The shorter excited-state
lifetime found with 2 compared to 1 and 3 suggests that
H-bonding between the hydroxy group and solvent molecules
is also probably operative. Further investigation of the solvent
dependence of the excited-state dynamics of 2 is in progress.
(iii) Intramolecular H-bond interaction between the hydroxyl
group on the Ar2 ring and the carbonyl oxygen is in principle
also possible in 4. The effect of this OH group can be evaluated
by comparing the excited-state dynamics of 4 and 3 (Tables 2
and 4). Whereas no significant difference can be observed in
polar solvents, substantially shorter fluorescence lifetime can
be seen for 4 in nonpolar solvents, with an about 10 times shorter
average lifetime. This effect is also reflected by the smaller
fluorescence quantum yield of 4 in CH (Table 1). These results
point to the existence of an intramolecular H-bond in nonpolar
media, favoring the nonradiative deactivation of the excited state.
The absence of a difference between 4 and 3 in polar solvents
could be explained by a twist of the phenol substituent relative
to the imidazolone ring, inhibiting intramolecular H-bonding.
Competition between inter- and intramolecular H-bonding could
also explain the very similar fluorescence lifetimes of 4 and 3
in protic solvents. Such competing H-bonding has recently been
reported for naphthalenediimide derivatives.105
5. Conclusion
(iv) Whereas in nonpolar solvents, the photophysics of 5
resembles that of 1 and 3, distinct excited-state dynamics is
observed with this compound in polar solvents. The solvato-
chromism of the first absorption band and of the fluorescence
spectrum is indicative of a substantially larger electric dipole
moment in the excited state than in the ground state of 5.
Moreover, the large change of the emission band shape with
solvent polarity (Figure 2) points to a different nature of the
emitting state, a locally excited state (LE) in nonpolar solvents,
and a CT state in polar solvents. According to the quantum
chemistry calculations (Figure 9), the LE state should already
possesses a non-negligible CT character. However, the CT state
should be characterized by a further degree of charge separation,
most probably favored by solvation and by structural rearrange-
ment of the excited-state, as is often the case in flexible
molecules undergoing intramolecular CT.106-108 Therefore, the
reconstructed time-resolved fluorescence spectra of 5 in ACN
(Figure 4C) can be interpreted as a transition from the LE to
the CT state, rather than as a simple dynamic Stokes shift. This
assignment is further reinforced by the TA dynamics (Figure
6) that could be very well reproduced assuming an A f B f
C f D reaction scheme with τAB ) 0.3 ps, τBC ) 0.9 ps, and
τCD ) 11.8 ps time constants in ACN. The strong similarity of
the spectra associated with species B and C suggests that A is
the LE state, B is a nonequilibrated CT state, C is the relaxed
CT state, and D is the ground state. In most cases, the
fluorescence lifetime of intramolecular CT states is on the
nanosecond time scale.106,109-111 As a consequence, the very fast
decay of the CT state found here with 5 is also most probably
due to IC via large amplitude motion, like in 1-4, rather than
The results presented here clearly show that some tuning of
the excited-state dynamics of GFP-chromophores can be
achieved beyond the simple variation of the medium rigidity.
For the compounds investigated here in liquid solutions, IC by
large amplitude motion is an important radiationless channel
of the excited state. Nevertheless, additional pathways can be
opened, such as inter- and intramolecular H-bond-assisted
deactivation, depending on the solvent and the substituent.
Occurrence of these processes further reduces the excited-state
lifetime of these chromophores. These results give a deeper
insight into the ultrafast nonradiative deactivation of p-HBDI,
the most investigated GFP chromophore model, as its structure
is very similar to those of the compounds studied here.
Furthermore, the nature of the excited state can be substan-
tially changed by introducing strong electron-donating substit-
uents. This combination of viscosity dependence of the fluo-
rescence quantum yield together with the polarity dependence
of the emission wavelength is not so common and therefore
these GFP-chromophores derivatives with CT character may
find potential applications as local molecular probes. Finally,
controlling the CT character of the excited state of these
molecules should in principle allow a rather fine-tuning of their
emission wavelength even in rigid media, where their emission
quantum yield is very large. These compounds may thus also
be potential candidates as fluorophores in electroluminescent
devices.
Acknowledgment. The authors would like to thank K.M.
Solntsev for fruitful discussions. I.P. acknowledges financial
support from NATO Security Through Science Programme (RIG

Dynamics of GFP-Inspired Imidazolone Derivatives
J. Phys. Chem. A, Vol. 114, No. 1, 2010 19
(36) Manring, L.; Peters, K. S. J. Phys. Chem. 1984, 88, 3516.
(37) Hicks, J.; Vandersall, M.; Babarogic, Z.; Eisenthal, K. B. Chem.
Phys. Lett. 1985, 116, 18.
(38) He, X.; Bell, A. F.; Tonge, P. J. FEBS Lett. 2003, 549, 35.
(39) Schellenberg, P.; Johnson, E.; Esposito, A. P.; Reid, P. J.; Parson,
W. W. J. Phys. Chem. B 2001, 105, 5316.
(40) You, Y.; He, Y.; Burrows, P. E.; Forrest, St. R.; Petasis, N. A.;
Thompson, M. E. AdV. Mater. 2000, 12 (22), 1678.
(41) Erlenmeyer, E. Ber. Dtsch. Chem. Ges. 1898, 31, 2239.
(42) Buck, J. S.; Ide, W. S. Organic Synthesis; Wiley: New York;
Collect. Vol. 2, 1943; p 55.
ref. No. 9811325) and European Science Foundation (DYNA
Exchange Grant 1893). This work was supported by Swiss
National Science Foundation (grant No. 200020-115942).
Supporting Information Available: Synthesis of 1-5,
fluorescence decay of 2 in a PMMA film, temperature depen-
dence of the nonradiative rate constants of 1 and 5 in MeOH
and MCH, and TA spectra recorded with 3 in CH. This
information is available free of charge via the Internet at http://
pubs.acs.org.
(43) Filler, R.; Rao, Y. S. AdV. Heterocycl. Chem. 1977, 21, 175.
(44) Carter, H. E. Org. React. 1974, 3, 198.
(45) Mukerjee, A. K.; Kumar, P. Heterocycles 1981, 16, 1995.
(46) Rao, Y. S.; Filler, R. Synthesis 1975, 12, 749.
(47) Kochetkov, N. P.; Budovskii, E. I.; Khomutov, R. M.; Karpeiskii,
M. Y.; Severin, E. S. J. Gen. Chem., USSR (Engl. Transl.) 1960, 30, 2573.
(48) Brocklehurst, K.; Price, H. S.; Williamson, K. Chem. Commun.
1968, 884.
(49) Morgenstern, A. P.; Schuitj, C.; Nauta, W. T. Chem. Commun.
1969, 321.
(50) Brocklehurst, K.; Bywater, R. P.; Palmer, R. A.; Patrick, R. Chem.
Commun. 1971, 632.
(51) Rao, Y. S. J. Org. Chem. 1976, 41, 722.
(52) Ko´cza´n, G.; Cs´ık, G.; Csa´mpai, A.; Balog, E.; Bo˝sze, S.; Soha´r,
P.; Hudecz, F. Tetrahedron 2001, 57, 4589.
(53) Harhash, A. H.; Kassab, N. A. L.; Elbanani, A. A. A. Indian
J. Chem. 1971, 789.
(54) Crawford, M.; Little, W. T. J. Chem. Soc. 1959, 729.
(55) Acheson, R. M.; Booth, D. A.; Brettle, R.; Harris, A. M. J. Chem.
Soc. 1960, 3457.
References and Notes
(1) Zimmer, M. Chem. ReV. 2002, 102, 3–759.
(2) Misteli, T.; Spector, D. L. Nat. Biotechnol. 1997, 15, 10–961.
(3) Cubbit, A. B.; Heim, R.; Adams, S. R.; Boyd, A. E.; Gross, L. A.;
Tsien, R. Y. Trends Biochem. Sci. 1995, 20, 448.
(4) Prendergast, F. G. Methods Cell Biol. 1999, 58, 1.
(5) Tsien, R. Y. Annu. ReV. Biochem. 1998, 67, 509.
(6) Chattoraj, M.; King, B. A.; Bublitz, G. U.; Boxer, S. G. Proc.
Natl. Acad. Sci. U.S.A. 1996, 93, 8362.
(7) Heim, R.; Prasher, D. C.; Tsien, R. Y. Proc. Natl. Acad. Sci. U.S.A.
1994, 91, 12501.
(8) Lossau, H.; Kummer, A.; Heinecke, R.; Poellinger-Dammer, F.;
Kompa, C.; Bieser, G.; Jonsson, T.; Silva, C. M.; Yang, M. M.; Youvan,
D. C.; Michel-Beyerle, M. E. Chem. Phys. 1996, 213, 1.
(9) Cubbit, A. B.; Woolenweber, L. A.; Heim, R. Methods Cell Biol.
1999, 58, 19.
(10) Perozzo, M. A.; Ward, K. B.; Thompson, R. B.; Ward, W. W.
J. Biol. Chem. 1988, 263, 7713.
(11) 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.
(12) Cody, C. W.; Prasher, D. C.; Westler, W. M.; Prendergast, F. G.;
Ward, W. W. Biochemistry 1993, 32, 1212.
(13) Webber, N. M.; Litvinenko, K. L.; Meech, S. R. J. Phys. Chem.
B 2001, 105, 8036.
(14) Mandal, D.; Tahara, T.; Webber, N. M.; Meech, S. R. Chem. Phys.
Lett. 2002, 358, 495.
(15) Litvivenko, K. L.; Webber, N. M.; Meech, S. R. J. Phys. Chem.
A 2003, 107, 2616.
(16) Mandal, D.; Tahara, T.; Meech, S. R. J. Phys. Chem. B 2004,
108, 1102.
(17) Webber, N. M.; Meech, S. R. Photochem. Photobiol. Sci. 2007,
6, 976.
(18) Altoe, P.; Bernardi, F.; Garavelli, M.; Orlandi, G.; Negri, F. J. Am.
Chem. Soc. 2005, 127, 3952.
(19) Usman, A.; Mohammed, O. F.; Nibbering, E. T. J.; Dong, J.;
Solntsev, K. M.; Tolbert, L. M. J. Am. Chem. Soc. 2005, 127, 11214.
(20) Martin, M. E.; Negri, F.; Olivucci, M. J. Am. Chem. Soc. 2004,
126, 5452.
(21) Stavrov, S. S.; Solntsev, K. M.; Tolbert, L. M.; Huppert, D. J. Am.
Chem. Soc. 2006, 128, 1540.
(22) Voityuk, A. A.; Kummer, A. D.; Michel-Beyerle, M. E.; Rosch,
N. Chem. Phys. 2001, 269, 83.
(23) Dong, J.; Abulwerdi, F.; Baldridge, A.; Kowalik, J.; Solntsev,
K. M.; Tolbert, L. M. J. Am. Chem. Soc. 2008, 130, 14096.
(24) Dong, J.; Solntsev, K. M.; Tolbert, L. M. J. Am. Chem. Soc. 2009,
131, 662.
(56) Kumar, P.; Mishra, H. D.; Mukerjee, A. K. Synthesis 1980, 836.
(57) Tripathy, P. K.; Mukerjee, A. K. Synthesis 1984, 418.
(58) Tripathy, P. K.; Mukerjee, A. K. Synthesis 1985, 285.
(59) Rao, P. S.; Venkataratnam, R. V. Indian J. Chem. 1994, 33B,
984.
(60) Puterova´, Z.; Sterk, H.; Krutosˇ´ıkova´, A. Molecules 2004, 9, 11.
(61) Royo, S.; Jime´nez, A. I.; Cativiela, C. Tetrahedron: Asymmetry
2006, 17, 2393.
(62) Conway, P. A.; Devine, K.; Paradisi, F. Tetrahedron 2009, 65,
2935.
(63) Cativiela, C.; Chueca, J.; Garcia, J. I.; Melendez, E. Heterocycles
1984, 22, 2775.
(64) Abbass, I. M.; Mosselhi, M. A.; Abdallah, M. A.; Shawali, A. S.
J. Chem. Res. 1995, (S), 190.
(65) Mukerjee, A. K.; Kumar, P. Can. J. Chem. 1982, 60, 317.
(66) Tu´ros´, G.; Csa´mpai, A.; Czugler, M.; Wamhoff, H.; Soha´r, P. J.
Organomet. Chem. 2001, 634, 122.
(67) Badr, M. Z. A.; El-Sherief, H. A. H.; Tadros, M. E. Indian
J. Chem. 1979, 18B, 240.
(68) Topuzyan, V. O.; Oganesyan, A. A.; Panosyan, G. A. Russ. J.
Org. Chem. (Eng.) 2004, 40, 1644.
(69) Ashare, R.; Mukerjee, A. K. Indian J. Chem. 1986, 25B, 762.
(70) Sanjayan, G. J.; Mukerjee, A. K. Indian J. Chem. 1994, 33B, 76.
(71) Tripathy, P. K. Indian J. Heterocycl. Chem. 2004, 14, 77.
(72) Tripathy, K.; Mukerjee, A. K. Indian J. Chem. 1986, 25B, 765.
(73) Abdallah, M. A.; Zayed, M. E.; Shawali, A. S. Indian J. Chem.
2001, 40B, 187.
(74) Nikolov, P.; Fratev, F.; Mintchev, St. Z. Naturforsch. 1983, 38
(A), 200.
(75) Reynolds, G. A.; Drexhage, K. H. Opt. Commun. 1975, 13, 222.
(76) Morandeira, A.; Fu¨rstenberg, A.; Gumy, J.-C.; Vauthey, E. J. Phys.
Chem. A 2003, 107, 5375.
(25) Yang, J.-S.; Huang, G.-J.; Liu, Y.-H.; Peng, S.-M. Chem. Commun.
2008, 1344.
(26) Gepshtein, R.; Huppert, D.; Agmon, N. J. Phys. Chem. B 2006,
110, 4434.
(77) Morandeira, A.; Engeli, L.; Vauthey, E. J. Phys. Chem. A 2002,
106, 4833.
(27) Follenius-Wund, A.; Bourotte, M.; Schmitt, M.; Iyice, F.; Lami,
H.; Bourguignon, J.-J.; Haiech, J.; Pigault, C. Biophys. J. 2003, 85, 1839.
(28) Bourotte, M.; Schmitt, M.; Follenius-Wund, A.; Pigault, C.;
Haiech, J.; Bourguignon, J.-J. Tetrahedron Lett. 2004, 45, 6343.
(29) Kummer, A. D.; Kompa, C.; Niwa, H.; Hirano, T.; Kojima, S.;
Michel-Beyerle, M. E. J. Phys. Chem. B 2002, 106, 7554.
(30) 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.
(31) Dong, J.; Solntsev, K. M.; Poizat, O.; Tolbert, L. M. J. Am. Chem.
Soc. 2007, 129, 10084.
(32) Solntsev, K. M.; Poizat, O.; Dong, J.; Rehault, J.; Lou, Y.; Burda,
C.; Tolbert, L. M. J. Phys. Chem. B 2008, 112, 2700.
(33) Weber, W.; Helms, V.; McCammon, J. A.; Langhoff, P. W. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 6177.
(78) Duvanel, G.; Banerji, N.; Vauthey, E. J. Phys. Chem. A 2007,
111, 5361.
(79) Banerji, N.; Duvanel, G.; Perez-Velasco, A.; Maity, S.; Sakai, N.;
Matile, S.; Vauthey, E. J. Phys. Chem. A 2009, 113, 8202.
(80) Furche, F.; Ahlrichs, R. J. Chem. Phys. 2002, 117, 7433.
(81) Becke, D. J. Chem. Phys. 1993, 5648.
(82) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(83) Scha¨fer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
5829.
(84) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995, 102, 346.
(85) Von Arnim, M.; Ahlrichs, R. J. Comput. Chem. 1998, 19, 1746.
(86) Deglmann, P.; Furche, F.; Ahlrichs, R. Chem. Phys. Lett. 2002,
362, 511.
(87) Deglmann, P.; Furche, F. J. Chem. Phys. 2002, 9535.
(88) Fu¨rstenberg, A.; Vauthey, E. J. Phys. Chem. B 2007, 111, 12610.
(89) Horng, M. L.; Gardecki, J. A.; Papazyan, A.; Maroncelli, M. J.
Phys. Chem. 1995, 99, 17311.
(34) Vengris, M.; van Stokkum, I. H. M.; He, X.; Bell, A. F.; Tonge,
P. T.; van Grondelle, R.; Larsen, D. S. J. Phys. Chem. A 2004, 108, 4587.
(35) Sundstro¨m, V.; Gillbro, T.; Bergstro¨m, H. Chem. Phys. 1982, 73,
439.

20 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Petkova et al.
(90) Pigliucci, A.; Duvanel, G.; Lawson Daku, M. L.; Vauthey, E. J.
Phys. Chem. A 2007, 111, 6135.
(91) Siemiarczuk, A.; Wagner, B. D.; Ware, W. R. J. Phys. Chem.
1990, 94, 1661.
(103) Fu¨rstenberg, A.; Vauthey, E. Photochem. Photobiol. Sci. 2005,
260.
(104) Sherin, P. S.; Grilj, J.; Tsentalovitch, Y. P.; Vauthey, E. J. Phys.
Chem. B 2009, 113, 4953.
(92) Wlodarczyk, J.; Kierdaszuk, B. Biophys. J. 2003, 85, 589.
(93) Ye, J. Y.; Hattori, T.; Inouye, H.; Ueta, H.; Nakatsuka, H.;
Maruyama, Y.; Ishikawa, M. Phys. ReV. B 1996, 53, 8439.
(94) Riddick J. A.; Bunger W. B. Organic SolVents; J. Wiley: New
York, 1970.
(95) Marcus, Y. Chem. Soc. ReV. 1993, 22, 409–416.
(96) Velsko, S. P.; Fleming, G. R. Chem. Phys. 1982, 65, 59.
(97) Kramers, H. A. Physica (Utrecht) 1940, 7, 284.
(98) Grote, H. F.; Hynes, J. T. J. Chem. Phys. 1980, 73, 2715.
(99) Hicks, J. M.; Vandersall, M. T.; Sitzmann, E. V.; Eisenthal, K. B.
Chem. Phys. Lett. 1987, 135, 413.
(100) Pigliucci, A.; Vauthey, E.; Rettig, W. Chem. Phys. Lett. 2009,
469.
(101) Flom, S. R.; Barbara, P. F. J. Phys. Chem. 1985, 89, 4489.
(102) Nishiya, T.; Yamauchi, S.; Hirota, N.; Baba, M.; Hanazaki, I. J.
Phys. Chem. 1986, 90, 5730.
(105) Banerji, N.; Fu¨rstenberg, A.; Bhosale, S.; Sisson, A. L.; Sakai,
N.; Matile, S.; Vauthey, E. J. Phys. Chem. B 2008, 112, 8912.
(106) Grabowski, Z. R.; Rotkiewicz, K.; Rettig, W. Chem. ReV. 2003,
103, 3899.
(107) Galievsky, V. A.; Druzhinin, S.; Demeter, A.; Jiang, Y.-B.;
Kovalenko, S. A.; Lustres, J. L. P.; Venugopal, K.; Ernsting, N. P.; Allonas,
X.; Noltemeyer, M.; Machinek, R.; Zachariasse, K. A. ChemPhysChem
2005, 6, 2307.
(108) Mohammed, O. F.; Vauthey, E. J. Phys. Chem. A 2008, 112, 3823.
(109) Kang, T. J.; Kahlow, M. A.; Giser, D.; Swallen, S.; Nagarajan,
V.; Jarzeba, W.; Barbara, P. F. J. Phys. Chem. 1988, 92, 6800.
(110) Herbich, J.; Kapturkiewicz, A. J. Am. Chem. Soc. 1998, 120, 1014.
(111) Banerji, N.; Angulo, G.; Barabanov, I. I.; Vauthey, E. J. Phys.
Chem. A 2008, 112, 9665.
JP903900B