Conformational analysis of polymethine dyes derived from the 2-azaazulene

Article abstract of DOI:10.1016/j.molstruc.2011.10.003

A systematic investigation of the conformational structure was performed for the series of symmetrical and unsymmetrical mono-, tri-, pentamethine cyanines, and styryl dyes bearing 2-azaazulenium terminal group. The rotation energy barriers of terminal gr

Full text of DOI:10.1016/j.molstruc.2011.10.003

Journal of Molecular Structure 1007 (2012) 52–62
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Conformational analysis of polymethine dyes derived from the 2-azaazulene
a
a
a
Aleksey B. Ryabitskii a,b, , Julia L. Bricks , Aleksey D. Kachkovskii , Vladimir V. Kurdyukov
⇑
a Institute of Organic Chemistry, National Academy of Sciences, Murmanskaya Str. 5, 02660 Kiev, Ukraine1
b Ukrainian–American Laboratory of Computational Chemistry, Kharkov, Ukraine–Jackson, MS, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 July 2011
Received in revised form 2 October 2011
Accepted 3 October 2011
Available online 25 October 2011
A systematic investigation of the conformational structure was performed for the series of symmetri-
cal and unsymmetrical mono-, tri-, pentamethine cyanines, and styryl dyes bearing 2-azaazulenium
terminal group. The rotation energy barriers of terminal groups were determined via the dynamic
variable temperature NMR experiments. The conformational transformation energy was calculated
by quantum chemical methods (B3LYP and M05-2X) both for the cases of considering the solvent
influence and not tacking it into account. Based on the comparison of theoretical and experimental
data, relative electron-donating abilities and geometrical features of the heterocyclic terminal groups
in 2-azaazulenium dyes were estimated. The arrangement of certain heterocyclic nuclei in order of
basicity by considering the results of the dynamic NMR investigations was proposed. Influence of
the conjugated chain length and the solvent nature on the conformational lability of the investigated
dye molecules was discussed.
Keywords:
Azaazulenes
Cyanine dyes
Styryles
Conformational analysis
Dynamic NMR
DFT
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
interaction with biomolecules (DNA), and hence for the spectral
properties of the formed complexes [11]. Therefore, study of the
Considerable attention is paid to the investigation of cyanine
dyes due to their usefulness in different fields of science and indus-
try [1,2]. These compounds have found numerous applications
such as photographic sensitizers [3], nonlinear optical materials
[4], fluorescent probes for the biomolecular labelling [5]. They
are able to interact with biomolecules via the covalent and nonco-
valent coupling. Thus, suitable fluorophores can be bound to the
double-stranded DNA molecules, resulting in the significant
enhancement of the fluorescence properties of the latter [6]. To
study interactions and structure of the complexes, different phys-
icochemical methods of analysis including X-ray, NMR, and FRET
were applied [7]. Depending on the structure, terminal groups
and the length of the polymethine chain the dye can bind DNA in
several ways [8]. Sterical effects of the terminal groups are very
important for the J-aggregation of symmetrical cyanines with
DNA [9,10].
peculiarities of the structure and physicochemical properties of
the cyanines is important for the prediction and analysis of their
bimolecular labelling properties.
The dependence of the dye spectral properties on the electronic
nature of terminal groups was examined for an extensive series of
symmetrical and asymmetrical cyanines in the middle of the 20th
century by Brooker [12]. The concept of the absorption maximum
‘‘deviation’’ and the corresponding scale of basicity were intro-
duced. The dependence of the positions of the absorption maxima
in the UV spectra on the differences of the electron-donating prop-
erties (basicity) of terminal groups was reported in [13]. Quantum-
chemical analysis method of terminal heterocyclic groups’ basicity
was developed by Dyadusha and Kachkovskii [14]. On the basis of
these studies one can draw conclusions regarding the influence of
the asymmetry of the polymethines electronic structure on their
physical properties.
The molecular structure of the cyanine dye and the asymmetry
of its electronic structure (the differences in the electron-donating
properties of the terminal groups) are crucial for the process of
However, optical methods are not the only option for the esti-
mation of the electronic asymmetry of the cyanines. NMR spectros-
copy is found to be one of the most useful instruments for the
investigation of the molecular electronic structure and conforma-
tions of the cyanine dye molecules [15,16]. Influence of the sterical
hindrance from the terminal groups on the trimethine cyanines
conformations was investigated [17]. The dependence of the
conformational transformations energy barriers (cis-trans-isomeri-
zation) on the terminal groups and the chromophore chain length
was determined experimentally by means of dynamic NMR [18].
⇑
Corresponding author at: Institute of Organic Chemistry, National Academy of
Sciences, Murmanskaya Str. 5, 02660 Kiev, Ukraine. Tel.: +380 44 449 46 13; fax:
+380 44 573 26 43.
E-mail address: bobry@ukr.net (A.B. Ryabitskii).
http://www.ioch.kiev.ua.
1
0022-2860/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2011.10.003

A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
53
Detailed data on the variable-temperature 1H NMR conformational
analysis for the series of trimethine cyanines derivatives of benz-
oxazole, benzothiazole, etc. is presented in the literature [19]. This
method was also successfully applied to the study of the photoiso-
merization of trimethine cyanines [20].
We also investigated styryl/methoxystyryl dyes derivatives of
2,6-di(terbutyl)-pyrylium OPy-St(R):
O
According to the X-ray diffraction data and the quantum-chem-
ical analysis of the constitution of polymethine cyanines, in the
absence of the sterical hindrance dye chromophores adopt all-trans
conformation [18]. Quantum-chemical calculations within the
AM1 approximation were applied to determine the conformational
transformations energy barriers for the cyanines and other linear
conjugated systems [21].
R
R=N(CH3)2,OCH3
OPy-St(R)
Earlier we have synthesized the series of monomethine [22]
and trimethine [23] cyanines bearing 2-azaazulenium moiety
and studied it using spectral and quantum-chemical methods.
In our previous paper [24] we presented the investigation results
of the conformational features of the unsymmetrical monome-
thine cyanine dye 2-[(2-butyl-1,3-dimethylcyclohepta[c]pyrrol-
6(2H)-ylidene)methyl]-3-ethyl-1,3-benzothiazol-3-ium perchlo-
rate. It was studied both in solution by means of NMR spectros-
copy and in the solid state by X-ray diffraction. Due to the data
obtained, it was proved that the dye molecule is practically pla-
nar in crystalline state. We described the effect of the intramo-
lecular rotation of molecular fragments around the
2.2. Synthesis
The synthesis, UV/Vis spectral characteristics, 1H and 13C NMR
data of 2-azaazulene monomethine cyanines AAz-0-AAz, AAz-0-
BIn, AAz-0-BO, AAz-0-BT, AAz-0-4Q have been described in [22],
the synthesis and properties of the corresponding trimethine cya-
nines – AAz-1-AAz, AAz-1-BIn, AAz-1-In, AAz-1-BT, AAz-1-4Q,
AAz-1-2Q, AAz-1-BIm, AAz-1-NPy – in [23]. The corresponding
new monomethine AAz-0-OPy and trimethines AAz-1-OPy and
AAz-1-BO have been synthesized in a similar way. The symmetri-
cal (AAz-2-AAz) and asymmetrical (AAz-2-OPy) pentamethine
cyanines as well as styryl dyes (AAz-St(R) and OPy-St(R)) were
synthesized according to the general approaches described in
[25]. All compounds were purified by column chromatography
and/or recrystallized before use, and their structures were con-
firmed by NMR.
CheterocycleACmethine bond, which is slow on the NMR time scale.
Both the bond order of the cyanine dye, around which rotation
occurs, and the energy barrier of this process were assumed to
depend on the electron-donating nature of the terminal groups.
The results of the quantum-chemical calculations supported the
experimental data. In the work presented below we show that
this approach can be used to draw a certain scale of the elec-
tron-donating properties of different terminal groups on the ba-
sis of the rotation energy derived by the dynamic NMR
experiments. Thus, we studied the dependence of the conforma-
tional transformations energy barriers on the nature of the ter-
minal groups and on the chromophore chain length for the
polymethine dyes derivatives of the 2-azaazulenium.
2.2.1. 2-Butyl-6-[(2,6-di-tert-butyl-4H-pyran-4-ylidene)methyl]-1,
3-dimethylcyclohepta[c]pyrrolium tetrafluoroborate AAz-0-OPy
Triethylamine (0.1 mL) was added to the solution of 2-butyl-1,
3-dimethyl-6-(methylthio)cyclohepta[c]pyrrolium iodide (0.175 g,
0.5 mmol) and 2,6-di-tert-butyl-4-methylpyrylium tetrafluorobo-
rate (0.147 g, 0.5 mmol). The solution was refluxed for 5 min, and
was then allowed to cool to room temperature. After adding of
diethyl ether the dye was precipitated, filtered off, and purified by
chromatography on silica gel column eluting with a mixture chloro-
form: methanol (50:1). Yield 0.1 g (39.6%); mp 225 °C (decomposi-
2. Experimental
tion). kmax = 580 nm, e
= 63,400 Mꢀ1 cmꢀ1 (in acetonitrile). 1H NMR
2.1. General methods
(CDCl3, 400 MHz) d: 7.86 (d, J = 11.3 Hz, 2H, 4-H 8-H AAz), 7.05 (d,
J = 11.3 Hz, 2H, 5-H 7-H AAz), 6.86–6.81 (br. s, 2H, Ar OPy), 6.23 (s,
1H, CAAzACH), 4.17–4.06 (m, 2H, NCH2), 2.58 (s, 6H, AAzACH3),
1.82–1.54 (m, 2H, NCH2CH2), 1.48–1.45 (m, 2H, CH2CH3), 1.31 (s,
18H, t-Bu), 0.92 (t, J = 7.3 Hz, 3H, CH2CH3). Calcd. for C29H40BF4NO
(505.44): C 68.91, H 7.98, N 2.77%; found: C 68.73; H 7.71; N 2.85%.
All starting materials and solvents for the synthesis and spec-
troscopic measurements were supplied by Aldrich Chemical Co.
and were used as received without further purification. UV/Vis
absorption spectra were recorded on a Shimadzu UV-3100 spectro-
photometer. The melting points were measured with a melting
point apparatus (Kleinfeld GmbH) and are uncorrected.
We chose representatives of the following groups of 2-aza-
azulene derivatives as objects of our research: unsymmetrical
mono-, tri-, and pentamethine cyanines AAz-n-Het (Het is sec-
ond terminal group, see Table 1) and styryl (methoxystyryl) dyes
AAz-St.
2.2.2. 2-Butyl-6-[3-(2,6-di-tert-butyl-4H-pyran-4-ylidene)prop-1-en-
1-yl]-1,3-dimethylcyclohepta[c]pyrrolium tetrafluoroborate
AAz-1-OPy
Triethylamine (0.1 mL) was added to a solution of 6-(2-anilino-
vinyl)-2-butyl-1,3-dimethylcyclohepta[c]pyrrolium tetrafluorobo-
rate (0.160 g, 0.38 mmol) and 2,6-di-tert-butyl-4-methylpyrylium
N
N
Het
n
R
R=N(CH3)2,OCH3
n=0,1,2
AAz-n-Het
AAz-St(R)

54
A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
Table 1
General structures and abbreviations of the heterocycles used as the second terminal groups in dyes AAz-n-Het.
Het
Abbrev.
AAz
Het
Abbrev.
In
Het
Abbrev.
4Q
n-Bu
N
N
N
N
OPy
BIn
BO
BT
2Q
t-Bu
O
N
O
Et
t-Bu
NPy
S
Ph
N
N
Me(Et)
N
Ph
BIm
N
N
tetrafluoroborate (0.113 g, 0.5 mmol) in acetic anhydride (3 mL).
The solution was heated at 60 °C for 5 min, and was then allowed
to cool to room temperature. The reaction mixture was diluted
with diethyl ether (10 mL). The precipitate was filtered off, washed
with diethyl ether, dried, and purified by recrystallization from
acetonitrile. Yield 0.145 g (72%), mp 219–221 °C. kmax = 705 nm,
zol-3-ium iodide (0.119 g, 0.41 mmol) in acetic anhydride (2 mL).
The solution was heated at 60 °C for 5 min, and was then allowed
to cool to room temperature. The precipitate was filtered off,
washed with diethyl ether, dried, and purified by recrystallization
from acetonitrile. Yield 0.120 g (55.6%), mp 219–221 °C.
kmax = 645 nm (in acetonitrile); 1H NMR (CD3CN, 400 MHz) d:
8.38 (deg. t, J = 13.4 Hz, 1H, CAAzACH@CH), 7.67–7.62 (m, 1H, Ar
BO), 7.61–7.55 (m, 1H, Ar BO), 7.55–7.47 (m, 2H, Ar BO), 7.18 (d,
J = 11.5 Hz, 1H, Ar AAz), 7.08 (d, J = 11.5 Hz, 1H, 5-H AAz), 6.98
(d, J = 11.5 Hz, 1H, Ar AAz), 6.34–6.25 (m, 3H, 7-H AAz
e
= 63,400 Mꢀ1 cmꢀ1 (in acetonitrile). 1H NMR (CDCl3, 400 MHz)
d: 8.41 (deg. t, J = 13.2 Hz, 1H,
CAAzACH@CH), 7.61 (br. d,
J ꢁ 11.2 Hz, 2H, 4-H 8-H AAz), 7.19 (br. s, 1H, Ar OPy), 6.60–7.20
(very br. signal, 1H, 5-H 7-H AAz), 6.36 (br. s, 1H, Ar OPy), 6.33
(d, J = 13.5 Hz, 1H, CAAzACH), 6.10 (d, J = 12.9 Hz, 1H, COPyACH),
4.10–3.97 (m, 2H, NCH2), 2.49 (s, 6H, AAzACH3), 1.25–1.48 (m,
2H, CH2CH3 t-Bu), 0.96 (t, J = 7.3 Hz, 3H, CH2CH3). Calcd. for
CAAzACH@CHACH), 4.26 (q, J = 7.3 Hz, 2H, NCH2CH3), 3.97–3.88
(m, 2H, NCH2CH2), 2.38 (br. s, 3H, AAzACH3), 2.35 (br. s, 3H,
AAzACH3), 1.67–1.57 (m, 2H, NCH2CH2), 1.44 (t, J = 7.3 Hz, 3H,
NCH2CH3), 1.44–1.33 (m, 2H, CH2CH3), 0.96 (t, J = 7.3 Hz, 3H,
CH2CH3). Calcd. for C27H31IN2O (531.48): C 61.60, H 5.94, I
24.11%; found: C 61.45, H 5.72, I 24.38%.
C31H42BF4NO (531.48): C 70.06, H 7.97, N 2.64%; found: C 69.85;
H 7.82; N 2.78%.
2.2.3. 2-Butyl-6-[(1E,3E)-5-(2,6-di-tert-butyl-4H-pyran-4-
ylidene)penta-1,3-dien-1-yl]-1,3-dimethylcyclohepta[c]pyrrolium
tetrafluoroborate AAz-2-OPy
2.2.5. 2-Butyl-6-[5-(2-butyl-1,3-dimethylcyclohepta[c]pyrrol-6(2H)-
ylidene)penta-1,3-dien-1-yl]-1,3-dimethylcyclohepta[c]pyrrolium
tetrafluoroborate AAz-2-AAz
A
mixture of 4-(2,6-di-tert-butyl-4H-pyran-4-ylidene)but-2-
enal (0.052 g, 0.2 mmol) and 2-butyl-1,3,6-trimethyl-cyclohep-
ta[c]pyrrolium tetrafluoroborate (0.064 g, 0.2 mmol) and acetic
anhydride (1.5 mL) was heated for 3 min at 100 °C. After cooling
to room temperature, the reaction mixture was diluted with
diethyl ether (10 mL). The precipitate was filtered off, washed with
diethyl ether, dried, and purified by recrystallization from acetoni-
trile. Yield 0.05 g (45%), mp 180–182 °C; kmax = 809 nm (in acetoni-
trile); 1H NMR (CDCl3, 400 MHz) d: 8.24 (deg. t, J = 13.0 Hz, 1H,
A solution of 2-butyl-1,3,6-trimethyl-cyclohepta[c]pyrrolium
tetrafluoroborate (0.315 g, 1 mmol), N-(3-anilinoprop-2-en-1-yli-
dene)benzenaminium chloride (0.130 g, 0.5 mmol), and anhydrous
sodium acetate (0.1 g) in acetic anhydride (7 mL) was heated under
stirring at 40 °C for 3 h. After cooling to room temperature the pre-
cipitated dye was filtered off and washed with acetic anhydride
(1 mL) and dry diethyl ether (10 mL). For purification the solution
of dye in methylene chloride (20 mL) was filtered, and filtrate was
evaporated in vacuum. Yield 0.140 g (24%), mp 175–177° (decom-
CAAzACH@CH), 8.17 (deg. t, J = 13.0 Hz, 1H, COPyACH@CH), 7.50
(d, J = 11.4 Hz, 2H, 4-H 8-H AAz), 6.99 (very br. d, 2H, 5-H 5-H
AAz), 6.49 (deg. t, J = 12.5 Hz, 1H, CAAzACH@CHACH), 6.31 (d,
J = 13.7 Hz, 1H, COPyACH), 6.03 (br. d, J = 13.5 Hz, 1H, CAAzACH),
4.07–3.96 (m, 2H, NCH2), 2.50 (s, 6H, AAzACH3), 1.72–1.60 (m,
2H, NCH2CH2), 1.47–1.35 (m, 2H, CH2CH3), 1.43–1.25 (br. s, 18H,
t-Bu), 0.99 (t, J = 7.3 Hz, 3H, CH2CH3). Calcd. for C33H44BF4NO
(557.51): C 71.09, H 7.95, N 2.51%; found: C 69.80, H 7.89, N 2.62%.
position); kmax = 924 nm,
NMR (acetone-d6 + CD3CN, 400 MHz) d: 7.68 (t, J = 12.9 Hz, 2H,
AAzACH@CH), 7.39 (d, J = 11.5 Hz, 4H, 4-H 8-H AAz), 6.77 (d,
J = 11.5 Hz, 4H, 5-H 7-H AAz), 6.58 (t, J = 12.2 Hz, 1H,
AAzACH@CHACH), 6.40 (d, J = 13.6 Hz, 2H, CAAzACH), 4.11–3.88
e
= 212,000 Mꢀ1 cmꢀ1 (in acetonitrile). 1H
C
C
(m, 4H, NCH2), 2.46 (s, 12H, AAzACH3), 1.59–1.71 (m, 4H,
NCH2CH2), 1.33–1.45 (m, 4H, CH2CH3), 0.95 (t, J = 7.1 Hz, 6H,
CH2CH3). Calcd. for C35H43BF4N2 (578.53): C 72.66, H 4.79, N
4.84%; found: C 72.52, H 5.02, N 4.99%.
2.2.4. 2-Butyl-6-[(1E,3E)-3-(3-ethyl-1,3-benzoxazol-2(3H)-
ylidene)prop-1-en-1-yl]-1,3-dimethylcyclohepta[c]pyrrolium iodide
AAz-1-BO
2.2.6. Preparation of AAz styryl dyes (general procedure)
Triethylamine (0.05 mL) was added to a solution of 6-(2-an
ilinovinyl)-2-butyl-1,3-dimethylcyclohepta[c]pyrrolium tetrafluo-
roborate (0.173 g, 0.41 mmol) and 3-ethyl-2-methyl-1,3-benzoxa
A solution of 1 mmol of 2-butyl-1,3,6-trimethyl-cyclohep-
ta[c]pyrrolium tetrafluoroborate and 1.5 mmol of p-dimethylami-
nobenzaldehyde or p-anisaldehyde in dimethylsulfoxide (2 mL)

A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
55
was heated at 50 °C for 15–20 min. The solvent was evaporated in
vacuum. The residue was triturated with diethyl ether (5 ꢂ 5 mL),
and dissolved in ethanol (2 mL). After cooling to room temperature
the precipitated dye was filtered off and crystallized from ethanol.
(no imaginary vibration for the ground state and one imaginary
vibration for the transition state) and zero point corrected ener-
gies. The rotation energy barriers around methine bonds were
defined as the difference between the energies of the transition
and ground states (1):
2.2.7. 2-Butyl-6-{2-[4-(dimethylamino)phenyl]vinyl}-1,3-
D
E ¼ Et:s: ꢀ Eg:s:
ð1Þ
dimethylcyclohepta[c]pyrrolium tetrafluoroborate AAz-St(NMe2)
e
= 48,200 Mꢀ1 cmꢀ1 (in
Wiberg bond indexes [30] were determined using the Natural
Yield 32%, mp >250 °C; kmax = 680 nm,
acetonitrile). 1H NMR (CDCl3, 400 MHz) d: 8.17 (d, J = 11.0 Hz, 2H,
4-H 8-H AAz), 7.80 (d, J = 15.3 Hz, 1H, CStACH), 7.62 (d, J = 8.6 Hz,
2H, 3-H 5-H St), 7.44 (d, J = 11.0 Hz, 2H, 5-H 7-H AAz), 6.95 (d,
J = 15.3 Hz, 1H, CAAzACH), 6.70 (d, J = 8.6 Hz, 2H, 2-H 6-H St),
4.20–4.11 (m, 2H, NCH2), 3.13 (s, 6H, N(CH3)2), 2.65 (s, 6H,
AAzACH3), 1.70–1.57 (m, 2H, NCH2CH2), 1.47–1.36 (m, 2H,
CH2CH3), 0.97 (t, J = 7.2 Hz, 3H, CH2CH3). Calcd. for C25H31BF4N2
(446.33): C 67.27, H 7.00, N 6.28%; found: C 67.07, H 6.88, N 6.35%.
Bond Orbital analysis [31]. Quantum-chemical calculations were
performed for the isolated cation. In some cases the influence of
the solvent was also taken into account. Geometry optimization
and calculation of the energy and vibration frequencies for the
ground and transition states were performed under the following
algorithms. We used parameters obtained for vacuum as the start-
ing geometry. The geometry optimization and calculation of the
vibrational frequencies were carried out using the Onsager model
[32]. Then we performed the geometry optimization by means of
the CPCM polarizable conductor calculation model [33]. The total
energy was defined as the sum of the electronic (PCM) and zero-
point (Onsager) energies. In our calculations the counter ions were
neglected due to the assumption that in the diluted solutions the
counter-ion has only weak influence on the cation conformation.
We assumed also that aggregation of the molecules does not take
place.
2.2.8. 2-Butyl-6-[2-(4-methoxyphenyl)vinyl]-1,3-
dimethylcyclohepta[c]pyrrolium tetrafluoroborate AAz-St(OMe)
Yield 29%, mp 142–143 °C; kmax = 498 nm,
e
= 23,900 Mꢀ1 cmꢀ1
(in acetonitrile). 1H NMR (CDCl3, 400 MHz) d: 8.42 (d, J = 10.8 Hz,
2H, 4-H 8-H AAz), 7.66 (d, J = 16.0 Hz, 1H, CStACH), 7.57 (d,
J = 8.6 Hz, 2H, 3-H 5-H St), 7.56 (d, J = 10.8 Hz, 2H, 5-H 7-H AAz),
6.99 (d, J = 16.0 Hz, 1H, CAAzACH), 6.90 (d, J = 10.8 Hz, 2H, 2-H 6-
H St), 4.27–4.18 (m, 2H, NCH2), 3.87 (s, 3H, OCH3), 2.70 (s, 6H,
AAzACH3), 1.75–7.65 (m, 2H, NCH2CH2), 1.50–1.35 (m, 2H,
CH2CH3), 0.98 (t, J = 7.3 Hz, 3H, CH2CH3). Calcd. for C24H28BF4NO
(433.29): C 66.53, H 6.51, N 3.23%; found: C 66.32, H 6.34, N 3.45%.
2.4. NMR experiments
All NMR experiments were carried out on Varian GEMINI 2000
spectrometer with 1H and 13C frequencies of 400.07 and
100.61 MHz, respectively. Tetramethylsilane was used as a stan-
dard for the d scale calibration. We used DMSO-d6, CD3CN, CDCl3,
acetone-d6, CD2Cl2 and some mixtures (see below) as solvents.
2.2.9. Preparation of OPy styryl dyes (general procedure)
A solution of 2,6-di-tert-butyl-4-methylpyrylium perchlorate
(1 mmol) and p-dimethylamino-benzaldehyde or p-anisaldehyde
(1.1 mmol) in acetanhydride (2 mL) was heated for 1 h at 130 °C.
After cooling, the reaction mixture was triturated with diethyl
ether (50 mL) and the obtained solid was filtered and crystallized
from the mixture of ethanol–diethyl ether (1:2).
Concentrations of the solutions constituted from 5 to 20 mmol Lꢀ1
.
One of the main problems of the NMR experiments was to achieve
the concentrations necessary for the accumulation of the accept-
able spectra. Therefore, choosing a solvent we considered first of
all the solubility of the substances. The crystallization and boiling
points of the solvents were also important because in certain cases
it was necessary to cool or heat the solution intensively in order to
achieve the definite values of reaction rate constants it was neces-
sary to cool or to heat the solution strongly.
The temperature of the experiments was determined accurate
to 0.5 K. The errors of the parameters calculation could be attrib-
uted to the temperature determination accuracy upon the thermo-
stating and to the precision of the line shape analysis.
2.2.10. 2,6-di-tert-Butyl-4-{-2-[4-
(dimethylamino)phenyl]vinyl}pyrylium perchlorate, OPy-St(NMe2)
Yield 71%, mp 192–194 °C; kmax = 577 nm,
e
= 78,400 Mꢀ1 cmꢀ1
(in acetonitrile). 1H NMR (CDCl3, 400 MHz) d: 8.26 (d, J = 15.3 Hz,
2H, CStACH), 7.89 (d, J = 9.0 Hz, 2H, 3-H 5-H St), 7.44 (s, 2H, Ar
OPy), 7.13 (d, J = 15.3 Hz, 1H, COPyACH), 6.72 (d, J = 9.0 Hz, 2-H 6-
H St), 3.15 (s, 6H, N(CH3)2), 1.45 (s, 18H, t-Bu). Calcd. for
C23H32ClNO5 (437.96): C 63.08, H 7.36, Cl 8.10%; found: C 62.84,
H 7.21, Cl 8.24%.
2.5. Dynamic NMR calculations
2.2.11. 2,6-di-tert-Butyl-4-[2-(4-methoxyphenyl)vinyl]pyrylium
perchlorate OPy-St(OMe)
The rate constants were determined by the visual comparison of
the theoretical and experimental spectra. It was realized by means
of WINDNMR 7.1 program [34], which afforded the simulation of
the theoretical spectra using the modified Bloch equation. Upon
the variation of the spectral parameters such as chemical shifts,
coupling constants, line widths and rate constants it was possible
to get the exact fitting of the theoretical and experimental spectra.
Thus, it was possible to determine the rate constant of the isomer-
ization process from the experimental spectra recorded at different
temperatures based on the signal widths of protons participating
in the exchange processes (for details see [34,35]). We chose
‘‘dddd’’ (for aromatic AAz protons), ‘‘AB Coupl’’ (for aromatic OPy
protons) or ‘‘2-spin’’ (for CH3 and tert-Bu groups) with the two
equal population (%a = %b = 50) as simulation patterns in WIND-
NMR program, as at the first approximation these spectra can be
regarded as a first-order system. Thermodynamic parameters of
the isomerization process were determined by means of Eyring
eq. (2) [35]:
Yield 56%, mp 261–263 °C; kmax = 446 nm,
e
= 37,800 Mꢀ1 cmꢀ1
(in acetonitrile). 1H NMR (CDCl3, 400 MHz) d: 8.34 (d, J = 15.8 Hz,
1H, CStACH), 7.89 (d, J = 8.8 Hz, 3-H 5-H St), 7.80 (s, 2H, Ar OPy),
7.73 (d, J = 15.8 Hz, 1H, COPyACH), 6.90 (d, J = 8.8 Hz, 2-H 6-H St),
3.85 (s, 3H, OCH3), 1.50 (s, 18H, t-Bu). Calcd. for C22H29ClO6
(424.92): C 62.19, H 6.88, Cl 8.34%; found: C 62.26, H 6.65, Cl 8.52%.
2.3. DFT calculation
The geometries of the ground and transition states were opti-
mized using the GAUSSIAN03 program set [26] within the DFT
approximation. A functional M05-2X [27] was chosen as the
most recommended for the determination of the kinetic and
thermodynamic parameters as well as B3LYP [28] method in
combination with 6-311G(d,p) basis sets [29] for the geometry
optimization. For the ground and transition states the vibration
analysis was \performed which yielded vibration frequencies

56
A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
ꢀ
ꢁ
GT–
RT
butyl)pyrylium groups. We used Wiberg bond indexes (WBI) from
kBT
h
ꢀ
D
kT ¼
exp
ð2Þ
the NBO theory as analogs of the bond orders. Traditionally, upon a
study of the linear conjugated systems, the values of bond lengths
in the conjugated chain are analyzed [36]. Fig. 2 demonstrates the
values of the bond lengths, WBI, and rotation energy values for
each bond of the PC for AAz-2-OPy:
where kT is the rate constant, h is the Plank constant, kB is the Boltz-
mann constant, and R is the gas constant. Enthalpy
H– and entro-
S– were calculated from the eq. (3):
D
py of activation
D
The alternation of the bond lengths and bond orders is observed
along the PC. The length of the bond 1 (CAAzACmethine) is maximal
and its order is minimal due to the considerable contribution of
the boundary structure 1.b. For the bonds with the higher popula-
tion the larger values of the rotation energies are observed.
As follows from the analysis of 1H NMR spectra of monomethine
cyanines AAz-0-Het, in the case of Het = 4Q, BT, BO, BIm (relatively
strong electron donors – highly basic terminal groups) the signals
of aromatic protons of AAz fragment appear as four broad non-
equivalent doublets. For example, for AAz-0-4Q we observe them
at 7.03 ppm (H8) and 6.86 ppm (H4), 6.90 ppm (H7) and
6.35 ppm (H5). In the case of relatively weakly basic terminal
groups Het two doublets of the corresponding two pairs of AAz
protons are observed: at 7.61 and 7.04 ppm for the symmetrical
cyanine AAz-0-AAz [22] and at 7.86 and 7.05 ppm for AAz-0-OPy.
NOESY experiments on the dye AAz-0-BO revealed that
‘‘through space correlation’’ between the methine proton
(6.42 ppm) and AAzAH5 (7.13 ppm) takes place, and not between
the methine proton and AAzAH7 (7.69 ppm). This phenomenon
can be explained by the hindered intramolecular rotation for some
cyanines. The activation energy of this process depends mainly on
the electron-donating abilities of the terminal groups and on their
spatial properties, and can be estimated qualitatively from the sim-
ple 1H NMR spectra. More accurate information can be obtained via
the variable temperature NMR experiments.
It should be mentioned that in our case the dynamic NMR
investigations are possible due to the fact that AAz fragment has
the C2v symmetry and thus the both corresponding conformers
are equivalent. In a similar way, it is possible to determine the
rotation energy barriers for other terminal groups with the same
symmetry, e.g. for OPy and BIm (bond 2 in structure 1.a), and with
the proper values of the activation energies. In the case of the ter-
minal groups with low symmetry (Cs) rotation conformers have
different energies. Therefore, most probably their relative popula-
tions will be considerably different. For such compounds the pop-
ulation of the energetically less favorable conformer is so poor that
usually it cannot be detected by 1H NMR.
k
T
D
RT
H
D
R
S
ln ¼ ꢀ
þ
þ 23:76
ð3Þ
Thermodynamic parameters were determined by the least-
squares method adjusted from the linear relationship between
the ln(k/T) and the inverse value of temperature 1/(RT). The abso-
lute errors of the parameter values were determined by the least-
squares method. Free energy of activation
D
G–T was calculated as
D
GT–
¼
D
H– ꢀ T
D
S–.
3. Results and discussion
3.1. Molecular geometry and the analysis of NMR spectra
It is possible to present the electronic structure of investigated
dyes simplistically from a position of the mesomerism theory as a
hybrid of two canonical structures (rotations around of the C-C
bonds 1 and 2n + 2 are shown by arrows), n – number of vinylene
groups:
H8
N
N
bond 2n+2
H7
Het
Het
n
n
H4
H5
bond 1
1.b
1.a
The contribution of each boundary forms to the molecular struc-
ture of a dye depends on the differences in the electron-donating
properties of both terminal groups AAz and Het. The bond popula-
tions (bond orders) of the polymethine chain (PC), the rotation
energy and the bond lengths change upon the variation of an elec-
tron-donating ability of the Het groups. For the given family of dyes
energy of the rotation around methine bonds equals to the value of
the energy barrier, which corresponds to the difference of energies
between the transition state (perpendicular conformation, Fig. 1b)
and the ground state (planar conformation, Fig. 1a).
Upon the rotation of the terminal group for 180° the degener-
ated isomer is formed. Thus, such a rotation process can be referred
to as a pseudo-isomerization. The bond population and the
rotation energy is a function of the electron-donating ability of
the terminal Het group.
To prove the possibility to detect the dynamic processes for the
other then AAz groups with C2v symmetry group we synthesized
and investigated three vinylogs AAz-n-OPy (n = 0,1,2), bearing
2,6-di(tert-butyl) residue. Indeed, at room temperature the signals
of aromatic protons of the OPy moiety were significantly broad-
ened; they appeared at ꢁ6.83 ppm for AAz-0-OPy (in CDCl3 at
room temperature). At the same time, the signals of AAz protons
were observed as two doublets (vide supra).
Cooling of the sample to low temperature resulted in the
appearance of four separate doublets of AAz protons (7.89, 7.66,
As an example, let us consider the constitution of pentamethine
cyanine AAz-2-OPy, bearing 2-azaazulene and 2,6-di(tert-
Fig. 1. The spatial structure of two conformations of the dye AAz-1-OPy according to M05-2X/6-311G(d,p). The ground state (planar conformation, (a)) and the transition
state (perpendicular conformation, (b)) are indicated.

A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
57
concerns to the energy of the ground state. It results in increase
of the calculated value of an energy barrier. The rotation energy
values for monomethine AAz-0-OPy obtained by methods M05-
2X and B3LYP and 6-311G(d,p) basis set are equal correspond-
ingly 55.2ꢀ1and 61.8 kJ molꢀ1
; for dye AAz-0-BT – 69.7 and
[24]; for trimethine AAz-1-BT: 74.0 and
71.7 kJ mol
82.3 kJ molꢀ1
.
For some cyanines and styryl dyes AAz-St the calculations of
rotation energy were performed taking into account the solvent
influence (Table 2). It was possible to get the required structural
regularities by the calculations of the most interesting cases which
have been mentioned in the Table 2. The calculations have been
performed not for all compounds because such procedures usually
are labor-intensive and time-consuming process. Geometry opti-
mization of the ground and transition states when taking into
account the solvent influence, predicts some decreasing of the full
molecule energy due to the solvation effect, where for the transi-
tion state such stabilizing is more pronounced then for the ground
state. It means that upon solvation, the decreasing of the rotation
energy takes place. For example, for symmetrical monomethine
cyanine AAz-0-AAz, stabilizing energies for the ground and transi-
tion states are equal 139.9 and 161.4 kJ molꢀ1 correspondingly. It
resulted in decreasing of the value of rotation energy on
21.5 kJ molꢀ1
For some dyes we performed the measurements in different sol-
vents using polar DMSO (dielectric constant = 46.7) and acetoni-
trile = 36.6) as well non-polar chloroform = 4.9) and
methylene chloride ( = 8.9) in order to fix the possible differences
.
e
(e
(e
Fig. 2. The bond lengths (a), Wiberg bond indexes (b), and the rotation energies (c)
for the PC bonds of pentamethine cyanine AAz-2-OPy (n = 2) according to M05-2X/
6-311G(d,p). ‘‘Bond’’ – means the number of bond in PC, see 1.a.
e
that could appear for media of different polarity. But minor differ-
ences are observed between the corresponding values in different
solvents (D
G– is not more than 1–2 kJ molꢀ1, see Table 2.). Thus
we can make a conclusion regarding a relatively slight influence
of medium polarity on the conformational behavior of dye mole-
cules. The calculations of rotation energy with regard to the solvent
7.25 and 7.71 ppm at ꢀ74 °C) and two signals of OPy protons (7.17
and 6.46 ppm with the meta-spin–spin coupling constants
4JHH < 1 Hz at ꢀ74 °C) in the spectra. Also we observed two sepa-
rate signals of two methyl groups of the AAz fragment (at 2.61
and 2.56 ppm) and two singlets of tert-Bu substituents (at 1.32
and 1.28 ppm). Thus, here we detected two parallel dynamic pro-
cesses for a single molecule. For tri-, and pentamethine cyanines
AAz-n-OPy, and for the styryl dyes we observed the same
phenomenon.
polarity influence predict larger range of difference in values (
DE:
7–10 kJ molꢀ1) than it follows from experiment.
It is possible to assume the rotation barrier values to be depen-
dent on the solution concentration since dye molecules could form
aggregates. But we did not perform the special experiments aiming
the investigation of the concentration dependence of the rotation
barriers due to the insufficient solubility of azaazulenium dyes.
At the same time, the experiments in the different solvents proved
3.2. Calculation model influence and solvent effects
the low solvent dependence of D
G– values. It allows us to assume
the corresponding low dependence of
factor.
D
G– on the concentration
The thermodynamic parameters of the CHetACmethine bond,
which were determined both theoretically and experimentally,
are presented in Table 2.
3.3. Monomethine cyanines
To search the regularities for correct analysis of received data it
would be reasonable to compare the calculated values of rotation
The symmetrical monomethine cyanine AAz-0-AAz has the
lowest value of the rotation energy barrier. Using a mixture of
acetone-d6 and CDCl3 (approx. 5:1 v/v) as a solvent (because the
solubility of this dye in the pure acetone is too low) it was still
impossible to attain the complete resolution of the signals of the
corresponding pairs of AAz moiety protons even at ꢀ70 °C. The
same situation took place in the case of the unsymmetrical dye
AAz-0-BIn. Here the measurements were complicated due to the
limited dye solubility at low temperatures. For the monomethine
cyanine containing less basic terminal BO group four non-equiva-
lent doublets of AAz protons are observed (7.68, 7.58, 7.50, and
6.74 ppm), which are slightly broadened due to the dynamic pro-
cess at room temperature in DMSO-d6. Only upon heating of the
sample to 100 °C we detected the coalescence of the aromatic pro-
tons. In this same as in many other cases of AAz cyanine dyes the
difference of the chemical shifts for the H5 and H7 proton pair can
exceed 300 Hz at 400 MHz 1H frequency (in the case of AAz-0-BO it
constituted 373 Hz in DMSO-d6). Evidently, this fact could be
energy
calculated magnitudes
parameters
G– determined for the same dyes are different. The
D
E and experimental values of free energy
D
G– [37]. But
D
E and the corresponding experimental
D
values of energy calculated for system in vacuum practically in
each case predict the greater values of energy, than experimental
values for a solution. Such difference is more pronounced for
weakly basic terminal groups, then for highly basic one, and runs
up to >30 kJ molꢀ1. This fact could be explained by that in calcula-
tion the influence of solvent is not considered. Some influence of
anion on the experimental result is also possible.
In order to try to explain this discrepancy and also to prove
the correctness of our assumptions regarding the isomerization
process, geometries of some dyes have been optimized by means
of different quantum-chemical methods (we used functionals
M05-2X and B3LYP) taking into account the influence of the
medium. Functional B3LYP regularly underestimate the total en-
ergy of a molecule in comparison with M05-2X, especially it

58
A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
Table 2
Calculated values of the rotation energy around the CHetACmethine bond (
D
E) (M05-2X/6-311G(d,p)), calculated energy values with the consideration of the solvent influence
(M05-2X/6-311G(d,p) CPCM (
DE (CPCM)), and experimental thermodynamic activation parameters.
Dye (Het)a
Solvent
D
E
D
E (CPCM)
D
H–
D
S–
D
G–298
AAz-0-AAz
AAz-1-AAz
AAz-2-AAz
AAz-0-OPy
AAz-0-OPy (OPy)
AAz-1-OPy
AAz-1-OPy (OPy)
AAz-1-OPy
AAz-1-OPy (OPy)
AAz-2-OPy
AAz-2-OPy (OPy)
AAz-2-OPy
AAz-2-OPy (OPy)
AAz-0-BIn
AAz-1-BIn
AAz-1-In
AAz-0-BO
AAz-0-BO
AAz-1-BO
AAz-0-BT
AAz-0-BT
AAz-1-BT
AAz-1-BT
AAz-1-BT
AAz-0-4Q
AAz-1-4Q
AAz-1-2Q
AAz-1-BIm
AAz-1-NPy
AAz-St(NMe2)
AAz-St(NMe2) (St)
AAz-St(OMe)
AAz-St(OMe) (St)
Py-St(NMe2) (OPy)
Py-St(NMe2) (St)
Py-St(OMe) (OPy)
Py-St(OMe) (St)
Acetone-d6 + CDCl3
Acetone-d6 + CDCl3
Acetone-d6 + CD3CN
CDCl3
CDCl3
CDCl3
57.4
77.5
78.5
55.2
64.2
69.6
85.1
69.6
85.1
75.3
96.5
75.3
96.5
48.5
71.1
70.0
77.6
77.6
77.7
69.7
69.7
74.0
74.0
74.0
63.0
78.3
82.1
97.6
100.1
50.4
52.3
38.8
36.8
65.3
62.7
52.6
44.2
35.8
55.3
36.6
42.6
51.9
57.7
68.4
50.3
58.5
53.6
62.2
38.4
44.6
–
–
–
–
–
–
–
–
–
–
–
59.0
65.9
–
–
–
40.9
–
31.1
30.5b
ꢀ26.3
38.3
49.8
44.6
42.9
52.4
49.1
59.8
51.8
57.9
50.1
57.2
47.7
52.4
39.8
51.4
57.1
73.0
72.4
65.6
63.7
64.9
64.2
65.0
63.6
63.3
72.6
73.9
–
52.2 0.9
43.6 0.8
44.4 2.5
50.5 1.3
51.3 0.5
67.8 1.9
50.1 0.6
53.6 0.8
43.0 0.8
56.7 0.8
42.5 0.8
47.4 0.8
36.4
63.6 0.9
60.5 0.6
84.2 3.2
79.1 3.4
62.9 0.6
61.1 0.8
59.1 0.5
79.3 1.1
72.8 3.4
84.4
8.0 3.4
ꢀ3.4 0.1
5.0 0.3
ꢀ6.4 0.1
7.3 2.0
26.9 6.5
ꢀ5.8 2.1
ꢀ14.5 2.7
ꢀ23.8 2.8
ꢀ1.7 1.1
ꢀ17.5 2.8
ꢀ16.7 2.8
ꢀ11.5
CDCl3
CD3CN
CD3CN
CDCl3
CDCl3
CD3CN
CD3CN
CDCl3
CD3CN
CD3CN
DMSO
40.7 3.4
11.4 2.0
37.5 9.6
22.3 10.6
ꢀ9.1 2.0
ꢀ8.9 2.5
ꢀ19.5 1.6
50.8 3.5
26.1 10.1
69.8
25.7 4.4
52.5 8.2
2.9 3.0
–
–
–
CDCl3
CD3CN
CD3CN
DMSO
CD3CN
DMSO
CD2Cl2
CDCl3
DMSO
DMSO
DMSO
DMSO
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
71.0 1.3
90.1 2.9
74.7 1.0
110 °Cc
78 °C
–
–
–
–
ꢀ70 °C
ꢀ70 °C
–
–
–
ꢀ63 °C
ꢀ63 °C
–
–
–
–
–
43.8 0.5
48.2 0.8
36.5 0.3
ꢀ92 °C
ꢀ14.5 2.3
8.9 5.7
0.3 1.7
–
48.1
45.6
36.4
–
a
On default, one means the rotation energy around bond 1 (formulae 1.a) Het = AAz. In parentheses we placed terminal groups different from AAz for which the values of
rotation energy around the bond 2 (1.a) was determined.
b
By italics the values determined with lowered accuracy are shown.
For compounds, for which it was not possible to determine the thermodynamic parameters owing to very great (or small) energy values, the temperature, to which the
c
solution was heated up (or the solution was cooled) are specified.
explained by the magnetic shielding of the H7 proton by the
conjugated system. The signals of the AAz H7 proton are usually
shifted upfield (in the case of AAz-0-BO it appears at 6.76 ppm).
It results in the observed wide temperature range (>100°) where
the dynamic process can be detected by NMR, allowing accurate
line shape analysis. It is worth mentioning that the 2-azaazuleni-
um unit has one distinctive feature: due to its electron-donating
properties and symmetry it yields such values of the rotation
energy barriers that they are the most suitable for running DNMR
investigations. Fig. 3 demonstrates the experimental and calcu-
lated values of the rotation energy barriers for the series of
monomethine cyanines.
3.4. Trimethine cyanines
Fig. 4 demonstrates graphically the values of Wiberg indexes for
theCAAzACmethine bond (bond1 – 1.a) in molecules of trimethinecya-
nines, rotation around which was investigated by the dynamic NMR.
For trimethine AAz-1-Het, where Het = OPy, In, Bin, as well as
for the symmetrical trimethine cyanine (Het = AAz), 1H NMR sig-
nals of the AAz H4 and H8 protons appear as broad doublets (for
Het = OPy at 7.60 ppm in CDCl3 at 20 °C) and signals of the H5
and H7 appear as considerably broadened peaks (for Het = OPy at
7.08 ppm in CDCl3) due to the fast rotation of the AAz fragment
at room temperature. To attain the conditions of slow exchange
we cooled the solution nearly to the crystallization point. Indeed,
In the series of monomethine dyes AAz-0-Het changes from
OPy to 4Q lead to the increase of the CAAzACmethine bond order
values, indicating the increase of the boundary structure 1.a
contribution.
According to the quantum-chemical calculations (M05-2X/6-
311G(d,p) as well as B3LYP [22]), monomethine cyanine cations
are nonplanar due to the considerable steric hindrance between
the heterocyclic terminal groups, with the exception of AAz-0-
BO, which is practically planar due to the small size of the oxazole
cycle. Therefore, aiming to reduce the influence of the steric factors
on the result of the analysis to get a clear insight in the effects of
the electron-donating properties of the terminal groups we have
to consider trimethine dye molecules.
Fig. 3. Values of the rotation energy around CAAzACmethine bond of the monome-
thine AAz-0-Het dyes according to the quantum chemical calculations M05-2X/6-
311G(d,p) (DFT),
DE, and experimental measurements (DNMR), D .
G–298

A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
59
above 60 °C the intramolecular rotation rate increases to such an
extend that the entire corresponding proton pairs become equiv-
alent. Besides that, the shift of peaks positions was detected
upon the temperature variation. The same pattern was observed
for all the other dyes.
Temperature variations did not induce any irregular changes in
the spectra e.g. appearance of additional signals or shifts of the sig-
nals, which were not connected with the temperature influence.
From our point of view, it indicates the absence of the dye mole-
cules aggregation. The similarity of the spectra, which were
recorded in the solvents with different polarities, also proves this
conclusion.
Fig. 4. Wiberg bond indexes for the CAAzACmethine bond (bond 1 in structure 1.a) for
the trimethine cyanines AAz-1-Het.
upon cooling two doublets divided into four separate doublets (for
Het = OPy at 7.70, 7.54, 7.44, 6.52 ppm in CDCl3 at ꢀ69 °C). In the
case of more basic terminal groups (BT, BO, 4Q, 2Q) we observed
the signals of the AAz protons as four non-broadened doublets
even at room temperature. Upon heating the broadening of dou-
blets followed by the coalescence was detected. For the most basic
terminal groups the activation energy is so high that we did not
manage to detect any appreciable broadening of the signals even
at temperatures above 100 °C and thus we could not measure the
rotation barriers experimentally. Such increase of the activation
energy is confirmed by the quantum-chemical calculations, which
predict its value in the range of 97–100 kJ molꢀ1 for NPy and BIm
(Table 2).
The dyes AAz-n-OPy are among the few cyanines for which
both dynamic processes were detected. Fig. 5 shows the series of
the variable temperature 1H NMR spectra recorded in CD3CN in
the temperature range from the crystallization to the boiling point
of the solvent.
At temperatures below ꢀ40 °C the intramolecular rotation
around both CAAzACmethine and COPyACmethine bonds became slow
on the NMR time scale, and we observed the signals of all
groups of protons, marked as H4AH12, separately. At higher
temperatures we detected the coalescence of the corresponding
pairs of signals. At ꢀ36 °C the coalescence of the proton signals
of two methyl groups of AAz moiety was observed. At 2 °C the
coalescence of tert-Bu group signals took place. At temperatures
It should be noted that rotation energies of the exocyclic
C
OPyACmethine bonds were estimated earlier for the series of hetero-
fulvenes and fell in the range of 58–92 kJ molꢀ1 [38].
The calculated values of the rotation energies (see Fig. 6) are in
good agreement with the bond orders: upon the increase of the
bond order the rotation energy value also increased. However, in
some cases experimental data differ from the theoretical both in
the absolute value and in the relative one.
8
N
13
7
O
10
4
9
11 12
5
Observed differences could be explained by the solvent influ-
ence, which would be discussed below, as well as by the peculiar-
ities of the calculation model. Despite these distinctions, the
general tendencies in the basicity variations upon the terminal
Het group changes are clearly visible.
It is important to note that when Het is more basic than AAz the
value of the rotation barrier around the CAAzACmethine bond in-
creases while the rotation barrier around the CHetACmethine bond
decreases upon the increase in difference of the terminal groups
Fig. 5. Variable temperature 1H NMR spectra of AAz-1-OPy in CD3CN. Coalescences of different signals at the corresponding temperatures are marked as ‘‘T coal’’. For the pairs
of H5AH7 and H12AH13 signals the significant difference of chemical shifts is observed, and in the region of their coalescence temperatures these peaks are ‘‘smeared’’.

60
A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
basicity. Thus, in the case of the most basic 2Q and 4Q groups the
largest values of the rotation barrier around the CAAzACmethine bond
are observed. We also confirm again our conclusion regarding the
low basicity of the 2-azaazulenium terminal group. On the basis
of results of the dynamic NMR investigations we obtained the fol-
lowing basicity order (electron donor ability) for the heterocyclic
terminal groups:
For the styryles the values of the rotation barriers for both
PhACmethine and CAAz(COPy)ACmethine bonds are in direct proportion
C
to the difference in basicities of the terminal groups. As could be
expected, the values of the rotation energy barriers for these bonds
of the styryl OPy-St(NMe2) are higher than for the methoxystyryl
OPy-St(OMe) (Table 2). The same difference in the rotation ener-
gies was also detected in the carbon NMR spectra of these com-
pounds recorded at ambient temperature in CDCl3. For the styryl
OPy-St(NMe2) signals of the carbon atoms C1 (109.94 ppm) and
C6 (134.71 ppm) are appreciably broadened due to the hindered
rotation around the C2AC3 and C4AC5 bonds, respectively
(Fig. 7). Such an effect was not observed for the methoxystyryl
molecules, which are characterized by lower rotation energies of
the corresponding bonds and sharp signals of atoms C1
(112.73 ppm) and C6 (133.09 ppm).
OPy < AAz < BIn < In < BT < BO < 4Q < 2Q < BIm;NPy
This sequence is analogous to the one presented by Brooker,
who arranged heterocyclic nuclei in order of basicity considering
the deviations in absorption maxima of para-dimethylaminostyryl
dyes [12]. There are only two exceptions in these sequences,
namely, according to [12] 4Q is more basic than 2Q, and BT is more
basic than BO.
3.6. The Influence of polymethine chain length
3.5. Styryl dyes
On the basis of the rotation energy values for the dye vinylogs
we can draw certain conclusions concerning the influence of the
length of polymethine chain on the characteristic features of the
conjugation in the system.
From the resonance–mesomerism theory point of view, the
main feature of the styryl dyes lies in more ‘‘polyene’’ than ‘‘cya-
nine’’ character of their conjugated system. It is exhibited in more
pronounced alternation of the bond lengths and bond orders, and
less pronounced alternation of the atomic electron density for
styryles. For the comparison, the magnitudes of Wiberg indexes
for the bonds of the conjugated chain constitute 1.259, 1.530,
1.227 for the styryl dye AAz-St(NMe2); 1.207, 1.605, 1.170 for
the methoxystyryl AAz-St(OMe); 1.337, 1.424, 1.383, 1.367 for
the cyanine AAz-1-OPy. From these data it is evident that for the
styryl AAz-St(NMe2) the bond alternation is less pronounced than
for the methoxystyryl AAz-St(OMe). Therefore for the styryl dyes
Only minor changes of the CAAzACmethine bond lengths were
observed upon going from the monomethine to pentamethine cya-
nines. Due to the M05-2X/6-311G(d,p) calculations in the case of
the symmetrical cyanines AAz-n-AAz these bond lengths are
1.407 Å, 1.407 Å, 1.405 Å, and in the case of the unsymmetrical
OPy-n-AAz they are 1.408 Å, 1.410 Å, 1.409 Å respectively. How-
ever, the bond populations and rotation energies are more sensi-
tive to the polymethine chain length. It was found that upon
going from the sterically hindered monomethine cyanines to the
unhindered trimethines with the same terminal groups the change
of the free activation energy DDG–298, calculated by the Eq. (4) cor-
relates with the torsion angle between the planes of the terminal
groups in the monomethine molecule (Fig. 8).
the values of the rotation energy around
CAAzACmethine and
CPhACmethinebonds are lower then for the cyanines. The calcula-
tions support this fact (Table 2). Unfortunately, we were unable
to determine experimentally the rotation barrier for the styryl
AAz-St(NMe2) due to its extremely low value, although at
T < ꢀ30 °C we detected some deceleration of the molecular frag-
ments rotation according to the NMR spectra. For the methoxysty-
ryl, no broadening of the corresponding signals was detected even
at the solvent crystallizing point at ꢀ63 °C.
DDG–298
¼
D
G–298ðtrimethineÞ ꢀ
D
G–298ðmonomethineÞ
ð4Þ
Among the dyes which have been investigated the minimal spa-
tial strain was observed for the molecules of the benzoxazolium
monomethine cyanine AAz-0-BO. It results in the stabilization of
the ground state and increase of the rotation energy. The perpen-
dicular transition state is not affected by the sterical features of
the terminal groups. In the case of AAz-n-BO the energies rises
in the row from the trimethine to the monomethine (DDG– < 0).
For the others monomethines the ground states are non-planar
and therefore energetically destabilized. It reduces the rotation
The solvation effects observed for the styryles are similar to that
observed for the cyanine dyes. According to the calculations, the
solvent reduces the energies of the ground and the excited states
as well as the rotation energy (Table 2). Di-tert-Bu-pyrylium termi-
nal group is less basic then the 2-azaazulenium one. It results in
the decrease of the alternations of the bond lengths and orders
upon the change to styryles OPy-St bearing di-tert-Bu-pyrylium
residue. We managed to measure experimentally the values of
the isomerization energy for these dyes. Besides that, we registered
also two parallel dynamic processes of the rotation around
CPhACmethine and COPyACmethine bonds. However, the exact value
for the first bond was not calculated, since the separation of the
corresponding signals was not achieved even at ꢀ92 °C.
Fig. 6. Values of the rotation energy around the CAAzACmethine bond for trimethines
AAz-1-Het according to the quantum chemical calculations (M05-2X/6-311G(d,p)),
D
Fig. 7. Fragments of 13C NMR spectra of the dyes OPy-St R = N(CH3)2 (a) and
R = OCH3 (b) in CDCl3 at T = 293 K.
E, and to the experimental measurements (DNMR), D .
G–298

A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
61
provided equipment. We are indebted to Dr. Vladimir V. Pir-
ozhenko for valuable discussion of theoretical and experimental
questions of dynamic NMR, Dr. Alexander B. Rozhenko and Dr. Oleg
A. Zhikol for advices in quantum-chemical calculations.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molstruc.2011.10.003.
Fig. 8. The differences of free energy activation DDG–298 (4) as a function of dihedral
(torsion) angle between the planes of the terminal groups of the monomethine
cyanine AAz-0-Het, and linear regression (solid line) calculated by the least-squares
method.
References
[1] A. Mishra, R.K. Behera, P.K. Behera, B.K. Mishra, G.B. Mishra, Chem. Rev. 100
(2000) 1973.
[2] A.A. Ishchenko, Yu.L. Slominsky, A.I. Tolamchov, J. Org. Pharm. Chem. 7 (2009)
3 (in Ukrainian).
[3] B.I. Shapiro, Spectral sensitization of silver halides in NIR region, in: S. Daehne,
U. Resch-Genger, O.S. Wolfbeis (Eds.), Near-Infrared Dyes for High Technology
Applications, NATO ASI Series, 3. High Technology, vol. 52, Kluwer Academic
Publishers, Dordrecht, Boston, London, 1998.
[4] S.R. Marder, Chem. Commun. (2006) 131.
[5] C. Sun, J. Yang, L. Li, X. Wu, Y. Liu, S. Liu, J. Chromatogr., B 803 (2004) 173;
G. Patonay, J. Salon, J. Sowell, L. Strekowski, Molecules 9 (2004) 40.
[6] B.A. Armitage, Top. Curr. Chem. 253 (2005) 55.
[7] D.G. Norman, R.J. Grainger, D. Uhrín, D.M.J. Lilley, Biochemistry 39 (2000)
6317.
[8] K.M. Sovenyhazy, J.A. Bordelon, J.T. Petty, Nucleic Acid Res. 31 (2003) 2561.
[9] T.Yu. Ogul’chanskya, M.Yu. Losytskyy, V.B. Kovalska, S.S. Lukashov, V.M.
Yashchuk, S.M. Yarmoluk, Spectrochim. Acta, A 57 (2001) 2705.
[10] R.A. Garoff, E.A. Litzinger, R.E. Connor, I. Fishman, B.A. Armitage, Langmuir 18
(2002) 6330.
Fig. 9. The rotation energies of the CAAzACmethine bond for the vinylogs AAz-n-AAz
and AAz-n-OPy (in CDCl3).
[11] S.M. Yarmoluk, S.S. Lukashov, T.Yu. Ogul’chansky, M.Yu. Losytskyy, O.S.
Kornyushyna, Biopolymers 62 (2001) 219.
[12] L.G.S. Brooker, A.L. Sklar, H.W.J. Creesman, G.H. Keyes, L.A. Smith, R.H. Sprague,
E. Van Lare, G. Van Zandt, F.L. White, W.W. Williams, J. Am. Chem. Soc. (1945)
1875.
[13] A.Ya. Ilchenko, Ukr. Khim. Zh. 40 (1976) 160 (in Ukrainian).
[14] G.G. Dyadusha, A.D. Kachkovskii, Ukr. Khim. Zh. 44 (1978) 948 (in Ukrainian).
[15] W. Grahn, Tetrahedron 32 (1976) 1931;
N. Katayama, Yu. Ozaki, Sh. Yasui, K. Iriyama, J. Mol. Struct. 274 (1992) 171.
[16] A.M. Kolesnikov, M.I. Povolotskii, Theor. Exp. Chem. 24 (1988) 220.
[17] E.I. Mayboroda, Yu.L. Slominsky, A.V. Turov, A.I. Tolmachev, Chem. Heterocycl.
Compd. 44 (2008) 78.
energy and receds the energies in the row from the tri- to monom-
ethine (DDG– > 0). Thus, DDG– parameter can be used to draw
certain conclusions on the spatial arrangement of the terminal
groups.
Further lengthening of the polymethine chain in the pentame-
thine cyanines leads to the decrease of the activation energy in
the case of the symmetrical cyanines AAz-n-AAz. Vice versa, in
the case of the unsymmetrical AAz-n-OPy minor increase of the
energy is observed (Fig. 9).
[18] A.M. Kolesnikov, F.A. Mikhailenko, Russ. Chem. Rev. 56 (1987) 275.
[19] P.M. Henrichs, S. Gross, J. Am. Chem. Soc. 98 (1976) 7169.
[20] S. Ghelli, G. Ponterini, J. Mol. Struct. 355 (1995) 193.
4. Conclusions
[21] A.D. Kachkovskii, N.M. Kovalenko, Theor. Exp. Chem. 33 (1997) 192;
A.D. Kachkovskii, O.A. Zhukova, Theor. Exp. Chem. 36 (2000) 150.
[22] J. Bricks, A. Ryabitskii, A. Kachkovskii, Eur. J. Org. Chem. (2009) 3439.
[23] J. Bricks, A. Ryabitskii, A. Kachkovskii, Chem. Eur. J. 16 (29) (2010) 8773.
[24] A.B. Ryabitskii, J.L. Bricks, A.D. Kachkovskii, A.N. Chernega, Y.G. Vlasenko, J.
Mol. Struct. 982 (2010) 79.
A systematic study of the conformational behavior of the
symmetrical and unsymmetrical cyanine dyes bearing a 2-azaazu-
lenium moiety in solution was carried out by means of quantum-
chemical and experimental methods.
[25] F.M. Hamer, in: A. Weissberger (Ed.), The Cyanine Dyes and Related
Compounds in the Chemistry of Heterocyclic Compounds, vol. 18,
Interscience, New York, NY, 1964, pp. 200–243. p. 200–243, 398.
[26] GAUSSIAN03; Revision E.01, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E.
Scuseria, M.A. Robb, J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin,
J.C. Burant, M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi,
G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida., T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M.
Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, C. Adamo, J. Jaramillo, R.
Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W.
Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J Dannenberg., V.G.
Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S.
Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong,
C. Gonzalez, J.A. Pople, Gaussian, Inc., Pittsburgh PA, 2004.
The direct correlation between the barriers to conformational
transformations and such parameters as electron-donating proper-
ties of the terminal groups and the polymethine chain length were
established experimentally. The values of the rotation energies for
the methine bonds are proportional to the difference in the elec-
tron-donating properties of the terminal groups of the unsymmet-
rical cyanines. Spatial hindrances of the bulky terminal groups
reduce the bond rotation energy in monomethine cyanines. The
styryl dyes are characterized by the minor values of the rotation
energies around the corresponding bonds compared to the cya-
nines. The data of the quantum-chemical calculations are in agree-
ment with those obtained experimentally.
It was proved by the calculations using the polarizable contin-
uum model that solvation stabilizes the ground and transition
states of the dyes and appreciably reduces the barrier values. The
solvent polarity has relatively weak influence on the absolute
energy values.
[27] Y. Zhao, D.G. Truhlar, Theor. Chem. Acc. 120 (2008) 215.
[28] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[29] M.S. Gordon, J.S. Binkley, J.A. Pople, W.J. Pietro, W.J. Hehre, J. Am. Chem. Soc.
104 (1982) 2797.
[30] K.B. Wiberg, Tetrahedron 24 (1968) 1083.
[31] E.D. Glendening, J.K. Badenhoop, A.E. Reed, J.E. Carpenter, J.A. Bohmann, C.M.
Morales, F. Weinhold, NBO 5.0, Theoretical Chemistry Institute, University of
Wisconsin, Madison, WI, 2001.
[32] M.W. Wong, M.J. Frisch, K.B. Wiberg, J. Am. Chem. Soc. 113 (1991) 4776.
[33] V. Barone, M. Cossi, J. Phys. Chem. A 102 (1998) 1995.
[34] H.J. Reich, WINDNMR: NMR Spectrum Calculation Version 7.1.12., Department
of Chemistry, University of Wisconsin, Madison, 2005. www.chem.wisc.edu/
areas/reich/plt/windnmr.htm.
Acknowledgments
We are grateful to Dr. Vladimir N. Fetukhin and Spoluka Chem-
ical Company (www.lifechemicals.com) for financial support and

62
A.B. Ryabitskii et al. / Journal of Molecular Structure 1007 (2012) 52–62
[35] I. Sandstrom, Dynamic NMR Spectroscopy, Academic Press, New York, London,
1982.
[37] J.W. Ochterski, Thermochemistry in Gaussian, Gaussian, Inc., 2000.
[38] I. Belsky, H. Dodiuk, Y. Shvo, J. Org. Chem. 42 (1977) 2734.
[36] A.D. Kachkovskii, A.B. Ryabitskii, Theor. Exp. Chem. 44 (2008) 232.