2014; 17(3): 550-556 © 2014Materials Research.
DOI:D httpI://dx.doi.org/10.1590/S1516-14392014005000056
Theoretical Investigation of Geometric Configurations
and Vibrational Spectra in Citric Acid Complexes
Rodrigo Marques Ferreiraa, Maycon Mottaa,b, Augusto Batagin-Netoa,
Carlos Frederico de Oliveira Graeffc, Paulo Noronha Lisboa-Filhoc, Francisco Carlos Lavardac*
aPrograma de Pós-Graduação em Ciência e Tecnologia de Materiais – POSMAT,
Universidade Estadual Paulista – UNESP, Bauru, SP, Brazil
bGrupo de Supercondutividade e Magnetismo, Departamento de Física,
Universidade Federal de São Carlos – UFSCar, São Carlos, SP, Brazil
cDepartamento de Física, Faculdade de Ciências,
Universidade Estadual Paulista – UNESP, Bauru, SP, Brazil
Received: April 15, 2013; Revised: March 22, 2014
The performance of advanced electronic ceramics is directly related to the synthesis route employed.
Sol-gel methods are widely used for this purpose. However, the physicochemical intermediate steps
are still not well understood. Better understanding and control of these processes can improve the final
quality of samples. In this work, we studied theoretically the formation of metal complexes between
citric acid and lithium or barium metal cations with different citric acid/metal proportions, using
Density Functional Theory electronic structure calculations. Infrared and Raman scattering spectra
were simulated for the more stable geometric configurations. Using this methodology, we identified
some features of complexes formed in the synthesis process. Our results show that the complexes can
be distinguished by changes in the bands assigned to C=O, COH-, and COO- group vibrations. An
estimate of the most stable complexes is made based on total energy.
Keywords: advanced electronic ceramics, density functional theory, infrared spectra simulations,
Raman spectra simulations, sol-gel, barium complex
1. Introduction
The performance of advanced electronic ceramics is
highly dependent on their synthesis route. In the so-called
sol-gel process, there are many variations in the routes used
for preparing samples. A deeper understanding of complexes
between α-hydroxycarboxylic acids and metals (M) that are
formed on those routes is desirable, since these compounds
are highly effective in obtaining heterometallic advanced
ceramics1. Several metallic ions with different applications
have been employed in this kind of route, such as lithium
ion in the manufacturing of cathodes for fuel cells2,3 and
barium ion in pellets of high temperature superconductors4-6,
magnetic materials7,8, and multiferroic materials9.
The modified polymeric precursors method (MPPM)1,4
involves the polyesterification process of a metal
chelate complex using a hydroxycarboxylic acid and a
polyhydroxy alcohol in an aqueous solution, which turns
into a polymeric gel. In this method, the most commonly
employed carboxylic acid is citric acid (CA)4. The MPPM
method has been widely used, since it allows for the
preparation of polycrystalline samples of excellent quality,
obtaining a highly homogeneous ceramic powder with
nanometer-scale particles, and it is reactive at temperatures
up to 30% lower than the ceramic powders obtained by
traditional routes1,5.
Several steps are involved in the MPPM procedure.
As mentioned, the main goal of this method is to obtain
a ceramic powder using polymeric precursors via
α-hydroxycarboxylic acid and ethylene glycol. There
are two sequential chemical reactions involved in these
methods1. The first, which is the purpose of our theoretical
studies, is the chelation between the α-hydroxycarboxylic
acid and metal ions obtained from simple salts, such as
oxides or carbonates. The second is the polyesterification
reaction with ethylene glycol, which results in a polymer.
These reactions occur in an aqueous medium to form a
sol. The gel is obtained after the evaporation of water by
raising the solution temperature to 90°C. Following this, heat
treatment at 400°C is carried out on the polymer, resulting
in a multicomponent oxide ceramic powder with extremely
homogeneous stoichiometry. The resulted powder is then
calcinated with several intermediate grindings in order to
remove the residual organic matter and form the desired final
phase. After calcination, the powder is pressed and sintered
to obtain ceramic pellets.
The polymer formed during synthesis is responsible for
both the intermediate and final structures, since it is believed
that the complexes tend not to break down during the
polymerization process. Consequently, better understanding
of the formation of metal complexes could provide valuable
information for improving the quality of the final samples. *e-mail: lavarda@fc.unesp.br
2014; 17(3) 551Theoretical Investigation of Geometric Configurations and Vibrational Spectra in Citric Acid Complexes
Although it is a widely used synthesis route, intermediate
physicochemical reactions and metal complexes are still
poorly known6, having been the subject of only a small
number of theoretical works10-12.
In this report, we have studied theoretically the citric
acid complexes formed by the lithium and barium ions.
For the lithium complexes, one citric acid molecule was
chelated with one or two metallic ions, whereas one
barium ion was chelated with one, two, or three citric acid
molecules in the barium complexes. The identification of
geometry configurations, the associated bands for infrared
(IR) and Raman spectra and the stability of such metallic
ion complexes were investigated, providing additional
information that can aid the preparation of electronic
ceramics.
2. Methodology
Our work has two distinct steps for each system studied.
The first step is determining the most stable structure
(geometry optimization) and the second is simulating
infrared and Raman shift spectra.
Prior to setting up initial geometries for further
optimization, Condensed-to-Atoms Fukui Indices analysis
(CAFIs)13 was carried out to evaluate the most likely
chelation sites for both metal ions on the CA structure. In
general, these indices describe how the frontier orbitals
are modified when the number of electrons is changed
and are employed to understand and predict the reactivity
of molecules and polymers14-16. According to the type of
reaction, three distinct CAFIs are defined. In our case
(metal incorporation), we considered electrophilic attack
(f
k
−) analysis:
f
k
− = q
k
(n) − q
k
(N−1) (1)
for atom k as a nucleophile, where q
k
(n) and q
k
(N−1) are
the electronic populations on the k-th atom of the cationic
and neutral species of the system studied, calculated for
neutral specie geometry.
After CAFIs analysis, the initial geometries of the metal
complexes were then optimized using the semiempirical
Parametric Method 617 (PM6), implemented in package
MOPAC200918. This method provides a very good structure
to be the initial geometrical configuration from which to
start the following phase, the ab initio quality search for the
most stable structure, for which we employed the Density
Functional Theory (DFT)19. For DFT calculations, the most
appropriate functional must be selected to ensure high-
quality results. Of the many available functionals, we chose
the hybrid functional B3LYP20, which has already been
used successfully in similar compounds20,21. Moreover, this
functional is widely used, since it has proven to be reliable,
generally yielding good-quality results22. The 6-31G basis
functions set23 was adopted for all atoms except barium, for
which we adopted the set LANL2DZ-ECP24,25, owing to the
high number of electrons of this metallic element. CAFIs,
Raman, and IR spectra were calculated with the same DFT
approach used in the geometry optimization.
In CAFIs calculations, the electronic populations
were evaluated by electrostatic potential fitting (ESP)26
and Mulliken population analysis27, and the restricted
open-shell approach (ROKS) was employed to avoid spin
contamination problems.
After obtaining the most stable structure for each
complex, we simulated the infrared and Raman shift spectra
to compare with the experimental data. The calculated
frequency values are usually multiplied by a scaling factor,
which varies according to the theoretical model used. For
DFT/B3LYP/6-31G and DFT/B3LYP/LANL2DZ, the scale
factors are 0.962 and 0.961, respectively28. The results
presented in this work were not multiplied by a scaling
factor.
All DFT calculations were performed with restricted
electronic orbitals, in vacuum, with the General Atomic
and Molecular Electronic Structure System (GAMESS)
package29. The software employed for the visualization of
complex structures and the evaluation of their properties
was Gabedit30 and Wxmacmolplt31.
3. Results and Discussion
First, it was necessary to define an initial geometry for
each complex studied. For this purpose, we needed to know
the oxidation state of the metal ions and the deprotonation
sites of citric acid. It was considered that lithium could enter
as a monovalent cation (Li+1) and barium as a divalent cation
(Ba+2)32. Deprotonation of CA was made, such that the final
complex was a closed-shell system.
Figure 1 shows the structure of CA and the CAFIs
calculated for eletrophilic attack employing Mulliken
population analysis. Regions in red and blue represent
reactive and non-reactive sites, respectively. Other colors
indicate sites with intermediary reactivity. Similar results
were obtained via ESP charge partitioning.
As can be seen, the central carboxylic group shows
higher CAFIs for eletrophilic attacks, defining a susceptible
region for metal chelation. Based on this result, the first
proton (H+) was removed from this group in all calculations.
In the complexes of citric acid with lithium, one H+ ion
was removed from CA for each lithium ion present. In the
case of CA/Li, after removing the first proton from CA, we
added one lithium ion Li+1 near the molecule. The complex
CA/2Li was modelled after the complex CA/Li optimization
by subtracting a second H+ from one of the side carboxylic
groups of the CA molecule. We then added a second lithium
ion near the complex. The obtained stable final structures
are shown in Figure 2 for both citric acid/metal ratios, and
agree well with previous results33.
Figure 1. Citric acid structural formula and Condensed-to-Atoms
Fukui Indices analysis. The most likely site for an electrophilic
attack is shown in red (color online).
552 Ferreira et al. Materials Research
The results obtained for the infrared spectra are
shown in Figure 3. For a pure CA molecule, it is possible
to observe bands corresponding to the vibrations of the
groups O-H (stretching) at 3650 cm-1, CH2 (stretching)
at 3140 cm-1, C=O (stretching) at 1750 cm-1, COOH
(symmetrical stretching), O-H (in-plane bending), and
C-H (wagging) between 1550 cm-1 and 1300 cm-1, C-OH
(stretching) at 1120 cm-1, C-OH (wagging) at 685 cm-1
and COOH (scissoring) at 560 cm-1[34]. Our results show
excellent agreement with experimental data35. For example,
the C=O, C-OH and COOH (1420 cm-1) bands show a
deviation of less than 1%.
Figure 3 shows the influence of the number of lithium
ions in the complex. One effect of the addition of Li occurs
in bands associated with C=O bonds. In pure CA, this strong
band is at 1750 cm-1, shifting to 1714 cm-1 and 1706 cm-1
when one or two lithium ions are added, respectively, owing
to the deprotonation of the carboxyl groups. Furthermore, it
is possible to see the presence of a new band at 1275 cm-1
for the CA/Li complex, corresponding to the symmetric
stretching of the COO- group. The band appears when
carboxyl acts as a monodentate ligand36,37; this gives rise
at the same time to an asymmetric stretching of the COO-
group at 1684 cm-1, which is hidden because of the strong
vibrational mode of the C=O. On the other hand, a carboxyl
group acts as a bidentate ligand in the CA/2Li complex, with
peaks at 1375 and 1631 cm-1[36-38]. The difference between
the asymmetrical and symmetrical stretching of the COO
group for the CA/Li and CA/2Li is in agreement with the
criteria presented by Kakihana et al.37 for the bidentate or a
monodentate character of the carboxylic group, compared
with a free carboxylic group (COO-), the difference between
the asymmetric and symmetric stretching vibrations being
298 cm-1 (1702 cm-1 and 1404 cm-1, respectively).
Another clear case of the influence of lithium appears
in bands associated with O-H and COH bonds, owing to
a hydrogen bond that appears between the hydrogen of
the hydroxyl (-OH) group and the oxygen of the carbonyl
(C=O), which belongs to a carboxylic group37, as shown in
Figure 2 by a dotted line. In the spectrum of CA, the band
associated with stretching vibrations of hydroxyl is located
at 3630 cm-1. The introduction of one lithium ion causes a
molecule distortion and a new band associated with the OH
group appears at 3380 cm-1. This group is bonded to the Li
ion (Li-O-H) and becomes weak bonded with the neighbor
carboxylate through a hydrogen bond. This influence
becomes stronger and shifts the band to 2860 cm-1 when two
lithium ions are present. Moreover, there is an increase in
the CO stretching vibration. In CA, the band associated with
such bonds is located at 1129 cm-1 and the maximum shift is
1147 cm-1 and 1163 cm-1 in CA/Li and CA/2Li complexes,
respectively. This means that there is competition between
the hydrogen bond and the COH bond. The calculated
bond order (which gives the strength of a bond) seems to
corroborate this conclusion. In Table 1, we can see that, as
the content of Li increases, the C=O bond becomes weaker
and the hydrogen bond becomes stronger.
The amount of lithium interferes with the bands in the
region between 500 and 1000 cm-1. In CA, the bands in
this region are well defined, with vibrations at 560 cm-1 and
685 cm-1 (COOH and COH), whereas new medium peaks
can be observed at 630 cm-1 owing to the LiO stretching
modes. As the number of lithium ions increases, there is an
increase in the number of bands, which will then overlap.
Figure 4 shows the Raman bands for the CA molecule.
The intense vibration at 3506 cm-1 is associated with the
asymmetric stretching of OH. It is also possible to observe
the CH2 symmetric and asymmetric stretching at 3120 cm-1
and 3085 cm-1, respectively. Other minor bands are also
observed near 1480 cm-1 for the CH
2
symmetric stretching
Figure 2. Optimized structures of complexes formed by one
molecule of citric acid with (a) one and (b) two lithium atoms; the
citric acid model is also shown for each complex (color online).
Figure 3. Simulation of infrared spectra for pure citric acid and the
complexes with one and two lithium ions (color online).
Table 1. Bond orders for selected groups.
Bond
complex
C=O OH
(hydrogen bond)
AC/Li 1.735 0.113
AC/2Li 1.506 0.202
2014; 17(3) 553Theoretical Investigation of Geometric Configurations and Vibrational Spectra in Citric Acid Complexes
and further bands appear at 1444 cm-1 and 1175 cm-1 that
are also associated with CH
2
. In the spectra of complexes of
CA/Li and CA/2Li, we observe LiO vibrations in the region
between 300 and 600 cm-1 (see Figure 4).
Comparison of the simulated Raman spectra of lithium
and CA complexes with pure CA shows that the intense
bands located in the region between 3600 and 3000 cm-1
in pure CA practically disappear in the complexes. The
interval 600-300 cm-1, which in pure CA shows some bands
that are rather weak and not well defined, in the complexes
has well defined and intense bands; as the amount of metal
increases, these bands become more intense. The bands are
attributed to the vibrations of the bonds between the lithium
and oxygen atoms, as previously stated.
The marked differences between the complexes CA/Li
and CA/2Li are better observed in the infrared spectra. The
Raman spectra are very similar, revealing few distinctions
except with regard to the intensities of the bands, which it
is not possible to determine experimentally. Finally, from
these observations we can say that the band that better
distinguishes the complexes CA/Li and CA/2Li is located
at 1275 cm-1 and 1355 cm-1, respectively, which refers to the
ligand character of the carboxylate group. Moreover, the
band at 2851 cm-1 in the infrared spectrum of the complex
CA/2Li indicates a stronger hydrogen bond of the OH
stretching vibration. However, it is known that this region is
experimentally masked because of the interference of H
2
O
present in the air or absorbed on the surface of the sample
while spectroscopy is performed.
In complexes between citric acid and barium, two H+
were subtracted from carboxyl groups for each barium ion
present. Considering the metallic ion size, as expected, the
CA molecule was less distorted with Ba2+ than it was with Li+
(1.35 and 0.68 Å atomic radii, respectively), which allowed
for the formation of hydrogen bonds to fully minimize the
energy. Figure 5 presents the optimized structures of the
different ratios between CA/Ba complexes, which are 1/1,
2/1, and 3/1. Kakihana and Yoshimura39 found that the CA/
Metal ratio 3/1 is the most stable.
Figure 6 shows a comparison of the simulated
infrared spectra of pure CA, CA/Ba, 2CA/Ba, and 3CA/
Ba complexes. Note that each proportion has its own
characteristics. The region between 3500 cm-1 and 2250 cm-1
refers to the OH stretching vibrations. The band around
3600 cm-1 is due to free OH bonds, whereas the peaks
at 3500 cm-1, 3000 cm-1, and 2500 cm-1 for the 2CA/Ba
complex and at 3500-3250 cm-1 and 2245 cm-1 refer to
OH influenced by hydrogen bonds. Experimentally, these
bands are hidden because of the water vibrations and
adsorbed CO
2
[34]. Furthermore, the spectrum for the CA/
Ba complex shows the same shift towards smaller values
of wavenumbers for the C=O bond at 1730 cm-1 as in the
CA/Li complex, and 2AC/Ba and 3AC/Ba show a defined
peak at 1750 cm-1, associated with the presence of more
COOH groups. The asymmetrical and symmetrical COO
stretching shows a characteristic monodentate ligand for
CA/Ba (1683 cm-1 and 1269 cm-1), monodentate for 2AC/
Figure 4. Simulation of Raman spectra for pure citric acid and
complexes with one and two lithium ions (color online).
Figure 5. Optimized structure for the complex of a barium atom
with (a) one, (b) two, and (c) three molecules of citric acid; the
citric acid model is also shown for each complex (color online).
Figure 6. Simulation of IR absorption spectra for pure citric acid
and complexes of barium (color online).
554 Ferreira et al. Materials Research
Ba (1707 cm-1 and 1290 cm-1), and bidentate for 3AC/Ba
(1606 cm-1 and 1383 cm-1)37. Moreover, when the number
of citric acid molecules increases, the quantity of BaO
and hydrogen bonds also increases, which creates a larger
distortion of the complex.
Another effect occurs in the bands associated with the
OH, CH, and COH bonds spread in the interval from 1000
to 1500 cm-1. In pure CA, these bands are well defined and
intense once all bonds are vibrating with nearly the same
frequency, as described above. With the addition of Ba2+
ion, some bonds experience more interference than others,
and this causes them to vibrate at different frequencies. This
effect can be observed in the spectra of the 2CA/Ba and 3CA/
Ba complexes through the spreading and decreasing of band
intensities related to these bonds. Further, vibrations below
1000 cm-1 due to COOH (scissoring), COH (wagging), and
BaO are spread out in this region.
Figure 7 shows a comparison of the Raman spectra
for the three simulated proportions. We can identify some
unique features in each case. In the CA/Ba case, unlike the
other cases, the spectrum shows the most intense bands in
the region within the range of 750–1500 cm-1. These bands,
as previously stated, are related to COH, CCO, and BaO
vibrations. As we have concluded for the CA+Li complexes,
the identification of the different types of CA+Ba complexes
can be better accomplished by comparisons of IR spectra.
An estimate for the most stable lithium and barium
citrates based on the total energy of the complexes was done
using the following expressions:
∆E = (CA/nLi)
total energy
– n.(Li+)
total energy
- K
1
(2)
and
∆E = (nCA/Ba)
total energy
– n.(CA)
total energy
- K
2
(3)
where n is the number of Li+ ions (1, 2) or is the
number of CA molecules (1, 2, 3) in the complexes.
As K
1
and K
2
are constants, which are irrelevant to
comparison of the different complexes, they were
set to zero; K
1
= (CA)
total energy
+ n.(H+)
total energy
and
K
2
= 2.(H+)
total energy
+ (Ba+2)
total energy
(where (H+)
total energy
is
equal to zero). ∆E is the variation of total energy between
the complex and the complex’s components: the larger |∆E|
indicates the most stable complex. The results, shown in
Figure 8, indicate that AC/Li and 3CA/Ba are probably the
most stable complexes, which in the case of barium citrates
is in agreement with experimental findings39.
Altogether, our results provide information about the
intermediate complexes formed in the MPPM synthesis
route. However, it is well known that in an experimental
spectrum certain factors have an influence, which were
not taken into account in our simulations, such as the
interference of the solvent, the presence of impurities, and
contaminants such as H
2
O and CO
2
. Thus, we believe that,
although the bands we mention may be disguised by these
interferences, they remain prominent in experimental results
and help to distinguish the different complexes.
4. Conclusions
In this study, we presented a set of features for lithium
and barium citric acid complexes that allow geometric
configurations to be identified from comparisons with the
associated bands for infrared and Raman spectra. In this
context, IR is preferred over Raman spectra, as it offers
features that enable better comparisons. The main features
are associated with the vibrations of the groups OH, C=O,
COOH, COO-, COH and M-O (M=metal). In comparing the
spectra, we focused our attention on the vibrations of the
groups of C=O, COO- and COH when they are in the region
1000 to 2000 cm–1, which is less susceptible to interference
from contaminants.
Further, the most stable complex was estimated based
on total energy calculations. For lithium citrates, the CA:Li
proportion 1:1 should be the most stable. For barium
citrates, the estimate indicates that the CA:Ba ratio 3:1 is
probably the most stable complex, which is supported by
experimental results.
Acknowledgements
This research was supported by resources supplied by
the Center for Scientific Computing (NCC/GridUNESP)
of São Paulo State University (UNESP). The authors
acknowledge the scholarships UNESP (2010/15588) and
Fundação de Amparo à Pesquisa do Estado de São Paulo
(FAPESP) (grant 2011/022455) (RMF), Coordenação de
Aperfeiçoamento do Pessoal do Ensino Superior (CAPES)
(ABN) and Instituto Nacional de Ciência e Tecnologia de
Materiais em Nanotecnologia (INCTMN) (CFOG/ABN),
and FAPESP for the grant 2007/080720 (PNLF/MM).
Figure 7. Simulation of Raman spectra for pure citric acid and
complexes of barium (color online).
Figure 8. Total energy difference estimate for the most stable lithium
and barium citrates complexes (color online).
2014; 17(3) 555Theoretical Investigation of Geometric Configurations and Vibrational Spectra in Citric Acid Complexes
References
1. Motta M, Deimling CV, Saeki MJ and Lisboa PN Fo.
Chelating agent effects in the synthesis of mesoscopic-size
superconducting particles. Journal of Sol-Gel Science and
Technology. 2008; 46(2):201-207. http://dx.doi.org/10.1007/
s10971-007-1673-0
2. Jafar MH. LiMn
0.3
Co
0.3
Ni
0.3
Cr
0.1
O
2
cathode materials prepared
via sol-gel and combustion methods: a comparison. AIP
Conference Proceedings. 2011; 1400(1):280-285. http://dx.doi.
org/10.1063/1.3663128
3. Wang DQ, Cao LY, Huang JF and Wu JP. Synthesis and
electrochemical properties of LiV
3
O
8
via an improved sol-gel
process. Ceramics International. 2012; 38(4):2647-2652.
http://dx.doi.org/10.1016/j.ceramint.2011.11.030
4. Chiang C, Shei CY, Wu SF and Huang YT. Preparation
of high-purity Tl-based ‘1223’ superconductor phase
by modified Pechini process in water solution. Applied
Physics Letters. 1995; 58(21):2435-2437. http://dx.doi.
org/10.1063/1.104865
5. Kakihana M. Sol-gel preparation of high temperature
superconducting oxides. Journal of Sol-Gel Science and
Technology. 1996; 6(1):7-55. http://dx.doi.org/10.1007/
BF00402588
6. Pathak LC and Mishra SK. A review on the synthesis of
Y-Ba-Cu-oxide powder. Superconductor Science and
Technology. 2005; 18:67-89. http://dx.doi.org/10.1088/0953-
2048/18/9/R01
7. Coey JMD, Viret M and von Molnár S. Mixed-valence
manganites. Advances in Physics. 1999; 48(2):167-293. http://
dx.doi.org/10.1080/000187399243455
8. Pullar RO. Hexagonal ferrites: a review of the synthesis,
properties and applications of hexaferrite ceramics. Progress
in Material Sciences. 2012; 57(7):1191-1334. http://dx.doi.
org/10.1016/j.pmatsci.2012.04.001
9. Eerenstein V, Mathur ND and Scott JF. Multiferroic and
magnetoelectric materials. Nature. 2006; 442(7104):759-765.
PMid:16915279. http://dx.doi.org/10.1038/nature05023
10. Zampieri M, Lazaro SR, Paskocimas CA, Ferreira AG,
Longo E and Varela JA. Structural analysis of Ti and Pb
citrate using NMR and FT-Raman signals and quantum
mechanics simulations. Journal of Sol-Gel Science and
Technology. 2006; 37:9-17. http://dx.doi.org/10.1007/s10971-
005-4892-2
11. Coskuner O and Jarvis EAA. Coordination studies of Al-
EDTA in aqueous solution. Journal of Physical Chememistry
A. 2008; 112(12):2628-2633. PMid:18293948. http://dx.doi.
org/10.1021/jp7098575
12. Kovács A, Nemcsok DS and Kocsis T. Bonding interactions
in EDTA complexes. Journal of Molecular Structure:
THEOCHEM . 2010; 950(1-3):93-97. http://dx.doi.
org/10.1016/j.theochem.2010.03.036
13. Yang W and Mortier WJ. The use of global and local molecular-
parameters for the analysis of the gas-phase basicity of amines.
Journal of American Chemical Society. 1986; 108(19):5708-
5711. PMid:22175316. http://dx.doi.org/10.1021/ja00279a008
14. Mineva T. Selectivity study from the density functional
local reactivity indices. Journal of Molecular Structure:
THEOCHEM. 2006; 762(1):79-86. http://dx.doi.org/10.1016/j.
theochem.2005.08.044
15. Cruz J, Martínez-Aguilera LMR, Salcedo R and Castro
M. Reactivity properties of derivatives of 2-imidazoline:
An ab initio DFT study. International Journal of Quantum
Chemistry. 2001; 85(4-5):546-556. http://dx.doi.org/10.1002/
qua.10018
16. Bronze-Uhle ES, Batagin A No, Lavarda FC and Graeff CFO.
Ionizing radiation induced degradation of poly (2-methoxy-
5-(2-ethyl-hexyloxy) -1,4-phenylene vinylene) in solution.
Journal of Applied Physics. 2011; 110:073510. http://dx.doi.
org/10.1063/1.3644946
17. Stewart JJP. Optimization of parameters for semiempirical
methods V: Modification of NDDO approximations
and application to 70 elements. Journal of Molecular
Modeling. 1989; 13(12):1173-1213. PMid:17828561
PMCid:PMC2039871. http://dx.doi.org/10.1007/s00894-
007-0233-4
18. Stewart JJP. MOPAC2009. Colorado Springs: Stewart
Computational Chemistry; 2008. Available from: . Access in: 03/02/2011.
19. Parr RG and Yang W. Density-functional theory of atoms and
molecules. New York: Oxford University Press; 1989.
20. Dunbar RC, Oomens J, Orlova G and Bohme DK. IRMPD
spectroscopic investigation of gas-phase complexes of
deprotonated penicillin G with Ba2+, Zn2+ and Cd2+. International
Journal of Mass Spectrometry. 2011; 308:330-337. http://
dx.doi.org/10.1016/j.ijms.2011.08.018
21. Stephens PJ, Devlin FJ, Chablowski CF and Frisch MJ.
Ab initio calculation of vibrational absorption and circular
dichroism spectra using density functional force fields. Journal
of Physical Chemistry. 1994; 98:11623-11627. http://dx.doi.
org/10.1021/j100096a001
22. Sousa SF, Fernandes PA and Ramos MJ. General performance
of density functionals. Journal of Physical Chemistry
A. 2007; 111:10439-10452. PMid:17718548. http://dx.doi.
org/10.1021/jp0734474
23. Hehre WJ. A guide to molecular mechanics and quantum
chemical calculations. California: Wavefunction; 2003.
24. Hay PJ and Wadt WR. Ab initio effective core potentials
for molecular calculations. Potentials for K to Au
including the outermost core orbitals. Journal of Chemical
Physics. 1985; 82:299-300. http://dx.doi.org/10.1063/1.448975
25. Feller D. The role of databases in support of computational
chemistry calculations. Journal of Computational
Chemistry. 1996; 17:1571-1586. http://dx.doi.org/10.1002/
(SICI)1096-987X(199610)17:13%3C1571::AID-
JCC9%3E3.0.CO;2-P
26. Levine IN. Quantum chemistry. New York: Prentice Hall; 1999.
27. Lewars EG. Computational chemistry: introduction to the
theory and applications of molecular and quantum mechanics.
New York: Springer; 2010.
28. National Institute of Standards and Technology (NIST).
Computational chemistry comparison and benchmark
database. Available from: . Access in: 15/06/2011.
29. Schmidt MW, Baldridge KK, Boatz JA, Elbert ST, Gordon
MS, Jensen JH et al. General atomic and molecular
eletronic-structure system. Journal of Computational
Chemistry. 1993; 14(11):1347-1363. http://dx.doi.org/10.1002/
jcc.540141112
30. Allouche AR. Gabedit: a graphical user interface for
computational chemistry softwares. Journal of Computational
Chemistry. 2011; 32(1):174-182. PMid:20607691. http://
dx.doi.org/10.1002/jcc.21600
31. Bode BM and Gordon MS. MacMolPlt: a graphical user
interface for GAMESS. Journal of Molecular Graphics and
http://dx.doi.org/10.1007/s10971-007-1673-0
http://dx.doi.org/10.1007/s10971-007-1673-0
http://dx.doi.org/10.1063/1.3663128
http://dx.doi.org/10.1063/1.3663128
http://dx.doi.org/10.1016/j.ceramint.2011.11.030
http://dx.doi.org/10.1063/1.104865
http://dx.doi.org/10.1063/1.104865
http://dx.doi.org/10.1007/BF00402588
http://dx.doi.org/10.1007/BF00402588
http://dx.doi.org/10.1088/0953-2048/18/9/R01
http://dx.doi.org/10.1088/0953-2048/18/9/R01
http://dx.doi.org/10.1080/000187399243455
http://dx.doi.org/10.1080/000187399243455
http://dx.doi.org/10.1016/j.pmatsci.2012.04.001
http://dx.doi.org/10.1016/j.pmatsci.2012.04.001
http://dx.doi.org/10.1038/nature05023
http://dx.doi.org/10.1007/s10971-005-4892-2
http://dx.doi.org/10.1007/s10971-005-4892-2
http://dx.doi.org/10.1021/jp7098575
http://dx.doi.org/10.1021/jp7098575
http://dx.doi.org/10.1016/j.theochem.2010.03.036
http://dx.doi.org/10.1016/j.theochem.2010.03.036
http://dx.doi.org/10.1021/ja00279a008
http://dx.doi.org/10.1016/j.theochem.2005.08.044
http://dx.doi.org/10.1016/j.theochem.2005.08.044
http://dx.doi.org/10.1002/qua.10018
http://dx.doi.org/10.1002/qua.10018
http://dx.doi.org/10.1063/1.3644946
http://dx.doi.org/10.1063/1.3644946
http://dx.doi.org/10.1007/s00894-007-0233-4
http://dx.doi.org/10.1007/s00894-007-0233-4
http://openmopac.net
http://openmopac.net
http://dx.doi.org/10.1016/j.ijms.2011.08.018
http://dx.doi.org/10.1016/j.ijms.2011.08.018
http://dx.doi.org/10.1021/j100096a001
http://dx.doi.org/10.1021/j100096a001
http://dx.doi.org/10.1021/jp0734474
http://dx.doi.org/10.1021/jp0734474
http://dx.doi.org/10.1063/1.448975
http://dx.doi.org/10.1002/(SICI)1096-987X(199610)17:13%3C1571::AID-JCC9%3E3.0.CO;2-P
http://dx.doi.org/10.1002/(SICI)1096-987X(199610)17:13%3C1571::AID-JCC9%3E3.0.CO;2-P
http://dx.doi.org/10.1002/(SICI)1096-987X(199610)17:13%3C1571::AID-JCC9%3E3.0.CO;2-P
http://cccbdb.nist.gov/vibscalejust.asp
http://cccbdb.nist.gov/vibscalejust.asp
http://dx.doi.org/10.1002/jcc.540141112
http://dx.doi.org/10.1002/jcc.540141112
http://dx.doi.org/10.1002/jcc.21600
http://dx.doi.org/10.1002/jcc.21600
556 Ferreira et al. Materials Research
Modelling. 1998; 16(3):133-138. http://dx.doi.org/10.1016/
S1093-3263(99)00002-9
32. Kakihana M, Arima M, Nakamura Y, Yashima M and
Yoshimura M. Spectroscopic characterization of precursors
used in the pechini-type polymerizable complex processing of
barium titanate. Chemistry of Materials. 1999; 11(2):438-450.
http://dx.doi.org/10.1021/cm9806681
33. Sambrano JR, Zampieri M, Ferreira AG and Longo E. Ab initio
study and NMR analysis of the complexion of citric acid with
ion lithium. Journal of Molecular Structure. 1999; 493:309-
318. http://dx.doi.org/10.1016/S0166-1280(99)00268-7
34. Socrates G. Infrared and raman characteristic group frequency.
New York: Wiley; 2011.
35. Smith AL. The coblentz society desk book of infrared spectra.
Kirkwood: The Coblentz Society; 1982. p. 1-24.
36. Chen HJ, Jian PC, Chen JH, Wang L and Chiu WY. Nanosized-
hybrid colloids of poly(acrylic acid)/titania prepared via in situ
sol-gel reaction. Ceramics International. 2007; 33(4):643-653.
http://dx.doi.org/10.1016/j.ceramint.2005.12.012
37. Kakihana M, Nagumo M, Okanoto M and Kakihana H.
Coordination structures for uranyl carboxylate complexes in
aqueous-solution studied by IR and carbon-13 NMR spectra.
Journal of Physical Chemistry. 1987; 91(24):6128-6136. http://
dx.doi.org/10.1021/j100308a015
38. Stokenhuber M, Mayer H and Lercher JA. Preparation
of barium titanate from oxalates. Journal of American
Ceramics Society. 1993; 76(5):1185-1190. http://dx.doi.
org/10.1111/j.1151-2916.1993.tb03738.x
39. Kakihana M and Yoshimura M. Synthesis and characteristics
of complex multicomponent oxides prepared by polymer
complex method. Bulletin of the Chemical Society of
Japan. 1999; 72(7):1427-1443. http://dx.doi.org/10.1246/
bcsj.72.1427
http://dx.doi.org/10.1016/S1093-3263(99)00002-9
http://dx.doi.org/10.1016/S1093-3263(99)00002-9
http://dx.doi.org/10.1021/cm9806681
http://dx.doi.org/10.1016/S0166-1280(99)00268-7
http://dx.doi.org/10.1016/j.ceramint.2005.12.012
http://dx.doi.org/10.1021/j100308a015
http://dx.doi.org/10.1021/j100308a015
http://dx.doi.org/10.1111/j.1151-2916.1993.tb03738.x
http://dx.doi.org/10.1111/j.1151-2916.1993.tb03738.x
http://dx.doi.org/10.1246/bcsj.72.1427
http://dx.doi.org/10.1246/bcsj.72.1427