Fluorescence spectroscopic properties of Nitro-Substituted diphenylpolyenes: Effects of intramolecular planarization and intermolecular interactions in crystals

Article abstract of DOI:10.1021/jp907441p

The steady-state absorption and fluorescence properties of (E,E,E)-1,6-diaryl-1,3,5-hexatrienes (2, aryl = 2-nitrophenyl; 3, aryl = 3-nitrophenyl; 4, aryl = 4-nitrophenyl) have been investigated in solution and in the crystalline state. The solid-state absorption spectra of 2 - 4 shifted to longer wavelengths than those in solution. A combination of theoretical calculations and single-crystal X-ray structure analyses shows considerable planarization of molecules in the solid state, which is mainly responsible for the spectral red shifts. The effects of intermolecular interactions on the absorption spectra appeared to be relatively small in these crystals. This is consistent with the monomeric origin of the solid-state emission. Molecule 2 was nonfluorescent in all solvents studied, probably due to the efficient nonradiative deactivation from ionic species produced by excited-state intramolecular proton transfer (ESIPT) along the C - HO-type hydrogen bonds. The fluorescence of 3, observed only in medium polar solvents, originated from an intramolecular charge transfer (ICT*) state, while that of 4 derived from locally excited (LE*) and/or ICT* states depending on the solvent polarity. All three molecules exhibited LE* fluorescence in the solid state. No observation of ICT* emission in crystals strongly suggests the twisted geometries for ICT* (TICT) of 3 and 4 in solution. The measurable fluorescence from crystal 2 can be attributed to the restricted torsional motions in the solid excited state.

Full text of DOI:10.1021/jp907441p

172
J. Phys. Chem. A 2010, 114, 172–182
Fluorescence Spectroscopic Properties of Nitro-Substituted Diphenylpolyenes: Effects of
Intramolecular Planarization and Intermolecular Interactions in Crystals
Yoriko Sonoda,*,† Seiji Tsuzuki,‡ Midori Goto,§ Norimitsu Tohnai,| and Masaru Yoshida†
Nanotechnology Research Institute and Technical Center, National Institute of AdVanced Industrial Science and
Technology (AIST), Higashi 1-1-1, Tsukuba, Ibaraki 305-8565, Japan, Research Institute of Computational
Sciences, AIST, Umezono 1-1, Tsukuba, Ibaraki 305-8568, Japan, and Department of Material and Life
Science, Graduate School of Engineering, Osaka UniVersity, 2-1, Yamadaoka, Suita, Osaka 565-0871, Japan
ReceiVed: August 3, 2009; ReVised Manuscript ReceiVed: October 25, 2009
The steady-state absorption and fluorescence properties of (E,E,E)-1,6-diaryl-1,3,5-hexatrienes (2, aryl )
2-nitrophenyl; 3, aryl ) 3-nitrophenyl; 4, aryl ) 4-nitrophenyl) have been investigated in solution and in the
crystalline state. The solid-state absorption spectra of 2-4 shifted to longer wavelengths than those in solution.
A combination of theoretical calculations and single-crystal X-ray structure analyses shows considerable
planarization of molecules in the solid state, which is mainly responsible for the spectral red shifts. The
effects of intermolecular interactions on the absorption spectra appeared to be relatively small in these crystals.
This is consistent with the monomeric origin of the solid-state emission. Molecule 2 was nonfluorescent in
all solvents studied, probably due to the efficient nonradiative deactivation from ionic species produced by
excited-state intramolecular proton transfer (ESIPT) along the C-H· · ·O-type hydrogen bonds. The fluorescence
of 3, observed only in medium polar solvents, originated from an intramolecular charge transfer (ICT*) state,
while that of 4 derived from locally excited (LE*) and/or ICT* states depending on the solvent polarity. All
three molecules exhibited LE* fluorescence in the solid state. No observation of ICT* emission in crystals
strongly suggests the twisted geometries for ICT* (TICT) of 3 and 4 in solution. The measurable fluorescence
from crystal 2 can be attributed to the restricted torsional motions in the solid excited state.
Introduction
CHART 1: Chemical Structures and Atom Numberings
of 1-4
Light absorption and fluorescence emission of organic dyes
in the solid state is a fundamental issue not only in molecular
science but also in material science.1 In particular, fluorescence
emission from organic solids is of great interest due to their
potential use as photoactive materials such as light-emitting
diodes (LEDs),2-6 solid-state lasers,7,8 and light-emitting elec-
trochemical cells.9
Despite the extensive studies, however, understanding and
even more predicting the absorption and fluorescence spectro-
scopic behavior of organic solids is still difficult.10 The
relationship between molecular structure and spectroscopic
properties is well-known in solution, while in the solid state,
the structure-property relationship is much more complicated.
The spectra in the solid state are in general red-shifted compared
to those in solution, and the shifts can mainly be attributed to
molecular planarization and intermolecular interactions in the
solid state.11-14 Flexible molecules in solution are often frozen
in more planar conformations in the solid state. The molecular
planarization leads directly to the extended delocalization of
the π-electronic system. The intermolecular interactions, which
are closely related to the molecular arrangements in the solid
state, can also affect the energy levels of the electronic systems
of the molecules. For better understanding of the absorption
and fluorescence behavior of organic molecules in the solid state,
systematic studies on the effects of intramolecular planarization
and intermolecular interactions on the spectroscopic properties
are required.
(E,E,E)-1,6-Diphenyl-1,3,5-hexatriene (DPH) (1, Chart 1) is
a highly fluorescent molecule with a one-dimensional polyenic
structure and is commercially available as a fluorescence probe
in biological membrane studies. The emission properties of 1
* Corresponding author: fax, +81-29-861-4673; e-mail, y.sonoda@
aist.go.jp.
† Nanotechnology Research Institute.
‡ Research Institute of Computational Sciences.
§ Technical Center.
| Osaka University.
10.1021/jp907441p  2010 American Chemical Society
Published on Web 11/23/2009

Nitro-Substituted Diphenylpolyenes
J. Phys. Chem. A, Vol. 114, No. 1, 2010 173
in solution have long attracted much attention because of its
dual fluorescence from the first two singlet excited states, S1
and S2.15,16 As for the emission properties in the solid state, on
the other hand, much remains to be clarified. We showed that
p,p′-disubstituted DPHs exhibited measurable fluorescence also
in the solid state.17 The fluorescence maximum (λf) moved to
longer wavelengths as the strength of electron-withdrawing
(EW) or electron-donating (ED) nature of substituents increased.
Our recent studies further show that, in a series of donor-acceptor
p-(n-alkoxy)-p′-nitro-substituted18 and ring-fluorinated19 DPHs,
the crystal packing patterns strongly affect not only λf but also
the origin of the solid-state fluorescence (e.g., monomeric,
excimeric, and ground-state aggregated species), reflecting the
magnitude of the intermolecular interactions in crystals.
We previously investigated the photophysical properties of
nitro-substituted DPH (4, Chart 1) in solution.20 Aromatic nitro
compounds are in general only weakly fluorescent or completely
nonfluorescent in solution, due to the efficient nonradiative decay
(NRD) processes such as singlet-triplet intersystem crossing
(ISC) and internal conversion (IC).21-27 In addition, they often
undergo intra- and intermolecular hydrogen abstraction, resulting
in rapid photodecomposition.27-29 Despite these general tenden-
cies of nitroaromatics, 4 exhibited relatively strong fluorescence
(quantum yield φf ) 0.61 in dichloromethane (DCM)) and was
proven to be photochemically very stable in solution. The
strongly red-shifted fluorescence of 4 in polar solvents was
assigned to the emission from an intramolecular charge transfer
state (ICT*), for which the twisting of C6H4(Ar)-NO2 single
bond was strongly suggested. If this is the case, then the solid-
state fluorescence behavior of 4 can differ significantly from
that in solution, as a result of molecular planarization along
with possible intermolecular interactions involving the nitro
groups in crystals.
C, 67.07; H, 4.38; N, 8.69. Found: C, 67.14; H, 4.29; N, 8.59.
νmax (KBr) 1602, 1567, 1510, 1474, 1441, 1338, 1301, 1277,
1007, 993, 956, 781, 741, 695, and 625 cm-1. 1H NMR (CDCl3)
δ 7.93 (2H, d, J 8.1, arom.), 7.71 (2H, d, J 7.8, arom.), 7.58
(2H, apparently (app.) t, J 7.6, arom.), 7.38 (2H, app. t, J 7.7,
arom.), 7.16 (2H, d, J 15.3, triene), 6.91 (2H, app. ddd, J 15.1,
6.9 and 3.0, triene), 6.65 (2H, app. dd, J 7.0 and 2.9, triene).
13C NMR (CDCl3) δ 147.9, 135.2, 133.6, 132.9, 132.4, 128.1,
128.0, 127.7, and 124.8. UV-vis λmax (acetonitrile (AN)) 327
nm (ꢀ ) 35000 M-1 cm-1), 373 nm (ꢀ ) 27900 M-1 cm-1);
λ
max (Tol) 331 nm (ꢀ ) 27100 M-1 cm-1), 385 nm (ꢀ ) 27800
M-1 cm-1).
(E,E,E)-1,6-Di(3-nitrophenyl)-1,3,5-hexatriene (3). The crude
product of the reaction between 3-nitrobenzaldehyde and the
1
salt contained a large amount of Z isomers as shown by H
NMR and UV-vis spectra. To induce Z-to-E isomerization, the
product was irradiated in DCM with Pyrex-filtered light at room
temperature in air. The solvent was evaporated and the resulting
yellow solid (predominantly E,E,E) was recrystallized from Tol
to give single crystals of 3 suitable for X-ray analysis. The purity
was checked by HPLC. mp 234-236 °C (lit.30 165-166 °C).
Anal. Calcd for C18H14N2O4: C, 67.07; H, 4.38; N, 8.69. Found:
C, 67.19; H, 4.37; N, 8.37. νmax (KBr) 1572, 1531, 1472, 1439,
1355, 1281, 1098, 1073, 1018, 995, 892, 828, 806, 735, 673,
and 624 cm-1. 1H NMR (CDCl3) δ 8.28-8.29 (2H, m, arom.),
8.06-8.10 (2H, m, arom.), 7.71 (2H, d, J 7.9, arom.), 7.50 (2H,
app. t, J 8.0, arom.), 7.02 (2H, app. ddd, J 15.4, 7.0 and 3.0,
triene), 6.69 (2H, d, J 15.5, triene), 6.62 (2H, app. dd, J 7.1
and 3.1, triene). 13C NMR (CDCl3) δ 148.8, 136.6, 134.5, 132.2,
131.5, 131.0, 129.6, 122.1, and 120.8. UV-vis λmax (AN) 352
nm (ꢀ ) 68000 M-1 cm-1); λmax (Tol) 360 nm (ꢀ ) 57700 M-1
cm-1).
It should be noted that the melting point of our crystalline
sample was higher by 70 °C than the value reported in the
literature.30 This is probably because the literature value was
for the sample containing Z isomer(s).
In this study, the absorption and fluorescence properties of
o,o′-, m,m′-, and p,p′-dinitro-substituted DPHs (2, 3, and 4,
respectively, in Chart 1) were investigated in solution and in
the crystalline state. These structural isomers are expected to
have different molecular planarity and different packing patterns
in crystals, depending on the position of the nitro substituent
on the benzene ring. We performed ab initio and DFT
calculations and single-crystal X-ray analyses to clarify the
structures of molecules in solution and in the solid state. Crystals
2-4 were all photostable. No Z-E isomerization or [2 + 2]
cycloaddition was observed in these crystals.
(E,E,E)-1,6-Di(4-nitrophenyl)-1,3,5-hexatriene (4). The crude
product of the reaction between 4-nitrobenzaldehyde and the
salt was recrystallized from Tol. The recrystallized sample,
which was unsuitable for X-ray analysis, was dissolved in AN.
Single crystals of 4 suitable for X-ray analysis were obtained
by very slow evaporation of the AN solvent at room temperature
in the dark. The purity was checked by HPLC. mp 240-241
°C (lit.32 195-197 °C, lit.30 218-219 °C, lit.31 229-231 °C).
Anal. Calcd for C18H14N2O4: C, 67.07; H, 4.38; N, 8.69. Found:
C, 67.04; H, 4.24; N, 8.66. νmax (KBr) 1588, 1509, 1335, 1178,
Experimental Section
1108, 994, 870, 833, 747, and 690 cm-1
146.8, 143.4, 135.3, 132.9, 131.7, 126.9, and 124.2. H NMR
data were reported previously.20 UV-vis λmax (AN) 410 nm (ꢀ
) 76400 M-1 cm-1); λmax (Tol) 410 nm (ꢀ ) 75700 M-1 cm-1).
.
13C NMR (CDCl3) δ
Materials. Compound 1 was purchased from Wako (scintil-
lation grade) and used without further purification. Compounds
2-4 were synthesized from nitrobenzaldehydes (TCI) and (E)-
2-butene-1,4-bis(triphenylphosphonium chloride) (TCI) by pro-
cedures similar to those described in the literature.30 Elemental
analyses were performed using a CE instruments EA-1100. IR
spectra were recorded on a Mattson Infinity Gold FT-IR
1
For 4, some different melting point values are reported in
the literature. We found that the melting point of the sample
obtained by simple recrystallization from Tol was 228-229 °C,
whereas the single crystals obtained as above had mp 240-
241 °C.
Photostability of 2-4. Although 2 was photostable in the
solid state, it was decomposed on prolonged irradiation in
solution, as shown by HPLC and UV-vis absorption spectra.
The fluorescence spectra in solution were therefore recorded
by minimum exposure to light to avoid any photoreactions30
during the measurements.
1
spectrometer. H and 13C NMR spectra were recorded on a
Varian Gemini-300 BB spectrometer (300.1 and 75.5 MHz,
respectively) with tetramethylsilane (TMS) as internal reference.
All solvents used in the measurements of absorption and
fluorescence spectra were of spectroscopic grade (Dojin).
(E,E,E)-1,6-Di(2-nitrophenyl)-1,3,5-hexatriene (2). The crude
product of the reaction between 2-nitrobenzaldehyde and the
phosphonium salt was recrystallized three times from toluene
(Tol) to give single crystals of 2 suitable for X-ray analysis.
The purity was checked by HPLC. mp 230-231 °C (lit.31
230-231.5 °C; lit.30 232-233 °C). Anal. Calcd for C18H14N2O4:
Compounds 3 and 4 were photostable in solution and in the
solid state. No Z-E photoisomerization from the E,E,E isomers
was observed at least in DCM or Tol.

174 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Sonoda et al.
TABLE 1: Absorption and Fluorescence Data of 2 and 3 in Solution
2a
3
solvent
λa (nm)
λa (nm)
λf (nm)
∆Ess (cm-1
)
φf
τs (ns)
methylcyclohexane
carbon tetrachloride
toluene
1,4-dioxane
tetrahydrofuran
chloroform
dichloromethane
acetone
N,N-dimethylformamide
acetonitrile
375, 328
380, 331
385, 331
381, 331
380, 331
385, 328
384, 328
375, 330
381, 333
373, 327
372, 328
373, 355, 340
379, 358, 342
379, 360, 346
377, 357, 343
377, 357, 343
378, 358, 343
377, 357, 342
374, 354, 338
378, 359, 344
371, 352, 338
370, 352, 338
nonfluorescent
7186
8285
9729
482
513
547
574
0.00042
0.0068
0.011
0.72
1.1
1.7
1.3
10589
0.0052
nonfluorescent
nonfluorescent
nonfluorescent
nonfluorescent
nonfluorescent
nonfluorescent
methanol
a Nonfluorescent in all solvents studied.
Measurements of Absorption Spectra. Absorption spectra
in solution were measured in air at room temperature using a
Shimadzu UV-3150 spectrometer. All solutions were highly
diluted ((1.0-3.5) × 10-5 M). Absorption spectra in the solid
state were obtained by Kubelka-Munk conversion of diffuse
reflectance spectra. The reflectance spectra were recorded on a
Jasco V-560 spectrometer equipped with an integrating sphere
accessory (model ISV-469). The sample solids were placed
between quartz plates (40 × 10 mm2).
Measurements of Fluorescence and Fluorescence Excita-
tion Spectra, Fluorescence Quantum Yields, and Lifetimes.
The corrected fluorescence and fluorescence excitation spectra
in solution and in the solid state were measured at room
temperature in air using a SPEX Fluorolog-3 spectrometer.
For the fluorescence measurements in solution, the excitation
wavelengths were set at 330 and 380 nm for 2, 355 nm for 3,
and 400 nm for 4, unless otherwise noted. Concentration of the
sample solutions was (1.0-3.5) × 10-6 M. Values of φf in
solution were determined using a solution of quinine sulfate in
1 N H2SO4 as a standard (φf ) 0.546).33 For 4, we measured
the fluorescence data in some selected solvents under air-
saturated conditions to find that they were practically identical
with those measured under degassed conditions which we
reported previously.20
Fluorescence and excitation spectra of the crystalline samples
were recorded using the front face geometry. The sample crystals
were placed between quartz plates (40 × 10 mm2) on the sample
holder. For all compounds, the excitation and emission wave-
lengths were set at 420 and 600 nm, respectively. The
measurements of the solid-state φf were performed at Osaka
University by using a Jasco FP-6500 spectrofluorometer with a
fluorescence integrate sphere unit (model ISF-513). The sample
crystals were encapsulated in a quartz cell (30 × 30 × 0.3 mm3)
under deoxygenated conditions. The excitation wavelength was
350 nm.
Fluorescence decay curves of 3 in solution and those of 2-4
in the solid state were obtained by the time-correlated single-
photon counting (TCSPC) method, using a HORIBA NAES 700
equipped with a subnanosecond nitrogen laser system (excitation
wavelength ) 337 nm). The monitor wavelengths were set at
λf as shown in Tables 1 and 6.
For the solid-state absorption and fluorescence measurements,
the samples were not ground to a powder in all cases.
Single Crystal X-ray Structure Analyses. The single crystal
X-ray diffraction measurements of 2 and 3 were performed at
183 K using a Bruker SMART CCD area-detector diffractometer
with graphite monochromated Mo KR radiation (λ ) 0.71073
Å). Data collection and reduction and empirical absorption
Figure 1. Absorption spectra of 2 in (a) methylcyclohexane, (b)
toluene, (c) dichloromethane, and (d) acetonitrile.
correction were carried out using SMART (Bruker2001),
SAINTPLUS (2001), and SADABS (2001).34 The structure was
solved by direct methods using SIR9235 and refined by full
matrix least squares on F2 with SHELXTL.36 The non-hydrogen
atoms were refined anisotropically. Hydrogen atoms were placed
in geometrically calculated positions and refining a riding model.
Computational Method. The Gaussian 03 program37 was
used for the ab initio and DFT calculations. The 6-311G** basis
set was used for all calculations unless otherwise noted. The
torsional potentials were calculated by the HF, MP2,38 and DFT
(B3LYP)39,40 methods using the HF level optimized geometries.
The torsional angle was fixed, and other geometrical parameters
were fully optimized in the torsional potential calculations. The
excitation energies were calculated using the CIS,41 CIS(D),42
and time-dependent DFT (TD-DFT)43 methods. The BHHLYP
and B3LYP functionals were used for the DFT calculations.39,40
The CIS and CIS(D) calculations were carried out using the
HF and MP2 level optimized geometries, respectively. The TD-
DFT calculations were carried out using the geometries opti-
mized by the DFT calculations with the same functionals. The
HOMOs and LUMOs shown in Figure 4 and Figures S4-S6
in Supporting Information were calculated at the HF/6-311G**
level using the MP2/6-31G* level optimized geometries. The
frozen-core approximation was applied for the MP2 calculations.
Results and Discussion
1. Absorption and Fluorescence Properties in Solution.
1.1. Absorption Properties. Figures 1 and 2 show the absorption
spectra of 2 and 3 in solution, respectively. Table 1 summarizes
the absorption maxima (λa) in various kinds of solvents with
different polarity. The data of 1 are shown in Figure S1
(Supporting Information) for comparison. Although the spectra

Nitro-Substituted Diphenylpolyenes
J. Phys. Chem. A, Vol. 114, No. 1, 2010 175
solvent polarity increased, λa red-shifted from 395 nm in MCH
to 422 nm in DMF. The spectrum became more structureless,
probably due to the increased solute-solvent interactions. Here,
the spectra in methanol and in AN were very similar, suggesting
that the structureless features in polar solvents were not due to
the intermolecular hydrogen bonds. The solvent effects on the
spectrum in 4 were thus small but significantly larger than those
in 2 and 3. In contrast to the cases of 2 and 3, however, the
plot of λa in wavenumber as a function of f(n2) shows no
correlation in 4 (Figure S3(a) in Supporting Information). On
the other hand, for solvents with similar n, the plot of λa versus
f(ꢀs) shows the decrease in the energy of λa with the increase in
f(ꢀs) (Figure S3(b) in Supporting Information). Although the
plot exhibits only a loose correlation, the slope of the linear fit
is calculated to be -4260 cm-1. The dependence of λa on f(n2)
and f(ꢀs) in 4 clearly differs from those in 2 and 3.
Figure 2. Absorption spectra of 3 in (a) methylcyclohexane, (b)
toluene, (c) dichloromethane, and (d) acetonitrile.
1.2. Ground-State Geometries and Transition Assignments.
Ground-State Geometries. The relative energies for the three
kinds of conformational isomers (conformers) of 2 (A, B, and
C in Chart S1 in Supporting Information) were calculated at
the MP2/6-311G**//MP2/6-311G** level.51 Conformer A is
shown to be 2-3 kcal/mol more stable than B and C, suggesting
that A is the major isomer and the populations of B and C are
less than 3% at room temperature. The optimized geometrical
parameters for A, whose structure is shown in Chart 1, are
summarized in Table 2. The C1-C6-C7-C8 (ArsCHd) and
O1-N1-C1-C6 (ArsNO2) torsional angles are -150.0° and
37.6°, respectively (MP2/6-311G**), showing somewhat twisted
conformation of the molecule. The O1 atom of the nitro group
and the H7 atom of the triene are in close proximity in the
optimized structure, which suggests the existence of attractive
electrostatic interaction between O1 and H7. In this case it is
very likely that there exists the weak intramolecular C-H· · ·O
hydrogen bond of the type described by Desiraju and Steiner.53,54
The major geometrical parameters for the hydrogen bond are
given in Table 3. The π-conjugation and the hydrogen bonding
decrease the torsional angles, while the steric repulsion between
O1 and H7 and between H5 and H8 atoms increases the angles.
The somewhat twisted conformation of 2 is thus a result of the
balance of these interactions.
of 4 have been reported previously by us20 and others,32,44 they
are shown in Figure S2 (Supporting Information) and sum-
marized in Table S1 (Supporting Information) for readers’
convenience.
In the spectra of 2, two broad bands were observed around
380 and 330 nm in all solvents studied. The relative intensity
of the band at 380 nm to that at 330 nm significantly decreased
with increasing solvent polarity, and the positions of λa moved
slightly. The spectrum thus showed only a small dependence
on the polarity of solvent. To understand the solvent-induced
spectral shifts more quantitatively, the relationship between
λa and the Onsager polarity functions, f(n2) ) (n2 - 1)/(2n2 +
1) and f(ꢀs) ) (ꢀs - 1)/(2ꢀs + 1), where n is the refractive index
and ꢀs is the static dielectric constant, was investigated.45-48 The
plots of λa in wavenumber as a function of f(n2) exhibit a weak
linear correlation (Figure S3(a) in Supporting Information).
Although the correlation is rather poor, the slopes of the linear
fits for the bands around λa ) 380 and 330 nm are calculated
to be -16030 and -4730 cm-1, respectively. This suggests that
the absorption around 380 nm is more strongly dependent on
f(n2) than that around 330 nm. On the other hand, the plots of
λa as a function of f(ꢀs) show no significant correlation, even
for the solvents having similar values of n (Figure S3(b) in
Supporting Information).
In 3, λa was observed at 373 nm (0-0) with its vibrational
progressions at 355 nm (0-1) and 340 nm (0-2) in methyl-
cyclohexane (MCH) solvent. Among them the 0-1 peak was
the strongest in intensity. The energy spacings were 1243 and
1359 cm-1, corresponding to the CdC and CsC stretches of
conjugated trienes.49,50 The solvent effects on the spectrum were
small. The spectral shapes did not change greatly, although the
spectrum was less structured in AN than in MCH. Even when
the solvent changed from low polar MCH to highly polar
AN, the peak positions shifted only slightly. Interestingly,
however, the plot of λa in wavenumber as a function of f(n2)
displays a reasonably good linear correlation with a slope of
-11360 cm-1 (Figure S3(a) in Supporting Information). Whereas,
if solvents of similar n values are considered, the plot of λa
versus f(ꢀs) shows a linear correlation with a slope of -490
cm-1 (Figure S3(b) in Supporting Information). From the results
it may be possible that the solvent-induced shift of this
absorption band is dominated mainly by dispersion interaction.46,47
The positions of λa of 3 were very near to those of 1 in all
solvents examined (Figure S1 in Supporting Information).
We also calculated the relative energies for the three
conformers for 3 (A, B, and C in Chart S2 (Supporting
Information)) to find that A was slightly (0.3-0.6 kcal/mol)
more stable than B and C. The small energy differences suggest
that molecule 3 exists as a conformational mixture at room
temperature in solution. Table 4 shows the results of optimiza-
tion for conformer A, which is found in the X-ray determined
crystal structure as described later. The results for molecule 4
are summarized in Table 5. As can be seen, the optimized
structures of 3 and 4 are more planar than the structure of 2.
Here it should be noted, however, that the torsional angles of
the ArsCHd single bonds in 3 (C6-C5-C7-C8) and 4
(C5-C4-C7-C8) given by the DFT calculations with
BHHLYP and B3LYP functionals are significantly smaller than
those calculated by the HF and MP2 methods. Similar discrep-
ancies between the MP2 and B3LYP structures are reported
for oligo(thiophene)s.55-57
Tables 2, 4, and 5 show the calculated values of bond length
alternation (BLA). BLA refers to the difference between the
averaged values of single (CsC) and double (CdC) bonds. It
is one of the most important structural parameters that charac-
terize the electronic and optical properties of one-dimensional
conjugated compounds such as polyenes and polyynes, since it
strongly affects the band gap of these molecules.58,59 For each
The spectrum of 4 in MCH showed λa at 395 nm (0-1) as
the highest peak and some vibrational structures. The spacings
of 1209 and 1393 cm-1 were similar to the values in 3. As the

176 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Sonoda et al.
B3LYP
TABLE 2: Major Geometrical Parameters for the X-ray and Optimized Structures of 2
X-ray
calcd valuesa
BHHLYP
molecule Ab
molecule Bb
HF
MP2
torsion angle (degree)
C1–C6–C7–C8
O1–N1–C1–C6
–163.9(1)
22.8(2)
–160.3(1)
36.1(2)
–147.5
28.0
–150.0
37.6
–153.5
24.8
–158.6
25.5
bond length (Å)
C6sC7 (r1)
C7dC8 (r2)
C8sC9 (r3)
C9dC9* (r4)
BLAc
1.469(2)
1.344(2)
1.437(2)
1.351(3)
0.107
1.469(2)
1.342(2)
1.438(2)
1.348(3)
0.110
1.480
1.327
1.459
1.330
0.142
1.465
1.358
1.442
1.362
0.094
1.463
1.335
1.440
1.338
0.116
1.463
1.352
1.438
1.355
0.098
a The 6-311G** basis set was used. b Crystallographically independent two molecules. Each molecule has a center of symmetry. c Bond
length alternation; BLA ) (r1 + r3)/2 - (2r2 + r4)/3. For atom numbering, see Chart 1.
TABLE 3: Intramolecular Hydrogen Bond Geometry for 2
TABLE 5: Major Geometrical Parameters for the X-ray
and Optimized Structures of 4
X-ray
calcdb
molecule Aa
molecule Ba
calcdb
X-raya
HF
MP2
BHHLYP B3LYP
d(O1· · ·H7)c (Å)
d(O1· · ·C7)c (Å)
C7H7O1 (deg)
2.269
2.768(2)
111.9
2.391
2.814(2)
106.6
2.432
2.781
96.9
torsion angle (degree)
C5sC4sC7sC8
C11sC12sC13sC14 -0.7(2)
O2sN1sC1sC6
18.0(2)
14.8
0.0
23.1
0.0
0.1
0.0
a Crystallographically independent two molecules. b At the MP2/
-1.4(2)
2.7(2)
8.1
-0.1
6-311G** level. c d: distance. For atom numbering, see Chart 1.
O3sN2sC16sC15
bond length (Å)
TABLE 4: Major Geometrical Parameters for the X-ray
and Optimized Structures of 3
C4sC7 (r1)
C7dC8 (r2)
C8sC9 (r3)
C9dC10 (r4)
C10sC11 (r5)
C11dC12 (r6)
C12sC13 (r7)
BLAc
1.459(2)
1.474
1.329
1.457
1.331
1.462
1.360
1.441
1.363
1.457
1.337
1.438
1.340
1.457
1.353
1.435
1.357
1.340(2)
1.437(2)
1.342(2)
1.437(2)
1.341(2)
1.457(2)
0.107
X-ray
calcda
molecule Ab molecule Bb HF MP2 BHHLYP B3LYP
torsion angle (degree)
C6sC5sC7sC8 -4.6(3)
O2sN1sC1sC6 -5.8(2)
-0.6(3) 11.0 22.4
0.1
0.0
0.1
0.136
0.091
0.110
0.092
-1.9(2) -0.7 -11.7 0.0
a Reference 72. b The 6-311G** basis set was used. c Bond length
alternation. For the X-ray structure, BLA ) (r1 + r3 + r5 + r7)/4 -
(r2 + r4 + r6)/3. For the optimized structures, BLA ) (r1 + r3)/2 -
(2r2 + r4)/3. For atom numbering, see Chart 1.
bond length (Å)
C5sC7 (r1)
C7dC8 (r2)
C8sC9 (r3)
C9dC9* (r4)
BLAc
1.470(2)
1.338(2)
1.447(2)
1.345(3)
0.119
1.466(2) 1.475 1.463
1.335(3) 1.329 1.359
1.435(2) 1.458 1.441
1.338(4) 1.331 1.363
1.459
1.460
1.352
1.437
1.356
0.095
1.336
1.439
1.339
0.112
0.115
0.137 0.092
mol (MP2/6-311G**//HF/6-31G*).63 These are close to experi-
mental values.62,63 The energy barrier for the ArsCHd bond
in 2 is higher than that in styrene. This is probably due to the
large steric repulsion between O1 and H8 in 2, which is absent
in styrene. On the other hand, the barrier height for the ArsNO2
rotation in 2 is lower than that in nitrobenzene. In nitrobenzene,
the planar structure is the most stable, as the steric repulsion
between the O atom of the nitro group and the aromatic H at
the ortho position relative to the nitro group is small.64 While
in 2, the planar structure is enhanced in energy and no longer
the energy minimum because of the steric repulsion between
O1 and H7. In this case, the somewhat twisted structure with
the ArsNO2 torsional angle of around 150° becomes the most
stable. This results in the low barrier height for the rotation of
ArsNO2 bond in 2 relative to that in nitrobenzene. Thus the
calculated barrier heights for the rotation of ArsCHd and
ArsNO2 bonds in 2 are consistent with those reported for
styrene and nitrobenzene, respectively.
Figure 3 also shows the torsional potentials for the ArsCHd
and ArsNO2 bonds calculated by the HF and B3LYP methods
using the HF level geometries for comparison. The barrier
heights for the rotation of ArsCHd bond obtained by the HF
and B3LYP methods are 4.8 and 5.9 kcal/mol, respectively. The
barrier heights for the ArsNO2 bond by the two methods are
2.7 kcal/mol (HF) and 2.9 kcal/mol (B3LYP). These values are
not greatly different from those obtained by the MP2 method.
Also the potential minima are found in a similar region of
a The 6-311G** basis set was used. b Crystallographically
independent two molecules. Each molecule has
a center of
symmetry. c Bond length alternation; BLA ) (r1 + r3)/2 - (2r2
r4)/3. For atom numbering, see Chart 1.
+
molecule, BLA from HF is larger than, and those from MP2
and B3LYP are smaller than those determined by X-ray
analyses. BLA from BHHLYP is in fairly good agreement with
the experimental values, as in the cases of a donor-acceptor
diphenylpolyene60 and polyacetylene oligomers.61
Torsional Potentials. Figure 3 shows the torsional potentials
for 2 that were calculated by rotating the ArsCHd or ArsNO2
torsional angle of the most stable conformer A at the MP2/6-
311G**//HF/6-311G** level. The barrier heights for the internal
rotation around these bonds are approximately 4.5 kcal/mol
(ArsCHd) and 2.0 kcal/mol (Ar-NO2). The low barrier heights
suggest that the rotation around these bonds occurs nearly freely
at room temperature in solution. In the figure, we see that the
ArsCHd angles for most of the molecules are in the range of
115-240° and the ArsNO2 angles are in the range of 0-70 or
115-180°. The potentials are shallow in a wide range of
torsional angles around the minima, suggesting that these bonds
have various torsional angles at room temperature in solution
(i.e., the presence of various kinds of conformers in equilibrium).
As reported earlier, the barrier height for the internal rotation
in styrene is calculated to be about 2.5 kcal/mol (MP4(SDQ)/
6-31G//HF/6-31G*)62 and that in nitrobenzene is about 4.8 kcal/

Nitro-Substituted Diphenylpolyenes
J. Phys. Chem. A, Vol. 114, No. 1, 2010 177
Figure 4. HOMO, LUMO, and LUMO+2 of 2.
acceptor diphenylpolyene.60 The values from TD-BHHLYP are
the nearest to the experiments among the results of the present
calculations. The calculated excitation energy of 1 is very similar
to the value of 3, in agreement with the experiments.
Assignments of Transition. The calculations show that the
two absorption bands around 380 and 330 nm observed in the
spectrum of 2 are mainly due to the electronic transition from
HOMO to LUMO and that from HOMO to LUMO+2,
respectively (Table S2 in Supporting Information). Figure 4
shows the HOMO, LUMO, and LUMO+2 in the optimized
structure. These orbitals are delocalized over the whole of the
molecule, indicating the effective π-delocalization between the
nitro group and DPH moiety. Clearly, significant CT from
the aromatic DPH to the nitro group occurs associated with
the HOMO-LUMO and HOMO-LUMO+2 transitions. In
LUMO+2, the π-orbitals of O1 and C7 atoms are in close
proximity to each other. This indicates that the cooperative twists
along the ArsNO2 and ArsCHd single bonds are important
for the HOMO-LUMO+2 transition. Further, this explains the
fact that no corresponding band was observed in the spectra of
3 and 4. For o,o′-dinitro-substituted 1,4-diphenyl-1,3-butadiene
(DPB), two absorptions are similarly observed at 340 and 278
nm in its solution spectrum.30
The absorption band centered at 355 nm in 3 is mainly due
to the HOMO-LUMO transition, although contribution from
HOMO-LUMO+2 is also significant in the TD-B3LYP
calculation (Table S3 in Supporting Information). Also in 3,
the HOMO-LUMO transition has CT character from DPH to
the nitro group. On the other hand, the absorption at 354 nm in
1 is assigned to the HOMO-LUMO transition with π-π*
character. Although the spectral features of 1 and 3 were very
similar, the nature of the transition was thus entirely different.
Theabsorptionat395nmin4isassignedtotheHOMO-LUMO
transition (Table S4 in Supporting Information). The HOMO
and LUMO indicate the effective π-delocalization between the
nitro group and DPH. The absorption is clearly due to the CT
transition from DPH to the nitro group. The HOMOs and
LUMOs of 1, 3, and 4 are displayed in Figures S4-S6 of the
Supporting Information.
Figure 3. Torsional potentials for (a) C1-C6-C7-C8 and (b)
O1-N1-C1-C6 bonds of 2 calculated by the HF, MP2, and B3LYP
methods using the HF level optimized geometries. The 6-311G** basis
set was used. For atom numbering, see Chart 1. The solid lines are
only guides for the eyes.
torsional angles for the ArsCHd and ArsNO2 rotation,
although the details are somewhat different.
The barrier heights for the internal rotation around ArsCHd
and ArsNO2 bonds in 3 and 4 will be close to the values in
styrene and nitrobenzene molecules, respectively. Considering
the low barrier heights calculated for styrene and nitrobenzene
mentioned above, we can expect that the rotation around
ArsCHd and ArsNO2 bonds in 3 and 4 is almost free in
solution.
Vertical Excitation Energies. To understand the electronic
transitions for the absorption bands of 2-4, vertical excitation
energies were calculated using the optimized structures for the
most stable conformers. The results are summarized in Tables
S2-S4 in Supporting Information.
For all three molecules, the CIS method gave considerably
higher excitation energies than the experimental values deter-
mined from the absorption spectra. The overestimation of the
excitation energies by CIS has been reported for other short
polyenes.65,66 The CIS(D) method gave the energies lower by
approximately 30 nm than those by CIS, approaching the
experimental values considerably. The improvement is probably
due to the fact that CIS(D) considers the double excitation,
which is important for polyene molecules65,66 but absent in
CIS.42,67
In contrast, the TD-B3LYP calculations yielded the excitation
energies lower than those from the experiments. Typically, TD-
DFT fails to describe precisely the excited states with substantial
charge transfer (CT) or double excitation character,60,67 espe-
cially in extended π-electron conjugation systems.66 In our case,
the BHHLYP functional gave the excitation energies higher than
those from B3LYP. Similar results are reported for a donor-
1.3. Fluorescence Properties. Molecule 2 was nonfluorescent
in all solvents studied. The optimized structure suggests the
existence of the C-H· · ·O intramolecular hydrogen bonds in
the ground state. In this case, the excited-state intramolecular
proton transfer (ESIPT) along the hydrogen bonds will occur
within a very short time domain,68,69 although the C-H· · ·O
hydrogen bonds are considered to be rather weak compared to

178 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Sonoda et al.
the typical ones like N-H· · ·O and O-H· · ·O. The resulting
ionic species from ESIPT in the singlet excited state will rapidly
deactivate nonradiatively by IC to the ground state. In the
efficient IC from such an ionic species, out-of-plane torsional
motions associated with the hydrogen bonds are shown theoreti-
cally to be important.69 Also, IC in 2 can involve the twisting
motion around the Ar-NO2 bond, as suggested for nitroperylene
in aprotic solvents like AN.24
methanol probably results from the intermolecular hydrogen
bonds between solute and solvent.
For 3, the plot of λf in wavenumber as a function of f(ꢀs) -
f(n2) exhibits no linear correlation. The plot for 4 shows the
decrease in the energy of λf with the increase in f(ꢀs) - f(n2)
(Figure S8 in Supporting Information). However, it exhibits only
a poor correlation, probably due to the complex nature of dual
fluorescence.
Molecule 3 exhibited fluorescence only in medium polar
solvents such as 1,4-dioxane and THF. Table 1 summarizes the
fluorescence data, and Figure 5 shows the absorption and
emission spectra in solution. The fluorescence spectra showed
no dependence on the excitation wavelength. The fluorescence
excitation spectra were essentially the same as the absorption
spectra and showed no dependence on the emission wavelength.
As the solvent polarity increased from carbon tetrachloride to
THF, λf red-shifted by 92 nm from 482 to 574 nm. The
observation is in sharp contrast to the fact that only a small
solvent dependence was observed in λa. As a result of this, the
Stokes shift (∆Ess) increased from 7186 cm-1 in carbon
tetrachloride to 10589 cm-1 in THF. Accordingly, the overlap
between the absorption and emission spectra was very small or
almost none. The position of λf moved to longer wavelengths
with increasing solvent polarity, suggesting that the emissive
state was more stabilized in more polar solvents. The emission
of 3 will therefore originate from an intramolecular charge
transfer state (ICT*), formed from an initially photoproduced,
locally excited state (LE*). Unlike the weakly structured
absorption spectra, the emission spectra were all broad and
structureless, although the spectrum in carbon tetrachloride was
rather noisy. These spectral features are also characteristic of
the ICT* emission.
The strong solvent dependence of φf was similar for 3 and 4.
The low φf in polar solvents can be attributed to the efficient
IC due to the lowered energies of ICT* (“the energy-gap low”).70
Whereas, in low polar solvents, ISC rates should probably be
considered. In the case of 4, the quantum yield of ISC increased
from <0.01 in AN to 0.89 in MCH.20
2. Absorption and Fluorescence Properties in the Solid
State. Figure 6 shows the absorption, fluorescence, and fluo-
rescence excitation spectra of crystals 2-4. The absorption and
fluorescence data are summarized in Table 6.
2.1. Absorption Properties. In the absorption spectrum of
2, λa was observed at 505 nm. Unlike that, two bands of different
electronic origins were observed in solution, only one broad
band was seen in the solid state.
In the spectrum of 3, we observed a main peak at 432 nm
and two weak peaks at 398 and 372 nm. The energy spacings
for them were 1756 and 1978 cm-1, somewhat larger than the
typical values for the CdC and CsC stretches of conjugated
trienes.49,50 It is unclear whether these weak peaks are the
vibrational progressions of the main peak or not.
In 4, a peak at 508 nm and its vibrational progressions at
474 and 441 nm were observed. The spacings of 1412 and 1579
cm-1 were similar to the values in solution.
For all three molecules, λa in the solid state red shifted
compared to those in solution. From λa of the longer wavelength
band of 2 and those of the 0-1 peaks of 3 and 4 in MCH
solution shown in Table 1 and Table S1 (Supporting Informa-
tion), and λa in the solid state shown in Table 6, the red shifts
for 2, 3, and 4 were calculated to be 6865, 5021, and 5631
cm-1, respectively. The shift for 2 was clearly larger than the
shifts for 3 and 4, thus leading to similar λa for 2 and 4 in the
solid state. Note that λa of 4 in solution was located at a longer
wavelength than that of 2.
2.2. Fluorescence Properties. The solid-state fluorescence
spectra were all broad and structureless. In contrast to the fact
that no emission from 2 was observed in solution, its fluores-
cence spectrum was measurable in the solid state. Although ∆Ess
of 3 was somewhat larger than those of 2 and 4 (Table 6), the
overlap of the absorption (or the fluorescence excitation) and
fluorescence spectra was considerably large for each molecule
(Figure 6).
Fluorescence lifetime (τs) and φf data are shown in Table 1.
The fluorescence decay curves were able to be analyzed by
monoexponential function to give τs of 1-2 ns. The value of
φf increased with an increase in solvent polarity to reach the
maximum in medium polar solvents and fell again. The low φf
can be attributed to small radiative rate constants (kf ) φf/τs)
on the order of 105-106 s-1, consistent with the weakly
forbidden nature of the ICT* transition.70
Despite the spectral similarity in the absorption of 1 and 3,
the highly solvent-dependent fluorescence behavior of 3 differed
entirely from that of 1 (Figure S1 in Supporting Information).
It is quite reasonable that the nitro groups in 3, which are absent
in 1, play an essential role in the formation of ICT*.
The fluorescence spectra of 4 are shown in Figure S7
(Supporting Information) and summarized in Table S1 (Sup-
porting Information). Although the emission properties of 4 in
solution have been reported previously,20 here we describe the
results briefly to compare them with those of 3. The fluorescence
spectrum was structured with λf of 460 nm in low polar MCH,
whereas in polar AN the spectrum became broad and struc-
tureless and λf dramatically red shifted to 580 nm. The
fluorescence origin of 4 was determined to be LE* in low polar
solvents such as MCH and carbon tetrachloride, LE* and ICT*
(dual fluorescence) in medium polar solvents such as THF and
chloroform, and ICT* in highly polar solvents such as DMF
and AN. These assignments were based on the steady-state and
picosecond time-resolved fluorescence data in various kinds of
solvents with different polarity. It is worth noting that the
position of λf in AN further red shifted to 654 nm in methanol,
and φf in this solvent was much lower than the value in AN
(Table S1 in Supporting Information). The observation in
As the origin of the solid-state fluorescence, monomeric (LE*
and ICT*) and excimeric (or excimer-like) species, and mo-
lecular aggregates having the ground-state energy minima and
thus their own absorption bands, would be possible. In the
present case, no dependence of fluorescence spectrum on the
excitation wavelength was observed, the fluorescence excitation
spectra were fundamentally the same as the absorption spectra,
and the excitation spectra showed no dependence on the
emission wavelength. These strongly suggest that the fluores-
cence of 2-4 derives not from the ground-state molecular
aggregates but from the monomeric or excimeric species.
Further,byconsideringtherelativelylargeabsorption-fluorescence
spectral overlaps, we can conclude that the fluorescence from
crystals 2-4 originates from the monomeric LE* states.

Nitro-Substituted Diphenylpolyenes
J. Phys. Chem. A, Vol. 114, No. 1, 2010 179
Figure 6. (a) Absorption, (b) fluorescence excitation, and (c) fluo-
rescence spectra of crystals 2-4.
TABLE 6: Absorption and Fluorescence Data for Crystals
2-4
compound λa (nm) λexc (nm) λf (nm) ∆Ess (cm-1
)
φf
a
b
2
3
4
505
432
508
503
439
504
570
559
613
2258
5259
3372
0.0046
0.0079
0.0068
a The maximum wavelength in the fluorescence excitation
spectrum. b Calculated from λa and λf.
The solid-state φf values of 2-4 were small (Table 6), and
their fluorescence decay curves were similar. Although the
perfect analysis of the decay curve was impossible due to small
φf and intrinsic inhomogeneity of solid samples, τs for crystals
2-4 were roughly estimated by single-exponential fitting to be
0.9-1 ns. Values τs of 1-2 ns are typical for the single-molecule
fluorescence of small organic dyes in the solid state.18,71
Figure 5. Absorption and fluorescence spectra of 3 in (a) carbon
tetrachloride, (b) toluene, (c) 1,4-dioxane, and (d) tetrahydrofuran.

180 J. Phys. Chem. A, Vol. 114, No. 1, 2010
TABLE 7: Crystal Data of 2-4
Sonoda et al.
dihedral angle between the DPH planes of the nearest molecules
is 8.28°. The molecular coplanarity in 4 is thus higher than that
in 2. The sheets stack further along the a axis with an interplane
distance of 3.47 Å by N · · ·O dipole-dipole and π-π stacking
interactions.
Thus, although linked by similar types of intermolecular
hydrogen bonds, the molecules are arranged very differently in
each crystal, depending on the position of the nitro substituent
on the benzene ring.
4. Effects of Intramolecular Planarization and Intermo-
lecular Interactions. 4.1. Absorption Properties. For all
molecules, λa in the solid state red shifted from those in solution.
The spectral red shifts should mainly be attributed to the
planarization of molecules and the intermolecular interactions
in crystals.
2
3
4a
formula
formula weight
C18H14N2O4
322.31
C18H14N2O4
322.31
C18H14N2O4
322.31
yellow, needle
crystal color, habit yellow, needle
yellow, needle
crystal size (mm) 0.45 × 0.07 × 0.05 0.30 × 0.10 × 0.07 0.30 × 0.25 × 0.02
crystal system
space group
a (Å)
b (Å)
c (Å)
R (deg)
ꢀ (deg)
monoclinic
P21/n
19.341(2)
3.8902(4)
20.212(2)
90
90.594(2)
90
1520.7(3)
4
triclinic
P-1
monoclinic
P21/c
3.8710(4)
21.440(2)
18.709(2)
90
95.555(2)
90
1545.4(3)
4
6.306(1)
10.849(2)
12.374(2)
107.847(3)
98.976(3)
103.827(3)
758.0(2)
2
γ (deg)
V (Å3)
Z
Dcalc (g/cm3)
T (°C)
1.408
-90
231
1.412
-90
236
1.385
-90
241
mp (°C)
R1 (I > 2σ(I))
The red shifts were 5021 cm-1 for 3 and 5631 cm-1 for 4. In
3, the molecules experienced almost complete planarization from
solution to the solid state, as shown by the X-ray analysis. This
results in more effective π-delocalization, leading to a consider-
able red shift of λa in the solid state. The degree of molecular
planarization in 4 was similar to (or somewhat smaller than)
that in 3, and thus the spectral shifts due to planarization would
not be very different for 3 and 4, although the red shift due to
planarization around Ar-NO2 may be somewhat different for
each. Therefore, the small difference of approximately 600 cm-1
between the total red shifts of 3 and 4 suggests that the shifts
due to intermolecular interactions are also similar for 3 and 4.
The intermolecular interactions in crystal 3 are expected to be
small, considering its herringbone structure. Consequently, the
effects of intermolecular interactions on the spectrum should
be similarly small in crystal 4 having π-stacked structure.
The spectral shift for 2 (6865 cm-1) was clearly larger than
that for 4. On comparison of the molecular arrangements in
crystals 2 and 4, it is unlikely that the intermolecular interactions
in 2 are much larger than those in 4 to induce a large spectral
difference between them. Thus, we can safely say that the
spectral shift due to the interactions in 2 is small as in 4.
Accordingly, the larger red shift in 2 should be attributed to
the larger shift due to molecular planarization in the solid state.
This is somewhat unexpected, because molecule 2, unlike 4, is
nearly planar but still distorted to some extent even in the solid
state.
The torsional potentials of 2 for the most stable conformer
A are shallow in a wide range of angles around the minima
(Figure 3). The flatness of the potential in the ground state
suggests that the observed absorption spectrum in solution is a
superposition of spectra for the various kinds of conformers
equilibrated in solution. For oligo(p-phenyleneethynylene)s, the
potential for the twisting of phenyl group in the excited state is
shown to be similar in shape to that of the ground state, but
more strongly dependent on the torsional angle than in the
ground state.73,74 The results are used to interpret their confor-
mation-dependent spectroscopic phenomena observed in en-
semble and single molecule spectroscopy. If we assume that
this is also the case for our present molecules, then the larger
red shift in 2 than in 4 may be explained by the larger angle-
dependence of potential for 2 in the excited state.
0.0444
0.0482
0.0400
a Reference 72.
3. Crystal Structures. Table 7 shows the single-crystal data
of 2-4 obtained by X-ray structure analyses.
3.1. Molecular Structures. The major geometrical parameters
for the structures of 2-4 are depicted in Tables 2, 4 and 5.
In the crystal structure of 2, torsional angles of ArsCHd
and ArsNO2 are -160 to -165° and 20-35°, respectively.
This indicates that the molecules are nearly but not completely
planar even in the solid state. Molecules 3 and 4 have more
planar conformations in crystals.
As in the optimized structure, the C-H· · ·O intramolecular
hydrogen bonds will exist in crystal 2. In Table 3, we compare
the geometrical parameters for the hydrogen bond in the MP2
optimized and X-ray determined molecular structures. Although
the angle of C-H· · ·O is somewhat larger in the X-ray
structures, the distances of d(O1· · ·H7) and d(O1· · ·C7) are not
largely different for these structures.
3.2. Molecular Arrangements. The crystal packing diagrams
for 2-4 are shown in Figure S9 in Supporting Information.
In the crystal structure of 2, there exist two crystallographi-
cally independent molecules, A and B. Each molecule has a
center of symmetry. Molecules A and B are linked to each other
by the intermolecular C-H· · ·O hydrogen bonds between the
O atoms of the nitro groups and the H atoms of the aromatic
rings or the trienes to form sheetlike structures. These sheets
are piled up along the b axis with the interplane distance of
3.49 Å for the stacking of two molecules of A and 3.55 Å for
those of B. For the nearest molecules of A and B in the sheet,
the dihedral angles for the least-squares planes of DPH (18
carbons) are 18.29° between A and B and 45.49° between A
and B′ (molecule B′ is related a screw axis with B; see Figure
S9 (Supporting Information)).
Also in 3, two crystallographically independent molecules,
A and B, are contained in the unit cell. Molecules A and B are
each linked through the intermolecular C-H· · ·O hydrogen
bonds between the O atoms of the nitro group and the H atoms
of the aromatic ring at the ortho position relative to the nitro
group. A pair of this type of hydrogen bonds makes a hexagonal
pattern between the nearest molecules along the long molecular
axis to form one-dimensional tapes of · · ·A· · ·A· · ·A· · · and
· · ·B· · ·B· · ·B· · ·. These tapes are arranged in a herringbone
fashion. The dihedral angle made by the DPH planes for the
nearest molecules of A and B is 83.49°.
Thus, the red shifts in the absorption spectra for 2-4 are
mainly attributed to the intramolecular planarization in the solid
state. The effects of intermolecular interactions on the spectra
appear to be relatively small in these crystals.
The structure of 4 was reported previously.72 As in 2, the
molecules in this structure are linked via intermolecular
C-H· · ·O hydrogen bonds to make sheets. In the sheets, the
4.2. Fluorescence Properties. Although 2 was completely
nonfluorescent in solution, the molecule exhibited measurable
fluorescence in the solid state. This can be attributed to the

Nitro-Substituted Diphenylpolyenes
J. Phys. Chem. A, Vol. 114, No. 1, 2010 181
restricted torsional motions in the solid state. As described
above, the intramolecular motions can possibly increase the IC
efficiency of 2 in solution. The enhancement of the solid-state
fluorescence intensity due to molecular planarization in the
excited state and the suppression of nonradiative torsional
deactivation has been considered for distyrylbenzenes.12
Conclusions
The absorption spectra of 2-4 in solution exhibited only a
small dependence on the solvent polarity; however, λa were
weakly correlated with the Onsager polarity functions. For these
molecules, theoretical calculations suggest that the internal
rotation around ArsCHd and ArsNO2 single bonds occurs
almost freely, and various kinds of conformers are present at
room temperature in solution. The calculated barrier heights for
the rotation of ArsCHd and ArsNO2 bonds in 2 are consistent
with those reported for styrene and nitrobenzene, respectively.
The solid-state absorption spectra of 2-4 were red-shifted
compared to those in solution. In the solid state, the torsional
motions were restricted and molecules were frozen in nearly or
almost completely planar conformations, as determined by the
X-ray analyses. We can therefore expect considerable planariza-
tion of molecules in the solid state, which is mainly responsible
for the spectral red shifts in the absorption. The molecular
arrangement in each crystal was significantly different; however,
the effects of intermolecular interactions on the absorption
spectra appeared to be relatively small. Consistently, the origin
of the solid-state fluorescence was all monomeric.
Although 2 was nonemissive in all solvents studied, the
molecule exhibited measurable fluorescence in the solid state.
The efficient NRD in solution would be due to very rapid IC
from the proton-transferred ionic state produced by ESIPT along
the CH · · ·O hydrogen bonds. In the efficient IC, out-of-plane
torsional motions possibly play an important role. The measur-
able fluorescence emission from crystal 2 can therefore be
attributed to the restricted torsional motions in the solid excited
state. The fluorescence of 3 in solution originated from ICT*,
and that of 4 from LE*and/or ICT* depending on the solvent
polarity. In contrast, they showed LE* fluorescence in the solid
state. No observation of ICT* emission in crystals strongly
suggests the twisted geometries for ICT* of 3 and 4 in solution.
Thus, 3 and 4 provide new examples for TICT molecules.
The fluorescence from crystals 2-4 was of all monomeric
origin, consistent with the relatively small effects of intermo-
lecular interactions on the absorption spectra. The LE* states
of 3 and 4 responsible for the solid-state emission are expected
to be almost completely planar, as those observed in the crystal
structures. Therefore, no observation of ICT* fluorescence in
the solid state strongly suggests that the ICT* states of 3 and 4
responsible for the red-shifted emission in solution have twisted
geometries (TICT).75
p-(Dimethylamino)benzonitrile (p-DMABN), one of the most
well-known TICT molecules in solution, showed a single
emission band originating from LE* in the crystalline state.76
The results show that ICT does not occur in the crystalline
p-DMABN, similar to our observations for 3 and 4. We should
note here, however, that the ICT* fluorescence of DMABN in
solution was only observed for its p-derivative, and the ICT*
emission from m-derivative has not been reported so far.77
In contrast to the case of 4, we previously observed that the
fluorescence behavior of p,p′-dicyano-substituted DPH was
essentially independent of the solvent polarity.20 Although the
EW strength of the cyano group is significantly weaker than
that of the nitro group, the difference in the fluorescence
behavior between them appears to be too large to be understood
only by the difference in the EW strength. To explain the results,
we assumed a distorted geometry for ICT* of 4. The twisting
of Ar-NO2 bond should lead to ICT* with strong CT character,
resulting from electronic decoupling due to nearly perpendicular
geometry of the π-planes of DPH and the nitro group, while
twisting of the Ar-CN bond does not lead to complete
decoupling between DPH and the cyano group because of the
intrinsic nature of the Ct N triple bond. The evidence supporting
this assumption has been provided by the present study.
Acknowledgment. We thank Dr. Y. Kawanishi (AIST) for
the use of the fluorescence spectrometer, and Dr. K. Ohta (AIST)
for valuable comments on the calculated results of vertical
excitation energies and transition assignments. We also ac-
knowledge Ms. S. Shibasaki (AIST) for the elemental analyses
of 2-4.
p-(n-Alkoxy)-p′-nitro-substituted DPHs also exhibited LE*
fluorescence of monomeric origin in the crystalline state.18
Absence of ICT* emission in the solid state shows the red-
shifted fluorescence in polar solvents to be originated from
TICT*. The involvement of TICT was also suggested in the
fluorescence behavior of nitro-cyano-substituted DPBs in solu-
tion.78 Thus, TICT* fluorescence has been observed for several
kinds of nitro-substituted diphenylpolyenes in solution and may
be common in this class of compounds.
Supporting Information Available: Crystallographic data
of 2 and 3 in CIF format; crystal and structure refinement data
of 2-4; chemical structures of conformational isomers of 2 and
3; absorption and fluorescence spectra of 1 and 4 in solution;
plots of λa and λf of 2-4 in solution versus the Onsager polarity
functions; HOMOs and LUMOs of 1, 3, and 4; crystal packing
diagrams of 2-4; vertical excitation energies of 2-4; and full
author list for ref 37. This material is available free of charge
via the Internet at http://pubs.acs.org.
The effects of the intramolecular planarization on the
fluorescence spectroscopic properties of 2-4 are considerably
large, while those of the intermolecular interactions appear to
be relatively small in these crystals. The results are in contrast
to our previous observations for alkoxy-nitro-substituted18 and
fluorinated19 DPHs, in which the intermolecular interactions in
crystals greatly affected the solid-state spectroscopic properties.
This suggests that, at least for one-dimensional conjugated
molecules like DPH, the magnitude of the intermolecular
interactions in crystals is highly sensitive to the relative positions
of the π-orbitals of the linear conjugated chains of the interacting
molecules. Depending on the molecular arrangements in crystals,
the interactions can be strong enough to induce emission from
excimeric or aggregated species as observed in some cases of
fluorinated DPHs,19 whereas, they can be rather weak only to
have limited effects on the spectra as in the present molecules.
References and Notes
(1) Zollinger, H. Color Chemistry, Syntheses, Properties and Applica-
tions of Organic Dyes and Pigments, 3rd ed.; Wiley-VCH: Weinheim,
Germany, 2003.
(2) Hung, L. S.; Chen, C. H. Mater. Sci. Eng., R 2002, 39, 143.
(3) Kulkarni, A. P.; Tonzola, C. J.; Babel, A.; Jenekhe, S. A. Chem.
Mater. 2004, 16, 4556.
(4) Kelley, T. W.; Baude, P. F.; Gerlach, C.; Ender, D. E.; Muyres,
D.; Haase, M. A.; Vogel, D. E.; Theiss, S. D. Chem. Mater. 2004, 16, 4413.
(5) Veinot, J. G. C.; Marks, T. J. Acc. Chem. Res. 2005, 38, 632.
(6) Burn, P. L.; Lo, S.-C.; Samuel, I. D. W. AdV. Mater. 2007, 19,
1675.
(7) Hide, F.; D´ıaz-Garc´ıa, M. A.; Schwartz, B. J.; Andersson, M. R.;
Pei, Q.; Heeger, A. J. Science 1996, 273, 1833.

182 J. Phys. Chem. A, Vol. 114, No. 1, 2010
Sonoda et al.
(8) Tessler, N.; Denton, G. J.; Friend, R. H. Nature 1996, 382, 695.
(9) Pei, Q.; Yu, G.; Zhang, C.; Yang, Y.; Heeger, A. J. Science 1995,
269, 1086.
(10) Abyan, M.; de Caro, D.; Fery-Forgues, S. Langmuir 2009, 25, 1651.
(11) Oelkrug, D.; Tompert, A.; Gierschner, J.; Egelhaaf, H.-J.; Hanack,
M.; Hohloch, M.; Steinhuber, E. J. Phys. Chem. B 1998, 102, 1902.
(12) An, B.-K.; Kwon, S.-K.; Jung, S.-D.; Park, S. Y. J. Am. Chem.
Soc. 2002, 124, 14410.
(13) Mizobe, Y.; Tohnai, N.; Miyata, M.; Hasegawa, Y. Chem. Commun.
2005, 1839.
(14) Mizobe, Y.; Hinoue, T.; Yamamoto, A.; Hisaki, I.; Miyata, M.;
Hasegawa, Y.; Tohnai, N. Chem.sEur. J. 2009, 15, 8175.
(15) Turek, A. M.; Krishnamoorthy, G.; Sears, D. F., Jr.; Garcia, I.;
Dmitrenko, O.; Saltiel, J. J. Phys. Chem. A 2005, 109, 293.
(16) Itoh, T. J. Chem. Phys. 2005, 123, 064302.
(17) Sonoda, Y.; Kawanishi, Y.; Ikeda, T.; Goto, M.; Hayashi, S.;
Yoshida, Y.; Tanigaki, N.; Yase, K. J. Phys. Chem. B 2003, 107, 3376.
(18) Sonoda, Y.; Goto, M.; Tsuzuki, S.; Tamaoki., N. J. Phys. Chem. A
2006, 110, 13379.
(19) Sonoda, Y.; Goto, M.; Tsuzuki, S.; Tamaoki., N. J. Phys. Chem. A
2007, 111, 13441.
(20) Sonoda, Y.; Kwok, W. M.; Petrasek, Z.; Ostler, R.; Matousek, P.;
Towrie, M.; Parker, A. W.; Phillips, D. J. Chem. Soc., Perkin Trans. 2
2001, 308.
(21) Mesaros, M.; Bonesi, S. M.; Ponce, M. A.; Erra-Balsells, R.; Bilmes,
G. M. Photochem. Photobiol. Sci. 2003, 2, 808.
(44) Cundall, R. B.; Johnson, I.; Jones, M. W.; Thomas, E. W.; Munro,
I. H. Chem. Phys. Lett. 1979, 64, 39.
(45) Onsager, L. J. Am. Chem. Soc. 1936, 58, 1486.
(46) Suppan, P. J. Photochem. Photobiol., A 1990, 50, 293.
(47) Banerji, N.; Angulo, G.; Barabanov, I.; Vauthey, E. J. Phys. Chem.
A 2008, 112, 9665.
(48) Gierschner, J.; Mack, H.-G.; Lu¨er, L.; Oelkrug, D. J. Chem. Phys.
2002, 116, 8596.
(49) Lin-Vien, D.; Colthup, N. B.; Fateley, W. G.; Grasselli, J. G. The
Handbook of Infrared and Raman Characteristic Frequencies of Organic
Molecules; Academic Press: San Diego, CA, 1991.
(50) Kohler, B. E. Chem. ReV. 1993, 93, 41.
(51) For actual solution, we should also consider the presence of s-cis-
s-trans isomers around the single bonds of the triene.15,52 However, only
all s-trans isomer was taken into account in the present case to avoid
calculations that are too complex.
(52) Bunker, C. E.; Lytle, C. A.; Rollins, H. W.; Sun, Y.-P. J. Phys.
Chem. A 1997, 101, 3214.
(53) Desiraju, G. R.; Steiner, T. The Weak Hydrogen Bond in Structural
Chemistry and Biology; Oxford University Press: Oxford, 1999.
(54) Desiraju, G. R. Chem. Commun. 2005, 2995.
(55) Ort´ı, E.; Viruela, P. M.; Viruela, R.; Effenberger, F.; Herna´ndez,
V.; Navarrete, J. T. L. J. Phys. Chem. A 2005, 109, 8724.
(56) Meng, S.; Ma, J. J. Phys. Chem. B 2008, 112, 4313.
ˇ
(57) Solc, R.; Lukesˇ, V.; Klein, E.; Griesser, M.; Kelterer, A.-M. J. Phys.
Chem. A 2008, 112, 10931.
(22) Morales-Cueto, R.; Esquivelzeta-Rabell, M.; Saucedo-Zugazagoitia,
J.; Peon, J. J. Phys. Chem. A 2007, 111, 552.
(23) Zugazagoitia, J. S.; Almora-D´ıaz, C. X.; Peon, J. J. Phys. Chem. A
2008, 112, 358.
(24) Mohammed, O. F.; Vauthey, E. J. Phys. Chem. A 2008, 112, 3823.
(25) Crespo-Herna´ndez, C. E.; Burdzinski, G.; Arce, R. J. Phys. Chem.
A 2008, 112, 6313.
(58) Yang, S.; Kertesz, M. J. Phys. Chem. A 2006, 110, 9771.
(59) Sonoda, Y.; Tsuzuki, S.; Tamaoki, N.; Goto, M. Acta Crystallogr.
2007, C63, o196.
(60) Wu, C.; Tretiak, S.; Chernyak, V. Y. Chem. Phys. Lett. 2007, 433,
305.
(61) Jacquemin, D.; Perpe`te, E. A.; Ciofini, I.; Adamo, C. Chem. Phys.
Lett. 2005, 405, 376.
(26) Zugazagoitia, J. S.; Collado-Fregoso, E.; Plaza-Medina, E. F.; Peon,
J. J. Phys. Chem. A 2009, 113, 805.
(62) Tsuzuki, S.; Tanabe, K.; Osawa, E. J. Phys. Chem. 1990, 94, 6175.
(63) Head-Gordon, M.; Pople, J. A. Chem. Phys. Lett. 1990, 173, 585.
(64) Takezaki, M.; Hirota, N.; Terazima, M.; Sato, H.; Nakajima, T.;
Kato, S. J. Phys. Chem. A 1997, 101, 5190.
´
(27) Arce, R.; Pino, E. F.; Valle, C.; Agreda, J. J. Phys. Chem. A 2008,
112, 10294.
(28) Il’ichev, Y. V. J. Phys. Chem. A 2003, 107, 10159.
(29) Naumov, P.; Sakurai, K.; Ishikawa, T.; Takahashi, J.; Koshihara,
S.; Ohashi, Y. J. Phys. Chem. A 2005, 109, 7264.
(30) Leznoff, C. C.; Hayward, R. J. Can. J. Chem. 1971, 49, 3596.
(31) Mitsudo, T.; Fischetti, W.; Heck, R. F. J. Org. Chem. 1984, 49,
1640.
(32) Johnson, I. D.; Thomas, E. W.; Cundall, R. B. J. Chem. Soc.,
Faraday Trans. 2 1985, 81, 1303.
(33) Melhuish, W. H. J. Phys. Chem. 1961, 65, 229.
(34) SMART Version 5.625, Bruker AXS, Madison, WI. SAINTPLUS
Version 6.22, Bruker AXS, Madison, WI. Sheldrick, G. M. SADABS,
Program for scaling and correction of area, detector data, University of
Go¨ttingen, Germany, 1996.
(35) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Brula,
M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435.
(36) Sheldrick, G. M., SHELXTL Version 6.12, Bruker AXS, Madison,
WI, 2000.
(37) Frisch, M. J.; et al. Gaussian 03, ReVision C.02; Gaussian, Inc.:
Wallingford, CT, 2004.
(65) Hsu, C.-P.; Hirata, S.; Head-Gordon, M. J. Phys. Chem. A 2001,
105, 451.
(66) Ma, H.; Liu, C.; Jiang, Y. J. Chem. Phys. 2005, 123, 084303.
(67) Grimme, S.; Neese, F. J. Chem. Phys. 2007, 127, 154116.
(68) Scheiner, S. J. Phys. Chem. A 2000, 104, 5898.
(69) Sobolewski, A. L.; Domcke, W. J. Phys. Chem. A 2007, 111, 11725.
(70) Klessinger, M.; Michl, J. Excited States and Photochemistry of
Organic Molecules; VCH Publishers, Inc.: New York, 1995; sections 1-3
and 5-2.
(71) Van Hutten, P. F.; Krasnikov, V. V.; Hadziioannou, G. Acc. Chem.
Res. 1999, 32, 257.
(72) Sonoda, Y.; Kawanishi, Y.; Goto, M. Acta Crystallogr. 2005, E61,
o1200.
(73) Sluch, M. I.; Godt, A.; Bunz, U. H. F.; Berg, M. A. J. Am. Chem.
Soc. 2001, 123, 6447.
(74) Li, N.; Jia, K.; Wang, S.; Xia, A. J. Phys. Chem. A 2007, 111,
9393.
(75) Grabowski, Z. R.; Rotkiewicz, K.; Rettig, W. Chem. ReV. 2003,
103, 3899.
(76) Demeter, A.; Zachariasse, K. A. Chem. Phys. Lett. 2003, 380, 699.
(77) Druzhinin, S. I.; Dubbaka, S. R.; Knochel, P.; Kovalenko, S. A.;
Mayer, P.; Senyushkina, T.; Zachariasse, K. A. J. Phys. Chem. A 2008,
112, 2749.
(78) Singh, A. K.; Darshi, M.; Kanvah, S. J. Phys. Chem. A 2000, 104,
464.
(38) Mφller, C.; Plesset, M. S. Phys. ReV. 1934, 46, 618.
(39) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(40) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(41) Foresman, J. B.; Head-Gordon, M.; Pople, J. A.; Frisch, M. J. J.
Phys. Chem. 1992, 96, 135.
(42) Head-Gordon, M.; Rico, R. J.; Oumi, M.; Lee, T. J. Chem. Phys.
Lett. 1994, 219, 21.
(43) Stratmann, R. E.; Scuseria, G. E.; Frisch, M. J. J. Chem. Phys.
1998, 109, 8218.
JP907441P