Spectro-ellipsometric studies of activated sol-gel thin films to detect Cu2+ ions in aqueous solutions

Article abstract of DOI:10.1016/j.jpcs.2006.03.016

Thickness, refractive index n and extinction coefficient k of eriochrome cyanine R (ECR) dye doped mesostructured silica film (ECRMSSF) dip-coated on silicon wafers without prior removal of surfactant have been determined in the visible range of the spectrum by spectroscopic-ellipsometry (SE). ECRMSSF was used as sensing detection layer of Cu²⁺ in aqueous solutions of known pH. The dependence of optical parameters on the pH range of the aqueous phase (pH=6-8) was investigated. The detection reversibility of ECRMSSF was demonstrated. An electron paramagnetic resonance spectrum demonstrated the complexation of Cu²⁺ by the entrapped ECR.

Full text of DOI:10.1016/j.jpcs.2006.03.016

ARTICLE IN PRESS
Journal of Physics and Chemistry of Solids 67 (2006) 1775–1780
www.elsevier.com/locate/jpcs
Spectro-ellipsometric studies of activated sol–gel thin films
to detect Cu2+ ions in aqueous solutions
Ã
Othman Belhadj Mileda, , Cedric Boissiereb, Clement Sanchezb, Jacques Livageb
´ ´
a
´ ´
Faculte des Sciences de Monastir, Departement de physique, Avenue de l’environnement, 5019 Monastir Tunisie
` ´ ´
Laboratoire Chimie de la Matiere Condensee,UMR 7574 Universite Paris VI, 4 place Jussieu, 75252 Paris, France
b
Received 15 August 2005; received in revised form 20 March 2006; accepted 29 March 2006
Abstract
Thickness, refractive index n and extinction coefficient k of eriochrome cyanine R (ECR) dye doped mesostructured silica film
(ECRMSSF) dip-coated on silicon wafers without prior removal of surfactant have been determined in the visible range of the spectrum
by spectroscopic-ellipsometry (SE). ECRMSSF was used as sensing detection layer of Cu2+ in aqueous solutions of known pH. The
dependence of optical parameters on the pH range of the aqueous phase (pH ¼ 6–8) was investigated. The detection reversibility of
ECRMSSF was demonstrated. An electron paramagnetic resonance spectrum demonstrated the complexation of Cu2+ by the entrapped
ECR.
r 2006 Elsevier Ltd. All rights reserved.
Keywords: A. Interfaces; A. Nanostructures; A. Thin films; B. Sol–gel growth; D. Optical properties
1. Introduction
contains ‘‘phase’’ information contained in D, which makes
the measurements very sensitive. The advantage of the
ellipsometric measurement is the ability to obtain the real
part and the imaginary part of dielectric function
simultaneously; therefore, the value of the complex
refractive index N (N ¼ n+ik) as a function of wavelength
can also be obtained. The real part n of refractive index
generally relates to dispersion, while the imaginary part k
provides a measure of the dissipative rate of the wave in the
medium [3]. Absorption coefficient (a) can be deduced from
the complex dielectric function (e) using the following
equation: a ¼ 4pk=l ¼ ð2p=lnÞe2, where e1 ¼ n2 ꢀ k2, e2 ¼
2nk and l is the wavelength (nm). The e1 and e2 format are
most appropriate when discussing the physics. The n and k
format are used extensively in engineering applications [2].
So, SE measurements can be used to confirm absorption
investigations and more importantly, give particular access
to optical parameters and thickness.
Spectro-ellipsometry (SE) is a non-destructive technique
well adapted to thin film characterisations [1–3]. Ellipso-
metry uses light of known polarisation incident on a
surface under study and detects the polarisation state of the
reflected light. Incident light is usually linearly polarised
and reflected light has elliptical polarisation, where
coordinates of incoming field vector E are the p- and
s-polarisations. Ellipsometry measures changes of the light
polarisation by the reflection at an interface. The measured
values are related to the ratio of Fresnel reflection
coefficient rp and rs for p and s-polarisation light,
respectively:
rp=rs ¼ tan ceiD,
(1)
where rp and rs are the reflection coefficients of the electric
field on the film surface with polarisation parallel and
perpendicular to the plane of incidence, respectively. From
Eq. (1), the ratio is seen to be a complex number, thus it
The goal of this work is to confirm, by SE measurements,
the previous spectrophotometric results of eriochrome
cyanine R (ECR) dye doped mesostructured silica Cu2+
detection (ECRMSSF-Cu2+) [4]. Results of cited studies
showed that ECR incorporated in mesostructured silica
Ã
Corresponding author. Tel.: +216 73 500 274; fax: +216 73 500 278.
E-mail address: Oth.Bhm@fsm.rnu.tn (O.B. Miled).
0022-3697/$ - see front matter r 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jpcs.2006.03.016

ARTICLE IN PRESS
O.B. Miled et al. / Journal of Physics and Chemistry of Solids 67 (2006) 1775–1780
1776
matrix without prior removal of surfactant is sensitive with
adjusted pH (ECRMSSF-pH) to Cu2+ (ECRMSSF-pH-
Cu2+). Cationic surfactant cetyltrimetylammonium bro-
mide (CTAB) seems to favour the trapping of dye in host
matrix limiting at the same time the leaching of the dye.
The ECR molecule, frequently used in chemical analysis as
an indicator for complexometry and particularly for Cu2+
ion detection [5–12], has two carboxylic groups, one
hydroxyl, one carbonyl and one sulphonic group which
ensure compatibility with polar solvents that are commonly
used in sol–gel synthesis [10].
solution containing 5 mmol/l of Cu2+ at l ¼ 580 nm (Fig.
5a) confirmed that ECR is well entrapped in the
mesostructured film with limited leaching.
2. Experimental details
A prehydrolysed solution was prepared by refluxing for
1 h an ethanolic solution containing TEOS, deionised water
and hydrochloric acid in the following molar ratio: 1
TEOS: 3 EtOH, 5 ꢁ 10ꢀ5 HCl: 1 H2O. Then, CTAB was
dissolved in ethanol and added to the prehydrolysed
solution together with an additional amount of water and
HCl. Typically, the final molar ratio was 1 TEOS: 20
EtOH: 0.004 HCl: 5 H2O: 0.1 CTAB. The pH of sol was
around 2. The dye ECR (1 mmol/l of sol) was introduced in
the sol and left under stirring for an hour. The resulting
solution had a red coloration and the pH of mixture was
around 4 and this pH accelerates solidification of the sol
(we can conclude that ECR acts as a basic indicator when it
is added to the described sol and thus can be explained by
the hydroxyl group of ECR). Stabilisation of the sol was
achieved by bringing back pH of the sol to 2 by adding one
drop of 2 M HCl. Sol has now a yellow coloration. The
doped and un-doped sols were then stirred for 4 days at
room temperature and a series of films were then dip-
coated on silicon wafers. Films were deposited at a velocity
of 3.1 mm/s at room temperature and left in an oven during
one night at 60 1C followed by 4 h at 130 1C.
With the aim of confirming the metal-complexing dye
ECR to Cu2+, an electron paramagnetic resonance (EPR)
Cu2+ test was performed using a Bruker spectrometer
(n ¼ 9:241 GHz) at room temperature. An amount of sol
was left at room temperature one night and the gel was
then dried at 130 1C and powdered. ECRMSS powder was
suspended in a Cu2+ solution (1 mmol/l) stabilised at
pH ¼ 8 and left under stirring for an hour and then filtered
and finally extensively rinsed with water and dried. The
powder has now a blue coloration.
Sol–gel processing is a versatile procedure for making
advanced materials. The recent discovery of mesostruc-
tured silica formed by the cooperative self assembly of
silicates and surfactants has opened up a new range of
possibilities. The highly porous nature of these materials
makes them excellent hosts for sensing molecules, since the
species to be sensed can easily diffuse towards the sensing
centre. Sol–gel technology has proved very useful to
prepare pH detectors [13–15], electrochemical sensors
[16,17] and especially optical sensors [4,18,19] for the
detection of the metal ions in aqueous solutions. In a
typical sol–gel process, the precursor is subjected to a series
of hydrolysis and polymerisation reaction to form a
colloidal suspension, or sol.
For this study, the sol was prepared via acidic hydrolysis
condensation of tetraethoxysilane (TEOS) in the presence
of CTAB as the structure-directing agent. The surfactant
was not removed and has been exploited for the low-
temperature synthesis of optical materials and it acts here
as a barrier to limit the leaching of the dye.
3. Results and discussion
Ellipsometric measurements were performed in the
visible range (350–650 nm) for different incident angles
651, 701 and 751. Film thickness of dip-coated silicon
wafers (ECRMSSF) was around 180 nm. For un-doped
mesostructured silica film (MSSF), the refractive index
versus wavelength is shown in Fig. 1. Refractive index n
decreases from n ¼ 1:50 (350 nm) to n ¼ 1:47 (650 nm)
which verifies the Cauchy equation in the absence of
absorption (k ¼ 0):
ECR entrapped in sol–gel composites displays
a
selectivity for complexing metal ions which is different
from that of the dye in solution [7]. In a recent paper,
Wright and Higginson [12] examine the effects of pH on a
gel prepared by sol–gel route using various values of the
water–alkoxide (tetramethoxysilane) ratio and showed that
ECR sensitivity increases on incorporation of 10%
methyltriethoxysilane. With the aim of detecting scandium
(Sc(III)), Park and Cha [8] conclude that the complex
between ECR and Sc(III) in the presence of surfactant
(CTAB) is very stable and more sensitive than in its
absence. They interpret this mechanism by the formation
of Sc(III)–ECR–CTAB ternary complex. Previous work [4]
using optical absorption performance of ECR to detect
Cu2+ after exposure for various time to a pH 8 buffered
nðlÞ ¼ A þ ðB=l2Þ þ ðC=l4Þ,
(2)
where the three parameters in the dispersion model are A,
B and C, and l is the wavelength. Refractive index of
MSSF seems greater than pure silica (n ¼ 1:457) in the
visible range. This could be due to the presence of CTAB.

ARTICLE IN PRESS
O.B. Miled et al. / Journal of Physics and Chemistry of Solids 67 (2006) 1775–1780
1777
1.5
1.495
1.49
0.005
0.0045
0.004
531nm
0.0035
0.003
1.485
1.48
0.0025
0.002
1.475
1.47
0.0015
0.001
1.465
350
400
450
500
550
600
650
Wavelength (nm)
0.0005
0
Fig. 1. Refractive index n of mesostructured silica film (MSSF) versus
wavelength showing monotonic decrease. Relative to pure silica (n ¼ 1:458
at l ¼ 600 nm), the surfactant seems to increase the refractive index.
350
400
450
500
550
600
650
Wavelength (nm)
Fig. 3. Extinction coefficient k versus wavelength (nm) of ECRMSSF (as
dip-coated) with a peak centred at 531 nm.
1.5
1.495
1.49
carrier solutions to the desired pH. Before each ellipso-
metric measurement, the films were dried. The SE results
relative to refractive index n and extinction coefficients k
versus wavelength were plotted in Figs. 4 and 5,
respectively, including those of ECRMSSF and MSS films.
In the visible range (350–650 nm), pH effect (when pH of
carrier solution increase from 6 to 9) decreases refractive
index of ECRMSSF and shifts the extinction coefficient
maximum from 532 to 432 nm. The absorbance at l ¼
432 nm increases as pH decreases. These results confirm the
previous spectrophotometer investigations [4] and are in
agreement with the work of Sommerdijk et al. [7] and
Villegas et al. [10].
1.485
1.48
543nm
1.475
1.47
In the following experiments, pH ¼ 8 was retained to
test the capability of ECRMSSF to detect Cu2+. After
exposure to pH ¼ 8, the film was successively immersed in
1 mmol/l of Cu2+ solution for 15 min, rinsed, dried and
then tested. Results are shown in Fig. 6. Absorption was
centred on 591 nm when the film was exposed to Cu2+
solution. We can summarise the behaviour of the film with
respect to wavelength by the following absorption maxima:
initial film (531 nm), exposed to pH ¼ 8 (432 nm) and
exposed to 1 mmol/l of Cu2+ (591 nm). A previous
absorption investigation [4] gives a peak centred at
582 nm when the dip-coated film was stabilised at pH ¼ 8
and exposed to Cu2+ solution. The two methods (spectro-
photometric and spectro-ellipsometric) seem confirmed the
ECR complexing metal ion Cu2+ around 590 nm. The
advantages that can actually be gained using the SE
method were the access to optical parameters as extinction
coefficient k(l) (which related to absorption coefficient)
and thickness.
1.465
350
400
450
500
550
600
650
Wavelength (nm)
Fig. 2. Refractive index n(l) of initial (before exposure to pH) ECR dye
doped mesostructured silica film (ECRMSSF). We note the appearance of
a dispersion peak at l ¼ 543 nm certainly due to the presence of the dye.
The ellipsometric spectra of refractive index n and the
extinction coefficient k of ECRMSSF before exposing to
pH are shown in Figs. 2 and 3, respectively. Refractive
index n shows a peak at 543 nm (n ¼ 1:475) certainly due to
the presence of dye. The corresponding extinction coeffi-
cient k shows a peak at 531 nm.
The response of ECRMSSF versus pH (ECRMSSF-pH)
of carrier solutions (distilled water) in the range pH ¼ 6–9
was performed using acetic acid and ammonia to adjust the

ARTICLE IN PRESS
O.B. Miled et al. / Journal of Physics and Chemistry of Solids 67 (2006) 1775–1780
1778
0.008
0.007
1.505
1.495
1.485
1.475
1.465
1.455
1.445
591nm
ECRMSSF stabilised at pH=8
and exposed to 1mmol/l
of Cu2+ for 15mn
0.006
0.005
0.004
0.003
0.002
0.001
0
430nm
pH=6
pH=7
pH=8
( a )
( b )
pH=9
( c )
( d )
( e )
ECRMSSF
( f )
350
400
450
500
550
600
650
Wavelength (nm)
350
400
450
500
550
600
650
Wavelength (nm)
Fig. 6. Extinction coefficient k(l) of ECRMSSF, ECRMSSF exposed to
different pH and ECRMSSF stabilised at pH ¼ 8 (ECRMSSF-pH ¼ 8)
and then exposed to 1 mmol/l of Cu2+ solution for 10 min (ECRMSSF-
pH ¼ 8-Cu).
Fig. 4. Comparison between refractive index n(l) of un-doped mesos-
tructured silica film (MSSF), ECR dye doped mesostructured silica film as
dip-coated (ECRMSSF) and ECRMSSF exposed to different pH: (a)
MSSF, (b) ECRMSSF (initial film before exposure to pH effect), (c)
ECRMSSF-pH ¼ 9, (d) ECRMSSF-pH ¼ 8, (e) ECRMSSF-pH ¼ 7 and
(f) ECRMSSF-pH ¼ 6.
0.007
0.006
0.005
0.007
0.006
0.005
0.004
432n
0.003
432nm
0.004
0.003
0.002
0.001
0
532nm
0.002
pH=6
pH=7
(a)
pH=8
pH=9
(b)
0.001
0
ECRMSSF
350
400
450
500
550
600
650
Wavelength (nm)
Fig. 7. Regeneration of film: extinction coefficient k versus wavelength of:
(a) ECRMSSF stabilised at pH ¼ 8 and then (b) successively exposed to
1 mmol/l of Cu2+ solution for10 min, followed by exposure to 2 M HCl
(placing the film in 2 M HCl displaced the copper ions and regenerated the
reagent) and finally stabilised at pH ¼ 8; regeneration of film was so
proved.
375
425
475
525
575
625
Wavelength (nm)
Fig. 5. Comparison between extinction coefficient k(l) of ECRMSSF and
ECRMSSF exposed to different pH in the range pH ¼ 6–9.
and again immersed in 1 mmol/l of Cu2+ solution for
20 min, rinsed, dried and then tested (Fig. 8). Reversibility
is proved with about 10 nm shift. Increase of peak intensity
was due to increased diffusion of Cu2+ ions to complexing
sites deeper within the bulk of the film in the greater
To test reversibility of detection, the same film was
immersed in acidic solution (2 M HCl) for 1 min (placing
the film in 2 M HCl displaced the copper ions and
regenerated the reagent), stabilised at pH ¼ 8 (Fig. 7)

ARTICLE IN PRESS
O.B. Miled et al. / Journal of Physics and Chemistry of Solids 67 (2006) 1775–1780
1779
0.014
0.012
0.01
584nm
0.008
0.006
0.004
0.002
0
(b)
591nm
(a)
350
400
450
500
550
600
650
700
Wavelength (nm)
Fig. 8. Reversibility of detection: extinction coefficient k versus wave-
length of: (a) ECRMSSF-pH ¼ 8-Cu2+ and (b) the same film of (a) was
used after ellipsometric test for regeneration (by exposition to 2 M HCl
solution) followed by stabilisation to pH 8 and finally re-exposed to Cu2+
solution.
Fig. 10. EPR test.
absorption increase not fully equilibrated after 5 min
exposure. This result is in good agreement with the work
of Sommerdijk [7] and Wright and Higginson [12] using
sol–gel-entrapped ECR. SE method used in the actual
work seems well adapted to dye-doped mesostructured thin
films absorption–dispersion investigations with a great
sensitivity.
The EPR spectra were recorded on a Bruker spectro-
meter at room temperature operating at 9.241 GHz and
100 kHz field modulation. The results are given in Fig. 10
which showed a well-resolved copper hyperfine features
and narrow EPR signals. The spectrum is characterised by
the parameters
1.52
1.51
1.5
1.49
(a)
1.48
1.47
1.46
1.45
1.44
(b)
(c)
542nm
g ¼ hn=beBres ðwith Bres ꢂ 3170 GaussÞ ꢂ 2:08,
jaj
618nm
¼ 2760 ꢀ 2620 ¼ 140 Gauss:
gbe
The EPR spectrum of Cu-ECR-MSS is typical of Cu2+
[20,21].
4. Conclusion
300
350
400
450
500
550
600
650
700
The ellipsometric spectra in the visible range of ECR dye
doped mesostructured silica sol dip-coated on silicon
wafers were studied for different incident angles 651, 701
and 751 by spectroscopic ellipsometry. The refractive index
n(l), extinction coefficient k(l) and thickness of the films
have been obtained. Analysis of dye-doped sol–gel films
shows the presence of two phenomena which are dispersion
and dissipation related, respectively, by the refractive index
n and the extinction coefficient k. The pH range (6–8)
seems to be the optimal absorption pH. Detection of Cu2+
ions by sol–gel dip-coated thin films on silicon wafers and
powder was examined, respectively, by spectro-ellipso-
metric measurements and RPE tests which confirm the
previous spectrophometric studies and evanescent optical
Wavelength (nm)
Fig. 9. Comparison between refractive indexes n(l) of: (a) ECRMSSF,
(b) ECRMSSF-pH ¼ 8 and (c) ECRMSSF-pH ¼ 8-Cu2+
.
exposure time to the Cu2+ solution. Refractive index n in
this case is illustrated by Fig. 9. We note a dispersion peak
at 618 nm when film was exposed to Cu2+
In this study, ECR seems well immobilised in the matrix.
In fact, previous work using optical absorbance of ECR to
detect Cu2+ as a function of exposure time to a pH 8
buffered solution containing 5 mmol/l of Cu2+ showed an
.

ARTICLE IN PRESS
O.B. Miled et al. / Journal of Physics and Chemistry of Solids 67 (2006) 1775–1780
1780
fibre wave sensing of the detection capability of ECR
activated mesostructured silica sol–gel matrix. The aims of
previous [4] and actual studies using spectroscopic inves-
tigations constitute a step to enhance sensitivity of optical
fibre Cu2+ continuous sensing based on optical evanescent
field and activated sol–gel detection layer. For enhancing
sensor sensitivity, the future investigations are planned to
explore the structure phase and pore size distributions of
dip-coated films, particularly the effect of the mole ratio of
dye molecule to surfactant [22].
[9] M.A. Javaid, P.J. Keay, A generic technique for coating doped
sol–gel films onto the inside of tubes for use as colorimetric sensors,
J. Sol–Gel Sci. Technol. 17 (2000) 55–59.
[10] M.A. Villegas, L. Pascual, S.E. Paje, M.A. Garcia, J. LIopis,
Eriochrome cyanine doped sol–gel coatings. Optical behaviour
against pH, J. Eur. Ceram. Soc. 20 (2000) 1621–1628.
[11] H. Dacres, R. Narayanaswamy, A new optical sensing reaction for
nitric oxide, Sensors Actuators B 90 (2003) 222–229.
[12] J.D. Wright, N.A.C. Higginson, Effects of matrix variations on pH
and Cu2+ sensing properties of sol–gel entrapped Eriochrome
Cyanine R, J. Mater. Chem. 14 (2004) 201–208.
[13] O. Belhadj Miled, H. Ben Ouada, J. Livage, pH sensor based on
a
detection sol–gel layer onto optical fiber, Mater. Sci. Eng.
References
C-Biomimetic Supramol. Syst. 21 (1-2) (2002) 183–188.
[14] O. Belhadj Miled, D. Grosso, C. Sanchez, J. Livage, An optical fibre
pH sensor based on dye doped mesostrucured silica, J. Phys. Chem.
Solids 65 (2004) 1751–1755.
[1] A. Bourgeois, A.B. Bruneau, S. Fisson, B. Demarets, D. Grosso,
F. Cagnol, C. Sanchez, J. Rivory, Determination of pore size
distribution in thin organized mesoporous silica films by spectro-
scopic ellipsometry in the visible and infrared range, Thin Solid Films
447–448 (2004) 46–50.
[15] M. Garcıa-Heras, C. Gil, N. Carmona, J. Faber, K. Kromka, M.A.
´
Villegas, Optical behaviour of pH detectors based on sol–gel
technology, Anal. Chim. Acta 540 (1) (2005) 147–152.
[2] W.H. Nosal, D.W. Thomson, L. Yan, S. Sakar, A. Subramanian,
J.A. Woollam, UV–Vis–infrared optical and AFM study of spin-cast
chitosan films, Colloids Surf. B: Biointerfaces 43 (2005) 129–135.
[3] G. Wang, F. Gan, Optical parameters and absorption studies of azo
dye-doped polymer thin films on silicon, Mater. Lett. 43 (2000) 6–10.
[4] O. Belhadj Miled, C. Sanchez, J. Livage, Spectroscopic studies and
evanescent optical fibre wave sensing of Cu2+ based on activated
mesostructured silica matrix, J. Mater. Sci. 40 (17) (2005) 4523–4530.
[16] A. Jimenez-Morales, P. Arand, J.C. Galvan, Nanocomposite
´ ´
materials based organopolysiloxane/macrocyle systems for electro-
chemical sensors, J. Mater. Process. Technol. 143–144 (2003) 5–10.
[17] A. Jimenez-Morales, J.C. Galvan, P. Aranda, A new silver-ion
´ ´
selective sensor based on a polythiacrown-ether entrapped by sol–gel,
Electrochim. Acta 47 (13-14) (2002) 2281–2287.
[18] J.D. Wright, N.A.C. Higginson, The adaptation of a piezo-optical
gas sensing system to detect copper ions in aqueous solution, Sensors
Actuators B: Chemical 99(1) (2004) 15, 134–140.
[5] M. Ahmad, R. Narayanaswamy,
A flow-cell optosensor for
monitoring aluminium(III) based on immobilised eriochrome cyanine
R (ECR) and reflectance spectrophotometry, Sci. Total Environ. 163
(1995) 221–227.
[19] M. Janotta, M. Karlowatz, F. Vogt, B. Mizaikoff, Sol–gel based mid-
infrared evanescent wave sensors for detection of organophosphate
pesticides in aqueous solution, Anal. Chim. Acta 496 (1-2) (2003)
339–348.
[6] S. Boudra, J.M. Bosque-Sendra, M.C. Valencia, Determination of
vanadium by solid-phase spectrophotometry after its preconcentra-
tion as an eriochrome cyanine R complex on a dextran-type
[20] R. Ahmamouch, S. Arsalane, M. Kacimi, M. Ziyad, Synthesis and
properties of copper-hafnium triphosphateCu1Hf2(PO4)3, Mater.
Res. Bull. 32 (6) (1997) 755–761.
exchanger, Talanta 42 (1995) 1525–1532.
[7] N.A.J.M. Sommerdijk, A. Poppe, C.A. Gibson, J.D. Wright,
Unexpected complexation behaviour of a sol–gel immobilised dye:
the development of an optical copper(II) sensor, J. Mater. Chem. 8
(3) (1998) 565–567.
[21] T. Breyer, F.W. Breitbarth, W. Volgesberger, EPR spectroscopic
investigation of the sol–gel transition in modified tetraethylorthosi-
licate gels, J. Colloid Interface Sci. 266 (2003) 153–159.
[22] M.H. Huang, F. Kartono, B. Dunn, J.I. Zink, G. Valverde, J. Garcia,
Hexagonal to lamellar mesostructural changes in silicate films caused
by organic additives, Chem. Mater. 14 (2002) 5153–5162.
[8] C.-i. Park, K.-W. Cha, Spectrophotometric determination of
scandium (II) in monazite after separation using amberlite IRC 718
chelating resin, Bull. Korean Chem. Soc. 20 (12) (1999).