Growth and characterization of Fe3+-doped bis(thiourea)zinc(II) chloride crystals

Article abstract of DOI:10.1016/j.saa.2011.05.029

Fe³⁺-doping at ～10 mol% in aqueous medium during crystal growth by slow evaporation solution method in bis(thiourea)zinc(II) chloride (BTZC) leads to form a new compound C₂H₈Cl ₂N₄S₂Zn

Full text of DOI:10.1016/j.saa.2011.05.029

Spectrochimica Acta Part A 79 (2011) 1648–1653
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
Growth and characterization of Fe3+-doped bis(thiourea)zinc(II) chloride crystals
K. Nithya, B. Karthikeyan, G. Ramasamy, K. Muthu, S.P. Meenakshisundaram∗
Department of Chemistry, Annamalai University, Annamalainagar 608002, India
a r t i c l e i n f o
a b s t r a c t
Fe3+-doping at ∼10 mol% in aqueous medium during crystal growth by slow evaporation solution method
in bis(thiourea)zinc(II) chloride (BTZC) leads to form a new compound C2H8Cl2N4S2Zn0.93Fe0.07 (BTZCF)
which crystallizes in orthorhombic structure with centrosymmetric space group Pnma though the parent
compound BTZC crystallizes in noncentrosymmetric structure with space group Pn21a. The interesting
feature observed in this new crystal is that though it crystallizes in centrosymmetric structure, it exhibits
positive SHG result (weak signal), quite likely due to possible surface effects or internal stress. The cal-
culated first-order hyperpolarizability is 1.457 × 10−30 esu which is ∼5.5 times that of urea. Fe3+-doping
enhances the transmittance to a significant extent. Comparison of the thermal analysis results by DSC
reveals the incorporation of dopant into the crystalline matrix. The high resolution XRD studies reveal
that the crystalline quality is improved considerably when the doping level is reached to ∼10 mol%.
© 2011 Elsevier B.V. All rights reserved.
Article history:
Received 30 October 2010
Received in revised form 7 May 2011
Accepted 16 May 2011
Keywords:
Crystal structure
High resolution X-ray diffraction
Single crystal growth
FT-IR spectroscopy
Thiourea
Computational calculation
1. Introduction
favourable for downward growth because the crystals would have
a large meniscus angle [17]. Carbon nanotubes (CNTs) filled with
Bis(thiourea)zinc(II) chloride (BTZC), Zn(CS(NH2)2)Cl2 is
a
metals like Fe, Co, Ni, etc. have attracted extensive research inter-
ests due to their unique electronic, magnetic and nonlinear optical
properties, as well as the potential applications in magnetic data
storage, xerography and magnetic resonance imaging [18,19]. The
influence of Fe-doping on the structural, crystallinity and optical
properties of ZnO and TiO2 thin films were reported [20,21]. In the
present investigation, we report the growth, crystal structure and
potential semiorganic NLO material, which exhibits a low angu-
lar sensitivity and hence, useful for type-II second harmonic
generation(SHG) [1–4]. An X-ray crystal structure analysis of
dichlorobisthiourea zinc [(NH2)2CS]2ZnCl2, has shown that the
complex is molecular, each zinc atom being tetrahedrally co-
ordinated to two chlorine and two sulphur atoms [5]. Influence
of pH on the growth and characteristic of nonlinear optical
BTZC single crystal has been reported [6–8]. BTZC crystals were
grown by Sankaranarayanan–Ramasamy method and character-
ized [9]. Nucleation studies of l-arginine doped with BTZC in
supersaturated aqueous solutions [10] and vibrational spectra [11]
have been investigated. It is well proven fact by both theory
and experiment that for having a SHG character, the material
should have a noncentrometric crystal structure. However, a few
reports are available in the literature which show that the cen-
trosymmetric crystals [12–16] or films [15] also exhibit NLO
properties.
optical properties of Fe3+-doped BTZC (C2H8Cl2N4S2Zn0.93Fe0.07
BTZCF) crystals.
;
2. Experimental procedure
2.1. Synthesis and crystal growth
BTZC was synthesized using AR grade zinc chloride and thiourea
in a stoichiometric ratio 1:2.
ZnCl2 + 2(CS(NH2)2) → Zn(CS(NH2)2)2Cl2
The influence of heavy Fe-doping on LiNbO3 fibers was investi-
gated by means of the infrared technique and polishing. LiNbO3
fibers doped with larger amounts of iron would exhibit larger
residual stress. Also, during growth, heavy Fe-doping would
flatten the curvature of the growth interface and increase the
chance of bubble-entrapment. Nevertheless heavier Fe-doping was
The purity was increased by successive recrystallization pro-
cesses. Crystals were grown by the slow evaporation solution
growth technique. Doping of Fe3+ (∼10 mol%) in the form of FeCl3
was done during the crystallization process. Photographs of pure
and doped crystals are shown in Fig. 1.
2.2. Characterization studies
∗
Single crystal X-ray diffraction studies were carried out using
Bruker AXS (Kappa Apex II) X-ray diffractometer. Single crystal
Corresponding author. Tel.: +91 4144 221670; mobile: +91 9443091274.
E-mail address: aumats2009@gmail.com (S.P. Meenakshisundaram).
1386-1425/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2011.05.029

K. Nithya et al. / Spectrochimica Acta Part A 79 (2011) 1648–1653
1649
Fig. 2. Powder XRD patterns of (a) BTZCF (∼10 mol%) and (b) pure BTZC.
Table 1
Values of lattice constants a (A), b (A), c (A) and cell volume V (A) .
3
˚
˚
˚
˚
Property
Pure BTZC
BTZCF (∼10 mol%)
a
b
c
12.80
12.62
5.78
12.99
12.72
5.87
Fig. 1. Photographs of as grown crystals: (a) BTZCF (∼10 mol%) and (b) pure BTZC.
V
934
971
3. Results and discussion
XRD data were collected on a diffraction system, which employs
graphite monochromated MoK␣ radiation (ꢀ = 0.71073 A). The
˚
3.1. X-ray analysis for crystal structure and perfection
structures were solved and refined by full matrix least squares
on F2 with Wingx software package utilizing SHELXS-97 and
SHELXL-97 modules. The plots for the structures were created
with DIAMOND software. All non-hydrogen atoms were refined
anisotropically.
The powder XRD pattern is recorded by PANalytical X’per PRO
X-ray diffractometer and analyzed by using ‘powder X’ refinement
software and it shows a single phase with a slight reduction in peak
intensities accompanied by a small shift in their positions (Fig. 2).
These observations could be due to change in structure. The radius
of the dopant, Fe3+ (64 pm) is nearly close compared with that of
Zn2+ (74 pm). Hence, it is reasonable to believe that the dopant
incorporation into the BTZC crystalline matrix is quite feasible. The
lattice constants ‘a’, ‘b’, ‘c’ and the cell volume ‘V’ are increased for
the doped specimens (Table 1) indicating a new structure which is
confirmed by the single crystal XRD.
The powder XRD pattern is recorded by PANalytical X’per PRO
X-ray diffractometer and analyzed by using ‘powder X’ refine-
ment software and it shows a single phase with a slight reduction
in peak intensities accompanied by a small shift in their posi-
tions. A multicrystal XRD developed at NPL [22] has been used to
record high-resolution diffraction curves (DCs). Fourier transform
infrared spectra were recorded using AVATAR 330 FT-IR spectrom-
eter. UV–vis spectra were recorded using CARY 5E UV–VIS–NIR
spectrophotometer. The surface morphology was observed using
a JEOL JSM 5610 LV scanning electron microscope which has a res-
olution of 3.0 nm and an acceleration voltage of 0.3–30 kV having
the maximum magnification of 2,00,000 times. ICP studies were
recorded by using Optima 5000 DV series spectrometer. DSC curves
were recorded on a SDT Q600 (TA instrument) thermal analyzer.
DSC analysis of pure and Fe3+ doped BTZC crystals was carried
out between 50 and 300 ◦C in the nitrogen atmosphere. The sec-
ond harmonic generation test on the crystal was performed by the
Kurtz powder SHG method. An Nd:YAG laser with modulated radi-
ation of 1064 nm was used as the optical source and directed on
the powdered sample through a filter. The HFC (Hartree–Fock) cal-
culation was performed using GAUSSIAN 03 [23] program package
on a personal computer without any constraint on the geometry
using LANL2 DZ basis set [24]. By the use of GAUSSVIEW molecular
visualization program [25] optimized structure of the molecule has
been visualized.
The crystal structure analysis by single crystal XRD was carried
out for BTZCF (∼10 mol% doping) and the ORTEP diagram is shown
in Fig. 3. The empirical formula C2H8Cl2N4S2Zn0.93Fe0.07 well sup-
Fig. 3. ORTEP diagram of BTZCF (∼10 mol%).

1650
K. Nithya et al. / Spectrochimica Acta Part A 79 (2011) 1648–1653
Table 2
Crystal data and structure refinement for BTZCF (∼10 mol%).
Empirical formula
Formula weight
C2H8Cl2N4S2Zn0.93Fe0.07
287.85
Temperature
Wavelength
293 (2) K
˚
0.71073 A
Crystal system, space group
Unit cell dimensions
Orthorhombic, Pnma
◦
˚
˚
a = 12.9953 (7) A ˛ = 90 , b = 12.7229 (6) A
◦
◦
˚
ˇ = 90 , c = 5.8754 (3) A ꢁ = 90
3
˚
Volume
971.43 (9) A
Z, Calculated density
Absorption coefficient
F(0 0 0)
4, 1.968 mg/m3
3.385 mm−1
575
Crystal size
Theta range for data collection
Limiting indices
0.30 mm × 0.20 mm × 0.20 mm
3.14–24.99◦
−15 ≤ h ≤ 15, −14 ≤ k ≤ 15, −6 ≤ l ≤ 6
4774/887 [R(int) = 0.0165]
100.0%
Reflections collected/unique
Completeness to theta = 24.99
Absorption correction
Max. and min. transmission
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2 sigma(I)]
R indices (all data)
Semi-empirical from equivalents
0.5509 and 0.4300
Full-matrix least-squares on F2
887/5/73
1.081
R1 = 0.0176, wR2 = 0.0428
R1 = 0.0204, wR2 = 0.0451
0.309 and −0.230 e A−3
Largest diff. peak and hole
ports the incorporation of a small but significant amount of dopant
(Table 2). The specimen crystallizes with centrosymmetric space
group Pnma and refined to a low R value of 0.0176 (final R indices).
This analysis clearly indicates that the incorporation of Fe3+ in BTZC
leads to change the space group from noncentrosymmetric (Pn21a)
[26] to centrosymmetric (Pnma) which in turn enforced to predict
the loss of second harmonic generation. However, the prediction
could not be found true as some weak SHG was found as observed
in the forthcoming NLO studies.
Fig. 5. FT-IR spectrum of (a) BTZCF (∼10 mol%) and (b) pure BTZC.
figure, the DC contains a single sharp peak and indicates that the
specimen is free from structural grain boundaries. The FWHM (full
width at half maximum) of the curve is 36ꢀꢀ which is close to that
expected from the plane wave dynamical theory of X-ray diffraction
for a perfect crystal [27].
Fig. 4b shows the high-resolution diffraction curve (DC)
recorded for a typical BTZC single crystal specimen using (2 0 0)
diffracting planes in symmetrical Bragg geometry by employing
the multicrystal X-ray diffractometer with MoK␣1 radiation [22].
The solid line (convoluted curve) is well fitted with the experi-
mental points represented by the filled circles. On deconvolution
of the diffraction curve, it is clear that the curve contains two addi-
tional peaks, which are 1018ꢀꢀ and 168ꢀꢀ away from the main peak.
These two additional peaks correspond to two internal structural
low angle boundaries whose tilt angles are 1018ꢀꢀ and 168ꢀꢀ from
the main crystal block. The FWHM (full width at half maximum) of
the main peak and the low angle boundaries are, respectively 344ꢀꢀ
and 730ꢀꢀ and 44ꢀꢀ. Though the specimen contains low angle bound-
aries, the relatively low angular spread of around 2000ꢀꢀ (around
half a degree) of the diffraction curve and the low FWHM values
show that the crystalline perfection is reasonably good. The effect
of such very low angle boundaries may not be very significant in
many device applications, but for applications like phase match-
ing, it is better to know these minute details regarding crystalline
perfection.
Fig. 4a shows the high-resolution diffraction curve (DC)
recorded for a typical SEST grown BTZC crystal with heavy Fe3+
-
doping (∼10 mol%) using (2 0 0) diffracting planes. As seen in the
3.2. FT-IR spectra
A close observation of FT-IR spectra of pure and doped speci-
mens reveal that doping generally results in small shifts in some of
the characteristic vibrational frequencies (Fig. 5). The CN stretching
frequencies of thiourea (1089 and 1472 cm−1) are shifted to higher
frequencies for pure and Fe3+-doped BTZC crystals (∼1102 and
∼1495 cm−1). The CS stretching frequencies (1417 and 740 cm−1
)
are shifted to lower frequencies (∼1410 and ∼716 cm−1) for pure
and doped samples. These observations suggest that metal coor-
dinate with thiourea through sulphur. This is well evidenced by
ORTEP diagram.
Fig. 4. HRXRD curves recorded for (a) BTZCF (∼10 mol%) and (b) pure BTZC.

K. Nithya et al. / Spectrochimica Acta Part A 79 (2011) 1648–1653
1651
Fig. 6. UV–vis–NIR spectrum of (a) BTZCF (∼10 mol%) and (b) pure BTZC.
Fig. 7. DSC spectrum of (a) BTZCF (∼10 mol%) and (b) pure BTZC.
3.3. UV–visible spectra
3.6. SEM and EDS
The UV spectrum reveals that the cut off wavelength is ∼350 nm.
Absorption is minimum in the 350–1200 nm region. The doping
does not destroy the optical transmission. Interestingly, heavy dop-
ing (∼10 mol%) of Fe3+ enhances the transmittance to a significant
extent as shown in Fig. 6.
The investigation of the influence of Fe3+-doping on the surface
morphology of BTZC crystal faces reveals the formation of structure
defect centers (Fig. 8a). SEM images are taken on the flat surface
indicating the external morphologies. The particles like images on
the surface of crystals could be due to surface roughness. SEM
photographs of the incorporation of Fe3+ in the crystalline matrix
result in more scatter centers than those of the undoped specimen
(Fig. 8b). BTZC shows the highest surface roughness due to bunched
steps or perhaps even macro steps. The characteristic flower like
morphology of BTZC crystal is not disturbed by doping.
3.4. ICP analysis
The amount of dopant in BTZCF specimens is estimated
using inductive coupled plasma (ICP) technique and the actual
dopant concentration in the crystalline matrices were found to be
The incorporation of Fe3+ into the crystalline matrix was con-
firmed by EDS performed on BTZCF (∼10 mol%) crystals (Fig. 9). It
appears that the accommodating capability of host crystal is limited
and only a small quantity is incorporated into the BTZC crystalline
matrix. Further, analysis of surface at different sites indicates that
the incorporation is non-uniform over the surface, connected with
adsorption mechanisms at ledges of steps. We have observed this
non-uniform incorporation of dopant in many studies [28,29].
0.004 ppm and 0.007 ppm, respectively for ∼1 and ∼10 mol% Fe3+
-
doped specimens, respectively. The foreign metal ion entering into
the BTZC crystal matrix is small but significant. Further, the final
dopant concentration within the host lattice is not proportional
to the concentration of the crystallization process since the host
crystal can accommodate the dopant only to a limited extent.
3.5. Thermal analysis (differential scanning calorimetry)
3.7. NLO studies
The DSC analysis of BTZC and BTZCF (∼10 mol%) crystals was car-
ried out between 50 and 300 ◦C in the nitrogen atmosphere (Fig. 7).
DSC shows that there is no physically adsorbed water in the molec-
ular structure of crystals. This study reveals the purity and the sharp
endothermic peaks indicate the melting point of the material. No
decomposition up to melting point ensures the suitability of the
material for application in lasers where the crystals are required to
withstand high temperatures. The melting point of BTZC (∼258 ◦C)
crystal is lowered considerably by doping with Fe3+ (∼249 ◦C) and
this observation indirectly supports the incorporation of Fe3+ in the
host crystal.
The doubling of frequency was confirmed by the green color of
the output radiation whose characteristic wavelength is 532 nm.
The relative measured SHG output is ∼0.16 times that of the stan-
dard KDP. Though it is very weak, it is interesting to observe the
SHG in the Fe3+-doped specimen BTZCF as it exhibits centrosym-
metry. Similar observation was also reported in centrosymmetric
dibenzyl croconate (DBCC) crystal which was found to exhibit a SHG
intensity 0.6 times that of urea [16]. Very weak signals result from
possible surface effects [30] or local noncentrosymmetry caused by
defects [31].

1652
K. Nithya et al. / Spectrochimica Acta Part A 79 (2011) 1648–1653
due to the Kleinman symmetry [25,32]. It can be given in the lower
tetrahedral format.
1
2˛ˇ
1
6
E = E0 − ꢂ˛F˛
−
F˛Fˇ
−
ˇ˛ˇꢁ F˛FˇFꢁ + · · ·
where E0 is the energy of the unperturbed molecule, F˛, the
field at the origin of ꢂ˛, ˛˛ˇ and ˇ˛ˇꢁ are the components of
dipole moment, polarizability and first order hyperpolarizabilities,
respectively.
ˇ0 = (ˇx2 + ˇy2 + ˇz2)1/2
and
ˇx = ˇxxx + ˇxyy + ˇxzz
ˇy = ˇyyy + ˇxxy + ˇyzz
ˇx = ˇzzz + ˇxxz + ˇyyz
The calculated first-order hyperpolarizability (ˇ) of the spec-
imen is 1.457 × 10−30 esu which is ∼5.5 times that of urea. The
maximum ˇ is due to the behaviour of non zero ꢂ values. In the
present study, micro ˇ is large but macro SHG is small. This type of
behaviour of high ˇ with small SHG has also been reported for the
compound ephendrinium [33]. It appears that molecular orienta-
tions cancel out charge transfer resulting in a very low SHG.
4. Conclusions
Fig. 8. SEM micrographs of (a) BTZCF (∼10 mol%) and (b) pure BTZC.
The
crystal
structure
of
the
Fe3+-doped
BTZC
(C2H8Cl2N4S2Zn0.93Fe0.07) has been reported for the first time. The
incorporation of dopant into the crystalline matrix is further con-
firmed by EDS, ICP, FT-IR and thermal analysis. The lattice constant
and cell volume are increased by doping with Fe3+ in comparison
with the pure BTZC. Doping does not destroy the optical trans-
mission. Fe3+-doping leads to change the noncentrosymmetric
structure of pure BTZC to a centrosymmetric structure. The first
order molecular hyperpolarizability is large for pure BTZC and the
macro SHG is small for doped specimen. The HRXRD studies reveal
that the specimen is free from structural grain boundaries.
Acknowledgements
Fig. 9. EDS spectrum of BTZCF (∼10 mol%).
The authors thank Dr. G. Bhagavannarayana, Materials Charac-
terization Division, NPL, New Delhi for HRXRD measurement. One
of the authors, K. Muthu is thankful to CSIR, New Delhi, for the
award of a Senior Research Fellowship.
Quantum mechanical calculations are used for predicting the
molecular NLO properties of molecules. Theoretical calculations
were performed using ab inito and Hartree–Fock level with
LANL2DZ as the basis set (Table 3). First hyperpolarizability is a
third rank tensor that can be described by 3 × 3 × 3 matrix. The 27
components of the 3D matrix can be reduced to 10 components
Appendix A. Supplementary material
CCDC 788613 contains the supplementary crystallographic
data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data request/cif.
Table 3
Calculated ˇ components and ˇtot (×10−30) value of pure BTZC (in esu).
References
1
2
3
4
5
6
7
8
9
ˇxxx
ˇxxy
ˇxyy
ˇyyy
ˇxxz
ˇxyz
ˇyyz
ˇxzz
ˇyzz
ˇzzz
ˇtot
205.717
176.153
−77.099
−330.925
198.247
0.738
[1] S. Velsko, Laser Program Annual Report, Lawrence UCRL-JC 105000, Lawrence
Livemore National Laboratory, Livemore, CA, 1990.
[2] G. Xing, M. Minhua, Z. Zongshu, D. Xu, Chin. J. Lasers 14 (1987) 357–363.
[3] H.O. Marcy, L.F. Warren, M.S. Webb, C.A. Ebbers, S.P. Velsko, G.C. Kennedy, G.C.
Caterlla, Appl. Opt. 31 (1992) 5051–5060.
[4] P.A. Angeli Mary, S. Dhanuskodi, Cryst. Res. Technol. 36 (2001) 1231–1237.
[5] N.R. Kunchur, M.R. Truter, J. Chem. Soc. 504 (1958) 3478–3484.
[6] R. Rajasekaran, P.M. Ushashree, R. Jayavel, P. Ramasamy, J. Cryst. Growth 229
(2001) 563–567.
−73.055
−4.308
−61.919
−155.342
[7] S. Moitra, T. Kar, Opt. Mater. 30 (2008) 1621–1624.
[8] R. Rajasekaran, R. Mohankumar, R. Jeyavel, P. Ramasamy, J. Cryst. Growth 252
(2003) 317–327.
10
11
1.457 × 10−30

K. Nithya et al. / Spectrochimica Acta Part A 79 (2011) 1648–1653
1653
[9] S.K. Kushwaha, N. Vijayan, G. Bhagavannarayana, Mater. Lett. 62 (2008)
[21] L. Xu, X. Li, J. Cryst. Growth 312 (2010) 851–855.
3931–3933.
[22] Krishan Lal, G. Bhagavannarayana, J. Appl. Crystallogr. 22 (1989) 209–215.
[23] Gaussian 03 Program, Gaussian Inc., Wallingford, CT, 2004.
[24] H.B. Schlegel, J. Comput. Chem. 3 (1982) 214–218.
[25] A. Frisch, A.B. Nielson, A.J. Holder, GAUSSVIEW User Manual, Gaussian Inc.,
Pittsburgh, PA, 2000.
[26] R. Uthrakumar, C. Vesta, C. Justin Raj, S. Krishnan, S. Jerome Das, Curr. Appl.
Phys. 10 (2010) 548–552.
[27] B.W. Batterman, H. Cole, Rev. Mod. Phys. 36 (1964) 681–717.
[28] S.P. Meenakashisundaram, S. Parthiban, G. Madhurambal, R. Dhanasekaran, S.C.
Mojumdar, J. Therm. Anal. Calorim. 94 (2008) 15–20.
[10] T. Balu, T.R. Rajasekaran, P. Murugakoothan, Physica B 404 (2009) 1813–1818.
[11] M. Quassaid, P. Becker, C. Carabatos-Nedelec, Phys. Status Solidi B 207 (1998)
499–507.
[12] G. Wensheng, G. Fang, W. Chunsheng, L. Qitao, Z. Guangyong, D. Wang, S.
Zhongshu, Sci. China Ser. B: Chem. 45 (2002) 267–274.
[13] K.E. Rieckhoff, W.L. Peticolas, Science 147 (1965) 610–611.
[14] T. Ishihara, K. Koshino, H. Nakashima, Phys. Rev. Lett. 91 (2003) 253901.
[15] S.K. Mohd Shakir, K.K. Kushwaha, M. Maurya, G. Arora, J. Bhagavannarayana,
Cryst. Growth 311 (2009) 3871–3875.
[16] H.Y. Chen, Q. Fang, H. Lei, W.-T. Yu, X.-F. Cheng, Z.-Q. Liu, J. Mol. Struct. 871
(2007) 1–5.
[29] G. Bhagavannarayana, S.K. Kushwaha, S. Parthiban, G. Ajitha, S. Meenakshisun-
daram, J. Cryst. Growth 310 (2008) 2575–2583.
[17] Y.J. Lai, J.C. Chen, J. Cryst. Growth 212 (2000) 211–216.
[18] R.S. Ruoff, Nature 372 (1994) 731–732.
[30] M. Fleck, A.M. Petrosyan, J. Cryst. Growth 312 (2010) 2284–2290.
[31] A.M. Petrosyan, J. Cryst. Growth 311 (2009) 4538.
[19] S.A. Majetich, J.O. Artman, M.E. McHenry, N.T. Nuhfer, S.W. Staley, Phys. Rev. B
48 (1993) 16845.
[20] J.H. Jeong, J.H. Ahn, B. Hoon Kim, J. Kor. Phys. Soc. 46 (2005) 559–561.
[32] D.A. Kleinman, Physiol. Rev. 126 (1962) 1977–1979.
[33] M. Sylla, M. Giffard, G. Mabon, N. Cubillan, O. Castellano, J. Hernandez, H. Soscun,
X. Nguyen Phu, Chem. Phys. 330 (2006) 387–393.