UV-vis spectroscopy and semiempirical quantum chemical studies on methyl derivatives of annulated analogues of azafluoranthene and azulene dyes

Article abstract of DOI:10.1016/j.saa.2010.04.006

Paper reports the measured optical absorption and fluorescence spectra of 4-(2-chlorophenyl)-7-methyl-1,3-diphenyl-1H-pyrazolo[3,4-b]quinoline (MCPDPPQ), as well as 6-methyl-1,3-diphenyl-3H-indeno[1,2,3-de]pyrazolo[3,4-b]quinoline (MDPIPQ) and 9-methyl-6-phenyl-6H-5,6,7-triazadibenzo[f,h]naphtho[3,2,1-cd] azulene (MPTNA) representing cyclized five- or seven-membered regioisomeric products of MCPDPPQ, respectively. The spectra has been recorded in solvents of different polarity and compared with the results of quantum chemical calculations performed by means of the semiempirical method PM3 in combination with molecular dynamics (MD) simulations. Cyclization of MCPDPPQ into MDPIPQ or MPTNA is accompanied by a significant red shift of the first optical absorption and fluorescence bands. While the solvent polarity rises all the dyes exhibit the blue shift of the first absorption band and the red shift of the fluorescence band. These trends have been reproduced within the semiempirical calculations in combination with the Lippert-Mataga dielectric polarization model and explained by specific orientations of the dipole moments in the ground and excited states. All dyes may be considered as candidates for the luminescent or electroluminescent applications. Depending on solvent polarity they emit light in the green-yellow range of the visible spectra.

Full text of DOI:10.1016/j.saa.2010.04.006

Spectrochimica Acta Part A 77 (2010) 16–23
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
UV–vis spectroscopy and semiempirical quantum chemical studies on methyl
derivatives of annulated analogues of azafluoranthene and azulene dyes
K.S. Danela, P. Ga˛siorskib, M. Matusiewiczb, S. Całusb, T. Uchaczc, A.V. Kitykb,∗
a Department of Chemistry, University of Agriculture, Balicka str. 122, 30-149 Kraków, Poland
b Faculty of Electrical Engineering, Cze¸ stochowa University of Technology, Al. Armii Krajowej 17, 42-200 Cze¸ stochowa, Poland
c Faculty of Chemistry, Jagiellonian University, Ingardena str. 3, 30-060 Kraków, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Paper reports the measured optical absorption and fluorescence spectra of 4-(2-chlorophenyl)-7-methyl-
1,3-diphenyl-1H-pyrazolo[3,4-b]quinoline (MCPDPPQ), as well as 6-methyl-1,3-diphenyl-3H-indeno
Received 20 January 2010
Received in revised form 16 March 2010
Accepted 8 April 2010
[1,2,3-de]pyrazolo[3,4-b]quinoline
(MDPIPQ)
and
9-methyl-6-phenyl-6H-5,6,7-triazadibenzo
[f,h]naphtho[3,2,1-cd]azulene (MPTNA) representing cyclized five- or seven-membered regioiso-
meric products of MCPDPPQ, respectively. The spectra has been recorded in solvents of different
polarity and compared with the results of quantum chemical calculations performed by means of the
semiempirical method PM3 in combination with molecular dynamics (MD) simulations. Cyclization of
MCPDPPQ into MDPIPQ or MPTNA is accompanied by a significant red shift of the first optical absorption
and fluorescence bands. While the solvent polarity rises all the dyes exhibit the blue shift of the first
absorption band and the red shift of the fluorescence band. These trends have been reproduced within
the semiempirical calculations in combination with the Lippert–Mataga dielectric polarization model
and explained by specific orientations of the dipole moments in the ground and excited states. All dyes
may be considered as candidates for the luminescent or electroluminescent applications. Depending on
solvent polarity they emit light in the green-yellow range of the visible spectra.
Optical absorption spectra
Fluorescence spectra
Semiempirical calculations
Organic dyes
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
1,3-diphenyl-1H-pyrazolo[3,4-b]quinoline (hereafter MBPDPPQ),
to synthesize the two regioisomers: 6-methyl-1,3-diphenyl-
In the past decade much attention has been focused on the
spectral properties of numerous low weight organic dyes in a con-
text of their broad range of applications in optoelectronics, mainly
as fluorescence dopants for dye lasers [1–3] or electroluminescent
devices such as organic light emitting diodes or electrolumines-
cent displays [4–6]. Synthesis of new molecules with desirable
fluorescence properties has always been a challenging and difficult
task. In our group we have been dealing for many years with
pyrazoloquinoline (PQ) dyes [7–21] being good candidates for
blue or blue-green fluorescence emitters in polymer layer host.
Being inspired by an interesting heptacyclic arene of rubicene
[22], we decided to start a new program focused on the synthesis
of annulated analogues of azafluoranthenes and heteroazulenes,
starting from the derivatives of PQ [23]. During our preliminary
studies we observed the possibility from the same starters, i.e.
4-(2-chlorophenyl)-7-methyl-1,3-diphenyl-1H-pyrazolo[3,4-b]
quinoline (hereafter MCPDPPQ) or 4-(2-bromophenyl)-7-methyl-
3H-indeno[1,2,3-de]pyrazolo[3,4-b]quinoline
(MDPIPQ)
and
9-methyl-6-phenyl-6H-5,6,7-triazadibenzo[f,h]naphtho[3,2,1-cd]
azulene (MPTNA). The chemical structures of MCPDPPQ
(MBPDPPQ), MDPIPQ and MPTNA dyes may be presented by
the following scheme
where depending on chemical reaction conditions, as described
in Ref. [23], one obtains from MCPDPPQ either MDPIPQ or
MPTNA molecule. Especially interesting is isomer MPTNA because
of its novelty including an additional fused seven-membered
ring. Similar arrangements of atoms can be exemplified in the
structures of acepleiadylene [24], triazadibenzo[cd, g]azulenes
[25], and in carbon nanotubes as so-called Stone-Wales defects
∗
Corresponding author. Tel.: +48 34 3250831.
E-mail addresses: kityk@ap.univie.ac.at, andriy.kityk@univie.ac.at (A.V. Kityk).
1386-1425/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2010.04.006

K.S. Danel et al. / Spectrochimica Acta Part A 77 (2010) 16–23
17
giving rise to different deformations of carbon nanotubes
[26].
products. Thereby using bromine derivative is found to be more
preferable.
In the present paper we report the optical absorption and flu-
orescence spectra of the starter compound MCPDPPQ as well as
its regioisomeric products, i.e. five-membered MDPIPQ or seven-
membered MPTNA cyclized molecular conformations. The aim
behind of these investigations is to study an influence of the
two types of the cyclizations on the electronic properties of
the regioisomers representing annulated analogues of azafluoran-
thene and azulene dyes. On the other hand, we analyze also the
change of the electronic structure caused by solvent-modified elec-
tron transfer interactions. For this reason the optical absorption
and photoluminescence spectra of starter compound MCPDPPQ
and its regioisomeric products MDPIPQ or MPTNA are recorded
in organic solvents of different polarity and consequently sub-
jected to quantum chemical analysis within the semiempirical
model PM3.
2.1. 4-(2-Chlorophenyl)-7-methyl-1,3-diphenyl-1H-pyrazolo
[3,4-b]quinoline (2a, MCPDPPQ)
Yellow solid; Rf = 0.57; Mp = 194–196 ◦C; H1 NMR (CDCl3,
300 MHz) ı 2.60 (s, 3H, Me), 7.06–7.34 (m, 11H), 7.52 (d, 1H, J =
9.0 Hz), 7.57 (t, 2H, J = 7.5 Hz), 8.05 (s, 1H), 8.61 (d, 2H, J = 8.7 Hz);
Anal. Calcd. for C29H20ClN3; C, 78.11; H, 4.52; N, 9.42. Found: C,
78.01; H, 4.88; N, 9.26.
2.2. 4-(2-Bromophenyl)-7-methyl-1,3-diphenyl-1H-pyrazolo
[3,4-b]quinoline (2b, MBPDPPQ)
Yellow solid; Rf = 0.57; Mp = 208–211 ◦C; H1 NMR (CDCl3,
300 MHz) ı 2.59 (s, 3H, Me), 7.06–7.25 (m, 9H), 7.31 (t, 1H, J =
7.5 Hz), 7.50–7.55 (m, 2H), 7.57 (t, 2H, J = 7.5 Hz), 8.05 (s, 1H), 8.62
(d, 2H, J = 8.7 Hz); Anal. Calcd. for C29H20BrN3; C, 71.03; H, 4.11;
N, 8.57. Found: C, 71.14; H, 4.34; N, 8.61.
2. Synthesis
General routines concerning the synthesis of heterocycles
dyes are presented in Ref. [23]. Commercially available reagents
(Aldrich, Fluka, Merck) were used without further purification.
2.3. 9-Methyl-6-phenyl-6H-5,6,7-triazadibenzo[f,h]naphtho
[3,2,1-cd]azulene (3, MPTNA)
1H NMR and 13C NMR were recorded using
a Mercury-Vx
Yellow solid; Rf = 0.63; Mp = 272–273 ◦C; H1 NMR (CDCl3,
300 MHz) ı 2.57 (s, 3H, Me), 7.24 (d, 1H, J = 1.5 Hz), 7.29 (t, 1H,
J = 7.5 Hz), 7.37–7.52 (m, 4H), 7.56 (t, 2H, J = 7.5 Hz), 7.73 (dd, 1H,
J = 7.8, 1.5 Hz), 7.79–7.83 (m, 1H), 7.85 (dd, 1H, J = 7.8, 1.5 Hz), 7.94
(s, 1H), 8.31–8.34 (m, 2H), 8.60 (d, 2H, J = 8.7 Hz); Anal. Calcd. for
C29H19N3; C, 85.06; H, 4.68; N, 10.26. Found: C, 84.90; H, 4.76; N,
10.01.
300MHz Varian operating at 300 MHz respectively in CDCl3
with tetramethylsilane (TMS) as an internal standard. Melting
points were measured on a MELT-Temp and are uncorrected.
The purity of the obtained compounds was checked by TLC. Ele-
mentary analyses were performed on a Perkin-Elmer 2400 CHN
analyzer.
Heating m-anisidine, pyrazolone and 2-halobenzaldehyde in
boiling ethylene glycol produced exclusively 2, which is consistent
with our previous findings [23], see scheme below.
The rationale to place the methyl group at 7-position of 2
is to exclude any steric obstacle close to the site of cycliza-
tion in case of formation of 4. The seven-membered 3 analogue
can be prepared from both haloderivatives 2a,b, but 4 is best
obtained from 2b using a palladium catalyst [23,27]. Treating chlo-
rine analogue 2a, as described either by Jackson et al. [28] or
Smet et al. [29], did not give quite satisfactorily results, because
most of a starting material indeed was not converted into final
2.4. 6-Methyl-1,3-diphenyl-3H-indeno[1,2,3-de]pyrazolo
[3,4-b]quinoline (4, MDPIPQ)
Orange solid; Rf = 0.46; Mp = 224–226 ◦C; H1 NMR (CDCl3,
300 MHz) ı 2.64 (s, 3H, Me), 6.92 (dq, 1H, J = 7.8, 0.6 Hz), 7.04
(td, 1H, J = 8.7, 1.2 Hz), 7.33 (t, 2H, J = 7.2 Hz), 7.54–7.63 (m,

18
K.S. Danel et al. / Spectrochimica Acta Part A 77 (2010) 16–23
6H), 7.76–7.84 (m, 4H), 8.54 (d, 2H, J = 8.7 Hz); Anal. Calcd. for
29H19N3; C, 85.06; H, 4.68; N, 10.26. Found: C, 85.09; H, 4.90; N,
C
10.36.
3. Experimental and calculation procedures
To probe the solvatochromic effect on the optical absorption
and fluorescence spectra the measurements have been carried
out using as solvents cyclohexane (CHX), tetrahydrofuran (THF)
and acetonitrile (ACN). The molar concentration of the dyes sub-
jected to investigations was about 10−5 M in each case. The
measurements were performed by means of Shimadzu UV-VIS
2101 scanning spectrophotometer in the range of 230–600 nm
using a standard 1 cm path length quartz cuvette for absorption
spectrometry. The steady-state fluorescence spectra have been
excited by a mercury lamp (ꢀex = 365 nm) and recorded in a sin-
gle photon counting mode. Accordingly, the setup consisted of a
conventional spectrofluorimeter equipped by the cooled photo-
multiplier EMI 955 8B. The fluorescence spectra were corrected
for the spectral sensitivity of the detecting system. The samples
(prepared in the darkness) were degassed before experiments
using the freezing–pumping–thawing technique. The fluorescence
quantum yields were determined from the steady-state measure-
ments using the solution of quinine sulphate in 0.05 M H2SO4
as an actinometer (˚fl = 0.53). The lifetimes were determined in
the time-resolved single photon counting regime by analyzing
the temporal decay of the fluorescence spectra being excited by
picosecond diode laser (IBH-UK, ꢀ = 400 nm, ꢁ = 70 ps pulse dura-
tion).
The quantum chemical analysis has been performed by means
of the software package HyperChem-8.0 within the semiempirical
method PM3 (Parametric Model 3). The geometrical optimization
in the ground state and the molecular dynamics (MD) simu-
lations has been performed using the same Hamiltonian, PM3.
ꢀ
ꢀ
The transition dipole moments [ꢂ0i ≡ ꢂ(S0 → Si)] and the exci-
tation energies [ꢃE0i = E(Si) − E(S0)] between the ground state
(S0) and the excited states (Si, i ≥ 1), representing the configu-
ration interaction (CI) singlet states, as well as the state dipole
ꢀ
ꢀ
moments [ꢂi ≡ ꢂ(Si)] were calculated taking into account only
the singly excited CI limited by a set of 16 occupied and 16
unoccupied orbitals. Such calculation procedure has been applied
either to a certain equilibrium molecular conformation (EMC)
in vacuo (T = 0 K) or combined with the MD simulations at
T = 300 K. In the second case we consider a molecular dynamic
trajectory generated within the standard MD procedure at a
certain simulation temperature and consisting of about 200 con-
sequent instantaneous conformations that were subjected to
further quantum chemical calculations. The optical absorption
spectra in this case is just an average over all the consequent
conformations. For each particular conformation of such MD
trajectory the band shape of simulated optical absorption spec-
Fig. 1. The equilibrium molecular conformations of MCPDPPQ, MDPIPQ and MPTNA
dyes as obtained within the semiempirical quantum chemical method PM3. ꢆ1, ꢆ2
or ꢆ0 are the torsion angles defining the angular orientation of the phenyl rings Ph1,
Ph2 or Ph0, respectively.
tra is assumed to be defined by the Gaussian function, i.e. ∝
2
ꢀ
ꢄ0i|ꢂ0i
|
exp{−2.773[(hꢄ − ꢃE0i)/ꢅ]2}, where ꢄ0i = ꢃE0i/h, h is
the Planck constant, ꢅ is the empirical model parameter describ-
ing the Gaussian lineshape broadening. It has been adjusted in
each case in order to reproduce experimental bandwidth of mea-
sured absorption bands. The total time of the simulated MD
trajectories has been chosen for all dyes as 50 ps. It is certainly
much longer compared to a vibration/rotation temporal periods
of the slowest molecular moieties like methyl (Me) or phenyl (Ph)
groups. Accordingly, the resulting spectra usually well converged
already after the calculations of 100–150 conformations. For this
reason the length of the MD trajectory, consisted of 200 confor-
mations, should be sufficient in order to satisfy the convergence
criteria.
4. Results and discussion
The geometry optimization of MCPDPPQ, MDPIPQ and MPTNA
dyes, performed within the semiempirical method PM3, leads to
the equilibrium conformations as shown in Fig. 1. Each of the
dyes are characterized by several equilibrium molecular geome-
tries with nearly the same total energy. Such equilibrium states
differ each other mainly by angular orientations of their phenyl
rings Ph0, Ph1 or Ph2 which are defined in our case by the torsion
angles ꢆ0, ꢆ1 or ꢆ2, respectively (see Fig. 1). Fig. 2(a)–(e) shows

K.S. Danel et al. / Spectrochimica Acta Part A 77 (2010) 16–23
19
Fig. 2. The angular diagrams of the ground state energy E(ꢆ) as calculated in the GP (EGP , blue color) and ACN solution (EACN , red color) and corresponding to it population
distribution p(ꢆ) in the GP (gray color) and ACN solution (black color). Blue dashed lines in sections (b) and (d) correspond to metastable states. (a) E(ꢆ1) and p(ꢆ1) for Ph1
unit of MCPDPPQ dye. (b) E(ꢆ2) and p(ꢆ2) for Ph2 unit of MCPDPPQ dye. (c) E(ꢆ1) and p(ꢆ1) for Ph1 unit of MDPIPQ dye. (d) E(ꢆ2) and p(ꢆ2) for Ph2 unit of MDPIPQ dye. (e)
E(ꢆ0) and p(ꢆ0) for Ph0 unit of MPTNA dye. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
the total molecular energy E vs the torsion angle ꢆ (ꢆ ≡ ꢆ0, ꢆ1 or
ꢆ2) that have been calculated for the restrained geometries, i.e. the
conformations being obtained within a forced optimization pro-
cedure for a sequence of fixed angular orientations of a particular
phenyl group. The blue and red lines correspond here to the energy
EGP in the gas phase (GP) and ACN solution (EACN), respectively.
In the GP the chlorophenyl unit of MCPDPPQ is oriented almost
orthogonally (torsion angle ꢆ ≈ 91◦) whereas the two other phenyl
rings, Ph1 and Ph2, being linked to PQ group by C–C and C–N bonds,
respectively, are characterized by the equilibrium torsion angles
ꢆ1eq ≈ 112 ( n · 180, n = 0, 1, 2, . . .) deg and ꢆ2eq ≈ 19 or −19
angles ꢆ2eq
stable energy equivalent conformations, and the two metastable
conformations with ꢆ2
≈
23 ( n · 180, n = 0, 1, 2, . . .) deg, representing two
≈
46 ( n · 180, n = 0, 1, 2, . . .) deg. The
Ph1 ring of MDPIPQ in the GP is characterized by nearly twisted
equilibrium angular position (ꢆ1eq ≈ 90 ( n · 180, n = 0, 1, 2, . . .)
deg, see Fig. 2(c)), however due to a very flat shape of E(ꢆ1) in
the vicinity of ꢆ1eq = 90◦ its exact angular position may be defined
with a relatively low precision, i.e. ꢃꢆ1
=
3◦. In the polar solvents
the equilibrium torsion angles are subjected to their modifica-
tions. To have an idea how the solvation influences them the total
molecular energy in ACN solution EACN(ꢆ) was evaluated from the
angular dependences of the ground state energy EGP(ꢆ) and dipole
moments ꢂg(ꢆ) (calculated in the GP) and the Onsager cavity radius
a0 [30] (taken as 4.4 Å for all dyes) using the following equation
[31,32]:
(
n · 180, n = 0, 1, 2, . . .) deg. In addition, the EGP(ꢆ2)-dependence
(Fig. 2(b)) indicates also on the two metastable conformational
states at ꢆ2
42 ( n · 180, n = 0, 1, 2, . . .) deg and is almost sym-
≈
metric with respect to the angle of 90 ( n · 90, n = 0, 1, 2, . . .) deg.
Contrary to this, due to a presence of the Cl atom, the Ph1 group
appears in substantially asymmetric double-well potential as one
may see in Fig. 2(a). Similarly, MPTNA (see Fig. 2(e)) is also char-
acterized by essentially asymmetric potential in regards to its Ph0
moiety and exhibits the equilibrium twist angles ꢆ0eq at about −17◦
and 45◦, respectively. Such asymmetry is obviously related with a
specific geometry of phenyl-azulene moiety of MPTNA having sub-
stantially bent shape in the ground state (see Fig. 1). On the other
hand, the phenyl-azafluoranthene moiety of MDPIPQ is perfectly
flat. Accordingly, the orientation of Ph2 ring in the GP corresponds
to a symmetric potential (see Fig. 2(d)) with the equilibrium torsion
ꢂ2g (ꢆ)
ε − 1
EACN(ꢆ) = EGP(ꢆ) −
(1)
a03
2ε + 1
where ε = 37.5 is the static dielectric constant of ACN. From
Fig. 2(a)–(e) one can realize that in ACN environment the Ph1 ring of
both MCPDPPQ and MDPIPQ tends to a more twisted orientations
whereas Ph2 unit of these molecules as well as the Ph group of
MPTNA tend to more planar ones. However, the changes in angular
orientations of aromatic groups are rather moderate for all dyes and
do not exceeds few degrees. On the other hand, in all the cases the

20
K.S. Danel et al. / Spectrochimica Acta Part A 77 (2010) 16–23
Fig. 3. The optical absorption spectra of MCPDPPQ (black color), MDPIPQ (red
color) and MPTNA (blue color) dyes as measured (a) and calculated (b) within the
semiempirical quantum chemical method PM3. Vertical lines in section (b) are the
calculated optical absorption spectra of equilibrium molecular conformations (EMC)
(ꢅ = 0 eV), solid curves are the optical absorption spectra being calculated within the
MD simulations at T = 300 K and Gaussian lineshape broadening (ꢅ = 0.24 eV). (For
interpretation of the references to color in this figure legend, the reader is referred
to the web version of the article.)
energy barriers between the equilibrium angular orientations of the
phenyl rings are of the order of kBT (kB is the Boltzmann constant)
or less this magnitude thus according to the Boltzmann statistic
one may expect rather large angular amplitudes in their rotational
dynamics at room temperature (T = 300 K). Our further quantita-
tive evaluations are based therefore on the population analysis both
in the GP and ACN solution. Following to it the population distri-
Fig. 4. Solvatochromism of MCPDPPQ, MDPIPQ and MPTNA dyes. Sections (a), (c)
and (e) show the optical absorption spectra in the region of the first absorption
band measured in CHX (blue) and ACN (red) solutions. Sections (b), (d) and (f) are
the calculated optical absorption spectra in the GP (blue) and ACN solution (red),
respectively. Here the vertical lines are the spectra calculated by PM3 method for
equilibrium molecular conformations (EMC) (ꢅ = 0 eV) and the solid curves are
the spectra being calculated within PM3 method in combination with the MD
simulations at T = 300 K and Gaussian lineshape broadening (ꢅ = 0.24 eV). (For
interpretation of the references to color in this figure legend, the reader is referred
to the web version of the article.)
bution p(ꢆ), being presented in Fig. 2(a)–(e), has been evaluated as
ꢀ
p(ꢆ) = p0 exp(−E(ꢆ)/kBT), where p0 = [ 2ꢇ exp(−E(ꢆ)/kBT)]−1 is
0
the normalization factor and E ≡ EGP or EACN. One can see that even
for the Ph1 group of MCPDPPQ and MDPIPQ its angular deviation
may be rather large, roughly up to 40◦ for MDPIPQ and about up
to 45◦ for MCPDPPQ. The situation with the Ph2 ring of these dyes
as well as the Ph0 group of MPTNA is even more dramatic since
the population p is considerable here over the complete angular
range. For this reason the molecular dynamics aspects of MCPDPPQ,
MDPIPQ and MPTNA dyes appear in the focus of our studies and will
be considered in further analysis of their spectroscopic properties.
Fig. 3(a) shows the optical absorption spectra of the MCPDPPQ,
MPTNA and MDPIPQ measured in weakly polar CHX solvent. In
the range of 230–600 nm the absorption spectrum of these dyes
are characterized by several broad absorption bands with the most
strongest ones in the region of 250–300 nm. MCPDPPQ exhibits the
first absorption band being unstructured and centered at about 398
nm. Cyclization of MCPDPPQ into MPTNA or MDPIPQ leads to its
significant red shift. For MPTNA or MDPIPQ dyes the first absorp-
tion band looks almost unstructured and exhibits nearly the same
spectral positions at about 449 nm. Semiempirical quantum chem-
ical analysis quite properly describes such trend for both dyes,
see Fig. 3(b). However the magnitudes of corresponding spectral
shifts appears to be somewhat smaller comparing to the obtained
experimental value ꢃꢀ(exp) ≈ 51 nm. In particular, the calcula-
tions for equilibrium conformations in the GP (T = 0 K, ꢅ = 0) give
ꢃꢀ(EMC) ≈ 28 nm whereas the spectra obtained within the MD
simulations (T = 300 K, ꢅ = 0.24 eV) reveal a bit larger spectral
shift, i.e. ꢃꢀ(MD) ≈ 31 nm.
While the solvent polarity changes from CHX to ACN the spec-
tral position of the first absorption band of MCPDPPQ, MDPIPQ and
MPTNA dyes exhibits a blue shift of about 6–8 nm, see Fig. 4(a),
(c) and (e), respectively. Such hypsochromic shift in the optical
absorption is accompanied by a bathochromic (red) shift of the flu-
orescence spectra (Fig. 5). MCPDPPQ and especially MDPIPQ reveal
relatively stronger solvatochromism of their fluorescence bands
comparing to MPTNA dye for which the red shift of the fluores-

K.S. Danel et al. / Spectrochimica Acta Part A 77 (2010) 16–23
21
where ꢄfl = 1/ꢀfml , n is the solvent refractive index. All the dyes
exhibit rather similar values of Mfl in different solvents. While
the solvent polarity changes from CHX to ACN the fluorescence
moment of MCPDPPQ, MPTNA or MDPIPQ dyes decreases accord-
ingly on about of 11, 6 or 15% only. This fact suggests that the
electronic structure and molecular conformation of the fluorescent
states should not change significantly with the solvent polarity.
In a simple approximation the solvatochromic shift can be eval-
uated within the Lippert–Mataga dielectric polarization model
[13,14]. In accordance to this approach the solvent-modified
wavenumbers of the optical absorption (ꢄabs = 1/ꢀabs) or fluores-
cence (ꢄfl = 1/ꢀfl) maxima are defined as [34,35]:
ꢀ
ꢀ
ꢀ
2ꢂg(ꢂe − ꢂg)F(ε, n)
ꢄabs = ꢄ0abs
−
(4)
hca30
ꢀ
ꢀ
ꢀ
2ꢂe(ꢂe − ꢂg)F(ε, n)
ꢄfl = ꢄ0fl
−
(5)
hca30
where
ε − 1
1 n2 − 1
F(ε, n) =
−
(6)
2
2ε + 1
2
2n + 1
is the solvent polarityfunction; ꢄ0abs and ꢄ0fl arethespectralpositions
of the absorption and fluorescence maxima in the GP, correspond-
ingly; c is the light speed in vacuo; n is the refractive index of the
ꢀ
solvent, ε is the static dielectric constant of the solvent, ꢂg and
ꢀ
ꢂe are the ground and the excited state dipole moments, respec-
tively. Fig. 4(b), (d) and (f) show the first optical absorbtion band
of MCPDPPQ, MDPIPQ and MPTNA dyes, respectively, being calcu-
lated by PM3 method in the GP (F = 0) as well as in ACN solution
(F = 0.393) using the solvatochromic correction as defined by Eq.
(4). One can see, that the semiempirical calculations in combination
with the dielectric polarization model, applied to EMC, describes
properly the hypsochromic trend in the solvatochromic behavior
of the first optical absorbtion band (see vertical lines in Fig. 4(b), (d)
and (f)). Moreover, with the Onsager cavity radius a0 = 4.4 Å they
give as well a reasonable value as for the shift itself. Also impor-
tantly, the solvatochromic shift keeps the same trends and exhibits
similar magnitudes in the MD simulations (see continuous curves in
Fig. 4(b), (d) and (f)). This finding means that a rotational dynamics
of phenyl rings, being characterized by quite strong angular ampli-
tudes in all the dyes at T = 300 K, indeed does not influences the
hypsochromic trend and only slightly modifies its magnitude.
Fig. 6 shows the wavenumbers ꢄmabs and ꢄmfl vs the solvent polarity
F being extracted from the measured optical absorbtion and fluo-
rescence spectra. The slopes of the linear plots allow to determine
Fig. 5. Steady-state normalized fluorescence spectra of MCPDPPQ (a), MDPIPQ (b)
and MPTNA (c) dyes being measured in CHX (blue) and ACN (red) solutions. (For
interpretation of the references to color in this figure legend, the reader is referred
to the web version of the article.)
cence is of several time weaker. In the highly and medium polar
solvents the photoluminescence spectra are unstructured. How-
ever, some vibronic structuring of the photoluminescence bands
occurs in weakly polar CHX being more evidently seen for MDPIPQ
and MPTNA dyes (see Fig. 5(b) and (c)).
The absorption and emission maxima of MPTNA and MDPIPQ
obtained in solvents of different polarity are reported in Table 1
together with the measured quantum yield ˚fl and the fluorescence
lifetime ꢁfl. The radiationless (knr) and radiative (kfl) rate constants
as well as the fluorescence transition moment Mfl have been evalu-
ated within a simple kinetic model of an irreversible excited charge
transfer state formation as in Ref. [33]:
1 − ˚fl
˚
fl
3
3
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
knr
=
,
kfl
=
(2)
(3)
directly the values of ꢂg(ꢂe − ꢂg)/a and ꢂe(ꢂe − ꢂg)/a0 which
are collected in Table 2. On the oth0er hand, they can be evalu-
ated having the ground and excited state dipole moments being
calculated within the semiempirical model PM3. Importantly, that
the quantum chemical calculations give for both dyes a proper
ꢁfl
ꢁfl
3hkfl
2
|Mfl|
=
64ꢇ4n3ꢄf3l
Table 1
The photophysical constants of MCPDPPQ, MPTNA and MDPIPQ dyes in different solutions.
Compound
MCPDPPQ
Solvent
ꢀambs (nm)
ꢀfml (nm)
ꢈfl
ꢁfl (ns)
knr (×107, s−1
)
kfl (×107, s−1
)
Mfl (D)
CHX
ACN
398
393
452
482
0.8
0.67
17.2
26.4
1.2
1.3
4.7
2.5
2.17
1.93
MDPIPQ
MPTNA
CHX
THF
ACN
449
446
441
511
546
562
0.37
0.191
0.103
11.9
12.4
7.4
5.3
6.5
1.2
3.1
1.5
1.4
2.13
1.70
1.80
CHX
THF
ACN
450
448
444
524
536
540
0.49
0.365
0.34
16.6
16.9
17.2
3.1
3.8
3.8
3.0
2.2
2.0
2.16
1.96
2.02
ꢀambs is the absorption maxima, ꢀmfl is the fluorescence maxima, ˚fl is the fluorescence quantum yield, ꢁfl is the fluorescence lifetime, knr is the radiationless rate constant, kfl
is the radiative rate constant and Mfl is the fluorescence moment.

22
K.S. Danel et al. / Spectrochimica Acta Part A 77 (2010) 16–23
ꢀ
ꢀ
and MPTNA dyes the vector difference ꢂe − ꢂg is close to be parallel
ꢀ
ꢀ
with ꢂe and nearly antiparallel with ꢂg. Comparing the calculated
and measured solvatochromic slopes for the first optical absorp-
3
ꢀ
ꢀ
ꢀ
tion band, being defined by ꢂg(ꢂe − ꢂg)/a , one must mention
rather good agreement for both dyes. Howev0er, in regards to other
3
ꢀ
ꢀ
ꢀ
value ꢂe(ꢂe − ꢂg)/a , which defines the solvatochromic shift of
the fluorescence ban0d, the quantitative discrepancy between the
experiment and theory must be admitted. The experiment reveals
much stronger solvatochromic red shift indicating thus on consid-
ꢀ
erably larger dipole moment of the excited state (|ꢂe|) involved
into a fluorescence emission, especially in MDPIPQ dye. One must
be emphasized, that the dipole moments have been calculated in
the GP, not for the solvated phase which are the conditions of the
experiment. In addition, after the excitation the nature of the low-
ꢀ
est excited state, and accordingly its dipole moment ꢂe, may be
subjected to significant modification if the solvent and/or molecu-
lar geometry relaxations lead to situations when: i) other strongly
polar excited state of solute becomes in solvent the lowest excited
singlet (fluorescent) state or ii) initially locally excited state appears
to be strongly modified due to solvent-dependent CI between con-
figurations of this and highly polar intramolecular charge transfer
(ICT) states, see e.g. Ref. [31,32]. To treat the problem properly, a
program which includes the solvation energy in the Hamiltonian
should be used. On the other hand, the fluorescent moment Mfl
exhibits rather weak F-dependence (see Table 1) for all dyes thus
the electronic structure and molecular conformation of the fluo-
rescent excited state should not change substantially if the solvent
polarity changes from CHX to ACN. Considering these facts together
one may conclude that significant modification of lowest excited
fluorescent state should occurs already on a route from the GP to a
weakly polar CHX.
Fig. 6. The solvatochromic shift of the optical absorption [ꢄmabs = (ꢀmabs
)
−1] and fluo-
rescence [ꢄmfl = (ꢀmfl
)
−1] maxima vs solvent polarity F.
Table 2
Slopes of the solvatochromic plots for MCPDPPQ, MDPIPQ and MPTNA dyes as
determined from optical absorption and fluorescence measurements and calculated
within the semiempirical model PM3.
3
3
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Compound
ꢂg (ꢂe − ꢂg )/ao (eV)
ꢂe(ꢂe − ꢂg )/ao (eV)
MCPDPPQ
Exp.
Calc.(MD)
Calc.(EMC)
−0.076
−0.085
−0.066
0.293
–
0.006
MDPIPQ
Exp.
Calc.(MD)
Calc.(EMC)
−0.086
−0.113
−0.078
0.377
–
0.044
5. Conclusion
MPTNA
Exp.
Calc.(MD)
Calc.(EMC)
In conclusion we have reported here the measured optical
absorption and fluorescence spectra as well as the basic photophys-
ical characteristics of newly synthesized five-membered MDPIPQ
and seven-membered MPTNA dyes being obtained from the same
starter compound MBPDPPQ (MCPDPPQ) via the cyclization reac-
tion. The spectra have been recorded in the organic solvents of
different polarity in order to analyze the change of the electronic
structure caused by solvent-modified electron transfer interac-
tions. The experimental results are compared with the quantum
chemical calculations performed within the semiempirical method
PM3. The geometry optimization of MCPDPPQ, MDPIPQ and MPTNA
dyes, leads in each case to several equilibrium conformations in
the ground state that differ each other mainly by angular orienta-
tions of their phenyl rings. Such states are characterized by the
same or nearly the same total energies being separated by the
energy barriers sizeable with the thermal energy (kBT) at T = 300 K.
Accordingly, the rotational dynamics of the phenyl groups is char-
acterized in each case by quite large angular amplitudes. For this
reason the calculations of optical absorbtion spectra has been
performed within the semiempirical calculations in combination
with the MD simulations. Cyclization of MCPDPPQ into MDPIPQ or
MPTNA is accompanied by a significant red shift of the first opti-
cal absorption and fluorescence bands. While the solvent polarity
changes from CHX to ACN all the dyes exhibit the hypsochromic
shift of the first absorption band and the bathochromic shift of the
fluorescence band. These findings appears to be consistent with
the Lippert–Mataga solvatochromic model. The semiempirical cal-
culations in combination with the dielectric polarization model
describe properly the hypsochromic trend in the solvatochromic
behavior regarding the first optical absorbtion band and give a rea-
sonable value as for the shift itself. Importantly, the solvatochromic
−0.064
−0.102
−0.079
0.12
–
0.04
signs for the slopes regarding the solvatochromic behavior and
explain why the bathochromic (red) shift of the fluorescence band
is accompanied by the hypsochromic (blue) shift of the absorp-
tion band. The vector diagram in Fig. 7 presents corresponding
geometrical interpretation. Following to it the opposite trends in
solvatochromic behavior may be explained by a specific orienta-
ꢀ
ꢀ
tion of the dipole moments ꢂg and ꢂe, in particular, in MDPIPQ
Fig. 7. The vector diagram of MCPDPPQ, MDPIPQ and MPTNA dyes. The ground
ꢀ
ꢀ
ꢀ
ꢀ
state moment ꢂg , excited state moment ꢂe and their vector difference ꢂe − ꢂg as
calculated within the semiempirical PM3 method.

K.S. Danel et al. / Spectrochimica Acta Part A 77 (2010) 16–23
23
shift keeps the same trend and also exhibits similar magnitudes in
the MD simulations at T = 300 K. This finding means that a rota-
tional dynamics of phenyl rings indeed does not influence the
hypsochromic trend of the solvatochromic shift of the first opti-
cal absorption band and only slightly modifies its magnitude. The
different solvatochromic trends of the optical absorbtion and fluo-
rescent bands are explained by specific orientations of the ground
and excited state dipole moments. On the other hand, quantitative
evaluations of the solvatochromic shifts give quite well agreement
with the experimental data only in regards the optical absorp-
tion spectra of all the dyes. In the case of the fluorescence spectra
the experiment suggests much larger state dipole moment of the
excited fluorescent state comparing to the one which follows from
the semiempirical calculations. Corresponding discrepancy is espe-
cially strong for MDPIPQ dye and is assumed to be related with a
significant modification of the nature of the lowest excited singlet
(fluorescent) state due to solvent and/or geometry relaxations in
the excited state of molecules. All the dyes reveal substantial quan-
tum yield in weakly polar solvents what may be of interest for the
luminescent or electroluminescent applications. Depending on a
solvent polarity they emit light in green-yellow range of the visible
spectra.
[8] E. Kos´cien´ , J. Sanetra, E. Gondek, B. Jarosz, I.V. Kityk, J. Ebothe, A.V. Kityk, Opt.
Commun. 242 (2004) 401.
[9] E. Kos´cien´ , J. Sanetra, E. Gondek, B. Jarosz, I.V. Kityk, J. Ebothe, A.V. Kityk, Spec-
trochim. Acta, Part A: Mol. Biomol. Spectrosc. 61 (2005) 1933.
[10] E. Kos´cien´ , J. Sanetra, E. Gondek, A. Danel, A. Wisła, A.V. Kityk, Opt. Commun.
227 (2003) 115.
[11] E. Gondek, E. Koscien, J. Sanetra, A. Danel, A. Wisla, A.V. Kityk, Spectrochim.
Acta, Part A: Mol. Biomol. Spectrosc. 60 (2004) 3101.
[12] S. Całus, E. Gondek, A. Danel, B. Jarosz, A.V. Kityk, Opt. Commun. 268 (2006) 64.
[13] S. Całus, E. Gondek, A. Danel, B. Jarosz, A.V. Kityk, Opt. Commun. 271 (2007) 16.
[14] E. Kos´cien´ , E. Gondek, A.V. Kityk, Spectrochim. Acta, Part A: Mol. Biomol. Spec-
trosc. 70 (2008) 1097.
[15] S. Całus, E. Gondek, M. Pokladlo, E. Kulig, B. Jarosz, A.V. Kityk, Spectrochim. Acta,
Part A: Mol. Biomol. Spectrosc. 67 (2007) 1007.
[16] S. Całus, E. Gondek, A. Danel, B. Jarosz, M. Pokladko, A.V. Kityk, Mater. Lett. 61
(2007) 3292.
[17] E. Gondek, A. Danel, B. Kwiecien´ , J. Sanetra, A.V. Kityk, Spectrochim. Acta, Part
A 74 (2009) 685.
[18] E. Gondek, A. Danel, B. Kwiecien´ , J. Niziol, A.V. Kityk, Mater. Chem. Phys. 119
(2010) 140.
[19] E. Kos´cien´ , E. Gondek, B. Jarosz, A. Danel, J. Nizioł, A.V. Kityk, Spectrochim. Acta:
Part A 72 (2009) 582.
[20] E. Gondek, I.V. Kityk, A. Danel, Mater. Chem. Phys. 112 (2008) 301.
[21] E. Kos´cien´ , E. Gondek, M. Pokladko, B. Jarosz, R.O. Vlokh, A.V. Kityk, Mater. Chem.
Phys. 114 (2009) 860.
[22] M. Smet, R. Shukla, L. Fülöp, W. Dehaen, Eur. J. Org. Chem. 12 (1998) 2769.
[23] K.S. Danel, A. Wisła, T. Uchacz, ARKIVOC 10 (2009) 71.
[24] V. Boekelheide, G.K. Vick, J. Am. Chem. Soc. 78 (1956) 653.
[25] M. Matschke, R. Beckert, E.U. Würthwein, H. Görls, Synlett 17 (2008) 2633.
[26] H.F. Hu, Y.B. Li, H.B. He, Diam. Relat. Mater. 10 (2001) 1818.
[27] H.A. Wegner, L.T. Scott, A. de Meijere, J. Org. Chem. 68 (2003) 883.
[28] E.A. Jackson, B.D. Steinberg, M. Bancu, A. Wakamiya, L.T. Scott, J. Am. Chem. Soc.
129 (2007) 484.
References
[1] V.S. Anthonov, K.L. Hohla, Appl. Phys. B 32 (1983) 9.
[2] S. Speiser, N. Shakkour, Appl. Phys. B 38 (1985) 191.
[3] J. Tu, N. Li, Y. Chi, S. Qu, C. Wang, Q. Yuan, X. Li, S. Qiu, Mater. Chem. Phys. 118
(2009) 273.
[4] X.H. Zhang, B.J. Chen, X.Q. Lin, Q.T.Y. Wong, C.S. Lee, H.L. Kwong, et al., Chem.
Mater. 13 (2001) 1565.
[5] G. Mao, A. Orita, L. Fenenko, M. Yahiro, C. Adachi, J. Otera, Mater. Chem. Phys.
115 (2009) 378.
[6] J.-A. Cheng, C.H. Chen, H-P.D. Shieh, Mater. Chem. Phys. 113 (2009) 1003.
[7] E. Gondek, S. Całus, A. Danel, A.V. Kityk, Spectrochim. Acta, Part A: Mol. Biomol.
Spectrosc. 69 (2008) 22.
[29] M. Smet, J. Van Dijk, W. Dehaen, Synlett 4 (1999) 495.
[30] L. Onsager, J. Am. Chem. Soc. 58 (1936) 1486.
[31] A. Onkelinx, F.C. De Schryver, L. Viaene, M. Van der Auweraer, K. Iwai, M.
Yamamoto, M. Ichikawa, H. Masuhara, M. Maus, W. Rettig, J. Am. Chem. Soc.
118 (1996) 2892.
[32] M. Maus, W. Rettig, S. Depaemelaere, A. Onkelinx, F.C. De Schryver, K. Iwai,
Chem. Phys. Lett. 292 (1998) 115.
[33] A. Kapturkiewicz, J. Herbich, J. Karpiuk, J. Nowacki, J. Phys. Chem. A 101 (1997)
2332.
[34] E. Lippert, Z. Naturforsch. A10 (1955) 541.
[35] N. Mataga, Y. Kaifu, M. Koizumi, Bull. Chem. Soc. Jpn. 28 (1955) 690.