Optical spectra of some 6-styrylo-1H-pyrazolo[3,4-b]quinolines

Article abstract of DOI:10.1016/j.saa.2005.05.017

It was proposed a method of synthesis several 6-styrylo1-1H-pyrazolo[3,4-b]quinolines as promising material for optoelectronics. Particularly, 6-styryl-1,3-diphenyl-1H-pyrazolo[3,4-b]quinolines were prepared by Wadsworth-Emmons reaction. One of advantageo

Full text of DOI:10.1016/j.saa.2005.05.017

Spectrochimica Acta Part A 63 (2006) 320–329
Optical spectra of some 6-styrylo-1H-pyrazolo[3,4-b]quinolines
I. Fuks-Janczareka, E. Gondekb, I.V. Kitykb,∗, K. Danelc,
c
b
c
´
´
L. Krzeminska , J. Sanetra , B. Kwiecien
a
Institute of Physics, J. Długosz University of Cze˛stochowa, Al.Armii Krajowej 13/15, Cze˛stochowa, Poland
b
Institute of Physics, Technical University of Cracow, ul.Podchorazych 1, Poland
c
Institute of Chemistry, Agricultural Academy of Cracow, Cracow, Poland
Received 25 March 2005; accepted 14 May 2005
Abstract
It was proposed a method of synthesis several 6-styrylo1-1H-pyrazolo[3,4-b]quinolines as promising material for optoelectronics. Partic-
ularly, 6-styryl-1,3-diphenyl-1H-pyrazolo[3,4-b]quinolines were prepared by Wadsworth–Emmons reaction.
One of advantageous of these materials is stability of their optical spectra versus the possible cis- and trans-transformation.
Theoretical and experimental studies of optical absorption and photoluminescence excitation spectra for styryloquinolines were done. They
show that the AM1 semi-empirical method gives better results compared to other approaches. We have found that the solvents do not play a
role in the behavior of the spectra. The backside-substituted groups do not influence substantially the observed optical spectra.
© 2005 Elsevier B.V. All rights reserved.
Keywords: 6-styrylo1-1H-pyrazolo[3,4-b]quinolines; Wadsworth–Emmons reaction; Semi-empirical method
1. Introduction
ever, principal limitation is related to the presence of aromatic
groups. So it is necessary to take into account the influence of
fitting parameters on the simulated spectra. For us it is impor-
tant to compare the spectra obtained by this two methods tom
work out recommendations for other similar compounds.
The article is as follows: chemical synthesis of styrylo-
quinolines is given in Section 2. In Section 3, we present
calculation procedure, molecular geometry optimization,
experimental and theoretical absorption and photolumines-
cence spectra. Conclusions are presented in Section 4.
Among large number of organic optoelectronic materials
considerable interest present 1H-pyrazolo[3,4-b]quinoline
anditsderivativeswhichrecentlyhavefoundtobeasaclassof
highly fluorescent materials in the blue spectral range [1–4].
These materials are potential candidates for use in optoelec-
tronic devices, optical fibers, photonics, medicine and light
emitting diodes.
Semi-empirical methods are grounded on restricted
Hartree–Fock approach [5,6]. The most frequently used
methods (MNDO, AM1, PM3) are all based on the neglect
of differential diatomic overlap (NDDO) integral approxima-
tion [5–8]. For the AM1 and PM3, the fitting parameterized
procedure is performed by the dipole moments. Particularly
AM1 method is fitted to transition dipole momentums. How-
2. Experimetnal
2.1. Chemical synthesis
The synthesis of 6-styryl-1,3-diphenyl-1H-pyrazolo[3,4-
b]quinolines 6-Styryl-1,3-diphenyl-1H-pyrazolo[3,4-b]qui-
nolines 4 were prepared according to the sequence of reaction
depicted on the Fig. 1.
∗
Corresponding author at: Solid State Department, Institute of Physics
AJD Czestochowa, J. Długosz University of Cze˛stochowa, Al.Armii Kra-
jowej 1315, PL42201 Cze˛stochowa, Poland.
Thus 6-methyl-1,3-diphenyl-1H-pyrazolo[3,4-b]quino-
line was prepared by condensation of p-toluidine with
Tel.: +48 601504268; fax: +48 223612228.
E-mail address: i.kityk@ajd.czest.pl (I.V. Kityk).
1386-1425/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2005.05.017

I. Fuks-Janczarek et al. / Spectrochimica Acta Part A 63 (2006) 320–329
321
Fig. 1. (a) NBS/CCl4; (b) P(OEt)3/xylene; (c) ArCHO/NaH.
5-chloro-4-formyl-1,3-diphenylpyrazole according to Brack
procedure [9]. 1 was heated in carbon tetrachloride with
NBS and AIBN yielding 2 [10]. 6-Bromomethyl derivative
2 was heated with triethylphosphite in xylene for 24 h
at 140 ◦C. The styryl derivaties 4 were prepared from
ester 3 and appropriate aldehydes in Wadsworth–Emmons
reaction.
Yellow crystals, yield 65%, mp 95.7 ◦C (toluene/hexane).
Anal. calcd. for C27H26N3O3P: C, 68.78; H, 5.56; N, 8.91.
Found: C, 68.81; H, 5.74; N, 8.73.
2.3. Synthesis of styryl derivatives 4 —general
procedure
Appropriate aldehyde (5 mmol) and sodium hydride (60%
in oil, 0.2 g, 5 mmol) were put into a 50 ml round-bottomed
flask with anhydrous DME (20 ml). To this mixture com-
pound 3 (2.36 g, 0.005 mol) in 10 ml DME was added drop-
wise at 0 ◦C with stirring. The solution was stirred at 70 ◦C
for 24 h. After completion of the reaction, some solvent was
removed and water was poured into reaction flask. The pre-
cipitate was filtered, dried and was further purified by column
chromatography.
2.2. (1,3-Diphenyl-1H-pyrazolo[3,4-b]quinolin-6-
ylmethyl)-diethyl phosphonate 3
This compound was prepared by reacting 2 (2.07 g,
5 mmol) with triethyl phosphite (0.73 g, 5.5 mmol) in xylene
(10 ml) at 90 ◦C for 24 h. The solvent was removed and
the oily residue was chromatographed over silica gel using
toluene/ethyl acetate (3:1) as eluent.
Fig. 2. Chemical structure of studied compounds styryloquinolines 1, 2, 3, 4.

322
I. Fuks-Janczarek et al. / Spectrochimica Acta Part A 63 (2006) 320–329
2.4. 1,3-Diphenyl-6-[(E)-styryl]-1H-pyrazolo[3,4-
b]quinoline 4a
Usually in semi-empirical methods, these integrals are
neglected or parameterized, and only valence shell electrons
are considered. The Hamiltonian operator takes the form [5]:
Yellow crystals, yield 71%, mp 204.6 ◦C (toluene). Anal.
calcd. for C30H21N3: C, 85.08; H, 5.00; N, 9.92. Found: C,
84.98; H, 4.89; N, 9.67.
ꢁ
ꢂ
N
N
v
Nv−1
v
ꢀ
ꢀ ꢀ
1
1
2
ˆ
Hval
=
=
− ∇i + V(i)
+
2
i=1 j=i+1 rij
i=1
N
v
Nv−1
Nv−1
2.5. 6-[(E)-2-(4-Methylphenyl)-vinyl]-1,3-diphenyl-1H-
pyrazolo[3,4-b]quinoline 4b
ꢀ
ꢀ ꢀ
1
ˆ core
Hval (i) +
(1)
i=1 j=i+1 rij
i=1
Yellow crystals, yield 65%, mp 238.40 ◦C (toluene). Anal.
calcd. for C31H23N3: C, 85.10; H, 5.30; N, 9.60. Found: C,
84.87; H, 5.17; N, 9.43.
where Nv is total number of valence electrons in the molecule,
V(i) the potential energy of the ith electron in the field of
nuclei and inner-shell electrons.
The AM1 and PM3 methods use Slater-type orbitals
(STOs) as basis set functions:
2.6. 6-[(E)-2-(2,4-Dichloro-phenyl)-vinyl]-1,3-
diphenyl-1H-pyrazolo[3,4-b]quinoline 4c
f = Nrn−1e−ξrYlm(θφ)
where Ylm is spherical harmonic special functions.
(2)
Yellow crystals, yield 73%, mp 225.8 ◦C (toluene). Anal.
calcd. for C30H19Cl2N3: C, 73.11; H, 3.89; N, 8.53. Found:
C, 73.22; H, 3.95; N, 8.45.
As a rule one do the following simplifying approximation:
ꢃꢃ
fz∗(1)fy(1)fm∗(2)fn(2)
dν1 dν2 = δzyδnm(zy|mn) (3)
r12
2.7. 1,3-Diphenyl-6-((E))-2-pyridin-4-yl-vinyl-1H-
pyrazolo[3,4-b]quinoline 4d
where δzy is Kronecker symbol. The Fyy terms in the secular
determinant have a form:
Yellow crystals, yield 45%, mp 273.5 ◦C (toluene). Anal.
calcd. for C29H20N4: C, 82.05; H, 4.75; N, 13.20. Found: C,
81.86; H, 4.56; N, 13.05.
The general structure of a synthesized styryloquinolines
derivate molecule is shown in Fig. 2.
ꢀ
Fyy = Uyy
−
CB(yy + sBsB)
B=A
ꢄ
ꢅ
A
ꢀ
1
+
Pzz (yy|zz) − (yz|yz)
2
z
B
B
3. Results and discussion
ꢀ ꢀ ꢀ
+
Ppq(yy|pq)
q
(4)
(5)
p
B=A
3.1. Calculation procedure
where the core integral Uyy is
The simulation of optical absorption spectra were
performed using AM1, PM3 semi-empirical approaches.
Molecular structures of the molecules were optimized by
molecular-mechanics force field method (MM+) which is
particularly useful for carbon-based systems.
ꢇ
ꢇ
ꢇ
ꢇ
ꢇ
ꢇ
ꢆ
1
2
Uyy
=
fy − ∇ + VA fy
ꢇ
ꢇ
2
The orbitals fz and fy are centered on atom A, and orbitals fp
and fq are centered on atom B. CB is the core charge on atom
B, i.e. atomic number of atom B minus the number of inner-
shell electrons, and (yy|sBsB) is a two-electron, two-center
overlapping integral. The sB orbital is the valence s orbital on
atom B. Pzz and Ppq are called density matrix elements and
are defined as
We used the MM+-force field method because:
• It is very fast.
• May be used for large organic molecules.
• It gives simultaneously an information about values of total
energy and molecular geometry.
(1/2)N
v
ꢀ
As a criterion of self-consistent convergence during the
numerical procedure we have chosen a value of energy gra-
dient to be equal to about 0.01 kcal/mol.
Following the calculated optimized molecular structure
quantum chemical calculations were done. Due to the rela-
tively large molecular formula and large amount of aromatic
rings we applied the semi-empirical quantum chemical meth-
ods. All the semi-empirical methods are fitted by choosing of
different forms of parameterization in the overlapping inte-
grals.
Pzz ≡ 2
cz∗j czj
(6a)
j=1
(1/2)N
v
ꢀ
Ppq ≡ 2
cp∗j cpj
(6b)
j=1
for closed-shell configurations. There are two types of pos-
sible presentation of the off-diagonal elements Fzy in the
secular determinant. The element in which the fz and fy
orbitals are on the same atom constitutes one type and is

I. Fuks-Janczarek et al. / Spectrochimica Acta Part A 63 (2006) 320–329
323
labeled FzAyA. The other type of off-diagonal element has the
fz and fp orbitals on different atoms and is labeled FzApB
.
ꢀ
1
AA
zy
F
= −
CB(zy|sBsB) + Pzz[3(zy|zy) − (zz|yy)]
2
B=A
B
B
ꢀ ꢀ ꢀ
+
Ppq(zy|pq)
(7a)
p
q
B=A
A
B
ꢀ ꢀ
1
1
2
AB
zp
F
=
[βz + βp]Szp
−
Pyq(zy|pq)
(7b)
2
y
q
Szp is the overlap integral <fz|fp> and it is calculated exactly.
The total energy of the molecule, Etotal, is the sum of the total
valence electronic energy, Eel, and the energy of repulsion
between the cores on atoms A and B.
ꢀ ꢀ
Etotal = Eel +
[CACB(sAsA|sBsB) + fAB
]
(8)
B>A
A
Unlike the AM1 method, the PM3 method treats the one-
center, two-electron interaction integrals (zz|yy) and (zy|zy) as
fittingparameters. Also, theprocedurethatisusedtooptimize
the atomic parameters differs significantly from the proce-
dure described above [11].
3.2. Molecular geometry optimization
The molecular dynamics simulations allow to optimize the
molecular geometry structure corresponding to the minimum
of the total energy as a criterion of self-consistent iteration
convergence, for the molecule geometry optimization and
electronic structure calculation we have chosen the conver-
genttotalenergycriteriontobeequalofabout0.014 eVwhich
corresponds to the model approach. Addition information
concerning the origin of absorption bands can be obtained
by considering the electrostatic potential. Fig. 3 shows the
total electrostatic potential for molecules derivatives calcu-
lated within AM1-procedure [12].
The substitution of the backside group changes drastically
electrostatic distribution within the molecular chain. So it
may be manifested also in the behavior of the optical spectra.
In general the band gap of the molecule 1–4 is the
orbital energy difference between the highest occupied—an
orbital that contains at least one electron molecular orbital—
(HOMO)andthelowestunoccupied—anorbitalthatcontains
at least one open space for an electron—molecular orbital
(LUMO) when the repeated unit number is infinite.
These orbitals are called the frontier orbitals, and deter-
mine the way the molecule interacts with other species. The
HOMO is the orbital that could act as an electron donor,
since it is the outermost (highest energy) orbital containing
electrons. The LUMO is the orbital that could act as the elec-
tron acceptor, since it is the innermost (lowest energy) orbital
that has room to accept electrons [13–15]. Using the semi-
empirical all-valence semi-empirical AM1 and PM3 methods
Fig. 3. Total electrostatic potential for molecules derivatives calculated
within AM1-procedure for styryloquinolines 1, 2, 3, 4. Balls: , carbon;
, hydrogen; , nitrogen; , chlorine.

324
I. Fuks-Janczarek et al. / Spectrochimica Acta Part A 63 (2006) 320–329
Table 1
Energy values of HOMO and LUMO for styryloquinolines
Methods
Molecule 1
Molecule 2
Molecule 3
Molecule 4
HOMO
(eV)
LUMO
(eV)
|ꢀE|
6.19
16.15 −20.44
20.59 −25.08
32.01 −35.85
HOMO
(eV)
LUMO
(eV)
|ꢀE|
6.87
16.84 −12.59
20.71 −15.63
29.74 −23.84
HOMO
(eV)
LUMO
(eV)
|ꢀE|
HOMO
(eV)
LUMO
(eV)
|ꢀE|
AM1
0
20
40
60
−7.60
−12.74
−16.21
−26.55
−1.41
3.41
4.38
−8.15
−1.28
3.60
4.37
−8.22
−1.40
3.48
4.34
6.82 −10.99
−4.89
0.58
2.88
6.90
12.30
19.29
30.06
16.07 −11.72
19.97 −17.04
28.96 −25.21
5.46
5.11
5.12
4.85
PM3
0
20
40
60
−9.22
−12.90
−16.33
−24.11
−2.55
2.72
3.91
6.67
−8.17
−1.49
3.01
3.93
6.68
−8.02
−1.56
2.96
3.81
6.46 −12.03
15.72 −15.54
19.40 −17.19
26.74 −29.20
−5.05
0.14
2.91
6.98
15.68
20.01
33.61
15.62 −12.65
20.24 −18.04
29.48 −21.20
15.66 −12.76
21.97 −15.59
26.14 −21.79
5.37
4.93
4.95
4.41
The HOMO and LUMO energies are originated from the atomic orbital energies of the carbazol group and agreed well with experimental data (5.1 eV).
has been calculated of theoretical values HOMO and LUMO
for molecule 1–4 (Table 1).
where j is the number of transition from ground state; Ej = hωj
the transition energy; µ(jx,y,z) the transition dipole component
hale Bern obtained directly within AM1 and PM3 procedure;
H the damping (value of about 0.12 eV, what gives the best
agreement in contour shape of calculated optical absorption
spectra compared to experimental data).
3.3. Experimental and theoretical absorption and
photoluminescence spectra
The optical spectra were simulated by the two semi-
empirical methods—AM1 and PM3. The electronic tran-
sition spectra were calculated considering only the singly
excited configuration interactions. Seven electrovolt limited
the criteria regarding the excitation energies or orbital occu-
pation criterion (3 occupied and 3 unoccupied orbitals).The
UV absorption spectra were recorded in tetrahydrofuran solu-
tionusingShimadzuUV–vis2101scanningspectrophotome-
ter in range 200–800 nm. Solution measurements were per-
formed using standard 1 cm path length quartz crucible for
absorption spectrometry.
The calculations were done for two type of isomers (cis-
and trans-). However, experimentally we are able to measure
only trans-compounds of the investigated molecule. From
the Fig. 4, one can clearly see that the AM1 method gives
better agreement compared to the PM3 ones. At the same
time it is a bit surprisingly that cis- and trans-isomerisation
should not play substantial role in the spectra simulated. Sev-
eral broadening of the UV-absorption spectra is caused by
electron-vibration interactions of the pure molecular elec-
tronic transitions.
Measured absorption spectra of styryloquinolines and the
calculated absorption spectra (H = 0.12 eV) are presented in
the Fig. 4. One can see that experimental absorption spectra
of styryloquinolines and calculated within the AM1 and PM3
methods are rather similar. They can be characterized gener-
ally by two relatively strong absorption bands. We observed
the theoretical peaks absorption (see Table 2).
The corresponding experimental spectra have in all the
cases an absorption threshold at about 241 nm which appears
to be in fairly good agreement with quantum chemical simu-
lations, namely for the spectra calculated by AM1 method. It
is also amazing that spectral positions of the most strongest
broad absorption bands at about 241 nm in the measured
spectra well coincide with the calculated spectra. Both
quanta–chemical methods give here sufficiently good agree-
ment, but AM1 method is better than PM3
Measured photoluminescence spectra of styryloquino-
lines and the calculated spectra (H = 0.12 eV) are presented
in the Fig. 5.
The intensity of photoluminescence spectra is determined
by the following expression:
The intensity of optical absorption is determined by the
following expression:
ꢊ
2
i=x,y,z(Mj(x,y,z) − M0(x,y,z)
)
n
ꢀ
IPL(ω) = ω4
(10)
(ω − ωj)2 + (H/2)2
ꢈ
ꢉ
ꢁ
ꢂ
n
2
ꢀ ꢀ
j=1
E − Ej
2
Ax,y,z ≈ ω
|µ(jx,y,z)| exp
−
(9)
where j is the number of transitions form ground state g
to excited state; ωj the corresponding transition frequencies
H
i=x,y,z
j=1
Table 2
The theoretical maximum peaks absorption of styryloquinolines 1–4
Theoretical peak
Molecule 1, λmax (nm)
Molecule 2, λmax (nm)
Molecule 3, λmax (nm)
Molecule 4, λmax (nm)
AM1a
PM3a
AM1a
PM3a
AM1a
PM3a
AM1a
PM3a
cis
trans
241
242
258, 290
260, 286
244
245
255
250
241
241
278
275
244
245
252
260
a
Method.

I. Fuks-Janczarek et al. / Spectrochimica Acta Part A 63 (2006) 320–329
325
(Ej/h);Mj(x,y,z), M0(x,y,z) theexcitedandreferencestatedipole
component, respectively and H is the damping.
We found that experimental and theoretical photolumines-
cence spectra in fairly good agreement.
The spectrum of photoluminescence (PL) was observed
in the range 400–700 nm.
From Fig. 5 one can see that the general view of the
theoretical photoluminescence spectra may be described
Fig. 4. Measured absorption spectra of the styryloquinolines (the experimental absorption spectra of the molecules dissolved in the tetrahydrofuran) and
calculated spectra within the semi-empirical quantum chemical PM3 and AM1 models. Inset shows the chemical structure of styryloquinolines obtained within
the geometrical optimization procedure has been carried out by (MM+) mode. The theoretically absorption spectra calculated of the investigated molecules in
the vacuum.

326
I. Fuks-Janczarek et al. / Spectrochimica Acta Part A 63 (2006) 320–329
Fig. 4. (Continued ).
Table 3
The theoretical maximum peaks potoluminescence of styryloquinolines 1–4
Theoretical peak
Molecule 1 λmax (nm)
Molecule 2 λmax (nm)
Molecule 3 λmax (nm)
Molecule 4 λmax (nm)
AM1a
PM3a
AM1a
PM3a
AM1a
PM3a
AM1a
PM3a
cis
trans
475
500
525
525
470
460
525
531
475
477
490
500, 525
475
480
530
525
a
Method.

I. Fuks-Janczarek et al. / Spectrochimica Acta Part A 63 (2006) 320–329
327
Fig. 5. Measured photoluminescence spectra of the styrylochinolines (the experimental absorption spectra of the molecules dissolved in the chloroform)
and calculated spectra within the semi-empirical quantum chemical PM3 and AM1 models. The theoretically photoluminescence spectra calculated of the
investigated molecules in the vacuum.

328
I. Fuks-Janczarek et al. / Spectrochimica Acta Part A 63 (2006) 320–329
Fig. 5. (Continued ).
by several maxima and are relatively similar. In particu-
larly we can see that the experimental photoluminescence
peaks obtains for molecule at about 475 nm. The same fig-
ure represented the theoretical photoluminescence peaks.
We observed the theoretical peaks photoluminescence (see
Table 3).
The absorption and photoluminescence spectra give
important information concerning the quantum chemical

I. Fuks-Janczarek et al. / Spectrochimica Acta Part A 63 (2006) 320–329
329
states and may serve as a powerful tool for understanding
the nature of the observed absorption and related lumines-
cence. At the same time, these calculations (experimental
and theoretical) may give an advice for directed changes of
the specific molecular structure.
groups do not influence substantially the observed optical
spectra.
Acknowledgment
This work was supported by the Ministry of Science
(KBN) and authors wish to express their sincere thanks for
grant 3 T11B 074 26 and 4 T09B 006 23.
4. Conclusion
A method of synthesis several 6-styrylo1-1H-pyrazolo
[3,4-b]quinolines was proposed. Particularly, 6-styryl-1,3-
diphenyl-1H-pyrazolo[3,4-b]quinolines were prepared by
Wadsworth–Emmons reaction. One of advantageous of these
materials is stability of their optical spectra versus the possi-
ble cis- and trans-transformation.
In this paper we presented experimental and theoretical
study of the optical absorption and photoluminescence exci-
tation spectra for styryloquinolines. We have calculated elec-
trostatic potential and value of the HOMO–LUMO energy
gap of these molecules, too.
A satisfaction agreement between the theoretical calcu-
lated and experimentally measured spectra was observed.
One of advantageous of these materials is stability of their
optical spectra versus the possible cis- trans-transformation.
Theoretical and experimental studies of optical absorption
and photoluminescence excitation spectra for styryloquino-
lines were done. They show that the AM1 semi-empirical
method gives better results compared to other approaches.
We have found that the solvents do not play a role in
the behavior of the spectra. The backside-substituted
References
[1] Z. He, G.H.W. Milburn, K.J. Baldwin, D.A. Smith, A. Danel, P.
Tomasik, J. Lumin. 86 (2000) 1–14.
[2] E. Kos´cien´, J. Sanetra, E. Gondek, B. Jarosz, I.V. Kityk, J. Ebothe,
A.V. Kityk, Opt. Commun. 242 (2004) 401–409.
[3] E. Gondek, E. Kos´cien´, J. Sanetra, A. Danel, A. Wisła, A.V. Kityk,
Spectrochim. Acta, Part A 60 (2004) 3101–3106.
[4] E. Gondek, E. Kos´cien´, J. Sanetra, A. Danel, A. Wisła, A.V. Kityk,
Opt. Commun. 227 (2003) 115–123.
[5] W. Kołos, J. Sadlej, Atom i Cze˛steczka, WNT, Warszawa, 1998.
[6] I.I. Guseinov, J. Mol. Struct. 422 (1998) 75–78.
[7] Walter Thiel, Mod. Meth. Algor. Quant. Chem. 3 (2000) 261–283.
[8] James J.P. Stewart, J. Mol. Model. 10 (2004) 155–164.
[9] A. Brack, Ann. Chem. 551 (1965) 105.
[10] K. Rurack, A. Danel, K. Rotkiewicz, D. Grabka, M. Spieles, W.
Rettig, Org. Lett. 4 (2002) 4647–4650.
[11] http://www.chm.davidson.edu/ronutt/che401/SemiEmp/SemiEmp.htm.
[12] Peter Politzer, Theor. Chem. Acc. 111 (2004) 395–399.
[13] C. Kubli-Garfias, J. Mendieta, R. Vazquez, J. Mol. Struct. 388 (1996)
35–41.
[14] Chia M. Chang, J. Mol. Struct. 622 (2003) 249–255.
[15] Jian Fang, Jie Li, J. Mol. Struct. 593 (2002) 179–185.